SEMICENTENNIAL PUBLICATIONS 
 
 OF THE 
 
 UNIVERSITY OF CALIFORNIA 
 
 1868-1918 
 
 <:^i 
 
..-a • W\ 
 
 \aahsV1aXu 
 
 MISCELLANEOUS STUDIES 
 
 IN Cl X ~ i ^ oi-i 
 AGRICULTURE 
 
 AND 
 
 BIOLOGY 
 
 UNIVERSITY OF CALIFORNIA PRESS 
 
 BERKELEY 
 
 1919 
 
 '•'''■•Vvtrn.'vo^,.'', -i 
 

 o; •t »• 
 
 SET 25 1920 
 

 c 
 
 CONTENTS 
 
 Pages 
 A Synopsis of the Aphiditlae of California, by Albert 
 
 F. Swain 1-221 
 
 Mutation in jMatthiola, by Howard B. Frost 223-333 
 
 Ocean Temperatures, their Relation to Solar Radiation 
 
 and Oceanic Circulation, by George F. McEwen 335-421 
 
 Changes in the Chemical Composition of Grapes during 
 Ripening, by F. T. Bioletti, W. V. Cruess, and H. 
 Davi 423-450 
 
A SYNOPSIS 
 
 OF THE 
 
 APHIDIDAE OF CALIFORNIA 
 
 BY 
 
 ALBERT F. SWAIN 
 
 [University of California Publications in Entomology, Vol. 3, No. 1, pp. 1-221, pis. 1-17] 
 
A SYNOPSIS 
 
 OF THE 
 
 APHIDIDAE OF CALIFORNIA 
 
 BY 
 
 ALBERT F. SWAIN 
 
 CONTENTS 
 
 PAGE 
 
 Introduction 2 
 
 Classification 4 
 
 External anatomy 5 
 
 Biology 8 
 
 Economic considerations 9 
 
 Synopsis 10 
 
 Family Aphididae 10 
 
 Subfamily Aphidinae 11 
 
 Group Callipterina 12 
 
 Tribe Phyllaphidini 12 
 
 Tribe Callipterini 16 
 
 Tribe Chaitophorini 32 
 
 Group Laclinina 39 
 
 Tribe Pterocommini 39 
 
 Tribe Laclmini 43 
 
 Group Aphidina 52 
 
 Tribe Macrosiphini 52 
 
 Tribe Apliidini 87 
 
 Subfamily Pemphiginae 138 
 
 Group Hormaphidina 139 
 
 Group Pemphigina 140 
 
 Group Schizoneurina 147 
 
 Group Vaeunina 150 
 
 Subfamily Phylloxerinae 151 
 
 Group Chermisina _ 151 
 
 Group Phylloxerina 152 
 
2 MISCELLANEOUS STUDIES 
 
 PAGE 
 
 Appendix 1. Keys to the genera and tribes of Aphididae ; a translation from 
 
 P. Van der Goot 154 
 
 Appendix 2. Host plant list _ _ 159 
 
 Addenda _.. 178 
 
 Explanation of plates 180 
 
 Index to genera and species 215 
 
 INTRODUCTION 
 
 In recent years considerable attention has been paid to the 
 Aphididae in the United States, and in Europe as well, and a large 
 amount of literature is the result. In California, "W. T. Clarke was the 
 first to make any systematic studies of these insects and his paper, 
 published in 1903, embodies the results of these studies. He listed 
 forty-three species, ten of which were described as new. Two or 
 three of his new species are known at present, but the remainder are 
 unknown. Unfortunately his collection was destroyed in the eartli- 
 quake of April, 1906, so now it is practically impossible to determine 
 his new species .with any degree of accuracy. Following this, there 
 was a period of six years in which there were no publications con- 
 cerning the Aphididae of California, except some economic bulletins 
 from the Experiment Station. In 1909, both B. 0. Essig and W. M. 
 Davidson published the results of their earlier studies. Since then 
 both have added papers occasionally. During 1912 and shortl.y before, 
 Harold Morrison made an extensive study of the species in the vicinity 
 of Stanford University. He has kindly placed a report of his studies 
 in the author's hands, with permission to publish the records in tliis 
 paper. The author has been studying the Californian species con- 
 tinuously since 1914. 
 
 At present there are about one hundred and eighty species known 
 to occur in California. This number will undoubtedly be greatly 
 increased as further studies are made, since to date only a compara- 
 tively small part of the state has been covered by collectors. Very 
 extensive collections have been made in Ventura County and in the 
 vicinity of Pomona College, Los Angeles "County, by Essig. The San 
 Francisco Baj^ region, particularly in the vicinity of the Universit}' 
 of California and Stanford University, has been carefully surveyed 
 for aphids, collections having been made by Clarke, Davidson, Essig, 
 Morrison, Ferris, and the author. Davidson, Clarke, and Essig have 
 made a few observations in the Sacramento Valley, particularly in 
 Placer and Sacramento counties. The author made a number of 
 
.4 SYNOPSIS OF THE APHIDIDAE 3 
 
 observations in the vicinity of Fi-esno during May and June, 1915, and 
 more or less extensive observations and collections during 1916 and 
 1917 in San Diego, Riverside, Orange, Los Angeles, and San Bernar- 
 dino counties. In addition to these, reports come to the College of 
 Agriculture occasionally from tlie State Insectary and the various 
 countj' horticultural commissioners. A summary of the above state- 
 ments shows that extensive collecting has been done onlj' in the 
 territory adjacent to San Francisco Bay, and throughout southern 
 California. The whole northern half of the state, the great interior 
 valleys of the Sacramento and San Joaquin rivers, and the desert 
 sections of the southeastern part of the state are as j'et unexplored. 
 Undoubtedly many interesting species will be found in these parts. 
 
 The author wishes to express his appreciation of the aid rendered 
 by various people during the past three years of study. To Harold 
 Morrison of the Federal Board of Horticulture is due especial thanks 
 for his assistance during the early part of the author's study, for his 
 collection notes, and for the use of his extensive collections of Stanford 
 Univei'sity vicinity and Indiana ; to E. 0. Essig of the University of 
 California for his continuous advice and assistance, for the use of 
 his large collections of Californian species, and for the reading of 
 this manuscript; to W. M. Davidson of the Bureau of Entomology', 
 U. S. Department of Agriculture, for his many notes and deter- 
 minations and for the use of his collection ; to A. C. Baker, J. J. 
 Davis, C. P. Gillette, A. S. Maxson. E. M. Patch, and H. F. Wilson 
 for their many determinations and suggestions ; to R. W. Doane of 
 Stanford University for the permission to work over his collection 
 of Utah aphids and for permitting his students to use the keys included 
 in this paper, thereby finding the weak points in the keys; and finally 
 to G. F. Ferris of Stanford University for collections and advice. 
 
 Ill this paper tlie author has brought together all the present 
 records of California Aphididae. He has included keys for the 
 determination of the subfamilies, groups, genera, and species, together 
 with such illustrations as are necessary for an understanding of the 
 keys. The discussion of each species includes a bibliography of the 
 California literature (exclusive of the merely economic and popular), 
 together with a citation of the original description and the best 
 available description, a list of host plants and localities, and a dis- 
 cussion of the synonomy, life history, and habits so far as they are 
 known. The descriptions of certain species are not readil.y accessible 
 and of others not at all adequate. Such species have been redescribed 
 
4 MISCELLANEOUS STUDIES 
 
 by the author in so far as it was possible to obtain specimens. Inci- 
 dentally it may be stated that the author has personally collected by 
 far the larger number of the species recorded in this paper. In other 
 eases the fact is noted. A host plant index (appendix 2) is also 
 included. 
 
 The sj-stem of classification followed is the one most generally 
 accepted by American aphidologists at the present time. The keys 
 to the species have been formulated by the author, those to the genera 
 and higher groups have to a large extent been adapted from other 
 workers, particularly Wilson and Essig (Aphidinae), Borner (Phyl- 
 loxerinae), and Tullgreu (Pemphiginae). The papers of Baker, 
 Clarke, Davidson, Davis, Essig, Gillette, Oestlund, Patch, Pergande, 
 Williams, Wilson, and other American aphidologists have been found 
 invaluable. Of the works of the European aphidologists, those of 
 Borner, Buckton, Del Guercio, Koch, Mordwilko, Tullgreu, and Van 
 der Goot have been in constant use. The classification suggested by 
 Van der Goot ("Zur Systematik der Aphiden," in Tijdschrift vom- 
 Entomologie, vol. 56, p. 1913) has proved interesting, and although 
 the author has not felt at liberty to accept it in full, a translation of 
 his keys to the groups and genera has been included herewith (appen- 
 dix 1), which, it is hoped, will be of assistance in the making of 
 determinations. 
 
 CLASSIFICATION 
 
 The Aphididae belong to the order Homoptera, being closely 
 related to the Psyllidae, or jumping plant lice, the Aleyrodidae, or 
 wliite flies, and the Coccidae, or scale insects. The Aphididae, or 
 plant lice, are small, soft-bodied insects, ranging from less tlian one 
 to five or six millimeters in length. Typically there are four forms : 
 the apterous and the alate viviparous females, and the sexual forms, 
 the oviparous females and the males. There is considerable variation 
 from the above in different groups and species, as will be pointed 
 out under the discussions of the various species. The alate viviparous 
 females are the individuals most commonly taken by the collector and 
 the ones that usually show the best characters for determinations. 
 In the keys in this paper all characters refer to the alate viviparous 
 females (the alates) unless otherwise mentioned. 
 
A SYNOPSIS OF THE APHIDIDAE 5 
 
 EXTERNAL ANATOMY 
 
 The body^ consists typically of three divisions, the head, thorax, 
 and abdomen. In the apterous forms the mesothorax and metathorax 
 are closely fused with the abdomen, while the prothorax and head 
 are distinct. In the alate forms the mesothorax and metathorax are 
 fused together and appear as a distinct division, the body appearing 
 to consist of four divisions, viz., the head, the prothorax, the meso- 
 thorax and metathorax, and the abdomen. 
 
 The head bears a pair of compound eyes, usually three ocelli, a 
 pair of three to six jointed antennae, and the beak. Of these, the 
 antennae show the best characters for determinations, not only of 
 species but of higher groups. They are either mounted on distinct 
 tubercles (Macrosiphini, certain Callipterini) or appear to arise from 
 the front of the licad. They consist of from three to six segments, 
 the terminal one of which is usually provided with a projection or 
 spur. They are six-segmented in the Aphidinae (except Essigella and 
 Cerosipha), five- or six-segmented in the Pemphiginae (except in the 
 stem mothers of certain genera), and three-segmented in the Phyl- 
 loxerinae (except in Chermisina, in which the alate forms have five- 
 segmented and the sexual forms four-segmented antennae). The 
 spur of the terminal segment maj' be equal to or longer than the 
 segment (Aphidinae, in the Macrosiphini it attains its greatest length, 
 often being as much as ten times the length of the base) ; it may be 
 merely a short thumblike process (Pemphiginae, Laehnini, and cer- 
 tain Callipterini) ; or it may be apparently lacking (Phylloxerinae). 
 The two basal segments are always short, and quite regular in all 
 species. The remaining segments show the greatest diversity, par- 
 ticularly in number, size, and shape. Sensoria are always present on 
 some of the segments. There is one primarj- sensorium always present 
 at the distal end of the terminal segment, and when the antennae con- 
 sist of more than three segments, one also at the distal end of the 
 penultimate segment. These sensoria are fairly large and clear (some- 
 times furnished with a hairy fringe) and are more or less circular. 
 The accessory sensoria are a group of small indistinct sensoria, which 
 number from three to six, and which are located in close proximity 
 
 1 Por a fuller discussion of the external characters consult the followinfc 
 papers: Vickerey, R. A., A comparative stutly of the external anatomy of plant 
 lice, 12th Kept.' Minnesota State Entomologist 1908 ; Sanborn, C. E., Kansas 
 Aphididae, Kansas Univ. Sci. Bull., vol. 3, 1904; Mordwilko, Alexander, Keys to 
 the groups and genera of the Aphididae, Ann. Mus. Zool. Imp. Acad. Sci. St. 
 Petersburg, vol. 13, pp. 3fi2-3l'.4, 1908. 
 
6 MISCELLANEOUS STUDIES 
 
 to the primary sensorium on the terminal segment. Secondary sen- 
 soria are usually present in the alate forms, but oftentimes absent 
 in the apterae of certain species. When present they are always on 
 the third segment, but in antennae consisting of five or six segments, 
 they ma.y be present upon the fourth, fiftli, and even sixth segments. 
 In the Pemphiginae they are arch-like or half rings, or form complete 
 rings about the segments. In the Aphidinae they are circular, oval, 
 or transversely linear, but are never rings or half rings. The shape 
 and number vary considerably, and are of specific importance. The 
 number maj- vary from as few as tliree or four {MyzocaUis maureri 
 Swain), to as many as forty to fifty on the third segment, and many 
 also on the fourth and fifth {Myzns hraggii Gillette). Unfortunately 
 these highly important characters were overlooked or not taken into 
 consideration by the earlier workers. The beak is four-jointed and 
 seems to arise from between the fore legs. It is always present (except 
 in the sexes of certain of the Phylloxerinae), but is seldom of specific 
 importance (except to distinguish Aphis baker i Cowen from Aphis 
 senecio Swain, and in certain of the Lachini). It may be very short, 
 as in Aphis hakeri Cowen, where it reaches only slightly beyond the 
 first coxa, or it may be very long as in Stomaphis, where it is from 
 one and one-half to two times as long as the body. In leaf-feeding 
 species it is u.sually sliort, while in bark-feeding forms it is longer. 
 This is naturally necessary, for those that live on thick bark must have 
 a longer beak in order to reach throiigh to the plant juices. 
 
 The thorax consists of three divisions, the last two of which are 
 usually more or less fused together, and considered as one ; the two 
 divisions being called, in this paper, the prothorax and the thorax. 
 On the lateral margins of tlie prothoi'ax there is sometimes a pair of 
 small tubercles. These are not present in all species, however, and 
 they differ considerably in size in the various species. There are 
 three pairs of fairly long and slender legs (except in Phylloxerinae. 
 where the legs are greatly atrophied, approaching those of the 
 Coecinae in size). Typically the legs consist of four joints, the 
 coxa, the femora, the tibia, and the tar.sus. In some genera the tarsi 
 may be atrophied {Atarsos, Mastapoda). The comparative lengths 
 of the first and second segment of the tarsi are sometimes of generic 
 importance (Laehnini), and the comparative lengths of the hind tarsi 
 and the cornicles are oftentimes of specific importance (Aphis, Ptero- 
 cmnnia). A small empodial hair is found between the claws in the 
 Aphidinae. In the Callipterina it is leaf-shaped or spatula-like. In 
 
A SYNOPSIS OF THE APEIDIDAM 7 
 
 the Aphidina and Lachnina it is hair-like, usually being as long as the 
 claws (except in the Pterocommini, in which it is considerably 
 shorter than the claws). The wings are membraneous and hyaline 
 (except in certain Callipterini, Lachnini, and Macrosiphini), and are 
 held roof-like over the body when at rest (except Moncllm, Phyllox- 
 erinae, Hormaphidina, in which they lie flat on the abdomen). The 
 veins of the fore wings are as follows: the costal and subcostal are 
 almost parallel with the anterior margin; the radial extends from 
 the posterior margin of the stigma to the outer margin of the wing, 
 being either curved or straight ; the discoidals, three in number, extend 
 from the subcostal to the posterior margin of the wing. The outer 
 or third discoidal {media, cuhitus of some authors) may be simple 
 (Hormaphidina, Pemphigina), absent (Phylloxerinae), once-branched 
 (Schizoneurina), or twice-branched (Aphidinae, except Tox.optera). 
 On the anterior margin of the fore wing is a dusky spot located be- 
 tween the wing margin and the subcostal veins, and between the 
 distal ends of the costal and subcostal veins, known as the stigma or 
 Pterostigma. It is usuallj' trapezoidal in shape, and does not extend 
 to the tip of the wing (except in Longistigma and Mindarus, in which 
 it reaches well beyond the tip of the wing). The hind wings have 
 one longitudinal and either one or two transverse veins.- In the 
 Pemphiginae and Phylloxerinae dorsal wax glands are sometimes 
 present on the thorax, in which ease their number, shape, and posi- 
 tion are of more or less specific importance. 
 
 The abdomen consists of nine more or less similar segments. The 
 coloration of the various segments, especially in species in which the 
 color is variegated, is sometimes of specific importance. In certain 
 species wax glands are present on the abdomen (Phylloxerinae, and 
 particularly the stem mothers of Pemphiginae) and may be of use 
 in making determinations. In the Aphidinae the presence or absence 
 and location of small lateral and dorsal tubercles are often important. 
 The anal segment consists of an anal plate and a cavida. The cauda 
 may not be .separated from the abdomen (Pemphiginae, Lachnina), or 
 it may be short and conical (Aphidini), short and globular, being 
 constricted in the middle (Callipterina), or it may be long and 
 ensiform or sickle-shaped (Macrosiphini). The anal plate is usually 
 well rounded, being half-moon-shaped, or it may be emarginate or 
 bilobed (Callipterina). On the sixth (or fifth?) segment is a pair 
 
 2 For a full discussion of the venation see Patch, Edith M., Homologies of the 
 wing veins of the Aphididae, Psyllidae, Aleurodidae, and Coccidae, Ann. Eutom. 
 Soe. Am., vol. 2, pp. 101-13G, June, 1909. 
 
8 MISCELLANEOUS STUDIES 
 
 of short tubular processes, the coruieles (honej' tubes, nectaries of 
 some authors). These are quite valuable characters, both specific and 
 generic. In the Phylloxerinae and most of the Pemphiginae they are 
 lacking, but in the Aphidinae they are always present, and show a 
 great diversity of form. They may be merely pores (certain Callip- 
 terini, Ccrosipha cupressl Swain, Lachmis taxifolm Swain), they may 
 be cj'lindrical, yet quite short (certain Callipterini, Chaitophorini) ; 
 they may be short and cylindrical or conical (Aphidini) ; they may 
 be truncate, cone-shape (Lachnini) ; they may be clavate and long 
 (certain Callipterina, Pteroeommini, Maerosiphini) ; or they may be 
 long and cylindrical (particularly in Macrosiphum and 3Iyzits). 
 
 BIOLOGY 
 
 Considerable variety is exhibited in the habits, life liistory, and 
 methods of reproduction, as well as in the structure and body form. 
 Reproduction is almost entirely parthenogenetic, although certain 
 species at certain times have a sexual rei^roduetion. Fewer species 
 have sexual reproduction in California than in colder climates, due 
 to the fact that mild weather throughout the winter permits them to 
 live over, and hence tlie eggs are unnecessary. Many species produce 
 generation after generation parthenogenetically, and are most abun- 
 dant in the spring and early summer, but gradually disappear toward 
 midsummer, due partially to their predaceous and parasitic enemies, 
 and partially, undoubtedly, to the heat of the summer. Other species 
 regularly produce sexual forms in the fall, which lay eggs that hatch 
 the next spring. The forms hatching from the eggs are wingless 
 (except in Callipterini) and usually of a different form from the 
 later generations, and are known as the fundatrix or stem mother. 
 The fundatrix is alwaj'S viviparous. Her progeny consists either of 
 all apterous or partly apterous and partly alate viviparous females 
 (fundatrigenia), which in turn produce other generations of funda- 
 trigeniae. The last asexual generation in the fall, which gives birth 
 to the sexual forms (sexuales), are known as sexupara, and are usually 
 alate. Oftentimes in the second or third and even fourth generation 
 there is a definite migration from one species of host plant to another, 
 where the aphids live over the summer (virgogenia), the sexupara 
 returning to the original species of host in the fall to give birth to 
 the sexuales, which lay their eggs there. Aphis nvalifoliae Fitch rep- 
 resents an example of this habit, the winter host being apple, the 
 summer plantain. Oftentimes the fall migrants (sexupara) of certain 
 
A SYNOPSIS OF THE APEIDIVAE 9 
 
 species differ considerably in structure from the spring migrants 
 (fundatrigenia). This is particularly noticeable in the Pemphiginae. 
 Many species are confined throughout the season to one species of host, 
 others to one or two or a few species, while still others may live on 
 any of a number of hosts (Aphis senecio Swain, Rhopalosiphum per- 
 sicae (Sulz.)). All sustenance is derived from the plant juices of 
 the various hosts, but each species is usually confined more or less 
 definately to feeding on some certain part of the plant. Some live 
 entirely upon the leaves, some on the stems of the leaves and small 
 twigs, some on the trunks and larger branches, some on the roots, 
 some on the flower heads and racemes of the host, and still others 
 feed on almost any part of the plant. The greater number of species 
 are free living, but certain of the Aphidinae form pseudogalls (Aphis 
 pomi De Geer, Aphis nmlifoliae Pitch, Phyllaphis coweni (Cockerell) ), 
 while the Pemphiginae and Chermisina form true galls. Nearly all of 
 the Pemphigina spend at least part of the season on various species 
 of Populus, the Schizoneurina on UIiiius, while the Lachnini and 
 Chermisina are practically confined to the conifers. The Ajihidinae 
 are found mostly on deciduous trees and herbaceous plants, although 
 some live on conifers (Myzaphis abictinus (Walker), Nectarosiphon 
 morriscmi Swain). 
 
 ECONOMIC CONSIDERATIONS 
 
 From an economic standpoint most of the species are of no 
 importance, although there are manj' that are well known pests of 
 cultivated crops. For example the woolly apple aphis (Eriosoma 
 lanigera) is a world-wide pest of considerable importance to the apple. 
 The green and the rosy apjile aphis (Aphis pomi, A. malifoliae) do a 
 large amount of injury in certain localities, and are extremely difficult 
 to control. The rose aphis (Macrosiphum rosac) is known the world 
 over, and although living unprotected and easily killed with any of 
 the common contact insecticides, it is recognized by everyone who has 
 grown roses in the doorj'ard as an extremely troublesome pest. The 
 walnut aphis (Chromaphis juglandicola), the cabbage aphis (Aphis 
 brassicae) , the green peach aphis or greenhouse aphis (Rhopalosiphum 
 persicae) are all well known pests. The common contact insecticides 
 are usually efficient for their control. Many species are kept well 
 in check by their predaceous and parasitic enemies, the ladybirds, the 
 syrphid flies, the laeewings, and the braconids. Of the ladybirds, 
 probablj^ the most efficient in California are Coccinella californic<i 
 
10 MISCELLANEOUS STUDIES 
 
 Mann., Hippodamia convergens Guerin, and Scymnus nehulosns Le- 
 conte. Of the syrpliid flies, those consuming the largest number of 
 aphids and the most abundant in the state^ are Catabomba pyrastri 
 Osten-Sackeu, Allograpta oMiqiui Say, Syrphtis arcuatus Fallen, S. 
 amerlcanus Wied., 8. opinator Will., and Eupeodes volucris Osten- 
 Saeken. Chrysopa calif ornic<i, Coq. and Sympherobms angustus Banks 
 are the most important aphid enemies among the lacewings. Among 
 the Braconidae there are two very common species in California, 
 Lysiphlebus tcstaccipes Cressou and Diaretus rapae Curtiss. Others 
 have been reared by the author and will be mentioned later. The 
 author wishes to thank Dr. L. 0. Howard and Mr. A. B. Gahan of 
 the Bureau of Entomology for their kindness in identifying the 
 various hymenopterous parasites of aphids sent to them. 
 
 SYNOPSIS 
 Family Aphididae Passeriui 
 
 Passerini, Gli Afidi, 1860. 
 The family Aphididae Passerini is divided into three subfamilies 
 (following Alexander Mordwilko), which are: Aphidinae Buckton, 
 Pemphiginae Mordwilko, and Phylloxerinae Dreyfus. Van der Goot 
 considers but two subfamilies: Aphidinae v. d. G. and Chermisinae 
 V. d. G. His subfamily Aphidinae includes both the Aphidinae and 
 Pemphiginae of Mordwilko, while his Chermisinae is the same as 
 Mordwilko 's Phylloxerinae. Following is a translation of Van der 
 Goot 's descriptions of the two subfamilies : 
 
 Subfamily Aphidinae v. d. G. : Body very often without distinct jiroups of 
 glands for the secretion of wax. Antennae usually six- or seven-jointed [when 
 the terminal process of the sixth segment is longer than the segment he considers 
 it as the seventh segment]. Only in a few cases are the apterous forms with 
 three-segmented antennae. The primary sensoria usually have a distinct "haar- 
 kranz" [hairy fringe?]. Cornicles almost always and eauda often present. Pore 
 wings with four veins, the cubitus or media I very often divided: hind wings 
 usually with two cross-veins. Vivi-oviparous: the sexuales mostly of the usual 
 form. 
 
 Subfamilj' Chertnisinae v. d. G. : Body almost always with distinct groups of 
 glands for the production of wax. Antennae three-segmented, often evidently five- 
 .segmented. Sensoria always without ' ' haarkranz. ' ' Cornicles always absent. 
 Fore wings with three veins; hind wings with only one small vein. Always only 
 oviparous: sexuales dwarfish, with or without beak. 
 
 3 Davidson, W. M., Syrphidae in California, Jour. Econ. Ent., vol. 9, pp. 454- 
 457, 1916. 
 
A SYNOPSIS OF THE APHIDIDAE 11 
 
 The latter subfamily has been cousidered by the author as Phyllox- 
 erinae Dreyfus; the former as two subfamilies, Aphidinae Buckton 
 and Pemphiginae Mordwilko. Mordwilko gives the following char- 
 acters for these two subfamilies : 
 
 Subfamily Pemphiginoe Mordw. : Antennae of the alate forms five- or six- 
 segmented, the third bearing a specifically definite number of transverse or arch- 
 like sensoria; sliort, usually not longer than the head and thorax. The apterous 
 parthenogenetie females have four- to six-segmented antennae, but these ar(i 
 sometimes reduced to three or even to two segments. The fore wings of the 
 alate forms have four transverse veins, of which the third or cubital vein [third 
 discoidal] is either simple or once-branched. The hind wings have one or two 
 transverse veins. The cornicles are either entirely absent or very slightly devel- 
 oped, and in the latter case may not be present in all the forms of one sjjecies. 
 
 Subfamily Aphidinae Buckton: Antennae always six-segmented, except in the 
 stem mother of some species, and in the genus Sipha Passerini. [This genus is 
 not represented in California. In EssiffcUa Del Guereio, Ccrosiplta Del Guercio, 
 and Trifidaphis Del Guercio, three Californian genera described since the publi- 
 cation of Mordwilko 's paper, the antennae are but five-segmented.] The last 
 antennal segment often ends in a long thread-like filament which may be longer 
 than the segment. Antennae with a long filament are mostly from half the 
 length of the body to longer than the body. The antennal filament is character- 
 istic only for this subfamily ; some genera of the groups Lachnina and Callipterina 
 have a very short filament, and the antennae are not longer than the head and 
 thorax. The sensoria are small and are shaped like dots, circles, or transverse 
 holes, but never archlike or half-rings. Segment 3 bears the largest number, 
 especially in the alate forms. The cubitus [third discoidal] of the fore wings is 
 usually twice-branched although there are some exceptions, as Toxoptera Koch. 
 Most species have long cylindrical cornicles which are often clavate in the middle. 
 Sometimes they may b'e greatly reduced or poorly developed, and, as in Lachnina 
 and Callipterina, they may be replaced by cupola-shaped elevations. A cauda 
 is usually present, being cdnicle, ensiform, or globular, although in Lachnina it is 
 not evident. The sexual forms have beaks, and become quite large. 
 
 Subfamily Aphidinae Buckton 
 
 Buckton, Mono, British Aphides, 1883. 
 
 This subfamily is divided into three groups, following Carl Borner 
 (Sorauer, Paul, Haudbuch der Pflanzenkranklieiten, vol. 3, p. 664, 
 1913). Borner considers the family Aphididae as a superfamily, and 
 divides it into four families ; so this subfamily Aphidinae he considers 
 a family, and the various groups as subfamilies. Below is a trans- 
 lation of his key : 
 
 1. Claws with spatula-like or leaf-shaped empodial hairs (fig. 1). Cornicles vari- 
 ously formed, bare. Pubescence of larvae as in Aphidina. The majority 
 of the species live free and monophagous on trees, only seldom on herbaceous 
 
 plants, and never migrate collectively Group Callipterina 
 
 — Claws with simple empodial hairs (fig. 2), often hard to see 2 
 
12 MISCELLANEOUS STUDIES 
 
 2. Antennae with short terminal joint (fig. 3), (except in Pterocommini, but then 
 the Cauda is not tail-like). Body ridges with more than six longitudinal 
 rows of hairs. Hairy covering mostly thick. Cauda not lengthened tail-like, 
 anal plate widely rounded (fig. 5). Wax glands either present or lacking. 
 Mostly strongly monophagous forms, at times of remarkable size. Pound 
 
 mostly on tree growths and without change of hosts Group Laclmina 
 
 — Terminal joint of antennae always with a long, slender filamentous projection 
 (fig. 4). Body ridges of young larvae at most with only six longitudinal 
 rows of hairs, which may be increased after the first molt. Cauda either 
 short or lengthened tail-like, anal plate widely rounded (fig. 6). Species 
 monophagous or polyphagous, many with a change of host plants. On 
 tree or herbaceous groivths Group Apludina 
 
 Group Callipterina Morel 
 (Subfamily Callipterinae Borner) 
 
 Mordwilko, Ann. Imperial Acad. Sci., St. Petersb., 1908. 
 
 Borner, in Sorauer, Handbuch der Pflanzenkrankheiten, vol. 3, p. 664, 1913. 
 
 According to Borner tliis group consists of two tribes, the Phyl- 
 laphidini and the Callipterini. lie divides the Callipterini into two 
 groups, the Callipterini and the Chaitophori. The author has followed 
 him to a certain extent, but has given each of the last two groups equal 
 rank with the Phyllaphidini, and thus considers this group, Callip- 
 terina, as consisting of three tribes. Below is a key to the same : 
 
 1. Wax glands with faceted pore fields present. Antennae as in Lachnina 
 
 (fig. 13). Pubescence delicate Tribe Phyllaphidini 
 
 — Wax glands lacking or without faceted pore fields. Pubescence often very 
 
 remarkable. Terminal joint of the antennae often lengthened into a bristle 
 (fig- 30) 2 
 
 2. Anal plate more or less emarginate or bilobed (fig. 7), except in Euceraphis 
 
 Koch Tribe Callipterini 
 
 — Anal plate widely truncate or rounded (fig. 8) Tribe Chaitophorini 
 
 Trilie Phillaphidini Borner 
 
 Borner, in Sorauer, Handbuch der Pflanzenkrankheiten, vol. 3, p. 664, 1913. 
 
 This tribe Phyllaphidini consists of but one genus, Phyllaphis 
 Koch, which is represented in California by three species. 
 
 1. Genus Phyllaphis Koch 
 Koch, Die Pflanzenlause, p. 248, 1857. Type Aphis fagi Linn. 
 Key to CALiroRNiA Species 
 1. Alate viviparous females unknown. Wing venation of alate males similar to 
 that of Eriosoma spp. (fig. 17). Forming pseudogalls on edges of leaves 
 or living free in masses of white floceulence on leaves of Quercus spp. 
 
 quercicola Baker 
 
A SYNOPSIS OF THE APBIDIDAE 13 
 
 — Alate viviparous females common. Venation normal, the third discoidal being 
 
 twice-branched. Not on Quercv^ spp 2 
 
 2. Antennae short, stout, with oval transverse sensoria (fig. 13). Forming galls 
 on Arctostaphylos spp. (and Arhutus spp.) _ coweni (Ckll.) 
 
 — Antennae longer and narrower with circular sensoria (figs. 9, 14—17). Living 
 
 under thick masses of white flocculence on Fagus spp fagl (Linn.) 
 
 1. Phyllaphis coweni (Ckll.) 
 
 Figure 13 
 Coekerell, Can. Ent., vol. 37, pp. 391-392. 1905. Pemphigus (orig. desc). 
 Davidson, Jour. Econ. Ent., vol. 4, pp. .559, 1911. Cryptosiphum tahoense 
 
 n.sp. (desc). 
 Davidson, Jour. Econ. Ent, vol. 5, p. 404, 1912 (list). 
 Essig, Pom. Jour. Ent. Zool., vol. 7, pp. 187-195, 1915 (desc). 
 - Records. — Arctostaphylos manzanita ; Oakville, Napa County, February, 1913 
 (E. L. Brannigan) ; Mount Diablo, Contra Costa County (Davidson) ; Jasper 
 Eidge, Santa Clara County, October, 1914 (B. A. Cornwell) ; Pine Hills, San Diego 
 County, June, 1916.* A. pumella, A. tomentosa, Lake Tahoe, August, 1911 
 (Davidson) : A. glauca, Alpine, San Diego County, June, 1916. 
 
 This species is found more or less abundantly throughout the state 
 wherever its host plants occur. Essig (1915) states it is found 
 throughout the Kocky, Sierra Nevada, and Coast Range mountains, 
 being more abundant in the central and northern parts of the state. 
 The author has found it to be extremely abundant in the Cuyamaca 
 and Laguna mountains in the extreme southern part of the state. 
 The insects can be found at any time of the year in the galls on 
 manzanita although most abundantly in the early fall. Collections 
 by the author in June showed that the stem mothers and young vir- 
 gogeniae only were present. A few weeks later the alate females were 
 abundant, while in August the sexuales begin to appear. However, 
 the alate viviparous females have been found in October and in 
 Februaiy. This species forms galls on the leaves, and flower and 
 fruit stalks of its host. Usually there is but one gall to a leaf, 
 although sometimes four or five may be found. When first formed 
 these galls are concolorous with the leaves; but as they become older 
 they turn more and more reddish in color, until when mature they are 
 a very bright red. 
 
 2. Phyllaphis fagi (Linn.) 
 
 Figures 9 to 12 
 Linnaeus, Syst. Nat., vol. 2, p. 735, 1735. Aphis (orig. desc). 
 Davidson, Jour. Econ. Ent., vol. 3, p. 376, 1910 (list). 
 Records. — Fagus sp., Palo Alto, 1910 (Davidson) ; Fagus sylvatica, Stanford 
 University, April to May, 1915. 
 
 * Records in which no collector 's name is mentioned refer to collections made 
 by the author. 
 
1-i MISCELLANEOUS STUDIES 
 
 This species has been taken only in the vicinity of Stanford Uni- 
 versity, where it infests copper beach {Fagus stjlvatica). It may be 
 easily recognized by the masses of whitish flocculence on the under 
 side of the leaves. Each mass contains one individual, which is 
 entirely hidden by it. In looking up the literature of this species 
 the author found that there has been no description of it published in 
 America, so below is included a brief description of specimens taken 
 near Stanford University on April 28 and May 29, 1915. 
 
 Alate viviparous female. — Prevailing color dark green, covered 
 with a whitish flocculence. Tliis flocculence consists of wax threads 
 as much as 3 mm. long. Head dusk.y, with frontal margin black. 
 Eyes red. Antennae dusky, except II and basal one-third of III, 
 which are pale. Beak pale with apex and joints dusky. Thorax 
 dusky green with lobes black. Abdomen dark green with a row of 
 black spots on each margin and abovit seven black transverse dorsal 
 bands. Cornicles black. Cauda and anal plate concolorous with 
 abdomen with distal margins slightly darker. First and second femora 
 pale with apices only dusky ; third femora dusky throughout. First 
 tibiae pale with apex dusky; second and thii'd dusky tlinnigliout. 
 Tarsi black. Wings hyaline, stigma gray. 
 
 Head twice as wide as long, furnished with many small wax 
 glands. Antennae reaching to the cornicles or to the base of the 
 Cauda, set on small tubercles (fig. 12). Ill is the longest segment, 
 followed by IV, V, and VI. VI spur is merely a thumb-like projec- 
 tion (fig. 16). The usual primary and accessory sensoria are present 
 on V and VI. Secondary sensoria are found only on III (fig. 9). 
 These are fairly large, almost circular, and placed in a single row 
 along the segment. They number from four to seven, five being the 
 average. The beak is short, reaching but slightly beyond the first 
 coxae. The wings are normal, with a twice-branched third diseoidal. 
 The cornicles are merely small pores. The cauda is short and knobbed, 
 the anal plate emarginate or bilobed (fig. 11). 
 
 Measurements: Body length 2.0 to 2.4 mm., width 0.8 to 1.04 mm., 
 antennae total 1.55 to 2.06 mm., Ill 0.591 to 0.77 mm., IV 0.34 to 0.47 
 mm., V 0.27 to 0.39 mm., VI 0.19 to 0.25 mm., cornicles (diameter) 
 0.05 mm. 
 
 Apterous viviparous female. — Prevailing color under flocculence 
 pale yellowish green. Light brown markings as follows : two rows 
 of four spots each across the prothorax, one large spot on each margin 
 and one on tlie dorsum of the thorax, four spots on each abdominal 
 
A SYNOPSIS OF THE APHWIDAE IH 
 
 segment, two doi"sal and two marginal. Antennae pale except VI, 
 apical two-thirds of V, and apical one-third of IV. Legs pale with 
 light brown spots at joints; tarsi black. Cauda small and conicle, 
 cornicles not evident. 
 
 Measurements: Bodj' length 2.9 to 3.0 mm., width 0.96 to 1.2 mm., 
 antennae total 1.26 mm., Ill 0.36 mm., IV 0.32 mm., V O.204 mm., 
 VI 0.205 mm. 
 
 3. Phyllaphis quercicola Baker 
 
 Figures 14 to 20 
 
 Clarke, Can. Ent., vol. 35, p. 248, 1903. Schizoneura querci (Fitch) (list). 
 Davidson, Jour. Econ. Ent., vol. 3, p. 374, 1910. S. querci (Fiteh) (list). 
 Davidson, Pom. Jour. Ent., vol. 3, p. 398, 1911. S. querci (Fiteh) (list). 
 Davis, Ent. News, vol. 22, p. 241, 1911. Phyllaphis querci (Fitch) (biblig.) 
 Davidson, Jour. Econ. Ent., vol. 7, p. 127, 1914. P. querci (Fiteh) (note). 
 Gillette, Ent. News, vol. 25, p. 274, 1914. Phyllaphis sp. (list). 
 Baker, Ent. News, vol. 27, p. 362, 191(!. P. quercicola n.n. for P. querci 
 (Fitch) of Davis. 
 
 Records. — Quercus agrifolia; Placer, Contra Costa, Santa Clara counties 
 (Davidson, Clarke); Stanford University, April, 1915; Berkeley, September, 1915; 
 Wynola, San Diego County, June, 1916; Charter Oak, Los Angeles County, Novem- 
 ber, 1916. Quercus lobata, Stanford University (Davidson) ; Q. wisligenii, Placer 
 County (Davidson) ; Q. dutnosa, San Diego, August, 1915; Quercus sp., Spreckels, 
 Monterey County, 1913 (Gillette). 
 
 This is a very common species of woolly aphis on the oaks, par- 
 ticularly the live oak, through southern and central California. 
 According to Davidson (1914) the stem mothers occur in pseudogalls 
 on the edges of the leaves. The second generation lice, when mature, 
 leave these galls to live on the upper and lower surfaces of the leaves, 
 unprotected except for their woolly covering. The sexes, apterous 
 oviparous females and alate males, occur late in the fall. The vivi- 
 parous generations are all apterous. The writer has observed the 
 stem mothers as late as August in San Diego County, while he has 
 found the viviparous females on the under side of the leaves as early 
 as mid-June in Berkeley. 
 
 The identity of this species has never been definitely established. 
 It was thought to be the species described by Fiteh (Kept. In.s. N. Y., 
 vol. 5, p. 804, 1859) as Eriusomu querci, but in 1916 Baker pointed 
 out the identity of Eriosania querci Fitch, proving it to be identical 
 with a species of Anoecia found on Cornus and formerly considered 
 to be A. corni Fab. Baker's decision is that the Quercus-Cornus 
 species of the eastern United States is Anoecia querci (Fiteh) and 
 
16 MISCELLANEOUS STUDIES 
 
 is distinct from our western one. In 1911 Davis described a species 
 of woolly aphis from oak under the name of Phyllaphis querci (Pitch) 
 stating that it is the same one as listed by Davidson. Baker proposes 
 the name Phyllaphis quercicola for this species described by Davis. 
 Consequently it is so listed in this paper. This species is not a typical 
 Phyllaphis, but it fits that genus better than any other so is placed 
 there provisionally. The figures (14-20) are from a specimen of alate 
 male in the Davidson collection in Stanford University. 
 
 Tribe Callipterini Wilson 
 
 Wilson, Can. Ent., vol. 42, p. 2.53, 1910. 
 
 The genera included in this tribe differ somewhat as considered by 
 various entomologists. Since Wilson has worked out the synonomy 
 of the various genera very well he is followed in preference to some 
 of the European authors, although there ai'e some points in which 
 he is mistaken. For instance, he places Ptcrooallis Passerini, Callip- 
 teroides Mordwilko, Tuberculatus Mordwilko, SuhcalUpterus Mord- 
 wilko, and Therioaphis Walker as synonyms of Myzocallis Passerini. 
 In regard to this, he states, "In 1894 Mordwilko used A. coryli Goetze 
 as the type of his genus Calliptcroides, but as this species..." He 
 is mistaken in this, for in the paper referred to, Mordwilko used 
 A. coryli Goetze as the type of the genus Myzocallis Passerini, and in 
 1908 he gave as the type of Callipteroides, Callipterus nigritarsus 
 He.yden {hetulae Koch). If nigritarsus Heyden is a synonym of 
 bctulae Koch, as Mordwilko indicates, then Calliptcroides is a synonym 
 of Euceraphis Walker, for C. hetulae Koch certainly falls into this 
 genus, as described by Wilson himself. The key to the California 
 genera below is adapted from Wilson's key (Can. Ent., vol. 42, pp. 
 253-254, 1910). 
 
 • Key to Californi.v Genera of Callipterini 
 
 1. Antennal tubercles prominent (fig. 21) ; antennae always exceedingly long.... 2 
 
 — Antennal tubercles wanting or very small (fig. 22) ; antennae variable, some- 
 
 times shorter than the body 3 
 
 2. Cornicles very long and large (figs. 23-24) 4 
 
 — Cornicles very short and more or less constricted in the middle 5 
 
 — • Cornicles little more than pores (fig. 25). Wings held horizontal at rest. 
 
 Monellla Oestlund 
 
 3. Cornicles distinct, usually being longer than broad in the middle (fig. 26) 6 
 
 — Cornicles little more than pores, and broader than long (fig. 25). Wings 
 
 held horizontal at rest MonelUa Oestlund 
 
.4 SYNOPSIS OF THE APBIBIDAE 17 
 
 4. Cornicles one-fourth the length of the body or more, swollen in the middle 
 
 (fig. 24) Drepanosiphum Koch 
 
 — ■ Cornicles large and nearly one-fourth the length of the body, swollen at the 
 base and tapering toward the middle (fig. 23) ....Drepanaphls Del Guercio 
 
 5. Inner side of antennal tubercles about one-half the length of the inner side 
 
 of the first antennal joint (fig. 29) Euceraphis Walker 
 
 — Inner side of antennal tubercles more than one-half the length of the inner 
 
 side of the first antennal segment (figs. 27-28) Calaphis Walsh 
 
 6. Antennae longer than body, except in CalUpterinella, with VI spur not much 
 
 shorter than VI base (fig. 31) 7 
 
 — • Antennae shorter than the body, with VI spur very short, often being little 
 more than a nail-like process (fig. 34) 9 
 
 7. VI spur considerably longer than VI base, being one and one-half to two 
 
 times as long. Anal plate emarginate but not deeply bilobed. 
 
 Callipteiinella Van der Goot 
 
 — VI spur about equal to or shorter than VI base. Anal plate deeply bilobed 8 
 
 8. VI spur and VI base subequal (fig. 31). Cornicles t^vice as long as broad 
 
 in the middle and constricted in the middle (figs. 26, 32). 
 
 Myzocallis Passerini 
 
 — VI spur shorter than VI base (fig. 30). Cornicles much broadened at base 
 
 (fig. 33) Eucallipterus Schouteden 
 
 9. VI spur less than one-half the length of VI base (fig. 34). Cornicles not 
 
 longer than broad at the base, and constricted in the middle (fig. 35). 
 
 Chromaphis Walker 
 
 — VI spur at least one-half as long at VI base (figs. 63, 66). Cornicles short, 
 
 about as long as broad and placed on a broad base CaUipterus Koch 
 
 2. Genus Drepanosiphum Koch 
 Koch, Die Pflanzenlause, p. 201, 1855. Type Aphis palantanoides Sehrank. 
 4. Drepanosiphum platanoides (Sehrank) 
 
 Figures 21, 24, 36 
 
 Sehrank, Fauna Boic, vol. 2, p. 1206, 1801. Aphis (orig. desc). 
 Wilson, Jour. Econ. Ent., vol. 2, p. 349, 1909 (desc. ala. vivi., ala. ovi. 
 
 females). 
 Davidson, Jour. Econ. Ent., vol. 3, p. 377, 1910 (list). 
 Essig, Pom. Jour. Ent., vol. 4, p. 759, 1912 (list). 
 
 Becords. — Acer macrophyJlum, A. nerjundo; Berkeley, 1915 (Essig) ; Stan- 
 ford University, October, 1914, April, 1915; A. pseudoplatanus, Stanford Univer- 
 sity, November, 1914 (Morrison); A. sacdharum, Berkeley, June, 1915; Platanus 
 racemosus. Stanford University (Davidson) ; Acer sp., San Lorenzo, 1908 (Wil- 
 son). 
 
 This is a very common species in the San Francisco Bay region 
 on various species of maples, and on box elder and western sycamore. 
 In April the alate and apterous viviparous females are abundant, 
 remaining' so throughout the summer and early fall. In tlie later 
 fall (October and November) the sexes appear. Just where the eggs 
 
18 MISCELLANEOVS STUDIES 
 
 are laid the author is unable to say. A curious fact is that the 
 oviparous females are alate as well as apterous. The author has never 
 seen the alate forms, but Wilson (1908) describes them. 
 
 :l. Genus Drepanaphis Del Cxuerein 
 
 Del Guereio, Eivista di patologia vegetable, vol. 4, pp. 49-.53, 1909. Type 
 Siplwnophora accrifolii Thomas. 
 
 5. Drepanaphis acerifolii (Thomas) 
 
 Figures 23, 37 
 
 Thomas, Illinois Lab. Nat. Hist., Bull. 2, p. 4, 1878. Siphonophora (orig. 
 
 desc.). 
 Clarke, Can. Ent., vol. 35, p. 249, 1903. Drepanosiphum (list). 
 Sanborn, Kan. Univ. Sci., Bull. 3, p. 45, 1904. Drepanosiphum (desc. ala.). 
 Davidson, Jour. Econ. Ent., vol. 2, p. 303, 1909. ' Drepmwsiphum (list). 
 Davidson, Jour. Econ. Ent., vol. 3, p. 380, 1910. Macrosiphum (list). 
 Essig, Pom. Jour. Ent., vol. 4, p. 760, 1912 (list). 
 Essig, Mon. Bull., Cal. Comm. Hort., vol. 3, p. 85, 1914 (list). 
 Essig, Mon. Bull., Cal. Comm. Hort., vol. 3, p. 445, 1914 (list). 
 
 Records. — Acer sp. : Stanford University (Davidson); Sacramento (Essig); 
 Hanford, Fresno County (B. V. Sharp) ; A. macrophyUum, A. saccharinum. Berke- 
 ley, July to October, 1915; Eiverside, October, 1916; A. dasijcarpam, A. plat- 
 anoides, Berkeley, 1915 (Essig); Quercu-s .sp. (live oak), Berkeley (Clarke) (?). 
 
 Tliis is as common a species on maple in the San Francisco Bay 
 region as the preceding one. It has also been taken in the Sacramento 
 and the San Joaquin valleys, and in southern California. It is a 
 species easily recognized by its dark markings and the dorsal tubercles 
 on the first and second abdominal segments. 
 
 4. Genus Calaphis AValsh 
 Figure 28 
 Walsh, Proc. Ent. Soc. Phila., vol. 1, p. 301, 1863. Type C. hetulclla ii.sp. 
 
 (). Calaphis betulaecolens (Fitch) 
 
 Figures 27-:'.S 
 
 Fitch, Cat. Honiop. N. Y., p. 66, 1851. Aphis (orig. desc). 
 
 Clarke, Can. Ent., vol. 35, p. 249, 1903. CalUpterus (list). 
 
 Davidson, Jour. Econ. Ent., vol. 2, p. 301, 1909. Callipteru.^ (list). 
 
 Davidson, Jour. Econ. Ent., vol. 3, p. 376, 1910. CalUpterus (list). 
 
 Essig, Pom. Jour. Ent., vol. 3, p. 556, 1911 (syn.). 
 
 Davidson, Jour. Econ. Ent., vol. 5, p. 404, 1912 (desc. sexes). 
 
 Essig, Pom. Jour. Ent., vol. 4, p. 760, 1912 (list). 
 
 Essig, Mon. Bull., Cal. Comm. Hort., vol. 3, p. 445, 1914 (list). 
 
 Baker, Proc. Ent. Soc, Washington, vol. 18, p. 186, 1916 (desc). 
 
 Records. — Betula sp., Alameda, Contra Costa, Santa Clara counties (Clarke, 
 Davidson, Essig, Morrison, and the author). 
 
A SYNOPSIS Of THE APHIDIDAE 19 
 
 This is a common species of aphid on birch (Betula spp.) in the 
 San Francisco Baj- region. In the early part of March the eggs begin 
 to liatch. In 1915 at Stanford Universitj- eggs began to hatch on 
 March 8, the process coiitimiing for several days. A month later 
 both alate and apterous females were quite abundant ; the alate 
 females being undoubtedly the stem mothers, the apterae belonging 
 to the second generation. Viviparous generations appeared through- 
 out the summer. During August the sexes, alate males and apterous 
 oviparous females, occurred. In 1914 the sexes and sexupara \v(^ro 
 noticed on August 28. Egg laying occurred shortly afterward, the 
 eggs being laid in the crotches of the twigs and under the curled 
 edges of the bark. Birch is the only recorded host plant. 
 
 5. Genus Euceraphis "Walker 
 
 Walker, The Zoologist, p. 2001, 1870. Type Aphis hetulac Koch. 
 
 Key to Californi.\ Species 
 
 1. Body light green; third joint of antennae with about 13-18 sensoria on basal 
 
 one-half (fig. 39) giUettei Dvdn. 
 
 — Body yellow with dark markings on head and thorax, and often with as 
 many as eight black transverse stripes on the abdomen (the number varies 
 between none and eight) ; third antennal segment with 19-25 sensoria 
 (fig. 40) betulae (Koeh) 
 
 7. Euceraphis betulae (Koch) 
 
 Figures 29, 40 
 
 Koch, Die Pflanzenliiuse, p. 217, 1855. Callipterus (orig. desc). 
 Davidson, Jour. Econ. Ent., vol. 5, p. 405, 1913 (dese. ovi. female). 
 Davidson, Jour. Econ. Ent., vol. 7, p. 129, 1914 (desc. stem mother). 
 
 Beeords. — Betula sp. : Oakland (Davidson); Palo Alto, March to April, 1915. 
 
 David.son lists this species, describing the stem mother and 
 oviparous female from the San 5'rancisco Bay region. The author 
 found it in Palo Alto during March and April. 1915, on Betula alha. 
 According to Davidson the stem mothers hatch from the eggs about 
 the middle of February, feeding on the .stems until the leaves open in 
 March. The viviparous generations occur during the summer. He 
 took the oviparous females in November. His description of the stem 
 mother gives three dusky traJisverse bands on the abdomen. The 
 author has found this to be variable, the number ranging from none 
 to eight. 
 
20 MISCELLANEOUS STUDIES 
 
 8. Euceraphis gillettei Davidson 
 
 Figure 39 
 
 Clarke, Can. Ent., vol. 35, p. 248, 1903. Lachmi^ alnifoliae Fitch (list). 
 Davidson, Jour. Econ. Ent., vol. 2, p. 300, 1909. L. alnifoliae Fitch (list). 
 Davidson, Jour. Econ. Ent., vol. 3, p. 375, 1910. L. alnifoliae Fitch (list). 
 Essig, Pom. Jour. Ent., vol. 4, p. 773, 1912. Lachn-us alnifoliae Fitch 
 
 (list). 
 Essig, Pom. Jour. Ent., vol. 4, p. 773, 1912. Lalinus alnifoliae Fiteh (note). 
 Davidson, Jour. Econ. Ent., vol. 8, p. 421, 1915 (orig. desc). 
 
 Secords. — AhiTis rhombifolia; Berkeley (Clarke), Stanford University, San 
 Jose, Walnut Creek (Davidson), Stanford University, March, 1915. 
 
 This species was reported from alder by Clarke and Davidson as 
 Lachnus alnifoliae Fitch. Essig, in his Host plant list of California 
 Aphididae, lists Callipterus alnifoliae (Fitch) on Alnus rhombifolia, 
 but later states that this citation should be Lachnus alnifoliae Fitch. 
 Therefore he referred to this new species of Davidson. The author 
 took both apterous and alate viviparous females of this species on 
 Alnus rhombifolia, along the banks of the San Franeisquito Creek, 
 near Stanford University, on March 19, 1915. During the spring it 
 was quite common there. 
 
 6. Genus Eucallipterus Sehouteden 
 Schouteden, Mem. Soc. Ent. Belg., vol. 12, 1906. Type Aphis tiliae Linn. 
 Key to California Species 
 
 1. Wings hyaline; III pale except at the apex, with 5-7 sensoria on the basal 
 
 one-fifth (fig. 41) flava (Dvdn.) 
 
 2. Wings vpith veins clouded; III with apical one-fifth and basal one-half dusky, 
 
 and with about 13-15 sensoria on the basal one-half (fig. 42). 
 
 tiliae (Linn.) 
 
 9. Eucallipterus flava (Davidson) 
 
 Figure 41 
 
 Davidson, Jour. Econ. Ent., vol. 5, p. 406, 1912. Euceraphis (orig. desc). 
 Davidson, Jour. Econ. Ent., vol. 8, p. 423, 1915 (desc. sexes). 
 
 Eecords. — Alnus rhoml}ifoli<i ; San Jose, Walnut Creek (Davidson), Stanford 
 University (Morrison). 
 
 This is an uncommon species in the San Francisco Bay region on 
 Alnus rhonihifolia, occurring on the under side of the leaves. The 
 author has never collected it, but has specimens from Davidson, taken 
 
A SYNOPSIS OF THE APHIDIDAE 21 
 
 in April, 1913, near Walnut Creek, Contra Costa County. According 
 to Davidson the sexes appear in October, egg laying occurring diiring 
 the first part of November. The eggs are laid at the axils of the new 
 buds and on the twigs or canes. These hatch the following spring, the 
 stem mothers being found in the early part of April. 
 
 10. Eucallipterus tiliae (Linn.) 
 
 Figures 7, 30, 33, 42, 50 
 
 Linnaeus, Syst. Nat., vol. 2, p. 734, 173.5. Ajihis (orig. desc). 
 Davis, Ann. Ent. Soc. Amer., vol. 2, p. 33, 1909. Callipterus (desc, biblic). 
 Davidson, Jour. Ecou. Ent., vol. 2, p. 302, 1909. Callipterus (list). 
 Davidson, Jour. Econ. Ent., vol. 3, p. 372, 1910. Callipterus (list). 
 Davidson, Jour. Econ. Ent., vol. 3, p. 370, 1910. Calliptenis (list). 
 Essig, Pom. Jour. Ent., vol. 4, p. 763, 1912 (list). 
 
 Becords. — Tilia amerieana, Tilia europea: Stanford University (Davidson), 
 Berkeley, August, 1914 (Es.sig) ; Stanford University, April to May, 1915; Berke- 
 ley, June, 1915. 
 
 In the San Francisco Bay region this very pretty aphid is quite 
 common on basswood or linden. The author has taken it throughout 
 April, MaJ^ and June. Essig found it abundantly in August. It is 
 very easily recognized when found at rest on the under side of the 
 leaves of its host bj- the two black lines extending from the front of 
 the head along the margins of the thorax and joining with the costal 
 margins of the wings. It so appears that these lines are continuous 
 from the front to the tip of the wings. 
 
 7. Genus Myzocallis Passerini 
 Passerini, Gli Afidi, p. 28, 1860. Type Aphis coryli Goetze. 
 Key to C.\liforni.\ Species 
 
 1. Wings hyaline 6 
 
 — Wings not h.valine, with portions shaded (fig. 262) 2 
 
 2. Costal cell of wings hyaline (figs. 266, 267) 3 
 
 — Costal cell of forewings dusky or shaded (figs. 263, 264) 5 
 
 3. First discoidal vein dusky, otherwise the wing is hyaline. VI with spur shorter 
 
 than base. Apical one-half of III dusky (fig. 47). Cornicles pale. Found 
 on Alnu-s spp alnifoliae (Fitch) 
 
 — Wings not as above (figs. 266, 267). VI with spur either equal to or longer 
 
 than base. Ill with less than apical one-half dusky 4 
 
 4. Cornicles pale. Abdomen without dusky dorsal markings. On Quercus spp. 
 
 mauieri Swain 
 
 — Cornicles dusky (fig. 62). Abdomen with dusky dorsal markings. On Casta- 
 
 nea spp. and Quercus spp davidsonl Swain 
 
22 MISCELLANEOUS STUDIES 
 
 5. Cornicles pale. Wings with greater portion cloudy (fig. 262). Antennae with 
 
 only tips of III to VI dusky. On Qiiercm spp discolor (Monell) 
 
 — Cornicles pale with apex dusky. Wings with dusky band along costal margin 
 
 (fig. 263). Antennae with tips of III and IV, apical one-half of V, and 
 all of VI and spur dusky. On Quercus spp bellus (Walsh) 
 
 6. Abdomen with four spine-like tubercles on the dorsum of the first segment. 
 
 VI with base and spur subequal, III being considerably longer than both. 
 Cornicles pale, small, and inconspicuous. On Ultmts spp. 
 
 ulmifolii (Monell) 
 
 — Abdomen without tubercles as above 7 
 
 7. Ill shorter than VI (base and spur). On Q-uercus spp punctatus (Monell) 
 
 — Ill not shorter than VI (base and spur) 8 
 
 8. VI with spur about twice as long as base (fig. 44). Cornicles pale. On 
 
 Cori/lm spp coryli (Goetze) 
 
 — ■ VI with spur at most only slightly longer than base 9 
 
 9. Ill with apex only dusky (figs. 57, 58). Cauda pale 10 
 
 — Ill dusky throughout (fig. 268) or with apex and a band near the base dusky 
 
 (fig. 48). Cauda dusky 11 
 
 10. Cornicles pale. Antennae longer than body. Sensoria on III (two or three 
 
 in number) small and located close to the base of the segment (fig. 57). On 
 Pasmiia spp pasaniae Dvdn. 
 
 — Cornicles dusky, at least apical one-half. Antennae not longer than the body. 
 
 Sensoria on III (five or more in number) fairly large and on basal two- 
 thirds of segment (fig. 58). On Quercus spp quercus (Kalt.) 
 
 11. Abdomen with dusky dorsal markings. Ill dusky throughout (fig. 268). Ou 
 
 Arundo spp arundinariae Essig 
 
 — Abdomen without dusky dorsal markings. Ill with apex and band near base 
 
 dusky (fig. 48). On Arutido s-pp arundlcolens (Clarke) 
 
 11. Myzocallis alnifoliae (Pitch) 
 
 Figure 47 
 Fitch, Cat. Homop. N. Y., p. 67, 1851. Lachnus (orig. desc). 
 Essig, Pom. Jour. Ent., vol. 4, p. 764 (762), 1912. M. nJni (Fabr.) (desc. 
 
 viviparae ) . 
 Baker, Jour. Eeon. Ent., vol. 10, p. 421, 1917 (note). 
 Eecords. — Alnus rhombifolin ; Santa Paula (Essig). 
 
 Only once has this species been taken in California, by Essig in 
 August, 1911, near Santa Paula, Ventura County. At that time it 
 was very abundant on the under side of tlie leaves, causing a large 
 amount of sooty mold. 
 
 12. Myzocallis arundlcolens f Clarke) 
 ■ Figures 22, 48, 51, 52 
 Clarke, Can. Ent., vol. 35, p. 249, 1903. Calliptenis (orig. desc). 
 Davidson, Jour. Eeon. Ent., vol. 2, p. 301, 1909. Callipterus (list). 
 Davidson, Jour. Eeon. Ent., vol. 3, p. 376, 1910. Callipterus (list). 
 Essig, Pom. Jour. Ent., vol. 4, p. 762, 1912 (list, in part). 
 Essig, Univ. Calif. Publ. Entom., vol. 1, p. 305, 1917 (desc.). 
 
J SYNOPSIS OF THE APHWIDAE 23 
 
 Records. — Bamboo, Berkeley (Clarke, Essig) ; Arundinaria japonica, Berkeley, 
 June, 1915. 
 
 In the San Francisco Bay region and in the Sacramento Valley 
 this species is often found infesting the upper and lower surfaces of 
 the leaves of various bamboos, particularl}' species of ArundUiaria, 
 Bamhitsa, and rhijUostachys, and the giant reed {Arundo donax). 
 Reports list it from Alameda, Sacramento, San Francisco, and Santa 
 Clara counties. The species described by Davidson (1914) as Eiwal- 
 liptcrus arundicolens (Clarke) and reported from southern California 
 by Essig (1912) proves to be distinct, and was described by Essig 
 (1917) as M. arimdinarme. The following brief description is from 
 a collection made by the author on June 9, 1915, from Arundinaria 
 japonica on the campus of the Universitj' of California in Berkeley. 
 
 Alate viviparous female. — (Second generation?) Prevailing 
 color, pale yellow. Head twice as wide as long, pale yellow, with 
 prominent red eyes. Antennal tubercles absent. Antennae longer 
 than liody; formula III, IV, V, VI spur, VI base, I, II. Segments all 
 pale except tlie margins of I and II, the apices of III, IV, V, and a 
 band about one-sixth tlie length of III a short distance from the base 
 of III (fig. 48), which are black, and VI which is slightlj-- dusky. 
 There are five or six transverse secondaiy sensoria on III, located in 
 the dark band. The usual primary sensoria are present on V and VI, 
 and the usual accessor sensoria on VI. Beak pale and short, reaching 
 on\\ to the middle of the fii-st coxae. Thorax and abdomen normal, 
 pale yellow, without tubercles or dusky markings. Cornicles (fig. 51) 
 pale, short, broader at base than at apex. Cauda short, constricted 
 in the middle, with distal end black. Anal plate (fig. 52) pale, deeply 
 bilobed. Wings normal, hyaline, with the first and second discoidal 
 veins and the base of the stigmal vein darker than the others. There 
 is a perceptible shading at the tip of each vein. 
 
 Measurements: Body length 1..326 to 2.023 mm. (av. 1.644 mm.), 
 width of thorax 0.51 to 0.68 mm. (av. 0.612 mm.), antennae total 
 2.839 to 3.077 mm. (av. 2.9299 mm.). Ill 0.8925 to 0.986 mm. (av. 
 0.9324 mm.), IV 0..578 to 0.663 mm. (av. 0.6423 mm.), V 0.527 to 
 0.561 mm. (av. 0.5403 mm.), VI base 0.306 to 0.323 mm. (av. 0.3103 
 mm.), VI spur 0.34 to 0.425 mm. (av. 0.3691 mm.), cornicle 0.595 to 
 0.765 mm. (av. 0.7002 mm.), cauda 0.153 mm., wing length 2.25 to 
 3.96 mm. (av. 2.9097 mm.), width 1.02 to 1.122 mm. (av. 1.071 mm.), 
 expansion 6.341 to 6.97 mm. (av. 6.6555 mm.). 
 
24 MISCELLANEOUS STUDIES 
 
 13. Myzocallis arundinariae Essig 
 
 Figure 268 
 
 Essig, Pom. Jour. Ent., vol. 4, p. 762, 1912. M. arundicolcns (Clarke) (in 
 
 part). 
 Davidson, Jour. Econ. Ent., vol. 7, p. 129, 1914. EucaUipterus arundi- 
 
 colens (Clarke) (desc. viviparae). 
 Essig, Univ. Calif. Publ. Entom., vol. 1, pp. 302-305, 1917 (orig. desc). 
 
 Records. — Arundo sp., San Francisco Bay region (Davidson) ; Arundinaria 
 japonica, Santa Barbara (Essig) ; Eiverside, January to May, 1917; Ai-undo 
 donax, San Diego, April to June, 1916. 
 
 This is the commonest bamboo-infesting species in southern Cali- 
 fornia and parts of central California. For some time it was con- 
 sidered as M. arundicolcns (Clarke) but this past j-ear Essig pointed 
 out the differences, describing it as a new species. 
 
 14. Myzocallis bellus (Walsh) 
 
 Figures 45, 46 
 
 Walsh, Proc. Ent. Soe. Phila., vol. 1, p. 299, 1862. Aphis (orig. desc). 
 Essig, Pom. Jour. Ent. Zool., vol. 7, pp. 195-200, 1915. Callipierus (desc). 
 
 Records. — Quercus agrifolia, Alhambra, Los Angeles County (Essig) ; Ventura 
 (Essig). 
 
 Two collections have been made of this species in California, both 
 in soutliem California, in January, 1912, in Alhambra, and in May, 
 1913, in Ventura. Both of these consisted only of the alate females 
 (stem mothers), and were described by Essig. 
 
 1.5. Myzocallis davidsoni Swain 
 
 Figures 60, 61, 62, 267 
 
 Clarke, Can. Ent., vol. 35, p. 249, 1903. Callipierus castaneac Fitch (list). 
 
 Davidson, Jour. Econ. Ent., vol. 3, p. 376, 1910. Callipterus castaneae 
 (BuCkton) (list). 
 
 Davidson, Jour. Econ. Ent., vol. 5, p. 405, 1912. CaJaphis castaneae (Buck- 
 ton) (desc sexuales). 
 
 Essig, Pom. Jour. Ent., vol. 4, p. 760, 1912. Calaphis castaneae (Fitch) 
 (list). 
 
 Swain, Trans. Am. Ent. Soc, vol. 44, p. 1, 1918 (orig. desc). 
 
 Records. — Castanea sp., Berkeley (Clarke, Essig, Swain), Stanford University 
 (Davidson, Swain), San Jose (Davidson) ; Quercus pedunculata, Berkeley (Swain, 
 Essig). 
 
 This species was first reported in California by Clarke as CaUip- 
 terus castaneae Fitch and later by Davidson as Callipterus castaneae 
 
A SYNOPSIS OF THE APHIDIDAE 25 
 
 Buckton. Kecently the author di'.scrihed the species from specimens 
 taken in Berkek'y on chestnut and oak. It cannot be the Callipterus 
 castaniac of Fitch, beeausc the hitter is really a Calaphis. It ma.y be 
 the same species that Buckton had when describinfi: his Callipterus 
 castaneae, in which case his name would be dropped as Fitch's species 
 has priority, and is replaced by the author's name, M. davidsoni. It 
 is more or less common throughout tlie San Francisco Bay region on 
 chestnuts, and in one case on two specimens of Qucrcus pcdunculata 
 in Berkeley. The stem mothers appear during the late spring, in 
 April and May. Viviparous generations are produced throughout the 
 summer, the sexuales occurring in October and November. 
 
 16. Myzocallis coryli (Goetze) 
 
 Figures 43, 44, 53, 54 
 
 Goetze, Ent. Beitrage, vol. 2, p. 311, 1778. Aphis (orig desc). 
 Qarke, Can. Ent., vol. 35, p. 249, 1903. Callipterus (list). 
 Davis, Jour. Econ. Ent., vol. 3, p. 417, 1910. Callipterus (desc). 
 Essig, Pom. Jour. Ent., vol. 4, p. 762, 1912 (list). 
 
 Becords. — Coryln-s sp., Berkeley (Clarke) ; Cori/lus rostrata, San Francisco Bay 
 region (Davidson) ; C. rostrata var. calif arnica, C. maxima, Berkeley, August, 
 1914, June to July, 1915. 
 
 In the San Francisco Bay region this species is quite common on 
 alder. During the seasons of 1914 and 1915 tlie author observed it 
 to be very abundant on species of alder on the University of California 
 campus. He has never found it in the south, however. 
 
 17. Myzocallis discolor (IMoncll) 
 
 Figures 262, 263 
 
 Monell, U. S. Geol. Geog. Surv., Bull. 5, p. 30, 1879. Callipterus (orig. 
 
 dese. ) . 
 Williams, Univ. Neb. Studies, vol. 10, p. 115, 1910. Callipterus (desc). 
 
 Record. — Quercus macrocarpa, Sacramento, October, 1916 (Davidson). 
 
 The author received specimens of this species from Davidson, 
 which were found in October, 1916, on Quercus macrocarpa in Sacra- 
 mento. The determination was made by Davis. Below are a few 
 descriptive notes to supplement Williams' description listed above. 
 
 Alate viviparovs female. — Antennae about as long as body, III 
 the longest segment, followed by IV, VI, and V. VI spur is slightly 
 longer than the base. The antennae are rather slender as compared 
 
26 MISCELLANEOUS STUDIES 
 
 with other species of this genus. Primary sensoria are present on 
 V and VI as usual, and accessory sensoria on VI. Tliere are about 
 seven secondary sensoria on III (fig. 262), which are more or less oval 
 to circular, and located on the basal two-thirds of the segment. The 
 cornicles, cauda, and anal plate are typical of the genus. 
 
 Measurements: Body length 1.28 to 1.37 mm., antenna total 1.41 
 mm.. Ill 0.459 mm., IV 0.306 mm., V 0.264 mm., VI 0.289 mm. (base 
 0.119 mm., spur 0.17 mm.), cornicles 0.68 mm., wing length 2.074 to 
 2.414 mm., width 0.68 to 0.833 mm. The two dusky transverse bands 
 across tlie fore wings (fig. 263) constitute the most distinguishing 
 character. The branching of the third discoidal is quite variable. 
 
 18. Myzocallis punctatus (ilonell) 
 
 Monell, U. S. Gk-ol. Geog. Surv., Bull. 5, p. 31, 1879. Callipterus (orig. 
 
 desc). 
 Clarke, Can. Ent., vol. .35, p. 249, 1903. Callipterus hyalinus Monell (list). 
 Essig, Pom. Jour. Ent., vol. 4, p. 762, 1912. M. hyalimts (Monell) (list). 
 Eeoord. — Quercus imbricata, Berkeley (Clarke). 
 
 This is a doubtful species, reported only bj' Clarke from Quercvs 
 imhricata in BerkeleJ^ It is the author's opinion that this is the same 
 species listed by Davidson as M. quercus (Kalt.). 
 
 10. Myzocallis maureri Swain 
 
 Figures 55, 56, 266 
 
 Swain, Trans. Am. Ent. Soc., vol. 44, p. 4, 1918 (orig. desc). 
 
 Records. — Quercus agrifolia, Berkeley (Swain) ; Quercus kelloggii, Julian, San 
 Diego County (Swain). 
 
 This species has been taken in Berkeley and in the Cuj'amaca 
 Mountains of San Diego County by the author. Essig has also taken 
 it in Berkeley. It is never abundant, but the author has observed it 
 several times and in several places in the localities mentioned. 
 
 20. Myzocallis pasaniae Dvdu. 
 
 Figure 57 
 
 Davidson, Jour. Econ. Ent., vol. 8, p. 424, 1915 (orig. desc). 
 
 Secords. — Pasania densiflora, Stevens Creek Canyon, Santa Clara County 
 (Davidson), Berkeley, February, 1915 (Essig). 
 
 This is a species found occasionally on tanbark oak in the San 
 Francisco Bay region. The author has never taken it but has speci- 
 mens from Davidson and Essig. 
 
A SYNOPSIS OF THE APHIDIDAE 27 
 
 21. Myzocallis quercus (Kalt.) 
 
 Figures 31, 32, 58 
 
 Kaltcnbach, Monog. d. Pflanzenlause, p. 98, 1843. Aphis (orig. desc). 
 Davidson, Jour. Econ. Ent., vol. 2, p. 302, 1909. Callipterus (list). 
 Davidson, Jour. Econ. Ent., vol. 3, p. 376, 1910. Callipterus (list). 
 Davidson, Pom. Jour. Ent., vol. 3, p. 399, 1911. Callipterus (list). 
 Essig, Pom. Jour. Ent., vol. 4, p. 762, 1912 (list). 
 Davidson, Jour. Econ. Ent., vol. 7, p. 130, 1914 (desc). 
 
 Records. — Querais agrifolia; Stanford University, San Jose, Penryn, Placer 
 County (Davidson) ; Q. lobata, Santa Clara County (Davidson) ; Berkeley, 1915 
 (Essig); Q. pedunculata, Berkeley, August, 1914; Q. douglasii, Stanford Univer- 
 sity, November, 1910, April, 1911 (Morrison); Q. rohur, Oakland (Davidson). 
 
 This is a variable species more or less eominon in the Sau Fran- 
 cisco Bay region and in the Sacramento Valley on various species of 
 oaks. When he first reported it Davidson was doubtful of its identitj'. 
 Later, however, it was identified by Peter Van der Goot^ as this 
 species. 
 
 22. Myzocallis iilmifolii (Monell) 
 
 Figure 59 
 
 Monell, U. S. Geol. Geog. Surv., Bull. 5, p. 29, 1879. Callipterv.9 (orig. 
 
 desc). 
 Davidson,' Jour. Eeon. Ent., vol. 2, p. 301, 1909. CaUiptenm (list). 
 Davidson, Jour. Eeon. Ent., vol. 3, p. 376, 1910. CalUptcru.s (list). 
 Essig, Pom. Jour. Ent., vol. 4, p. 762, 1912 (list). 
 
 Hecords, — Vlmus sp., Stanford University (Davidson), Ulmus americana, Wal- 
 nut Creek, October, 1913 (Davidson). 
 
 Davidson reports this as common on elms in the San Francisco 
 Bay region. However, the author has never collected it. The follow- 
 ing brief descriptive notes are from an alate viviparous female, taken 
 in Walnut Creek by Davidson. The most distinguishing character is 
 the presence of a pair of small but prominent tubercles on the mid- 
 dorsum of the first and second abdominal segments. The usual 
 primarj^ and accessory sensoria are present on V and VI. Secondary 
 sensoria (fig. 59) are present on the basal one-half to two-thirds of 
 III. These are transversely linear or oval, and number about six. 
 The cornicles are very short, being fully as broad at the apex as long. 
 Cauda and anal plate normal. Wings normal, radial vein indistinct, 
 first discoidal curving toward base of wing. Body length 1.836 mm., 
 width of thorax 0.578 mm., antennae total 1.309 to 1.326 mm.. Ill 
 
 5 In 1917 George Shinji (Ent. News, vol. 27, February, 1917) described three 
 species, M. essiggi n.sp., M. woodworthi n.sp., and M. hyalinus (Monell), all of 
 which are undoubtedly but varieties of this species, M. quercus (Kalt.). 
 
28 MISCELLANEOUS STUDIES 
 
 0.442 mm.. IV 0.255 to 0.272 mm., Y 0.221 to 0.2465 mm., VI 0.255 mm. 
 (base 0.136 mm., spur 0.119 mm.), cornicles height 0.034 mm., diam- 
 eter at apex 0.034 mm., wing length 1.581 to 1.768 mm., width 0.663 
 to 0.68 mm., expansion 3.825 mm. 
 
 8. Genus Chromaphis "Walker 
 
 Walkei-, The Zoologist, p. 2001. 1870. Type Lachiius juglandicoJa Kalt. 
 
 23. Chromaphis juglandicola (Kalt.) 
 
 Figures 34, 35 
 
 Kaltenbach, Monog. d. Pflanzenlause, p. 151, 1843. Larhnus (oi'ig. desc). 
 Essig, Pom. Jour. Ent., vol. 1, p. 51, 1909. Callipteras (desc. vivi.). 
 Essig, Pom. Jour. Ent., vol. 4, p. 763, 1912 (list). 
 Davidson, U. S. Dept. Agr., Bull. 100, pp. 2-19, 1914 (desc. all forms). 
 
 Secords. — Juglans regia; San Francisco Bay region, southern California. 
 
 This walnut aphis is the most abundant and injurious of the 
 species attacking walnut in California. It is more or less abundant 
 throughout the San Francisco Bay region, vvliile in southern Cali- 
 fornia during certain seasons it is an important pest. Davidson 
 (1914) has described all the forms and studied the life history care- 
 fully, so but little comment is necessary. In 1915 tlu^ author 
 observed the young stem mothers on March 22 in Sunnyvale, Santa 
 Clara County. Three weeks later the second generation was well 
 advanced. From tlie first of May on, in 1916, the viviparae were 
 abundant on walnuts throughout San Diego County, from nursery 
 stock in San Diego to a few cultivated trees at Santa Ysabel (altitude 
 3000 feet). From the middle of October until well into December, 
 1916, the sexuales were found throughout Los Angeles and Riverside 
 counties. 
 
 9. Genus Callipterus Koch 
 Koch, Die Pflanzenlause, p. 208, 1855. Type Aphis juglandis Kalt. 
 
 The two members of this genus in California have been considered 
 heretofore as species of Moncllm Oestlund (genus 10), but according 
 to Davis" they can not lie so considered for in Monellm the wings are 
 laid flat on the abdomen when at rest. This is found only in Moncllia 
 caryella (Fitch). Incidentally it may be remarked that the species 
 known by that name in California does not have that habit, so 
 should really be placed in this genus, Calli-ptems Koch. However, as 
 it is identical with eastern specimens, except for this habit, the author 
 
 6 Essig, E. O., Beneficial and Injurious Insects of California, Mon. Bull. Cal. 
 Comm. Hort., vol. 4, p. 83, 1915. 
 
A SYNOPSIS OF THE AFHIDIDAE 29 
 
 has tliought best to retain it in MoneUis, at least for the time being. 
 
 Key to California Species 
 
 1. VI spur about equal to or slightly longer than VI base. Tibiae mostly pale. 
 
 caryae Monell 
 
 — - VI spur shorter than VI base. Tibiae entirely dark. Considerably larger than 
 
 preceding species calif omicus (Essig) 
 
 2J:. Callipterus calif omicus (Essig) 
 Figures 63, 64 
 
 Essig, Pom. Jour. Ent., vol. 4, p. 767, 1912. Monellia (orig. desc). 
 Davidson, U. S. Dept. Agr., Bull. 100, p. 34, 1914. MonelUa (list, key to 
 walnut aphids). 
 Seeords. — Juglanx coUfonUca (California black walnut); Santa Paula. 
 
 In 1912 Essig described this species from .specimens taken near 
 Santa Paula in July. 1911. No otlier definite collections are known 
 to the writer, although Essig reports it as more or less abundant on 
 the California black walnut throughout the southern part of the state. 
 Davidson has not found it in the San Francisco Bay region, nor has 
 the author ever observed it, either in the bay region or in southern 
 California. 
 
 25. Callipterus caryae IMonell 
 
 Figures 65, 66 
 
 Monell, U. S. Geol. Geog. Surv., Bull. 5, p. 31, 1879 (orig. desc). 
 Clarke, Can. Ent., vol. 35, p. 249, 1903 (list). 
 Davidson, Jour. Econ. Ent., vol. 2, p. 301, 1909 (list). 
 Davidson, Jour. Econ. Ent.,- vol. 3, p. 376, 1910 (list). 
 Essig, Pom. Jour. Ent., vol. 4, p. 764, 1912. Monellia (list). 
 Davidson, U. S. Dept. Agr., Bull. 100, pp. 19-20, 1914. Monellia (desc. 
 all forms). 
 
 Seeords. — Juglans regia, J. calif ornica; Berkeley, Stanford University, San 
 Jose, San Francisco Bay region. 
 
 This species is more or less common in the San Francisco Bay 
 region on walnuts. Davidson has described all the foi'ms and noted 
 its life history. The author has not taken the species. 
 
 10. Genus Monellia Ocsthnid 
 
 Oestlund, Minn. Geol. Nat. Hist. Surv., Bull. 4, p. 44, 1887. Type Aphis 
 caryella Fitch. 
 
 This genus, as described by Oestlund, differs from CalUpterus par- 
 ticularly in the position of the wings when the insects are at rest. In 
 Callipterus they are held roof-like over the body as is usual in aphids. 
 but ill Monellia they are laid flat on the abdomen. It includes but 
 the one species, M. caryella (Fitch). 
 
30 
 
 MISCELLANEOUS STUDIES 
 
 26. Monellia caryella (Fitch) 
 
 Figures 25, 67, 68 
 
 Fiteh, Insects N. Y., vol. 1, p. 163, 1855. Aphis (orig. dese. apt. vivi.). 
 Fitch, Ins. N. Y., vol. 3, p. 448, 1856. Callipterus (first dese. ala. vivi.). 
 Davidson, Jour. Econ. Ent., vol. 7, p. 132, 1914 (list). 
 Davidson, U. S. Dept. Agr., Bull. 100, pp. 26-34, 1914 (dese. all forms). 
 
 Records. — Juglans calif ornica, J. Nigra, J. regia; San Jose, Walnut Creek 
 (Davidson) ; Stanford University, May to June, 1915. 
 
 A more or less common species on both the native black walnut, 
 and the cultivated walnut in the San Francisco Bay region. This 
 species, while very similar to the preceding species, is probably the 
 more common of the two. The following- table of differences is taken 
 from Davidson ■/ 
 
 Form 
 Alate viviparous female 
 
 Callipicrus cari/ae Monell MonelUa caryella (Fitch) 
 
 Antennal joint III very Antennal joint III quite 
 slightly thickened bas- noticeably thickened 
 
 ally. 
 
 Sensoria 
 
 on antennal 
 
 joint III occupying 
 
 basal half or two- 
 thirds. 
 
 for its basal half. 
 
 Seu.soria on antennal 
 joint III occupying 
 basal third. 
 
 Pupa of viviparous 
 
 female 
 Oviparous female 
 
 Antennal joint VI and Antennal joint VI one- 
 its spur or filament third as long again as 
 subequal, or VI less its spur or filament, 
 than spur. 
 
 ■ Dusky knee spots often Dusky knee spots absent, 
 present. 
 
 Four longitudinal rows Six longitudinal rows of 
 
 of capitate spines. capitate spines. 
 
 Smaller than viviparous Larger than viviparous 
 
 female. female. 
 
 Four longitudinal rows Six longitudinal rows of 
 
 of cipitate spines. capitate spines. 
 
 This species is distinct from the preceding and according to Mor- 
 rison, who has examined eastern species, is structurally identical 
 except in the matter of the wings. He writes as follows : 
 
 'Davidson, W. M., Walnut aphides in California, IT. S. Dept. Agr., Bull. 100, 
 p. 28, 1914. 
 
A SYNOPSIS OF THE APHIDIDAE 31 
 
 I made a very careful study of specimens from California, sent me by David- 
 son, and of specimens collected both in Indiana and New York (type locality). 
 I was unable to find any structural differences that would definitely separate the 
 two lots of specimens, with the exception of the position of the wings. These are 
 laid flat when at rest in the eastern specimens, but are not so in the Californian 
 specimens, according to Davidson. In spite of this apparent agreement, I feel 
 that the two must be distinct. 
 
 If this is the ease, that the wings are not laid flat at rest, this 
 species must belong to the genus CalUptcrvs, and therefore cannot be 
 Monellia caryeUa (Fitch). However, the author has not had an 
 opportunitj- to study this carefully, so leaves it as it is, calling this 
 California species Monellia caryella (Fitch). 
 
 Because of the fact that all the species of aphids on walnut are 
 so closely related, and so very similar in structure, a key to separate 
 them, one from another, is given here. This key is adapted from 
 Davidson.^ 
 
 1. Cornicles quite evident, about as long as wide. 
 
 Chromaphis juglandlcola (Kalt.) 
 — ■ Cornicles barely perceptible, considerably wider than long 2 
 
 2. Tibiae of alate viviparae entirely dusky CaUipterus caUfornicus (Essig) 
 
 — Tibiae of alate viviparae mostly pale - 3 
 
 3. VI spur longer than VI base. Oviparous females with four longitudinal rows 
 
 of capitate hairs CalUpterus caryae Monell 
 
 — VI spur shorter than VI base. Oviparous females with six longitudinal rows 
 
 of capitate hairs MoneUia caryella (Fitch) 
 
 11. Genus Callipterinella Van der Goot 
 
 Van der Goot, Zur Systematic der Aphiden, 1913. Type Aphis (CalUpterus) 
 ietulariv^ Kaltenbaeh. 
 
 27. Callipterinella annulata (Koch) 
 
 Koch, Die Pflanzenlause, p. 1855. Chaitophorus (orig. desc). 
 Gillette, Jour. Econ. Ent., vol. 3, p. 367, 1910, Chaitophorus hetulae (Buck- 
 ton) (list). 
 Davidson, Jour. Econ. Ent., vol. 10, p. 292, 1917 (desc). 
 
 Becords. — Betula alba; Oakland, Walnut Creek (Davidson). 
 
 This species has been reported by Davidson as infesting the leaves 
 and shoots of the white birch in the San Francisco Bay region. It is 
 unknown to the author. 
 
 8 Ibid., p. 35. 
 
32 MISCELLANEOUS STUDIES 
 
 Tribe Chaitophorini Wilson (Lachnidea Mordw. and 
 Chaitopheri Mordw.) 
 
 Wilson, Can. Ent., vol. 42, pp. 385-387, 1910. 
 
 This tribe as considered by Wilson contains the following genera : 
 Arctaphis, Chaitophorus, Symydobius, Thomasia, and Sipha. The 
 author has followed Wilson's classification, having added, however, 
 two genera described later by Essig: viz., Micrella and Fullawaya. 
 Essig's genus Eichochaitophorm is a synonj-m of Arctaphis Walker 
 (see discussion under no. 27). Mordwilko's groups Lachnoidea and 
 Chaitophori are both included in this one tribe. In the former, Mord- 
 wilko includes Symydobivs and Pterochloriis, and in the latter, 
 Cl'Odobius, Mfl-anoxanthiis, and Chaitophorus. Botli Cladobius and 
 Melanoxanthus are included in this paper in the tribe Pterocomniini, 
 being synonyms of the genus Pteroconima Buckton. Following is a 
 description of tlie tribe Cliaitophorini as given by Wilson {op. cit.) : 
 
 Antennae, except in Sipha, always six-segmented ; in Siplia there are but five. 
 Length variable; antennal tubercles wanting; antennae, legs, and body covered 
 with hair-like bristles. Fore wings with two oblique veins and cubitus always 
 twice forked; hind pair with two cross veins. Nectaries (cornicles) variable in 
 length and size, but never longer than one-tenth the length of the body. The 
 genera in this tribe are somewhat similar to those in the tribe Callipterim, but 
 are easily distinguished by the shorter and heavier antennae and legs, as well as 
 by the finer and more hair-like bristles. 
 
 The following kej' to the Californian genera has been adapted from 
 Wilson and E.ssig: 
 
 1. Spur of sixth antennal segment at least three times as long as the segment 2 
 
 — Spur not three times as long as the segment. Cauda broadly rounded and 
 
 without knobbed tip 4 
 
 2. Spur more than five times as long as the segment; cornicles longer than the 
 
 base of the sixth segment Chaitophorus Koch 
 
 — Spur of sixth segment not more than five times as long as the segment; corn-. 
 
 icles not. longer than the base of the sixth segment 3 
 
 3. Cauda a knob on a quadrangular base (fig. 69). Spur about five times as 
 
 long as sixth segment Arctaphis Walker 
 
 — Cauda tapering to a blunt tip which is usually straight across, not being 
 
 rounded or constricted at the base (fig. 70). Spur but slightly more than 
 three times as long as the sixth segment _ Micrella Essig 
 
 4. Spur of sixth segment shorter or scarcely longer than the segment; antennae 
 
 nearly as long as the body Symydobius Mordwilko 
 
 — Spur considerably longer than sixth segment; antennae about one-half the 
 
 length of the body 5 
 
 5. Cornicles absent; body with lateral tubercles Fullawaya Essig 
 
 — Cornicles present; lateral body tubercles wanting Thomasia Wilson 
 
A SYNOPSIS OF TEE APHIVIDAE 33 
 
 GeiiTis Chaitophorus Koeli 
 Koch, Die Pflanzenlause, p. 1, 1854. Type Apliis accris Linn. 
 
 Tlu're are at present no .species of this genus in California; most 
 of the species hitherto placed in it are now considered as belonging 
 to the genus Thomasia Wilson. 
 
 12. Genus Arctaphis Walker 
 Walker, The Zoologist, p. 2000, 1870. Type aphis populi Linn. 
 
 This genus as defined by Wilson is represented in California by 
 two species: A. viminalis (Monell) and A. popiilifolii (Essig). The 
 latter was placed by Essig in a new genus, Eichocliaitophorus, but 
 there is not enough difference between these to warrant a new genus. 
 
 Key to California Species 
 
 1. Wings hyaline. Three-nine large sensoria on third antennal segment (fig. 71). 
 
 IV half as long against as V populifoUi (Essig) 
 
 — Wings subhyaline. About ten rather small sensoria on III. IV but very 
 little longer than V viminalis (Monell) 
 
 28. Arctaphis populifolii (Essig) 
 
 Figures 69, 71 
 
 Essig, Pom. Jour. Ent., vol. 4, p. 722, 1912. Eichocliaitophorus (orig. 
 
 dese.). 
 Davidson, Jour. Econ. Ent., vol. 3, p. 37-5, 1910. Chaitophorus populifoliae 
 (Fitch) (desc. male). 
 Davidson, Pom. Jour. Ent., vol. 3, p. 399, 1911. Chaitophorus populifoliae 
 
 (Fitch) (list). 
 
 Records. — Populus trichocarpa, Santa Paula (Essig), Berkeley, September, 
 1915; Populus fremontii, Stanford University and Penryn, Placer County (David- 
 son) ; Menlo Park, San Mateo County, October, 1914 (Morrison) : Berkeley, Sep- 
 _tember, 1915; El Cajon, San Diego County, June, 1916; Riverside, October, 1916. 
 
 In 1912 Essig described this species from specimens taken on 
 Populus trichocarpa at Santa Paula, and placed it in a new genus, 
 Eichochaitophorus. He separated this genus from Arctaphis for the 
 following reasons : 
 
 According to Wilson the Cauda [in Arctaphisl is a knob on a quadrangular 
 base. The anal plate is broadly rounded. In the new genus [Eich-ochaitophorus] 
 the style has a distinct neck and is situated on a very distinct conical base. The 
 anal plate is deeply notched in the middle so as to make it somewhat forked as 
 in the genus Callipterus, 
 
34 MISCELLANEOUS STUDIES 
 
 Although the anal plate is somewhat notched, there is scarcely 
 difference enough to warrant the forming of a new genus. In fact, 
 in many specimens one cannot tell whether or not a notch is present. 
 As to the Cauda, consisting of the tip, a distinct neck, and a distinctly 
 conical base, this is not greatly different from a cauda consisting of 
 a knobbed tip on a quadrangular base. The only practical difference 
 is in the base, being conical in one and quadrangular in the other. In 
 populifolii (E.ssig) the base seems to be conical, yet one cannot be 
 certain unless the specimen is mounted exactly. 
 
 This species, A. populifolii (Essig), as stated above, was described 
 from specimens taken on Populus trichoC'arpa at Santa Paula. In 
 1910 Davidson found a species on Populus fremonti at Stanford Uni- 
 versity, and the following j-ear at Penryn, Placer County, which he 
 listed as Chaitophorus popitlifoliae (Fitch). A careful studj' of 
 specimens from Davidson and the cotj'pes of Essig 's species convinced 
 the author that they were identical. Morrison writes that Davidson's 
 specimens are not C. populifoUae (Fitch), so Essig's species is distinct. 
 In September, 191.5, the author obsei'ved a great number of specimens 
 of this species on a weeping elm (Ulmus sp.) in Berkele.y, which was 
 in close proximity to some populars. However, none were seen to be 
 feeding on the elm, all being restless and wandering over the leaves 
 and branches. In southern California this is often found infesting 
 the empty galls of Thecabius popiiUmonilis Riley, such having been 
 observed in San Diego and Riverside counties. 
 
 29. Arctaphis viminalis (Monell) ? 
 
 Monell, TJ. S. Geol. Geog. Surv., Bull. 5, p. 31, 1879. Callipterus (orig. 
 
 desc). 
 Clarke, Can. Eiit, vol. 35, p. 248, 1903. Chaitophorus (list). 
 Davidson, Jour. Econ. Ent., vol. 3, p. 375, 1910. Chaitophorus (list). 
 Davidson, Pom. Jour. Ent., vol. 3, p. 398, 1911. Chaitophorus (list). 
 Essig, Pom. Jour. Ent., vol. 4, p. 716, 1912. Tliom-asia (list, key to Califor- 
 
 nian species of Thomasia). 
 Patch, Maine Agr. Exp. Sta., BuU. 213, p. 80, 1913. Chaitophorus (desc.). 
 
 Records — Sali-x spp. ; Watsonville, Santa Cruz County, and Newcastle, Placer 
 County (Clarke) ; Penryn, Placer County, and Stanford University (Davidson). 
 
 This species has been reported from Placer, Santa Clara, and 
 Santa Cruz counties on various species of willow. The true Chai- 
 tophonis viminalis Monell is an Arctaphis, but whether or not the 
 western species is the same as the eastern is a question. The author 
 has never seen specimens of either and is therefore unable to make 
 
A SYNOPSIS OF THE APHIDIDAE 35 
 
 any further coniment. He once thought the western species was iden- 
 tical with Thomasia salicicola (Essig), to which Morrison considers 
 it very closely related, but Davidson assures him the two are distinct. 
 
 13. Genus Micrella Essig 
 
 Essig, Pom. Jour. Ent., vol. 4, p. 71fi, 1012. Type if. vioncUa n.sp. 
 
 30. Micrella monella Essig 
 
 Figures 70, 72 
 Essig, Pom. Jour. Eut., vol. 4, p. 717, 1912 (orig. desc). 
 Secords. — SaUx lasuAepis, Oxnard (Essig) ; S. laevigata, Santa Paula (Essig). 
 
 This species wa.s taken twice by Essig, who described it, in 1910 
 near Oxnard, and in 1911 near Santa Paula. Since then it has never 
 again been found. The author has had access to cotype specimens in 
 Essig 's collection. 
 
 14. Genus FuUawaya Essig 
 Essig, Pom. Jour. Ent., vol. 4, p. 735, 1912. Type F. saliciradicis n.sp. 
 31. FuUawaya saliciradicis Essig 
 
 Figure 7.5 
 Essig, Pom. Jour. Ent., vol. 4, p. 737, 1912 (orig. desc). 
 Eecord. — Salix laevigata, Santa Paula, August, 1911 (Essig). 
 
 On the roots of willow near Santa Paula, Essig once found a large 
 number of aphids, the greater part of which were apterae, although 
 a few alates were present. Unable to identifj' them with any known 
 species, or to fit them into anj' genus, he described tliem as this species. 
 Since tlu'U they have not been taken. The author has had access to 
 cotype specimens in Essig 's collection. 
 
 15. Genu.s Thomasia Wilson 
 Wilson, Can. Ent., vol. 42, p. 386, 1910. Type Chaitophorus populicola Thomas. 
 
 This genus is separated from Chaitophorus principally by the 
 comparative lengths of the antennae and the comparative lengths of 
 the spur of the sixth antennal segment. In Chaitophorus (type Aphis 
 aceris Linn.) the antennae are almost as long as the body, and the 
 spur of the sixth segment is over five times as long as the base. In 
 this genus the antennae are but about one-half as long as the body, 
 and VI spur is but slightly longer than VI base. 
 
36 MISCELLANEOUS STUDIES 
 
 Key to California Species 
 
 1. Wings hyaline 2 
 
 — Wings with veins clouded (fig. 273) populicola (Thomas) 
 
 2. Ill longer than VI (including spur) negundinis (Thomas) 
 
 — Ill not longer than VI (including spur) 3 
 
 3. IV with secondary sensoria crucis Essig 
 
 — IV without secondary sensoria salicicola Essig 
 
 32. Thomasia crucis Essig 
 
 Figure 7(5 
 Essig, Pom. Jour. Ent., \ol. 4, p. 742, 1912 (orig. desc). 
 Secords. — Salix macrostaoluia, Santa Paula, August, 1911 (Essig). 
 
 Essig once found this species ou the leaves of willow near Santa 
 Paula. Since then it has never again been taken. The author has 
 had access to cotype specimens in Essig 's collection. 
 
 33. Thomasia negundinis (Thomas) 
 
 Thomas, 111. Lab. Nat. Hist., Bull. 2, p. 10, 1878. Chaitophorus (orig. 
 
 desc). 
 Sanborn, Kan. Univ. Sei., Bull. 3, p. 3."), 1904. Chaitophonui (desc). 
 Davidson, Jour. Econ. Ent., vol. 3, p. 37fi, 1910. Chaitophonis (list). 
 Essig, Pom. Jour. Ent., vol. 4, p. 716, 1912 (list). 
 
 Records. — Acer negundo, Stanford University (Davidson); Salt Marshes, Palo 
 Alto, May, 1912 (Morrison). 
 
 This species of Thomasia is quite common on box elder in the 
 vicinity of Stanford University and Palo Alto. The author has never 
 taken specimens, nor had access to any. Morrison writes that 
 although lie has never had access to eastern specimens of T. negundinis 
 (Thos.) for comparison he is not able to convince himself that the 
 western species is negundinis. The author is unable to form any 
 opinion at present, having never seen specimens, hence lists the species 
 as Davidson has done. 
 
 3-t. Thomasia populicola (Thomas) 
 
 Figures 77, 275 
 
 Thomas, 111. Lab. Nat. Hist, Bull. 2, p. 10, 1878. Chaitophonis (orig. 
 
 desc. ) . 
 Essig, Pom. Jour. Ent., vol. 1, p. 98, 1909. Chhitophorus (desc). 
 Essig, Pom. Jour. Ent., vol. 4, p. 716, 1912 (list). 
 
 accords. — Populus s-pp., Salix spp., Santa Paula (Essig) ; Riverside, May, 1917; 
 Populus sp.. Canton, Broadwater County. Montana, July, 191.5, E. W. Haegele; 
 Edna Canon, Boxelder County, Utah, August, 1916, R. W. Doane. 
 
A SYNOPSIS OF TSE APHIDIDAE 37 
 
 This species has been reported by Essig from Ventura County. 
 The author lias never taken tlie alatcs, but has had the opportunity 
 of examining Essig's specimens, and specimens from Montana and 
 Utah taken by Haegele and Doane. It is easily distinguished from 
 other members of the genus bv the broad, dark wing veins. 
 
 35. Thomasia salicicola (Essig) 
 
 Figure 78 
 
 Essig, Pom. Jour. Ent., vol. 3, p. 532, 1911. Chaitophorus (oi'ig. desc). 
 Davidson, Jour. Eeon. Ent., vol. 3, p. 375, 1910. Chaitophorus nigrae Oest- 
 
 hmd (?) (list). 
 Davidson, Pom. Jour. Eut., vol. 3, p. 398, 1911. Chaitophorus nigrae Oest- 
 
 lund (?) (list). 
 Essig, Pom. Jour. Ent., vol. 4, p. 619, 1912 (note). 
 Essig, Pom. Jour. Ent., vol. 4, p. 716, 1912 (list). 
 
 Becords. — Salix laevigata, Santa Paula (Essig), Salix nigra, Lakeside, San 
 Diego County, April, 1916; Populus trichocarpa, Santa Paula (Essig); Salix sp., 
 San Jose, Stanford University, Penryn, Placer County (Davidson), Fillmore, Ven- 
 tura County, March, 1911 (Essig). 
 
 Essig reported this from Ventura Count}', and the author has 
 found it in San Diego County. It was observed to be in large colonies 
 on the leaves and leaf petioles of the tender growth of willow, in 
 company with Siphocorijne carpreue (Fabr.). Specimens taken by 
 Davidson and listed as Chaitophorus nigrae Oestlund prove to be 
 identical with this species. 
 
 16. Genus Symydobius Mordwilko 
 
 Mordwilko, Rap. Lab. Zool. Kap. Imp. Varch. Univ., 1895. Type Aphis 
 oblonga Heyden. 
 
 Key to California Species 
 
 1. Anal plate half-moon-sliaped 2 
 
 — Anal plate bilobed (fig. 271) ; cornicles pale, appearing white in life; antennae 
 
 ivith about six to eight secondary sensoria on III, and one or two on IV 
 (fig. 272 ) chrysolepis Swain 
 
 2. Spur of VI but a short thumb-like projection; sensoria on III numbering about 
 
 six to ten; none on IV agrifoliae Essig 
 
 — Spur of VI longer, being equal to or longer than base of VI ; fifteen to twenty 
 
 sensoria on III, one or two on IV (figs. 73, 74) 3 
 
 3. Antennae for the most part dark, being dark brown or black; spur and base 
 
 of VI equal, lateral abdominal tubercles present in aptcrae. 
 
 macrostachyae Essig 
 
 — Antennate for the most part pale, being light brown or amber; spur of 
 
 VI usually slightly longer than base ; apterae without lateral abdominal 
 tubercles sallcicorticis Essig 
 
38 MISCELLANEOUS STUDIES 
 
 36. Symydobius agrifoliae Essig 
 
 Essig, Univ. Calif. Publ. Entom., vol. 1, pp. 311-317, 1917 (orig. desc). 
 Records. — Quercus agrifolia; Santa Paula (Essig). 
 
 This interesting aphid was taken in Ventura County on live oak 
 during 1911. It differs from other members of this genus in the 
 extremely short spur of the sixth anteunal segment. The coloration 
 is very similar to that of the next species, but the length of VI spur 
 and the fact that the anal plate is not bilobed serves to distinguish it. 
 
 37. Symydobius chrysolepis Swain 
 Figures 269 to 274 
 Swain, Trans. Am. Ent. Soc, vol. 44, p. 6, 1918 (orig. desc). 
 Eecords. — Quercus chrysolepis; Alpine, San Diego County (Swain). 
 
 This is a medium sized, brownish colored aphid found in 1916 
 infesting the terminal twigs and leaf petioles of maul oak in San 
 Diego County. Its pale white cornicles are very conspicuous, and 
 serve as a distinguishing character. The anal plate is bilobed, a char- 
 acter not found in other members of the genus, and one which may 
 be sufficient for the separation of the species (and S. albuiphus Davis, 
 in which the anal plate is also bilobed) from Symydobius into a new 
 genus. However, the author believes it best to retain them in this 
 genus at present.. The apterous females were found to be heavily 
 parasitized by the chalcid fly, Clostcrocerus utahomis Crawford var. 
 californicus Girault. 
 
 38. Sjonydobius macrostachyae Essig 
 Figure 73 
 Essig, Pom. Jour. Ent., vol. 4, p. 727, 1912 (orig desc). 
 Records. — Salix macrostachi/a ; Santa Paula (Essig), Fresno, June, 1915. 
 
 Twice has this species been taken, once by Essig near Santa Paula 
 and once by the author along the San Joaquin River near Fresno. 
 It is found in fairly large colonies on the younger stems of willow. 
 These colonies consist for the most part of apterae, only a very few 
 alates being present. 
 
A SYNOPSIS OF THE APHIDIDAE 39 
 
 39. Symydobius salicicorticis Essig 
 
 Figure 74 
 Essig, Pom. Jour. Ent., vol. 4, p. 731, 1912 (orig. desc). 
 Jtecord. — Salix laevigata; Santa Paula (Essig). 
 
 Together with specimens of Fullawaya saliciradicis Ftssig, this was 
 taken on willow along the Santa Clara River near Santa Paula in 
 August, 1911. The colonies are found on the bark near the surface 
 of the ground either just above or just below it. Essig reports that it 
 is preyed upon quite extensively by the larvae of an undetermined 
 species of syrphus fly. The author has had access to cotype specimens 
 in Essig 's collection. 
 
 Group Lachniria Passerini 
 
 Passerini, Gli Afidi, 1860. 
 
 In this group there are included two tribes, Lachnini Del Guercio 
 and Pteroeommini "Wilson, following "Wilson. Mordwilko places but 
 the one tribe Lachnini in this group, including the genus Pterocamma 
 Buckton in the tribe Chaitophori. However, to the author the group- 
 ing followed here seems more natural. The following key is adapted 
 in part from Borner (Sorauer, Pflanzenkrankheiten, vol. 3, p. 665, 
 1913) : 
 
 Sixth antennal segment with a short, thick (thumb-like) projection. Cornicles 
 conical (fig. 91) or wart-like. Empodial hair short, and oftentimes indistin- 
 guishable (fig. 79) Tribe Lachnini 
 
 Sixth antennal segment with a slender projection (VI spur) which is about as 
 long as the segment (VI base). Cornicles cylindrical or clavate (figs. 81, 82). 
 Empodial hair practically as long as the claws (fig. 80). Tribe Pteroeommini 
 
 Tribe Pteroeommini Wilsoii 
 
 Wilson, Ann. Ent. See. Am., vol. 8, pp. 347-358, 1915. 
 
 This tribe, as considered by Wilson, contains but the one genus, 
 Pterocamma Buckton. In a former paper (Can. Ent., vol. 43, p. 384, 
 1910) he recognized two genera: the one, Melanox-antherium Schoute- 
 den, in which the cornicles were swollen or vasiforai, and the other, 
 
40 MISCELLANEOVS STUDIES 
 
 Pterocaiiuna Buekton, in which the cornicles were cylindrical. He 
 states in his later paper : "... after having further studied the group 
 I am of the opinion that such a divsion is illogical, and if a division 
 is necessary each species should form a different genus. It, therefore, 
 seems more practical to confine all the species to a single genus. ' ' The 
 characters of this tribe and genus are as follows : 
 
 Antennae with six segments and reaching near the base of tlie abdomen. 
 Wings normally with venation as in Aphis. Nectaries [cornicles] short, but 
 clavate. Cauda short and broadly rounded at the tip as in Lachnini. Entire 
 body, antennae, and legs covered with long hairs as in Lachnini. As has already 
 been pointed out by Oestlund, this group appears intermediate between the Ch-aito- 
 phorini and the Lachnini. Their habits and actions being in different ways similar 
 to both. 
 
 17. Genus Pterocomma Buekton 
 
 Buekton, Monog. Brit. Aphides, vol. 2, p. 143, 1879. Type P. pilosa 
 Buekton. 
 
 Key to California Species 
 
 1. Cornicles abruptly constricted at distal end, and without a distinct flange 
 
 (fig. 81), the diameter of the opening being less than the diameter of the 
 smallest part of the cornicle. Wing veins broad and shaded. 
 
 flocculosa (Weed) 
 
 — Cornicles not so abruptly constricted and with a distinct flange. Wing 
 
 veins normal 2 
 
 2. Cornicles about twice as long as their greatest diameter ....smithiae (Monell) 
 
 — Cornicles considerably longer than greatest diameter, and longer tlian hind 
 
 tarsus populifoliae (Fitch) 
 
 40. Pterocomma flocculosa (Weed) 
 Figure 81 
 
 Weed, Insect Life, vol. 3, p. 291, 1891. Melanoxaiitlius (orig. desc.). 
 Wilson, Ann. Ent. Soc. Am., vol. 8, p. 350, 1915 (desc). 
 
 Records. — Salix sp., Berkeley, March, 1915; 1916 (Essig). 
 
 In his paper on Pterocomma Wilson states that this species does 
 not occur on the Pacific Coast. However, in March, 1915, the author 
 found it rather abundantly on willow on tiie campus of the University 
 of California in Berkeley. During the 1916 season Essig observed it 
 to be quite common in Berkeley. Tlie species is easily recognized in 
 life by the white cottony flocculence covering the colonies on the bark. 
 
A SYNOPSIS OF THE APHIDIDAE 41 
 
 41. Pterocomma populifoliae (Fitch) 
 
 Figures 82, 83 
 
 Fitch, Cat. Homop. N. Y., p. 66, 1851. Aphis (orig. desc). 
 
 Davidson, Jour. Econ. Ent., vol. 2, p. 300, 1909. Cladobius rufuhis n.sp. 
 
 (desc). 
 Davidson, Jour. Econ. Eut., vol. 3, p. 375, 1910. Cladobius rufulus Dvdn. 
 
 (list). 
 Essig, Pom. Jour. Ent., vol. 4, p. 786, 1912. Melanoxantherium rufulum 
 
 (Dvdn.) (desc). 
 Wilson, Ann. Ent. Soe. Am., vol. 8, p. 353, 1915. Pterocomma populca 
 
 (Kalt.) (desc). 
 Baker, Can. Ent., vol. 48, pp. 280-282, 1916 (desc). 
 
 Eecords. — Salix sp. ; Stanford University (Davidson); Santa Paula (Essig); 
 Walnut Creek, March, 1915 (Davidson) ; Grossniont, San Diego County, March, 
 1916; Lakeside, San Diego County, April, 1916; Stanford University, May, 1912 
 (Morrison); Populus sp. ; Stanford University (Davidson); Palo Alto, March, 
 1915; Populus caroliniana, Banning, Riverside County, April, 1917. 
 
 Tliis is a widely distributed species in California on various species 
 of poplars and willows. Davidson first found it in 1909, describing 
 it as a new species. In 1915 Wilson stated that it was synonymous 
 with P. populea (Kalt.), but specimens sent him by the author he 
 determined as P. bicolor (Oestlund). According- to his paper the 
 cornicles of populca (Kalt.) are about equal in length to the hind 
 tarsi. Californian specimens have the cornicles considerably longer 
 than the hind tarsi, but not twice as long as he states they are in 
 iicolor (Oestlund). His figures of the antennae show that in populca 
 VI base and spur are subequal, and in bicolor the spur is considerably 
 longer than the base. The latter is true for the Californian species. 
 His color notes of populra fit the Californian species very well. Baker 
 identified Aphis populifoliae Fitch as a Pterocovimu and places 
 rufulus (Davidson) as a synonym. From a study of specimens taken 
 in Santa Paula, Grossmont, Lakeside, Stanford University, and Wal- 
 nut Creek, the author finds that Baker's description of populifoliae fits 
 this species verj' well. Below are the measurements in microns of 
 four alate specimens, together with the measurements of cornicles, 
 antennae, and hind tarsi of one from Lakeside. (This was preserv'ed 
 for several months in alcohol before being mounted for study, and had 
 shrunk considerablj'. ) 
 
 An examination of the following table shows that in the California 
 specimens the cornicles are always considerably longer than the hind 
 tarsi, but never twice as long, and that the spur of six is always longer 
 than the base, except in one case. This specimen is considerably 
 
42 
 
 MISCELLANEOUS STUDIES 
 
 bD-C Ol Cl O O 'O 
 
 C f n '-' GO O CO t^ 
 
 ;^ ^ lO '^ T^ o; y:: 
 
 ^ £ CO CO lO TtH -^ 
 
 ffi- 
 
 O00OrHC^COt-<MC3 
 rHT-l(M(MC<lCM(M(M 
 
 O O -rt* --44 Oi ?D 
 Tf -^ Ir- t^ 00 O 
 CO CO CO CO (M CO 
 
 ^ to t^ 00 (M to O t^ 
 C0C0C0c0t:^O"*l0 
 r-^i-l>-t(M(MCOCOCO 
 
 t- .-' rH T*^ -^ 
 00 !M O O O 
 r- eg <M (M (M 
 
 •--oo--*^oi>-t-oo 
 
 C^J CO r- 1 -t irD IQ tJ* "* 
 
 wcgcMcococococo 
 
 . OlOCOOt^t--^!— ti>-o 
 p* !>• 1'3 !M •** »0 »0 t^ Oi lO -t* 
 
 '-' OICJCOCOCOCOCOCOCOCO 
 
 QOooo.-i{3icgaoc<i'*i-( 
 
 t^t-f-HCOOli-Ht^l-t-^O 
 
 lo lo lO L-^ :d :o LT to lo o 
 
 
 CJOlt-OtOOCOCO 
 
 ■^"rt<t-00C5OO'-* 
 ^rHT-l,-iCI'M(MCl 
 
 . - o 
 
 (M T-H 
 
 
 >*-^ Oi 00 
 
 13 tie lO rH 
 
 Cm a c-l C-l 
 
 o 
 
 a 
 w 
 
 o 
 
 li .t c. .& 
 
 1 „- 5 ^. ^ s ^ ^ 
 
 a: ^ c/i ^ h-1 
 
 
 (-< 
 
 ^^ G 
 
 to 
 
 > 
 
 f3 
 
 Oi 
 
 
 c3 p 
 
 •"^ 
 
 
 S <^ 
 
A SYNOPSIS OF THE APHIDIDAE 43 
 
 smaller than the others and has many more secondary sensoria, being 
 a male. From this evidence this species is the same as Baker lists as 
 P. populifoliae (Fitch) and should be so considered. The author has 
 reared a number of specimens of a species of Aphidius from material 
 obtained near Stanford University in May, 1915. 
 
 42. Pterocomma smithiae (Monell) 
 
 Moiiell, U. S. Geol. Geog. Surv., Bull. 5, p. 32, 1879. Cliaitophorus (orig. 
 
 desc). 
 Davidson, Jour. Ecou. Eut., vol. 2, p. 300, 1909. Cladobius salicti (Harris) 
 
 (list). 
 Davidson, Jour. Econ. Ent., vol. 3, p. 375, 1910. Cladobius salicti (Harris) 
 
 (list). 
 Essig, Pom. Jour. Eut., vol. 4, p. 780, 1912. Melanoxantherium salicti 
 
 (Harris) (list). 
 Wilson, Ann. Ent. Soc. Am., vol. S, p. 3.55, 1915 (desc). 
 
 Mecctrds. — Salix spp., Stanford University (Davidson, Morrison). 
 
 Both Davidson and Morrison have taken this species in the vicinity 
 of Stanford University on various species of willow. According to 
 Wilson, with whom Morrison and Baker agree, this is P. smithiae 
 (IMonell), the salicti of Harris being synonymous. The sexuales were 
 observed by Davidson in October, the eggs hatching in January. 
 
 Tribe Lachnini Del Guercio 
 
 Del Guercio, Redia, vol. 5, 1908. 
 
 This tribe is represented in California by three genera, viz., Essig- 
 ella Del Guercio, Tubcrolachnus Mordwilko, and Lachmts Burmeister, 
 while there are six genera included in the tribe as it is here considered. 
 Following is a brief characterization of the tribe adapted from Mord- 
 wilko: 
 
 The body and appendages are very hairy, and usually quite large. The eauda 
 is absent, the cornicles cupola-shaped, being black or brown in color. Sometimes 
 they are reduced to mere pores or not fully developed [Lachnus taxifoUa Swain]. 
 The antennae in general are not longer than the head and thorax, six-jointed 
 [except in Essigella Del Guercio], with the spur of the sixth segment very short, 
 not being as long as the segment itself. The beak is almost always elongated, 
 generally reaching to or beyond the middle of the abdomen. All this group possess 
 the anatomical peculiarity that the narrowed hind end of the stomach is covered 
 with the intestine. The stigma of the fore wing is elongate linear [in Longi- 
 stigma Wilson it reached past the tip of the wing (fig. 89)]. The cubitus is twice- 
 branched. 
 
44 MISCELLANEOUS STUDIES 
 
 All the California species with the exception of Tuberolachnus vimi- 
 nalis (Fonsc), which lives on willow, are found on eouifei-s — Piims 
 sp., Pseudotsuga sp., or Picea sp. 
 
 Following is a key to the genera, adapted from Del Guercio, Wil- 
 son and Essig. In this key are included not only the California 
 genera but the other three as well, in that an understanding of the 
 characters is thus made easier. 
 
 1. Antennae six-segmented - 2 
 
 — Antennae five-segniented (fig. 83) Essigella Del Guercio 
 
 2. Stigma exceptionally long, reaching beyond the tip of the wing (fig. 84). 
 
 Longistigma (Wilson) 
 
 — • Stigma not exceptionally long, not reaching beyond tlic tip of the wing 
 
 (fig. 85) " - 3 
 
 3. First joint of the hind tarsus much shorter than half the second (fig. 86) .... 4 
 — ■ First joint of the hind tarsus equal to or slightly longer than half the second 
 
 (fig. 87) Eulachnus Del Guercio 
 
 4. Abdomen with horn-like tubercle on median dorsum between the cornicles, 
 
 (Sometimes this cannot be made out in specimens mounted in balsam, but 
 it is always readily discernible in fresh or alcoholic material). 
 
 Tuberolachmus Mordwilko 
 
 — Abdomen without horn-like tubercle 5 
 
 5. Bases of first and second discoidal close together; third discoidal often very 
 
 faint; wings slightly if ever clouded (fig. 85) Lachnus Burmeister 
 
 — Bases of first and second discoidals not so close together as in Lachnus Burm. ; 
 
 third discoidal plain ; wings often darkly clouded Pterochlorus Bondani 
 
 18. Genus Essigella Del Guercio 
 
 Del Guercio, Eer. di patal. veg., vol. 3, p. 328, 1909. Type Lachiius cali- 
 fiirnicus Essig. 
 
 43. Essigella californica (Essig) 
 
 Figures 3, 5, 83 
 
 Essig, Pom. Jour. Ent., vol. 1, p. 1, 1909. Larhnus (orig. desc). 
 
 Del Guercio, Pom. Jour. Ent., vol. 1, p. 73, 1909 (translation by C. F. 
 
 Baker of Del Guercio 's paper listed above). 
 Essig, Pom. Jour. Ent., vol. 4, p, 773, 1912 (list). 
 Essig, Pom. Jour. Ent., vol. 4, p. 780, 1912 (desc). 
 
 Becords. — Pinus radiata; Claremont, Los Angeles County, and Santa Paula 
 (Essig); Pinus saiiniana, Stanford University, March, 1915; Finns spp., Stan- 
 ford University, March and April, 1912 (Morrison) ; Ontario, San Bernardino 
 County, January, 1917. 
 
 This curious little aphid, described b.y Essig from specimens taken 
 in Claremont, Los Angeles County, on Pinus radiata, has since been 
 found in several parts of the state. "Wilson has taken it in Oregon 
 on Pseudotsuga taxifolia, and Patch in Maine on Pinus strobus. It 
 
A SYNOPSIS OF THE APHIDIDAE 4S 
 
 is a small, slender, long-legged aphid, that clings fast to the pine 
 needles and is extremely difficult to see. However, if a branch of 
 pine is struck sharply and with considerable force over a white paper 
 or cloth, a large number of these aphids will jar off. 
 
 19. Genus Tuberolachnus Mordwilko 
 
 Mordwilko, Ann. Mus. Zool. d. I'Acad. Imp. Sci., vol. 13, p. 374, 1908. 
 Type Aphis viminalis Fonso. 
 
 44. Tuberolachnus viminalis (Fonsc.) 
 
 Figure 8(5 
 
 Boyer de Ponscolmbe, Ann. Ent. Soc. France, vol. ]0, p. 162, 1841. Ai>liis 
 
 (orig. desc). 
 Davidson, Jour. Econ. Ent., vol. 2, p. 299, 1909. Laehnus (list). 
 Davidson, Jour. Econ. Ent., vol. 3, p. 374, 1910. Laehnus (list). 
 Davidson, Pom. Jour. Ent., vol. 3, p. 398, 1911. Laehnus dentatus Le 
 
 Baron (list). 
 Essig, Pom. Jour. Ent., vol. 4, p. 774 (772), 1912 (list). 
 
 Records. — Salix spp., Stanford University and Penryn, Placer County (David- 
 son) ; Ventura County (Essig); Stanford University, November, 1914; Berkeley, 
 July, 1915; Eiverside, July, 1916. 
 
 This extremely large aphid, which lives in large colonies on the 
 branches of various species of willows, is found throughout the San 
 Francisco Bay region, Sacramento Valley, and southern California, 
 although it is not at all common. Davidson reports considerable 
 parasitization by a species of Epherdkis, and Essig infection from 
 some bacterial or fungus disease. The large size and the presence of 
 a dorsal abdominal tubercle are distinguishing characters. 
 
 20. Genus Laehnus Burmeister 
 
 Burmeister, Handbuch d. Entomologie, p. 91, 1835. Type Laehnus fadatus, 
 n.sp. 
 
 This is the third largest genus of aphids in regard to the number 
 of species in California. All the species are to be found on various 
 conifers, usually feeding through the bark of the branches or trunk. 
 Characters for distinguishing the species are hard to obtain, and 
 those used by the author in the following key are of no value except 
 with specimens of the alate viviparae. This key is not at all adequate, 
 and is offered here merely as an aid. The author understands that 
 Wilson is preparing a monograph of this genus, which will undoubt- 
 edly prove quite valuable. 
 
46 MISCELLANEOVS STUDIES 
 
 Key to California Species 
 
 1. Beak reaching considerably beyond the third coxa 2 
 
 — Beak at most bareh' reaching to the third coxa 8 
 
 2. Beak reaching almost to or even beyond the tip of the abdomen 3 
 
 — - Beak not reaching to the tip of the abdomen 4 
 
 3. First joint of hind tarsus more than cue-third as long as the second joint. 
 
 Legs black except the base of the femora and a broad ring near the base 
 of the tibiae ponderosa Williams 
 
 — First joint of hind tarsus scarcely more than one-fourth as long as the second 
 
 joint. Legs pale at the base of the femora and tibiae, black at tips. 
 
 oregonensis Wilson 
 
 4. Body exceptionally large, being over 4 mm. long, usually about 5 mm., and 
 
 over 2 mm. wide _ 5 
 
 — Body of average size, being from 2.5 mm. to 3 mm. long, and from 0.73 to 
 
 1.2 mm. wide 7 
 
 5. Third segment of antennae with many sensoria (eight or more), (figs. 88, 
 
 89) ; 6 
 
 — Third joint of antennae with but few or no sensoria, at most with one or two. 
 
 First joint of hind tarsus a little less than half as long as the second. 
 On Finns sabiniana sabinlanus n.sp. 
 
 6. Third joint of antennae with about 8rl2 sensoria (fig. 88). Tibiae with a pale 
 
 ring near the base. First joint of hind tarsus scarcely more than one-third 
 the length of the second. On Picea sp vanduzei n.sp. 
 
 — Third joint of the antennae with 19-20 sensoria (fig. 89). Tibiae without 
 
 pale ring near base. First joint of hind tarsus almost one-half the length 
 of the second. On Pinus sp. and Abies sp ferrisi Swain 
 
 7. Beak not reaching to the middle of the abdomen. Segment three of the 
 
 antennae almost as long as the fourth, fifth, and sixth together. Apex 
 of stigma meeting the margin of the wing in an acute angle, and not 
 terminated by a distinct vein (fig. 92). On Pseudotsuria iaxifoUa. 
 
 pseudotsugae Wilson 
 — • Beak reaching beyond the middle of the abdomen. Third antennal segment 
 not nearly so long as the fourth, fifth, and sixth together. Apex of stigma 
 meeting the wing margin in an obtuse angle, and terminated by a distinct 
 vein (fig. 93). Apterous viviparous females with a distinctive pattern on 
 dorsum of abdomen. On Thuya occidentalis tujafilinus (Del Guercio) 
 
 8. First joint of hind tarsus longer than one-fourth the second 10 
 
 — ■ First joint of hind tarsus less than one-fourth the second 9 
 
 9. Third antennal segment without sensoria (fig. 94). Body robust, being of the 
 
 usual Lachnus shape. Third discoidal twice-branched, only occasionally 
 once-branched. On Abies grandis occidentalis Davidson 
 
 — Third antennal segment with several irregular sensoria (fig. 9.")). Body long 
 
 and narrow, being somewhat the shape of Essigella californica (Essig). 
 Third discoidal simple or once-branched. On Pinus sp. 
 
 pini-radiatae Davidson 
 10. Cornicles very poorly developed, seemingly absent in some cases (fig. 103). 
 Segment three of antennae with five-seven large circular sensoria which are 
 hardly distinguishable (fig. 106). On Pseudotsuga taxifolia. 
 
 taxifolia Swain 
 
 — Cornicles normal (fig. 97), being quite conspicuous. Third antennal segment 
 
 with two-four clearly defined sensoria (fig. 101). On Picea glehni. 
 
 glehnus Essig 
 
A SYNOPSIS OF THE APHIDIDAE 47 
 
 45. Lachnus ferrisi Swain 
 Figures 89, 91 
 
 Davidson, Jour. Econ. Ent., vol. 2, p. 299, 1909. Lachnus ahietis Fitch 
 
 (list). 
 Davidson, Jour. Econ. Ent., vol. 3, p. 374, 1910. Lachnus niittis Fitch 
 
 (list). 
 Eseig, Pom. Jour. Ent., vol. 4, p. 773, 1912. Laohnus ahietis Fitch (list). 
 Swain, Trans. Am. Ent. Soc, vol. 44, p. 9, 1918. 
 
 Kecords. — Allies concolor, Stanford University (Davidson) ; Pi7ius sp., Stan- 
 ford University (Swain). 
 
 This large laclniid, recently described by the author, has been 
 found only in the vicinity of Stanford University, in 1909 and 1910 
 by Davidson on lowland fir, and in 1915 by Ferris on some young 
 pine trees. Since then it has not been observed. 
 
 46. Lachnus glehnus Essig 
 Figures 96, 97 
 Essig, Pom. Jour. Ent. Zool., vol. 7, pp. 180-187, 1915 (orig. desc). 
 Becord. — Picea glehni, Sacramento (Essig). 
 
 Essig described this species from specimens taken on a Japanese 
 spruce in Capitol Park, Sacramento, in 1912. At the time it wa-s so 
 abundant that control measures were deemed necessary'. The author 
 has had access to the type specimens in E.ssig's collection. 
 
 47. Lachnus occidentalis Davidson 
 
 Davidson, Jour. Eeon. Ent., vol. 2, p. 300, 1909 (orig. desc. apterae). 
 Davidson, Jour. Econ. Ent., vol. 3, p. 374, 1910 (list). 
 Essig, Pom. Jour. Ent., vol. 4, p. 773, 1912 (list). 
 WUson, Can. Ent., vol. 44, p. 193, 1912 (desc. all forms). 
 
 Becords. — Abies grandis, Stanford University (Davidson, Morrison, Ferris 
 and the author) ; Ahies concolor, Corvallis, Oregon (Wilson). 
 
 This species is practically always present on a lowland fir tree in 
 the cactus garden of the Stanford University grounds. Wilson has 
 found it in the vicinity of Corvallis, Oregon, on white fir. Davidson 
 states that it is heavily preyed upon by the larvae of Syrphus arcuatus 
 and Syrphus opinator. 
 
48 MISCELLANEOUS STUDIES 
 
 48. Lachnus oregonensis "Wilson 
 Wilson, Trans. Am. Ent. Soc., vol. 12, p. 103, 1915 (orig. desc). 
 Becord. — Pinus contorta, Oregon and California (Wilson). 
 
 There ha.s been no published record of tliis species from California. 
 Wilson wrote the autlior some time ago that he had taken it in this 
 state, although he gave no definite locality. The author has never seen 
 specimens. 
 
 49. Lachnus pini-radiatae Davidson 
 
 Figure 95 
 
 Davidson, Jour. Econ. Eut., vol. 2, p. 299, 1909 (orig. dese.). 
 Davidson, Jour. Econ. Ent., vol. 3, p. 374, 1910 (list). 
 Davidson, Pom. Jour. Ent., vol. 3, p. 398, 1911 (list). 
 Essig, Pom. Jour. Ent., vol. 4, p. 773, 1912 (list). 
 Essig, Pom. Jour. Ent., vol. 4, p. 785, 1912 (descriptive note). 
 
 Mecords. — Pinus radiata, Stanford University (Davidson), August, 1914, April, 
 1915 (author), March, 1916 (K. B. Brown); Pimis ponderosa, Bowman, Placer 
 County, November, 1911 (H. H. Bowman), Berkeley, March, 1915 (Geo. Shinji) ; 
 Pinus sahiniana, Penryn, Placer County (Davidson). 
 
 This is a fairlj' small, slender-bodied, long-legged lachnid found 
 infesting the needles of various pines in the San Francisco Bay region 
 and in the Sacramento Valley. They are easily recognized on the 
 needles by the wliitisli mass of flocculenee which covers their bodies. 
 
 50. Lachnus ponderosa Williams 
 
 Figure 104 
 
 Williams, Univ. Neb. Studies, vol. 10, p. 106, 1910 (orig. dese.). 
 Davidson, Jour. Econ. Ent., vol. 7, p. 127, 1914 (list). 
 
 Record. — Piyius ponderosa jeffreyi, Tallae, Eldorado County (Davidson). 
 
 Davidson's is the onlj^ report of this species in California. The 
 identification of his specimens was verified by Davis. One specimen 
 the author saw was quite small, being much smaller than the others 
 taken by Davidson. 
 
 51. Lachnus pseudotsugae Wilson 
 Figures 92, 98 
 Wilson, Can. Ent., vol. 44, pp. 159, 302, 1912 (orig. dese.). 
 Record. — Psetidotsuga taxifolia; Oregon, California (Wilson). 
 
A SYNOPSIS 01'' THE APHIDIDAE 49 
 
 Wilson wrote tlie author some time ago that he had taken this 
 species in California, although he gave no definite locality or collec- 
 tion record. The autlior lias had the opportunity to study cotype 
 specimens. 
 
 52. Lachnus sabinianus n.sp. 
 
 Eecord. — Pinus sahiniana, San Francisco (Compere). 
 
 In March, 1915, Harold Compere of the California State Insectary 
 found a small infestation of a species of Lachnus on Digger Pine in 
 the Golden Gate Park, San Francisco. Since this one collection, the 
 species has not again been observed. Being unable to identify the 
 species with any described in America, a description is herewith 
 appended, the species being named after its host plant, Pinus sabin- 
 i-ana. All the specimens, including the types are in the collections 
 of E. 0. Essig and of the University of California, Berkeley. The 
 specimens were all mounted in Canadian balsam before color notes 
 were taken, so those in tlie following description are only approxi- 
 mately correct. 
 
 Alate viviparous female. — Rich chestnut-amber to dark brown. 
 Antennal segments I and II, amber; III, yellowish with tips darker; 
 IV, V, and VI, dark yellow to dusky. Prothorax, chestnut-brown. 
 Thoracic lobes very dark brown to black. Beak, pale with tips dusky. 
 Cornicles, black. Cauda and anal plate with distal margins black. 
 Femora, chestnut-brown with base amber; tibiae, brown with amber 
 ring near the base ; tarsi, amber. Wing veins, grayffi stigma, dusky 
 gray. 
 
 Measurements: Body 4.2 mm. long and 1.7 mm. wide at thorax. 
 Antennae reach to base of abdomen, without secondary sensoria. I, 
 0.10 mm. ; II, 0.09 mm. ; III, 0.50 mm. ; IV, 0.25 mm. ; V, 0.19 mm. ; 
 VI, 0.08 mm. ; total, 1.21 mm. Beak reaches to the base of the cor- 
 nicles. Cornicles medium sized and of the usual Lachmts shape, 
 being 
 
 Apterous viviparous female. — Chestnut-brown in color with black 
 dorsal spots on abdomen. Antennal segments I and II, dark; III, 
 dusky yellow with tip dark; IV, V, and VI slightly darker. Beak 
 reaches to the base of tlie cornicles. Coxae, black ; femora, black with 
 basal one-fifth paler; tibiae, black with pale ring near base; tarsi, 
 black. Cornicles, black and conspicuous. They measure 5.2 mm. in 
 length and 3.3 mm. in width. 
 
50 MISCELLANEOVS STUDIES 
 
 53. Lachnus taxifolia Swain 
 
 Figures 99-103 
 
 Swain, Trans. Am. Ent. Soc, vol. 44, p. 11, 1918. 
 
 Secords. — Pxeudutsuga taxifolia, Sacramento (Essig), Berkeley and San Fran- 
 cisco (Shinji). 
 
 This is a fairly common species found in colonies on the branches 
 and trunks of Douglas fir in the San Francisco Bay and Sacramento 
 Valley. It is interesting pai-ticularly because of the atrophied cor- 
 nicles. 
 
 54. Lachnus tujafilinus (Del Guereio) 
 
 Figures 93, 105 
 
 Del Guereio, Kedia, vol. 5, p. 287, 1909. LachneiUa (orig. desc). 
 
 Essig, Pom. Jour. Ent., vol. 3, p. 541, 1911. Lachnuj< juniperi DeGeer 
 
 (dese.). 
 Essig, Pom. Jour. Ent., vol. 4, p. 773, 1912. Lachnus juniiieri DeGeer 
 
 (list). 
 Davidson, Jour. Econ. Eut, vol. 7, p. 127, 1914 (list). 
 
 Records. — Thuya occidentalis, Claremont, Santa Paula (Essig) ; Palo Alto, 
 Walnut Creek (Davidson); Stanford University, March, 1912 (Morrison); San 
 Diego, March, 1916; Riverside, October, 1916, March, 1917. 
 
 This oddly marked Lachnus is more or less common throughout 
 California wherever arborvitae is cultivated. The apterous females 
 are the most common, and are easily recognized by the odd markings 
 on the dorsum of the abdomen (see Essig 's illustrations). Occasion- 
 ally the alate females are found, Davidson finding some in April, 
 Morrison and the autlior in March. The author has observed the 
 larvae of CoccineUa calif ornica feeding on them in Riverside. 
 
 55. Lachnus vanduzei n.sp. 
 
 Figure 88 
 Records. — Picea sp., Berkeley, September, 1914 (Essig, E. P., Van Duzee). 
 
 In September, 1914, E. P. Van Duzee collected a few specimens 
 of a large Lachnus on a species of spruce in Strawberry Canyon, near 
 Berkelej\ Later in the same month Essig found specimens on the 
 same tree. The following fall the atithor hunted for the species, 
 
/I SrXOPSIS OF THE APBIDIDAE 51 
 
 but was imable to find any specimens, the tree on whieli it was first 
 found liaving been cut down. In the following description the color 
 notes are not absolutely accurate, as they were taken from material 
 mounted in balsam. This species is named after its first collector, 
 Mr. E. P. Van Duzee, of the University of California. Type speci- 
 mens are in the eolUetion of tiie University of California. 
 
 Alatc viviparous female. — The alate viviparous females are of a dark 
 muddy color, as near as can be judged from the mounted specimens. 
 The antennae are : 1 and II, dusky ; III and IV, pale with apical half 
 dusky; V, pale witli the apex or apical third duskj^; VI, pale with the 
 apex and sjiur dusky. The measurements of the segments are : I, 
 0.09 mm.; II, 0.07 mm.; Ill, 0.5 mm.; IV, 0.26 mm.; V, 0.27 mm.; 
 VI, 0.16 mm. The sensoria are located as follows: III, 10-12; IV, 
 2-3; V, 2-3; VI, 1. They are large and circular, and quite evenly 
 distributed in a line on each .segment. The beak reaches to the base 
 of the Cauda. The coxae are black, the femora amber on the basal 
 half and black on the apical, the tibiae are black with an amber ring 
 near the base, the tarsi are black. The first joint of the hind tarsus 
 is not one-third the length of the second, the first measuring 0.08 mm., 
 and the second 0.26 mm. The wings are quite large, with a very 
 distinct stigma. The costal vein is grayish-brown, the subcostal 
 brown. The stigma is long and brown, the stigmal vein being pale 
 brown and slightly curved throughout its entire length. Tlie first 
 and second discoidals are distinct and pale brown, the second dis- 
 coidal being slightlj' curved near the tip. The third discoidal is indis- 
 tinct and twice-branched, the angles of the branches being very acute. 
 Apterous viviparovs female. — Prevailing color, amber-brown, with 
 the abdomen mottled gray, brown, and black. The head is brown 
 with anterior margin amber. The antennae are colored as follows : 
 I, amber; II, amber; III, amber with tip dusky; IV, amber with tip 
 dusky; V, amber with apical two-thirds dusky; VI, dusky. The beak 
 reaches to the base of the cauda. The femora are brown with the 
 bases amber, the tibiae and tarsi brown. The fii*st joint of the hind 
 tarsus is scarcely more than one-third the length of the second. In 
 four tarsi measured, the relative lengths of the joints were : 0.07 to 
 0.23 mm. ; 0.08 to 0.23 mm. ; 0.08 to 0.28 mm. ; and 0.07 to 0.25 mm. 
 The cornicles are conspicuous and dark, the cauda well rounded and 
 dark on its posterior edge. The lengths of the antennal segments are : 
 I, 0.1 mm.; II, 0.1 mm.; Ill, 0.56 to 0.57 mm.; IV, 0.21 to 0.23 mm.; 
 V, 0.22 to 0.28 mm. ; VI, 0.15 to 0.16 mm. 
 
52 MISCELLANEOUS STUDIES 
 
 Group Aphidina Wilson 
 
 Wilson, Ann. Ent. Soe. Am., vol. 3, p. 314, 1910. 
 
 This group as considered by Wilson consists of three tribes : 
 Trichosiphini, Macrosiphini, and Aphidini. The first of these con- 
 tains two genera found onlj^ in the Asiatic islands, so it will not be 
 considered in this paper. This group contains quite closely related 
 genera, and in many cases it is quite hard to distinguish between 
 them. Following is a brief extract from Wilson's paper (cited above) : 
 
 In studying closely related genera the development of the external characters 
 may be placed in five divisions: (1) the antennae and spur; (2) the antennal 
 tubercles; (3) the development of the nectaries [cornicles]; (4) the development 
 of the Cauda; (5) the development of the wing venation. In a gi-oup of insects 
 as pliable as the present one, any one or two of these characters may be either 
 under- or over-developed and it is necessary to place the genera according to the 
 greatest development. Of all the characters which show this variation tlie wings 
 show what may be true of all these characters. 
 
 The two tribes have been separated from one another on the character 
 of the antennal tubercles, as Wilson says in the same paper : 
 
 The division is made between species with distinct antennal tubercles and 
 those having none or at the most indistinct tubercles. However, should a certain 
 species have distinct antennal tubercles with the other characters [of the Macro- 
 siphini] wanting, then it would have to go into the next tribe [Aphidini']. 
 
 The keys to the tribes and genera below have been formulated by the 
 author, following, however, those of Wilson, Van der Goot, and 
 Mordwilko. 
 
 1. Antennal tubercles well formed. Antennae usually as long as or longer than 
 the body. Apterae often with sensoria on the third antennal segment. 
 Body never with lateral tubercles on the seventh abdominal segment. Cor- 
 nicles variable but usually about one-fourth the length of the body or 
 
 longer Tribe Macrosiphini 
 
 — Antennal tubercles absent or more or less indistinct. Antennae seldom longer 
 than the body. Apterae seldom with sensoria on the third antennal seg- 
 ment. Body with lateral tubercles on at least the seventh abdominal seg- 
 ment Tribe Aphidini 
 
 Tribe Macrosiphini Wilson 
 
 Wilson, Ann. Ent. Soe. Am., vol. 3, p. 314, 1910. 
 To a large extent the author has followed Wilson in the placing 
 of the genera, but in a few eases he has not. This is noticeable in 
 To.Toptera, which is considered by Wilson as belonging to this tribe, 
 
A SYNOPSIS OF THE APHIDIDAE 53 
 
 while the author feels that it is better associated with the Aphidini, 
 inasmuch as the anteuual tubercles are very small and more or less 
 indistinct and as the antennae are scarcely as long as the body. Van 
 der Goot's genus, Myzaphis, has been accepted for the two species, 
 Myzus rosarum (Walker) and Aphis ahietina Walker, and is included 
 with the Aphidini. The species Aphis nymphaeae Linn., which Wil- 
 son uses as the type of Rhopalosiphinn, has been taken from this genus 
 and placed in Siphocorync, chiefly because of the apparent absence 
 of antenna! tubercles and of the presence of distinct tubercles on the 
 seventh abdominal segment. Therefore Aphis persioae Sulzer takes 
 the place as type of the genus Rhopalosiphum. 
 
 Key to California Genera 
 
 1. Cornicles cylindrical, or at most but very slightly swollen on one side (figs. 
 
 122, 152) 4 
 
 — Cornicles distinctly swollen toward apex, or clavate (figs. 109, 113, 119) 2 
 
 2. Antemial tubercles very large and tapering but not gibbous on the inner side ; 
 
 the bases of the antennae being more or less approximate (fig. 107). 
 
 Nectarosiphon Schouteden 
 
 — Antennal tubercles distinct, but not large and tapering as above, being more 
 
 or less toothed or gibbous on the inner side ; the bases of the antennae not 
 approximate (figs. 108, 111) 3 
 
 3. Antennal tubercles short and wedge-shaped, the outer side not evident (fig. 
 
 108). Cauda ensiform and of medium size. Antennae at most but slightly 
 longer than the body , Rhopalosiphum Koch 
 
 — Antennal tubercles short, but not wedge-shaped (fig. 111). Antennae con- 
 
 siderable longer than the body. Cauda very large and long. 
 
 AmphoTophora Buckton 
 
 4. Antennal tubercles large and as long on the outer as on the inner side (fig. 
 
 106) 5 
 
 — Antennal tubercles with outer side shorter than inner, or not evident (figs. 112, 
 
 115, 116) 7 
 
 5. Cornicles tapering, longer than cauda which is ensiform (fig. 152). Wing 
 
 venation regular, with third discoidal twice-branched. 
 
 Macrosiphum Passerini 
 
 — Cornicles and cauda variable. Wing venation irregular and very striking with 
 
 veins either wanting or combined, and shaded 6 
 
 6. Antennal tubercles with short upper inner angle. Cauda shorter than cornicles 
 
 and tapering. Stigmal and third discoidal veins meet in a broad dark 
 band, giving the wing the appearance of having a closed triangular cell 
 (fig. 110) Idlopterus Davis 
 
 — Antennal tubercles with small rounded tubercle at the upper inner angle. 
 
 Cornicles slightly constricted in the middle and at the tip. Wing venation 
 variable, but usually the stigmal and third discoidal veins are partly 
 joined and form a distinct, closed, four-sided cell Pentalonia Coquerel 
 
 7. Antennal tubercles and first antennal segment with a strong tooth on the 
 
 inner side of each (figs. 115, 116). Cauda short and tapering (fig. 118). 
 Cornicles cylindrical and tapering slightly with tip outcurved (fig. 117). 
 
 BhoTodon Passerini 
 
54 MISCELLANEOUS STUDIES 
 
 — Autennal tubercles with a distinct but not prominent blunt projection forming 
 the inner angle (tig. 112), but the prominent teeth as above are lacking. 
 Cauda short, tapering, and usually triangular (fig. 121). Cornicles as 
 above, being cylindrical, with a slight tapering from base to apex, and 
 often slightly outcuryed at tip (fig. 122) Myzus Passerini 
 
 21. Genus Amphorophora Buckton 
 
 Buckton, Monog. Brit. Aphides, 187C. Type A. ampuUata n.sp. 
 Key to Califobnian Species^ 
 
 Cornicles pale, or at most slightly dusky, swollen and vasiform (fig. 113). VI 
 spur longer than III, the latter with 35-45 sensoria rubi (Kalt.) 
 
 Cornicles black, greatly dilated in apical one-half (fig. 161). VI spur shorter 
 than III, latter with 13-17 sensoria latysiphon Davidson 
 
 56. Amphorophora latysiphon Davidson 
 
 Figure 161 
 
 Davidson, Jour. Econ. Ent., vol. 5, p. 408, 1912 (orig. desc). 
 
 Becords. — Vinca major, San Jose (Davidson) ; Courtland, Contra Costa County 
 (Davidson); Stanford University, 1912 (Morrison, Essig). Convolvulus arvensis, 
 San Jose (Davidson). Solanum tuherosum, Walnut Creek, Contra Costa County, 
 igi.") (Davidson). 
 
 This species has been found sparingly in the San Francisco Bay 
 region on periwinkle, morning-glory, and potato tubers, although it 
 has never seemed to be common. The author has not collected it, his 
 only specimens being some taken by Essig on periwinkle near Stan- 
 ford University. The odd shape of the cornicles is a distinguishing 
 character. 
 
 .57, Amphorophora rubi Kalt. 
 
 Figures 111, 113, 162 
 
 Kaltenbach, Monog. d. Pflanzenliiuse, p. 23, 1843. Aphis (orig. desc). 
 Davidson, Jour. Econ. Ent., vol. 5, p. 411, 1912 (list). 
 Shinji, Can. Ent., vol. 49, p. 52, 1917 (list). 
 
 Becords. — Bubics pan-ifiorus; San Jose (Davidson) : Buhus spp., Walnut Creek, 
 1915 (Davidson) ; Berkeley (Shinji). 
 
 This species has been taken a few times on thimble-berry in the 
 San Francisco Bay region. Davidson writes that he has also found 
 it on blaekberrv and loganberry in the vicinity of Walnut Creek, 
 
 9G. 0. Shinji (Can. Ent., vol. 49, p. 51, 1917) described an aphid from CicuUa 
 virosa var. calif ornica in Berkeley, which he called Amphorophora cicutae n.sp. 
 The author has never seen specimens, so does not feel that he can recognize this 
 as a good species. Of some half dozen new (?) species described by Shinji the 
 author has found none, on examining specimens, that are good species, hence he 
 cannot recognize this one at present. 
 
A SYNOPSIS OF THE APEIDIDAE 55 
 
 Contra Costa County. The author has recently received specimens 
 from Gillette of an alate viviparous female and apterous oviparous 
 females taken in the vicinity of Fort Collins, Colorado. Inasmuch as 
 the descriptions of this species are inadequate and not readily acces- 
 sible it has been thouglit best to give here brief descriptions of the 
 different forms. As no color notes were received with the specimens 
 they must necessarilj' be omitted. 
 
 Alate viviparous female (from Fort Collins, Colorado). — Antennae 
 half as long again as the body, dusky, and placed on small but distinct 
 tubercles. From the mounted material it appears as if III were 
 dusky, IV, pale with extreme tip dusky; V, pale with apical one-third 
 dusky; and VI dusky. VI spur is the longest segment, followed by 
 III, IV, V, VI base, I, and II. The usual primarj' and accessory 
 sensoria are present on VI base, and the primary sensorium on V. 
 Secondary sensoria are present only on III. Thfese are small, circular, 
 irregular-sized, and irregularly placed along the whole length of the 
 segment. The number (35 to 40) is such as to make the segment 
 appear tuberculate. The beak is quite large and long, reaching to or 
 slightl.y bej'ond the third coxae. The thorax is dusky. The wings 
 fairly large, and normal. The second branch of the third discoidal 
 vein arises nearer to the base of the first branch than to the apex of 
 the wing. Normally the measurements are as follows : From the base 
 of the second branch of the third discoidal to the tip of the wing is 
 about 0.8 mm., from the base of the first branch to the base of the 
 second 0.4 mm., from the apex of the first branch to the apex of the 
 second 0.29 mm. In one case the base of the second branch was 1.02 
 mm. from the apex of the wing, and but 0.0-34 mm. from the base of 
 the second, while the apices of the two branches were but 0.187 mm. 
 apart. The legs are long, femora pale with apical one-fourth dusky, 
 tibiae and tarsi du.sky. The abdomen is pale with some slight dorsal 
 dark markings, these being indistinct in the mounted specimens. The 
 cornicles are fairly long, clavate on the apical one-half or two-thirds, 
 dusky throughout, and with the extreme tip reticulated. In length 
 they are somewhat shorter than III, but longer than IV. The cauda 
 is pale, short, and triangular, being about equal in length to the 
 hind tarsi. 
 
 Measurements : body length, 1.785 mm. ; antennae total, 2.788 mm. ; 
 III, 0.68 mm. ; IV, 0.51 mm. ; V, 0.408 to 0.425 mm. ; VI, base, 0.12 
 mm. ; VI, spur, 0.867 to 0.884 mm. ; cornicles, 0.578 to 0.646 mm. ; 
 Cauda, 0.102 mm.; hind tarsi, 0.102 mm.; wing length, 3.128 mm.; 
 width, 1.292 mm. ; expansion, 6.8 mm. 
 
56 MISCELLANEOUS STUDIES 
 
 Apterous ovipurous female (Fort Collins, Colorado). — Pale 
 throughout, with many small hairs scattered over the body. Most 
 of these haii-s are simple, but some especially on the front of the head 
 and on the bases of the antennae, are capitate. Antennae slightly 
 longer than the body, pale, with VI and the apices of the other seg- 
 ments dusky. VI spur and III are subequal or either one may be 
 slightly longer than the other. These are followed by IV, V, VI base, 
 I, and II. The usual primary and accessory sen.soria are present on 
 VI base, and the primarj- sensorium of V. Secondary sensoria are 
 present onlj- on III, and number about nine or ten. These are small, 
 circular, but varying in size, and are arranged in a more or less 
 even line along the basal one-half to two-thirds of the segment. Beak 
 pale, with tip duskj', quite large and long, reaching to or beyond the 
 third coxae. Thorax and legs normal, except the hind tibiae which 
 are quite long, and' furnished with a large number of sensoria. 
 These sensoria cover practically the whole joint. Cornicle very long 
 and large, curved outward, pale, with apex dusky, and with distinct 
 reticulations at the extreme tip. They are markedly larger than in 
 the alate viviparous females, being considerably longer than the third 
 antennal segment, and in some eases even half as long again. The 
 Cauda is small, pale, and triangular, although somewhat larger in the 
 viviparous female. 
 
 Measurements : body length, 2.04 mm. ; width of thorax, 0.595 mm. ; 
 antennae total, 2.446 mm. ; III, 0.646 to 0.697 mm. ; IV, 0.442 to 0.459 
 mm. ; V, 0.356 to 0.374 mm. ; VI, base, 0.136 mm. ; VI, spur, 0.663 mm. ; 
 cornicles, 0.918 to 0.952 mm. ; cauda, 0.187 mm. ; liiud tarsi, 0.136 mm. 
 
 22. Genus Idiopterus Davis 
 
 DavLs, Aun. Ent. Soe. Am., vol. 2, p. 198, 1909. Type, 7. neprelepidis n.sp. 
 
 58. Idiopterus nephrelepidis Davis 
 
 Figure 110 
 
 Davis, Ann. Ent. Soc. Am., vol. 2, p. 198, 1909 (orig. tlesc). 
 Davidson, Jour. Econ. Ent., vol. 3, p. 376, 1910 (list). 
 Essig, Pom. Jour. Ent, vol. 3, p. 538, 1911 (list). 
 
 Records. — Nephrolepis exaltata, Santa Paula (Essig), Palo Alto, April, 1915, 
 San Diego, March to May, 1916; Riverside, February, 1917: Cyrtonium fulcotum, 
 Berkeley, March, 1915 (Essig); ferns (unidentified species of house ferns), Stan- 
 ford University (Davidson, Morrison); Viola sp., Claremont (Essig). 
 
 This small black aphid is often found in houses and nurseries, and 
 occasionally out of doors, on the fronds of various kinds of house 
 
A SYNOPSIS OF THE APHIDIDAE 57 
 
 ferns, particularly the Boston fern. Essig has also found it on violets 
 in the vicinity of Pomona College. The alate females have the wings 
 beautifully marked with black and white. 
 
 23. Genus Macrosiphum Passerini 
 Passerini, Gli Afiili, 1860. Type Aphis rosac Linn. 
 Key to California Species 
 Alate viviparous females 
 
 1. Cornicles slightly clavate on one side, somewhat as in Ehopalosiphum. 
 
 tulipae (Monell) 
 
 — Cornicles not clavate 2 
 
 2. Ill as long as V and VI (base and spur) sonchella (Monell) 
 
 — Ill not as long comparatively 3 
 
 3. Ill pale, IV, V, and VI dusky jasminum (Clarke) 
 
 — Not so ; if IV, V, and VI are dusky then III is also, except perhaps the base ; 
 
 or if III is pale throughout then at least the greater part of IV and V are 
 also pale 4 
 
 4. VI (base and spur) shorter than III, but V and VI together are longer than 
 
 III baccharadis (Clarke) 
 
 — VI (base and spur) not shorter than III 5 
 
 5. Secondary sensoria on III, IV, and V. Cornicles not reticulated. 
 
 heucherae (Oestlund) 
 
 — No secondary sensoria on V 6 
 
 6. Secondary sensoria on both III and IV. Cornicles with tips at least reticulated 
 
 (fig. 152) 7 
 
 — No secondary sensoria on IV 10 
 
 7. Cornicles and cauda subequal in length, the former being more or less bottle- 
 
 shaped sanbomi Gillette 
 
 — Cornicles longer than cauda 8 
 
 8. Cauda light green. Secondary sensoria only occasionally present on IV and 
 
 then very small and indistinct rosae (Linn.) 
 
 • — Cauda dark (brown or black). Seven or more distinct secondary sensoria 
 on IV 9 
 
 9. Body with capitate setae, especially on head and antennae. 
 
 artemisiae (Fonsc.) 
 
 — Body without capitate setae. Abdomen with dark dorsal markings. 
 
 lactucae (Kalt.) 
 
 10. Cornicles with at least tips reticulated (fig. 132) 14 
 
 — Cornicles with no reticulations (fig. 156) 11 
 
 11. Body with fan-shaped setae artemlslcola (Williams) 
 
 — Body without fan-shaped setae 12 
 
 12. Distal two-thirds of cornicles black orthocarpus (Dvdn.) 
 
 — Only tip of cornicles black 13 
 
 13. Cornicles long and slender. About 18 secondary sensoria in a row on III 
 
 (fig. 130) pisi (Kalt.) 
 
 — Cornicles shorter and heavier. About 25 to 30 sensoria scattered irregularly 
 
 along III (fig. 157) dirhodum (Walker) 
 
 14. Cornicles with more than apical one-half reticiJate (fig. 149). 
 
 ludovlcianae (Oestlund) 
 
 — Cornicles with less than apical one-half reticulated (fig. 128) 15 
 
58 MISCELLANEOUS STUDIES 
 
 15. Cornicles dusky for practically their entire length 20 
 
 — Cornicles with less than apical one-half dusky 16 
 
 16. Cornicles considerably longer than III, with apical portion curved outward. 
 
 About a dozen, meilium-sized sensoria in a straight line along basal two- 
 thirds of III (fig. 131) caUfomicum (Clarke) 
 
 — Cornicles not considerably longer than III 17 
 
 17. Cornicles and VI spur subequal, the former fairly long, slightly curved 
 
 outward and slightly swollen before the tip (fig. 128) stanleyl Wilson 
 
 — Cornicles considerably shorter than VI spur, and not swollen before the tip 18 
 
 18. Secondary sensoria in a fairly straight line on III. Body not pulverulent 19 
 - — Body covered with a slight pulveruleuce. Ill with about 30 fairly large- 
 sized sensoria, more or less scattered along the entire length (fig. 143). 
 
 albifrons Essig 
 
 19. Cornicles about half the length of VI spur and considerably shorter than III, 
 
 the latter with about 20 to 30 secondary sensoria pteridis Wilson 
 
 — - Cornicles about two-thirds as long as VI spur and slightly shorter than III, 
 the latter with about 15 sensoria (fig. 133) cucurbitae (Thomas) 
 
 20. Ground or basal color of abdomen green 21 
 
 — • Ground or basal color of abdomen red, brown, or black 2-!: 
 
 21. Cornicles green, sometimes dusky at apex solanifoUi (Ashmead) 
 
 — Cornicles black 22 
 
 22. Ill with a small number (9-15) of secondary sensoria on basal one-half 
 
 (fig. 135); longer than VI spur granarium (Kirby) 
 
 — Ill with some 30 or more sensoria scattered along its entire length (figs. 151, 
 
 159) ; subequal to or shorter than VI spur 23 
 
 23. Cornicles and III subequal. Tibiae with apices only dusky rosae (Linn.) 
 
 — Cornicles longer than III. Tibiae dusky throughout. 
 
 rudbeckiae (Pitch) n.var. madia 
 
 24. Cauda pale 25 
 
 — Cauda dusky 27 
 
 25. Ill and VI spur subequal rosae (Linn.) 
 
 — • III shorter than VI spur 26 
 
 26. Cauda about one-half as long as cornicles, the latter shorter than IV. 
 
 chrysanthemi (Oestlund) 
 
 — Cauda slightly more than one-half as long as cornicles, the latter equal to or 
 
 longer than IV rudbeckiae (Fitch) 
 
 27. Ill and VI spur subequal 28 
 
 — Ill longer than VI spur taraxici (Kalt.) 
 
 28. Body yellowish-brown in color; legs same except tarsi and tips of tibiae and 
 
 femora which are dusky to black Valerianae (Clarke) 
 
 — ■ Body dark reddish-brown to black in color; legs dusky throughout. 
 
 ambrosiae (Tliomas) 
 
 Apterous viviparotis /fnioJesio 
 1. Cornicles clavate on one side, somewhat as in Bltopalosiphnm. 
 
 tulipae (Monell) 
 — • Cornicles not so, being cylindrical or subcylindrical 2 
 
 10 Only the species of which there are specimens available to the author, or of 
 which there are adequate descriptions, are included in this key. The species rep- 
 resented in the author's collection are marked with an asterisk (*). The author 
 recognizes the great difficulty in separating the apterae of various species, par- 
 ticularly in this genus, and offers this key merely as a slight aid toward the recog- 
 nition of the better known species. 
 
A SYNOPSIS OF THE APHIDIDAE 59 
 
 2. Ill without or at moet with only a few secondary sensoria (0-12) 11 
 
 — Ill with several (over 12) secondary sensoria scattered along the greater part 
 
 of its length 3 
 
 3. Cornicles short and tapering, being somewhat bottle-shaped and not distinctly 
 
 longer than the Cauda sanboml Gillette* 
 
 — Cornicles normal, being cylindrical and considerably longer than tlie Cauda 4 
 
 4. Ill and IV with secondary sensoria heucherae (Oestlund) 
 
 — IV without secondary sensoria 5 
 
 5. General body color dark, being red, wine, brown or black 6 
 
 — General color lighter, usually being a shade of green 8 
 
 6. Cauda black. Legs black, except the bases of the femora taraxlci (Kalt.) 
 
 — Cauda pale. Legs with at least the bases of the femora and tibiae not black 7 
 
 7. Legs green, except tarsi and apices of femora and tibiae. Cauda not more than 
 
 half the length of the cornicles. Not more than ten to twelve sensoria on 
 
 the basal one-third of III rosae (Linn.)* 
 
 — • Legs black, except bases of femora and tibiae, which are light brown. Cauda 
 more than half the length of the cornicles. A considerable number of 
 sensoria scattered over more than the basal one-half of III. 
 
 nidbeckiae (Fitch)* 
 
 8. Cornicles subequal to or shorter than III. Body covered with a whitish pul- 
 
 verulence 9 
 
 — Cornicles distinctly longer than III. Body without whitish pulverulence .... 10 
 
 9. Cornicles, except tip, and Cauda green; the former subequal in length to III 
 
 and about twice as long as cauda albifrons Essig* 
 
 — Cornicles black, cauda yellow or light brown; the former considerably shorter 
 
 than III and not twice as long as cauda ludovicianae (Oestlund)* 
 
 10. Cauda quite broad and blunt at end. Cornicles with not more than apical one- 
 
 sixth reticulated rosae (Linn.) * 
 
 — • Cauda slender and pointed. Cornicles with apical one-fourth reticulated. 
 
 rudbeckiae (Fitch) n.var. madia* 
 
 11. Body covered with capitate or fan-shaped setae 12 
 
 — Body without specialized setae 14 
 
 12. Setae with fan-shaped tips and thickly covering the body. Cornicles slender 
 
 and imbricated for their entire length arte'misicola (Williams)* 
 
 — Setae capitate and only sparsely covering body 13 
 
 13. Cornicles fairly stout, with tips reticulated, and about twice as long as cauda. 
 
 artemisiae (Fonsc.) 
 
 — Cornicles slender, with no reticulations, and considerably more than twice the 
 
 length of the cauda pteridis Wilson 
 
 14. Cornicles with tips at least reticulated 16 
 
 — Cornicles with no reticulations 15 
 
 15. Cornicles very long and slender. Antennae considerably longer than body. 
 
 pisi (Kalt.)* 
 
 — Cornicles shorter and heavier. Antennae at most but slightly longer than 
 
 body dirhodum (Walker)* 
 
 16. Cornicles for the most part dusky or black 17 
 
 — Cornicles mostly pale or green 19 
 
 17. Cornicles and III subequal. Body not pulverulent 18 
 
 — Cornicles considerably shorter than III. Body more or less pulverulent. 
 
 ludovicianae (Oestlund)* 
 
60 MISCELLANEOUS STUDIES 
 
 18. Ill with but two or four sensoria near base; longer than VI spur. 
 
 granarium (Kirby)* 
 
 — Ill with six or so sensoria on basal one-half; shorter than or equal to VI 
 
 spur rosae (Linn.)* 
 
 19. Cornicles longer tlian III 20 
 
 — Cornicles at most subequal to III 21 
 
 20. Antennae pale, except VI and the apices of III to V. Cornicles slightly 
 
 swollen near distal end Stanley! Wilson* 
 
 — • Antennae dusky, except III, basal part of IV, and perhaps the extreme base 
 of V. Cornicles long, slender, and out-curved caUfornicum (Clarke)* 
 
 21. Cauda broad, and blunt, with the sides almost parallel and about half as long 
 
 as cornicles lactucae (Kalt.) * 
 
 — Cauda slender-pointed, and more than half as long as cauda 22 
 
 22. VI spur and III subequal solanifoUi (Ashmead)* 
 
 — VI spur considerably longer than III ., cucurbitae (Thomas)* 
 
 59. Macrosiphum albifrons Essi<j 
 
 Figures 143, 144 
 
 Davidson, Jour. Econ. Ent., vol. 2, p. 304, 1909. Macrosiphum sp. (list). 
 Essig, Pom. Jour. Ent., vol. 3, p. 543, 1911 (orig. desc). 
 
 Mecords. — Lupinus sp., Santa Paula (Essig) ; Stanford University (Davidson) ; 
 Jasper Eidge, Coast Range Mountains, Santa Clara County, April, 1912 (V. G. 
 Stevens) ; Berkeley, April, 1915 (Geo. Shinji) ; Mount Hood, Oregon, August, 
 1916 (E. A. McGregor). 
 
 This large, flocculent aphid is found occasionally infesting various 
 lupines tliroughout the Pacific Coast, from southern California north, 
 well into Oregon. The author has specimens from Berkeley and 
 Oregon, although he has never collected it himself. 
 
 60. Macrosiphum ambrosiae (Thomas)? 
 
 Thomas, 111. Lab. Nat. Hist., Bull. 2, p. 4, 1878. Siphonophora (orig. 
 
 desc.). 
 Sanborn, Kans. Univ. Sci., Bull. 3, p. 74, 1904 (desc). 
 
 Records. — Eelianthus annum; Orange (T. D. A. Cockerell) ; San Diego, April, 
 1916. 
 
 In 1915 the author received a few specimens of this species from 
 T. D. A. Coclserell from Orange, and in 1916 he collected it once on 
 sunflower in Exposition Park, San Diego. At first it was thought to 
 be M. sonchi (Linn.), and was so reported by Cockerell. Since then 
 it was identified by J. J. Davis as probably M. anibrosiae (Thomas). 
 
A STNOPSIS OF THE APEIDIDAE 61 
 
 61. Macrosiphum artemisiae (Fonsc.) 
 Figures 142, 145 
 
 Boyer de Fonseolnibe, Ann. Ent. Soc. France, vol. 10, p. 162, 1841. Aphis 
 
 (orig. dcsc). 
 Essig, Pom. Jour. Ent., vol. 3, p. 54(i, 1911. Macrosiphum frigidae (Oest.) 
 
 (desc). 
 Davidson, Jour. Econ. Ent., vol. 3, p. 133, 1914. Macrosiphum frigidae 
 
 (Oest.) (list). 
 Wilson, Trans. Amer, Ent. Soc, vol. 41, p. 97, 1915 (desc). 
 
 Records. — Artemisi-a californica; Santa Paula (Essig) ; Walnut Creek, Contra 
 Costa County (Davidson). 
 
 Occasioiuilly tliis species is foiiiul infesting the tender shoots of 
 the common California sage brusli. It is characterized by the presence 
 of capitate hairs scattered sparsely over the body, particularly of the 
 apterous female. The .s.ynonomy above is after Wilson, who lists 
 M. frigidae (Oestlund) as a synonym of artemisiae (Fonsc). 
 
 62. Macrosiphum artemisicola (Williams) 
 
 Figures 146, 147 
 
 Williams, Univ. Neb. Studies, vol. 10, p. 73, 1910. Siphonophora (orig. 
 desc. ) . 
 • Wilson, Trans. Am. Ent. Soc, vol. 41, p. 96, 1915 (desc). 
 
 Records. — Artemisia tridentata, A. vulgaris; Oregon (California) (Wilson). 
 
 Although there is no published record of the presence of this 
 species in California it is included here on Wilson's authority. He 
 stated to the author that he had found it in California, although he 
 failed to give any date or locality record. This is characterized bj' 
 the fan-shaped setae which thickly cover the body of the apterae, and 
 which are present on the ventral side of the abdomen of the alates. 
 The author has specimens taken by R. W. Haegele in the sum.mer of 
 1915 on Arlemi.sia sp. near Canton, Montana. 
 
 63. Macrosiphum baccharadis (Clarke) 
 Clarke, Can. Ent., vol. 35, p. 254, 1903. Neetarophora (orig. desc). 
 Record. — Baccharis sp., Berkeley (Clarke). 
 
 This species is one of those desci'ibed by Clarke, but since then 
 unknown. It is possible that it is M. ritdbeckiae (Fitch), which is so 
 common on Baccharis throughout California. 
 
62 MISCELLANEOUS STUDIES 
 
 64. Macrosiphum calif ornicum (Clarke) 
 
 Figures 131, 132 
 
 Clarke, Can. Ent., vol. 35, p. 2.54, 1903. Nectarophora (orig. dese. apterae). 
 Davidson, Jour. Econ. Ent., vol. 2, p. 304, 1909 (list). 
 Daridson, Jour. Eeon. Ent., vol. 3, p. 380, 1910 (list). 
 Davidson, Pom. Jour. Ent., vol. 3, p. 398, 1911 (list). 
 
 Essig, Pom. Jour. Ent., vol. 3, p. 548, 1911. M. laevigatae, n.sp. (orig. 
 desc). 
 
 Records. — Salix sp. ; Newcastle, Placer County (Clarke); Stanford University 
 and Penryn, Placer County (Davidson) ; Stanford University, November, 1914 
 (Morrison), May, 1915; Berkeley, April, 1915 (Shinji); August, 1915, Salix 
 laevigata; Santa Paula (Essig) ; Riverside, May, 1917. 
 
 Clarke described the apterous females of a species of Nectarophora 
 (Macrosiphum) from specimens taken on willow in Placer Connty. 
 Becanse of the extremely long cornicles it is possible to identify this 
 with specimens taken since throughout the San Francisco Bay region 
 on various species of willows. Essig 's M. laevigatae from Santa 
 Paula is the same species, having been compared by the author with 
 specimens from Stanford University and Berkeley. Morrison has 
 taken the males and oviparous females of this species in the vicinity 
 of Stanford University in November, 1914. The author has reared 
 specimens of Aphidius polygonaphis Fitch, and Prnon sinudans Prov. 
 from tills species taken in Berkeley. 
 
 65. Macrosiphum chrysanthemi (Oest.) 
 
 Oestlund, 14th Eep. Geol. Surv. Minn., vol. 22, 1886. Siphonophora (orig. 
 
 desc). 
 
 Davidson, Jour. Econ. Ent., vol. 5, p. 411, 1912 (list). 
 
 Record. — Undetermined species of Compositae ; Courtland (Davidson). 
 
 This is a doubtful species taken by Davidson at one time from an 
 undetermined compo.site near Courtland. The author is entirely 
 unacquainted with the species. 
 
 66. Macrosiphum cucurbitae (Tlios.) 
 
 Figures 133, 134 
 
 Thomas, 8th Ann. Rep. Illinois St. Ent., p. 66, 1879. Sip}ionophora (orig. 
 desc. ) . 
 
 Record. — Cucuriita sp., Hayward, Alameda County, July, 1915 (Roy E. Camp- 
 bell) ; Los Angeles, May, 1917. 
 
A SYNOPSIS OF THE APHIBIDAE 63 
 
 In July, 1915, Roy E. Campbell of the Bureau of Entomology, 
 sent the author specimens of a Macrosiphum sp. from squash in Hay- 
 ward. In 1917 the autlior found the same species abundantly on 
 squash in Los Angeles. These the author identified as being specimens 
 of 31. cucurbifae (Thomas). Later J. J. Davis verified the deter- 
 mination. This is a new record for California. As the available 
 descriptions of this species are quite inadequate, tlie author gives 
 herewith a few descriptive notes taken from these specimens. 
 
 Alate viviparous feviale. — Antennae longer than the body, placed 
 on distinct frontal tubercles, dusky except I, II, and extreme base 
 of III. The spur of VI is the longest segment, followed by III, which 
 is about four-fiftlis as long. IV and V are subequal, and almost as 
 long as III. Tlie usual primary and accessory sensoria are present 
 on V and VI. Secondary sensoria are present on III (fig. 133), being 
 small, circular, numbering about 14 to 15, and arranged in a fairly 
 even row along the whole length of the segment. Beak pale with 
 dusky tip, reaching to tlie second coxae. Thorax and abdomen green, 
 the thoracic lobes not conspicuously darkened. Cornicles (fig. 134) 
 green with apical one-third dusky, equal to or slightly longer than 
 III, imbricated with tip reticulated. Cauda large, pale, vasiform, 
 slightly more than half the length of the cornicles, reaching to their 
 apices. Wings and legs normal. 
 
 Measurements : Body length, 2.3 mm. ; antennae total, 3.25 to 3.35 
 mm. ; III, 0.685 to 0.714 mm. ; IV, 0.629 to 0.646 mm. ; V, 0.603 to 
 0.612 mm. ; VI, base, 0.136 to 0.153 mm. ; VI, spur, 0.935 to 0.696 mm. ; 
 cornicles, 0.714 to 0.731 mm. ; cauda, 0.408 mm. 
 
 67. Macrosiphum dirhodum ("Walker) 
 
 Figures 156, 157 
 
 Walker, Ann. Nat. Hist., (2), vol. 3, p. 43, 1848. Aphis (orig. desc). 
 Theobald, Jour. Econ. Biol., vol. 8, p. 128, 1913 (desc.). 
 Patch, Maine Agr. Exp. Sta., Bull. 233, p. 268, 1914 (note). 
 Gillette, Jour. Econ. Ent., vol. 8, p. 103, 1915 (note). 
 
 Record. — Rose, Santa Ysabel (3000 feet altitude), San Diego County, May, 
 1916; Riverside, April, 1917. 
 
 The author found this species sparingly on rose near Santa Ysabel, 
 San Diego County, in May, 1916, and again in April, 1917, in River- 
 side. According to Gillette, this species passes the winter on rose, 
 and the summer on various grains and grasses, as M. rosac (Linn.) 
 
64 MISCELLANEOUS STUDIES 
 
 may do. These are the only records of it in California. The author 
 has compared it with specimens taken by R. W. Doane in 1915 on 
 grain in Utah. 
 
 68. Macrosiphum granarium (Kirby) 
 
 Figures 135, 148 
 
 Kirby, Linn. Soc. Trans., vol. 4, p. 238, Aphis (orig. desc). 
 Davidson, Jour. Econ. Ent., vol. 5, p. 411, 1912 (list). 
 Theobald, Jour. Econ. Biol., vol. 8, p. 58, 1913 (desc). 
 Davidson, Mon. Bull., Cal. Comm. Hort., vol. 6, p. 65, 1917 (note). 
 
 Mecords. — Graminaceae (various species) ; San Jose (Davidson) ; Stanford 
 University, January to May, 1915; Berkeley, March, 1915: Typha latifolia 
 (Davidson). 
 
 This is a more or less common species of Macrosiphum on various 
 grains and grasses in the San Francisco Bay region during the winter 
 and spring. In late spring and earlj' summer, as the grasses begin 
 to dry out, it leaves them for the cat-tail rush or California tule 
 (Davidson, 1917). In the late fall or early winter it returns to the 
 grains and grasses, where it passes the winter in the viviparous forms. 
 
 69. Macrosiphum heucherae (Thomas) 
 
 Thomas, 8th Ann. Rep. Illinois St. Eut., p. 66, 1879. Siphonophora (orig. 
 
 desc). 
 Davidson, Jour. Econ. Ent., vol. 8, p. 427, 1915 (desc). 
 
 Record. — Heuchera hartwegi, Eedwood Canyon, Contra Costa County (David- 
 son). 
 
 In the latter part of May, 1914, Davidson found all the forms, 
 including the apterous and alate viviparous females, the apterous 
 oviparous females, the alate males, and eggs on the flower stalks of 
 alum root in Contra Co.sta County. Since his description no record 
 has been made concerning the species. The author is unacquainted 
 with it, having never seen specimens. 
 
 70. Macrosiphum jasmini (Clarke) 
 Clarke, Can. Ent., vol. 35, p. 252, 1903. Nectarophora (orig. desc). 
 Secord. — Jessamine, Berkeley (Clarke). 
 
 Since Clarke's description of the apterous viviparous females of 
 this species it has never been found. Its identity is, therefore, un- 
 known to the author. 
 
A SYNOPSIS OF THE APHIDIDAE 65 
 
 71. Macrosiphum lactucae (Kalt.) 
 
 Kaltenbach, Monog. J. Pflanzenlause, p. 199, 1857. Ncctarophora (orig. 
 
 desc). 
 Sanderson, Can. Ent., vol. 33, p. 69, 1901. Ncctarophora (desc). 
 Essig, Univ. Calif. Publ., Entom., vol. 1, p. 328, 1917 (list). 
 
 Secord. — Cicorium intijhus, Rutherford, Napa County, 1916 (Essig). 
 
 This species has been taken only by Essig on chicory in Napa 
 County during June, 1916. As its detei-mination is doubtful the 
 autlior gives herewith a brief description of the alate female. 
 
 Body pale to green, with the following parts more or less dusky : 
 head, antennae, prothorax, thoracic lobes, apex of beak, tarsi, apical 
 one-fifth to one-fourth tibiae, apical one-half femora, cornicles, anal 
 plate, marginal spots on the abdominal segments, submarginal spots 
 of the second and third abdominal segments, dorsal bands on the 
 fourth and fifth, and the dorsum of the remaining abdominal seg- 
 ments. Eyes red. 
 
 The antennal tubercles are prominent and project rectangularly 
 inward. A prominent frontal tubercle is present on the apex. The 
 antennae are about half as long again as the body. The usual primary 
 and accessor}- sensoria are present. On III there are from thirty-five 
 to forty-five circular secondarj- sensoria ; on IV fi'om five to fifteen 
 secondary sensoria. These two segments appear tubereulate. The 
 beak reaches beyond the second coxae. The cornicles are longer than 
 the Cauda, and subequal in length to the fourth antennal segment. 
 They are subcylindrical and fairly stout. The cauda is long and ensi- 
 form, reaching to the tip of the cornicles. The wings and venation 
 are normal. 
 
 Measurements (of three specimens) : Body length, 1.836 to 1.955 
 mm. ; width of thorax, 0.765 to 0.833 mm. ; antennae, total, 2.805 to 
 2.992 mm. ; III, 0.680 to 0.697 mm. ; IV, 0.441 to 0.527 mm. ; V, 0.391 
 to 0.425 mm. ; VI, base 0.085 to 0.119 mm. ; VI, spur 0.952 to 1.105 
 mm. ; cornicles, 0.441 to 0.493 mm. ; cauda, 0.238 to 0.272 mm. ; hind 
 tarsi, 0.136 to 0.153 mm. ; wing, length, 3.145 to 3.315 mm. ; width, 
 0.952 to 1.139 mm. ; expansion, 7.36 to 7.87 mm. 
 
 72. Macrosiphum ludovicianae (Oestund) 
 
 Figures 136, 148 
 
 Oestlund, Minn. Geol. Nat. Hist. Surv., vol. 14, p. 23, 1886. Siplwnopliora 
 
 (orig. desc.). 
 Davidson, Jour. Econ. Ent., vol. 7, p. 136, 1914 (list). 
 Wilson, Trans. Am. Ent. Soc, vol. 12, p. 98, 101.5 (desc). 
 
66 MISCELLANEOUS STUDIES 
 
 Becords. — Artemisia hcterophylla ; Walnut Creek, Contra Costa County (David- 
 son), Berkeley, 1915 (Shinji) ; Artemisia dracunculoides, Convolvulus sp., Stachys 
 bullata, Berkeley, 1915 (Shinji). 
 
 This species is quite common in tlie San Francisco Bay region on 
 various species of sagebrush. George Shinji has taken it also on hedge- 
 nettle and bindweed in Berkeley. It is distinguished from other sage- 
 infesting species of Macrosiphwnt by the fact that the body of the 
 apterous females is covered with pointed setae as opposed to the fan- 
 shaped setae of M. artemisicola (Williams), and the capitate setae 
 of M. artemisiae (Fonsc). 
 
 73. Macrosiphum orthocarpus Davidson 
 
 Davidson, Jour. Econ. Ent, vol. 2, p. 304, 1909 (orig. desc). 
 Davidson, Jour. Econ. Ent., vol. 3, p. 380, 1910 (list). 
 
 'Record. — Orthocarpus purpurascens ; Stanford University (Davidson). 
 
 Since Davidson found the specimens on owl-clover from whicli he 
 described this species, it has not again been taken. 
 
 7-t. Macrosiphum pisi (Kalt.) 
 
 Figures 130, 150 
 
 Kaltenbach, Monog. d. Pflanzenlausc, p. 23, 1843. Aphis (orig. desc). 
 Davidson, Jour. Econ. Ent., vol. 2, p. 304, 1909 (Ust). 
 Davidson, Jour. Econ. Ent., vol. 3, p. 380, 1910 (list). 
 Essig, Pom. Jour. Ent., vol. 2, p. 336, 1910. Nectarophora (desc). 
 Branigan, Mon. Bull. Cal. Comm. Hort., vol. 4, p. 285, 1915. M. destruc- 
 tor (Johnson) (list). 
 Davis, V. S. Dept. Agr., Bull. 276, p. 11, 1915 (list). 
 
 Records. — Pisum sativum.; Clareniont, Santa Ana, and Ventura (Essig) ; Ala- 
 meda County (Brannigan) ; El Cajon, San Diego County, May, 1916: Lathyrus 
 odoratus; Stanford University (Davidson, Morrison); San Diego, October, 1916: 
 Viola sp.; Claremont, Santa Ana, Ventura (Essig); Medicago sp. ; Holtville, 
 Imperial County (V. L. Wildermuth) : Psorales macrostachya ; Santa Paula 
 ' (Essig). 
 
 Tlie pea aphis is quite common throughout the state, especially on 
 garden and sweet peas. It has been taken a few times on other plants, 
 such as alfalfa, violets, and leather-root, but it is uncommon. This 
 species is readily distinguished by its bright, shining green color, large 
 size, and long, slender, imbricated, but non-reticulated cornicles. 
 
A SYNOPSIS OF THE APBIDIDAE 67 
 
 75. Macrosiphum pteridis Wilson 
 
 Figures 317, 318 
 
 Wilson, Trans. Am. Ent. Soe., vol. 41, p. 101, 1915 (orig. desc). 
 
 Records. — Pieris aquilina; Walnut Creek, Contra Costa County, 1915 (David- 
 son). 
 
 This species has been found by Davidson on the fronds of common 
 brake in the San Francisco Baj' region. Wilson reported it as present 
 throughout southern and western Oregon. There are a few specimens 
 of the alate females in the author's collection, received from Davidson. 
 
 76. Macrosiphum rosae (Linn.) 
 
 Figures lOli, 151, 152 
 
 Linnaeus, Syst. Nat. vol. 4, p. 73, 1735. Aphis (orig. desc.). 
 
 Clarke, Can. Ent., vol. 35, p. 254, 1903. Nectarophora (list). 
 
 Davidson, Jour. Econ. Ent., vol. 2, p. 304, 1909 (list). 
 
 Davidson, Jour. Econ. Ent., vol. 3, p. 380, 1910 (list). 
 
 Davidson, Pom. Jour. Ent., vol. 3, p. 399, 1911 (list). 
 
 Essig, Pom. Jour. Ent., vol. 5, p. 550, 1911 (desc). 
 
 Carnes, Mon. Bull. Cal. Coram. Hort, vol. 1, p. 398, 1912 (list). 
 
 Hecords. — Rose; through California from Humboldt County south to San Diego 
 County (Clarke, Davidson, Morrison, Essig, Ferris, Shinji, the author). 
 
 This is the common pink and green aphid of roses, known the 
 world over. The apterae are found most abundantly in the late win- 
 ter and early spring on the buds and stems of rose. As the alates are 
 matured they fly away, supposedly either to other rose bushes or to 
 various grains and grasses. This past spring (1917) it has been very 
 abundant in the vicinity of Riverside, but the previous spring (1916) 
 in San Diego it was rare. There the most abundant rose aphis was 
 Myzaphis rosarum (Walker). 
 
 77. Macrosiphum rudbeckiae (Fitch) 
 
 Fitch, Cat. Homop. N. Y., p. 66, 1851. Aphis (orig. desc). 
 Essig, Pom. Jour. Ent, vol. 3, p. 400, 1911. Aphis (desc)." 
 Davidson, Jour. Econ. Ent., vol. 7, p. 137, 1914 (list). 
 
 Records. — Ambrosia psilostachya ; Santa Paula (Essig) ; Bacchuris vimin-alis; 
 Santa Paula (Essig), Riverside, September, 1916; Dipsacus fullonum; San Jose 
 (Davidson): Helianthus aiinuus ; Riverside, September, 1916; SaUx sp. ; Chrysan- 
 themum; Arlington, Riverside County, September, 1916; undetermined species of 
 Compositae ; Redwood Canyon, Contra Costa County, July, 1914 (R. W. Haegele). 
 
 11 In the drawings accompanying this description by Essig the following mis- 
 takes are noticeable: the third discoidal vein of the forewings is twice-branched 
 instead of once-branched, and the third antenal segment of the apterous female 
 bears several secondary sensoria instead of none, as figured. 
 
68 MISCELLANEOUS STUDIES 
 
 Tliis reddish-eolored Macrosiphum is distributed abundantly 
 through the San Francisco Baj' region and southern California on 
 various Compositae. In one ease the author found it doing consider- 
 able damage to ehr^-santhemums by stunting and distorting the buds. 
 Once he found it infesting the tender leaves and stalks of willow. 
 The author reared specimens of Diarctus rapae Curt, from an infesta- 
 tion of this species taken on willow. 
 
 77a. Macrosiphum rudbeckiae (Pitch) var. madia n.var. 
 
 Figures 153, 154 
 
 In September, 1915, the author found a species of Macrosiphum 
 infesting the heads of tarweed {Madia sativa) on the campus of the 
 University of California, Berkeley. Specimens of Praon simiilans 
 Prao. were reared from this collection. Mounted specimens are almost 
 identical with .V. rudbeckiae (Fitch), but in life they differ in the 
 coloration. Because of this it has been thought best to describe it 
 herewith as a color variety of M. rudheckiae, naming the variety. 
 madia, after its host plant. 
 
 Host: Madia sativa. Date: Septcmlier 12, 1915. 
 
 Locality: Berkelej', California. Collection number: AFS 70-15. 
 
 Alate viiuparous female. — Prevailing color: dark-green, slightly 
 pruinose. Head brownish (fuscous), about as long as broad, with 
 distinct antennal tubercles. Antennae black, except I and II and the 
 base of III, which are concolorous with the head. The spur is slightly 
 longer than III; IV is next in length, followed by V, VI, and I, which 
 are subequal, and II, which is the shortest segment. The spur is about 
 six times as long as the baise of VI. The usual primary sensoria are 
 present on V and VI, and the usual aeces.sory sensoria on VI. IV is 
 without sensoria, III has 25-35 irregularly arranged, various-sized 
 secondary sensoria placed along the whole length of the segment 
 (fig. 154). The thorax is fuseous; the prothorax with rather distinct 
 lateral tubercles. The beak is slightly dusky with the apical one-third 
 lilaek, reaching to the second coxae. The abdomen is greenish with a 
 slight pulveralence, making it appear pruinose. The cornicles are 
 long, slightly tapering, black except the ba-sal one-third, which is 
 concolorous with the abdomen, apical one-fifth reticulate (fig. 153). 
 The Cauda is long and pointed, pale (slightly reddish?), about one- 
 half as long as the cornicles. The legs are black except the basal half 
 of the femora and the coxae, which are greenish. The wings and 
 venation are normal. 
 
A SYNOPSIS OF THE APHIDIDAE 69 
 
 Measurcmtiits : Body length (exclusive of cauda), 2.11 mm.; width 
 of thorax, 0.91 mm. Antennae : total, 2.07 mm. ; I, 0.12 mm. ; II, 
 0.09 mm. ; III, 0.76 mm. ; IV, 0.55 mm. ; V, 0.47 mm. ; VI, 0.12 mm. ; 
 sjiur. 0.78 mm. ; cornicles, 0.91 mm. ; eauda, 0.45 mm. ; beak, 0.89 mm. ; 
 hind tarsns, 0.14 mm. Wing: length, 3.6 mm.; width, 1.25 mm.; 
 expansion, 8.11 mm. 
 
 78. Macrosiphum sanborni Gillette 
 
 Figures 1-11, 1S3 
 
 Sanborn, Ivans. Univ., Sci. Bull. 3, p. 73, 190-1. Macrosiphum chrysaiUhemi 
 
 (dese. ala. vivi.). 
 Gillette, Can. Eut., vol. 11, p. 65, 1908 (orig. desc. apt. vivi.). 
 
 Secords. — Chrysanthemum; Stanford University, May, 1915; Riverside, March, 
 1917. 
 
 Twice has the author found this species: once a small infestation 
 in the greenhouse of Stanford University, and once abundantly out 
 of doors in Riverside. It is an interesting species in that it does not 
 fit well into any known genus. Except for the cornicles it fits Macro- 
 siphum. and has been so considered. The cornicles are, however, short, 
 being searcelj' longer than the cauda, and are somewhat bottle-shaped, 
 being considerably smaller at tlie apex than at the base. 
 
 79. Macrosiphum solanifolii (Ashmead) 
 
 Figures 137-140, 159-160 
 
 Ashmead, Can. Ent., vol. 12, p. 91, 1881. Siphonophora (orig. desc). 
 
 Clarke, Can. Ent., %'ol. 35, p. 252, 1903. Nectarophora citrifolii (Ashmead) 
 (list). 
 
 Davidson, Jour. Econ. Ent., vol. 31, p. 380, 1910. Macrosiphum citrifolii 
 (Ashmead) (list). 
 
 Essig, Pom. Jour. Ent., vol. 3, p. 592, 1911. Macrosiphum citrifolii (Ash- 
 mead) (desc). 
 
 Davidson, Jour. Econ. Ent, vol. 5, p. 411, 1912 (list). 
 
 Patch, Maine Agr. Exp. Sta., Bull. 242, 1915 (desc). 
 
 Records. — Citrus sp.; Azusa, Los Angeles County (Clarke); Lindsey, Tulare 
 County (Clarke) ; Santa Paula (Essig) ; Disporum hookeri; Berkeley, May, 1915 
 (Shinji) : Solanum nigrum; Stanford University, October, 1916 (Ferris) : Fuchsia 
 sp. ; Berkeley, July, 1915: Sonclius asper and S. oleraceus; Stanford University, 
 February, 1915: apple; Stanford University, May, 1915; El Cajou, San Diego 
 County, July, 1916: Atriplcx sp.; Berkeley, September, 1915: Oxalis corniculata, 
 Biverside, February, 1917: Deinandra fasciculaia. Riverside, February, 1917: 
 Erodium moschatum; Pasadena, April, 1917 (R. E. Campbell); Riverside, April, 
 1917. 
 
70 MISCELLANEOUS STUDIES 
 
 This "piuk and green aphid of potato" is distributed throughout 
 California on a large varietj- of plants. It is recognizable bj' the long 
 reticulated cornicles and black antennae. When the author first exam- 
 ined specimens of Macrosiphum citrifolii (Ashmead) in Essig's collec- 
 tion he was struck with its resemblance to this species. In fact, after 
 considerable study he could not find any constant differences. This was 
 in 1915 in Berkeley. This past spring (1917) he had the opportunity 
 in Riverside of making some transfer tests with specimens from oxalis. 
 Migrants wei-e placed under muslin bags on sucker growth of orange. 
 It was observed that these settled there readily and produced young, 
 demonstrating that the citrus species is the same as the other. On 
 the strength of this Macrosiphum citrifolii (Ashmead) is listed as a 
 synonym of this species. 
 
 80. Macrosiphum sonchella (Monell) ? 
 
 Monell, TJ. S. Geol. Geog. Surv., Bull. 5, p. 21, 1879. Siplionophora (orig. 
 
 desc). 
 Clarke, Can. Ent., vol. 35, p. 252, 1903. Nectarophora (list). 
 Davidson, Jour. Econ. Ent., vol. 2, p. 304, 1909 (list). 
 Davidson, Jour. Econ. Ent., vol. 2, p. 380, 1910 (list). 
 
 Rccurcls. — Sonchus sp.; Berkeley, Newcastle, and Palo Alto (Clarke); Stan- 
 ford University (Davidson). 
 
 According to Jlorrison the species listed as this by Davidson is 
 not Macrosiphum sonchella (Monell), although he cannot say what 
 it is. Consequently Clarke probably referred to the same specires as 
 did Davidson. As the author has never seen specimens he can make 
 no statement as to its identity, so lists it as it has been heretofore. 
 
 81. Macrosiphum stanleyi Wilson 
 
 Figures 128, 158 
 Wilson, Troe. Ent. Soc. Brit. Columbia, January, 1915 (orig. desc.). 
 Record. — Samhitcvs callicarpa californica; Berkeley, June, 1915. 
 
 From the early part of June, 1915, until the middle of August, 
 this species was very abundant on an elderberry tree in the Botanical 
 Gardens of the University of California. By the latter part of August 
 all specimens had disappeared. Since then the author has never seen 
 the species. J. J. Davis kindly identified these specimens. 
 
A SYNOPSIS OF THE APHIDIDAE 71 
 
 82. Macrosiphum taraxici (Kalt.) 
 
 Kalteubacli, Moiiog. d. Pflanzenliiuse, p. 30, 1743. Aphis (orig. desc). 
 Theobald, Jour. Econ. Bio!., vol. 7, p. 77, 1913 (desc). 
 
 Eecord, — Taraxacum officinale; California (Wilson). 
 
 H. F. Wilson stated to the author that he liacl taken this species 
 on dandelion (Taraxacum officinale) in California, although he gave 
 no date or locality record. 
 
 83. Macrosiphum tulipae (Monell) 
 
 Monell, IT. S. Geol. Geog. Surv., Bull. 5, p. 19, 1879. Siphoiwphora (orig. 
 
 desc). 
 Davidson, Jour. Econ. Ent., vol. 3, p. 380, 1910 (list). 
 
 Eecords. — Tulipa sp. ; Stanford University (Davidson); Liriodendroii sp. ; 
 Berkeley, 1915 (Essig, Shinji). 
 
 This species is not known to the author. It has been found on 
 tulips and on the tulip trees in the San Francisco Bay region by 
 Davidson, Essig, and Shinji. 
 
 84. Macrosiphum Valerianae (Clarke) 
 
 Clarke, Can. Ent., vol. 35, p. 253, 1903. Nectaropliora (orig. desc). 
 Eecord. — Valeriana officinialis; Berkeley (Clarke). 
 
 In 1903 Clarke described this species from specimens taken on 
 heliotrope in Berkeley. Since then it has not again been found. 
 
 24. Genus Myzus Passerini 
 Passerini, Gli Afidi, 1860. Type Aphis rihes Linnaeus. 
 This genus is very closely related to Rhopalos-iphuni Koch, the 
 principal difference being in the shape of the cornicles. However, 
 some species fall easily into one or the other genus, depending entirelj' 
 upon what form one has. In this respect Rhopalosiphum pcrsicae 
 (Sulz.) is particularly noticeable, the spring migrants having the 
 clavate cornicles of Rhopalosiphum, the fall migrants having the 
 cylindrical cornicles of Myzus. The author has followed Van der 
 Goot in taking out of this genus M. rosarum (Walker) and placing it 
 in the genus Myzaphis v.d.G. The antennal tubercles are lacking, 
 thus placing the species in the Aphidini instead of the Macrosiphini. 
 There are at present ten species of Myzus known to occur in Califor- 
 nia. Following is a key to them : 
 
72 MISCELLANEOUS STUDIES 
 
 Key to the California Species 
 Alate viviparous females 
 
 1. Seeonilary sensoria present on III only 5 
 
 — Secondary sensoria present on other segments as well as on III 2 
 
 2. Secondary sensoria on III, IV, and V S 
 
 — Secondary sensoria on III and IV, none on V fragaefolii Cockerell 
 
 3. Cornicles dusky for entire length cyiiosbati (Oestlund) 
 
 — • Cornicles mostly pale 1 
 
 4. Thoracic lobes distinctly darker than general body color, being black or dark 
 
 brown aquilegia Essig 
 
 — Thoracic lobes at most only slightly darker than body, being a pale brown. 
 
 braggii Gillette 
 
 5. Body black throughout cerasi (Fabricius) 
 
 — Body not black throughout 6 
 
 6. Cornicles pale except at extreme tip 7 
 
 — Cornicles with more than tip dusky 8 
 
 7. VI spur longer than III, the latter with but 9 to 12 sensoria. 
 
 varians Davidson 
 
 — VI spur at most equal to III, the latter with 18 to 26 sensoria. 
 
 lycopersici (Clarke) 
 
 8. Cornicles longer than either IV or V 10 
 
 — Cornicles not longer than either IV or V 9 
 
 9. VI spur longer than II circumflexum (Buckton) 
 
 — VI spur shorter than III ribifolii Davidson 
 
 10. Ill with 15 to 25 sensoria (fig. 178) rhamni (Fonsc.) 
 
 — Ill with but 9 to 12 sensoria varians Davidson 
 
 Apterous viviparotis femiile 
 
 1. Body covered with capitate hairs 2 
 
 — Body not covered with capitate hairs except on head and antennae 5 
 
 2. Secondary sensoria on III 3 
 
 — No secondary sensoria on III 4 
 
 3. Cornicles dusky ribifolii Davidson 
 
 — Cornicles pale except tip aqtuilegiae Essig 
 
 4. Cornicles almost twice as long as III. Body fairly large sized. 
 
 braggii Gillette 
 — ■ Cornicles but slightly longer than III. Body small sized. 
 
 fragaefolii Cockerell 
 
 5. Secondary sensoria on III 6 
 
 — No secondary sensoria on III 7 
 
 6. VI spur longer than III. Several sensoria scattered along the whole length 
 
 of III cynosbati (Oestlund ) 
 
 — VI spur at most equal to HI. Only a few (1-3) sensoria at base of III. 
 
 lycopersici (Clarke) 
 
 7. Ill longer than cornicles. Dorsum of abdomen with dusky ninrkings, shaped 
 
 somewhat as a horseshoe circumflexum (Buckton) 
 
 — Ill at most equal to cornicles. Abdomen not marked as above 8 
 
 8. Body black throughout cerasi (Fabricius) 
 
 — Body not black throughout 9 
 
 9. VI spur almost twice as long as III varians Davidson 
 
 — VI spur but slightly longer than III rhamni (Fonsc.) 
 
A STNOPSIS OF THE APHIDIDAE 73 
 
 85. Myzus aquilegiae Essig 
 
 Shinji, Can. Ent., vol. 40, p. 49, 1917. Mysm sp. (list). 
 
 Essig, Univ. Calif. Publ. Entom., vol. 1, p. 314, 1917 (orig. desc). 
 
 Records. — Aquilegia truncata; Berkeley, 1916 (Essig) : A. vulgare, Inverness, 
 Marin County (Shinji). 
 
 This .species was recently described by Essig from specimens found 
 on columbine on tlie campus of tbe University of California, Berkeley. 
 The author has had access to cotype specimens, although he has never 
 collected it himself. 
 
 86. Myzus braggii Gillette 
 
 Figure 176 
 
 Gillette, Can. Ent., vol. 11, p. 17, 1908 (orig. desc). 
 
 Davidson, Jour. Econ. Ent., vol. 5, p. 409, 1912. Phorodon carduinum 
 (Walker) (list). 
 
 Becords. — Cynara scolymus; Courtland, Oakland, and San Jose (Davidson) ; 
 Riverside, January and February, 1917. 
 
 The author found this species during tlie early spring of 1917 
 infesting the leaves of artichoke in Riverside. The determination of 
 specimens was verified by C. P. Gillette. Davidson reported Phorodon 
 carduinum (Walker) from artichoke in the San Francisco Bay region. 
 His specimens were determined by J. Monell, but P. Van der Goot 
 was doubtful as to its identity. Davidson himself has decided that 
 the species is Myzus braggii Gillette. There is no doubt but that the 
 species on artichoke in California is 31. braggii Gillette, but whether 
 or not this is the same as P. carduinum (Walker) is uncertain. 
 
 87. Myzus cerasi (Fabrieius) 
 
 Figures 112, 121, 122, 179, 307 
 
 Fabrieius, Syst. Nat., p. 734. Aphis (orig. dese.). 
 
 Clarke, Can. Ent., vol. 35, p. 2-52, 1903 (list). 
 
 GOlette, Jour. Econ. Ent., vol. 1, p. 362, 1908 (desc). 
 
 Newman, Mon. Bull. Calif. Comm. Hort., vol. 4, p. 446, 1915 (list). 
 
 Shinji, Can. Ent., vol. 49, p. 49, 1917 (list). 
 
 Becords. — Pnmvs cerasi; Susanville, Lassen County (Newman) ; Berkeley, 
 1914, 1915, and 1916 (Essig, Shinji) ; Riverside, 1914 (Sharp) ; Fresno, June, 
 1915: Prunvs domestica; Berkeley (Clarke). 
 
74 MISCELLANEOUS STUDIES 
 
 The black cherry aphis is found occasionally throughout Califor- 
 nia, but seldom in large enough numbers to be injurious. It infests 
 the terminal leaves of cherry, and sometimes other species of Prunus, 
 causing them to curl to a certain extent. Eggs are laid in the late fall 
 and early winter in the crevices of the bark and near the bases of the 
 buds. These hatch the following spring about the time the buds are 
 opening. The first few generations consist entirelj^ of apterous 
 females. In the early summer the alate females appear, and con- 
 tinue to do so in each succediug generation until fall. In fact, after 
 the first of July, or thereabouts, the majority of the lice produced 
 are alate until the sexes appear in the fall. The first alate females 
 taken by the author were on June 7, 1915. However, on April 25, 
 1916, Essig found a few alate females in Berkeley. In August, 1914, 
 the apterae were also found in Berkeley. 
 
 Van der Goot makes tliis species out of the genus Myzns, using it 
 as the type of his genus Myzoides. The author is inclined to follow 
 him inasmuch as this is quite different from other members of this 
 genus, approaching Aphis in its robust form and separated from 
 that only by the length of the cornicles and presence of antennal 
 tubercles. However, it has so long been considered as a species of 
 Myzus that it is best to leave it so. It is not a good policy usually to 
 form a new genus for one species, especially when it has for so long 
 been considered as a member of another genus. 
 
 88. Myzus circumflexus (Buckton) 
 
 Figure 175 
 
 Buckton, Monog. Brit. Aphides, vol. 1, p. 130, 187.5. Siplwnophora (orig. 
 
 desc. ) . 
 Gillette, Can. Ent., vol. 40, p. 19, 1908. M. vincae, n.sp. (desc). 
 Davidson, Jour. Econ. Ent, vol. 3, p. 380, 1910. M. vincae Gill. (list). 
 Shinji, Can. Ent., vol. 49, p. 49, 1917 (list). 
 
 Bccord. — Vinca major; Stanford University (Davidson, Morrison), Berkeley, 
 1915 (Shinji), Los Angeles, March, 1917; Aesculiis californicus, Alopecurus 
 pratensis, Asparagus, spp., Ceanothus sp., Ccrastium viscosum, Cheirantlms chicri, 
 Cyrtonium falcatum, Digitalis purpurea, Fuchsia sp., Gladiolus sp., Plantago sp., 
 Senecio mikanioides, Sisymbriitm sp., Solanum spp., Stachys hullata, Tropaeolum 
 sp., Symphoricarpus racemosus; Berkeley, 1915, 1916 (Essig, Shinji) : Viola tri- 
 color; Stanford University, March, 1915; Berkeley, 1915 (Essig, Shinji): 
 Biohardia africana; Pomona, 1909 (Essig); Stanford University, March, 1915; 
 Berkeley, March, 1915 (Essig) ; San Diego, May, 1916; Los Angeles, March, 1917. 
 
 This very common aphid is found in the spring on a large variety 
 of host plants throughout California. At times it may become so 
 
A STNOPSIS OF THE A PHI DID AE 75 
 
 abundant as to cause some considerable damage to its liost. On March 
 4, 1917, the author observed it on periwinkle in Los Angeles in such 
 numbers as to stunt the flowers and to cause all the plants to appear 
 black and stickj'. The apterae of this species are readilj' recognized 
 by the black horseshoe-shaped marking on the doi'sum of the abdomen. 
 
 89. Myzus cy-nosbati (Oestlund) 
 
 OestluBd, Minn. Geol. and Nat. Hist. Surv., Bull. 4, p. 81, 1887. Nectaro- 
 
 phora (orig. desc). 
 Davidson, Jour. Econ. Ent., vol. 10, p. 294, 1917 (note). 
 Shinji, Can. Ent, vol. 49, p. 49, 1917. M. ribis (Linn.) (list). 
 
 Eec&rds. — Sibes vulgare ; Walnut Creek (Davidson); Bibes glutinosum, B. 
 memiesii; Berkeley, April, 1915 (Shinji). 
 
 This species has been taken bvit a few times in the San Francisco 
 Bay region ; once on cultivated red currant in company with Aphis 
 neomexicana pacifica, once on wild flowering currant, and once on 
 wild canyon gooseberry. Furthermore, only the sexapura (migrants) 
 and sexuales have been taken. Davidson writes that this is true 
 cynosbati of Oestlund and not the species described by Davis (Ann. 
 Ent. Soc. Am., vol. 2, p. 38, 1909), as MacrosipJntm cynosbati (Oest.), 
 wliich is not that species but some other. Shinji listed M. ribis 
 (Linn.), but liis specimens prove to be the sexuales of this species. 
 
 90. Myzus fragaefolii Cockerell 
 
 Figure 177 
 
 Cockerell, Can. Ent., vol. 33, p. 101, 1901 (orig. desc). 
 Davidson, Jour. Econ. Ent., vol. 7, p. 135, 1914 (desc. sexuales). 
 
 Becords. — Fragaria chiloensis; Walnut Creek, Contra Costa County (David- 
 son) ; Berkeley, March to September, 1915; Palo Alto, April, 1915; Ontario, 
 April, 1917; Buena Park, Orange County, May, 1917 (E. K. Bishop); Santa 
 Barbara, May, 1917; Eialto, San Bernardino County, May, 1917 (A. B. Snow): 
 F. calif ornicus ; Pine Hills, San Diego County, June, 1916. 
 
 On the under side of the leaves of native and cultivated straw- 
 berries this small yellowish aphid is often found, both in the San 
 Francisco Bay region and in southern California. Seldom does it 
 become abundant, although several records of its abundance were 
 received from various parts of the south during the spring of 1917. 
 Several growers have thought it bad enough to spray for it. During 
 the late winter (January and February) the sexuales appear and 
 the eggs are laid. These hatch in a short time, and during the rest 
 of the year tlie alate and apterous viviparae are found. 
 
76 MISCELLANEOUS STUDIES 
 
 91. Myzus lycopersici (Clarke) 
 
 Clarke, Can. Ent., vol. 35, p. 253, 1903. N ectaropliora (orig. desc). 
 Davis, Can. Ent., vol. 46, p. 123, 1914 (desc). 
 
 Becord. — Lycopersicum eseulentum; Berkeley (Clarke). 
 
 Only once has this species been found in California. Davis in 
 1914 described a species from tomato in Idaho, Montana, and Oregon 
 which he believed to be this one. It may be, and it may not be so. 
 That can never be decided for the t j'pes of Clarke 's species are all lost. 
 
 92. Myzus rhamni (Clarke) 
 
 Figure 178 
 
 Clarke, Can Ent., vol. 35, p. 254, 1903. Nectarophora (orig. desc). 
 Shinji, Can. Ent., vol. 49, p. 49, 1917. M. rhamni (Boyer) (list). 
 
 Records. — Bhavmus californicus; Berkeley (Clarke), Berkeley, March, 1915 
 (Shinji). 
 
 In March, 1915, George Shinji took a species of Myzus from coffee- 
 berry in Berkeley. This fits Clarke's description of Nectarophora 
 rhamni, in so far as the description goes. The author considers it to 
 be the same species as described by Clarke, inasmuch as it was collected 
 in tlie same locality and on the same host plant. 
 
 Wilson (Can. Ent., vol. 44, p. 156, 1912) describes a species from 
 Bhamnus purshmn-a in Oregon as M. rhamni (Boyer), listing Clarke's 
 species as a synonym. This is the same species as taken by Shinji 
 in Berkeley, but it is doubtful if it is the species described by Boyer 
 de Fonscolombe. Specimens in the author's collection from Ehamnus 
 in Colorado are determined bj' Gillette and Bragg to be Aphis rhamni 
 Fonsc. These are certainly different from the coast species, the former 
 being an Aphis closely related to A. euonomii Fabr., the latter a 
 Myzns. Prom this evidence the author cannot follow Wilson in 
 placing Nectarophora rhamni Clarke as a synonym of Aphis rhamni 
 Fonsc., considering both as Myzus, but he considers them as distinct, 
 Clarke's species being a Myzus, Fonscolombe 's an Aphis. 
 
 93. Myzus ribifolii Davidson 
 
 Davidson, Jour. Econ. Ent., vol. 10, p. 294, 1917 (orig. desc). 
 
 Becord. — Ribes glutinosum; Redwood Canyon, Contra Costa County (David- 
 son). 
 
A SYNOPSIS OF THE APHIDIDAE 77 
 
 Davidson recently described all forms of this species from speci- 
 mens taken during March, April, and May, 1913, 1914, and 1915, on 
 wild flowering currant in Redwood Canyon, Contra Costa County. 
 The author is unacquainted with the species. 
 
 94. Myzus varians Davidson 
 
 Davidson, Jour. Econ. Ent., vol. 5, p. 409, 1912 (orig. desc). 
 Becord. — Clematis ligusticifolia; San Jose (Davidson). 
 
 Davidson found this species on the under side of the leaves of 
 wild clematis, or Yerhade chivato, near San Jose, and later in Walnut 
 Creek, Contra Costa County. The author is unacquainted with the 
 species. 
 
 25. Genus Nectarosiphon Schouteden 
 
 Schouteden, Aphidologische Notizen, Leipzig, 1901. Type Macrosiplmm 
 rubicola Oestlund, n.n. for Macrosiphum Oestlund, preoccupied. 
 
 Key to California Species 
 1. Body quite large, being about 3 to 4 mm. in length. Wings with dusky spot 
 
 near tip rubicola (Oestlund) 
 
 — ■ Body not so large, being only about 1.5 mm. long. Wings \rithout dusky spot 
 near tip morrisoni Swain 
 
 95. Nectarosiphon rubicola (Oestlund) 
 
 Figures 107, 109, 123 
 
 Oestlund, Minn. Geol. Nat. Hist. Surv., vol. 14, p. 27, 1886. Macrosiphum 
 
 (orig. dese.). 
 Davidson, Jour. Econ. Ent., vol. 7, p. 136, 1914. Amphorophora (list). 
 
 Eecords. — Subus nutkanus; Contra Costa County (Davidson) ; Berkeley (Essig, 
 Shinji). 
 
 This species is sometimes found infesting the tender leaves and 
 shoots of thimbleberry in the San Francisco Bay region. The most 
 distinctive character which readily separates it from Amphorophora 
 rubi (Kalt.) is the presence of a dusky patch near the tip of the fore- 
 wing. This was originally described by Oestlund as the type of his 
 genus Macrosiphum. However, this name was preoccupied by Macro- 
 siphum, Passerini, so Schouteden proposed the name Nectarosiphon 
 for this genus. Davidson listed this species as Amphorophora, and 
 Morrison writes that he has never been able to satisfy himself why 
 
78 MISCELLANEOUS STUDIES 
 
 this is not Amphorophora instead of Ncctarosiphon. There is con- 
 siderable difference in the antennal tubercles of this species and 
 species of Amphorophora, although otherwise they are quite similar. 
 The author believes that slight as the difference is it should be recog- 
 nized for it is througli the shape and size of the antennal tubercles 
 that the different genera of the Macrosiphini are recognized in a 
 large part. In this species the tubercles are large and distinct and 
 neither gibbous nor toothed on the inner side, and with the outer side 
 quite evident, while in Amphorophora they are small and distinctly 
 toothed on the inner side, with the outer side a mere line, or not at 
 all evident. 
 
 96. Nectarosiphon morrisoni Swain 
 
 Figures 124 to 127 
 
 Swain, Trans. Am. Ent. Soc, vol. 44, p. 8, 1918. 
 
 Records. — Cupres.ftts macrocarpa; San Francisco (Compere, Morrison), San 
 Diego (Swain) : C. gnaduhipensis; San Diego (Swain). 
 
 In Golden Gate Park, 8an Francisco, and in Exposition Park, San 
 Diego, this species has been taken on cypress. The small, slender, 
 long-legged apterae are found infesting the terminal leaves of the 
 host. Occasionally an alate female is seen. In San Diego, the apterae 
 were found in company with Cerosipha cupressi Swain. 
 
 26. Genus Pentalonia Coquerel 
 
 Coquerel, Ann. Ent. Soc. France, vol. 7, p. 239, 1860. Type P. nigro- 
 nervosa n.sp. 
 
 97. Pentalonia nigronervosa Coquerel 
 
 Coquerel, Ann. Ent. Soc. France, vol. 7, p. 239, 1860 (orig. desc). 
 Wilson, Jour. Econ. Ent., vol. 2, p. 346, 1909 (desc). 
 
 Mecord. — Fdargonium sp.; Stanford University (Morrison). 
 
 The following note concerning this species is from Morrison: 
 Pentalonia nigronervosa Coquerel. See Wilson, Jour. Econ. Ent., 1909. In 
 the Davidson collection (belonging to Stanford University) there is a single 
 glycerine jelly mount of this species. I have been able to see enough of it to be 
 certain of its identity with that described by WUson in the Journal (above). The 
 record is from geranium, and Davidson once told me that he found it in alcohol 
 in the laboratory [of Stanford University] at the time he began his study of the 
 Aphididae. I believe the record should be published. 
 
A SYNOPSIS OF THE APEIDIDAE 79 
 
 27. Geiins Phorodon Passpriiii 
 Passeriui, Gli Afidi, 1860. Type P. humuli Schr. 
 
 No attempt has been made to formulate a key to the California 
 species of this genus, owing to the fact that the author has specimens 
 of but one species, and that the description of the other is quite inade- 
 quate. Four species have been reported from this state, two of which 
 prove to be species of other genera and one of which is very doubtful. 
 Phorodon carduinuin (Walker) as reported by Davidson, is Myztis 
 hraggi Gillette. Phorodon galeopsidk (Kalteubach), also reported by 
 Davidson, is B.hopalosiphum hippophaaes Koch. There is much 
 diversity of opinion concerning the specific determination of these 
 species and of Myzus clacagni Del Guercio. One might refer to Gil- 
 lette's paper on Bhopalosiphum hippophoaes Koch and 3Iyzus braggii 
 Gillette. Davis writes that he is not prepared to be quoted. Davidson 
 lists P. galeopsidu and R. hippophoaes as synonyms. He states that 
 his specimens listed as P. carduinum Walker were determined by 
 Monell, but that Van der Goot is doubtful, while he himself believes 
 them to be M. braggii Gillette. He has been followed in so listing 
 them. This then leaves but two species reported from California. 
 
 98. Phorodon humuli (Schrank) 
 
 Figures 115 to 118 
 
 Sehrank, Fauna Boica, vol. 2, p. 110, 1801-02. Aphis (orig. desc). 
 
 Clarke, Can. Ent., vol. 35, p. 252, 1903 (list). 
 
 Clarke, Calif. Agri. Exp. 8ta., Bull. 160, 1904 (econ.). 
 
 Parker, U. S. Dept. Agri., Bull. Ill, 1913 (econ.). 
 
 Vosler, Men. Bull., Cal. Comm. Hort., vol. 2, p. 668, 1913 (list). 
 
 Eecords. — Humulris spp. ; Berkeley (Clarke); Placer County (Vosler); Berke- 
 ley, July to September, 1915: Prunus domesUca; Berkeley, March to April, 1915 
 (Essig, Shinji) ; (Parker). 
 
 This is the common hop plant louse found throughout the central 
 part of the state. D\iring the summer it is common on hops, but in 
 the fall the sexupara migrate to plum, where the eggs are laid. These 
 eggs hatch the following spring into stem mothers which feed on the 
 opening buds of plum. During later generations, probably about the 
 third or fourth, alate fundatrigeniae appear, which leave the plum 
 and migrate to hop. Here the summer generations are produced until 
 well into the fall. Parker states that the normal life cycle is as just 
 stated, but that it is also possible, and it occasionally occurs, that this 
 
80 MISCELLANEOVS STUDIES 
 
 aphid maj- live the entire .year upon hops, or on plum, generation after 
 generation of parthenogenetic females being produced. 
 
 99. Phorodon scrophulariae Thomas 
 
 Thomas, Ann. Eep. 111. St. Ent., vol. 8, p. 72, 1879 (orig. desc). 
 Clarke, Can. Ent., vol. 35, p. 252, 1903 (list). 
 
 Record. — Scrophularia sp., Berkeley (Clarke). 
 
 This is a doubtful species, reported bj' Clarke as present on 
 Scrophularia in Berkeley, and by Dr. Thomas in 1879 on a species 
 of plant which he thought to be Scrophularia in Illinois. Since 
 Clarke's record it has never been found, although Morrison states 
 that he has .spent considerable time examining the conmion Scrophu- 
 laria plants in the vicinity of Stanford University, but to no avail. 
 The author attempted to find it many times in the vicinity of San 
 Diego during 1916, and in the vicinity of Riverside in 1917, with no 
 success. 
 
 28. Genus Rhopalosiphum Koch 
 Koch, Die Pflanzenlause, p. 23, 1854. Type Aphis persicae Sulz. 
 
 This genus is very closely related to Myzvs, and is distinguished 
 only by the shape of the cornicles. This distinction is variable, how- 
 ever, as in some species certain forms have the clavate cornicles of 
 Rhopalosiphum while other forms have the cj'lindrical cornicles of 
 Jlyzus. This is particularly true in the case of Rhopalosiphum 
 persicae (Sulz.) and Myzus braggii Gillette. However, most aphidol- 
 ogists separate these two genera, so the author feels that it is best 
 to do so. 
 
 Key to California Species 
 Alate viviparous females 
 
 1. Grounil color dark (olive-^een, wine, brown, and so forth) 2 
 
 — Ground color light, usually green (this does not refer to the dark markings 
 
 on head, thorax, or abdomen, but rather to the ground color of the 
 abdomen) 4 
 
 2. Wing veins with smoky borders and tips (fig. 164). IV with a few small 
 
 sensoria '. violae Pergande 
 
 — Wing veins without smoky borders or tips, and IV without sensoria. 
 
 rhols Monell 
 
 3. Antennae distinctly tuberculate, with sensoria on both III and IV (figs. 170, 
 
 279) 4 
 
 — Antennae not tuberculate, and IV without sensoria, or at most with but a 
 
 few small ones (figs. 167, 168) 5 
 
A SYNOPSIS OF TEE APHIDIDAE 81 
 
 4. VI spur slightly longer than III (figs. 279, 281). Cornicles quite large and 
 
 heavy (figs. 282, 284) lactucae (Kalt.) 
 
 — • VI spur about twice as long as III. Cornicles comparatively small and slender 
 (fig. 165 ) blppophoaes Koch. 
 
 5. First discoidal vein with distinct, smoky border, second discoidal bordered 
 
 slightly so (fig. 166) nervatum Gillette 
 
 — ■ First and second discoidal without smoky borders 6 
 
 6. Abdomen with dusky dorsal markings. Ill with a few (10-12) sensoria 
 
 (fig. 168) persicae Sulz. 
 
 — Abdomen without dusky dorsal markings. Ill with many (24-30) sensoria 
 
 (fig. 167) corylinum Davidson 
 
 Apterous viviparous females^^ 
 
 1. Ground color dark (olive-green, wine, brown) 2 
 
 — • Ground color light (green, and so forth) 3 
 
 2. Cornicles large and stout, longer than III rhois Monell 
 
 — Cornicles smaller and more slender, shorter than III vlolae Pergande 
 
 3. Cornicles longer than III 4 
 
 — Cornicles shorter than III 5 
 
 4. VI spur considerably longer than III, and subequal to cornicles. 
 
 nervatum Gillette 
 
 — VI spur about equal to III, and distinctly shorter than cornicles. 
 
 hippophoaes Koch. 
 
 5. Ill with secondary sensoria lactucae (Kalt.) 
 
 — Ill with no secondary sensoria persicae (Sulz.) 
 
 100. Rhopalosiphum corylinum Davidson 
 
 Figure 167 
 
 Davidson, Jour. Econ. Ent., vol. 7, p. 134, 1914 (orig. desc). 
 Secords. — Corylus rostrata; Walnut Creek, Contra Costa County (Davidson) : 
 Physocarpus capitatus; (Davidson). 
 
 This species was originall.y described from specimens of alate 
 viviparae and pupae taken on wild hazelnut near Walnut Creek. 
 Davidson writes that he has found it quite common on nincbark in 
 the San Francisco Bay region. The author has never taken the 
 species, but has had access to cotype specimens in Essig's collection. 
 
 101. Rhopalosiphum hippophoaes Koch 
 
 Figures 165, 170 
 
 Koch, Die Pflanzenlause, p. 28, 1854 (orig. desc). 
 
 Davidson, Jour. Econ. Ent., vol. 7, p. 136, 1914. Plwrodon galeopsidis 
 
 Kalt. (list). 
 Gillette, Jour. Econ. Ent., vol. 8, p. 373, 1915 (synonomy). 
 
 Becord. — Polygonum sp. ; San Jose (Davidson). 
 
 12 E. corylinum Dvdn. is omitted from this key as the apterous female was 
 never described and specimens are not available to the author. 
 
82 MISCELLANEOVS STUDIES 
 
 Davidson reported this species as present on knotweed in the 
 vicinity of San Jose, under the name P. galeopsidis Kalt. Later he 
 followed Gillette in placing it as a synonym of R. hippophoaes Koch. 
 The author has never collected it, hut has had access to specimens 
 from Davidson in San Jose, and Davis in Oak Park, Illinois. For a 
 full discussion of the synonymy of this species see Gillette's paper 
 listed above. 
 
 102. Rhopalosiphum lactucae (Kalt.) 
 
 Figures 277 to 28-5 
 
 Kaltenbach, Monog. d. Pflanzenlause, p. 37, 1843. Aphis (orig. desc). 
 Davidson, Jour. Econ. Ent., vol. 3, p. 277, 1910 (list). 
 
 Eccords. — Sonchiut spp.; Stanford University (Davidson); Stanford Univer- 
 sity, May to July, 1915; Walnut Creek, May, 1915 (Davidson); Berkeley, July, 
 1915; Lemon Grove, San Diego County, January, 1916; Eiverside, January to 
 May, 1917; Los Angeles, April, 1917: Asclepias sp.; Corvallis, Oregon, November, 
 1913 (Moznette). 
 
 This is a common species infesting the heads of sow thistle 
 throughout the San Francisco Bay region and southern California. 
 In November, 1913, G. F. Moznette took it on milkweed in Corvallis, 
 Oregon. This collection consisted entirely of alate females, that may 
 have been the sexupara. Inasmuch as the identity of this species is 
 doubtful there is given below a brief description drawn from speci- 
 mens of nine alates and eight apterae taken on Sovchus spp. at Stan- 
 ford University in May, 1915, in "Walnut Creek in May, 1915, in 
 Berkeley in July, 1915, and in Lemon Grove in Januai-y, 1916, and 
 on Asclepias sp. in Corvallis, Oregon, in November, 1913. This latter 
 collection is by George F. Moznette of Corvallis. 
 
 Alate viviparous female. — Prevailing color is apple green with 
 the head dark green to black, the prothorax apple green, the thoracic 
 lobes black. The abdomen is apple green with three pair of dusky 
 marginal spots on segments one, two, and three, respectively, and 
 with a larger dusky patch on the dorsum of segments four, five, and 
 six, being between the cornicles. The cornicles and cauda are luteous 
 with the extreme tip of the former dusky. The antennae are dusky 
 throughout. The legs are luteous with the tarsi and tips of the femora 
 and tibiae dusky. 
 
 The head is about twice as broad as long, with a distinct frontal 
 tubercle (fig. 278). The antennae are set on distinct tubercles and 
 are between one and one-fourth to one and one-half times as long 
 
A SYNOPSIS OF THE APHWIDAE 83 
 
 as tlie body. Tlio relative lengths of the segments are as follows: the 
 spur is the longest, being followed by III, which is subequal but never 
 longer. IV is about one-half the length of the spur and slightly longer 
 than V. II is slightly longer than VI, which is about equal to I. 
 Sensoria are arranged as follows (figs. 279-281) : on V and VI are 
 the usual primary and accessory sensoria ; on V in addition to the 
 primary sensoria. there are at times a.s many as seven small circular 
 secondary sensoria, located about the middle of the segment. The 
 number of these sensoria range from none to seven, two and three 
 being the usual number; on IV there are from six to twelve irregular 
 secondarj- sensoria (fig. 280), placed irregularly along the whole 
 length of the segment ; on III there are between thirt.y and forty 
 irregularly placed and irregularlj- sized sensoria (fig. 279) scattered 
 along the whole length of the segment. The usual number is from 
 thirty-six to thirty-nine. The prothorax is without lateral tubercles. 
 The beak is of medium length, reaching to slightly beyond the second 
 coxae. The cornicles (fig. 282) are fairly large and clavate on one 
 side. At the widest point they are slightly less than one-fifth the 
 length. The tip is .slightly wider than the base. They are about the 
 same length as the fourth antennal segment, although in some cases 
 they may be slightly longer, and in others slightly shorter, but in all 
 eases longer than the fifth antennal segment. The cauda (fig. 283) is 
 long and fairly large, not quite reaching to the tip of the cornicles, 
 being about one-half as long as the cornicles and one-half as long 
 again as the hind tarsi. The wings and venation are nomial, the 
 forewings being about twice as long as the body. 
 
 Measurements : Body length, 1.48 to 1.87 mm. ; width, 0.73 to 0.82 
 mm. ; antennae total, 2.35 to 2.51 mm. ; III, 0.544 to 0.697 mm. ; IV, 
 0.306 to 0.425 mm.; V, 0.218 to 0.357 mm.; VI, 0.085 to 0.119 mm.; 
 spur, 0.68 to 0.799 mm. ; cornicles, 0.323 to 0.459 mm. ; cauda, 0.187 to 
 0.255 mm. ; hind tarsi, 0.136 to 0.153 mm. ; wing length, 3.4 to 3.8 mm. ; 
 wing width, 1.2 to 1.5 mm. ; wing expansion, 8.0 to 8.3 mm. The 
 average measurements are as follows : body length, 1.74 mm. ; width, 
 0.768 mm. ; antennae total, 2.445 mm. ; III, 0.645 mm. ; IV, 0.382 mm. ; 
 V, 0.328 mm.; VI, 0.107 mm.; spur, 0.753 mm.; cornicles, 0.403 mm.; 
 cauda, 0.248 mm. ; hind tarsi, 0.139 mm. ; wing length, 3.6 mm. ; width, 
 1.32 mm. ; expansion, 8.1. 
 
 Apterous viviparous fern-ale. — Prevailing color pale green with the 
 head paler, being almost luteous or of a pale j-ellowish green color. 
 The eyes are red. The antennae, except the apices of segments three 
 
84 MISCELLANEOUS STUDIES 
 
 to six inclusive, the legs, except the tarsi and tips of the tibiae, the 
 Cauda, and the cornicles, except the tip, are all luteous. Sensoria are as 
 follows: on V and VI the usual primary sensoria, on VI the accessory 
 sensoria, and on III (fig. 278), from nine to eleven small, circular 
 irregularly placed secondary sensoria. IV is without sensoria. The 
 antennae are considerably longer than the body, the spur and III 
 being subequal and the longest segments. Sometimes the spur is 
 slightly longer than III. V is about one-half as long as III or the 
 spur, and about four-fifths as long as IV. I and VI are subequal, 
 being about one-seventh as long as the spur. The cornicles (fig. 284), 
 are clavate, quite large, usually being slightly more than one-fifth 
 the length of the body and over three times the length of the hind 
 tarsi. The cauda (fig. 285) is long, sickle-shaped, and a little more 
 than one-half as long as the cornicles. 
 
 Measurements: Body length, 1.7 to 2.18 mm.; widtli of abdomen, 
 0.82 to 1.73 mm.; antennae total, 2.32 to 2.48 mm.; Ill, 0.646 to 
 0.714 mm. ; IV, 0.391 to 0.425 mm. ; V, 0.323 to 0.34 mm. ; VI, 0.102 
 mm. ; spur, 0.646 to 0.782 mm. ; cornicles, 0.459 to 0.493 mm. ; cauda, 
 0.238 to 0.272 mm.; hind tarsi, 0.136 to 0.153 mm. The average 
 measurements are as follows : body length, 1.87 mm. ; width, 0.99 mm. ; 
 antennae total, 2.39 mm. ; III, 0.674 mm. ; IV, 0.408 mm. ; V, 0.334 
 mm. ; VI, 0.102 mm. ; spur, 0.7099 mm. ; cornicles, 0.473 mm. ; cauda, 
 0.255 mm. ; hind tarsi. 0.1445 mm. 
 
 103. Rhopalosiphum nervatum Gillette 
 
 Figures 166, 169, 171 
 
 Gillette, Can. Ent., vol. 40, p. 63, 1908 (orig. desc). 
 
 Davidson, Jour. Econ. Ent., vol. 3, p. 378, 1910. B. arbuti, n.sp. (desc). 
 
 Davidson, Jour. Econ. Ent., vol. 7, p. 134, 1914 (list). 
 
 Secords. — Arbutus menziesii; Stanford University, San Jose, Walnut Creek 
 (Davidson); Sacramento (Essig) ; Stanford University, February to May, 1915; 
 Berkeley, September, 1915: Arbutus unedo ; Redlands, February, 1917; Bosa spp. ; 
 Walnut Creek (Davidson); Berkeley, February, 1915 (Essig). 
 
 In 1910 Davidson described a species of Rhopalosiphum, which he 
 named arhvti, from specimens taken on madrone in the vicinity of 
 Stanford University. Since then it has been found quite commonly 
 on madrone throughout the San Francisco Bay region, and once on 
 a strawberry tree in Silva Park, Redlands. It was noticed by the 
 author that the alate females were very scarce at all times, although 
 the apterae and nymphs were often quite abundant. Later, when 
 
A S7N0PSIS OF THE APHIDIDAE 85 
 
 studying specimens while working up a key to the species of Rhopalo- 
 siphum, he found tliat structurally this species was identical with 
 Ehopalosiphum nervatum Gillette. The latter had been taken on roses 
 in the San Francisco Bay region. The identical structure and the 
 scarcity of alates on madrone led to a belief that they were the same 
 species. However, it was too late in the season (October, 1915) to 
 trj' any transfer tests. No opportunity was found to try migration 
 tests until in Februai-y, 1917, when the species was taken in Redlands. 
 Two alate females were reared in the laboratory and then placed 
 under a muslin bag on a rose bush, out of doors. A few daj's later 
 these were examined and several young larvae observed. No further 
 observations were made for two weeks, when it was found that the 
 bag had been ripped off by the severe winds. Although this test was 
 not a complete success the author feels confident of the identity of 
 this species. 
 
 104. Rhopalosiphum persicae^' (Sulz.) 
 
 Figures 108, 119, 120, 168 
 
 Sulzer, Kan. Ins., p. 10.5, 1761. Aphis (orig. desc). 
 
 Clarke, Can. Ent., vol. 35, p. 2.52, 1903. Shopalosiphum dianthi (Schrank) 
 
 (list). 
 Gillette, Jour. Econ. Ent., vol. 1, p. 359, 1908. Myzus (desc.). 
 Davidson, Jour. Econ. Ent., vol. 2, p. 303, 1909. B. dianthi (Schrank), 
 
 B. achjirantes Monell, and Mi/eus (list). 
 Davidson, Jour. Econ. Ent., vol. 3, p. 377, 1910. B. tulipae Thomas (list). 
 Davidson, Jour. Econ. Ent., vol. 3, p. 378, 1910. B. dianthi (Schr.) (list). 
 Davidson, Jour Econ. Ent., vol. 3, p. 379, 1910. Myzus (list). 
 Essig, Pom. Jour. Ent., vol. 3, p. 598, 1911. Myzus (desc). 
 
 Secords. — Throughout California by Clarke, Davidson, Essig, Ferris, Morrison, 
 and the author on Aiutilon sp., Amaranthus retrofiexus, Amsiyickia respectabilis, 
 Bougainvillaea sp., Brassica spp., Capsella bursa-pastoris. Capsicum annuum, 
 Catalpa sp., Chenopodium murale, Citrus spp., Cynoglossum grande, Cyticus pro- 
 liferus, Geranium carolinianum, Hedera helix, Lycopersioum esculentum, Malva 
 parviflorus, Oxalis oregona, Prunus spp., Banunculus californicus, Eaphanus 
 sativU'S, Bumex spp., Sambu/^us glauca, Sanicula inenziesii, Senecio vulgare, 
 Solanum tuberosum, Sonchus spp., Tropaeolum sp., Tulipa sp., Vinca major. 
 
 This green peach aphis is one of the most common aphids found 
 in the state. It is most abundant in the spring, at which time it will 
 be found on almost any plant. According to Gillette various species 
 
 13 George Shinji (Can. Ent., vol. 49, p. 49, 1917) recently described an aphid 
 from specimens taken on Godetia amacna in Berkeley, which he named Myzus 
 godetiae n.sp. The author has not seen specimens of this species, but from the 
 description and figures, it is in all probability Bhopalosiphum. persicae (Sulz.). 
 
86 MISCELLANEOUS STUDIES 
 
 of Prmuis are the winter hosts iu Colorado, while during the sununer 
 it migrates to other plants. In California, however, winter eggs are 
 not laid, the viviparous females living the year round. So far as tlic 
 author has observed in over three years, only the form with elavate 
 cornicles is found in California. 
 
 105. Rhopalosiphum rhois Monell 
 
 Figure 173 
 
 Monell, V. S. Geol. Geog. Surv., Bull. 5, p. 27, 1879 (orig. desc). 
 Davis, Can. Ent., vol. 46, p. 165, 1914. R. howardi (Wils.) (desc). 
 Essig, Univ. Calif. Publ. Ent., vol. 1, p. 330, 1917. K. howardi (Wils.) 
 
 (list). 
 Ibid., p. 334, 1917 (list). 
 
 Records. — Rhus divcrsiloba; Berkeley, April, 1915; Avena satira, Berkeley, 
 (Essig). 
 
 This species has been taken in Berkeley on poison oak and grasses. 
 Essig reported it recently as R. hmmirdi (Wils.), but according to 
 Gillette" this is a synonym of R. rhois Monell, RMis being the winter 
 host, and various species of Gramlnacrac the summer hosts. 
 
 This species does not seem to be a typical Rhopalosiphum, being 
 quite close to Siphocoryne nympharae Linn., but yet not fitting the 
 generic description of Siphocoryne exactl.v. Consequently it is best 
 to list it as has been done heretofore as Rhopalosiphum. 
 
 lOf). Rhopalosiphum violae Pergande 
 
 Figures 164, 174 
 
 Pergande, Can. Ent, vol. 32, p. 29, 1900 (orig. desc). 
 Essig, Pom. Jour. Ent., vol. 1, p. 4, 1909 (desc). 
 Davidson, Jour. Econ. Ent., vol. 2, p. 303, 1909 (list). 
 Davidson, Jour. Econ. Ent., vol. 3, p. 277, 1910 (list). 
 
 Records. — Viola spp.; Claremont, Santa Paula (Essig); Stanford University 
 (Davidson); Palo Alto, May, 1915; Santa Ana, February, 1917; Eiverside, April, 
 1917. 
 
 This beautiful little aphid is found more or less abundantly in the 
 spring on the under side of the leaves of violets throughout the state. 
 Tlie dark red color and broad black wing veins serve to distinguish it 
 readily from other aphids. 
 
 14 Gillette, Jour. Econ. Ent., vol. 8, p. 100, 1915. 
 
A SYNOPSIS OF THE APHIDIDAE 87 
 
 Tribe Aphidini Wilson 
 
 Wilson, Ann. Ent. Soc. Am., vol., 3, p. 331, 1910. 
 Following is a brief characterization of this tribe, from Wilson : 
 
 The characters which separate this tribe from tlie previous one [Macrosiphinij 
 are taken as follows: Antennae shorter than the body, or when as long as the 
 body the cornicles and cauda are very short; antennal tubercles, when present, 
 are indistinct, or else the cornicles and cauda are small; when the cornicles are 
 very long or large the development is limited and the other characters are used 
 to place the genera. 
 
 The California genera included bj- Wilson in this tribe are Ai}his, 
 Cerosipha, Coloradoa, Hyalopierus, Liosomaphis, and Siphocoryne 
 [Hyadaphis]. In addition to these the author includes Toxoptera 
 because of the small and indistinct antennal tubercles and the short 
 cornicles, and Myzaphi^ because of the absence of antennal tubercles. 
 The kej' to the California genera has been formulated by the author, 
 following Wilson, Mordwilko, and Van der Goot. 
 
 1. Antennae five-segmented Cerosipha del Guercio 
 
 — Antennae six-segmented - 2 
 
 2. Cornicles much shorter than cauda Hyalopterus Koch 
 
 — • Cornicles about as long as or longer than cauda 3 
 
 3. Cornicles cylindrical, tapering, or conical, not distinctly clavate (fig. 182), 
 
 except in Coloradoa and Myzaphis, in which they may be slightly clavate 
 
 at the apex (fig. 315) 5 
 
 — ■ Cornicles distinctly clavate (figs. 183, 184) 4 
 
 4. Cornicles long and strongly clavate on one side (fig. 184). Antennae shorter 
 
 than body, with VI spur not longer than III liiosomapliis Walker 
 
 — Cornicles slender and but slightly clavate (fig. 183). Antennae never much 
 
 shorter than body, with VI spur longer than III (fig. 2.58) or with a supra- 
 caudal tubercle (figs. 2.5.5, 256) Siphorcoryne Passerini 
 
 5. Third discoidal vein but one-branched (fig. 276). Body without lateral 
 
 tubercles. Cauda long and prominent, being about as long as cornicles. 
 
 Toxoptera Koch 
 
 — Third discoidal vein twice-branched. Body with or without lateral tubercles. 
 
 Cauda usually distinctly shorter than cornicles 6 
 
 6. Front of head with a very distinct tubercle (figs. 308, 313). Body long 
 
 without lateral tubercles. Cornicles long and often slightly swollen near 
 apex Myzaphis Walker and Coloradoa Wilson 
 
 7. Front of head without prominent tubercle (fig. 233). Body more rounded 
 
 with lateral tubercles on prothorax and seventh abdominal segment, and 
 oftentimes on some of the anterior abdominal segments Aphis Linn. 
 
88 MISCELLANEOUS STUDIES 
 
 29. Geinis Aphis Linn. 
 Linnaeus, Syst. Nat., 1748. Type Aphis rumicis Linn. 
 Key to Califoenia Species 
 Alate viviparous females 
 
 1. Abdomen with floeculent masses of wax. Antennae considerably sliorter than 
 
 body, and VI spur shorter than III alamedensis Clarke 
 
 — • Abdomen without such floeculent masses of wax (except perhaps Aphis cooki 
 Essig) 2 
 
 2. Antennae one and one-half times as long as the body, or more. 
 
 houghtonensis Throop 
 
 — ■ Antennae not so much longer than body; when longer, which is seldom, but 
 
 slightly so 3 
 
 3. Abdomen pale yellowish green. Found only on Morus sp mori Clarke 
 
 — Abdomen darker being black, dark green, yellow. Not found on Morus sp. 4 
 
 4. Abdomen dark-green with an orange band between the cornicles. 
 
 angelicae Koch 
 
 — Abdomen without such an orange band between the cornicles (sometimes there 
 
 is a slight orange or reddish coloring between the cornicles of the apterae 
 of Aphis avenae Fabr., but it is not constant) 5 
 
 5. Abdomen sage-green with faint lateral spots. Ill with apical one-half con- 
 
 spicuously darkened and with six large sensoria. VI spur less than one-half 
 as long as III. On Atriplex spp tetrapteralis Cockerell 
 
 — Not with above combination of characters 6 
 
 6. IV with secondary sensoria (fig. 244) 7 
 
 — IV without secondary sensoria (fig. 204) 27 
 
 7. Cornicles and hind tarsi subequal 8 
 
 — Cornicles considerably longer than hind tarsi 17 
 
 8. VI spur shorter than III 9 
 
 — VI spur equal to or longer than III 13 
 
 9. Cornicles short and tapering 10 
 
 — Cornicles short and incrassate pseudobrassicae Davis 
 
 10. V with secondary sensoria. Body slightly pulverulent cooki Essig 
 
 — • V without secondary sensoria. Body not pulverulent 11 
 
 11. Less than 12 secondary sensoria on III, arranged in a more or less even line 12 
 
 — About 20 to 25 secondary sensoria on III, arranged irregularly along segment 
 
 (fig. 244) senecio Swain 
 
 12. Ill with 9 to 12 sensoria. V and VI base subequal, each being shorter than 
 
 IV lithospermi Wilson 
 
 — Ill with' 5 to 9 sensoria. IV and V subequal, each being longer than VI 
 
 base vibumicolens n.sp. 
 
 13. Cornicles shorter than hind tarsi. A large black species in life being marked 
 
 with white bars and cross bands on the abdomen albipes Oestlund 
 
 — • Cornicles and hind tarsi subequal. Body color greenish 14 
 
 14. Koot-infesting species. Antennae short, scarcely reaching the middle of the 
 
 abdomen middletonil Thomas 
 
 — Aerial species. Antennae reaching to base of the cornicles, or as long as 
 
 body 15 
 
 15. Cornicles incrassate. A mediimi-sized species pseudobrassicae Davis 
 
 — Cornicles cylindrical and tapering slightly. A smaller-sized species 16 
 
A SYNOPSIS OF THE AFHIDIDAE 89 
 
 It). Cauda shorter than himl tarsi. Ill witli 11 to 15 sensoria scattered irregu- 
 larly along segment (fig. 294) marutae Oestlund 
 
 — Cauda longer than hind tarsi. Ill with 5 to 9 more or less evenly' arranged 
 
 sensoria viburnicolens n.sp. 
 
 17. Cornicles equal to or longer than III 18 
 
 — Cornicles not as long as III 19 
 
 18. Cauda, cornicles, and III subequal. Second branch of third discoidal vein 
 
 very near to apex of wing spiraecola Patch 
 
 — Cauda considerably shorter than cornicles or III, the last two being subequal. 
 
 Second branch of third discoidal about midway between base of first 
 branch and apex of wing oenotherae Oestlund 
 
 19. Fore wing with the second branch of the third discoidal arising very near 
 
 to the apex of the wing. (In a few eases the second branch is not found, 
 but never in both wings) (fig. 191) avenae Fabr. 
 
 — Venation of fore wing normal (fig. 187) 20 
 
 20. Antennae longer than body persicae-niger Smith 
 
 — Antennae not longer tlian body 20 
 
 21. A pair of small tubercles present on the middle of the seventh and eighth 
 
 abdominal segments malifoliae Fitch 
 
 — Such tubercles not present 22 
 
 22. V with secondary sensoria. VI spur longer than III 23 
 
 — V without secondary sensoria. VI spur at most equal to III 25 
 
 23. Beak scarcely reaching second coxae maidis Fitch 
 
 — Beak reaching beyond second coxae, even to or beyond the third 24 
 
 24. Cornicles longer than cauda (figs. 194, 19.5) and more than twice as long as 
 
 hind tarsi sambucifoliae Fitch 
 
 — ■ Cornicles and cauda subequal ; tlie former not more than twice as long as 
 hind tarsi neomexicana Cockerell var. pacifica Davidson 
 
 25. Cauda and hind tarsi subequal. Ill with a few large sensoria (fig. 232). 
 
 Abdomen green with dark dorsal markings ramona Swain 
 
 — Cauda longer than hind tarsi. Ill with several sensoria. Abdomen black 
 
 or dark brown 26 
 
 26. Cornicles more than twice as long as hind tarsi, often almost three times as 
 
 long. VI spur and cornicles subequal, hind tarsi and VI base subequal. 
 
 hederae Kalt. 
 
 — Cornicles never more than twice as long as hind tarsi, usually considerably 
 
 less. Hind tarsi usually slightly longer than VI base, and VI spur longer 
 than cornicles euonoml Fabr. 
 
 27. Cornicles distinctly knobbed, the tip being widened to twice the width of 
 
 the rest of the cornicles frigidae Oestlund 
 
 — Cornicles normal 28 
 
 28. Fore wing with the second branch of the third discoidal arising very near the 
 
 apex of the wing (fig. 188) salicicola Thomas 
 
 — Fore wing with venation normal (fig. 187) 29 
 
 29. Cornicles distinctly longer than cauda 31 
 
 — Cornicles at most equal to cauda 30 
 
 30. Cornicles short and swollen throughout apical one-half (fig. 203). Antennae 
 
 as long as or longer than the body brassicae Linn. 
 
 — Cornicles short and slender, and slightly clavate on one side. Antennae 
 
 scarcely two-thirds as long as the body atripllcis Linn. 
 
 31. Abdomen without lateral tubercles on anterior segments. Cauda short and 
 
 broad, with rounded tip, and almost as long as the cornicles cardui Linn. 
 
 — Abdomen with lateral tubercles on at least one of the anterior segments .... 32 
 
90 MISCELLANEOUS STUDIES 
 
 32. VI spur shorter than III 33 
 
 — VI spur not shorter than III 34 
 
 33. Cornicles about three times as long as cauda medicaginis Koch 
 
 — Cornicles not three times as long as cauda 34 
 
 34. Cauda more than one-half as long as cornicles 35 
 
 — Cauda not more than one-half as long as cornicles 38 
 
 35. Ill with four or five fairly large sensoria oregonensls Wilson 
 
 — Ill with many irregular sensoria 36 
 
 36. Ill with 20 or more sensoria, IV with none 37 
 
 — Ill with less than 20 sensoria, usually 14 or 15. IV usually with one or two, 
 
 or more sensoria euonomi Fabr. 
 
 37. IV about one-third longer than V. Cornicles about four times as long as 
 
 broad at base. On Heraclium spp heraclil Cowen 
 
 — IV but about one-sixth longer than V. Cornicles about three times as long 
 
 as broad at base. On Tucca sp Yuccae Cowen 
 
 38. A few (about 10) equal-sized sensoria on III (fig. 222). A large yellow 
 
 species with distinct dark markings nerii Fonsc. 
 
 — About 20 irregular sensoria on III (fig. 211). Not yellow 39 
 
 39. Cornicles slightly more than twice as long as hind tarsi 40 
 
 — Cornicles not twice as long as hind tarsi carl Essig 
 
 40. Hind tarsi slightly longer than cauda ceanothl Clarke 
 
 — Hind tarsi shorter than cauda comifoUae Fitch 
 
 41. VI spur one and one-half or more times as long as III setariae Thomas 
 
 — VI spur never so much longer than III 42 
 
 42. Ill with a few large circular sensoria (5-10) (figs. 226, 290) 43 
 
 — Ill with several (15 or more) irregular sensoria 45 
 
 43. Beak reaching to or beyond third coxae. IV never with sensoria. 
 
 gossypli Glover 
 
 — Beak not reaching third coxae 44 
 
 44. VI spur longer than III (fig. 226). Small size pomi de Geer 
 
 — VI spur subequal to or shorter than III (figs. 289, 290). Medium to large 
 
 size cerasifoliae Fitch 
 
 45. Cornicles twice as long as cauda. Femora of all three pairs of legs similarly 
 
 colored carl Essig 
 
 — Cornicles longer than cauda, but not twice as loag. Femora of first pair of 
 
 legs pale, of second and third pair black euonomi Fabr. 
 
 Apterous viviparous femalesi^ 
 
 1. Cornicles shorter than hind tarsi 2 
 
 — • Cornicles equal to or longer than hind tarsi 4 
 
 2. VI spur longer than III. White bars and bands on abdomen in life. 
 
 alblpes Oestlund 
 
 — VI spur not longer than III. Abdomen not as above 3 
 
 3. Cornicles and cauda subequal. Beak not reaching to second coxae. Pul- 
 
 verulent brassicae Linn. 
 
 — Cornicles shorter than cauda. Beak reaching to or beyond second coxae. Not 
 
 pulverulent atripUcls Linn. 
 
 15 In this key only those species are included of which there are specimens in 
 the author's collection or of which there are adequate descriptions available. 
 The following species are therefore omitted: Aphis alamedensis Clarke, A. hough- 
 tonensis Throop, A. inori Clarke, A. neomexicana Cockerell, A. oenotherae Oest- 
 lund, and A. tetrapteralis Cockerell. 
 
A SYNOPSIS OF THE APSIDIDAE 91 
 
 4. Cornicles and hind tarsi subequal : 5 
 
 — Cornicles longer than hind tarsi 10 
 
 5. Secondary sensoria on III and IV. Root species _ mlddletonii Thomas 
 
 — No secondary sensoria. Aerial species 6 
 
 6. Ill longer than VI spur 7 
 
 — Ill shorter than or at most equal to VI spur 8 
 
 7. IV and cornicles subequal llthospermi Wilson 
 
 — IV shorter than cornicles. Pulverulent cooki Kssig 
 
 8. IV and cornicles subequal. Antennae considerably more than one-half the 
 
 length of the body 9 
 
 — IV shorter than cornicles. Antennae at most one-half the length of the body. 
 
 senecio Swain 
 
 9. Cornicles twice as long as cauda and slightly swollen before the tip. 
 
 avenae Fabr. 
 
 — Cornicles not twice as long as cauda, cylindrical, and tapering toward tip. 
 
 marutae Oestlund 
 
 10. Cornicles less than twice as long as hind tarsi 11 
 
 — Cornicles twice as long as or longer than hind tarsi 21 
 
 11. Secondary sensoria on III and IV. Root-infesting species. 
 
 mlddletonii Thomas 
 
 — No secondary sensoria. Aerial species 12 
 
 12. VI spur longer than III 1 13 
 
 — VI spur at most equal to III 16 
 
 13. Ground color, black or dark brown 14 
 
 — Ground color, a shade of green 15 
 
 14. VI spur one and one-half to two times as long as III. Apex only of femora 
 
 dusky setarlae Thomas 
 
 — VI spur but slightly longer than III. Apical one-half of femora dusky. 
 
 medecaginis Koch 
 
 15. Pale green. Cornicles and cauda subequal. Dark, mottled green. Cornicles 
 
 twice as long as cauda or longer avenae Fabr. 
 
 16. VI spur considerably shorter than III 17 
 
 — VI spur almost as long as III 18 
 
 17. Cornicles swollen toward tip pseudobrassicae Davis 
 
 — Cornicles cylindrical and tapering toward tip ramona Swain 
 
 18. Cornicles but slightly longer than hind tarsi 19 
 
 — Cornicles about one and one-half times as long as hind tarsi 20 
 
 19. Dark green. Cornicles at least three times as long as broad at base. 
 
 maidis Fitch 
 
 — Pale green. Cornicles at most twice as long as broad at base. 
 
 senecio Swain 
 
 20. Dark green to reddish yellow. On Yu^ca spp yuccae Cowen 
 
 — Black or very dark brown with black dorsal bands and spots. On various 
 
 plants euonomi Fabr. 
 
 21. Cornicles distinctly knobbed at tip frlgidae Oestlund 
 
 — Cornicles normal 22 
 
 22. VI spur longer than II 23 
 
 — VI spur at most equal to III 25 
 
 23. Pale green with dusky dorsal abdominal markings calendulicola Monell 
 
 — Not colored as above, either not green, or if green with dusky dorsal abdom- 
 
 inal markings 24 
 
92 MISCELLANEOUS STUDIES 
 
 24. Bright yellow with black markings. Cornicles at least three times as long as 
 
 hind tarsi nerii Fonsc. 
 
 — Dark green with black markings. Cornicles but about twice as long as hind 
 
 tarsi cardui Linn. 
 
 25. Cornicles longer than III 26 
 
 — Cornicles at most equal to III 29 
 
 26. Ill considerably longer than VI spur 27 
 
 — Ill subequal to or but slightly longer than VI spur 28 
 
 27. Black. Cornicles about three times as long as hind tarsi. Ill one and one- 
 
 half times as long as VI spur sambucifoliae Fitch 
 
 — Green, pale to apple. Cornicles about four times as long as hind tarsi. Ill 
 
 almost twice as long as VI spur saUcicola Thomas 
 
 28. Cornicles subequal to or but slightly longer than III, and about twice as 
 
 long as Cauda prunorum Fabr. 
 
 — Cornicles one and one-half to two times as long as III, and about four times 
 
 as long as cauda oregonensis Wilson 
 
 29. Cornicles considerably shorter than III 30 
 
 — Cornicles subequal to or but slightly shorter than III 37 
 
 30. Ill and IV spur subequal persicae-niger Smith 
 
 — Ill longer than VI 31 
 
 31. Cornicles at least twice as long as cauda 36 
 
 — • Cornicles not twice as long as cauda 32 
 
 32. Pale green, pulverulent cerasifoliae Fitch 
 
 — Dark green, brown, or black, not pulverulent 33 
 
 33. Cornicles about three times as long as hind tarsi 34 
 
 — Cornicles not three times as long as hind tarsi 35 
 
 34. Ill with a few small secondary sensoria hederae Kalt. 
 
 — No secondary sensoria cornifoliae Fitch 
 
 35. Cornicles considerably more than twice as long as hind tarsi. Lateral abdom- 
 
 inal tubercles only on first and seventh segments heraclii Cowen 
 
 — Cornicles at most but slightly more than twice as long as hind tarsi. Lateral 
 
 tubercles usually on more than first and seventh segments ....euonomi Fabr. 
 
 36. Antennae about as long as body. Cornicles more than twice as long as 
 
 Cauda carl Essig 
 
 — Antennae but about one-half as long as body. Cornicles but about twice as 
 
 long as Cauda gossypii Glover 
 
 37. Ill considerably longer than VI spur 38 
 
 — Ill and VI spur subequal 40 
 
 38. A pair of dorsal abdominal tubercles on sixth and seventh segments. 
 
 malifoliae Fitch 
 — ■ No dorsal abdominal tubercles on sixth and seventh segments 39 
 
 39. Cornicles green, cylindrical, tapering slightly toward tip, and fairly straight. 
 
 Cauda about one and one-half times as long as hind tarsi. Abdomen with- 
 out dusky dorsal markings ramona Swain 
 
 — Cornicles black, cylindrical, curved outward. Cauda and hind tarsi subequal. 
 
 Abdomen with dusky dorsal markings seanothi Clarke 
 
 40. VI spur slightly longer than III. Cornicles and cauda subequal. 
 
 vibumicolens n.sp. 
 
 — VI spur slightly shorter than III. Cornicles one and one-half times as long 
 
 as cauda pomi De Geer 
 
A SYNOPSIS OF TEE APHIDIDAE 93 
 
 107. Aphis alamedensis Clarke 
 Clarke, Cau. Eut., vol. 35, p. 2.51, 1903 (orig. desc). 
 Kccord. — Pnutus domestica ; Berkeley (Clarke). 
 
 Tills is an unknown species described from specimens taken by 
 Clarke on greengage plum in Berkeley. Davidson suggests that it 
 might be Aphis cardui Linn, {pruni Koch) from its brief description. 
 
 108. Aphis albipes Oestlund 
 
 Figures 198 to 200 
 
 Oestlund, Geol. Nat. Hist. Surv. Minn., Bull. 4, p. 52, 1887 (orig. desc). 
 WUliams, Univ. Neb. Studies, vol. 10, p. 119, 1910 (desc). 
 Davidson, Jour. Econ. Ent., vol. 3, p. 376, 1910 (list). 
 
 Eecords. — Symphoricarpus raceviosus; Stanford University (Davidson) ; Con- 
 gress Springs, Santa Clara County, July, 1915 (McCracken) ; Berkeley, July, 1915 
 (Shinji). 
 
 This species is found at times curling the leaves of snowberrj- in 
 the San Francisco Bay region. Dr. McCracken noted in connection 
 with the infestation at Congress Springs, "they are quite prettily 
 patterned with white bars and cross-bars." This is usually enough 
 to distinguish them. 
 
 109. Aphis angelicae Koch. 
 
 Koch, Die Pflanzenliiuse, p. 521, 1854 (orig. desc). 
 Wilson, Jour. Econ. Ent., vol. 2, p. 348, 1909 (desc). 
 
 Record. — Angelica sp., Hcdera sp. ; California (Wilson). 
 
 Wilson reported this species from California, but gave no locality 
 or date. It is unknown to the author. 
 
 no. Aphis atriplicis Linn. 
 
 Linnaeus, Fauna Sweden, p. 1000, 1761 (orig. desc). 
 Hayhurst, Ann. Ent. Soc Am., vol. 2, pp. 88-100, 1909 (desc). 
 Davidson, Jour. Econ. Ent., vol. 5, p. 407, 1912 (desc sexuales apterous 
 viviparae). 
 Davidson, Jour. Econ. Ent., vol. 7, p. 133, 1914 (desc. fundatrix). 
 
 Sccords. — Chenopndium album, C. murale; San Jose, Walnut Creek (David- 
 son). 
 
94 MISCELLANEOUS STUDIES 
 
 This has been reported twice from pigweed or goosefoot in the 
 San Francisco Bay region, where Davidson states that it is very 
 common. The sexes occur in October. Davidson believes that there 
 is an alternate host, but as to what it might be, he is uncertain. The 
 author has never collected specimens, but has had access to material 
 taken by R. W. Doane on Chenopodium in Utah in August, 1916. 
 
 m. Aphis avenae Fabr. 
 
 Figures 191, 201, 202 
 
 Fabricius, Ent. Syst., p. 736, 1775 (orig. desc). 
 Clarke, Can. Ent., vol. 35, p. 254, 1903 . Nectarophora (list). 
 Davidson, Jour. Eeon. Ent., vol. 3, p. 377, 1910. Siphocori/ne (list). 
 Essig, Pom. Jour. Ent., vol. 3, p. 465, 1911. A. padi Linn. (list). 
 Essig, Pom. Jour. Ent., vol. 4, p. 790, 1912. A. inaidis Fitch (desc). 
 Smith, Men. Bull. Cal. Comm. Hort., vol. 3, p. 116, 1914 (list). 
 Davidson, Men. Bull. Cal. Comm. Hort., vol. 6, p. 65, 1917 (note). 
 
 Secords. — Graminaceae (various spp.) ; California, December to May (David- 
 son, Essig, Morrison, author) : Phalaris arundinaccn ; Stanford University, May 
 to July, 1915: Dracaena draco; Stanford University, June, 1915: Musa sapientum ; 
 San Diego, March, 1916: Typha latifolia; (Davidson). 
 
 This is an abundant species througli the state, occurring during 
 the late winter and spring on grasses and grains, migrating to other 
 hosts as these become ripened and dried. 
 
 The life history of this species, according to Davis (U. S. Dept. 
 Agr., Bull. Ill, April, 1914), is somewhat as follows: 
 
 The spring colonics on grains and grasses originate from viviparous females 
 which passed the winter on the grains and grasses, or from spring migrants from 
 the apples or related fruits; i.e., the progeny of the aphids hatching from eggs 
 laid the previous fall on such trees. As the weather becomes cooler they seek the 
 lower parts or the roots of wheat and other plants of the grass family, and 
 here pass the winter as viviparous females ; or the winged fall migrants from the 
 grain may seek such trees as the apple, where the true sexes are produced. 
 
 Undoubtedly the most common method of wintering over in Cali- 
 fornia is on the roots and lower parts of the grains and grasses. This 
 species has never been collected on apples or other related trees in 
 this state, nor have the eggs ever been observed. During the early 
 spring it is found abundantlj' on the grains and small grasses, in 
 January and February in the southern part of the state, and during 
 April and May in the central part. As the grains ripen and the 
 stalks and leaves become hardened, it seems that the aphids migrate 
 to other varieties of grass which remain soft and green later, as 
 canary grass and reed grass and corn, or even to such hosts as the 
 
A SYNOPSIS OF THE APHIDIDAE 95 
 
 dragon tree, cat-tail riisli, and the banana. But the winter is spent 
 as viviparous females on the grains and grasses. 
 
 This species has been confused many times with other species 
 infesting grains, such as Macrosiphum grutiarium (Kirby) and Tox- 
 optera graminum (Rond.). As the latter does not occur in this state 
 it cannot be confused here with Aphis avenae Fahr. Clarke listed this 
 as Ncctarophara avoiae Fabr., so it appears that he might have had 
 Mdcrosiphum. granarium (Kirby) in mind, as it is highly improbable 
 tliat he could have confused J^p/u's avenae Fabr. with a species of 
 Macrosiphum [Nectnrophora). The cornicles of aveiKie Fabr., the 
 absence of antennal tubercles, and the irregular venation make it 
 quite easily distinguishable. The cornicles are quite short, as com- 
 pared with a species of Macrosiphum, and distinct antennal tubercles 
 are entirely lacking. The third discoidal vein of the forewing is 
 typically twice-branched, but the second is close to the apex of the 
 wing, and sometimes is entirely lacking. The only other species of 
 Aphis in this state with this character is Aphis salicicola Thomas, 
 found on willows. These two are readilj* distinguished from each 
 other by the comparative lengths of the cornicles, which are consider- 
 ably longer in salicicola Thomas than in avenae Fabr. 
 
 112. Aphis brassicae Linnaeus 
 
 Figures 203, 204 
 
 Linnaeus, Syst. Nat., vol. 2, p. 734, 1735 (orig. dese.). 
 Clarke, Can. Ent., vol. 35, p. 250, 1903 (list). 
 Davidson, Jour. Ecou. Eut., vol. 2, p. 302, 1909 (list). 
 Davidson, Jour. Econ. Ent., vol. 3, p. 876, 1910 (list). 
 Davidson, Pom. Jour. Ent., vol. 3, p. 399, 1911 (list). 
 Essig, Pom. Jour. Ent., vol. 3, p. 523, 1911 (desc). 
 
 Secord. — Cruciferae (various spp.); throughout California. 
 
 During tlie late winter and spring cruciferous plants are often 
 heavily infested with this species. Of the cultivated plants cabbages 
 and radishes seem to be most heavily infested ; while the wild mustard 
 and radish often have the entire flower clusters covered with these 
 aphids. Oftentimes in the colonies of this species are also found 
 Aphis pseud-ohrassicae Davis, Rhopcdosiphum lactncae (Kalt.), and 
 R. pcrsicae (Sulz.). In southern California the colonies are always 
 attacked by the braconid fl.v, Diaretus rapae Curtiss, and a large per- 
 centage of the individuals destroyed. As summer comes on these para- 
 sites and such predatore as syrphids and ladybirds usually get the 
 best of the aphids, which disappear to a large extent until fall. 
 
96 MISCELLANEOUS STUDIES 
 
 113. Aphis calendulicola ilonell 
 
 Monell, U. S. Geol. Geog. Surv., Bull. 5, p. 23, 1879 (orig. desc). 
 Clarke, Can. Ent, vol. 35, p. 250, 1903 (list). 
 
 Record. — Calendula officinale; Berkeley (Clarke). 
 
 Tlii.s species has uot beeu recognized since Clarke's report of it on 
 marigold. It is possible that he had Aphis senecio Swain, which is 
 very common on marigolds throughout the state. 
 
 1]4. Aphis cardui Linn. 
 
 Figures 208, 209 
 
 Linnaeus, Syst. Nat., vol. 2, p. 735, 1735 (orig. desc). 
 
 Games, Men. Bull. Cal. Comin. Hort., vol. 1, p. 399, 1912. Aphis pruni 
 
 (list). 
 Davidson, Jour. Econ. Ent., vol. 5, p. 407, 1912 (list). 
 Patch, Maine Agr. Exp. Sta., Bull. 233, p. 263, 1914 (desc). 
 
 Eecords. — Cirsiv,m sp.; San Jose (Davidson); Berkeley, June, 1915: Prunus 
 domestica ; Orangevale, Sacramento County (Carnes) ; Walnut Creek (David.son) ; 
 Berkeley, March, 1916 (Essig). 
 
 According to Patch this tiiistle aphid is the same as the one infest- 
 ing plums and formerlj^ known as A. pruni Koch. Both are abundant 
 in the San Francisco Bay region, pruni being found in the fall and 
 spring on plum, cardui during tlie summer on thistle. The author 
 has attempted no transfer tests, so accepts Patch's statement as 
 authority for the synonym}-. It is certain that structurally these are 
 strictly identical. 
 
 n."). Aphis cari Essig 
 
 Essig, Univ. Calif. Publ. Eiitom., vol. 1, pp. 317-321, 1917 (orig. desc). 
 
 Record. — Canmi kellogr/ii ; Rutlierford, Napa County (Essig); Angelica 
 tomentosa ; Berkeley (Essig). 
 
 Essig recently described this from specimens taken on wild anise 
 in Rutherford. The author has seen cotype specimens, but has never 
 collected the species. 
 
 116. Aphis ceanothi Clarke 
 
 Figures 210, 211 
 
 Clarke, Can. Ent., vol. 35, p. 250, 1903 (orig. desc). 
 Davidson, Jour. Econ. Ent,, vol. 2, p. 302, 1909 (list). 
 Davidson, Jour. Econ. Ent., vol. 3, p. 377, 1910 (list). 
 Essig, Pom. Jour. Ent., vol. 3, p. 525, 1911. Aphis ceanothi-hirsuti n. sp. 
 (desc). 
 
A SYNOPSIS OF THE APHIDIDAE 97 
 
 Bccords. — Ceaiwthus intcgrrrimus; Colfax, Placer County (Clarke) ; Witch 
 Creek, San Diego County, June. lOKi: C. cuncatus; Stanford University (David- 
 son), November, 1910 (Morrison), October, 1915 (R. A. Vickerey) : C. thysiflorus ; 
 Bear Creek Gulch, Santa Clara County, April, 1911 (Morrison) : C. hirsuti; Santa 
 Paula (Essig). 
 
 This is a widely distributed species, liaving been found on Ceano- 
 thus as far north as Plaeer County, and as far south as San Diego 
 County. It is seldom abundant, however. The species that Essig 
 described as A. ceanoihi-hi):siiti n.sp. is undoubtedly the same as 
 Clarke described. 
 
 117. Aphis cerasifoliae Fitch 
 
 Figtures 288 to 292 
 
 Fitch, Bept. Ins. N. Y., vol. 1, p. 131, 1855 (orig. desc). 
 Patch, Maine Agr. Exp. Sta., Bull. 233, p. 260, 1914 (desc). 
 
 Becord. — Prunus emorginata; Wynola, San Diego County, June, 1916. 
 
 This aphid was found abundantlj' curling the terminal leaves of 
 wild cherry near Wynola (3700 feet altitude), San Diego County, in 
 June, 1916. Alate and aptei-ous viviparous females as well as nymphs 
 were abundant in the curled leaves. The apterae and nymphs were 
 .slightlj' pulverulent. This species corresponds very closely to Aphis 
 cerasifoliae Fitch as described by Patch {op. cit.), although there are 
 some minor differences. Following is a copy of Patch's description of 
 the Maine specimens of this species : 
 
 This well defined species is common on both the native choke cherry, Prumis 
 virginiana, and the western P. demissa Walp. introduced in a nursery row on our 
 campus. 
 
 Aptcrotis female. — Head, pale green or water wliitisli, beak short, extending 
 to second coxae, eyes, antennae with I, II and III concolorous with head, distal 
 half darker to black, III with no sensoria, proportions as shown in figure; pro- 
 thorax pale green, lateral tubercles present; thorax green with dark green mid- 
 dorsal line, femora and tibiae pale and tarsi black; abdomen pulverulent, pale 
 green with dark green median line and dark green transverse lines between seg- 
 ments, lateral tubercles present, cornicles pale with dusky tips, slender, slightly 
 tapering, and approximately twice the tarsus in length, cauda white with dark tip ; 
 conical, being broad at base and abruptly tapering. 
 
 Nymphs and pupae are also pulverulent and have dark green niiddorsal and 
 transverse intersegmental line, though these are not always well defined in the 
 pupa which has two lateral dark green lines on thorax. 
 
 Alatc female. Head black, beak short, not reaching to second coxae, eyes 
 black, antennae dark. III with from about 12 to 18 large sensoria about the size 
 of the terminal one on V, IV with from none to several sensoria like those on III, 
 proportions of joints as shown in the figure; prothorax green with black trans- 
 verse band, lateral tubercles present; thorax black, wings iridescent with slender 
 brown veins and largo dusky stigma with pointed tip ; commonly though not 
 
98 MISCELLANEOUS STUDIES 
 
 always with second branch very short, abdomen glabrous, rather bright though not 
 vivid green, median line dark green, sutural lines dark green ending in marginal 
 green dots, cornicles dark, cauda greeu. 
 
 Aphis cerasifolUie is gregarious on the ventral surface of the terminal leaves 
 badly curling and deforming them. A copious amount of honeydew is present, 
 and ants are usually found attending a colony of this species. 
 
 The specimens from W,yiiola agree very well with this description, 
 although as stated above, there are a few minor points of difference. 
 However, as Dr. Patch writes: "It seems too close to cerasifoliae to 
 give it a distinct name," and "if the appearance in life answers my 
 description of cerasifoliae I should be inclined to call it that. It hap- 
 pens to be a species as characteristic alive as dead." Following are 
 the notes the author took of its appearance alive, before he suspected 
 its identity: "Alates, apterae and nymphs abundant on terminal 
 leaves curling them badlJ^ Large amount of honeydew and many 
 ants in attendance. Apterae and nymphs pulverulent." These notes 
 agree exactly with Patch 's notes, cited above. 
 
 Following is a brief description of specimens taken at Wj-nola on 
 July 8 : 
 
 Apterous viviparous female. — Prevailing color pale apple green, 
 pulverulent. Head luteous. Thorax and abdomen pale green with 
 middorsal longitudinal stripe darker green. Antennae with the three 
 basal joints luteous, tlie three apical joints .shading into black. Pri- 
 mary sensoria on V and VI, accessory sensoria on VI, no secondary' 
 sensoria. Ill and spur are subequal, or III slightly the longer. IV 
 and V subequal and a little more than one-half as long as III. In 
 some cases IV is slightly longer than V. VI is about one-fourth as 
 long as its spur, longer than I, which in turn is longer tlian II. Tjie 
 antennae are longer than the body. Cornicles long, slightly tapering, 
 pale with tip dusky, about eciual in length to the fifth antennal seg- 
 ment and about twice the length of the hind tarsus. Cauda long, 
 conical, and about two-thirds the length of the cornicles, pale with tip 
 dusky. Lateral tubercles are present on the first and seventh abdom- 
 inal segments and on one other of the abdominal segments, in some 
 cases on the second, in otherslon the third, and in others on the fourth. 
 
 Measurements (of specimens mounted in Canadian balsam) : Body 
 length, 1.5 to 1.53 mm.; body width (abdomen), 0.247 mm.; antennae 
 total, 1.445 to 1.734 mm. (av. 1.6082 mm.) ; I, 0.085 to 0.117 mm. 
 (av. 0.0987 mm.) ; II, 0.068 mm.; Ill, 0.408 to 0.467 mm. (av. 0.4335 
 mm.) ; IV, 0.238 to 0.306 mm. (av. 0.272 mm.) ; V, 0.221 to 0.233 mm. 
 (av. 0.224 mm.) ; VI, 0.1105 to 0.119 mm. (av. 0.1169 mm.) ; spur, 
 
A SYNOPSIS OF THE APEIDIDAE 99 
 
 0.408 to 0.45 mm. (av. 0.4186 mm.); cornicles, 0.221 to 0.255 mm. 
 (av. 0.2401 mm.); caiida, 0.15 mm.; hind tarsi, 0.12 to 0.135 mm. 
 (av. 0.1275 mm.). 
 
 Alate viviparous female. — Prevailing color pale to apple green. 
 Head, antennae, thorax, marginal spots on abdomen, cornicles, tip of 
 Cauda, femora, and tarsi all black. Antennae (fig. 289, 290) with the 
 usual primary sensoria on V and VI and the usual accessory sensoria 
 on VI. IV without sensoi'ia and III with from 6 to 11 fairly large 
 circular sccondarj- sensoria, the usual number being 8 (fig. 290). In 
 this character it differs most markedly from the Main specimens, 
 which have from 12 to 18 sensoria on III and from none to several 
 on IV. The antennae are slightly shorter than the body although 
 practically of the same length. Ill is the longest segment, closely 
 followed by the spur, then by IV, V., VI, I and II. Ill and the spur 
 are subequal, or either one or the other may be slightly the longer. 
 In Patch's drawing V is a little longer than IV. In the California 
 specimen IV is always slightly the longer of the two. In all the 
 California specimens the antennal segments are all a little shorter 
 than in the Maine material. Lateral tubercles are present on the pro- 
 thorax ; they are always present on the seventh abdominal segment, 
 and may be present on any of the first few segments of the abdomen 
 as well. In one case they were observed on the second and seventh 
 segments, in another on the second, third, and seventh, in still another 
 on the fourth, fifth, and seventh, and in a fourth case on the first, 
 second, third, fourth, and seventh segments (fig. 292). The wings 
 and venation are normal, with the second branch of the cubitus arising 
 nearer to the tip of the wing than to the base of the first branch (fig. 
 291 ) . However, it is not quite so close to the wing tip as in the Maine 
 specimens. The cornicles (fig. 292) are long and cylindrical. They 
 are equal to or slightly shorter than V, and from one and one-half to 
 two times as long as the hind tarsi. The cauda (fig. 292) is more or 
 less ensiform, about one-half as long as the cornicles, reaching to the 
 tip of the cornicles, and subequal to or slightlj- shorter than the hind 
 tarsi. 
 
 Measurements (of specimens mounted in Canadian balsam) : Body 
 length, 1.53 to 1.65 mm. (av. 1.585 mm.) ; width of thorax 0.697 to 
 0.765 mm. (av. 0.731 mm.), antennae total, 1.568 mm.; I, 0.068 to 
 0.085 mm. (av. 0.0765 mm.) ; II, 0.051 mm.; Ill, 0.331 to 0.408 mm. 
 (av. 0.3644 mm.) ; IV, 0.238 to 0.289 mm. (av. 0.2817 mm.) ; V, 0.221 
 to 0.247 mm. (av. 0.2295 mm.) ; VI base, 0.085 to 0.111 mm. (av. 
 
100 MISCELLANEOUS STUDIES 
 
 0.1015 mm.) ; VI spur, 0.391 mm.; cornicles, 0.204 to 0.2-46 mm. (av. 
 0.2179 mm.) ; cauda, 0.103 to 0.119 mm. (av. 0.1084 mm.) ; hind tarsi, 
 0.136 mm. 
 
 118. Aphis cooki Essig 
 
 Figures 212 to 214 
 
 Essig, Pom. Jour. Eut., vol. 2, p. 323, 1910. Aphis gossypii Glover (desc). 
 Essig, Pom. Jour. Eut., vol. 3, p. 587, 1911 (orig. desc). 
 
 Record. — Citrus sp., Pomona (Essig). 
 
 In 1909, C. H. Varj', county horticultural inspector in Pomona, 
 found a few orange trees heavily infested with this aphid. Prompt 
 control measures were taken and since then it has never again been 
 observed. Essig first tliought it to be Aphis (jossijpii Glover and de- 
 scribed it under that name. Later, however, he found it to be an 
 undescribed species, so named it Aphis cooki n.sp. after Dr. A. J. Cook. 
 
 119. Aphis cornifoliae Fitch 
 
 Fitch, Cat. Homop. N. Y., p. 65, 18,51 (orig. desc). 
 
 Records. — Cornus pu&esccHS, Sanicula menziesii ; San Francisco Bay region 
 (Davidson). 
 
 A species comparing very favorably with this has been taken bj' 
 Davidson a number of times in the San Francisco Bay region. The 
 fall and winter is spent on dogwood, the summer on gambleweed. 
 Davidson writes as follows : 
 
 This aphid [from «S'a»!c«7a] certainly appears to be very close to what I have 
 called (after Gillette) ctirnifoliac. Moreover, I have noticed that the two plants, 
 dogwood and Sanicula, frecjueutly grow near each other and that there appeared 
 to be a migration of alates from the former just about the time there was a 
 migration of the alates to the latter. 
 
 This migration took place the latter jiart of April in 1916. 
 
 120. Aphis crataegifolii Fitch 
 
 Fitch, Cat. Homop. N. Y., p. 66, 1851 (orig. desc). 
 Sanborn, Kan. Univ. Sci. Bull. 3, p. 53, 1904 (desc). 
 Davidson, Jour. Econ. Ent., vol. 3, p. 377, 1910 (list). 
 
 Record. — Crataegus oxycantha; San Jose, Palo Alto (Davidson). 
 
 This lias been reported more or less abundant on hawthorne in the 
 San Francisco Bay region. According to A. C. Baker this is a good 
 and distinct species and not a synonym of Aphis ponii De Geer, as 
 formerly believed. 
 
A SYNOPSIS OF THE APHIDIDAE 101 
 
 12]. Aphis euonomi Fabr. 
 
 Figures 182, 187, 190. 205 to 207, 23G, 237 
 
 Fabricius, Syst. Eiit., p. 736, 1794 (orig. desc). 
 
 Davidson, Jour. Ecoii. Eut., vol. 2, p. 302, 1909. A. rumicis Linn, (list, in 
 
 part?). 
 Essig, Mon. Bull. Cal. Comm. Hort., vol. 4, p. 446, 1915. A. rumicis Linn. 
 
 (list). 
 
 Eecords — AUh-aea rosea, Berkeley, June, 1915; Hijibiscus moscheutos, Berkeley, 
 July, 1915: Moiiteiius hoaria, Berkeley, July, 1915; Mesemiryanthemum equilat- 
 erale, Stanford University, June, 1915 ; Silyhum marianum, Stanford University, 
 July, 1&15: TJrtica holoserica, Menlo Park, San Mateo County, January, 1915: 
 Calendula officinale, Orange, February, 1917: Anthemis sp., Pasadena, April, 1917: 
 Papaver sp., El Cajon, San Diego County, May, 1916 (Aphis papaveris Fabr.?) : 
 Vici^i falta, Stanford University (Davidson), Oxnard (Essig, 1915), Montebello, 
 Los Angeles County, December, 1916, Riverside, January to May 1917 (Apliis 
 fabac Seop.?): Eumex spp., Palo Alto, January, 1912 (Davidson), Stanford 
 University, March. 1912 (Morrison), March, 1915, Ventura County, May, 1917: 
 Phascolus spp., Ventura County, May, 1917 (Aphis rumicis Linn.?). 
 
 There ha.s been a great deal of confusion regarding the identity 
 of this species of aphid, and as yet its synonomj^ is not worked out 
 satisfactorily. The following is offered only provisionally b.v the 
 autlior. The eonmion black aphid has usually been considered as 
 Aphis riuiiici^s Linn., Aphis euonomi Fabr. being taken as a synonym, 
 but according to Gillette, Linnaeus' description calls for an aphid 
 "brassy brown in color, and not black according to the popular opin- 
 ion; and its food plant should be species of Rumcx." He considers 
 the common black species to be Aphis euonomi Fabr., as does Mord- 
 wilko in the European form. The author follows these two aphidol- 
 ogists in placing Aphis rumicis Linn, of American authors (and later 
 European authors) as a synonym of Aphis euonomi Fabr. He (i.e., 
 Gillette) writes, "whether or not it is synonymous with rumicis we 
 are not certain, but we very much doubt this being the case." As 
 long ago as 1894, Osborn and Sirrine (Iowa Agr. Sta., Bull. 26, p. 
 904, 1894) proved that the species which wintered in Iowa on 
 Euoni/mMS migrated to Rumex and other plants in the summer. In 
 California the author has been unable to find it at any time vipon 
 Euonynius, although this is a very common ornamental plant, especi- 
 ally in the vicinity of Riverside. This ma\' be due, however, to the 
 mild winter climate of southern California, which permits plant lice to 
 live throughout the winter, thus not necessitating the laying of eggs. 
 Concerning the identity of the California species the author believes 
 the form described briefly below to be Aphis euonomi Fabr. The 
 
102 MISCELLANEOUS STUDIES 
 
 one following is probably the same species, and is the one described 
 as Aphis papavcris by Fabricius. The species from Vicia faba is 
 probably the species described as Aphis fahae Scop., which may be 
 synonymous with Aphis euonomi Fabr., but again may not be. The 
 author tried a few transfer tests this spring (1917) with the form 
 from Vicia, attempting to colonize it on Hedera helix and on Rumex 
 spp., with negative results. Of course, this does not prove that it will 
 not colonize on these plants, although tlie author has come to the conclu- 
 sion that the Hedera species is entirely different, being Aphis hederae 
 Kalt. Dr. Patch" in her interesting paper on aphid ecology makes 
 the following statement regarding migration tests, which, it seems to 
 the author, it is well to remember when making such tests : 
 
 If an investigator fails in one hundred attempts to colonize thistle with 
 migrants from plum, that will not be a safe reason for him to conclude that he is 
 not working with Aphis cardui, or that this thistle aphid has nothing to do with 
 the leaf deformations of the plum iu the spring. It has been my experience that 
 negative data with a'phids under such conditions are just no data at all. If the 
 structural characters are such as warrant the migration test in the first place, they 
 warrant a patient continuation even in the face of repeated failures. 
 
 On the other hand (and this is a most encouraging and stimulating circum- 
 stance in connection with aphid migration tests), a single success goes a long way 
 to prove the case. Barring complications, a single success is enough, and repe- 
 titions and verifications are needed only as safeguards in that respect. 
 
 The third description is from specimens taken on Rumex spp. and 
 although slightly different from the one considered as Aphis euotimni 
 Fabr., it may be the same, and it may be Aphis runiicis Linn., but of 
 this the author is doubtful. 
 
 In the bean fields of Ventura County, this black bean aphis is very 
 abundant, and often does considerable damage. In May, 1917, the 
 bean plants were just beginning to appear, and as yet were not 
 infested with the aphis. However, the native dock was quite heavily 
 infested. It seems that the aphis lives over the winter on dock and 
 perhaps on other native plants, migrating in the early summer to the 
 beans. Here it lives throughout the summer, returning to dock when 
 the beans have been harvested and the plants plowed under. Horti- 
 cultural Commissioner A. A. Brock, of Ventura County, places great 
 hope in the efficiency of Hippodamia convergens Guerin as a con- 
 trolling factor. In the spring of 1917 he collected a vast number of 
 these ladybird beetles in Sespe Canyon and turned them loose in the 
 bean fields just as the aphids were beginning to appear. At the 
 present time the results are unknown. 
 
 18 Patch, Edith M., Concerning problems in Aphid ecology, Jour. Econ. Ent., 
 vol. 9, pp. 44-51, 1917. 
 
A SYNOPSIS OF THE APHIDIDAE 103 
 
 The following brief description was made from specimens col- 
 lected from the first six host plants listed above, and is the one con- 
 sidered as Aphis euonomi Fabr. 
 
 Alate viviparous female. — Color apparently black, but on close 
 examination it seems that the ground color is a very dark brown, 
 covered with a blackish tinge, with the following parts decidedly 
 black: head, antennae, thoracic lobes, marginal spots and transverse 
 bands on the abdomen, coi"nicles, tarsi, coxae, tips of tibiae, and apical 
 one-half to two-thirds of the middle and hind femora. The tibiae and 
 fore femora are pale, appearing whitish in life. The antennae are 
 shorter than the body, III being the longest segment, followed closely 
 by VI spur. In one case VI spur was slightly longer than III and 
 in another equal to III. In all other specimens III was the longer 
 segment. IV and V are subequal, V usually being slightly the 
 shorter. There are from eleven to twenty-one secondary sensoria on 
 III, of irregular size. These are .scattered along the whole length of 
 the segment, the distal five or six being in a more or less even line. 
 The usual number is about twelve to fourteen. The number of 
 secondary sensoria on IV range from none to seven, the modal number 
 being two. In one specimen only were sensoria absent from IV ; in 
 another, one antenna had seven, the other having two, while in a 
 third, one antenna had five, the other six. When there are more than 
 two or three sensoria, they are all quite small, and can be clearly 
 distinguished only by the higher power of a microscope. Two is the 
 usual number, being located about the middle of the segment. V is 
 usually without secondary sensoria, the primary sensorium being 
 always present, however. lu one specimen the antennae had one or 
 two very small secondary sensoria on V, and in another specimen one 
 antenna had one small sensorium, the other none. The usual primary 
 and accessory sensoria are present on VI base. Lateral abdominal 
 tubercles are always present on the seventh segment, usually on the 
 first, and often on the second, third, fourth, or fifth. There are 
 always at least three pair of these tubercles, and oftentimes more. 
 One specimen had tubercles on the first, second, third, fourth, and 
 seventh segments. The cornicles are black, imbricated, and taper 
 noticeably from base to apex. They are quite constant in length, the 
 variation being not more than 0.05 mm. in all the specimens examined. 
 They are about half as long again as the hind tarsi. The cauda is 
 concolorous with the abdomen, short and conical or ensifonn, and 
 subequal in length to the hind tarsi. The wings are normal, with the 
 typical Aphis venation. 
 
104 MISCELLANEOUS STCVIES 
 
 Measurements: Body length, 1.53 to 1.989 mm. (av. 1.74 mm.); 
 width of thorax, 0.68 to 0.918 mm. (av. 0.765 mm.) ; antennae total, 
 1.122 to 1.36 mm. (av. 1.272 mm.) ; III. 0.289 to 0.425 m.m (av. 0.3648 
 mm.) ; IV, 0.1955 to 0.272 mm. (av. 0.2266 mm.) ; V, 0.187 to 0.221 
 mm. (av. 0.1885 mm.) ; VI, base 0.102 to 0.136 mm. (av. 0.1119 nun.) ; 
 VI, spur 0.289 to 0.357 mm. (av. 0.3145 mm.) ; cornicles, 0.1785 to 
 0.221 mm. (av. 0.2118 mm.) ; cauda, 0.136 to 0.162 mm. (av. 0.14875 
 mm.) ; hind tarsus, 0.136 to 0.152 mm. (av. 0.1372 mm.). 
 
 Specimens taken by the author in May, 1916, on Papaver sp. (cul- 
 tivated poppy) near El Cajon, San Diego County, seem to liim to be 
 Aphis papaveris Fabr. (Clenera Insectorum. p. 303, 1717), and prob- 
 ably' are the same as the above species, although they may be different. 
 There are from thirteen to fifteen irregular seeondai-y sensoria on III 
 as above, but IV and V are without secondarj- sensoria, with one 
 exception, in whicli there was one small sensoriuin near the middle of 
 IV. The eauda is ecpial to the hind tarsi, tlie cornicles being longer, 
 and about the same comparative length as above. The third antennal 
 segment appears to be longer in comparison than above in some speci- 
 mens. Lateral abdominal tubercles are present on the first, third, and 
 seventh abdominal segments. 
 
 Measurements: Body length, 1.486 to 1.908 nun. (av. 1.711 mm.) ; 
 width of thorax, 0.595 to 0.765 mm. (av. 0.68 mm.) ; antennae total, 
 1.224 to 1.343 mm. (av. 1.2878 mm.) ; III, 0.323 to 0.374 mm. (av. 
 0.3536 mm.) ; IV, 0.2125 to 0.22 mm. (av. 0.2193 mm.) ; V, 0.187 to 
 0.204 mm. (av. 0.2024 mm.) ; VI, base 0.102 to 0.119 mm. (av. 0.1054 
 mm.) ; VI, spur 0.255 to 0.34 mm. (av. 0.2992 mm.) ; cornicles, 0.187 
 to 0.221 mm. (av. 0.204 mm.) ; eauda. 0.136 to 0.153 ram. (av. 0.142 
 mm.) ; hind tarsus, 0.119 mm. 
 
 Specimens taken by the author near Montebello, Los Angeles 
 County, in December, 1916, and in Riverside from January to May, 
 1917, on Vicia faha seem to be somewhat different from the fore- 
 going, yet are very nearly identical. Gillette considers that they 
 might possibly be Aphis fabac Scop., which may or may not be 
 synonymous with Aphis eumiomi Fabr. Superficially, the coloring 
 seems to be the same, although on close observation it appears to be 
 a very' dark green in ground color, covered with a blackish tinge. The 
 legs are colored as above, however. 
 
 Specimens from Rume.r appear to have considerably more brown 
 in the ground color than the preceding varieties. Secondary sensoria 
 are located as follows: III, 14 to 24 (av. 18) ; IV, 4 to 7 (av. 5) ; V, 
 
A SYNOPSIS OF THE APHIDIDAE 105 
 
 I to 4 (av. 3). Lateral abdominal tubercles could be found only on 
 the first and seventh segments. 
 
 Alate viviparous female. — Measurements: Body lengtli, 1.768 to 
 2.142 mm. (av. 1.942 mm.) ; width of thorax, 0.782 to 1.054 mm. 
 (av. 0.918 mm.) ; antennae total, 1.445 to 1.581 mm. (av. 1.496 mm.) ; 
 III, 0.357 to 0.408 unu. (av. 0.394 mm.); IV, 0.255 to 0.323 mm. 
 (av. 0.286 mm.) ; V, 0.204 to 0.255 mm. (av. 0.233 mm.) ; VI, base 
 0.136 to 0.153 mm. (av. 0.139 mm.); VI, spur 0.289 to 0.323 mm. 
 (av. 0.306 iimi.) ; cornicles, 0.187 to 0.255 mm. (av. 0.219 mm.) ; cauda, 
 0.136 to 0.17 mm. (av. 0.153 mm.) ; hind tarsus, 0.119 to 0.153 mm. 
 (av. 0.147 mm.). 
 
 Apterous viviparous female. — Measurements : Body length, 2.278 to 
 2.448 mm. (av. 2.3403 mm.) ; antennae total, 1.309 to 1.598 mm. 
 (av. 1.4382 mm.); Ill, 0.306 to 0.408 mm. (av. 0.3502 mm.); IV, 
 0.221 to 0.306 mm. (av. 0.2618 mm.) ; V, 0.206 to 0.255 mm. (av. 0.238 
 mm.) ; VI, ba.se 0.119 to 0.17 mm. (av. 0.1394 mm.) ; VI, spur 0.289 to 
 0.34 mm. (av. 0.306 mm.) ; cauda, 0.17 to 0.204 mm. (av. 0.187 mm.) ; 
 cornicle, 0.255 to 0.323 mm. (av. 0.289 mm.); hind tarsus, 0.153 to 
 0.17 mm. (av. 0.167 mm.). 
 
 122. Aphis frigidae Oestluiul 
 
 Oestlund, Geol. Nat. Hist. Snrv. Minn., vol. U, p. 46, 1886 (orig. desc). 
 Davidson, Jour. Econ. Ent., vol. 7, p. 132, 1913 (desc. stem mother). 
 
 Eecords. — Artemisia californica; Walnut Creek, Contra Costa County (David- 
 son). 
 
 In company with Macrosiphum artemisiae (Fonse.) this species is 
 found on sagebrush in the San Francisco Bay region. Wilson reports 
 it from Oregon, so probably it is distributed along the coast from the 
 bay north. In the course of observations in southern California 
 during a period of two j^ears the author has been unable to find any 
 aphids infesting sagebrush. 
 
 123. Aphis gossypii Glover 
 Figures 192, 193, 215 
 
 Glover, Pat. Off. Ree., p. 62, 1854 (orig. desc.). 
 
 Clarke, Can. Ent., vol. 35, p. 250, 1903 (list). 
 
 Essig, Pom. Jour. Ent., vol. 1, p. 47, 1909. Aphis citri Ashmead (desc.). 
 
 Essig, Pom. Jour. Ent., vol. 3, p. 590, 1911 (desc). 
 
 Cook, Mon. Bull. Cal. Comm. Hort., vol. 1, p. 65, 1912 (list). 
 
 Carnes, Mon. Bull. Cal. Comm. Hort., vol. 1, p. 398, 1912 (list). 
 
 Weldon, Mon. Bull. Cal. Comm. Hort., vol. 2, p. 597, 1913 (list). 
 
 Davidson, Mon. Bull. Cal. Comm. Hort., vol. 6, p. 65, 1917 (note). 
 
106 MISCELLANEOUS STUDIES 
 
 Secords. — Cucumis spp. ; Newcastle, Placer County, Watsonville, Santa Cruz 
 County (Clarke); Imperial County (Weldon) ; San Diego County, June, 1916: 
 Cucurbita spp.; Alpine, San Diego County, June, 1916: Citrus spp.; Santa Paula, 
 Claremont (Essig), Acampo, San Joaquin County (Carnes), San Diego, March, 
 1916 (E. R. McLean); Whittier, May, 1917: Eeracleum lanatum ; Berkeley, 
 March, 1915 (Essig): Begonia; Stanford University, February, 1912 (Morrison), 
 Eiverside, January, 1917; Punica granatum, Stanford University, April, 1911 
 (Davidson): Seliantlms; Santa Ysabel, San Diego County, May, 1916: Pcrsea 
 gratissima ; Avondale, San Diego County, August, 1916; Chrysanthemum; 
 Ontario, January, 1917; Esclischoltzia calif ornica ; Ontario, January, 1917: 
 Anthemis spp.; Pasadena, AprU, 1917 (R. E. Campbell): Pyrus spp.; Santa 
 Cruz County (Volck), Nevada County (Norton). 
 
 The melon or cotton aphis is distributed throughout the state and 
 is found on a large number of host plants. On melons it is often a 
 considerable pest, particularly in the Imperial Valley. In the apple 
 sections of Santa Cruz and Nevada counties it often becomes abundant 
 enough upon the young trees to cause considerable damage, according 
 to County Horticultural Commissioners Volok and Norton. In San 
 Diego County the author found an infestation on young avocado trees 
 which was very severe. Oftentimes it becomes quite abundant in 
 nurseries and greenhouses. 
 
 124. Aphis hederae Kalt. 
 
 Kaltenbaeh, Monog. d. Pflanzenlause, p. 89, 1843 (orig. dese. ). 
 
 Davidson, Jour. Econ. Ent., vol. 2, p. 302, 1909. A. rumi-cis Linn, (list in 
 
 part). 
 Essig, Pom. Jour. Ent, vol. 2, p. 335, 1910 (desc). 
 Davidson, Jour. Econ. Ent., vol. 3, p. 376, 1910. A. immicis Linn. (list). 
 
 Becords. — Hedcra helix; Stanford University (Davidson), March, 1912 (Mor- 
 rison); Claremont, Los Angeles County (Essig) ; San Jose, May, 1911 (Davidson, 
 Morrison); Oakland, November, 1916 (Davidson); Berkeley, April, 1915; Lemon 
 Grove, San Diego County, March, 1916; Riverside, October, 1916: Chcnopodium 
 sp., Walnut Creek, Contra Costa County, May, 1915 (Davidson). 
 
 Throughout the San Francisco Bay region and southern Califor- 
 nia a small dark brown to black aphid is often found in colonies on 
 the tender shoots of English ivy. Essig described it as Aphis hederae 
 Kalt., but later it was believed to be Aphis rumici-s Linn. {A. euonami 
 Fabr.). However, a careful study of a large series of specimens of 
 this aphid from ivy and of A. euonmni Fabr. from a number of dif- 
 ferent host plants has convinced the author that they are distinct. 
 Gillette is of the same opinion. Consequently the species from ivy 
 in California is Aphis hederae Kalt. In the author's collection there 
 is a specimen from Chenopodium sp. taken by Davidson that appears 
 to be the same species. The most noticeable difference between this 
 
A SYNOPSIS OF THE APHIDIDAE 107 
 
 and Aphis cucinumi Fabr. is iu the length of the cornicles, which are 
 very much longer in this species. Measurements of specimens of the 
 alates from Oakland, Walnnt Creek, San Jose, and Riverside are 
 herewith given : 
 
 ileasurt'nients : Body length, 1.411 to 1.768 mm. (av. 1.621 mm.) ; 
 width of thorax, 0.714 to 0.782 mm. (av. 0.748 mm.) ; antennae total, 
 1.411 to 1.549 mm. (av. 1.499 min.) ; III, 0.323 to 0.391 mm. (av. 0.365 
 mm.) ; IV, 0.272 to 0.323 mm. (av. 0.2914 mm.) ; V, 0.221 to 0.272 mm. 
 (av. 0.2518 mm.) ; VI, base 0.119 to 0.136 mm. (av. 0.311 mm.) ; VI, 
 spur 0.306 to 0.34 mm. (av. 0.323 mm.) ; cauda, 0.136 mm.; cornicle, 
 0.306 to 0.34 mm. (av. 0.3252 mm.) ; third tarsus, 0.119 to 0.136 mm. 
 (av. 0.1237 mm.). 
 
 It will be seen that tlie cornicles are considerably more than twice 
 as long- as the hind tarsi, in some cases practically tliree times, while 
 in A. euonomi Fabr., the.v are scarcely twice as long as the hind tarsi. 
 In A. euonomi Fabr. the hind tarsi are longer than the base of VI, 
 while the cornicles are shorter than VI spur. In A. hederae Kalt. VI 
 spur and the cornicles are subequal or on the average the cornicles are 
 very slightly longer, while VI base and the hind tarsi are also sub- 
 equal, the tarsi being shorter on the average. The secondary sensoria 
 in A. hederae Kalt. are small, irregular in size, and are scattered more 
 or less irregularly along III but in a fairly even row along IV and V. 
 They appear very much the same as in A. euonomi Fabr. There are 
 from thirteen to twenty on III, seventeen being the average; from 
 five to nine on IV, seven and eight being the usual number; and 
 usually one on V, although in a few cases there appear to be none. 
 
 125. Aphis heraclei Cowen 
 
 Coiven, Hemip. Colo., p. 120, 1895 (orig. desc). 
 
 Essig, Univ. Calif. Publ. Eutom., vol. 1, p. 339, 1917 (list). 
 
 Record. — Heracleum montezznmum ; Berkeley (Essig). 
 
 Recentlj' Essig reported having taken this species on Heracleum 
 in Berkeley. The author has specimens from Essig, although he has 
 never collected it himself. This is the only report of the species since 
 Cowen 's original report and description. 
 
 126. Aphis houghtonensis Troop? 
 
 Troop, Ent. News, vol. 17, p. 59, 190G (orig. desc). 
 Davidson, Jour. Econ. Ent., vol. 7, p. 132, 1914 (list). 
 
 Record. — Ribes sanguineum; Contra Costa County (Davidson). 
 
108 MISCELLANEOUS STUDIES 
 
 Davidson reported a species of Aphis infesting the terminal leaves 
 of wild currant in the canj'ons of Contra Costa County. He identified 
 it provisionally as this species as he was uncertain. The author is 
 unacquainted with it. 
 
 127. Aphis lithospermi Wilson 
 Wilson, Trans. Am. Ent. Soc, vol. 41, p. 100, 1915 (orig. desc). 
 Becord. — Lithospermum pilosum; California (WUson). 
 
 There is no definite record of this species in California, but it is 
 listed here because Wilson added it to a list of the California Aphi- 
 didae submitted to him by the author. 
 
 128. Aphis maidis Fitch 
 
 Figures 216 to 218 
 
 Fitch, Insects N. Y., vol. 1, p. 318, 1855 (orig. desc). 
 Clarke, Can. Ent., vol. 35, p. 251, 1903 (list). 
 Davidson, Jour. Econ. Ent., vol. 5, p. 408, 1912 (list). 
 
 Records. — Corn; Watsonville, Berkeley (Clarke); San Jose (Davidson); Lake- 
 side, San Diego County, April, 1916; Chula Vista, San Diego County, August, 
 1916: sorghum; Julian, San Diego County, August, 1916 (H. M. Armitage) ; 
 Corona, Eiverside County, September, 1916. 
 
 Onlj' oceasionalh' is this corn aphis found in California, where it 
 infests the ears and tassels and leaves of corn and some of the sor- 
 ghums. Never has it been observed as injurious as is sometimes 
 reported from the middle western states. 
 
 129. Aphis malifoliae Fitch 
 
 Figures 248 to 250 
 
 Fitch, Trans. N. Y. State Agr. Soc, vol. 5, p. 14, 1854 (orig. desc). 
 Clarke, Can. Ent., vol. 35, p. 252, 1903. Aphis sorbi Kalt. (list). 
 Carnes, Mon. Bull. Cal. Comm. Hort., vol. 1, p. 400, 1912. A. sorbi Kalt. 
 
 (list). 
 Weldon, Mon. Bull. Cal. Comm. Hort., vol. 3, p. 188, 1914. A. sorbi Kalt. 
 
 (list). 
 Baker and Turner, Jour. Agr. Res., vol. 7, pp. 321-343, 1916 (complete 
 
 account). 
 
 Records. — Pyrus malus, P. communis ; Central and northern California; Orange 
 County, May, 1917. 
 
A SYNOPSIS OF THE APHIDIDAE 109 
 
 This is one of the most injurious of our California species of Aphis, 
 being found in practically all of the apple-growing regions of tlie 
 state, and in most of them necessitating some control measures. It 
 has been reported on apple and pear in the following counties : Hum- 
 boldt, Orange, Placer, Sacramento, Santa Clara, Shasta, Tehama, 
 Nevada, Inyo, Santa Cruz, and Alameda. Probably it is present 
 wherever apples are grown, with the exception of the southern Cali- 
 fornia districts where it has never been observed. The apple is the 
 primarj- host, and only occasionally has it been taken on pear. In 
 May, 1917, Roy K. Bishop found it in Orange County, this being the 
 first report of it south of the Tehachapi. 
 
 The life histor.y of this Aphis in California is as follows : 
 In the fall and early winter the eggs are laid in the crotches of the 
 twigs. These hatch in the following spring, the exact time depending 
 upon the weather conditions but it is usually as the buds are begin- 
 ning to show green, or as they are beginning to open. The author has 
 observed the young stem mothers on the young buds of the apple in 
 the latter part of March, although he has never been able to find the 
 eggs, either those yet unhatched or those from which the stem mothers 
 have already hatched. Horticultural Commissioner Weatherby of 
 Humboldt County writes that he has found the eggs hatching as early 
 as February 24. He goes on to state that the eggs of Aphis pomi 
 De Geer do not hatch until considerably later. Horticultural Com- 
 missioner Norton of Nevada County has made the following observa- 
 tions : 
 
 The eggs of Aphis sorbi [inalifoliae] are laid on the buds, or sometimes on 
 the spurs close to the buds. At first they are hard to see as they are small and 
 light green, but later they turn to a shiny black, when they can be more readily 
 detected. The young aphids hatch as soon as the buds begin to swell, which time 
 varies with the season. I have found them sometimes as early as the first of 
 March and at other times as late as the middle of April. 
 
 The stem mothei*s feed upon the plant juices through the buds, 
 sometimes appearing on the outer surface of the buds and at other 
 times crawling down into the unfolding leaves, as is the ca.se with 
 Aphis pomi De Geer. In a few weeks these are mature and begin to 
 deposit live young. All of this second generation are apterous females 
 so far as the author has been able to observe. On April 12, 1915, he 
 found several colonies of these aphids in the apple orchard at Stanford 
 University, each colony consisting of a stem mother and several young 
 apterous viviparous females. These females mature in a few weeks 
 
110 MISCELLANEOUS STUDIES 
 
 and a third generation is begun. The most usual place to find the 
 second and third generations is in the curled terminal leaves of the 
 plant. These leaves are curled very similarly to those by the green 
 apple aphis {Aphis pomi De Geer), but they are curled a great deal 
 tighter. Winged females may appear in this third generation, but 
 it is most usual to find them in the fourth. Horticultural Commis- 
 sioner Volck of Santa Cruz County states that he has counted four 
 generations before the summer migration. During May, 1915, the 
 author collected many colonies of this Aphis and placed them in vials 
 in the laboratory. Many others he attempted to colonize on some 
 apple seedlings. Owing to various causes he was unable to make any 
 successful colonizations on the apple trees, one of the chief causes 
 being the destructive work of coccinellid larvae. Also during the 
 first few days of June he was forced to be absent from town and on 
 his return found that the gardener had "cleaned" the trees, for 
 "they were all covered with lice." Until May 25 no alate females 
 had been found, but on that date two appeared in the laboratory. On 
 May 10, 1917, alates were found in Orange County. 
 
 These alate females of the fourth (perhaps sometimes they appear 
 in the third) generation migrate from the apple to some unknown 
 host. At Stanford University in 1915 the migration began about 
 the first of June and continued for some two or three weeks. On 
 June 20 only two or three colonies, each consisting of but a very few 
 individuals, were found where a month before there had been literally 
 hundreds. The curled leaves still hung on the trees and in each 
 curled leaf the moulted skins of the aphid were abundant. From 
 Commissioner Norton of Nevada County comes the statement that he 
 has known the migrants "to leave the trees as early as the middle 
 of J\ine, but the migration usually takes place between the first and 
 the fifteenth of Julj-. Where they go I have never been able to find 
 out, as I have never observed them on any other host plant." 
 
 According to 0. E. Bremner, Horticultural Commissioner of 
 Sonoma County, the migration takes place there during June. This 
 is the same a-s in Santa Clara County. In Orange County in 1917 
 the alate females appeared about the first of May. Migration began 
 almost immediately and continued for two or three weeks. By May 24 
 only a very few aphids remained. This is fully a month earlier than 
 migration takes place north of the Tehachapi. Incidentally the spring 
 of 1917 was exceedingly cool and the summer very late. In normal 
 years one would expect the aphids to leave the apple two or three 
 weeks earlier. 
 
A SYNOPSIS OF THE APHIDIDAE 111 
 
 Tlie summer host plant of this aphid is as yet unknown in Cali- 
 fornia. During June, 1915, the author spent many hours in search 
 of this host plant, hut to no avail. He examined every kind of plant 
 within two or three hundred yards of the apple orchard at Stanford 
 University, but on none was he able to find any aphid that could pos- 
 sibly be the summer form of Aphis malifolme Fitch. Bremner reports 
 having found isolated individuals on pigweed (Amaranthus retro- 
 flexus) in Sonoma County, but believes this to be accidental for he has 
 never observed them to deposit young on this plant. Davidson writes 
 that he has been able to colonize them in the laboratory on the leaves 
 of plantain {Plantago spp.), in fact has been able to have them repro- 
 duce in such large numbers as to kill the plants. On May 28, 1915, 
 the author placed two alate females from apple leaves on each of two 
 specimens of Plantago hirteUa under bell jars in the laboratory at 
 Stanford University. On returning to towai on June 10 he found that 
 the plants were in a dying condition, owing to a lack of proper care 
 during his absence. However, he found many young lice present, all 
 of which were alive and feeding. The adult alate females had already 
 died. By June 16 the lice had moulted once, but then the plants were 
 practically dead. He left Stanford within a few days not to return, 
 so was unable to begin fresh experiments along this line. In his search 
 for the alates in the field he was particularly careful to examine 
 closely every plantain plant in the vicinity, but could find no trace of 
 this aphid on them. Davidson also reports the same lack of success. 
 Consequently, although the alates will deposit young on plaintain 
 in the laboratory it cannot very well be the natural summer host in 
 this state. Baker and Turner have proven that Plantago lancrolata 
 is the summer host in Virginia. W. H. Britain has observed a definite 
 migration to plaintain in Nova Scotia (Proe. Ent. Soc. Nova Scotia, 
 vol. 1, pp. 16-30, 1915). Incidentally he has been able to breed it 
 throughout the summer on apple. In Orange County, in the vicinity 
 of the known infestations, the author was unable to find any plaintain 
 whatsoever. On inquiring of Roy K. Bishop, the county horticultural 
 commissioner, it was learned that plaintain is very scarce in that 
 county, except very near to the coast, and that it is exceedingly doubt- 
 ful if there is any in the vicinity of the known aphid infestations. 
 
 The fall migrants begin to return to the apple some time during 
 the fall and deposit living males and females. From Nevada County 
 comes the report that the migrants return to the apple "between the 
 twentieth of September and the first of October. " Davidson has taken 
 the oviparous females and the alate males on December 5 (1912) at 
 
112 MISCELLANEOUS STUDIES 
 
 Sebastopol ; Morrison has taken the sexes at Stanford University on 
 December 16 (1910) ; Moznette of the Oregon station has taken the 
 migrants as late as the middle of November at Corvallis, Oregon. 
 Consequently, egg laying probably occurs from the middle of October 
 well into December in the various parts of California. Commissioner 
 Norton states: "The first eggs that I have seen were observed about 
 the fifteenth of October. However, they continue egg laying, in 
 favorable years, well along into November." 
 
 The injury caused by this aphid is done entirely in the spring of 
 the year, before the summer migration, and consists in the curling of 
 the terminal leaves. The colonies are found usually in the leaves 
 surrounding a cluster of apples, and although most of the feeding is 
 on the leaves themselves oftentimes they feed upon the fruit. In 
 such a case the fruit (according to Weldon, "Apple Growing in Cali- 
 fornia," Mon. Bull. Cal. Comm. Hort., p. 86, 1915) "is injured to 
 such an extent that it becomes stunted and not only fails to mature, 
 but is distorted so badly that the variety may not be recognizable." 
 In Nevada County, Commissioner Norton reports : ' ' The purple aphis 
 unless controlled lessens tlie apple crop from ten to fifteen per cent. 
 This is a higher percentage, undoubtedly, than is common throughout 
 the state, but it shows how serious the pest may be. 
 
 130. Aphis marutae Oestlund 
 
 Figures 293 to 299 
 
 Oestlund, Minn. Geol. Nat. Hist. Surv., vol. 14, p. 40, 1886 (orig. desc.). 
 
 Mecords. — Silybum marianum; Gros.smont, San Diego County, April, 1916: 
 Centaurea melitensis ; El Cajon, San Diego County, May, 1916. 
 
 In April, 1916, the author observed a small aphid on milk thistle 
 near Grossmont, San Diego County, and later on tacalote in the 
 El Cajon Valley. It infested the smaller leaves, the leaf petioles, and 
 the base of the flowers. Large numbers of ants were in attendance, 
 but it was preyed upon extensively by the larvae and adults of Cocci- 
 nella californica. A considerable number of adults of LyslpMehus 
 testaceipes Cresson were reared from colonies of this aphid. Being 
 unknown to the author specimens were sent to J. J. Davis and E. 0. 
 Essig, both of whom determined the species to be Aphis marutae Oest- 
 lund. Inasmuch as Oestlund 's descriptions are the only ones avail- 
 able, a brief description is given below of specimens taken May 1, 
 1916, on Silyhum marmnum in San Diego County. 
 
A SYNOPSIS OF TEE APHIDIDAE 113 
 
 Alatc viviparous female. — Prevailiug color pale to olive green. 
 Head and prothorax dark olive green, thoracic lobes almost black. 
 Abdomen pale green with marginal spots and patch on dorsum dusky. 
 Legs pale except tarsi, apex of tibiae, and apical two-thirds of femora. 
 Antennae, cornicles, and cauda dusky. Beak pale at base and dusky 
 at tip. 
 
 Head (fig. 293) not quite as long as broad, with a prominent 
 tubercle at apex of front and small but distinct projections from head 
 on inner side of first antennal .segments. Antennae about same lengtli 
 as body or slightly longer or slightly shorter (figs. 294-295). Ill and 
 the spur are about equal or III slightly longer, never shorter than 
 spur. IV about one-half as long as III. V either shorter or equal to 
 IV. VI shorter than V and about one-third as long as spur. I and II 
 subequal and slightl.y shorter than VI. The usual primary sensoria 
 are present on V and VI and the accessory sensoria on VI. Ill is 
 tuberculate and IV is slightly so. IV has from two to six small, 
 circular secondary sensoria and III from eleven to fifteen irregularly 
 placed (fig. 294). The beak reaches considerably beyond the second 
 coxae, in some cases almost to the third. 
 
 The prothorax is without lateral tubercles. The wings are about 
 twice as long as the body with normal venation. The stigmal vein is 
 curved its entire length, the second branch of the cubitus arises about 
 midway between the tip of the wing and the base of the first branch. 
 
 The abdomen is without lateral tubercles in so far as the author 
 can discern. The cornicles (fig. 299) are short and taper slightly 
 from base to apex. They are about equal in length to the third tarsi, 
 are almost one-half as wide at base as long, and about one-third 
 as wide at apex as long. The cauda (fig. 298) is shoi't and blunt 
 (conical) and about two-thirds as long as the cornicles. The anal 
 plate is half-moon-shaped and dusky at its distal edge. 
 
 Measurements (of specimens in Canada balsam) : Body length, 
 0.918 to 1.02 mm. (av. 0.9248 mm.) ; width (thorax), 0.34 to 0.442 mm. 
 (av. 0.4082 mm.) ; antennae total, 0.885 to 1.02 mm. (av. 0.942 mm.) ; 
 I, 0.034 to 0.051 mm. (av. 0.037 mm.) ; II, 0.034 to 0.051 mm. (av. 
 0.048 mm.) ; III, 0.225 to 0.2975 mm. (av. 0.2601 mm.) ; IV, 0.117 to 
 0.17 mm. (av. 0.152 mm.) ; V, 0.1105 to 0.136 mm. (av. 0.1346 mm.) ; 
 VI, 0.068 to 0.102 mm. (av. 0.0833 mm.) ; spur, 0.204 to 0.272 mm. 
 (av. 0.2295 mm.) ; cornicles 0.0850 to 0.119 mm. (av. 0.0978 mm.) ; 
 cauda, 0.0595 to 0.068 mm. (av. 0.0624 mm.) ; hind tarsi, 0.085 to 
 0.102 mm. (av. 0.0901 mm.) ; wing length, 1.921 to 1.955 mm. (av. 
 1.928 mm.) ; wing width, 0.661 mm.; wing expansion, 4.556 mm. 
 
114 MISCELLANEOUS STUDIES 
 
 Apterous viviparous female. — The apterae are quite similar to the 
 alates except that the thorax is not dark, and that the second, third, 
 and basal three-fourths of the fourth antennal segments are pale. 
 There are no secondary sensoria (fig. 296) and no lateral tubercles 
 on prothorax and abdomen (fig. 297). The individuals are slightlj^ 
 larger and the proportions of the antennal segments differ slightly 
 from the alates. The measurements of specimens mounted in Canada 
 balsam are as follows : 
 
 Measurements: Body length, 1.00 to 1.04 mm. (av. 1.026 ram.); 
 width (abdomen), 0.595 to 0.629 mm. (av. 0.6064 mm.) ; antennae 
 total, 0.561 to 0.697 mm. (av. 0.6151 mm.) • III, 0.102 to 0.136 mm. 
 (av. 0.1218 mm.): IV. 0.0765 to 0.1105 mm. (av. 0.0906 mm.); V, 
 0.068 to 0.085 mm. (av. 0.0765 mm.) ; VI, 0.595 to 0.0765 mm. (av. 
 0.068 mm.) : spur, 0.1615 to 0.1785 nnn. (av. 0.1711 mm.) ; cornicles, 
 0.0765 to 0.11 mm. (av. 0.0935 mm.) ; cauda, 0.0595 mm.; hind tarsi, 
 0.102 mm. (Description from nine specimens of apterae). It will 
 be noticed that in the apterae the antennae are but about two-thirds 
 as long as the body, while in the alates they are almost as long as the 
 body. Furthermore, in the apterae the spur of the sixth antennal 
 segment is always longer than III while in the alates it is equal to III 
 at the most, and in many cases shorter. 
 
 131. Aphis medicaginis Koeh 
 
 Figure 189 
 
 Koch, Die Pflanzenlause, p. 94, 1854 (orig. desc). 
 Davidson, Jour. Econ. Ent., vol. 2, p. 302, 1909 (list). 
 Davidson, Jour. Econ. Ent., vol. 3, p. 376, 1910 (list). 
 Essig, Pom. Jour. Ent., vol. 3, p. 527, 1911 (desc). 
 
 Becords. — Medicago hispida; Stanford University (Davidson), April, 1914 
 (K. W. Haegele) : Astragalus leucopsis; Nordhoff, Ventura County (Essig) : Vicia 
 faha, lima bean, Pasadena (E. E. Campbell). 
 
 This small dark Aphis has been found occasionally in California, 
 particularly on alfalfa and beans. Such other plants as loco weed, 
 licorice, sagebrush, locust, and others are said to be hosts. The author 
 has never collected it himself, but lias had access to specimens taken 
 by Essig, Haegele, and Campbell. Davidson has reared the braconid 
 fly, Lysipheebus testaceipes Cresson, from this aphid. 
 
A SYNOPSIS OF THE APHIDIDAE 115 
 
 132. Aphis middletonii Thomas 
 
 Figures 219, 220 
 
 Thomas, 8th Ann. Rep. 111. St. Ent., p. 99, 1S79 (orig. desc). 
 
 Bccords. — AmarantJn^s rctroflcxus; Santa Paula, August, 1911 (Essig) : Ran- 
 tinculus calif orniciis; Julian, Sau Diego County, June, 1916: Hemuonia rudis; 
 Stanford University, 1916 (Ferris) : Hclmnthit.s unituui; Riverside, September, 
 1916. 
 
 In the fall of the year this species is rather eommou on the roots 
 of various plants in California. The individuals are small green 
 aphids, covered with a slight pulvei'ulence. The^' are very similar 
 to Aphis maidis-radicis Forbes, with which they have often been con- 
 fused, and differ particularly in the presence of seeondarj- sensoria 
 on the fourth antennal segment of the apterae. Below are a few 
 descriptive notes taken from specimens mounted in balsam, collected 
 in 1916 in Julian and Riverside, and in 1911 near Santa Paula: 
 
 Alate viviparous female. — Greenish, pruiuose. Head, antennae, 
 tliorax, marginal spots on abdomen, cornicles, cauda, apical one-half 
 femora, apices tibiae, tarsi, and apex of beak, black. Antennae reach 
 to the base of the second abdominal segment ; III being the longest 
 segment, followed by VI spur. IV and V are subequal, VI base 
 slightly shorter. Tlie usual primary and accessory sensoria are pres- 
 ent. Secondary sensoria occur on III and IV (fig. 220). There are 
 nine to twelve on III, and one to four on IV. The average numbers 
 are eight and two respectively. The beak reaches to the third coxae. 
 Prominent lateral tubercles are present on the first and seventh 
 abdominal segments, as well as on the prothorax. The cornicles are 
 short and taper slightly toward the apex. They are subequal in 
 length to the hind tarsi, and very slightly larger than the cauda. The 
 wings are normal, with the second branch of the third discoidal arising 
 nearer to the apex of the wing than to the base of the first branch. 
 
 Measurements: Bod.v length, 1.65 to 1.7 mm. (av. 1.674 mm.); 
 width of tliorax, 0.561 mm.; antennae total, 0.816 to 0.918 mm. (av. 
 0.884 mm.) ; III, 0.204 to 0.255 mm. (av. 0.2338 mm.) ; IV, 0.11 to 
 0.119 mm. (av. 0.1169 mm.) ; V, 0.11 to 0.136 mm. (av. 0.1275 mm.) ; 
 VI, base 0.085 to 0.102 mm. (av. 0.0986 mm.) ; VI, spur 0.204 mm.; 
 cauda, 0.102 mm.; cornicles, 0.1275 to 0.136 mm. (av. 0.1332 mm.) ; 
 hind tarsus, 0.119 to 0.136 mm. (av. 0.1303 mm.) ; wing length, 1.904 
 to 2.38 mm. (av. 2.159 mm.); width, 0.731 to 0.85 mm. (av. 0.815 
 mm.) ; expansion, 4.3 to 5.1 mm. (av. 4.717 mm.). 
 
116 MISCELLANEOUS STUDIES 
 
 Aptermis viviparous female. — These are very similar to the alate 
 females, only slightly larger. The antennae are duskj^ throughout 
 except the base of III. They reach to the base of the first abdominal 
 segment. Ill is the longest segment. VI spur is next, being about 
 two-thirds a.s long. IV, V, and VI base are subequal, with V some- 
 what shorter than the others. The usual primarj- and accessory seu- 
 soria are present on V and VI. Ill has two or three small secondary 
 sensoria located in the apical one-third of the segment. IV has from 
 one to three in the apical one-half. The prothorax and the first and 
 seventh abdominal segments each have a pair of conspicuous lateral 
 tubercles. The cornicles are black and somewhat larger than in the 
 alates, being slightly longer than the hind tarsi. The cauda is a 
 little shorter than the hind tarsi. 
 
 Measurements: Body length, 1.632 to 1.785 mm. (av. 1.708 mm.) ; 
 width of tliorax, 0.748 to 0.85 mm. (av. 0.799 mm.) ; antennae total, 
 0.867 to 0.969 mm. (av. 0.9265 mm.) ; III, 0.2465 to 0.289 mm. (av. 
 0.2635 mm.) ; IV, 0.102 to 0.136 mm. (av. 0.119 mm.) ; V, 0.102 to 
 0.119 mm. (av. 0.1105 mm.) ; VI, base 0.119 mm.; VI, spur 0.1615 to 
 0.187 mm. (av. 0.17 mm.); cornicles, 0.153 to 0.17 mm. (av. 0.1615 
 mm.) ; Cauda 0.119 mm. ; hind tarsus, 0.136 mm. 
 
 133. Aphis mori Clarke 
 Clarke, Can. Ent., vol. 35, p. 2.51, 1903 (orig. desc). 
 Record. — Moms sp., Berkeley (Clarke). 
 
 This is a rather doubtful species, described by Clarke from speci- 
 mens taken on mulberry in Berkeley. Since the original description 
 it has never again been observed. 
 
 134. Aphis neomexicana Ckll. var. pacifica Dvdn. 
 
 Figures 300, 302 
 
 Davidson, Jour. Econ. Ent., vol. 10, p. 293, 1917 (orig. desc. var.). 
 Hecords. — Eibes rubrum ; Walnut Creek, Contra Costa County, and San Jose 
 (Davidson). 
 
 Davidson described this variety from .specimens found curling the 
 leaves of cultivated red currant in Walnut Creek in June, 1915. 
 What he takes to be the same species he had already collected in San 
 Jose in May, 1912. The author has specimens from him, but has never 
 collected any himself. 
 
A SYNOPSIS OF THE APHIDIDAE 117 
 
 135. Aphis nerii Fonse. 
 
 Figures 221, 222 
 
 Boyer de Fonscolombe, Ann. Eut. Soc. France, vol. 10, p. 167, 1841 (orig. 
 
 desc). 
 Davidson, Jour. Econ. Ent., vol. 3, p. 377, 1910 (list). 
 
 Davidson, Jour. Eeon. Ent., vol. 3, p. 377, 1910. A. lutescens Monell (list). 
 Davidson, Pom. Jour. Ent., vol. 3, p. 399, 1911. A. lutescens Monell (list). 
 Essig, Pom. Jour. Ent., vol. 3, p. 401, 1911. A. lutescens Monell (desc). 
 Essig, Pom. Jour. Ent., vol. 3, p. 530, 1911 (desc.) 
 Branigan, Mon. Bull. Cal. Comm. Hort., vol. 4, p. 53, 1915 (list). 
 
 Eecords. — Asclcpias mexicana; Stanford University (Davidson) ; Stanford 
 University, October, 1910 (Morrison) ; Penryn, Placer County (Davidson) ; south- 
 ern California (Essig); Berkeley, July to September, 1915: Nerium oleander; 
 southern California (Essig); Sacramento (Branigan) ; Berkeley, August to Decem- 
 ber, 1915; San Diego, 1916. 
 
 In the late .spring, summer, and early fall milkweeds throughout 
 the state are often seen to be infested with a bright yellow and black 
 aphid. In the fall and early winter this same species is found infest- 
 ing oleanders. Where oleanders are present but no milkweeds this 
 aphid can be found from spring until winter on the oleander, as 
 observed during 1916 in San Diego. 
 
 Heretofore the species on oleander and milkweed have been con- 
 sidered as distinct, the former being called A. lutescens Monell, the 
 latter A. nerii Ponsc. According to a note from J. J. Davis the 
 species on milkweed could not be A. lutescens Monell. Following are 
 extracts from his letters concerning this point: 
 
 I am wondering whether you have ever found winged specimens on Asclepias 
 that do not bear the black markings at the base of the cornicles. All the speci- 
 mens that I have collected and which Mr. Monell has collected in recent years have 
 these black markings at the base of the cornicles in the winged forms. However, 
 in referring to an old note from Mr. Monell, he says that it would seem hardly 
 possible that he could have missed these dark spots if they had been present in 
 the specimens from which he drew his description for Aphis lutesceiis, and re- 
 marks further that he is not sure that he has ever seen A. lutescens alive since he 
 first described it. I am wondering if lutescens is not really asclepiadis of Pass- 
 erini and whether our other common species on Asclepias and Nerium is not 
 nerii Fonsc. 
 
 During the summer of 1915 the author found this species on 
 Asclepias in the Botannical Gardens at the University of California. 
 During July and August it was quite abundant; in fact, it was 
 especially thick on the stems and undersides of the leaves and blossoms. 
 However, in the latter part of Augu.st it seemed to be getting less 
 
118 MISCELLANEOUS STUDIES 
 
 and less numerous. No sign of parasites was present, and the pre- 
 daceons enemies were not more abundant than usual, so a searcli for 
 the cause was made. Within fifty feet of the milkweed plants several 
 oleanders were found and on them was noticed a large yellow species 
 of Aphis. This supposedly was Aphis ncrii Fonsc. In the laboratory 
 the author could find no structural difference whatsoever between this 
 species and the one on Asclcpias, so he continued"to watch them care- 
 fully on the hosts. As the days passed the Asclepias became freer and 
 freer of the infestation, while the Nerium became more and more 
 heavily infested. This continued through September and into October, 
 b}' which time the Asclepias had died down and incidentally no aphids 
 were left. The Nerium was very heavily infested then. This was 
 taken a.s a good proof that these were the same species. Later Essig 
 told tlie author that the summer before (1914) he had made transfer 
 tests in the laboratory of specimens from Asclcpias to Nerium and 
 that they thrived there and bred well. This fact and the observations 
 above mentioned were noted in a letter to Davis. Following is his 
 
 I have your letter relative to Aphis aselepiadis and nerii, and am interested 
 in your observations. In 1914, Theobald described a species under the name of 
 Aphis nigrepes, which he now places as a variety of aselepiadis. He considers 
 nerii as distinct from aselepiadis because the latter lacks the black patches at 
 the base of the cornicles. Passerini 's aselepiadis is entirely different from Fitch 's 
 Aphis aselepiadis. Fitch's name has priority for, as you will notice, it was 
 described in 1851. This being the case, Passerini 's name will have to fall and 
 be replaced by Aphis lutescens of Monell, which according to Mr. Monell 's data 
 does not bear the black patches around the base of the cornicles. 
 
 This would seem to indicate tliat the California species on Asclepias 
 is Aphis nerii Fonsc. and not .1. liifescDis Monell, as brought out by 
 Essig 's experiment and by the author's observation. Consequently 
 this Californian species is Aphis nerii Fonsc, with Aselepias for its 
 summer host and Nerium. for the winter host. 
 
 136. Aphis oenotherae Oestlund 
 
 Oestlund, Minn. Geol. Nat. Hist. Surv., Bull. 4, p. 62, 1887 (orig. desc). 
 Clarke, Can. Ent., vol. 35, p. 252, 1903 (list). 
 
 Record. — Oenothera hectiana; Epilobium sp., Berkeley (Clarke). 
 
 In 190.3 Clarke recorded finding tliis species on primrose and 
 willow herb in Berkeley. Since then it lias not been observed in Cali- 
 fornia. Tlie author has had the opportunity to study specimens from 
 Minnesota, taken by A. C. Maxson. 
 
A SYNOPSIS OF THE APHIDIDAE 119 
 
 137. Aphis oregonensis Wilson 
 
 Wilsou, Traus. Am. Ent. Soc, vol. 41, p. 92, 1915 (orig. desc). 
 Record. — Artemisia tridcntata, California (Wilson). 
 
 Wilson stated to the author that lie had taken this .species in 
 California although he gave no loealit.y or date records. On the 
 strength of his statement it is included among the California aphids. 
 The author has never seen specimens of it. 
 
 138. Aphis persicae-niger Smith 
 
 Figures 223, 224 
 
 Smith, Ent. Am., p. 101, 1890 (orig. desc). 
 
 Clarke, Can. Ent., vol. 3.5, p. 2.52, 1903 (list). 
 
 Gillette, Jour. Econ. Ent., vol. 1, p. 308, 1908 (desc). 
 
 Weeks, Mon. Bull. Cal. Comm. Hort., vol. 1, p. 244, 1912 (list). 
 
 Jones, Mon. Bull. Cal. Comm. Hort., vol. 1, p. 318, 1912 (list). 
 
 Carnes, Mon. Bull. Cal. Comm. Hort., vol. 1, p. 399, 1912 (list). 
 
 Wood, Mon. Bull. Cal. Comm. Hort., vol. 2, p. 570, 1913 (list). 
 
 Secords. — Prunus spp. ; throughout California. 
 
 This species is ordinarily found infesting the tender twigs and 
 leaves of peach in the spring and early summer. Occasionally it is 
 found on nectarine, plum, and cherry. There are two records of its 
 occurrence on cherry known to the author; one in San Jose in May, 
 1912, by Davidson, and one in El Cajon, San Diego County, in May, 
 1916, by the author. Definite reports of its presence on peach come 
 from Los Angeles, Placer, Riverside, San Benito, San Bernardino, 
 San Diego, Santa Clara, and Tehama counties. In May, 1916, the 
 author observed it doing considerable damage to a j^oung peach orchard 
 in the El Ca.ion Valley, San Diego County. Many of the twigs and 
 some of the larger branches were killed back for several inches, due 
 to the ravages of this insect. 
 
 The Hippodaiiila ladybird and the larvae of a syrphid fly were 
 abundant and devouring vast numbers of the aphids. However, it is 
 not often that this appears abundant enough to cause anj- great 
 amount of damage. 
 
 Its life history, although not thoroughly worked out, is interesting. 
 The following brief summary is from Essig:'' 
 
 The insect winters over on the roots of the peach trees, where it may also be 
 found in the summer. The first aphids appear above ground very early in the 
 
 1' Essig, E. 0., Beneficial and injurious insects of California; ed. 2. Suppl. 
 Mon. Bull. Cal. Comm. Hort., vol. 4, pp. 91-92, 1915. 
 
120 MISCELLANEOUS STUDIES 
 
 spring and begin attacking the tender leaflets, shoots and suckers, usually those 
 at the base of the tree or nearest the ground. These first plant lice are all wing- 
 less. As soon as the buds, young fruit, and leaves appear they are promptly 
 attacked, the entire crop often being entirely ruined. The leaves are curled and 
 weakened, while the young fruit is so distorted as to be killed or rendered unfit 
 for market. During the months of April and May winged migratory females 
 appear, which start colonies on other trees. The work continues until about the 
 middle of July, when most of the lice leave the tops and again go to the roots. 
 
 139. Aphis pomi De Geer 
 
 Figures 225 to 227 
 
 De Geer, Memoires, vol. 3, p. 173, 1773 (orig. desc). 
 
 Davidson, Jour. Econ. Ent., vol. 2, p. 301, 1909 (list). 
 
 Davidson, Jour. Econ. Ent., vol. 3, p. 377, 1911 (list). Aphis mali Fabr. 
 
 Weatherby, Mon. Bull. Cal. Comm. Hort., vol. 1, p. 318, 1912 (list). 
 
 Carnes, Mon. Bull. Cal. Comm. Hort., vol. 1, p. 399, 1912 (list). 
 
 Branigan, Mon. Bull. Cal. Comm. Hort., vol. 4, p. 285, 1915 (list). 
 
 Hurdley, Mon. Bull. Cal. Comm. Hort., vol. 4, p. 445, 1915 (list). 
 
 Baker and Turner, Jour. Agr. Res., vol. 5, pp. 955-995, 1916 (complete 
 
 account). 
 
 Records. — Piiriis mains; Crataegus oxycantha; Catalpa sp. ; California. 
 
 In California this species has been reported on apple and haw- 
 thorn {Crataegus sp.) at Stanford University by Davidson and 
 Morrison ; in Humboldt County by Weatherby ; at Santa Rosa by 
 Carnes ; and by others in Orange, Placer, Sonoma, Santa Cruz, San 
 Bernardino, and Monterey counties. Horticultural Commissioner 
 Armitage states that it has never been found in San Diego County, 
 and Horticultural Commissioner Norton verites that it is unknown in 
 Nevada County. These are the only two of the apple growing regions 
 of the state in which it is not known. The author has found it at 
 Stanford University on apple, catalpa, pear, and hawthorn, and at 
 Marysville on catalpa. Gillette lists loquat, quince, and flowering 
 crab as additional hosts. It seems to prefer the apple to other hosts, 
 and it is on the apple that its greate.st injury is done. Gillette states : 
 "Among the apple trees it has its preference. Missouri Pippin seems 
 to be its first choice, while Rome Beauty, Black Twig, Ben Davis, and 
 a few others are second choice, and the Northern Sp.y is scarcely 
 attacked." The fact that the Northern Spy is almost immune is 
 interesting in that this variety is also quite immune to the devastations 
 of the woolly aphis (Eriosonm lanigfra Hausman). 
 
 The life historj^ of this aphid is quite similar to that of many other 
 species, and is as follows: 
 
A SYNOPSIS OF THE APHIDIDAE 121 
 
 The eggs are laid in the fall of the yeai-, probably during the latter 
 part of October, throughout November, and on into December. They 
 are laid for the most part on the smooth bark of the suckers and water 
 sprouts of the newer shoots. The author has found them in the 
 crotches of the twigs and stems where the bark is rougher, but this is 
 not the usual place. These eggs hatch in the spring about the time 
 the buds begin to show green. In California this is usually during 
 March, although some seasons it is as early as the middle of February, 
 depending entirely upon the weather conditions. These stem-mothers 
 at first feed on the young buds, until the latter have opened enough 
 to allow the aphids to crawl down into the curled leaves. Here they 
 feed for two or three weeks, when they mature and begin depositing 
 living young. This second generation consists chiefly of apterous 
 females, whicli mature in from two to four weeks and in turn produce 
 young. The following generations are in large part alate females 
 which migrate to other trees and there form new colonies. The alates 
 are most common at Stanford University during the latter part of 
 May and during the month of June. After June they seem to lessen 
 in number, perhaps due to the predaceous and parasitic enemies. The 
 first alates that the author has found in the spring were taken at 
 Stanford University on April 13, 1914. In the fall, often as early as 
 October, sexual males and females begin to appear, the males being 
 apterous, the females alate. These mate and very soon the female 
 lays its eggs. Egg laying begins usually in the latter part of October, 
 just as the leaves are beginning to fall, and continues into December 
 after the trees are bare. These eggs hatch in the spring into stem 
 mothers, and the life cycle is completed. 
 
 140. Aphis prunorum Dobr. 
 
 Figures 228 to 230 
 
 Dobrowljansky, Zur Biol. d. Blattlause d. Abstbaume u. BUrenstaucher, 1913 
 
 (orig. desc). 
 Patch, Maine Agr. Exp. Sta., Bull. 233, p. 2(52, 19U (desc. note). 
 
 Records. — Prunus domestica; Walnut Creek (Davidson) ; San Francisco, April, 
 1915 (Shinji). 
 
 A species of Aphis, supposed to be this species, has been taken on 
 prune and plum in the San Francisco Bay region. It agrees very well 
 with Dr. Patch's description listed. However, it may prove to be 
 sj^nonymous with Siphocoryne nymphaeae (Linn.). 
 
122 MISCELLANEOUS STUDIES 
 
 141. Aphis pseudobrassicae Davis 
 
 Figure 231 
 
 Davis, Can. Ent., vol. 46, p. 231, 1914 (orig. desc). 
 
 Records. — Brassica spp. ; Walnut Creek (Davidson), San Diego, Eiverside: 
 Saphaiius sp., Riverside, September, 1916, June, 1917: Matthiola annua, Riverside, 
 February to May, 1917. 
 
 Oftentimes in the spring this false cabbage aphis is found in large 
 colonies on radish, mustard, and so forth. Davidson has taken it in 
 the San Francisco Bay region, and the author throughout southern 
 California. The first few times that it was observed by the author 
 colonies of Aphk brassicae Linn, were also abundant. This led the 
 author to doubt its validity, and to undertake some breeding experi- 
 ments. In February, 1917, two colonies were started, each from one 
 alate female. They were followed through three generations, with the 
 result that all tlie individuals proved to be this species. At the same 
 time a colony of Aphis brassicae Linn, was started from one alate. 
 All tiie progeny of this individual proved to be the same. A. pscudo- 
 brassicue Davis differs from A. brassicae Linn, in the following major 
 points ; 
 
 A. pseudobrassicae Davis: A. brassicae Linn.: 
 
 Apterae not puh'erulent. Apterae pulverulent. 
 
 Cornicles of apterae longer than hind Cornicles of apterae shorter than 
 
 tarsi. hind tarsi. 
 
 IV of alates with sensoria. IV of alates without sensoria. 
 
 1-42. Aphis ramona Swain 
 
 Figures 232 to 235 
 
 Swain, Trans. Am. Ent. See., vol. 44, p. 14, 1918 (orig. desc). 
 
 Secords. — Eamona stachyoides ; Nordhoff and Santa Paula, Ventura County 
 (Swain). 
 
 This species has been taken twice in Ventura County by Essig. 
 It was described by the author from the specimens taken by Essig on 
 black sage. 
 
 143. Aphis rubiphila Patch 
 Patch, Maine Agr. Exp. Sta., Bull. 233, p. 269, 1914 (orig. desc). 
 Records. — Subiis spp.; San Jose, May, 1916 (Davidson). 
 
 In the summer of 1916 Davidson found a species of Aphi.s infesting 
 loganberries and blackberries in San Jose, which was determined by 
 Dr. Patch as A. rubiphila Patch. Essig believes this to be a synonym 
 
A SYNOPSIS OF THE APHIDIDAE 123 
 
 of A. gossypii Glover, but as the author has not had an opportunity 
 to study specimens he believes it best to recognize it as a distinct 
 species at present. 
 
 144. Aphis salicicola Thomas 
 Figures 188, 238, 237 
 
 Thomas, 111. Lab. Nat. Hist., Bull. 2, p. 8, 1879 (orig. desc). 
 Williams, Univ. Neb. Studies, vol. 10, p. 139, 1910 (desc). 
 Davidson, Jour. Econ. Ent., vol. 5, p. 408, 1912 (list). 
 
 Records. — Salix laevigata; Berkeley, June, 1915: Salix, sp. ; San Jose (David- 
 son). 
 
 This is an uncommon species, found in the San Francisco Bay 
 region on willow. The individuals are found in large colonies on 
 the terminal shoots and leaves. These colonies consist in large part 
 of apterae, there being but a very few alates. The species is quite 
 easily recognized by the long cornicles and by the very short second 
 branch of the third diseoidal vein. 
 
 145. Aphis sambucifoliae Fitch 
 
 Figure 240 
 
 Fitch, Cat. Homop. N. Y., p. 66, 185 (orig. desc). 
 Sanborn, Kan. Univ. Sci. Bull. 3, p. 52, 1904 (desc). 
 
 Eecords. — Sambucus glauca; Oakland, April, 1915 (Essig) ; Berkeley, July, 
 1915. 
 
 In 1915 this species was taken twice, once by Essig in Oakland 
 and once by the author in Berkeley. This medium-sized black aphid 
 occurs in large colonies on the tender shoots and flower heads of the 
 common elderberry. In southern California the author has examined 
 hundreds of elderberry trees for this form, but has never found it. 
 Only once has he found any aphid on elderberry in the south, and 
 these proved to be Rhopalosiphum persic-ae (Sulz.). 
 
 146. Aphis senecio Swain 
 
 Figures 2, 4, 6, 241 to 245 
 
 Davidson, Jour. Econ. Ent., vol. 2, p. 302, 1909. Aphis sp. (list). 
 Davidson, Jour. Econ. Ent., vol. 3, p. 377, 1910. A. halceri Cowen (list). 
 Davidson, Jour. Econ. Ent., vol. 7, p. 133, 1914. A. hakeri Cowen (list). 
 Swain, Trans. Am. Ent. Soc, vol. 44, p. 16, 1918. 
 
 Eecords. — Abutilon sp. ; Stanford University, February, 1915: Ambrosia 
 psilostachya ; Berkeley, 1915 (Essig): Amsinckia spp. ; Stanford University, 1909 
 (Davidson), 1912 (Morrison); Berkeley, 1915 (Essig): Anthemis spp.; San 
 
124 MISCELLANEOUS STUDIES 
 
 Francisco Bay region, 1914 (Davidson) ; Pasadena, May, 1917 (Eoy B. Camp- 
 bell) : Artemisia spp.; San Francisco Bay region, 1914 (Davidson); Berkeley, 
 1915 (Essig): Aster sp.; San Diego, January, 191G; Ontario, January, 1917: 
 Baccharis pilularis; Berkeley, 1915 (Essig), Stanford University, 1916 (Ferris): 
 Calendula officinale; Berkeley, 1915 (Essig); San Diego, March, 1916; Riverside 
 and Orange, February, 1917: Chrysanthemum sp. ; Berkeley, 1914 (Essig); Octo- 
 ber, 1915 ; Menlo Park, San Mateo County, March, 1915 ; San Diego, January, 
 1916; La Jolla, February, 1916; Ontario, January, 1917: Cytisus proliferus; 
 Berkeley, 1915 (Essig): Gnapholium sp. ; Walnut Creek, 1914 (Davidson): Grin- 
 dclia cuneifolia; Walnut Creek, 1915 (Davidson) : Selianthus anmius ; San Fran- 
 cisco Bay region, 1914 (Davidson): Sumex sp. ; Stanford University, March, 
 1915: Salix sp.; Berkeley, 1915 (Essig): Senecio spp.; Stanford University, 1909, 
 1910, 1914 (Davidson) ; Santa Paula, 1911 (Essig) ; Palo Alto, February, 1915. 
 
 This is a very common species throughout California, occurring 
 on man.y host plants, particularly the Compositae. It is found most 
 commonly in the early spring on asters, marigolds, and chrysanthe- 
 mums in southern California, and on German ivy and amsinckia in the 
 San Francisco Baj' region. For sometime it was believed to be Aphis 
 bakeri Cowen, but its variety of host plants so widely different from 
 tho.se of bakeri, led to its being identified as a distinct species. It is 
 one of the most common in the state, as a glance at the collection 
 records will show. 
 
 1-47. Aphis setariae Thomas 
 
 Figures 246, 247 
 
 Thomas, 111. Lab. Nat. Hist., Bull. 2, p. 5, 1878 (orig. desc). 
 Williams, Univ. Neb. Studies, vol. 10, p. 141, 1910 (desc). 
 
 Record. — Prunus domestica; San Francisco Bay region (Davidson). 
 
 In some parts of the country this plum louse becomes abundant 
 enough to cause serious damage, but it has never been observed to be 
 so in California. Davidson writes that he has found it sparinglj' a few 
 times in the San Francisco Bay region. The author has never collected 
 it, but has had access to specimens from Morrison, taken in Indiana. 
 
 148. Aphis spiraecola Patch 
 
 Patch, Maine Agr. Exp. Sta., Bull. 233, p. 270, 1914 (orig. desc). 
 
 Records. — Spiraea spp.; Stanford University, 1912 (Morrison) ; Walnut Creek, 
 Contra Costa County, 1916 (Davidson). 
 
 In the San Francisco Bay region there is a small aphid very 
 similar to Aphis pomi De Greer found attacking meadowsweet. David- 
 son and Morrison, who have both observed it, believe it to be this 
 species. The following brief descriptive notes are from alate females 
 
A SYNOPSIS OF THE APHIDIDAE 125 
 
 taken b.y Dr. Patch on cultivated spiraea in Orono, Maine. These 
 notes are included here as there is no adequate description of this 
 species, the only ones^* being very meager notes indeed . 
 
 Alate viviparous females. — Body rather long and narrow, head 
 normal with no antennal tubercles. Antennae shorter than body, 
 reaching to about tlie base of the fourth abdominal segment. VI spur 
 the longest segment, followed by III, which is about two-thirds as 
 long. Following III are IV, V, and VI base. The usual primary 
 sensoria are present on V and VI, and the accessory sensoria on VI. 
 The secondary sensoria are fairly large and circular. There are six 
 or seven in an even line along the whole length of III. On IV there 
 may be one or two near the middle, or there may be none.- Prominent 
 lateral tuliercles are present on the protliorax and on the first and 
 seventh abdominal segments. The cornicles are fairly long, slender, 
 and taper slightly toward the apex. They are from one and one-half 
 to two times as long as the hind tarsi, and subequal to or very slightly 
 longer than the cauda. The cauda is fairly long, ensiform, slightly 
 constricted before the tip. The wings are normal, witli the second 
 branch of the third discoidal nearer the apex of tlie wing tlian the 
 base of the first branch. 
 
 Measurements : Body length, 1.19 to 1.33 mm. ; width of thorax, 
 0.544 to 0.561 mm.; antennae total. 0.85 to 0.918 mm.; Ill, 0.17 to 
 0.1785 mm. ; IV, 0.136 to 0.153 mm. ; V, 0.1275 to 0.1445 mm. ; VI, 
 base 0.0935 to 0.102 mm. ; VI, spur 0.238 to 0.255 mm. ; cornicles, 0.1785 
 to 0.187 mm. ; cauda, 0.17 mm. ; hind tarsus, 0.102 mm. ; wing length, 
 1.97 to 2.04 mm. ; width, 0.748 to 0.782 mm. ; expansion, 4.55 mm. ; 
 from base of first branch of third discoidal to wing tip, 0.578 to 0.68 
 mm. : from base of second branch to wing top, 0.17 to 0.255 mm. 
 
 149. Aphis tetrapteralis Cockerell 
 Cockerell, South. Cal. Acad. Sci., Bull. 1, p. 4, 1902 (orig. deso.). 
 Eecord. — Atriplex canescens tctraptera; La Jolla (Cockerell). 
 
 This species has been observed but once, when described by 
 Cockerell. He writes: "It differs fi'om Aphis atriplices Linn, by its 
 smaller size, mode of life, and shorter cornicles. It seems to be 
 related to Aphis monardae Oestlund." In 1916 the author spent 
 considerable time hunting for this species in the vicinity of La Jolla, 
 but in vain. 
 
 '8 Patch, Eilith M., Maine Aphids of the Eose Family. Maine Agr. Exp. Sta., 
 Bull. 23.'!, p. 270, 1914, Aphis spirnecola n.n.; Gillette-, C. P., Plant louse notes. 
 Family Aphididae. Jour. Econ. Ent., vol. 3, p. 404, 1910. Aphis spiraeella Sehout. 
 
126 MISCELLANEOVS STUDIES 
 
 150. Aphis viburnicolens n.sp. 
 
 Eecords. — Viburnum tinus ; Riverside, February to May, 1917; Eedlands, Feb- 
 ruary, 1917; Orange, February, 1917: Laurus rotoundifolia, Riverside, March, 
 1917. 
 
 Ill the early spring there is a small green and black aphid that 
 attacks in great numbers the racemes of laurustinus and laurel in 
 Southern California. In fact, it is so abundant at times as to serioiisly 
 injure the plants by preventing them from flowering. The leaves 
 and buds are verj- stick.y and covered with the sooty mold fungus. 
 During April, 1917, all the aphids left the laurel and laurustinus, but 
 the alternate host has as yet not been observed. Specimens were sent 
 to Gillette and Patch for determination, but neither could identify 
 them. Dr. Patch wrote as follows : 
 
 This insect is not spiraecoJa, a slide of which I am sending you. 
 
 spiraecola sp. 
 
 Cornicles longer than III Cornicles shorter than III 
 
 VI spur longer than III VI spur subequal to III 
 
 VI spur longer than IV and V VI spur subequal to IV and V 
 
 IV subequal to V IV longer than V 
 
 I do not know this species. I do not have spiraeella Schout. for comparison. 
 
 Gillette stated concerning tliis species: "This is a species of Aphis 
 close to, but almost certainly distinct from, spiracdla Schout., and so 
 far as we know, may be new. ' ' 
 
 From this it would appear that the species from laurustinus and 
 laurel is a new species, and it is described herewith as such.'" Cotype 
 specimens are in the author's private collection, in the collection of 
 the University of California in Berkeley, and of the Citrus Experi- 
 ment Station in Riverside. 
 
 Alate viviparovs female. — Prevailing color green. Head and 
 thorax dusky brown to black. Antennae dusky to black. Beak light 
 brown with tip black. Tibiae, femora of fore legs, and basal one-half 
 of femora of middle and hind legs brown ; tarsi, tips of tibiae, tips of 
 fore femora, and apical one-half of middle and hind femora black. 
 Abdomen pale to apple green, sometimes with a few dusky marginal 
 spots. Cornicles and cauda black. 
 
 1" The species reported by Davidson (Jour. Econ. Ent., vol. 3, p. 377, 1910) 
 as Aphis mali Fabr. from Laurus laurustinus (Fibunmm tinus?) and by Essig 
 (Injurious and Beneficial Insects of California, Men. Bull. Cal. Comm. Hort., 
 Supp. vol. 4, p. xlvi, 1915) as Aphis pomi De Geer from laurustinus, are probably 
 this species. 
 
A SYNOPSIS OF THE APHIDIDAE 127 
 
 Head normal, with frontal and antenual tubercles absent. An- 
 tennae short, reaching only to the second abdominal segment. Ill 
 and VI spur subequal ; IV and V subequal and about three-fourths as 
 long as III or VI spur. The usual primary sensoria are present on V 
 and VI, and the accessory sensoria on VI. Secondary sensoria are 
 found on III and IV, from five to nine on the former and from one 
 to four on the latter. Cornicles short, subeylindrical, and tapering 
 from base toward apex. Cauda fairly long, ensiform, with a slight 
 constriction in the middle, the cauda is slightly longer than the hind 
 tarsi, and the cornicles a little longer than the caiula. Lateral tuber- 
 cles are present on the prothorax, and on the first, fourth, and seventh 
 abdominal segments. The cornicles are subequal to IV or V. The 
 hind tarsi are somewhat longt'r than VI base. The wings ai"e fairl,y 
 large, witli regular venation, the second joint of the third diseoidal 
 arising about half way between the tip of the wing and the base of 
 the first joint. 
 
 Measurements: Bodj' length, 1.214 to 1.479 mm. (av. 1.372 mm.) ; 
 width of thorax, 0.476 to 0.578 mm. (av. 0..5338 mm.) ; antennae total, 
 0.733 to 0.918 mm. (av. 0.8925 mm.) ; III, 0.187 to 0.230 mm. (av. 
 0.2067 mm.) ; IV, 0.136 to 0.161 mm. (av. 0.1473 mm.) ; V, 0.119 to 
 0.153 mm. (av. 0.1416 mm.) ; VI, base 0.085 to 0.102 mm. (av. 0.0877 
 mm.) ; VI, spur 0.204 to 0.230 mm. (av. 0.216 mm.) ; cornicles, 0.127 
 to 0.153 mm. (av. 0.1422 mm.) ; cauda, 0.110 to 0.136 mm. (av. 0.1252 
 mm.) ; hind tarsi, 0.102 to 0.119 nun. (av. 0.1023 mm.) ; wing length, 
 1.921 to 2.397 mm. (av. 2.167 mm.) ; width, 0.799 to 0.935 mm. (av. 
 0.8704 mm.) ; expansion, 4.42 to 5.304 mm. (av. 4.875 mm.). 
 
 Aptermis viviparous female. — General color green with the follow- 
 ing dusky to black : head, antennae, apex of beak, cornicles, cauda, 
 distal margin anal plate, tarsi, and tips of tibiae. Legs, except tarsi 
 and tips of tibiae, duskj' brownish green. Antennae reach to the ba.se 
 of the second abdominal segment. The various segments are propor- 
 tionally the same as in the alates. The beak reaches to the distal 
 margin of the first coxae or almost to the apical margin of the third 
 coxae. Lateral body tubercles are present on thesprothorax and first, 
 second, and seventh abdominal segments. Sometkaes they are also 
 present on the third, fourth, or fifth abdominal segments as well. The 
 cornicles and cauda are subequal, each slightly longq^ than the hind 
 tarsi, and of the same form as in the alates. 
 
 Measurements: Body length, 1.326 to 1.462 mm. (av.\1.3685 mm.) ; 
 width of thorax, 0.595 to 0.68 mm. (av. 0.6975 mm.) ; antennae total 
 
128 MISCELLANEOUS STUDIES 
 
 0.731 to 0.782 mm. (av. 0.748 mm.); Ill, 0.153 to 0.187 mm. (av. 
 0.170 mm.) ; IV, 0.119 to 0.136 mm. (av. 0.1224 mm.) ; V, 0.119 mm. ; 
 VI, base 0.085 mm.; VI, spur 0.136 to 0.1995 mm. (av. 0.1632 mm.) ; 
 cornicles, 0.153 to 0.1995 mm. (av. 0.170 mm.) ; caiida, 0.136 to 0.170 
 mm. (av. 0.162 mm.) ; hind tarsi, 0.102 to 0.119 mm. (av. 0.114 mm.). 
 
 151. Aphis yuccae Cowen 
 Figures 303 to 305 
 
 Cowen, Colo. Agr. Exp. Sta., Bull. 31, p. 122, 1895 (orig. dese.). 
 Williams, Univ. Neb. Studies, vol. 10, p. 145, 1910. Aphis yuccicola n.sp. 
 (desc). 
 
 Records. — Tmcc-o moliavensis; Moorpark, Ventura County, April, 1916 (F. M. 
 Trimble); San Diego, May, 1916. 
 
 Ill April, 1916, Horticultural Inspector F. M. Trimble of Ventura 
 County sent the author a few specimens of the alate and apterous 
 viviparous females of this species, taken on Spanish dagger in Moor- 
 park. In the latter part of tlie next month the author found a few 
 apterae on tlie leaves of Spanisli dagger in Golden Hill Park, San 
 Diego. There were only a few individuals present at that time, but 
 there was evidence of an earlier heavy infestation. Following are a 
 few notes to supplement Williams' excellent description of this species. 
 
 Ill is the longest segment of the antennae, followed by VI spur, 
 which is about three-fourths as long. IV is next, being a little over 
 one-half as long as III and about five-sixths as long as VI spur. V 
 is slightly sliorter than IV and is followed closely by VI base, which 
 is about one-half the length of the spur. The usual primary sensoria 
 are present on V and VI and the accessory sensoria on VI (fig. 303). 
 The apterae have no secondary sensoria, while the alates along the 
 whole length of III (fig. 304) have about twenty-five irregularl.y 
 placed sensoria of irregular size. VI is without sensoria. Lateral 
 tubercles are present on the prothorax and on the first and seventh 
 abdominal segments. The cornicles (fig. 305) are long and slightly 
 tapering, being but slightly shorter than tlie spur of the sixth antennal 
 segment and about twice as long as the hind tarsi. The cauda (fig. 
 305) is ensiform or sickle-shaped and about three-fourths as long as 
 the cornicles. In length it is about equal to the fifth antennal seg- 
 ment and one-half again as long as the hind tarsi. 
 
A SYNOPSIS OF THE APHIDIDAE 129 
 
 Alate viviparous females. — Measurments : Body length, 1.78 to 1.9 
 mm. (av. 1.86 mm.) ; width, (thorax), 0.95 mm.; antennae total, 1.38 
 to 1.51 mm. (av. 1.449 mm.) ; III, 0.34 to 0.425 mm. (av. 0.391 mm.) ; 
 IV, 0.238 to 0.273 mm. (av. 0.256 mm.) ; V, 0.212 to 0.229 mm. (av. 
 0.219 mm.) ; VI, 0.136 to 0.17 mm. (av. 0.155 mm.) ; spur, 0.255 to 
 0.306 mm. (av. 0.289 mm.) ; cornicles, 0.255 to 0.2975 mm. (av. 0.275 
 mm.) ; Cauda, 0.2125 to 0.238 mm. (av. 0.225 mm.) ; hind tarsi, 0.153 
 mm.; wing length, 3.06 to 3.4 mm. (av. 3.19 mm.) ; wing width, 1.27 
 to 1.46 mm. (av. 1.338 mm.) ; wing expansion, 7.48 mm. 
 
 30. Genus Toxoptera Kooh. 
 
 Koch, Die PflanzenlUuse, p. 253, 1857. Type Aphis aurantii Fonsc. 
 
 152. Toxoptera aurantii (Fonsc) 
 
 Figures 114, 163, 27G 
 
 Boyer de Fonscolombe, Ann. Ent. Soc. France, vol. 10, 1841. Aphis (orig. 
 
 desc). 
 Essig, Pom. Jour. Ent., vol. 3, p. 601, 1911. T. aurantiae Koch (desc). 
 Davis, XJ. S. Dept. Agr., Bur. Ent., Tech. Ser., Bull. 25, pt. 1, p. 8, 1912. 
 
 Becords. — Citrus spp. ; throughout citrus sections of southern and central Cali- 
 fornia (Essig, author) ; San Jose (Davidson). 
 
 This is the common black louse of the citrus trees, and is found at 
 almost any time of the year on the younger and more tender leaves 
 of various species of Citt-us. It is more or less heavily preyed upon 
 by the braconid fly, Lijstphlebus testaccipes Cresson. In fact, the 
 author ha-s noticed several infestations in which fully ninety-five per 
 cent of the individuals were parasitized. Besides these the sj'rphid 
 flies cause great havoc among colonies. Of these the author has reared 
 Allograpta obliqua Say from a colony taken in the vieinit.y of El 
 Cajon, San Diego County. Never does this species become abundant 
 enough to seriously damage trees, due undoubtedly to the effective 
 work of its predacious and parasitic enemies. Only in the spring 
 are they found to any great extent, although occasionally tliroughout 
 the year small infestation can be noticed. 
 
130 MISCELLANEOUS STUDIES 
 
 31. Genus Hyalopterus Koeli 
 
 Koch, Die Pflanzenlause, p. 17, 1854. Type Aphis arundinis Fabricius (A. 
 pruni Fabr.). 
 
 153. Hyalopterus arimdinis (Fabr.) 
 
 Figures 181, 185, 186 
 
 Fabricius, Ent. Syst., vol. 4, p. 212, 1749. Aphis (orig. desc). 
 
 Clarke, Can. Ent, vol. 35, p. 247, 1903 (list). 
 
 Davidson, Jour. Econ. Ent., vol. 2, p. 303, 1909 (list). 
 
 Davidson, Jour. Eeon. Ent., vol. 3, p. 377, 1910 (list). 
 
 Essig, Mon. Bull. Cal. Comm. Hort., vol. 2, p. 569, 1913 (list). 
 
 Essig, Mon. Bull. Cal. Comm. Hort., vol. 3, p. 624, 1913. A. prunifoliae 
 
 Fiteh (list). 
 Weldon, Mon. Bull. Cal. Comm. Hort., vol. 2, p. 630, 1913 (list). 
 Weldon, Mon. Bull. Cal. Comm. Hort., vol. 3, p. 378, 1914 (list). 
 Patch, Maine Agr. Exp. Sta., Bull. 233, 266, 1914 (desc.). 
 Davidson, Mon. Bull. Cal. Comm. Hort., vol. 6, p. 64, 1917 (note). 
 
 Bccords. — Prumis spp., Phalaria arundiiuicoa. Phragmitcs comnninis, Typha 
 latifolia: central California. 
 
 During the spring and early summer of the year this ' ' mealy-plum 
 louse" is often very abundant on various species of Prunus in the 
 central part of the state, especially in the San Francisco Bay region 
 and the Sacramento Valley. As summer continues all the aphids 
 desert the plum for other host plants, where thej' remain until fall. 
 The summer hosts in California so far known are reed grass, canarj' 
 grass, and tule, or cat-tail rush. In the Santa Clara Valley there is a 
 feeling among the prune growers that this aphid is the cause of the 
 splitting of the prunes, which is often quite extensive. However, this 
 remains to be proven. 
 
 32. Genus Liosomaphis Walker 
 Walker, The Zoologist, p. 1119, 1868. Type AphU bcrberidis Kalt. 
 
 154. Liosomaphis berberidis (Kalt.) 
 
 Figures 184, 251, 252 
 
 Kaltenbach, Monog. d. Pflanzenlause, p. 85, 1843. Aphis (orig. desc). 
 Davis, Ann. Ent. Soc. Am., vol. 1, p. 254, 1908. Rhopalosiphum (desc). 
 Davidson, Jour. Econ. Ent., vol. 3, p. 378, 1910. Rhopalosiphum (list). 
 
 Becords. — Berberis vulgaris: Stanford University (Davidson); February to 
 May, 1915 ; Berkeley, June to August, 1915. 
 
 This species is found throughout the year on the lower sides of 
 the leaves of barberry in the San Francisco Bay region. The apterae 
 are often very abundant, but the alates are always quite scarce. This 
 
A SYNOPSIS OF THE APHIDIDAE 131 
 
 species is similar to species of Rhopalosiphum, particularly in the 
 shape of the cornicles and cauda, but owing to the absence of antennal 
 tuberek'S it falls into the tribe Aphidini instead of Macrosiphini. 
 Hence Walker's genus Liosomaphis is maintained for this species. 
 
 33. Genus Siphocorjoie Passerini 
 Passerini, Gli Afidi, 1860. Type Aphis paxtinacae Linn, {xylostei Schrank). 
 
 There has been much diversity of opinion concerning this genus, 
 some aphidologists considering it as Siphocoriine Passerini, some as 
 Hyadnphis Kirkaldy, and some as a synonym of Rhopalosiphum Koch. 
 This last is incorrect as this is most certainly not a Macrosiphini for 
 the antennal tubercles are lacking. In 1904 Kirkaldy proposed the 
 name Ilijndaphis to replace Siphocoryne, but in the author's opinion 
 this is uncalled for, so he maintains the original name, Siphocoryne 
 Passerini. 
 
 There have been reported from various parts of California eight 
 species of Siphocoryne as follows: caprcae (Pabr. ), conii (Dvdn.), 
 foeniculi (Schrank), nymphaeae (Linn.), pastinacae (Linn.), salicis 
 (Monell), umhellul-ariae (Dvdn.), and xylostei (Schrank). There 
 are, however, really but three species; capreae (Fabr.), nymphaeae 
 (Linn.) and pastinacae (Linn.). According to Gillette,^" S. salicis 
 Monell is a synonym of S. capreae (Fabr.), and xylostei (Schr.) of 
 pastinacae (Linn.). Davidson^^ states that S. conii (Dvdn.) is a 
 synonym of xylostei (Schr.), and therefore it is the same as pastinacae 
 (Linn.). Morrison writes that the specimens Davidson called S. foeni- 
 culi (Sehr.) are capreae (Fabr.), and those he described as Hyadaphis 
 umheUulariae n.sp. are jS'. pastinacae (Linn.). These two species, 
 pastinacae (Linn.) and capreae (Fabr.), have been greatly confused 
 but Gillette-- has worked out their sjoionj-my quite satisfactorilj-. The 
 following key for distinguishing them is from his paper. 
 
 Joints 4, 5, 6, and antennal spur subequal, the spur usually distinctly the 
 longest, cornicles fully three-fourths as long as third joint of the antenna, a small 
 tubercle on the alate form and a large one on the apterous individuals always 
 present capreae 
 
 20 Gillette, C. P., Two Shopalosiphum species and Aphis pulverulens n.sp.. 
 Jour. Econ. Ent., vol. 4, pp. 320-32.5, 1911. 
 
 21 Davidson, W. M., Plant louse notes from California, Jour. Econ. Ent., vol. 7, 
 p. 133, 1914. 
 
 22 Gillette, C. P.. Two Hhopalosiphum species and Aphis pulverulens n.sp., 
 Jour. Econ. Ent., vol. 4, pp. 320-325, 1911. 
 
132 MISCELLANEOUS STUDIES 
 
 Joint 623 of the antenna distinctly shorter than 5, the fourth still shorter 
 and its spur nearly as long as joints 4, 5, and 6 combined, cornicles seldom much 
 exceeding one-half the third joint of the antenna in length, and a supra-caudal 
 tubercle or spine entirely absent pastinacae 
 
 Aphis nymphaeae Linn, has usually been considered by American 
 aphidologists as a species of Rhopalosiphuin. but the presence of lat- 
 eral body tubercles, the short, robust body, and the absence of antennal 
 tubercles place it in the Aphidini rather than the Macrosiphini. 
 Therefore, it must be considered as belonging to this genus. Baker-* 
 has recently recognized it as belonging here. 
 
 Key to Californian Species 
 
 1. A small spine or tubercle present at the distal end of the body just above the 
 
 cauda (figs. 2-55, 2.56) capreae (Fabr.) 
 
 — No supra-caudal tubercle or spine 2 
 
 2. General color pale green. VI spur as long as IV, V and VI base combined. 
 
 Cornicles at most but slightly more than one-half the length of III. 
 
 pastinacae (Linn.) 
 
 — General color dark brown, wine, or black. VI spur not as long as IV, V and 
 
 VI base combined, although longer than any two together. Cornicles and 
 III subequal nymphaeae (Linn.) 
 
 155. Siphocoryne capreae Fabr. 
 
 Fabricius, Ent. Syst., p. 211, 1794. Aphis (orig. desc). 
 Clarke, Can. Ent., vol. 3.5. p. 252, 1903. S. foemculi (Pass.) (list). 
 Davidson, Jour. Eeon., vol. 2, p. 303, 1909. S. salicis Monell (list). 
 Davidson, Jour. Econ. Ent., vol. 3, p. 377, 1910. S. focniculi (Pass.), (list). 
 Davidson, Jour. Eeon. Ent., vol. 3, p. 377, 1910. S. salicis Monell (list). 
 Essig, Pom. Jour. Ent., vol. 3, p. 534, 1911. Eyadaphis pastinacae (Linn.) 
 (desc). 
 
 Records. — Foeniculum vulgare; Berkeley and Newcastle (Clarke), Stanford 
 University (Davidson): Carum spp. ; Cicuta virosa ; Santa Paula, Berkeley 
 (Essig) : Salix laevigata; Santa Paula (Essig), Brea Canyon, Los Angeles County, 
 AprO, 1917; Riverside, May, 1917: Salix nigra; Lakeside, San Diego County, 
 April, 1916: Salix sp., Stanford University (Davidson). 
 
 This species is found more or less abundantly in the spring on the 
 tender shoots and leaves of willows, migrating in early summer to 
 various species of Umbelliferae. It is more common than S. pastinacae 
 (Linn.), which species is also found on Umbelliferae in the summer, 
 but which passes the fall, winter, and spring on lioneysuckle. 
 
 23 In all the author's specimens. VI is shorter than V, which in turn is shorter 
 than IV, while VI spur is nearly as long as the three together. 
 
 2* Baker, A. C. and Quaintance. A. L. Aphids injurious to orchard fruits, 
 currant, gooseberry and grape, U. S. Dept. Agr., Farmers' Bulletin 804, p. 21, 
 1917. 
 
A SYNOPSIS OF THE APHIDIDAE 133 
 
 156. Siphocorjnie nymphaeae Linn. 
 
 Figure 172 
 
 Linnaeus, Syst. Nat., vol. 2, p. 734, 1735. Aphis (orig. desc,.). 
 Davidson, Jour. Econ. Ent., vol. 3, p. 377, 1910. Bhopalosiplium (list). 
 Essig, Pom. Jour. Ent., vol. 4, p. 793, 1912. Ehopalosiphum (desc). 
 Davidson, Mon. Bull. Cal. Comm. Hort., vol. 6, p. 65, 1917. Ehopalosiphum 
 (note). 
 
 Eecoi-ds. — Polygonum sp., Alisma sp., Potamogetoii sp. ; San Francisco Bay 
 region (Davidson) : Typha latifolia; Santa Paula (Essig), San Francisco Bay 
 region (Davidson): Nymphaea sp.; San Francisco Bay region (Davidson), 
 Fresno, June, 1915: Prunus domestica ; Berkeley, 1916 (Essig). 
 
 This aphid occurs tlirougliout the summer montlis on various semi- 
 aquatic plants, lily, tule, and so forth. In the fall it migrates to 
 plum, where eggs are laid. The first two or three generations in the 
 spring occur ou plum, but about June there is a migration to its sum- 
 mer host plants. So far it has been found in southern California 
 only in Ventura County. 
 
 The species listed as Aphis prunorum Dobr. (see no. 140) may be 
 this species. Essig believes it is, but the author is not certain so does 
 not list it as a synonym. 
 
 157. SiphocorjTie pastinacae Linn. 
 Figures 266 to 270 
 
 Linnaeus, Syst. Nat., p. 451, 1735. Aphis (orig. desc). 
 Davidson, Jour. Econ. Ent., vol. 2, p. 304, 1909. S. xylostei (Schr.) (list). 
 Davidson, Jour. Econ. Ent., vol. 2, p. 304, 1909. S. conii n.sp. (desc). 
 Davidson, Jour. Econ. Ent., vol. 3, p. 377, 1910. S. xylostei (Sehr.) and 
 
 S. conii Dvdn. (list). 
 Davidson, Jour. Econ. Ent., vol. 4, p. 599, 1911. Hyadaphis umbellulariae 
 
 n.sp. (desc). 
 David.son, Pom. Jour. Ent., vol. 3, p. 399, 1911. S. conii Dvdu. (list). 
 Davidson, Jour. Econ. Ent., vol. 7, p. 133, 1914. S. xylostei (Schr.) (list). 
 
 Eecords. — Lonicera sp.; Stanford University (Davidson), Claremont (Essig), 
 Berkeley, AprU, 1915: Umbellularia calif orni-ca; San Jose (Davidson): Conium 
 macalatum ; Stanford University, Penryn, Placer County, and San Jose (David- 
 son). 
 
 This aphid occurs on honeysuckle during the winter and spring, 
 and on various semiaquatic plants in the summer. It has been taken 
 in southern California, in the San Francisco Baj" region, and in the 
 Sacramento Valley. 
 
134 MISCELLANEOUS STUDIES 
 
 34. Genus Myzaphis Van der Goot 
 
 Van der Goot, Ziir Sjstematik der Aphiden, Tijdserift voor Entomologie, 
 vol. 56, p. 96, 1913. Type Aphis rosarum Walker. 
 
 The author believes that this genus of Van der Goot's should be 
 accepted for the two following species: Aphis abietiim Walker and 
 Aphis rosarum Walker. A. rosarum has usually been considered as 
 belonging to the genus Myzus, but the absence of antennal tubercles 
 excludes it from that genus (see figs. 306-308, 313). The cornicles 
 and cauda are not typical of Aphis, and these together with the dis- 
 tinctive frontal tubercle on the head and the absence of lateral bod.y 
 tubercles distinguish it from Aphis. Consequently this genus should 
 be recognized. Following is a key for separating the two known 
 species, both of which occur in California: 
 
 Cornicles slightly clavate (figs. 312, 315), shorter than III. Ill tuberculate, IV 
 
 without sensoria (fig. 309). Found on Hosa spp rosarum (Walker) 
 
 Cornicles cylindrical (fig. 197), equ.il to or longer than III. Ill with 9 to 12 
 rather large secondary sensoria, IV with 1 to 4 (fig. 196). On conifers. 
 
 abietina (Walker) 
 
 • 158. Myzaphis abietina (Walker) 
 
 Figures 196, 197 
 Walker, Ann. Mag. Nat. Hist., vol. 3, p. 301, 1848. Aphis (orig. desc). 
 Wilson, Proc. Ent. Soc. Brit. Columbia, June, 1915 (desc). 
 Record. — Picea excelsa; San Francisco, March, 1915 (Compere). 
 
 The only report of this species in America is that of Wilson, who 
 found it on spruce {Pirca sp.) at Vancouver, British Columbia. On 
 March 26, 1915, Harold Compere of San Francisco took a number of 
 specimens of this species on the twigs of Norway spruce {Picea 
 excelsa) in Golden Gate Park, San Francisco. The specimens are in 
 Essig's and tlie author's collections. 
 
 159. Myzaphis rosarum (Walker) 
 Figures 308 to 317 
 Walker, Ann. Mag. Nat. Hist, vol. 3, 1848. Aphis (orig. desc). 
 Davidson, Jour. Econ. Ent., vol. 3, p. 379, 1910. Myzus (list) 
 Becords. — Rosa spp.; Stanford University (Davidson); Santa Paula (Essig), 
 San Diego, March to July, 1916. 
 
 This species has been reported in the San Francisco Bay region 
 by Davidson and in Santa Paula by Essig. In the Bay region it is 
 rather scarce and is second to Macrosiphum rosae (Linn.) in abun- 
 dance on roses. The author has taken it at Stanford University in 
 1915, and in San Diego several times in 1916. In San Diego in 1916 
 it was by far the most abundant rose-infesting aphid. The author 
 
A SYNOPSIS OF THE APHIDIDAE 135 
 
 has observed it iu such numbers on roses as to cover the undersides 
 of practically all the leaves and the calyx cups of the flowers. Iu some 
 cases the buds were stunted and the flowers unshapely from its effect. 
 In the rose garden of the Panama-California International Exposition 
 these aphids were of considerable importance, necessitating continual 
 care to keep them under control. 
 
 Since there is no adequate description of this species in the Ameri- 
 can aphid literatm-e the author describes it herewith. The following 
 description was drawn from ten specimens of alate and eight of 
 apterae, collected in Santa Paula, Stanford University, and San Diego. 
 
 Alate viviparous female. — Color notes (taken from notes made 
 at the time of collection of specimens at Stanford in March, 1915) : 
 Head, antennae, and thoracic plates black. Abdomen pale apple green 
 with smoky blotch on dorsum. Legs : apical two-thirds of femora 
 smoky, basal one-third pale, tibiae pale except dusky tip, tarsi dusky. 
 Cornicles green (dusky), cauda pale apple green. 
 
 Head is twice as wide as long with a fairly distinct tubercle on the 
 front (fig. 308). Antennal tubercles are lacking or very indistinct. 
 Antennae reach almost to the base of the third abdominal segment 
 (figs. 309, 310). Ill is the longest segment, followed by IV, spur, V, 
 and VI. The spur and IV are practically equal. Of sixteen antennae 
 examined, in three, the spur and IV were equal, in ten, IV was slightly 
 longer than the spur, while in three, the spur was slightly longer than 
 
 IV. V is slightly shorter than the spur, and VI slightly shorter than 
 
 V. However, IV, spur, V, and VI are all almost equal. On V and VI 
 are the usual primary sensoria, and VI the accessory sensoria (fig. 
 300). Ill is tuberculate, being furnished with a large number of 
 irregularly placed secondary sensoria (fig. 309). IV is without any 
 sensoria. The beak reaches almost to the second coxae. 
 
 The prothorax is without lateral tubercles. The wings are normal, 
 being about twice the length of the body. The second branch of the 
 cubitus arises nearer the apex of the wing than the base of the first 
 branch (fig. 311). In but one of seventeen specimens examined was 
 the origin of the second branch of the cubitus nearer the base of the 
 first branch than the tip of the wing. In this specimen the measure- 
 ments were : 0.561 mm. from tip of wing to base of first branch and 
 0.289 mm. from tip of wing to base of second branch. 
 
 The abdomen is long and narrow and is without lateral body 
 tubercles. The cornicles (fig. 312) are long, being but slightly shorter 
 than the third antennal segment, and over twice as long as the hind 
 tarsi. They are slightly clavate on the inner side. The cauda (fig. 
 
136 MISCELLANEOUS STUDIES 
 
 312) is long and pointed (ensiform), being slightly more than one-half 
 as long as the cornicles and about one-half as long again as the hind 
 tarsi. 
 
 Measurements: Body length, 1.19 to 1.41 mm. (av. 1.28 mm.); 
 width of thorax, 0.459 to 0.527 mm. (av. 0.487 mm.) ; antennae total, 
 0.85 to 1.156 mm. (av. 1.027 mm.) ; III, 0.255 to 0.34 mm. (av. 0.317 
 mm.) ; IV, 0.1275 to 0.2295 mm. (av. 0.1768 mm.) ; V, 0.119 to 0.17 
 mm. (av. 0.1365 mm.); VI, 0.085 to 0.119 mm. (av. 0.1095 mm.); 
 spur, 0.119 to 0.204 mm. (av. 0.1695 mm.) ; cornicles, 0.238 to 0.306 
 mm. (av. 0.2574 mm.) ; cauda, 0.136 to 0.187 mm. (av. 0.1588 mm.) ; 
 hind tarsi, 0.119 to 0.136 mm. (av. 0.1205 mm.) ; wing length, 2.482 
 to 2.72 mm. (av. 2.5483 mm.) ; wing width, 0.884 to 1.02 mm. (av. 
 0.9396 mm.) ; wing expansion 5.423 to 5.967 mm. (av. 5.5836 mm.). 
 From tip of wing to base of first branch of cubitus 0.561 to 1.037 mm. 
 (av. 0.8041 mm.) ; from tip of wing to base of second branch of cubitus, 
 0.17 to 0.34 mm. (av. 0.2907 mm.). 
 
 Apterous viviparous female. — Head about as long as broad with a 
 large prominent tubercle on the front, this tubercle being considerably 
 larger than in the alate form ; in some individuals it is fully as large 
 as the first antennal .segment (fig. 313). Antennal tubercles small but 
 distinct, similar to those of the alate. Antennae (fig. 314) short, 
 reaching only to the third coxae. Ill is the longest segment, followed 
 by the spur, IV, VI, and V. These are all subequal, the fornuila of 
 the averages being spur, IV, VI, and V. The formulae for seven 
 antennae are S, VI (V, IV) ; S, (VI, V, IV) ; S, V, IV, VI; S, IV (V, 
 VI) ; S, (IV, V), VI; IV (V, VI, S) ; (S, VI, IV), V. The usual 
 primary sensoria are present, but there are no secondary sensoria. 
 The beak is short, reaching only to the second coxae. 
 
 The prothorax is without tubercles. The thorax is normal, as are 
 the legs. The abdomen is long and narrow, without lateral tubercles, 
 and without long capitate hairs as found in some species of Myziis. 
 The cornicles (fig. 315) are long, cylindrical, and slightly tapering 
 toward the apex, or slightly clavate at apex. They are over twice as 
 long as the third antennal segment and over three times as long as 
 the hind tarsi (fig. 317), and half as long again as the cauda. The 
 Cauda (fig. 316) is long and ensiform, being slightly more than twice 
 the length of the hind tarsi, and about two-thirds the length of the 
 cornicles. 
 
 Measurements: Body length, 1.275 to 1.615 mm. (av. 1.428 mm.) ; 
 width of thorax, 0.493 to 0.748 mm. (av. 0.6375 mm.) ; antennae total, 
 0.544 to 0.731 mm. (av. 0.6239 mm.) ; III, 0.153 to 0.238 mm. (av. 
 
A SYNOPSIS OF THE APHIDIDAE ■ 137 
 
 0.178 mm.) ; IV, 0.068 to 0.119 mm. (av. 0.0855 mm.) ; V, 0.068 to 
 0.102 mm. (av. 0.0833 mm.) ; VI, 0.068 to 0.119 mm. (av. 0.085 mm.) ; 
 spur, 0.085 to 0.136 mm. (av. 0.117 mm.) ; cornicles, 0.306 to 0.442 
 mm. (av. 0.3655 mm.) ; eauda, 0.204 to 0.272 mm. (av. 0.2338 mm.) ; 
 hind tarsi, 0.102 mm. (Note : no color notes were taken of the apterae 
 at the time of collection and as all the specimens were killed in alcohol, 
 dehydrated in xylene and mounted in Canadian balsam, it is impossible 
 to give any color notes.) 
 
 35. Genus Coloradoa Wilson 
 
 Wilson, Ann. Ent. Soc. Am., vol. 3, p. 323, 1910. Type Aphis rufoniaculaia 
 Wilson. 
 
 This genus was described by Wilson in 1910 to contain the species 
 Aphis rufo-rnaculata Wilson. After examining specimens of this 
 species recently, the author is of the opinion that Coloradoa and 
 Myzaphis are synonymous, for there does not seem to be enovigh differ- 
 ence between this species and the two species of Mijzaphis to warrant 
 a separation of genera. However, the author does not feel certain 
 concerning the point, so lists both these genera. Should they later 
 prove to be synonymous, Myzaphis would have to be dropped and 
 replaced by Colorarloa. There is but one species belonging to this 
 genus. 
 
 160. Coloradoa rufomaculata Wilson 
 
 Wilson, Ent. News, vol. 14, p. 261, 1908. Arihis (orig. desc). 
 Secord. — Chrysanthemum, cultivated; Sacramento, April, 1917 (Davidson). 
 
 The author has recently received specimens of this species from 
 Davidson taken on chrysanthemum in Sacramento. 
 
 36. Geuus Cerosipha Del Guercio 
 
 Del Guercio, Nouve relazione agraria di Pirenze, vol. 2, p. IIB, 1909. Type 
 C. passeriniana n.sp. 
 
 ]61. Cerosipha cupressi Swain 
 Swain, Trans. Am. Ent. Soc, vol. 44, p. 19, 1918 (orig. desc). 
 Becords. — Cupressus guadelupensis ; San Diego, 1916; Eiverside, 1917; C. 
 macrocarpa, San Diego, 1916. 
 
 This species, recently described by the author, has been taken by 
 him several times in San Diego and Riverside on blue cypress and 
 Monterey cypress. It is an extremely interesting little aphid, differ- 
 ing considerably from any other species known to the author, both 
 in habits and appearance. Its five-jointed antennae, long cauda, 
 atrophied cornicles, and convexitj'- of abdomen are quite distinctive. 
 
138 • MISCELLANEOUS STUDIES 
 
 Subfamily Pemphiginae Mordwilko" 
 
 Mordwilko, Ann; Mus. Zool. Imp. Acad. Sci. St. Petersburg, vol. 13, pp. 
 362-364, 1908. 
 
 A summary of Mordwilko "s description of tliis subfamily has 
 already been given. The latest and probably the most complete sj'S- 
 tematic work on this subfamilj' that has been done is that of Dr. Albert 
 Tullgren of Stockholm, Sweden, in liis paper, " Aphidologische 
 Studien I" in 1909. Tullgren divides this subfamily into six tribes, 
 viz: Vacunina, Hormaphidina, Mindarina, Pemphigina, Sehizoneu- 
 rina, and Anoeciina. In the tribe Vacunina he places Vacunu Heyden 
 and Glyphina Koch ; in Hormaphidina is the one genus Hmnwmelistes 
 Shimmer; in Mindarina is the one genus Mindarus Koch; in Pem- 
 phigina he places Asiphum Koch, Pachijpappa Koch, Prociphilus 
 Koch, Thecabhis Koch, and Pemphigus Hartig; in Schizoneurina he 
 places the two genera, Schizoneura Hartig, and Tetraneura Hartig; 
 and finalh' in the Anoeciina is found the one genus Anoccia Koch. It 
 can be seen that he uses several of Koch 's genera which have not here- 
 tofore been generally used, namely: Prociphilus Koch, Thecabius 
 Koch, Asiphum Koch, and so forth. Lately there has been a tendency 
 among American aphidologists to accept these genera, and thus to 
 divide up the larger genus Pemphigus into these smaller ones. Mord- 
 wilko in his keys divides this subfamily into four groups, namely: 
 Hormaphidina, Pemphigina, Schizoneurina, and Vacunina. In Hor- 
 maphidina he includes besides the genus Hamamilcstcs Shimmer, the 
 genera Ilurmuphis Osten-Sacken and C< rata phis Lichtenstein. In 
 Pemphigina he includes Pentaphis Heyden, Tetraneura Hartig, 
 Pemphigus Hartig, Aplone.ura Passerini, Rhizoctonus Horvath, and 
 Paraclrtus Heyden. In Schizaneurina he places Lowia Lichtenstein, 
 Colopha Monell, Paehi/pappa Koch, Sehizeineura Hartig, Anoecia Koch, 
 and Mindarus Koch. In Vacunina he includes but the one genus 
 Vacvna Heyden, which he does not separate from Glyphina Koch. 
 
 There is considerable difference in the classifications of these two 
 authors, bat as far as we are concerned here in California our genera 
 are placed about the same by both. Following is a translation of 
 Mordwilko 's key to the groups: 
 
 25 The author has innlcr way a more exhaustive study of tliis subfamily, par- 
 ticularly of the species of Pcniphiflus and Frocipliilus. As this research is still 
 in progress, however, it was tliouglit best to omit any report of it, the author here 
 confining himself merely to the records of the presence of the various species in 
 California. It was hoped to have this study completed at the present time, but 
 the unprecedented conditions of this season have made it necessary to delay further 
 study for the time being. 
 
A SYNOPSIS OF THE AFHIDIDAE 139 
 
 1. Winged forms with a cucurbit-shaped cauda. Nymphs that failed to molt 
 
 with three-joiutcd antennae. Winged forms with three to five-jointed 
 antennae, which are coarsely ringed from the third on. Wingless partheno- 
 genetic females presenting the appearance of the larvae of other families, 
 as of some kinds of Coccidae, or of species of Aleyrodes. Sexual forms 
 with beaks Group Hormaphidina 
 
 — Winged forms without distinct cauda. Nymphs that failed to molt with four 
 
 to five-jointed antennae. Antennae of winged females five-to six-jointed. 
 Sensoria may be found on the third and following joints, often in the 
 form of arches or half rings, but never as complete rings 2 
 
 2. Cubitus [third discoidal vein] of the fore wings simple. Cornicles, which are 
 
 pore or pointlike, present only in some species, and then not in all forms. 
 
 Group Pemphigina 
 
 — Cubitus [third discoidal vein] of fore wings once-branched. Cornicles mostly 
 
 point or pore-like 3 
 
 3. Antennae of winged forms six-jointed. Wings held roof-like when at rest. 
 
 Group Schizoneurina 
 
 — Antennae of winged forms five-jointed. Wings held flat when at rest. 
 
 Group Vacunina 
 
 Grou}) Hormaphidina Mordw. 
 
 Mordwilko, Ann. Mus. Zool. Imp. Acad. Sci. St. Petersburg, vol. 13, pp. 
 364-365, 1908. 
 
 The antennae of the winged forms are 5- to 3-joiuted (?). With the exception 
 of the first two joints they are closely and entirely ringed. Even in the genus 
 Morvwphis O.-S., where the antennae are 3-jointed, they may probably be con- 
 sidered morphologically as of five joints. The wings are held flat at rest. There 
 are four transverse veins on the fore wings, the third of which [third discoidal] 
 is simple. The first two [first and second discoidals] originate at the same point 
 on the subcosta. The hind wings have one or two transverse veins, in the latter 
 case both originating at the same point. The wingless parthenogenetie females 
 on the alternate host plants (for example on birch) are mostly circular in shape, 
 and have small wax tubes around them. Other forms are coccid-like. The sexual 
 forms have beaks. The cornicles are absent. 
 
 This is a description as gjiven by Mordwilko in tiie above mentioned 
 paper. Below is a key to the genera, as given by Mordwilko and by 
 Van der Goot, tlir latti r of whom inelndes in this group tlie two 
 genera IIama>nelestes Shimmer and Cerataphis Licht. 
 
 1. Antennae of winged females plainly five-jointed 2 
 
 — Antennae of winged females only three-jointed Hormaphis O.-S. 
 
 2. Antennae always five-jointed. Front of head always with two little horns. 
 
 Third discoidal once-branched Cerataphis Lichtenstein 
 
 — Antennae of apterous forms three- or four-jointed. Front without horns. 
 
 Third discoidal simple HamameUstes Shim. 
 
140 MISCELLANEOUS STUDIES 
 
 37. Genus Cerataphis Liehtenstein 
 
 Lichtensteiu, Bull. Societe eiit. de France, vol. 2, p. 16, 1882. Type Coccus 
 lataniae Boisd. 
 
 162. Cerataphis lataniae Boisduval 
 
 Boisduval, Ent. Hort., 1867. Coccus (orig. desc). 
 Davidson, Jour. Econ. Ent., vol. 5, p. 404, 1912 (list). 
 Essig, Univ. Calif. Publ. Entom., vol. 1, p. 342, 1917 (list). 
 Secords. — Fern, Stanford University (Davidson); orchid, Oakland (Essig). 
 This coecid-like species has been reported twice in the San Fran- 
 cisco Bay region, by Davidson and by Essig. Morrison and the author 
 have also taken it on the same ferns on which Davidson found it in 
 the Stanford University nursery. 
 
 Group Pemphigina Liehtenstein 
 
 Below is a key to the California genera of this group, adapted 
 from Mordwiiko, Tullgren and Del Guercio. Del Gucrcio described 
 a genus in 1909 for Pemphigus radicicola Essig, which he called 
 Trifidaphis. 
 
 1. Antennae of alate females five-jointed Trifidaphis Del Guar 
 
 — Antennae of alate females si.\-jointed 2 
 
 2. Stem mothers with five-jointed antennae. Wax-gland plates on head always 
 
 present and usually large. Spring and fall migrants with wax-gland plates 
 always on mesothorax and abdomen, and usually on head. Dorsal pores 
 never present 3 
 
 — Stem mothers with four- jointed antennae. Head normally without wax-gland 
 
 plates. Dorsal pores sometimes present. Stem mothers and spring migrants 
 (fundatrix and fundatrigenia) at first live in the same closed galls. 
 
 Pemphigus Hartig 
 
 3. Secondary sonsoria furnished with hairy fringe (Wimperkranz). Wax-gland 
 
 plates generally large. In stem mothers there appear four very large pro- 
 notal wax-gland plates, placed in a transverse row. All plates have a 
 clearly chitinized border. Stem mother and migrants live together. 
 
 Prociphilus Koch 
 
 — Secondary sensoria without hairy fringe (Wimperkranz). Wax-gland plates 
 
 generally small. In stem mothers there are six pronotal plates, of which 
 the four middle ones are arranged in the form of a trapezium. In the 
 winged fall migrants (sexupara) there are also transverse abdominal gland 
 plates, which are without clearly chitinized borders. Stem mothers and 
 spring migrants live in separate galls Thecablus Koch 
 
A SYNOPSIS OF THE APHIDIDAE 141 
 
 38. Genus Trifidaphis Del Guercio 
 
 Del Guercio, Kiv. di patal. veg., vol. 3, p. 20, 1909. Type Pemphigus radi- 
 cicola Essig. 
 
 163. Trifidaphis radicicola Essig 
 
 Essig, Pom. Jour. Ent., vol. 1, p. 8, 1909. Pemphigus (orig. deac). 
 Baker, Pom. Jour. Ent., vol. 1, p. 74, 1909. (Translation of Del Guer- 
 cio 's description of the genus.) 
 Essig, Pom. Jour. Ent., vol. 2, p. 283, 1910 (list). 
 Essig, Pom. Jour. Ent., vol. 4, p. 699, 1912 (list). 
 
 Becords. — Amarantlius retroftexus, Solanum douglasii; Claremont, Santa Paula 
 (Essig). 
 
 Essig described this species from specimens taken on the roots of 
 Amaranthvs rctroflcxus and Solanum douglasii in Santa Paula and 
 Claremont. Later Del Guercio described a new genus for this species 
 based on the venation and the antennae. It seems that the type speci- 
 men of this species had but five-jointed antennae and so of course 
 it could not belong to the genus Pemphigus. On an examination of 
 eight specimens, including the type specimen and seven eotj^pes, the 
 autlior finds that the number of joints in the antennae are variable. 
 The type specimens had both antennae with but five joints. Six 
 antennae had but five joints, six had six distinct joints, and four had 
 five joints in which the division into six could be made out. This 
 divison was in the third joint at about one-third the distance from the 
 apex. Consequently one could say that this species was tj'pically five- 
 jointed, but with some specimens with the third joint divided into two, 
 or it could be said that it was typically six-jointed, but in some speci- 
 mens a reducton occurred through the joining of the third and fourth 
 segments. As but a few specimens were examined the author is not 
 willing to state which is the more common, hence leaves this as a valid 
 genus, although he is of the opinion that this really belongs to the 
 genus Prociphilus Koch. 
 
 39. Genus Pemphigus Hartig 
 
 Hartig, Jahresb. u. d. Eortsohr. d. Forstwiss. u. forstliche Naturk., vol. 1, 
 p. 645, 1837. Type Aphis iursarius Linn. 
 
 This genus is represented in California by three well known 
 species,-" P. betae Doane, P. populi-caulis Fitch, and P. popuU-tram- 
 
 2s There has been taken several times a species forming elongate leaf galls on 
 Populus fremontii, both in the San Francisco Bay region by Davidson and in San 
 Diego County by the author, that structurally seems to be identical with P. populi- 
 caulis Fitch, but its gall is quite distinct, being more or less similar to that of 
 P. betae Doane. Further study may reveal the identity of this form. 
 
142 MISCELLANEOUS STUDIES 
 
 i-erstts Eiley. All of these species, during' at least a part of their life 
 cycles, infest various species of Populus, where thej- form more or less 
 distinctive galls. 
 
 Key to Fundatrigeniae^j 
 
 1. Secondary sensoria present only on III. Galls formed on leaf petioles, with a 
 
 transverse opening on the outside of the curve populi-transversus Riley 
 
 — Secondary sensoria on other segments as well as on III 2 
 
 2. Secondary sensoria on III to VI inclusive. Galls formed by the tAvisting of 
 
 the petiole with an oblique opening on the inside of the curve. 
 
 populi-cauUs Fitch 
 
 — Secondary sensoria on 111 and IV.23 Gall formed on tlie under side of the 
 
 leaves, being more or less elongate and opening on the upper side. 
 
 betae Doane 
 
 164. Pemphigus betae Doane 
 
 Doane, Ent. News., vol. 11, p. 390, 1900 (orig. desc). 
 
 Clarke, Can. Ent., vol. 35, p. 248, 1903 (list). 
 
 Davidson, Jour. Econ. Ent., vol. 2, p. 299, 1909 (list). 
 
 Davidson, Jour. Econ. Ent., vol. 3, p. 372, 1910 (list). 
 
 Williams, Univ. Neb. Studies, vol. 10, p. 92, 1910. P. halsamifcrae n.sp. 
 
 (desc. fundatrigenia). 
 Essig, Pom. Jour. Ent., vol. 4, p. 299, 1912 (list). 
 Maxson, Jour. Econ. Eut., vol. 9, p. 500, 1916 (note). 
 
 Records. — Beta vulgaris: San Francisco Bay region, Monterey County, Sacra- 
 mento Valley. {Eumex spp., Chcnopodiuin spp., etc.?) 
 
 Under the name P. bctac Doane, Clarke, Davidson, and Essig have 
 reported a species of aphid infesting the roots of sugar beets, dock, 
 Chenopodium, and other plants throughout California. 
 
 Originally this species was described from " specimens taken on 
 sugar beet in "Washington, but later-" it was proven that a species 
 forming elongated leaf galls on Poindus halsmiiifera in the spring 
 migrated to beets, and was identical with this species. In 1916 Maxson 
 (cited above) states that his investigations point to the fact that in 
 Colorado there are more than one species of Pemphigus attacking the 
 sugar beet, one of which is this species that forms the elongate leaf 
 gall on poplar in tlie spring, and which is known now as P. hetae 
 Doane. 
 
 ~^ At present only a key to the alate migrants or fundatrigeniae occurring in 
 galls on poplar is given. It is hoped that later, keys to all forms may be formu- 
 lated. At present, however, the life histories of the species are not sufficiently 
 known. 
 
 28 The sexupara or ^late migrants from beets to popUars have secondary sen- 
 soria on III to V inclusive. These form no galls on poplar, however. 
 
 2» Parker, The life history of the sugar-beet root louse, Jour. Econ. Ent., vol. 
 7, pp. 13fi-141, 1914; 
 
 GUlette, Notes on some Colorado aphids having alternate host plants, Jour. 
 Econ. Ent., vol. 8, p. 97, 1915. 
 
A SYNOPSIS OF THE APHIDIDAE 14.1 
 
 These observations of Maxson's together with those made by the 
 author lead to the conclusion that all the reported cases of infestation 
 of beets and other hosts by P. hetae Doane in California do not neces- 
 sarily refer to this species. Never have the fundatrix or fundatrigenia 
 been taken on poplar in California. This strengthens the point that 
 the aphids on beets and other hosts may not all be P. hetae Doane. 
 Further studies and observations will have to be made before this 
 point can be settled, however. 
 
 165. Pemphigus populicaulis Fitch 
 
 Fiteh, Rep. Ins. N. Y., vol. 5, p. 845, 1859 (orig. desc). 
 
 Clarke, Can. Ent., vol. 35, p. 248, 1903 (list). 
 
 Davidson, Jour Eeon. Ent., vol. 2, p. 299, 1909 (list). 
 
 Davidson, Jour. Econ. Ent., vol. 3, p. 372, 1910 (list). 
 
 Davidson, Jour. Econ. Ent., vol. 3, p. 372, 1910. P. popuU-transversus Riley 
 
 (list). 
 Davidson, Pom. Jour. Ent., vol. 3, p. 398, 1911. P. popuU-transversus Riley 
 
 (list). 
 Essig, Pom. Jour. Eut., vol. 4, p. (i99, 1912 (list). 
 Essig, Pom. Jour. Ent, vol. 4, p. 708, 1912 (desc). 
 Davidson, Jour. Eeon. Ent., vol. 8, p. 420, 1915 (se.xuales). 
 
 Records. — Populus fremontii, P. trichocarpa ; from Placer County to San Diego 
 County (Clarke, Davidson, Essig. Morrison, and the author). 
 
 The species infests cottonwoods througliout the state, forming a 
 gall by the twisting of the leaf petiole. The sexuales are found, 
 according to Davidson, under the bark where the eggs are also laid. 
 The author has found the species in San Diego County, having taken 
 the fundatrix, virgogenia, and fundatrigenia in galls in May, 1916, 
 and the dead sexupara at the same time in old galls. These latter 
 probably- died without ever leaving the galls. 
 
 166. Pemphigus populi-transversus Riley 
 
 Riley, U. S. Geog. Geol. Surv., Bull. 5, p. 15, 1880 (orig. desc). 
 Essig, Univ. Calif. Publ. Entom., vol. 1, p. 343, 1917 (list). 
 
 Eecords. — Populus fremontii, Berkeley, September, 1914 (Essig), Riverside 
 September to October, 1916, May to July, 1917. 
 
 This species forms large galls on the leaf petioles of poplar some- 
 what similar to the preceding species, differing in that the opening is 
 on the opposite side of the gall, and is transverse rather than oblique. 
 Essig "s specimens were determined b}^ Gillette, the author's by Max- 
 son. Davidson reported a species under this name from Stanford 
 
144 MISCELLANEOUS STUDIES 
 
 University, but later wrote the author that he was mistaken in his 
 determination, the species being P. populicaulis Fitch instead. 
 
 Just recentl.y the author received specimens of the sexupara of this 
 species from J. R. Parker, Bozeman, Montana. These were taken by 
 S. H. Jones in Port Allen, Louisiana, in September, 1915, on the roots 
 of cabbages. Jones notes that cabbage and other cruciferous plants 
 are the alternate ho.st of this species. This spring the author received 
 a large number of apterae of a species of Pemphigus taken in Orange 
 County on the roots of cabbage. A specific determination of the 
 species was impossible but it may have been this one. 
 
 40. Genus Thecabius Koch 
 Koch, Die Pflanzenlause, p. 294, 1857. Type Pemphigus affini^ Kalt. 
 
 This genus is very similar to Prociphihis, and by some authors, 
 particularly Baker, ^^ is considered as synonymous. However, for 
 present purposes tlie author proposes to retain it for the three species 
 included herewith. 
 
 Key to CALiroBNiAN Species 
 
 1. Antennae short, barely reaching to the metathorax, and not ouc-third as long 
 
 as the body. Ill but slightly longer than VI popuU-monilis Riley 
 
 — ■ Antennae longer, reaching beyond the base of the abdomen, and about one-half 
 as long as the body. Ill considerably longer than VI 2 
 
 2. V and VI with secondary seusoria populi-condupllfoUus Cowen 
 
 ■ — ■ VI without secondary sensoria califomicus Davidson 
 
 167. Thecabius califomicus (Davidson) 
 
 Davidson, Jour. Econ. Ent., vol. 3, p. 372, 1910. Pemphigus ranunculi n.sp. 
 
 (orig. desc). 
 Davidson, Jour. Econ. Ent., vol. 4, p. 414, 1911, renamed Pempliigus cali- 
 
 fornicus Dvdn. 
 Essig, Pom. Jour. Ent., vol. 4, p. 699, 1912. Pemphigus (desc. ala. and 
 
 apt. female). 
 Davidson, Jour. Econ. Ent., vol. 7, p. 127, 1914 (note). 
 
 RecordJi. — Uanunculus califomicus ; San Francisco Bay region (Davidson, Mor- 
 rison, Essig, author) : ? Populus sp. ; Walnut Creek, Contra Costa County, May, 
 1915 (Davidson) : Fraxinus oregona; Walnut Creek (Davidson). 
 
 This aphid is found quite abundantly on the roots and stems of 
 the small California buttercup in the San Francisco Bay region. 
 According to Davidson there is a migration during April from butter- 
 
 ' Baker, A. C, Identity of Eriosome pyri, Jour. Agr. Ees., vol. 5, p. 1118, 1916. 
 
A SYNOPSIS OF THE APHIDIDAE 14.1 
 
 cup to ash. There may be a migration to poplar as well, for the 
 author has specimens that seem to be this species taken by Davidson 
 on poplar. Gillette" places this species as a synonym of T. populi- 
 conduplifoUus Cowen, which attacks both Ranunculus and Populus 
 in Colorado. Davidson, however, is convinced that they are distinct. 
 
 168. Thecabius populiconduplifolius (Cowen) 
 
 Cowen, Colo. Agr. Exp. Sta., Bull. 31, p. 11.5, 1895. Pemphigus (orig. 
 
 desc). 
 Davidson, Jour. Econ. Ent., vol. 3, p. 374, 1910. Pemphigus (li.st). 
 Essig, Pom. Jour. Ent., vol. 4, p. 699, 1912. Pemphigus (list). 
 Gillette, Annals Ent. Soc. Am., vol. 7, p. 61, 1914 (desc. and life history). 
 
 Becord. — Populus trichocarpa; Stanford University (Davidson). 
 
 This species was reported b}^ Davidson on poplar at Stanford 
 University. Since then no further records of its occurrence in the 
 state have been made. In Colorado, Gillette finds that the common 
 buttercup, Ramincuhis sp., is an alternate host and so considers the 
 preceding species as a synonj'm. This may be possible, but it is quite 
 doubtful. 
 
 169. Thecabius populimonilis (Riley) 
 
 Eiley, U. S. Geol. Surv., Bull. 5, p. 13, 1879. Pemphigus (orig. desc). 
 Davidson, Jour. Econ. Ent., vol. 3, p. 374, 1910. Pemphigus (list) 
 Davidson, Pom. Jour. Ent., vol. 3, p. 398, 1911. Pemphigus (list). 
 Essig, Pom. Jour. Ent., vol. 4, p. 699, 1912. Pemphigus (list). 
 Gillette, Ann. Ent. Soc. Am., vol. 6, p. 48.5, 1913 (desc. and life history). 
 
 Becords. — Populus spp. ; Tulare and Placer counties (Davidson); Santa Paula 
 (Essig), Riverside, 1916-1917. 
 
 Throughout central and southern California this species is found 
 on various species of Populus where it forms more or less globular 
 galls on the upper side of the leaves near the margins. In the vicinity 
 of Riverside the j'oung stem mothers began to appear in April (1917). 
 When first observed in September, 1916, nearly all the galls were 
 empty while a few contained alate migrants (sexupara probably). 
 According to Gillette the eggs are laid on the ti'unks of Populus, thus 
 the entire life cycle is passed on the one host plant. This is rather 
 unusual for the Pemphiginae of this section. 
 
 31 Gillette, C. P., Some Pemphiginae attacking species of Populus in Colorado, 
 Ann. Ent. Soc. Am., vol. 7, pp. 61-65, 1914. 
 
146 MISCELLANEOUS STUDIES 
 
 41. Genus Prociphilus Koch 
 Koch, Die Pflanzenlause, p. 279, 1837. Type Aphis bumeliae Schrank. 
 Key to Califoenian Species 
 
 1. Stigma of forewings conspicuously darkened. V witli a few annular secondary 
 sensoria, VI with or without any. Dorsal thoracic wax plates small and 
 
 oval alnifoliae (Williams) 
 
 — Stigma not oonspicuously darkened. V and VI without annular secondary 
 sensoria. Dorsal thoracic wax plates quite large and triangular. 
 
 venafuscus Patch 
 
 170. Prociphilus alnifoliae (Williams) 
 
 Williams, Univ. Neb. Studies, vol. 10, p. 91, 1910. Pemphigus (orig. desc). 
 Baker, Jour. Agr. Ees., vol. 5, p. 1118, 1916 (note). 
 
 Becords. — Heterorncles arbutifoliae; Sespe, Ventura County, March, 1915 (S. 
 H. Essig); May, 191.5 (C. P. Clausen). 
 
 There has been no record of this species from California heretofore, 
 but the author has specimens taken on California holly or Christmas 
 berry in Sespe Canyon during ilarch and May, 1915, by S. H. Essig 
 and C. P. Clausen. 
 
 m. Prociphilus venafuscus Pati'h 
 
 Patch, Ent News, vol. 20, p. 319, 1909. Pemphigus (orig. desc). 
 
 Essig, Pom. Jour. Ent., vol. 3, p. 553, 1911. Pemphigus fraxini-dipetalae 
 n.sp. (orig. desc). 
 
 Essig, Pom. Jour. Ent., vol. 4, p. G99, 1912. Pempliigus fraxini-dipetalae 
 Essig (list). 
 
 C'hilds, Mon. Bull. Cal. Comm. Hort., vol. 3, p. 220, 1914. Pemphig-iis 
 fraxini-dipetalae Essig (list). 
 
 Wilson, Trans. Am. Ent. Soc, vol. 41, p. 85, 1915. Prociphilus fraxini- 
 dipetalae (Essig) (note). 
 
 Davidson, Jour. Econ. Ent., vol. 8, p. 421, 1915. Prociphilus fraxini- 
 dipetalae (Essig) (list). 
 
 Baker, Jour. Agr. Ees., vol. 6, pp. 1118-1119, 1916 (desc. notes, synonymy). 
 
 Becords. — Fraxinus dipetala; Santa Paula (Essig), Contra Costa and Santa 
 Clara counties (Davidson): F. oregona: Oregon (Wilson); Berkeley, April, 1915: 
 Aesculus calif ornicus ; Sacramento (Childs) : Pscudotsvga taxifolia; Oregon (Wil- 
 son). 
 
 Occasionally this verj- large aphid is found infesting the leaves 
 of ash in the San Francisco Bay region and in the mountains of 
 southern California. In early summer it leaves the ash, and according 
 to Wilson infests the roots of Douglas fir in Oregon. At one time 
 Leroy Childs found a few specimens on buckej^e in the vicinity of 
 Sacramento, but it is probable that these were accidental there. 
 
A SYNOFSIS OF THE AFUIDIDAE 147 
 
 Group Schizoneurina Liehteusteiu 
 
 This group as considiTcd by JMordwilko contains tlie following 
 genera: Loicia Lieht., Colopha Monell, Pachypappa Koch, Schizoneura 
 Hartig, Anoecia Koch, and Mindnnta Kocli. Tullgren places in his 
 tribe Schizoneurina the two genera, Schizoneura Hartig, and Tetra- 
 nrura Hartig. Pachypappa Koch he places in his tribe Pemphigina, 
 and he has a separate tribe for each of the genera Anoecia Koch and 
 Mindarus Koch, calling them respectively tribe Anoeciina and tribe 
 Jlindarina. Below is a translation of Mordwilko's key. 
 
 1. Wings laid flat on back when at rest Lbwia Licht. 
 
 — Wings held roof -like when at rest 2 
 
 2. Stigma of forewings trapezoidal in shape, reaching only to the beginning of 
 
 the curve around the end of the wing, never extending to the tip of the 
 wing. Eadial vein originating from the posterior exterior corner of the 
 
 stigma 3 
 
 — ■ Stigma linear, very long, reaching to the wing tip on the front side of the 
 wing, and even following the backward curve of the exterior side of the 
 wing to some extent. Radial vein starting almost at the beginning to the 
 interior edge of the stigma. Sexual forms with beaks Mindarus Koch 
 
 3. Hind wings with one transverse vein Colopha Monell 
 
 — Hind wings with two transverse veins 4 
 
 4. Both transverse veins originating from the same point on the longitudinal 
 
 veins Pachypappa Koch 
 
 — Transverse veins of hind wings originating separately 5 
 
 5. Bodies of apterous and alate forms with little hair, and covered at least on the 
 
 dorsum of the abdomen with waxy powder. Cornicles pore-like (point -like). 
 Sexual forms without beaks Eiiosoma Leach 
 
 — Bodies of apterous and alate forms very hairy and not covered with waxy 
 
 powder or granules (only the stem mothers are weakly pulverulent). Cor- 
 nicles comparatively large, tuberculate (cone-like). Sexual forms with 
 beaks .' Anoecia Koch 
 
 The genera Lowia Licht., Pachypappa Koch, and Anoecia Koch 
 are not represented in California. Colopha Monell and Mindarus 
 Koch are both represented by their type species. It has been proven 
 that Eriosoma Leach has priority over Schizoneura Hartig, so that 
 genus is now known by that name. It is represented in California by 
 three or four species at present. 
 
148 MISCELLANEOUS STUDIES 
 
 42. Genus Colopha Jlonell 
 
 Monell, Can. Ent., vol. 9, p. 102, 1877. Type Byrsocrypta -ulmicola Fitch. 
 
 172. Colopha ulmicola (Fitch) 
 
 Fitch, Eept. Ins. N. Y., vol. 4, p. 63, 1858. Byrsocrypta (orig. desc). 
 Davidson, Jour. Eeon. Ent., vol. 2, p. 299, 1909 (list). 
 Patch, Maine Agr. Exp. Sta., Bull. 181, 196, 1910 (desc). 
 
 Record. — Ulmus sp. ; Stanford University (Davidson). 
 
 Davidson recorded this species from elm at Stanford University in 
 1909. Since then it has not been found again. 
 
 ■43. Genus Eriosoma Leach 
 
 Leach, Trans. Hort. Soc. London, vol. 3, p. 54, 1820. Type Aphis lani- 
 gerum Hausman. 
 
 Until quite recenth' tliis genus has been known as Schizoneura 
 Hartig, but as Baker^- has pointed out, the name Eriosoma has 
 priority-. In California there are three distinct species represented, 
 with a possible fourth. One of these is known only on elm, one on 
 apple (and elm), and one on pear (and elm). 
 
 The following key to the fall migrants is adapted partially from a 
 table of Baker and Davidson.'^ 
 
 1. Body naked except caudal segment. Distal sensoria of V and VI with fringe. 
 
 languinosa (Hartig) 
 
 — Body with some woolly covering. Distal sensoria without fringe 2 
 
 2. Wing veins narrow without brown margins. Ill longer than IV, V, and VI 
 
 together lanigerum (Haus.) 
 
 — Wing veins broad with bro^vnish margins. Ill not so long as IV, V, and VI. 
 
 americana (Riley) 
 
 178. Eriosoma americana (Riley) 
 
 Riley, U. S. Geol. Surv., Bull. 5, p. 4, 1879. Schizoneura (orig. desc). 
 Clarke, Can. Ent., vol. 35, p. 248, 1903. Schizoneura (list). 
 Patch, Maine Agr. Exp. Sta., Bull. 220, p. 268, 1913. Schizoneura (desc 
 note). 
 
 Records. — Ulmus americana; Berkeley (Clarke) ; Walnut Creek, June, 1915 
 (Davidson) ; Palo Alto, May, 1915. 
 
 This leaf-curling aphid of the American elm is found in the San 
 Francisco Baj^ region, and in some cases is very abundant. In May 
 
 32 Baker, A. C, The woolly apple aphis, U. S. Dept. Agr., Office Sec'y, Report 
 101, pp. 11-12, 1915. 
 
 33 Baker, A. C, and Davidson, W. M., Woolly pear aphis. Jour. Agr. Res., 
 vol. 6, p. 358, 1916. 
 
A SYNOPSIS OF THE JPHIDIDAE 149 
 
 and June, 1915, it was especially so on a row of elms on the campus 
 of Stanford University. At that time stem mothers, nj-mphs, and 
 alate spring migrants were present in the galls. By the la.st of June 
 all of these had flown away, leaving the galls empty. According to 
 Baker elm is the only host plant of this species. 
 
 174. Eriosoma lanigerum (Hausman) 
 
 Hausman, Mag. Ins., vol. 1, p. 440, 1802. Aphis (oi-ig. desc). 
 Davidson, Jour. Econ. Ent., vol. 2, p. 299, 1909. ScUizoneum (list). 
 Daviflson, Jour. Econ. Ent., vol. 3, p. 374, 1910. Scldzoneura (list). 
 Baker, U. S. Dept. Agr., Office See 'y, Report 101, pp. 11-16, 1915 (desc. 
 and biology). 
 
 Secord. — Pyrus mains, throughout the state. 
 
 Wlierever apple trees are found in the state this woolly aphis is 
 also found ; the white mas.ses on the trunks and leaves being very con- 
 spicuous, the colonies on the roots more injurious but less conspicuous. 
 In California only the apple lias been found to be attacked. The 
 winter is passed by young nymphs on the roots. As the warmer 
 weather of spring comes these migrate up the trunks and out on 
 the branches and twigs. Here they feed throughout the summer. In 
 tlie fall there is a downward migration, and occasionally a fall 
 migrant is seen. Whether or not these fly to elms as in other parts of 
 the countrj-, is not known, but none have ever been observed on elm. 
 
 175. Eriosoma languinosa (Hartig) 
 
 Hartig, Zeitschr. Ent., vol. 3, p. 359, 1841. Aphis (orig. desc). 
 
 Baker and Davidson, Jour. Agr. Ees., vol. 6, pp. 351-360, 1916. E. pyricola 
 
 n.sp. (desc). 
 Baker and Davidson, Jour. Agr. Roa., vol. 10, pp. 65-74, 1917. E. pyricola 
 
 B. & D. (desc. and biology). 
 Becords. — Pyrus communis, XJlmus campestris ; central California. 
 
 In 1916 Baker and Davidson described a species of Eriosoma that 
 attacks the roots of pears throughout the central part of the state, 
 naming it E. pyricola. Later Davidson found that a species common 
 on IJlmus campestris was the alternate form of this species. This elm 
 form checks up very favorably witli specimens of E. languinosa Hartig 
 from Europe, and is undoubtedly identical. Thus the name pyricola 
 will have to be dropped in favor of langninosa. These elm galls are 
 of a rather peculiar shape, and, as Patch writes, they have the appear- 
 ance of a bonnet. 
 
150 MISCELLANEOUS STUDIES 
 
 44. Genus Mindarus Koch 
 Koch, Die Pflanzenlause, p. 277, 1857. Type M. abittiiius n.sp. 
 
 176. Mindarus abietinus Koch 
 
 Koch, Die Pflanzenlause, p. 278, 1837 (orig. desc). 
 
 Clarke, Can. Ent., vol. 35, p. 248, 1903. Schisoiieura panicola Thos. (list). 
 
 Patch, Maine Agr. Exp. Sta., Bull. 182, p. 242, 1910 (desc). 
 
 Records. — Piniis radiata ; Berkeley, Palo Alto (CHarke) : Abies cilicia ; Stan- 
 ford University, May, 1915. 
 
 Thi.s aphid, easily recognizi'd by tliu extremely long stigma of the 
 fore wings, has been fonnd in the San Francisco Bay region infesting 
 the shoots of Monterey pine and Cilieian fir. 
 
 Group Vacunina ]\Ic»rdwilko 
 
 This group contains but two genera, Vactoia Heyden and (rlyphin-a 
 Koch. Mordwilko does not recognize Ghjphina as distinct from 
 Vacuna, althongh Tnllgi-en does. The latter separates the two genera 
 as follows : 
 
 1. Last abdominal tergite formed into a knob-shaped tail. Integument bare, 
 
 and at most partially set with short lancet-shaped hairs Vacuna Heyd. 
 
 — Last abdominal tergite half-moon shaped, strongly swollen, but scarcely, 
 if at all, separated from the base. Integument set with stiff bristle-like 
 hairs and in apterous females with grain-like elevations ....Glyphina Koch^i 
 
 45. Genus Vacuna Ileyden 
 
 Heyden, Ent. Beitr., vol. 2, p. 289, 1837. Type Aphis drjiopJiila Sflirank. 
 
 177. Vacuna dryophila (Schrank) (?) 
 
 Schrank, Fauna Boica, vol. 1, p. 113, 1801. Aphis (orig. desc). 
 Davidson, Jour. Econ. Ent., vol. 7, p. 128, 1914. Chaitophorus sp. (desc). 
 Davidson, Jour. Econ. Ent., vol. 10, p. 290, 1917 (desc). 
 
 Record. — Quercns lob at a ; Walnut Creek (Davidson). 
 
 Recently Davidson described this species from specimens taken on 
 valley oak in Contra Costa County, where he had observed it for three 
 years. The single alate female he has taken does not appear identical 
 with European specimens of V. dryophila. so he lists the species under 
 this name provisionally. 
 
 34 This genus is not represented in California. 
 
A SYNOPSIS OF THE APHIVIDAE 151 
 
 Subfamily Phylloxerinae Dreyfus 
 
 This subfamily consists of two groups, the Chermisina and the 
 Phylloxerina. Below is a key to these two groups taken from Van 
 der Goot : 
 
 1. Body alwaj's with wax glands. Antennae of adults three-jointed, seemingly 
 five-jointed, with three large sensoria. Gonapophyses appearing as three 
 
 short lips. Sexuales dwarfed, with beak Group Chermisina 
 
 — Body usually without wax glands. Antennae of adults three-jointed, with two 
 large sensoria. Gonapophyses seem to be lacking. Sexuales dwarfed, with- 
 out beak Group Phylloxerina 
 
 Group Chermisina Borner 
 
 This groiip consists of three genera, Pineus Shimmer, Cnapholodes 
 Macq., and Chcrmcs Linn, as it is generally considered, although some 
 authors add more, as Gillcttca Del Guercio and Gucrcioja Mordw. In 
 California but one of these genera is represented, and that hy but 
 two species. 
 
 46. Genus Chermes Linnaeus 
 
 Linnaeus, Syst. Nat., vol. 10, IT.'iS. Type Chermes sambuci Linn. 
 
 178. Chermes cooleyi Gillette 
 
 Gillette, Proc. Acad. Nat. Sci. Phila., vol. 69, p. 3, 1907 (orig. desc). 
 Davidson, Jour. Eeon. Ent., vol. 2, p. 299, 1909. C. coweni Gill. (list). 
 Davidson, Jour. Econ. Ent., vol. 3, p. 372, 1910. C. coweni Gill. (list). 
 Brannigan, Mon. Bull. Cal. Comm. Hort., vol. 4, p. 285, 1915 (list). 
 
 Records. — Pseudotsvga taxifolia, Pinus pinea ; San Francisco Bay region, Sac- 
 ramento Valley. 
 
 This species was first reported in California by Davidson, wlio 
 found it on Douglas fir at Stanford University. Essig lists it from 
 San Francisco, San Mateo, and Santa Clara counties on Douglas fir. 
 In 1915 it was reported twice, once in Sacramento on Douglas fir, and 
 once on Italian stone pine. The author has specimens from E. J. 
 Vosler taken in Sacramento where it was found infesting the twigs 
 and needles of Italian stone pine. Only the apterous females were 
 present, however. 
 
152 MISCELLANEOUS STUDIES 
 
 179. Chermes pinicorticis Pitch 
 
 Fitch, Trans. N. Y. State Agr. Soc, vol. 14, p. 971, 1855. Coccus (orig. 
 
 desc). 
 Storment, 20th Ann. Eep. Illinois St. Ent., appendix, 1898 (desc). 
 Da\'idson, Jour. Econ. Ent., vol. 2, p. 299, 1909 (list). 
 Davidson, Jour. Econ. Ent., vol. 3, p. 372, 1910 (list). 
 
 Becord. — Finns pinaster maritima; Stanford University (Davidson). 
 
 This species, which is uuknown to the author, was reported as 
 present at Stanford University on Pimis pinaster mMritima, where 
 it was so abundant as to sometimes kill the young trees. For a com- 
 plete description see Storment 's jjaper listed above. 
 
 Group Phylloxerina Borner 
 
 There are two genera in tliis tribe, as considered by Borner and 
 Mordwilko, although tlie American authors have generally taken cog- 
 nizance of but one, namely, Phylloxera Boyer. Below is a key from 
 Mordwilko to these genera. 
 
 1. Neither wingless females nor any other forms secreting any -naxj material. 
 
 Phylloxera Boyer 
 — Wingless females secreting a waxy powder Phylloxerina Borner 
 
 -t7. Geiins Phylloxera Boyer 
 
 Boyer de Fon.scolmbe, Ann. Ent. Soc. France, vol. 3, p. 222, 1834. Type 
 P. quercus Boyer. 
 
 ISO. Phylloxera vitifoliae Fitch 
 
 Fitch, Eept. Ins. N. Y., vol. 1, p. 58, 1855 (orig. desc). 
 
 Planchon, C.-R. Acad. Sci. Paris, vol. 67, pp. 588-594, 1868. P. vastatrix 
 
 (desc). 
 Clarke, Can. Ent., vol. 35, p. 248, 1903. P. vastatrix Plan (list). 
 Davidson, Jour. Econ. Ent., vol. 2, p. 299, 1909. P. vastatrix Plan. (list). 
 
 Davidson, Jour. Econ. Ent., vol. 3, p. 372, 1910. P. vastatrix Plan. (list). 
 Records. — Grape; Central and Northern California. 
 
 This is the only species of this genus reported in California. It 
 is one of the most destructive species of plant lice in this section of 
 the country, having in its time practically wiped out the grape indus- 
 try of Santa Clara Valley, and of many other parts of the state. It 
 seems that in California this species infests the roots only of the grape, 
 the forms that produce the leaf galls in the eastern parts of the 
 country not being found here. 
 
A Sl'NOPSIS OF THE APEIDIDAE 153 
 
 48. Genus Phylloxerina Borner 
 
 Borner, Arbeiter aus d. kais. biol. Anst. f. Land- und Porstwirtschaft, 
 vol. 6, pp. i-v, 81-320, 1908. Type Phylloxera salicis Liun. 
 
 Thi.s genus is represented in California by two species, one found 
 on the stems of cottonwood {Pupulus sp.) and the other on the stems 
 and exposed roots of willow {Salix sp.). 
 
 181. Phylloxerina popularia (Pergande) 
 
 Pergrande, Proc. Davenport Acad. Sci., vol. 9, p. 266, 190-t. Phylloxera 
 
 (orig. desc). 
 Davidson, Jour. Econ. Ent., vol. 8, p. 420, 1915. Phylloxera (list). 
 
 Sccords. — Populus spp.; Walnut Creek (Davidson), Merced (Beers). 
 
 The only report of this species in California is the one of Davidson 
 who found it on Populus frcmonti and Populus trichocarpa at "Walnut 
 Creek. On October 14, 1915, A. A. Beers of Merced sent some speci- 
 mens to the author from balm of Gilead {Populus halsamifera) in 
 Merced. These were all apterous females, and were found in great 
 masses of white wax on the smaller branches and twigs. These reports 
 are the onlj^ ones since its original report from Texas and Louisiana 
 bv Pergande. 
 
 182. Phylloxrina salicola (Pergande) 
 
 Pergande, Proc. Davenport Acad. Sci., vol. 9, p. 267, 1904. Phylloxera 
 
 (orig. desc). 
 Davidson, Jour. Econ. Ent., vol. 8, p. 419, 191.5. Phylloxera (list). 
 
 Becords. — Salix spp.; Walnut Creek (Davidson); Pasadena (Smith). 
 
 This species was also reported from Walnut Creek by Davidson 
 on arroyo willow {Salix lasiolepis) where he found it on the stems 
 and exposed roots. On October 13, 1915, A. G. Smith sent the author 
 specimens from an ornamental willow {Salix sp.) in Pasadena, where 
 he found it very abundantly that fall. The specimens were all 
 apterous females, and M'ere found in the midst of considerable masses 
 of wax. This species has only been reported from Illinois, District 
 of Columbia, and California. 
 
154 MISCELLANEOUS STUDIES 
 
 APPENDIX 1 
 Keys to the Genera and Tribes of APHIDIDAE 
 
 BY 
 
 P. VAN DER GOOT, 1913 
 
 SubfamUy APHIDINAE 
 
 1. Antennae seven-jointed (better six-jointed). The last true joint with a dis- 
 
 tinct, more delicate continuation (terminal process). This continuation 
 almost as long as, or even much longer than the last segment; if shorter, 
 the Cauda is distinctly wart-shaped, and the number of rudimentary gona- 
 pophj-ses is always two. Cornicles almost always well formed and clearly 
 projecting. Wings with twice-branched cubitus, only once-branched in 
 exceptional eases 2 
 
 — Antennae mostly six-jointed, the last joint with a short projection, this being 
 
 usually distinctly shorter than half the last segment. Cornicles scarcely 
 projecting, very often only appearing as pores or entirely absent. Wings 
 with a simple or once-branched cubitus 5 
 
 2. Cauda wart-like, occasionall.v not so, or scarcely separated, but then the number 
 
 of rudimentary gonapophyses is always distinctly two 3 
 
 — Cauda sickle-shaped or knobbed, not wart-like, only very seldom absent. Rudi- 
 
 mentary gonapophyses always three Siphonophorina 
 
 3. Cornicles very long, almost cylindrical. Rudimentary gonapophyses tlirce. 
 
 Drepanosiphina 
 
 — Cornicles very sliort, somewhat clubbed. Rudimentary gonapophyses two or 
 
 four 4 
 
 4. Xumber of rudimentary gonapophyses four. Body never with long clubbed 
 
 hairs Chaitophorina 
 
 — Number of rudimentary gonapophyses two. Body often with kiioliI)ed hairs. 
 
 Tarsi always with two pulvillae [Haftlappehen] Callipterina 
 
 5. Cauda wart-like 6 
 
 — Cauda not wart-like, usually absent 7 
 
 6. Anal plate bilobed. Sensoria of alate females linear Hormaphidina 
 
 — Anal plate simple. Sensoria of alate females circular Vacunina p.p. 
 
 7. Cauda distinctly sickle-shaped Mindarina' 
 
 — - Cauda only scarcely or not at all separated 8 
 
 8. Antennae five-jointed. Cornicles very short, only slightly projecting. Body 
 
 without distinct wax gland groups Vacunina p.p. 
 
 — Antennae six-jointed, those of the apterous forms often only four- or five- 
 
 jointed. Cornicles often only pores or entirely lacking. Body often with 
 wax gland 9 
 
 9. Body with long, mostly fine hairs; without distinctly facetted wax gland 
 
 plates. Primary sensoria almost always without hairy edges 10 
 
 — Body naked; very often with distinctly facetted wax-gland plates. Primary 
 
 sensoria often with hairy edges 11 
 
 10. Rudimentary gonapophyses three. Wings mostly with twice-branched cubitus. 
 
 Cornicles always prominent Lachnina 
 
 — • Rudimentary gonapophyses none. Wings with simple or once-branched cubitus. 
 
 Cornicles often absent Anoeciina 
 
A SYNOPSIS OF THE APHIDIDAE 155 
 
 11. Rudimentary gonapopliyses three. Facets of wax-glanil plates almost equal- 
 sized. Wings with simple cubitus. Sensoria of alate forms long oval, net 
 linear PempMgina 
 
 — ■ Rudimentary gonapophyses none. Wax gland plates always with at least 
 one large central facet. Sensoria of alate forms linear. Wings with simple 
 or once-branched cubitus Schizoneurina 
 
 Subfamily CHEBMISINAE 
 1. Rudimentary gonapophyses appearing as three short cones. Wax glands 
 almost alwa}'s present Chermislna 
 
 — Rudimentary gonapophyses seemingly lacking. Wax glands mostly absent. 
 
 Phylloxerina 
 
 Group SiPHONOPHORINA 
 
 1. Apterous forms with a few sensoria on the third antenna! segment. Autennal 
 
 tubercles usually well formed. Body almost never with lateral tubercles, 
 in any case these are never formed on the seventh abdominal segment -.. 2 
 
 — Apterous forms without Sensoria on the third antennal segment. Anteunal 
 
 tubercles often small or absent. Body with lateral tubercles i 
 
 2. Cornicles almost cylindrical, or rarely somewhat swollen on the side, but then 
 
 the body is covered with capitate hairs 3 
 
 — Cornicles distinctly clavate. Body almost bare, never with capitate hairs. 
 
 Rhopalosiphum Koch 
 Type Amphoropliora anipuUata Buckton. 
 
 3. Body of apterous forms with long capitate hairs. First antennal joint 
 
 drawn out, somewhat tooth-shaped on the inner side Myzus Passerini 
 
 Type Aphis riiis Linn. 
 
 — Body of apterous forms bare or without capitate hairs. First antennal joint 
 
 never drawn out, tooth-like Macrosiphum Passerini 
 
 Type Aphis millifolii Fabr. 
 i. Body of apterous forms with capitate hairs. First antennal joint more or 
 
 less toothed on inner side Capitophorus n.gn. 
 
 Type Fhorodon carduinixm Walker. 
 — • Body of apterous forms without capitate hairs. First antennal joint not 
 toothed 5 
 
 5. Body with many long delicate hairs. Cornicles short, somewhat swollen. 
 
 Cladobius (Koch.) Pass. 
 Type Aphis populea Kalt. 
 
 — Body bare or almost so fi 
 
 6. Cornicles almost as long or longer than cauda 7 
 
 — Cornicles much shorter than cauda Ifi 
 
 7. Cornicles always distinctly clavate Siphocoryne Pass. 
 
 Type Aphis avenae Fabr. 
 
 — Cornicles cylindrical or conical S 
 
 8. Antennal tubercles well formed, very distinctly toothed on the inner side. 
 
 Tubercles on the side of the body always absent 9 
 
 — Antennal tubercles mostly small or lacking, never distinctly toothed. Body 
 
 often with lateral tubercles 10 
 
 9. Antennal tubercles very strongly toothed, the first joint being distinctly 
 
 toothed on the inner side Phorodon Pass. 
 
 Type Aphis humuli Schr. 
 — • Antennal tubercles only slightly toothed, first antennal joint being rounded 
 
 or flat on the inner side, never toothed Ovatus n.gn. 
 
 Tj'pe Ovatus mespili v. d. G. 
 
15G MISCELLANEOUS STUDIES 
 
 10. Antennal tubercles well formed, strongly rounded on the inner side. 
 
 Type Aphis cerasi Fabr. '^ ' 
 
 — Antennal tubercles small or lacking, never drawn out distinctly on inner 
 
 side 11 
 
 11. Body with small tubercles on the middle of the seventh and eighth abdominal 
 
 segments, and often also on the head and prothorax Dentatus n.gn. 
 
 Type Aphis sorbi Kalt. 
 
 — Body without tubercles on the middle of the seventh and eighth abdominal 
 
 segments 12 
 
 12. Cubitus of fore wing only once-branched Toxoptera Koch 
 
 Type Toxoptera graminum (Eond.). 
 
 — Cubitus of fore wing always twice-branched. Body often with lateral tuber- 
 
 cles 13 
 
 13. Cornicles short, always distinctly conical. Cauda very short, broad with 
 
 rounded tip, usually approximately the length of the cornicles, or entirely 
 lacking. Lateral tubercles lacking or only indistinctly formed on the an- 
 terior abdominal segments 15 
 
 — Cornicles long, almost cj-lindrical. Cauda sickle- or club-shaped, usually dis- 
 
 tinctly shorter than cornicles ..14 
 
 14. Body long without lateral tubercles. Front often with a very distinct tubercle 
 
 in the middle Myzaphis n.gn. 
 
 Type Aphis rosarnm Walker. 
 — • Body more rounded, with lateral tubercles. Front usually flat, never with a 
 distinct tubercle Aphis Linn. 
 
 Type Aphis rumicis Linn. 
 
 15. Cauda distinctly separated, almost as long as broad Brachycaudus n.gn. 
 
 Type Aphis myosotidis Koch. 
 
 {Aphis cardui Linn, belongs in this genus.) 
 
 — Cauda lacking or scarcely separated, much shorter than broad ...Acaudus n.gn. 
 
 Type Aphis lychnidis Linn. 
 
 16. Cornicles distinctly longer than broad. Cauda usually not conical 17 
 
 — • Cornicles extremely short, scarcely projecting, cylindrical, usually nearly as 
 
 long as broad. Cauda always conical with broad base 19 
 
 17. Cornicles only a little longer than broad, distinctly conical. Cauda sickle- or 
 
 club-shaped Longicaudis n.gn. 
 
 Type Hijalopterus trirhodus (Walker). 
 
 — Cornicles cylindrical, at least twice as long as broad 18 
 
 18. Body with lateral tubercles on first and seventh abdominal segments. Cauda 
 
 small, club-shaped Hyalopterus Koch 
 
 Type Aphis pruni Fabr. 
 
 — Body without lateral tubercles on first and seventh abdominal segments. Cauda 
 
 conical Semiaphis n.gn. 
 
 Type Aphis carotae Koch. 
 
 19. Body long, without lateral tubercles. Antennae short, at the most about half 
 
 the length of the body Brachycolus Buckton 
 
 Type Brachycolus stellariae (Hardy). 
 
 — Body oval with lateral tubercles on prothorax, first and seventh abdominal 
 
 segments. Antennae at least about three-fourths as long as the body. 
 
 Type Aphis thalictri Koch. J v 'S • 
 
 Group Drepanosiphina 
 Genus Drepanosiphum Koch. Type Drepanosiphum platanoides Schrank. 
 
A SYNOPSIfi OF THE APHIDWAE 157 
 
 Group Callipterina 
 
 1. Antennae six-jointed. Cornicles merely pores. Body always with wax 
 
 glands Phyllaphis Koch 
 
 Type Phyllaphis fagi (Linn.). 
 — • Antennae seven-jointed, the terminal process at least one-half .as long as the 
 preceding joint. Cornicles always distinctly projecting. Body almost 
 always without wax glands, the.se always of similar shape 2 
 
 2. Seventh antennal joint distinctly longer than sixth 3 
 
 — Seventh antennal joint only as long as, or shorter than sixth 6 
 
 3. Cornicles but slightly projecting. Antennae curved as in beetles. 
 
 Bradyaphis Mord. 
 Type Bradyaphis antennata (Kalt.). 
 
 — Cornicles distinctly prominent. Antennae straight 4 
 
 4. Anal plate only slightly emarginate, never bilobed. Body with tolerably stiff 
 
 hairs, these are never capitate. Apterous forms always with sensoria on 
 
 third antennal joint Callipterinella n.gn. 
 
 Type Callipteriis hctularins Kalt. 
 
 — Anal plate distinctly bilobed. Body bare or with capitate hairs 5 
 
 5. Apterous forms without sensoria on third antennal joint. The body always 
 
 with capitate hairs Callipterus Koch 
 
 Type Callipterus coryli (Goetze). 
 
 — Apterous forms with a few sensoria on third antennal joint. Boily with dis- 
 
 tinct tubercles Tuberculatus Mord. 
 
 Type Tuberculatus hetulicolus (Kalt.). 
 
 6. Anal plate distinctly bilobed S 
 
 — Anal plate always simple 7 
 
 7. Cauda wart-like, distinctly separated from base. Apterous forms without 
 
 sensoria on third antennal joint CaUipteroides Mord. 
 
 Type Callipterus hetulae Koch. 
 
 — Cauda scarcely visible. Sensoria present on third antennal joint of apterous 
 
 forms Symydobius Mord. 
 
 Type Symydobius ohlongus (Heyden). 
 
 8. Seventh antennal joint nearly as long as the sixth. Wings with only very 
 
 small black spots at the tip of the veins Subcallipterus Mord. 
 
 Type Callipterus alni (Fabr.). 
 
 — Seventh antennal joint nearly half as long as sixth. Wings black spotted. 
 
 Pterocallis Pass. 
 Type Pterocallis tiliae (Linn.). 
 
 Group Chaitophorina 
 
 1. Body with long delicate hairs. Antennae seven-jointed. Cornicles well de- 
 
 veloped 2 
 
 — Body with short thorn-like hairs. Antennae six-jointed, the terminal process 
 
 always distinctly longer than the preceding joint. Cornicles only slightly 
 
 projecting Sipha Passerini 
 
 Type Sipha glyceriae (Kalt.). 
 
 2. Tarsi with two " Haf tliippchen " [i.e., the empodial hair is spatula-like]. 
 
 Chaltophorinella n.gn. 
 Type Chaiiophorus testudinntus (Thornton). 
 
 — Tarsi without " Haftliippehen " [i.e. the empodial hair is bristle-like]. 
 
 Chaltophorus Koch 
 Type Chaiiophorus leucomelas Koch. 
 
158 MISCELLANEOUS STUDIES 
 
 Group Lachnina 
 
 1. Wings usually with twice-branched cubitus, the radius always straight. Cauda 
 
 not at all or only slightly separated Laclinus 111. 
 
 Type Lachnus juniperi De Geer. 
 
 — Wings with once- or twiee-branchel cubitus and with a curved radius ; the mem- 
 
 brane usually with dusky markings. Cauda usually slightly separated 2 
 
 2. Beak distinctly longer than the body, strongly retractile. Cubitus but once- 
 
 branched, the wings only slightly darkened Stomaphis Buckton 
 
 Type Stomaphis quercus (Linn.). 
 
 — Beak clearly shorter than body and only slightly retractile. Wings beautifully 
 
 spotted with dark brown 3 
 
 3. Cubitus twice-branched Dryobius Koch 
 
 Type Dryobius croaticus Koch. 
 
 — Cubitus once-branched Schizodryobius n.gn. 
 
 Type Laclinus exsiccator Hart. 
 
 Tribe ANOECIINA 
 
 1. Hind tarsi elongate Trama Heyden 
 
 Type Trama radicis Koch. 
 — • Hind tarsi not elongate 2 
 
 2. Cubitus once-branched. Cornicles present, quite prominent. Margin of body 
 
 with peculiar non-faceted "wax-gland" (?) plates Anoecia Koch 
 
 Type Anoecia corrii (Fabr.). 
 — ■ Cubitus not branched. Cornicles absent. Wax gland plates not present. 
 
 TuUgrenia v. d. G. 
 Type TuUgrenia phaseoli (Pass.). 
 
 Tribe HORMAPHIDINA 
 
 1. Antennae always five-segmented. The fronds almost without exception with 
 
 two little horns. Cubitus once brandied Cerataphis Lieht. 
 
 Type C. lataniae Boisd. 
 — • Antennae of the apterae often only three-segmented. Fronds without protuber- 
 ances. Cubitus simple Hamamelistes Schim. 
 
 Type H. ietulae Mordw. 
 
A SYNOPUla OF THE APHIDIDAE 159 
 
 APPENDIX 2 
 Host Plant List op California APHIDIDAE^^ 
 
 Abies (fir) 
 
 4.5. Lachnus fcrrisi Swain 
 47. Lachnus occidentalis Dvdn. 
 176. Mindaru-s obietinus Koch 
 AhtUilon (Indian mallow) 
 
 146. Aphis senecio Swain 
 104. Slwpalosiphum pcrsicac (Sulz.) 
 Acer (maple, box elder) 
 
 5. Drepanaphis acerifolii (Thomas) 
 4. Drepannsiphum plaiannides (Sehrank) 
 33. Thomtisia negundinis (Thomas) 
 Achillea (yarrow) 
 
 79. Macrosiphuin solanifolii (Ashmead) 
 Acgopodium (goutweed) 
 
 108. Siphocorync capreae (Pabr.) 
 Aesculus (California buckeye) 
 
 88. Myzus circumflexus (Buckton) 
 171. Prociphilus venafuscus Patch 
 Alder, see Alnus 
 Alfalfa, see Mcdicago 
 Alfilerilla, see Erodium 
 Alisma (water plantain) 
 
 156. Siphocoryne ni/mphaeae (Linn.) 
 Almond, see Prunus 
 Alnus (Alder) 
 
 9. Eucallipterus jlava (Davidson) 
 8. Euceraphis gillettei Davidson 
 11. MyzocaUis alnifoliae (Fitch) 
 Alopecurus (foxtail) 
 
 68. Macrosiphum granarium (Kirby) 
 88. Myzus circumflexus (Buckton) 
 Althaea (hollyhock) 
 
 121. Aphis eunnomi Fabr. 
 Alum root, see Heuchera 
 
 35 In the following list only the generic and common names of the plants are 
 employed, the various species of plants being omitted. Although in certain cases 
 aphids are restricted to certain species, as Eriosoma languinosa Hartig on Pyrus 
 communis but not on Pyrus malus, these are in the minority. The botanical 
 names are taken from the following works, with preference as in the order listed : 
 
 Bailev, L. H., Standard cyclopedia of horticulture, vols. 1-6, New York, Mac- 
 millan, 1914-1917. 
 
 Eobinson, B. L., and Fernald, M. L., Gray 's New manual of botany, ed. 7, 
 Cambridge, Harvard University, 1908. 
 
 Jepson, W. L., A flora of western middle California, ed. 2, San Francisco, Cun- 
 ningham, 1911. 
 
 Abrams. LeRoy, Flora of Los Angeles and vicinity, Palo Alto, Stanford Uni- 
 versity Press, 1904. 
 
160 MISCELLANEOUS STUDIES 
 
 Amaraiithus (pigweed) 
 
 123. Aphis gossypii Glover 
 
 132. Aphis middletonii Thomas 
 
 104. Shopalosiplium persicae (Sulz) 
 
 163. Trifidaphis radicicola (Essig) 
 Amhrosia (ragweed) 
 
 146. Aphis scnecio Swain 
 77. Macrosiphum rudbeckiae (Fitch) 
 Ampelodesma 
 
 68. Macrosiphum granarium (Kirby) 
 Amsiiiclcia (amsinckia) 
 
 146. Aphis seneoio Swain 
 
 104. Bhopalosiphiim persicae (Sulz.) 
 Angelica (angelica) 
 
 109. Aphis angi'licae Koch 
 
 115. Aphis cari Essig 
 
 108. Siphocoryne capreae (Fabr.) 
 Anise, Wild, see Carum 
 Anthemis (chamomile) 
 
 121. Aphis euonomi Fabr. 
 146. Aphis senecio Swain 
 123. Aphis gossjipii Glover 
 
 Apple, see Pyrus 
 Apricot, see Prunus 
 Aquilegia (columbine) 
 
 85. Mysus aquilegiae Essig 
 Arbor vitae, see Thuja sp. 
 Arbutus (mailrone, strawberry tree) 
 
 103. Hhopalnsiphum nervatuvi Gillette 
 Arctostaphylos (maiizanita) 
 
 1. PhyUapliis coiveni (Cockerell) 
 
 103. Bhopulosiphum nervatum Gillette 
 Artemisia (sagebrush, oldman, California mugAvort, etc.) 
 
 122. Aphis frigidae Oestlund 
 131. Aphis medicaginis Koch 
 137. Aphis oregonensis Wilson 
 146. AjMs senecio Swain 
 
 61. Macrosiphum artemisiac (Fonsc.) 
 
 62. Macrosiphum artemisicola (Williams) 
 72. Macrosiphum ludovicianiae (Oestlund) 
 
 Artichoke, see Cynara 
 Arundinaria (bamboo) 
 
 12. MyzocaUis arundicolens (Clarke) 
 
 13. Myzocallis arundinariae Essig 
 Arundo (giant reed) 
 
 12. MyzocaUis arundicolens (Clarke) 
 
 13. Mjisocallis arundinariae Essig 
 Asclepias (milkweed) 
 
 123. Aphis gossypii Glover 
 135. Aphis nerii Fonsc. 
 
 102. Rhopalosiphum lactucae (Kalt.) 
 
A SYNOPSIS OF THE APHIDIDAE 161 
 
 Ash, see Fraxiinis 
 
 Arparagus (asparagus, smilax, asparagus feru) 
 
 123. Aphis gossypii Glover 
 88. Myzus circamflexus (Buekton) 
 Aster (aster) 
 
 132. Aphis middletonii Thomas 
 
 146. Aphis senccio Swain 
 Astragalus (loco weed) 
 
 131. Aphis medicaginis Koch 
 Atriplex (oraehe) 
 
 147. Aphis tetrapteralis Coekerell 
 
 79. Macrosiphum solanifolii (Ashmead) 
 Avocado, see Persca 
 Avena (oats) 
 
 111. Aphis avenae Fabr. 
 
 (i8. Macrosiphum granarium (Kirby) 
 105. Uhopalosiphum rhois Monell 
 Baccharis (groundsel) 
 
 146. Aphis senecio Swain 
 
 63. Macrosiphum baccharadis (Clarke) 
 
 77. Macrnsiplivm rudheclme (Fiteh) 
 104. Uhopalosiphum persicae (Sulz.) 
 Bamboo, see Arundinaria, Bamhusa, and Phyllostachys 
 Bamhusa (bamboo) 
 
 12. Mrtzocallis arundicolens (Clarke) 
 Banana, see Musa 
 Barberry, see Berberis 
 Barley, see Hordrum 
 Basswood, see Tilia 
 Bean, see Pliaseolus 
 Bean, Blackeye, see Vigna 
 Bean, Broad, see Vicia 
 Beech, see Fagus 
 Beet, see Beta 
 Begonia (begonia) 
 
 123. Aphis gossypii Glover 
 Bell, fairy, see Dipsorum 
 Berberis (barberry) 
 
 154. Liosomaphis herberidis (Kalt.) 
 Beta (beet, sugarbeet) 
 
 124. Aphis gossypii Glover 
 164. Pemph'gns betae Doane 
 
 Betula (birch) 
 
 6. Calaphis betulaecolens (Fitch) 
 27. CalUpterinella annulata (Koch) 
 
 7. Euceraphis betulae (Koch) 
 Birch, see Betula 
 
 Blackberry, see Subus 
 Bougainvillea (bougainvillea) 
 
 104. Uhopalosiphum persicae (Sulz.) 
 Boxelder, see Acer 
 
162 MISCELLANEOUS STUDIES 
 
 Brassica (cabbage, mustard, turnip, etc.) 
 
 112. Aphis hrcissicac Linn. 
 
 144. Aphis pseudohrassicae Davis 
 
 166. Pemphigus popuU-transversiis Riley (?) 
 
 102. Ehopalosiphum lactucac (Kalt.) 
 
 104. BhopalosipMim I'^rsicue (Sulz.) 
 
 Broom, see Cytisus 
 
 Buckeye, California, see Aesculus 
 
 Buekton, see Bhamnxis 
 
 Bur clover, see Medicago 
 
 Buttercup, see Sanuncuhis 
 
 Cabbage, see Brassica 
 
 Calendula (marigold) 
 
 113. Aphis calendulicola Monell 
 121. Aphis euonomi Fabr. 
 146. Aphis senecio Swain 
 
 California buckeye, see Aesculus 
 California holly, see Heteromeles 
 California mugwort, see Artemesia 
 California poppy, see Esclischoltzia 
 California sagebrush, see Artemisia 
 California tule, see Tjipha 
 Calla, see Zanicdeschia 
 Camellia (camellia) 
 
 152. Toxoptcra auraiitii (Fonsc.) 
 Canary grass, see Phalaris 
 Cantaloupe, see Cucumis 
 Capsella (shepard's purse) 
 
 112. Aphis brassicac Linn. 
 
 123. Aphis gossypii Glover 
 
 141. Aphis pseudohrassicae Davis 
 
 104. Ehopalosiphum persicae (Sulz.) 
 Capsicum (pepper piraeuto) 
 
 104. Ehopalosiphum persicae (Sulz.) 
 Caragana (pea tree) 
 
 131. Aphis medicaginis Koch. 
 Carum (wild anise) 
 
 115. Aphis cari Essig 
 
 155. Siphocor]ine caprcae (Fabr.) 
 Castanea (chestnut) 
 
 15. My^ocallis castanicola Baker (davidsoni Swain) 
 Catalpa (catalpa) 
 
 123. Aphis gossypii Glover 
 
 139. Aphis pomi de Gee'r 
 
 104. Ehopalosiphum persicae (Sulz.) 
 Cauliflower, see Brassica 
 Ccanothus (mountain lilac) 
 
 116. Aphis ceanothi Clarke 
 Centaurea (taealste) 
 
 130. Aphis marutae Oestlund 
 Centranthus (red valerian) 
 
 76. Macrosiphum rosae (Linn.) 
 104. Ehopalosiphum persic-ae (Sulz.) 
 
A SYNOPSIS OF THE APHIDIDAE 163 
 
 Chaerophyllum 
 
 155. Siphocoryne capreae (Fabr.) 
 Chamomile, see Anthemis 
 Chaparral broom, see Baecharis 
 Charlock, see Brassica 
 Cheeseweed, see Malva 
 Cheiranthus (wallflower) 
 
 88. Myzus circumflexus (Buckton) 
 Chenopodium (lamb's-quarters, pigweed) 
 
 110. Aphis atriplicis Linn. 
 
 123. Aphis gossypii Glover 
 
 124. Aphis hederae Kalt. (?) 
 164. Pemphigus betae Doane (?) 
 104. Ehopalosiphum persicae (Sulz.) 
 
 Clierry, see Prunus 
 Cherry, wild, see Prunus 
 Chestnut, see Castanea 
 Chestnut, Horse, see AescnUis 
 Chicory, see Chicorium 
 Christmas berry, see Hcteromeles 
 Chrysanthemum (chrysanthemum) 
 
 56. Amphorophora latysiphon Davidson 
 
 123. Aphis gossypii Glover 
 
 146. Aphis senecio Suvain 
 
 160. Coloradoa rufomaculata Wilson 
 
 77. Macrosiphum rudheckiae (Fitch) 
 
 78. Macrosiphum sanhorni Gillette 
 Cichorium (chicory) 
 
 71. Macrosiphum lactticae (Kalt.) 
 Cicuta (water hemlock) 
 
 155. Siphocoryne capreae (Fabr.) 
 Cirsium (thistle) 
 
 144. Aphis cardui Linn. 
 CitruUus (watermelon) 
 
 123. Aphis gossypii Glover 
 Citrus (citrus, orange, lemon, etc.) 
 
 118. Aphis cool'i Essig 
 
 123. Aphis gossypii Glover 
 
 131. Aphis medicaginis Koch 
 
 79. Macrosiphum solanifolii (Ashmead) 
 104. Ehopalosiphum persicae (Sulz.) 
 152. Toxoptera aurantii (Fonsc.) 
 
 Clarlia (clarkia) 
 
 104. Ehopalosiphum persicae (Sulz.) 
 Clewiatis (clematis) 
 
 94. Myzus varians Davidson 
 Clover, see Trifolium 
 Clover, Sweet, see Melilotus 
 Coffeeberry, see Ehamnus 
 Columbine, see Aquilegia 
 Compositae (various species) 
 
 131. Aphis medicaginis Koch 
 
 65. Macrosiphum chrysanthcmi (Oestlund) 
 
 77. Macrosiphum rudbeckiae (Fitch) 
 
164 MISCELLANEOUS STUDIES 
 
 Conium (poison hemlock) 
 
 159. Siphocoryne pastinac<ie (Linn.) 
 Convolvulius (morning glory) 
 
 56. Amphorophora latysiphon Davidson 
 123. Aphis gossypii Glover 
 72. Macrosiphum ludovicianae (Oestlund) 
 Corn, see Zea 
 Coriius (dogwood) 
 
 119. Aphis cornifoliae Fitch 
 123. Aphis gossypii Glover 
 
 Coryhis (hazelnut) 
 
 16. Myzocallis coryli (Goetze) 
 
 100. Bhopalosiphum corylinum Davidson 
 Cotoneaster (cotoneaster) 
 
 139. Aphis pomi de Geer 
 Cotton, see Gossypium 
 Cottonwood, see Popuhts 
 Cow parsnip, see Heracleuin 
 Cowpea, see Vigiia 
 Cowslip, see Primula 
 Crab apple, see Pyrus 
 Cranesbill, see Geranium 
 Crataegus (hawthorn) 
 
 120. Aphis crataegifoliae Fitcli 
 139. Aphis pomi de Geer 
 
 Cruciferae (various spp. ) 
 
 112. Aphis brassicae Linn. 
 
 141. Aphis pscudobrassicae Davis 
 Cucumber, see Cucumis 
 Cucumis (cucumber, muskmelon, cantaloupe, etc.) 
 
 123. Aphis gossypii Glover 
 Cucurbita (squash, gourd, pumpkin, etc.) 
 
 123. Aphis gossypii Glover 
 66. Macrosiphum cucurhitae (Thomas) 
 Cuprcssus (cypress) 
 
 161. Cerosipha cuprcssi Swain 
 
 90. Macrosiphum morrisoni Swain 
 Currant, see Bibcs 
 Cydonia (quince) 
 
 139. Aphis pomi De Geer 
 Cynara (artichoke) 
 
 8fi. Myzus braggii Gillette 
 C ynoglossum (houndstongue) 
 
 104. Shopalosiphum persicae (Sulz.) 
 Cypress, see Cupressus 
 Cyrtomium (holly fern) 
 
 162. Cerataphis lataniac (Boisd.) 
 .58. Idiopterus neithrclepidis Davis 
 88. Myzus cireumfiexus (Buckton) 
 
 Cytisus (broom) 
 
 146. Aphis senecio Swain 
 
 104. Bhopalosiphum persicae (Sulz.) 
 
A SYNOPHIS OF THE APHIDIDAE ]65 
 
 Dandelion, see Taraxamm 
 Datura (jimson weed) 
 
 123. Aphi^ gossypii Glover 
 Deinandra 
 
 79. Maerosiphwm solanifolii (Ashmead) 
 Digitalis (foxglove) 
 
 88. Myzus circumflexus (Buckton) 
 Dipsamis (fuller's teasel) 
 
 7fi. Macrosiphum rosae (Linn.) 
 
 77. Macrosiplnim rudbeckiae (Fitch) 
 
 104. HJiopalosiplium persicae (Sulz.) 
 Di^porum (fairy bell) 
 
 79. Macrosiphum salanifolii (Ashmead) 
 Dock, see Sumex 
 Dogwood, see Cornws 
 Douglas fir, see Pseudotsuga 
 Dracaena (dragon tree) 
 
 ill. Aphis avenae Fabr. 
 Dragon tree, see Dracaena 
 Elderberry, see Samhucus 
 Elm, see Vlmus 
 Elymus (wild rye) 
 
 68. Macrosiphum granarium (Kirby) 
 English ivy, see Eedera 
 Epilohium (fireweed) 
 
 1.36. Ajyhis oenothcrae Oestlund 
 Eriobotrya (loquat) 
 
 139. Aphis pomi de Geer 
 Erodium. (alfilerilla) 
 
 79. Macrosiphum solanifolii (Ashmead) 
 
 104. Shopalosiphum persicae (Sulz.) 
 Erysimum (western wallflower) 
 
 1.55. Siphocoryne capreae (Fabr.) 
 Escallonia (escallonia) 
 
 104. Phopalosiphum persicae (Sulz.) 
 Eschscholt^ia (California poppy) 
 
 123. Aphis gossypii Glover 
 Everlasting, see Gnaph-alium 
 Fagiis (beech) 
 
 2. PhyllapMs fagi (Linn.) 
 Fairybell, see Dipsorum 
 Fennel, see Foeniculum 
 Fenugreek, see Trigonella 
 Fern, asparagus, see Asparagus 
 Fern, Boston, see Nephrolepis 
 Fern, holly, see Cyrtomium 
 Fig marigold, see Mesevibryanthemum 
 Figwort, see Scrophularia 
 Fir, see Abies 
 
 Fir, Douglas, .see Pseudotsuga 
 Fireweed, see Epilobium 
 Foeniculum (fennel) 
 
 155. Siphocoryne capreae (Fabr.) 
 
166 MISCELLANEOUS STUDIES 
 
 Foxglove, see Digitalis 
 
 Foxtail, see Alopec^lrus ' 
 
 Fragaria (strawberry) 
 
 90. Myzus fragaefolii Cockerell 
 
 Fraxinus (ash) 
 
 171. Prociphilus venafuscus Patch 
 
 167. Thecahius californicus (Davidson) 
 
 Fuller 's teasel, see Dipsacus 
 
 Fuchsia (fuchsia) 
 
 79. Macrosiphum solanifoUi (Ashmead) 
 88. Myzus circumflexus (Buckton) 
 
 Gambleweed, see Sanicula 
 
 Geranium, see Pelargonium 
 
 Geranium (cranesbill) 
 
 104. Bhopalosiphuin persicae (Sulz.) 
 German ivy, see Senccio 
 
 Gladiohis (gladiolus) 
 
 88. Myzus circumflexus (Buckton) 
 Glycyrrhisa (liquorice) 
 
 131. Aphis medicaginis Koch 
 Gyiaplialium (everlasting) 
 
 146. Aphis senccio Swain 
 60. Macrosiphum amlirosiae (Thomas) 
 Gooseberry, see Eiics 
 Goosefoot, see Chenopodium 
 Gossypium (cotton) 
 
 123. Aphis gossypii Glover 
 Gourd, see Cucurbita 
 Goutweed, see Aegopodium 
 Graminaceae (various species) 
 
 111. Aphis avenae Fabr. 
 
 67. Macrosiphum dirhodum (Walker) 
 
 68. Macrosiphum granarium (Kirby) 
 
 105. Ehopalosiphum rhois Monell 
 Grape, see Vitis 
 
 Grindelia (marsh grindelia) 
 
 146. Aphis senecio Swain 
 Hawthorn, see Crataegus 
 Hazelnut, see Corylus 
 Hcdera (English ivy) 
 
 109. Aphis angelic^ae Koch 
 
 124. Aphis hederae Kalt 
 
 104. Ehopalosiphum persicae (Sulz.) 
 Hedge mustard, see Sisymbrium 
 Hedge nettle, see Staehys 
 Eelianthus (sunflower) 
 
 123. Aphis gossypii Glover 
 
 132. Aphis middletonii Thomas 
 146. Aphis senccio Swain 
 
 60. Macrosiphum amirosiae (Thomas) 
 77. Macrosiphum riidheckiae (Fitch) 
 104. Ehopalosiphum persicae (Suli.) 
 
A SYNOPSIS OF THE APHIDIDAE 167 
 
 Hemlock, Poison, see Conium 
 Hemlock, Water, see Cicuta 
 Heracleum (cow parsnip) 
 
 123. Aphis goss!/i}ii Glover 
 
 125. Aphis heraclei Cowen 
 Heteromeles (California holly, Christmas berry) 
 
 170. Frociphilus alnifoli-ae (Williams) 
 
 103. Ehopalosiphum nervatum Gillette 
 Heuchcra (alum root) 
 
 69. Macrosiphum heuclicrae (Thomas) 
 Hibiscits (rose mallow) 
 
 121. Aphis euonomi Fabr. 
 Holly fern, see Cyrtonium 
 Hollyhock, see Althaea 
 Holly, mountain, see Heteromeles 
 Honey flower, see Melianthus 
 Honeysuckle, see Lonicera 
 Hop, see Humulus 
 Hordeum (barley) 
 
 111. Aphis avei)ae Fabr. 
 68. Macrosiphum granarium (Kirby) 
 Houndstongue, see Cynoglossiim 
 Humulus (hop) 
 
 123. Aphis gossypii Glover 
 98. Phorodon humuli (Schrank) 
 Hydrangea (hydrangea) 
 
 123. Aphis gossypii Glover 
 Indian mallow, see Abut don 
 Ironweed, see Veronina 
 Ivy, Engislh, see Hedera 
 Ivy, German, see Senecio 
 Jasminum (jessamine) 
 
 70. Macrosiphum jasmini (Clarke) 
 Jessamine, see Jas-ininnm 
 
 Jimpson weed, see Datura 
 Juglans (walnut) 
 
 24. Callipterus calif amicus (Essig) 
 
 25. Callipterus caryae Monell 
 
 23. Chromaphis juglandieola (Kalt.) 
 
 26. Monellia caryella (Fitch) 
 Knotweed, see Polygonum 
 Lactuca (lettuce) 
 
 79. Macrosiphum solanifolii (Ashmead) 
 Lamb 's-quarters, see Chenopodium 
 Lathyrus (sweet pea) 
 
 74. Macrosiphum iii-si (Kalt.) 
 Laurel, see Laurus 
 Laurel, California, see Umiellularia 
 Laurestinus, see Viburnum 
 Laurus (laurel) 
 
 150. Aphis vibumicolens n.sp. 
 
168 MISCELLANEOUS STUDIES 
 
 Lavatera (tree mallow) 
 
 104. Ehopalosiplmm persicae (Sulz.) 
 Leatlier root, see Psorales 
 Lemon, see Citrus 
 Lepidium (peppergrass) 
 
 123. Aphis gossypii Glover 
 Lettuce, see Lactuca 
 Ligusticum (lovage) 
 
 155. Siiihocoryne caprcae (Fabr.) 
 Lilac, see Syringa 
 Lilac, Mountain, see Ceanotlius 
 Lilium (lily) 
 
 123. Aphis gossypii Glover 
 88. Myzus circumflexus (Buckton) 
 Lily, see Lilium 
 Lily, Water, see Nymphaea 
 Linden, see Tilia 
 Liquorice, see Glycyrrliira 
 Liriudendron (tulip tree) 
 
 104. Ehopalosiphum persicae (Sulz.) 
 Lithospermum 
 
 127. Aphis lithospenni Wilson 
 Loco weed, see Astragalus 
 Locust, see Soiiiiia 
 Loganberrj-, see Huhus 
 Lonicera (honeysuckle) 
 
 157. Siphocoryne pastinacae (Linn.) 
 Loquat, see Eriobotrya 
 Lovage, see Ligusticum 
 Lupinus (lupine) 
 
 59. Macrosiphum albifrons Essig 
 Lycopersicum (tomato) 
 
 91. My:us li/copf-rsicae (Clarke) 
 
 104. Bhopalnsiphum persicae (Sulz.) 
 Madia (tarweed) 
 
 77a. Macrosiphum rudhecMae (Fitch) var. madia n.var. 
 Madron, see Arbutus 
 Mallow, Indian, see Aiutilon 
 Mallow, Eose, see Hibiscus 
 Mallow tree, see Lavatera 
 Malva (cheeseweed) 
 
 121. Aphis euonotni Fabr. 
 
 123. Aphis gossypii Glover 
 
 104. Ehopalosiphum persicae (Sulz.) 
 Manzanita, see Arctostaphylos 
 Maple, see Acer 
 Marigold, see Calendula 
 Marigold, fig, see Mesembriianthrmum 
 Matthiola (ten-weeks' stock) 
 
 141. Aphis pscudobrassicae Davis 
 Mayten, see Maytetius 
 Maytenus (mayten) 
 
 121. Aphis euonomi Fabr. 
 
A SYNOPSIS OF THE APHIDIDAE 169 
 
 Medicago (alfalfa, bur clover, etc.) 
 
 131. Aphis viedicaginis Koch 
 74. Macrosiplium pisi (Kalt.) 
 Meliantlius (honey flower) 
 
 104. Bhopalosiplmm persicae (Sulz.) 
 Melilotus (sweet clover) 
 
 131. Aphis medicaginis Koch 
 Mesembryanthemum (fig marigold) 
 
 121. Aphis euonomi Fabr. 
 Milk thistle, see Silyhum 
 Milkweed, see Asclepias 
 Morning glory, see Convolvulus 
 Morus (mulberry) 
 
 133. Aphis mori Clarke 
 Mountain holly, see Heteromeles 
 Mountain lilac, see Ceanothus 
 Mugivort, California, see Ai-temisia 
 Mulberry, see Morus 
 Musa (banana) 
 
 111. Aphis avcnac Fabr. 
 Muskmelon, see Cucumis 
 Mustard, see Brassica 
 Mustard, Hedge, see Sisymhrium 
 Mustard, Teasel, see Erysimum 
 Nasturtium, see Tropaeolum 
 Nectarine, see Prunus 
 Nephrolcpis (Boston fern) 
 
 58. Idioptcrus nephrelepidis Davis 
 Nerium (oleander) 
 
 13.5. Aphis nerii Fonsc. 
 Nettle, Hedge, see Stachys 
 Nettle, Stinging, see Urtica 
 Nightshade, see Solamim 
 Ninebark, see Physocarpus 
 Nymphaea (water lily) 
 
 123. Aphis gossypii Glover 
 
 156. Siphocoryne nympliaeae (Linn.) 
 Oak, see Quercus 
 Oak, Poison, see Bhus 
 Oak, Tanbark, see Pasania 
 Oats, see Avena 
 Oenothera (evening primrose) 
 
 13(i. Aphis oenothcrae Oestlund 
 Oldnian, see Artemisia 
 Oleander, see Nerium 
 Orange, see Citrus 
 Orache, see Atriplex 
 Orchidaceae (orchids) 
 
 162. Cerataphis lataniae (Boisd.) 
 Orthoearpus (owl clover) 
 
 73. Macrosiphum orthocarpi Davidson 
 Owl clover, see Orthoearpus 
 
170 MISCELLANEOUS STUDIES 
 
 Oxalis (oxalis) 
 
 97. Macros! phum solanifolii (Ashmead) 
 104. EhopaJosiiihum pcrsicae (Sulz.) 
 
 152. Toxoptera aurantii (Fonsc.) 
 Pansy, see Viola 
 
 Papaver (poppy) 
 
 121. Aphis euonomi Fabr. {papaveris Fabr.f) 
 Parsley, see Petroselinum 
 Parsnip, see Pastinaca 
 Parsnip, Cow, see Hcracleum 
 Fasania (tanbark oak) 
 
 20. Mi/^ocallis pasaniae Davidson 
 Pastinaca (parsnip) 
 
 156. Siphocoryne pastinacae (Linn.) 
 Pea, see Pisum 
 Pea, Cow, see Vigna 
 Pea, Sweet, see Luthyrus 
 Pea tree, see Caragona 
 Peach, see Pruiius 
 Pear, see Pyrus 
 Pelargonium (geranium) 
 
 97. Pcntalonia nigronervosa Coiiiiellct 
 
 104. Hhopalosiplmm pcrsicae (Sulz.) 
 Pentstemon (pentstemon) 
 
 88. Myzus cireumflexus (Buckton) 
 
 104. Ithopalosiphum pcrsicae (Sulz.) 
 Pepper, see Ctipxicum 
 Peppergrass, see Lepidium 
 Periwinkle, see Vinca 
 Persea (avacado) 
 
 123. Aphis gossypii Glover 
 Petroselinum (parsley) 
 
 155. Siphocoryne capreac (Fabr.) 
 Phalaris (canary grass) 
 
 111. Aphis avenac Fabr. 
 
 153. Hyalopterus arundinis (Fabr.) 
 Pliascolus (bean) 
 
 121. Aphis euonomi Fabr. (rumicis Linn.f) 
 131. Aphis medicaginis Koch 
 74. Macrosiphum pisi (Kalt.) 
 Phragmites (reed grass) 
 
 153. Hyalopterus anindinis (Fabr.) 
 Phyllostachys (bamboo) 
 
 12. My::ocaUis arundicolciis (Clarke) 
 Physocarpus (ninebark) 
 
 100. Ehopalosiplium coryliniim Davidson 
 Picea (spruce) 
 
 46. Lachnus glehnus Essig 
 55. Lachnus vanduzei n.sp. 
 158. Myzaphis aiietina (Walker) 
 Pigweed, see Amaranthus, and Chenopodium 
 Pimento, see Capsicum 
 
A SYNOPSIS OF THE APRIDIDAE 171 
 
 Pimpinella 
 
 155. Siphocorync capreac (Fabr.) 
 Pine, see Pinus 
 
 Pinus (pine) 
 
 178. Chermcs cooleyi Gillette 
 
 179. Chermes pinicorticis Fitch 
 43. Essigella calif ornica (Essig) 
 45. Laclmus ferrisi Swain 
 
 48. Laclmus oregonensis Wilson 
 
 49. Lachinis pini-radiatac Davidson 
 
 50. Lachnus ponderosa Williams 
 52. Lachnus sabinianus n.sp. 
 
 53a. Lachnus tomcntusa (De Geer) (Addenda) 
 176. Mindants abietinus Koch 
 Pisttm (pea) 
 
 74. Macrosiphum pisi (Kalt.) 
 Pittosporum (pittosporum) 
 
 139. Aphis pomi De Geer 
 
 79. MacroMphum solanifolii (Ashmead) 
 104. Slwpulosiphum pcrsicae (Sulz.) 
 Plantago (plantain) 
 
 123. Aphis gossjipii Glover 
 129. Aphis malifoliae Fitch (?) 
 88. Myzus circnmflerns (Buekton) 
 Plantain, see Plantago 
 Plantain, Water, sec Alisma 
 Platanus (western sycamore) 
 
 4. Drepanosiphum platanoides (Sehrank) 
 Plum, see Prunus 
 Polygonum (knotweed) 
 
 100. Shopalosiphum liippopliacs (Koch). 
 
 156. Siphocoryne nymphaeae (Linn.) 
 Pomegranate, see Punica 
 
 Pondweed, see Potamogeton 
 
 Poplar, see Populus 
 
 Poppy, see Papavcr 
 
 Poppy, California, see Esclischolt-ia 
 
 Populns (poplar, eottonwood) 
 
 28. Arctaphis populifolii (Essig) 
 ■ 164. Pemphigus betae Doane 
 
 165. Pemjihigus populi-caulis Fitch 
 
 166. Pemphigus populi-transrersus Eiley 
 181. Phylloxerina pop^^laria (Pergande) 
 
 40. Ptcrocomma populifoliae (Fitch) 
 
 167. Thccabius californicus (Davidson) 
 
 168. Thecabius populiconduplifolixis (Cowen) 
 
 169. Thecabius populi-monilis (Eiley) 
 
 34. Thomasia populicola (Thomas) 
 
 35. Thomasia salicola (Essig) 
 Potamogeton (pondweed) 
 
 156. Siphocoryne nymphaeae (Linn.) 
 Potato, see Solanim 
 
172 MISCELLANEOUS STUDIES 
 
 Primrose, Evening, see Oenothera 
 Primula (cowslip) 
 
 5(5. Amplioropliora latysiphon Davidson 
 Prune, see Prunus 
 Pruntis (almond, apricot, cherry, nectarine, peach, plum, prune) 
 
 107. Aphis alamedensis Clarke 
 
 114. Aphis cardui Linn. 
 
 117. Aphis cerasifoliae Fitch 
 
 138. Aphis persicae-niger Smith 
 140. Aphis prunorum Dobr. 
 
 153. Ei/alopterus ar^lndinis (Fabr. 
 
 87. Myzus cerasi (Fabr.) 
 
 98. Phorodon humuli (Schrauk) 
 104. Shopalosiphum persicae (Sulz.) 
 156. Siphocoryne nymphaeae (Linn.) 
 Pseudotsuga (Douglas fir) 
 
 178. Chermes cooleyi Gillette 
 
 43. Essigella calif ornica (Essig) 
 
 51. Lachnus pseudotsuga Wilson 
 
 53. Lachnus taxifolia Swain 
 171. Prociphilus venafuscus Patch 
 Psorales (leather root) 
 
 74. Macrosiphum pisi (Kalt.) 
 Pteris (brake) 
 
 75. Macrosiphum pteridis Wilson 
 Punica (pomegranate) 
 
 123. Aphis gossypii Glover 
 Pumpkin, see Cucurtita 
 Pyrus (apple, pear) 
 
 123. Aphis gossypii Glover 
 
 129. Aphis malifoliae Fitch 
 
 139. Aphis pomi De Geer 
 
 175. Eriosoma languinosa Hartig (pyricola B. & D.) 
 174. Eriosoma lanigcrum (Hausraan) 
 Qucrcus (oak) 
 
 5. Drepanaphis accrifolii (Fitch) (?) 
 
 14. Myzocallis ielliis (Walsh) 
 
 15. Myzocallis castanicola Baker (davidsojii Swain) 
 
 17. Myzocallis discolor (Monell) 
 
 18. Myzocallis pmictatus (Monell) 
 
 19. Myzocallis californicus Baker {maurcri Swain) 
 21. Myzocallis querctis (Kalt.) 
 
 3. Phyllaphis quercicola Baker 
 
 36. Symydohius agrifoliae Essig 
 
 37. Symydohius chrysolcpis Swain 
 177. Vacuna dryophila Sehrank (?) 
 
 Quince, see Cydonia 
 Eadish, see Eaphanus 
 Eagweed, see Amhrosia 
 Eamona (black sage) 
 
 142. Aphis ramona Swain 
 
A SYNOPSIS OF TEE APHIDIDAE 173 
 
 Sanunculus (buttercup) 
 
 132. Aphis middletonii Thomas 
 104. Ehopalosiphum persicae (Sulz.) 
 
 167. Thecabius calif ornicus (Davidson) 
 
 168. Thecabius populiconduplifolius (Cowen) 
 Baphanus (radish) 
 
 112. Aphis brassicae Linn. 
 
 141. Aphis pseudobrassicae Davis 
 
 104. Ehopalosiphum persicae (Sulz.) 
 Eeed, Giant, see Aruiido 
 Keed grass, see Phragmites 
 Bliamnus (buckthorn, coffeeberry) 
 
 123. Aphis gossypii Glover 
 
 92. Myeus rhamnus (Clarke) 
 Rhus (poison oak) 
 
 lO.o. Ehopalosiphum rhois Monell 
 Bibes (currant, gooseberry) 
 
 126. Aphis lioughtonensis Troop 
 134. Aphis neo-mexicana Ckll. var. pacifica Dvdn. 
 89. Myzus cynosbati (Oestlund) 
 
 93. Mygus ribifolii Davidson 
 Bobinia (locust) 
 
 131. Aphis medicaginis Koch 
 Eosa (rose, -wild and cultivated) 
 
 67. Macrosiphum dirhodum (Walker) 
 
 76. Macrosiphum rosae (Linn.) 
 159. MyiiOphis rosarum (Walker) 
 
 103. Mygus nervatum Gillette 
 Rose, see Eosa 
 
 Rose mallow, see Hibiscus 
 
 Bubus (blackberry, loganberry, thimbleberry) 
 
 143. Aphis rubiphila Patch 
 
 57. Amphorophora rubi (Kalt.) 
 
 95. Nectarosiphon rubicola (Oestlund) 
 Bumex (dock, sorrell) 
 
 121. Aphis euonomi Fabr. {rumicis Linn.) 
 
 123. Aphis gossypii Glover 
 
 146. Aphis seneeio Swain 
 
 164. Pemphigus bctae Doane (?) 
 
 104. Ehopalosiphum persicae (Sulz.) 
 Rye, Wild, see Elymus 
 
 Sagebrush, see Arteinisia 
 Sage, Black, see Eamona 
 Salix (willow) 
 
 144. Aphis salicicola Thomas 
 146. Aphis seneeio Swain 
 
 29. Arctaphis viminalis (Monell) 
 31. Fullau-aya saliciradicis Essig 
 
 64. Macrosiphum californicum (Clarke) 
 
 30. Micrella monella Essig 
 
 182. Phylloxerina salicola (Pergande) 
 40. Pterocomma flocculosa (Weed) 
 
174 MISCELLANEOVS STUDIES 
 
 41. Ptcrocomma populifoUae (Fiteh) 
 
 42. Pterocomma smithiae (Monell) 
 155. Siphocoryne capreae (Fabr.) 
 
 38. Slimy duhius macrostachyae Essig 
 
 39. Symiidobius salicicorticis Essig 
 32. Thomasia crucis Essig 
 
 34. Thotnasia papulicola (Thomas) 
 
 35. Thomasia salicola (Essig) 
 
 44. Tuherolachnus viminaU,s (Fonsc.) 
 Samhucus (elderberry) 
 
 145. Aphis samhucifoliac Fiteh 
 
 81. Macrosiphum stanleyi Wilsou 
 104. Rhopalosiphum persicac (Sulz.) 
 Sanicula (gambleweed) 
 
 119. Ai^his cornifoliae Fitcli 
 104. Ehnpalosiphum pcrsicae (Sulz.) 
 Scrophularia (figwort) 
 
 99. Pkorodon scrophulariae Thamos 
 Senecio (German ivy, ivy seneeio) 
 144. Apliis senecio Swain 
 
 88. Myzus circumflexus (Buckton) 
 104. Bhitpidosiphum persicae (Sulz.) 
 Shepherd 's-pursc, sec Capsella 
 Silyhum (milk thistle) 
 
 121. Aphis euonomi Fabr. 
 130. Aphis marutae Oestlund 
 Sisymbrium (hedge mustard) 
 
 88. Myzus eircumflexus (Bucktou) 
 Smilax, see Asparagus 
 Snowball, see Virburnum 
 Snowberry, see Symphoricarpos 
 Solanum (potato, nightshade) 
 
 'iG. Amphorophora latysiphon Davidson 
 79. Macrosiphum solanifolii (Ashmead) 
 88. Myzus circumflexus (Buckton) 
 102. Bhopalosiphum lactucae (Kalt.) 
 104. Rhopalosiphum persicae (Sulz.) 
 163. Trif'daphis radicicola (Essig) 
 Sonchus (sow thistle) 
 
 79. Macrosiphum solanifolii (Ashmead) 
 
 80. Macrosiphum sonchella (Monell) 
 102. Ehopalosiphum lactucae (Kalt.) 
 104. Bhopalosiphum persicae (Sulz.) 
 
 Sorghum 
 
 129. Aphis maidis Fiteh 
 Sorrell, see Eumex 
 Sow thistle, see Sonchus 
 Spinacia (spinach) 
 
 123. Aphis gossyj'ii Glover 
 
 104. Bhopalosiphum persicae (Sulz.) 
 Spirea (spirea) 
 
 148. Apliis spiraecola Patch 
 
A SYNOPSIS OF THE APHIDIDAE 175 
 
 Spruce, see Picea 
 
 Squash, see Cucurhita 
 Stacliijs (hedge nettle) 
 
 73. Macrosiphum ludoviciaiiae (Oestlund) 
 88. Myzus circumfltxus (Buekton) 
 Stock, Ten-week, see Matthiola 
 Strawberry, see Fragaria 
 Sti'awberry tree, see Arhutus 
 Sugar beet, see Beta 
 Sunflower, see Heliantlius 
 Sweet clover, see Melilotus 
 Sweet pea, see Lathyrus 
 Sycamore, Western, see Platanns 
 Symphoricarpos (snowberry) 
 
 108. Aphis albipes Oestlund 
 Syringa (lilae) 
 
 131. Aphis medicaginis Koch 
 Tacalote, see Centaurea 
 Taraxacum (dandelion) 
 
 82. Macrosiphum taraxici (Kalt.) 
 Tarweed, see Madia and Hemitonia 
 Teasel, Fuller 's, see Dipsacus 
 
 Teasel, mustard, see Erysimum 
 Thimbleberry, see iv«&«« 
 Thistle, see Cirsium 
 Thistle, Milk, see Silyhum 
 Thistle, Sow, see Sonclnis 
 Thuja (arbor vitae) 
 
 .54. Lachnus tiijafiiinus (Del Guercio) 
 Tilia (linden, basswood) 
 
 10. Eucallipterus tiliae (Linn.) 
 Tomato, see Lycopersicum 
 Trifoliiim (clover) 
 
 llfia. Aphis hal'eri Cowen 
 131. Aphis medicaginis Koch 
 Trigonella (fenugreek) 
 
 69. Macrosiphum pisi (Kalt.) 
 Triticum (wheat) 
 
 111. Aphis avenae Fabr. 
 
 64. Macrosiphum granarium (Kirby) 
 Tropaeolum (nasturtium) 
 
 121. Aphis euonomi Fabr. 
 
 88. Myzus circumflcxus (Buekton) 
 104. Ehopalosiphum persicae (Sulz.) 
 Tule, California, see Typha 
 Tulip, see Tulipa 
 Tulip tree, see Liriodendron 
 Tulipa (tulip) 
 
 83. Macrosiphum tulipae (Monell) 
 104. Shopalosiphum persicae (Sulz.) 
 
 Turnip, see Brassica 
 
176 MISCELLANEOUS STUDIES 
 
 Typha (California tule) 
 
 111. Aphis avenac Fabr. 
 153. Eyalopienis arundinis (Fabr.) 
 68. Macrosiphum granarium (Kirby) 
 
 156. Siphoeoryne nymphaeae (Linn.) 
 Ulmus (elm) 
 
 28. Aretaphis populifolii (Essig) (?) 
 
 172. Colopha ulmicola (Fitch) 
 
 173. Eriosoma americana (EUey) 
 
 175. Eriosoma languinosa Hartig {pyricoJa B. & D.) 
 
 174. Eriosoma lanigcrum (Hausman) 
 
 79. Macrosiphum solanifolii (Ashmead) 
 22. Myzocallis ulmifolii (Monell) 
 Umiellularia (California laurel) 
 
 88. Myzus circumflexus (Buckton) 
 104. Ehopalosiphum. pcrsicae (Sulz.) 
 
 157. Siphoeoryne pastinacae (Linn.) 
 Urtica (stinging nettle) 
 
 121. Apliis cuonomi Fabr. 
 Valerian, Red, see Ccntranthus 
 Valeriana 
 
 84. Macrosiphum Valerianae (Clarke) 
 Vernoiiia (ironweed) 
 
 123. Aphis gossypii Glover 
 Vetch, see Vicia 
 
 Vihurnum (lauristinus, snowball) 
 121. Aphis euonomi Fabr. 
 139. Aphis pomi Dc Geer 
 150. Aphis viburnicolens u.sp. 
 Vicia (horse bean, vetch) 
 
 121. Aphis euonomi Fabr. {fabae Scop.) 
 131. Aphis medicaginis Koch 
 74. Macrosiphum pisi (Kalt.) 
 Vigna (blackeye bean, cowpea) 
 
 131. Aphis medicaginis Koch 
 Vineca (periwinkle) 
 
 56. Amphorophora latysiphon Davidson 
 88. Mysus cireumfiexus (Buckton) 
 104. Ehopalosiphum pcrsieae (Sulz.) 
 Viola (pansy, violet) 
 
 58. Idiopterus nephrelepidis Davis 
 74. Macrosiphum pisi (Kalt.) 
 88. Myzus cireumfiexus (Buckton) 
 106. Ehnpalosiphum vioUie Pergande 
 Vitis (grape) 
 
 180. Phylloxera vitifoliae (Fitch) 
 Wallflower, iee Chciranthus 
 "Wallflower, Western, see Erysimum 
 Walnut, see Juglans 
 Water hemlock, see Cicuta 
 Watermelon, see CitruUus 
 Water plantain, see Alisma 
 
A SYNOP^ila OF THE APHIDIDAE 177 
 
 Wheat, see Triticu7n 
 Willow, see Salix 
 Yarrow, see Achillea 
 Yucca (yucea) 
 
 151. Ai'ltis yuccae Cowen 
 Zantedeschia (calla) 
 
 88. Myzus circumflexus (Buckton) 
 Zea (corn) 
 
 111. Aphis avenae Fabr. 
 
 128. Aphis maidis Fitch 
 Zizia 
 
 155. Siphocoryne capreae (Fabr.) 
 
178 MISCELLANEOUS STUDIES 
 
 ADDENDA 
 
 Since the preparation of this manuscript there have appeared a few papers^i' 
 in which there are some new records for certain of the California Aphididae 
 and in which there are notes concerning the synonymy of some of the species. 
 These records are noted here and are listed in the Host Plant Index (appendix 2). 
 
 2. PhyUaphis fagi (Linn.) on Fagus tricolor, Oakland (Essig, p. 321). 
 
 7. Euceraphis betulae (Koch) on Betula populifoHu laciniata and B. papy- 
 rifera (Essig, pp. 322-323). 
 
 10. EucaUipterus tiliae (Linn.) on TiUa tomcntosn, Berkeley (Essig, p. 323). 
 Baker places this species in the genus MyzocalUs, for although it is quite distinct 
 from the type of Mii^ocaUis. various species form definite connections leading to 
 this one. 
 
 15. MyzocalUs castanicola Baker (Baker, p. 424). This name has been sug- 
 gested by Baker to replace M. castaneae (Buekton) (preoccupied by castaneae 
 (Fitch)). Therefore the name suggested by the author, M. davidsoni Swain, 
 must be dropped. E.ssig (p. 323) lists M. castaneae (Fitch), but he refers to this 
 species. 
 
 19. MyzocalUs caUforuicus Baker (Baker, pp. 421-422). This is the same 
 species as described by the author under the name, MyzocalUs niaureri Swain, 
 which name will have to be dropped, and replaced by M. ealifornicus Baker. 
 
 53a. Lachnus tomentosus (De Geer), on Finns radiata, Berkeley (Gillette, 
 pp. 140-141). This species is very similar to L. pini-radiatae Davidson, accord- 
 ing to Gillette. The author finds on looking over his specimens that some of them 
 labeled L. pini-radiatae Dvdn. are this species, particularly those taken on the 
 campus at Berkeley. 
 
 56. Amphorophora latysiphon Davidson, on Clirysantlunnim and Primula sp., 
 Berkeley (Essig, p. 329). 
 
 68. Macrosiphum granarium (Kirby), on Alopccurns pratensis, Ampdodesma 
 tenax, and Elyinus sji., Martinez (Essig, p. 328). 
 
 76. Macrosiphum rosae (Linn.) on Dipsaciis fulloiium and Ccntrantlms ruber, 
 Berkeley (Essig, p. 329). 
 
 79. Macrosiphum solanifolii (Ashmead), on Achillea millefolium and Pitto- 
 sporum tubira, Berkeley, and on Ulmus americanu^, San Francisco (Essig, p. 329). 
 
 88. Myzus circuraflexus (Buekton), on Lilium spp., Pentstemon spectabilis, 
 and Umbelhdaria ealifornica. Berkeley (Essig, p. 335). 
 
 102. Rhopalosiphum lactucae (Kalt.). Dobrovliansky lists this as a synonym 
 of E. ribis (Buekton), giving the latter name preference. 
 
 30 Baker, A. C, Eastern aphids, new and little known, II, Jour, Econ, Ent., 
 vol. 10, pp. 421-433, 1917. 
 
 Baker, A. C. The correct name for our apple-grain aphis. Science, vol. 46, 
 pp. 410-411, 1917. 
 
 Davidson, W. M., The reddish-brown plum aphis, Jour. Econ. Ent., vol. 10, 
 pp. 330-353. 1917. 
 
 Dobrovliansky, V. V., A list of aphids found on cultivated plants in the gov- 
 ernment of Kharkov, in Pests of Agriculture, Kharkov, Bull, 1916; reviewed in 
 Rev. Appl. Ent., vol. 5, pp. 561-562, 1917. 
 
 Essig, E. O., Aphididae of California, Univ. Calif. Publ. Entom., vol. 1, pp. 
 301-346, 1917. 
 
 Gillette, C. P., Some Colorado species of the genus Luehniis. Ent. Soc, Am,, 
 vol, 10, pp. 133-146, 1917. 
 
 Van der Goot, P.. Zur Kenntnis der Blattliiuse Java's, in Contrib. a la fauna 
 der Indes neerlandaises, vol. 1, pp. 1-301, 1916. 
 
A SYNOPSIS OF THE APEIDIDAE 179 
 
 104. Rhopalosiphum persicae (Sulzer), on Bnccharis douglasii, CentraiUhus 
 ruber, Clarkia ctiytin.s, Dipsdcus fullonum, Escallonia pulverulcnta, Rclianthus 
 animus, Lavatcra assurgentiflora, Liriodendron Udipifera, Mtiianthus major, 
 Pcntstcmon spcctabilis, Pittosporum spp., and Vmbcllularia califurnica, Berkeley 
 (Essig, pp. 331-332). 
 
 111. ApMs avenae Pabr. It would appear from a study of Baker's paper in 
 Science tliat the common California species is Aphis prunifoliae Fitch. It is 
 certain that it is distinct from A. cerasifoliae Fitcli, which has been taken here 
 once and is described in this paper. If it is possible, as Baker says, that A. 
 cerasifoliae Fitch is a synonym of A. padi Linn., then our common species must 
 be known as A. prunifoliae Fitch. From the brief description of Aphis (Siphon- 
 aphis) padi Linn, given by Van der Goot (pp. 71-72) it would appear that our 
 species may be distinct, differing slightly in the comparative lengths of the 
 cornicles and cauda. Consequently the author favors accepting the name. Aphis 
 pimnifoliac Fitoh, for this species. 
 
 123. Aphis gossypii Glover, on Asclepias speciosa, A. vestita, Lilium spcciosum 
 rubrum, Lonicera sp., and Ehamnus purshiana, Berkeley and Oakland (Essig, 
 pp. 338-339). 
 
 131. Aphis medicaglnis Koch, on Citrxis sp., Sacramento, ami on Vigna sin- 
 ensis, Moorjiark (Essig, p. 340). 
 
 139. Aphis pomi De Geer, on Cotoneaster franchetii, Pittosporum eugenioides, 
 and Viburnum tinus, Berkeley (Essig, p. 341). The author is inclined to believe 
 this to be Aphis viburnicolens n.sp. (see no. 150) which is quite similar to Ajyhis 
 pomi De Geer, but which is common on Fiburimm and related plants. He has not, 
 however, seen Essig 's specimens, so can not state positively whether or not it is 
 this species. 
 
 140. Aphis prunoTum Dobr. Dobrovliansky places this species as a synonym 
 of Siphocoryne nymphaeae (Linn.). This author noted the similarity of these 
 two, but was not certain of their identity, so listed them as distinct species. 
 
 141. Aphis pseudobrassicae Davis. Dobrovliansky believes this to be a syno- 
 nym of Aphis erysinii Kalt. 
 
 146. Aphis senecio Swain. Essig (p. 337) lists Aphis bal-eri Cowen from 
 Trifolium pratense. This proves to be the true Aphis bal-eri Cowen and not 
 A. senecio Swain, which is the species that has been hitherto called A. bal-eri 
 Cowen In California. 
 
 152. Toxoptera aurantii (Fonsc.) on Camellia japonica, Oakland (Essig, p. 
 330). 
 
 153. Hyalopterus arundinis (Fabr.). Both Dobrovliansky and Van der Goot 
 list this as a synonym of B. pruni (Fabr.) giving the later preference. Accord- 
 ing to Hunter, arundinis should have priority, but it is entirely possible that the 
 dates he gives are incorrect. This point the author is unable to settle as he has 
 not access to Fabrieius' works. 
 
 156. Siphocoryne nymphaeae (Linn.). Davidson gives a brief account of the 
 habits and biology of this species, as well as a description of the various forms. 
 
 175. Eriosoma languinosa Hartig (pyricola Baker and Davidson). The 
 species listed by Essig (p. 345) as Eriosoma sp. on Vlmus campestris in Berkeley 
 and in Hayward is this species. 
 
 115. Aphis cari Essig. Davidson recently remarked to the author that he 
 could see no difference betiveen this species and Aphis heliantliii Monell. It is 
 quite possible that these are synonyms. 
 
EXPLANATION OF PLATES 
 PLATE 1 
 
 Fig. 1. Myzocalli^ asclepwdis (Fitch), tarsus and claw. 
 
 Fig. 2. Aphis senecio Swain, tarsus and claw. 
 
 Pig. 3. Essigella calif ornica (Essig), sixth antennal segment and spur. 
 
 Fig. 4. Aphis senecio Swain, sixth antennal segment and spur. 
 
 Fig. 5. Essigella californica (Essig), eauda and anal plate (lateral view). 
 
 Fig. 6. Aphis senecio Swain, eauda and anal plate (lateral view). 
 
 Fig. 7. Eucallipterus tiliae (Linn.), eauda and anal plate. 
 
 Fig. 8. Thomasia populicola (Thos.), eauda and anal plate. 
 
 Fig. 9. Phyllaphis fagi (Linn.), third antennal segment. 
 
 Fig. \0. Phyllaphis fagi (Linn.), sixth antennal segment. 
 
 Fig. 11. Phyllaphis fagi (Linn.), eauad and anal plate. 
 
 Fig. 12. Phyllaphis fagi (Linn.), front of head and antennal tubercles. 
 
 Fig. 13. Phyllaphis coweni (Ckll.), Antenna. 
 
 Fig. 14. Phyllaphis quercicola Baker, third antennal segment. 
 
 Fig. 1.5. Phyllaphis quercicola Baker, fourth antennal segment. 
 
 Fig. Iti. Phyllaphis quercicola Baker, fifth antennal segment. 
 
 Fig. 17. Phyllaphis quercicola Baker, sixth antennal segment. 
 
 Fig. 18. Phyllaphis quercicola Baker, forcwing. 
 
 Fig. 19. Phyllaphis quercicola Baker, eauda and anal plate. 
 
 Fig. 20. Phyllaphis querci, tarsal claw. 
 
 Fig. 21. Drepanosiphum platanoides (Schr.), antennal tubercles. 
 
 [180] 
 
( SWAIN ] PLATE 1 
 
PLATE 2 
 
 Fig. 22. Mi/zocallis arundicoUns (Clarke), antennal tubercles. 
 
 Fig. 23. Drcpaiuiphis acerifolii (Thomas), cornicle. 
 
 Fig. 24. Drepanosiphum platanoides (Schr.), cornicle. 
 
 Fig. 25. Moncllia caryella (Fitch), cornicle. 
 
 Fig. 26. My^ocallis ielius (Walsh), cornicle. 
 
 Fig. 27. Calaphis betulaccolrns (Fitch), antennal tubercles. 
 
 Fig. 28. Calaphis bctulclla Walsh, antennal tubercles. 
 
 Fig. 29. Euceraphis hetulae (Koch), antennal tubercles. 
 
 Fig. 30. Ewalliptcrus tili-ae (Linn.), sixth antennal segment and spur. 
 
 Fig. 31. Myzocaliis quercus (Kalt.), sixth antennal segment and spur. 
 
 Fig. 32. Myzocaliis quercus (Kalt.), cornicle. 
 
 Fig. 33. Eucallipterus tiliae (Linn.), cornicle. 
 
 Fig. 34. Chromaphis juglandicola (Kalt.), sixth antennal segment and spur. 
 
 Fig. 35. Chromaphis juglandicola (Kalt.), cornicle. 
 
 Fig. 36. Drepanosiphum platanoides (Schr.), third antennal segment. 
 
 Fig. 37. Drepanaphis acerifolii (Thomas), third antennal segment. 
 
 Fig. 38. Calaphis betulae-colens (Fiteh), third antennal segment. 
 
 Fig. 39. Euceraphis gillettei Dvdn., base of third antennal segment. 
 
 Fig. 40. Euceraphis betulae (Koch), base of third antennal segment. 
 
 [182] 
 
SWAIN 1 plate: 2 
 
PLATE 3 
 
 Fig. 41. Eucallipterus flava (Dvdn.), base of third antennal segment. 
 
 Fig. 42. Eucallipterus tiliac (Linn.), third antennal segment. 
 
 Fig. 43. MyzocalUs coryli (Goetze), third antennal segment. 
 
 Fig. 44. MyzocalUs coryli (Goetze), sixth antennal segment and spur. 
 
 Fig. 45. MyzocalUs hellus (Walsh), sixth antennal segment and spur. 
 
 Pig. 46. MyzocalUs iellus (Walsh), third antennal segment. 
 
 Fig. 47. MyzocalUs alnifoliae (Fitch), third antennal segment. 
 
 Fig. 48. MyzocalUs arundicolens (Clarke), third antennal segment. 
 
 Fig. 49. Eucallipterus tiliac (Linn.), cornicle. 
 
 Fig. 50. Eucallipterus tiliac (Linn.), anal plate. 
 
 Fig. 51. MyzocalUs arundicolens (Clarke), cornicle. 
 
 Fig. 52. MyzocalUs arundicolens (Clarke), anal plate. 
 
 Fig. 53. MyzocalUs coryli (Goetze), cornicle. 
 
 Fig. 54. MyzocalUs coryli (Goetze), anal plate. 
 
 Fig. 55. MyzocalUs californicus Baker, third antennal segment. 
 
 Fig. 56. MyzocalUs californicus Baker, sixth antennal segment and spur. 
 
 Fig. 57. MyzocalUs pasaniae Dvdn., third antennal segment. 
 
 Fig. 58. MyzocalUs quercus (Kalt.), third antennal segment. 
 
 Fig. 59. MyzocalUs ulmifoUi (MoneU), third antennal segment. 
 
 Fig. 60. MyzocalUs castanicola Baker, third antennal segment. 
 
 Fig. 61. MyzocalUs castanicola Baker, eauda and anal plate. 
 
 Fig. 62. MyzocalUs castanicola Baker, cornicle. 
 
 Fig. 63. Calliptcrus californicus (Essig), sixth antennal segment ami spur. 
 
 Fig. 64. CalUpterus californicus (Essig), third antennal segment. 
 
 Fig. 65. CalUpterus caryae Monell, third antennal segment. 
 
 [1S4J 
 
oO OjLiiL 
 
 ^--.^^^^iiyiV>M!M^ 
 
 c_ji-—^ ^ 
 
 ""c c S> ^?'~~ 
 
 ^^ 
 
 55 
 56 
 
 3^512 
 
 aTTnrxBixx 
 
 43 
 
 f.nO„^r)()„ 
 
 3r 
 
 t: — c-^' — p-r-°'^ _;i_ 
 
 [ SWAIN ] PLATE 3 
 
PLATE 4 
 
 Fig. 66. Callipterus caryae Monell, sixth antennal segment and spur. 
 
 Fig. 67. Monellia carycUa (Fitch), sixth antennal segment and spur. 
 
 Fig. 68. Monellia caryella (Fitch), third antennal segment. 
 
 Fig. 69. Arctaphis populifolii (Essig), eauda. 
 
 Fig. 70. Micrella monclla Essig, cauda. 
 
 Fig. 71. Arctaphis populifolii (Essig), third antennal segment. 
 
 Fig. 72. Micrella vionella Essig, third antennal segment. 
 
 Fig. 73. Syviydobius macrostachyae Essig, third antennal segment. 
 
 Fig. 74. Symydobi)ts salicicorticis Essig, third antennal segment. 
 
 Fig. 75. Fullawaya saliciradicis Essig, third antennal segment. 
 
 Fig. 76. TlwinaMa crucis Essig, third antennal segment. 
 
 Fig. 77. Thomasia populicola (Thomas), third antennal segment. 
 
 Fig. 78. Thomasia salicicola (Essig), third antennal segment. 
 
 Fig. 79. Lachnus ferrisi Swain, tarsal claw. 
 
 Fig. 80. Pterocomma popiilifoliae (Fitch), tarsal claw. 
 
 Fig. 81. Pterocomma floeculosa (Weed), cornicle. 
 
 Fig. 82. Pterocomma populifoliae (Fitch), cornicle. 
 
 Fig. 83. Essigella californica (Essig), antenna. 
 
 Fig. 84. Longistigma sp., front wing. 
 
 L1S6J 
 
rmsjmsiisim^simsu,^ 44 
 
 w 
 
 t 
 
 
 
 70 
 
 fcS 
 
 yrf7^;te^ ^j^^yn nr liS^^^^]jSS^ 
 
 y^ <•» a 
 
 r 
 
 ■■T 
 
 £gSSP 3}:a®^I^ 
 
 77 
 
 : SWAIN ] PLATE 4 
 
PLATE 5 
 
 Fig. 85. Lacknii^ sp., front wing. 
 
 Fig. 8G. Tuberolachims viminalis (Fonsc), hind tarsus. 
 
 Fig. 87. Eulachnus rileyi Davis, hind tarsus. 
 
 Fig. 88. Lachnus vanduzei n.sp., third antennal segment. 
 
 Fig. 89. Laclinus ferrisi Swain, first, second, and third antennal segments. 
 
 Fig. 90. Laohnus ferrisi Swain, fourth, fifth, and sixth antennal segments. 
 
 Fig. 91. Laclinus ferrisi Swain, cornicle. 
 
 Fig. 92. Lachnus pseudotsugae Wilson, tip of front wing. 
 
 Fig. 93. Lachnus tujafiUnus (Del Guereio), tip of front wing. 
 
 Fig. 94. Lachnus occidentalis Dvdn., third antennal segment. 
 
 Fig. 95. Laclinus pini-radiatae Dvdn. (?), third antennal segment. 
 
 Fig. 9(i. Lachnus glchnus Essig, third antennal segment. 
 
 Fig. 97. Lachnus glelinus Essig, cornicle. 
 
 Fig. 98. Lachnus pseudotsugae Wilson, third antennal segment. 
 
 Fig. 99. Lachnus taxifolia Swain, hind tarsus. 
 
 Fig. 100. Lachnus taxifolia Swain, fourth, fifth and sixth antennal segments. 
 
 Fig. 101. Lachnus taxifolia Swain, first, second, and third antennal segments. 
 
 [188] 
 

 
 SWAIN I PLATE 5 
 
Fig. 
 
 102. 
 
 Fig. 
 
 103. 
 
 Fig. 
 
 104. 
 
 Fig. 
 
 105. 
 
 Fig. 
 
 106. 
 
 Fig. 
 
 107. 
 
 Fig. 
 
 108. 
 
 Fig. 
 
 109. 
 
 Fig. 
 
 110. 
 
 Fig. 
 
 111. 
 
 Fig. 
 
 112. 
 
 Fig. 
 
 113. 
 
 Fig. 
 
 114. 
 
 Fig. 
 
 115. 
 
 Fig. 
 
 116. 
 
 PLATE 6 
 
 Laclinus taxi folia Swain, wing. 
 
 Liiclmus taxifoUa Swain, cornicle of apterous female. 
 
 Lachnus pondcrosa Williams, third antennal segment. 
 
 Laclinus tujafillnus (Del Guercio), third antennal segment. 
 
 Macrosiplmm rosae (Linn.), antennal tubercles. 
 
 Nectaro-siphon rubicola (Oest.), antennal tubercles. 
 
 Rhopalosiphum persicae (Sulz.), antennal tubercles. 
 
 N ectarosiphon rubicola (Oest.), cornicle. 
 
 Idiopterus ncplirclepidis Davis, wing. 
 
 Amphorophora rubi (Kalt. ), antennal tubercles. 
 
 Myzus cerasi (Fabr.), antennal tubercles, 
 
 Amphorophora ruhi (Kalt.), cornicle. 
 
 Toxoptera aurantii (Fonsc), cornicle. 
 
 Phorodon humuli (Sehr.), antennal tubercles of alate females. 
 
 Phorodon humnli (Schr.), antennal tubercles of apterous females. 
 
 |l!m| 
 
[SWAIN ] PLATE 6 
 
PLATE 7 
 
 Fig. 117. Phorodon humuli Schr., cornicle. 
 
 Fig. 118. Phorodon humuli Schr., eauda. 
 
 Pig. 119. Ehopalosiphum persicae (Sulz.), cornicle. 
 
 Pig. 120. lihopalosiphum peswae (Sulz.), cauda. 
 
 Pig. 121. Miizus cerasi (Fabr.), cornicle. 
 
 Pig. 122. Myzus cerasi (Fabr.), cauda. 
 
 Fig. 123. N ectarosiphon rubicola (Oest.), cauda. 
 
 Pig. 124. N ectarosiphon morrisoni Swain, antennal tubercles. 
 
 Fig. 125. Nectarosiplion morrisoni Swain, third antennal segment. 
 
 Fig. 126. N ectarosiphon morrisoni Swain, cauda. 
 
 Pig. 127. N ectarosiphon morrisoni Swain, cornicle. 
 
 Pig. 128. Macrosiphiim stanlcyi Wilson, cornicle. 
 
 Fig. 129. Macrosiphum solanifolii (Ashm.) (from Snnchus), cornicle. 
 
 Fig. 130. Macrosiphum pi^i (Kalt.), cornicle. 
 
 Fig. 131. Macrosiphum calif ornicum (Clarke), third antennal segment. 
 
 Fig. 132. Macrosiphum calif ornicum (Clarke), cornicle. 
 
 Fig. 133. Macrosiphmn cucurbitae (Thomas), third antennal segment. 
 
 Pig. 134. Macrosiphum cucurbitae (Thomas), cornicle. 
 
 Pig. 13.5. Macrosiphum granarium (Kirby), third antennal segment. 
 
 Pig. 136. Macrosiphum ludovicianae (Oest.), third antennal segment. 
 
 Fig. 137. Macrosiphum solanifolii (Ashm.), cornicle. 
 
 Fig. 138. Macrosiphum solanifolii (Ashm.), third antennal segment. 
 
 [192] 
 
[ SWAIN 1 PLATE 7 
 
PLATE 8 
 
 Fig. 139. Macrosiplium solanifolii (Ashm.) (from Citrns), cornicle. 
 
 Fig. 140. Macrosiplium solanifolii (Ashm.) (from Citrus), third antennal 
 segment. 
 
 Fig. 141. Macrosiphum sanboryii Gillette, cornicle of apterous female. 
 
 Fig. 142. Macrosiphum artemisiae (Fonsc), cornicle. 
 
 Fig. 143. Macrosiphum albifrons Essig, third antennal segment. 
 
 Fig. 144. Macrosiphum albifrons Essig, cornicle. 
 
 Fig. 145. Macrosiplium artemisiae (Fonsc), third antennal segment. 
 
 Fig. 146. Macrosiphum artemisicola (Williams), third antenal segment. 
 
 Fig. 147. Macrosiphum artemisicola (Williams), cornicle. 
 
 Fig. 148. Macrosiphum granarium (Kirby), cornicle. 
 
 Fig. 149. Macrosiphum ludovicianae (Oest.), cornicle. 
 
 Fig. 150. Macrosiphum pisi (Kalt.), third antennal segment. 
 
 Fig. 151. Macrosiphum rosae (Linn.), third antennal segment. 
 
 Fig. 152. Macrosiphum rosae (Linn.), cornicle. 
 
 Fig. 153. Macrosiphum rudbechi-ae (Fitch), cornicle. 
 
 Fig. 154. Macrosiphum rudheclciae (Fitch), third antennal segment. 
 
 Fig. 155. Macrosiphum sanborni Gillette, cauda apterous female. 
 
 Fig. 156. Macrosiphum dirhodum (Walker), cornicle. 
 
 Fig. 157. Macrosiplium dirhodum (Walker), third antennal segment. 
 
 Fig. 158. Macrosiphum stanleyi Wilson, third antennal segment. 
 
 Fig. 159. Macrosiphum solanifolii (Ashm.) (from Souclius), third antennal 
 segment. 
 
 Fig. 160. Macrosiphum solanifolii (Ashm.) (from Sonchti-s), cauda. 
 
 I 194] 
 
[SWAIN ] PLATE 8 
 
Fig. 161. 
 
 Fig. 162. 
 
 Fig. 163. 
 
 Fig. 164. 
 
 Fig. 165. 
 
 Fig. 166. 
 
 Fig. 167. 
 
 Fig. 168. 
 
 Fig. 169. 
 segment. 
 
 Fig. 170. 
 
 Fig. 171. 
 ment. 
 
 Fig. 172. 
 
 Fig. 173. 
 
 Fig. 174. 
 
 Fig. 175. 
 
 Fig. 176. 
 
 Fig. 177. 
 
 Fig. 178. 
 
 Fig. 179. 
 
 Fig. 180. 
 
 Fig. 181. 
 
 Fig. 182. 
 
 Fig. 183. 
 
 Fig. 184. 
 
 Fig. 185. 
 ments. 
 
 Fig. 186. 
 
 PLATE 9 
 
 Amplwrophora latysiphon Dvdn., cornicle. 
 
 Amphoropliora rubi (Kalt.), cauda. 
 
 Toxoptera aurantii (Fonsc), third antennal segment. 
 
 Ehopalosiphum violae Pergande, wing. 
 
 Bhopalosiphum hippophaes Koch, cornicle. 
 
 Shopalosiphum nervatum Gillette (from Arbutus), wing. 
 
 Shopalosiphum corijUnum Dvdn., third antennal segment. 
 
 Bhopalosiphum persicae (Sulz.), third antennal segment. 
 
 Jthopdlosiphum nervatum Gillette (from Arhutus), third antennal 
 
 Bhopalosiphum hippophaes Koch, third antennal segment. 
 Bhopalosiphum nervatum GUlette (from rose), third antennal seg- 
 
 Siphocori/ne niimphaeae (Linn.), third antennal segment. 
 
 Bhopalosiphum rhois Monell, third antennal segment. 
 
 Bhopalosiphum violae Pergande, third antennal segment. 
 
 Myzus circumflexus (Buckton), third antennal segment. 
 
 Mysus braggii Gillette, third antennal segment. 
 
 Myzus fragaefolii Ckll., third antennal segment. 
 
 Myzus rhamni (Fonsc), third antennal segment. 
 
 Myzus cerasi (Fabr.), third antennal segment. 
 
 Myzus ribis (LLnn.), third antennal segment. 
 
 Hyalopterus arundinis (Fabr.), cornicle. 
 
 Aphis cuonomi Fabr., cornicle. 
 
 Siphocoryne caprcae (Fabr.), cornicle. 
 
 Liosomaphis berberidis (Kalt.), conricle. 
 
 Hyalopterus arundinis (Fabr.), third and fourth antennal seg- 
 
 Hyalopterus arundinis (Fabr.), cauda. 
 
 [196] 
 
-J^^T T^T^.o.o ° ° l^r^"^^ " "3^ 
 
 SWAIN 1 PLATE 9 
 
PLATE 10 
 
 Fig. 187. Aphis euonomi Fabr., wing. 
 
 Fig. 188. Aphis salicicola Thomas, wing. 
 
 Fig. 189. Aphis medicaginis Koch, third and fourth antennal segments. 
 
 Fig. 190. Aphis euonomi Fabr. (?), third and fourth antennal segments. 
 
 Fig. 191. Aphis avenue Fabr., wing. 
 
 Fig. 192. Aphis gossijpii Glover, cornicle. 
 
 Fig. 193. Aphis gossypii Glover, cauda. 
 
 Fig. 194. Aphis samhucif oliae Fitch, cauda. 
 
 Fig. 195. Aphis samhucif oliae Fitch, cornicle. 
 
 Fig. 196. Myzaphis abietina (Walker), third and fourth antennal segments. 
 
 Fig. 197. Myzaphis abietina (Walker), cornicle. 
 
 Fig. 198. Aphis albipes Oest., eoniiele. 
 
 Fig. 199. Aphis albipes Oest., cauda. 
 
 Fig. 200. Aphis albipes Oest., third and fourth antennal segments. 
 
 Fig. 201. Aphis avenae Fabr., cornicle. 
 
 Fig. 202. Aphis avenae Fabr., third and fourth antennal segments. 
 
 Fig. 203. Aphis brassicae Linn., cornicle. 
 
 Fig. 204. Aphis brassicae Linn., third and fourth antennal segments. 
 
 Fig. 205. Aphis euonomi Fabr., cornicle. 
 
 Fig. 206. Aphis euonomi Fabr., cornicle. 
 
 Fig. 207. AiJhis euonomi Fabr., third and fourth antennal segments. 
 
 Fig. 208. Aphis cardui Linn., third and fourth antennal segments. 
 
 Fig. 209. Aphis cardui Linn., cornicle. 
 
 Fig. 210. Aphis ceanolhi Clarke, cornicle. 
 
 Fig. 211. Aphis ceanothi Clarke, third and fourth antennal segments. 
 
 Fig. 212. Aphis cookii Essig, third and fourth antennal segments. 
 
 Fig. 213. Aphis cookii Essig, cauda and anal plate. 
 
 Fig. 214. Aphis cookii Essig, cornicle. 
 
 Fig. 215. Aphis gossypii Glover, third and fourth antennal segments. 
 
 Fig. 216. Aphis maidis Fitch, cauda. 
 
 Fig. 217. Aphis maidis Fitch, antenna. 
 
 Fig. 218. Aphis maidis Fitch, cornicle. 
 
 Fig. 219. Aphis middletonii Thomas, cornicle. 
 
 Fig. 220. Aphis middletonii Thomas, third and fourth antennal segments. 
 
 Fig. 221. Aphis nerii Ponsc, cornicle. 
 
 Fig. 222. Aphis nerii Fonsc, third and fourth antennal segments. 
 
 Fig. 2231 Aphis persicae-niger Smith, cornicle. 
 
 Fig. 224. Aphis persicae-niger Smith, third and fourth antennal segments. 
 
 [198] 
 
cz^ 
 
 
 I SWAIN 1 PLATE 10 
 
PLATE 11 
 
 Fig. 225. 
 
 Fig. 226. 
 
 Fig. 227. 
 
 Fig. 228. 
 
 Fig. 229. 
 
 Fig. 230. 
 
 Fig. 231. 
 
 Fig. 232. 
 
 Fig. 233. 
 
 Fig. 234. 
 
 Fig. 235. 
 
 Fig. 236. 
 
 Fig. 237. 
 
 Fig. 238. 
 
 Fig. 239. 
 
 Fig. 240. 
 
 Fig. 241. 
 
 Fig. 242. 
 
 Fig. 243. 
 
 Fig. 244. 
 
 Fig. 245. 
 
 Fig. 24G. 
 
 Fig. 247. 
 
 Fig. 248. 
 
 Fig. 250. 
 
 Fig. 251. 
 
 Fig. 252. 
 ments. 
 
 Fig. 253. 
 
 Fig. 254. 
 and spur. 
 
 Fig. 255. 
 females. 
 
 Fig. 256. Siphocoryne caprcae (Fabr.), cauda and supra-caudal spine of 
 apterous females. 
 
 Fig. 257. Siphocoryne pa.^tinacae (Linn.), tliird and fourth antenna! seg- 
 ments. 
 
 Fig. 258. Siphocoryne pastinacae (Linn.), fifth and sixth antennal segments 
 and spur. 
 
 Fig. 259. Siphocoryne pastinacae (Linn.), eauda of apterous female. 
 
 Fig. 260. Siphocoryne pastinacae (Linn.), cauda of alate female. 
 
 Fig. 261. Siphocoryne pastinacae (Linn.), cornicle. 
 
 Aphis pomi De Geer, cauda. 
 
 Aphis pomi De Geer, antennae. 
 
 Aphis pomi De Geer, cornicle. 
 
 Aphis prunorum Dobr., cauda. 
 
 Aphis prunorum Dobr., third and fourth autenual segments. 
 
 Aphis prunorum Dobr., cornicle. 
 
 Aphis pscudobrassicae Davis, tliird and fourth antennal segments. 
 
 Aphis ramona Swain, antenna. 
 
 Aphis ramona Swain, front of head. 
 
 Aphis ramona Swain, cauda and anal plate. 
 
 Aphis ramona Swain, cornicle. 
 
 Aphis cuonomi Fabr., cornicle. 
 
 Aphis cuonomi Fabr., third and fourth antennal segments. 
 
 Aphis salicicola Thomas, cornicle. 
 
 Apliis salicicolii Tliomas, third and fourth antennal segments. 
 
 Apliis samhucifoliae Fitch, third and fourth antennal segments. 
 
 Aphis senecio Swain, cauda. 
 
 Aphis senecio Swain, cornicle. 
 
 Aphis senecio Swain, front of head. 
 
 Aphis senecio Swain, third and fourth antennal segments. 
 
 Aphis senecio Swain, fifth, sixth antennal segments, and spur. 
 
 Aphis setarae Thomas, cornicle. 
 
 Aphis setarae Thomas, third and fourth autenual segments. 
 
 Aphis malifoliae Fitch, cornicle. 
 
 Apliis malifoliae Fitch, fourth antennal segment. 
 
 Liosomaphis licrheridis (Kalt.), front of head. 
 
 Liosomaphis bei-beridis (Kalt.), third and fourth autenual seg- 
 
 Siphocoryne capreae (Fabr.), third and fourth antennal segments. 
 Siphocoryne capreae (Fabr.), fifth and sixth antennal segments 
 
 Siphocoryne capreae (Fabr.), cauda and supra-caudal spine of alate 
 
 [200] 
 
; SWAIN ) PLATE 1 1 
 
PLATE 12 
 
 Fig. 262. Myeocallis discolor (Monell), fore wmg. 
 
 Fig. 263. Myzoeallis discolor (Monell), third antennal segment. 
 
 Fig. 264. Myzoeallis bellus (Walsh), fore wing. 
 
 Fig. 265. Myzoeallis bellus (Walsh), third antennal segment. 
 
 Fig. 266. Myzoeallis ealifornicus BaTcer (maureri Swain), fore wing. 
 
 Fig. 267. Myzoeallis castanicola Baker (davidsoni Swain), fore wing. 
 
 Fig. 268. Myzoeallis arandinariae Essig, third antennal segment. 
 
 [202] 
 
2 tr 
 [SWAIN J PLATE 12 
 
PLATE 13 
 
 Pig. 269. Siimydobius chrysolepis Swain, head. 
 
 Fig. 270. Symydobius ohrysolepis Swain, cornicle. 
 
 Fig. 271. Symydobius chrysolepis Swain, anal plate. 
 
 Fig. 272. Symydobiibs chrysolepis Swain, antenna. 
 
 Fig. 273. Symydobius chrysolepis Swain, fore wing. 
 
 Fig. 274. Symydobius chrysolepis Swain, hind ^vLng. 
 
 Fig. 275. Thomasia popuUcola (Thomas), fore wing. 
 
 [204] 
 
275 
 
 [ SWAIN ] PLATE 13 
 
PLATE 14 
 
 Fig. 276. Toxoptera auratitii (Fonsc), fore wing. 
 
 Fig. 277. SJiopalosiphum lactucae (Kalt) head. 
 
 Fig. 278. Ehopalosiphum lactucae (Kalt.), third antennal segment, aptera. 
 
 Fig. 279. Ehopalosiphum lactucae (Kalt.), third antennal segment, alate. 
 
 Fig. 280. Ehopalosiphum lactucae (Kalt.), fourth and fifth antennal seg- 
 ments, alat€. 
 
 Fig. 281. Ehopalosiphum lactucae (Kalt.), sixth antennal segment, alate. 
 
 Fig. 282. Ehopalosiphum lactucae (Kalt.), cornicle, alate. 
 
 Fig. 283. Ehopalosiphum lactucae (Kalt.), eauda, alate. 
 
 Fig. 284. Ehopalosiphum lactucae (Kalt.), cornicle, aptera. 
 
 Fig. 284o. Ehopalosiphum lactucae (Kalt.), Cauda, aptera. 
 
 [206] 
 
2 5^43 
 
 [ SWAIN 1 PLATE 14 
 
PLATE 15 
 
 Fig. 285. Aphis viburnicolens n.sp., third antennal segment. 
 
 Fig. 286. Aphis viiurni-colens n.sp., cornicle. 
 
 Fig. 287. Aphis viburnicolens n.sp., Cauda. 
 
 Fig. 288. Aphis cerasifoliae (Fitch), head. 
 
 Fig. 289. Aphis cerasifoliae (Fitch), fifth and sixth antennal segments. 
 
 Fig. 290. Aphis cerasifoliae (Fitch), third and fourth antennal segments. 
 
 Fig. 291. Aphis cerasifoliae (Fitch), end of wing. 
 
 Fig. 292. Aphis cerasifoliae (Fitch), side of abdomen showing Cauda, cor- 
 nicle, and lateral tubercles on segments one, two, three, four, and seven. 
 
 [208] 
 
Z'iK 
 
 iZ9 
 
 2S-7 
 
 [ SWAIN ] PLATE 15 
 
PLATE 16 
 
 Fig. 293. Aphis viarutae Oest., head. 
 
 Fig. 294. Aphis marutae Oest., third and fourth antennal segments. 
 
 Fig. 295. Aphis marutae Oest., fifth and sixth antennal segments. 
 
 Fig. 296. Aphis vmrutae Oest., antenna, aptera. 
 
 Fig. 297. Aphis marutae Oest., end of abdomen, aptera. 
 
 Fig. 298. Aphis marutae Oest., cauda, alate. 
 
 Fig. 299. Aphis marutae Oest., cornicle, alate. 
 
 Fig. 300. Aphis neomexicana Ckll., var. pacifica Dvdn., tliird and fourth 
 antennal segments. 
 
 Fig. 301. Aphis neomexicana Ckll. var. pacifica Dvdn., cornicle. 
 
 Fig. 302. Aphis neomexicana CkU. var. pacifica Dvdn., cauda. 
 
 Fig. 303. Aphis yuccac Cowen, fourth, fifth, and sixth antennal segments. 
 
 Fig. 304. Aphis yuccae Cowen, third antennal segment. 
 
 Pig. 305. Aphis yiu'cae Cowen, tip of abdomen. 
 
 [210] 
 
dis 
 
 X94 
 
 Z96 
 
 300 
 
 S99 
 
 i;9r 
 
 30; 
 
 50Z 
 
 504 
 
 [ SWAIN 1 PLATE 16 
 
PLATE 17 
 
 Fig. 306. Myzus ribis (Linn.), head. 
 
 Fig. 307. Myzus cerasi (Fabr.), head. 
 
 Fig. 308. Myzaphis rosarum (Walker), head, alate. 
 
 Fig. 309. Myzaphis rosarum (Walker), third and fourth antenual segments. 
 
 Fig. 310. Myzaphis rosarum (Walker), fifth and sixth antennal segments. 
 
 Fig. 311. Myzaphis rosarum (Walker), tip of wing. 
 
 Fig. 312. Myzaphus rosarum (Walker), end of abdomen. 
 
 Fig. 313. Myzaphis rosarum (Walker), head, aptera. 
 
 Fig. 314. Myzaphis rosai-um (Walker), antenna, aptera. 
 
 Fig. 315. Myzaphis rosarum (Walker), cornicle, aptera. 
 
 Fig. 316. Myzaphis rosarum (Walker), cauda, aptera. 
 
 Fig. 317. Myzaphis rosarum (Walker), hind tarsus, aptera. 
 
 [212] 
 
307 
 
 309 
 
 31 3 
 
 3(7 
 
 SWAIN I PLATE 17 
 
A SYNOPSIS OF THE APEIDIDAE 
 
 215 
 
 INDEX TO GENERA AND SPECIES 
 
 abietes, hachnus, 47. 
 abietina, Myzaphis (Aphis), 134. 
 abietinus, Mindarus, 1.50. 
 aeei'ifolii, Drepanaphis (Siplwno- 
 
 phora, Macrosiphum), 18. 
 acliyrantcs, Sliopalosiplium, 80. 
 agrifoliae, Symydobius, 38. 
 alameilensis, Aphis, 93. 
 albifrons, Macrosiphum, 60. 
 albipes. Aphis, 93. 
 obit, Myzocaliis, 21. 
 ahiifoliae, CaUipterus, 20. 
 almfoliae Lachnus. 20. 
 alnifoliae Myzocallis, 22. 
 aInifoUae Prociphilus (Pemphigus), 
 
 146. 
 ambrosiae, Macrosiphum (Siphono- 
 
 phora), 60. 
 americana, Eriosoma (Schizoneura), 
 
 148. 
 Amphorophora, 54. 
 
 cicutae, 54. 
 
 latysiphon, 54, 178. 
 
 rubi, 54. 
 
 rubicola, 77. 
 angelicae, Aphis, 93. 
 anuulata, Callipterinella (Chaitopho- 
 
 rus), 31. 
 Aphis, 88. 
 
 abietina, 134. 
 
 alamedensis, 93. 
 
 albipes, 93. 
 
 angelicae, 93. 
 
 artemisiae, 61. 
 
 arundinis, 130. 
 
 atriplicis, 93. 
 
 aurantii, 129. 
 
 avenae, 94, 179. 
 
 bakeri, 123, 124. 
 
 bakeri, 6, 179. 
 
 bellus, 24. 
 
 berberidis, 130. 
 
 betulaecolens, 18. 
 
 brassieae, 95. 
 
 calendulicola, 96. 
 
 capreae, 132. 
 
 cardui, 96. 
 
 cari, 96, 179. 
 
 caryella, 30. 
 
 ceanothi, 96. 
 
 ceanothi-hirsut i, 96. 
 
 cerasi, 73. 
 
 cerasifoliae, 97.. 
 
 citri, 105. 
 
 cooki, 100. 
 
 cornifoliae, 100. 
 
 coryli, 25. 
 
 crataegifoliae, 100. 
 
 dirhodum, 63. 
 dryophila, 150. 
 euonomi, 101. 
 fabae, 102, 104. 
 fagi, 13. 
 frigidae, 105. 
 gossyini, 100. 
 g'ossypii, 105, 179. 
 g^ranarium, 64. 
 hederae, 106. 
 heraclei, 107. 
 houghtonensis, 107. 
 humuli, 79. 
 juglandis, 28. 
 lactucac, 82. 
 languinosa, 149. 
 lanigervm, 149. 
 lithospermi, 108. 
 lutescens, 117. 
 maidis, 94. 
 maidis, 108. 
 man. 120. 
 malifoliae, 108. 
 marutae, 112. 
 medicaginis, 114, 179. 
 middletonii, 115. 
 mori, 116. 
 neomexicaua, 116. 
 nerii, 117. 
 nymphaeae, 133. 
 oenotherae, 118. 
 oregonensis, 119. 
 padi, 94. 
 
 papaveris, 102, 104. 
 pastin-acae, 133. 
 pcrsicae, 85. 
 persicae-niger, 119. 
 pisi, 66. 
 platanoides, 17. 
 pomi, 120, 179. 
 pomi, 109. 
 populifoliae, 41. 
 pruni, 96. 
 
 prunorum, 121, 179. 
 prunifoliae, 130, 179. 
 pseudobrassicae, 122, 179. 
 quercus, 27. 
 ramona, 122. 
 rhamni, 76. 
 rosae, 67. 
 rosarum, 134. 
 rubi, 54. 
 rubiphila, 122. 
 rudbeckiae, 67. 
 rufomaculata, 137. 
 rMmicis. 101, 106. 
 salicicola, 123. 
 sambucifoliae, 123. 
 senecio, 123, 179. 
 setariae, 124. 
 
216 
 
 MISCELLANEOUS STUDIES 
 
 sorbi, 108. 
 
 spiraecola, 124. 
 
 spiraeeHa, 125, 126. 
 
 taraxici, 71. 
 
 tetrapteralis, 125. 
 
 tilMC, 21. 
 
 viburnicolens, 126, 179. 
 
 viminalis, 45. 
 
 yuccae, 45. 
 
 yuccicola, 128. 
 aquilegiae, Myzus, 73. 
 arbuti, Hhopalosiphum, 84. 
 Arctaphis, 33. 
 
 populifolii, 33. 
 
 viminalis, 34. 
 artomisicola, Macrosiphum {Siphono- 
 
 phora), 61. 
 artemisiae, Macrosiphum (Aphis), 61. 
 arundicolens, Eucallipterus (Myzocal- 
 
 lis), 24. 
 arundicolens, Myzocallis (Callipterus) , 
 
 22 
 arundinariae, Myzocallis, 24. 
 arundinis, Hvalopterus (Aphis), 130, 
 
 179. 
 atriplicis. Aphis, 93. 
 aurantiae. Toxoptera. 129. 
 aurantii, Toxoptera (Aphis), 129, 179. 
 avouae. Aphis (Nectaropliora, Sipho- 
 corj/nc), 94, 179. 
 
 baccharadis, Macrosiphum (Nectaro- 
 
 phora), 61. 
 bak-eri. Aphis, 123. 
 bakeri. Aphis, 6, 179. 
 balsamifenie. Pemphigus, 142. 
 bellus, Myzocallis (Aphis, Callip- 
 
 tems), 24. 
 berberidis, Liosomaphis (Aphis, Uho- 
 
 palosipbum), 130. 
 betae. Pemphigus, 142. 
 betulae, Cliaitophorus, 31. 
 betulae, Euceraphis (Callipterus) , 19, 
 
 178. 
 betulaecolens, Calaphis (Aphis, Cal- 
 
 liptcrtis), 18. 
 braggii, Myzus, 73. 
 brassicae. Aphis, 95. 
 Byrsocrypta, 148. 
 ulmicola, 148. 
 
 C 
 calendulicola. Aphis, 96. 
 Calaphis, 18. 
 
 betulaecolens, 18. 
 rastancae, 24. 
 ealifornica, Essigella (L(ichnus), 44. 
 californicum, Maerosipliuni (Nectaro- 
 
 phora), 62. 
 californicus, Callipterus (Monellia), 
 
 29. 
 californicus Myzocallis, 178. 
 californicus Theeabius (Pemphigus), 
 144. 
 
 Callipterinella, 31. 
 
 annulata, 31. 
 Callipterus, 28. 
 
 alnifoliae, 20. 
 
 arundicolens, 22. 
 
 beUus, 24. 
 
 betulae, 19. 
 
 betulaecolens, 18. 
 
 californicus, 29. 
 
 earyae, 29. 
 
 caryella, 30. 
 
 castaneae, 24. 
 
 coryli, 25. 
 
 discolor, 25. 
 
 hyalinus, 26. 
 
 juglandicola, 28. 
 
 juglandis, 28. 
 
 punctatus, 26. 
 
 quercus, 27. 
 
 tiliae, 21. 
 
 ulmifolii, 27. 
 
 viminalis, 34. 
 eapreae, Siphocoryne (Aphis), 132. 
 cardui. Aphis, 96. 
 oarduinum, Phorodon, 73. 
 cari. Aphis, 96, 179. 
 earyae, Callipterus (Monellia), 29. 
 caryella, Monellia (>4p/ws, Callip- 
 terus), 30. 
 castaneae, Calaphis (Callipterus), 24. 
 castaneae, Myzoc-aJlis, 178. 
 castanicola, Myzocallis, 178. 
 ceanothi. Aphis, 96. 
 ceanothi-hirsuti. Aphis, 96. 
 cerasi, Myzus (^pJii^), 73. 
 cerasifoliae, Aphis, 97. 
 Cerataphis, 140. 
 
 lataniae, 140. 
 Cerosipha, 137. 
 
 cupressi, 137. 
 Chaitophorus, 33. 
 
 annulata, 31. 
 
 betulae, 31. 
 
 negundinis, 36. 
 
 nigrac, 37. 
 
 populicola, 36. 
 
 populifoliae, 33. 
 
 salicioola, 37. 
 
 smithiae, 34. 
 
 ■!!tffli(wi(is, 34. 
 Chermes, 151. 
 
 cooleyi, 151. 
 
 coweni, 151. 
 
 pinicorticis, 152. 
 Chromaphis, 28. 
 
 juglandicola, 28. 
 chrysanthemi, Macrosiphum (Siphono- 
 
 phora), 62. 
 chrysanthemi, Macrosiphum, 69. 
 chrysolepis, Symydobius, 38. 
 cicutae. Ampliorophora, 54. 
 eircumflexus, Myzus (Siphonophora). 
 citri. Aphis, 105. 
 
A SYNOPSIS OF THE APHIDIDAE 
 
 217 
 
 citrifoUi, Macrosipltum {Ncctaro- 
 
 phora), 69. 
 Cladobius, 41. 
 
 rufulus, 41. 
 
 salicti, 43. 
 Coccus. 140. 
 
 latanioc, 140. 
 
 pinicorticis, 152. 
 Colopha, 148. 
 
 ulmicola, 148. 
 Coloratloa, 137. 
 
 rufoniaculata, 137. 
 conii, Siplwcoriinc, 133. 
 cooki, Aphis, 100. 
 cooleyi, Clierme.>i, 151. 
 cornifoliae, Aphis, 100. 
 coryU, Myzocallis {Aphis, Callip- 
 
 terus), 25. 
 corylinum, Rhopalosiphum, 81. 
 cowciii, Chcrmes. 151. 
 coweui, Phyllaphis (Pemphigus), 13. 
 crataegfifolii. Aphis, 100. 
 p.rucis, Thoiiiasia, 36. 
 Cryptusiphum, 13. 
 
 tahoense, 13. 
 cucurbitae, Macrosiphum (Siphono- 
 
 phora), 62. 
 cupressi, Cerosipha, 137. 
 cynosbati, Myziis (Ncctarophora), 75. 
 
 D 
 
 davidsoni, Myzocallis, 24, 178. 
 detitatn-s, Lachiius, 45. 
 destructor, Macrosiphum, 66. 
 dianth', Rhopalosiphum, 85. 
 dirhodum, Macrosiphum {Aphis), 63. 
 discolor, Myzocallis {Cnllipterus), 25. 
 Drepanaphis, 18. 
 
 acerifolii, 18. 
 Drepanosiphum, 17. 
 
 acerifolii, 18. 
 
 platanoides, 17. 
 dryophila, Vacuna (Aphis, Cliaito- 
 phorus), 150. 
 
 E 
 Eichoclmitophorus, 33. 
 
 populifolii, 33. 
 Eriosoina, 148. 
 
 americaua, 148. 
 
 languinosa, 149, 179. 
 
 lanigerum, 149. 
 
 pyricola, 149. 
 Essigella, 44. 
 
 ealifornica, 44. 
 essigi, Myzocallis, 27. 
 Eucallipterus, 20. 
 
 arundicolens, 24. 
 
 flava, 20. 
 
 tiliae, 21, 178. 
 Euceraphis, 19. 
 
 betulae, 19, 178. 
 
 flava. 2u. 
 
 gillettei, 20. 
 euonomi, Aphis, 101. 
 
 fagi, Phyllaphis, 13, 178. 
 
 ferrisi, Laclinus, 47. 
 
 flava, Eucallipterus (Euceraphis), 20. 
 
 flocculosa, Pterocomma (Melanoxan- 
 thus), 40. 
 
 foeniculi, Siphocoryne, 132. 
 
 fragaefolii, Myzus, 75. 
 
 fraxiihi-dipetalae, Prociphilus (Pem- 
 phigus), 146. 
 
 frigidae. Aphis, 105. 
 
 frigidae, Macrosiphum, 61. 
 
 FuUawaya, 35. 
 salieiradicis, 35. 
 
 G 
 
 galeopsidis, Phorodon, 81. 
 gillettei, Euceraphis, 20. 
 glehnus, Lachnus, 47. 
 godetiae, Myms, 85. 
 gossypii. Aphis, 100. 
 gossypii. Aphis, 105, 179. 
 granarium, Macrosiphum (Aphis), 64, 
 178. 
 
 H 
 hederae, Aphis, 106. 
 heraclei. Aphis, 107. 
 lieuclierae, Macrosiphum (Siphono- 
 
 phura), 64. 
 hippophoaes, Rhopalosiphum, 81. 
 houghtouensis. Aphis, 107. 
 howardi. 'Rhopalosiphum, 86. 
 humuli, Phorodon (Aphis), 79. 
 Hyadaphis, 132. 
 
 pastinacoe, 132. 
 
 umhellulariae, 133. 
 hyalinu^, Myzocallis (Callipterus), 26. 
 Hyalopterus, 130. 
 
 arundinis, 130, 179. 
 
 Idiopterus, 56. 
 nephrelcpidis, 56. 
 
 jasmini, Macrosiphum (Ncctaro- 
 phora), 64. 
 
 juglandicola, Chromaphis (Lachnus, 
 Callipterus), 28. 
 
 juglandis, Callipterus (Aphis), 28. 
 
 juiiiperi, Lachnus, 50. 
 
 L 
 
 Lachmclla, 50. 
 
 tujafiliiius, 50. 
 Laclinus, 45. 
 
 ahicii.'i. 47. 
 
 alnifoliae, 20, 22. 
 
 californicus, 44. 
 
 dental'!^, 45. 
 
 ferrisi, 47. 
 
 glehnus, 47. 
 
 juglan-dicola, 28. 
 
 junipcri, 50. 
 
218 
 
 MISCELLANEOUS STUDIES 
 
 occidentalis, 47. 
 
 oregonensis, 48. 
 
 pini-radiatae, 48, 178. 
 
 ponderosa, 48. 
 
 pseudotsugae, 48. 
 
 sabinianus, 49. 
 
 taxifolia, 50. 
 
 tomentosus, 178. 
 
 tujafilinus, 50. 
 
 vanduzei, 50. 
 
 viiniiuilis, 45. 
 lactucae, Macrosiphum {Nectaro- 
 phora), 65. 
 
 lactuea Rhopalosiphum (Aphis), 82. 
 laevigatac, Macrosiphum, 62. 
 langiiinosa, Eriosoma {Aphis), 149, 
 
 179. 
 langerum, Eriosoma (Aphis, Schizo- 
 
 neura), 149. 
 lataniae, Cerataphis (Coccus), 140. 
 latysiphoii, Amphorophora, 54, 178. 
 Liosomaphis, 130. 
 
 berberidis, 130. 
 lithospermi. Aphis, 108. . 
 ludoTicianae, Macrosiplium (Sipho)io- 
 
 phora), 65. 
 lutcscens. Aphis, 117. 
 lycopersici, Myzus (Ncctarophora), 
 76. 
 
 M 
 
 Macrosiphum, 57. 
 acerifolii, 18. 
 albifroDS, 60. 
 ambrosiae, 60. 
 artemisiae, 61. 
 artemisieola, 61. 
 baccharadis, 61. 
 californicum, 62. 
 idirysanthemi, 62. 
 chi-yaanthcmi, 69. 
 citrifolii, 69. 
 cucurbitae, 62. 
 destructor, 66. 
 dirhodum, 63. 
 frigidac, 61. 
 grauarium, 64, 178. 
 heueherap, 64. 
 jasmiui, 64. 
 lactucae, 65. 
 laevigatae, 62. 
 ludovicianae, 65. 
 orthoearpus, 66. 
 pisi, 66. 
 pteridis, 67. 
 rosae, 67. 
 rubicola, 77. 
 rudbeckiae, 67. 
 rudbeckiae var. madia, 68. 
 sanborni, 69. 
 sulanifolii, 69, 178. 
 sonchclla, 70. 
 snnchi, 60. 
 stanleyi, 70. 
 
 taraxici, 71. 
 tulipae, 71. 
 yalerianae, 71. 
 macrostachyae, Symydobius, 38. 
 madia, Macrosiphum (rudbeckiae), 68. 
 maidis, Aphis, 84. 
 maidis. Aphis, 108. 
 mali. Aphis, 120. 
 malifoliae, Aphis, 108. 
 marutae, Aphis, 112. 
 maureri, Myzocallis, 26, 178. 
 medicaginis. Aphis, 114, 179. 
 Melanoxantherium, 41. 
 rufulum, 41. 
 salicti, 43. 
 Melanoxanthus, 40. 
 
 ftocculosa, 40. 
 Micrclla, 35. 
 
 monella, 35. 
 middletonii. Aphis, 115. 
 Miiidarus, 150. 
 
 abietinus, 150. 
 monella, Micrella, 35. 
 Moncllia, 29. 
 californicii-s, 29. 
 caryae. 29. 
 caryclla, 30. 
 mori. Aphis, 116. 
 morriaoui, Neetarosiphon, 78. 
 Myzaphis, 134. 
 abietina, 134. 
 rosarum, 134. 
 Myzocallis, 21. 
 'aliii. 21. 
 ainifoliae, 22. 
 arundicolens, 22. 
 aruiidicolens, 24. 
 arundinariae, 24. 
 bellus, 24. 
 californicus, 178. 
 castaneae, 178. 
 castanicola, 178. 
 eoryli, 25. 
 davidsoni, 24, 178. 
 discolor, 25. 
 cssigi, 27. 
 hyalimis, 26. 
 maureri, 26, 178. 
 pasaniae, 26. 
 punctatus, 26. 
 quercus, 27. 
 ulmifoUi, 27. 
 woDdu'ortlii, 27. 
 Myzus, 71. 
 aquelegiae, 73. 
 braggii, 73. 
 cerasi, 73. 
 circumflexus, 74. 
 cynosbati, 75. 
 f'ragaefolii, 75. 
 godetiae, 85. 
 lycopersici, 76. 
 pcrsicae, 80, 85. 
 rhamni, 76. 
 
A SYNOPSIS OF THE APHIDIDAE 
 
 219 
 
 ribes, 75. 
 ribifolii, 76. 
 rosarum, ] 34. 
 varians, 77. 
 vincae, 74. 
 
 N 
 
 Nectarophora 
 
 avenae, 94. 
 
 iacoharadis, 61. 
 
 californicum, 62. 
 
 citrifolii, 69. 
 
 cynosbati, 75. 
 
 jasmini, 64. 
 
 ZactMcae, 65. 
 
 lycopcrsici, 76. 
 
 pwi, 66. 
 
 rlmmni, 76. 
 
 rosae. 67. 
 
 so)icheJIa, 70. 
 
 Valerianae, 71. 
 Nectai'osiphon, 77. 
 
 morrisoni. 78. 
 
 rubicola, 77. 
 neguntlinis, Thomasia (Chaitopliorits) , 
 
 36. 
 neomexieana, Aphis, 116. 
 npphrelepirlis, Idiopterus, 56. 
 uerii, Aphis, 117. 
 nervatum, Ehopalosiphum, 84. 
 niyrac, Chaitophorus, 37. 
 nigronervoaa, Pentalonia, 78. 
 nymphaeae, Siphocoryne {Aphis, Eho- 
 palosiphum), 133, 179. 
 
 O 
 
 occidentalis, Lachnus, 47. 
 ocnotherae, Aphis, 118. 
 orcgonensis. Aphis, 119. 
 oregoneusis, Laclmus, 48. 
 orthocarpus, Macrosiphum, 66. 
 
 padi, Aphis, 94. 
 
 panicola, Schizoneura, 150. 
 
 pasaniae, Myzoeallis, 26. 
 
 pastinaeae, Siphocoryne {Aphis, Hy- 
 
 adaphis), 133. 
 pastinaeae, Hyadaphis, 132. 
 Pemphigus, 141. 
 
 aliiifoliac, 146. 
 
 balsamiferae, 142. 
 
 betae, 142. 
 
 californieus, 144. 
 
 coweni, 13. 
 
 fraxini-dipetalae, 146. 
 
 populicaiilis, 143. 
 
 populiconduplif alius, 145. 
 
 populimanilis, 145. 
 
 populi-transversus, 143. 
 
 populi-trausversus, 143. 
 
 radieicola, 141. 
 
 ranunculi, 144. 
 
 vcnafuscus, 146. 
 
 Pentalonia, 78. 
 
 uigronervosa, 78. 
 persieae, Rhopalosiphum {Aphis, 
 
 Myzus), 179, 185. 
 persicae-niger, Aphis, 119. 
 Phoroclon, 79. 
 
 earduinum, 73. 
 
 galeopsidis, 81. 
 
 humuli, 79. 
 
 scrophulariae, 80. 
 Phyllaphis, 12. 
 
 coweni, 13. 
 
 fagi, 13, 178. 
 
 querci, 15. 
 
 quercicola, 15. 
 Phylloxera, 152. 
 
 popularia, 153. 
 
 .lalicola, 153. 
 
 va-siatrix, 152. 
 
 vitifoliae, 152. 
 Phylloxerina, 153. 
 
 jjopularia, 153. 
 
 salicola, 153. 
 pinicorticis? Chermes, 152. 
 pini-radiatae, Lachnus, 48, 178. 
 pisi, Macrosiphum {Aphis, Nectaro- 
 phora), 66. 
 platanoides, Drepanosiphum {Aphis), 
 
 17. 
 pomi. Aphis, 109. 
 pomi, Aphis, 120, 179. 
 ponderosa, Lachnus, 48. 
 popularia, Phylloxerina {Phylloxera), 
 
 293. 
 populca, Pterocomma, 41. 
 populicaulis. Pemphigus, 143. 
 populieola, Thomasia {Chaitophorus), 
 
 36. 
 populiconduplifolius, Thecabius {Pem- 
 phigus), 145. 
 populifoliae, Chaitophorus, 33. 
 populifoliae, Pterocomma {Ajthis), 41. 
 populifolii, Aretaphis {Eichocluiito- 
 
 jihorus), 33. 
 populimonilis, Thecabius {Pc/nphi- 
 
 gu^'), 145. 
 populi-transversus. Pemphigus, 143. 
 populi-transversus, Pemphigus, 143. 
 Prociphilus, 146. 
 
 aluifoliae, 146. 
 
 fraxini-dipetalae, 146. 
 
 vcuafuscus, 146. 
 pruni. Aphis, 96. 
 prunifoliac, Aphis, 130, 179. 
 prunorum. Aphis, 121, 179. 
 pseudobrassicae. Aphis, 122, 179. 
 pseudotsugae, Lachnus, 48. 
 pteridis, Macrosiphum, 67. 
 Pterocomma, 40. 
 
 flocculosa, 40. 
 
 popuJea. 41. 
 
 populifoliae, 41. 
 
 smithiae, 43. 
 
MISCELLANEOUS STUDIES 
 
 puuctatus, Mvzocallis (CaUqitents), 
 
 26. 
 pyricola, Eriosoma, 149. 
 
 Q 
 
 querci, Phyllaphis, 15. 
 querci, Schizoneura, 15. 
 quercicola, Phyllaphis, 15. 
 quercus, M.vzocallis (Aphis, Callip- 
 terus), 27. 
 
 R 
 radicicola, Trifidaphis (Pemphigus), 
 
 141. 
 rainona, Aphis, 122. 
 r<inuncuH. Pemphigus, 144. 
 rhamni. Aphis, 76. 
 rhanini, Myzus (Nectarophora), 76. 
 rhois, Rhopalosiphimi, 86. 
 Rliopalosiphum, 80. 
 arhi/rantis, 80. 
 (trbuti. 84. 
 berberidis, 130. 
 eorylinum, 81. 
 diaiithi. 85. 
 liippophoaes, 81. 
 h^ncardi, 86. 
 lactucae, 82. 
 ncrvatum, 84. 
 niimpliaene, 133. 
 persicae, 85, 179. 
 rhois, 86. 
 iuJipae, 85. 
 violae, 86. 
 riliifolii, Myzus, 76. 
 ribes, Myzus, 75. 
 
 rosae, Maerosiphum (Aphis, Nectaro- 
 phora), 67. 
 rosarum, Myzaphis (Aphis, Myzus), 
 
 134. 
 nibi, Amphorophora (Aphis), 54. 
 rubicola, Nectarosiphura (Maero- 
 siphum) (Ampliorophora), 77. 
 rubiphila. Aphis, 122. 
 riulbeekiae, Maerosiphum (Aphis), 67. 
 rudbeckiae var. madia, Maerosiphum, 
 
 68. 
 rufomaculata, Coloradoa (Aphis), 137. 
 rufulum, Melanoxanthcrium, 41. 
 rufuht-s, Cladobius, 41. 
 rumicis, Aphis, 101. 
 
 sabinianus, Lachuus, 49. 
 
 salieicola. Aphis, 123. 
 
 salic.icola, Thomasia (Chaitophorus), 
 37. 
 
 salicieorticis, Symydobius, 39. 
 
 saliciradicis, Fullawaya, 35. 
 
 .mlicis, Siphocoryne, 132. 
 
 salicola, Phylloxerina (Phylloxera), 
 153. 
 
 salicti, Cladobius (Melanoxan- 
 thcrium), 43. 
 
 sambucifoliae. Aphis, 123. 
 sanborni, Maerosiphum, 69. 
 Schizoneura, 148. 
 amerieana, 148. 
 lanigerum, 149. 
 panicola, 150. 
 qiierci, 15. 
 serophulariae, Phorodon, 80. 
 seneeio, Aphis, 123, 179. 
 setariae. Aphis, 124, 
 Siphocoryne, 131. 
 avenae, 84. 
 eapreae, 132. 
 conii, 133. 
 focnicidi, 132. 
 nymphaeae, 133, 179. 
 pastinacae, 133. , 
 
 salicis, 132. 
 xylostei, 133. 
 Siphonophora, 60. 
 acerifolii. 18. 
 ambrosiae, 60. 
 artnnisicola. 61. 
 chrysaiitlu mi, 62. 
 eirrumflcxus, 74. 
 eururbitae, 62. 
 hcuelierac, 64. 
 ludovicianMe, 65. 
 solanifoUi, 69. 
 sonchrlla. 70. 
 tuHpac, 71. 
 smithiae, Pterocomma (Chaitoplwrus) , 
 
 43. 
 solanifolii, Maerosiphum (Siphono- 
 phora), 69, 179. 
 sonohella, Maerosiphum (Siphono- 
 phora) (Nectarophora), 70. 
 sonehi, Maerosiphum, 60. 
 sorbi. Aphis, 108. 
 spiraecola, Aphis, 124. 
 spiraeelUi. Aphis, 125, 126. 
 stauleyi, Maerosiphum, 70. 
 Symydobius, 37. 
 agrifoliae, 38. 
 ehrysolepis, 38. 
 inacrostaehyae, 38. 
 salleicortieis, 39. 
 
 tahoense, Cryptosiphum, 13. 
 taraxici, Maerosiphum (Aphis), 71. 
 taxifolia, Lachuus, 50. 
 tetrapteralis. Aphis, 125. 
 Thecabius, 144. 
 
 californieus, 144. 
 
 populiconduplifolius, 145. 
 
 populimonilis, 145. 
 Thomasia, 35. 
 
 crucis, 36. 
 
 negundinis, 36. 
 
 populicola, 36. 
 
 salieicola, 37. 
 
 viminalis, 34. 
 
A SYNOPSIS OF THE APBIDIVAE 
 
 221 
 
 tiliae, Eueallipterus {Aphis, CalUp- 
 
 term), 21, 178. 
 tomeutosus, Lachnus, 178. 
 Toxoptera, 129. 
 
 aurantiae, 129. 
 
 aurantii, 129, 179. 
 Trifidaphis, 141. 
 
 ratlieicola, 141. 
 ■Tuberolachnus, 45. 
 
 viminalis, 4.5. 
 tujafilinus, Lachnus {Lachneilla), 50. 
 tulipae, Macrosiphura {Siphonophora) , 
 
 71. 
 tulipae, Hhopalosiphum, 85. 
 
 U 
 ulmieola, Colopha {Brysocrypta), 148. 
 ulmifolii, Myzocallis (Callipterus), 27. 
 umbeUulariae, Hyadaphis, 133. 
 
 V 
 
 Vacuna, 150. 
 
 dryophila, 150. 
 Valerianae, Macrosiphum {Nectaro- 
 pliora), 71. 
 
 vanduzei, Lachnns, 50. 
 varians, Myzus, 77. 
 vastatrix. Phylloxera, 152. 
 venafuscus, Prociphilus (Pemphigus), 
 
 146. 
 virburnicolens, Aphis, 126, 179. 
 viminalis, Arctaphis {Callipterus, Chai- 
 
 tophorus, Thmnasia), 34. 
 viminalis, Tuberolachnus {Lachnus) 
 
 {Aphis), 45. 
 vinoae, Myzus, 74. 
 violae, Ehopalosiphum, 86. 
 vitifoliae, Phylloxera, 152. 
 
 W 
 
 looodworthi, Myzocallis, 27. 
 
 X 
 
 xylostci. Siphocoryne, 133. 
 
 yuccae. Aphis, 128. 
 yuccicola. Aphis, 128. 
 
MUTATION IN MATTHIOLA 
 
 BY 
 
 HOWARD B. FROST 
 
 [Universit)- o( California Publications in Agricultural Sciences, Vol. 2, No. 4, pp. 81-190] 
 
MUTATION IN MATTHIOLA 
 
 BY 
 
 HOWARD B. FROST 
 
 CONTENTS 
 
 PAGE 
 
 Introduction 81 
 
 Genetic, literature relating to Mattliiola S4 
 
 Methods 85 
 
 Experimental data 89 
 
 The occurrence of apparent mutants 89 
 
 Characteristics and heredity of mutant tj'pes 92 
 
 1. The early type _ 92 
 
 2. The smooth-leaved type 118 
 
 3. The large-leaved type 125 
 
 4. The crenate-leaved type 127 
 
 5. The slender type 135 
 
 6. The narrow-leaved type 141 
 
 7. Miscellaneous aberrant types 143 
 
 8. Some probabilities of random sampling 145 
 
 General discussion 153 
 
 Summary 159 
 
 Literature cited _ 161 
 
 INTRODUCTION 
 
 It is hardly safe to use the term mutation without first defining it. 
 In this paper it will be taken to mean a genotypic change, or a change 
 in essential hereditary constitution, due neither to immediate cross 
 fertilization nor to segregation in a heterozygous parent. No attempt 
 will be made to restrict the term to any of the known or supposed 
 types of such genotypic change ; a limitation of this kind, which 
 restricts the generally accepted sense of a widely used term, seems to 
 tend to confusion rather than to clearness. 
 
 rail 
 
224 MISCELLANEOUS STUDIES 
 
 If we use the term factor mutation,'^ (Babeock, 1918) where the 
 cytological change occurs within a locus, transforming a factor into a 
 different factor, two analogous terms will apply where the cytological 
 change is external to the locus. When the cytological change consists 
 of a loss, reduplication, or transposition of one or more loci it may 
 be called a Jocus inufaiion, and when the change consists in such 
 phenomena affecting a whole chromosome it may be called a chromo- 
 some mutation. If the term mnlaiion i.s applied to the cytological 
 change itself, the last two types of mutation may be grouped together 
 as extralocus mutations, while the first type consists of intraloeus 
 mutations. Examples of factor mutation are white eye in Drosophila. 
 and probably the rubrinervis type in Ocnotliera; an example of locus 
 mutation is (possibly) "deficiency" in Drosophila: and examples of 
 chromosome mutation are Oenothera gigas and 0. lata. 
 
 It is now evident that the immediate problem with Oenothera relates 
 to the mechanism of heredity in the genus. There are two sharply 
 opposed views. One is that recently emphasized by Atkinson (1917, 
 p. 254), when he says, "The evidence from Oenothera cultures points 
 more and more to the conclusion of Shull that 'a hereditary mechanism 
 must exist in Oenothera fundamentally different from that which dis- 
 tributes the Mendelian unit-characters.' " The opposing view is 
 represented by Muller's (1918) strictly Mendelian explanation for 
 Oenothera, based on "an Oenothera-Vike ca.se in Drosophila" ; he says. 
 "The striking parallel between the above behavior and that exhibited 
 in Oenothera makes it practically certain that this, too, is a complicated 
 case of balanced lethal factors." 
 
 A notable feature of the extensive genetic study of Ornothrra is 
 the lack of progress toward any definitely supported explanation of 
 its hereditary mechanism which is not Mendelian. The only definite 
 non-Mendelian hypothesis of chromosome behavior so far proposed, 
 aside from "merogony" and other hypotheses (Goldschmidt, 1916) 
 apparently possible but not proved for Oenothera, which as,sume loss 
 of chromatin after fertilization, seems to be Swingle's (1911) 
 "zygotaxis, " proposed for the apparently parallel case of Citrus. 
 This .suggestion that F^ hybrids may differ, apart from non-uniformity 
 of the Pj gametes, because of the establishment of permanently differ- 
 ent arrangements of the chromosomes in the fertilized egg, still seems 
 to be purely speculative. 
 
 "With more or less "Oenothera-like" cases in other genera, the only 
 definite progress in analysis seems to have resulted from the assump- 
 
 ■ MuUer (1918) has recently used point mutation in the same sense. 
 
 [82] 
 
MVTATIOS IN MATTHJOLA 223 
 
 tion of Mendelian segregation. AYith Oenothera itself, the trend of 
 the evidence tends to favor this form of explanation. 
 
 This fact is strikingly illustrated by two papers of de Vries (1918. 
 1919) which have appeared since the present paper was written, 
 especially as MuUer's (1918) complete report on the beaded- wing 
 case in Drosophila (see especially pp. -471-474, 4S9, and 498-499) 
 indicates that de Vries had hardly yet realized the full possibilities of 
 the balanced- factor hypothesis. In the light of Muller's masterly 
 demonstration of these possibilities, we may be confident that "mass 
 mutation"' is merely ordinary segregation, and that the "unisexual" 
 crosses of Oenothera are really "Mendelian" in their essential phe- 
 nomena. Some difference of usage respecting the inclusiveness of 
 the term Mendelian may be involved here, it is true, since apparently 
 de Vries would apply it only to cases where strictly homologous factors 
 are opposed in homologous chromosomes. Since, however (^Muller, 
 1918), there are good reasons for expecting the occurrence of grada- 
 tions of similarity and of synaptic attraction between opposed loci, and 
 hence of gradations of linkage, the criterion of Mendelian behavior 
 should obviou.sly be the occurrence of segregation between homologous 
 chromosomes, whatever their degree of similarity or amount of cross- 
 ing over. We have no reason to assume that an "unpaired" factor 
 in a parent would so divide as to be included in all gametes; on the 
 other hand, we have learned of a mechanism capable of insuring, in 
 certain particular cases, the inclusion of a certain factor or group of 
 factors either in every functional gamete or in every viable zygote. 
 
 No doubt, as Davis (1917) says, "A great forward step will be 
 taken in Oenothera genetics when types of proven purity have been 
 established . . . ." Meanwhile, cases of "Oenothera-Uke" heredity in 
 species known to possess the Mendelian mechanism deserve most 
 thorough investigation. Special interest consequently attaches to the 
 peculiar inheritance of certain apparent nnitations of the ten-weeks 
 stock {Matthiola annua Sweet), a species in which various character- 
 istics are typically Mendelian. A remarkable series of aberrant forms 
 in this species'' has been briefly discussed in two preliminary com- 
 munications (Frost, 1912 and 1916), and the present paper gives a 
 fuller account of the same phenomena.* 
 
 3 In the variety "Snowflake, " a glabrous, double-producing form with white 
 flowers. 
 
 •• While this paper was in ]iress Blakeslee and Avery (1910), have reported 
 the occurrence of apparent mutations in Datura, which seem to be similar in 
 almost every respect to those here discussed. 
 
 IS3] 
 
226 MISCELLANEOUS STUDIES 
 
 Apparent mutants were first found in the course of work on 
 another problem, the relation of temperature to variation (Frost, 
 1911), conducted at Cornell University. Studied incidentally at 
 first, these new forms were later given special attention. About nine 
 thousand plants, of which about two thousand were progeny of 
 mutant -type parents of peculiar heredity (nearly one-fourth of the 
 latter representing crosses with Snowtiake), have been examined 
 altogether. Some of these plants have been grown at Riverside, where 
 hybridization studies with mutant types are in progress. The present 
 account considers the origin and characteristics of these types, their 
 inheritance with self pollination, and the rather meager available data 
 relating to their behavior in crossing. 
 
 In connection with the work at Cornell, special acknowledgment 
 is due to the late Professor John Craig, and to Dr. H. J. Webber 
 and Dr. H. H. Love. Facilities for work were furnished by the depart- 
 ments of Horticulture and Plant Breeding of the New York State 
 College of Agriculture. 
 
 GENETIC LITERATURE RELATING TO MATTHIOLA 
 
 The work of Correns (1900) on Mattliiola furnislied one of the 
 earliest confirmations of IMendel 's law, and also pointed to complica- 
 tions not found by ilendel. The earlier literature, according to Correns, 
 gives no indication of the study of Mattliiola hybrids beyond the first 
 generation. 
 
 In his later paper on aberrant hybrid ratios, the same author 
 (1902) discusses complications in maize and in 3IattIiioIa. After 
 referring the deviations foinid in maize to selective pollination, he 
 considers a suggestion of de Vries relating to environmental modi- 
 fication of Mendelian ratios, and him.self suggests the possibility of 
 selective elimination of gametes. He says (pp. 171-172), "Solche 
 Einfliisse brauchten nieht alle Sorten Keimzellen des Bastardes gleich- 
 massig zti treffen, sondern sie kiinnten eine Sorte starker angreifen als 
 die andere." 
 
 Von Tsehermak (1904, 1912) has made extensive studies of 
 Mattliiola hybrids, considering mainly, as did Correns, pubescence and 
 flower color. The latter of these papers on hybrids in the genera 
 Mattliiola, Pisinn, and Phaseolus represents a careful analytical test 
 of the "factor hypothesis" of segregating inheritance, leading to the 
 conclusion that the applicability of this hypothesis is strongly eon- 
 
 [841 
 
MUTATION IN MATTHIOI.A 227 
 
 firmed by the results secured. This work, with that of Miss Saunders, 
 leaves no possibility of doubt that the typical Mendelian mechanism is 
 present in Matthiola. 
 
 The most extensive genetic work on Matthiola is evidently that of 
 Miss Saunders, reported by herself (1911, 1911a, 1913, 1913a, 1915, 
 1916) and by Bateson and Saunders, with others (1902, 1905, 1906, 
 1908). This also is work on heredity in hybrids, with special emphasis 
 on the factorial interpretation of the various complications relating 
 to pubescence and to "doubleness" of flowers. 
 
 Goldsehmidt (1913) has explained the inheritance of doubleness 
 by sex linkage and lethal action of a femaleness factor in pollen 
 formation, and his interpretation has been criticized by Miss Saunders 
 (1913). I (Frost, 1915) have presented a somewhat different lethal- 
 factor seheme. and Mi.ss Saunders (1916) has since restated her views 
 and criticized mine. 
 
 Muller (1917) has cited the inheritance of doubleness as a case of 
 "balanced factors," in apparent agreement with my formulation. 
 
 Apparently no one but the present writer (Frost. 1912, 1916; see 
 also review by Bartlett, 1917) has reported experimental evidence of 
 any notable tendency to apparent mutation in the genus, although 
 de Vries (1906, p. 338) mentions the occasional occurrence of vigorous, 
 rigidly upright individuals (a gigas type?), known at Erfurt as 
 "generals," and refers to the rare mutative occurrence of single 
 flowers on branches of double-flowering plants. Doubleness, and color 
 variations in considerable number, have evidently arisen under culti- 
 vation, probably through mutative changes. 
 
 METHODS 
 
 The general cultural methods employed for the first three genera- 
 tions have been very briefly described elsewhere (Frost, 1911). 
 
 The plants of the first four years were grown in pots in the green- 
 house. The plants of the first generation came from one or both of 
 two packets of commercial seed planted in the fall of 1906, and all 
 plants in the later cultures (possibly excepting series 18) were 
 descendants of these. The cultures will in general be designated by 
 the year in which the seed was sown ; the field and greenhouse cultures 
 of 1911 are indicated by 1911F and 1911 H respectively. 
 
 Part of the seed planted, especially in 1908, came from unguarded 
 flowers. The seed lots where thi.s occurred will be indicated in the 
 
 (851 
 
228 MISCELLANEOUS STUDIES 
 
 tabulation of parental data by italic figures, while protection possibly 
 defective will be indicated by an asterisk. It is not probable that 
 much vicinism occurred in the greenhouse cultiires. since this plant 
 is well adapted to self fertilization. 
 
 In the first year's (1906) cultures the plants in each experimental 
 environment were separately numbered. Eaeli plant was designated 
 bj- its number preceded b.y two letters indicative of the environment. 
 For greenhouse temperature these letters were C (cool), M (medium 
 temperature), and W (warm) ; for potting soiP they were S (sand), 
 L (unfertilized "loam"), and G ("good" soil, fertilized). Thus CSl 
 CS2, AVG9, etc.. were pedigree numbers of the first generation, and 
 CG2-M8 and WG9-C10 of the second generation. A few syncotyle- 
 donous plants outside the regular cultures of 1907 were called "WG9- 
 synl. etc. 
 
 For the work at Riverside a new system of numbering was adopted, 
 better suited to ordinary pedigree cultures, and the numbers from 
 this system are used below in the individual treatment of all but one 
 of the mutant types ("early"). This is essentially Webber's (1906, 
 p. 308) system, except that each initial or P, individual of a series is 
 indicated by a letter; a full description has been published (Frost, 
 1917). With 3Iattliiola each type or cross between two types that is 
 tested receives a scries number, the apparent mutants themselves 
 always being taken as the initial individuals of their selfed series. 
 
 The cultures of 1908 included progeny of various parents, one being 
 WG9-C10, an early and few-noded plant suspected of being a mutant. 
 
 The cultures of 1910 consisted of a second-generation test of WG9- 
 ClO, and a first-generation test of other possible mutants, with control 
 lots. The plants were all grown on one bench in one greenhouse 
 (house C). from thirty lots of fifteen seeds each, lots 1-17 relating to 
 WG9-C10. The parents descended from WG9-C10 (see table 7) were 
 selected as those with fewest internodes, a medium number of inter- 
 nodes, and most" internodes in each house of the 1908 cultures, earli- 
 ness of flo.wering being considered when parents were alike in number 
 of internodes. The control parents were both few-noded and many- 
 noded, relatively to their sibs. 
 
 In 1911 eighty progenj^ lots were grown in the field at Ithaca. 
 Lots 1 to 28, transplanted from the greenhouse, paralleled the test of 
 
 •> Soil experimentally varied onlj- in the 1906 cultures, temperature varied in 
 the two following years also. 
 
 For house M, not the highest, which was e.xeeptioually high, but the next 
 to the highest. 
 
 [861 
 
MirAridX IX MATTIIIOI.A 229 
 
 WG9-C10 made in lDlO-11 ; all available progeny of WG9-C10, except 
 the crenate-leaved apparent mutant "WG9-C10-C10, were tested, with 
 check lots between as before. Soil differences and unavoidable differ- 
 ences between lots in time of transplanting combined with hot weather 
 and drought to reduce the value of the results. The remaining fifty- 
 two lots, all field-sown, included a further test of the heredity of 
 aberrant types other than early. Most of these lots, however, were 
 progeny of Snowtlake parents, grown to obtain evidence on the relation 
 of temperature to nuitation and on the inheritance of doi;bleness of 
 flowers, and therefore the results are not reported here. 
 
 The 191 IH cultures eon.stituted a coldframe and greenhouse prog- 
 eny test of mutant types, mainly in the second generation, the plants 
 being grown in flats. 
 
 There was added in 1912-13 a small greenhouse test bearing on the 
 supposed mutative origin of WG9-C10, in view of the apparent possi- 
 bility that WSl or WLIO, in the same house with the unbagged 'WGd, 
 might have been heterozygous for the early type — cross pollination 
 then giving the apparent mutant. 
 
 Further progeny tests of the mutant types have been made in the 
 field at Eiverside, beginning in the fall of 1913. Mainly on account 
 of the unsuitability of the usually hot and dry climate of River- 
 side, the cultures have been largely experimental and always on a 
 small scale, and germination or development ha.s sometimes been un- 
 satisf actor}'. Cultures have been started in October, November, 
 January, and February, and a trial culture in progress at the time 
 of writing was started in August. Some of the plants of the 1915-16 
 cultures were kept until the summer of 1917, and many of them 
 flowered for the first time when about a year old. 
 
 In the cultures of 1913, growth was largely unsatisfactoi-y. and 
 with part of the plants aphid injury interfered more or less with the 
 classification of types. In the cultures of 1914, the seeds were largely 
 lost through toxic effects favored by very shallow planting (as at 
 Ithaca) and strong evaporation from the soil. In subsequent planting, 
 the seeds, planted singlj- in small paper pots, were dropped into 
 relatively deep holes punched in the soil, and covered with sand. 
 
 The only field-grown plants closely resembling those grown in the 
 greenhouse at Ithaca, it may be noted, have been those of the 1917 
 cultures, grown in a lathhouse with added shade from muslin. 
 
 In the cultures of 1915-16, with partial shade and more frequent 
 irrigation than before, development was in general good; but even 
 
 (871 
 
230 
 
 MISCELLANEOUS STUDIES 
 
 S 
 a 
 
 Eh 
 
 S 
 
 -^ s 
 
 
 I 
 
 G 
 
 2 
 
 o 
 
 u 
 
 c 
 
 'i 
 -1 
 
 c 
 
 > 
 
 o 
 
 E 
 
 .s 
 
 6 
 S5 
 
 C 
 d 
 
 3 
 
 £ 
 
 a 
 a 
 
 < 
 
 
 5.47 ± .93« 
 .96 ± 1.06 
 .91 ± -.73 
 
 
 c 
 
 3 
 
 z 
 
 ^ cc ^ ^ »o c^ ^ 
 
 
 
 ; ; h ; : : 
 
 
 tUBUIB 
 
 poB snoia 
 -lunu saA-ea'j; 
 
 ::!—(:»—( ; 
 
 
 
 : : f-H : f-H 
 
 c 
 
 1 
 
 aapuais 
 
 : f-i ": : ^H ^ 
 
 adXi 
 paABa(-MOJJB^ 
 
 : :^ rt (N 
 
 pad^> p9A-Bdi 
 -8;Buajo-im3S 
 
 : t— t ; : T-H 
 
 9dX^ 
 paABai-di«uaJO 
 
 rf — irt : :0 -H ^ 
 
 adXj 
 
 paABdi-q;oomg 
 
 M ': : (N ib -H -H 
 
 3I 
 -1 
 
 iooo-*t^^ CO 
 
 OOCOMi- t^ '1< 
 
 
 None 
 
 None 
 
 None 
 
 None 
 
 None 
 
 Selection in 
 
 notebook" 
 
 Selection at 
 
 trans- 
 planting 
 
 s 
 
 a 
 
 1 
 
 
 
 
 CG2 
 WSl 
 WLIO 
 WG9 
 All above 
 All above 
 
 CG2-F, 
 
 and 
 WG9-F, 
 
 05 
 
 CD 
 
 o 
 o 
 
 o M ^ 
 
 so ^ 
 c - 
 
 
 ■ > 
 
 S ^ 
 > M 
 
 y i2 ■ 
 
 13 to •-! 
 
 (U o 
 
 a ~ 
 
 
 
 
 5< =3 O rt ^j _c 
 
 
 3 =S 
 bo (V 
 
 ^ = j= £.a 
 
 
 o o o 
 
 c e 
 
 <: H 
 
 
 1881 
 
Mrr.lTlON IN MATrillOLA ■ 231 
 
 here the mutant types, with one exception, often failed to grow satis- 
 factorily or to set seed. Infection, probably by Fusarimn, evidently 
 was the cause of the death of many of these plants in their second 
 season. 
 
 "With all cultures grown after (probably) those of 1906, special 
 care was taken to secure random samples of seed, and after 1908 no 
 plants were rejected. The only exception to this statement is the 
 rejection of one pot out of every fifteen, by number and systematically, 
 in the first twenty progeny lots of 1910. For the earlier cultures, a 
 certain amount of selection must be recorded, as follows. In 1906 
 the small and the largest plants were omitted at potting, and probably 
 any weak and abnormal seedlings had been omitted at the preceding 
 transplanting. In 1907 all markedly weak, late, or abnormal seedlings, 
 as determined mainly by the appearance of the cotyledons, were 
 omitted at the first transplanting; and the same was done in 1908, 
 except that certain lots from old seed were unselected.' 
 
 These last lots were arranged at transplanting in such a way that 
 the weak and abnormal plants came at the end in each lot. 
 
 EXPERIMENTAL DATA 
 The Occurrence of Apparent Mutants 
 
 In the cultures of 1906, 88 plants were grown to maturity, none 
 of these being suspected of mutation. In the cultures of 1907, among 
 170 plants one striking variant appeared; this plant, ^69-010, was 
 exceptionally small and early in blooming. 
 
 In the cultures of 1908, 714 plants were available, including ap- 
 parent mutants in several hereditary lines as indicated in table 1. A 
 striking feature of the results is the scarcity of apparent mutants 
 among the seedlings classed as strictly normal at transplanting ; prob- 
 ably the scarcitj- in the preceding years was due mainly to the rejection 
 of abnormal seedlings (see "ilethods"). The first, second, and fourth 
 of these forms have been common in later cultures, while the third 
 and fifth have been rarer; the last three, if seen at all elsewhere, have 
 not being recognized as belonging to the same types as these three 
 plants. 
 
 ■ One tiny plant from WG9, probably not viable, was discarded. 
 
 r89J 
 
232 
 
 MJSCELLANEOrS STUDIES 
 
 Table 2 shows the mimbers and percentages of apparent nmtants 
 found in the enltures of 1910 and 1911F. Since the early type seems 
 to differ from Snowflake only in size and earliness, and is probably 
 inherited without special complications, the available progeny of early- 
 type parents are included in the totals. The progeny of all parents 
 recognized as belonging to other aberrant types are omitted. The 
 second column under "Percentage of mutants" omits doubtful types 
 and individuals, but includes some individuals for which some doubt 
 was indicated in the original records. One rare type of 1911, large- 
 
 Table 2 
 Aberrant types: occurrence among progeny of SnoicflaJce and early parents. 
 Apparent selective elimination at or after germination in 
 field-sown cultures." 
 
 
 
 Percentage of apparent mutants 
 
 Cultures 
 
 Progeny examined'' 
 
 
 
 
 
 
 
 AH counted 
 
 Doubtful omitted 
 
 Greenhouse, 1910 
 
 338 
 
 5.03 ^ .82-^ 
 
 4.14 -+- .77 
 
 Field, 1911, seed 
 
 
 
 
 house-sown 
 
 2072 
 
 5.31 ± .33 
 
 4 63 ± .31 
 
 All above 
 
 2410 
 
 5.27 + .31 
 
 4,56 ^ .29 
 
 Same,Snowflake par- 
 
 
 
 
 ents only 
 
 1364 
 
 4.33 ± .41 
 
 3.74 ± .38 
 
 Field,1911,seed field- 
 
 
 
 
 sown (parents all 
 
 
 
 
 Snowflake ) 
 
 3927 
 
 2.34 ± .24 
 
 1.55 ± .22 
 
 " Germination in greenhouse-sown lots, counting only plants examined for 
 type, 93.2 per cent; in field-sown lots, 4.5.1 per cent. 
 
 •> Including some plants of uncertain type, indicated for some lots (wlien 
 apparently not Snowflake) in tables 1 and 3. 
 
 ' For the calculation of these probable errors the percentages on the third 
 line are used as p. 
 
 leaved, here omitted, has proved to be genetic, but its deterniinatiim 
 in these cultures was in general uncertain. A stricter criterion for 
 the second column, elimination of all individuals not considered posi- 
 tively determined, was used in the calculations for the table.s for the 
 inheritance of the separate mutant types. 
 
 Evidently the more rigorous field conditions of 1911 eliminated 
 many of the "mutants" at or soon after germinaticm. The "coefficient 
 of mutability" with good germination, as was the ease with the un- 
 selected cultures of 1908. seems to be near 5 per cent, a surprisingly 
 high figure if immediate true mutation is responsible. 
 
 Before the aberrant types are considered separately, we may 
 examine (table 3) a detailed illustration of their occurrence in lai'ger 
 cultures. It seems probable, from this evidence, that any descendant 
 of 'WG9 was capable of producing any of the mutant types so far 
 
MITATIOX IX MATTUIOLA 
 
 233 
 
 >> 
 
 3 3 
 I I 
 
 Field lot 
 
 
 Generation 1 
 
 O 
 
 o 
 
 O 
 
 O 
 
 Generntion 2 
 
 
 
 CJeneration 3 
 
 
 Total progeny of 
 determinable type 
 
 H-tOw H-^: tOtCH-: tC: tCCOWi—: k-: H-H-h-i— 
 
 Smooth-leaved 
 
 Small-smooth- 
 leaved 
 
 Crenate-leaved 
 
 Semi-crenate- 
 leaved 
 
 Pointed-crenate- 
 leaved 
 
 Medium-smooth- 
 leaved 
 
 into: »-' ten-* : •-- ; tO <— h- : i-" : i— h- : 
 
 Narrow-leaved 
 
 COH-: to 
 
 Narrow-dark- 
 leaved 
 
 Slender 
 
 Small-eonvex- 
 leaved 
 
 Compact 
 
 Curly-leaved 
 
 Pointed-Iight- 
 leaved 
 
 Large-leaved 
 
 .-0 • ■ -o ■ 
 
 Medium-large- 
 leaved 
 
 Large-thick- 
 leaved 
 
 Small stout- 
 capsuled 
 
 Jagged-leaved 
 
 CC04iOCCjTtsiOCnC:i«50i^**'rf^CO>t»tO>— 'Ci00aiCOrfi.CCtOtnCntsSCO 
 
 All counted 
 
 OiOCCOOCntOOWCntOtt^CC*».tC4^tOH-Cn'^CnCC4^WtOCnrfi.i--CO 
 
 Doubtful 
 omitted 
 
 (911 
 
234 
 
 MISCELLANEOUS STUDIES 
 
 discovered ; the occurrence of the various types suggests a random 
 distribution among the progeny lots. This conclusion is confirmed, 
 and extended to 062, by the field-sown lots of 1911. 
 
 Various parents belonging to mutant types have given other 
 mutant types among their progeny. There is some reason, as table 4 
 indicates, to suppose that parents of the early type have a more 
 marked tendency to produce these other types than have Snowflake 
 parents.' 
 
 Table 4 
 
 IDIO and 1911F; sown in greenhouse. Apparent mutants among descendants 
 
 of WG9-C10 and other ancestors, comparing early parents (pure or 
 
 heterozygous) with Snowflake parents. 
 
 
 Type of parent 
 
 Progeny 
 
 Ancestry 
 
 Total 
 examined 
 
 Percentage of mutants 
 
 
 All counted 
 
 Doubtful omitted 
 
 WG9-C10 
 
 Pure Snowflake 
 
 Both 
 
 Both 
 
 1 Early 
 
 ( Snowflake 
 Snowflake 
 Snowflake 
 Both 
 
 1046 
 
 558 
 
 806 
 
 1364 
 
 2410 
 
 6.50 ± .47" 
 
 4.30 ± .64 
 4.34 =t .53 
 4.33 ± .41 
 5.27 ± .31 
 
 5.64 ± .44 
 
 3.41 =»= .60 
 3 97 ± .50 
 
 3.74 ± .38 
 4,56 ± .29 
 
 " For the calculation of these probable errors the percentages on the last line 
 are used as p. 
 
 Characteristics and Heredity op Mutant Types 
 1. THE EARLY TYPE 
 
 So far as is known, WG9-C10 (figs. 1, 2) was the only apparent 
 mutant of the early type in the cultures. Since, however, this 
 type visibly differs from Snowflake only or mainly in quantitative 
 characters, it cannot be positively identified without comparative 
 progeny tests, and therefore may have been represented by mutant 
 individuals not used as parents. WG9-C10 was much smaller pro- 
 portionately than were its progeny ; this difference was probably due 
 to an embryonic abnormality, early blind termination of the main 
 axis, which was occasionally observed elsewhere and probably occurred 
 in this case. Plants of this type, as compared with Snowflake, are, in 
 general, fewer-noded, smaller, and earlier in blooming. 
 
 The principal data from the cultures of 1908 are shown in tables 
 5 and 6, which also indicate the later conclusions as to the segregation 
 of the early type in the cultures of this year ; figures 3 and 4 illustrate 
 
 8 Inspection of the data in detail indicates that this difference is not due to 
 the possible tendency in parents grown in the warm house to more frequent 
 apparent mutation. 
 
 [921 
 
MUTATION IN MATTHIOLA 
 
 235 
 
 Table 5 
 Cultures of 190S. Time from sowing to emergence of corolla of earliest flower 
 of primary cluster. Frequency distributions.'^ 
 
 
 Singles 
 
 Doubles 
 
 
 Housfi C 1 
 
 House M 1 
 
 House W 
 
 House C 
 
 House M 1 
 
 House W 
 
 Parents; 
 
 WG9- 
 ClO 
 
 Rest 
 
 WG9- 
 ClO 
 
 Rest 
 
 WG9- 
 CIO " 
 
 est 
 
 WG9- „ 
 €10 R 
 
 est 
 
 WG9- 
 ClO 
 
 Rest 
 
 WG9- 
 ClO 
 
 Rest 
 
 Da:,.- '' 
 
 
 
 
 
 
 
 
 
 
 
 
 
 110 
 
 
 
 It 
 
 
 
 
 
 
 
 
 
 
 111 
 
 
 
 It 
 
 
 
 
 
 
 
 
 
 
 
 112 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 113 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 114 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 115 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 116 
 
 
 
 
 1 
 
 
 
 
 
 1 
 
 
 
 
 
 117 
 
 
 
 
 
 It 
 
 
 
 
 
 
 
 
 
 118 
 
 
 
 It 
 
 2 
 
 
 
 
 
 
 
 
 
 
 119 
 
 
 
 
 3 
 
 
 
 
 
 
 
 
 
 
 120 
 
 It 
 
 
 
 4 
 
 
 
 
 
 
 
 
 
 
 121 
 
 
 
 It 
 
 2 
 
 
 
 
 
 
 1 
 
 
 1 
 
 122 
 
 
 
 
 3 
 
 
 
 
 
 
 
 
 2 
 
 123 
 
 
 
 
 8 
 
 It 
 
 
 
 
 1 
 
 1 
 
 
 2 
 
 124 
 
 
 
 1 
 
 4 
 
 
 
 
 
 
 1 
 
 
 ') 
 
 125 
 
 
 
 
 7 
 
 It 
 
 
 
 
 
 5 
 
 r 
 
 3 
 
 126 
 
 
 
 
 16 
 
 
 
 
 
 
 7 
 
 
 7 
 
 127 
 
 
 
 
 3 
 
 "it 
 
 2 
 
 
 
 
 9 
 
 
 2 
 
 128 
 
 
 
 
 4 
 
 
 3t 
 
 
 
 
 8 
 
 
 5 
 
 129 
 
 
 
 
 7 
 
 It 
 
 4 
 
 
 
 
 8 
 
 
 7 
 
 130 
 
 
 
 
 1 
 
 It 
 
 3 
 
 
 
 
 12t 
 
 
 4 
 
 131 
 
 
 i 
 
 
 2 
 
 
 4 
 
 
 
 2 
 
 12 
 
 
 3 
 
 132 
 
 It 
 
 
 
 1 
 
 
 2 
 
 
 
 
 7 
 
 
 8 
 
 133 
 
 Ift 
 
 
 
 1 
 
 1 
 
 
 
 
 
 4 
 
 
 7 
 
 134 
 
 1 
 
 •2 
 
 
 2 
 
 
 1 
 
 
 
 
 2 
 
 
 7 
 
 135 
 
 
 4 
 
 
 4 
 
 
 4 
 
 
 
 
 6 
 
 
 6 
 
 136 
 
 
 1 
 
 1 
 
 1 
 
 
 3 
 
 1 
 
 
 
 
 
 4 
 
 137 
 
 
 7 
 
 
 2 
 
 
 3 
 
 
 3 
 
 
 1 
 
 
 2 
 
 138 
 
 
 10 
 
 
 1 
 
 
 9 
 
 
 4 
 
 
 1 
 
 
 1 
 
 139 
 
 
 18t 
 
 
 1 
 
 1 
 
 2 
 
 
 8 
 
 
 1 
 
 
 3 
 
 140 
 
 1 
 
 7 
 
 
 It 
 
 
 4 
 
 1 
 
 3 
 
 
 
 
 3 
 
 141 
 
 
 10 
 
 
 
 
 1 
 
 1 
 
 9 
 
 
 1 
 
 
 3 
 
 142 
 
 
 4 
 
 
 
 
 
 
 1 
 
 5 
 
 
 1 
 
 
 1 
 
 143 
 
 
 4 
 
 
 
 1 
 
 5 
 
 2 ] 
 
 
 
 
 
 
 2 
 
 144 
 
 
 4 
 
 
 
 
 2 
 
 
 9 
 
 
 
 
 2 
 
 145 
 
 
 
 
 
 
 3 
 
 
 6 
 
 
 
 
 3 
 
 146 
 
 
 
 
 
 
 
 1 
 
 5 
 
 
 
 
 2 
 
 147 
 
 
 
 
 
 
 
 
 4 
 
 
 
 
 
 148 
 
 
 
 
 
 
 2 
 
 
 1 
 
 
 
 
 i' 
 
 149 
 
 
 
 
 
 
 2 
 
 
 1 
 
 
 
 
 
 150 
 
 
 l' 
 
 
 
 
 1 
 
 
 2 
 
 
 
 
 
 151 
 
 
 1 
 
 
 
 
 
 
 1 
 
 
 
 
 1 
 
 152 
 
 
 
 
 
 
 
 
 
 
 
 
 2 
 
 153 
 
 
 i 
 
 
 
 
 
 
 
 
 
 
 1 
 
 154 
 
 
 
 
 
 
 
 
 
 
 
 
 
 155 
 
 
 1 
 
 
 
 
 
 
 l' 
 
 
 
 
 1 
 
 •Daggers (t) indicate the position and number of apparent mutants. Double 
 daggers ({) indicate inheritance of parental type (here, early); all single progeny 
 of WG9-C10 here reported have been tested for inheritance of this type. The 
 conventional statistical constants corresponding to the house totals of tables 5 
 and 6 have been published (Frost, 1911); the means for flowering given there 
 are too high by one half-day. 
 
 ^ To time of observation (upper limit of one-day class). 
 
 [931 
 
236 
 
 MISCELLANEOV.'i STUDIES 
 
 Table 5. Cultures of 1908 {Continued) 
 
 
 
 
 Siiif 
 
 ■les 
 
 
 
 
 
 Doubles 
 
 
 
 
 House C 
 
 House M 
 
 House W 
 
 House C 
 
 House M 
 
 House W 
 
 Parents: 
 
 WG9- 
 CIO 
 
 Rest 
 
 WG9- 
 CIO 
 
 Rest 
 
 WG9- 
 ClO 
 
 Rest 
 
 WG9- 
 ClO 
 
 Rest 
 
 WG9- 
 ClO 
 
 Rest 
 
 WG9- 
 ClO 
 
 Rest 
 
 Days '' 
 156 
 
 157 
 158 
 159 
 160 
 161 
 162 
 163 
 164 
 165 
 166 
 167 
 168 
 169 
 170 
 171 
 172 
 173 
 
 
 It 
 
 
 .... 
 
 
 
 
 
 E 
 
 
 
 i' 
 
 It 
 .... 
 
 ° Daggers (t) indicate the position and number of apparent mutants. Double 
 daggers (}) indicate inheritance of parental type (here, early); all siiirjlc progeny 
 of WG9-C10 here reported have been tested for inheritance of this type. The 
 conventional statistical constants corresponding to the house totals of tables 5 
 and 6 have been published (Frost, 1911); the means for flowering given there 
 are too high by one half-day. 
 
 *■ To time of observation (upper limit of one-day class). 
 
 the difference in earlincss between the early and Snowtlake types. The 
 parents gronped under "rest" include CG2 and WG9 tliemselves, with 
 four progeny of the former and eight of the latter. Of these fourteen 
 parents, not one has produced exceptionally few-noded jirogeny like 
 those of WG9-C10. 
 
 Apparently WG9-C10 wa.s heterozygous for a "few-nodedness" 
 factor not carried by any of the other parents tested. Neither in 
 the 1907 cultures nor in the 1908 cultures now under consideration 
 did the data siiggest that WG9 itself was similarly heterozygous. 
 Tables 5 and 6 include the first 30 progeny of "WG9, for each hou.se, 
 as arranged at the lirst transplanting," 88 plants altogether; including 
 the remaining plants, mainly weak or abnormal at transplanting, the 
 total is 116. One of the F, plants (WG9-syn3-M10) was very sug- 
 gestive of the early type, but (tables 12 and 13) it gave only Snow- 
 flake progeny in a small test. 
 
 !> See page 89. Two plants not jiroducing a normal main inflorescence are 
 omitted. 
 
 1941 
 
.1/; r Alios IS M.ITTIIIULA 
 
 237 
 
 Table G 
 
 Cullures of 190S. Number of main-stem internodes bcloic first flower-bearing 
 node. Frequency distributions.'- 
 
 
 Singles 
 
 Doubles 
 
 
 House C 
 
 House M 
 
 House W 
 
 House Q 
 
 House M 
 
 House W 
 
 Parents: 
 
 WG9- 
 
 ClO 
 
 Rest 
 
 WG9- 
 ClO 
 
 Rest 
 
 WG9- 
 ClO 
 
 Rest 
 
 WG9- 
 ClO 
 
 Rest 
 
 WG9- 
 ClO 
 
 Rest 
 
 WG9- 
 ClO 
 
 Rest 
 
 InternodeH 
 
 
 
 
 
 
 
 
 
 
 
 
 
 16 
 
 It 
 
 
 
 
 
 
 
 
 
 
 
 
 17 
 
 
 
 
 
 
 
 
 
 
 
 
 
 18 
 
 
 
 
 
 
 
 
 
 
 
 
 
 19 
 
 I'tt 
 
 
 
 
 
 
 
 
 
 
 
 
 20 
 
 It 
 
 
 
 
 
 
 i' 
 
 
 
 
 
 
 21 
 
 
 
 2tt 
 
 
 
 
 
 
 
 
 
 
 22 
 
 
 
 It 
 
 
 
 
 1 
 
 
 r 
 
 
 
 
 23 
 
 
 
 
 
 
 
 
 
 1 
 
 
 
 
 24 
 
 
 
 
 
 
 
 
 9 
 
 
 
 
 
 25 
 
 1 
 
 2t 
 
 It 
 
 1 
 
 
 
 2 
 
 25 
 
 
 
 
 
 26 
 
 
 
 
 
 
 
 
 29 
 
 
 1 
 
 
 
 27 
 
 1 
 
 i 
 
 
 r 
 
 
 
 1 
 
 22 
 
 
 2t 
 
 
 
 28 
 
 
 17 
 
 
 
 
 
 
 6 
 
 
 9 
 
 1 
 
 
 29 
 
 
 24 
 
 
 1 
 
 
 
 
 
 
 14 
 
 
 
 30 
 
 
 13 
 
 
 1 
 
 
 
 
 l' 
 
 2 
 
 22 
 
 
 i" 
 
 31 
 
 
 8 
 
 
 2 
 
 
 .... 
 
 
 
 
 27 
 
 
 
 32 
 
 
 2 
 
 
 7 
 
 
 
 
 
 
 5 
 
 
 
 33 
 
 
 1 
 
 
 15 
 
 It 
 
 
 
 
 
 3 
 
 
 
 34 
 
 
 
 1 
 
 16 
 
 
 
 
 
 
 4 
 
 
 i' 
 
 35 
 
 
 2 
 
 
 19 
 
 It 
 
 
 
 
 
 
 
 5 
 
 36 
 
 
 
 
 4 
 
 It 
 
 
 
 
 
 1 
 
 
 6 
 
 37 
 
 
 
 
 1 
 
 It 
 
 
 
 
 
 
 
 8 
 
 38 
 
 
 
 
 8 
 
 
 
 
 
 
 
 
 
 5 
 
 39 
 
 
 
 
 3 
 
 
 
 
 
 
 
 
 
 13 
 
 40 
 
 
 
 
 1 
 
 
 
 
 
 
 
 
 
 6 
 
 41 
 
 
 
 1 
 
 1+ 
 
 
 
 
 
 
 
 
 
 8 
 
 42 
 
 
 It 
 
 
 
 It 
 
 
 
 
 
 
 
 
 6 
 
 43 
 
 
 
 
 
 
 If 
 
 
 
 
 
 
 
 12 
 
 44 
 
 
 
 
 
 
 
 
 
 
 
 
 
 9 
 
 45 
 
 
 
 
 
 It 
 
 
 
 
 
 
 
 
 6 
 
 46 
 
 
 
 
 
 
 4 
 
 
 
 
 
 
 
 3 
 
 47 
 
 
 
 
 
 
 3 
 
 
 
 
 
 
 
 3 
 
 48 
 
 
 
 
 
 
 4 
 
 
 
 
 
 
 
 
 49 
 
 
 
 
 
 i' 
 
 3 
 
 
 
 
 
 
 
 i' 
 
 50 
 
 
 
 
 
 
 6 
 
 
 
 
 
 
 
 
 51 
 
 
 
 
 
 i' 
 
 3 
 
 
 
 
 
 
 
 2t 
 
 52 
 
 
 
 
 
 
 10 
 
 
 
 
 
 
 
 1 
 
 53 
 
 
 
 
 
 
 6 
 
 
 
 
 
 
 
 
 54 
 
 
 
 
 
 
 2 
 
 
 
 
 
 
 
 2 
 
 55 
 
 
 
 
 
 1 
 
 3 
 
 
 
 
 
 
 
 
 56 
 
 
 
 
 
 
 7 
 
 
 
 
 
 
 
 i 
 
 57 
 
 
 
 
 
 
 8 
 
 
 
 
 
 
 
 
 58 
 
 
 
 
 
 
 1 
 
 
 
 
 
 
 
 
 59 
 
 
 
 
 
 
 1 
 
 
 
 
 
 
 
 \ 
 
 60 
 
 
 
 
 
 
 3 
 
 
 
 
 
 
 
 
 61 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 62 
 
 
 
 
 
 
 r 
 
 
 
 
 
 
 
 
 ' See note o to table 5. 
 
 [95] 
 
238 • MISCELLANEOUS STUDIES 
 
 The differentiation of the eai'ly race is very marked ; with the 
 singles, in fact, the later cultures indicate no case of overlapping in 
 the 1908 cultures, in either character, between extracted pure Snow- 
 flake and pure or heterozj-gous early. The total sterility of the doubles 
 necessarily leaves their constitution somewhat in doubt. 
 
 The ci;ltures of 1908 so far suggest that WG9-C10 was a mutant. 
 To be reasonably certain, however, we must have further evidence 
 (1) on the fact and mode of inheritance of the suppo.sed new type, 
 and (2) on the possibility that either WG9 or some other plant of 
 the cultures of 1906 brought the character into the cultures. We shall 
 now consider somewhat extensive evidence bearing on these points, 
 concluding with a special test of the possibility of vieinism. 
 
 When I la.st saw the warm-house plants of 1906, three were known 
 to be singles, and all but two of the rest were recorded as certainly 
 or probably doubles. Seed was secured from these three singles only, 
 and presumably no other singles occurred in the house. Since this 
 seed was all from unguarded flowers, we must con.sider the possibility 
 that WSl or WLIO, the other warm-house singles, brought the early 
 factor into the cultures. It is also barely po.ssible tliat pollen was 
 brought to WG9 from some plant not in this greenhouse. 
 
 These two parents were tested in supplementary cultures, in house 
 C in 1907, and in house W in 1908. The 1907 progeny averaged 
 slightly earlier than those of other parents, but this may have been 
 due to their position, whieh was much nearer a partition separating 
 the house'" from a warm greenhouse. Unfortunately the internodes 
 were not recorded. 
 
 In the 1908 cultures these lots were potted two days later than 
 most of the other lots and one day later than the WG9 lot, and for 
 some unknown reason the WLIO lot wilted badly for some days. The 
 parents in question gave singles (16 and 11 plants respectively) which 
 when compared with progeny of CG2 and WG9 (23 and 15) might 
 suggest that the parents were heterozygous for the early type. The 
 results with the similar numbers of doubles decidedly disagree with 
 these, and suggest that cultural accidents produced the differences; 
 the WSl lot was not exceptional, while all the WLIO progeny were 
 grouped near the lower end of the range of the other lots. In view of 
 all the facts, the data hardly deserve tabular presentation, but they 
 raise a question requiring further study; a later test is reported below. 
 
 10 A temporary substitute for the regular house C. 
 
MVTATION IN MATTHIOLA 
 
 239 
 
 ^ 
 
 ^ 
 
 l-< 
 
 t/J 
 
 
 
 o 
 
 ^^ 
 
 ^ 
 
 f« 
 
 rn 
 
 
 
 n 
 
 O 
 
 o 
 
 H 
 
 % 
 
 
 to 
 
 K" 
 
 K- 
 
 
 s; 
 
 o 
 
 o 
 
 CO 
 
 o 
 
 to 
 
 
 
 §■ 
 as 
 
 CD 
 
 1 
 
 o 
 
 to 
 
 B 
 O 
 
 '- " " ^ tj g (O to 
 
 o 55 1^' 
 
 a 
 
 
 ,00 
 3 CO 
 
 23 
 
 ffi o 
 < 
 
 >*- 
 
 O O 
 
 to CO 
 
 CO 
 D 
 
 ^ I 
 
 O CO 
 
 - 
 
 o 
 
 CD 
 
 CD 
 
 3J 
 P 
 
 o 
 
 
 o 
 
 CO 
 
 B 
 O 
 
 o q o o 
 
 
 02 
 B 
 O 
 
 as 
 
 
 CO 
 B 
 O 
 
 Bl 
 
 g 
 
 '"to 5 -J 
 
 Ip 
 
 CD S CD 
 
 1 I 1 1 ^^ r.l '— '^ rrf^ fT. . 1 /"t. ,-^^ ^~* ''^ ^1 t,-\ .f. ^r\ /•^\ ^^ WI CW f^-, ^.-, 
 
 H-i-riw 
 
 CBCOi: 
 
 B 3 ; 
 
 CO 
 
 p p 
 
 CD CD ■ 
 
 £,33 
 
 2S 
 
 |&- 
 < o 
 as 
 ^a 
 
 CD ^ 
 
 BBppgSBppBBgggBB 
 
 aai-ig^S^ai-^aag^^^ aa 
 p P £-&P p p si. p p 
 
 ?r!C 
 
 CD CD 
 
 CD 
 
 CD CD ■ 
 
 CD CD 
 
 > 
 
 '■a 
 
 cn I-" tx>a< tt^to*- 02 os to o m en w o >*» rf^ 
 
 ^ CO 00 to 4^ 1-' 1^ ><^ W to ^ en to CO CO 0> *>. CO to to CO to to CO CO to to I— W CO 
 CO rf^OOCnWCO^tOCOCn^^4*-^COOOCntOCOcOM4^CnH-coCO^J0050Cn 
 
 5 » 
 
 13 
 
 ce 
 
 CO C0-^COC5COrfi.MO5MMO5>*^^03CnM00CnCOCOO5C5*t»-^^G000CO^ 
 
 c 
 3 
 
 H- 00^ toco O ^ ^M C5 05 *.■ O en 00 OOM >;k 00 01 O GO 00 O ^ ^O: 05Cn OJ 
 
 [97] 
 
240 
 
 MISCELLANKOrH STUDIES 
 
 In the cultures of 1910 and 1911F. all the 1908 progeny of WG9- 
 ClO were tested. On account of the variable nature of the quanti- 
 tative character involved, an elaborate study was necessary. Only 
 small eultui'es could be grown in the greenhouse ; these were supple- 
 mented by larger lots in the field in 1911, but inhibition of flowering 
 by the hot summer, together with the effects of disease and soil varia- 
 tions, made the field results erratic and necessitated special methods 
 of treatment of the evidence. 
 
 singles 
 Doubles 
 
 • 
 
 _ _ 
 
 _^ _^ ^ _ _ 
 
 • ^^s= ; 
 
 — =^ — ^^=1=111^^ — ,^:z:_!_=^ zzz^^^^ 
 
 ^^ ^^ == 
 
 — *- - _ _ — 
 
 iV13 M') 
 
 J L 
 
 I 
 Cl 
 
 1 
 
 C2 Co Cl MU M9 M4 .M2 .M7 C3 
 
 I I I J I I I I I 
 
 €10 
 
 C5 
 
 CIO 
 
 C9 
 
 I 
 
 C7 \V() 
 
 I I 
 
 Wo WIO Mo MS 
 
 I I J I 
 
 CIO 
 
 I 
 
 09 
 
 WCI9 
 
 Ancestry 
 
 Chart 1. Cultures of 1910. Internodes: parental values and progeny means 
 (respectively shown by dots and lines) for progeny lots 1 to 17, omitting 
 aberrant progeny. Parental values should be compared only for the same house. 
 
 Table 7 gives the available data for the parents of the 1910 cultures, 
 and the numbers of progenj^ available for quantitative data. The 
 order of the pedigree numbers here is the same as that of the progeny 
 lots on the greenhouse bench. For convenience, the 1910 tests of other 
 mutant types, together with tests of several Snowflake parents, are 
 included in the table (lots 18 to 30). 
 
 [981 
 
MITATIOX l.\ MATTIllOLA 241 
 
 The plants were grown in house C of the previous work. Two or 
 three plants (one shown in fig. 25) were extremely vigorous, pre- 
 sumably because of some accidental soil difference ; aside from these, 
 a few apparent mutants, and a few plants otherwi.se abnormal, the 
 plants were fairly uniform except where heterozygosis was to be 
 expected. 
 
 The data for time of flowering, as with the 1908 cultures, show the 
 same main features as the internode data, and only the latter will be 
 considered in detail. The types were again more widely different in 
 internodes than in earliness, a fact which seems to indicate that the 
 early type grows more slowly than Snowflake. 
 
 So large and so regular are tlie differences in internodes that the 
 means of the.se very small lots seem worthy of presentation (chart 1).'^ 
 Apparently the few-noded character was carried, among the nine 
 parents descended from WG9-C10, by all except the three parents 
 having the highest number.s in their respective houses. 
 
 Tables 8 and 9 give the internode frequencies for the singles and 
 doubles respectively, by separate progeny lots and by groups of similar 
 ancestry. The range of variation for the check lots, omitting the 
 indicated apparent mutants and other apparently abnormal plants, is 
 rather surprisingly small, as is the case with the cool-house cultures 
 of 1908. The three late progeny of WG9-C10 give lots closely corre- 
 sponding in range to the check lots, only one individual falling below 
 the range of the combined check lots. The six early and medium 
 progeny of WG9-C10, on the other hand, give distributions of far 
 greater range than do the check parents, extending to much low-er 
 values. 
 
 Tables 10 and 11 give the ordinary statistical constants for the 
 grouped lots. The mean ninnber of internodes, for both singles and 
 doubles, is about 25 per cent lower in the progeny of the six few- 
 noded parents, the difference being not far from ten times as great as 
 its probable error. The increase in variability with the progeny of 
 the early parents is also striking, and the difference is about five to six 
 times its probable error. With time to flowering, it may be noted, the 
 differences are similar to those with internodes, but somewhat less 
 marked in the ca.se of the mean ; the flowering data are not given here. 
 
 It is plain that the previous conclusion as to the heterozj-gous nature 
 of WG9-C10 is sustained. The elimination of the apparent mutants 
 
 11 Calculated with the apparent mutants ami four other apparently aljnornial 
 plants eliminated; see tables 8 and 9. 
 
 [991 
 
242 
 
 MISCELLANEOUS STUDIES 
 
 5» 
 8 
 
 
 
 2^ 
 
 No 
 
 
 cq rt rt -H : ■^oo-^foi mn ■not !» 
 
 : :^ 
 
 ; CO -^ Oi 05 1>- CO 1— I ; ^^ 
 
 : ^ Oi ; M rH 
 
 i-H CO lO O ^ t-t C<1 C^ CC ^ C<l ^ ^ CC ^ 04 
 
 :rtN<N ii— 1 
 
 ^ : 1-H (N : ^ : ^ 
 
 :T-<rt (N ,-( :rt rt 
 
 :i-i :C0 
 
 :i-i(N •^(N 
 
 :tH(N :rt : 
 
 :rt :iM :t-h 
 
 :tO : :-^ 
 
 :^ : :■* : CO 
 
 
 S: ^CDt-^GOaiO'-<(NCO-^»OCDt^OOOiO^C<lCO'^ 
 ■5 ^-H— <— <— <MC^MC>J{MC^C^M01(NCOCOCOCOCO 
 
 ;:i <» 
 
 
 3s 
 
 ^ 'Ojl 
 
 
 IV 
 
 a; 
 
 >H J3 
 
 C13 o 
 
 c a 
 
 ^ bo 
 
 O (» 
 
 s a 
 
 OJ c ™ 
 
 (-. - a 
 
 CL, 1^ it! 
 
 o -r o 
 
 CO a> bO 
 
 N 
 
 -?, S o: 
 
 
 -e-o-- 
 
 ^C oj 'u ^^ 
 
 a; > 
 
 Eh S 
 
 e 
 
 "I 
 So- 
 
 -°a( 
 
 fiooi 
 
MUTATION IN MATTEIQLA 
 
 243 
 
 
 
 -C 
 
 e 
 
 m^ 
 
 
 Ei^ 
 
 Cm 
 
 <nH 
 
 B? 
 
 C-i 
 
 10 
 
 O^ 
 
 }■. 
 
 
 s 
 
 A 
 
 li. 
 
 <Bt 
 
 — • 0) 
 
 <; a: 
 
 •Z.S 
 
 13 
 
 
 3n3 
 2 g 
 
 C o 
 
 i- »; 
 
 iSfe 
 
 ■^ CS 
 
 3 ■" 
 
 
 S -g 
 
 3 3 
 
 -» "^ 
 
 
 ^ bo 
 
 c 
 
 OS o 
 
 
 
 S J3 
 
 o. 
 
 ca -« 
 
 «iH 
 
 Oh 
 
 O OQ 
 
 &F1 
 
 
 
 
 U4 
 
 3 -O m 
 
 
 <D rt OT 
 
 O 
 
 be S " 
 
 
 o 3 
 
 h » 
 
 
 0) +J 
 
 p- 1" t: 
 
 -O .« 
 
 F-=3 
 
 y a 
 
 >;,:£■ f 
 
 3 O 
 
 
 3 
 
 bc-a o 
 
 
 t- a; ^ 
 
 as 
 
 30, la 
 ate-Ieav 
 zygous 
 
 -♦^ CS 
 
 
 ■S.2 
 ^ a 
 
 IS 
 
 ere 
 
 aeter 
 
 5 S °^ S 
 
 a -a 
 
 x^ ■* 
 
 -^ r^ 
 
 
 C3 ^ 
 
 'w r^ '^ 
 
 
 3 » . 
 
 T3 m 
 
 •3 M^^ 
 
 .Srt 
 
 .s^.s 
 
 
 
 
 E-^f^ 
 
 bc^ 
 
 30 
 
 rt 3 
 
 SV2 
 
 
 .2^5 
 
 « 03 
 
 ■^ S 
 
 J303 
 
 H C 
 
 f^J, ° 
 
 - s 
 
 ' ""'s 
 
 J= 
 
 > S 
 
 [101 J 
 
244 
 
 M ISC ELLA NEOr.S S TV PIES 
 
 
 
 
 ^=1 
 
 ^ ^ ^ OQ CC -* '-H 
 
 ^ rH CO 
 
 
 
 
 (N 
 
 
 2 
 
 o 
 
 
 ■ : rH C^ : ,-( i CO »0 00 t^ >0 ^ -^ ^ -.::: CO 
 
 CO : IN 
 
 
 
 
 
 
 o 
 O 
 
 5 
 
 ^ 
 
 i 
 
 ^ 
 
 1— < 1— 1 
 
 IN 
 
 
 5 
 
 
 : i-» 1-H r-^t^O '■ -^ 
 
 ; : »-4 : : : : : 
 
 
 =1 
 
 2:^ 
 
 '-' CC CC "^ >0 CO ^ --t T^* ^ CO "^ ^ (N <-H ^ 
 
 ;-(—:::::: 
 
 
 O 
 
 5 
 
 o 
 
 5 
 
 1 
 
 
 
 
 
 
 :'-'iM :COiM :C^ 
 
 
 ■ ■ 
 
 
 s 
 
 
 ;::::;::rH^;c^— i:::::::::::;: 
 
 
 
 ::::;::.-*; ii-H tN ^ CO :::::::::::: 
 
 «9 
 
 o 
 
 5 
 
 o 
 
 5 
 
 ::::::::::;i-icccc::'-i::::::::: 
 
 
 
 ; CO ^-i :: »-H f—( :: I— 1 ::::;:::::::::: : 
 
 
 ■a 
 
 ^ : : ; : : : : I'^J ;;;;;;;:;;;;!; 
 
 0-. ^ 
 w - 
 
 8 
 
 O 
 
 ::::::l::::r-rirMM:::::::;:;:: 
 
 < .2 
 Eh rf3 
 
 s 
 
 ! : M M M^ ; ics,^^ ; M M M : ! : ; ; 
 
 s 
 
 o 
 
 5 
 
 
 :::::i:::-<:iN— iio::: 
 
 M^ ; ; ; ; i 
 
 S 
 
 e 
 
 N 
 
 s 
 
 ::C<l(N::::rt::»-i,-.rt;::;:;:::::: 
 
 
 
 
 
 
 
 :::iN«rt:rt^;:::::::::::::::: 
 
 
 10 
 
 O 
 
 2 
 
 ;::::::::::i-iI^»h::: 
 
 :::::: :»H 
 
 
 o 
 
 S 
 
 
 
 o 
 O 
 
 5 
 
 :::;::::— i::-H«C<l:::::::::::: 
 
 
 O 
 
 
 
 3 
 
 
 
 ::;;rt:::-i;(M::;^;; 
 
 : 1— 1 : : : : : : 
 
 
 5 
 
 en 
 
 
 :::::: :-t— 
 ::::::: i—t 
 
 
 :::::::::::: <N C^ :: : 
 
 
 CO 
 
 ::;:;:;;;i:r-<;iN-H-H: 
 
 .— 1 :::::: : 
 
 
 
 c 
 C3 
 
 CO 
 
 d 
 O 
 
 ■g^ .-l.-irtrt,-HrtC<lC<IC^(NC^IMC^IMOJ(NCOCO00COC0COC0C0C0C0 
 
 5& 
 
 [1021 
 
MUTATION IN MATTIIIOLA 
 
 245 
 
 Doubles 
 
 Singles 
 
 
 
 , « , 
 
 
 A 
 
 
 
 
 1 
 
 >> ^ 
 
 >> 
 
 ^ 
 
 Q 
 
 
 
 
 
 
 = = o 
 
 «— 1 — . 
 
 d 
 
 3 
 
 
 ^^ CO 
 
 
 CD 
 
 
 
 rf>-*. 
 
 >(^ >t^ 
 
 
 
 
 
 
 
 
 
 X ,--- 
 
 X 
 
 
 
 
 > 
 
 3 
 
 ~§ ot-< 
 
 •-t 
 
 Of 
 
 3 
 
 1 
 
 g s| 
 
 0' 
 c 
 
 «p 
 
 
 
 °s- 
 
 10 
 
 ■< 
 
 0:1 
 
 CA 
 
 
 
 
 ><j->-, 
 
 >< 
 
 
 
 
 
 ~§ Ht-it-i 
 
 S50 
 2 
 
 Ht-'tr' 
 
 
 g-^&g. 
 
 0' 
 
 fD P P 
 
 
 
 
 
 
 
 C ,^CD fD 
 
 £^ 
 
 ,^f^ <^ 
 
 
 
 Cfi "^ 
 
 M 
 
 <; 
 
 
 
 r+ 
 
 r^ 
 
 
 
 
 << 
 
 << 
 
 
 
 
 
 •a 
 
 
 
 
 f^ 
 
 CD 
 
 
 
 
 cc 
 
 OT 
 
 
 
 
 
 
 
 :2 
 
 
 
 
 ^ c 
 
 H- W t>5 0i 
 
 GC <I 
 
 !<:»>- en 
 
 33 
 
 to »+-. CD en h4^ 
 
 *- to 
 
 COCn *. 
 
 
 
 
 D S 
 
 
 
 
 '<o 
 
 
 
 
 
 10 to to to ts3 
 
 to to 
 
 to to to 
 
 
 CJiCn Oi Oi 
 
 M CO 
 
 loco CO 
 
 
 i—^COrf^Cn 
 
 CO 1*^ 
 
 CjiOCO 
 
 
 01 00 to M^ 
 
 1— M 
 
 oioo 
 
 << 
 
 If If H- If K- 
 
 If If 
 
 If If If 
 
 p 
 
 3 
 
 i— to h^^ to to 
 
 toco 
 
 ai 4^ CO 
 
 
 00 to coco ^ 
 
 <7>t0 
 
 I— to to 
 
 
 wco*. to to 
 
 Oi it^ 
 
 en loeo 
 
 
 CO tOCJ> P—O 
 
 tob 
 
 CO CO*- 
 
 aix 
 
 1— to CJi tf- CJi 
 
 tOH- 
 
 "-COM 
 
 2 p 
 
 If If If If If 
 
 If If 
 
 If If If 
 
 sr.p 
 §3. 
 
 U H- CO to H- 
 
 i- to 
 
 It. to to 
 
 
 CO Oi Cn OCO 
 
 00 CO 
 
 coco to 
 
 
 h- h^ to t-" 
 
 1— ' 1— » 
 
 to ►- 
 
 
 en tO^0Ol-' 
 
 00 CO 
 
 O3 00H- 
 
 
 
 Cn h** MO '-' 
 
 ^cn 
 
 to to 02 
 
 II 
 
 en CO >t- 00 c^ 
 
 o>- 
 
 H- 03C0 
 
 If If If If If 
 
 If If 
 
 If If If 
 
 p 2. 
 C". ftp 
 
 . . !^. . 
 
 
 to h- 
 
 £, 
 
 enOSMMM 
 
 0-4 
 
 bo~j 
 
 
 to to 1*^00 CO 
 
 00 00 
 
 COtOM 
 
 
 
 «5 
 
 s- s 
 
 a. s 
 
 1^ 
 
 [io:M 
 
246 
 
 MISCELLANEOUS STUDIES 
 
 s 
 
 '5 
 
 m 
 -«! 
 Eh 
 
 
 1 
 
 C^ C^ lO 
 
 cooo> 
 
 
 CO oco 
 
 Tt* 00 o 
 
 
 
 
 o 
 
 
 
 ^ n 
 
 
 
 ^ 2 
 
 ■H 11 -H 
 
 ■H ■« -H 
 
 '3.S 
 
 
 
 ts s 
 
 CO o r^ 
 
 (DiOlM 
 
 1> 
 
 t^C<l ■* 
 
 -<# 1-H TJH 
 
 O 
 
 -•^^ 00 o 
 
 CD OOO 
 
 
 (N 
 
 (N 
 
 
 OiOiO 
 
 »— 1 »-l CO 
 
 
 OC^ CO 
 
 -HC^CO 
 
 •n c 
 
 
 
 S.2 
 
 11 -H ■« 
 
 •« •« 11 
 
 rt > 
 
 I^ Ol o 
 
 OOO o 
 
 M-o 
 
 CO CO »o 
 
 O I-H W 
 
 
 rtlN'1' 
 
 rt ntp 
 
 
 CO (N C^ 
 
 coot^ 
 
 
 ^ -^ Tt* 
 
 ^ CO ^ 
 
 
 •« -H ■« 
 
 •H -H ■« 
 
 tu 
 
 
 
 ^ 
 
 (M o t^ 
 
 NOO 
 
 
 OO Oi 
 
 o to »o 
 
 
 CO OS ^ 
 
 to coo 
 
 
 <M (M C<1 
 
 IN(N(N 
 
 
 
 
 °>, 
 
 
 
 S o 
 
 OHO 00 
 
 CO --^ CD 
 
 T)< I-ICO 
 
 lOlM CO 
 
 1^ 
 
 
 
 
 
 . >> 
 
 -, »^ >> 
 
 
 
 GJ OJ^— 
 
 Oj Of-- 
 
 
 M 
 
 ^ -*J fc^ 
 
 
 
 
 e3 cS 03 
 
 "c3 o3 o3 
 
 
 d 
 
 0' 
 
 ►3.-1W 
 
 iJhJW 
 
 
 O 
 
 
 
 & 
 
 O) 
 
 -So 
 
 3o 
 
 
 
 ^-H 
 
 03—1 
 
 s 
 
 g 
 
 fJO 
 
 iJO 
 
 c 
 
 o 
 
 ^-v~^ 
 
 ^ — ' 
 
 < 
 
 
 
 
 
 
 
 
 
 
 05 
 
 02 
 
 
 
 o 
 
 o 
 
 
 o 
 
 ^ 
 
 ^ 
 
 
 
 sai^uig 
 
 ssiqnoa 
 
 [1041 
 
MVT.triON IN MATTHIOLA 247 
 
 and the other abnormal plants presumably gives a better comparison 
 as to mean and variability, but the conclusion is the same in either 
 ease. The three many-noded (late) parents descended from WGO- 
 ClO give no definite indication of being genetically different from the 
 "check" lots not descended from WG9-C10, while the variability con- 
 stants are sufficient, taken alone, to make probable the genetic differ- 
 entiation of the fewer-noded progeny of WG9-C10. Apparently all 
 the fewer-noded progeny of WGQ-CIO that were tested — seven, when 
 WG9-C10-C10, a erenate-leaved apparent mutant (tables 12 and 13), 
 is included — were either simplex or duplex for presence of an earliness 
 factor or factors. 
 
 The variabilitj^ of all the thirty progeny lots, taken together, is 
 high, as might be expected, though decidedly below that of the progeny 
 of early parents. This high variability is due only in very small part 
 to the progeny of the five or six supposedly mutant parents ; the last 
 thirteen lots, alone, are mucli less variable than the mixed early lots. 
 The portion of the cultures containing these progeny lots from 
 aberrant parents was conspicuous for irregularity of germination, and, 
 on the whole, a relatively low rate of germination. 
 
 A few of the last thirteen lots give more evidence bearing on the 
 origin of WG9-C10. The early WG9-.syn3-M10 (tables 12 and 13) 
 gives no evidence of genotypic differentiation from its ordinary sib, 
 WG9-syn3-Mll ; WSl-WjlG, another phenotypically early parent, 
 also failed to transmit earliness to its progeny. CG2-C2-C6, on the 
 other hand, although itself an ordinary plant, shows a rather sus- 
 picious tendency to the production of early and few-noded progeny, 
 but better evidence would be required for any positive conclu-sion. 
 'WG9-C10-C10 appears, from the data in tables 12 and 13 and from 
 observation of the flowering of plants of the next generation in the 
 1911H cultures, to have been heterozygous for the early type, as well 
 as for the erenate-leaved type. We find in this test no definite indi- 
 cation that the early type has appeared elsewhere than in WG9-C10 
 and its descendants. 
 
 The P, progeny of WG9-C1, an abnormal plant whose F^ progeny 
 were unusually and uniformly early but not few-noded, have been 
 included with the other check lots without question. This treatment 
 seems justified by the flowering data, which do not indicate any 
 repetition of the precocious development of the first-generation plants ; 
 the peculiarities of the P; cultures, if not a mere cultural accident, 
 presumably depended on the very abnormal development of the parent. 
 
 [1051 
 
248 
 
 MISCELLANEOUS STUDIES 
 
 Table 12 
 
 1910, greenhouse, lots 18 to SO. Number of main-stem internodes below first 
 flower-bearing node. Frequency distributions for singles.'^ 
 
 Ances- 
 try 
 
 Gen. 1 
 
 CG2 
 
 WG9 
 
 wsi 
 
 WLIO 
 
 Gen. 2 
 
 02 
 
 W4 
 
 syn (M) 3 
 
 C228 
 
 W2 
 
 CIO 
 
 W2 
 
 W7 
 
 W2I6 
 
 W225 
 
 W22O 
 
 Gen 3 
 
 CG 
 
 W3 
 
 017 
 
 MIO 
 
 Mil 
 
 
 M7 
 
 010 
 
 W2 
 
 05 
 
 
 
 
 Ivter- 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 nodes 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 21 
 
 1 
 
 
 
 
 
 
 
 
 It 
 
 
 
 
 
 
 22 
 
 
 
 
 
 
 
 
 It 
 
 
 
 
 
 
 23 
 
 
 
 
 
 
 
 
 
 It 
 
 
 
 
 
 24 
 
 i 
 
 
 
 
 
 
 
 
 
 
 
 
 
 25 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 26 
 
 
 1 
 
 
 
 
 
 
 
 
 it 
 
 2tt 
 
 
 
 
 27 
 
 i 
 
 
 
 i 
 
 
 
 
 
 It 
 
 It 
 
 2 
 
 
 2 
 
 28 
 
 1 
 
 2 
 
 i 
 
 
 i 
 
 
 
 
 It 
 
 
 4 
 
 i 
 
 3 
 
 29 
 
 
 
 1 
 
 2 
 
 2 
 
 i 
 
 
 
 1 
 
 
 2 
 
 3 
 
 30 
 
 
 1 
 
 3 
 
 3 
 
 2 
 
 1 
 
 1 
 
 It 
 
 1 
 
 
 
 
 
 31 
 
 
 
 1 
 
 
 1 
 
 
 1 
 
 
 
 
 
 
 
 
 32 
 
 
 
 
 
 
 1 
 
 i 
 
 
 
 
 1 
 
 
 
 33 
 
 
 1 
 
 
 
 It 
 
 
 
 
 
 
 
 
 
 34 
 
 
 
 i 
 
 
 
 i 
 
 
 
 
 
 
 
 
 35 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 36 
 
 
 
 
 
 
 1 
 
 
 
 
 
 
 
 
 37 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 38 
 
 
 
 
 
 
 
 
 
 
 
 
 
 1 
 
 39 
 
 
 
 
 
 
 
 It 
 
 
 
 
 
 
 
 40 
 
 
 
 i 
 
 
 
 
 
 
 
 
 
 
 
 41 
 
 
 
 
 
 
 It 
 
 
 
 
 
 
 
 
 
 42 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 43 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 44 
 
 
 
 
 
 
 
 It 
 
 
 
 
 
 
 
 
 ' See table 7 and notes to tables 5 and 8. 
 
 Table 13 
 Same as table 13, for doubles.' 
 
 
 Gen. 1 
 
 CG2 
 
 WG9 
 
 wsi 
 
 WLIO 
 
 .Ances-« 
 
 try 1 
 
 Gen. 2 
 
 02 
 
 W4 
 
 syn (M) 3 
 
 C228 
 
 W2 
 
 010 
 
 W2 
 
 W7 
 
 W2I6 
 
 W225 
 
 W220 
 
 
 Gen. 3 
 
 C6 
 
 W3 
 
 017 
 
 iMlO 
 
 Mil 
 
 
 M7 
 
 CIO 
 
 W2 
 
 OS 
 
 
 
 
 
 ^ Inter- 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 nodes 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 16 
 
 
 
 
 
 
 
 
 1 
 
 
 
 
 
 
 
 17 
 
 
 
 
 
 
 
 
 2 
 
 
 
 
 
 
 
 18 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 19 
 
 1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 20 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 21 
 
 
 
 
 
 
 
 
 3 
 
 
 
 
 
 
 
 22 
 
 1 
 
 
 
 
 2 
 
 
 
 2 
 
 
 
 
 
 
 
 23 
 
 
 1 
 
 
 2 
 
 
 
 i:; 
 
 
 
 
 3t + 
 
 i 
 
 
 
 24 
 
 2 
 
 1 
 
 
 
 
 
 1:: 
 
 
 It 
 
 
 
 2 
 
 
 
 25 
 
 4 
 
 
 
 
 i 
 
 3tt 
 
 
 2 
 
 It 
 
 it 
 
 i 
 
 1 
 
 i 
 
 
 26 
 
 2 
 
 
 2 
 
 2 
 
 1 
 
 
 
 
 1 
 
 
 2 
 
 It 
 
 
 
 27 
 
 
 2 
 
 
 1 
 
 2 
 
 2 
 
 
 
 3 
 
 3 
 
 1 
 
 2 
 
 
 
 28 
 
 
 
 1 
 
 
 1 
 
 1 
 
 
 
 2 
 
 5 
 
 
 1 
 
 
 
 29 
 
 
 
 2 
 
 1 
 
 
 1 
 
 
 
 
 2 
 
 
 
 
 
 30 
 
 
 
 
 
 
 
 3tt 
 
 
 
 
 
 
 
 
 31 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 32 
 
 
 
 i 
 
 
 
 
 1 
 
 
 
 it 
 
 
 
 
 
 33 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 34 
 
 
 
 
 
 
 
 i 
 
 
 1 
 
 
 
 
 
 ° See table 7 and notes to tables 5 and 8. 
 
MrTATIOX rx MATTBIOLA 
 
 249 
 
 with its aborted main axis and very late production of a tiowering 
 shoot. 
 
 Table 14 shows the general plan of the house-sown field cultures 
 of 1911. The progeny of WG9-C10 were arranged as before in the 
 order of their numbers of internodes for each house of the 1908 cul- 
 tures, beginning with the lowest numbers. The parental values for 
 tiowering and internodes are the values indicated by "$" in tables 5 
 
 T.\BI,E 14 
 
 1911; field, plants transplanted from greenhouse. Ancestry, seed, and 
 
 numbers of progeny." 
 
 
 
 
 Number of 
 
 Numbers of plants 
 
 "or data 
 
 Lot 
 
 .\ncestry 
 
 Seeds 
 
 plants alive 
 
 33 days after 
 
 sowing 
 
 on mutation and f 
 
 owering 
 
 
 Gen. 1 
 
 Gen. 2 
 
 Gen. 3 
 
 sown 
 
 Totalb 
 
 Singles 
 
 Doubles 
 
 ^) 
 
 
 
 C5 
 
 /C8 
 \C10 
 
 80 
 
 79 
 
 77 
 
 34 
 
 43 
 
 
 
 71 
 
 71 
 
 70 
 
 36 
 
 34 
 
 31 
 
 
 
 
 rc2 
 
 80 
 
 80 
 
 79 
 
 39 
 
 39 
 
 4 , 
 
 5 1 
 
 
 
 CIO 
 
 C5 
 
 80 
 
 79 
 
 78 
 
 40 
 
 38 
 
 
 
 ■ C8 
 
 80 
 
 79 
 
 78 
 
 35 
 
 43 
 
 6J 
 
 
 
 
 CI 
 
 80 
 
 78 
 
 76 
 
 36 
 
 40 
 
 7\ 
 
 
 
 C5 
 
 /W18 
 \W24 
 
 80 
 
 79 
 
 77 
 
 37 
 
 40 
 
 8/ 
 
 
 
 80 
 
 76 
 
 75 
 
 41 
 
 34 
 
 91 
 10 
 
 
 
 
 
 M4 
 
 SO, 26 
 
 77 
 
 76 
 
 30 
 
 45 
 
 
 
 
 
 M9 
 
 80 
 
 80 
 
 78 
 
 36 
 
 41 
 
 11 1. 
 
 12 '' 
 
 
 
 CIO 
 
 
 M6 
 
 80,63 
 
 77 
 
 77 
 
 34 
 
 42 
 
 
 
 
 
 M2 
 
 80,7/ 
 
 78 
 
 77 
 
 36 
 
 40 
 
 13 
 
 14 J 
 
 
 
 
 
 M7 
 
 80 
 
 80 
 
 80 
 
 37 
 
 42 
 
 
 
 
 
 M8 
 
 80 
 
 74 
 
 70 
 
 33 
 
 37 
 
 
 WG9 
 
 
 
 
 
 
 
 
 
 15 1 
 16/ 
 
 
 
 C9 
 
 /C3 
 
 iC7 
 
 80 
 80 
 
 78 
 76 
 
 78 
 75 
 
 30 
 32 
 
 48 
 43 
 
 171 
 
 
 
 
 
 
 W6 
 
 80 
 
 74 
 
 70 
 
 31 
 
 39 
 
 18 
 
 
 
 
 
 
 W4 
 
 80, ^/ 
 
 70 
 
 65 
 
 26 
 
 38 
 
 19 
 
 
 
 
 
 
 WU 
 
 80 
 
 78 
 
 76 
 
 34 
 
 42 
 
 20 
 
 
 
 
 
 
 W9 
 
 80, 1.9 
 
 76 
 
 72 
 
 24 
 
 47 
 
 21 
 
 ■ 
 
 
 
 CIO 
 
 
 W5 
 
 SO 
 
 79 
 
 75 
 
 33 
 
 42 
 
 22 
 
 
 
 
 
 
 W8 
 
 80, 14 
 
 76 
 
 74 
 
 36 
 
 36 
 
 23 
 
 
 
 
 
 
 W7 
 
 SO 
 
 75 
 
 72 
 
 32 
 
 40 
 
 24 
 
 
 
 
 
 
 W3 
 
 80 
 
 73 
 
 71 
 
 37 
 
 34 
 
 25 J 
 
 
 
 
 
 
 WIO 
 
 74 
 
 72 
 
 66 
 
 32 
 
 34 
 
 26 
 
 
 
 CIO 
 
 
 80 
 
 60 
 
 59 
 
 . 27 
 
 32 
 
 271 
 28/ 
 
 
 
 C9 
 
 /WIO 
 
 \W24 
 
 80 
 
 74 
 
 73 
 
 33 
 
 39 
 
 
 
 80 
 
 80 
 
 78 
 
 32 
 
 45 
 
 ' For plan of arrangement and parental data, see page 86 and tables 5 and 6. 
 Seeds from unguarded flowers are indicated by italic figures; where two numbers 
 are given the first is the total. 
 
 •> Including twelve plants (all late mutants) with which determination of the 
 form of flower was impossible. 
 
 [1071 
 
250 
 
 MISCELLANEOUS STUDIES 
 
 and 6, in the order there given, except that the arrangement by inter- 
 nodes reverses the two-day difference in earliness of the parents of 
 lots 19 and 20; for convenient comparison, the parental and parent- 
 lot internode values are included in table 19. 
 
 Two progeny lots were set in each of the fourteen rows; probably 
 the soil was less favorable at the east end of the plot, and hence for 
 the even-numbered lots, at least in about the last seven rows out of the 
 fourteen. 
 
 The plants were beginning to grow very rapidly when moved to the 
 field. On account of deficient soil moisture and excessive heat, the 
 transplanting was slow and in part purposely delayed, covering a 
 period of five days. Lots 21 to 28 were set three days later than lots 
 
 80 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 70 
 
 OPO 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 (JO 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 40 
 
 
 ?o 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 in 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 1 
 
 2 
 
 3 
 
 4 
 
 5 
 
 6 7 8 9 
 
 10 
 
 11 
 
 12 
 
 13 
 
 14 
 
 (C) 
 
 
 
 (C) 
 
 
 (C) 
 Row number 
 
 
 
 
 
 (C) 
 
 Chart 2. 1911, field; lots transplanted from greenhouse. Percentages of 
 progeny lots not flowering by November 3, for singles. Apparent mutants and 
 injured plants eliminated. Odd-nurabered lots represented by solid line. (C) 
 indicate check rows. The curves'are broken between rows 10 and 11, where a 
 cultural difi^erence enters. 
 
 11 to 20, and the later loss of roots resulting seems, in spite of rain 
 coming the next day, to have seriously delayed flowering. Lots 1 and 2 
 wilted badly after transplanting, and some difference in soil con- 
 ditions in the flats, rather than a genetic difference, was doubtless 
 responsible for the exceptional lateness of these lots. Lot 20 lost an 
 exceptionally large leaf area as a result of transplanting. A fungus 
 disease (a slow stem rot) was more common on lots 20 to 24 than 
 elsewhere; it doubtless killed some young plants and delayed or pre- 
 vented flowering in some other cases. Pos.sibly the soil was poorer in 
 the later rows. 
 
 [108] 
 
MUTATION IN MATTE lOLA 
 
 251 
 
 Table 15 
 1911, Held; plants transplarited from greenhouse. Plants alive November 3, 
 not having flowered. Singles. 
 
 
 
 Nou- 
 
 Non-flowering, 
 
 
 Non-flowering 
 
 Non-flowering, 
 
 Row 
 
 Lot 
 
 flowering 
 plants 
 
 Snowflake and 
 early types » 
 
 Lot 
 
 plants 
 
 Snowflake and 
 early types" 
 
 1 
 
 1 
 
 27 
 
 26 
 
 2 
 
 29 
 
 28 (27) 
 
 2 
 
 3 
 
 5 
 
 2 
 
 4 
 
 7 
 
 5 
 
 3 
 
 5 
 
 9 
 
 9 
 
 6 
 
 11 
 
 11 
 
 4 
 
 7 
 
 7 
 
 5 
 
 8 
 
 17 
 
 17 
 
 5 
 
 9 
 
 
 
 
 
 10 
 
 2 
 
 2(1) 
 
 6 
 
 U 
 
 1 
 
 
 
 12 
 
 9 
 
 7 
 
 1 
 
 13 
 
 19 
 
 18(17) 
 
 14 
 
 20 
 
 19 
 
 8 
 
 15 
 
 12 
 
 12 
 
 16 
 
 14 
 
 14 
 
 9 
 
 17 
 
 1 
 
 1 
 
 18 
 
 8(7?) 
 
 7(6?) 
 
 10 
 
 19 
 
 
 
 
 
 20 
 
 3 
 
 3 
 
 11 
 
 21 
 
 8 
 
 8 
 
 22i> 
 
 11 
 
 10 
 
 12 
 
 23 
 
 21 
 
 20 
 
 24 
 
 23 
 
 23 
 
 13 
 
 25 
 
 IR 
 
 M (12) 
 
 26 
 
 14 
 
 12 
 
 14 
 
 27 
 
 23 
 
 22 
 
 28 
 
 24 
 
 24 
 
 ■ Omitting non-flowering apparent mutants. For the numbers in parenthesis, 
 "doubtful mutants" are classed as mutants. Two plants accidentally seriously 
 injured, in lots 14 and 25, were counted out with the mutants. 
 
 " The stem-rot disease (see p. 108) was evidently worst in lot 22; some two or 
 three of the worst infected plants (included above) were nearly or quite dead 
 by November 3. 
 
 Table 16 
 
 Same as table 15, for doubles. 
 
 Row 
 
 Lot 
 
 Non- 
 flowering 
 plants 
 
 Non-flowering, 
 
 Snowflake and 
 
 early types* 
 
 Lot 
 
 Non-flowering 
 plants 
 
 Non-flowering. 
 
 Snowflake and 
 
 early types'* 
 
 1 
 
 1 
 
 4 
 
 2 
 
 2 
 
 
 
 
 
 2 
 3 
 
 3 
 5 
 
 3 
 3 
 
 1 
 
 3 
 
 4 
 6 
 
 1 
 
 
 
 1(0) 
 
 
 4 
 
 7 
 
 3 
 
 2(1) 
 
 8 
 
 3 
 
 1 
 
 5 
 6 
 
 7 
 
 9 
 11 
 13 
 
 2 
 
 4 
 
 
 
 
 2(1) 
 
 10 
 12 
 14 
 
 
 
 
 1 
 
 
 
 
 1 
 
 8 
 
 15 
 
 2 
 
 2 
 
 16 
 
 6 
 
 5 
 
 9 
 10 
 11 
 12 
 13 
 
 17 
 19 
 21 
 23 
 25 
 
 1 
 
 
 3 
 5 
 1 
 
 
 
 1 
 5 
 1 
 
 18 
 20 
 22 
 24 
 26 
 
 3 
 2 
 
 1 
 4 
 7 
 
 3(2) 
 
 2 
 
 1 
 
 4 
 
 5 
 
 14 
 
 27 
 
 1 
 
 1 
 
 28 
 
 10 
 
 8 
 
 ' See notes to table 15. 
 
 [109] 
 
252 
 
 MISCELLANEOUS STUDIES 
 
 Altogether, these cultures are doubtless much less reliable for their 
 size thau the greenhouse tests of the early type, but they nevertheless, 
 with clue consideration of the points just mentioned, seem to permit 
 of fairly safe conclusions for most of the parents. 
 
 The plants were examined for flowering every other afternoon from 
 July 4 to November 3, inclusive (73 to 195 days from sowing) . A very 
 large part of the plants flowered in July, some in August, and a few 
 still later. Evidently the high summer temperature largely iiiliibitcd 
 flowering ; many of the singles and a few of the doubles entirely failed 
 to flower. 
 
 uu 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 8(J 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 70 
 
 fin 
 
 
 
 fJ/£ 
 
 V 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 50 
 
 ODD 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 30 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 20 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 \ 
 
 2 
 
 3 
 
 4 
 
 5 
 
 6 7 8 9 
 
 10 
 
 11 
 
 12 
 
 13 
 
 1-1 
 
 (C) 
 
 
 
 (C) 
 
 
 CC) 
 Row number 
 
 
 
 
 
 (C) 
 
 Chart 3. 1011, field; lots transplanted from greenhouse. Percentages of 
 progeny lots with primary cluster flowering or aborted by October 10-16, for 
 singles. Lines as in chart 2. 
 
 Figures 5 and G .show the plants in July. Growth was usually 
 vigorous through the sea.sou, but the internodes were very short, the 
 branches numerous, and the region of the terminal inflorescence often 
 abortive, so that determination of the number of main-stem internodes 
 was not practicable. The emergence of the earliest corolla on the plant 
 was recorded at the bi-dinrnal oliservations. and at two periods during 
 the season the aborted primary clusters were noted. 
 
 The data show very definitely the transmission of "earliness" by 
 the fewer-noded progeny of WG9-C10. Tables 15 and 16 show the 
 numbers of plants alive, without having flowered, on November 3 ; the 
 figures are thus a measure of lateness. The two progeny lots in each 
 row are given one line in each of the tables, in order to facilitate 
 separate eoniparison of the fourteen lots in eaeli end half of the jilot. 
 
 [1101 
 
MriATIOX IX ilATriUOI.A 
 
 253 
 
 The last cohimn, with the apparent mutants omitted, no doubt gives 
 the best comparison. The data for the singles, reduced to percentages, 
 are also given in chart 2. 
 
 The doubles, which are often earlier to flower than the singles under 
 unfavorable climatic conditions, flowered so generally that table 16 
 presents no significant differences. The singles (table 15), however, 
 give definite evidence of segregation ; the lots in row-s 2, 5, 6, and 9 to 
 11 all show a tendency to early flowering. Lot 26, consisting of Fj 
 
 Table 17 
 
 1911, field; j}Jaiits transplanted from greenhouse. Singles with primary 
 inflorescence flowering or aiorted as indicated.'^ 
 
 Row 
 
 Lot 
 
 Aborted by 
 July 29 
 
 Flowering or 
 aborted by 
 Oct. 10-16 
 
 Lot 
 
 Aborted by 
 July 29 
 
 Flowering or 
 aborted by 
 Oct. 10-16 
 
 1 
 
 1 
 
 
 
 
 
 2 
 
 
 
 0(2) 
 
 2 
 
 3 
 
 3 
 
 5 
 
 12 (13) 
 
 2 
 
 19 (22) 
 2 
 
 4 
 6 
 
 20 
 
 4 
 
 24 (26) 
 8(9) 
 
 4 
 
 7 
 
 2(3) 
 
 3(4) 
 
 8 
 
 1 
 
 1(2) 
 
 .5 
 6 
 
 7 
 
 9 
 11 
 13 
 
 22 
 25 
 
 1 
 
 27 
 
 29 (30) 
 2(4) 
 
 10 
 
 12 
 14 
 
 26 
 17 
 
 2(3) 
 
 33 
 22 
 
 2(3) 
 
 8 
 
 }5 
 
 4 
 
 5(6) 
 
 16 
 
 0(1) 
 
 1(2) 
 
 9 
 10 
 11 
 12 
 13 
 
 17 
 19 
 21 
 23 
 25 
 
 19 (20) 
 25 (27) 
 
 4 
 
 2 
 
 
 
 20 (23) 
 29(31) 
 
 6(8) 
 3(4) 
 3(4) 
 
 18 
 20 
 22 
 24 
 26 
 
 9 
 U 
 
 7 
 1 
 3 
 
 12 
 
 12 
 
 9 
 
 1(2) 
 4(5) 
 
 14 
 
 27 
 
 1 
 
 3(4) 
 
 28 
 
 
 
 0(1) 
 
 "In this table and also in table 18 the numbers in parenthesis include the 
 probable but somewhat doubtful cases. 
 
 progeny of "WG9-C10, is decidedly earlier than the adjacent lots. 
 Lot 25 also appears early, however. 
 
 Tables 17 and 18 give a direct measure of earliness, relating to the 
 primary inflorescence alone. The clusters visibly aborted were in 
 general relatively far advanced, and those aborted at the earlier date 
 correspond to decidedly early flowering; consequently the flowering 
 and aborted clusters are classed together as early. Chart 3 gives the 
 percentages for singles. 
 
 Here the data for the doubles show fairly consistent differences in 
 the number aborted at the earlier date, while the October totals are 
 
 [111] 
 
254 
 
 MISCELLANEOUS STUDIES 
 
 less regular. There are contrasts similar to those of table 15 up to 
 lot 26, which is late, while the check lots 27 and 28 are early. The 
 singles show the type differences very strikingly throughout lots 1 
 to 20, while lots 21, 22, and 26 give less positive indications of the 
 pre.sence of the earl.y factor. 
 
 Table 19 gives the numbers of singles flowering, in primary in- 
 florescence or elsewhere, by November 3, when growth had practically 
 stopped. The indications are in general the same as with the data 
 already discussed, with better evidence than usual that lots 21 and 22 
 
 Table 18 
 Same as table 17, for doubles.' 
 
 Row 
 
 Lot 
 
 Aborted by 
 July 29 
 
 Flowering or 
 aborted bv 
 Oct. 10-16 
 
 Lot 
 
 Aborted by 
 July 29 
 
 Flowering or 
 aborted by 
 Oct. 10-16 
 
 1 
 
 1 
 
 15 
 
 30 
 
 2 
 
 6 
 
 22 (23) 
 
 2 
 3 
 
 3 
 5 
 
 23 
 12 
 
 33 (34) 
 29 (30) 
 
 4 
 6 
 
 21 
 16 
 
 35 
 30(31) 
 
 4 
 
 7 
 
 17 
 
 30 
 
 8 
 
 8 
 
 22 (24) 
 
 5 
 6 
 
 7 
 
 9 
 11 
 13 
 
 25 
 
 25 (26) 
 16 
 
 41 
 
 40(41) 
 28 
 
 10 
 12 
 14 
 
 24 
 23 
 22 
 
 41 
 40 
 31 
 
 8 
 
 15 
 
 27 
 
 37 (39) 
 
 16 
 
 11 (12) 
 
 29 (32) 
 
 9 
 10 
 11 
 12 
 13 
 
 17 
 19 
 21 
 23 
 25 
 
 20 
 21 
 22 
 
 16(17) 
 17 
 
 35 
 
 41 (42) 
 35 
 
 27 (28) 
 25 
 
 18 
 20 
 22 
 24 
 26 
 
 20 (21) 
 21 
 18 
 10 
 9 
 
 32 (33) 
 42 (44) 
 27 (28) 
 16 (17) 
 15 
 
 14 
 
 27 
 
 21 
 
 29 (30) 
 
 28 
 
 20 
 
 25 (27) 
 
 ■ See note to table 17. 
 
 po.ssessed the early factor. The mean time of flowering is irregular, 
 but shows some effect of the earliness factor. Lot 26 is late as to 
 number flowering, but early as to mean. 
 
 Table 20, for doubles flowering by August 1, no doubt gives more 
 reliable ipeans; these means disagree with our scheme only in lot 26 
 and perhaps lot 22. 
 
 According to tables 17-20, the fewer-noded check parent of each 
 check row has usually given the earlier progeny. In fact, the agree- 
 ment of parental and progeny differences, throughout the cultures, is 
 decidedly remarkable. It is unfortunate that the later parents were 
 always placed in the east half of the row, especially in view of the 
 fact that there was indication of important differences in soil and 
 
 [1121 
 
MUTATION IN MATTHWLA 
 
 255 
 
 Table 19 
 
 1911, Held: plants transplanted from greenhouse. Time from sowing to 
 emergence of earliest corolla. Singles. 
 
 
 Parent-lot 
 
 
 Parental 
 
 Progeny flowering 
 by Nov. 3 
 
 
 Parental 
 
 Progeny flowering 
 by Nov. 3 
 
 Row 
 
 internode 
 mean 
 
 Lot 
 
 internode 
 number 
 
 
 
 Lot 
 
 internode 
 number 
 
 
 
 Number 
 
 Days to 
 flowering 
 
 Number 
 
 Days to 
 flowering 
 
 1 
 
 29 60 
 
 1 
 
 29 
 
 7 
 
 147 14 
 
 2 
 
 32 
 
 7 
 
 128.57 
 
 2 
 3 
 
 21.40 
 
 3 
 5 
 
 16 
 25 
 
 34 
 26 
 
 91.94 
 119.46 
 
 4 
 6 
 
 20 
 27 
 
 33 
 25 
 
 105.45 
 104.08 
 
 4 
 
 49.57 
 
 7 
 
 46 
 
 30 
 
 103 13 
 
 8 
 
 54 
 
 24 
 
 105.67 
 
 5 
 6 
 7 
 
 27.33 
 
 9 
 11 
 13 
 
 21 
 22 
 34 
 
 30 
 33 
 
 18 
 
 91.73 
 
 98.85 
 100 67 
 
 10 
 12 
 14 
 
 21 
 25 
 41 
 
 34 
 27 
 13 
 
 91.12 
 108.30 
 120.62 
 
 8 
 
 28.50 
 
 15 
 
 27 
 
 17 
 
 112 94 
 
 16 
 
 29 
 
 18 
 
 118.00 
 
 9 
 10 
 11 
 12 
 13 
 
 42 . 56" 
 
 17 
 19 
 21 
 23 
 25 
 
 33 
 36 
 42 
 49 
 55 
 
 30 
 34 
 25 
 11 
 15 
 
 100.27 
 97.35 
 129.36 
 122.00 
 136.40 
 
 18 
 20 
 22 
 24 
 26 
 
 35 
 37 
 45 
 51 
 
 18 
 21 
 25 
 14 
 13 
 
 109.67 
 117.81 
 121.76 
 151.57 
 121.08 
 
 14 
 
 47.80 
 
 27 
 
 46 
 
 10 
 
 159.40 
 
 28 
 
 56 
 
 8 
 
 162.50 
 
 • This parent-lot value does not apply to lot 26, which consists of progeny of 
 WG9-C10 itself. 
 
 Table 20 
 Same as table 19, for doubles flowering by August 1. 
 
 Row 
 
 Lot 
 
 Progeny flowering by Aug. 1 " 
 
 Lot 
 
 Progeny flowering by Aug. 1 
 
 Number 
 
 Days to flowering 
 
 Number 
 
 Days to flowering 
 
 1 
 
 1 
 
 38 
 
 90.26 
 
 2 
 
 33 
 
 91 03 
 
 2 
 3 
 
 3 
 5 
 
 36 
 39 
 
 80.22 
 84.46 
 
 4 
 6 
 
 36 
 39 
 
 80 00 
 84 10 
 
 4 
 
 7 
 
 36 
 
 81.28 
 
 8 
 
 31 
 
 84.32 
 
 5 
 6 
 
 7 
 
 9 
 11 
 13 
 
 42 
 42 
 37 
 
 75.86 
 
 77.90 
 
 • 84.32 
 
 10 
 12 
 14 
 
 41 
 40 
 35 
 
 76.59 
 80.25 
 84 80 
 
 8 
 
 15 
 
 46 
 
 83.87 
 
 16 
 
 33 
 
 85.21 
 
 9 
 10 
 11 
 12 
 13 
 
 17 
 19 
 21 
 23 
 25 
 
 37 
 42 
 39 
 33 
 32 
 
 ' 80.43 
 78.24 
 85.95 
 89.03 
 
 88 56 
 
 18 
 20 
 22 
 24 
 26 
 
 34 
 39 
 30 
 26 
 21 
 
 84.12 
 83.85 
 87.93 
 88.85 
 89.05 
 
 14 
 
 27 
 
 35 
 
 88.97 
 
 28 
 
 29 
 
 90.28 
 
 * Only 48 more doubles altogether flowered by November 3, and 23 of these 
 were in the even-numbered lots 20 to 28. 
 
 riisi 
 
256 
 
 MISCELLANEOUS STUDIES 
 
 probably in the incidence of disease, favoring the plants in the west 
 half. The internode data of 1910, however, show a similar tendency. 
 Small genetic differences are suggested, though it would be remarkable 
 if they were so uniformly present in these plants of a single line of a 
 usually selfed species, descendants of parents and a common grand- 
 parent grown under glass. 
 
 If such differences exist in the race, conceivably some combination 
 due to cros.sing might simulate an early mutation. The evidence as a 
 whole, however, does not favor such an origin for our early type ; it is 
 widely divergent from the Snowflake type, and seems to depend on 
 a single main factor difference from Snowflake. 
 
 Table 21 
 CuJtiires of 1913. Ancestry and parental data. 
 
 
 
 Parental data 
 
 
 Lot 
 
 Parent 
 
 
 
 
 Seeds sown 
 
 
 
 Probable type 
 
 Days to 
 flowering^ 
 
 Inter- 
 nodes^ 
 
 
 1 
 
 WS1-W,16 
 
 Snowflake" 
 
 120 .'i 
 
 38 
 
 15 
 
 2 
 
 WG9-C10-W6 
 
 Early 
 
 116.5 
 
 33 
 
 15 
 
 3 
 
 WL10-VV.2 
 
 Snowflake 
 
 139.5 
 
 51 
 
 15 
 
 4 
 
 WL10-W.3 
 
 Snowflake" 
 
 120,5 
 
 38 
 
 15 
 
 5 
 
 WSl-W-1 
 
 Snowflake 
 
 141 5 
 
 57 
 
 15 
 
 6 
 
 WL10-Wa4 
 
 Snowflake" 
 
 12R.5 
 
 38 
 
 15 
 
 7 
 
 WL10-W..7 
 
 Snowflake 
 
 145.5 
 
 54 
 
 15 
 
 8 
 
 WG9-C10-W8 
 
 Early 
 
 129 . 5 
 
 45 
 
 15 
 
 9 
 
 WSI-W0I2 
 
 Crenate-leaved"''' 
 
 119.5 
 
 34 
 
 7c 
 
 •Suspected before testing of belonging to the early type; first parent also 
 tested in 1910. 
 
 •" A heterozygote between the erenate-Ieaved and Snowflake types. 
 
 ' Probably open pollinated. 
 
 '' All the parents grew in the same house at the same time. 
 
 The essential feature of the supplementary cultures of 1912, since 
 no seed of WLIO remained, was a test of two pairs of early and late 
 progeny of WLIO (lots 3 and 4. 6 and 7, table 21), in comparison with 
 two control lots — one (lot 2) from a known early parent, descended 
 from "WG9-C10, and one (lot 5) from a late descendent of WSl. 
 Incidentally, WSl-WJG and WGg-ClO-WS were retested, and the 
 few available .seeds of WSl-W-,12 were used to test that phenotypieally 
 early parent. 
 
 The results are given in tables 22 and 23 and chart 4. The very 
 low individual from "WSl-WolC came from a very weak embryo, and 
 should be disregarded; the exceptionally high general range of this 
 lot, which was also visibly behind all othei-s in development, was prob- 
 
 1.1141 
 
MUTATION IN MATTHIOLA 
 
 257 
 
 Table 22 
 
 Cultures of 19li. Number of main-stem internodes below first flower-bearing 
 node. Frequency distributions for singles." 
 
 ' See note a to table 5. 
 
 Cultures of 191S. 
 
 Table 23 
 Same as table SS, 
 
 for doubles.' 
 
 
 Gen. 1 
 
 WSl 
 
 WG9 
 
 w 
 
 1.10 
 
 WSI 
 
 WLIO 
 
 WG9 
 
 WSl 
 
 Ancestry - 
 
 Gen. 2 
 
 Wo 16 
 
 CIO 
 
 W22 
 
 W23 
 
 Wol 
 
 W2I4 
 
 V 
 
 f27 CIO 
 
 Wo 12 
 
 Gen. 3 
 
 
 W6 
 
 
 
 
 
 
 W8 
 
 
 Internodes 
 
 18 
 
 it 
 i' 
 
 2 
 
 1 
 
 2t 
 
 i' 
 
 i' 
 
 3 
 
 i' 
 
 1 
 
 4 
 
 1 
 
 
 i ' 
 
 3 
 3 
 1 
 
 
 
 It 
 
 ".'. 1" 
 
 2" 
 
 3 1 
 2 
 
 
 19 
 
 
 
 
 20 
 
 
 21 
 
 
 22 
 
 
 23 . 
 
 
 24 
 
 
 25 
 
 
 26 
 
 
 27 
 
 2 
 
 1 
 
 2 
 1 
 
 
 28 . 
 
 u 
 
 1 
 
 29 
 
 30 
 
 31 
 
 
 32 
 
 
 
 
 33 
 
 
 34 
 
 
 35 
 
 
 
 
 
 
 Gen. 1 
 
 WSl 
 
 WG9 
 
 WLIO 
 
 WSl 
 
 WLIO 
 
 WG9 
 
 WSl 
 
 Ancestry < Gen. 2 
 
 Woie 
 
 CIO 
 
 Wo 2 
 
 Wo 3 
 
 Wol 
 
 WoU 
 
 Wo 7 
 
 CIO 
 
 W212 
 
 Gen. 3 
 
 
 W6 
 
 
 
 
 
 
 W8 
 
 
 Internodes 
 
 12 
 
 lA 
 
 1 
 
 2" 
 3t 
 
 1 
 
 1 
 1 
 
 2 
 2 
 1 
 
 i' 
 
 
 
 
 i' 
 i' 
 
 3 
 
 1 
 
 i' 
 
 1 
 
 1 
 
 1 
 
 2 
 3 
 
 i 
 
 2' 
 2 
 
 1 
 
 2 
 
 
 13 
 
 
 
 
 
 
 
 
 14 
 
 
 15 
 
 
 16 
 
 
 17 
 
 
 18 
 
 
 19 
 
 
 20 
 
 
 21 
 
 
 22 
 
 
 23 
 
 It 
 
 3 
 3 
 2 
 
 
 24 
 
 1 
 
 5 
 
 1 
 
 
 25 
 
 2 
 3 
 
 1 
 
 1 
 
 26 
 
 1 
 
 27 
 
 
 28 
 
 
 
 
 
 It 
 
 29 
 
 30 
 
 
 31 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 " See note a to table 5. 
 
 [1151 
 
258 
 
 MISCELLANEOUS STUDIES 
 
 ably due to some cultural accident, perhaps to an excess of moisture 
 in this row of pots. 
 
 The lots of plants may seem rather absurdly small for their pur- 
 pose, but the uniformity of development here, with the marked normal 
 divergence in internodes of the types in question, seems to justify a 
 fair degree of confidence. Ten plants here were probably worth fifty 
 in the field. 
 
 31 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 I 
 
 ingl 
 oub: 
 
 es 
 
 68 
 
 
 
 
 
 _ 
 
 
 
 29 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 i 
 
 » 
 
 
 
 
 
 
 
 
 
 
 
 2.S 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 O" 
 
 
 
 '' 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 _ — . 
 
 -— . 
 
 
 
 ■26 
 25 
 
 ■s 
 
 ■g 24 
 
 c 
 
 
 
 
 
 
 
 1 
 
 » 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 — . 
 
 .__ 
 
 
 
 
 
 
 
 1 
 
 
 
 
 
 
 
 .__ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 — •< 
 
 •— 
 
 21 
 
 
 
 """■ 
 
 """ — 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 20 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 19 
 
 IS 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 17 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 16 
 
 
 
 ( 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 w^iie W6 
 
 I I 
 
 wsi CIO 
 
 WG9 
 
 Wi2 
 
 Wi.3 
 
 Wsl 
 
 I 
 WSI 
 
 Ancestry 
 
 WiU 
 
 W27 
 
 will 
 
 W8 
 
 1 
 CIO 
 
 I 
 
 WG9 
 
 Wil2 
 
 I 
 WSI 
 
 Chart 4. Cultures of 1912. Internodes: parental values and progeny means, 
 shown as in chart 1. The true parental values are twice those indicated by the 
 ordinate figures, which apply directly to the progeny values. 
 
 This test, with that of 1910, shows very positively that WSl-WjlG 
 was only phenotypically few-noded. Evidently WGO-CIO-WS, the 
 parent of field lot 22, really carried the earliness factor, as was some- 
 what doubtfully inferred from the field results; the five progeny of 
 WS1-W„12, on the other hand, though from a fewer-noded parent, 
 have values that make the presence of the earliness factor improbable. 
 
 On the main point at issue the evidence seems satisfactory. Neither 
 of the two very early and few-noded progeny of WLIO represented 
 
Mr 'I. ir ION IN MATrillOLA 259 
 
 shows in its progeny any evidence of belonging to the early type ; the 
 means are slightly lower than for the many-noded sibs of these parents, 
 but far less so than with the parents descended from "W69-C10. 
 
 We conclude, then, that WG9-C10 was probably a monohybrid, 
 and that the early-bearing gamete entering into its composition was of 
 unknown but presumably mutative origin. 
 
 Most of the extracted late or many-noded parents may now be 
 selected with practical certainty. WGQ-CIO-CS and CI (lots 5 and 6 
 in the 1911P cultures) and WG9-C10-M7 and M8 (lots 13 and 14) 
 were genetically very similar to the check parents, as has already been 
 concluded for two of them from the greenhouse cultures; presumabh" 
 they were pure Snowflake. 
 
 The data for WG9-C10 itself (lot 26) seem to indicate that the 
 results from the last eight lots are of very doubtful value ; still, they 
 show, especially in the original individual records, some evidence of the 
 earliness factor which must be present in part of the individuals. The 
 poor and slow germination of the old seed available may have had an 
 important influence on the result ; many of the early embryos may have 
 been non-viable, and the seedlings may have been weaker than those 
 from fresh seed. The 1911 data and observation of the plants in the 
 field suggest that WG9-C10-W7, W3, and WIO (lots 23, 24, and 251 
 are the only remaining extracted late parents, WG9-C10-W5 and W8 
 (lots 21 and 22) carrying the earliness factor, as the four parents just 
 preceding them in the cultures obviously did. Tables 22 and 23 con- 
 firm this conclusion for WG9-C10-W8. 
 
 It is presumably impossible to make a positive separation of the 
 parents homozygous for the presence of the early factor. The green- 
 house data suggest that WG9-C10-M4 was a pure early individual ; 
 the field data (see lot 9) agree, and suggest that WG9-C10-M9 (lot 
 10) and perhaps WG9-C10-M6 (lot 11) belong in the same cla.ss. 
 WG9-C;iO-C2, C5, and CIO (lots 3, 4, and 40)'- were all evidently 
 heterozygous. Of the parents grown in house W, it would seem that 
 only WG9-C10-W11 (field lot 19) was homozygous early. We have, 
 provisionally, for the available single progeny of WG9-C10 : 
 
 House C House M House W Total 
 
 Pure early 3 14 
 
 Hybrid early 3 15 9 
 
 Pure late 2 2 3 7 
 
 20 
 
 '2 Statistical data given for the last only for the 1910 cultures, not for this 
 field lot. 
 
 [1171 
 
260 MISCELLANEOUS STUDIES 
 
 This corresponds well enough with the monohj'brid expectation of 
 5 : 10 : 5 ; in fact, the deviation is just such as would be expected if there 
 was occasional cross pollination of the unprotected flowers of WG9- 
 ClO from Snowflake plants. The large proportion of evidently pure 
 late parents is strong evidence for the monohybrid nature of WGO-CIO. 
 
 The proportions of the two types among the doubles can only be 
 estimated. The 1908 data suggest that 5 of the 10 doubles there 
 reported were early ; this number, with the 13 singles so classed, makes 
 a total of 18 early-type plants out of 30. The ratio is slightly nearer 
 to 1 : 1 than to 3:1, and the former proportion would suggest the 
 peculiar type of inheritance found with the mutant types yet to be 
 described. The evidence of the 1910 distributions, however, shows that 
 the early type largely predominates in the next generation witli both 
 singles and doubles, and apparently this is true even when we exclude 
 the progeny of the one parent classed as pure early. 
 
 The early factor can be positively detected only by progeny tests. 
 No test has shown the presence of this factor elsewhere than in WG9- 
 ClO and part of its descendants. WG9-C10 produced the early and 
 Snowflake types among 20 single progeny nearly in the typical mono- 
 hybrid proportions. Inspection of the double progeny in two genera- 
 tions suggests similar or possibly somewhat lower proportions there. 
 A vicinistic origin for "WG9-C10 is improbable. Presumably, then, 
 the early tj'pe arose from Snowflake by a single factor mutation, the 
 dominant mutant factor being inherited without special complications. 
 We shall now consider certain apparently mutant types which are 
 characterized by peculiar genetic behavior. 
 
 2. THE SMOOTH-LEAVED TYPE 
 
 This type was first observed in the cultures of 1908 (table 1) and 
 has occurred frequently in later cultures (table 3). It is perhaps the 
 mutant type of most frequent occurrence among progeny of Snow- 
 flake or early parents; 2410 unseleeted progeny from house-sown seed 
 of such parents (see table 28) included 28 apparent mutants (14 
 singles, 11 doubles, and 3 undetermined), a mutation coefficient of 
 1.16 ± .15 per cent. 
 
 As grown in the greenhouse at Ithaca, this type (fig. 7, tables 12 
 and 13) was often many-noded, with correspondingly late flowering. 
 Its most striking peculiarity, shown especially by young seedlings and 
 not evident in the figures, was a lack of buckling between the veins 
 
 [118] 
 
MUTATION IN MATTHIOLA 
 
 261 
 
 S^ '^ 
 
 
 o 
 
 :i so . 
 
 a.S 
 o " 
 
 C c» 
 
 o- &■ 
 
 rt- CD 
 
 C 35 
 
 orq B 
 c c 
 §3. 
 
 &. cr 
 
 o ^ 
 
 f5 rti 
 
 3 S 
 
 S-H. 3- 
 
 3 Ct) 
 
 e b 
 
 3 
 
 =. 
 
 flQ 
 
 
 C 
 
 r! 
 
 P 
 
 
 •^ 
 
 
 D- 
 
 r^ 
 
 
 
 
 t3 
 
 
 (JQ 
 
 » 
 
 
 
 
 ^ 
 
 CD 
 
 "1 
 
 Cu 
 
 
 o 5". 
 
 
 o = 
 31 
 
 gggggg^Sfe 
 
 All P, 
 smoot 
 26b-l 
 
 26c-S 
 
 to to to 
 
 CD 35 en 
 
 
 ? 
 
 
 
 
 i T 
 
 o er p 
 
 
 3 
 
 
 .^00 =^ 
 
 S-~ 
 
 On- 
 
 
 
 Cfi 
 
 
 in 
 
 
 
 
 
 
 > 
 
 -1 w ►-* h- ^ s> 
 J5COO « to C 
 
 > 
 
 ce CO CD O CD — 
 
 
 
 
 
 CO CD CO CD CD 
 
 
 o 
 
 
 
 
 
 
 
 
 
 
 
 -- CO CO w »— 
 
 
 CO 1— CJ 1— 1— 
 
 H- Oh- O^ 
 
 
 
 
 
 n a 
 
 
 -^ ffi-^ a a 
 
 ^R^^R^^ 
 
 
 5 
 
 
 
 
 
 CD' CD" 
 
 
 
 
 
 
 
 I—' h-» 
 
 
 
 
 
 
 
 
 4^ 4^ 
 
 
 
 
 
 to 
 
 en 
 
 to 
 
 I-- •-• to 
 
 »-^ 
 
 
 
 
 o 
 
 SD to O CJitO O 
 
 Ci 
 
 I-- ^ to 4^ 
 
 4- ^^ fs 
 
 
 
 
 GO 
 
 O CO_Cn O 0_p 
 
 Oi 
 
 00 00 O 4* 
 
 to o e« 
 
 
 
 
 
 
 
 
 
 a- 
 
 
 "cn 
 
 Cr, ^^ 
 
 
 t« 
 
 
 
 
 
 o 
 
 
 
 ■^ 
 
 en 
 
 
 
 
 
 
 
 
 
 3? 
 
 H 
 
 
 
 ts3 
 
 OOtOOOGO 
 
 Ol 
 
 CO to oco 
 
 eoi o o 
 
 ^5 
 
 o 
 
 s 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 P 
 
 
 
 h-' 
 
 to 
 
 1 
 
 ^ 
 
 
 ^S' 
 
 
 
 
 -J 
 
 to to ~J J^ H- CO 
 
 
 en Cn 1— 
 
 to to en 
 
 B'% 
 
 5' 
 
 
 
 W 
 
 O OCO H-CC ^ 
 
 lO 
 
 en 05 00 en 
 
 H- CO en 
 
 
 s. 
 
 
 
 
 
 
 
 
 m 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 d" 
 
 
 
 
 
 
 
 : O 
 
 -4 
 
 C3 O H-CO 
 
 to h- 
 
 
 
 
 
 
 
 
 
 
 
 CB 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 a 
 
 O 
 
 o 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 -1 
 
 
 
 
 
 
 
 
 
 
 D" 
 
 
 
 
 
 
 +- 
 
 IX 
 
 Oi to OOi 
 
 *- to 
 
 
 (T 
 
 
 ffl 
 
 
 
 
 
 
 JT 
 
 
 
 3 
 
 
 a 
 
 CO 
 
 ►-• H- CO 
 
 „ 
 
 
 
 O 
 
 OOOOOCO 
 
 
 CD ^ ,-' to 
 
 00 o: 00 
 
 > 
 
 
 
 
 
 
 
 
 
 
 2 
 
 
 
 a 
 
 CO 
 
 to 
 
 
 ^ 
 
 
 p* 
 
 3 
 
 
 
 4i 
 
 to 
 
 o 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 GC 
 
 H- CO to ** 
 
 CO 
 
 H- to 
 
 
 
 
 
 
 
 -JH-CO *.C: to 
 
 *- 
 
 to o tooo 
 
 to o 
 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 5" 
 
 
 
 
 00 
 
 xw'coto rf^ 
 
 CO 
 
 1— to 
 
 
 
 ^ 
 
 
 
 
 Cn 
 
 — - to 4^_oi en 
 
 M 
 
 ■SS i5 
 
 
 
 cT 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ~J 
 
 H- CO h- 1— *. 
 
 CO 
 
 h-* h- > 
 
 
 
 o 
 
 D 
 O 
 
 
 
 O 
 
 O ^CCCO ►-■-J 
 
 CD 
 
 03 CD 4^00 
 
 CO en 
 
 
 
 
 
 
 
 
 
 
 c 
 
 f 
 
 
 
 00 
 
 ►- 1- Oi 
 
 4^ 
 
 H- to 
 
 
 
 cr 
 
 3i 
 
 
 
 w 
 
 Ci w o 
 
 _io 
 
 00 o 
 
 
 
 P 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 H-i 
 
 
 
 
 
 
 
 
 02 
 
 H- H- M CO >-» en 
 
 M 
 
 to *- ^ 
 
 H- to ** 
 
 
 
 
 
 
 C« 00 ^— M ^H- 
 
 4^ 
 
 GO O 05-J 
 
 H- to 4^ 
 
 > 
 
 
 
 
 
 
 
 
 
 
 
 
 
 H^C^Q"^^!^ 
 
 ■00 
 
 CO *. 00 
 
 H- »^ 
 
 
 
 
 
 C5 
 
 tO^MOCD Ol 
 
 o 
 
 ^ CO •- 
 
 OO 05 
 
 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 ■--^ 
 
 
 
 
 
 
 
 
 w 
 
 ^- H- H-O *- 
 
 CO 
 
 H- to ►- 
 
 O H- 
 
 *< <: 
 
 I 
 
 
 
 
 
 
 
 
 "D : 
 
 
 
 
 .1^ 
 
 r- '-0 
 
 ^ 
 
 ^ ^M 
 
 
 S £ 
 
 5 
 
 
 
 ' 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 g 
 
 to 
 
 [1191 
 
262 MISCELLANEOVS STUDIES 
 
 of the leaves, and of general convexity of the upper surface of the 
 leaves. Mature plants developed under favorable conditions in the 
 greenhouse closely resembled Snowflake; the leaves, however, were 
 noticeably brittle, and the dry capsules so brittle that it was often 
 necessary, as it was not with Snowflake, to shell the seeds individually. 
 Probably the tibrovascular system is in some way defective; Oenothera 
 rubrinervis, which is also brittle (de Vries, 1906, lecture 18), has 
 thin-walled bast fibers. 
 
 In the field cultures, both at Ithaca (fig. 5) and at Riverside, under 
 conditions less favorable on the whole to the initiation of flowering, 
 this tj'^pe (fig. 8) diifered much more widely from Snowflake. Flower- 
 ing was excessively delayed, and the plants often remained low, with 
 few branches, and rosette-like, with thin, rather narrow leaves. Small 
 brown dead spots, possibly due to excessive transpiration, occurred so 
 frequently on the leaves as to constitute a good diagno.stic character 
 for the type. Another peculiarity observed in the field is a reflexed 
 position of the tip of the j'oung leaf when first visible — Snowflake 
 leaves being completely erect from the first. 
 
 In the 1914 cultures, with better development than in other field 
 cultures, some smooth-leaved plants (figs. 9 and 10) were again more 
 like Snowflake, though later and evidently more leafy. 
 
 Six smooth-leaved parents have been used in progeny tests, three 
 of these being apparent mutants and three being Fj progeny of two 
 of those mutants. The results are presented in tables 24 and 25 ; those 
 tables require a brief explanation, which will apply also to the similar 
 tables for other types. 
 
 For the plan of the new pedigree numbers here used, see "^Methods. " 
 The initial plants of a series are designated as the Pi generation in 
 the tables, their progeny as Fi, ete. In table 24 the cultures are 
 aiTanged according to their generations and their pedigree numbers 
 under each generation; the smooth-leaved parents (P, or of the P, 
 type) are given first, followed by the extracted Snowflake parents. 
 In table 25 "good germination" indicates that in all lots included 
 (taken as grown, not as summed by parents in table 24) the number 
 of plants determined exceeds 50 per cent of the number of seeds sown. 
 and vice versa; the weighted mean percentages obtained by dividing 
 the total numbers of plants by the respective total numbers of seeds 
 are given for each table in a footnote. 
 
 All six smooth-leaved parents (tables 24 and 25) gave mixed 
 progeny, part smooth-leaved and part Snowflake, The surprising 
 
 [120] 
 
Mrr.iTioy in matthiol.i 
 
 263 
 
 fact is that the parental (smooth-leaved) type appears not in three- 
 fourths of the progeny, but in only about one-fourth. 
 
 The extracted Snowflake parents tested behave like pure recessives, 
 showing no intluence of their smooth-leaved ancestry. Only the 
 aberrant ratio seems inconsistent with the assumption that the smooth- 
 leaved individuals tested were ordinary heterozygous dominants. 
 
 The relatively weak growth of this type and the apparently poor 
 germination of the seed produced by it suggast that normal segregation 
 may be masked by selective elimination. Possibly the smooth-leaved 
 
 T.A.BLE 2.5 
 
 Smooth-leaved type: heredity. Summary. 
 
 
 Progeny 
 
 
 Cultures 
 
 Seeds 
 
 Plants 
 
 Parents 
 
 Total 
 
 examined 
 
 Smooth-leaved 
 
 
 Undeter- 
 
 Deter- 
 
 
 
 
 
 
 mined 
 
 mined 
 
 Number 
 
 Per cent 
 
 All smooth- 
 
 
 
 
 
 
 
 leaved 
 
 Ithaca 
 
 304, ;2/ 7 
 
 7 
 
 156 
 
 40 
 
 25.6 ± 2.4 
 
 All smooth- 
 
 
 
 
 
 
 
 leaved 
 
 Riverside 
 
 196 
 
 1 
 
 78 
 
 23 (24) 
 
 30.8 ± 3.4 
 
 All smooth- 
 
 
 
 
 
 
 
 leaved (6) 
 
 All 
 
 500,5/7 
 
 8 
 
 234 
 
 63 (64) 
 
 27.4 ± 2.0 
 
 All P, smooth- 
 
 
 
 
 
 
 
 leaved (3) 
 
 All 
 
 244, ?/r 
 
 3 
 
 115 
 
 32 
 
 27.8 ± 2.8 
 
 All Fi smooth- 
 
 
 
 
 
 
 
 leaved (.3) 
 
 .\11 ■ 
 
 2.56 
 
 5 
 
 119 
 
 31 (32) 
 
 26.9 ± 2.S 
 
 All smooth- 
 
 Germination 
 
 
 
 
 
 
 leaved 
 
 good 
 
 293, 138 
 
 8 
 
 187" 
 
 55 (.56) 
 
 29.9 ± 2.2 
 
 All smooth- 
 
 Germination 
 
 
 
 
 
 
 leaved 
 
 poor 
 
 207, 79 
 
 
 
 47a 
 
 8 
 
 17.0 ± 4.4 
 
 All Sno\vflake 
 
 
 
 
 
 
 
 (5, Fi and F.) 
 
 AH 
 
 208, -50 
 
 2 
 
 173 
 
 
 
 
 
 ■ Respectively 63.8 and 22.7 per cent of the numbers of seeds planted. 
 
 factor is lethal when homozygous, as is often the case (MuUer. 1918) 
 with dominant mutant factors in Drosnph ila ; the data for germi- 
 nation, however, indicate that two-thirds of the mature embryos can 
 hardly belong to the mutant type. We might expect, in view of the 
 weak growth of .smooth-leaved plants, that partial elimination of 
 heterozygotes would also occur. That this is the case is suggested, 
 though the numbers are small, by the lower proportion of the mutant 
 type with poor germination (table 25; see also tables 39 and 40) ; it 
 should be noted, however, that transferring the first lot of table 24, 
 the only lot between 50 and 73 per cent, to the "poor" total, makes 
 the percentages practically identical." 
 
 '3 See also table 2 and the second paragraph under "Occurrence of Mutants." 
 
 [121] 
 
264 MISCELLANEOUS STUDIES 
 
 In connection with the qnestion of lethal action we must consider 
 the inheritance of doubleness of flowers. Snowflake seed regularly 
 gives a mixture of singles and doubles, about 53 per cent being doubles. 
 The doubles, which are totally sterile, are probably (Frost, 1915) pure 
 reeessives (dd) for a single-double factor pair. The singles are always 
 heterozj'gous (Dd) ; crosses with pure single races fSaunders, 1911) 
 show that the approximately 1 : 1 ratio and the failure to produce pure 
 singles, with self pollination, are due to the fact that all the functional 
 pollen is doubleness-carrying (d). The excess of doubles over 50 per 
 cent has been explained by Miss Saunders (1911) as due to hetero- 
 zygosis of the singles for two linked complementary factors necessary 
 to singlenass, and by the present writer (Frost, 1915) as due to lower 
 viability of the "single" gametes or embryos. The absence of func- 
 tional single-carrying pollen is apparently due to a lethal factor acting 
 after separation of the microspore tetrads, since the tetrads themselves 
 appear normal. 
 
 In any consideration of factors linked with the single-double pair, 
 this semisterility of the pollen must be remembered. For example, 
 any dominant factor completely coupled with D in pollen formation 
 would be totally absent from the functional pollen, and the zygotes 
 produced by selfing would show directly the .strength of linkage in the 
 ovules. 
 
 The available data for the smooth-leaved type (table 24) are far 
 from con.stituting an adequate test of linkage, but they suggest that 
 the factors are independent. Certainly no high degree of linkage is 
 indicated by the totals, nor do the detailed data .suggest that smooth- 
 leavedness is coupled with singleness in some parents and with double- 
 ness in others. 
 
 We must admit that the peculiar inheritance of this type is not 
 yet positively explained. Evidently larger cultures are needed, and 
 crossing with the Snowflake type and with other commercial varieties ; 
 cytological .study may also be required. Certain comparisons and 
 speculative possibilities deserve mention, however, especially since the 
 types yet to be discussed furnish additional evidence bearing on them. 
 We may compare the smooth-leaved and double types, as follows : 
 
 DouBi.E Smooth-leaved 
 
 1. A rare mutation of pure single 1. Apparently a common mutation 
 
 ("normal"). of pure Snowfliike ("normal"). 
 
 2. Recessive; extracted reeessives 2. Apparently dominant; extracted 
 
 are sterile mutant-type plants. reeessives are fertile normal 
 
 plants. 
 
MUTATION IN MATTHIOLA 
 
 2fi5 
 
 Double 
 
 3. Homozygous dominants not pro- 
 duced by hybrids, because func- 
 tional pollen carries recessive 
 factor only. 
 
 4. Eecessive (mutant) type the 
 
 more vigorous. 
 
 Dominant factor or another fac- 
 tor very closely linked with it 
 is incompatible with formation 
 of functional pollen. 
 
 Eecessive type exceeds the ex- 
 pected equality by about 3 per 
 cent among some 7000 indi- 
 viduals. 
 
 Smooth-leaved 
 3. Homozygous dominants perhaps 
 not produced by hybrids.'* 
 
 4. Recessive (normal) tyjie the 
 more vigorous; difference much 
 greater than with single and 
 double. 
 
 .5. Relation of dominant factor to 
 viability of pollen not yet de- 
 termined. 
 
 6. Eecessive type exceeds equality 
 by about 23 per cent among 
 234 individuals. 
 
 The most probable hypothesis for .smooth-leavedness, theu, would so 
 far seem to be essentially the same as for doubleness — complete elimina- 
 tion of the weaker type in pollen formation, and partial elimination in 
 embryo-sac formation. Reciprocal crosses with Snowflake are obviously 
 necessary; as we shall soon see, three of the other mutant types have 
 already proved to he carried by both eggs and sperms. 
 
 The case of Oenothera lata (Gates, 1915) suggests the possibility 
 that the smooth-leaved form might arise by reduplication of a chromo- 
 some. With ordinary 0. lata the pollen is sterile, but pollination by 
 0. lamarckiana gives about 15-20 per cent of lata. This deficiency of 
 lata individuals is due, it seems, to a frequent loss of the extra 
 chromosome at meiosis in lata ovules, with a resulting formation of 
 more than 50 per cent of seven-chromosome {lamarckiana) eggs. 
 
 If the smooth-leaved type originates through duplication of a 
 chromosome, we might .suppose that other types of similar heredity 
 involve other pairs of chromosomes. The apparent parallel with 
 0. lata, which Bartlett (1917) has noted, was long ago suggested by 
 the data, but with at least two or three types to be described linkage 
 phenomena have seemed to conflict with this interpretation. Possibly 
 different processes have produced different mutant types as with 
 Oenothera ; as we have considered types siiggestive of 0. rubrinervis 
 (early) and of 0. lata (smooth-leaved), we ma.v consider next a form 
 which in appearance is remarkably suggestive of 0. gigas. 
 
 '■• This possibility is only suggested by these cultures, but it becomes highly 
 probable when the data for other types are considered. 
 
 [ 123 I 
 
266 
 
 MISCELLANEOUS STUDIES 
 
 
 
 
 
 u n 
 
 u o 
 
 
 
 
 
 
 
 
 ^ °' 
 
 
 
 
 
 
 
 
 0-& 
 
 MOOOOOOOOOrtrtO-H^C 
 
 ^ oooo o 
 
 CD 
 
 O 
 
 
 
 
 
 
 
 ^ 
 
 „ 
 
 
 
 
 
 
 
 
 t^ ,--,,-^ 
 
 CO 
 
 
 
 
 
 
 
 ^^O^ ^ rf-^ ^ 
 
 -H,-,,— , OOC 
 
 t~ 
 
 
 
 
 
 
 
 CC CC w CO t^ ^ CO ^ 
 
 C!S2- CI-G 
 
 
 
 
 
 
 
 ■< 
 
 
 
 
 
 
 
 
 
 lOOOO-*OOrHiOTtHCOOOOC>Ct~Ot~CO 
 
 ^ t^ •^t^ -^ C3 -^ 
 
 o 
 
 >o 
 
 
 
 
 
 
 (M rt rt 
 
 o c^ 
 
 tC 
 
 f-H 
 
 
 
 
 
 
 
 *-( 
 
 '"' 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 aj 
 
 
 
 
 
 
 
 
 ^ 
 
 
 1^,-v .-^ ,^ 
 
 CC^^ — Ol 
 
 b- 
 
 
 
 
 
 
 3 
 O 
 
 Q 
 
 ^CO o cc 
 
 SS- SC- 
 
 s 
 
 
 
 
 
 s 
 
 
 
 
 
 
 
 
 o 
 
 C^OlCC^t^OOT^CC'^'-^Tj^CC'r'M'-i 
 
 O CM — ' .-< UO C2 
 
 O 
 
 00 
 
 
 
 
 l-H 
 
 lO 
 
 00 
 
 
 06 
 
 
 
 
 
 
 
 
 
 1 
 
 
 
 
 
 
 8 
 
 
 
 
 
 
 
 
 
 
 
 
 _aj 
 
 
 00 --^ o 
 
 GO 
 
 
 
 
 
 "m 
 
 -H io f c^ ro 
 
 3', ^ -H 
 
 r- 
 
 
 u 
 
 
 
 
 a 
 
 
 
 
 
 Co 
 
 
 
 
 m 
 
 coocOfM^ioc^JOr^csjc^^-^otcc^i 
 
 CM lO CO CO -*! lO 
 
 o 
 
 t^ 
 
 
 
 
 
 '" 
 
 rt- .-H 
 
 r- 
 
 
 
 
 
 
 
 
 
 ■ea 
 
 
 -2 
 
 
 
 
 ^^ 
 
 ^^ 
 
 
 '« 
 
 
 a 
 
 
 
 
 05 --^ 
 
 <o 
 
 
 
 
 <a 
 
 
 
 o'^^^^ --.o --^o^^-^ .-~. 
 
 O— , CO 
 
 r- 
 
 
 s 
 
 
 S 
 
 
 
 ■^ CO t^ G: lO ^ CD T-H r-H Ci CO 
 
 CtS £1 
 
 
 
 
 
 
 
 
 
 
 ■M 
 
 
 
 
 "" 
 
 00(MOt~C<lCJX02C0O(Na:Ot^C0C.l 
 
 O CO CD C) O: lO 
 
 co 
 
 o 
 
 
 
 
 
 
 CO --I —1 
 
 C5 CM 
 
 lo 
 
 
 e 
 
 
 
 
 
 
 
 
 
 g 
 
 
 
 
 
 
 
 
 
 C 
 
 
 -d 
 P 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 s 
 
 
 
 CO '-^ C^l 
 
 'f 
 
 
 
 o 
 
 
 u 
 
 ^ 
 
 ^cojJcD cor-, 2<, S, d 
 
 S2^ "^. 
 
 X 
 
 
 k> 
 
 
 
 3 
 
 
 
 
 
 a 
 
 Pa 
 
 
 & 
 
 O 
 
 Q 
 
 Tt((NCO>OC^^-Xt^COCO-"!j"OlOOO 
 
 r^ CO CO CM CO ^ 
 
 CM 
 
 o 
 
 a 
 
 
 
 ,3 
 
 I— 1 
 
 ^ rf 
 
 t^ 
 
 
 
 
 
 
 
 
 
 
 8 
 
 
 
 
 
 
 c 
 
 
 
 
 _a; 
 
 ^ ^^ ,-v ,~,--^^--,-^ 
 
 CM 
 
 f^ 
 
 
 
 
 
 
 bC 
 
 mSr Ol f?0 CD OQ 
 
 -# 
 
 t^ 
 
 
 
 
 
 C 
 
 
 
 
 
 ^ 
 
 
 
 
 W 
 
 cooc^^r-^^c^c<ico^to»o^c^c^^ 
 
 »0 CO CO O CD C^l 
 
 O 
 
 o 
 
 1 
 
 2i 
 
 
 
 
 
 c^ 
 
 CO •-' 
 
 r~ 
 
 
 
 
 -3 
 
 S 5 
 
 co5Dt-t~0'ir^coCJTt<c»h-acOrtl^ 
 
 O 00 ^ CD C5 ■<1* 
 
 h* 
 
 uo 
 
 
 
 
 
 t^ S 
 
 t^ —1— (lN^-HT-lr-lrH,-l^rtlM.-C 
 
 CO ^ ^ rH lO 
 
 "O 
 
 1— > 
 
 w 
 
 
 
 C 
 
 ae 
 
 
 CM 
 
 CO 
 
 
 Pi* 
 
 
 
 ca 
 
 
 
 
 
 
 >i 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 
 
 ■^ 
 
 
 
 O 
 
 ^-3 
 
 II 
 
 
 
 
 
 
 
 
 (M„OrfOOOOOOOMOOOO 
 
 i<ooooo 
 
 CD 
 
 o 
 
 1 
 
 
 
 H 
 
 d6 
 
 
 
 
 
 <D 
 
 
 
 
 
 ^ 
 
 
 C!l 
 
 
 
 
 « 
 
 
 # 
 
 
 S^ 
 
 
 
 
 05 —1 
 
 
 
 
 13 
 
 
 -2 
 
 
 * *-* »-( CO 
 
 
 CO 
 
 
 (^ 
 
 
 'Z 
 
 
 t- >-. -* . 
 
 
 
 « 
 
 
 
 S 
 
 
 
 o"^ 
 
 r« 
 
 
 
 tc 
 
 
 CMS5 
 
 CM 
 
 
 
 
 
 rH (M (N 'i' CO 
 
 CO ^ 
 
 
 
 
 
 TjTo 
 
 
 
 
 
 
 
 
 i-H f-H 
 
 
 
 
 
 
 o 
 
 
 Oil CDCDCDCDCD'OCDCDCDOCDO 
 
 CD O CD CD CD 
 
 
 CO 
 
 
 
 ' ,t 
 
 
 r-HiO i-H^Ht-Ht-).— ti— *»-H^H^Hi— 't-*.-H 
 
 ^- ^- .— 1 rH rH 
 
 
 
 
 
 3 
 
 
 -^ 1 1 1 1 1 1 1 1 1 1 1 1 
 
 1 1 1 1 1 
 
 
 1 
 
 
 
 ^ 
 
 
 CO C:^ rt^ -t ^ lO lO to UO lO »0 »0 >0 »0 UO lO lO 
 
 »C 'O »o »o l-O 
 
 
 lO 
 
 
 
 a 
 
 
 
 ^H r-l .— 1 t-H T-H 
 
 
 T-l 
 
 
 
 u 
 
 
 Ci ^GiClCnOOOOC-. ClOOlOlOlOC:'-; 
 
 <J3CiOl Oi Gi 
 
 < 
 
 o 
 
 
 
 
 
 ^ i-H l-H i-H i-( 
 
 
 
 
 
 
 OOO 00 1 
 
 
 v-^ 
 
 
 
 
 
 CO CO CM CM ^ 
 
 
 V 
 
 
 03 
 
 
 CBt^cDl^COiOt^CMcOt^-lCOr- 
 
 III ," ?'•— 
 
 lo-^ 
 
 
 a 
 
 
 1 1 1 1 1 1 1 1 1 1 1 1 1 1 
 
 o3 ao3cS!:3cJric3cSc3c3c333c3c!oSci 
 
 ooopo5cnC^_5 
 
 (D 
 
 tN- 03 
 
 1 ^ 
 
 
 £ 
 
 
 03 c« ri c! S — 
 
 c3 
 
 d^ 
 
 
 ^ 
 
 
 00 00 00 00 CC OC CC 00 00 cr 00 00 CO X 00 00 GO 
 
 00 00 CC CO CO —: 
 
 C-) CM CM CM CM < 
 
 2S S 
 
 CM Q 
 
 
 Cu 
 
 
 M C<l CM N ei IN (N C^ N 01(N C<1 CM (N O) C<1 !N 
 
 
 
 
 
 
 
 m 
 
 
 
 
 
 
 
 
 
 — 
 
 [124 1 
 
UVTAriOX IN MAirHIOLA 267 
 
 It should, however, first be noted that, as will appear later, phe- 
 nomena of apparent linkage in the ease of certain other types (erenate, 
 slender, and narrow) suggest that these forms commonly arise from 
 Snowflake by segregation rather than by immediate mutation. The 
 obvious objection to this hypothesis is the fact that the apparently 
 mutant types seem to be dominant to the ' ' normal ' ' or Snowflake type. 
 This objection can be met by assuming the presence of dominant in- 
 hibiting factors in the Snowflake parents that give apparent mutants.'"' 
 
 If the apparent mutants of the smooth-leaved type are thus pro- 
 duced by crossing over in a set of balanced factors, the lethal "balanc- 
 ing" the smooth-leaved factor itself may be distinct from that which 
 sterilizes the singleness-carrying pollen. In considering the results 
 here reported, therefore, we must always bear in mind the possible 
 presence of several unidentified lethal factors. If the apparent absence 
 of linkage between the smooth and double factors is not misleading, we 
 must suppose that these factors are carried by different pairs of 
 chromosomes; considerations advanced by Muller (1918, pp. 479-482), 
 however, make it rather probable that the commoner types of apparent 
 mutants here discussed are all due to factors carried by one pair of 
 chromosomes, the pair containing the factor for doubleness and its 
 normal allelomorph. 
 
 3. THE LARGE-LEAVED TYPE 
 
 A double of this t.ype probably occurred in the 1907 cultures, 
 though its appearance attracted so little attention that no record was 
 made. In the field cultures of 1911 (table 3) several individuals sug- 
 gested a gigas type, though there seemed to be intergradation with 
 Snowflake. In the 1912 cultures a single with leaves "long, rather 
 narrow, thick" developed normally and produced an abundance of 
 good seed; from this individual f28a) all cultures of this type are 
 descended. 
 
 This type is stout and coarse throughout, and late to flower. The 
 leaves are strikingly long, thick, and rigid, though as a rule relatively 
 
 i-"" A letter suggesting this explanation was received from Dr. Muller soon 
 after the same idea had been outlined in the "General Discussion" section below. 
 Dr. Muller kindly gave further attention to difficulties at first encountered by 
 the present writer, materially assisting in the formulation of an apparently 
 tenable form of the hypothesis. Since, however, this scheme may seem "far- 
 fetched" and unduly complex, it appears desirable to leave the original discus- 
 sion of the individual types substantially unchanged. When the difficulties 
 encountered by the assumption of frequent true mutation have been more fully 
 presented, the need for some such addition to the scheme will be more evident. 
 
 [1251 
 
268 
 
 MISCELLANEOUS STUDIES 
 
 narrow ; under unfavorable weather conditions the flowers are often few 
 and defective, while the leaves are resistant and long-lived (fig. 11). 
 Figures 12 and 13 show well the coarse leaves and lateness of well 
 developed large-leaved plants in the 1915-16 cultures, the plants in 
 the latter figure being several weeks the older. 
 
 The results of the progeny tests are given in tables 26 and 27. All 
 the twenty large-leaved individuals tested have given mixed progeny; 
 the proportion of the mutant type, though much larger than with 
 
 Table 27 
 Large-leaved type: heredity. Summary. 
 
 
 Progeny 
 
 
 Cultures 
 
 Seeds" 
 
 Plants 
 
 Parents 
 
 Total examined 
 
 Large-leaved 
 
 
 
 
 
 
 
 
 
 mined 
 
 mined 
 
 Number 
 
 Per cent 
 
 28a 
 
 1013, 1914, 
 
 
 
 
 
 
 
 & 1915-16 
 
 122 
 
 2 
 
 73 
 
 38 (40) 
 
 54.8 ± 3.9 
 
 28:i-F, (3) 
 
 1914 
 
 120 
 
 2 
 
 40 
 
 14(19) 
 
 47.5 ± 5.3 
 
 28a-F, (12) 
 
 1915-16 
 
 288 
 
 2 
 
 190 
 
 76 (90) 
 
 47.4 ± 2.4 
 
 28a-F2 (4) 
 
 1915-16 
 
 90 
 
 
 
 54 
 
 25 (26) 
 
 48.1 ± 4.6 
 
 28a-F, & F, (19) 
 
 .\11 
 
 498 
 
 4 
 
 284 
 
 115(135) 
 
 47.5 ± 2.0 
 
 All large-leaved 
 
 
 
 
 
 
 
 (20) 
 
 .A.11 
 
 620 
 
 6 
 
 357 
 
 153 (175) 
 
 49.0 ± 1.8 
 
 Large-leaved 
 
 Germination 
 
 
 
 
 
 
 
 good 
 
 360 
 
 3 
 
 260'> 
 
 115(131) 
 
 50.4 ± 2.1 
 
 Large-leaved 
 
 Germination 
 
 
 
 
 
 
 
 ])oor 
 
 260 
 
 3 
 
 97" 
 
 3S (44) 
 
 45.4 =t 3.4 
 
 Snowflake (1,F,) 
 
 191.5-16 
 
 24 
 
 
 
 15 
 
 
 
 
 
 ■ Mainly from unguarded flowers; see table 26. 
 
 " Respectively 72.2 and 37.3 per cent of the numbers of seeds planted. 
 
 smooth-leaved, approximates to 50 per cent, not 75 per cent, with little 
 indication of selective elimination with poor germination."' 
 
 Here plainly, as with smooth-leaved, no pure mutant-type parent 
 has yet been tested. Since this is also true of the other types, aside 
 from early, that have been somewhat extensively tested, and fifty-three 
 mutant-type parents in all have given Snowflake progeny, it is prob- 
 able that homozygous individuals of these types seldom or never 
 develop. The actual adult ratio with large-leaved is plainly not 2 : 1, 
 but rather 1 : 1, a fact that would suggest absence of the mutant-type 
 factor or factors from the pollen. The small trial cultures .started 
 in 1917, however, show that the type is carried by both sperm and 
 eggs. 
 
 i« Since hybrids are of the mutant type in appearance, the possible cross 
 pollination by Snowflake parents could hardly give Snowflake progeny with any 
 pure large-leaved parent. It may, however, have reduced slightly the proportion 
 of large-leaved progeny from heterozygous parents of this type. 
 
 fl261 
 
MUTATION IN MATTHIOLA 
 
 269 
 
 If we are dealing here with a type cytologieally like Oenothera 
 gigas, or rather the triploid semigigas, abnormal distributions of 
 chromosomes may occur at meiosis, giving unpredictable genetic 
 results. There has been special difficulty, as the numbers of doubtful 
 individuals in table 26 suggest, in separating large-leaved from Snow- 
 flake, though in part of the cases the diffei'ence is extreme. Possibly 
 some of the doubtful individuals are genetic intermediates due to 
 irregular meiosis in triploid nuclei ; such irregularities in division 
 (Gates, 1915) occur with Oenothera. Both cytological examination 
 and crosses with Snowflake are plainly required. 
 
 Table 28 
 
 Crenate-leaved type: numbers of apparent mutants and association of the 
 type with singleness of flowers. 
 
 
 Progeny of Snowflake and early parents 
 
 
 
 
 
 
 Culture 
 
 Total 
 examinedi* 
 
 
 
 
 
 Single 
 
 Double 
 
 All 
 
 Coefficient of 
 mutation 
 
 1908 
 
 725" 
 
 6 
 
 1 
 
 7 
 
 .97 =t .22 
 
 1910 
 
 338 
 
 3 
 
 
 
 3 
 
 .89 ± .32 
 
 191 IF, seed house- 
 
 
 
 
 
 
 sown 
 
 2072 
 
 13 
 
 3 
 
 16 
 
 .77 ± .13 
 
 All above 
 
 313.5 
 
 22 
 
 4 
 
 26 
 
 .83 ± .11 
 
 All unsclected 
 
 2410 
 
 16 
 
 3 
 
 19 
 
 .79 ± .12 
 
 * See note 6 to table 2. 
 ■• See note c to table 1. 
 
 4. THE CRENATE-LEAVED TYPE 
 
 This type (tables 1 and 3) is one of the three aberrant types of 
 most frequent occurrence in the cultures here described, having con- 
 stituted (table 28) about .79 per cent of the progeny of Snowflake 
 and early parents. A large majority of the individuals have been 
 singles, as table 28 shows. If the apparent mutants are produced by 
 some process of segregation of factors, evidently the crenate and single 
 factors were usually coupled in this material; if they are produced 
 by immediate factor mutation, or are individually due to some change 
 in a particular locus, evidently that locus is linked with the single- 
 double locus and the change is more frequent in the single-carrying 
 chromosomes; and finally, if they are due to reduplication or loss of 
 a chromosome, the apparent linkage remains to be explained. 
 
 The margins of Snowflake leaves vai-y from entire or slightly 
 sinuate to coarsely and irregularly dentate or serrate, this character- 
 istic being subject to much environmental modification and varying 
 
270 
 
 MlSCELLANEOrS STUDIES 
 
 
 
 
 
 
 
 
 
 
 J^ u 
 
 ^ 
 
 
 
 
 
 
 
 
 
 O 
 
 
 
 
 u £ 
 
 
 G 
 
 CO 
 
 CO «-! CD 
 
 G- 
 
 
 
 ^ a 
 
 
 
 
 
 
 
 
 o5 
 
 rt o 
 
 o —■ 
 
 (M 
 
 -HC-)-HIN (MO Tf CO 
 
 
 
 
 
 
 ^ ^ 
 
 
 "^ 
 
 
 
 
 
 
 
 (M 
 
 
 CO 
 
 
 
 
 
 oo 
 
 
 O 
 
 O O -:)• 
 
 d 
 
 
 
 
 
 CI CO 
 
 CO 
 
 
 G fc. S 
 
 G 
 
 
 
 
 -< 
 
 
 
 
 
 
 
 
 
 
 02C/C' 
 
 —1 o 
 
 00 
 
 03rt(M003 — O lO (N 
 
 00 
 
 
 
 
 
 -^M 
 
 IN CO 
 
 02 
 
 ,-( CO ^ ^ ^ CO lO 
 
 
 
 
 0) 
 
 
 
 
 
 
 e<i 
 
 
 
 
 
 
 ^ ^ 
 
 
 
 ^ 
 
 
 
 
 
 
 o 
 
 
 
 ia 
 
 
 lO 
 
 
 ^ 
 
 (M CO 
 
 en 
 
 
 
 ec 
 
 .2 
 
 £J 
 
 
 2- 
 
 CO ^ 
 
 *— 1 
 
 
 
 ^ 
 
 3 
 O 
 
 P 
 
 
 
 
 
 
 
 
 
 
 IN^ 
 
 o r- 
 
 CO 
 
 00 CO o f~ o 00 lo r^ ^ 
 
 CO 
 
 
 
 c 
 
 CJ 
 
 
 -^ 
 
 -H oq lO -*i 
 
 00 
 
 i-H 
 
 
 
 
 
 
 
 
 
 
 
 
 
 o 
 
 
 
 
 
 CO 
 
 
 
 
 "if 
 
 lO 
 
 
 t-^ 
 
 CI 
 
 CO 
 
 
 
 
 c 
 
 
 
 
 
 
 
 
 
 i» 
 
 o^ 
 
 o c^ 
 
 CO 
 
 ^ lO CO O C CO IC 00 CO 
 
 C7 
 
 
 CD 
 
 
 
 
 
 
 
 ~ 
 
 
 ** 
 
 
 
 
 
 
 
 00 
 
 
 A 
 
 
 
 
 
 
 -V ^^ IQ 
 
 CN 
 
 
 
 
 
 
 
 O CO lO CO 
 
 
 
 cu 
 
 
 < 
 
 
 
 
 
 
 
 
 
 
 ^o 
 
 rf Ci 
 
 o 
 
 M 
 
 00 CO »0 to iO -^ Ci CO 00 
 
 CO 
 
 IM 
 
 1-H 
 
 
 
 13 
 
 
 
 
 
 
 
 >) 
 
 
 tt) 
 
 
 
 
 
 
 
 c 
 
 
 > 
 
 
 
 
 
 
 
 
 
 ca 
 
 0) 
 
 
 
 
 
 
 
 
 j> 
 
 
 
 
 
 --N -^ 
 
 iO 
 
 
 
 
 ^ 
 
 
 
 
 S- "^ 
 
 CO 
 
 
 
 V 
 
 3 
 
 
 
 
 
 
 &< 
 
 
 
 O 
 
 Q 
 
 OC-1 
 
 O -H 
 
 CO 
 
 d ^ C<l fh ^ ^ ■•^ CO 00 
 
 CO 
 
 
 
 e 
 
 
 
 
 ^~* 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 
 
 c 
 
 
 
 
 
 
 
 
 
 
 ji 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 *u 
 
 IM t^ 
 
 CO 'T 
 
 o 
 
 CO-^CO-^^COiC O O: 
 
 cc 
 
 
 
 
 c 
 
 
 
 
 M CO 
 
 00 
 
 
 
 
 cc 
 
 
 
 
 
 
 
 
 •2 
 
 
 "0 02 
 
 LO C^ 
 
 _^ 
 
 o o 00 r^ CO 00 ci o GO 
 
 Tt< 
 
 
 
 C 
 
 -S £ 
 
 C^ CO 
 
 C^ -J- 
 
 CO 
 
 CO CO ^ ^ ^ ^ ^ o o 
 
 t- 
 
 
 
 
 Q6 
 
 
 
 r-l 
 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 _^ 
 
 
 
 
 
 
 
 
 03 
 
 
 
 
 
 
 
 
 
 "o 
 
 -ol 
 
 OC^l 
 
 M O 
 
 -^ 
 
 COO^OOO 'i* CO 
 
 00 
 
 
 
 E- 
 
 :5S 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 *" 
 
 *-< 
 
 **" 
 
 
 
 
 
 
 oc 
 
 
 OC- 
 
 
 
 
 c 
 
 
 IM 
 
 »-i 
 
 N 
 
 
 
 
 
 
 
 
 
 
 
 
 10^05 
 
 O lO 
 
 cr. Co 
 
 ^OOOOOO oc 00 
 
 
 
 tK 
 
 
 CO O ■* 
 
 CO rt 
 
 <N •-. 
 
 t^iOCSdmClIN CO CD 
 
 
 
 
 
 f-H >-s 
 
 r-t I— 1 
 
 ■* ^ 
 
 f-H T-H 
 
 >o 
 
 
 
 
 
 
 -t< t< 
 
 
 
 
 
 
 
 
 
 
 
 £ 
 
 
 ■a 
 
 
 
 O c: 
 
 
 
 
 
 fe =« 
 
 fe°^fe 
 
 
 =« ffiKKffiffiffitffi^^ 
 
 
 
 
 
 --■ IM coo 
 
 « o— ■ 
 
 
 CO^CO^^^^'-'^COWCO 
 
 
 
 3 
 
 
 
 
 
 
 
 
 Q 
 
 
 c: CE m CK o cr. a> 
 
 < 
 
 o: o o c"j c^ c: cr. c:- Cw o c; o 
 
 < 
 
 
 
 
 
 OJ 
 
 
 CD 
 
 
 
 
 
 
 
 +J 
 
 c 
 u 
 
 
 
 0. 
 
 
 1 1 1 1 1 1 1 1 1 >^ 
 
 - c 
 
 ^ 
 
 
 c3 J2 o 
 
 -o — 
 
 o 
 
 h si a a a-a-a-T3 -a — 
 
 C^ INC<1!N!NIMC-)C-J IM — 
 CJ 04Ct(NM(N'NCl (N «f 
 
 V 
 
 
 
 
 
 M C^(N 
 
 
 
 [128) 
 
MUTATION IN MATTHIOLA 
 
 o 
 e 
 
 3 
 
 ■5 S 
 
 Eh 
 
 
 
 
 
 fcS 
 
 
 
 
 
 
 
 
 
 
 
 S ^ 
 
 CD S^ -- 
 
 
 
 
 j= a 
 
 
 — I>\ —1 -t\ 
 
 
 
 oi- 
 
 ^ O Cvi O O O O CO ^o^ 
 
 o 
 
 •~^ 
 
 
 
 
 
 
 CI 
 
 
 
 
 
 ^ ^ 
 
 
 
 
 
 
 
 '-s lO 
 
 
 
 
 
 
 IC c<3 — 
 
 t^ ^--v 
 
 IN 
 
 
 
 
 
 JJ, CO CJ 
 
 00 -*'»' 
 
 £J 
 
 
 
 
 < 
 
 
 
 
 
 
 
 
 CO CO C^ ^ ^ I^ CD CO OS i-H 00 O^ 1 
 
 
 
 4> 
 
 
 (N ^ CO 
 
 00 OS T- 
 
 co 
 
 I— 1 
 
 
 
 ^ 
 
 
 
 
 ^ 
 
 
 
 ^, w 
 
 
 
 
 C3 
 
 
 rt 00 ^ 
 
 CO O ^^ 
 
 CO 
 
 
 
 cc 
 
 J3 
 
 3 
 O 
 
 « 
 
 IN (N C^ 
 
 I^ CO CO 
 
 
 
 
 
 
 
 
 
 
 o 
 
 OOt^"— li— I5DIO Ol "C"— ICO ^1 
 
 
 
 X 
 
 -HrtC<l 
 
 CD O: .- 
 
 1— 1 
 
 
 
 
 
 
 
 
 
 
 
 o 
 
 
 p^-. 
 
 
 
 
 
 "m 
 
 
 to i— < 
 
 CO 
 
 
 
 
 c 
 
 
 
 
 
 
 
 in 
 
 Tf CO »C O O ^ ^ T^ lO lO *c 1 
 
 
 
 
 
 
 —1 ira 
 
 
 
 « 
 
 
 
 
 Oi 
 
 
 
 c 
 
 
 
 
 CC 
 
 
 
 03 
 
 
 — 
 
 
 
 
 
 E 
 
 
 ■< 
 
 
 
 
 
 
 
 
 O'J'—IOOOO "O -hoc 
 
 o 
 
 
 
 
 
 rt IN 
 
 CO 00 
 
 
 >. 
 
 
 "B 
 
 
 
 
 
 c 
 
 
 > 
 
 
 
 
 
 a; 
 
 
 0] 
 
 
 
 
 
 CC 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 O 
 
 
 
 J3 
 
 
 
 
 
 »o 
 
 
 
 t-, 
 
 CM 
 
 
 i 
 
 3 
 
 o 
 
 
 
 
 
 CO ^-' 
 
 
 
 
 c 
 
 Q 
 
 tOiN O 
 
 r-l 05 : 
 
 
 
 
 t. 
 
 
 
 
 
 
 
 o 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 — 
 
 
 
 
 
 
 
 
 
 
 
 "m 
 
 ■OIN Tfi 
 
 
 
 
 ^ 1(3 
 
 
 
 
 
 
 c 
 
 i—t 
 
 
 
 
 C) IN 
 
 
 
 
 
 
 IE 
 
 
 
 
 
 
 OJ 
 
 fc-S 
 
 i^ t^ 00 rt ^) r^ cc 
 
 00 CO ^ o- 
 
 CO 
 
 
 
 
 al 
 
 CO »-l »-o 
 
 CI CO -H 
 
 — . CO 
 
 CJ 
 
 
 
 «J 
 
 
 
 
 
 
 
 a; 
 
 
 
 
 
 u 
 
 
 
 
 
 
 
 S'O 
 
 
 
 
 
 
 i;: 
 
 S "» 
 
 
 
 
 
 
 
 
 •S.5 
 
 -H o coooo o 
 
 ■* CD OO 
 
 o 
 
 
 
 H 
 
 £s 
 
 
 '^ 
 
 
 
 
 
 00 
 
 
 
 
 
 
 
 
 
 1= 
 
 
 
 r-^- 
 
 
 
 
 
 ^ (N OOOOOOO 
 
 GO -H* 
 
 
 
 w 
 
 
 ON O -HN 
 
 CI IN OOOOO 
 
 X 
 
 
 
 
 T— 1 
 
 CI rfO Tt< o 
 
 CD 
 
 
 « 
 
 
 CO CD CD 
 
 
 
 
 3 
 
 
 1 1 1 
 
 X 
 
 
 
 ^ 
 
 
 CO »o CO -^ -^ <c »o 
 
 ■^ ^ 
 
 
 
 
 
 
 
 
 
 u 
 
 
 Oi ^ O^ Cj Gi C^ Oi 
 
 55 OS O 
 
 < 
 
 
 
 
 ,— 1 1— I ^H 1— < ^H I— 1 r-t 
 
 
 
 
 
 ^ ^ 
 
 
 
 COOC C<1 CD OC 
 
 QJ 
 
 
 
 
 00 ^ IN CO CD 05 
 
 r^ 1 ' 1 L 1 1 
 
 [ Oa »0 »0 »0 IC lO 
 
 
 CO 
 
 ^ 
 
 
 V 1 1 1 1 1 1 
 
 
 f^ 
 
 
 o -O T3 -a -0 -O T3 
 
 — o_^ ^-C 
 
 zz^ 
 
 
 
 C^J (N C^ 0-) 0-1 C^ IN 
 
 -J ■-; cic) 
 
 — C3 
 
 
 
 
 
 C^CQ ININ 
 
 IN 
 
 Ol 
 
 IN 
 
 ■< •< o 
 
 IN 
 
 <« 
 
 
 
 CC 
 
 
 be 
 
 c 
 
 rt 
 
 
 m 
 
 p. 
 
 
 rt 
 
 _, 
 
 
 [S 
 
 3 
 
 
 a 
 
 ^ 
 
 
 rrl 
 
 
 
 
 n 
 
 ert 
 
 Ch 
 
 
 
 
 
 
 '_^ 
 
 
 
 
 
 
 
 <1> 
 
 
 
 
 s 
 
 3 
 
 o 
 
 «^ 
 
 d 
 
 
 o 
 
 u 
 
 a; 
 
 o 
 
 
 
 
 CQ 
 
 H 
 
 o 
 
 [129 1 
 
272 MISCELLANEOUS STUDIES 
 
 with the position of the leaves on the plant. In the crenate-leaved 
 type this character is much accentuated, as can be seen by comparing 
 figure 14 witli figures 1 and 3 ; a warm greenhouse (fig. 14, upper line) 
 gave very marked serration, while a cool greenhouse (lower plant, and 
 also fig. 15) produced leaves much more nearly entire. 
 
 Under the much more extreme conditions of insolation, temperature, 
 and humidity at Riverside, this type was often much dwarfed in com- 
 parison with Suowflake (figs. 16 and 17; see also fig. 23). In general, 
 growtli is weaker than with Snowflake and the stems more slender. 
 Buds and flowers are often produced in great abimdance, but the 
 capsules are relatively few, small, and few-seeded. See tables 12 and 
 13 for internode data. 
 
 The progeny tests (table 29) show a slightly higher proportion of 
 nuitant-type progeny than occurred with smooth-leaved. A striking 
 new feature appears for the first time in these results, the regular 
 presence of linkage, or an association sinndating linkage, with the 
 single-double allelomorphs. Further, in all tlie four apparent mutants 
 tested the erenate factor seems to be coupled with singleness, while 
 among the sixteen F^ and F^ erenate parents there seem to be no 
 crossovers.'" We seem to be justified, for reasons just given, in 
 summing tli6 progeny as in the tables. Two things appear at once in 
 table 29: (1) there is a great excess of total doubles over the usual 
 53 per cent; (2) there is a much greater excess of doubles with Snow- 
 flake than of singles with erenate; (3) the supposed double-recessive 
 class (Snowflake double) is about two and one-half times as large as 
 the double-dominant class (erenate single). 
 
 Table 30 adds two features of special interest. First, there is good 
 evidence of selective elimination with poor germination ; compare the 
 remaining percentages with those for "Ithaca, field," "1915," "Pj," 
 and "Germination poor," and see tables 39 and 40; the only excep- 
 tional ease is the low percentage for the thirty plants of 1915-1(1. It 
 would be surprising if the slow and weak growth of the erenate plants 
 did not lead to such a result. Second, there is evidence that the 
 erenate individuals are smaller than Snowflake even before germina- 
 tion. The seeds of erenate i)arents are le.ss uniform in size than those 
 of Snowflake parents; small seeds are numerous, and even the larger 
 ones probably weigh decidedly less than normal Snowflake seeds. 
 "With five erenate parents included in the cultures of 1913, random 
 
 I' With four of the parents the tests are obviously entirely inadequate; one 
 other, 22d-9, gives no indication of linkage among nineteen progeny. 
 
 {i:w] 
 
MUTATION IN MATTE lOLA 
 
 w CO 
 
 v; o 
 
 ^^^^^^^^^fe:^^^^ 
 
 
 
 
 1 
 
 g 3 ;?- - - ;^ ?s ;? ;? ?3 ?3 ;? ?3 
 
 
 
 
 
 
 
 
 
 ■t:? 
 
 -biT^r^Q rD OJ <-^irr-TTr*ir-t-r*(-*<-^ 
 
 
 zjOoc^iS^oococ&d^oaoc^ 
 
 
 p 
 
 ■^ ?» P g^^^_^^,,„_ 
 
 
 ^•B 
 
 ^ s-s-s-g^^scsssjs 
 
 
 B 
 
 
 w 
 
 ^ '^O't^ 
 
 
 
 
 
 
 
 
 (O 
 
 
 
 
 
 
 i»OOS*ii*S»i»td'-'*-"-"-|i-<'-' 
 33 |r*'"?^gg 
 
 
 
 
 
 
 
 a, gn 
 a, fie 
 a, al! 
 
 16 
 side, 
 
 inati 
 inati 
 
 
 
 
 §§ s, Eg 
 
 
 i 
 
 
 •am ~ p- 
 
 
 
 
 o o O 
 
 
 
 
 o o c 
 
 
 
 
 •-s a. w 
 
 
 
 
 » 
 
 
 
 
 ,_! 
 
 
 
 Cn t^ K) 
 
 
 
 
 O 1s3f- 
 
 
 
 
 h- '£1^ 
 
 
 
 
 - M ^2 Oi- - OS >- O. Oi W ^ 
 
 
 
 
 Cito^^todtoto^^^^oto^^ 
 
 
 cc 
 
 
 GoSoCSOOOooX^OliOWOiJM 
 
 
 
 
 - -) - - ^^ ta ^ 
 
 
 09 
 
 
 (■a t«. ^a 5- -* 
 
 
 
 
 ►^ >-, ^ Oo ►-. 
 
 
 
 
 ►- OoOo 
 
 
 
 
 
 G 
 
 
 
 
 s 
 
 
 
 
 
 a. 
 
 
 
 
 t— I 1— ' I-* 1— ' 
 
 1" 
 
 
 
 ? 
 
 0>-Oirf^CO*-05tOOl— 1— ^-Orf* 
 
 1 
 3 
 
 5 
 
 -3 
 
 o 
 
 
 o 
 n 
 
 3 
 
 
 s. 
 
 
 
 ^ 
 
 
 
 
 
 p 
 
 
 
 
 d 
 
 3 
 
 
 
 
 o 
 
 5' 
 
 
 
 Oi h- W H- 05 4^ Goto ►-* 
 
 
 2 
 
 
 
 lO304itO^MWH-C»:iOCSi— Cnen 
 
 
 a. 
 
 
 
 COh4iCD004^H-WCSOC;'H-MOOO 
 
 3. 
 
 2 
 
 
 2 
 
 
 
 
 
 
 
 
 
 p" 
 
 
 
 
 
 a 
 
 
 
 
 
 
 
 
 
 
 :a 
 
 
 
 
 
 
 
 ►—^CObOtOOOtvS H-Ol en 
 
 2; 
 
 
 
 
 00>^OiW05*^C04i.COC;i-'^4^ 
 
 c 
 
 
 
 
 ^ ,, ^ , ^ , ,, , ,,-^ , ,^ ^ ^_^ 
 
 3 
 
 
 
 
 |_ H- H- h- h- *.^0 Ol 
 
 tr 
 
 
 
 
 ►-■^ to CCN5 --'^OJ C-. 
 
 fB 
 
 
 
 
 ■^00 GO OOS CC^'-^ 
 
 -1 
 
 
 
 
 
 
 
 
 
 
 
 D 
 
 
 
 
 
 P 
 
 
 
 
 
 S" 
 
 
 
 i-WbOCOl—tOOOi— •— wto^-co 
 
 
 
 
 
 OWtsi^I4^:D«20Wait00^01 
 
 ? 
 
 
 
 
 ►- J- CO to 'X o CJ o; o ^ o o *^ 
 
 H-K-H-H-lflflfH-H-lflfH-lf 
 
 n 
 
 3 
 
 
 
 
 coi—iot-toi— H-a.OH-to*'to 
 
 
 
 
 
 4^C0^02^tOCnC5tOOiH-OCn 
 
 a 
 
 
 [1311 
 
274 
 
 MISCELLANEOUS STUDIES 
 
 samples of seed were sorted, and tlie smaller and larger seeds planted 
 separately. 
 
 Table 31 gives the data from this test. Here is practicallj' con- 
 clusive evidence (.see tables 39 and 40) that the smaller seeds much 
 more often contain embryos of the crenate type.^* Since the embryo 
 of a Matthiola seed occupies practically all the space within the seed 
 coats, it is evident that even as embryos Snowflake plants exceed 
 
 Table 31 
 
 Cultures of 1913. Crenaie-Jeaved type: proportions from smaller and larger 
 
 seeds of crenate parents. 
 
 
 Seeds 
 
 Progeny 
 
 Parent 
 
 Total 
 deter- 
 mined 
 
 r- * 1 A 
 
 
 
 
 
 
 
 
 
 Snowflake 
 
 Other 
 
 
 Size 
 
 Number 
 
 Number 
 
 Per cent 
 
 types 
 
 22a-l 
 
 Smaller 
 
 21 
 
 6 
 
 4(5) 
 
 83.3 ± 
 
 12.9 
 
 
 
 1 
 
 22a-l 
 
 Larger 
 
 29 
 
 23 
 
 3 
 
 13.0 ± 
 
 6 6 
 
 19 (20) 
 
 
 
 22a-,') 
 
 Smaller 
 
 17 
 
 11 
 
 4 
 
 36.4 =t 
 
 9 5 
 
 6 
 
 1 
 
 22a-5 
 
 Larger 
 
 .3.3 
 
 28 
 
 2 
 
 7,1 ± 
 
 6.0 
 
 25 
 
 1 
 
 22b 
 
 Smaller 
 
 13 
 
 8 
 
 5 
 
 62.5 ± 
 
 11.2 
 
 3 
 
 
 
 22b 
 
 Larger 
 
 36 
 
 31 
 
 4 
 
 12.9 ± 
 
 5.7 
 
 25 (27) 
 
 
 
 22d-12 
 
 Smaller 
 
 30 
 
 24 
 
 17 
 
 70.8 ± 
 
 6.5 
 
 5 
 
 2 
 
 22d-12 
 
 Larger 
 
 70 
 
 57 
 
 11(12) 
 
 21 1 ± 
 
 4 2 
 
 42 (44) 
 
 1 
 
 22d-15 
 
 Smaller 
 
 32 
 
 24 
 
 17 
 
 70.8 ± 
 
 6.5 
 
 3 
 
 2(4) 
 
 22d-15 
 
 Larger 
 
 68 
 
 54 
 
 18 
 
 33.3 ± 
 
 4.3 
 
 34 (35) 
 
 1 
 
 AH 
 
 Smaller 
 
 113 
 
 73" 
 
 47 (48) 
 
 65.8 ± 
 
 3 7 
 
 17 
 
 6(8) 
 
 AU 
 
 Larger 
 
 236 
 
 193« 
 
 38 (39) 
 
 20.2 ± 
 
 2 3 
 
 145(151) 
 
 3 
 
 All 
 
 All 
 
 349 
 
 266 
 
 85 (87) 
 
 32.7 ± 
 
 19 
 
 162(168) 
 
 9(11) 
 
 ° Respectively 64.6 and 81.8 per cent of the numbers of seeds planted. 
 
 crenate plants in size. This fact, obviousl.y, is further evidence in 
 favor of the hypothesis of partial selective elimination of crenate 
 heterozygotes during embryonic development. 
 
 It may be worth noting that the 73 plants from the smaller seeds 
 include 6 (8) apparent mutants of other types (mutation coefficient 
 11.0 per cent), while the 193 plants from the larger seeds include 
 only 3 apparent nuitants (1.6 per cent). 
 
 Before we can profitably discuss these data further, we must con- 
 sider the results from cross pollination (tables 32 and 33). The 
 numbers, though small, make it very probable that both eggs and sperms 
 carry the crenate factor. Further, it appears from series 20 that only 
 a small portion of the sperms carry this factor, as we should expect 
 from its apparent linkage with singleness. If homozygotes are non- 
 viable, the combined crenate percentages of reciprocal crosses should 
 
 IS The ])Oorer germination of the smaller seeds suggests that the disparity 
 between the two lots of seeds in the proportion of crenate embryos was even 
 greater than the cultures indicate. 
 
 [132 
 
MUTATION IN UATTUIOLA 
 
 275 
 
 9: 
 
 » £ 
 
 
 ►3 
 
 
 
 •a 
 
 3-3 X 
 
 > 
 
 to to to 
 
 r> t-H to to to 10 to 
 
 
 to Ti 
 
 toio h3 
 
 
 
 
 1 
 
 "<: s= 3 
 
 
 — 3 00000 
 
 — S 3-CRm OjO-f; 
 
 - • 
 
 
 
 era = " 
 
 — CCTP to 
 
 
 
 ? 
 
 
 o^a.cr» 
 
 
 W M P 
 —.OTT 
 
 
 
 
 
 CO 
 
 a 
 
 
 
 •-OCO C 
 
 a2 :^ 
 
 ^^^^^^^ 
 
 
 
 
 
 CD CO CO CO CD 
 
 to CO 
 
 
 
 
 
 
 
 
 
 
 
 
 if^U 
 
 Ol 1*^C0 
 
 C;i Cn CT Cn Cn cn 4^ 
 1 1 1 1 1 1 
 
 CO CO CO CO 
 
 
 r^ 
 
 
 8= 
 
 03 
 
 ►-» h-* H- h-' 1— ' 1— I 
 05 05 05 35 O^ 
 
 
 
 3 
 
 
 to 
 
 4^ 
 
 
 
 
 02 
 
 
 ~a M 
 
 0: CO 
 
 to to H- H- C.T (-• J5 
 
 K) M 4^ 
 
 
 fO 
 
 
 H- c;i 
 
 1*- to 03 
 
 0001 MOOOCO 
 
 CO ^ CO 
 
 
 CL 
 
 
 o. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 3? 
 
 H 
 
 
 CO 1-' 
 
 ^— Cn 
 
 0000000 
 
 en Oi— rf*. 
 
 II 
 
 
 
 3 
 
 
 
 
 
 
 
 
 ^ 
 
 to 
 
 ,_, 
 
 
 %.^ 
 
 
 
 OJ to 
 
 
 K- (-• i-i >;*. h- H- 
 
 05 »0 
 
 
 5' 
 
 
 
 *. 01 
 
 ■35kJ uOO 
 
 CO CO >^ to CT rf^ »^ 
 
 CO k^ rfi-Cn 
 
 I7 
 
 s. 
 
 
 
 to 
 
 
 
 to 00 
 
 CO H- ►-» ^ 
 
 
 Oi 
 
 4*. h- 
 
 M 
 
 
 
 
 
 
 
 
 
 
 
 
 5' 
 
 
 
 
 
 
 
 
 
 
 
 
 'E- 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 (T 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 : OCO 
 
 COOOh-O to 
 
 
 
 
 
 3 
 
 
 ? 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 c 
 
 2; 
 
 
 
 
 !R 
 
 
 
 4^ 
 
 
 
 
 
 " 
 
 
 fT> 
 
 
 
 
 
 
 
 
 
 
 3 
 
 
 
 
 
 Q 
 
 to 
 < 
 
 
 "< 
 
 
 
 
 
 
 
 
 to 
 
 00 to^ 
 
 0>-" 1— H- too 
 
 CJ» Orf^t— 
 
 
 
 
 
 
 
 
 
 > 
 
 
 !2 
 
 
 H- CO 
 
 CO 1-^ 
 
 
 Oi en 
 
 
 
 3 
 
 
 
 ""' to 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 M 
 
 
 01 
 
 •M 
 
 .ik >— • 
 
 CO tc 
 
 
 
 
 
 10 CO 
 
 H-Ot0 4- 
 
 ^- -4 Cn C71 ^ VI CO 
 
 I-* tc -^ 
 
 w 
 
 
 
 
 
 
 
 
 5' 
 
 
 
 
 ?■ 
 
 -1 
 
 ifa 1— *» 
 
 CO 00 
 
 «_ 
 
 
 
 
 CO 
 
 M 
 
 to 00 — ' 
 
 
 
 
 
 
 
 
 ^ ^ 
 
 ^ 
 
 
 
 
 
 
 ^ 
 
 
 
 
 
 
 
 05 H— • 
 
 1—0 
 
 rf-l-' 10 
 
 >^ CO h-1 
 
 
 
 
 
 
 
 ^ 
 
 Oi ►-* to CO 
 
 00 ~1 en to rf^ C5 
 
 OCO tCrf^ 
 
 
 
 
 
 
 
 
 
 c 
 2: 
 
 ^ 
 
 
 
 a: w 
 
 
 Cn 00 to »-. 
 
 en CO 
 
 rs 
 
 
 
 4J. 
 
 
 ■~-' CO 
 
 )--• »{^ 
 
 P 
 
 
 
 
 '-^ 
 
 
 
 T 
 
 
 
 
 ^ 
 
 1— ' 
 
 
 
 
 1— to 
 
 l-l-J 
 
 CC h- h- h- CO ►- 
 
 M en tc 
 
 
 
 
 
 CO i-" 
 
 H- rf*. ^ 
 
 CO -J to CO h- 
 
 CO *f* ^*i. K- 
 
 
 
 
 
 
 
 
 
 > 
 
 
 
 
 K ''^ 
 
 H- ^- 
 
 CD H- ►-rf^ w 
 
 00 cn>o 
 
 
 
 
 
 to to 
 
 w CD 
 
 4^ CO H- CO ^ 
 
 ^^ as to 
 
 
 
 
 
 tJ ~' 
 
 
 
 
 
 
 
 
 
 ^^ 
 
 
 
 
 
 
 
 CO 
 
 00005 
 
 coooo>- w 
 
 4:*OCOh- 
 
 
 ) 
 
 
 
 
 
 
 00 
 
 CO to 
 
 
 en 
 
 Ni 
 
 2'!; 
 
 
 
 
 
 
 1^ 
 > 
 
 CO 
 
 t-a 
 
 [133] 
 
276 
 
 MISCELLANEOUS STUDIES 
 
 exceed the percentage from selfed parents ; the expected high pro- 
 portion with series 21, however, might well be realized with adequate 
 numbers and good germination. 
 
 In spite of the small totals, it is very probable that linkage similar 
 to that of the selfed cultures prevails with series 21. "Where the 
 crenate type is the pollen parent (series 20) linkage ratios are on our 
 hypothesis impossible, since the eggs are all Snowflake and the sperms 
 all double ; the data, however, though statisticallj'' inconclusive, sug- 
 gest that the excess of singles with crenate and of doubles with Snow- 
 flake is greatly reduced but not abolished. 
 
 Table 33 
 Hybridisation of the Snoii-flal'e and crenate-leaved types. Summary. 
 
 
 Progeny 
 
 
 Cultures 
 
 Seeds 
 
 Plants 
 
 Parents 
 
 Total examined 
 
 Crenate 
 
 
 Undeter- 
 mined 
 
 Deter- 
 mined 
 
 Number 
 
 Per cent 
 
 20aa, bb, & cb 
 20dc, ed, & ic 
 20de,ff,gf,gg,&hd 
 All of series 20 
 21aa, bb, & dd 
 Snowflake par- 
 ents of hybrids 
 (5) 
 
 1913 
 1914 
 1915-16 
 
 All 
 All 
 
 All 
 
 123 
 163 
 120 
 406 
 75 
 
 271,. 50 
 
 5 
 
 
 5 
 1 
 
 3 
 
 93 
 14 
 
 103 
 
 210 
 
 25 
 
 134 
 
 5(6) 
 
 6 
 
 11(12) 
 2(3) 
 
 (1) 
 
 6.5 =>= 1.6 
 
 
 5.8 ± 1.5 
 
 5.7 ± 1.1 
 
 12.0 =*= 4.4 
 
 .7 ± .5 
 
 If we may ignore the doubtful correlation just mentioned a fairly 
 adequate complete hypothesis for the selfing ratio is possible. Assume 
 (1) a gametic ratio'" of 5DC -.IdC -.'iDc-.bdc, or 16% per cent of 
 crossing over ; (2) non-viability of homozygous crenate (CC) ; (3) low 
 viability of simplex crenate (C'c), eliminating an average of 60 per 
 cent of this type; and (4) coupling of D and C in all parents tested. 
 Evidence has already been prasented for assumptions (1), (2), and 
 (3), except as to the intensity of linkage, while (4), as will be seen, 
 is not at all improbable. 
 
 Random fertilization under these conditions, excepting (3), would 
 give 26DdCc (crenate single) -\- lOddCc (crenate double) + 5Ddcc 
 (Snowflake single) -\- 25ddcc (Snowflake double). The other two 
 clas.ses, 5DdCC and IddCC, would be non-viable pure crenate. Adding 
 assumption (3) gives the following comparison: 
 
 19 Representing the singleness ami doubleness factors by D and d, and the 
 crenate factor and its "normal" allelomorph by C and c. 
 
 [1341 
 
UVXATION IN MATTHWLA 277 
 
 DdCc ddCc Ddcc ddcc 
 
 , Theoretical ratio (« = 44.4) 10.4 i 5 25 
 
 Calculated for n — 540 126 49 61 304 
 
 Observed ()i = 540)» 125 51 57 307 
 
 This fit surely caunot be criticised, whatever may be thought of 
 the devices employed to obtain it ! With cross pollination the agree- 
 ment is fairly good in the case of series 20, which gives the only fairly 
 reliable data. We are assuming 16% per cent of crossover dC sperms: 
 elimination of .60 of 16% per cent, or 10 per cent of the total, gives 
 .06%/.90 = 7.4 per cent expected crenate, as against 5.9 per cent 
 observed. Series 21 is supposed to have 50 per cent of C eggs in 
 the ratio 7>DC AdC ; elimination of .60 of this proportion, or 30 per 
 cent of the total, would leave .20/.70 = 28.6 per cent, against 12.0 
 per cent in the very inadequate material observed. An adequate test 
 of the hypothesis obviously requires large hybrid cultures, from 
 vigorous seed sown under favorable conditions for germination. 
 
 A scarcity of crossover crenate singles follows from the hypothesis ; 
 they constitute only one twenty-sixth of the total number of viable 
 crenate single progeny of crenate parents. No direct evidence indi- 
 cating that the crenate and double factors are ever coupled in singles 
 has yet been discovered. 
 
 If the supposed crenate mutants are due to immediate factor muta- 
 tion, however, it seems strange that the same locus is changed more 
 readily in a singleness chromosome than in one carrying the doubleness 
 factor, in a ratio similar to the linkage ratio of later generations. 
 If the apparent mutants are really segregates from a balanced-lethal 
 combination, the observed original coupling of crenate with single 
 might be an accident of sampling involved in the original choice of 
 material ; other initial parents might give the reverse coupling. 
 
 5. THE SLENDER TYPE 
 
 This type is comparativel.y rare as an apparent mutant from Snow- 
 flake or early; the 3135 plants reported in table 28 gave only 4 (6) 
 mutants (2 singles and 4 doubles, 2 of the latter perhaps Snowflake), 
 a mutation coefficient not over .19 per cent. This type seems to occur 
 more frequently among progeny of crenate. a tj'pe similar in some 
 
 -" Omitting 29 plants classed as neither crenate nor Snowflake, which as 
 probably non-crenate should perhaps be added to Snowflake, and also 64 plants 
 (13 crenate and 51 Snowflake) with flower data incomplete. Complete data for 
 the total of 633 plants would plainly give a somewhat poorer fit, but this could 
 be improved by assuming a slightly greater elimination of Ccii zygotes. 
 
 [13.S1 
 
278 
 
 MISCELLANEOUS STUDIES 
 
 S 
 
 n 
 ■< 
 Eh 
 
 
 
 CJ S 
 
 
 
 s s 
 
 s s 
 
 S 
 
 CO 
 
 
 
 :£ a 
 
 
 
 
 
 
 
 
 
 O^ 
 
 o 
 
 o 
 
 O ^ -HO 
 
 (N OOC^ 
 
 «— 1 
 
 *"* 
 
 
 
 
 
 
 
 
 ^ 
 
 
 ^ 
 
 
 
 
 
 
 
 
 ^v Ol 
 
 
 t^ 
 
 
 
 
 
 
 CO 
 
 CO -H 
 
 t^ >o 
 
 
 Tj< 
 
 
 
 
 
 
 o 
 
 S S- 
 
 s ^ 
 
 
 
 
 
 
 -< 
 
 
 
 
 
 
 
 
 
 
 
 o 
 
 05 
 
 O "O t^ 00^ 
 
 ^ Tj< IC 05 
 
 C^ 
 
 t- 
 
 
 
 
 
 
 lO 
 
 >0 CO"-! 
 
 to M ^ 
 
 t— I 
 
 CO 
 
 
 
 4) 
 
 
 
 
 
 
 
 f-H 
 
 
 
 
 
 ^ 
 
 
 ^ 
 
 
 
 
 aj 
 
 
 
 
 y~^ CO 
 
 
 
 
 
 "« 
 
 
 C^l 
 
 C<» !^ 
 
 ■CO to 
 
 
 "Tf* 
 
 
 
 C 
 
 S 
 
 
 to 
 
 to CO 
 
 to ^ 
 
 
 1-H 
 
 
 
 1 
 
 1 
 
 
 
 
 00 
 
 00 M t^ CO w 
 
 00-^ -rt* Tj< 
 
 CI 
 
 CI 
 
 
 
 C 
 
 
 »c 
 
 lO CO ^ 
 
 iO C<l '* 
 
 
 CO 
 
 
 
 
 
 
 
 l-H 
 
 
 1-H 
 
 
 
 
 Ji 
 
 
 
 
 
 
 
 
 
 
 ■a 
 
 
 
 CO 
 
 s s 
 
 
 S" 
 
 
 
 
 B 
 
 
 
 
 
 
 
 
 
 
 s 
 
 
 f— < 
 
 -H c^o o o 
 
 (N Orf ^ 
 
 o 
 
 T)l 
 
 
 s 
 
 
 
 
 
 
 
 
 
 
 c 
 
 
 
 
 - o 
 
 
 CO O 05 
 
 Tf 
 
 s 
 
 
 _rt 
 
 
 
 
 G 
 
 £J G 
 
 S- cit 
 
 c 
 
 s 
 
 
 S 
 
 
 ■< 
 
 
 
 
 
 
 
 
 
 
 
 r~ 
 
 t^ r}^ t-M to O 
 
 -H COOO o 
 
 o> 
 
 o 
 
 
 
 
 
 
 " 
 
 -. -H 
 
 CO .-HCO 
 
 .— 1 
 
 lO 
 
 >. 
 
 
 Oi 
 
 J> 
 
 
 
 ,-^ lO ■ 
 
 o CI 
 
 S" 
 
 £' 
 
 Q> 
 
 
 -o 
 
 XJ 
 
 
 s 
 
 £ ^ ^ 
 
 IN CO 
 
 
 
 ^ 
 
 
 c 
 
 3 
 
 
 
 
 
 
 
 £ 
 
 
 _« 
 
 o 
 
 00 
 
 00 Tt<-< -H : 
 
 tO-H(Nt^ 
 
 CI 
 
 tC 
 
 CU 
 
 
 CQ 
 
 
 
 
 -H M 
 
 
 " 
 
 
 
 
 
 
 _ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 -2 
 
 
 
 IN 
 
 CO ^ 
 
 
 oc 
 
 
 
 
 bC 
 
 
 
 
 t-H -h 
 
 
 CO 
 
 
 
 
 c 
 
 
 
 
 
 
 
 
 
 
 tn 
 
 
 Ol 
 
 05 O O -* : 
 
 'I'lN lOO 
 
 CO 
 
 CO 
 
 
 
 "S 
 
 V s; 
 
 
 CO 
 
 Tt* to O w ^ 
 
 00 r^ -^ CO 
 
 00 
 
 lO 
 
 
 
 c 
 
 1 = 
 
 
 00 
 
 00 »0 W CO 
 
 O -rt^ rt< 
 
 CO 
 
 o 
 
 
 
 "a 
 
 X 
 
 Q'S 
 
 
 
 
 ^ IN 
 
 
 CI 
 
 S-o 
 
 
 
 
 
 
 
 
 
 "d 
 
 ■£ ** 
 
 
 
 
 
 
 
 
 
 ■h> 
 
 -S.S 
 
 t-H 
 
 o 
 
 -H OO ^o 
 
 -HOOIN 
 
 1 — 1 
 
 1 — 1 
 
 
 
 H 
 
 5^ 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 o 
 
 <^> 
 
 ^? 
 
 o 
 
 
 
 
 
 <3^ 
 
 ^* 
 
 00 
 
 ^^ 
 
 ^!f- 
 
 
 1 
 
 
 
 
 
 
 
 
 
 
 o; 
 
 co' 
 
 (N ^ O O O: 
 
 CO^COO 
 
 ca 
 
 oc 
 
 
 cc 
 
 
 
 r^ 
 
 05 aJC-> t^ CO 
 
 CI -iMO-H 
 
 
 03 
 
 
 
 
 
 " 
 
 
 Ol lO 
 
 " 
 
 CO 
 
 
 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 Tt^ 
 
 
 
 
 
 s 
 
 
 
 o 
 
 .o ^ 
 
 CO 
 
 
 
 
 3 
 
 
 
 .=*J 
 
 ffi^.ffiW 
 
 1 
 
 
 
 
 J:; 
 
 
 CO 
 
 2 fa 
 
 — ' CCi-i -- CO Tt- 
 
 CO lO 
 
 
 
 
 3 
 
 
 
 
 
 
 
 
 o 
 
 
 o- 
 
 en — 1 
 
 "t; OS Ci oi CI C5 C5 
 
 <I^ ,-H ,-H .-H T-t 1— < .— 1 
 
 •-; 02 en— ; 
 
 ^ 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 2 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 _ 
 
 * 
 
 " 
 
 
 
 
 k4 
 
 CO fc- 
 (- . 1 OJ 
 
 ^ t- 
 
 Is 
 
 
 
 
 
 ^ w OS 
 
 - CcOcO OJ 
 
 
 ca 
 
 
 
 pi 
 
 &H JJ 1 1 "3 
 
 -s oj 
 
 T3'0 
 
 Ph 
 
 
 C3 
 
 XI _ 
 
 '3 jb ^j= ^ 
 
 
 
 
 
 
 
 
 
 >o — 
 
 
 5 "" -0^555 
 
 < (NIN < < 
 
 
 
 CJ 
 
 rt 
 
 
 <D 
 
 a. 
 
 p. 
 
 X 
 
 .n 
 
 rt 
 
 
 C« 
 
 
 
 ■I- 
 
 ~* 
 
 ■ 
 
 o 
 
 
 § 
 
 tn 
 
 J3 
 
 
 
 -♦-» 
 
 1> 
 
 o 
 
 %4 
 
 ^ 
 
 a 
 
 
 Fh 
 
 « 
 
 <D 
 
 s 
 
 M O Q 
 
 [1361 
 
MUTATION IN MATTHIOLA 279 
 
 respects, and vice versa. Under favorable conditions this type may 
 closely resemble Snowflake, but is decidedly more slender in stems, 
 leaves, and pedicels. A characteristic drooping of flowers and branches 
 is well shown by two plants in figure 18 ; the single is 25b of the 
 tables. The progeny of 25b shown in figure 19 illustrate a variability 
 of the "slender" characteristics which has suggested the presence of 
 genetic differences among plants classed as slender. The leaves often 
 resemble those of crenate more closely than do Snowflake leaves. 
 
 In the fleld at Ithaca flowering was markedly earlier than with 
 Snowflake, and the type seems to be earlier on the whole. The River- 
 side conditions have commonly given a decided dwarfing as compared 
 with Snowflake, though not to the extreme degree that this has occurred 
 with crenate (figs. 20 and 21). 
 
 The results of selfing tests are reported in tables 34 and 35. The 
 distributions have the same general characteristics as with crenate, 
 with some remarkable differences. The excess of doubles with Snow- 
 flake is very much greater, the ratio being about 30 : 1 ; with slender, 
 however, the excess of singles is slight in the grand total and perhaps 
 significantly variable with different parents. 
 
 Plant 25b-ll, the "extreme" individual of figure 19, appears to 
 give a real excess of slender over Snowflake, and of double slender 
 over single slender, though the numbers are much too small for cer- 
 tainty. The two parents classed as "extreme" are (tables 39 and 40)^^ 
 quite probably genetically different from the other slender parents. 
 It should be noted that plant 25b-6-8-6, progeny of one of the parents 
 described as "extreme," has also given a relatively high proportion of 
 slender progeny. Perhaps the "extreme" form is heterozygous for a 
 second slenderness factor similar to the original one. 
 
 The percentages of mutant-type progeny are (table 39) much more 
 variable than with smooth, large, or crenate, and (table 40) there is 
 no good evidence of selective elimination ; both these facts may depend 
 on genetic differences among the parents tested. 
 
 The great niodifiability of the various types, including Snowflake, 
 indicated by a comparison of, for instance, flgures 14, 15, and 16, 
 greatly complicates the positive determination of types. In the cul- 
 tures of 1911H and 1913, where crowing in flats or aphis injury in 
 the field interfered with normal development of some plants, the im- 
 pression was obtained that the slender type occurred in several grades 
 
 21 In the calculation of the probability of simple sampling, f is taken as 3 
 (the number of cultures), not 2 (the number of parents). 
 
 [137] 
 
280 
 
 MISCELLAKEOrS STVDIICS 
 
 ^ 
 
 
 
 
 
 r-Tj<o5r~ oocBO^o Ttocto-HC^ 
 
 
 
 
 c 
 
 M^c^m -*(NoiMm ^Mcoiracsi 
 
 
 
 
 <i> 
 
 ■fl-H-H-H •«■«•«•«•« -H-fl-H-H-H 
 
 
 
 
 £ 
 
 
 
 
 L. 
 
 
 N i-t C<1 ■<** Tji Tj^ re C^ CO Tt^COCOCDCv) 
 
 
 
 
 
 
 
 
 H 
 
 
 
 
 
 a3 
 
 
 
 
 
 X- 
 
 s 
 
 So £"5<' o5coo?^£" 5?S5f5<"S 
 
 •^ooc^o oor^ot^^ ^ f* oj oi o 
 
 
 c 
 •a 
 
 
 -H 'MC^C-<Tt<CD— ICO CM^C^I— no 
 
 
 
 
 
 
 
 
 -d 
 
 
 
 
 
 (y 
 
 
 
 
 
 c 
 
 
 
 
 
 
 !i a 
 
 
 
 
 e 
 
 ^T-^CJCO^-^^CO-rt^GO ^lOOOOOlO 
 
 
 
 -u 
 
 
 t^ lO C^ t^ -rf C^ -^ 00 O IC O 1^ CO O 
 
 
 
 a) 
 
 
 — ■ ^ e^ —1 —1 c~5 
 
 
 
 C 
 
 
 
 
 
 1 
 
 Q 
 
 
 
 
 a 
 
 
 
 
 
 X 
 
 
 
 
 
 0) 
 
 
 
 a 
 
 
 
 •o 
 
 
 a> 
 
 
 SJ 
 
 s 
 
 
 EC 
 
 
 ■^ 
 
 
 O 
 
 
 o 
 
 1 
 
 »-* c »-• ^ o o ^ c^ ^ ^ o^^^^ 
 
 
 
 
 1^) ^©^O <3^^^^0 
 
 
 •:: 
 
 
 00 cooo~^ -*^co^-<f 
 
 
 s 
 
 
 >o r^ (N c^Tco CO CO o^c^Tco tSc^r^cSoc 
 
 
 CO 
 
 
 W CO ^ 00 CO lO O r-H o C^ OCl^^O 
 —>iyi'-l C^iC«C^ -HCO-tCO 
 
 
 
 
 m T3 t. 
 
 
 
 
 3 O O 
 
 
 
 
 o o o 
 
 
 
 
 J3 ^ bCC. 
 
 
 
 
 is "S CO 
 
 
 O 
 
 
 Ithaca, gn 
 
 Ithaca, fie 
 
 Ithaca, all 
 
 1913 
 
 1914 
 
 1915-16 
 
 Riverside, 
 
 All 
 
 All 
 
 All 
 
 All 
 
 Germinati 
 
 Germinati 
 
 All 
 
 All 
 
 
 
 
 
 
 fr! <>' ^ 
 
 
 
 c* ■ — — ' 
 
 
 
 T3 
 
 
 
 ??5^^ -a -a 
 
 
 
 -^"-^ M a a 
 
 a 
 
 
 Ti< ,-1 ^ lO IN ^ r~ oo a> <D fa -3'S 
 
 <u 
 
 
 -_-v-._-_.~_---^^~_-^ T3 HH 
 
 
 
 cacaacc:c'"'"'°r^cc£g 
 a; Qj 0) Oj <y oj o_aj - - ^'^^^ (i.-- 
 
 Ph 
 
 
 
 
 
 
 
 
 
 'm tfl '^ 'm '^ '^ '^ "^ Ph Pi< &H ^'^"x'S 
 
 
 
 
 
 <^<<<^<<<<^ ^<i^? 
 
 [138 
 
MVTATIOS I.\ M.irrillOLA 28] 
 
 probably unlike genetically. In the 1916 cultures, on the other hand, 
 with better development, this type seemed substantially as uniform 
 as the others. 
 
 If we ignore these possible genetic differences and attempt to 
 apply the scheme worked out for crenate, difficulties appear at once. 
 First, the scarcity of Snowfiake singles would indicate much closer 
 linkage than with crenate, while the relative abundance of slender 
 doubles apparentl.y contradicts this supposition. Second, the in- 
 adequate results from crossing with Snowflake (table 36) suggest 
 that the sperms carry the supposedly crossover slender factor at least 
 as often a.s do the eggs. While crenate as pollen parent gives results 
 agreeing tolerably with the hypothesis, slender gives results differing 
 from these in the wrong direction. 
 
 No doubt, however, the disagreements can be over emphasized. Both 
 crenate and slender as seed parent seem to give the expected relations 
 between singles and doubles, and series 23 also does this with the 
 Snowflake progeny. Oljviously tlie functional sperms and eggs of these 
 mutant-type parents exhibit different ratios between types, and the 
 peculiar results in other respects with slender may be related to the 
 added complication suggested above. The astonishing feature of the 
 data, of course, is the great excess of single slender over double slender 
 in series 23 — an excess which suggests an actual significant excess of 
 singles in the totals of all types given by this cross — while with selfed 
 slender there is a great total deficiency of singles. We may at least 
 feel confident that the modifications of the single-double ratio, with 
 this type and with crenate, are due to lethal action which also affects 
 the proportions of viable slender and crenate gametes or zygotes. 
 
 If differential viability before germination is an important factor 
 with these types, very probably it differs according as Snowflake or 
 the mutant type is the seed parent, and according to the parental 
 environment. In other words, partial selective elimination during 
 seed formation maj' vary with the environment of the embryos, accord- 
 ing as this environment is affected by either the genetic constitution 
 or the external environment of the seed parent. Until such uncer- 
 tainties are eliminated, we are hardly justified in ruling out, for 
 the types discussed, the probability that regular segregation and (in 
 the last two cases) true linkage are concerned in these phenomena. In 
 fact, the definite differences in ratios between reciprocal crosses and 
 between at least one of the crosses and selfing encourage further 
 attempts at satisfactory factorial analysis. 
 
 [1391 
 
282 
 
 MISCELLANEOUS STUDIES 
 
 a 
 
 
 o 
 =2 
 
 tq 
 
 
 
 
 ^ 
 
 1 
 
 
 
 
 
 
 5§ 
 
 
 
 
 5* 
 
 0(N IN Tf rt —1 
 
 
 
 = „ ^ 
 
 
 
 
 
 ^ ^0 ^ 
 
 
 
 
 
 io ioCi ca 
 
 
 
 
 < 
 
 'M CO in «CC 
 
 
 
 a 
 
 
 i-H CO 00 ^ IM 
 
 
 
 
 
 ^ 
 
 _o 
 
 
 
 
 <a 
 
 ^^ ICO CO 
 
 
 
 EC 
 
 3 
 
 »f3 CC C^ 
 
 
 
 S 
 
 3 
 
 
 
 
 o 
 
 O 
 
 Q 
 
 ^ CO -* CO Oi 
 
 
 
 CO 
 
 ^ 00 ^ T-H ^ 
 
 
 
 
 <y 
 
 
 
 
 
 "m 
 
 coc 
 
 
 
 
 c 
 
 G3t 
 
 
 
 
 s 
 
 
 
 
 
 ^ CO c^ 0: ^ 
 
 
 
 
 
 coco 
 
 
 
 oj 
 
 CC 0; lOCO -^ 
 
 
 
 
 
 
 
 
 
 
 
 O 
 
 10 !M (N Tf CO 
 
 
 
 
 
 O 
 
 •H -H -H -H -H 
 
 
 5 
 
 
 
 
 
 
 c 
 
 
 
 a 
 
 CD CO "O -rf 
 
 
 31 
 
 
 
 s 
 
 
 
 K 
 
 
 -*; 
 
 
 -i (nI -i t~ 
 
 
 
 
 
 <NC^1C1 
 
 
 
 a 
 c 
 
 
 l-r 
 
 a 
 
 
 1-^ 
 
 r-CC 
 
 £JS- 
 
 m m IM IM 
 
 
 M 
 
 
 
 01 c^ 
 
 
 
 £ 
 
 
 
 3 
 
 o 
 Q 
 
 ■ 3S 
 
 a> 
 
 
 
 
 
 
 
 
 
 
 b£ 
 
 : Tji Tt< M N 
 
 
 
 
 C 
 
 M C^ 
 
 
 
 
 W 
 
 
 
 •^1^ 
 
 
 
 
 "S 
 
 U 0) 
 
 CD CO CO IN -< t^ 
 
 
 
 
 IN CO (M Ol O) 
 
 
 
 X 
 
 QS 
 
 i-< 
 
 
 
 
 
 
 
 
 
 OJ 
 
 1 
 
 
 
 
 "3 
 
 s-g 
 
 
 
 
 H 
 
 
 -H -• -H 
 
 
 m 
 
 
 
 
 "S 
 
 tH ITS 00 CO t^ 
 
 
 
 01 
 
 ^ fj CM "O 10 CO 
 
 
 
 M 
 
 -^ rtiN r- 
 
 
 
 c 
 coco CO -2 
 
 
 
 m 
 
 
 
 
 V 7' T* =i 
 
 
 
 3 
 
 1 1 _-■ 1 r- 
 
 ■-^ to 'o "■ >o . i; 
 
 
 
 
 r-^ »-l F-l <: i-H M t- 
 
 0; 05C;; '''So 
 
 
 
 6 
 
 
 
 
 ~ 
 
 
 
 T+l 
 
 
 
 ^ C<i 
 
 
 
 ,&c 
 
 ries2 
 and 
 
 I 
 
 
 -° «> 23 
 
 
 
 ^ Cfi C<) 
 
 p. 
 
 
 ^4-1 in 
 03 cj c3 C3.2 
 
 ?3 C^ 3.' « C3 b 
 
 CO CO CO—; rr 
 
 CJ C<< M < Cl t/3 
 
 
 
 
 
 
 ja 
 
 <v 
 
 _ 
 
 J3 
 
 .-a 
 
 m 
 
 
 60 
 
 
 
 
 
 » 
 
 ni 
 
 3 
 
 ^ 
 
 a 
 
 
 
 
 
 
 
 x> 
 
 a 
 
 >>^ 
 
 _>> 
 
 
 s ® 
 
 
 0) 
 
 T3 > 
 
 
 
 a n> 
 
 
 
 
 
 m 
 
 "w -^ 
 
 
 a 
 
 6CO 
 
 
 4^ 
 
 =3'^ 
 
 
 
 
 
 D 
 
 ^ a> 
 
 
 t« 
 
 is 
 
 
 CJ 
 
 
 P^ 
 
 
 a 
 
 ?^" 
 
 
 <u 
 
 
 
 
 CJ 3 
 
 
 
 
 sr, 
 
 ■^ 
 
 
 T-f 
 
 a 3 
 
 rt 
 ^ 
 
 S 
 
 S^ 
 
 
 
 
 
 rt J= 
 
 
 
 ^B 
 
 ^ 
 
 
 
 
 a> 
 
 ■73 
 
 CO « 
 
 £« 
 
 n) 
 
 ■a 
 
 
 tn 
 
 a 
 
 
 P 
 
 
 
 
 ^P. 
 
 3 
 
 
 
 
 CJ 
 
 a 
 
 
 P. 
 
 t4H (h 
 
 OJ 
 
 
 
 n 
 
 '0 
 
 ^fT3 
 
 
 
 3 ci 
 
 QJ 
 
 «w 
 
 a J3 S'co 
 
 IB H .S rt 
 
 I UOI 
 
MVTATION IN MATTHIOLA 
 
 283 
 
 6. THE NARROW-LEAVED TYPE 
 
 As table 37 indicates, this type competes with crenate for second 
 place in frequency of occurrence in the Ithaca cultures ; in fact, when 
 only the strictly unseleeted cultures are considered the percentage is 
 very close to that for smooth-leaved. A feature of special interest is 
 the apparent association of the mutant type with doubleness. 
 
 In a cool greenhouse this type (fig. 22) varied from exceptionally 
 late and many-noded to ordinary in both characters. The leaves (see 
 also fig. 18) were typically narrow, rather strictly entire, often rolled 
 backward or twisted, and typically more ascending than those of 
 
 Table 37 
 
 Narrow-leaved type. Numbers of apparent mutants and association of Hie type 
 
 with doubleness of flowers. 
 
 
 Progeny of Snowflake and early parents 
 
 Culture 
 
 Total 
 examined" 
 
 Narrow-leaved 
 
 
 Single 
 
 Double 
 
 All 
 
 Coeffii'ient of 
 mutation 
 
 1908 
 
 1910 
 
 191 IF, house-sown 
 
 Al'. above 
 
 All unselectnd 
 
 725" 
 
 338 
 
 2072 
 
 3135 
 
 2410 
 
 
 
 1 
 
 7 
 8 
 8 
 
 2 
 4 
 
 12 
 IS 
 16 
 
 2 
 
 6 
 
 20 
 
 28 
 
 26 
 
 .28 ± .26 
 
 1.78 ± ..38 
 
 .97 ± .15 
 
 .89 ± .12 
 
 1.08 ± .14 
 
 ■ See note 6 to table 2. 
 ■• See note c to table 1. 
 
 Snowflake. The apex of the leaf is often more acute than w'ith Snow- 
 flake, and many leaves are mucronate or at least end in a sharp, rigid 
 tip. 
 
 A striking characteristic is the narrowness of the sepals, resulting 
 in frequent early separation at the edges, partially exposing the petals 
 in immature buds. 
 
 Under the less favorable field conditions the plants often remain 
 long as dwarf rosettes, and flower late and feebly if at all. Figures 23 
 and 24 show comparatively well developed plants in the field. 
 
 The type is on the whole very distinct in the field, though there 
 has been some question whether a greenhouse plant such as that in 
 figure 18 is genetically different from those with short and rigid 
 leaves (figs. 22 and 24) ; the very great variability in leaf form due 
 to external conditions makes such a question very difficult without 
 extensive progeny tests. It is now (1918) probable that narrow-dark 
 (p. 143) was not distinguislied from narrow in the greenhouse. 
 
28-t 
 
 MISCELLANEOUS STUDIES 
 
 n 
 
 
 
 
 t* " 
 
 
 ^1 
 
 
 
 
 ^ a 
 
 IC 
 
 
 
 
 0-& 
 
 ^ --H C<) O CO 
 
 
 
 .__, _ _, ,-^ 
 
 
 
 
 
 0(N IN OJ O 
 
 
 
 
 
 <N ^ CO W CO 
 
 
 
 
 ^ 
 
 
 
 
 
 :^ ^ r^ t_i lo 
 
 
 
 
 
 -H -H c^ c-1 r-1 
 
 
 
 
 
 
 
 
 
 ^ 
 
 o to oc 
 
 
 
 * 
 
 Xi 
 
 
 
 
 =9 
 
 
 
 
 o 
 
 o 
 
 
 
 
 a 
 
 
 
 y: :o -^ ^ o 
 
 
 
 Zfl 
 
 T-H T-1 .— ( 
 
 
 ,-, _ _ ,-^ 
 
 
 
 
 s 
 
 O— ^ to in' c? 
 
 
 
 
 "m 
 
 dS d :^ w 
 
 
 
 
 c 
 
 
 
 
 
 i» 
 
 CC O CO o o 
 
 
 
 
 
 o 
 
 -*< ^ c; 
 
 
 
 
 
 •^ 
 
 
 : CO »C Cl 
 
 
 
 
 
 "o 
 
 
 
 
 
 
 
 a 
 
 
 i -H -H -H 
 
 
 5 
 
 s 
 
 
 < 
 
 8 
 
 
 i CC rH rt 
 O f O 
 
 
 S 
 
 t 
 
 
 P-* 
 
 r-l (N 
 
 
 
 >i 
 
 
 
 
 
 
 
 
 
 ^- C-- 
 
 
 
 
 
 £ ^. 
 
 
 >. 
 
 
 
 
 1- 
 
 w CO ^ Ji 3 
 
 c 
 
 
 t 
 
 
 
 lO O) 
 
 ^ 
 
 
 a 
 
 
 
 o 
 
 
 0. 
 
 ^ 
 
 ^« 
 
 p^ 
 
 
 
 3 
 
 O 
 
 q 
 
 r-id m 'S 't^ 
 
 CO 
 
 ^ 
 
 
 
 
 
 'm 
 
 Or^ ^ O I-; 
 
 
 
 
 c: 
 
 
 
 
 
 
 
 
 
 
 CO 
 
 
 
 II 
 
 Clio 1~ C5 ^ 
 
 
 
 c 
 
 C-J rH CO C^ -r^ 
 
 
 
 's 
 
 oe 
 
 
 
 
 « 
 
 
 
 
 
 V 
 
 
 
 
 
 
 
 la 
 
 v'O 
 
 
 
 
 o 
 
 
 OO O O lO 
 
 
 
 ^ 
 
 11 
 
 
 
 
 ».-^ 
 
 
 
 
 ^^ 
 
 
 
 
 - en 00 
 
 
 
 
 ^ lO to CO C2 
 
 
 M 
 
 
 ■* Tf 00 
 
 
 
 CO CO 
 
 
 
 
 ^ ^ ^ 
 
 
 
 
 O Cl r-H 
 
 
 ? 
 
 
 -H -HO-. 
 
 
 
 
 »a =8 '^ 
 
 
 
 S 5 K =^ 
 
 rtCO < -- CO 
 
 
 
 ,-lr-< i-H ,-( 
 
 
 
 
 OO a: Ci 
 
 
 
 
 rt rt r-l ^ 
 
 
 
 C-- 
 
 
 
 V "3 
 
 
 
 — c^ 
 
 
 
 33a 
 33b 
 AUnarnnv- 
 
 leavcd 
 "Short stoi 
 
 (1, P.) 
 27a, "small 
 convex-leav 
 
 c 
 
 
 
 
 
 
 
 
 
 
 'T3 '^ 
 
 ■^ S ■$ 
 
 m O 'A 
 
 ri42! 
 
MITATIOX IX MATTUIOLA 285 
 
 The few singles have produced few seeds, and these were higldy 
 variable in size. The capsule often has a defective septum, more or less 
 of the distal portion being absent. Germination was poor in the small 
 cultures secured (table 38, upper part), with only 10.8 per cent of the 
 mutant type among the progeny. 
 
 This case agrees in most respects with those previously discussed, 
 but adds one point of interest in the occurrence of apparent coupling 
 of mutant type with doubleness rather than singleness. Seed appears 
 to be less abundant and less well developed than with any of the pre- 
 ceding mutant types, facts probably significant in relation to the low 
 percentage of narrow progeny from narrow parents, though the large 
 probable error of the percentage must be considered. 
 
 7. MISCELLANEOUS ABERRANT TYPES 
 
 As part of the aberrant individuals occurring in the greenhouse 
 were either doubles or singles that produced no seed, while practically 
 no seed was produced by any plants in the field at Ithaca or by even 
 .some of the commoner mutant types at Riverside, the opportunity for 
 progeny tests has been almo.st entirely limited to the types so far 
 discussed. 
 
 The narrow-dark-leaved type (table 3) was common and distinct 
 in the field at Ithaca, where it constituted about .48 per cent of the 
 2072 plants from house-sown seed, and has been readily identified 
 in several cas&s at Riverside. It was not distinguished in the green- 
 house cultures, but was very probably included under narrow-leaved. 
 Possibly a single described as "small-convex-leaved" belonged to this 
 type, though two field plants were given this name as distinct from 
 narrow-dark; according to a photograph (fig. 25, second plant from 
 left), another greenhouse plant (a double) may have been similar to 
 narrow-dark-leaved. The narrow-dark-leaved type (figs. 26 and 27) 
 has narrow dark-green leaves, strongly convex upward, and evidently 
 tends to compactness of growth and lateness of flowering ; under field 
 conditions it seems decidedly more like Snowflake than like narrow- 
 leaved. 
 
 The 44 progeny (table 38) secured from the greenhovise single 
 mentionad above included 2 (4) narrow-dark-leaved individuals and 
 3 (5) other plants not Snowflake (the last including two smooth, one 
 large, one slender, and one .semicrenate), besides five undetermined 
 plants. Plainly the type of the parent is still in doubt. 
 
 ri43] 
 
286 MISCELLANEOUS STUDIES 
 
 Another very different greenhouse plant, described as "stout 
 dwarf" (fig. 25, third from left), gave among 29 progeny (table 38) 
 5 (7) individuals evidently not Snowfiake, which may have been 
 narrow-dark or may have belonged to another type tliat was somewhat 
 similar under the conditions of the tests. The parent resembled 
 Snowflake except in its short interuodes and short, stout capsules. 
 
 Four other plants suspected of mutation apparently entirely failed 
 to repeat their type in their progeny, perhaps because of the smallness 
 of the house cultures. One of these was the plant, much branched 
 for the warm greenhouse, third fi'om the riglit in figure 18; another 
 was a very late plant with a remarkably large number of main-stem 
 leaves ; the others were a plant with unusually small flowers and one 
 with some of the leaves somewhat spatulate. Possiblj- all of these were 
 Snowflake, though the second, which gave poor germination, probably 
 was not. All these four plants have been included as Snowflake 
 parents for tables showing numbers of apparent mutants. 
 
 The small-smooth-leaved type is well shown in figure 25 (first and 
 fifth from the left). It is the smallest and weakest of the fairly common 
 and definitely identified types ; it has small, very smooth leaves, and is 
 late in blooming. The two plants .shown were both singles, but they 
 set no seed. 
 
 The semicrenate-leaved type (table 3) differed slightly but appar- 
 ently definitely from Snowflake, somewhat resembling crenate-leaved 
 in leaf form. The one "pointed-erenate-leaved" plant of table 3 may 
 have been crenate-leaved. The "compact" and "curly-leaved" plants 
 of this table have not been identified with any aberrant types in other 
 cultures. "With the remaining six types of table 3 all the individuals 
 have been questioned as possibly Snowflake; it is now practically cer- 
 tain that some of those in the second, third, and fourth groups 
 belonged to the large-leaved type since studied, but the apparent inter- 
 gradation with Snowflake makes any attempt at a definite reclassi- 
 fication from the records a matter of doubtful value. 
 
 The second plant from the right in figure 25 was remarkable for 
 its short stem and few but large leaves. Several other more or less 
 exceptional individuals have appeared in the cultures, especially among 
 some plants with abnormal cotyledons, selected from large numbers of 
 greenhouse seedlings in the 1908 cultures, which were examined for 
 syncotyledony. Some of these were very weak plants which finally 
 died without flowering. 
 
 F144] 
 
MUTATJOX I.\ MATTlllULA 287 
 
 The fluctuations in habit, leaf form, etc.. within the type are such 
 that the determination of familiar types is often a matter of some 
 uncertainty, as is shown b.y data that have been presented. It may 
 well be that among the doubtful types are included several definite but 
 comparative rare mutant forms, which occurred too infrequently to 
 afford adequate material for positive classification. 
 
 8. SOME PEOBABILITES OF RANDOM SAMPLING 
 
 For compactness of presentation and convenience of comparison 
 the material in tables 39 and 40, to which some incidental references 
 have already been made, is collected here rather than scattered through 
 the discussions of the various types, concerned. Some statements as 
 to methods are also necessary in connection with each of the topics 
 here treated. 
 
 First, it should be noted that the percentages previously given 
 have regularity been accompanied by the probable errors of simple 
 sampling. These probable errors have been calculated by the formula 
 
 E per cent = .6744898 y HH . wlicrc p is the percentage of the mutant 
 
 n 
 type ("successes"), q is 1 — p, and // is the size of the sample (the 
 
 number of plants concerned). 
 
 In the heredity tables for each type, p has uniformly been taken 
 as the percentage of the total of the lots compared, or p^. 
 
 For the "mutation coefficient" the percentage of the grand total 
 of unselected house-sown lots has regularly been used. Evidently the 
 few .selected progeny included in tables 1, 28. and 37 shovild be omitted. 
 All the percentages here are so low that the probable errors deserve 
 little confidence, even though n is usually fairly large. The rather 
 close agreement of the percentages of all apparent mutants in the 
 three distinct lots of unselected house-sown cultures suggests that 
 they represent fairly well the population value for the potentialities 
 of the seeds; and even if the mean percentage of the total of the lots 
 for the main comparisons is actually nearer, it is safer to use the 
 larger probable errors resulting from the method here employed. 
 Furthermore strict use of p„ would sometimes require several slightly 
 different probable errors for the same percentage, for use in different 
 comparisons in the same table. 
 
 [145] 
 
288 MISCELLANEOUS STUDIES 
 
 If the probable error of the difference of any two percentages in 
 the same table is to be obtained, therefore, formulae corresponding to 
 those given by Yule (1911, pp. 26-1—267) are applicable. 
 
 Now, it is possible in some of these cases to calculate the actual 
 standard deviation of the percentage in subsamples which make up 
 an aggregate sample. Table 39 givas such actual standard deviations, 
 in comparison with the corresponding theoretical or expected standard 
 
 deviations given bv , 
 
 , = W PI 
 
 " per cent ^^^^ • 
 
 >i — 3 
 Table 39 
 Standard deviations of percentages of mutant types. Values derived from 
 
 'PI, 
 
 compared xoith values expressing the actual 
 
 variability of subso 
 
 mples. 
 
 
 
 N 
 
 / 
 
 n 
 
 p 
 
 Standard deviation of samples 
 
 if mean size n 
 
 Type of parent and grouping of 
 progeny 
 
 Actual 
 
 Theoretical,.* 
 
 PQ 
 n-3 
 
 Difference 
 E<r 
 
 Smooth-leaved type: 
 All lots by parentage 
 AH lot.s as grown 
 
 Germination good 
 
 234 
 234 
 
 187 
 
 6 
 
 12 
 
 7 
 
 39.0 
 19 5 
 
 26.7 
 
 27.35 
 27.35 
 
 29.95 
 
 7.5 
 11 3 
 
 10 9 
 
 7.4 ± 
 11.0 ± 
 j 9.4 =t 
 1 9.2" 
 
 14 
 
 1,5 
 1.7 
 
 + ,1 
 -f- .2 
 -1- .9 
 
 Germination poor 
 
 47 
 
 5 
 
 9.4 
 
 17.02 
 
 5.2 
 
 fl4.9 ± 
 
 \17.6 
 
 3.2 
 
 - 3.0 
 
 Large-leaved type: 
 All lots by parentage 
 All lots as grown 
 
 Germination good 
 
 357 
 357 
 
 260 
 
 20 
 22 
 
 14 
 
 17.85 
 16.2 
 
 18.6 
 
 49.02 
 49.02 
 
 50.38 
 
 10.7 
 10 9 
 
 11.3 
 
 13 ± 
 13.7 ± 
 
 f 12.7 ± 
 112. 7 
 
 1.4 
 1.4 
 1.6 
 
 - 16 
 
 - 2.0 
 
 - .9 
 
 Germination poor 
 
 97 
 
 8 
 
 12.1 
 
 45.36 
 
 8.7 
 
 /16.5 =>= 
 116.6 
 
 2.8 
 
 - 2.8 
 
 Crenate-leaved type: 
 All lots by parentage 
 All lots as grown 
 
 Germination good 
 
 633 
 633 
 
 549 
 
 20 
 
 28 
 
 20 
 
 31.65 
 22.6 
 
 27.45 
 
 29.86 
 29.86 
 
 32.42 
 
 10.6 
 12.5 
 
 10.7 
 
 8.6 ± 
 10.3 =t 
 j 9.5 ± 
 \ 9.3 
 
 .9 
 
 .9 
 
 1.0 
 
 -1- 2.2 
 + 2.4 
 + 1.2 
 
 Germination poor 
 
 84 
 
 8 
 
 10.5 
 
 13.10 
 
 10.5 
 
 f 12.3 ± 
 116.7 
 
 2.1 
 
 - .9 
 
 Seed-size test, smaller seeds 
 
 73 
 
 5 
 
 14.6 
 
 65.75 
 
 13.2 
 
 (13.9 ± 
 \13 8 
 
 3.0 
 
 - .2 
 
 Same, larger seeds 
 
 193 
 
 5 
 
 38.6 
 
 20.21 
 
 9 4 
 
 ( 6.7 ± 
 I 7.9 
 
 14 
 
 + 1.9 
 
 Same, all seeds, by p.arentage 
 Same, all seeds, as grown 
 Slender type: 
 
 All lots by ijarentage 
 All lots as grown 
 
 266 
 266 
 
 243 
 243 
 
 5 
 10 
 
 8 
 13 
 
 53.2 
 26.6 
 
 30.4 
 
 18.7 
 
 32.71 
 32.71 
 
 32.51 
 32.51 
 
 10.3 
 22.9 
 
 17.5 
 19.7 
 
 6.6 =t 
 
 9.7 ± 
 
 9.0 ± 
 11.8 ± 
 
 1.4 
 1.5 
 
 1.5 
 1.6 
 
 -1- 2.6 
 
 -1- 8.8 
 
 -1- 5.7 
 -1- 4.9 
 
 Germination good 
 
 165 
 
 7 
 
 23.6 
 
 33.33 
 
 14.9 
 
 I 10.4 ± 
 1 10.3 
 
 1.9 
 
 + 2.4 
 
 Germination poor 
 
 78 
 
 6 
 
 13 
 
 30.77 
 
 27.2 
 
 j 14,6 ± 
 114.8 
 
 2.8 
 
 -1- 4.5 
 
 Parents ' ' extreme 
 
 38 
 
 3 
 
 12 7 
 
 63 16 
 
 14 4 
 
 ( 15.5 ± 
 1 15.1 
 
 4.3 
 
 - .3 
 
 Parents "ordinary" 
 
 205 
 
 10 
 
 20.5 
 
 26.83 
 
 14.7 
 
 ( 10.6 ± 
 111.2 
 
 16 
 
 -f 2.6 
 
 Narrow-leaved type: 
 All lots as grown 
 
 37 
 
 3 
 
 12.3 
 
 10 81 
 
 8.1 
 
 10.2 ± 
 
 2.8 
 
 - .75 
 
 ' The sei'ond values for some cases in this coliimii are derived from p„ (see text). 
 
 [14fil 
 
MUTATION IN MATTHIOLA 289 
 
 For example, table 27 gives the percentage of large-leaved plants 
 among the 357 progeny of the 20 large-leaved parents as 49.0 ± 1.8 
 
 per cent. This probable error is given by .674-1898 \ — , where 
 
 p = 49.0 per cent, g=;51.0 per cent, and w= 357. These 357 
 progeny, as table 39 indicates, came from 20 parents which contribnted 
 an average of 17.85 progeny each, and the actual standard deviation 
 of the percentage in tiicse 20 sibships was 10.7 per cent. 
 
 Obviously the expected standard deviation of simple sampling for 
 comparison must represent samples not of 357 plants each but of 17.85 
 plants each. Now a percentage is obviously a mean (of values all 
 either or 1). Since "Student" (1908) has shown that the theoretical 
 .standard deviation of the mean in samples is given more exactly by 
 
 O" variate ,i i 0"variate 
 
 than by a mean = 
 
 Vw — 3 \'n 
 
 (the value for the normal curve conventionally used for the probable 
 error of the mean) and since '71, the mean size of sample, is small 
 enough to make the correction a matter of considerable importance. 
 
 V M — 3 is here used. Since o-variate^ VP9- we have (Tmean = 
 
 V PI where «= 17.85. Thi.s gives a theoretical standard devia- 
 'ot — 3 
 
 tion of 13.0 per cent.-- 
 
 It is true (Yule, 19n, p. 260) that the ordinary method of calcu- 
 lation of the actual standard deviation is not satisfactory for means 
 when the samples vary in size. A method has been used, however, which 
 obviates this difficulty, so that comparison with the results given by 
 
 is strictly legitimate. Each squared percentage deviation 
 
 i 
 
 « — 3 
 ha.s been weighted by multiplying it by the number of individual 
 
 plants which it represents, and the summation of squared deviations 
 
 has then been divided, not by 2/, the number of samples, but by 
 
 2/ X «. the number of samples multiplied by the mean weight or 
 
 average size of sample (in other words, by N, the total number of 
 
 individuals).-' 
 
 -- In tlie calculations for table .SO p has been taken as the percentage fjiven 
 in this table, to two flecimal places, while with all other numbers employed in 
 calculation, including n — 3, three or more decimal places have been used as 
 needed. 
 
 -■■! Algebraic proof of the correctness of the method has kindly been furnished 
 by Frank L. Griffin, Professor of Mathematics, Reed College, Portland, Oregon. 
 If it develops that this rather obvious device has not been suggested for the 
 purpose, it is to be i)resented elsewhere with the mathematical proof. When the 
 variates are not grouped in classes the calculation is substantially as easy as 
 without weighting, while the theoretical value is found with much less work 
 than by the method given by Yule (1911, p. 260), which requires the harmonic 
 mean of the sample sizes. 
 
 11471 
 
290 
 
 MISCELLANEOUS STUDIES 
 
 <51 
 
 
 K 
 
 OB 
 
 
 
 
 
 
 w. 
 
 
 £ 
 
 <: 
 
 te:'§ 
 
 
 
 a 
 
 
 
 
 
 
 0; 
 
 .2 
 
 
 
 
 
 » 
 
 
 
 
 
 
 C 
 
 .2 
 
 
 
 
 
 a> 
 
 '> 
 
 
 
 
 
 
 1 
 
 -3 
 
 (N r^ .-Hio O 'JD 
 
 
 O —I 
 
 
 ■•3 
 
 OT 
 
 oce<i «o 05rt-i n 
 
 o 
 
 1--^ cJ 
 
 
 K aj 
 
 c 
 
 ic -c 0-. o: 
 
 
 ,— » ^H 
 
 
 = ? 
 
 '5 
 
 i-* c^ ^ ^. 
 
 ■^ 
 
 
 
 c ^ 
 
 t 
 
 ^ c^ re 
 
 
 
 
 3-° 
 
 13 
 
 ^ c^ 
 
 
 
 
 qS 
 
 TS 
 
 
 
 
 
 0-9 
 eg 
 
 o 
 
 
 
 
 
 o a 
 
 
 
 
 
 
 S-j 
 
 
 
 
 
 
 
 
 
 
 
 
 f, c 
 
 c 
 
 
 
 
 
 gg 
 
 o 
 
 
 
 
 
 
 
 cs 
 
 
 
 
 n S 
 
 
 O CO 
 
 
 
 
 :? c 
 
 > 
 
 r^ oq 
 
 
 
 
 Si 
 
 V 
 
 "U 
 
 2 7 
 
 
 
 
 ^ Q.. 
 
 
 ■^ 00 
 
 
 
 
 t 
 
 
 t^ o 
 
 
 
 
 ^ 
 
 
 
 CC-^ lOCC (XllCO o 
 
 o 
 
 t^ c 
 
 
 
 COCJ ^o ocoo o 
 
 ■M 
 
 lO en 
 
 
 tl 
 
 a; 
 
 OI^ ^o occo o 
 
 
 iC O' 
 
 
 E 
 
 a 
 5 
 
 ^IM C-IO OCOO O 
 
 °. 
 
 o o 
 
 
 C3 
 
 
 
 
 
 >1 
 
 w 
 
 O 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ~ 
 
 
 
 
 
 
 J3 
 
 
 
 
 
 
 01 
 
 
 
 
 
 
 J2 
 o 
 
 
 a 
 .2 
 
 
 
 
 £ 
 
 
 S 
 
 
 
 
 
 
 '? 
 
 
 l— 1 
 
 t-H ^H 
 
 
 'aT 
 
 OJ 
 
 
 
 
 
 
 t! 
 
 ■d 
 
 
 
 
 
 
 C 
 
 
 lO "O 
 
 
 
 
 
 
 '5 
 
 a. CC C IM (MM 
 
 
 o 
 
 O O 
 
 C^l ^ O 'f^ 
 
 
 
 ■2^ 
 
 s 
 
 CO "^ 
 
 
 .—1 
 
 to 
 
 
 5* 
 
 
 
 (N 
 
 
 
 oi u 
 
 ■§ 
 
 
 
 
 
 "32 
 
 
 
 
 
 &£ 
 
 •o 
 
 
 
 
 
 2 m 
 
 O 
 
 
 
 
 
 ft— 
 
 
 
 
 
 
 il 
 
 
 
 
 
 
 
 
 
 
 ga 
 
 1 
 
 
 
 
 
 
 
 
 
 
 
 il 
 
 s 
 
 
 
 
 
 aj 
 
 
 
 
 
 
 
 O 
 
 "o 
 
 o cc o o ^ c^ 
 
 
 CD 
 
 T? o 
 
 
 1- 
 
 ■s<;d —km to 00 
 
 
 
 ^ oc 
 
 
 
 Q> 
 
 CO ■* a-. CO o to 
 
 
 
 
 
 ^^ 
 
 O 
 
 cq lO Tf O O 1^ 
 
 
 O 
 
 ,-. c 
 
 
 
 3 
 
 
 
 
 
 
 J3 
 
 
 
 
 
 
 o 
 
 
 
 
 
 
 
 c 
 
 
 v 
 
 
 
 
 .2 n, 
 
 
 
 
 
 
 
 
 Oj ti 
 
 
 
 
 "S m "^ aj ""^ a; "~ 
 
 & 
 
 .1 c.-> 
 
 
 - 
 
 
 
 
 ^ fe^- 
 
 
 _S 
 
 
 "lj -tJ 
 
 o *- \ 
 
 
 "o 
 
 
 ^^^■^2 ^^K^ 
 
 
 
 
 
 
 •=^0 <D II S 11 -§ >. II 
 
 
 m -a II 
 
 
 
 I. o bf a a -^ 
 
 > c 
 
 
 
 
 ood vs. poo 
 With sm 
 With lar 
 Same, /„ 
 With ore 
 Same, /„ 
 With sle 
 With all 
 
 
 
 
 
 mall vs. 1 
 With 
 
 Extreme 
 
 parents : 
 
 With 
 
 Same, 
 
 
 
 
 o 
 
 
 1 Oj 
 
 1 : 1 
 
 6 a 
 -*j Pi 
 
 O ID 
 » O 
 
 o 
 
 9 II 
 
 g § 
 
 to •<-• 
 
 03 
 
 o 
 
 
 -M 
 
 
 rl 
 
 
 
 fl 
 
 > 
 
 rrl 
 
 
 
 nr! 
 
 
 
 to r:2 
 
 Fh 
 
 ^. 
 
 p^ 
 
 rt 
 
 "— ' 
 
 ^ 
 
 ^. 
 
 fl 
 
 o 
 
 
 
 
 
 [1481 
 
MlTA'nUX IX MATTHIOLA 291 
 
 In the calculation the deviations are taken from zero, and with 
 these small numbers of samples the percentages are not thrown into 
 classes; it si^fficos, then, to S(]nare eacli number of "successes," divide 
 by the corresponding total of individuals, add the quotients, and 
 divide by the grand total of individuals, correcting this weighted 
 mean squared deviation by .subtracting the square of the weighted 
 mean percentage (percentage of grand total). If s is the number 
 of successes and n is the total number of individuals in the sub- 
 
 sample, and M is the weighted mean percentage, then .1/=: — , and 
 
 _ \~yl 
 
 "■percent ^—t 11 -.„ 
 
 Table 39 gives, for the most important comparisons of heredity 
 percentages, the total number of progeny (yV)> the number of cidtural 
 groups or (with the first line for each type) the number of parents (/), 
 the average size of the groups of progeny (w), and the mean per- 
 centage of the mutant type (p). This serves as a summary of some 
 of the most important statistical data already presented relating to 
 the inheritance of these types, and also shows the basis of the remain- 
 ing part of this table and of table 40. For comparison of actual and 
 theoretical standard deviations the theoretical value has been calculated 
 from the actual percentage as given in this table. For comparison of 
 means (table 40) the percentage of the corresponding total (p„) has 
 also been used, this theoretical standard deviation being the second in 
 the table in the cases where the two values are not identical. 
 
 Since small changes in a percentage have little effect on its 
 theoretical standard deviation, we are fairly well justified in taking 
 the latter, as calculated from the actual percentage in each ease, to be 
 the "population" value. Consequently, the difference between the 
 theoretical and actual standard deviations has been expressed in each 
 ca,se as a multiple of the probable error of the theoretical value. 
 
 Aside from the last line for crenate-leaved, where there is an 
 obvious artificial reason for high variability, there is no very significant 
 difference except with slender. In this case, the deviation of 5.7 times 
 the probable error (line 1) is probably largely due to the genetic 
 differentiation of "extreme" and "ordinary" parents suggested by 
 their appearance and by the wide difference in the heredity per- 
 centages ; the differences become moderate when the progeny of the 
 two classes of parents are .separated. 
 
 (1491 
 
292 MISCELLANEOUS STUDIES 
 
 In the two cases (smooth-leaved and crenate-leaved types) where 
 the percentages of mutant types differ greatly with good and poor 
 germination, separation according to germination gives a mean value 
 of the standard deviation decidedly lower than the value for all lots 
 taken together. In the case of the large-leaved type there is little 
 change, while the considerable reduction with the slender type is 
 probably due to unequal separation of lots from parents genetically 
 different. 
 
 Table 40 shows the simple-sampling probability of the most striking 
 differences of heredity percentages, aside from the characteristic 
 differences between different types. "Student's" (1917) table of 
 pr()])abilities of mean deviations with small samples is used, with 
 interpolation by second differences. Where the standard deviation 
 of the difference is required it is found from the theoretical values 
 given in table 39 by the formula (Yule, 1911, pp. 264-265) 
 
 C difference V "^l ~l~ O'; 
 
 J Pogo 
 
 Po g« 
 
 ji, — 3 «2 — 3 
 
 when one statistical population is assumed (table 40, columns 2 and 3). 
 When two popiilations are assumed (table 40, columns 4 and 5) the cor- 
 responding formula using p,i7, and p„q., is employed. In the one case 
 where this is possible (the seed-size test), it is also calculated from the 
 actual differences of the pairs of percentages in the separate tests, each 
 difference being weighted with the total number of progeny from the 
 parent concerned. Where two values of / (the n of "Student's" 
 table) are involved, the smaller is taken, giving understatements of the 
 probabilities involved ; in the two cases where the difference is more 
 than 2. the values are recalculated, with / as the nearest smaller 
 integer to the geometric mean of the two actual numbers (that is 
 with /o= V/i/2)- In the ca.se where the probabilities of four devia- 
 tions all in the same direction are combined, the four chances of 
 occurrence are nuiltiplied together; that is, if the i(l-|-a) of 
 "Student's" table is P, and 1 — P is F, then F,.„.,.^ = F,-F„-F^-F,. 
 "Student" (1908, p. 1) says, "The usual method of determining 
 the probability that the mean of the population lies within a given 
 distance of the mean of the sample, is to assume a normal distribution 
 about the mean of the sample . . . ." AVhen this is done with a differ- 
 ence of means, it is at once evident that only half of the chances of 
 deviations as great as the distance of the given difference from zero 
 difference lie below zero difference: the other half of the chances of 
 
 (isoi 
 
MUTATION IN MATTUWLA 293 
 
 such deviations lie in the opposite direction and represent positive 
 differences still greater than the sample difference. In other words, 
 it' the implications of a sample difference are to be given full weight, 
 thi.s dift'erence must be considered the most probable value of tlie 
 theoretical "true" difference between two assumed distinct statistical 
 populations. In the present case we wish to know the probability that 
 the "'true'' or theoretical-population means differ in tlie same sense 
 as the observed sample means. This involves calculation of the proba- 
 bility of deviations in one direction (beyond zero difference) from 
 the .sample difference. If the sample difference of means is considered 
 as positive, then the negative "tail" of the theoretical frequency 
 curve of sample differences (this curve being centered at the observed 
 sample difference) must be compared with the rest of the curve. The 
 positive portion of the curve the i (1 -(- o)-* of the tables, then gives 
 the chances favoring the hypothesis that the sample means truly 
 represent the population means. The odds in favor of the hypothesis 
 are therefore given by the formula 
 
 Values calculated from this fornnila are given in columns 4 and 5 of 
 table 40. 
 
 When other considerations than the sample evidence are to be taken 
 as determining the most probable value of the "true" mean, the case 
 is different. For example, if the probability that our sample per- 
 centages are mere sampling deviations from some theoretical Mendelian 
 value were in question, that theoretical value mu.st be taken as the 
 population mean and only the magnitude of the deviations must be 
 considered. 
 
 When a difference of means is considered from this latter stand- 
 I)oint, it is assumed that the two samples come from one statistical 
 population, and hence that zero is the most probable value of the 
 population difference. If we choose to assume tliat the most probable 
 value of the population difference in our eases is zero, we must 
 calculate the odds against a deviation of the observed amount in 
 either direction from zero difference. The formula for these odds is 
 
 „ ^ (l+„)_^ (l_a) 
 
 U, = „ ~ : or -L . 
 
 2X^ (1— a) 
 
 -■i Tlie whole area of the frequency eurve is taken as unity, and a is the area 
 enclosed by any given deviation in both directions from the mean. 
 
 [1.511 
 
294 MISCELLANEOUS STUDIES 
 
 Values from this formula are given in columns 2 and 3 of table 40 ; 
 their magnitude in three eases, however, and the uniform agreement 
 of the direction of difference with the exjiectation from biological 
 evidence which has been discussed, weigh heavily in each test against 
 the assumption of random sampling from a single statistical population. 
 
 It does not appear necessary, however, thus to weigh the evidence 
 in detail before deciding which formula is suited to the ease. There 
 is no evident theoretical value from which these percentages are 
 reasouablj' likely to be sampling deviations. This being the case, and 
 granting such general possibilities as that of differential viability, it 
 seems most reasonable to use the former (Oj) formula. That is, we 
 .should give full weight to tln' implications of a sample deviation 
 luiless there is some definite reason for assuming that some other value 
 better represents the mean of the theoretical statistical population. 
 
 It must be remembered that the actual probabilities of sampling 
 deviations do not necessarily correspond closely with the probabilities 
 of random sampling. With the material in table 40, however, aside 
 from the germination comparison in the case of the slender type, 
 table 89 suggests a fair agreement with the conditions of random 
 sampling. The actual standard deviations of the subsamples do not 
 in general differ widely from the corresponding theoretical values, and 
 the differences are negative about as often as positive. 
 
 The hypothe.sis of selective elimination with poor germination is 
 strongly .sustained (table 40), although only one difference (with the 
 crenate type) has much statistical signiticance when considered alone. 
 If we may multiply together the members of the four ratios in column 
 3 of the table, the combined odds (using the /„ values) are 130:1 
 against occurrence of these four deviations as accidents of simple 
 sampling, when magnitude of deviation alone is considered. If 
 direction of deviation alone is considered the random chance of these 
 four deviations all in the same direction is obviously (^)*, or the odds 
 favoring the elimination hypothesis are 15:1. Combination of these 
 two chances indicates a high probability for the hypothesis. When 
 the two-population formula is used in calculating the standard devia- 
 tion of the difference (columns 4 and 5) the value of P is consider- 
 ably reduced in some cases, and the combined odds obtained from 
 F.^- F.,--F^-Ft are very high. Evidently the best single expression 
 of the simple-sampling odds, though possibly somewhat too high, is the 
 value given last in column 5. or 123,093:1. 
 
 With the seed-size test of crenate the odds are 499 : 1 with the 
 theoretical standard deviation of the difference, or 1666:1 with the 
 
 [152] 
 
Mrr.iTioN IN iiArriiioLA 295 
 
 actual standard deviation. When the relatively small size and weak 
 growth of erenate seedlings are also taken into account, the relativel.y 
 small average size of erenate embryos may be considered to be 
 demonstrated beyond reasonable doitbt. 
 
 With "extreme" and "ordinary" slender parents the odds de- 
 cidedly favor the hypothesis of genetic differentiation of parents, in 
 spite of the small numbers involved. We must remember that definite 
 statistical differentiation of lots of progeny grown under uniform con- 
 ditions does not necessarily demonstrate genetic differences (differences 
 in output of gametes) between the parents; in this case, however, the 
 difference in the appearance of the parents and in the single-double 
 ratio among the progeny also suggest genetic differentiation. 
 
 GENERAL DISCUSSION-' 
 
 It might be argued with some plausibility that the available 
 evidence hardly justifies conventional factorial analysis, or at least that 
 the data indicate strongly the presence of marked factorial incon- 
 stancy. The aberrant types occur in very small proportions among 
 the progeny of selfed Snowflake parents, in much larger proportions 
 from "mutant-type" parents, and in intermediate proportions from 
 crosses with Snowflake. It might be supposed that the Snowflake type 
 has a slight tendency to nuitate to the other types, and that these have 
 a much more marked tendency to mutate back to Snowflake. Varioiis 
 considerations, however, especially the occurrence of apparently 
 regular linkage phenomena, seem to favor the general form of 
 hypothesis which has been presented. 
 
 As we have seen, it is well known from the behavior of various 
 factors that the typical Mendelian mechanism is present in Maiihinla. 
 It cannot be argued here, as sometimes with Oenothera, that the 
 genetic behavior of the genus or species is fundamentally non- 
 Mendelian. Since the Mendelian mechanism is demonstrably present, 
 and [Muller's (1918) work on beaded wings in Drosophila seems to 
 establish the adequacy of this mechanism in a closely parallel case, 
 surely conventional factorial analysis .should be carried as far as pos- 
 sible; in fact (iMuller. 1918, p. 423), a Mendelian explanation should 
 not be abandoned for anything .short of positively contradictory 
 evidence. 
 
 2jMuIler's (1918) complete report on the beailed-wing case in Drosophila 
 appeared several months after the present paper had gone to the publisher. 
 Certain conclusions given below, very similar to Muller 's but not credited to 
 him, were therefore reached independently. 
 
 f 1.531 
 
296 MISCELLANEOUS STUDIES 
 
 In the Drosophila case just mentioned, the " principal'' factor for 
 the character in question is "dominant for its visible effect and 
 recessive for a lethal effect," so that no pure beaded individuals 
 appear among the progeny of beaded. The original race regularly 
 gave progeny partly heterozygous beaded and partly homozygous 
 normal, while after a long period of selection 4 true-breeding beaded 
 race appeared. This latter form, it proved, fails to give normals not 
 because of being duplex for beaded — it is still simplex — but because 
 of its possession of another factor, known only by its lethal effect 
 when homozygous, which is carried by the chromosome bearing the 
 normal allelomorph of the factor for beaded. The locus of this reces- 
 sive lethal factor gives in general about 10 per cent of crossovers with 
 the locus of beaded, but in this case, because of the presence of a factor 
 "which almost entirely prevents crassing over" between the loci of 
 the two lethal factors, viable non-beaded zygotes are very rarely 
 produced. Thus every zygote receiving either two beaded-carrying 
 chromosomes or two non-beaded-carrying chromosomes of the pair 
 concerned fails to develop, and all the insects produced are necessarily 
 heterozygous for both lethal factors. 
 
 A point of special interest in this ca,se is the fact that by certain 
 crosses individuals can be produced which give certain types among 
 their progeny in very small percentages. MuUer suggests that part 
 at least of the supposed mutants of Oenothera may be due to crossing 
 over between chromosomes carrying lethal factors, by which certain 
 recessive factors are permitted to come to expression in viable zygotes. 
 
 For the inheritance of doubleness of flowers in Matthiola he gives 
 a "balanced-factor" explanation essentially identical with mine (Frost, 
 1915). 
 
 There seems to be little reason to doubt that the differential factors 
 for these aberrant Matthiola types have originated by mutation. On 
 the analogy of Drosophila we might expect that the true mutations 
 woiild be relatively rare, and that most of the apparent mutants, in 
 cases where they appear frequently, would be due to segregation, 
 appearing as the result of crossing over in chromo.somes carrying 
 balanced lethal factors. The evidence seems to indicate, however, that 
 the differential factors for the mutant types at all extensively studied 
 are dominant for their visible effects and usually (probably imper- 
 fectly) recessive for a lethal effect, the mutant factors thus being 
 genetically similar to the factor for beaded wings in Drosophila. 
 This would seem to imply the occurrence of certain mutations in pro- 
 
 ri54i 
 
MUTATION IN MATTHWLA 297 
 
 portions as high as about 1 per cent, and a general mutation coefficient 
 of perhaps 4.5 per cent, wliile the only jNIendelian alternative would 
 seem to be some more complex scheme whose satisfactory formulation 
 might require much more extensive hybridization data. 
 
 To be more specific: (1) these types are not single reeessives, since 
 they are not homozygous but split into the mutant and "normal" 
 types; (2) they are not simple eases of multiple reeessives, as has 
 been proposed by Ileribert-Nilsson (1915) for Oenothera mutations, 
 since what is on that hypothesis the full dominant type reappears with 
 selling; (3) if these types are single dominants, as they appear to be, 
 they cannot (barring the action of inhibiting factors) arise from the 
 pure recessive "normal" or Snowfiake type by segregation, but only 
 by immediate mutation ; (4) they are not simple cases of comple- 
 mentary dominant factors, since they occur among the progeny of 
 selfed parents. 
 
 "We might assume that a "mutant" type depends on two pairs of 
 factors, one homozygous and the other heterozygous, while both pairs 
 are heterozygous iu the "mutating" Snowflake parent. Thus the 
 
 crenate tvpe might have the zygotic formula —, -r , where d is the 
 
 d ci 
 
 factor for double flowers, (' a dominant factor for crenate, and I a 
 
 dominant inhibitor of C, all three loci being situated in the same 
 
 eliromosome, at distances of, say, 16 and 4 units apart, in the order 
 
 indicated. A Snowflake parent producing crenate progeny would 
 
 then be^; — - or ., ' , aud crossover combinations would produce the 
 dci del 
 
 apparently mutant crenate progeny. The crenate progeny would 
 
 behave as heterozygous dominants when selfed, and if CC zygotes 
 
 were non-viable would yield constant Snowflake and inconstant 
 
 crenate ; the extracted Snowflake singles, having the composition 
 
 Dci 
 
 — r^, could not throw crenate individuals except bv true mutation of 
 del 
 
 c to C. "With selfed Snowflake, if we assume 16 per cent and 4 per 
 cent of crossing over in the two positions, and a 60-per-cent selective 
 elimination of crenate zygotes, all CC zygotes being non-viable, sub- 
 stantially the observed percentages of crenate singles and doubles 
 result.^" 
 
 2" See page 125, footnote. This scheme agrees fairly "nell with the results 
 from crossing, and gives almost exactly the observed proportion of total doubles 
 (a little over 53 per cent) for selfed Snowflake. Its adequate presentation must 
 be reserved for a later paper. 
 
 [155] 
 
298 MISCELLANEOUS STUDIES 
 
 Formerly (Frost, 1916) the hypothesis of frequent dominant 
 mntations seemed the more probable, but there is apparently non- 
 conformable evidence. It is true that the peculiar behavior of the 
 slender type might conceivably depend on an occasional mutation in 
 another locus, or an exchange (Shull, 1014) or duplication of loci. 
 giving two similar or identical factors for slender. An apparently 
 fatal objection, however, is the fact that the supposed mutants seem 
 to show linkage with singleness or doubleness at their origin from 
 Snowflake as well as in subsequent generations — a fact which strongly 
 suggests segregation in the former case. 
 
 If the apparent mutants are really due to segregation complicated 
 by lethal action, the origin of the complex heterozygosis indicated for 
 Snowflake is doubtful ; it may be due to hybridization, but more 
 probably to a gradual accumulation of mutant factors in balanced- 
 lethal combinations. On the analog}^ of ^MuUer's Drosopliila case, 
 aspecially, it might be expected that the latter would be the true 
 explanation, particularly since self fertilization seems to be the rule 
 in Matthiola. On this basis the term mutant type is used with some 
 confidence in this paper, while the aberrant individuals have been 
 called apparent mutants. 
 
 We must not forget that some of the mutant types may arise, as 
 with Oenothera gigas and 0. lata, by non-disjunction, or reduplication 
 of chromosomes, and that this fact may determine tlieir heredity. 
 This is not to be expected with the types whose factors show apparent 
 coupling with singleness or doubleness, but it might be true of the 
 apparently unlinked smooth-leaved type. A preliminary study of 
 several types shows that the usual somatic number of chromosomes 
 is probably fourteen, but that positive counts are difficult. While it 
 might be very hard to demonstrate the regular presence of one extra 
 chromosome in an individual or a type, it should be easy to decide 
 between the diploid and triploid numbers. The large-leaved type is 
 so strongly suggestive of 0. gigas that it would not be surprising to 
 find the triploid number in the material now on hand for examination. 
 
 In a preliminary paper on these types the writer (Frost, 1916) 
 discussed some possible relations of mutation, heterozygosis, and 
 partial sterility, with special reference to Oenothera, mentioning the 
 po&sibility that special prevalence of heterozygosis in the genus may 
 be, "in large part, a result rather than a eause of mutation." This 
 suggestion is evidently justified even if much of the supposed mutation 
 of Oenothera is really segregation, since it is highly probable that 
 
 [156] 
 
MUTATION IN MATTHIOLA 299 
 
 the peculiar plienomena depend on lethal factors or combinations of 
 factors originally due to mutation. 
 
 Another possibility there mentioned, advanced by Belling (1914) 
 and since specially discussed by Goodspeed and Clausen (1917), is 
 that of the occurrence of lethal combinations of certain factors which 
 in other combinations may be in no way prejudicial to normal develop- 
 ment. As the latter paper shows, it is probable that in certain 
 crosses between "good species" most of the new combinations brought 
 together in the formation of Fj gametes are incompatible witli the 
 production of functional gametes. Perhaps in the case of Oenothera 
 there ma.y exist within a species factors lethal in any combination 
 when homozygous, and other factors lethal only in certain com- 
 binations. 
 
 A balanced-factor explanation for the inheritance of doubleness-' 
 in Matthiola, a ease which ]\Iuller (1918) discusses, seems to have been 
 first definitely stated by Goldsehmidt (1913), though he failed to pro- 
 vide for one feature of the evidence, the deviation of the heredity 
 ratio from 50 per cent. As has been shown (Frost, 1915), this 
 peculiarity may be due to greater viability of the homozygotes (sterile 
 doubles) during embryonic development, since the doubles are more 
 viable in the mature seeds and more vigorous in later development 
 (Saunders, 1915). In this case the "normal" factor is completely 
 eliminated in favor of the mutant (sterile-double) factor in the 
 formation of the sperms, and probably is partially eliminated in the 
 formation of either the eggs or the embryos or both. 
 
 Here the normal singleness (sporophyll) factor D may act as a 
 lethal in the heterozygous parent, possibly from its general relations 
 of growth vigor in the presence of the more vigorous d-carrying cells. 
 If the lethal factor is situated in a distinct locus, evidently crossovers 
 are at most extremely rare. It is true that Mi-ss Saunders (1911) 
 finds that Fj hybrids with pure single forms produce functional 
 single-carrying pollen ; with the pure single forms from which the 
 original "double-throwing" mutants arose, however, this might not 
 be true, or a lethal change may have occurred in the singleness factor 
 itself rather than in a factor coupled with it. The Drosophila ease 
 would suggest a lethal change in another locus of the single-carrying 
 chromosome. 
 
 In my paper of 1915 this lethal change in one chromosome ap- 
 parently accompanying the mutation of D to d in the homologous 
 
 -" For a brief outline of the genetic behavior of doubleness see the discussion 
 of the experimental data for the smooth-leaved type. 
 
 [157] 
 
300 MISCELLANEOUS STUDIES 
 
 chromosome was considered puzzling. Evidently, however, it may 
 have occurred in one chromosome before D mutated to d in the other, 
 and even then may have produced its lethal effect. It is evident 
 that if doubleness should arise in the absence of the lethal effect it 
 would tend to be eliminated by the return of one-third of the singles 
 to the homozj'gous condition in each generation. In fact, it is possible 
 that the lethal change arose later than doubleness, as in the Droso- 
 phila ease, or was brought in later by cross pollination, and happened 
 to be preserved as a result of horticultural selection for a high pro- 
 portion of double.s. 
 
 A parallel-column comparison between the double type and the 
 types especially discu.ssed above has already been given, in connec- 
 tion with the smooth-leaved type. It will now be seen that this com- 
 parison seems to apply to all mutant types, except early, that have 
 been genetically tested, the principal differences between these types 
 relating to the heredity percentage and the apparent presence or 
 absence of linkage with the single-double factors. 
 
 Prom the standpoint of its relation to genetic analysis the double- 
 ness factor is remarkably similar to the sex factor in animals. There 
 are two types in each generation, one heterozygous and the other 
 evidently homozygous, and these types are produced by the fertiliza- 
 tion of two kinds of eggs, produced in equal or nearly equal numbers, 
 by a single kind of sperm. Although one of the somatic types is 
 sterile, and the uniformity of the sperms produced by the other is due 
 (evidently) to lethal action, the opportunity for chromosome analysis 
 is similar to that with sex chromosomes. 
 
 We may say that the doubleness factor and its normal allelomorph 
 (d and D) are carried by chromosome pair I. Already we know 
 several other pairs of factors evidently carried by this pair of chromo- 
 somes. These are, to name only the mutant or possibly mutant 
 member of each pair of factors: P (pale sap color) and W (colorless 
 plastids), both studied by Miss Saunders (1911, 1911a) ; C (crenate- 
 leaved), S (slender; possibly two factors), and ^V (narrow-leaved). 
 As we have seen, the last three of these are probably lethal when 
 homozygous, and one or more unidentified lethal factors may be con- 
 cerned in the breeding results, while the doubleness factor affects the 
 race much like a recessive lethal, since all dd individuals are completely 
 sterile. 
 
 1158) 
 
MUTATlOy IN MATTHIOLA SOI 
 
 SUMMARY 
 
 This paper describes the occurrence, characteristics, and heredity 
 of certain aberrant plant types which decidedly resemble some of the 
 "mutant" types produced bj' OeiiotJiera lamarckiana. The parent 
 form is Matthiola annua Sweet, of the horticultural variety "Snow- 
 flake." 
 
 These aberrant forms may be called mutant types, since it is highly 
 probable that they are originally produced by mutation. The aberrant 
 individuals may be termed apparent mutants, since it may be con- 
 sidered uncertain whether they iisually arise by immediate mutation 
 or by segregation. The case acquires special significance because indi- 
 viduals belonging to the mutant types, although the species is known 
 to be typically Mendelian with respect to various characters, give 
 erratic heredity ratios suggestive of Oenothera. 
 
 At least eight types have been somewhat carefully studied, and six 
 of these have shown their heritability in progeny te.sts. Several other 
 types have been named, but for various reasons their distinctness is 
 more or less doubtful. 
 
 Some of the commoner types have each been produced by many 
 parents, and in several pure lines isolated from the original com- 
 mercial variety. The apparent mutants other than the early type com- 
 pose about four or five per cent of the progeny of Snowflake and early 
 parents, the separate types ranging down from about one per cent. 
 
 Most of the mutant types are in general inferior to Snowflake in 
 vigor, and the difference in development is greatly increased by certain 
 unfavorable environmental conditions. The proportion of apparent 
 mutants in cultures from Snowflake parents appears to be definitely 
 lower where germination is comparatively poor. 
 
 The mutant types differ from Snowflake and from each other in 
 various respects. The early type is practically a smaller and earlier 
 Snowflake. The other mutant types, on the other hand, differ markedly 
 from Snowflake in vigor, fertility, and various form and size char- 
 acters. Each type is named from some conspicuous characteristic 
 difi'erence from Snowflake. but usually various other differences can 
 readily be found. 
 
 Somewhat extensive progeny tests have been made for five of the 
 mutant types, and a little evidence secured for two or three other types. 
 
 I 1.5fl I 
 
302 MISCELLAXEOUS STUDIES 
 
 The early type is probably due to a single dominant mutant factor 
 Segregating normally from the corresponding Snowfiake factor; the 
 quantitative nature of its differences from 'Snowflake, however, makes 
 positive determination of this point a matter of great difficulty. 
 
 At least five other types plainly reproduce themselves, but about 
 50 to 70 per cent of the progeny are usually Snowflake; no true- 
 breeding individual of any generation of any of these types has yet 
 been tested. Genetic work with most of those types has been much 
 hampered or even prevented by low vigor and fecundity, and the 
 aggregate data from progeny of parents of four types strongly indi- 
 cate selective viability at germination. It has been determined by 
 crossing that in three of the types the mutant factor (or factors) is 
 carried both by eggs and by sperms. From these facts it seems prob- 
 able that homozygotes of the mutant types are non-viable, and that 
 severe selective elimination occurs during embryonic development ; 
 or. in other words, that the mutant factor is imperfectly recessive for 
 a lethal effect. 
 
 In three types there appears to be linkage with the factor pair for 
 singleness and doubleness of flowers, the mutant factor being coupled 
 with singleness in the tested apparent mutants of two types, and with 
 doubleness in the third type. With two other types these factors 
 seem to be independent. No reversal of coupling has been found in 
 later generations of the former two types, but on the scheme presented 
 crossover singles should be scarce. 
 
 For one type (crenate-leaved) a hypothesis based on the facts stated 
 gives very closely the ratio obtained from selfed parents. Reciprocal 
 crosses with Snowflake conform less clo.sely to the requirements of the 
 hypothesis, but do not definitely contradict it. The slender type, 
 which shows similar apparent linkage, seems to disagree definitely 
 with the hypothesis; there is strong evidence, however, that slender 
 individuals may differ genetically among themselves. 
 
 A more complex scheme providing also for the usual origin of these 
 types from Snowflake by segregation is briefly outlined. 
 
 The selfing ratios are very suggestive of duplication of a chromo- 
 some (non-disjunction), as in Oenothera lata, but it is hard to 
 reconcile the cases of apparent linkage with this hypothesis. It seems 
 probable that these three linked types have originated and are trans- 
 mitted in the same general way as the double-flowered type, and that 
 all of these four mutant factors (including double) represent changes 
 of some sort within a chromosome of the same pair, which may be 
 
 [1601 
 
ULTAIION IN UATIUIOLA :{(i:i 
 
 iiuinbered I. ^liss Sanndcr's work shows that two flower-coh)!- factors 
 also beh)ng to this linked group. 
 
 The large-leaved type strikingly resembles Oenothera gigas, and it 
 may prove to be triploid in nuclear constitution. In that case segrega- 
 tion may be irregular and genotypieally intermediate individuals may 
 be more or less frequently produced. 
 
 It is jirdbable that further study of these types will help to explain 
 the rcmarK'nlile ijciietic beliavinr of Oenallivra and of ('Urns. 
 
 LITERATURE CITED* 
 
 Atkinson, George P. 
 
 ]ni7. Quadruple liybrids in the F, generation from Oenothera miianx and 
 Ooiothera pycnocarpa, with F~ generations and back- and inter- 
 crosses. Genetics, vol. 2, pp. 21.3-260, 11 tables, 1 diagr., 1.5 figs. 
 
 Babcock, Ernest B. 
 
 191S. The role of factor mutations in evolution. Am. Naturalist, vol. 52, 
 pp. n6-12S. 
 Barti.ett, 11. 11. 
 
 1917. Mutation in Mafthiola aiinmi. a " Mendelizing" species. [A review 
 of paper of same title by IT. B. Frost.] Hot. Gaz., vol. 63, pp. 82-83. 
 Bateson, Wlliam, and Saunders, Edith E. 
 
 1902. Experimental studies in the physiology of heredity. I. Experiments 
 with plants. Matthiola. III. Discussion. Roy. Soe. London, Re- 
 liorts to the Evolution Committee, vol. 1, pp. 32-87, 12.'5-160, 1.5 
 tables. 
 Bateson William, Saunders, Edith R., and Punnett, Reginald C. 
 
 190.5. Experimental studies in the physiology of heredity. Matthiola. Roy. 
 Soc. London, Reports to the Evolution Committee, vol. 2, ]ip. 5-44, 
 tables. 
 1906. Ibid. Stocks. Ibicl.. vol. 3, pp. 38-53, 4 tables, 2 figs. 
 Bateson, William, Saunders, Edith R., Punnett, Reginald C, and Killbt 
 (Miss) H. B. 
 1908. Experimental studies in the ph.ysiology of heredity. Stocks. Roy. 
 Soc. London, Reports to the Evolution Committee, vol. 4, pp. 3.5-40, 
 3 tables. 
 Belling, John. 
 
 1914. The mode of inheritance of semi-sterility in the offspring of certain 
 hybrid plants. Zeitschr. f. indukt. Abstam.- u. Vererbungsl., vol. 12, 
 pp. 303-342, tables, 17 figs. 
 Bi.akeslee, Albert P., and Avery, B. T. .Tr. 
 
 1919. Mutations in the jimson weed. Jour. Heredity, vol. 10, pp. 111-120, 
 11 figs. 
 
 * An asterisk prefixed to the date indicates that the pa]ier cited has not been 
 seen by the present writer. 
 
304 MISCELLANEOVS Sl'VDIKS 
 
 CoRBENS, Carl E. 
 
 ]900. uber Levkojenbastanle. Zur Kenntniss der Grenzen der Mendel 'schen 
 
 Regehi. Bot. Centralbl., vol. 84, pp. 97-113. 
 1902. Seheinbare Ausnahme von der Mendel 'schen Spaltungsregel fiir Bas- 
 tarde. Deutseli. bot. Ges.. Ber., vol. 20. pp. 159-172, 4 tables. 
 
 Davis, Bradley, M. 
 
 1917. A criticism of the evidence for the mutation theory of de Vries from 
 
 the behavior of species ot Oenothera in crosses and in selfed lines. 
 
 Nat. Acad. Sci., Proe., vol. 3, pp. 705-710. 
 Frost, Howard B. 
 
 1911. Variation as related to the temperature environment. Am. Breeders' 
 
 Assoc, Ann. Eept., vol. 6, pp. 3S4-395, 4 tables, 4 charts. 
 
 1912. The origin of an early variety of Matthiola by mutation. Ibid., 
 
 vol. 8, pp. 536-545, 5 tables. 
 
 1915. The inheritance of doubleness in Matthiola and Petunia. I. The 
 
 hypotheses. Am. Naturalist, vol. 49, pp. 623-636, 1 fig., 2 diagr. 
 
 1916. Mutation in Matthiola annua, a " Mendelizing" species. Am. Jour. 
 
 Bot., vol. 7, pp. 377-383, 3 figs. 
 
 1917. A method of numbering plants in 2Jedigree cultures. Am. Naturalist, 
 
 vol. 51, pp. 429-437. 
 
 Gates, E. E0cgles. 
 
 1915. The mutation factor in evolution. Jjondon, Macmillan, xiv + 353 pp., 
 
 1 map, 114 figs., bibl. 
 
 GOLDSCHMIDT, RlCH.\RD. 
 
 1913. Der Vprorbnngsmodns der gefiillten TiCNkojenrassen als Fall geschleehts- 
 
 begrenzter Vererbung? Zeitschr. f. indukt. Abstam.- u. Verer- 
 bungsl., vol. 10, pp. 74-98, diagr. 
 
 1916. Nochmals iiber die Merogonie der Opnotlierabastarde. Genetics, vol. 1, 
 
 pp. 348-353, 1 pi. 
 
 GooDSPEED, Thomas H., and Ci ausen, E. E. 
 
 1917. Mendelian-factor differences versus reaction-system contrasts in hered- 
 
 ity. Am. Naturalist, vol. 51, pp. 31-46, 92-101. 
 
 Heribert-Nilsson, N. 
 
 *19]5. Die Spaltungserscheinungen der Oenothera himarclHana. Lunds Uuiv. 
 Arsskrift, vol. 12, pp. 4-131. (Review by Ben C. Uelmick in Bot. 
 Gaz., vol. 63, 1917, pp. 81-82.) 
 
 Muller, Hermann .T. 
 
 1917. An Oenotlicra-like case in Drosopliila. Nat. ,\ead. Sci., Proc, vol. 3, 
 
 ]ip. 619-626. 
 
 1918. Genetic variability, twin hybrids and constant hylirids, in a case of 
 
 balanced lethal factors. Genetics, vol. 3, jip. 422-499, 1 table, 1 fig., 
 1 diagr. 
 
 Saunders, Edith E. 
 
 1911. Further experiments on the inheritance of doubleness and other char- 
 acters in stocks. Jour. Genetics, vol. 1, pp. 303-376, 8 tables. 
 1911a. The breeding of double flowers. Fourth Intern. Conf. on Genetics, 
 
 Proc, pp. 397-405, diagr. 
 ■ 1913. Double flowers. Roy. Hort. Soc, Jour., vol. 38, pt. 3, pp. 469^82. 
 
 fui-ji 
 
MUTATION IN MATTHIOLA 305 
 
 1913a. On the mode of inheritance of certain characters in double-throwing 
 
 stocks. A reply. Zeitschr. f. indukt. Abstam.- u. Yererbungsl., vol. 
 
 10, pp. 297-310. 
 1915. A suggested explanation of the abnormally high records of doubles 
 
 quoted by growers of stocks (Matthiola). Jour. Genetics, vol. 5, 
 
 pp. 137-143, 3 tables. 
 191 G. On selective partial sterility as an explanation of the behavior of the 
 
 double-throwing stock and the petunia. Am. Naturalist, vol. 50, 
 
 pp. -1S6-498. 
 
 Siiui.L, George H. 
 
 1914. Duplicate genes for capsule form in Bursa bursa-pastoris. Zeitschr. 
 f. indukt. Abstam.- u. Vererbungsl., vol. 12, pp. 97-149, 5 tables, 
 7 figs. 
 
 ' ' Student. ' ' 
 
 1908. Tlie ]irobahle error of a mean. Biometrika, vol. 6, pp. 1-25, tables, 
 4 diagr. 
 
 1917. Tables for estimating the probability that the mean of a unique series 
 
 of observations lies between — ro and any given distance of the 
 mean of the population from which the sample is drawn. Ibid., 
 vol. n, pp. 414-417, tables. 
 
 Swingle, Walter T. 
 
 1911. Variation in first-generation hybrids (imperfect dominance): its pos- 
 
 sible exjilanation through zygotaxis. Fourth Intern. Conf. on 
 Genetics, Proc, pp. 381-393, 10 figs. 
 
 TSCHKKMAK, ErICH VON. 
 
 *1904. Wcitcre Kreuzungsstudien an Erbsen, Levkojen u. Bohnen. Zeitschr. 
 f. d. landw. Yersuchswesen in Oesterreich, 1904, pp. 533-638. 
 
 1912. Bastardierungsversuehe an Erbsen, Levkojen, und Bohnen mit Riiclc- 
 
 sicht auf die Faktorenlehre. Zeitschr. f. indukt. Abstam.- u. Yerer- 
 bungsl., vol. 7, pp. 81-234, tables. 
 
 Webber, Herbert J. 
 
 1906. Pedigree records used in the plant-breeding work of the Department 
 of Agriculture, in L. H. Bailey, Plant Breeding (New York, Mac- 
 millan), pp. 308-319. 
 
 DE Vries, Hugo. 
 
 1906. Species and varieties: their origin by mutation. Ed. 2, Cliieago, Open 
 Court Pub. Co., xviii + 847 pages. 
 
 1918. Twin hybrids of Oenothera hoolceri T. and G. Genetics, vol. 3, pp. 397- 
 
 421, 14 tables. 
 
 1919. Oenothera riilirinervis, a half mutant. Bot. Gaz., vol. 67, pp. 1-26, 
 
 tables. 
 Yule, G. TJdnt. 
 
 1911. An introduction to the theory of statistics. London, Charles Griffin & 
 Co., xiii + 376 pages, 53 figs. 
 
 ri6:!l 
 
306 MISCELLANEOUS STUDIES 
 
 PLATE 22 
 
 The Eaely Type 
 
 Fig. 1. March 20, 1908. The single progeny of WG9. Plants from house M 
 to the reader's left from stake, from house W to right of stake, from house C 
 below. WG9-C10, the early apparent mutant, is the middle plant in the lower 
 row. The stake indicates inches. 
 
 Fig. 2. About May 1, 190S. WG9-C10 at the left, WG9-C9 (Snowflake) at 
 the right. 
 
 ri64i 
 
MVTATIOS L\ .U.irrUlULA 
 
 307 
 
 
 
 Fig. 1 
 
 Fig. 2 
 
 [ FROST 1 PLATE 22 
 
308 MISCELLANEOUS STUDIES 
 
 PLATE 23 
 
 The Eakly Type 
 
 Fijr. 3. April S, liiOi). The single progeny of \VG9-C9 (Snowflake); arrange- 
 ment as in figure 1. 
 
 Fig. 4. April 9, 1909. The single progeny of WG9-C10 (heterozygous early). 
 Warm-house plants partly at right of stake in lower row; arrangement other- 
 wise as in figure 3. Compare with figure 3, house by house. 
 
 [1661 
 
MVTATIOS IN M.ITTIIIIIL.I 
 
 :',m 
 
 Fisr. 3 
 
 Fig. 4 
 
 I FROST I PLATE 23 
 
310 MISCELLANEOUS STUDIES 
 
 PLATE 24 
 
 The Early Type 
 
 Fig. 5. July 19, 1911. Lots 1 to 10, with lots 11 to 1-t mostly in sight at 
 the right. Odd-numbered lot in nearer (west) half of each row. 
 
 Fig. 6. July 19, 1911. Lots 19 to 28, with lots 15 to 18 mostly in sight at 
 the left. 
 
 [16S1 
 
ilCrATIOX IS MATTHIOLA 
 
 311 
 
 Fig. 5 
 
 F\'J. 6 
 
 1 FROST 1 PLATE 24 
 
312 MISCELLANEOUS STUDIES 
 
 PLATE 25 
 
 The Smooth-leaved Type 
 
 Fig. 7. April 27, 1909. Smooth-leaved apparent mutants. Compare with 
 figures 3 and 4 as to earliness, noting the difference in date. 
 
 Fig. 8. May 29, 1914. Progeny of a smooth-leaved parent. Plant at right 
 Snowflake single, the others smooth. 
 
 1170] 
 
MriAiiox /.v M.irrii iDi.A 
 
 Fisr. 7 
 
 Fig. 8 
 
 FROST 1 PLATE 25 
 
3U MISCELLANEOUS STUDIES 
 
 PLATE 26 
 
 The Smooth-leaved Type 
 
 Fig. 9. June 28, 1913. Progeny of a smooth-leaved parent. Sniodtli single 
 at left, Snowflake double at right. 
 
 Fig. 10. Same date and parent as with figure 9. From left to right: Snow- 
 flake double (also shown in figure 9), Snowflake single, smooth double. 
 
 [IT'] 
 
Ml T Alios I.\ MATTlllOLA 
 
 315 
 
 Fig. 9 
 
 Via. 10 
 
 FROST I PLATE 26 
 
316 MISCELLAXEOrS STUDIES 
 
 PLATE 27 
 
 The Large-leaved Type 
 
 Fig. 11. August 29, 1S14. Progeny of a large-leaved parent (28a), near 
 the close of the hot Riverside summer. From left to right: large single, large 
 double, Snowflake single (two, the first injured by aphids). 
 
 11741 
 
UVTATION IX MATTHIOLA 
 
 317 
 
 r FROST 1 PLATE 27 
 
318 MISCELLANEOVS STUDIES 
 
 PLATE 28 
 
 The Large-leaved Type 
 
 Fig. 12. July 8, 1916. Progeny of a large-leaved parent. Middle plant 
 Suowflake; the rest large; all single. 
 
 Fig. 13. July 8, 1916. Progeny of a large-leaved parent, more than a month 
 older than those shown in figure 12. From left to right: large double,' Snow- 
 flake double, large single. 
 
 (1761 
 
MITATIOX IX MATTHIOLA 
 
 319 
 
 Fiff. \2 
 
 
 !^B^^^^^Hi«> 
 
 
 p^^^ 
 
 Fig. 13 
 
 I FROST I PLATE 2£ 
 
320 MISCELLANEOUS STUDIES 
 
 PLATE 29 
 
 The Crenate-i.eaved Type 
 
 Fig. 14. April 6, 1909. Crenate-leaved apparent mutants. Note tlie varia- 
 tion in leaf serration, and especially the slightness of the serration (or crenation) 
 with the one cool-house plant (below). 
 
 Fig. 1.5. April 14, 1911. Progeny of a crenate-leaved parent, grown in a 
 cool greenhouse. The first two plants at the right are Snowflake, the rest 
 crenate. 
 
 [178] 
 
MUTATION L\ M ATTII lOL.I 
 
 321 
 
 Fitr. 14 
 
 Fig. 15 
 
 FROST 1 PLATE 29 
 
322 MISCELLANEOUS STUDIES 
 
 PLATE 30 
 
 The CRE.VATE-i.EAVEn Type 
 
 Fig. 16. July 8, 1916. Progeny of a crenate-leaved parent. Prom left to 
 right: crenate single (two), crenate double, Snowflake double. 
 
 Pig. 17. July 8, 1916. Snowflake X erenate-leaved, P,. From left to right: 
 smooth, Snowflake single, crenate double (two). 
 
 [1801 
 
MUTAI'IUX IX MATTIIIOLA 
 
 323 
 
 Fie. 16 
 
 Fig. 17 
 
 FROST 1 PLATE 30 
 
324 MISCELLANEOUS STUDIES 
 
 PLATE 31 
 
 The Slender Type 
 
 Fig. IS. April 27, 1909. Miscellaneous aberrant individuals, with two 
 typical Snowflake plants (third from the left above, second from the left 
 below). In ujiper row: second from left, narrow double; second from right, 
 slender double. In lower row at left, slender single (2.5b). 
 
 Fig. 19. April 14, 1911. Progeny of a slender jiarent (2.5b). Two at the 
 right Snowflake, the rest slender. 
 
 11821 
 
MirATlUS IX MATTIIIULA 
 
 325 
 
 &►.-<*• *-»Mi.*JBi. 
 
 
 
 
 'dC-^S 'dtnX^oW^^ ^* 
 
 5/A 
 
 
 Jl 
 
 
 Fi". 18 
 
 IS 
 
 
 
 
 1 
 
 -:cA9, 
 
 
 _B " 1 ,^ ll- 
 
 ^Mk::=» t,.._^li_-. — il r. 
 
 
 
 Fi". Ill 
 
 FROST 1 PLATE 31 
 
326 MISCELLANEOUS STVDIES 
 
 PLATE 32 
 
 The Slender Type 
 
 Fig. 20. June 3, 1914. Progeny of slender jiarents. From left to right: 
 slender single, slender double, Snowflake double. 
 
 Fig. 21. July 7, 1916. Snowflake X slender, F,. Middle plant Snowflake; 
 tlie otliers slender; all single. 
 
 11841 
 
ilVTATlOX IX MATTIIIOLA 
 
 327 
 
 Ficr. 20 
 
 Fig. 21 
 
 I FROST 1 PLATE 32 
 
328 MISCELLANEOUS STUDIES 
 
 PLATE 33 
 
 The Narrow-leaved Type 
 
 Fig. 22. April 13, 1011. Narrow-leaved apparent mutants. 
 
 Fig. 23. June 3, 1914. A narrow-leaved apparent mutant among progeny 
 of a erenate-Ieaved parent. From left to right: narrow double, crenate single 
 (two). 
 
 [186] 
 
Mrr.iriox ix matthiola 
 
 312!) 
 
 [ FROST ] PLATE 33 
 
330 MISCELLANEOVS STUDIES 
 
 PLATE 3-4 
 
 The Narrow-leaved and Small-smooth-leaved Types 
 
 Fig. 24. June 28, 1915. A narrow-leaved apparent mutant among F, progeny 
 from Snowflake X slender. Narrow double at left; the rest SnovviJake single. 
 
 Fig. 25. April 14, 1911. Miscellaneous aberrant plants, some being apparent 
 mutants. From the left: first and fifth small-sniootli, tliird stout dwarf, seventh 
 slender. See text. 
 
 11881 
 
Ml T.iriOX l.\ M.ITTIIIOLA 
 
 331 
 
 Fig. 24 
 
 Fig. 25 
 
 FROST I PLATE 34 
 
332 MISCELLANEOUS STUDIES 
 
 PLATE 35 
 
 The Nakrow-d.vek-leaved T\pe 
 
 Fig. 2G. June 3, 1914. A narrow-dark-leaved apparent mutant among 
 progeny of a narrow-leaved parent. Third plant from left narrow-dark single; 
 the other three Snowflake double. 
 
 Fig. 27. June 28, 1915. Progeny of a "small-convex-leaved(f ) " parent 
 (27a). From left to right: narrow-dark single, Snowflake double, smooth single. 
 
 (1901 
 
ilLTATlO.X I.\ MAlTillOLA 
 
 333 
 
 Kiir. -M 
 
 
 V\s. 
 
 FROST I PLATE 35 
 
OCEAN TEMPERATURES 
 
 BY 
 
 GEORGE F. McEWEN 
 
OCEAN TEMPERATURES, THEIR RELATION 
 
 TO SOLAR RADIATION AND 
 
 OCEANIC CIRCULATION 
 
 QUANTITATIVE COMPARISONS 
 OF CERTAIN EMPIRICAL RE- 
 SULTS WITH THOSE DEDUCED 
 BY PRINCIPLES AND METHODS 
 OF MATHEMATICAL PHYSICS 
 
 GEORGE F. McEWEN 
 
 Oceanographer of the Scripps Institution for Biological Research 
 of the Uni^'ersity of California 
 
CONTENTS 
 
 Introduction. The place of mathematical methods in researches on oceano- 
 graphic problems 337 
 
 Solar radiation and surface temperature, assuming the average rate of flow of 
 the water to be zero 338 
 
 Preliminary discussion, and statement of certain generally accepted 
 conclusions as to the process by which the water gains and loses heat 338 
 
 Statement of assumptions, mathematical formulation of the problem and 
 its solution 339 
 
 Determination of the numerical values of the constants in the solution.... 344 
 
 Observed and theoretical lag of temperature maxima and minima behind 
 the radiation ma.xima and minima; comparison of computed and 
 observed normal temperatures 348 
 
 Numerical estimates of the coefficient of absorption of solar radiation 
 in sea water 350 
 
 Deduction of the change in surface temperature produced by a horizontal flow 
 of water 352 
 
 Preliminary discussion, statement of assumptions, and mathematical 
 formulation of the problem 352 
 
 Solution for the case in which the flow is constant 356 
 
 Solution for the case in which the flow is a periodic function of the time ,,. 359 
 
 Solution for the particular case in which the time interval is so small 
 that the solar radiation may be assumed to depend only upon the 
 latitude 360 
 
 The rate of horizontal flow in the North Pacific off the California coast from 
 latitude 40° N to 30° N and in the North Atlantic off the west coast of Africa 
 from latitude 30° N to 20° N 362 
 
 The rate of flow off the coast of California deduced from surface tem- 
 peratures 362 
 
 The rate of flow deduced from temperature data compared with that ex- 
 pected from the empirically ascertained relation of winds to currents 
 and with direct observations on currents 363 
 
 The surface current prevailing for a short time interval near the north- 
 west coast of Africa, estimated from surface temperatures, com- 
 pared with direct observations and with results deduced from the 
 empirically ascertained relation of winds to currents... 364 
 
336 MISCELLANEOUS STUDIES 
 
 The relation of the temperature to time, depth and rate of vertical flow in the 
 depth interval from 40 to 600 meters 367 
 
 Statement of assumptions and mathematical formulation of the problem 367 
 
 Solution for the case in which the vertical flow is constant 369 
 
 Solution for the case in which the vertical flow is a periodic function of 
 the time 372 
 
 Numerical values of the constants in the solution, determined from tem- 
 perature observations in the Pacific near San Diego 374 
 
 Comparison of theoretical and observed monthly temperatures at depths 
 from 40 to 600 meters in the San Diego region 382 
 
 Solution of the problem of temperature reduction due to upwelling, with 
 numerical appHcations relative to the 40 meter level in the San Diego 
 region 382 
 
 Deduction of the change in surface temperatures due to a vertical flow of water 
 near the surface 388 
 
 Statement of assumptions and mathematical formulation of the problem 
 and solution for the case in which the flow is constant 388 
 
 Solution for the case in which the flow is a periodic function of the time. . 391 
 
 Theoretical reduction of the surface temperature for each month in the San 
 Diego region due to upwelling, and comparison with observations 393 
 
 Deductions relative to oceanic circulation in the San Diego region, based on 
 Ekman's hydrodynamical theory 397 
 
 Deduction of the upwelling velocity in the San Diego region from the observed 
 relation of salinity to depth, and comparLson with that deduced from tem- 
 perature data 406 
 
 Conclusion 415 
 
 Literature cited 419-421 
 
OCEAN TEMPEltATVnES 337 
 
 INTRODUCTION 
 
 The Place op IMathematical Methods in Researches on 
 oceanographic problems 
 
 The present paper deals with the formulation and solution of 
 several quantitative problems suggested by data on ocean winds, 
 temperatures, and circulation. Before formulating these problems 
 a brief general discussion of the place of mathematical methods in 
 oceanographie researches is given. 
 
 The process of testing physical laws in the laboratory is greatly 
 facilitated by devising appropriate experiments in which the variables 
 are largely under the control of the investigator. Even under these 
 favorable conditions the actual phenomena are too complex for de- 
 tailed representation in a mathematical formula, and an appropriate 
 simplification by abstraction is required to formulate problem.s that 
 are amenable to mathematical treatment. This is true in a much 
 greater degree of the more complex phenomena occurring in nature ; 
 yet a rigorous mathematical treatment of natural problems capable of 
 yielding results in agreement with oliservations, while in general more 
 difficult, is necessary and fully justifies the increasing attention being 
 given to terrestrial and cosmic physics. 
 
 The actual phenomena of heating and cooling of the water in the 
 ocean are far too complex to be considered in detail. Therefore, in 
 order to apply rigorous mathematical reasoning to these phenomena, it 
 is necessary to devise a comparatively simple ideal system which would 
 behave in essentially the same way as the actual one with reference to 
 the observations in question. Certain problems can then be formulated 
 definitely in such a way as to permit of the precise calculation of 
 results, the comparison of which with observations tests the practical 
 value of the abstract system. 
 
 It is fortunate for the problems considered in this paper that the 
 proper choice of the simple assumptions needed in devising the ideal 
 system is facilitated hy certain general results of numeroi;s and extended 
 ocean as well as laboratorij observations. An abstract system foinided 
 on such assumptions would in general agree much better with the 
 conditions in nature than one in which the assumptions were hypo- 
 thetical or carried over from some other field. Evidently deductions 
 from any group of simple assumptions cannot have the same degree 
 
338 MISCELLANEOUS STUDIES 
 
 of certainty as direct observations, since the ideal system cannot con- 
 form accurately to all the details of the phenomena of the actual one. 
 However, if in this way a logical and reasonably accurate description 
 of a wide range of physical quantities is ol)taincd there is good reason 
 to believe that deductions or predictions relative to ciuantities not yet 
 observed will be in agreement with the facts. This is especially true 
 if two or more lines of reasoning converge to the same conclusion. 
 
 WJien it is impossible or impracticable to make tlic appropriate 
 direct observations the theoretical resuUs must be regarded as the 
 best estimates, even though it is not impossible that future observations 
 may shoxv important deviations from theory. Finally, while the exist- 
 ing observations may be logically described by means of the ideal 
 system, and deductions based on it, extended results reached by apply- 
 ing purely deductive methods to the ideal system are not substitutes 
 for a correspondingly extended series of new observations. The neces- 
 sity for making observations will always exist. 
 
 Solar Radi.\tion and Surface Temperature, Assuming the 
 
 Average Rate of Flow^ op the Water to be Zero 
 
 Preliminary discussion, and .statement of certain generally accepted 
 
 conclusions as to the way in which the water gains and loses heat. 
 
 In order to have a basis for estimating the effect of circulation on 
 ocean temperatures it is necessary to work oiit quantitatively the rate 
 at wliicli tlic heat of tlie water is gained and lost luulcr the more 
 sim{)le condition of no flow. This will be done by devising an ideal 
 ocean, based on assumptions agreeing as nearly as possilile with the 
 following conclusions which are founded on numerous and widely 
 extended ocean observations. 
 
 1. The primary source of heat is the radiant energy of the sun, 
 both direct and diffuse, that penetrates the water (Murray. 1912, p. 
 225, Gehrke, 1910, p. fi7. nclland-IIaiiscn. 1911-12, pp. (i-1-66). 
 
 2. Absorption of this radiation directly heats the water in the 
 upper layers (Nanseu, 1913, pp. 21-22), and only a small fraction of 
 this radiant energy penetrates below 25 meters (Kriimmel, 1907, pp. 
 253-270, Helland-Hansen, 1911-12. pp. 65-fi8, and Knott, 1903-05). 
 
 3. There is always a complex vertical circulation (Helland-IIansen, 
 1911-12, p. 68. Gehrke, 1910, p. 68, Nansen, 1913, p. 21, ilurray. 
 1912, p. 226) due to a lack of balance of the many forces acting on 
 the water particles. The resultant vertical flow through a finite sec- 
 tion due to this motion may be very small and may be either u|)ward 
 
OCEAN TliMl'EKATVliKS 339 
 
 or downward. That is, at the same time, some portions of the water 
 are moving upward and other.s downward, thus tending to mix up the 
 water at diiferent levels. In this problem the resultant of the upward 
 and downward flow will be assumed to be zero. 
 
 4. This "mixing process" is most intense in the layers nearest 
 the surface, owing to wave motion and other surface disturbances due 
 to wind, but is present in some degree at all depths (Gehrke, 1909, 
 p. 12, Murray, 1898, p. 127). 
 
 5. The amount of heat transferred from one level to another by 
 conduction through the water is a negligible fraction of that carried 
 by the water particles themselves as a result of the mixing process 
 (Gehrke. 1909, p. 12). 
 
 6. The mean animal rate of change of temperature with respect 
 to latitude is practically independent of the depth within the upper 
 hundred meters. This is revealed by a study of the average tempera- 
 tures of the North Pacitie, tabulated with respect to latitude, longitude, 
 and depth (Schott, 1910. p. 14). 
 
 7. At the time of year when the surface temperature is a minimum 
 there is practically no variation of the temperature with respei^t to 
 depth in the upper thirty meters (McEwen, 1916, p. 272). 
 
 Statement of assumplions: mathematical formulation of llu problem 
 
 and its solution. 
 Let K^f^(L,t) equal Q,. the amount of radiant energy available 
 per montli per unit area of horizontal surface at the latitude L and 
 time t, where K^ is proportional to the solar constant and /, {L,t) 
 is a function of the latitude L and the time t. Let Q equal the amount 
 of radiant energy used directly in heating the water, that is, the 
 amount passing into the water. Also let 
 
 y equal the distance in meters from the surface of the water, the 
 
 positive direction being downwards, 
 or equal the specific heat of sea water per unit volume, 
 6 equal the temperature, centigrade. 
 
 t equal the time, the unit being 1 month, and t equal 1 for January, 
 ^1 equal a temperature assumed to depend only on the latitude and 
 
 depth y, and 
 pi equal the average transmission coefficient of sea water for the 
 solar radiation, that is, tlie proportion of radiation at any level 
 that passes through unit thickness of water measured from that 
 level. 
 
340 
 
 MISCELLANEOUS STUDIES 
 
 Since the solar radiation consists of a series of waves of varjing 
 length, each having a different coeffieient of transmission (Murray, 
 1912, p. 248, Kriimmel, 1907, p. 263), the use of a single average value 
 is only a simple approximation to the true relation. 
 
 No analysis will be attempted of the complex way in which the 
 heat in an element of volume, specified by given values of y and L, 
 is lost. This loss depends upon evaporation at the surface and on 
 the mixing process at all depths, and the rate of evaporation increases 
 as the temperature increases. Also heat tends to flow from regions 
 
 Fig 
 
 of high temperature to those of low temperature (Gehrke, 1910. p. 68). 
 It seems reasonable to suppose, therefore, that the rate of loss would be 
 greater, the greater the temperature. 
 
 Although the precise manner in which the rate of loss of heat 
 depends upon the temperature is not known, some definite form of 
 relation must be assumed in order to formulate the temperature 
 problem mathematically. For simplicity assume the rate of loss at 
 any depth to be proportional to {6 — ^j ) at that depth, where 6^ is a 
 function of the depth and latitude only. Consider now the time rate 
 at which heat is gained and lost in a given rectangular element of 
 volume of unit cross section and thickness dy whose upper sxirface is 
 at the depth y (fig. 1). 
 
 The rate of change of heat in this volume element is evidently 
 
 d[(T(dy)e] 
 dt 
 
 -o{,dy)~ 
 
 (1) 
 
 since the volume specific heat multiplied by the volume of the element 
 equals the change in the amount of heat per degree change of tem- 
 peratui-e. 
 
OCEAN TEMPEEATVEES 341 
 
 The rate of gain of heat in this element of volume due to the 
 absorption of solar radiation equals the difference between the rate 
 at which the radiant energy passes in through its upper surface and 
 out through its lower surface. At the upper surface the rate is 
 ^/8i» and at the lower surface it is Q^^"*''". Therefore the rate of 
 gain diie to absorbed radiation is 
 
 (?;3/ — (?^/*''« = (?/?/ (l — A"^) = — <?(loge/?,)/?A?:V (2) 
 since 
 
 1 — /J/^^i— [1 +(iog/3t)'?.'/] ==— (iog/8.)f?y. 
 
 The rate of loss of heat will be assumed to be k(x{9 — ^Jr/y where 
 k is a function of y only. 
 
 Equating the rate of change of heat in the element to the rate of 
 gain from solar radiation less the rate of loss, we have the following 
 differential equation 
 
 (3) 
 
 (41 
 after division by achj. 
 
 Let L^Li-\- X where L^ is a standard latitude chosen arbitrarily 
 and X is the distance in degrees from this position, x is positive for 
 latitudes higher than L, and negative for lower latitudes. The function 
 fj{L,t), (p. 339). then becomes f{x,t), which expresses the way in 
 which the radiation varies with respect to latitude and time. The 
 precise form ot f{x,t) is unknown; however, estimates of the amount 
 of radiant energy available at the earth's surface made by Angot 
 (Hann, 1915, p. 40) can be closely approximated to within a ten-degree 
 interval of latitude by an expression of the form 
 
 Qi ^= /I'l [ (di -\- a^x) cos at -{■ a^x + 1] (5) 
 
 TT . . . 
 
 where a = — , and the coefficient of cos at is negative. 
 
 n 
 
 Assuming the amount of energy Q that enters the water to be pro- 
 portional to the amount available ^i 
 
 Q = K[ {a^ + a^_x) cos at + a.,x + 1] (6) 
 
 .(cZy)^=. 
 
 —Q(\ogli,)py(hj — -ka{e- 
 
 -6,)dy 
 
 diich becomes 
 
 M 
 dt 
 
 CT 
 
 -6,) 
 
342 MISCELLANEOUS STUDIES 
 
 where the constant K s K^, since the amount of energy used cannot 
 exceed the amount available. Equation (4) then becomes 
 
 66 Eilogp,)^,". 
 
 [(a, + a,z) cos at -\- a.,x + 1] —k{e — e^) (7) 
 K log 13, Kb, 
 
 dt 
 Let 
 
 K Incr R K h 
 
 B (8) 
 
 where B is a constant and 
 
 p, = e-^- (9) 
 
 where h, is the absorption coefficient (Kriimmel, 1907, p. 263). 
 Equation (7) then reduces to the oi'dinary linear differential equation 
 
 flft 
 
 — j^l-e=B[ia,+ a,x) cos at + a,x + 1] e-'« + M, (10) 
 
 Therefore 
 
 e=e-"|e-''"/c'■■'B(a,+ff,,r)cosa/rff+f-''^"jV*■'(a,.r+l+/,^(9Jd«+F(.r,.v)| 
 
 (11) 
 where F(.r,y) i.s an arbitrary function of x and .//. Integrating 
 equation (11) gives 
 
 6=Be-^^y{a,-^a„x) ~-, 1 p-^-^ \r6,+e-^<F{x,y) 
 
 a"-|-/i'" K 
 
 (12) 
 which can be readily transformed into 
 
 ^ 5(a, +a„.r)e-"'" , , , , .„ ( Ba,x , B] , „ ^qi 
 
 fl= VI '- eos (a# — £)+<'-''-» -^ —r^+ — ^ + ^1 (13) 
 
 whore 
 
 tan£ = - (14) 
 
 and only the periodic part of the integral is retained. 
 
 If 6, is assumed to be independent of .r the latitude gradient (j 
 of the mean annual temperature is, from equation (13"). 
 
 g = e-i>^v{^\ (15) 
 
 Therefore, since g is independent of \j (sixth statement, p. 339) 
 
 k^l;€-"^v (16) 
 
 where k, i.s a constant. That is, 
 
OCEAN TEAirKBATPEES 343 
 
 Corresponding to the time of year <„ when the surface temperature 
 has the minimum vahie 6g equation (13) becomes 
 
 Since the eoelficient of cos (at — t) is negative (see p. 341) cos {at^ — e) 
 must equal plus 1; therefore from the seventh general statement 
 (p. 339) 
 
 B(a,±a^^^ (19) 
 
 where 6, is a constant. 
 
 Making use of the results just found equation (13) becomes 
 
 e = m±M^[cO. iat-.)-l] +^ + f + ^3 (20) 
 
 where 
 
 ^--=i=v^^ ^''^ 
 
 The small variation in temperature with respect to depth in the upper 
 six meters indicates that these upper water layers are very thoroi;ghly 
 mixed (]Miehael and ]McEwen, 1915. 191(i). Accordingly temperatures 
 in this 6 meter interval will be computed by using 3 meters, the 
 average value of the depth. That is, (y — 3) will be substituted for 
 1/ when the depth exceeds 6 meters and the constant value (6 — 3)= 3 
 for all depths between and 6 meters. From equation (20) it follows 
 that at the time of minimum temperature the temperature is inde- 
 pendent of the depth y in accordance with the seventh general state- 
 ment (p. 339). But the latitude gradient which is the part of the 
 coefficient of x in equation (20) not involving the time is 
 
 Ba„c-'''y , Ba, 
 
 This is not in accordance with the observed fact that the latitude 
 gradient is independent of ;/.■ but the relative error will depend ni)on 
 the ratio of the first term to the second term, and may not be important. 
 As will be seen later (p. 345) it proves to be a negligible error if we 
 add to 6^ of equation (19) the term Ba^xe-^ 
 
 Therefore equation (20) with this modification and the use of (y — 3) 
 for the depth ij gives the approximate form of the relation between 
 
344 MISCELLANEOUS STUDIES 
 
 temperature, time, depth, and latitude, in accordance with the assump- 
 tions and general results of temperature observations already stated. 
 The modified equation is 
 
 "— ^ 1^ - ' .[cos (o^— e)— l]+^7-^.T M ' = ' " 
 
 (22) 
 Also from equations (19) and (20) 
 
 B Ea r B(a -4- a T')e-6i(!/-3) 
 
 (g— gJ^^ + £^ + -^li^l± gg^l^ COs(a/-c) (23) 
 
 '>'i ''1 ^/a--\-k- 
 
 where 
 
 a a 
 
 tan e = 
 
 A- k^e-o^'v-^' 
 
 Determination of the numerical values of the constants in the solution. 
 
 In applying mathematical methods to physical problems the 
 functional relation between the variables involves certain constants 
 which must be determined from observed values of these variables. 
 The constants in equation (22) are 
 
 flj flo a^ a A-j O.f B and 6,. 
 
 Also from equation (8) we have 2? = — —. The first four constants 
 are found by fitting the function 
 
 J^i [ (^1 + O"!^) cos at -\- a^x -\- 1] 
 
 (equation 5) to the estimated values of solar radiation, as given for 
 example by Angot's tables (Hann, 1915, p. 40). The next three 
 require observations on temperature. For example, they can be found 
 from the observed values of the normal annual range of temperature 
 at a .series of latitudes and from the mean annual temperature at 
 these latitudes. 
 
 The coefficient of absorption h^ can be estimated from direct meas- 
 urements of the intensity of radiation in the ocean at different depths 
 (Grein, 191.3), or from observations on water samples taken to the 
 laboratory (Petersen, 1912, p. 39). Also an indirect estimate can be 
 made from temperature and .solar radiation data (pp. 350-352) by 
 means of equation (22). However, in the problem of .surface tem- 
 perature in which the vertical flow is neglected the value of b^ is not 
 required. 
 
OCEAN TEMPEEATVSES 
 
 345 
 
 Choosing 30° N for tlie standard latitude L^. tlie data on solar 
 radiation taken from Angot's tables and based on the atmospheric 
 transmission coefficient of 0.60 gives the following values of the first 
 foiir constants : 
 
 a., : 
 
 -.0128 a3 = — .0159, a=-= 
 
 6 
 
 0.523. 
 
 ttj = — .47 
 
 A later and more accurate estimate of the radiant energy' available at 
 the ocean's surface (Schmidt, 1915, p. 121) gives smaller values, on 
 the average, than those of Angot; also instead of the value — .0159 
 
 53 
 
 for ^3 his re.sults give — "C -in, ^ — .0244, which will be iised in 
 in this paper. 
 
 Kriimmel (1907. p. 413) gives the observed normal mean annual 
 temperature and the normal annual temperature range for the surface 
 of the ocean corresponding to a series of latitudes. These values are 
 in part presented in the following table. 
 
 Table 1 
 
 Observed values of the normal mean annual temperature and the normal annual 
 range for the surface of the ocean at latitudes 10° N to SO N 
 
 North latitude 
 
 10° 
 
 20° 
 
 30° 
 
 40° 
 
 50° 
 
 
 
 Mean Annual temperature . 
 
 Annual range 
 
 Half Annual range 
 
 27.2 
 2.2 
 11 
 
 2.5.4 
 3 6 
 
 1.8 
 
 21.3 
 6.7 
 3. 35 
 
 14.1 
 
 10.2 
 
 5.1 
 
 7 9 
 
 8.4 
 4.2 
 
 X 
 
 -20° 
 
 -10° 
 
 00° 
 
 + 10° 
 
 +20° 
 
 
 From the tabulated values it follows that the gradient of the half 
 range is .175 and the gradient of the mean annual temperature is .72 
 from latitudes 30° N to 40° N. Let m^ equal c'"'^ where 3 is the mean 
 depth of the upper homogeneous layer. From these values and 
 equation (22) we have 
 
 Ba^yriy 
 
 
 Va" + A-fmi" 
 
 = —.175 
 
 3 35_Ba^r _^/_fc^y-| 
 
 and 
 
 Ba, 
 
 = —.72 
 
 (24) 
 
 (25) 
 
 (26) 
 
346 MISCELLANEOUS STUDIES 
 
 in which the last members of equations (24) and (25) are approxi- 
 
 mately correr-t if l~ — -) is small. 
 
 Substituting the numerical values of a^ a. a, and a from page 345 
 we have 
 
 From equations (2-4) and (27) neglecting -(— ^ — -) we have the first 
 
 approximation 
 
 29.5 AvHi X 1-910 X (—-47) =—3.35 (28) 
 
 from which 
 
 Therefore 
 
 and 
 
 ^''■»'-^ 29.5XL910X.47 =-^^^^- 
 
 1 _^ 1/2 (A!!liy = 1.0292 
 k^m^ = 1.0292 X .1265 = .1302 (29) 
 
 ^ ^ 29.5 X. 1302 ^3j4 ^3^^^ 
 
 Similarly from equations (25) and (27) we have 
 29.5 k,tn, X 1-910 X -0128 = .175 
 
 from which 
 
 Therefore 
 
 7. „, -^^ 9407 
 
 '^~ 29.5 X 1-190 X .0128 
 
 and 
 
 1 + 1/2 (^hUlij '=1.1075 
 k,m, = 1.1075 X .2427 = .2688 (31) 
 
 g ^ 29.5 X .2688 ^ 7.927 .^^^ 
 
 The difference between the values of B and k^ found by th&se two 
 methods of computation is due to the fact that the ratio between the 
 
OCEAN TEMPERATVRES 347 
 
 half i-ange of radiation at the standard latitude to the gradient of 
 the half range of the radiation is not quite consistent with the ratio 
 of the half range of temperature at that latitude to its gradient. 
 
 It seemed best to take the average of the two values and to apply 
 corrections to a^ and a„, making them agree with the temperature 
 data; then the difference between the corrected values a\ and a\ and 
 the original values will indicate the magnitude of this discrepancy 
 between temperature data and solar radiation data. Accordingly 
 
 12(M^+I:^)=.^ = B = ^ (33) 
 
 and 
 
 1/2 (.1302 + .2688) = .1995 = k^m, = .20. (34) 
 
 From equations (2-1) and (25) the new values a\ and a'„ of a^ 
 and a„ are 
 
 3.35V.274+.04^-3.35V.314__31S ^^.^ 
 
 ' Bm^ 5.9 
 
 and 
 
 -.175V.314__Q^gg_ (36) 
 
 5.9 
 
 Substituting these niunerical valuations in equation (22) and deter- 
 mining ^3 from the observed surface temperature 21?3 when x equals 
 we have, expressing the angle in degrees 
 
 — (—.318 — .0166a-) c-^^^y-^'> 
 
 e= I" [cos (30f — €)— 1] 
 
 y.274 + .04(9!) 
 
 _. 72. -i£i£mi+ 17.95 (37) 
 
 V.274 + .04 
 
 where 
 
 tan e = 2.62 ,^\^ . (38) 
 
 For the surface temperature, put y ^= 6 and we have 
 _5.9(-.318-.0166x) [,,, ^,ot -e9y-l]-.12.- '-'^-''''^^ 
 
 .56 L-" V""- -/ -J •■— 5g 
 
 (39 
 -f 17.95 = — (3.35 + . 175x) cos (30< — 69) ° + 21.30 — .72x. 
 
348 MISCELLANEOUS STUDIES 
 
 In the same way the following values of the constants for the 
 latitude interval from 20° N to 30° N were determined: 
 
 «', = — .343, a'„ = — .0159, ag^— .0161, ;,-im,= .217, 
 
 — i = the latitude gradient = — .41, 
 
 — .155 = the latitude gradient of the half range, 
 „ 5.535 , . .523 m, m^ 
 
 Hli 217 6-*''"-=" g-lh'U'^i 
 
 Substituting these numerical values in equation (22) gives for the 
 siirface temperature 
 
 ^ = — (3.35 + .155x) cos (30< — 67.5)° + 21.30 — .41.r (40) 
 
 Observed and theoretical lag of temperature maxima and minima 
 
 behind the radiation maxima and minima. Comparison 
 
 of computed and observed normal temperatures. 
 
 The value of «=69°, corresponding to the latitude interval 30° N 
 to 40° N, was deduced from theory; and since 30° corresponds to 
 
 69 
 one month, 69° corresponds to —^2.3 months, the theoretical lag 
 
 oU 
 
 J. ,, ( maximum ) , , , i • t ,i ( maximum ] j- j.- 
 
 or the ^ • • f temperature behuid the ^ • ■ radiation. 
 
 ( minimum ) '■ ( minimum \ 
 
 The theoretical time of the maximum temperature is therefore 
 8.3, or about halfway between August and September ; while the 
 theoretical time of the minimum temperature is 2.3, or about halfway 
 between February and March. According to Kriimmel (1907, p. 407) 
 from numerous and extended oceanic observations the average time 
 of the lowest temperature is February (t = 2) and that of the highest 
 is August it = 8). 
 
 Again, from the three curves (Makaroff, 1894, pi. 26) giving the 
 mean monthly surface temperature observed in the North Pacific 
 between latitudes 30° N and 45° N the minimum temperature occurred 
 when 
 
 t = 1.8, 3.0 and 2.5 
 
 and the maximum temperature occurred when 
 
 t = 8.1, 8.3 and 7.8 
 
 respectively. The average of the above values of t is 2.4 for the 
 minimum and 8.1 for the maximum. Thus the predicted value of the 
 
OCEAN TEMPEBATISES 
 
 349 
 
 lag agrees very closely with the obseryed value. Since this value was 
 computed from that of tlie period of the temperature change, which 
 is accurately l<nown to be twelve months, and from the foregoing 
 determination of the value of k^nii this agreement between theory and 
 observation affords strong evidence in favor of the reliability of the 
 vahie .2 adopted for }t\i)ij. which is an important constant in the 
 investigation of ocean currents presented later. 
 
 From numerous surface temperature observations in the Pacific at 
 long. 173° W, lat. 20° N (Puis, 1895, pis. 1-4), off Madeira in the 
 Atlantic, lat. 32° 30' N (Kriimmel, 1907, p. 407) and off Yokohama 
 and at long. 140° W in the Pacific, lat. 35° N (Kriimmel, 1907, p. 408; 
 Thorade, 1909. pis. 1-3) it was found that the mean annual tempera- 
 tures agreed well with the normal values for the latitude. Therefore 
 there is good reason to suppose that the condition giving rise to the 
 temperatures at these places approximates closely to the normal con- 
 dition. Thus a comparison between the theoretical monthly tempera- 
 tures given by equations (39) and (40) with the observed vahies will 
 give a still more detailed test of the theory (table 2). 
 
 Table 2 
 Theoretical and ohserved normal surface temperatures at a series of latitudes 
 
 Latitude 20° N 
 
 21° 18' N 
 
 30°N 
 
 t 
 
 o 
 
 18 °7 
 18.0 
 18.2 
 
 19 2 
 20.8 
 22.6 
 23.9 
 24.6 
 24.4 
 23 3 
 21.6 
 20.3 
 
 32° 30' N 
 
 35° N 
 
 40°N 
 
 j3 
 
 "c 
 
 O 
 1 
 
 2 
 3 
 
 4 
 
 5 
 
 6 
 
 7 
 
 8 
 
 9 
 
 10 
 
 11 
 
 12 
 
 s 
 
 H 
 24 °0 
 23.6 
 24.0 
 24.7 
 25.6 
 26.5 
 26.8 
 27.2 
 26.8 
 26.1 
 25.2 
 24.3 
 
 -a 
 (1) 
 
 > 
 
 tD 
 
 73 
 J2 
 O 
 
 24 °1 
 24.2 
 24.7 
 24.7 
 24.9 
 25.9 
 26.3 
 26.7 
 26.7 
 26.4 
 2.5.8 
 24.8 
 
 i 
 
 Q 
 
 + °1 
 + .6 
 + .7 
 + .0 
 — .7 
 -.6 
 -.5 
 -.5 
 -.1 
 + .3 
 + .6 
 + .5 
 
 g 
 
 23 °3 
 22.9 
 23.3 
 24.1 
 25.2 
 26.1 
 26.5 
 26.9 
 26.5 
 25.7 
 24.6 
 23.7 
 
 1 
 1 
 ? 
 
 21 °6 
 
 20.8 
 21.6 
 22.1 
 23.0 
 24.4 
 25.3 
 25.8 
 2.5.1 
 24.4 
 23 
 21.1 
 
 c 
 
 a 
 
 -1°7 
 -2.1 
 -1.7 
 -2.0 
 -2.2 
 -1.7 
 -1.2 
 -11 
 -1.4 
 -13 
 -1.6 
 -2.6 
 
 t 
 
 g 
 
 H 
 16 °6 
 15.7 
 16 
 17.1 
 18.9 
 20.9 
 22.4 
 23.2 
 23.1 
 21 9 
 20.1 
 18,1 
 
 1 
 1 
 
 O 
 18 °0 
 17.2 
 17.1 
 17.9 
 18.7 
 20.3 
 21.8 
 22.9 
 23.2 
 22.1 
 20.6 
 19.0 
 
 S 
 c 
 
 5 
 
 + 1^4 
 + 1.5 
 + 1.1 
 +0.8 
 -0.2 
 -0 6 
 -0.6 
 -0.3 
 +0.1 
 +0.2 
 +0.5 
 +0.9 
 
 1 
 
 14 °4 
 13.6 
 13.8 
 15.1 
 17.0 
 
 19 2 
 21.0 
 21.9 
 21.6 
 
 20 3 
 18.4 
 16.2 
 
 -a 
 
 1 
 
 13 °4 
 13.8 
 13.4 
 14.2 
 17.2 
 18.5 
 21.6 
 23.5 
 22.7 
 20.0 
 17.2 
 16.4 
 
 V 
 
 a 
 
 5 
 
 -1°0 
 + .2 
 
 - .4 
 
 - .9 
 + .2 
 
 - .7 
 + .6 
 + 1.6 
 + 1.1 
 
 - .3 
 -1.2 
 + .2 
 
 H 
 lOfl 
 9.1 
 9.3 
 10.9 
 13.3 
 15.9 
 18.1 
 19.1 
 18.9 
 17.3 
 14.9 
 12.3 
 
 Mean 
 
 annual 
 
 values 
 
 25.4 
 
 25.5 
 
 + .1 
 
 24.9 
 
 23.2 
 
 -1.7 
 
 21.3 
 
 19.5 
 
 19.9 
 
 +0.4 
 
 17.7 
 
 17.5 
 
 _ 2 
 
 14.1 
 
 *Air temperatures at Honolulu (Monthly Weather Review, 1903, pp. 225-226). 
 
350 
 
 MISCELLANEOUS STUDIES 
 
 It appears that the theoretical resultfs agree well with observation. 
 If the difference between the theoretical and observed mean annual 
 temperatures due to the average departure of the local conditions 
 from the normal is applied as a correction to the observed monthly 
 temperatures the agreement between theory and observation is very 
 close. 
 
 Numerical estimates of the coefficient of absorption of solar radiation 
 
 in sea water. 
 A lower limit of the value of the absorption coefficient h^ can be 
 determined from quantities depending on surface temperatures and 
 the amount of solar radiation at the siirface of the ocean by the fol- 
 lowing method. The value of A'l (p. 344) obtained from Angot's data 
 (Hann, 1915, p. 40) was v 
 
 B:i=5.18X10'\ 
 
 where A is the solar constant. The more accurate result of Schmidt's 
 later investigation (1915, p. 121) (p. 345) is 
 
 A\= 217 X 30 X 10* = 3.255 X lO'A. 
 
 Since B 
 
 Kb, 
 
 K . 
 
 and i?//^^5.9 (p. 347) 
 
 94 X 10" X 5.9 5.54 
 
 K. 
 
 mj). 
 
 m^b. 
 
 XIO". 
 
 But since -r;r- is the ratio of the amoiuit of energv supplied to the water 
 
 by solaj- radiation to the amount available at the surface, -v;^ < 1. 
 
 A, 
 
 Therefore 
 
 5.54 X 10" 
 
 5.54 
 
 A\ m,b, X 3.255 X lO'A 32.55 m,b,\ 
 Using the accepted value 2.00 for A we have 
 m,b, = b,e-^''^ > .0851. 
 Table 3 
 
 <1. 
 
 b, 
 
 
 
 .05 
 
 .10 
 
 .12 
 
 .15 
 
 .20 
 
 .25 
 
 .30 
 
 be~''" 
 
 
 
 .043 
 
 .074 
 
 .083 
 
 .096 
 
 .110 
 
 .118 
 
 .122 
 
OCEAN TEMPEliATVUES 351 
 
 From the values of 6,c"^''> given iu table 3 it follows that b^ > .12 
 or e-''i = /?j < .887 where /Jj equals the proportion of incident light 
 that passes through one meter of sea water. Direct observations of 
 the proportion of solar radiation passing through samples of sea water 
 taken from the Nordlichen Ostsee and the Bottensee (Petersen, 1912, 
 p. 39) give values of /?j varying from .60 to .86, which are less than the 
 upper limit .887 deduced from theory. 
 
 The variation with respect to depth of the heat absorbed by the 
 water tends to maintain a temperature gradient which would be 
 greater the smaller the transmis.sion coefficient, and the mixing process 
 tend.s to reduce the gradient by transferring heat from warm to cooler 
 layers. That is, the rate at which heat is supplied to a given layer is 
 equal to that due to direct absorption of radiation plus the amount 
 due to the alternating vertical circulation of the water. But the rate 
 of gain of heat was assumed in the theory to be due entirely to the 
 absorption of radiation ; and therefore the estimate of the value of the 
 transmission coefHeient deduced from observed temperatures at dif- 
 ferent depths would be larger than the true value. This conclusion is 
 confirmed by the following computation, based on temperature observa- 
 tions near San Diego (McEwen, 1916, pi. 26). The general equation 
 (22) (p. 344), is of the form 
 
 e = R^e-''^'i'-^' [cos {at — e) — l] +E„ 
 
 where E^ and E„ are con.stants, for a given latitude. Therefore, 
 ^^g-6i(6-3) equals the half range of temperature at the surface, 
 ^^g-6i(io-3) eqi^als the half range at the depth of 10 meters and 
 [Eje-^'i'''-^' — Eje-'^i'"-^'] equals the difference between the mean 
 annual temperature at the surface and at the depth of 10 meters. If 
 there is a vertical flow (p. 374) the general temperature equation 
 reduces to the same form (equation 155. p. 390), and can therefore be 
 applied to temperatures in the San Diego region. 
 
 Substituting the observed average values of these quantities 
 (McEwen, 1916, pi. 26) gives 
 
 jj^g-sii --. 3jg ]jj|]j! range at surface (41) 
 
 2?,e-'''= 2.70 half range at depth of 10 meters (42) 
 
 i?j [c'^^'i — e"''*i]=.40 difference in mean annual tem- 
 
 pei'ature at surface and at 10 meters. (43) 
 
352 MISCELLANEOUS STUDIES 
 
 From equations (41) and (42) 
 
 e"i= 14^=1.168 or h, = m9 
 2.70 
 
 and 
 
 EJe-3ii_c-'''i]=Bi(.1285) = .45 or 7?i = 3.50. 
 
 From equation (43) and the value, 3.5, already found for R^ 
 
 e-si-. — e-'i-i = .1^1^.1142 and &i = .034 
 
 which agrees approximately with the value .039 obtained from the 
 first equation. The average value .0365 should be lased instead of the 
 large value of /j, exceeding .12 (p. 351), in order to obtain the actual 
 rate at which tlie water gains heat as a result of both absorption and 
 mixture of water from other layers. That is, the rate of gain of 
 heat in the actual sj^stem takes place as if there were no such mixture 
 of the water and the coefficient of absorption were less than the true 
 value. Hence, as far as the rate of gain and loss of heat is concerned 
 we can substitute this more simple ideal sy.stem for the actual one. 
 We have now determined all of the constants of the original differential 
 equation (4), on page 341, which expresses the time rate of change of 
 heat in an element of volume, on the assumption that the average 
 flow, either vertically or horizontally, is zero. The modified tempera- 
 ture resulting from any additional factor, for example a current, can 
 be deduced by solving the above differential equation, to whicli has 
 been added the rate of change of heat due to this factor. 
 
 Deditction of the Change in Surface Temperature Produced 
 BY A Horizontal Flow op Water 
 
 Preliminary discussion. Statement of assumptions and mathematical 
 formidation of the problem. 
 
 It is well known, as stated by a prominent British hydrographer, 
 Wharton (1894, pp. 699-712), that "the most obvious phenomenon of 
 the ocean is the constant horizontal movement of its surface water, 
 which in many parts takes well defined directions." 
 
 The work of both practical seamen and scientists has after many 
 years revealed the essential features of the main ocean currents, and 
 in a few limited regions a fairly detailed knowledge of the currents 
 has been obtained. However, all investigators agree that the esti- 
 
OCKAX rKMVKnATVRKS 353 
 
 mation of the direction and rate of tlow of water in tlie ocean is 
 attended with many difficulties. Some of the metliods of malving snch 
 estimates will now be briefly reviewed. 
 
 The most direct and widely used method is the comparison of the 
 position of a ship every noon determined from astronomical observa- 
 tion and from the log and course during the previoixs twenty-four 
 hours. The set of a current estimated in this way is subject to large 
 errors, unless special care is taken in making the observations. Under 
 ordinary conditions such estimates of currents less than ten miles in 
 twenty-four hours are quite uncertain (Kriimmel, 1911, p. 420). 
 
 Another metliod of studying currents is to use drift bottles enclos- 
 ing slips of paper on which to enter information as to when and 
 where they were found. A sufficient number of records of the initial 
 and final positions of these bottles and the corresponding time intervals 
 will, under favorable conditions, yield information especially as to 
 the average direction of the surface drift. This method is be.st 
 adapted to small enclosed seas as the bottles may then be easily 
 recovered soon after reaching the shores. For the open oceans it is 
 not satisfactory. Other floating objects, such as wrecks, icebergs, trees, 
 and plankton also furnish some information about the horizontal 
 circulation. 
 
 Under favorable conditions the current at a given place can be 
 measured directly by means of a current meter or by observing a 
 floating object designed to move with the current. In the open ocean 
 the difficulty of holding a ship in a reasonably fixed position usually 
 renders these methods impracticable. 
 
 Investigations of the causes of ocean cvirrents and their relation 
 to these causes provide indirect methods of determining them. Ocean 
 currents are directly due to various external forces, the wind or 
 friction of a neighboring current, differences in pressure resulting 
 from evaporation, precipitation and differences in specific gravity, and 
 are modified by the deflecting force due to the earth's rotation and 
 by internal friction of the water. Thus, any theory of ocean currents 
 capable of yielding even a rough approximation to the quantitative 
 relations between the complex system of causes and the resulting 
 motion of the water would necessarily be highly complicated. As a 
 matter of fact, great difficulties always arise in attempts to establish 
 a connection between practical hydrography and theoretical hydro- 
 dynamics, and deductions of currents from their causes are quite 
 uncertain except in special cases in which the conditions in the ideal 
 
354 ■ MISCELLANEOUS STUDIES 
 
 problem agree well with those in nature. The application of theory 
 to practical problems is rendered especially difficult, first, because of 
 lack of knowledge of the frietional resistance to the motion of sea 
 water, and, second, because of the uncertainty regarding the current 
 produced by a wind of given velocity and direction. 
 
 One of the most important needs now is a comprehensive pro- 
 gramme of observations at sea, of the currents themselves and their 
 causes, supplemented by attempts to formulate hydrodynamical prob- 
 lems whose solution shall be consistent with the observations. ^Mueh 
 credit is due to the pioneer investigators, Zoppritz, Mohn, Bjerknes, 
 Sandstrom, Ekman, Jacobsen, and others, for their development of 
 methods of dealing- with such problems. 
 
 Another important aid to the determination of oceanic circulation 
 is found in the fact that a current consists of water particles tending 
 to preserve their temperature and salinity as they move along. These 
 characters change slowly and thus serve to depict tlie currents some- 
 what as do floating objects that are readily identified. 
 
 This part of the paper presents an attempt to develop, along the 
 line suggested in the following translation from Kriiimnel (1911, 
 p. 439), a method of deducing currents from the temperature dis- 
 tribution : 
 
 No simple rule has been formulated for determining currents from tempera- 
 ture charts. But it is conceivable if not certain that a systematic investigation 
 of the so-called individual temperature changes will give a reliable basis for 
 the estimation of currents from temperatures. (We must distiuguish between 
 the annual temperature range, corresponding to definite geographical positions, 
 and the practically uninvestigated temperature changes which one and the same 
 water particle undergoes along the great horizontal current systems. In a 
 continuous current, for example, the Gulf Stream, these individual temperature 
 changes which must be distinguished from changes at a given position may 
 run through the whole range from tropical heat to the freezing point.) The 
 problem is, however, very difScult, and a cursory comparison of the current 
 charts in the Atlantic as jirepared from the Log Book of the "Seewarte" 
 reveals the great complexity of these closely inter-related phenomena. In 
 general, in connection with all water motions time is an all important factor. 
 Rapid and slow currents behave very differently as regards their heat content, 
 and can therefore give rise to widely different types of isotherms. No constant 
 angle between stream lines and isotherms can be proposed; the angle can vary 
 between 0° and 00°. The most frequent ease is that of stream lines cutting 
 the isotherms obliquely. 
 
 In general, the rate of change of heat in an element of volume can 
 be expressed by adding to the right-hand member of the differential 
 equation (3) the rate of change due to other factors not considered on 
 
OCEAN TEMPEllATVEKS 
 
 355 
 
 page 339, and the solution of the new e(iuation will give the tem- 
 perature under the new conditions. The rate of change of heat due 
 to a horizontal flow H of the water can be readily derived as follows : 
 consider a rectangular element of volume (fig. 2) of unit length per- 
 pendicular to the direction of flow and of breadth dz measured in the 
 dire.ction of flow and thickness chj normal to the direction of flow. 
 
 -i 
 
 Fig. 2. 
 
 Then the rate at which heat enters into the element less the rate at 
 which it is removed will be 
 
 Haedy ~Ha{e + d6)dy = —HadOdy = —Ha ~ dzdy ( 44 ) 
 
 which is the time rate of change of heat in the element due to the flow 
 of water. Multiplying equation (3) by dz to make it apply to the 
 element of volume now considered and adding the above expression 
 for rate of change of heat gives tlie new equation 
 
 <T~dydz^—Q{\ogcP,)P^(hjdz - haie — dj d.ydz~E,r~ dy dz 
 
 dt 
 
 dz 
 
 (45) 
 Dividing through by adydz and .substituting the value of Q and ^, 
 from page 341 we have 
 
 ^= Be-M[ (a -)_ a,a) cos at + a.r + l]—l-{d—0,)— H^ 
 at - .. ^2 
 
 (46) 
 
 which is the same as the temperature equation (10) with the term 
 
356 MISCELLANEOUS STUDIES 
 
 — n~ — ) added, x is the distance north or south from the latitude 
 dz ' 
 
 chosen for reference, z is the distance from the same point measured 
 in the direction of flow, making an angle \^ with the x direction ; there- 
 fore X equals nz where i/( is measured from the positive (north) direc- 
 tion of X, and n equals cos i/'. Making this substitution in equation 
 (46) gives 
 
 ^=5[(ai-f a,»2) cosa/-|-o.3?(2-fl]e-''i!'— A-(6i — ^J — H^ 
 Of " oz 
 
 (47) 
 Solution for the case in which the flow is constant. 
 To solve equation (47) let 
 
 = 0' jf- 0" -{- ff" 
 
 where 6' is the solution already found (equation 22, p. 344) correspond- 
 ing to H^O, 6" a function of y and t only is to be determined, and 
 8'" is a general solution of the part left after substituting {6'-\-6"). 
 Substituting the value 6' -\- 6" + G'" for 6 in equation (47) we have 
 
 ^+^ +^ = B[{a, + a,nz\ cos at -f a,nz + 1]^"^ 
 
 _],(0'+0"+0"'-0^)-H^ H~-H^ (48) 
 
 dz dz dz 
 
 From the definitions of B', 6" and 6'" this equation reduces to 
 
 66" . 66'" ^_j.^._j.ff„_jj^ -H^ (49) 
 
 dt dt dz dz 
 
 which can be broken up into two equations 
 
 ^=_A.r-H^ (50) 
 
 dt dz 
 
 and 
 
 dt ' "" ' " dz 
 
 ^ + 7,r' + ff^ = 0. (51) 
 
 From equation (22) which gives the value of 6' we have 
 de' Ba.,n , Ba:,nf-'"^!'-^^ . , . x n , Ba^ne'"" 
 
 - ■' ' - [cos (a# — e) — 1|-| = 
 
 =,n Ui — K.e-"^'"-^^ [1 — iA^cosat + B.sinat)] | (52) 
 
OCEAN TEMPERATUBES 357 
 
 where g, /i,, A^ and B^ are constants having the following values : 
 Ba., , Ba.e-'"^ _ Ba„ 
 
 ^i \/a.- -\- k- ' y/a- -f A;- 
 
 k 
 J.1 = — ^ and Bj ^ 
 
 Va" + A'" Va' + ''■'" 
 
 Substituting the above value of- — in equation (50) gives 
 
 03 
 
 -+A-(9" = — H |sf— e-''i<!'-=»A', [1— (^iCOSai+BiSinaO] |n 
 
 (53) 
 remembering that A- equals AjP"''''"'^' and for depths between zero and 
 six meters the constant value y equals 6 is to be used for ij, while for 
 other depths the actual value of the depth is to be used for ij (p. 343). 
 H, the horizontal velocity, may be any function of the time, but it 
 is assumed to be independent of z and y. Having in mind a numerical 
 application to be made later it will be convenient to let 11 equal the 
 periodic function of the time 
 
 il = -ffi ( 1 -|- a^ sin a< -f a. cos at) 
 where H^, a^ and a, are constants. Equation (53) then becomes the 
 oi'dinary linear differential equation of the first order in 6" and / 
 
 ^-^ke" = —n,il ^g — K,{^—) [l—(A,COSat + B,Smat)]^ 
 
 X (1 + «4 sin at -\- a- cos at) (o-i) 
 
 Solving by the corresponding standard formula we have 
 
 e" = —e-''t / I ffjH [fif — — 7u(l — AjCosai + BisinaO] 
 
 X [I + a, sin at + a, cos af]c>'' + C I dt (55) 
 
 where C is arbitrary but independent of t and z. Under these con- 
 ditions Ce'''' will evidently be included in a general solution of equa- 
 tion (51), and will therefore be neglected in the expression for 6". 
 Equation (55) can be directly integrated with the aid of well known 
 standard forms and the result for II equal to a constant velocity Hj is 
 
 ~ l^i^~ V'^^T^J" (a= + A-=) = 
 
 X [2aA- sin at + (A-= — a=) cos at] (56) 
 
358 MISCELLANEOUS STUDIES 
 
 Equation (51) in wliich n.^{l -\- a^ sin at -\- a -^ cos at) is substituted 
 for H becomes 
 
 ~ +M'" +-H"i(l + a, sin at + a- cos at) ^ ^^ (57) 
 
 or ■ ' oz 
 
 In the special case where o^ = a,, = the solution is 
 
 0"'=e-'^^f{t-^) (58) 
 
 where / (/ tt~^} ^^ ^^ arbitrary function of U n"^) • ^^^^^ 
 
 solution can be easily verified by substitution in equation (57). 
 
 For a constant velocity H equals 77^, the general solution of 
 the differential eciuation (47) is the sum of the three quantities 
 6' -j- 6" -\- 6'" already found, and can be put in the form 
 
 .= {^<"- + "-'"^""^'^COs(a^-.)-l]+g^+g+^-^-'"""^ , , ] 
 
 X l'2al{smat+{k-—a-) COSat] i + | e" ^i / (<- ^ ~" j i (59) 
 
 Suppose the relation of the temperature to the time at a given 
 position ^„ is known and that there is a constant horizontal velocity 
 Hi from that point in any given direction. From equation (59) the 
 temperature at any time and at any point along the stream line 
 down stream from the point z equals s,, can be found by giving the 
 
 arbitrary function / l<— ^ tt~" ) *^"*'^^ values that when z equals ^o 
 
 6^6' -\- 6" -\- 6'" will equal the observed temperature, wliich is a 
 known function of the time at that point. All of the consstants in 
 the equation are given on page 347. Tlierefore /(<') being known, 
 when t'. the time at the position, z equals z„ is known, the arbitrary 
 function is determined. For a time t and a value (z — z^) of the 
 distance from Zg the expre.ssion 
 
 f{t-^^) = f{t') (60) 
 
 where (,_^_^o) = (,.) (61) 
 
 since, in general, the function is determined by the values of the 
 independent variable / z — z„ \ 
 
OCEAN TKMVEKATrUES 
 
 359 
 
 Solution for the case in which the flow is a periodic function 
 
 of the time. 
 
 If the flow is the periodic function of the time 
 
 H = E, (1 + a, fi'm at ^ a, cos at) ( 62 ) 
 
 the term 6' (equation 22, p. 344) will be the same as before, but 6" will 
 be given by equation' (55) where a^ and a- are retained. The integra- 
 tions can be readily performed with the aid of well known standard 
 forms and the result is 
 
 ' =-T-L(^-^^^TF^r2(?+w^'"=+'^"^^j 
 
 Ba e-''i<"-''i 
 
 a-+ Ir 
 
 COS at - 
 
 \l;-a,~Saka, — 2a-a,] cos 2at 1 (63) 
 
 The solution of equation (57) when a^ and a-, ai-e retained, found 
 by Lagrange's method, is 
 
 e"'=e-^*f^ 
 
 I t ■■ cos at -\ — ^'sin at ) 
 
 z 
 
 (64) 
 
 where /j [ ] is an arbitrary function of 
 
 ( 
 
 t ■* cos at -\-- sin at ]— Yf 
 
 This solution can be verified by substitution in equation (57), and 
 can be readily changed into the more suitable form 
 
 A-(z — Zo) A-(ai cos at — «:, sin aO 
 
 Bi 
 
 s. J 
 
 / it — - cos at + - sin af J 
 
 (65) 
 
360 MISCELLANEOUS STUDIES 
 
 which reduces to equation (58) when a^ and a^ equal zero. The 
 temperature at any time and place down stream from the position 
 where z equals z^ can be found if the relation of the temperature to 
 the time is known where 2 equals z„, by giving the arbitrary function 
 /■[ ] values such that for z equals z^, 6 = 6' +6" + 6'" will equal 
 the observed temperature which is a known function of the time at 
 that position. Thus the arbitrary function is determined .since its 
 value is known for a series of values of the independent variable 
 
 (f — cos at' -\ — sin ai' ) , using f for the time where 2 equals z^. 
 a a / 
 
 For any other value of the time t' and for a distance (2 — 2„) down 
 stream 
 
 /[(^_^COSa< + -^sinaf)-^^] 
 
 =f[t'— -^ COH at' +^smat'l, (66) 
 
 l_ a o — I 
 
 where 
 
 = (/'_-^C'OSar + -^siuar). (67) 
 
 \ a a ' 
 
 Solution for the particular case in wliirli the time interval is so small 
 that the solar radiation may be assumed to depend 
 only on the latitude. 
 In certain cases it will be convenient to take a time interval so 
 short that the insolation may be regarded as independent of the time 
 and the current may be assumed to have a constant velocity from a 
 po.sition 2 equals 2„ where the temperature may be assumed constant. 
 Under these conditions the temperature at any point distant (2 — 2,,) 
 down stream will be independent of the time if sufficient time has 
 elapsed for an element of volume of the water passing through tlie 
 position 2„ and having the given constatit temperature to move through 
 a distance equal to or greater than (2 — z^). For this steady state, 
 equation (47) becomes 
 
 B[b.+ b,m]e-''^^''-^^—k(e — 6,)—n, ^ = (68) 
 
 where h^^l -\- a^ cos at.^ 
 
 &3 = a„ cos at^ -\- a^ 
 
OCEAN riCMVEliATVUES 361 
 
 and fj is the average of the values of t for the beginning and end of 
 the time interval. Let $ = 6'-\-d" where 6' is the solution when 
 J7, = 0. Then 
 
 Substituting (i9' + (9") inequation (68) gives 
 
 (69) 
 
 ^^r = -f- (70) 
 
 dz H, dz 
 
 For 6' use the normal value determined from the expression j | 
 
 of equation (59) for t^t^ then for surface temperatures using the 
 value 6 for y (see page 343) 
 
 66' Ba.,n , Ba„nc-^^^ , . s l^^\ 
 
 T~ = —r — I — cos (at, — e) = g, (<1) 
 
 where tan t ^^t- 
 
 This is consistent with equation (fiO) since 6^ may be any function of z. 
 Equation (70) then becomes 
 
 f::+^r=-., (71) 
 
 Integrating equation (71) gives 
 
 e" = ec-^^-a, --^^ (72) 
 
 where 6 is arbitrary. Adding the two solutions 6' and 6" gives 
 
 + {^.-^?^--^} (73) 
 
 where the expression \ [ is the normal temperature. To deter- 
 
 mine the temperature at a given position, distant (z — z„) from the 
 initial position ^i,, give 6 such a value that the expression for 6 in 
 equation (73) will reduce to the given temperature when z = z„. Then 
 substitute this value and the given value of z in equation (73). 
 
362 MISCELLANEOUS STUDIES 
 
 Consider two parallel stream lines, A and B, the velocity being 
 Ha along the first and i^B along the second, then the temperature in 
 A for any value of z minus the temperature in B for the same value 
 of z is 
 
 ej,—eB=eAO- ha —Obc- bb +|'(i7B— 5"a)=a9 (74) 
 
 Denote 77a— Fb by AH. tlien 
 
 Ae = —-~AE+eAe-nB+AH —due- hb (75) 
 
 Also if -=^-is small we liave approximately 
 -cia 
 
 (76) 
 
 The Rate of Horizontal Flow in the North Pacific off the 
 California Coast from Lat. 40° N to 30° N and in the North 
 Atlantic off the West Coast of Africa from Lat. 30° N to 
 20° N. 
 
 The rate of flow deduced from surface temperatures. 
 
 From the hydrographic cliarts (Thorade, 1909) of the region of 
 the Pacific off North America, it appears that that the average direc- 
 tion of the surface drift from Cape Mendocino, Lat. 40° N, does not 
 at any season differ greatly from a straight line determined by the 
 points. Lat. 40° N, Long. 124° W. and Lat. 30° N, Long. 126° W. 
 Assuming that there is a surface drift in thi.s constant average direc- 
 tion which i.s proportional to the average wind velocity over this 
 course, will some numerical value of the drift account for the monthl.y 
 temperatures at the down-stream end of the line? From the 
 monthly isotherms worked out by Thorade (1909), the observed 
 mean monthly temperatures at any point of the region can be found. 
 From these observed monthly temperatures at the upper end of the 
 line and a mean valuQ of the drift velocity determined by trial, the 
 temperatures at the down-stream end will be computed according to 
 the theory on page 359. A comparison of these theoretical tempera- 
 tures with the observed ones and of this theoretical value of the drift 
 with estimates made in other ways will indicate the practical value of 
 the theory. The observed temperatures taken from Thorade 's chart 
 (1909), and the numerical values of the other cpiantities computed 
 
Table 4 
 
 Computation of temperatures alonq a horizontal stream-line in the northeastern Fa 
 
 Z = Z„^ 1 
 
 Month = t 
 
 
 1 
 
 2 
 
 
 Observed temperature = 6* 
 
 
 11.3 
 
 10 7 
 
 1( 
 
 Normal temperature ^ 6' 
 
 
 10 14 
 
 9 OG 
 
 c 
 
 6" (from equation 63) 
 
 
 -3.38 
 
 -3 38 
 
 ■ — >' 
 
 $•" = ^ — ( (9' + e" ) = e- a *"* "' "''" "'• "'" "' 
 
 ^f{f' — ^COSar+%mal') 
 a a 
 
 4.54 
 
 5 02 
 
 4 
 
 - (a* cos at' — as sin a(') 23 sin at' 
 
 e-a — e 
 
 
 .887 
 
 .819 
 
 
 f, f, .23 sin al' 
 
 *f ( V— %os at' +%m at') = e 0'" 
 
 a a 
 
 
 5.11 
 
 6 14 
 
 e 
 
 *f — -'cos at' + ^sin at'=t'— 1.15 sin at' 
 
 a a 
 
 
 -.42 
 
 1.0 
 
 1 
 
 : = 
 
 Normal temperature ^ 0' 
 
 18.7 
 
 18.0 
 
 18 
 
 e' + 6" 
 
 15.32 
 
 14 62 
 
 14 
 
 {t —-'cos at + -'sin at)--^^ =[i~ 1-15 sin at — 8.85 
 
 a a Hi \_ _ 
 
 -8.77 
 
 -7.35 
 
 -6 
 
 f (t ^COSa< + — .sm an— 
 
 •' L ^ a 'a Hi _ 
 
 4.5 
 
 2 4 
 
 1 
 
 -^ — («! COS at — Or. sin aO -i on _:_ .23 sin at 
 
 .167 
 
 .154 
 
 
 e"'=:.188e- -^Ssinafy.^ -j 
 
 .8 
 
 .4 
 
 
 0' 4- (9" + 6"' = 61 = the computed temperature 
 
 16.1 
 
 15 
 
 14 
 
 Observed temperature 
 
 16.7 
 
 16 5 
 
 16 
 
 Computed minus observed temperature 
 
 -0.6 
 
 -1.5 
 
 -1 
 
 
 t' = 
 
 -9 
 6.27 
 
 -8 
 
 * These two lines are continued here for negative values of t' 
 
 4 
 
 
 
 -10.15 
 
 -9 
 
Tablb 4 
 Computation of temperatures along a horizontal streamline in the northeastern Pacific, where the 
 
 current varies penodically with the time 
 
 Month = t 
 
 Observed temperature = 6 
 Normal temperature = 0' 
 g" (from equation 63) 
 
 .,,, /) la' -i—A"\ -,-- (Oi cos aC — Or, sin of) » / ., 0,^ #• . '^s • ii\ 
 
 - (04 cos at' — Ob sin at') __ 23 sin at' 
 
 ro 
 
 *' "'\ 
 
 a, . 
 
 .23 sin at' 
 
 *f(f— %os at' +-'sin at') = e 6" 
 
 > ^ a a 
 
 *f_-%os at' + ^sin at'=t'— 1.15 sin at' 
 
 Normal temperatiire = 6' 
 
 ' + ' 
 
 lt-'%osat + -'sin at)-— ^= r< — 1.15 sin ai — 8.35l 
 
 a a Hi L_ _| 
 
 y[^(i_%OSa<+JsinaO-^] 
 
 — ^-= (at cos at — Od sin at) < qq .:_ .23 sm at 
 
 6 ill n — .J-oo e 
 
 r=.188e- •''''"'"/I ] 
 
 e' -(- 61" + e"' = 61 = the computed temperature 
 
 Observed temperature 
 
 Computed minus observed temperature 
 
 * These two lines are continued here for negative values of f 
 
 13.1 
 
 12.4 
 
 14.9 
 
 12.27 
 
 -3.85 
 
 -3.61 
 
 -2.05 
 
 3 48 
 
 1.127 
 
 1.00 
 
 1.82 
 
 3.48 
 
 11.58 
 
 12.0 
 
 = 
 
 18.7 
 15.32 
 -8.77 
 4.5 
 .167 
 
 18.0 
 
 14.62 
 
 -7.35 
 
 2.4 1.5 
 
 18.2 
 14.59 
 -6.5 
 
 19.2 
 
 15.35 
 
 -5.35 
 
 .154 
 
 16.1 
 
 16.7 
 
 -0.6 
 
 t' = 
 
 15.0 
 
 16.5 
 
 -1.5 
 
 ,149 
 
 .1 
 
 14.8 
 
 16.2 
 
 -1.4 
 
 20.8 
 
 16.21 
 
 -3.93 
 
 154 . 167 
 
 .10 
 
 15.4 
 
 16 1 
 
 -0 7 
 
 6.27 
 
 -10.15 
 
 4.82 
 
 -9.0 
 
 2.58 
 
 -7.58 
 
 22.6 
 
 17.77 
 
 -2.35 
 
 -.75 
 
 .188 
 
 05 .14 
 
 16.2 
 
 16.3 
 
 -0.1 
 
 -6 
 1.10 
 -6.0 
 
 17.6 
 
 17.0 
 
 +0.6 
 
 .18 
 
 -4.42 
 
 23.9 
 
 18.84 
 
 ■1.2 
 
 212 
 
 .35 
 
 19 1 
 
 18 3 
 
 -1-0.8 
 
 -4 
 
 -1 70 
 
 -3 
 
 24.6 
 
 19.54 
 
 .65 
 
 5.3 
 
 .230 
 
 12 
 
 20.7 
 
 18 9 
 
 -t-1.8 
 
 -3 
 
 -1 62 
 
 -1.85 
 
 24 4 
 
 19 57 
 
 1.8 
 
 6.3 
 
 ,236 
 
 15 
 
 21.1 
 
 19.4 
 
 +1,7 
 
 -2 
 
 23.3 
 
 18.71 
 
 2 66 
 
 5 7 
 
 .230 
 
 1.3 
 
 21 6 
 
 17.75 
 
 3 23 
 
 4.4 
 
 20.0 
 
 16.39 
 
 3.65 
 
 4.0 
 
 .212 
 
 .93 
 
 20.0 18.7 
 
 18.9 
 
 + 1.1 
 
 -1 
 
 .89 
 
 ■1.0 
 
 1.82 
 
 -.42 
 
 18.2 
 
 .188 
 
 17.2 
 
 17.2 
 
 +0.5 0.0 
 
 3.48 
 
cific, where the current varies periodically with the time 
 
 3 
 
 4 
 
 5 
 
 6 
 
 7 
 
 8 
 
 9 
 
 10 
 
 11 
 
 12 
 
 1.7 
 
 11 
 
 11.0 
 
 12 2 
 
 13.2 
 
 12.0 
 
 12.0 
 
 13.8 
 
 13.1 
 
 12.4 
 
 1.34 
 
 10.9 
 
 13.3 
 
 15.93 
 
 18.06 
 
 19 14 
 
 18.86 
 
 17.3 
 
 14.9 
 
 12.27 
 
 ;.6l 
 
 -3.85 
 
 -4.59 
 
 -4.83 
 
 -5.06 
 
 -5.06 
 
 -4.83 
 
 -4.59 
 
 -3.85 
 
 -3.61 
 
 -.97 
 
 3.95 
 
 2 29 
 
 1.10 
 
 20 
 
 -2.08 
 
 -2.03 
 
 -1.09 
 
 -2 05 
 
 3.48 
 
 794 
 
 .819 
 
 .887 
 
 1.00 
 
 1.127 
 
 1.221 
 
 1 259 
 
 1.221 
 
 1.127 
 
 1.00 
 
 ..27 
 
 4.82 
 
 2.58 
 
 1 10 
 
 .18 
 
 -1.70 
 
 -1 62 
 
 .89 
 
 1.82 
 
 3.48 
 
 85 
 
 3.0 
 
 4.42 
 
 6.0 
 
 7.58 
 
 9 
 
 10 15 
 
 11.0 
 
 11.58 
 
 12.0 
 
 .2 
 
 19.2 
 
 20.8 
 
 22.6 
 
 23.9 
 
 24.6 
 
 24.4 
 
 23.3 
 
 21 6 
 
 20.0 
 
 .59 
 
 15 35 
 
 16 21 
 
 17.77 
 
 18.84 
 
 19.54 
 
 19.57 
 
 18.71 
 
 17 75 
 
 16.39 
 
 .5 
 
 -5 35 
 
 -3.93 
 
 -2 35 
 
 — .77 
 
 .65 
 
 1.8 
 
 2.65 
 
 3.23 
 
 3.65 
 
 .5 
 
 -.8 
 
 -.3 
 
 -.75 
 
 -1.2 
 
 5.3 
 
 6.3 
 
 5.7 
 
 4.4 
 
 4.0 
 
 .149 
 
 154 
 
 .167 
 
 .188 
 
 .212 
 
 .230 
 
 236 
 
 .230 
 
 .212 
 
 .188 
 
 1 
 
 .10 
 
 -.05 
 
 .14 
 
 .35 
 
 1.2 
 
 1.5 
 
 13 
 
 .93 
 
 .8 
 
 .8 
 
 15.4 
 
 16.2 
 
 17,6 
 
 19 1 
 
 20.7 
 
 21.1 
 
 20.0 
 
 18.7 
 
 17.2 
 
 .2 
 
 16 1 
 
 16.3 
 
 17.0 
 
 18.3 
 
 18.9 
 
 19.4 
 
 18,9 
 
 18 2 
 
 17.2 
 
 .4 
 
 -0 7 
 
 -0.1 
 
 +0.6 
 
 +0.8 
 
 + 1.8 
 
 + 1.7 
 
 + 1.1 
 
 +0.5 
 
 0.0 
 
 
 -7 
 
 -6 
 
 -5 
 
 -4 
 
 -3 
 
 -2 
 
 -1 
 
 
 
 
 .82 
 
 2 58 
 
 1.10 
 
 .18 
 
 -1.70 
 
 -1 62 
 
 .89 
 
 1.82 
 
 3.48 
 
 
 
 
 -7.58 
 
 -6.0 
 
 -4.42 
 
 -3.0 
 
 -1.85 
 
 -1.0 
 
 - 42 
 
 
 
 
OCEAX TEMPEliATVRES 363 
 
 from the theory (pp. 355-360) are eondensed in table 4, iji which 
 the following values of the constants are nsed: «4^0, a.^ = — .6, 
 «!==— .318, a, = —.0244, a, = —.0166, 7i = .20, and ir = — 1.2 
 (1 — .6 cos 30° equals the drift, in degrees per month equals 
 1.(1 to 3.8 miles in twenty-four liours. The mean wind velocity in miles 
 per hour over the cour.se considered (Mooi-e, 1908-11) is approximately 
 F = 10[l — .6cos (300°]. 
 
 Tlic rate of flow deduced from temperature data compared icith that 
 expected from the empirical relation of winds to currents and ivith 
 direct ohservations on currents. 
 
 As stated by Helland-Hansen in his paper on physical oceanography 
 
 (:Murray and Hjort, 1912, p. 247) : 
 
 The wind may produce a current, particularly in the surface layers, thus 
 altering the direction and velocity of the existin<f current. We know very 
 little, however, about the relation between wind and current, through lack of 
 detailed observations, although the question is naturally of the first importance 
 from an oceanographical point of view, as well as from its bearings on the 
 conditions of everyday life. This is one of the principal tasks for the ocean- 
 ographer of the future; such observations are difficult to make, no doubt, but 
 with modern methods much can be done. 
 
 However, numerous ohservations of winds and currents have been 
 made. And, although the relation of wind and current varies with 
 the wind velocity, the latitude, coast line, depth, and distribution of 
 specific gravity, some progress has been made in estimating the drift 
 that a given wind velocity will produce. A careful investigation of 
 this question based on Ekman's theory (1905, 1906) and a large mass 
 of available data made by Thorade (1914) yielded the following 
 results. In case the coast is sufficiently distant and the effect of the 
 pi"es.sure gradient due to differences in specific gravity is small, the 
 drift will be directed at an angle of 45° to the right of the wind 
 direction in the northern hemisphere. The relation of the drift to the 
 wind velocity estimated by Thorade (1914, p. 387) is 
 
 „ .0259\/V -- ..meters „ ^ . mi. ,^^, 
 
 H= ^ , I <4.3 <8.74t (77) 
 
 Vsin<> sec. hour 
 
 and 
 
 ^_m26y 5 meters ^^^mi^ 
 
 Vsin</> sec. hour 
 
 where V is the wind velocity. 27 is the current in meters per second, 
 and (f> is the latitude. 
 
364 
 
 MISCELLANEOUS STVDIES 
 
 The mean latitude of the drift eompTited from temperature data 
 (p. 362) is 35° and its direction was 45° to the right of the mean wind 
 velocity in accordance with Ekman's theory (1906) and Thorade's 
 estimate from observations (1914). From equation (78) and the 
 observed value of "T (p. 363) the drift would be 3.9 miles in twenty- 
 four hours if it were due entirel.y to the observed wind.s, uninfluenced 
 by the coast and differences in specific gravity. This estimate is of the 
 same order as 2.4. that made from temperature observations (p. 363). 
 Again, direct observations of the drift having a southerly component 
 (Thorade. 1914, p. 283) near the head of the stream, Lat. 40° N to 
 50° N, gave the values presented in table 5. 
 
 T.\BLE 5 
 Observed surface drift, and values computed from temperature data 
 
 Month 
 
 4 
 
 5 
 
 10 
 
 Mean 
 
 Observed drift in 24 hours 
 
 3.69 
 
 2,0.5 
 
 2.41 
 
 2.6 
 
 
 
 No. of observations 
 
 21 
 
 24 
 
 54 
 
 
 
 
 Computed drift in 24 hours 
 
 3.12 
 
 3.6.5 
 
 1.68 
 
 2.8 
 
 Thus the theoretical drift estimated from temperature data agrees 
 as well with the observations as could be expected. And it appears 
 from the comparisons made that estimates of the drift from tempera- 
 ture data will prove to be as reliable as those made by other methods. 
 
 The surface current during a short time interval near the northivest 
 coast of Africa, estimated from surface temperatures, and com- 
 pared, irith direct observations, and with results deduced from the 
 empirically ascertained relation of udnds to currents. 
 
 From a series of direct measurements by means of a float designed 
 especially for the purpose (Schott et al., 1914), the average flow 
 between latitudes 20° N and 28° N, off the west coast of Africa, was 
 found to be nearly parallel to the coast and toward the southwest. 
 These current measurements were accompanied by observations of 
 surface temperatures and winds, and the stations were distributed 
 along a line nearly parallel to the average surface drift and about 
 150 miles offshore. All of the observations were made during the 
 
OCEAN TEMPESATUllES 365 
 
 short time interval from June 2 to June 15, 1911, and are therefore 
 appropriate for the application of the theory developed on pages 360- 
 362 for estimating surface currents from temperatures. 
 
 The mean position during the three days, June 2, 3, and 12, was 
 Lat 30° N, Long. 14?6 W, and that during the three days, June 13, 
 14, and 15, was Lat. 24° N, Long. 17?3 W. The distance between these 
 positions is 6.74, the unit being a degree of latitude, and the mean 
 surface temperatures were respectively 18?32 and 19?12, each value 
 being the average of eighteen observations. In equation (73), page 
 361, (z — Z(,) is the distance, measured in degrees of latitude in the 
 direction of the drift, and the initial position is in this case at Lat. 
 30° N. The direction of the drift was found to be to the south at an 
 angle of about 27° to the right (west) of the meridian, therefore the 
 change in latitude corresponding to the distance {z — z^) along the 
 line of the flow is (cos27°) (z — z^) ^ .891 (z — z^) . From these 
 values and the numerical values of the constants given on page 346, 
 equation (73) becomes 
 
 ^ = 22.6— (.891) (.41) (2 — So)+^e '" h' '" +(.891) (1.88FJ 
 
 which gives the temperature at any point along the stream line, the 
 mean velocity being H^ degrees per month. 
 
 To determine Hj substitute the two mean temperatures with the 
 corresponding values of {z — z^), thus obtaining two equations 
 
 18.32 :== - 22.6 +1 + 1.685fl^i 
 and 
 
 _ 1.461 
 
 19.12 = 22.6+ (.365) (6.74) -\- de^ir+l.&SbE^ 
 Eliminating 6 gives the equation 
 
 1.461 
 
 — (4.28 + 1.685ff,)f~"ffr+ 1.68577, = — 5.94 
 
 from which the value of 77,, found by trial is 
 
 77, = — 9.4 degrees per month = — .78 miles per hour. 
 
 Using this value of 77, the theoretical temperature at any point along 
 the stream line is 
 
 „ „^ .,„_ , , ,, ^ .0231(2 — z„) 
 
 6 = 6./o — .36o {z — Zq) + ll.o7e 
 
 where (z — z„) is negative since the latitude decreases in the down- 
 stream direction. 
 
366 
 
 MISCELLANEOUS STUDIES 
 
 Direct estimates of the drift were made at six stations along this 
 line from angular measurements relative to a float, the ship being 
 manoeuvered in such a way as to keep the sounding cable as nearly 
 vertical as possible. The values obtained at each station in the order 
 from north to south are 1.0, 0.7, 0.9, 0.8, 0.9, 1.3 miles per hour in a 
 southwesterly direction. Each value is the mean of about twenty-five 
 
 Pig. 3. Geometrical construction for determining 
 the surface current jiroduced by wind. 
 
 the effect of a coast oii 
 
 observations. The components parallel to a line from the first to the 
 last station having the mean direction of the observed drift are 0.99, 
 0.7, 0.9, 0.7, 0.86, and 1.11 miles per hour, and the mean value is 0.88. 
 
 From the four estimates based on "dead reckoning" and the 
 position of the ship determined from astronomical observations at 
 noon, the drift appeared to be directed to the west of the direction 
 determined by the "float" method. The values are 0.4, 0.5, 0.4, 0.4 
 and the components parallel to the mean direction of the drift found 
 by the float method are 0.3, 0.2, 0.16, 0.19, the mean value is 0.21. 
 
 The wind blew steadily from the northeast, the observed velocities 
 in miles per hour being 28, 23, 28, 34, 3, 18, 13, 28, 28, 28, 34, and 34; 
 the mean is 25. 
 
 From equation (78), page 363, using 24°, the mean latitude of the 
 stream line for <^, the drift due to a wind velocity of V miles per hour 
 would be .01975F miles per hour. Using the value 25 for V the un- 
 
OCEAN TEMPERA TVHES 
 
 36T 
 
 disturbed drift due to the wind would be 0.494 directed at an angle 
 of 45° to the right of the wind direction; this direction of drift is 
 nearly the same as that obtained from dead reckoning. If the .same 
 wind velocity prevailed over the whole coastal belt, a correction to 
 the above estimate of the drift must be made (Ekman, 1906, p. 23). 
 Tile computation can be carried out graphically as follows (fig. 3). 
 Let OT be the direction of the wind, and OA represent in magnitude 
 and direction the "undisturbed drift" computed from equation (78). 
 If a circle is described through A tangent to OT, and a line AD is 
 drawn parallel to the coast then OD will represent in magnitude and 
 direction the corrected surface drift. In the ca.se imder consideration 
 the corrected estimate OD is twice the value of OA and makes an angle 
 of about 27° to the right (west) of the observed mean direction. 
 Therefore the component parallel to this observed direction is 
 (0J9) cos 27° == 0.89 (0Z)) = (0.89) 2 (0A)=: 1.78 (0.4). 
 The results corresponding to various values of the wind (T) are 
 presented in the following list. 
 
 V 
 
 OA 
 
 OD 
 
 (OD) co,s 27 
 
 1.3 
 
 .257 
 
 .51 
 
 .46 
 
 18 
 
 .356 
 
 .71 
 
 .63 
 
 25 
 
 .494 
 
 .99 
 
 .88 
 
 34 
 
 .672 
 
 1.34 
 
 1.20 
 
 Finally it is evident that the velocity of 0.78 miles per hour 
 deduced from surface temperatures agrees well with the estimates 
 made bv the other methods. 
 
 The Rel.\tion of Temperature to Time, Depth and Rate of 
 Vertical Flow in the Interval from 40 to 600 IMeters 
 
 Statement of assumptions and mathematical formulation of the 
 
 problem. 
 
 It has been found (p. 351) that the direct heating of the sea water by 
 the absorption of solar radiation is proportional to e"*'" where 6, > .12 
 and jj is the depth in meters. Hence at the depth exceeding 40 meters 
 this direct heating effect would be less than 1 per cent of that at the 
 surface. Also the temperature range at that depth woidd bear the 
 same proportion to that at the .surface if the variation in rate of 
 gain of heat were due only to the variation in this rate of absorption. 
 
368 MISCELLANEOUS STUDIES 
 
 However, observation shows that there is a seasonal variation of 5° 
 at 40 meters and exceeding 1° at 100 meters (Murray, 1898, p. 127; 
 MeEwen, 1916, p. 268) ; thus something other than the direct absorp- 
 tion of solar radiation must be the main factor in heating the water 
 of these lower levels. 
 
 These facts show that there must be a transfer of heat between 
 the upper and lower level, but the ordinary process of heat conduc- 
 tivity, as illustrated by laboratory experiments on still water, is 
 wholly inadequate to effect this transfer at a sufficiently rapid rate 
 (Wegemann, 1905a, 1905Z)). It is now generally recognized that this 
 transfer of heat results from an alternating vertical (p. 338) circula- 
 tion of the water (Helland-Hansen, 1911-12, pp. 68, 69), in which at 
 any given instant certain portions of the water are moving upward 
 while otliers are moving downward. The resultant flow of a given 
 column of water may be either upward or downward, or may be zero. 
 "Without analyzing the complicated process by which heat is trans- 
 ferred from one level to another in the ocean, it will be assumed to 
 be similar to ordinary conduction. But the coefficient of conductivity 
 corresponding to conditions in tlie ocean will depend mainly on the 
 intensity of the circulation or mixing process (Gehrke, 1910, p. 68; 
 Jacobsen, 1913, p. 71), and might be called the "coefficient of con- 
 vective conductivity" to distinguish it from the ordinary laboratory 
 coefficient. In the following investigation this coefficient of conduc- 
 tivity will be used and the direct effect of solar radiation will be 
 neglected. If the resultant vertical flow is zero, the well known partial 
 differential equation 
 
 -^=.= ^ (79) 
 
 applies, where is the temperature, t is the time, y is the distance 
 below the surface, and p.- is the diffusivity, a constant proportional 
 to conductivity. If the resultant vertical flow is w it follows, as on 
 pages 354-355 that the time rate of change of temperature due to this 
 
 flow is { — -iv ——) and the temperature equation then becomes 
 \ oy ' 
 
 d6 „ d-d 66 ,Q^^ 
 
 — =/x- -— r — M'-T- (80) 
 
 dt '^ dy- dy 
 
 Equation (80) is a special case of the general equation of the conduc- 
 tivity in a moving medium (Winkelmann, 1906, p. 444). Equation 
 (79), a special case of Pourtier's equation of the flow of heat in a 
 
OCEAN TEMPER ATUEES 369 
 
 stationary medium, has been applied to the problem of temperature 
 distribution in the ocean by Wegemann (1905a, 19056), using the 
 laboratory value of ju,-, but the theoretical results were of an entirely 
 different order of magnitude from those given by observation. 
 
 Jaeobsen (1913, p. 71) has successfully applied the equation of 
 the form (79) to some data on the distribution of salinity and cur- 
 rents in the sea near Denmark. He determined fr, the Mischungs- 
 intensitat from field observations, using the idea that salt content, 
 quantity of motion, temperature, and other properties of sea water 
 varj^ because of the alternating changes in the position of the water 
 particles. The writer is, however, not aware of any application of 
 equation (80) to ocean ographic problems. 
 
 Solution for the case in rchich the vertical flow is constant. 
 
 If the vertical velocity has the constant value ii\ then we have 
 (p. 368) to find a solution of the following linear partial differential 
 equation with constant coefficients 
 
 69 .6^6 . 66 . .„,, 
 
 -TT— /i--T-:r+«'i-T— =0 (81) 
 
 6t 6y- 6y 
 
 satisfying certain bovmdary conditions. To determine the temperature 
 at any depth, having given that at the upper level y = y.. we must 
 have a solution reducing to the given function of the time t (in this 
 paper it will be a periodic function of at the upper level and having 
 a given constant value at the lower boundary. A convenient method 
 of solution is to assume 
 
 e = Me<'v+^* (82) 
 
 and substitute in equation (81). The result is 
 
 l)-\-iv^a — /x-a-^0. (83) 
 
 Therefore 6°^'+'" is a solution of equation (81) for all values of the 
 constant M and for all values of a and 6 satisfying equation (83). Let 
 
 a^di ± hj, (84) 
 
 where a^ and h^ are real, then from equation (83) we have 
 
 h^[l>.-{a^'—h^-) — u\a,] ± [(2a,/i= — wj&ji (85) 
 
370 MISCELLANEOUS STUDIES 
 
 and, if the solution is to be a periodic function of the time having 
 
 o 
 the period — where a^ is positive 
 
 »i 
 
 /j-'ia^- — &1-) — u\ai = (86) 
 
 {2a,ix- — ii\)h, = a, (87) 
 
 Solving equation (86) for a^ and equation (87) for b^ we have 
 
 and 
 
 w, =p ^/u\^ + 4^'b,- 
 Oi= 2~2 ^ ' 
 
 \=7r^ (89) 
 
 Since the temperature and the amplitude of the temperature 
 decrease as the depth increases, the exponent o,?/ and hence a, must 
 be negative (y is positive in the direction from the upper surface 
 downward). Therefore only the negative sign is admissible before the 
 radical in equation (88) and 
 
 is definitely determined by given values of n\. /x- and ±&i. Solving 
 equation (87) for a^ gives 
 
 therefore because of equation (90) ' ., must be negative, or b^ 
 
 2b,ii- 
 
 must be negative since a^ is assumed to be positive. From equations 
 (90) and (91) we have 
 
 „^==Vwr+V^' (92) 
 
 and 
 
 where only the plus sign is admissible since b^" is necessarily positive. 
 Substituting this value of &/ in equation (90) gives 
 
 "^~2^^~ 2^=V2- ^^ 
 
OCKAN ■iKMVElATVEKS 371 
 
 substituting this value of a, in equation (89) gives 
 
 h =: "i _ (95) 
 
 which agrees with the result already found on page 370, that —^^ 
 must be negative. From pages 369 and 370, e^Me'"'+^', where 
 a^a^±bii and b^ ± 0^= ±{2a^fji." — m'J&i- 
 
 The solution of equation (81) is therefore 
 
 6i = i/e'"2'±('"!'+°i"' (96) 
 
 where M and a^ are arbitrary constants, a^ and h^ are given by equa- 
 tions (94) and (95) and the same value of a^ is to be used with either 
 the plus or the minus sign before the expression in brackets. From 
 the properties of imaginary exponents equation (96) can be put in 
 the real periodic form 
 
 e = ("^y { A.shi {b,y + a,t) +B, COS {b,y + aj)] (97) 
 
 where A^, B^ and a^ are arbitrary constants. 
 
 Also since the differential equation is linear the sum of any number 
 of such expressions will be a solution. Therefore the following more 
 general expression 
 
 n = oo . 
 
 6=^2 e"-" iAnsm {b„tj + aj)+ B„ cos {b„y + a„t) | (98) 
 n = 
 is a solution, where A„, B„ and a„ are arbitrary constants and a„ and 
 b„ have the values 
 
 
 and 
 
 V2 
 
 K— ^ """ (100) 
 
 — ^w-c + V"'i* + 16ia*a,r 
 
 1 'a~ W 
 
 Denoting — - \/-^ bv A„ and — ' bv X the following approximate 
 fx. ^ 2 ■ /i^ ■ 
 
 expression for a„ and bn can be easily derived 
 
 a„=^+(l + /,,r + 7,/)A„ (101) 
 
 &„=(1_7(„=— 7,/)A„ (102) 
 
372 MISCELLANMOVS STUDIES 
 
 W - 
 where hn^=^ —4 — r, is small. If the velocity u\ equals 0, 
 
 a„^bn^^ — ~\"^ ^^'^ equation (98) reduces to the well known 
 
 form 
 
 )l=co 1 , , — ^ 
 
 «^0 
 
 ij^ \] (103) 
 
 A solution independent of the time results, if in equation (83) we 
 
 make h^O and a= — , or b = a^0. The result is 
 A*" 
 
 Wi 
 
 e = Ce'''''+D = Ce^i'+D ,(104) 
 
 where C and D are arbitrary constants. This expression for 6 can be 
 added to the right-hand member of equation (98), giving the more 
 general solution 
 
 (9 = Ce'^''+I> + e'"44,sin {b,ij + aj) + B.cos {b,y + a,t) I (105) 
 
 Solution for the case in which the vertical flow is a periodic 
 function of the time. 
 
 If tlie vertical velocity has the value w = iVi[l-\- r cos at] the 
 general equation (80) becomes 
 
 ^=,= |»-.,,[l + .-™.,l|^ (106) 
 
 Let 
 
 0^gay + fit) (107) 
 
 and substitute in equation (106), the result is 
 
 ^^— i^-a- + aw, [1 + r cos at] = (108) 
 
 from which 
 
 f{t) = {iJ.-a-—aw,)t — -^^^-^s\nat + c (109) 
 
 where c is an arbitrary constant. 
 
OCEAN TEMFKJiArUBES 373 
 
 Let 
 
 a^a^±hii or a= = ai= — bi'±2aj)ii (HO) 
 
 then 
 
 fl = Me"'^ - '''>'' + [/i'(V~''i'± 20,6,0 — «,"',+ 6,i<',i]< — -^^sinaf (HI) 
 
 the arbitrary constant of equation (109) being so chosen, that when 
 r equals zero, equation (111) reduces to equation (82) derived for a 
 constant velocity. 
 
 Eearranging the terms in the exponent of e in equation (HI) gives 
 
 (112) 
 
 Let 
 
 h,i2f,-a, — u\)=a, (113) 
 
 and 
 
 fj-'ia^- — &1-) — ait(.''i = (114) 
 
 as before (p. 370). Substituting in equation (112) and making use of 
 the properties of imaginary exponents gives 
 
 ^(ga,y—^^ sin alj j ^i siu la,f + b^tj ^—^ siu a< j 
 
 + Bcos(a,f + 6,y-^i^sina/)| (115) 
 
 wliere a, and &, have the values given on page 370 and A^, B^ and a^ 
 are arbitrary. The values 
 
 &j = ai = and «! = or A 
 
 are also consistent with equations (113) and (114) and lead to the 
 solution 
 
 (7eX!'+ Z) — C (1 — e-^"*'" "') e 
 
 \v 
 
 where C and D are arbitrary constants and A has the value — ^. 
 
374 MISCELLANEOUS STUDIES 
 
 The differential equation being linear, this solution can be added to 
 the right-hand member of equation (115) giving the solution 
 
 _"i«'i'' , { A ■ r i , 7 b,w,r . ^~| 
 _j_ gaii/ ^^"1"' j .4isni aj+h^u i-^-smaf 
 
 -|-BiCOS aj-^b^y — ^sina< I 
 
 = Ce^vJ^D + I 4, sin ^aj + &i(/ — ^i^^sin at~\ 
 
 + B, cosfa,^ + b,y _i^i!^ sin atl [ -(l — f"^ ■*'" A 
 
 I Ce^!' + ?"■!'( J, sin a,t-{-b^y i^^sin af 
 
 4- Bi cos 
 
 a,f + &,y— ^^i^sina/l") I (116) 
 
 which reduces to equation (105) corresponding to a constant velocity 
 
 when r^O. If the product of A. «i or /j, by f-^-sina* j is small 
 
 equation (116) can be transformed into the following apjiroximate 
 form, i-etaining only the first powers of the small quantities. 
 
 e = C gX!/ + Z) — C ('^^^sin aA c ^'J 
 + e'''viA,s'miaJ.-\-b,y) + S, cos {aj + b,y)+^r {B,b, — A,a,) 
 
 f cos (a — a J — ?),,(/) — cos (a + a,f + b^y) j — (Bi«i + ^i&J 
 
 ( sin (a + aj^ + ^1^) +sin (a aj — b^y) \ i (117) 
 
 Numerical values of the constants in the solution, determined from 
 temperature observations in the Pacific near San Diego. 
 
 It is well known that the waters of certain inshore regions, includ- 
 ing that off the west coast of North America, have a temperature 
 significantly below the normal for the latitude. Various explanations 
 of this phenomena off the California coast have been offered, but 
 (Holway. 1905, and McEwen. 1912, 1914, 1916) the only one .so far 
 proposed that is consistent with all of the known facts is that of an 
 upward flow of cold water from lower levels. 
 
OCEAN TEMPESATVRES 
 
 375 
 
 Assuming this upward flow to be the only cause of the temperature 
 
 reduction, the theory developed on pages 372-374 will now be applied 
 
 to the series of temperature observations made off the Coronado Island 
 
 about twenty miles from San Diego (McEwen, 1916, pp. 267, 268 and 
 
 pi. 26). It follows from Ekman's theory (p. 401) that the vertical 
 
 velocity off the California coast is proportional to the component of 
 
 the wind velocity parallel to the coast, and decreases in magnitude as 
 
 the distance from the coast increases. For this rea.son the velocitj' 
 
 estimated from the temperature data mentioned above will be less 
 
 than that nearer the coast. 
 
 Table 6 
 
 Average of observed and computed monthly temperatures of San Diego 
 
 at a series of depths. 
 
 Month 
 
 1 
 
 2 
 
 3 
 
 4 
 
 5 
 
 6 
 
 7 
 
 8 
 
 9 
 
 10 
 
 11 
 
 12 
 
 mean 
 
 
 
 Depth in meters 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 15.0 
 
 14,3 
 
 14.6 
 
 15,2 
 
 16,1 
 
 17.8 
 
 19,7 
 
 20,6 
 
 20,2 
 
 18,8 
 
 16.8 
 
 15,5 
 
 17,0 
 
 
 
 C 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 O 
 
 15.0 
 
 14.6 
 
 14.5 
 
 14.7 
 
 15,4 
 
 16,6 
 
 18,2 
 
 19.5 
 
 19,9 
 
 18,8 
 
 16,8 
 
 15,4 
 
 16,6 
 
 10 
 
 C 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 14.2 
 
 14.2 
 
 14.0 
 
 13,6 
 
 13,0 
 
 14,2 
 
 15,0 
 
 16,0 
 
 18,2 
 
 19,0 
 
 16,2 
 
 15,0 
 
 15,2 
 
 20 
 
 C 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 13.8 
 
 13.8 
 
 13.5 
 
 12,6 
 
 11,0 
 
 12,2 
 
 13,0 
 
 14,0 
 
 17,8 
 
 18,0 
 
 15,8 
 
 14,5 
 
 14,2 
 
 30 
 
 C 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 13.4 
 
 13.4 
 
 13.0 
 
 11 
 
 10 5 
 
 10,8 
 
 11,8 
 
 12,7 
 
 14,8 
 
 16,8 
 
 15 4 
 
 14,0 
 
 13,1 
 
 40 
 
 C 
 
 13.4 
 
 11.9 
 
 10.6 
 
 9,9 
 
 9,9 
 
 10,6 
 
 11 9 
 
 13,4 
 
 14,7 
 
 15,5 
 
 15,5 
 
 14,7 
 
 
 
 O 
 
 13 1 
 
 13.0 
 
 12,3 
 
 10 5 
 
 10 
 
 10,5 
 
 11 
 
 12,0 
 
 13,6 
 
 15,0 
 
 15 1 
 
 13,5 
 
 12,5 
 
 50 
 
 C 
 
 13.2 
 
 11.9 
 
 10 6 
 
 9,8 
 
 9,8 
 
 10,4 
 
 11,6 
 
 12,9 
 
 14,2 
 
 14,9 
 
 15,0 
 
 14,4 
 
 
 
 
 
 12. S 
 
 12 6 
 
 11.7 
 
 10,0 
 
 9 9 
 
 10,2 
 
 10,7 
 
 11,0 
 
 12,8 
 
 14 3 
 
 14 7 
 
 13 2 
 
 12 
 
 60 
 
 C 
 
 13.1 
 
 US 
 
 10.6 
 
 9,9 
 
 9,7 
 
 10,2 
 
 11,2 
 
 12,5 
 
 13,7 
 
 14,4 
 
 14,6 
 
 14,1 
 
 
 
 
 
 12.4 
 
 12 1 
 
 11 
 
 9,9 
 
 9,8 
 
 10,0 
 
 10,5 
 
 10 8 
 
 12,2 
 
 13,8 
 
 14,1 
 
 12,9 
 
 11,6 
 
 70 
 
 c 
 
 12.9 
 
 11.7 
 
 10.6 
 
 9,8 
 
 9,6 
 
 10 
 
 10 9 
 
 12,0 
 
 13,2 
 
 13,9 
 
 14 2 
 
 13,8 
 
 
 
 
 
 12.0 
 
 11.7 
 
 10.7 
 
 9,8 
 
 9,7 
 
 9,8 
 
 10,3 
 
 10,7 
 
 11,8 
 
 13,0 
 
 13,7 
 
 12,7 
 
 11 3 
 
 80 
 
 c 
 
 12.7 
 
 11.6 
 
 10.6 
 
 9,8 
 
 9,5 
 
 9,8 
 
 10,6 
 
 11,6 
 
 12,7 
 
 13,5 
 
 13,7 
 
 13,4 
 
 
 
 
 
 11.7 
 
 11.3 
 
 10 3 
 
 9,6 
 
 9,6 
 
 9,8 
 
 10 2 
 
 10 5 
 
 11 2 
 
 12,8 
 
 13,2 
 
 12,3 
 
 11 
 
 90 
 
 c 
 
 12.5 
 
 11 5 
 
 10 5 
 
 9,8 
 
 9,5 
 
 9,7 
 
 10,3 
 
 11 3 
 
 12,3 
 
 13,0 
 
 13,3 
 
 13,1 
 
 
 
 
 
 11.3 
 
 11,0 
 
 10.0 
 
 9,5 
 
 9,5 
 
 9.7 
 
 10,0 
 
 10,3 
 
 11,0 
 
 12,4 
 
 12,8 
 
 11,9 
 
 10,8 
 
 100 
 
 c 
 
 12.2 
 
 11.4 
 
 10 4 
 
 9,7 
 
 9,4 
 
 9,5 
 
 10,1 
 
 10,9 
 
 11,9 
 
 12,6 
 
 13,0 
 
 12,8 
 
 
 
 o 
 
 10.3 
 
 9.9 
 
 9 2 
 
 9 
 
 9 
 
 9,2 
 
 9,5 
 
 9,8 
 
 10 2 
 
 10 9 
 
 11 
 
 10,7 
 
 9,9 
 
 150 
 
 c 
 
 11.2 
 
 10.7 
 
 10 
 
 9,4 
 
 9,0 
 
 8,9 
 
 9,1 
 
 9,6 
 
 10 2 
 
 10,9 
 
 11,3 
 
 11,4 
 
 
 
 
 
 9.5 
 
 9,0 
 
 8.8 
 
 8,7 
 
 8,8 
 
 8.9 
 
 9,1 
 
 9,4 
 
 9,7 
 
 10,2 
 
 10,4 
 
 10,0 
 
 9,4 
 
 200 
 
 c 
 
 10.1 
 
 9.9 
 
 9.5 
 
 9,1 
 
 8,7 
 
 8,5 
 
 8,5 
 
 8.7 
 
 9.1 
 
 9,6 
 
 10,0 
 
 10,2 
 
 
 
 
 
 8.4 
 
 8.1 
 
 8.3 
 
 8,2 
 
 8,2 
 
 8.2 
 
 8,4 
 
 8.7 
 
 9,0 
 
 9,4 
 
 9,4 
 
 8,7 
 
 8,6 
 
 300 
 
 c 
 
 8.5 
 
 8.5 
 
 8.4 
 
 8,2 
 
 8,0 
 
 7,8 
 
 7,7 
 
 7.7 
 
 7,8 
 
 8,0 
 
 8,2 
 
 8,4 
 
 
 
 
 
 7.8 
 
 7.7 
 
 7,6 
 
 7,5 
 
 7,6 
 
 7,6 
 
 7,7 
 
 7,8 
 
 8,0 
 
 8,4 
 
 8,4 
 
 7,8 
 
 7,8 
 
 400 
 
 c 
 
 
 
 7.4 
 
 7.5 
 
 7.5 
 
 7,5 
 
 7,4 
 
 7,2 
 
 7,1 
 
 7,0 
 
 7,0 
 
 7.1 
 
 7,2 
 
 7,3 
 
 7,0 
 
 500 
 
 c 
 
 6.8 
 
 6.8 
 
 6.8 
 
 6,8 
 
 6,7 
 
 6,7 
 
 6,7 
 
 6,6 
 
 6 6 
 
 6 6 
 
 6,7 
 
 6,7 
 
 
376 
 
 MISCELLANEOUS STUDIES 
 
 From table 6, which gives the observed temperature averages at 
 different depths and months, the constants of ecjuation (116) or the 
 simpler approximate form equation (117) will now be determined. 
 The mean annual temperature (9„, is given by the first two terms 
 
 which can be put in the linear form 
 
 'Logc{0,n — D)=hogcC + Xij. 
 
 (118) 
 
 (119) 
 
 Assuming different values of D, plotting the results, and selecting the 
 value of D, for which the points fell most nearly on a straight line, 
 resulted in the following values of the con.stants: 
 
 Z) = 5.6, 
 
 or 
 
 C = 8.3, X = 
 : 5.6 + 8.3 e-""*" 
 
 -.004 
 
 (120) 
 
 where y is the depth in meters. 
 
 The satisfactory agreement between the computed and observed 
 values of 6„„ shown by table 7, proves that the form of the function 
 deduced from theory differs but little from the true form. 
 
 Table 7 
 Computed and ohserved mean annual temperatures at a series of depths from 
 
 40 to 700 meters 
 
 Depth 
 
 40 
 
 50 
 
 60 
 
 70 
 
 80 
 
 90 
 
 100 
 
 150 
 
 200 
 
 300 
 
 400 
 
 500 
 
 600 
 
 700 
 
 Om computed 
 ^mob.served. 
 Difference... 
 
 12°7 
 13.1 
 -.4 
 
 12°4 
 12.5 
 - 1 
 
 12?1 
 12.0 
 
 + .1 
 
 11?9 
 11.6 
 
 + .3 
 
 11°6 
 11.3 
 
 + .3 
 
 11?4 
 11.0 
 
 + .4 
 
 11?2 
 10.8 
 
 + .4 
 
 10^2 
 9.9 
 
 + .3 
 
 9?3 
 9.4 
 -.1 
 
 8?1 
 8.6 
 -.5 
 
 7?3 
 7.8 
 -.5 
 
 6?7 
 7.0 
 -.3 
 
 6?4 
 6.3 
 
 + .1 
 
 6°0 
 5.5 
 
 + .5 
 
 The time of minimum wind velocity is in December, that of the 
 maximum is in July (McEwen, 1912, p. 265), and the magnitude of 
 the wind velocity, and therefore the vertical velocity of the water, 
 
 is approximately proportional to 1 + r cos — where t = 12 corre- 
 sponds to the time halfway between Jime and July and r = 0.2. 
 Also since, as showii by table 6, the temperatures have the same period 
 
 as the wind, o, = a = - in equation (117). In order to determine 
 6 
 
 the remaining constants substract from the observed temperature for 
 
OCEAN TEMPEEATVUES 377 
 
 each month and depth the observed mean annual temperature for 
 that depth. Then subtract the expression 
 
 from each of these values using a provisional value of u\. Then fit 
 
 Me"^" cos {at + b,lj — e) 
 the equivalent of the expression 
 
 e°i!')4isin {at + &iy)+B, cos {at + bji) - 
 
 (p. 374) to these remainders, thus determining M, e'. a^ and h^. 
 The last part of equation (117) is neglected at first and its value 
 
 estimated later. Assuming w, to be — 31 -^ the espres.sion 
 
 month 
 
 ■ C ^"^ '' (sin aOe'^" becomes — .38 ( sin -^ je' 
 The following formulae corresponding to the special case when 
 a:=ai= — are useful in determining a^, &i, w^ and /*-, and follow 
 
 from equations (89, 91, 94 and 95) remembering that A= —^. 
 J41 .524 _ .524 
 
 (121) 
 
 "•■■'''' ,^h = ,r^^ (122) 
 
 {X^y = ^, J^. ^ (123) 
 
 _\Mf^'i>.)r-) ^i 
 
 «! can readily be determined from the rate at which the annual 
 range varies with the depth. A was determined from the mean annual 
 temperature at different depths. These values substituted in equation 
 (122) give firbi from which n" and h^ can be found by substituting 
 in equations (123) and (122), and finally the product kfx- gives the 
 velocity tt'i.* 
 
 * If \ and fc, are given, the following equations derived from equations 90 
 and 95 can be used for computing the remaining quantities: 
 
 — a, — Xa, 
 
 "b,V(2b.)-' + X'' ' 6.VT26,)' + X=' ' 2 2 
 
 , a,=;.-5V(26,r + X' 
 
378 MISCELLANEOUS STUDIES 
 
 The constants obtained in this way are listed below : 
 
 A =—.004. a^ = — .008, ^- = 7760, ?)i = — .00560, «■ =— 31 E?!££5 
 
 month 
 
 A,= — —-*/ — =— -0058, .¥ = 4.40, e' = — 6r. 
 M > 12 
 
 Assuming t = 1 for January, and using the same origin for 
 determining the time in the expression for wind velocities, the expres- 
 sion for the temperature becomes 
 
 ^ = 5.6 + S.Se-""*" + 4.4e— "^J- cos (SOt — .32y + 61) ° 
 — .HSe--'">'y cos (30f + 75) ° (124) 
 
 remembering that C = 8.3 and D = 5.6 (p. 376). 
 
 The value of /.i-^7760, when expressed in c. g. s. units is 
 
 ^^•^Sox 24X3600-^°' 
 
 which is about 25,000 times the laboratory value (Wegemann, 1905a, 
 p. 139). But this quantity y.- is the same as the Misclningsintensitdt 
 (Jacobsen, 1913, p. 71), which is a measure of the rate of transfer of 
 salts, heat, or other properties of sea water arising from the mixing 
 of water particles in the alternating circulation (p. 368). Suppose the 
 diffusion of salts and the molecular conductivity of heat to be negligible 
 in comparison to the rate of transfer due to the alternating or 
 reciprocal changes in the positions of the water particles : then as 
 Jacobsen (1913, p. 71) says, the value of this coefficient, the 3Iischungs- 
 iiifciisifaf, determined from au.y of the properties should be the same 
 under tlie same conditions. He found values ranging from 1.9 to 3.8 
 from observations on currents and the distribution of salinities in the 
 sea near Denmark. From observations on tidal currents and salinities 
 in a neighboring region he obtained the values ranging from 0.3 to 
 11.4. The value 30 obtained from temperatures in the San Diego 
 region is of the some order of magnitude, but the intensity of the 
 circulation in the two regions would probably be different, hence the 
 coefficients would be expected to differ. 
 
 The idea regarding the alternating motion of water in the ocean 
 held by Kriimmel and Ruppin (1905. p. 36) may be summarized as 
 follows : The coefficient of viscosity determined from laboratory experi- 
 ments, in which the motion of the water is slow and takes place along 
 
OCEAN TEMPEItATrSES 379 
 
 parallel surfaces, varies from .008 to .02 for a wide range of tempera- 
 tures and salinities. But the idea of laminar flow no longer holds 
 when one considers the motion of whole volumes of water, hundreds 
 of meters in thiclaiess, throughout which there is a pressure gradient, 
 as is often the case in oceanographie problems. In such cases one 
 should not use inncre Reibung (viscosity), but Massenwidersland 
 (hydraulic friction). 
 
 The character of the motion is no longer simply laminar (one of 
 simple sliding to and fro, parallel to a given plane) but the particles 
 of fluid leave their surfaces and move in vortices along stream lines 
 transverse to the laminar motion. Thus a much greater resistance is 
 developed. For example, the values of the coefficient of viscosity 
 obtained by Nansen in his ocean researches are 200 to 40,000 times 
 the laboratory value. 
 
 The same general idea has been successfully applied to certain 
 meteorological problems i-elative to wind, temperature, and humidity 
 by G. I. Taylor (1915). He found that the transfer, in a vertical 
 direction, of heat and water vapor in the atmosphere followed the law 
 of heat conductivity in solids, and that the effect of friction on the 
 motion of the air could be taken into account by substituting in the 
 general equations of motion a quantity called "eddy viscosity" for 
 the laboratory value of the coefficient of viscosity. 
 
 From the observed relation of air temperature to height off the 
 coast of Labrador, lie obtained values of the coefficient ; of " eddy 
 conductivity" from .57 X lO'' to 3.4 X 10^, corresponding to wind 
 velocities varying from 2 to 3.4 Beaufort. Also from observations on 
 the relation of wind velocity and direction to height he obtained values 
 of the coefficient of eddy viscosity varying from .77 X 10^ to 6.9 X 10^. 
 The values of these coefficients are more than 10,000 times the labora- 
 tory values, the ratio being of the same order of magnitude as that 
 obtained for sea water. 
 
 In general, comparisons of his theoretical results with observations 
 indicated a very satisfactory agreement. An especially good agree- 
 ment was found between the predicted and the obsei'ved values of the 
 angle between the wind and the horizontal pressure gradient at 
 different levels. 
 
380 
 
 MISCELLANEOUS STUDIES 
 
 JANUAEY 
 
 JULY 
 
 100 
 350 
 
 130 200 
 
 400 Meters 
 
 350 Meters 
 
 Figs. 4, 5, 6, 7, 8, and 9. Curves Bhowintr the theoretical relation of ocean 
 temperatures to depth in a region approximately eight miles west of the 
 Coronado Islands. The crosses ( + ) corresjiond to observed temperatures. 
 
OCEAX TEMPEBATIEES 
 
 381 
 
 20° 
 19 
 
 
 
 
 
 Al 
 
 KII, 
 
 
 : 
 
 OCTOBER 
 
 18 - 
 
 
 
 
 
 
 
 
 - 
 
 A f 
 
 17 _ 
 
 
 
 
 
 
 
 
 - 
 
 \ 
 
 16 - 
 
 
 
 
 
 
 
 
 - 
 
 \ 
 
 15 - 
 14 _ 
 13 _ 
 12 _ 
 
 11 _ 
 
 ^ 
 
 * 
 
 
 
 
 
 
 _ 
 
 \ 
 
 10 _ 
 
 
 vLi. 
 
 ^ 
 
 
 ~T 
 
 ' 4 T 
 
 -^— _ 
 
 - 
 
 ^^x»^ 
 
 9 
 
 
 
 1 
 
 t"^^^^^ * 
 
 8 _ 
 
 
 
 
 
 
 
 
 _ 
 
 ^-— ^-_ ^ ^^^ 
 
 7 
 
 
 
 
 
 
 
 
 
 "" t ~~ -. 
 
 
 
 1 
 
 
 
 1 
 
 
 1 
 
 1 
 
 1 1 1 1 1 1 1 
 
 50 
 
 150 
 
 210 
 
 
 250 
 
 50 
 300 
 
 100 
 360 
 
 150 200 
 
 400 Meters 
 
 250 
 
 350 Meters 
 
 IS"-, 
 
 MAY 
 
 _ 
 
 NOVEMBER 
 
 17 - 
 
 
 
 _ 
 
 X 
 
 16 - 
 
 ^^ 
 
 
 _ 
 
 \ 
 
 15 - 
 
 + + 
 t 
 
 
 _ 
 
 x^ 
 
 14 - 
 
 
 
 
 — 
 
 x^^ 
 
 13 - 
 
 J 
 
 
 
 _ 
 
 ^\^ 
 
 12 - 
 11 _ 
 
 * 
 
 V 
 
 
 - 
 
 \^^ 
 
 10 _ 
 
 Vj 
 
 ,__v 
 
 - 
 
 ^~^^v.^^^ 
 
 9 - 
 8 - 
 
 7 
 
 
 "^ '^^^ TT-f- 
 
 ______^ 
 
 ^^^^^--....^^^ 
 
 
 
 " 
 
 — - — — _ ^~'~~"~~-^~^^__ 
 
 
 
 
 J 1 1 
 
 1 
 
 1 1 1 1 1 1 1 
 
 5J 
 
 10 J 
 
 
 25,) 
 
 50 
 300 
 
 100 
 350 
 
 ISO 200 
 
 400 Meters 
 
 250 
 
 300 
 
 350 Meters 
 
 DECEMBER 
 
 100 
 350 
 
 150 200 
 
 400 Meters 
 
 1 r 
 
 300 350 Meters 
 
 Figs. 10, 11, 12, 13, 14, and 15. Curves showing the theoretical relation of 
 ocean temperatures to depth in a region approximately eight miles west of the 
 Coronado Islands. The crosses ( + ) correspond to observed temperatures. 
 
382 MISCELLANEOUS STUDIES 
 
 Comparison of theoretical and observed monthly temperatures at 
 depths from 40 to 600 meters in the San Diego region. 
 The values computed from equation (124) and entered under the 
 observed temperatures given in table 6 are seen to agree well with the 
 mean of the observed values, thus proving the approximate correctness 
 of the form of the function deduced from theory. These computed 
 values and those from table 12 for the surface are also shown graph- 
 ically by figures 4 to 15. on which are entered a number of points 
 corresponding to actual observations (Michael and ilcEweu, 1915, 
 1916). 
 
 Solution of the prohlmi of temperature reduction due to upwclling 
 with application relative to the 40 meter level in 
 the San Diego region. 
 In the relation of mean annual temperature to depth 
 
 e,„=Cc>^!>->rD = <i> (125) 
 
 Wi 
 
 deduced from the differential equation (80), A= — ^ equals the 
 
 1^ 
 
 velocity divided by the difFusivity, but C and D are con.stants of 
 integration. From observations of the mean annual temperature at 
 a series of depths thase constants can be determined as was done on 
 pages 372 to 376, and they correspond to the particular physical con- 
 ditions under which the observations were made. For the same value 
 of D, the deep water temperature, but a different value of one of the 
 physical conditions, say the velocity Wj, what will the temperature <^ 
 be? To answer this question it is necessary to know the relation of 
 each constant to the velocity M',. The relation of X to n\ is known 
 and it remains to find the relation of C to u\. 
 
 In the limiting case in which A = 0. denote the new value of the 
 constants by C. D' and A'; then expanding the exponential gives 
 
 cf>' = C'e^'^ + D'= C'X'u + C + D' = B'g + /)/ (126) 
 
 where B'^C'X' is the constant temperature gradient corresponding 
 to zero vertical velocity. 
 
 ^^* C=— A(A) (127) 
 
 A 
 
 ^""^ D^D'f.W (128) 
 
 where /,(0) =/„(0) = 1, since, as A = 0. C = (" and D= D'. 
 
OCEAN TEMPEBATUSES 383 
 
 Substituting in equation (125) gives 
 
 .^ = + ^A(A)e^'' + Z?'A(A) (129) 
 
 where the forms of the functions /i(A) and /^(X) are to be deter- 
 mined. At the greatest depth, i/, for which the theory is valid, assume 
 the temperature to have the constant value <^i for all values of X. 
 What effect will a vertical velocity have on the temperature above this 
 level? The right-hand members of the equations (126) and (129) are 
 equal for y = y^, since <^i is assumed to be independent of \ at that 
 depth, that is, 
 
 4>^=-^ fr{k)c^y^+ D'f,{\) =B'rj, + D' + C (130) 
 
 A 
 
 Therefore from equations (129) and (130) 
 <#. = ^A(A)eX"+i)7.(A)— •f^A(A)eX^'— i)73(A) + B'2/i+^'+C" 
 
 A A 
 
 = ^/\(A)[eX''-eX«']+ B'y, + P/. (131) 
 
 A 
 
 Subtracting the general value of <^ given by equation (131) from 
 the particular value <^' corresponding to the case of no upwelling 
 given by equation (126) gives 
 
 ^'_^ = B'(y — 2/J— ^A(A)(f^"— f^^") = A<^ (132) 
 A 
 
 the reduction in temperature due to the upwelling velocity rt\ = — ju-A. 
 
 It remains to determine B' and the form of the function f^iX). 
 
 The temperature change due to the variation of velocity with the 
 
 time was found to be approximately 
 
 — Ce'^« (l — e- —■ ^i° "' ) 
 
 (equation 116), where the velocity is 
 
 w = iVi{l -\- r cos at) 
 
 and the value of r in the remaining terms is neglected, for the fol- 
 lowing reasons. The values of the constants A^ and B^ depend mainly 
 on the seasonal variation in temperature due to radiation ; if there 
 were no such variation they would be zero, in which case the variation 
 in temperature with respect to time would be due entirely to that of 
 the wind. The temperature reduction is therefore approximately 
 
384 
 
 MISCELLANEOUS STUDIES 
 
 where R is the constant average reduction. A variation of the velocity 
 from its minimum to its maximum value, that is, from to 2u\ as at 
 varies from to 'Itt produces a variation of temperature reduction from 
 
 lR+e^y{l — e^\ C to R+e^yfl~e-^\ cl 
 
 that is, as the velocity varies from to 2u\ the temperature reduction 
 increases bv the amount 
 
 \«A , R' 
 
 -)-?]} 
 
 from which the temperature reduction due to tlie velocity u\ is 
 
 A^=C{c^y)l(e^^-e-^\^Ce^y,mh{^\ (133) 
 
 The approximate temperature reduction deduced by two inde- 
 pendent methods is given by equations (132) and (133), respectively. 
 Equating these two values and using ecjuation (127) gives 
 
 Ce^" sinh 
 B' 
 
 A (A) e'^" sinh (^\ 
 
 A 
 
 = A<^ 
 
 (134) 
 
 Solving for tlie unknown function /](A) gives 
 
 A (A): 
 
 ^(.y — yi) 
 
 e\y — gXi/i _|_ g\!/ sinh 
 
 (^-?) 
 
 (135) 
 
 Since A is assumed to be independent of y. the variation of the 
 right-hand member of equation (135) with respect to y is a measure 
 of the error in the two expressions for A<^. Also a comparison of the 
 
 AC 
 
 theoretical temperature gradient. B'-- 
 
 7.(A) 
 
 (equation 126). which 
 
 would be expected in case of no upwelling with observations of deep 
 water temperature in sucli regions affords an additional test of the 
 theorj-. The following values of the constants of equations (125) and 
 
OCEAN TEMPEEATVBES 
 
 385 
 
 (133) computed on pages 376 to 378, C = 8.3, D^5.6, X-- 
 
 -.004, 
 
 iv, = —n, 2A = 600, ^^ = 
 
 KlVi 
 
 .237 and sinh ^^^^^ = .24 are used in 
 
 a 
 
 table 8, which gives the relation of f, (A) (equation 135) to y. 
 
 Table 8 
 The variation of f, (X) ivith respect to y. 
 
 y 
 
 Hy-y^) 
 
 fX!/ — fX!/i_|_eX!/sinh( — ] 
 
 /iW 
 
 30 
 
 2.28 
 
 1 009 
 
 2 26 
 
 40 
 
 2 24 
 
 0.965 
 
 2. 32 
 
 50 
 
 2.20 
 
 924 
 
 2.38 
 
 60 
 
 2.16 
 
 0.885 
 
 2.44 
 
 70 
 
 2.12 
 
 0.846 
 
 2.51 
 
 SO 
 
 2.08 
 
 0.809 
 
 2.57 
 
 90 
 
 2.04 
 
 0.744 
 
 2.63 
 
 100 
 
 2 00 
 
 0.740 
 
 2.70 
 
 200 
 
 1 60 
 
 466 
 
 3.59 
 
 300 
 
 1 20 
 
 282 
 
 4.26 
 
 400 
 
 80 
 
 160 
 
 5 00 
 
 500 
 
 0.40 
 
 076 
 
 5.26 
 
 600 
 
 0.00 
 
 0.022 
 
 00 
 
 700 
 
 
 
 
 800 
 
 
 
 
 
 
 Table S .shows the variation of A (A) with respect to depth to be 
 less than 20 per cent in the depth interval from 30 to 100 meters, 
 hence the two methods (equation 134) of estimating A(/) are in good 
 agreement within this interval. The value of /i(A), corresponding to 
 30 meters, the .smallest value of y for which the theory is valid, gives 
 the best estimate of /i(A), and hence of B', since the variation of /i(A) 
 with respect to y is least for small values of y. Substituting the 
 
 numerical values gives 
 
 AC 
 
 B'-- 
 
 7i(A 
 
 -=—.0148 
 
 (136) 
 
 the mean annual theoretical temperature gradients that would be 
 expected at latitude 32° 30' if there were no up welling and the other 
 conditions remained the same as those prevailing when the observations 
 in the San Diego region were made. The observed temperature 
 gradient at the depth of 600 meters not near shore would be inde- 
 pendent of seasonal variations and would be but little affected by 
 horizontal currents, and is in the depth interval of the observations 
 from which the constants of the theoretical formula for mean annual 
 temperatures were computed. 
 
386 
 
 MISCELLANEOUS STVDIES 
 
 Table 9 
 Vertical temperature gradients in degrees per meter at the depth of 600 meters. 
 
 Indian Ocean 
 
 South Atlantic Ocean 
 
 North Atlantic Ocean 
 
 Lat. 30°to35S 
 
 Lat. 30° S 
 
 Lat. 30° N 
 
 -.004 
 
 -.0135 
 
 -.0025 
 
 -.006 
 
 -.0160 
 
 -.0080 
 
 -.007 
 
 -.0160 
 
 -.0090 
 
 -.0085 
 
 -.0165 
 
 - 0110 
 
 -.0095 
 
 -.0210 
 
 - 0150 
 
 -.0115 
 
 -.0260 
 
 -.0155 
 
 -.0130 
 
 
 -.0160 
 
 -.0140 
 
 
 -.0175 
 
 
 
 
 
 North Pacific Ocean 
 
 
 
 Lat. 30° to 35° N 
 
 
 - 0075 
 
 -.0140 
 
 -.0215 
 
 -.008 
 
 -.0150 
 
 - 0215 
 
 -.0095 
 
 -.0150 
 
 -.0220 
 
 -.0095 
 
 -.0150 
 
 -.0235 
 
 -.0100 
 
 -.0155 
 
 -.0245 
 
 - 0100 
 
 -.0155 
 
 - 0250 
 
 -.0105 
 
 -.0160 
 
 -.0260 
 
 -.0120 
 
 -.0170 
 
 -.0290 
 
 -.0120 
 
 -.0195 
 
 -.0305 • 
 
 -.0125 
 
 -.0205 
 
 -.0315 
 - . 0350 
 
 
 
 
 Table 9 shows the observed temperature gradient at the depth of 
 600 meters and at the approximate latitude 32° 30', corresponding to 
 widely different positions between latitudes 30° S to 35° S and 30° N 
 to 35° N, and their average is probably a good approximation to the 
 normal gradient at 600 meters. The average of the 22 observations 
 in the Indian and Atlantic oceans (Schott, 1902, pp. 158-160) is 
 — .0126 degrees per meter, and the average of the 31 observations in 
 the North Pacific (Makaroff, 1894, pp. 456-464) is —.0179. The 
 theoretical result — .0148 agrees well with these observations. 
 
 The reduction of the mean annual temperature at a given level y 
 for a given velocity u\ is proportional to CeM' (equation 133), that is, 
 the reduction is proportional to the difference between the tempera- 
 ture at the depth y and the constant D. Therefore the temperature 
 reduction corresponding to a given month is proportional to the differ- 
 ence between the temperature at that time and the same constant D. 
 
 That is, 
 
 A<i>i _ 4>t — D 4>t~D 
 
 A^ ~ \Ce^y + D) — D ~ <t>,u—D 
 
 (137) 
 
OCEAN TEMPEh'.irUEES 
 
 387 
 
 where <t>t is the temperature at the time t, A<j)t is the corresponding 
 reduction, <^„, is the mean annual temperature, and A<^ is the reduction 
 of the mean annual temperature. Substituting the numerical values 
 for ij equals 40 meter.s, from page 385, equation (137) reduces to 
 
 ^<^' = iIt^(8.3) (.852) (.24)= (^1^) 1.7 = . 227<^, _ 1.27 
 
 (138) 
 Svibstituting the observed values of <^, at the depth y equals 40 from 
 table 6 in equation (138) gives the values of A<^( entered in the 
 second line of table 10. 
 
 Table 10 
 The monthly temperature reduction at the depth of 40 meters near San Diego. 
 
 t 
 
 1 
 
 2 
 
 3 
 
 4 
 
 5 
 
 6 
 
 7 
 
 8 
 
 9 
 
 10 
 
 11 
 
 12 
 
 AOt 
 
 1.75 
 
 -.10 
 
 1 65 
 
 1.75 
 
 -.27 
 1.48 
 
 1.68 
 
 -.37 
 
 1.31 
 
 1.23 
 -.37 
 0.86 
 
 1.11 
 
 -.27 
 0.84 
 
 1.18 
 
 -.10 
 
 1.08 
 
 1.41 
 
 -.10 
 
 1.51 
 
 1 61 
 
 .27 
 1.88 
 
 2 09 
 
 .37 
 
 2.46 
 
 2.55 
 
 .37 
 
 2.92 
 
 2.23 
 
 .27 
 
 2.50 
 
 1.91 
 .10 
 
 2.01 
 
 But the vertical velocit 
 
 y is u\ ( 1 + .2 cos -M, 
 
 and because of its 
 
 ,X"'i'' 
 
 variation with respect to time a correction equal to e"'^ C ~^ sin at 
 (equation 116) must be added to these values. The correction for 
 this ease is 
 
 .386-1" cos (30f + 75)''=.324cos(30i+75)» = A<^r 
 
 (last term of equation 124. p. 378), and the values are entered in the 
 third line of table 10. Finally the fov;rth line gives A<^( -j- A<^,. = A(9( 
 the total temperature reduction at the depth j/ ^ 40 meters. 
 
 Computing C from equations (136) and (135) and substituting 
 the result in equation (133) gives the expression 
 
 (y — 600) (—.0148) sinh (^^^ e^wk 
 ■'° + '"''' (^4062 j ' "'" 
 
 e 7760 e 7760 
 
 for the mean annual reduction in temperature at the depth ij xlue 
 to the mean annual velocity of upwelling u\. The values of this ex- 
 pression corresponding to a series of values of u\ (expressed in meters 
 per month are presented in table 11 for the depth y = 40 meters. 
 
388 
 
 MISCELLANEOUS STUDIES 
 
 Table 11 
 
 Theoretical temperature reduction at the depth of 40 meters corresponding to a 
 
 series of values of the vertical velocity «',. 
 
 W^ 
 
 
 
 10 
 
 20 
 
 30 
 
 40 
 
 50 
 
 60 
 
 70 
 
 80 
 
 90 
 
 100 
 
 Temperature 
 reduction 
 
 0°0 
 
 0°38 
 
 0°95 
 
 1°67 
 
 2°48 
 
 3°34 
 
 4°1S 
 
 5°00 
 
 5°80 
 
 6°48 
 
 7°07 
 
 From plates 24 and 36 (MeEwen, 1916), and from an examination 
 of the temperature data (Michael and MeEwen, 1915, 1916) it appears 
 that the temperature reduction at the 40 meter level, inshore averages 
 from 2° to 3° more than that ten miles offshore. Also surface tem- 
 peratures as much as 9° below the normal have at times been found 
 close inshore in July or August, when the normal temperature is 
 about 23°, while the reduction of the surface temperature ten miles 
 offshore does not at any time exceed 3?5 (table 12). If, corresponding 
 to a temperature reduction of 1?7 ten miles offshore, the inshore 
 reduction is 1?7 plus 2?5 equals 4?2. the corresponding velocity of 
 upwelling inshore would, from table 11, be about twice as great as 
 that ten miles offshore. 
 
 Deduction op the Change in Surface Temperatures Due to a 
 
 Vertical Flow op Water near the Surpace 
 
 Statement of assumptions and mathematical formulation of the 
 
 2)rohlem. and solution for the case in which the 
 
 flow is constant. 
 
 The temperature reduction at the 40 meter level due to upwelling 
 
 applies also to the surface water, but owing to the upwelling in this 
 
 upper level the surface temperature will be still more reduced. The 
 
 time rate of change of temperature in the case of no resultant flow 
 
 given by the differential equation (10) on page 342 plus the term 
 
 — u^ — will be the modified rate due to the vertical flow »•„ as on 
 
 page 354. Hence the new differential equation corresponding to a 
 vertical flow near the surface is 
 
 dd 
 
 ^ = Be-'"y[{a, + a.a) cosat + a,.T + l] — kie—e,)—w „ 
 
 OS 
 
 dy 
 
 (139) 
 
 where k = k^e' 
 
 iiV 
 
OCEAN TEMPEBATVRES 
 
 389 
 
 In ease »'o is constant, assume for the form of the solution 
 
 e = A(y)smaf + /,(y)eosa< + /3(2/)x + /-,(y)+e, (140) 
 
 where 6^ is independent of t and x. Substitute this expression in 
 equation (139), thus obtaining the following equations: 
 
 [". 
 
 (sinaOUA(y)— A;,e-''"/i(y)— «• 
 
 
 = (141) 
 
 (cos 
 
 aO [-aA()/) + Bf-''.''(«,+a,x)-A-,e-".-''/,(y)-«'„^?^] =0 
 
 (142) 
 x\^a,Ber^^y-Ke-'>^yf,{y)~ic,'^-xc,,'^l^'^ =0 (143) 
 
 [ 
 
 Be-''^>'~k,e-''^t'f,iy)—w, 
 
 , rf^i 
 
 dij 
 
 " (hi J 
 
 (144) 
 
 in which r is regarded as a constant. 
 
 In ease of no vertical velocity the variation of (6 — 0^) with respect 
 to time is small compared to its mean value, which is independent of 
 the time (p. 344). Therefore, assuming this to be true for the present 
 case, a good approximation would result if the constant part were 
 multiplied by the correct factor ],\e~'"y and the variable part by an 
 average value k. Making this change in equations (141) to (144) 
 we have the much simpler ones 
 
 af.Ay)-kf;(y)-w,^h^=o 
 
 dlJ 
 
 — o.fAll)—J;f.,{il)—u\ 
 
 df,{y) 
 
 dy 
 
 k.e-^^"fAy) + "'o^^^= a^-Be-^'^ 
 
 k,e-^>^yf. 
 
 .'»'+- (^-t)^ 
 
 : Be-^^i' 
 
 (145) 
 
 -Be-^^y{a, + a._x) (146) 
 
 (147) 
 (148) 
 
 From equations (145) and (146) 
 
 Va=+(fc,) 
 where l\,^^k — »•„&! and tan Co^ 
 
 cos (a< — e,) + /3(y)x + /,(y) + e, (149) 
 
 K 
 
390 MISCELLANEOUS STUDIES 
 
 From equation (147) 
 
 Ba. 
 and from equation (148) 
 
 f.{y)=~ (150) 
 
 dd 
 where-T-^equals — b^M^e-"'-^ which gives 
 
 0,=M,e-^^«+e\_ (152) 
 
 where 0\ is a constant of integration. Substituting these values in 
 equation (149) gives 
 
 ^^g(a,+ a..)e-^.v^^^ („^-e.) + ^-+ ^^ + '^^-^'r" + ^ + 3/..-^ 
 
 Va^+(A%)- ^'' ^'^ 
 
 (153) 
 As on page 343 
 
 ,,__ B(a, + a,r )_ ^^^^^ 
 
 yja' + iKV- 
 and since the change in temperature due to upwelling near the surface 
 
 is due to the gradient — , only terms involving y should be functions 
 dy 
 
 of ?('(,. Moreover, as «•(, increases indefinitely the temperature at the 
 
 lower boundary of this upper layer should approach the constant value 
 
 Therefore, as on page 344 
 
 ■ ^^"-' .,e-^^. + ^ + e, (155) 
 
 which reduces to equation (22), (p. 344), when M'n = 0, except that 6., 
 takes the place of d^ and the average value of k" takes the place of 
 [fc,e"''''^"-'"]-. The latter quantities are small compared to a- and 
 therefore the difference between their values makes biit little difference 
 in the result. This difference or error comes from the approximation 
 made in solving the differential equation (139), (p. 388). 6., is found 
 by subtracting the temperature reduction due to upwelling at and 
 below the surface (3 meter level) from 0^, therefore in case of no 
 
OCEAN TEUPESATUEES 
 
 391 
 
 upwelling (9, equals ^3. A second approximation to the solution of 
 
 equation (139) can be found bv substituting for — - its value from 
 
 dy 
 
 the first approximation. The solution is then reduced to 
 
 6= _ - _ g + ^ [cos (at — t.,) 1]-| J- 
 
 Ba^xe^^ B_ 
 
 (156) 
 
 where 
 
 tan e, = 
 
 
 k 
 
 a'+(k,y 
 
 If M'0^0, equation (156) is identical with equation (22). 
 
 Solution for the case in ^vhich the flow is a periodic fimction 
 
 of the time. 
 
 If the vertical velocity equals 
 
 M'o[l + »' COS (at — £1)] 
 
 the time rate of change of temperature can be obtained from the dif- 
 ferential equation (139) in which the last term is multiplied by 
 
 [1 -|- r COS {at — fi)]. 
 
 The new differential equation is therefore 
 
 ^=Be-'>^y[ia,+aa)cosat+a,x+l]—k{e~d,)—iv„[l+rcos(at—e,)f^ 
 at ' oy 
 
 (157) 
 
 Let d' be the solution already found when r = 0, and let 6" be the 
 correction due to r, then 6^^d'+d", and the result of substituting 
 in equation (157) is 
 
 -^ + ke" + w^— + w^r cos (at — t,)— + ?(•„/• cos (af — fj— = 
 
 (158) 
 
392 MISCELLANEOUS STUDIES 
 
 Equation (155), using y instead of {y — 3), gives approximately 
 d' and 
 
 ^y ^a- + ihy 
 
 [cos (a^ — £,)— lle""'^ (159) 
 
 Substituting this value of in equation (158) and letting 
 
 0" ^= ve'^'-v results in the ordinary differential equation 
 
 -77+ At — H'oOif [«■„)• cos (at £j] -_ [cos {at — u) — 1] 
 
 — ru^Jj^v cos, {at — ej (160) 
 
 where v is a function of t only. Let 
 
 A- — tvjj^ = Av, and Ji ., = — ■ 
 
 Va^+Av 
 
 then equation (160) can be put in the form 
 
 ^ + A-,t^ = — rJ7„ jcos (af — ej— ^[eos (2a/ — £3) + cos ej ■ 
 
 + rWgb^v cos {at ej (161) 
 
 where e^ + tj ^ £3 and £, — t^ = t^. 
 
 When the vertical velocity is directed upward (/■„ is negative, 
 therefore 
 
 A-,= (A- — H'„^'i)> I n^i I 
 
 and the last term of equation (161) can be neglected in the first 
 approximation, which is 
 
 v = e-'"-*^—rM, I f^='feos {at — e,)— -cos (2a; — £,) — ^cos £X/i + C - 
 — ( Fa COS £, + l\ sin £, ~1 ■ , FA", cos £, — a sin £,~| 
 
 = -'--^^^ { L ^F+k? J "° "' + [ ' a- + A/ J-^^^ "' 
 
 r2a cos£,+ fc, sin£,~| . „ J 
 
 rA\cos£s — 2asin£,,~] cos £4 ) ,.,„, 
 
 - 1 2{ia^-+kj) r' ^"'-^kT \ (1^2) 
 
 The arbitrary constant C is 0, since from physical considerations 
 the solution must be a periodic function of the time. Substituting 
 
OCEAN TEMPESATVttES 
 
 393 
 
 {ij — 3) for y in the exponential vc''''" and adding the result to from 
 equation (156) gives the following approximate value of the tem- 
 perature in case the velocity of upwelling is w„[l -\- r cos {at — tj] 
 
 ( i?(a, + a,,.r)f-'"'"-^' 
 
 [■ 
 
 e= - ' ' -_ = cos (a< — £.,)—! 
 
 M 
 
 = +A:= 
 
 +^„ + , 
 
 rir„Bbi{aj-]-a^.r) \ Fa cos e, + /«■„ sin e, ~| . 
 
 Va= + (E) 
 
 Fa COS £, + /«■„ sin £, ~I • 
 
 , r2acos£, + A", sin e,~| . ^ , , r^'-eose„ — 2asin6,~] 
 
 l\ cos €, — a sin £i 
 
 COS at 
 
 Ba.,.rc-^'" , B 
 
 where 
 
 i (.g-6i(!/-3) 
 
 (163) 
 
 tan ( 
 
 a=+(A-,)^ 
 
 (164) 
 
 Theoretical Reduction op the Surface Temperature for Each 
 iMoNTH IN the San Diego Region, Due to Upwelling ; 
 AND Comparison with Observations 
 
 The theoretical relation of ocean temperatures to time and depth 
 developed in pages 368-381 was found to agree well with observations 
 from 40 to 700 meters in the San Diego region. The theory developed 
 in pages 388-393 is valid for only the upper ten meters ; but no satis- 
 factory theory for the intermediate interval from 40 to 10 meters or 
 to the surface has been worked out. Now since the temperature reduc- 
 tion at the surface depends upon the upwelling in all three intervals, 
 it is neeessarj^ to estimate the reduction in this intermediate interval 
 for which we have no theory. A method of making this estimate is 
 included in the following plan of computing the theoretical tempera- 
 ture reduction at the surface. 
 
394 MISCELLANEOUS STUDIES 
 
 An explanation of s.ymbols used in making the computations will 
 be given for reference : 
 
 6' equals the normal surface temperature. 
 
 6 equals the theoretical surface temperature when the effect of 
 upwelling is considered. 
 
 6' equals the mean annual normal temperature at the surface. 
 
 Ob equals the constant temperature at the depth 600. 
 
 A6' equals the total theoretical reduction of the surface temperature. 
 
 Ad's equals the theoretical reduction of the temperature at the 
 depth of .3 meters which corresponds to surface conditions 
 (p. 343). 
 
 A^' equals the theoretical reduction of the surface temperature due 
 to upwelling in the interval from 3 to 600 meter.s.. 
 
 A^' equals the mean annual temperature reduction at the 3 meter 
 level due to upwelling in the interval from 3 to 600 meters. 
 
 A4>s equals the mean annual temperature reduction due to up- 
 welling at the 3 meter level. 
 
 A^' equals the total theoretical reduction of the mean annual tem- 
 peratvire at the depth of 3 meters. 
 
 Ai/'', equals the mean annual temperature reduction due to upwell- 
 ing in the interval from 40 to 3 meters. 
 
 A^'ta equals the mean annual tempf^rature reduction due to up- 
 welling in the interval from 600 to 40 meters. 
 
 Throughout the interval fi'om 100 to 40 meters the temperature 
 reduction increases at a constant rate by the amount .36 (table S). 
 the velocity is practically constant {^ii\), and the mean annual 
 
 2 3 
 
 temperature gradient is ^. Tii tlic interval from 40 to 3 meters the 
 
 Ij Q 13 1 3 9 
 
 mean annual gradient is — =-^ (table 6), but the mean 
 
 velocity is approximately - (1 -\- q)y, where qu\ is the velocity at 
 
 the depth 3 meters. Assuming provisionally that q^O.l (p. 403) the 
 mean velocity in this interval is .55?r,. The temperature reduction 
 in any depth interval is proportional to the length of the interval, as 
 was shown to be the case in the interval from 100 to 40 meters, and 
 
OCEAN TEMPEEATUEES 395 
 
 is proportional to the temperature gradient (p. 386). Therefore if the 
 velocity were the same in the interval from 40 to 3 meters, as at the 
 levels below 40 meters, the following relation would hold 
 
 
 A4,\ _ V37 /'^' _3.9 
 .36-/2.3\go 2.3 
 
 or Af. = (.3G)(^||)=.61. 
 
 But using the provisional estimate 0.55m'i of the mean velocity in this 
 interval we have Aij/\ = .55 X -61 = .34. Using the principle (p. 386) 
 that for a given velocity the temperature reductions are proportional 
 to the temperature gradients 
 
 Act>' = J^'~^^'\ (A^')^ ^'~'^'^'~^"T^^'V a^') (165) 
 
 e'—0b—Ae' d'—6i—Ae' 
 
 Solving for A(f>' gives 
 
 /|^_^^^__Ae\\ -^ (166) 
 
 \6'—e,—Ae's ) 
 
 From page 387. and the value of Ai//'^ we have A(/)'=1.7 + .34^2.04. 
 From page 347, the normal temperature is 
 
 e'=— 3.79 cos (30f — 69) ° + 19.5 (167) 
 
 therefore 
 
 A,' = (-3.79 COS (30^ -69)° + 13.2 -A^.^^ ^ ^^ 
 
 since ^,,„„=6?3 (p. 376). 
 
 (168) 
 
 The observed mean annual .surface temperature is 17?0 (p. 375) but 
 the normal value le.ss Ac^' equals (19?5 — 2?04) equals 17?46, which 
 is ?46 higher than the observed value. This indicates that there is a 
 still further temperature reduction of the surface temperature due 
 to upwelling at the 3 meter level. From page 390, Oo^^i — ^4>' • 
 therefore, if A<^' is added to both members of equation (163) we can 
 replace 6. by 6^, and the value of ^ + A<^' differs from the normal 
 surface temperature solely because of the iTpwelling at the 3 meter 
 level. Therefore 6' — (0 -\- A<f>') equals the temperature reduction 
 Ai9'., due to upwelling at this depth. Since only the surface tempera- 
 
396 
 
 MISCELLANEOUS STUDIES 
 
 tures are required A" can be put equal to k. then l\ will equal k„. 
 Prom the value of 6' (equation 24) and equation (163) we have for 
 the value of A6's = 6' — (S + A<#>') 
 
 A6'',, = Be-^'"(ai + a,x) 
 1 i 
 
 cos (ai — e) eos(a< — e^) 
 
 + 
 
 Va- + Av 
 
 •WiyBb, ( a + a„x ) ( cos e^ fa cos ei + A-j sin ei~| 
 
 - ^ L a=+A,= J 
 
 x=+A,= 
 
 2A., 
 
 sin 
 
 r ^'^ COS €, — asine,~| , , rSa COS e, + A"o sin €,"] 
 
 ^"' "' -[ ' a- + Av r + L 2(4:- + A-.-) J 
 
 [ A, cose, — 2a -sin t, "I 
 
 ^^"^^ + [ 2(4a= + V) J 
 
 COS (2aO -c'-'" 
 
 (169) 
 
 Leaving iCg and therefore A, and u uudeterniiued for the present, 
 equation (169) can be reduced to the form 
 
 ..ff =9 19 i J 1 [ I (■00775HV) eose, 
 
 ^" ■' ^- ^ ] .560 V"^274T/^ i ^:;(-2^4 + ^^^'^ 
 
 f cos(30< — €.) cos (30^ — 69) jl .OlSSwy 
 
 "^"" 1V.274 + V -560 r(-274 + Av) 
 
 sin at -\- 
 
 .506 + .259A, 
 .274 + A-,= 
 
 .966A..— .1361 
 .2/4 + A,- 
 
 + h' J 
 
 ri.046 CO.S £., -\- A, sin ( 
 L 2(1.096 + A,=) 
 
 + 
 
 COS 2a/ '- 
 
 sin 2a/ + 
 
 :., COS e^ — 1.046 sin £3"] 
 2(1.096 + A:/) J 
 
 ;i70) 
 
 using the following values of the remaining constants: fti = — .318, 
 a„ = — .0166, 5e-=".= 5.9. e, = ]95, e=69, £3=(e, +£,), 6,== (£,, — cj, 
 r=.2, a = .523. ft, = .0365. (pp. 347, 351, 376), A = .2, A, = 
 
 .2 — .0365(c„, tan £, =— (p. 394). Assume w^ equals — 3, then 
 
 A^',= .448 — 3.78 cos (30/ — 60) + 3.48 cos (30/ — 69) 
 
 — .113 1 1.58 sin 30/ + .441 cos 30/ 1 =.448 —.69 cos 30/ — .18 sin 30/ 
 
 (171) 
 
 The values of the temperatures and their reductions are given in 
 table 12. 
 
OCEAN TEMPKEATUBES 
 
 397 
 
 Table 12 
 Normal surface temperatures and temperature reductions in the San Viego region. 
 
 t= 
 
 1 
 
 2 
 
 3 
 
 4 
 
 5 
 
 6 
 
 7 
 
 8 
 
 9 
 
 10 
 
 11 
 
 12 
 
 means 
 
 6' 
 
 16.55 
 
 15.75 
 
 15.95 
 
 17.11 
 
 18.91 
 
 20.86 
 
 22.45 
 
 23.25 
 
 23.05 
 
 21.89 
 
 20.09 
 
 18.14 
 
 19.50 
 
 ^0/ 
 
 -.24 
 
 -.06 
 
 + .27 
 
 + .64 
 
 + .96 
 
 + 1.14 
 
 + 1.14 
 
 + .96 
 
 + .63 
 
 + .26 
 
 -.06 
 
 -.24 
 
 + .45 
 
 A<^' 
 
 1 66 
 
 1.52 
 
 1.50 
 
 1.63 
 
 1.86 
 
 2,15 
 
 2.40 
 
 2.56 
 
 2.58 
 
 2.45 
 
 2.22 
 
 1.93 
 
 2.02 
 
 *^<t>r 
 
 -.10 
 
 -.27 
 
 -.37 
 
 -.37 
 
 -.27 
 
 -.10 
 
 + .10 
 
 + .27 
 
 + .37 
 
 + .37 
 
 + .27 
 
 + .10 
 
 .00 
 
 AB' 
 
 1 32 
 
 1.19 
 
 1.40 
 
 1.90 
 
 2.55 
 
 3.19 
 
 3.64 
 
 3.79 
 
 3.58 
 
 3.08 
 
 2.43 
 
 1.79 
 
 2.47 
 
 A6' obs. 
 
 1.55 
 
 1 35 
 
 1.35 
 
 1.91 
 
 2.81 
 
 3.06 
 
 2.75 
 
 2.65 
 
 2.85 
 
 3.09 
 
 3.29 
 
 2.64 
 
 2.50 
 
 6 comp: 
 
 15.23 
 
 14.56 
 
 14.55 
 
 15.21 
 
 16.36 
 
 17.67 
 
 18.81 
 
 19.46 
 
 19.47 
 
 18.81 
 
 17.66 
 
 16 35 
 
 17.03 
 
 6 obs. 
 
 15 00 
 
 14.30 
 
 14.60 
 
 15.20 
 
 16.10 
 
 17.80 
 
 19.70 
 
 20.60 
 
 20.20 
 
 18.80 
 
 16.80 
 
 15.50 
 
 17.00 
 
 Difference 
 
 + 23 
 
 + .26 
 
 -.05 
 
 + .01 
 
 + .26 
 
 -.13 
 
 -.89 
 
 -1.14 
 
 -.73 
 
 + .01 
 
 + .86 
 
 + .85 
 
 + .03 
 
 •From line 3, table 10 (p. 387). 
 
 The agreement between the predicted monthly temperatures and 
 the observed averages is very satisfactory. Moreover, from the rela- 
 tion of the vertical velocity to depth deduced from hydrodynamieal 
 principles, and shown by table 13, the velocity u\ at the 3 meter level, 
 would be .116h\, where it'i is the velocity below the 40 meter level. 
 That is, the value of w^ to be expected from that deduced from deep 
 water temperautres (p. 403) is .116 X ( — 31) equals — 3.6, which 
 agrees well with the value — 3.0 deduced from the surface temperature. 
 
 Deductions Relative to Oceanic Circulation in the San Diego 
 Region Based on Ekman's Hydrodynamical Theory 
 
 Zfippritz (1878) obtained some theoretical results relative to oceanic 
 circulation, from the general equations of motion of a viscous fluid, 
 and some of his conclusions have been widely used by oeeanographer.s 
 and geographers; but a critical examination made in the light of later 
 observations showed that his conclusions do not apply to conditions 
 found in nature. These erroneous conclusions are due to hi.s failure 
 to take into account the deflecting force of the earth's rotation and 
 to his use of the laboratory value of the coefficient of viscosity. The 
 importance of the effect of the earth's rotation on currents in the air 
 and ocean was pointed out long ago by Hadley, Coriolis, and Ferrel ; 
 but with the exception of free currents, that is. currents moving by 
 their own inertia, the deflecting force due to the earth's rotation wa.s 
 thought to be so .small that it could be neglected until Bjerknes 
 (1901) first made clear the importance of this deflecting force iii the 
 ca.se of forced currents. 
 
398 MISCELLANEOUS STUDIES 
 
 Ekman (1905) also used the general equations of the motion of a 
 viscous fluid, but included the deflecting force due to the earth's 
 rotation, and used in place of the coefScient of viscosity a constant 
 whose value was estimated by applying his formal solution of the 
 equations to field data. That is, he used a virtual value of the co- 
 efficient of viscosity in order to take into account the effect of the 
 irregular vortex motion which greatly increases the magnitude of the. 
 mutual reaction between the adjacent water layers. 
 
 On the simple assumption that the depth of the region considered 
 is large and the coast is at a sufficiently great distance Ekman (1905, 
 p. 7) deduced the effect of a wind, constant in magnitude and direc- 
 tion, over the whole region. His results for the northern hemisphere 
 are. for the velocity of the water, perpendicular and parallel respec- 
 tively to that of the wind, 
 
 U' = Foe-"!' cos (~ a!j\ (172) 
 
 r'=V,e-"«sm(j^ — aij\ (173) 
 
 Where V„ is the absolute velocity of the water at the surface y is the 
 depth below the surface and o is a constant. The value of a (Ekman, 
 
 1905 
 
 , p. (5) is n = J 'I'" ^'^" '^ where o) is the angular velocity 
 
 of the earth. <f> is the latitude, /x- is the virtual coefficient of viscosity, 
 and g is the density of the water. From equations (172) and (173) 
 it follows that the surface water velocity (where y equals zero) makes 
 an angle of 45 degrees to the right of the wind velocity, and the angle 
 increases as y increases. "When y has such a value (denoted by D) 
 that the water velocity has the opposite direction to that of the sur- 
 face, that is, when 
 
 ay = aD = 7r=DJ1^iI}R± (174) 
 
 , ^ J qio sin <i> 
 
 the magnitude of the velocity is r"^ or .043 times its surface value. 
 D is called the "depth of frictional influence," since the water velocity 
 below that depth produced by a wind over the open ocean is but a 
 small fractoin of that at the surface. From an estimate of the relation 
 between the wind velocity and its tangential pressure and the eorre- 
 
OCEAN TEMPEEATDEES 
 
 399 
 
 spending ocean eiirreut produced, Ekman (1905, p. 42) concluded that 
 
 the surface velocity Vg would be approximately 
 D would have the approximate value 
 
 7.6 
 
 ■0127 
 Vsin 4> 
 
 D 
 
 V sin</> 
 
 meters 
 
 /( and that 
 
 (175) 
 
 where /; is the wind velocity in meters per second. Thus for a wind 
 velocity of ten miles per hour (5.1 meters per second) and at the 
 
 Fig. 16. Components of the water velocity U' and V perpendicular and 
 parallel res]ieetively to the wind, and the velocity of Vc in a direction perpen- 
 dicular to the coast. 
 
 7 6X51 
 latitude 35° D would equal " =^ 51 meters, and for a velocity 
 
 V sin (t> 
 of fifteen miles per hour D would equal 75 meters. Solving equation 
 
 (174) for fi.- gives 
 
 „ D- . .0000729, . ,„. 
 fi'^^qw sin <f> = 7, (sni <^)D- 
 
 which equals 217 in c. g. s. imits, when T) equals 7500 cm. and 
 <^ equals 35°. 
 
 The relation of vertical velocity to depth will now be deduced from 
 the results of Ekman 's theory. 
 
 Let V make an angle A with the coast (fig. 16) and V the angle 
 (90-)- A) ° with tlie coast, then the velocity perpendicular to the 
 coast is, from equations (172) and (173), 
 
 y sin A -K C/' sin (90 -(- A) = Fc 
 
400 MISCELLANEOUS STUDIES 
 
 where a velocity directed away from the coast is regarded as positive. 
 That is. 
 
 Vc = "Foe""" - cos A cos 
 
 n — 0// j+ sin A sill H— "A c 
 = V^e-'y cos (j — ay —A J ( 176 ) 
 
 Let z equal the distance perpendicular to the coast, then Vc from 
 equation (176) is the limiting value of the velocity perpendicular to 
 the coast as z increases. Also, adjacent to the coast where z equals 
 zero, the velocity perpendicular to the coast must be zero. Let the 
 velocity perpendicular to the coast at the distance z be given by 
 the equation 
 
 V=Vcf(z) (177) 
 
 where /(0)=0 and f{z) =1 as z increases. The removal of 
 water at and near the surface due to a flow away from the coast 
 decreases the pressure at the lower levels, and gives rise to a com- 
 pensating flow of deep water toward the coast. Therefore in a coastal 
 region where the surface water flows away from the coast there is a 
 compensating upward flow of deep water. 
 
 Denoting the vertical velocity by W, the equation of continuity is 
 
 dV dW 
 
 ^+^=0 (178) 
 
 dz dy 
 
 neglecting the variation of the component parallel to the coast. Tliere- 
 fore from equation (177), assuming 
 
 W^f,(z)f,{y) (179) 
 
 we have 
 
 dfM 
 
 + /.<.-)%^ 
 
 f]80) 
 
 dz dy 
 
 To .solve equation (180) let 
 
 ^^ = Mf,iz) (181) 
 
 where M is a constant. Then from equation (176) and (180) 
 
 '^^'''■'l=—MVr^—MV, €-<"> cos ( T — ay — A ] (182) 
 
 dy 
 
 (^l-ay-x) 
 
OCEAN TEMPEEATVUES 401 
 
 fi-om \vliic-h we have 
 
 fAu) = —^1^0 i e-"" cos r^^—ay — xyhj + C, 
 
 = -j sin A sin ay — cos A cos ay r e-"" -\- C^ 
 
 MV \/2 
 
 — ° ^ -e-oi/cos (ay + A) +C, (183) 
 
 2« 
 where C, is the constant of integration which must have the value 
 
 MV \/Y 
 ^ COS A in order that /^(O) may equal zero. From equa- 
 tions (176, 177, 179, and 181 ) it follows that the horizontal and vertical 
 components of the velocity are respectively 
 
 V = f(z)V„e-'"-' cofi 
 
 W = — A(3)* ^°^ 
 
 (I-«.'v-a) 
 
 = + / 6-"^ cos ( - — ay — X](hi { MVo 
 'a J \4 J ) 
 
 (184) 
 
 4r{f""(4'-«"-0 
 
 , cos A ,, 
 
 dy j^l^o 
 
 V2a ) 
 
 = ^L-".vcos(ai/ + A)— cosA !-.^(^ (185) 
 
 V2ffl ] \ f?2 
 
 Since the horizontal velocity of the siirfaee water is proportional to the 
 wind velocity (pp. 363, 398) it follows from equation (185) that the 
 vertical velocity of the water is also proportional to the wind velocity. 
 
 The differential equation of a stream line is in general -^^ equals 
 the slope of the curve ecjuals — equals 
 
 r (it \ , cos A )rf/(2 
 
 — -, / (■-•'y cos I a\\ — A \hj ^=\ -^r- 
 
 dy _}J \4 / • \/2a \ dz 
 
 dz 
 
 e-'"'jcos/'- — ay — aM/(2) (186) 
 
 which can be reduced to the exact differential equation 
 
 df{z) 
 
 e-"" cos ( - — ay — A | rZy 
 
 /(2) " 
 whose solution is 
 
 \ ( e-'v cos (- —ay ~ \\dy — ^^=1 (187) 
 [cos A — c-"" cos (ay + X)]f{z) = C„ (188) 
 
402 MISCELLANEOUS STUDIES 
 
 where C„ is a constant of integration corresponding to a given stream 
 line. From equation (185) the upward flux through a horizontal area 
 of unit width and length z measured perpendicularly to the coast is 
 
 S-'— 
 
 [cos A — e-"!' cos (az + X)]f{z) 
 
 aV2 
 Eembering that /(0)^0 and /(^) = 1 as z = oa 
 
 (189) 
 
 {\y,,^_y^^[<^o^^-'^-''^o^«'y + ^)]f(^) (190) 
 
 Jo «V2 
 
 and that the maxinuun numerical value for a given value of y is 
 
 C';ra,=.-YA^^l^^ri^9ll!^I>+m (191) 
 
 Jo «V2 
 
 which approaches tlie value — ("^S^) as V increases, the ratio R 
 
 of the upward flux tlirough a horizontal area at the depth y of unit 
 width and length z measured perpendicularly to the coast to the total 
 upward flux is 
 
 j^_ J„ "^^ ^ [cosA — f-'"'cos (gjy + A)1/(2) ,-^^2) 
 
 V„ cos A cos A 
 
 Therefore the parameter C„ in the equation of the stream line (equa- 
 tion 188) ecjuals the ratio R multiplied by cos A, and the flux between 
 any two adjacent stream lines of a series in which tlie increments of 
 C, are equal is constant. 
 
 The mean wind velocity of the 5 degrees square of the U. S. Coast 
 Pilot Charts (Moore, 1908-11) west of San Diego was found to be 
 about fifteen miles per hour in a southeasterly direction, approxi- 
 mately parallel to the coast. Therefore for the San Diego region the 
 angle A in equation (185) is zero, and the vertical velocity at the 
 depth )/ is, from equation (185), proportional to [1 — c""^ cos ay] 
 
 where a =^ -= — (p. 398). The values of this function are tabu- 
 
 lated with respect to depth in table 13. 
 
OCEAN TEMPEEATVBES 
 
 403 
 
 Table 13 
 
 Tabulation of the function I J g- 75 cos 
 
 7r!l \ 
 
 75 / 
 
 y 
 
 UL TTlJ 
 
 l-e-75 cos=^ 
 
 y 
 
 IIL irv 
 1-e-io cos=^ 
 
 
 
 
 
 18 
 
 .658 
 
 1 
 
 .043 
 
 19 
 
 .685 
 
 2 
 
 .085 
 
 20 
 
 .710 
 
 3 
 
 .116 
 
 21 
 
 .737 
 
 4 
 
 .166 
 
 22 
 
 .760 
 
 5 
 
 .207 
 
 23 
 
 .783 
 
 6 
 
 .246 
 
 24 
 
 .805 
 
 7 
 
 .288 
 
 25 
 
 .825 
 
 8 
 
 .325 
 
 30 
 
 .913 
 
 9 
 
 .361 
 
 35 
 
 .977 
 
 10 
 
 .399 
 
 40 
 
 1 02 
 
 11 
 
 .436 
 
 50 
 
 1.067 
 
 12 
 
 .471 
 
 60 
 
 1.065 
 
 13 
 
 .515 
 
 70 
 
 1.052 
 
 14 
 
 .535 
 
 80 
 
 1.034 
 
 15 
 
 .568 
 
 90 
 
 1.019 
 
 16 
 
 .600 
 
 100 
 
 1.007 
 
 17 
 
 .630 
 
 infinity 
 
 1.000 
 
 The relation of the velocity to depth was deduced from hydro- 
 dynamical eonsidei-ations, but its relation to distance from the coast, 
 which requires the determination of the function fiz) (equation 177. 
 p. 400), did not result from the foregoing reasoning, but will now be 
 considered. Prom equation (177), page 400 /(0)=0 and f{z)= 1 
 as z increases, and for large values of y off San Diego, equation (185) 
 becomes 
 
 y„ df(z) 
 
 W=W^- 
 
 «\/^ 
 
 dz 
 
 (193) 
 
 where (pp. 377, 388) TFi = — 31 meters per month where z equals 10 
 miles, and equals double that value or 62 meters per month, where 
 z equals zero. From page 363, for an average wind velocity of 15 miles 
 per hour or 7.5 meters per second 
 
 ^^ .0126X7.5 TO' . A 
 
 Kn = =: .12d meters per second 
 
 Vsin 35° 
 
404 MISCELLANEOUS STUDIES 
 
 equals 324.000 meters per month and a equals ;^^ (p. 402). Substi- 
 
 tuting these numerical values in equation (193) and expressing z in 
 miles gives 
 
 dz 
 
 and 
 
 dfiz) 
 
 ^^(^)— 0210 for z = 0, 
 
 = .0105 for 2 = 10. (194) 
 
 dz 
 
 While these conditions, which f(z) and must satisfy, do not 
 
 determine the functions precisely, they suffice for a rough estimate. 
 The following form 
 
 /(2) = l_A-,e-"i^— (1 — /.-Je-''^' (195) 
 
 has the value zero when z equals zero and approaches 1 as 2 increases 
 indefinitely for all positive values are of h^ and lu, and difEerentiating 
 with respect to z 
 
 .^^ = hXe-'"' + h„{l — k, ) e-''^-' (196) 
 
 dz 
 
 The above expression for — ^— satisfies the conditions expressed by 
 
 equation (194) for the following values of the eon.stants found by trial : 
 
 /,^ = 01, /,i = .93, /u=.17, (1 — A-i)=.07. 
 Therefore from equation (193) the vertical velocity at any depth ?/ 
 equals 
 
 W = — 2960 ^^4^ = — 29G0 f.0093r-'"'— .0119e-''--) (197) 
 dz 
 
 and from equation fl92) the ratio of the upward ilux within a dis- 
 tance z from the coast to the total flux is proportional to 
 
 f(z)=l — .9.3e-»'- — .07f-'-' (198) 
 
 where z is the distance from the coast in miles. The values of f{z) 
 
 df(^) 
 and -~ — are tabulated with respect to z in table 14. 
 
 az 
 
OCEAN TEMPEBA'TDEES 
 
 40o 
 
 Table 14 
 Tabulation of the functions f{z) and 
 
 df{^) 
 dz 
 
 z 
 
 f{z) 
 
 dm 
 
 dz 
 
 z 
 
 /(2) 
 
 dm 
 
 dz 
 
 
 
 
 
 .0212 
 
 40 
 
 .380 
 
 .0062 
 
 1 
 
 .020 
 
 .0192 
 
 50 
 
 .430 
 
 .0057 
 
 2 
 
 .040 
 
 .0175 
 
 70 
 
 .530 
 
 .0046 
 
 3 
 
 .060 
 
 .0161 
 
 100 
 
 .660 
 
 .0035 
 
 4 
 
 .075 
 
 .0150 
 
 200 
 
 .870 
 
 .0013 • 
 
 5 
 
 .090 
 
 .0139 
 
 300 
 
 .950 
 
 .00046 
 
 10 
 
 .150 
 
 .0105 
 
 400 
 
 .980 
 
 .00020 
 
 15 
 
 .190 
 
 .0089 
 
 500 
 
 .994 
 
 .00006 
 
 20 
 
 .240 
 
 .0080 
 
 700 
 
 .9992 
 
 .00000 
 
 30 
 
 .310 
 
 .0070 
 
 1000 
 
 .99995 
 
 .00000 
 
 Tluis it appears that 90 per cent of the upward flux is confined to 
 a coastal belt about 250 miles wide. Finally the stream line equation 
 (188) becomes 
 
 e- tFcos i!L 
 (5 
 
 [1 — .93e- 
 
 .07* 
 
 C„ 
 
 (199) 
 
 for the San Diego region, and the stream lines corresponding to C, 
 equal 0, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, and 1.0 are shown in 
 figure 17. 
 
 Meters Water surface Cs 
 
 
 
 150 
 Miles 
 Fig. 17. Traces of theoretical surfaces of flow on a vertical plane perpen- 
 dicular to the coast. The flow included between any two consecutive surfaces 
 is one-tenth of the total amount. 
 
406 MISCELLANEOUS STUDIES 
 
 Figure 17 represents the component of the hypothetical circulation 
 in a plane perpendicular to the coast corresponding to a uniform wind 
 .over the whole region, in which the bottom bends sharply upward at 
 the coast, but gives some idea of the actual circulation. In examining 
 the figure it must be noted that the vertical is very much greater than 
 the horizontal scale. In fact, if the horizontal scale were the same 
 as the vertical one actually used the diagram would be about one and 
 one-half miles in length. 
 
 Deduction of the Upwelling Velocity off San Diego from the 
 Observed Relation of Salinity to Depth 
 
 If the rates of molecular diffusion of salts and conduction of heat 
 are relatively very small as compared with the rate of transfer due to 
 the alternating circulation (p. 368), the differential equation 
 
 ^ = ^=^_W^ (200) 
 
 (equation 80, p. 368) applies in general where the constant fj.- is a 
 measure of tlie rate of transfer, and the dependent variable is the salt 
 concentration or temperature. An application to temperature data 
 has already been made, and we have only to replace 6 by the salinity 
 <Sf in the temperature ecpiation and its solution already worked out 
 (pp. 375-378) in order to obtain the corresponding formulae for 
 salinity. However, the salinity data are too incomplete to furnish 
 reliable estimates of averages for each month, and it seemed best to 
 use the data taken in the same region ^Section 40 J for each of the 
 three months, Augu.st, 1912, February, 1913. and April. 1913, which 
 correspond to an interval of less than one year. These data are 
 presented in table 15. 
 
OCEAN TEMPEllAl'UriES 
 
 4U7 
 
 Table 15 
 OVuerved Salinities for Aucjust, 13U, February, WIS, and Aprii, WIS, 
 
 in Section 40^ 
 
 Depth 
 
 Salinities 
 
 y 
 
 ((=2) 
 
 (( = 4) 
 
 ((=8) 
 
 Mean annual 
 
 February 
 
 April 
 
 August 
 
 values 
 
 
 
 33.47 
 
 33.58 
 
 33.75 
 
 33.61 
 
 10 
 
 33.47 
 
 33.58 
 
 33.69 
 
 33.58 
 
 20 
 
 33.44 
 
 33 58 
 
 33.58 
 
 33.51 
 
 30 
 
 33 42 
 
 33.61 
 
 33.54 
 
 33.48 
 
 40 
 
 33.46 
 
 33.66 
 
 33.57 
 
 33.52 
 
 50 
 
 33.50 
 
 33 73 
 
 33.65 
 
 33.57 
 
 60 
 
 33.54 
 
 33.79 
 
 33.73 
 
 33.63 
 
 70 
 
 33.57 
 
 33.85 
 
 33.57 
 
 33.67 
 
 SO 
 
 33.62 
 
 33.90 
 
 33.81 
 
 33.72 
 
 90 
 
 33.66 
 
 33.94 
 
 33.85 
 
 33.76 
 
 100 
 
 33.70 
 
 33.99 
 
 33.88 
 
 33.79 
 
 150 
 
 34.03 
 
 34.19 
 
 34.07 
 
 34.05 
 
 200 
 
 34.25 
 
 34,30 
 
 34.17 
 
 34.21 
 
 300 
 
 34.32 
 
 34.33 
 
 34.23 
 
 34 27 
 
 400 
 
 34.33 
 
 34.34 
 
 34.28 
 
 34.30 
 
 500 
 
 34 36 
 
 34.35 
 
 34.32 
 
 34 34 
 
 600 
 
 
 
 
 34.40 
 
 800 
 
 
 
 
 34.48 
 
 1100 
 
 
 
 
 34.53 
 
 From the formula for salinity (replacing 6 by S in equation 117) 
 it follows that the mean of any two salinities corresponding to a tinT? 
 interval of six months would equal approximately the mean annual 
 salinity. Accordingly, the means of the salinities for February and 
 August are assumed to be the mean annual salinities in this case. The 
 constants C, D, and A of equation (117) were found as in the case 
 of temperature data by fitting the equation 
 
 S,„ = Ce->^v+D (201) 
 
 to the observed mean annual salinities. .Si,,,. Then the observed mean 
 annual salinity was subtracted from each of the entries under February 
 and April, and the expression 
 
 C{^^\e 
 
 e'^i' cos (30^ + 75)° 
 
 was subtracted from each of these, using the same numerical values 
 for M'l, r and a as on page 377, and the value of (J determined above. 
 The remaining expression 
 
 il/e"'" cos {at -\-h^y — e') 
 
408 
 
 MISCELLANEOUS STUDIES 
 
 (p. 377) was fitted to these remainders. The values of the constants 
 thus found are D == 34.55, C = — 1.25, A = —.005, M = —.6, 
 .€' = —65. Hi = —.0075, ^1== 10,600, n',=—53, &i = — .00495. The 
 value of &i determined from equation (121) exceeds numerically the 
 value — .00175 determined from the observations. Expressing the 
 angle in degrees, these values are — .28 and — .10 respectively. The 
 values fi- = 10,600 and \\\ =^ — 53 obtained from the salinity data are 
 in good agreement with the values 7760 and — 31 obtained from the 
 more complete and extensive temperature data (p. 378). The com- 
 puted and observed values of the remainders and of the mean annual 
 salinities are entered in table 16 as an additional test of the theory. 
 The computed remainders were obtained from 
 
 __6e-oo75y(,os (30« — .ly-f 65)° 
 
 and the computed mean annual salinities were obtained from 
 
 34.55 — 1.25 e-""-'". 
 
 Table 16 
 
 Computed and Observed Remainders for February and April, and the Computed 
 
 and Observed Mean Annual Salinites 
 
 Depth 
 
 Remainders 
 
 Mean 
 
 annual 
 
 t = 2 
 
 Feb'. 
 
 « = 4 
 
 April 
 
 Salinities 
 
 y 
 
 Computed 
 
 Observed 
 
 Computed 
 
 Observed 
 
 Computed 
 
 Observed 
 
 50 
 
 .14 
 
 -07 
 
 .32 
 
 .16 
 
 33.53 
 
 33.57 
 
 100 
 
 .07 
 
 -.09 
 
 .21 
 
 .20 
 
 33.74 
 
 33.79 
 
 150 
 
 .02 
 
 - 02 
 
 .13 
 
 .14 
 
 33.96 
 
 34.05 
 
 200 
 
 .00 
 
 .04 
 
 .08 
 
 .09 
 
 34.04 
 
 34.21 
 
 300 
 
 -.01 
 
 .05 
 
 .03 
 
 .06 
 
 34.22 
 
 34.27 
 
 400 
 
 -.01 
 
 .03 
 
 .01 
 
 .04 
 
 34.33 
 
 34.30 
 
 500 
 
 -.01 
 
 .02 
 
 .00 
 
 .01 
 
 34.40 
 
 34.34 
 
 GOO 
 
 
 
 
 
 34.44 
 
 34.40 
 
 800 
 
 
 
 
 
 34.48 
 
 .34.48 
 
 1100 
 
 
 
 
 
 34.50 
 
 34.53 
 
 The mean velocity of upwelling can also be estimated from the 
 salinity distribution in the upper 30 meter layer, and by an entirely 
 different method. A comparison of this value with the two estimates 
 made with the aid of theoretical results ah-eady presented affords a 
 severe test of the theories and gives an idea of the reliability of the 
 
OCEAN TEMPEBATVBES 
 
 409 
 
 Water Surface 
 
 Coast 
 
 results. lu dealing with mean annual values we can assume all con- 
 ditions to be independent of the time, from which it follows that the 
 total amount of water in a given volume remains constant and the 
 total amount of salts remain constant. Tlierefore the rate of flow of 
 
 water and salts into the volume must 
 equal the rate of flow out of the 
 volume. 
 
 This principle will be applied to 
 two difEerent volumes, thus giving two 
 estimates of the velocity of the up- 
 welling. First, consider a vertical 
 column (fig. 18) whose cross section 
 is a square of unit area and who.se 
 base is at the depth y., where the 
 salinity has its minimum value (Mc- 
 Ewen, 1916, p. 272). 
 
 The explanation of symbols used 
 
 follows : 
 
 8 = the salinity at any depth y. 
 
 Sg = the salinity at the surface. 
 
 S^ = the salinity at the depth .y„. 
 
 To 1^1 = the vertical velocity at the 
 depth y,, W^ is the maximum 
 value, and corresponds to large 
 values of y (table 13). 
 
 y = the horizontal velocity. 
 
 E =^ the rate of evaporation at the 
 surface. 
 
 bottom 
 
 Fig. 18. Eectangular volume of A flow into the volume is regarded 
 
 ■(vater from the depth y, of mini- 
 mum salinity to the surface, used as negative, and a flow out IS regarded 
 
 in determining the velocity of up- positive, the vertical distances and 
 
 welling from salinity. *^ ' 
 
 velocities are regarded as positive 
 
 when directed downward from the .siirface, and horizontal distances 
 
 and velocities are positive when directed away from the coast. 
 
 Because of the invariability of the amount of water in the volume 
 
 r„^Y, — E +(vdA = 
 
 (202) 
 
410 MISCELLANEOUS STUDIES 
 
 where dA is an element of the vertical surface enclosing the column, 
 and the integral is taken over the whole vertical surface. Similarly 
 because of the invariability of the amovint of salts in the volume 
 
 r„W,S„ +j'vSdA = 0. (203) 
 
 Let 
 
 S = 8 + AS (204) 
 
 where <S' is the constant mean salinity for the whole volume and AS 
 is a variable increment. Then equation (203) becomes 
 
 r„W,S., + S CvdA + Cv(AS)dA = ' (205) 
 
 and substituting the value of | Yd A from equation (202) we have 
 
 r„W,S._ — 'S{rJY, — E) + fv(AS)dA = (206) 
 
 or solving for W^ 
 
 — SE— CviAS)dA 
 
 W,^ ^ (207) 
 
 i\{S. — S) 
 
 An estimate of | V(AS)dA can be made as follows: Let the 
 
 vohnne be so turned that two of its parallel faces are parallel to the 
 coast line and therefore perpendicular to the horizontal velocity V 
 directed away from the coast and given by equation (p. 401) 
 
 r = f{z) r^e-oy cos C^ - an ^ (208) 
 
 Then neglecting the variation of the salinity in a direction parallel to 
 the coa.st, the integral 
 
 ry(A^')(L4=.- fv,{AS),du+ Cv,{AS),dy 
 
 (209) 
 
 where Y^ and (AS), correspond to the face next to the coast and "F, 
 and (AS), to the face farthest from the coast. From a study of our 
 salinity observations (ilcEwen. 1916, especially plates 20, 21, 22, and 
 24) made from five to fifteen miles offshore, it appears that the hori- 
 zontal gradient parallel to the coast is negligible as compared to that 
 perpendicular to the coast, thus justifying equation (209). The 
 numerical values of the horizontal salinity gradient per meter esti- 
 mated from our observations are given for a series of depths in table 17. 
 
OCEAN TEMPERATDEliS 
 
 411 
 
 Table 17 
 Mean horizontal aalinity gradient per meter during the Kummer for a seriet 
 
 of depths 
 
 Depth, y 
 
 
 
 5 
 
 10 
 
 1.5 
 
 20 
 
 25 
 
 30 
 
 Salinity 
 gradient, A(S' 
 
 6X10-' 
 
 6X10-" 
 
 3X10-" 
 
 
 
 -10-" 
 
 -3X10-" 
 
 -6X10-" 
 
 Our salinity data indicate that (AS') is practically zero in winter, 
 hence the mean annual value would be about half of that entered in 
 the table. 
 
 Owing to the .small value of (T^j — V.,) compared to the mean 
 value V and because AS'^(AS)o — (A.S')j equation (209) can be 
 written in the form 
 
 j'v(\S)dA^j'v(AS')d,j 
 From equations (207, 208, and 210) we have iinally 
 
 I {AS') c-"" cos (|— «yj '^U 
 
 (210) 
 
 -SE — VJ{z) 
 
 W,^ 
 
 (211) 
 
 ■i\iS,— S) 
 
 In order to check the above results the same principle will be 
 applied to a different volume (fig. 19). The stream lines (fig. 17) 
 being traces of surfaces of flow on a plane perpendicular to the coast, 
 two such planes, two surfaces of flow, and two horizontal planes 
 inclose a volume such that the component of the velocity along a line 
 parallel to the coast is the same for each vertical plane. Hence the 
 vertical flux through a horizontal section of this volume must be 
 independent of the depth of the section, in order that the total quantity 
 of water inclosed by these surfaces may be constant. Consider the 
 volume inclosed by two surfaces of flow, two vertical faces perpen- 
 dicular to the coast and parallel to the plane of the paper at unit 
 distance apart (fig. 19), and two horizontal sections at the depths y^ 
 and !/, of which the upper form.s the base of a rectangular prism 
 extending upward to the surface of the water. 
 
 Let t\W^ be the mean vertical velocity at the depth y,, and )\Wi 
 that at the depth y„, then i\W^B^ must equal r„W,B„ where 7?. ^'iid 
 fi, are the areas of the upper and lower sections respectively, whence 
 the section areas B, and B„ must satisfy the equation 
 
 B„ r, 
 
 (214) 
 
412 
 
 MISCELLANEOUS S2UVIES 
 
 For the whole volume inclosed, from the base B., to the water 
 surface, the condition of the constancy of the quantity of water re- 
 quires that 
 
 r^W^B^ — EB^ +fv,dA, — Cv.clA, = (215) 
 
 Coast 
 
 Watei Scjrfac 
 
 BoTVom 
 
 Fig. 19. Volume of water included in part by two surfaces of flow from the 
 depth 2/j of minimum salinity to a depth ,i/„ used in determining the velocity 
 of upwelling from salinity. 
 
 where (dA.^) is an element of area of the vertical face of the prism 
 next to the coast, (dA^) is an element of the other parallel face, and 
 Fg and V^ are the corresponding horizontal velocities. 
 
 Similarly, in order that the total amount of salts may remain 
 constant, 
 
 r^W,S.B. +j'v,S,dA,— fv,S,dA, = (216) 
 
 where S^ and S^ are the salinities corresponding to the elements of 
 area {dA^) and (cZAJ. 
 
OCEAN TEMPEEATUBES 413 
 
 Let 
 
 O3 ^= Oj -\- AiSg 
 
 and 
 
 8, = S, + ^S, (217) 
 
 where <Si is the constant mean salinity for the prisniatio volume from 
 the surface to the depth y, and ASg and AS^ are variable increments. 
 Then equation (216) becomes 
 
 — fv,{AS,)dA,- fv,{AS,)dA, = 0, (218) 
 
 and substituting the value of 
 
 i^fv,dA,-fv,dA:,^ 
 
 from equation (215) we have 
 
 V{A8')dy 
 
 W,= ^ ^- ^ (219) 
 
 r,{8, — 8,) r,{S,~8,) 
 
 where (A»S") is defined on page 411, and finally, substituting for V 
 the value given by equation (208) 
 
 w,=- -^-^ 
 
 VJ{z) j { AS' )€-"» CO J ^^ay\du 
 
 r,(S,— SJ rA8,-S,) (220) 
 
 If the depth of the upper section is at the level y^ we must sub- 
 stitute y., for 2/1 and r, for t\ in equation (220), which then becomes 
 identical with equation (211). But in equation (220) y^ can have 
 any value between the limits, zero and y,, where y„ is the depth of 
 minimum salinitj', and estimates of the velocity based on different 
 values of y, should give the same result. Some divergence of these 
 values in any actual case is to be expected, since the different esti- 
 mates are based on different observations that are subject to errors 
 of measurement and since the actual relation of the velocity to depth 
 may differ from the theoretical relation (p. 403). 
 
 In table 18, where y„=: 324000 (p. 404) and —E for the latitude 
 of San Diego is .0754 meters per month (Schmidt, 1915, p. 121) are 
 presented the results based upon the mean value of {AS'), that i.s, 
 half the value entered in table 17, the mean annual salinities as shown 
 by plate 11 and table 3 (McEwen, 1916) and the values of t\ and 
 /(lO) from tables 13 and 14. 
 
414 
 
 MISCELLANEOUS STUDIES 
 
 
 h 
 
 ICfN-^ClGCCOCC-^U^C^ 
 
 COOGOOOCO^UOCDO^ 
 ']' 1 1 1 1 1 1 1 1 1 
 
 Vcfiz) 
 
 {AS')e-''ii cos {^-ay\ly 
 
 o 
 
 O=0t^00-*02C<5(NCO 
 OC0C000CDCD':0t^^»O 
 
 r, (Sz-ST) 
 
 .05 
 
 (AS')e-"!'cosn -0.1/ jrfy 
 
 o 
 
 IOCOC01^I^I--C-100C^ 
 
 ^(MCliOOlOGOOO-^CO 
 
 coooooooooocc 
 
 1 {AS' )e-'"-' cos (j - wi) ibj 
 
 C0-^00^05C:OOC0O 
 
 1 
 
 1? 
 
 1 
 
 '^(N'-(OiCOI><Nt^t^iO 
 
 'j^ 1 i 1 1 1 1 1 1 1 
 
 1 
 
 
 (N'^^!NC^'^^(NC^^c^'^^(N 
 
 C 
 
 
 
 1 
 
 CO 
 
 co^ooot^co»OTt<Tj4(ro 
 
 ^^^ooooooo 
 
 1 1 1 1 1 1 1 1 1 1 
 
 \^ 
 
 lOiOOOiCOiCOOrt^O^ 
 
 cococococococccccoco 
 cocococococococococo 
 
 
 
 s 
 
 CCC001(MtOOO^'<*<(^0 
 1-H 1— < .— 1 c^ C>1 c^ cc 
 
OCEAN TEMrEHArUEES 415 
 
 Except f(ir tli(> first two values of y^ near the surface, where the 
 water is most disturbed, and for the hirgest value of y„ where the 
 diiference {S„ — i?,) is so small that it is subject to the largest pro- 
 portional error, the computed values of the velocity W^ shown in the 
 last column are in good agreement. And the mean of the central 
 values, which are subject to the least error, is about — 35, which is 
 the best estimate from the available data and agrees well with the 
 values — 31 and — 53 found before, page 408. 
 
 Conclusion 
 
 In the case of no average flow of the water the form of a function 
 giving the rate of gain of heat and of another giving the rate of loss 
 of heat from a small volume of water at a given latitude and depth, 
 was developed from a few simple assumptions, suggested by laboratory 
 experiments as well as field observations. Equating the sum of these 
 two expressions to the product of the specific heat by the volume by 
 the rate of change of temperature resulted in a differential equation 
 whose solution gave the temperature from the surface to a depth of 
 ten meters as a known function of time, depth, latitude, and certain 
 phj'sical constants, under the conditions of no average flow of water. 
 
 From observations on the relation to latitude of the mean animal 
 surface temperature and the annual temperature range and the relation 
 to latitude of the mean annual solar radiation and its annual range, all 
 of the physical constants of the formula were computed. 
 
 The lag between the time of the temperature maxima and minima 
 and tlie time of the maximum and minimum values of the solar radia- 
 tion deduced from these constants agreed well with the observed value. 
 Also the mean monthly temperature at a region whose mean annual 
 temperatures agree well with the normal value for the latitude, that 
 is, the value corresponding to no average flow, would be expected to 
 agree with those computed from the ftu-mula for normal temperature. 
 A comparison of .such computed and observed temperatures for a 
 series of latitudes from 20° N to 40° N indicated a very satisfactory 
 agreement. 
 
 If the rate of change of temperature due to some factor not in- 
 cluded in the above reasoning is known tliis quantity can be added 
 to the differential equation already derived. Since all of the constants 
 of the original differential equation are known the solution of the 
 modified equation will give the temperature due to the new factor. 
 
416 MISCELLANEOUS STUDIES 
 
 If, for example, there is a flow of the water, the rate of flow multiplied 
 by the temperature where the water enters a given element of volume 
 gives the rate at which the heat is carried into the volume. The rate 
 at which the heat leaves the volume, computed in the same way, 
 subtracted from the rate at which it enters gives the rate of change 
 of heat in the volume due to the corresponding ocean current, whether 
 horizontal or vertical. A term expressing this rate of change of heat 
 in the case of a horizontal current was added to the differential 
 equation, and the solution furnished a means of estimating the magni- 
 tude of horizontal currents from surface temperatures witliout con- 
 sidering the causes of the currents. 
 
 Numerical applications of this formula to a region of the North 
 Pacific off the California coast and of the North Atlantic off the 
 African coast gave estimates of the horizontal flow in good agreement 
 with direct observations and with what would be expected from the 
 observed wind velocity. 
 
 ^ Conclusive evidence of the presence of cuiTents directed upward 
 from the bottom along the California coast which cause reduction in 
 temperature has been published before; but to test this conclusion 
 further it was assiuned that the reduction of the temperature of the 
 coastal water was due entirely to the upwelling of deep water, and 
 the temperature distribution at depths exceeding 40 meters was 
 assumed to result from a flow of heat according to Fourier's well 
 known conductivity equation, in which a term expressing the rate of 
 loss of heat due to upwelling was added. The formal solution of this 
 equation contained certain physical constants whose evaluation re- 
 quired the observed monthly temperatures at a series of depths. Our 
 temperature data for a deep water region twenty miles offshore from 
 San Diego were sufficient for making approximate estimates of all of 
 the constants, of which the velocity of upwelling and the term corre- 
 sponding to conductivity are of special interest. The latter constant 
 depends largely upon the eddy motion, or alternating circulation, 
 which tends to mix the water and has been called eddy conductivity or 
 Mischungsintensitat, and was fovind in this case to be several thousand 
 times the laboratory value. In dealing with salinities the same 
 formula can be vised; and a similar constant appears, which also 
 depends largely on the mixing motion of the water, and would be 
 expected to have practically the same value as that determined from 
 temperatures. Our salinitiy data, though not so complete as the 
 temperature data, confirmed tliis conclusion and gave approximately 
 
OCEAN TEMPEEATUEES 
 
 417 
 
 the same velocity of upwelling. Moreover, in applying hydrodynamieal 
 equations to problems of oceanic circulation the coefficient of viscosity 
 must be replaced by another constant depending on the eddy, or 
 turbulent motion, and having a nuich greater value than the laboratory 
 vahie obtained from observations on a slow laminar flow free from 
 irregular motions. Similar results have also been found by G. I. 
 Ta.ylor in certain recent studies of the temperature, water vapor, and 
 velocity in the atmosphere. The results of laboratory experiments 
 and theories based on them were helpful but could not provide the 
 numerical values requi red ; iu each case field observations were neces- 
 sary. Furthermore, since the eddy conductivity, or Mischungs- 
 inten-sitdt, is not a physical constant of the substance, sea water or 
 air, but depends upon the intensit.y and character of the circulation, 
 its value will vary accordingly. The following approximate values of 
 these constants, the coefficient of viscosity, diffusion, and conductivity 
 under laboratory conditions and estimated from field observations in 
 the ocean and the atmosphere, illustrate the great differences between 
 field and laboratory conditions. 
 
 Table 19 
 
 Estimates of the coefflcients of viscosity, diffusion, and heat conductivity made 
 
 from field observations in the ocean and atmosphere compared 
 
 tvith values obtained in laboratory experiments 
 
 Sea Water 
 
 Observ-er 
 
 Coefficient 
 
 Laboratory value 
 in c. g. s. units 
 
 Value from field 
 
 obsen-ations in 
 
 c. g. s. units 
 
 Ratio of the field 
 to the laboratory 
 value 
 
 Ekman 
 
 Vi.scosity' 
 
 .014 
 
 217 
 
 15,500 
 
 Jacob.sen 
 
 Diffusion 
 
 .0000125 
 
 .3 to 11.4 
 
 24,000 to 
 320,000 
 
 McEwen 
 
 Diffusion 
 
 .0000125 
 
 40 
 
 
 McEwen 
 
 Conductivity^ 
 
 .0012 
 
 30 
 
 25,000 
 
 Air 
 
 Taylor 
 Taylor 
 
 Viscosity' 
 Conductivity^ 
 
 .13 
 
 .20 
 
 770 to 6,900 
 570 to 3,400 
 
 6,000 to 50,000 
 3,000 to 17,000 
 
 ' The laboratory value of the "kinetic coefficient" of viscosity, or the coeflficient of viscosity 
 divided by the density, is given since it is the constant in the equations of motion, which is 
 formally equivalent to the one given by field observations. 
 
 = The laboratory value of the thermometric conductivity is given since that is the constant 
 in the equation of heat conductivity, which is formally equivalent to the one given by field 
 observations. 
 
 The estimation of the effect of upwelling on the surface tempera- 
 ture made necessary the consideration of results obtained for depths 
 
418 MISCELLANEOUS STUDIES 
 
 below the 40 meter level and the solution of the original differential 
 equation for surface temperatures after adding a term giving the rate 
 of temperature change due to upwelling. The monthly values deduced 
 in this wa.y for the San Diego region agree very well with those 
 afforded by the observations. 
 
 From the magnitude of the vertical velocity found from tempera- 
 tures and from certain results deduced from Ekman's hydrodynamical 
 theory, the distribution of the horizontal and vertical velocity of the 
 water in a vertical plane perpendicular to the coast was deduced and 
 represented graphically. 
 
 An independent estimate of the velocity of upwelling made from 
 the distribution of salinities in the upper 30 meter layer and of the 
 rate of evaporation at the surface agreed well with the other two 
 estimates. Moreover, the estimates of the velocity of upwelling from 
 the temperature or salinity distribution did not depend upon the 
 cause of the upwelling ; Imt it is an interesting fact that such a vertical 
 current would be expected along the California coast from Ekman's 
 hydrodynamical theory. 
 
 I wish to express my obligation to Dr. W. E. Ritter of this institu- 
 tion, and to my laboratory assistant, ]Mr. Nephi W. Cummings, for 
 his aid in making the computations and for his suggestions while 
 preparing the manuscript. 
 
 Transmitted June 26, 1918. 
 Scripps Institution for Biological Ecscarch 
 of the University of California, 
 La JoUa, California. 
 
OCEAN rKMPEKA'VVEES 419 
 
 LITERATUEE CITED 
 Bjerknes, V. P. K. 
 
 1901. Cirkulation relative zu der Erde. Ofversikt af Koiifjl. Vet.-Akad. 
 Forhandl., 58, 739-757. 
 Ekman, V. W. 
 
 1905-06. On the influence of the earth 's rotation on ocean currents. Arkiv 
 for MatematiU, Astrononii och F3-sik, 2, 1-53, 1 pi. and 10 figs. 
 
 1906. Beitrage zur Theorie der Meerestroraungen. Ann. d. ITydrog. u. 
 
 inarit. Meteor., 34, 423-430, 472-484, 527-540, 566-583, 38 figs. 
 Gehrke, Johan. 
 
 1909. Beitrag zur Hydrographie des finnischen Meerbusens. Finnlandische 
 
 hydrographisch-biologische Untersuchungen, III. (Leipzig, Engel- 
 maun), pp. 1-S9, pis. 1-3, 7 figs, in te-xt. 
 
 1910. Beitrage zur Hydrographie des Ostseebassins. Cons, pernian. explor. 
 
 Mer., Circ. 52, 1-195, 2 pis. 
 
 Grein, K. 
 
 1913. Untersuchungen uber die Absorption des Lichts in Seewasser. I. 
 Die selektive Absorption des Sonnenlichts im Meer nebst einer 
 Besehreibung des Instruments und der Methode. Ann. Inst. 
 Oceanogr., 5, fasc. 6, pp. 1-23, 7 figs, in text. 
 Hann, Julius. 
 
 1915. Lehrbueh der Meteorologie. (Leipzig, Tauehnitz), xiv + 847 pp., 28 
 pis., 69 figs, in text. 
 Helland-Hansen, B. 
 
 1911-12. The ocean waters, an introduction to physical oceanography. I, 
 General jiart (methods). Intern. Rev. d. ges. Ilydrobiol. u. Hydrog., 
 3, Hydrog. Suppl., pp. 1-84, figs. 1-46 in text. 
 Jacobsen, .T. p. 
 
 1913. Beitrag zur Hydrographie der diinisehen Gewasser. Medd. Komm. 
 Ilavandersogelser (Hydrografi), 1, no. 2, 94 pp., 14 pis., 17 figs, in 
 text. 
 Krummel, O. 
 
 1907. Handbuch der Ozeanographie. (Stuttgart, Engelhorn), 1, xv + 526 
 
 pp., 60 figs, in text. 
 
 1911. Ilandbuch der Ozeanographie. (Stuttgart, Engelhorn), 2, xvi + 766 
 
 pp., 182 figs, in text. 
 Krummel, O., and Ruppin, E. 
 
 1905. Ueber die innere Eeibuug des Seewasser. Wiss. Meeresuntersuch- 
 ■ ungen, N. F., 9, 29-36. 
 Knott, C. G. 
 
 1903-05. Ocean temperatures and solar radiation. Proc. Royal Soc. Edin- 
 burgh, 25, part 1, pp. 173-184. 
 Makarofp, S. 
 
 1894. Le "Vitiaz" et 1 'ocean Pacifique. (St. Petersburg, Kummar), 1, 
 xliii + 337 pp. 
 
420 MISCELLANEOUS STUDIES 
 
 McEwEN, G. F. 
 
 1912. The distribution of ocean temperatures along the west coast of North 
 America deduced from Ekman 's theory of the upwelling of the 
 cold water from the adjacent ocean depths. Intern. Eev. d. ges. 
 Hydrobiol. u. Hydrog., 5, 243-286, 21 figs, in text. 
 
 1914. Peculiarities of the California climate explained on the basis of 
 
 general principles of atmospheric and oceanic circulation. U. S. 
 Mon. Weather Rev., 42, 14-23, 13 figs. 
 1916. Summary and interpretation of the hydrographic observations made 
 by the Scripps Institution for Biological Research of the Univer- 
 sity of California, 1908-191.5. Univ. Calif. Puil. Zool., 15, 255-336, 
 pis. 1-38. 
 
 Michael, E. L., and McEwen, G. F. 
 
 1915-16. Hydrographic, plankton and dredging records of the Scripps Insti- 
 tution for Biological Research of the University of California, 
 1901-12. Univ. Calif. Publ. Zool., 15, 1-206, 4 figs, and 1 map in 
 text; ibid., 16, 207-254, 4 figs, in text. 
 
 MoHN, H. 
 
 1887. The Norwegian North Atlantic expedition, 1876-1878: The North 
 Sea, its depths, temperature and circulation. (Christiania, Gron- 
 dahl), 212 pp., 48 pis. 
 
 Moore, W. I. 
 
 1908-11. Meteorological charts of the North Pacific Ocean. U. S. Weather 
 Bureau, Washington, D. C. 
 
 Murray, Sir joun. 
 
 1898. On the annual range of temperature in the surface waters of the ocean 
 and its relation to other oeeanographic phenomena. Geog. Jour., 
 12, 113-137. 
 
 Murray, Sir John, and Hjort, Johan. 
 
 1912. The depths of the ocean. (London, Macmillan), xx + 821 pp., 9 pis., 
 
 575 figs, in text. 
 Nansen, F. 
 
 1913. The waters of the north-eastern North Atlantic. Intern. Rev. d. ges. 
 
 Hydrobiol. u. Hydrog., 4, 1-139, pi. 17, figs. 1-52 in text. 
 
 Petersen, Johannes 
 
 1912. Hydrographie und Meteorologie Finnlands und der benachbarten 
 Meere, nach "Atlas de Finlande." Ann. Hydr. u. marit. Meteor., 
 40, 131-145, 4 figs, in text. 
 
 PULS, Casar. • 
 
 1895. Oberflachentemperaturen und Stromungsverhaltnisse des Aequatorial- 
 giirtels des Stillen Ozeans. Aus dem Archiv der deutschen See- 
 warte, 18, 1-38, pis. 1-4. 
 
 Schmidt, Wilhelm. 
 
 1915. Strahlung und Verdunstung an freien Wasser fliichen. Ein Beitrag zum 
 
 Warmehaushalt des Weltmeers und um Wasserhaushalt der Erde. 
 Ann. Hydr. u. marit. Meteor., 43, 111-124. 
 
OCEAN TEMPEMATURES 421 
 
 SCHOTT, G. 
 
 1895. Die jahrliche Temperaturachwankung des Oceanwassers. Petermauns 
 
 Mittelungen, 41, 153-159, pi. 10. 
 1902. Oceanographie und maritime Meteorologie. Wissenschaftliche Ergeb- 
 
 nisse der deutsclien Tief see-Expedition auf dem Dampfer ' ' Val- 
 
 divia," 1898-1899. (Jena, Fischer), 1, xii + 403 pp., 26 pis. and 
 
 35 figs in text. 
 ScHOTT, Gerhard, und Schu, Fritz. 
 
 1910. Die Wiirmeverteilung in den Stillen Ozeans. Ann. Hydr. u. marit. 
 
 Meteor., 38, 1-26, 15 pis. 1 fig. in text. 
 
 SCHOTT, G., SCHULZ, B., UND PeRLEWITZ, P. 
 
 1914. Die Forschungsreise S.M.S. " Miiwe, " im Jahre 1911. Aus dem Archiv 
 
 der deutsclien Seewarte, 37, vi + 104 pp., 8 pis. and 16 figs, in text. 
 
 Taylor, G. I. 
 
 1915. Eddy motion in the atmosphere. Philos. Trans. Ro3'al Soc. London, 
 
 215, ser. A, pp. 1-26, figs. 1-5 in text. 
 
 THOR.'iDE, H. 
 
 1909. Ueber die Kalifornische Meeresstromungen, Oberflachentemperaturen 
 und Stromungen an der WestUiiste Nordamerikas. Ann. Hydr. u. 
 marit. Meteor., 37, 17-34, 6.3-77, pis. 5, 10, 11, 5 figs, in text. 
 
 1914. Die Geschwindigkeit von Triftsstromungen und die Ekmansche 
 Theorie. Ibid., 42, 379-391, 3 figs, in text. 
 
 Wegemann, G. 
 
 190on. Die vertikale Temperaturverteilung im Weltmeere durch Warmeleit- 
 ung. Wissenschaftliche Meeresuntersuchungen, herausgegeben von 
 der Konimission zur Untersuchung der deutschen Meere in Kiel 
 und der Biologischen Anstalt auf Helgoland. Abteilung Kiel, 
 N. F., 8, 137-143. 
 
 19056. Ursachen der vertikalen Temperaturverteilung im Weltmeere unter 
 besonderer Beriicksichtigung der Warmeleituhg. Ann. Hydr. u. 
 marit. Meteor., 33, 206-211. 
 
 Wharton, W. J. L. 
 
 1894. Presidential address. Section E, Geography. Rept. Brit, .\ssoc. Adv. 
 Sei. 
 
 WiNKELMANN, A. 
 
 1906. Handbueh der Physik. (Leipzig, Barth), 3, xiv + 1178, 206 figs, in 
 text. 
 
 ZOPPRITZ, K. 
 
 1878. Hydrodynamische Probleme in Eeziehung zur Theorie der Meeres- 
 stromungen. Wiedenianns Anrialen, 3, 582-607. 
 
CHANGES IN THE 
 CHEMICAL COMPOSITION 
 OF GRAPES DURING RIPENING 
 
 BY 
 
 F. T. BIOLETTI, W. V. CRUESS, and H. DAVI 
 
 [University of California Publications in Agricultural Sciences, Vol. 3, No. 6, pp. 103-130] 
 
CHANGES IN THE 
 
 CHEMICAL COMPOSITION OF GRAPES 
 
 DURING RIPENING 
 
 BY 
 
 F. T. BIOLETTI, W. V. CRUESS, and H. DAVI 
 
 The investigations reported in this paper were undertaken to 
 determine the changes in chemical composition of vinifera varieties 
 of grapes in California during the growing and ripening stages. A 
 survey of the literature indicated that, although tlie subject had been 
 quite fully investigated in Europe with vinifera varieties and in 
 America with the native varieties, very little had been published iipon 
 the ripening of vinifera varieties under California Conditions. A 
 great many analyses of different varieties of grapes have been made 
 by chemists of the University of California Experiment Station, nota- 
 bly by G. E. Colby, and are reported in the publications of this station.' 
 A paper by G. E. Colby^ gives data upon the nitrogen content of a 
 number of varieties of ripe vinifera grapes. Most of the analyses, 
 however, do not show the changes in composition during ripening. 
 
 Of the more recent European investigations^ some deal with the 
 changes in general composition, others are confined to a discussion of a 
 single component, such as sugar, or coloring matter, or acid principles. 
 
 The changes in composition of American varieties of grapes during 
 ripening have been studied quite thoroughly by "W. B. Alwood* and 
 his associates. These investigations gave particular attention to the 
 
 1 Hilgard, E. W., The composition and classification of grapes, musts, and 
 wines. Rept. of Viticultural Work, Univ. Calif. E.^cper. Sta. Rep., 1887-93, pp. 
 3-360. 
 
 2 Colby, G. E., On the quantities of nitrogenous matters contained in Cali- 
 fornia musts and wines. Ibid., pp. 422-446. 
 
 3 KcHiofer, W., The gi-ape in the various stages of maturity; trans, by E. 
 Zardetti. Gior. Vin. Ital., vol. 34 (1908), no. .30, pp. 475-477. 
 
 Barberon, G., and Changeant, F., Investigations on the development and 
 
 (103 1 
 
424 MISCELLANEOUS STUDIES 
 
 increase in sugar content and changes in acidity during the period 
 in which the grapes were under observation. Alwood and other mem- 
 bers of the Bureau of Chemistry, United States Department of Agri- 
 culture, have also published a ninnber of reports* on the general 
 composition of American varieties of grapes as affected by season, 
 locality, etc. 
 
 The most notable changes taking place during ripening were found 
 by the European and American investigators mentioned above to be : 
 (1) increase in total sugar ; (2) decrease in ratio of glucose to fructose ; 
 (3) decrease in total acid; (4) increase in ratio of cream of tartar to 
 total acid due to decrease in total acid ; (5) decrease in tannin ; and (6) 
 increase in coloring matter. The cream of tartar and protein change 
 very little in percentage during ripening, although, according to the 
 
 composition of varieties of grapes in Abraon-Durso. Ann. Soc. Agr Sci. et Ind., 
 Lyon (8), vol. 1 (1903), pp. 97-159. 
 
 Laborde, J., The transformation of the coloring matter of grapes during 
 ripening. C. E. Acad. Sci. (1908), vol. 17, pp. 753-755. 
 
 Martinand, V., On the occurrence of sucrose and saccharose in different parts 
 of the grape. C. R. Acad. Sci. (1907), vol. 24, pp. 1376-79. 
 
 Boos, L., and Hughes, E., The sugar of the grape during ripening. Ann. 
 Falsif. (1910), vol. Ill, p. 395. 
 
 Bouffard, A., Observations in regard to the proportion of sugar during ripen- 
 ing. Ann. Falsif. (1910), vol. JII, pp. 394-5. 
 
 Zeissig, Investigations on the process of ripening on one-vear-old grape wood. 
 Ber. k. Lehranst. Wien, Obst-u. Garten-bau (1902), pp. 59-64. 
 
 Koressi, F., Biological investigations of the ripening of the wood of the 
 grape. Rev. Gen. Bot., vol. 13 (1901), no. 149, pp. 19,3-211; no. 150, pp. 251-264; 
 no. 151, pp. 307-325. 
 
 Brunet, R., Analvsis and composition of the grape during ripening. Rev. 
 de Viticulture, vol. 37, pp. 15-20. 
 
 Garina, C, Variations in the principal acids of grape juice during the process 
 of maturing. Canina, Ann. R. acad. d 'agricultura di Torino, vol. 57 (1914), 
 p. 233. Cf. Ann. Chim. applicata, vol. 5 (1914), pp. 65-6, See also Ann, r, acad. 
 d 'agr. di Torino, vol. 57, pp. 233-90. 
 
 Baragolia, W. I., and Godet, C, Analytical chemical investigations on the 
 ripening of grapes and the formation of wine from them. Landw. Jahrb., vol. 
 47 (1914), pp. 249-302. 
 
 Riviere, G., and Bailhache, G., Accumulation of sugar and decrease of acid 
 in grapes. Chem. Abs. Jour. (1912), p. 1022; Jour. Soc. Nat. Hort. France (4), 
 pp. 125-7; Bot. Cent., 1912, pp. 117, 431. 
 
 Pantanelli, Enzyme in must of overripe grapes. Chem. Abs. Jour., vol. VI 
 (1912), p. 2447. 
 
 4 Alwood, W, B., Hartmann, J. B., Eoff, J. R., and Sherwood, S. F., Develop- 
 ment of sugar and acid in grapes during ripening. U. S. Dept. Agrie. Bull. 335, 
 April 11, 1916. 
 
 The occurrence of sucrose in grapes. Jour. Indust., vol. II, Eng. Chem. 
 
 (1910), pp. 481-82. 
 
 Sugar and acid content of American native grapes. 8th Inter. Cong. 
 
 Appl. Chem. (1912), Sect. Vla-XIv, pp. 33, 34. 
 
 Enological Studies: the chemical composition of American grapes grown 
 
 in Ohio, New York, and Virginia. IT. S. Dept. Agric. Bur. Chem. Bull. 145, 1911. 
 
 Crystallization of cream of tartar in the fruit of grapes. U. S. Dept. 
 
 Agric. Jour. Agric. Research (1914), pp. 513, 514. 
 
 Alwood, W. B., Hartmann, B. G., Eoff, J. R., Sherwood, S. F., Carrero, J, C, 
 and Harding, T. J., The chemical composition of American grapes grown in the 
 central and eastern states. XT . S. Dept. Agrie. ( 1916) Bull. 452. 
 
 11041 
 
CHEMICAL COMI'OSITION OF Gl^.tl'IiS 425 
 
 investigations rcferreti to, there is a slight increase in both of these 
 constituents. 
 
 In the investigations reported in the present paper, particular 
 attention was given to increase in total solids and sugar, decrease in 
 total acid, and changes in protein and cream of tartar in the must or 
 juice of the grapes. The ripening of the leaves was traced by noting 
 the changes in starch, sugar, acid, and protein content. 
 
 Sampling. — During 1914 and 1915 samples of fruit were taken 
 from the time the grapes had reached full size but were still hard and 
 green until they had become overripe. During 1916 the first samples 
 were taken shortly after the berries had set and before the seeds had 
 formed. The last samples were taken when the grapes had become 
 overripe. Samples of leaves were also taken in 1916 on the same 
 dates that samplings of the grapes were made. The samples were 
 taken at intervals of approximately one week. They were in all cases 
 taken from the experimental vineyard at Davis.'* 
 
 Five-pound samples of grapes were used. The grapes were picked 
 from the first crop, except in 1914, when a comparison of the ripening 
 of first and second crops was made. An ordinary five-pound grape 
 basket was filled with leaves at each sampling. The samples of grapes 
 and leaves were shipped from the vineyard to the laboratorj' at 
 Berkeley, where the grapes were placed in an Enterprise fruit crusher 
 and pressed. The juice was sterilized in bottles at 212° F. The leaves 
 were ground in an Enterprise food chopper and sterilized at 212° F 
 in wide mouth, air tight bottles. The samples were then reserved for 
 chemical examination. 
 
 In 1914 it was found that there was considerable irregularity in 
 the variation of samples from week to week. For example, instead 
 of an increase of total solids during the periods between samplings, a 
 slight decrease was found in a few samples. During the 1915 season 
 it was therefore considered of interest to note what effect certain 
 factors might have upon the composition of samples taken on the 
 same date. 
 
 1. Effect of Age of Vine. The entire first crop from three large 
 old vines and from three small young vines, all of the Muscat variety, 
 was picked, crushed, and pressed. Analyses of the juices were made 
 with the following results : 
 
 ^ The authors wish to express their appreciation of the assistance of F. C. 
 Flossfeder, of the University Farm at Davis, who gathered most of the samples 
 reported upon in this paper. 
 
 11051 
 
426 MISCELLANEOUS STUDIES 
 
 Table 1 — Effect of Age of Vine on Balling and Acid of Must of Muscat 
 
 Grapes 
 
 Vine Balling Acid 
 
 Small, no. 1 24.7 .67 
 
 Small, no. 2 27.7 .49 
 
 Small, no. 3 27.6 .67 
 
 Large, no. 1 22.0 .88 
 
 Large, no. 2 23.5 .75 
 
 Large, no. 3 23.6 .76 
 
 Average, small 26.7 .61 
 
 Average, large 23.0 .81 
 
 Difference 3.7 — .20 
 
 The results show rather strikingly that young vines ripen their 
 fruit earlier than do mature vines. This fact makes it essential that 
 samples, to be comparative, must be taken from vines of the same age. 
 
 2. Comparison of Grapes from North and South Sides of Vines. 
 The whole first crop from three large Muscat vines was picked. The 
 bunches from the north and south sides of each vine were kept sep- 
 arate. The}- were crushed, pressed, and analyzed for Balling and acid 
 content. 
 
 Table 2 — Comparison of B.vlling and Acid of Juice from Grapes Picked from 
 North and South Sides of Vines 
 
 Vine and side of vine Balling Acid 
 
 1-N 21.3 .92 
 
 as 22.7 .84 
 
 2-N 23.5 .81 
 
 2-S 23.5 .80 
 
 3-N 23.1 .81 
 
 3-S 24.1 .71 
 
 Average, N side 22.63 .85 
 
 Average, S side 23.43 .78 
 
 Difference 80 —.07 
 
 The tests indicate that grapes located on the south side of the vine 
 ripen more rapidl.y than those on the north side. This difference is 
 apparently due to the fact that the south side of the vine receives 
 more heat than the north side. 
 
 3. Effect of Location of Bunch on Cane. Grapes of first crop, from 
 canes showing two bunches each, were picked and the bunches from 
 near tlie bases of the canes kept separate from those near the tip of 
 the cane. They were crushed, pressed, and analyzed for Balling and 
 acid. 
 
 [1061 
 
CHEMICAL COMPOSITION OF GRAPES 427 
 
 Table 3 — Effect of Location op Bunch on Cane 
 
 Nearest base of cane Nearest tip of cane 
 
 Vine Balling Acid Balling Acid 
 
 Muscat, no. 1, cane 1 25.1 .73 23.7 .83 
 
 Muscat, no. 1, cane 2 25.6 .79 24.8 .80 
 
 Muscat, no. 2, cane 1 25.1 .85 24.6 .87 
 
 Muscat, no. 2, cane 2 25.2 .78 24.7 .85 
 
 Muscat, no. 3, cane 1 23.0 .79 22.6 .82 
 
 Muscat, no. 3, cane 2 24.5 .73 23.8 .73 
 
 Muscat, no. 4, cane 1 ., 24.2 .90 25.2 .90 
 
 Muscat, no. 4, cane 2 24.5 .68 23.8 .83 
 
 Tokay, cane 1 21.2 .67 21.2 .80 
 
 Tokay, cane 2 23.0 .63 22.4 .76 
 
 Sultanina, cane 1 23.3 .61 22.3 .62 
 
 Sultanina, cane 2 22.5 .61 23.0 .63 
 
 Sultana, cane 1 23.2 .78 21.6 .70 
 
 Sultana, cane 2 21.1 .90 20.0 1.20 
 
 Palomino, cane 1 25.1 23.5 
 
 Palomino, cane 2 22.0 23.7 
 
 Means 24.9 .75 23.1 .81 
 
 The data indicate that bunches at the base of the cane ripen in most 
 cases more rapidly than those near the tip. although this relation does 
 not always hold and may be reversed in some instances. 
 
 4. Variation in Balling Degree of Must from Bunches of Similar 
 Appearance and Size from Same Vineyard and Gathered on Same 
 Date. A five-pound basket of grapes of first crop and selected for 
 similarity of color, .size of bunch, and general appearance was picked 
 from each of a number of vines in the same vineyard. Vines of 
 similar size and appearance were chosen. Several varieties were rep- 
 resented in the experiment. Tests of Balling degree only were made. 
 
 Table 4 — Variation in Balling in Must From Grapes of Same Variety Picked 
 Prom Different Vines of Similar Appearance 
 
 Variety 
 
 Vine 
 number 
 
 Balling 
 
 Mean 
 Balling 
 
 Maxiimira 
 variation 
 
 Cornichon 
 
 3 
 
 14.5 
 
 
 
 Cornichon 
 
 6 
 
 15.0 
 
 
 
 Cornichon 
 
 9 
 
 14.2 
 
 
 
 Cornichon 
 
 11 
 
 14.7 
 
 
 
 Cornichon 
 
 
 16.1 
 
 14.9 
 
 1.9 
 
 Emperor ■ 
 
 Emperor 
 
 Emperor 
 
 10 
 11 
 13 
 
 12.0 
 14.5 
 
 1.5.2 
 
 
 
 Emperor 
 Emperor 
 
 14 
 
 17 
 
 i5.n 
 
 15.0 
 
 UA 
 
 3.5 
 
 Malaga 
 Malaga 
 
 5 
 6 
 
 18.5 
 17.2 
 
 
 
 [1071 
 
•128 
 
 MISCELLANEOUS STUDIES 
 
 Table 4 — {Continued) 
 
 
 Vine 
 
 
 Mean 
 
 Maximum 
 
 Variety 
 
 number 
 
 Balling 
 
 Balling 
 
 variation 
 
 Malaga 
 
 7 
 
 19.7 
 
 
 
 Malaga 
 
 9 
 
 18.5 
 
 
 
 Malaga 
 
 11 
 
 19.2 
 
 18.6 
 
 2.0 
 
 Muscat 
 
 * 
 
 21.7 
 
 
 
 Muscat 
 
 * 
 
 21.1 
 
 
 
 Muscat 
 
 • 
 
 20.9 
 
 
 
 Muscat 
 
 « 
 
 21.5 
 
 
 
 Muscat 
 
 ♦ 
 
 21.7 
 
 21.4 
 
 .8 
 
 Palomino 
 
 3 
 
 19.5 
 
 
 
 Palomino 
 
 4 
 
 21.0 
 
 
 
 Palomino 
 
 6 
 
 21 2 
 
 
 
 Palomino 
 
 7 
 
 20.7 
 
 
 
 Palomino 
 
 9 
 
 18.8 
 
 20.2 
 
 2.4 
 
 Sultanina 
 
 » 
 
 22.5 
 
 
 
 Sultanina 
 
 # 
 
 21.5 
 
 
 
 Sultanina 
 
 * 
 
 18.7 
 
 
 
 Sultanina 
 
 * 
 
 22.0 
 
 
 
 Sultanina 
 
 * 
 
 22.6 
 
 21.5 
 
 3.9 
 
 Tokay 
 
 * 
 
 19.8 
 
 
 
 Tokay 
 
 * 
 
 19.3 
 
 
 
 Tokay 
 
 » 
 
 18.7 
 
 
 
 Tokay 
 
 * 
 
 20.7 
 
 
 
 Tokay 
 
 * 
 
 19.5 
 
 19.6 
 
 2.0 
 
 Pedro Zumbon 
 
 7 
 
 21.5 
 
 
 
 Pedro Zumbon 
 
 4 
 
 21.2 
 
 
 
 Pedro Zumbon 
 
 6 
 
 20.6 
 
 
 
 Pedro Zumbon 
 
 3 
 
 18.5 
 
 
 
 Pedro Zumbon 
 
 5 
 
 19.8 
 
 20.3 
 
 3.0 
 
 Emperor 
 
 15 
 
 18.1 
 
 
 
 Emperor 
 
 8 
 
 15.8 
 
 
 
 Emperor 
 
 14 
 
 16.2 
 
 
 
 Emperor 
 
 9 
 
 16.8 
 
 
 
 Emperor 
 
 16 
 
 16.3 
 
 16.6 
 
 2.3 
 
 Cornichon 
 
 4 
 
 17.3 
 
 
 
 Cornichon 
 
 9 
 
 16.3 
 
 
 
 Cornichon 
 
 10 
 
 17.9 
 
 
 
 Cornichon 
 
 11 
 
 17.8 
 
 
 
 Cornichon 
 
 13 
 
 18.0 
 
 17.5 
 
 1.7 
 
 Malaga 
 
 4 
 
 18.3 
 
 
 
 Malaga 
 
 5 
 
 20.4 
 
 
 
 Malaga 
 
 6 
 
 20.0 
 
 
 
 Malaga 
 
 8 
 
 20.1 
 
 19.7 
 
 1.8 
 
 Mean variation, six ripest varieties 2.32 
 
 Mean variation, six least ripe varieties 2.20 
 
 Average variation, whole series 2.30 
 
 ' Adjacent vines. 
 
 [1081 
 
cuEMivAi. vvMroanios of ghapks 429 
 
 The data illustrate the difficulty of selecting five-pound lots of the 
 same variety that will represent average samples. 
 
 5. Effect of Location of Berries on the Bunch. All of the bunches 
 of the first crop were taken from two Muscat vines. The bunches 
 were cut into top and bottom halves. These lots were cruslied sep- 
 arately, pressed, and the juices analyzed. 
 
 T.A.BLE 5 — Effect of Location of 'Berries on Bunch 
 
 Sample Balling Acid 
 
 Vine no. 1, stem end of bunch 23.6 .76 
 
 Vine no. 1, apical end of bunch 22.7 .87 
 
 Vine no. 2, stem end of bunch 21.3 .92 
 
 Vine no. 2, apical end of bunch 21.3 .93 
 
 The results show that considerable variation in composition of the 
 berries may exist within the same bunch. 
 
 6. Effect of Thoroughness of Pressing. About ten pounds of Mus- 
 cat grapes were crushed and lightly pressed. The pulp and skins left 
 from this pressing were then thoroughly crushed and pressed a second 
 time. The juices from the two lots were analyzed separately. 
 
 Table 6 — Effect of Thoroughness of Pressing 
 
 Sample Balling Acid 
 
 First pressing 22.8 .78 
 
 Second pressing 22.8 .79 
 
 There was practically no difference between the juices from lightly 
 and thoroughly pressed grapes of the same lot. 
 
 The data from the above six tests indicate that it is a very difficult 
 matter to select grapes that will represent a fair average sample of 
 the grapes to be studied. The size and age of the vine, the side of 
 the vines, the location of the bunch on the cane, and individual vines, 
 all affect the composition of the juice from the grapes very materially, 
 and these factors should be taken into account when samples are 
 taken. 
 
 Preservation of Samples and Preparation for Analysis. — In 1914 
 the samples of juice were preserved with HgCl,, 1 :1000. In 1915 and 
 1916 the samples were sterilized at 100° G. Before analysis the bottles 
 were heated to 100° C for an hour to dissolve any cream of tartar which 
 might have separated. The juices were filtered before analysis. Con- 
 siderable coagulation of dissolved solids took place during sterilization. 
 
 1 109 I 
 
430 MISCELLANEOUS STUDIES 
 
 Methods of Analysis. — The samples were analyzed by the methods 
 in use in the Agricultural Chemistry Laboratory and the Nutrition 
 Laboratory of this station. A brief description of the methods follows : 
 
 1. Total Solids. The juice was filtered clear and cooled below 
 15° C. The specific gravity was determined by a pyenometer at 
 15?5 C. The corresponding total solids, or extract, was found from 
 Windisch's tables in Leach's Food Anahjsis, page 697. This table 
 gives the extract as "grams per 100 grams"; that is, per cent by 
 weight. To calculate the corresponding grams per 100 c.c, the per 
 cent by weight was multiplied by the specific gravity. This gives a 
 figure not very much greater than grams per 100 grams in juices of 
 low specific gravity, but gives a figure as much as 2 per cent greater 
 where the total solids are much above 20 per cent. The two methods 
 of reporting total solids has in the past led to much unnecessary 
 confusion. It is therefore urged that the reader bear in mind the 
 distinction between the two methods when reading the discussions in 
 this paper or examining the curves. 
 
 2. Sugar. The sample was filtered ; an aliquot was treated with 
 lead acetate ; diluted to mark ; filtered ; lead removed witli anhydrous 
 Na„CO..,, and the sugar determined in an alicjuot by the gravimetric 
 method, using Soxhlot's modification of Fchling's solution. The Cu.^O 
 was weighed directly after drying at 100° C. The corresponding 
 sugar as invert sugar was obtained from Munsou and Walker's table 
 in Leach's Food Anahjsis. The grams of invert sugar ]ier 100 c.c. 
 found in this way was divided by the specific gravity of the must to 
 obtain the corresponding grams per 100 grams of juice. 
 
 3. Total acid was determined by titration of a 10 c.c. sample with 
 N/10 NaOII, using phenolphthaleiii as an indicator, and is reported 
 as tartaric acid, grams per 100 c.c. 
 
 4. Cream of tartar was estimated by a method suggested by Pro- 
 fessor D. R. Hoagland of the Division of Agricultural Chemistry. 
 Ten c.c. of the juice was incinerated at a low heat in a muffle furnace 
 until well carbonized, but not to a white ash. (Excessive heating 
 results in loss of K by volatilization.) The KXO3 formed by incin- 
 eration was leached out with hot water and a known excess of N/10 
 HCl added. This was titrated back with N/10 NaOH, using methyl 
 orange as an indicator. The K.CO3 is obtained by difference and 
 calculated back to cream of tartar, assuming that all of the KoCO, is 
 formed by the oxidation of cream of tartar, KH(C^H^O,;)- It is 
 
 I II" I 
 
CHEMICAL COMPOSITION OF GRAPES 431 
 
 reported as grams KI^C^H^O,,) per 100 c.c, and also as tartaric 
 acid. 
 
 5. Free Tartaric Acid was obtained by difference between total acid 
 and cream of tartar calculated as tartaric acid. It is reported as 
 grams per 100 c.c. 
 
 6. Protein in the juice was determined by the usual Kjeldahl- 
 Gunning method upon a 10 c.c. sample. It is reported as grams per 
 100 c.c. 
 
 7. Moisture in the leaves was determined by drying the sample at 
 100° C. 
 
 8. Sugar in the leaves was estimated by leaching the dried sample 
 with cold water and determining sugar by the gravimetric Pehling 
 method in the filtrate. 
 
 9. Starch in the leaves was determined by hydrolysis of the dried 
 ground sample with dilute HCl at 100^ C, followed by filtration and 
 the usual gravimetric Fehling method for juice described above. 
 
 10. Protein in the leaves was determined by the Kjeldahl-Gunning 
 method on .5 gram samples. 
 
 11. Acid in the leaves was estimated by leaching in hot w-ater and 
 titrating in the presence of the leaves, using litmus paper as indicator. 
 
 Analyses of Musts from Grape-Ripening Samples, 191 i, lOl'j, 1916. 
 The data from the analyses have been assembled in the following 
 tables. Owing to the size of the tables, abbreviations have been 
 necessary for the headings of the columns. 
 
 EXPLAXATIOXS OP FeaDIXGS OF TABLES 
 
 1. Sp. gr. ^ Specific gravity at 15?5 C. 
 
 2. T. S. G. = Total solids in grams per 100 grams. 
 
 3. T. S. C. = Total solids in grams per 100 c.c. 
 
 4. S.G.^ Sugar in grams per 100 c.c. 
 
 5. S. I. ^ Sugar in grams per 100 grams. 
 
 6. Tl. A. = Total acid in grams per 100 c.c. 
 
 7. C.T. = Cream of tartar in grams per 100 c.c. 
 
 8. C. T. T. = Cream of tartar as tartaric acid, grams per 100 c.c. 
 
 9. T. A. = Total free acid as tartaric obtained by subtracting cream of tartar 
 as tartaric from total acid as tartaric. 
 
 10. P. ^Protein, grams per 100 c.c. 
 
 11. S. = Sum of sugar, cream of tartar, tartaric acid, and protein in grams 
 per 100 c.c. 
 
 12. T. S. — S. = Total solids (T. S. C.) — S (preceding column). 
 
 run 
 
432 
 
 MISCELLANEOUS STUDIES 
 
 Table 7 — Grape Ripening Tests, 1914 
 (Grapes from Davis) 
 
 Malaga 
 
 
 
 
 
 
 
 
 
 
 
 
 
 First crop: 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Variety 
 and date 
 
 1 
 Sp- gr. 
 
 O 
 
 T. S. G, 
 
 3 
 
 , T, S. C 
 
 4 
 , S. 6, 
 
 5 
 S, I. 
 
 6 
 
 Tl, A, 
 
 7 
 C. T, 
 
 8 
 C.T.T 
 
 9 
 , T, A, 
 
 10 
 
 p. 
 
 11 
 S. 
 
 12 
 T, S, S, 
 
 Aug. 19 
 
 1.0396 
 
 10.25 
 
 10.65 
 
 7.32 
 
 7.04 
 
 2.78 
 
 .35 
 
 .13 
 
 2.65 
 
 .21 
 
 10.53 
 
 .12 
 
 Aug 26 
 
 1.0413 
 
 10.69 
 
 11.13 
 
 7.84 
 
 7.53 
 
 2.65 
 
 .36 
 
 .14 
 
 2.51 
 
 .25 
 
 10.96 
 
 .17 
 
 Aug. 26 
 
 1.0595 
 
 15.42 
 
 16.33 
 
 13.37 
 
 12.62 
 
 .77 
 
 .48 
 
 .19 
 
 .58 
 
 .55 
 
 14.98 
 
 1.41 
 
 Aug. 26 
 
 1.0613 
 
 15.87 
 
 16.84 
 
 14.31 
 
 13.50 
 
 1.46 
 
 .31 
 
 .12 
 
 1.34 
 
 .33 
 
 16.29 
 
 .55 
 
 Aug. 26 
 
 1.0694 
 
 18.01 
 
 19.25 
 
 16.59 
 
 15.52 
 
 1.00 
 
 .36 
 
 .14 
 
 .86 
 
 .38 
 
 18.19 
 
 1.06 
 
 Aug. 31 
 
 1.0732 
 
 19.00 
 
 20.39 
 
 17.65 
 
 16.45 
 
 .87 
 
 .55 
 
 .22 
 
 .65 
 
 .45 
 
 19.30 
 
 1.09 
 
 Sept. 23 
 
 1.0736 
 
 19.10 
 
 20.50 
 
 17.83 
 
 16.60 
 
 .74 
 
 .38 
 
 .15 
 
 .59 
 
 .52 
 
 19.32 
 
 1.18 
 
 Oct. 5 
 
 1.0965 
 
 25.12 
 
 27.54 
 
 24.89 
 
 22.70 
 
 .72 
 
 .50 
 
 .20 
 
 .52 
 
 .57 
 
 26.48 
 
 1.06 
 
 Second crop: 
 
 
 
 
 
 
 
 
 
 
 
 
 Aug. 10 
 
 1.0213 
 
 5.51 
 
 5.62 
 
 2.07 
 
 2.03 
 
 3.22 
 
 .23 
 
 .09 
 
 3.13 
 
 .17 
 
 5.60 
 
 .02 
 
 Aug. 31 
 
 1.0495 
 
 12.82 
 
 13.45 
 
 9.58 
 
 9.13 
 
 2.51 
 
 .40 
 
 .16 
 
 2.35 
 
 .28 
 
 12.61 
 
 .84 
 
 Sept. 14 
 
 1.0532 
 
 13.78 
 
 14.51 
 
 11.89 
 
 11.30 
 
 2.07 
 
 .37 
 
 .15 
 
 1.92 
 
 .31 
 
 14.49 
 
 .02 
 
 Sept. 23 
 
 1.0670 
 
 17.43 
 
 18.60 
 
 15,29 
 
 14.33 
 
 1.54 
 
 .50 
 
 .20 
 
 1.35 
 
 .29 
 
 17.43 
 
 1.17 
 
 Sept. 23 
 
 1.0869 
 
 22.59 
 
 24.,55 
 
 22.04 
 
 20.19 
 
 1.07 
 
 .45 
 
 .18 
 
 .89 
 
 .41 
 
 23.79 
 
 .76 
 
 Oct. 5 
 
 1.0930 
 
 24.20 
 
 26.45 
 
 23.90 
 
 21.87 
 
 .94 
 
 .48 
 
 .19 
 
 .75 
 
 .41 
 
 25.54 
 
 .91 
 
 Tokay 
 
 
 
 
 
 
 
 
 
 
 
 
 
 First crop: 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Aug. 2 
 
 1,0454 
 
 11.75 
 
 12,28 
 
 8.73 
 
 8.35 
 
 2.63 
 
 .46 
 
 .18 
 
 2.45 
 
 .32 
 
 11.96 
 
 .32 
 
 Aug. 10 
 
 1.0624 
 
 16.08 
 
 17.08 
 
 14,28 
 
 13.44 
 
 1.56 
 
 .45 
 
 .18 
 
 1.38 
 
 .27 
 
 16.38 
 
 .70 
 
 Aug. 19 
 
 1.0682 
 
 17.69 
 
 18.90 
 
 15.94 
 
 14.92 
 
 1.32 
 
 .45 
 
 .18 
 
 1.14 
 
 .27 
 
 17.80 
 
 1.10 
 
 Aug. 3i 
 
 1.0849 
 
 22.09 
 
 23.97 
 
 21.87 
 
 20.16 
 
 .63 
 
 .59 
 
 .23 
 
 .40 
 
 .40 
 
 23.26 
 
 .71 
 
 Sept. 4 
 
 1.0865 
 
 22.49 
 
 24,44 
 
 22.21 
 
 20.44 
 
 .77 
 
 .43 
 
 .17 
 
 .60 
 
 .32 
 
 23.56 
 
 .88 
 
 Sept. 4 
 
 1.0912 
 
 23.72 
 
 25.88 
 
 23.44 
 
 21.48 
 
 .59 
 
 .64 
 
 .25 
 
 .44 
 
 .41 
 
 24.93 
 
 .95 
 
 Sept. 23 
 
 1.0937 
 
 24.38 
 
 26.66 
 
 24.15 
 
 22.08 
 
 .58 
 
 .49 
 
 .19 
 
 .30 
 
 .39 
 
 25.33 
 
 1.33 
 
 Oct. 14 
 
 1.0991 
 
 25.80 
 
 28.36 
 
 25.55 
 
 23.25 
 
 .45 
 
 .54 
 
 .21 
 
 .24 
 
 .45 
 
 26.78 
 
 1.58 
 
 Oct. 14 
 
 1.1000 
 
 26.04 
 
 28.64 
 
 25.78 
 
 23.44 
 
 .52 
 
 .58 
 
 .23 
 
 .29 
 
 .58 
 
 27.23 
 
 1.41 
 
 Second crop 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Aug. 19 
 
 1.0657 
 
 17.04 
 
 18.16 
 
 15.03 
 
 14.10 
 
 1.91 
 
 ..50 
 
 .20 
 
 1.70 
 
 .32 
 
 16.55 
 
 .61 
 
 Sept. 14 
 
 1.0701 
 
 18.19 
 
 19.47 
 
 16.68 
 
 15,59 
 
 1.29 
 
 .52 
 
 .21 
 
 1.11 
 
 .33 
 
 18.64 
 
 .83 
 
 Sept. 23 
 
 1.0769 
 
 19.95 
 
 21.48 
 
 19.22 
 
 17.85 
 
 1.01 
 
 .48 
 
 .19 
 
 .82 
 
 .40 
 
 20.92 
 
 .56 
 
 Oct. 14 
 
 1.0911 
 
 23.70 
 
 25.86 
 
 23.43 
 
 21.47 
 
 .69 
 
 .60 
 
 .24 
 
 .45 
 
 .40 
 
 24.88 
 
 .98 
 
 Table 8 — Grape Ripening Tests, 1915 
 (Grapes from D.ivis) 
 
 Cornichon 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Variety 
 and date 
 
 1 
 Sp. gr. 
 
 2 
 t, S, G, 
 
 3 
 T. S. C. 
 
 4 
 S, G. 
 
 5 
 S, I. 
 
 6 
 
 Tl, A, 
 
 7 
 C, T, 
 
 8 
 C. T, T. 
 
 9 
 T, A, 
 
 10 
 
 P, 
 
 11 
 
 S, 
 
 12 
 
 T. S. S. 
 
 Aug. 22 
 
 1.0324 
 
 8.38 
 
 8.65 
 
 3.99 
 
 3.86 
 
 3.05 
 
 .58 
 
 .23 
 
 2.82 
 
 .38 
 
 7.77 
 
 .88 
 
 Sept. 1 
 
 1.0514 
 
 13.31 
 
 13.99 
 
 10.70 
 
 10.18 
 
 1.62 
 
 .61 
 
 .25 
 
 1.37 
 
 .42 
 
 13.10 
 
 .89 
 
 Sept. 15 
 
 1.0688 
 
 17.85 
 
 19.08 
 
 15.94 
 
 14.91 
 
 .97 
 
 .70 
 
 .28 
 
 .69 
 
 .43 
 
 17.76 
 
 1.32 
 
 Sept. 22 
 
 1.0723 
 
 18.76 
 
 20.12 
 
 16.97 
 
 15.83 
 
 .94 
 
 .71 
 
 .28 
 
 .66 
 
 .46 
 
 18.80 
 
 1.32 
 
 Sept. 29 
 
 1.0737 
 
 19.13 
 
 20.54 
 
 18,31 
 
 17.05 
 
 .87 
 
 .75 
 
 .30 
 
 .61 
 
 .66 
 
 20.33 
 
 .21 
 
 Oct. 7 
 
 1.0781 
 
 20.28 
 
 21.86 
 
 19.41 
 
 18.02 
 
 .71 
 
 .73 
 
 .29 
 
 .42 
 
 .48 
 
 21.04 
 
 .82 
 
 Oct. 14 
 
 1.0843 
 
 21.91 
 
 23.76 
 
 20.40 
 
 18.81 
 
 .78 
 
 .68 
 
 .27 
 
 .62 
 
 .66 
 
 22.36 
 
 1.40 
 
 Oct. 22 
 
 1.0873 
 
 22.70 
 
 24.68 
 
 21.06 
 
 19.37 
 
 .75 
 
 .78 
 
 .31 
 
 .44 
 
 .46 
 
 22.74 
 
 1.94 
 
 [1121 
 
CHEMICAL COMPOSITION OF GKAPES 
 
 433 
 
 Emperor 
 
 Table 8 — (Continued) 
 
 "Variety 
 and date 
 
 1 
 Sp. gr. 
 
 2 
 
 T. s" G. 
 
 3 
 T. S. C. 
 
 4 
 S. G. 
 
 5 
 
 S. I. 
 
 6 
 
 Tl. A. 
 
 7 
 C. T. 
 
 8 
 C.T.T. 
 
 9 
 T. A. 
 
 10 
 
 p. 
 
 11 
 
 S. 1 
 
 12 
 . A. S 
 
 Aug. 19 
 
 1.0420 
 
 10.87 
 
 11.33 
 
 6.96 
 
 6.68 
 
 2.33 
 
 .38 
 
 .15 
 
 2.18 
 
 .38 
 
 9.90 
 
 1.43 
 
 Sept. 1 
 
 1.0479 
 
 12.40 
 
 12.99 
 
 9.82 
 
 9.37 
 
 1.89 
 
 .40 
 
 .16 
 
 1.73 
 
 .62 
 
 12.57 
 
 .42 
 
 Sept. 7 
 
 1.0560 
 
 14.51 
 
 15.32 
 
 11.48 
 
 10.87 
 
 1.70 
 
 .47 
 
 .19 
 
 1.57 
 
 .54 
 
 14.00 
 
 1.32 
 
 Sept. 15 
 
 1.0632 
 
 16.37 
 
 17.40 
 
 14.88 
 
 14.00 
 
 1.40 
 
 .53 
 
 .21 
 
 1.18 
 
 .54 
 
 17.13 
 
 .27 
 
 Sept 22 
 
 1.0652 
 
 16.91 
 
 18.01 
 
 15.46 
 
 14.51 
 
 .93 
 
 .48 
 
 .19 
 
 .74 
 
 .55 
 
 17.23 
 
 .78 
 
 Sept. 29 
 
 1.0672 
 
 17.43 
 
 18.60 
 
 16.37 
 
 15.34 
 
 .91 
 
 .48 
 
 .19 
 
 .72 
 
 .66 
 
 18.23 
 
 .37 
 
 Oct. 7 
 
 1.0744 
 
 19.31 
 
 20.75 
 
 17.82 
 
 16.59 
 
 .79 
 
 .58 
 
 .23 
 
 .56 
 
 .51 
 
 19.47 
 
 1.28 
 
 Oct. 14 
 
 1.0765 
 
 19.86 
 
 21.38 
 
 18.37 
 
 17.06 
 
 .79 
 
 .59 
 
 .24 
 
 .56 
 
 .63 
 
 20.15 
 
 1.23 
 
 Oct. 22 
 
 1.0792 
 
 20.57 
 
 22.20 
 
 19.81 
 
 18.36 
 
 .75 
 
 .63 
 
 .25 
 
 .49 
 
 .66 
 
 21.59 
 
 .61 
 
 Malaga 
 
 Aug. 
 
 19 
 
 1.0546 
 
 14.14 
 
 14.91 
 
 12.47 
 
 11.82 
 
 2.05 
 
 .36 
 
 .15 
 
 1.90 
 
 .75 
 
 15.48 
 
 .57 
 
 Aug. 
 
 2^ 
 
 1.0651 
 
 16.86 
 
 17.96 
 
 14.53 
 
 13.64 
 
 1.66 
 
 .46 
 
 .18 
 
 1.48 
 
 .90 
 
 17.37 
 
 .59 
 
 Sept. 
 
 1 
 
 1.067S 
 
 17.59 
 
 18.78 
 
 16.75 
 
 15.69 
 
 1.38 
 
 .44 
 
 .18 
 
 1.20 
 
 .89 
 
 19.28 
 
 .50 
 
 Sept 
 
 7 
 
 1.0719 
 
 18.66 
 
 19.50 
 
 17.00 
 
 15.86 
 
 1.29 
 
 .44 
 
 .18 
 
 1.11 
 
 .70 
 
 19.25 
 
 .25 
 
 Sept. 
 
 15 
 
 1.0758 
 
 19.68 
 
 21.17 
 
 18.17 
 
 16.89 
 
 1.21 
 
 .62 
 
 .25 
 
 .96 
 
 .70 
 
 20.45 
 
 .72 
 
 Sept. 
 
 22 
 
 1.0760 
 
 19.81 
 
 21.32 
 
 18.39 
 
 17.09 
 
 1.18 
 
 .61 
 
 .25 
 
 .93 
 
 .74 
 
 20.67 
 
 .65 
 
 Sept. 
 
 29 
 
 1.0812 
 
 21.20 
 
 22.92 
 
 18.48 
 
 17.09 
 
 1.07 
 
 .58 
 
 .23 
 
 .84 
 
 .75 
 
 20.65 
 
 2.27 
 
 Oct. 
 
 7 
 
 1.0838 
 
 21.78 
 
 23.61 
 
 21.03 
 
 19.40 
 
 1.07 
 
 .65 
 
 .26 
 
 .81 
 
 .73 
 
 23.22 
 
 .39 
 
 Oct. 
 
 14 
 
 1.0970 
 
 25.25 
 
 27.70 
 
 24.58 
 
 22.41 
 
 .59 
 
 .83 
 
 .33 
 
 .26 
 
 .88 
 
 26.55 
 
 1.15 
 
 Muscat 
 
 Aug. 19 
 
 1.0615 
 
 15.94 
 
 16.92 
 
 13.93 
 
 13.12 
 
 1.70 
 
 .36 
 
 .15 
 
 1.55 
 
 .70 
 
 16.54 
 
 .38 
 
 Aug. 25 
 
 1.0744 
 
 19.31 
 
 20.75 
 
 17.96 
 
 16.72 
 
 1.21 
 
 .62 
 
 .25 
 
 .96 
 
 .62 
 
 20.16 
 
 .59 
 
 Sept. 1 
 
 1.0805 
 
 20.91 
 
 22.59 
 
 19.50 
 
 18.05 
 
 .79 
 
 .63 
 
 .25 
 
 .54 
 
 .63 
 
 •21.30 
 
 1.29 
 
 Sept. 7 
 
 1.0827 
 
 21.47 
 
 23.25 
 
 20.39 
 
 18.83 
 
 .76 
 
 .65 
 
 .26 
 
 .50 
 
 .66 
 
 22.20 
 
 1.05 
 
 Sept. 15 
 
 1.0917 
 
 23.85 
 
 26.04 
 
 23.49 
 
 21.52 
 
 .96 
 
 .58 
 
 .23 
 
 .73 
 
 .58 
 
 25.38 
 
 .66 
 
 Sept. 22 
 
 1.0954 
 
 24.14 
 
 26.44 
 
 24.54 
 
 22.40 
 
 .77 
 
 .62 
 
 .25 
 
 .52 
 
 .85 
 
 26.53 
 
 .09 
 
 Sept. 29 
 
 1.1048 
 
 27.30 
 
 30.16 
 
 27.01 
 
 24.45 
 
 .72 
 
 .72 
 
 .29 
 
 .44 
 
 .72 
 
 28.89 
 
 1.27 
 
 Oct. 7 
 
 1.1079 
 
 28.12 
 
 31.15 
 
 28.28 
 
 25.53 
 
 .66 
 
 .59 
 
 .23 
 
 .43 
 
 .66 
 
 29.96 
 
 1.19 
 
 Pedro Zum 
 
 ion 
 
 
 
 
 
 
 
 
 
 
 
 
 Aug. 19 
 
 1.0555 
 
 14.38 
 
 15.18 
 
 11.96 
 
 11.33 
 
 1.81 
 
 .68 
 
 .27 
 
 1.54 
 
 .33 
 
 14.51 
 
 .67 
 
 Aug. 25 
 
 1.0588 
 
 15.24 
 
 16.14 
 
 13.77 
 
 13.01 
 
 1.09 
 
 .57 
 
 .23 
 
 .86 
 
 .53 
 
 15.73 
 
 .41 
 
 Sept. 1 
 
 1.0642 
 
 16.64 
 
 17.71 
 
 15.61 
 
 14.67 
 
 .58 
 
 .52 
 
 .21 
 
 .37 
 
 .43 
 
 16.93 
 
 .78 
 
 Sept. 7 
 
 1.0693 
 
 17.98 
 
 19.23 
 
 16.55 
 
 15.48 
 
 .84 
 
 .48 
 
 .19 
 
 .65 
 
 .73 
 
 18.41 
 
 .82 
 
 Sept. 15 
 
 1.0708 
 
 18.37 
 
 19.67 
 
 18.17 
 
 16.97 
 
 .56 
 
 .58 
 
 .23 
 
 .33 
 
 .64 
 
 19.72 
 
 .05 
 
 Sept. 22 
 
 1.0912 
 
 23.72 
 
 25.88 
 
 23.02 
 
 21.10 
 
 .53 
 
 .87 
 
 .35 
 
 .19 
 
 .64 
 
 24.72 
 
 1.16 
 
 Sultana 
 
 Aug. 19 
 Aug. 25 
 Sept. 1 
 Sept. 7 
 Sept. 22 1.0902 23.39 25.50 23.10 21.19 1.24 .50 
 
 1.0673 
 
 17.80 
 
 19.00 
 
 15.63 
 
 16.64 
 
 1.69 
 
 .33 
 
 1.0746 
 
 19.37 
 
 20.82 
 
 17.96 
 
 16.71 
 
 1.44 
 
 .37 
 
 1.0815 
 
 21.17 
 
 22.90 
 
 20.26 
 
 18.73 
 
 1.14 
 
 .54 
 
 1.0893 
 
 23 22 
 
 25.29 
 
 23.02 
 
 21.13 
 
 .78 
 
 .44 
 
 Sept. 29 1.0922 23.99 26.20 24.04 
 
 :.oi 
 
 .80 .41 
 
 .13 
 
 1.56 
 
 .32 
 
 17.84 
 
 1.16 
 
 .14 
 
 1..30 
 
 .38 
 
 20.01 
 
 .81 
 
 oo 
 
 .92 
 
 .50 
 
 22.22 
 
 .68 
 
 .18 
 
 .60 
 
 .34 
 
 24.40 
 
 .89 
 
 .20 
 
 1.04 
 
 .38 
 
 25.02 
 
 .48 
 
 .17 
 
 .63 
 
 .42 
 
 25.50 
 
 .70 
 
 [113] 
 
434 
 
 MISCELLANEOUS STUDIES 
 
 Table 8 — (Continued) 
 
 Sultanina 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Variety 
 and date 
 
 1 
 Sp. gr. 
 
 2 
 T. S. 6. 
 
 3 
 T. S. C. 
 
 4 
 S. G. 
 
 5 
 S.I. 
 
 6 7 
 Tl. A. C. T. 
 
 8 
 C. T. T. 
 
 9 
 T. A. 
 
 10 
 
 p. 
 
 11 
 
 S. T 
 
 12 
 .A. S 
 
 Aug. 19 
 
 1.0673 
 
 17.46 
 
 18.64 
 
 15.87 
 
 14.87 
 
 1.27 
 
 .44 
 
 .18 
 
 1.09 
 
 .42 
 
 17.82 
 
 .82 
 
 Aug. 2.5 
 
 1.0743 
 
 19.26 
 
 20.69 
 
 18.30 
 
 17.03 
 
 1.19 
 
 .47 
 
 .19 
 
 1.00 
 
 .37 
 
 20.14 
 
 .55 
 
 Sept. 1 
 
 1.0771 
 
 20.02 
 
 21.56 
 
 18.98 
 
 17.62 
 
 .85 
 
 .49 
 
 .20 
 
 .65 
 
 .42 
 
 20.54 
 
 1.02 
 
 Sept. 7 
 
 1.0892 
 
 23.20 
 
 25.27 
 
 22.42 
 
 20.58 
 
 .72 
 
 .80 
 
 .32 
 
 .40 
 
 .62 
 
 24.24 
 
 1.03 
 
 Sept. 15 
 
 1.0927 
 
 24.12 
 
 26.36 
 
 23.62 
 
 21.62 
 
 .79 
 
 .76 
 
 .30 
 
 .39 
 
 .45 
 
 25.22 
 
 1.14 
 
 Sept. 22 
 
 1.0984 
 
 25.62 
 
 28.14 
 
 25.71 
 
 23.41 
 
 .60 
 
 .58 
 
 .23 
 
 .37 
 
 .45 
 
 27.11 
 
 1.03 
 
 Sept. 29 
 
 1.1049 
 
 27.33 
 
 30.20 
 
 27.41 
 
 24.81 
 
 .54 
 
 .51 
 
 .20 
 
 .34 
 
 .42 
 
 28.68 
 
 1.52 
 
 ToKay 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Aug. 19 
 
 1.0598 
 
 15.50 
 
 16.43 
 
 14.41 
 
 13.60 
 
 1.74 
 
 .41 
 
 .16 
 
 1.58 
 
 .29 
 
 16.69 
 
 .26 
 
 Aug. 25 
 
 1.0676 
 
 17.54 
 
 18.73 
 
 15.63 
 
 14.64 
 
 1.24 
 
 .39 
 
 .15 
 
 1.09 
 
 .69 
 
 17.80 
 
 .93 
 
 Sept. 1 
 
 1.0757 
 
 19.65 
 
 21.14 
 
 18.17 
 
 16.89 
 
 .84 
 
 .47 
 
 .19 
 
 .66 
 
 .44 
 
 19.74 
 
 1.40 
 
 Sept. 7 
 
 1.0781 
 
 20.28 
 
 21.86 
 
 19.11 
 
 17.73 
 
 .79 
 
 .45 
 
 .18 
 
 .61 
 
 .37 
 
 20.54 
 
 1.32 
 
 Sept. 15 
 
 1.0785 
 
 20.39 
 
 21.99 
 
 19.26 
 
 17.86 
 
 .74 
 
 .48 
 
 .19 
 
 .55 
 
 .40 
 
 20.69 
 
 1.30 
 
 Sept. 22 
 
 1.0798 
 
 20.73 
 
 22.,38 
 
 20.17 
 
 18.68 
 
 .59 
 
 .51 
 
 .20 
 
 .39 
 
 .36 
 
 21.43 
 
 .95 
 
 Sept. 29 
 
 1.0823 
 
 21.38 
 
 23.14 
 
 20.76 
 
 19.18 
 
 .85 
 
 .58 
 
 .23 
 
 .62 
 
 .28 
 
 22.24 
 
 .90 
 
 Oct. 7 
 
 1.0830 
 
 21.57 
 
 23.36 
 
 20.87 
 
 19.27 
 
 .69 
 
 .63 
 
 .25 
 
 .44 
 
 .42 
 
 22.36 
 
 1.00 
 
 Oct. 14 
 
 1.0851 
 
 22.12 
 
 24.00 
 
 21.53 
 
 19.84 
 
 .65 
 
 .69 
 
 .28 
 
 .38 
 
 .36 
 
 22.96 
 
 1.04 
 
 Oct. 22 
 
 1.0895 
 
 23.28 
 
 25.36 
 
 22.91 
 
 21.03 
 
 .66 
 
 .72 
 
 .29 
 
 .37 
 
 .37 
 
 24.37 
 
 .99 
 
 Burger 
 
 Table 9 — Grape Ripening Tests, 1916 
 
 Variety 
 and date 
 
 1 
 Sp. gr. 
 
 2 
 T. S. G. 
 
 3 
 T. S. C. 
 
 4 
 S. G. 
 
 5 
 S.I. 
 
 6 
 TI. A. 
 
 7 8 
 C. T. C. T. T 
 
 9 
 T. A. 
 
 10 
 P. 
 
 11 
 
 S. 
 
 12 
 T. S. S 
 
 June 12 
 
 1.0212 
 
 5.48 
 
 5.59 
 
 1.29 
 
 1.55 
 
 2.95 
 
 .55 
 
 .22 
 
 2.73 
 
 .44 
 
 5.27 
 
 .32 
 
 June 19 
 
 1.0195 
 
 5.04 
 
 5.88 
 
 .87 
 
 .88 
 
 2.88 
 
 .51 
 
 .21 
 
 2.67 
 
 .45 
 
 4.51 
 
 1.37 
 
 June 27 
 
 1.0220 
 
 5.69 
 
 5.82 
 
 1.25 
 
 1.28 
 
 2.94 
 
 .33 
 
 .13 
 
 2.81 
 
 .45 
 
 4.87 
 
 .95 
 
 July 7 
 
 1.0220 
 
 5.69 
 
 5.82 
 
 1.11 
 
 1.28 
 
 2.98 
 
 .49 
 
 .20 
 
 2.78 
 
 .31 
 
 4.86 
 
 .96 
 
 July 10 
 
 1.0200 
 
 5.17 
 
 5.27 
 
 .93 
 
 .95 
 
 3.32 
 
 .57 
 
 .23 
 
 3.09 
 
 .37 
 
 4.97 
 
 .30 
 
 July 19 
 
 1.0205 
 
 5.30 
 
 5.41 
 
 1.03 
 
 1.05 
 
 3.13 
 
 .55 
 
 .22 
 
 2.91 
 
 .35 
 
 4.86 
 
 .55 
 
 July 27 
 
 1.0225 
 
 5.82 
 
 5.95 
 
 1.13 
 
 1.15 
 
 2.93 
 
 .48 
 
 .19 
 
 2.74 
 
 .34 
 
 4.71 
 
 1.24 
 
 Aug. 3 
 
 1.0258 
 
 6.67 
 
 6.84 
 
 2.14 
 
 2.19 
 
 2.71 
 
 .63 
 
 .25 
 
 2.46 
 
 .40 
 
 5.68 
 
 1.16 
 
 Aug. 7 
 
 1.0330 
 
 8.53 
 
 8.83 
 
 3.36 
 
 3.46 
 
 2.67 
 
 .87 
 
 .35 
 
 2.32 
 
 .47 
 
 7.12 
 
 1.21 
 
 Aug. 16 
 
 1.0391 
 
 10.11 
 
 10.51 
 
 5.90 
 
 6.13 
 
 2.41 
 
 .95 
 
 .38 
 
 2.03 
 
 .46 
 
 9.57 
 
 .94 
 
 Aug. 23 
 
 1.0422 
 
 10.92 
 
 11.38 
 
 6.03 
 
 6.27 
 
 2.10 
 
 .98 
 
 .39 
 
 1.71 
 
 .63 
 
 9.59 
 
 1.89 
 
 Aug. 30 
 
 1.0529 
 
 13.70 
 
 14.42 
 
 9.95 
 
 10.42 
 
 1.15 
 
 1.03 
 
 .41 
 
 .74 
 
 .49 
 
 12.70 
 
 1.72 
 
 Sept. 5 
 
 1.0645 
 
 16.73 
 
 17.81 
 
 14.51 
 
 1.5.43 
 
 1.01 
 
 1.07 
 
 .43 
 
 .68 
 
 .61 
 
 17.79 
 
 .02 
 
 Sept. 12 
 
 1.0717 
 
 18.61 
 
 19.94 
 
 16.27 
 
 17.36 
 
 .95 
 
 .98 
 
 .39 
 
 .56 
 
 .82 
 
 19.72 
 
 .22 
 
 Sept. 20 
 
 1.0765 
 
 19.86 
 
 21.37 
 
 17.44 
 
 18.73 
 
 .87 
 
 1.06 
 
 .42 
 
 .45 
 
 .62 
 
 20.86 
 
 .51 
 
 Sept. 26 
 
 1.0808 
 
 20.99 
 
 22.68 
 
 18.48 
 
 19.99 
 
 .81 
 
 1.01 
 
 .40 
 
 .41 
 
 .83 
 
 22.24 
 
 .44 
 
 Cornichon 
 
 
 
 
 
 
 
 
 
 
 
 
 
 June 12 
 
 1.0202 
 
 5.22 
 
 5.32 
 
 .91 
 
 .93 
 
 3.15 
 
 .64 
 
 .26 
 
 2.89 
 
 .32 
 
 4.78 
 
 .54 
 
 June 19 
 
 1.0200 
 
 5.17 
 
 5.27 
 
 .86 
 
 .88 
 
 2.96 
 
 .62 
 
 .25 
 
 2.71 
 
 .42 
 
 4.63 
 
 .64 
 
 June 27 
 
 1.0193 
 
 4.99 
 
 5. OS 
 
 .84 
 
 .86 
 
 2.89 
 
 .39 
 
 .16 
 
 2.73 
 
 .56 
 
 4.54 
 
 .44 
 
 July 7 
 
 1.0201 
 
 5.19 
 
 5.29 
 
 .87 
 
 .89 
 
 2.88 
 
 .44 
 
 .18 
 
 2.70 
 
 .52 
 
 4..55 
 
 .74 
 
 July 10 
 
 1.0206 
 
 5.32 
 
 5.43 
 
 .85 
 
 .87 
 
 3.27 
 
 .54 
 
 22 
 
 3.05 
 
 .53 
 
 4.99 
 
 .44 
 
 July 19 
 
 1.0225 
 
 5.82 
 
 5.95 
 
 1.28 
 
 1.30 
 
 3.11 
 
 .57 
 
 .23 
 
 2.88 
 
 ..55 
 
 5.30 
 
 .65 
 
 July 27 
 
 1.0242 
 
 6.25 
 
 6.40 
 
 1.63 
 
 1.66 
 
 2.94 
 
 .54 
 
 oo 
 
 2 72 
 
 .44 
 
 5.26 
 
 .14 
 
 Aug. 3 
 
 1.0373 
 
 9.65 
 
 10.00 
 
 5.00 
 
 5.19 
 
 2.87 
 
 .59 
 
 .24 
 
 2.63 
 
 .56 
 
 8.97 
 
 1.03 
 
 [1141 
 
CHEMIC.il COMl'OSITION OF CUAl'KS 
 
 43-) 
 
 Table 9 — {Continued) 
 
 Variety 
 and date 
 
 1 
 
 Sp. gr. 
 
 2 
 T. s'. G. 
 
 3 
 T. S. C. 
 
 4 
 
 S. fi. 
 
 S.''l. 
 
 (i 
 Tl. A. 
 
 C. T. C 
 
 8 
 T.T 
 
 I) 
 T. A. 
 
 10 
 
 V. 
 
 11 12 
 S. T. A. S 
 
 Aug. 7 
 
 l.OS?.^ 
 
 9.70 
 
 10.06 
 
 5.28 
 
 5.48 
 
 2.79 
 
 .65 
 
 .26 
 
 2.53 
 
 .66 
 
 9.32 
 
 .64 
 
 Aug. 16 
 
 1.0434 
 
 11.23 
 
 11.71 
 
 6.30 
 
 6.57 
 
 2.75 
 
 1.06 
 
 .43 
 
 2.32 
 
 .53 
 
 10.48 
 
 1.23 
 
 Aug. 23 
 
 1.0635 
 
 16.47 
 
 17.51 
 
 12.19 
 
 12.96 
 
 1.85 
 
 1.06 
 
 .43 
 
 1.42 
 
 .58 
 
 16.02 
 
 1.49 
 
 Aug. 30 
 
 1.06S5 
 
 17.77 
 
 18.97 
 
 14.75 
 
 15.61 
 
 1.16 
 
 1.10 
 
 .44 
 
 .72 
 
 .63 
 
 18.06 
 
 .91 
 
 Sept. 5 
 
 1.0694 
 
 18.01 
 
 19.25 
 
 15.03 
 
 16.07 
 
 .93 
 
 .90 
 
 .36 
 
 .57 
 
 .58 
 
 18.12 
 
 1.13 
 
 Sept. 12 
 
 1.0757 
 
 19.65 
 
 21.09 
 
 16.37 
 
 17.60 
 
 .87 
 
 1.14 
 
 .46 
 
 .41 
 
 .78 
 
 19.97 
 
 1.12 
 
 Sept. 20 1.0786 20.41 22.00 17.52 18.88 
 Sept. 26 1.082S 21..52 23.30 18..52 20.03 
 
 .84 .94 .37 .44 .59 20.85 1.15 
 .72 .83 .,33 .39 .85 22.10 1.20 
 
 Muscat 
 
 June 12 
 
 1.0203 
 
 5.25 
 
 5.35 
 
 .91 
 
 .93 
 
 2.93 
 
 .65 
 
 .26 
 
 2.71 
 
 .38 
 
 4.67 
 
 .68 
 
 June 19 
 
 1.0199 
 
 5.14 
 
 5.24 
 
 .70 
 
 .72 
 
 3.37 
 
 .63 
 
 25 
 
 3.12 
 
 .44 
 
 4.91 
 
 .33 
 
 June 27 
 
 1.0210 
 
 5.43 
 
 5.54 
 
 1.33 
 
 1.36 
 
 3.33 
 
 .48 
 
 19 
 
 3.14 
 
 .49 
 
 5.47 
 
 .07 
 
 July 7 
 
 1.0210 
 
 5.43 
 
 5.54 
 
 1.63 
 
 1.66 
 
 3.32 
 
 .54 
 
 22 
 
 3.10 
 
 .45 
 
 5.75 
 
 .21 
 
 July 10 
 
 1.0195 
 
 5.04 
 
 5.14 
 
 1.33 
 
 1.36 
 
 3.60 
 
 .55 
 
 22 
 
 3.38 
 
 .36 
 
 5.65 
 
 .51 
 
 July 19 
 
 1.0251 
 
 6.49 
 
 6.65 
 
 2.55 
 
 2.61 
 
 3.40 
 
 .58 
 
 23 
 
 3.17 
 
 .49 
 
 6.85 
 
 .20 
 
 July 27 
 
 1.0308 
 
 7.97 
 
 8.22 
 
 3.56 
 
 3.67 
 
 2.67 
 
 .66 
 
 26 
 
 2.01 
 
 .45 
 
 6.79 
 
 1.43 
 
 Aug. 3 
 
 1.0488 
 
 12.64 
 
 13.26 
 
 9.72 
 
 10.19 
 
 1.77 
 
 .68 
 
 27 
 
 1.50 
 
 .46 
 
 12.83 
 
 .43 
 
 Aug. 7 
 
 1.0582 
 
 15.68 
 
 16.58 
 
 12.72 
 
 13.53 
 
 1.60 
 
 .73 
 
 29 
 
 1.31 
 
 .55 
 
 16.12 
 
 .46 
 
 Aug. 16 
 
 1.0803 
 
 20.86 
 
 22.53 
 
 16.81 
 
 18.15 
 
 1.16 
 
 .94 
 
 38 
 
 .78 
 
 .51 
 
 20.38 
 
 1.70 
 
 Aug. 23 
 
 1.0910 
 
 23.67 
 
 25.82 
 
 20.20 
 
 22.04 
 
 .82 
 
 1.04 
 
 42 
 
 .40 
 
 .56 
 
 24.04 
 
 1.78 
 
 Aug. 30 
 
 1.0972 
 
 25.30 
 
 27.75 
 
 21.87 
 
 22.99 
 
 .65 
 
 1.21 
 
 49 
 
 .16 
 
 .58 
 
 25.94 
 
 1.81 
 
 Sept. 5 
 
 1.1023 
 
 26.64 
 
 29.36 
 
 23.28 
 
 24.74 
 
 .60 
 
 1.17 
 
 47 
 
 .13 
 
 .65 
 
 26.69 
 
 2.67 
 
 Sept. 12 
 
 1.1101 
 
 28.70 
 
 31.85 
 
 25.95 
 
 27.83 
 
 .56 
 
 1.35 
 
 54 
 
 .02 
 
 .69 
 
 29.89 
 
 1.96 
 
 Sept. 20 
 
 1.1122 
 
 29.25 
 
 32.72 
 
 26.43 
 
 29.39 
 
 .68 
 
 1.56 
 
 63 
 
 .05 
 
 .58 
 
 31.27 
 
 1.45 
 
 Sept. 26 
 
 1.1133 
 
 29.54 
 
 32.89 
 
 26.68 
 
 29.70 
 
 .56 
 
 1.39 
 
 56 
 
 .00 
 
 .59 
 
 31.57 
 
 1.32 
 
 
 Table 10 — Catawba 
 
 Grape Ripening 
 
 Tests 
 
 
 
 
 (Table from U. S. Dept. Agric 
 
 Bulletin 335, by 
 
 "W. B. 
 
 Alwood) 
 
 
 Cataxcha 
 
 
 
 
 
 
 
 
 
 
 1912: 
 
 
 
 
 
 
 
 
 
 
 Variety 
 and date 
 
 1 
 
 Sp. er. 
 
 T. s". G. 
 
 3 
 T. S. C. 
 
 4 
 
 S. I. 
 
 5 
 S. G. 
 
 6 
 Tl. A. 
 
 7 
 0. T. 
 
 8 9 
 r. T. T. Days 
 
 Sept. 4 
 
 1.0329 
 
 8.51 
 
 8.84 
 
 3.60 
 
 3.72 
 
 3.68 
 
 .39 
 
 .16 
 
 
 
 Sept. 9 
 
 1.0419 
 
 10.84 
 
 11.29 
 
 6.68 
 
 6.96 
 
 3.02 
 
 .41 
 
 .16 
 
 5 
 
 Sept. 12 
 
 1.0515 
 
 13.34 
 
 14.03 
 
 9.35 
 
 9.78 
 
 2.48 
 
 .46 
 
 .18 
 
 8 
 
 Sept. 17 
 
 1.0537 
 
 13.91 
 
 14.66 
 
 10.38 
 
 10.95 
 
 2.12 
 
 .45 
 
 .18 
 
 13 
 
 Sept. 24 
 
 1.0569 
 
 14.74 
 
 15.58 
 
 11.33 
 
 11.96 
 
 1.74 
 
 .53 
 
 .21 
 
 20 
 
 Oct. 1 
 
 1.0614 
 
 15.92 
 
 16.89 
 
 12.75 
 
 13.48 
 
 1.63 
 
 .54 
 
 .22 
 
 27 
 
 Oct. 7 
 
 1.0663 
 
 17.20 
 
 18.34 
 
 13.79 
 
 14.71 
 
 1.53 
 
 .61 
 
 .24 
 
 33 
 
 Oct. 16 
 
 1.0725 
 
 18.82 
 
 20.18 
 
 15.35 
 
 16.46 
 
 1.34 
 
 .61 
 
 .24 
 
 42 
 
 Oct. 23 
 
 1.0716 
 
 18..58 
 
 19.90 
 
 15.01 
 
 16.09 
 
 1.28 
 
 ..59 
 
 .24 
 
 47 
 
 Oct. 29 
 
 1.0769 
 
 19.97 
 
 21.50 
 
 16.49 
 
 17.75 
 
 1 22 
 
 .57 
 
 .23 
 
 53 
 
 Nov. 4 
 
 1.0790 
 
 20..'52 
 
 22.14 
 
 16.77 
 
 18.08 
 
 1.28 
 
 .71 
 
 .28 
 
 59 
 
 Nov. 8 
 
 1.0755 
 
 19.60 
 
 21.07 
 
 16.39 
 
 17.61 
 
 1.09 
 
 .52 
 
 .21 
 
 63 
 
 1115] 
 
436 MISCELLANEOUS STUDIES . 
 
 Curves of Total Solids, Sugar, Total Acid, Free Acid, and Cream 
 of Tartar. — In order to present the data in a form in which they may 
 be readily studied, graphs have been constructed using time in days 
 as abscissae and the above constituents expressed in grams per 100 c.c. 
 as ordinates. The curves represent the data for 1914, 1915, and 1916. 
 For comparison, curves of the changes in composition of Catawba 
 grapes reported by W. B. Alwood in the United States Department 
 of Agriculture Bulletin 335 have been included. The acid principles 
 have been plotted to a scale five times as great as that used for total 
 solids and sugar in order that the variations in acidity might be more 
 apparent. 
 
 Discussion of Graphs of Total Solids, Sugar, Total Acid, Cream of 
 Tartar, and Free Acid. — (1) Total Solids and Sugar. The data are 
 more complete for 1916 than for 1914 or 1915, and include the period 
 during which the berries are growing to full size as well as the ripen- 
 ing period itself, during which the rapid increase in sugar occurs. 
 The curves for 1916, therefore, are of more interest than those for 
 1914 and 1915. In the case of the Burger variety, total solids and 
 sugar remained constant for approximately forty days after the tests 
 were started. There was then a slight rise in these components for 
 a period of about ten days. From that point on the rise in total solids 
 and sugar was very rapid and fairly uniform. The behavior of the 
 Cornichon was very similar. 
 
 The Muscat began ripening about ten days earlier than the Burger 
 and Cornichon, and proceeded much more rapidly up to aboiit the 
 ninetieth day after the experiment was started. There was then a 
 slowing up in the increase in total solids and sugar corresponding to 
 the period of over-ripeness. This slower increase in total solids is 
 also evident in the curves for Emperor, Muscat, Sultana, and Tokay 
 for the 1915 season, and would undoubtedly show in all eases if the 
 observations were continued sufficiently. 
 
 The effect of the season upon the rate of ripening is shown by a 
 comparison of the Cornichon and Muscat varieties for 1915 and 1916. 
 All varieties ripened more slowly in 1915 than in 1916, resulting in 
 steeper curves for 1916. However, owing to the fact that sampling 
 was started later in 1914 and 1915 than in 1916, the curves for the 
 former two years show only the changes taking place during the latter 
 half of the ripening period. No very close comparisons therefore can 
 be made of the three years. 
 
 The Catawba reported by Alwood, and for which curves appear 
 
 riifii 
 
CHEMICAL COMPOSITION OF GSAPES 
 
 437 
 
 
 n. 
 
 
 
 
 Mfll 1 
 
 lG/1 
 
 it^Ch 
 
 OF' 1 
 
 Q/4^ 
 
 
 
 
 a/ 
 
 R= 
 
 ffyfThrit 
 
 rfrrrf 
 
 Trr. /at 
 
 mr/c. n. 
 
 ■uLIm'-& 
 
 7/w nf 
 
 Jrir^r 
 
 
 /^ /ninr.r 
 
 
 X4- 
 
 n 
 
 T,ih/ 
 
 , V'^'i ' 
 
 7r^/i d/t 
 
 ynr C^/ 
 
 ^.•% f^.r 
 
 /nCr.e 
 
 
 / 
 
 
 
 
 
 5 
 
 
 
 
 
 
 
 
 y , 
 
 
 
 
 X4 
 
 
 
 
 
 
 
 
 
 ^ y 
 
 
 
 
 a 
 
 
 
 
 
 
 
 
 y/ 
 
 / 
 
 
 
 
 a. 
 
 ^ 
 
 
 
 
 
 
 ^ ^ 
 
 "" / 
 
 
 
 
 
 1^ 
 
 
 
 
 
 
 ^ 
 
 
 / 
 
 
 
 
 
 lb 
 
 
 
 
 j,i5 
 
 ^A*^ 
 
 " 
 
 ^^ 
 
 
 
 
 
 
 
 T 
 
 
 
 <M^ 
 
 
 _^^0^ 
 
 
 
 
 
 
 
 M 
 
 
 
 A 
 
 ^jj^ 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 ^>^ 
 
 
 =i.rf 
 
 ii^ 
 
 
 
 
 
 
 
 
 li. 
 
 
 ^^^ 
 
 lO 
 
 ^ ■ 
 
 SX- 
 
 "^^^>< 
 
 ^ 
 
 
 
 
 
 
 
 
 
 fi 
 
 
 
 -^ "" 
 
 <^ 
 
 
 
 
 
 
 
 
 
 ( 
 
 " 
 
 
 
 
 
 %> 
 
 
 
 
 
 
 
 
 1 
 
 
 
 
 
 ^^> 
 
 
 
 
 
 
 
 4 
 
 
 
 
 
 
 ^"^ 
 
 C^^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 "^ 
 
 
 
 i 
 
 
 
 K 
 
 
 
 
 
 
 
 
 ^r 
 
 
 
 
 
 
 
 
 
 ^ 
 
 3 — -A 
 
 ' 1 
 
 7 A 
 
 7 X~ 
 
 !■ J 
 
 5 i 
 
 ^ ^ 
 
 ^ -sJ 
 
 ? 321 3b^ 
 
 r/ME/ND/f^S 
 
 -m — 73 :^ 
 
 T/M£ IN D/F/S 
 
 Fig. 1 — Malaga first and second crops, 1914. 
 
438 
 
 MISCELLANEOUS STVDIES 
 
 ~3 7v 73 To :e3' 
 77A<£" /A' C/f-yS 
 
 Fig. 2 — Tokay first and second crops, 1914. 
 
 [1181 
 
CIIKMHAI. fOMrOSiriOX OF (,I;.I1'KS 
 
 439 
 
 TH fS 
 
 Fig. 3 — Cornii'liou and Emperor, 1915. 
 
 "7^ 73 ^? ■5V~ 3^ 
 
 T/M£ IN DFt-yS 
 
 [1191 
 
440 
 
 MISCELLANEOUS STUDIES 
 
 Fig. 4^Malaga and Muscat, 1915. 
 
 [1201 
 
CHEMICAL COiirOlsniUN OF GllAl'ES 
 
 441 
 
 Fig. 5 — Pedro Zuinbon and Sultana, 1915. 
 
 [1211 
 
442 
 
 illSCELLANEOrS STUDIES 
 
 -m — ^ 
 
 T/ME IN Dft-ys 
 
 
 it. 
 
 TaiByLMs. 
 
 /J= To'a/ flcld fPer lortcr/c oAd Gea/ v of l^rfar ( S>7J /fer /0O)i 
 
 iz n -^ — 35" 
 
 TIME. IN D/f-ys 
 
 Fig. 6 — Sultanina and Tokay, 1915. 
 
 ri221 
 
CHKUICAL COMl'OtilTlOS OF Gli.ll'ES 
 
 448 
 
 -2& 3» — ^» 33 — ar 
 
 T//^r /A/ £5/9X5 
 
 Fig. 7 — Burger and Corniclion, 1916. 
 
 TOO — W 
 
 [123] 
 
444 
 
 MISCELLANEOUS STUDIES 
 
 -^ 72' 35 '^ 3er 
 
 time: IN onys 
 
 Fig. 8— Muscat, 1916. 
 
 BW 
 
 
 C/fT/JWS/J J9JZ. U.5JDjBJDcpi:-^^/-^3SMnanoGa 
 
 fl= ^rfonc /Jcf /^ ^e r /Jr,ty nm:/ CrfO'^ f^^ Inr tat 
 G-. Tnfhl ^nl/d', nnri 3 iiqnr Cym.', f^r ino G/ m 
 
 Fig. 9— Catawba (U. S. Dept. Agric. Bull. 335). 
 
 [1241 
 
CHEMICAL COMPOSITION OF GRAPES 445 
 
 ill figure 9, ripened more slowly than the Vinifera varieties. For 
 example, during a period of fifty days, the total solids increased only 
 4 per cent. It can not be said from the data at hand whether this 
 slow ripening is due to the conditions under which the grapes were 
 grown or to the variety. 
 
 By reference to figures 1 and 2 it may be seen that the general 
 form of the ripening curves is the same for the first and for second 
 crop. In one case, the Malaga, the curves are almost identical for 
 the period common to both, i.e., from 10.6 Bal. to 26.3 Bal., showing 
 an equal rate of ripening. In the other, the Tokay, the curve of the 
 second crop, from 18.2 Bal. to 24.6 Bal., is much flatter than that of 
 the first, indicating a rate of ripening with the latter of about two 
 and a half times that of the former. This diiference can be accounted 
 for by the cooler weather during the time the second crop Tokay was 
 ripening, which was about ten days later than in the case of the second 
 crop Malaga. The slower ripening is probably due both to the direct 
 effect of the cool weather and to the decreased activity of the leaves 
 at lower temperatures. 
 
 (2) ("hanges in Total Acid, Cream of Tartar, and Free Acid. 
 Owing to the fact that the analyses were started in 1914 and 1915 
 after ripening had commenced, the curves for these years show a 
 decrease in acid throughout the period of the tests. In 1916, however, 
 a rise in total acid occurred during the growing stage, as shown by 
 a rise in the curve during the first thirty days of the experiment. 
 Although this rise is not very large, it is quite definite, and occurs 
 in all three varieties tested. The rise was most positive in the case 
 of the Muscat grape, and amounted to .67 per cent acid as tartaric. 
 From the point of maximum acidity, the total decreases slowly until the 
 period of rapid ripening sets in. The total acid then decreases very 
 rapidly for a time and more or less in proportion to the increase in 
 total solids and sugar. As the grapes near maturity, the rate of de- 
 crease of total acid becomes less and the total remains practically , 
 constant after the grapes have reached maturity. 
 
 The cream of tartar in general increases very slightly during the 
 periods of growth and ripening. 
 
 The increase in total acid during the first stages of growth is due 
 to increase in the free acid. Since the cream of tartar remains almost 
 constant throughout the ripening period, the curve of the free acid 
 is practically parallel with that of the total acid. 
 
 As the grapes approach maturity, the cream of tartar calculated as 
 
 [125] 
 
446 
 
 MISCELLANEOVS STUDIES 
 
 tartaric acid approaches the total acid, and in one case, (Musct, 1916), 
 actually became equal to the total acid, indicating that in this instance 
 no free acid remained. 
 
 Second crop grapes were found to be higher in free acid than 
 first crop grapes of the same total solids and sugar content. The 
 Catawba grape grown under eastern conditions (fig. 9) exhibits rela- 
 tively high free acid. Alwood** has found this free acidity in eastern 
 grapes to be due largely to malic acid. No attempt was made in the 
 analyses of the California samples to identify the various acids making 
 up the free acidity which was calculated as tartaric acid. 
 
 Mean Differences Between Total Solids and Sugar. — The following 
 table contains figures representing the differences between total solids 
 and sugar at the various percentages of total solids indicated at the 
 tops of the columns. The data represent a range of total solids from 
 5 per cent to 30 per cent. The figures were taken from the data 
 reported in tables 7 to 9, and represent several varieties of grapes. 
 Only a few determinations of total solids and sugar were available 
 for the lower concentrations (5 per cent to 1.5 per cent), and therefore 
 the figures for this range may not represent averages so accurately 
 as the figiires above 15 per cent total solids. 
 
 Between 5 per cent and 11 per cent solids, the average difference 
 between total solids and sugar remains practically constant. From 
 11 per cent to 17 per cent total solids, the mean difference decreases 
 quite rapidly. Prom 17 per cent to 30 per cent, the difference remains 
 fairly constant. The variations noted after 17 ])er cent total solids 
 
 T 
 
 
 
 
 
 Mimn 1 
 
 'llU.StSti 
 
 /7*f.T 
 
 "irtn 
 
 rrr 
 
 ^ofnl 
 
 2a/M: 
 
 
 
 
 
 > 
 
 
 
 aacL2i 
 
 
 
 ,■7 m 
 
 i 7)0 ;, 
 
 ' 7^A 
 
 / <lJ 
 
 d^ 
 
 
 4 
 
 ^~~^ 
 
 '■^ 
 
 '\. 
 
 
 
 
 
 
 
 
 
 
 
 i 
 
 
 
 \ 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 h 
 
 
 
 s 
 
 
 
 
 
 
 
 
 
 
 
 S 
 
 
 
 
 
 ^•^^^^ ' 
 
 ■^^-( 
 
 
 ^ 
 
 ^ 
 
 ^ y 
 
 .^ 
 
 <e 
 
 I 
 
 
 
 
 
 
 
 i 
 
 "H 
 
 ■^ 
 
 
 
 
 \i 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 5 
 
 ' 5 
 
 U^ 
 
 i%_^ 
 
 •■9"'=^. 
 
 
 ^ /; 
 
 / X) 
 
 X 
 
 J ^~ 
 
 X 
 
 7 Xi 
 
 \ 37 
 
 Fig. 10 — Mean dififerences between total solids and sugar between 5 per cent 
 and 30 per cent total solids. 
 
 6 XJ. S. Dept. Agric. Bull. 335. 
 
CIIKMICM. CUMrUNiriOX OF OUAI'KH 
 
 447 
 
 >(>' 
 
 1 
 
 
 5 
 
 
 ■g 
 
 w 
 
 1^ 
 
 
 
 
 M 
 
 
 Co 
 
 
 CO 
 
 
 
 
 *^ 
 
 
 
 
 Ol 
 
 
 h-» 
 
 
 *. 
 
 *■ 
 
 rf^ 
 
 h^ 
 
 K) 
 
 to 
 
 CO 
 
 1^ 
 
 >f" 
 
 to 
 
 Ol 
 
 
 
 h;^ 
 
 ^ 
 
 CO 
 
 ;;! 
 
 
 
 Oi 
 
 Oi 
 
 
 -1 
 
 
 
 rf* 
 
 ht- 
 
 M 
 
 % 
 
 
 ::::::: i i *^ 
 
 to 
 
 ^ 
 
 
 a- 
 
 
 rf^ 
 
 rf^ 
 
 )^ 
 
 ^ 
 
 Bi 
 
 > 
 
 :;:::;;:: «D 
 
 no 
 
 I—* 
 
 ■^ 
 
 
 
 
 
 
 
 tr" 
 
 
 rf^ 
 
 CO 
 
 ^ 
 
 o 
 
 M 
 
 
 o 
 
 GO 
 
 lO 
 
 o* 
 
 1— ' 
 
 to 
 
 CO 
 
 CO 
 
 1-1 
 
 ^ 
 
 
 
 
 tn 
 
 
 i^ 
 
 
 
 
 
 
 
 
 
 
 
 
 > 
 
 CO 
 
 CO 
 
 CO 
 
 l_l 
 
 
 5^ 
 
 ; i : i i : i : : oo 
 
 Ol 
 
 to 
 
 •;^ 
 
 p_ 
 
 w 
 
 CO 
 
 CO 
 
 to 
 
 to 
 
 CO 
 
 en 
 
 a 
 
 & 
 
 
 to CO CO 
 
 to 
 
 
 0> 
 
 p 
 
 a 
 
 5^ 
 
 • ; : : i : : b if'- w 
 
 CO 
 
 rf^ 
 
 o 
 
 to CO to t>0 to CO CO fO 
 to O -<1 CO -^ CO O CO 
 
 w 
 
 to to to to CO CO JO 
 CD •<) bo -^l CO CO to 
 
 cotococototocotocototo >-• 
 
 O CO to CD bi --3 I-' CO CO CO bi ''^ 
 
 CO to CO to to to CO to CO to W to 
 
 CO **- o bo hp- rf*- H* ^ to ^ en ° 
 
 tf' to to to to CO to to to to CO J-- to 
 
 O "no OT '^ ^ CO If* <D ^ CO H-* '-~5 *"* 
 
 COtOCOtOtOtOtOtOCOtO 10 
 
 : o^obobiQo'ojuibio 
 
 to 
 
 to CO CO to to CO 
 ^ o OS bi *rf^ o 
 
 to to 
 
 cototoi-*totocotoco M 
 bicn--icobobicnb>to 
 
 CO to to to CO to to to " 
 
 i rfifc rfk. oi I*^ '^ en bi bi 
 
 to to to to to to 
 
 • : ; i CO en -^ bo no ^ 
 
 CO to to to to tJ 
 
 I ■ rfi. M en CI no ~^ 
 
 , to to to ^ 
 
 I i : i I : '** b> no 
 
 to CO to ta 
 
 i : : : ■ . : co bo bo * 
 
 O 
 
 CO 
 no 
 
448 
 
 MISCELLANEOUS STUDIES 
 
 was reached are probably within the experimental error. The large 
 difference between the total solids and sugar noted during the first 
 stages of ripening is no doubt due to the high acid content of the 
 unripe grapes. The fact that the difference remains fairly constant 
 after the grapes have become mature is to be expected, because the 
 cream of tartar, total acid, and protein remain fairly constant as 
 maturity is approached and during the periods of maturity and over- 
 ripeness. 
 
 -?» 7t> 33 7» yn to 7n Ttr 
 
 Fig. 11 — Variation in uon-coagulable protein content for three varieties, 1916. 
 
 Protein. — The total nitrogen content of the various samples was 
 multiplied by 6.25 to convert it into its protein equivalent. Owing 
 to the fact that the samples were sterilized by heat and filtered before 
 analysis, the figures represent only the protein not coagulated by heat. 
 
 The curves. show that there is a slow increase in protein content 
 during growth and ripening and the greatest increase occurs during 
 the period of most rapid increase of sugar and most rapid decrease 
 of acid. The increase amounted to about .2 per cent in the case of the 
 Muscat and .6 per cent in the ease of the Cornichon. The increase 
 seems to be quite definite, although the protein curves are not so 
 i-egular as those of total solids, sugar, and total acid. 
 
 [1281 
 
CHEMICAL COMPOSITION Of GUAl'KS 449 
 
 Summary op Changes in Must of Grapes During Growth and 
 Ripening op Berries 
 
 1. Total Solids. — The total solids remain fairly constant during 
 the period of growth, corresponding to the period between setting of 
 the berries and the time at which the berries have reached almost 
 full size but are still hard and green. From this point on, there is a 
 rapid increase in total solids due to increase in sugar. 
 
 After the period usually considered as full maturity is reached, 
 the increase in total solids is slow. The question may be raised as to 
 whether this last increase is due to an actual synthesis and secretion 
 of sugar or other solids, or simply to evaporation of water. The fact 
 that there is no change in the curve of the acid decrease at this time 
 indicates that the same processes are continuing and that the increased 
 Balling degree represents an actual increase of solids. This view is 
 fortified by observations regarding the increase of weight of solids 
 during the ripening of raisin grapes. It has been shown that the 
 weight of dried grapes shows a continuous increase up to the highest 
 degree observed, 28.75 Balling.^ 
 
 ' 2. Sugar.- — The total sugar during the growth period comprises 
 only a small amount of the total solids. During ripening, the sugar 
 rapidly increases and then constitutes a much greater proportion. 
 During ripening, the sugar curve follows the total solids curve closely. 
 It is more or less the mirror image of the total acid curve multiplied 
 by five, i.e., increases as the acid decreases. 
 
 3. Total Acid and Free Acid. — During the early stages of the 
 growth of the berries, the acidity increases owing to an increase of 
 free acid. This is a fact that the authors have not found mentioned 
 in the literature. During ripening, the total and free acid rapidly 
 decrease. After maturity is reached, the decrease is very slow. 
 
 4. Cream of Tartar. — There is a very slow, but usually fairly defi- 
 nite, increase in cream of tartar during ripening. This increase is 
 very much less than the decrease in free acid, and therefore can not 
 account for any great part of this decrease. 
 
 ' Bioletti, Frederic T., Relation of the maturity of the grapes to the quantity and 
 quality of the raisins. Proc. Inter. Cong, of Viticulture, San Francisco, 1915, 
 pp. 307-314. 
 
 [129] 
 
450 MISCELLANEOUS STUDIES 
 
 5. Protein. — The protein not coagulated by heat increased defi- 
 nitely during growth and ripening, although the increase was not so 
 regular nor so marked as the increase in sugar or the decrease in total 
 acid. 
 
 6. Difference Between Total Solids and Sugar. — This factor re- 
 mained constant for the lower percentages of total solids, decreased 
 during the rapid ripening stage, and remained constant through 
 maturity and over-ripeness. 
 
 [1301 
 
I 
 
 LIBRARY OF CONGRESS 
 
 003 080 442 9