LIBRARY OF THE UNIVERSITY OF ILLINOIS AT URBANA-CHAMPAIGN 6 30 U > o 3 evi n. s. no. 61-6 6 cop. 3 'NATURAL HISTORY SURVEY Digitized by the Internet Archive in 2015 https://archive.org/details/papersoncottonbo6166unit U. S. DEPARTMENT OF AGRICULTURE, BUREAU OF ENTOMOLOGY— BULLETIN No. 63. L. O. HOWARD, Entomologist and Chief of Bureau. PAPERS ON THE COTTON BOIL WEEVIL AND RELATED AND ASSOCIATED INSECTS. CONTENTS AND INDEX. Issued July 17, 1909. WASHINGTON: GOVERNMENT PRINTING OFFICE. 1909. U. S. DEPARTMENT OF AGRICULTURE, BUREAU OF ENTOMOLOGY— BULLETIN No. 63. L. O. HOWARD, Entomologist and Chief of Bureau. PAPERS ON THE COTTON BOLL WEEVIL AND RELATED AND ASSOCIATED INSECTS. I. HIBERNATION AND DEVELOPMENT OF THE COTTON BOLL WEEVIL. By E. DWIGHT SANDERSON, Formerly State Entomologist of Texas. II. NOTES ON THE BIOLOGY OF CERTAIN WEEVILS RELATED TO THE COTTON BOLL WEEVIL. By W. DWIGHT PIERCE, Special Field Agent. III. AN ANT ENEMY OF THE COTTON BOLL WEEVIL. By W. E. HINDS, In Charge of Cotton Boll Weevil Laboratory. IV. A PREDATORY BUG REPORTED AS AN ENEMY OF THE COTTON BOLL WEEVIL. By A. C. MORGAN, Special Field Agent. V. NOTES ON THE PEPPER WEEVIL. By F. C. PRATT, Assistant. VI. THE STRAWBERRY WEEVIL IN THE SOUTH-CENTRAL STATES IN 1905. By A. W. MORRILL, Special Field Agent. VII. THE COTTON STALK-BORER. By A. C. MORGAN, Special Field Agent. WASHINGTON: GOVERNMENT PRINTING OFFICE. 1909. BUREAU OF ENTOMOLOGY. L. 0. Howard, Entomologist and Chief of Bureau. C. L. Marlatt, Entomologist and Acting Chief in Absence of Chief. R. S. Clifton, Executive Assistant. C. J. Gilliss, Chief Cleric. F. H. Chittenden, in charge of truck crop and stored product insect investigations. A. I). Hopkins, in charge of forest insect investigations. W. D. Hunter, in charge of southern field crop insect investigations. F. M. Webster, in charge of cereal and forage insect investigations. A. L. Quaintance, in charge of deciduous fruit insect investigations. E. F. Phillips, in charge of bee culture. D. M. Rogers, in charge of gipsy moth field work. A. W. Morrill, in charge of white fly investigations. W. F. Fiske, in charge of gipsy moth laboratory. R. S. Woglum, in charge of hydrocyanic acid gas investigations. R. P. Currie, in charge of editorial work. Mabel Colcord, librarian. Southern Field Crop Insect Investigations. W. D. Hunter, in charge. W. D. Pierce, R. A. Cushman, C. E. Hood, E. S. Tucker, engaged in cotton boll weevil investigations. F. C. Bishopp, J. D. Mitchell, H. P. Wood, engaged in cattle tick life history investi- gations. A. C. Morgan, C. R. Jones, engaged in tobacco insect investigations. I). L. Van Dine, engaged in sugar-cane and rice insect investigations. F. C. Pratt, engaged in cactus insect investigations. IV LETTER OF TRANSMITTAL U. S. Department of Agriculture, Bureau of Entomology, Washington , D. C., June 10, 1909. Sir: I have the honor to transmit herewith, for publication as Bulletin No. 63, seven papers dealing with the cotton boll weevil and related and associated insects. These papers, which were issued sep- arately during the early part of the year 1907, are as follows: Hiber- nation and Development of the Cotton Boll Weevil, by E. Dwight Sanderson; Notes on the Biology of Certain Weevils Related to the Cotton Boll Weevil, by W. Dwight Pierce; An Ant Enemy of the Cotton Boll Weevil, by W. E. Hinds; A Predatory Bug Reported as an Enemy of the Cotton Boll Weevil, by A. C. Morgan; Notes on the Pepper Weevil, by F. C. Pratt ; The Strawberry Weevil in the South- Central States in 1905, by A. W. Morrill; The Cotton Stalk-Borer, by A. C. Morgan. Respectfully, L. O. Howard, Chief of Bureau. Hon. James Wilson, Secretary of Agriculture. CONTENTS: Page. Hibernation and development of the cotton boll weevil . . E. Dwight Sanderson . . 1 Introduction 1 Hibernation 1 Date of entering hibernation 1 Number of weevils entering hibernation 5 Development of immature stages during normal period of hibernation. 8 ♦ Mortality of hibernating weevils * . . 11 Time of greatest mortality during hibernation 14 Places of hibernation 17 Time of emergence from hibernation 19 Prevention of hibernation 22 Summer broods of the weevil 25 Mortality of summer broods . 28 Rate of increase of the weevil 31 Injury to squares in relation to the natural increase of squares 33 Notes on the biology of certain weevils related to the notton boll weevil W. Dwight Pierce . . . 39 Introduction ... 39 Anthonomus disjunctus Lee 41 Anthonomus fulvus Lee 41 Anthonomus squamosus Lee 42 IAxus musculus Say ... 43 Orihoris crotchii Lee 44 An ant enemy of the cotton boll weevil. ( Solenopsis geminata Fab. var. xyloni McC.) W. E. Hinds.. 45 Introduction 45 Description 45 An enemy of the cotton leaf- worm and bollworm 46 Its work as an enemy of the cotton boll weevil : 46 Distribution 48 A predatory bug reported as an enemy of the cotton boll weevil. ( Apiomerus spissipes Say) A. C. Morgan.. 49 Introduction 49 Life history 49 Food . . 50 Length of life cycle 53 Natural enemies 54 Distribution . 54 Conclusion ........ 54 a The seven papers constituting this bulletin were, issued in separate form on Jan- uary 15, February 5, February 5, February 8, February 9, January 22, and February 9, 1907, respectively. VII VIII PAPERS ON THE COTTON BOLL WEEVIL, ETC. Page. Notes on the pepper weevil. ( Anthonomus xneotinctus Champ.) . . F. C. Pratt . . 55 Occurrence in Texas. 55 Origin 55 Food plants 56 Natural enemies 56 Proliferation 57 Remedies 58 The strawberry weevil in the South-Central States in 1905. ( Anthonomus sig- natus Say) A. W. Morrill . . 59 The strawberry weevil in Texas 59 The strawberry weevil in Louisiana 60 The strawberry weevil in Arkansas 61 Varieties of strawberries grown 62 The cotton stalk-borer (Ataxia cry pta Say) A. C. Morgan.. 63 Distribution 63 Host plants 63 Life history 64 Injury and ecor&mic importance % 64 Natural enemies 65 Remedy 65 Bibliography 66 ILLUSTRATIONS. PLATES. Page. 'late I. Work of Lixus musculus and Orthoris crotchii. Fig. 1. — Gall of Lixus musculus and exit hole of adult on stem of Polygonum pennsyl- vanicum. Fig. 2. — Gall of Lixus musculus and entrance hole (closed) of pyralid caterpillar on stem of Polygonum. Fig. 3. — Pod of Mentzelia nuda showing two egg punctures of Orthoris crotchii. Fig. 4. — Pod of Mentzelia opened, showing a cluster of Orthoris crotchii cells, and the cocoons of Tetrastichus. Figs. 5, 6. — Interior of pod of the same, showing several cells of Orthoris •crotchii. Fig. 7. — An isolated cell of Orthoris crotchii , showing the manner in which the seeds are eaten 42 II. Work of the pepper weevil ( Anthonomus xneotinctus) . Fig. 1. — Emergence holes of adult in chilli pepper. Fig. 2. — Full-grown larvae in situ in chilli pepper. Fig. 3. — Pupae in cells in chilli pep- per, Fig. 4. — Undeveloped bell-pepper pod, showing numerous egg and feeding punctures. Figs. 5, 6. — Malformed bell-pepper pods caused by egg punctures 56 III. The cotton stalk-borer ( Ataxia crypta). Fig. 1.— Adult or beetle. Fig. 2. — a, Galleries; b, work. Fig. 3. — Emergence holes of adults. Fig. 4. — a, Larva in situ ; b, pupa in situ. Fig. 5. — a, Gal- lery running into root of plant ; b, castings behind larva 64 TEXT FIGURES. ig. 1. Daily temperatures at College Station, Tex., October 1, 1903, to June 15,1904 4 2. Rainfall and temperature records for College Station and Victoria, Tex., in 1903-4, compared with normals : . . 5 3. Rainfall and temperature records of College Station and Victoria, Tex., for 1902-3, compared with normals 13 4. Rainfall and temperature records, Hallettsville, Tex., 1899-1901, com- pared with normals for each year 15 5. Rainfall and temperature records, Hallettsville, Tex., 1902-1904, com- pared with normals for each year 16 6. Comparison of normal temperatures, October 15 to June 15, for various districts in Texas 20 7. An ant enemy of the cotton boll weevil: Cotton squares showing emer- gence hole of weevil and entrance holes of Solenopsis geminata var. xyloni for comparison 47 8. Apiomerus spissipes: Adult, egg mass, nymph 50 9. Apiomerus spissipes: Egg 50 10. The pepper weevil {Anthonomus xneotinctus ) : Adult 56 IX INDEX. Page. cer negundo, food plant of Ataxia crypia 64 labama argillacea. ( See Leaf-worm, cotton.) mbrosia, food plant of Ataxia crypta 63, 64 nt enemy of boll weevil 45-48 nthonomus aeneolus, host of Catolaccus incertus. 40 on Solanum rostratum 40 aeneotinctus 55-58 enemies 56-57 food for Apiomerus spissipes 51, 52, 54 Sinea diadema 51, 52, 54 plants 39, 56 influence of proliferation 57 in Texas 55 origin 55-56 remedies 58 disjunctus, habits 41 on Heterotheca subaxillaris , 40, 41 parasites 41 Julvus, habits 41-42 host of Bracon mellitor . . 40, 41 on Callirrhoe involucrata 40, 41-42 grandis ( see also Boll weevil). as food for Apiomerus spissipes 51, 52, 54 Sinea diadema 51, 52 host of Bracon mellitor 40, 47 Eurytoma tylodermatis 40 on cotton 39 search for parasites 59 scutellaris, on wild plum in Texas. . . 40 signatus , food plants 39, 60 in Arkansas 61 Louisiana] 60 South-Central States in 1905 59-62 Texas 59-60 squamosus, habits 42-43 host of Bracon mellitor 40, 43 Eurytoma tylodermatis 40,43 on Grindelia squarrosa nuda 40, 42-43 parasites 43 Aphis, cotton. (S ee Aphis gossypii.) gossypii, not proper food for Apiomerus spissipes 52 Apiomerus spissipes, cannibalism 53 copulation 49 distribution 54 i einr 49—50 68 PAPERS ON THE COTTON BOLL WEEVIL, ETC. Apiomerus spissipes, enemies : enemy of boll weevil . food incubation length of life cycle. . . life history oviposition Apion, eaten by Apiomerus spissipes Apricot, food plant of Ataxia crypta Ataxia crypta bibliography distribution economic importance enemies 61 host plants ... life history remedy sordida, bibliographic reference Bees, eaten by Apiomerus spissipes Blackberry, food plant of Anthonomus signatus Boll weevil (see also Anthonomus grandis). ant enemy, Solenopsis geminata xyloni biology of related weevils, notes bug, predatory, enemy control by crop rotation early planting, explanation fall destruction of stalks 7, 6 < 61 51 31 4 l 31 41 grazing leaf- worm planting early varieties of cotton Solenopsis geminata xyloni development of immature stages during normal period of hiberna- tion d a hibernation and development II as affected by rainfall 5— 6l temperature 3— 6| condition of immature stages I emergence, time 191 entrance, date I mortality in cages I the field Ill time when greatest. Ml places. 1/1 of first appearance after emergence 241 prevention I increase, rate 311 mortality at different stages of development I during hibernation, in cages I the field Ill time when greatest 14| in buried squares I of summer broods 2b| INDEX. 69 Page. 1 weevil, percentage emerging from infested squares - - 28 places of first appearance after emergence from hibernation preyed upon by Apiomerus spissipes . progress of infestation compared with production of cotton fruit. . . 88-88 summer broods mortality d lworm, preyed upon by Solenopsis geminata xyloni 4 ^ t elder. (See Acer negundo.) con mellitor , parasite of Anthonomus xneotinctus fulvus 40 ’ 41 grandis - 40,47 squamosus 40, 43 y, predatory, reported enemy of boll weevil 4 ^ ^ 4 rning of stalks in control of boll weevil ao 41-49 '.lirrhoe involucrata, food plant of Anthonomus fulvus 40, 41-4Z mibalism in Apiomerus spissipes \olaccus incertus, parasite of Anthonomus xneolus xneotinctus ^ sp., parasite of Anthonomus squamosus — 4 ^ ndomyiid gall-maker, eaten by Apiomerus spissipes ^3 'ambycobius cyaneiceps, parasite of Lixus musculus. tton aphis. (See Aphis gossypii.) boll weevil. ( See Boll weevil and Anthonomus grandis.) early varieties to avoid damage from boll weevil, explanation 38 planting to avoid damage from boll- weevil, advantage explained . . 38 fall destruction of stalks in boll-weevil control ^ food plant of Anthonomus grandis y Ataxia crypta ’ leaf- worm. (See Leaf-worm, cotton.) squares, injury from boll weevil in relation to natural increase of squares . 88-88 stalk-borer. (See Ataxia crypta.) varieties, production of fruit ^ ^ yield ” 7 yield at College Station, Tex., m 1904 ^ op rotation in control of boll weevil lcumber beetle, twelve-spotted. (See Diabrotica 12-punctata.) lltural experiments in control of pepper weevil " esmoris scapalis on Sideranthus rubiginosus - evelopment of boll weevil during normal period of hibernation 8- ewberry, food plant of Anthonomus signatus ™ iabrotica 12-punctata, eaten by Apiomerus spissipes * " 'troxys sp., parasite of Ataxia crypta 'urytoma tylodermatis , parasite of Anthonomus grandis squamosus 40, 48 Lixus musculus 40, 43 all destruction of stalks in control of boll weevil ^ 'ig, food plant of Ataxia crypta ^ Fire-ant,” name for Solenopsis geminata lies, house, eaten by Apiomerus spissipes - - - ’ bod supply as controlling numbers of boll weevil in the fall Uyptomorpha rugator, parasite of Lixus musculus - - - - hazing, factor in control of boll weevil — Trindelia squarrosa nuda, food plant of Anthonomus squamosus. ’ 70 PAPERS ON THE COTTON BOLL WEEVIL, ETC. Hackberry, food plant of Ataxia crypta Hadronotus n. sp., parasite of Apiomerus spissipes Helianthus, food plant of Anthonomus squamosus Ataxia crypta Heterospilus websteri , parasite of Ataxia crypta , Heterotheca subaxillaris, food plant of Anthonomus disjunctus 4^ Hibernation of boll weevil, as affected by rainfall 5 _^ temperature 3-6 condition of immature stages development of immature stages during normal period g. emergence, time j 19 . entrance, date /. mortality in cagejs \ the field time when greatest number entering places of first appearance after emergence prevention Hinds, W. E., paper, “An Ant Enemy of the Cotton Boll Weevil ( Solenopsis geminata Fab. var. xyloni McC.)” Homalodisca triquetra, refused by Apiomerus spissipes “Hormiga brava, ” Spanish-American name for Solenopsis geminata Increase of boll weevil, rate Ladybirds, eaten by Apiomerus spissipes Lamia crypta , bibliographic reference i — Ataxia crypta Leaf-worm, cotton, factor in control of boll weevil preyed upon by Solenopsis geminata xyloni Leptinotarsa decemlineata , refused by Apiomerus spissipes Lixus musculus, habits host of Eurytoma tylodermatis parasites Locusts, eaten by Apiomerus spissipes Mentzelia nuda, food plant of Orthoris crotchii Microbracon ( Bracon ) nuperus, parasite of Orthoris crotchii Morgan, A. C., paper, “A Predatory Bug Reported as an Enemy of the Cotton Boll Weevil ( Apiomerus spissipes Say) ” “The Cotton Stalk-Borer ( Ataxia crypta Say) ” Morrill, A. W., paper, “The Strawberry Weevil in the South-Central States in 1905 ( Anthonomus signatus Say) ” Mortality of boll weevil at different stages of development hibernating in cages 11 - 14r 17- 24r 45- 31- 43- 40, 49-, 63-i 59-i 2 the field 11 - in buried squares summer broods 28- time when greatest during hibernation 14- Neocatolaccus tylodermx , parasite of Lixus musculus Orthoris crotchii , habits * parasites Peach, food plant of Ataxia crypta Pentatoma ligata , not proper food for Apiomerus spissipes ( INDEX. 71 Page. Pepper, bell or sweet, chilli, and tabasco varieties, food plants of Anthonomus seneotinctus - 56 food plant of Anthonomus seneotinctus. 39, 56 proliferation 57 weevil. (See Anthonomus seneotinctus.) wild or bird, not food plant of Anthonomus seneotinctus 56 Pierce, W. Dwight, paper, “Notes on the Biology of Certain Weevils Related to the Cotton Boll Weevil ” 39-44 Plowing out in control of boll weevil 7 Plum, wild, food plant of Anthonomus scutellaris 40 YPolygonum pennsylvanicum , food plant of Lixus musculus 43 Potato beetle. (See Leptinotarsa decemlineata.) Pratt, F. C., paper, “Notes on the Pepper Weevil ( Anthonomus seneotinctus Champ.)” 55-58 Rainfall as affecting hibernation of boll weevil 5-6, 13 Redbud, food plant of Anthonomus signatus 39 Sanderson, E. Dwight, paper, “Hibernation and Development of the Cotton Boll Weevil” 1-38 Sharpshooter. (See Homalodisca triquetra.) Sideranthus rubiginosus , food plant of Desmoris scapalis 40 Sinea diadema , food 51, 52 Solanum rostratum , food plant of Anthonomus seneolus 40 Solenopsis geminata , enemy of Anthonomus seneotinctus 56 xyloni, description 45 distribution 48 enemy of boll weevil 45-48 cotton leaf-worm and bollworm 46 Stalk-borer, cotton. (See Ataxia crypta.) Strawberry, food plant of Anthonomus signatus 39, 60 varieties grown in South-Central States 62 weevil. (See Anthonomus signatus .) Temperature as affecting hibernation of boll weevil 3-6, 13 time of emergence of boll weevil from hibernation. . . 19-20 Tetrastichus sp., parasite of Orthoris crotchii 44 Thrips, eaten by Apiomerus spissipes 53 Triphleps insidiosus , eaten by Apiomerus spissipes 53 Wasps, eaten by Apiomerus spissipes 51 Xanthium, food plant of Ataxia crypta 63 o U. S. DEPARTMENT OF AGRICULTURE, BDREAC OF ENTOMOLOGY— BULLETIN NO. 63, Part I. L. O. HOWARD, Entomologist and Chief of Bureau. PAPERS ON THE COTTON BOLL WEEVIL AND RELATED AND ASSOCIATED INSECTS. HIBERNATION AND DEVELOPMENT OF THE COTTON BOLL WEEVIL. BY E. DWIGHT SANDERSON, Formerly Slate Entomologist of Texas. Issued January 15 , 1907 . WASHINGTON: GOVERNMENT PRINTING OFFICE. 1907. LETTER OF TRANSMITTAL. U. S. Department of Agriculture, Bureau of Entomology, Washington , D. 6V, September 15 , 1906. Sir: I have the honor to transmit herewith the manuscript of a cor tribution by Prof. E. Dwight Sanderson on the hibernation an development of the cotton boll weevil. This paper embodies til results of a large amount of work carried on by Professor Sandersol while State entomologist of Texas, and was written by him in the fan of 1904. In view of its practical bearing and the fact that much cl the data it contains is new, I recommend its publication as Part I cl Bulletin No. 63 of this Bureau, the bulletin as a whole to be entitle! “Papers on the Cotton Boll Weevil and Related and Associate Insects,” and to be made up of separate papers by various persor engaged in cotton boll weevil investigations. Respectfully, F. H. Chittenden, Acting Chief of Bureau. ' lion. James Wilson, Secretary of Agriculture. CONTENTS. introduction - Hibernation Date of entering hibernation Number of weevils entering hibernation Development of immature stages during normal period of hibernation Mortality of hibernating weevils Time of greatest mortality during hibernation Places of hibernation Time of emergence from hibernation Prevention of hibernation Fall destruction of stalks Crop rotation Where the weevils first appear . Summer broods of the weevil Mortality of summer broods Rate of increase of the weevil Injury to squares in relation to the natural increase of squares Page. 1 1 1 5 8 IP 14 17 19 22 22 24 24 25 28 31 33 hi ILLUSTRATIONS. TEXT FIGURES. Page. Fig. 1 . Daily temperatures at College Station, Tex., October 1, 1903, to June 15,1904 4 2. Rainfall and temperature records for College Station and Victoria in 1903-4, compared with normals 5 3. Rainfall and temperature records of College Station and Victoria, Tex., for 1902-3, compared with normals .y 13 4. Rainfall and temperature records, Hallettsville, Tex., 1899-1901, com- pared with normals for each year 15 5. Rainfall and temperature records, Hallettsville, Tex., 1902-1904, com- pared with normals for each year 16 6. Comparison of normal temperatures, October 15 to June 15, for various districts in Texas 20 IV J. S. D. A., B. E. Bui. 63, Part I. C. B. W. I., January 15, 1907. PAPERS ON THE COTTON BOLL WEEVIL AND RELATED AND ASSOCIATED INSECTS. HIBERNATION AND DEVELOPMENT OF THE COTTON BOLL WEEVIL. By E. Dwight Sanderson, Formerly State Entomologist of Texas. INTRODUCTION. The most economical method of controlling* the boll weevil is to destroy its food supply by burning or grazing the green cotton stalks in the fall before frost occurs, so as to reduce by starvation the num- bers which go into hibernation. As it is known that a large proportion of weevils die during hibernation* it may be readily understood that a much greater degree of benefit may be derived by destroying the weevils in the fall than by attempting to destroy them in the spring. The time at which the weevils enter hibernation, the numbers which attempt to hibernate, the mortality occurring during the winter and the factors which influence this mortality, and the time of emergence in the spring are, therefore, matters of practical importance upon which we have made careful observations. HIBERNATION. DATE OF ENTERING HIBERNATION. ! Table I gives a summary of eight lots of weevils which were con- I fined in cages at different dates during October and November, 1903. Altogether some 1,300 weevils were used in these experiments. Begin- ning with October 7 one cage was started each week until November 17. On November 18 a sharp freeze occurred, and almost immediately thereafter all weevils went into hibernation. It will be seen that in lots 1, 2, and 3 all of the weevils died prior to November 18. During the beginning of these experiments a plentiful supply of squares was furnished the weevils for food, but as they showed an inclination toward hibernating about the third week in October this supply was discontinued. In all cases the weevils which hibernated successfully l 2 PAPERS ON THE COTTON BOLL WEEVIL, ETC. did not begin hibernation until about November 17 or 18. This was also true of other weevils under observation in the laboratory. The daily notes concerning one or two of the outdoor cages started in Octo- ber are of interest in this connection. For example, lot 3 comprized 100 weevils put in hibernation cage with food October 16. The hiber- nation cages were wooden boxes 2 feet high and 1 foot square, with the sides about one-half open and covered with wire window screening, except the front, which was of glass. In the bottom of each cage was placed plenty of rubbish, consisting of fallen cotton leaves, rotten wood, etc. The weevils were thus exposed to the outside temperature, but were largely protected from rain. These cages were placed in various situations in a small grove and pasture at the apiary of the Agricultural and Mechanical College, College Station, Tex. The rec- ords kept of this lot and of lot 2 are as follows: October 17, 18, no hibernation, weevils active; October 19, same, quiet in bottom of cage; October 21, sign of hibernation evident, some weevils congregate about dead rubbish on floor of cage; October 22, 23, 24, similar conditions, some seem to be hiber- nating; October 25, a few more in hibernating attitude; October 26, weevils are undoubtedly hibernating, shown by dormant attitude under rubbish; October 27, about all hibernating — if they were let out they would not hibernate, as some were let out, and they joined those in the field and commenced feeding; October 29, hoi day arouses weevils to activity; October 30, think weevils die at the rate of 1 to 4 pei day. November 2, warm weather makes them’ exceedingly active; this continues until the 8th, when cool weather checks their activity; November 15, all weevils dead. Lot 2. — One hundred weevils placed in cage October 8 and fed with squares; Octo- ber 9, weevils lively, flying about in cage; October 10, 11, no signs of hibernation October 12, less active; October 13, those on rubbish dead or dying; October 14, nc signs of hibernation; October 18-23, weevils active, no hibernation; October 25, al but 8 dead, those hibernating under rubbish; October 29, weevils active; Octobei 29-31, November 1, 2, sluggish; November 5, all dead. Table I . — Percentage of mortality of hibernating weevils. Lot. Date placed in cage. Number of wee- vils. Date hi- bernated. To November 18, 1903. November 18-30, 1903. December, 1903. Died in Decem- ber. Date. Alive. Dead. Date. Alive. c3 a> ft o3 « Alive. 'd c3 0) bi K jn Total De- cember!. Per Per Pa- Pa- Per Per Pa- Pa- 1903. ct. ct. ct. ct. ct. ct. ct. ct. 1 October 7 100 Nov. 1 100.0 I Oct. 24 a 2 October 8 !Oct. 25 92. 0 100 )Nov. 3 99.0 (Nov. 5 100.0 « 3 October 16 100 /Oct. 30 (6) |Nov. 15 100.0 4 October 23 100 Nov. 18 Nov. 18 100. 0 14 7 71 22 71-i- 71 5 October 25 ....do... 9.0 J 4 67 33 , )- 82 71 a 6 November 1 . . . 200 ....do... 50.0 '50.0 23 50 50 \26 27 18 14 82 61 25 61 + 11 7 November 9 . . . 200 ....do ... Nov. 15 93.5 6.5 8 November 17.. 200 ....do... 25 100 :::: 5+7 ! Average . . . .,nu ....do ... 8.0 1 “I 4,5,6 , Average . . . 000 ....do ... 21.0 . . . J 75 + 51-1 1 | 1 I « Furnished with squares for food. b One to 4 dying per day. HIBERNATION AND DEVELOPMENT OF BOLL WEEVIL. 3 Table I . — Percentage of mortality of hibernating weevils — Continued. Lot. Date placed in cage. Number of J weevils. Date hi- bernated. February, 1904. March, 1904. Total sur- vived. Date active. Date. Alive. Dead. Date. Alive. Dead. 1 a 2 « 3 4 5 a 6 8 5+7 4, 5, 6 1903. October 7 100 100 100 100 300 200 200 200 500 600 Per cl. Per ct. Per ct. Per ct. Per ct. Mar. 22 Mar. 18 October 8 October 16 Or+.nhpr 93 Nov. 18 ....do... ....do... do ( b ) 2 ( b ) October 25 November 1 . . November 9 . . November 17 . Average . . Average . . 7.3 91.6 22 5.6 94.4 5.6 21 24 10.5 7.0 89.5 93.0 ....do ... 7.0 .. ..do ... 6.2 do . . . a Furnished with squares for food. b Accidentally disturbed and discontinued. It will be seen from these experiments that weevils can not be forced into hibernation before the temperature declines to that point at which the} 7 normally hibernate, and that if deprived of food or confined much before that time most of them will die. The first weevils to go into hibernation were those of lot 4, placed in cages on October 23, but these did not actually hibernate until November 18. The case of lot 8, of 200 weevils placed in a cage November IT, in which all died by November 25, is difficult to explain. Our notes upon the condition of the weevils in the fields during the same fall show practically the same conditions as those existing in the cages. The very careful record by Mr. A. F. Conradi, made in fields in the neighborhood of }the college, is briefly as follows: September 6, 7, abundant, very few perfect squares, dry weather; September 8, 9, abundant, mating pairs common; September 10, abundant, practically all squares punctured, mating pairs less common; September 11, 12, less numerous, fewer mating pairs, few perfect squares; September 13, less abundant, perfect squares more com- mon, mating pairs scarce; September 14, 15, less abundant, perfect squares more common; September 16-18, fewer weevils, perfect squares more common; Septem- ber 19, less destructive than at any time since the 10th, perfect squares abundant; September 20-28, similar conditions; September 29, 30, conditions growing worse, more weevils, fewer perfect squares; October 5, numerous, eating vigorously; Octo- ber 10, exceedingly abundant and destructive, a few dozen perfect squares found with great difficulty, smallest squares often punctured in several places; October 11, 'abundant, several on plants near field, one in door yard on house, mating common; ! October 12, 13, abundant, few mating, few perfect squares; October 14, many weevils I found clustering about the base of bolls of all sizes just inside of the involucre, no signs of hibernation; October 16, 100 weevils found on 95 plants in thirty minutes; October 18, 19, weevils more sluggish, unusual numbers congregate behind involucres of larger squares; October 20, the size of weevils is exceedingly variable, more so than at other seasons, general depression in activity; October 22, some adults mating, no (signs of hibernating; October 23-26, less numerous, no mating pairs, weather cool, jbeetles seem to be leaving; November 2, abundant in certain localities in the field ; ;.on 5 plants they average one to a square, then for a dozen or more none are found; I mating common, but little egg laying for several days. 9223— No. 63, pt 1-07- — ■ 2 4 PAPERS ON THE COTTON BOLL WEEVIL, ETC. October 3, weather warm, weevils that had assumed a hibernating attitude lively again, mating pairs and oviposition common; October 4, much more active, mating pairs and number of eggs increasing, collected 100 weevils from 132 stalks; October 6, conditions similar to middle of September, collected 50 weevils from 46 stalks; October 7, weevils numerous, mating common, more egg punctures, collected 100 weevils from 92 stalks and 67 from 80 stalks; October 9, at apiary, 100 weevils on 98 stalks at east end, 50 on 100 stalks, west end, 48 on 90 stalks at the middle of the field; October 11, 90 weevils on 176 stalks; October 13, average of 1 weevil to 2 x a> H S Em stalks; October 15, 16, weevils active, eggs common; October 17-21, low temperature, with sharp freeze on the night of the 18th, when the cotton froze and the supply of [ food was exhausted; October 22, after the 18th weevils went into hibernation rapidly, I but diminution not markedly noticed until the 22d, when there was but 1 to 10 i stalks. The weevils were not injured by the freeze of the 18th, as many I were found in the field on the 19th. By referring to figure 1, giving). ' HIBERNATION AND DEVELOPMENT OF BOLL WEEVIL. 5 the daily temperature, a it will be seen that the activity of the weevils corresponds very closely with the rise and fall in temperature. Hunter and Hinds 6 state that hibernation will commence as the mean average temperature falls to between 55° and 60° F., which our observations of last fall corroborate. They also state that at Victoria hibernation will normally commence about Decern ber 1. By referring to figure 2 it will be seen that the normal temperature at Victoria drops to 60° at about December 1 and at College Station from November 10 to 15. At Dallas hibernation would be several days earlier and for northeastern Texas generally about November 1 to 10. From the observations available, therefore, it would seem safe to assert that hibernation will usually occur when the mean average temperature drops below 60° F. Fig. 2. — Rainfall and temperature records for College Station and Victoria in 1903-4 compared with normals. NUMBER OF WEEVILS ENTERING HIBERNATION. The number of weevils found in an infested field in the fall at about the time hibernation would normally begin is governed largely by the food supply and is influenced also by the temperature and rainfall during September and October. a Concerning the method of plotting the curves it should be stated that the vertical lines upon which the nodes of the curve occur are marked as being the 15th of the month. The mean temperature as given by the Weather Bureau is reported at the end of each month. If we should consider the node of the curve as occurring at the end of the month, we would misrepresent the facts, as during the spring and fall the temperature is on a gradual rise or decline. Consequently if the mean monthly temperature is given for the 15th of the month, the exact temperature for dates between these nodes can be more correctly ascertained from the curve. This applies also to the figures showing rainfall. &Bul. 45, Div. Ent., U. S. Dept. Agric., p. 81. 6 PAPERS ON THE COTTON BOLL WEEVIL, ETC. If during September and early October there should be a normal or excessive rainfall and the plants keep growing, there will be plenty of squares upon which the weevils may breed and multiply until frost occurs. If, on the other hand, during this period the weather should be hot and dry, the plants will stop growing and those infested squares which fall to the ground will be quickly dried up and but a relatively small number of weevils will be produced. This was strikingly shown in observations upon the number of weevils on cotton at the college in the fall of 1903 as compared with the fall of 1901. During Septem- ber, 1904, there was an average of not over 3 squares to a plant, whereas in September, 1903, there was an average of about 40 squares to a plant. At this time during these two seasons there was approxi- mately the same number of weevils in the field; but it can be readily seen that with so few squares in which to breed, in September, 1904, there would be far fewer weevils developed after September to go into hibernation. Mr. Louis Teltschick states that in Lavaca County, Tex., at the time he turned cattle in to graze his cotton, October 17, 1903, there were about 2 weevils to a stalk. Concerning this he writes as follows: I will state in this connection that in my opinion the bulk of the weevils had then left the cotton. I think that during the last three years the majority of weevils began leaving the cotton as early as September 1. They were undoubtedly com- pelled to do this by a scarcity of squares. To be more explicit I give you the follow- ing figures: On August 28, 1904, I estimated the number of weevils in my field at 12 per stalk. While I made no such estimate in 1902 or 1903, considering the extent of the injury worked by them in those years, I think that their numbers on the same date might safely be put at 6 for 1902 and 9 for 1903. When I cut the stalks early in October, 1902, I found that there were about II weevils per stalk, and in October, . 1903, when I turned the cattle in, there were about 2 weevils per stalk. You will see that according to these estimates from 75 to 80 per cent of the weevils infesting the fields on August 28 in those years had disappeared by October 15, leaving out of account all weevils hatched between the two dates. Where they went or what be- came of them I am unable to say. They may have gone to uninfested regions, or they may have died or gone into hibernation quarters when squares became too scarce to supply all of them. These observations merely confirm our previous statement that the number of weevils in the fall is controlled entirely by the food supply. One of the principal factors in cutting off the food supply of the boll weevil in the fall is the cotton leaf-worm or so-called “army worm” ( Alabama argillacea Hbn.). Whereas this insect was formerly j considered to be the great scourge of the cotton planter, it may now be said to be an almost positive advantage in territory affected by the boll weevil, as its injury does not ordinarily occur until after the crop has been made, and all further fruiting would be prevented by the boll J weevil. As soon as the leaf worms have stript the cotton of foliage I and squares the weevils either die or leave for fields where the}’' may j obtain food. This was strikingly shown in an examination of four I HIBERNATION AND DEVELOPMENT OF BOLL WEEVIL. 7 plats at Smith & Carson’s plantation on September 14, 1904. In plats 1 1-3, where the leaf worm had stript the plants clean of foliage, but . 8 weevils were found to 100 stalks and but 40 per cent of the squares were punctured. On plat 4, immediately adjoining these plats, which had not been so seriously injured by the leaf worm, there were 27 squares to a plant, 148 weevils to 100 stalks, and 70 per cent of the squares were punctured. Mr. Teltschick has given us some interesting observations upon the leaf worm, as follows: As far back as I can remember — some twenty-odd years — the people here ( Lavaca County) had to fight the leaf worm every year. Arsenic, Paris green, and London purple were used with varying success. At first the worms made their appearance as early as June and frequently hatched in such numbers as to enable them to sweep a field clean in less than 48 hours. Gradually, however, this wave of worms sub- sided; that is to say, they made their appearance later and later in the season, and became fewer in numbers, until finally when the boll weevil made its first appearance, their coming was no longer viewed with alarm. On the contrary they were often welcomed by the farmers, inasmuch as they cleared away the thick foliage of the cotton plant, thus often saving a considerable number of the bolls at the bottom, which would otherwise have rotted for the want of light and air. At the time of the weevils coming here (seven or eight years ago) the worms were no longer regarded as a menace to the cotton crop; they made their first appearance then either late in August or early in September. They kept this up practically until in 1902 and 1903 they failed to come at all. This year (1904) they reappeared early in September and completely stript cotton of its foliage. The second destruc- tive brood of worms is just now (October 2, 1904) hatched, and there is every reason to believe that they will keep the stalks bare until frost as they did in former years. It will have to be admitted that during 1902-3, when cotton grew and formed squares undisturbed up to the first frost, which occurred in December of both years, an opportunity was given the weevil to go into hibernating quarters later in the season and in greater numbers than would have been the case had the further growth of cotton been prevented either by man’s or nature’s means. Thus the leaf worm accomplishes practically the same result as is obtained by grazing stalks with cattle and may be considered a most valuable ally. The number of weevils which go into hibernation may of course be decreased by the thoro grazing of the cotton with cattle before frost, or by cutting, plowing out, and burning the stalks. It is safe to say that by a thoro fall clearing of the ground the number of weevils entering hibernation can be decreased at least 75 per cent. Weevils in all stages are thus immediately destroyed, and if the land is cleared of cotton a month before the normal hibernation period the weevils then escaping are almost certain to starve, as has been shown in Table I, before it becomes cold enough for them to hibernate. Under normal conditions, as noted in Table III, column 3, the num- ber of weevils to a stalk at the time of hibernation will vary from 1 for every 2 stalks to 2 for each stalk in badly infested fields, the num- ber depending, as before stated, entirely on the food supply and 8 PAPERS ON THE COTTON BOLL WEEVIL, ETC. weather conditions. There seems to be no reason to believe that any difference will be found in the number of weevils in infested fields in northern Texas as compared with southern Texas except as it may be controlled by these factors, for after the third brood becomes matured the number of weevils depends entirely upon these conditions. DEVELOPMENT OF IMMATURE STAGES DURING NORMAL PERIOD OF HIBERNATION. In southern Texas during an open dry winter larvae, pupae, and adult weevils may frequently be found in the injured bolls hanging upon the cotton stalks. In feeding upon the bolls the larvae frequently eat a hole thru the septum between two locks and form a small cavity just under the inner surface of the lock. These feeding cavities become lined by the dried and hardened excrement, so that when a lock is broken to pieces the weevil cells may be separated out and have quite the appearance of seeds. It is within such cells that the imma- ture stages pass the winter in the bolls. Occasionally, however, the larvae eat into the seed, so that immature stages may be found actually within the seeds. This is undoubtedly not the usual habit and there seems to be practically no danger of disseminating weevils actually secreted inside of the seeds. W e have been unable to make a personal examination of these conditions, but on February 20, 1903, Mr. W . T. Allgood, of Devine, Medina County, Tex., wrote us sending some cotton bolls and seed, in which we found 4 weevils alive, 1 dead, 1 newly transformed, 1 in a seed, and also 2 seeds which seemed to have been eaten out by weevil larvae, but which contained only larvae which may have been parasitic upon the weevil. The 4 live weevils were kept in a tube in our laboratory until about April 1, when they die , owing to the unsuitable conditions. tux Several of the infested seeds were evidently entered before the shell was fully hardened and the normal development of the seed pre- vented, whereas in other cases the weevils were in seeds fully devel- oped. It seems quite reasonable that the larv* should bore into the seeds during the early winter, for as the cotton matures the seeds* would be more succulent. However, it seems exceedingly improba- ble that weevils could be disseminated by being carried in the seed, for they occur within the seed only on unopened bolls which hangt upon the plant, and very rarely would cotton in which weevils had entered the seed ever be ginned. Early in March of the same year Mr. Allgood sent the writer quite i, quantity of bolls, which were examined March 12 with the following results: Out of 200 bolls 165 contained no weevils in any stage, alivi or dead, but of these, 20 per cent had been noticeably injured >y weevils. Forty, or 20 per cent, contained weevils alive or dead nl some stage. In these 40 bolls there were 40 live and 11 dead pup* HIBEKN ATION AND DEVELOPMENT OF BOLL WEEVIL. 9 30 live and 40 dead adults, and 5 dead larvae. Many of these weevils had just transformed from pupae and were still soft and light brown. Some transformed after being received. But one live larva was found in all the material examined. Mr. Allgood wrote the writer that these bolls had been collected from stalks in a neighbor’s field in which there were 7,000 plants to the acre, that each plant held from 12 to 15 bolls, and that the bolls sent in were picked at random from the field. Upon this basis, disregarding the live pupa? but counting the 30 weevils which were found alive as having survived the winter, there would have been 10,500 weevils per acre in the spring. If but one-third of these survived there would have been 3,500 per acre, or more than have ever been recorded in any field observations. If, as seems quite possible, but one-fifth survived, there would have been 2,100 per acre, which is approximately the number which survived in Lavaca County under the favorable weather conditions of the winter of 1903-4. Mr. Allgood wrote that cattle were then grazing in these fields, but it is doubtful if the cattle would have eaten many of the dry hardened bolls. In another lot of 48 bolls and forms sent b} 7 Mr. Allgood at the same time, we found 18 uninhabited large bolls, 15 forms, and 15 small bolls. These contained 17 live weevils, 4 live pupse, 4 dead pupa3, and 2 dead larva;, or one live weevil to every 3 bolls or forms. These weevils were confined in a wooden box in the laboratory and were still alive April 10, but later died, as no food could be supplied them at that time under temperature conditions such that they would normally have emerged. Doubtless they would all have emerged early in April at Devine. The temperature at Pearsall, the nearest weather station to Devine, during the winter was as follows: Decem- ber, 1902, mean 54°, lowest 30°; January, 1903, mean 51.8°, highest 80°, lowest 29°; February, 1903, mean 52.8°, highest 77°, lowest 23°; March, 1903, mean 60.6°, highest 82°, lowest 38°. It will be seen that there were but few days during the winter in which the temperature was below freezing. The total mean tempera- ture for these four months at Pearsall was but 63° less than the total normal mean for southwestern Texas during this time, and the rain- fall at Pearsall was but 6.34 inches greater than the normal for south- western Texas, no normal records for Pearsall being given. The total rainfall of these months is normally but 7.51 inches for southern Texas, so that the rainfall was nearly double the normal and the winter would appear to have been exceedingly unfavorable for the hibernation of weevils in their usual places of shelter. However, weevils which developed in the unopened bolls were almost entirely protected from the excessive rainfall, and the temperature being nor- mal they would probably survive, while those weevils which went into hibernation in all usual places in the fall would doubtless have suffered an excessive mortality due to the unusual rainfall. 10 PAPERS ON THE COTTON BOLL WEEVIL, ETC. Immature stages and adult weevils are often found in early winter in the unopened bolls hanging on the stalks thruout the infested regions. We received live adult weevils in bolls from Hamilton, Hamilton County, in December, 1903, and they have been observed here and elsewhere, but we have no record of weevils having been found alive in these situations in early spring in central Texas, and it is exceedingly doubtful if many of them would survive the winter except in southern Texas. Table II. — Condition of immature stages of weevils after date of normal hibernation. o h3. Date. 1 Number of squares or bolls examined. 1 2 3 4 1903. Nov. 13 Nov. 12 Nov. 18 Nov. 20 a 550 a 98 a 50 a 200 5 Nov. 22 a 100 6 Nov. 28 a 100 7 Nov. 26 a 200 8 Nov. 30 1 o 200 9 Dec. 4 « 200 10 ....do ... a 200 11 do ... a 200 12 Dec. 12 Ja200 \cl00 13 14 15 16 Dec. 5 ....do... Dec. 4 do . . . a 250 c 250 a 500 c 200 Contain- ing eggs. Containing larvse. H 1 0) In Per cent of squares show- weevil injury, but no stage present. Per cent of total. Per cent unhatched. Per cent of total. Per cent less than half grown. Per cent over half grown. Per cent alive. Per cent dead. Containing pupae, cent of total. OG 0> ! ga oj a> C7 1 — 02 O a> li o O CL| > 9. 15 25.5 0 40.5 25 8 24 10 14 0 io 12 2 100 9 4 100 100 4 2 100 100 1 b 100 & 100 1 b 99 b 100 b 100 12.8 100 b 87. 2 0 42 57 43 100 0 b 95. 8 4 100 100 0 b 96 Remarks. Collected Nov. 7. Squares frozen Nov. 18. Collected from plant Nov. 15. Collected from ground Nov. 22. Collected from ground Nov. 8. One-half collected from ground, one-half from plant, Nov. 1. One-half collected from ground, one-half from plant, Nov. 15. One-half collected from ground, one-half from plant, Nov. 18. One-half collected from ground, one-half from plant, Nov. 22. One-half collected from ground, one-half from plant, Nov. 24. One-half collected from ground, one-half from plant, Nov. 28. a Squares. b Squares with no stage of weevil in them. Percentage injured unknown, but probably 40 per cent as in lot 1. c Bolls In November and December, 1903, Mr. Conradi made careful obser- vations at College Station upon the condition of the immature stages after the freeze of November 18. The results of these observations are shown in Table II. It will be noticed that no pupa? or adults were found in over 3,600 squares and bolls examined, showing that egg laying had temporarily ceased during some period before Novem- ber 1. The most probable explanation of this remarkable absence of HIBERNATION AND DEVELOPMENT OF BOLL WEEVIL. 11 pupse and adults in the presence of eggs and larvae is that an outbreak of the cotton leaf-worm ( Alabairta argillacea Hbn.) had completely defoliated the plants during the latter part of September, preventing entirely the formation of squares, and thus stopping abruptly for a short time the multiplication of the weevils in squares. Subsequent^ squares were formed, and when these became large enough the repro- duction of weevils was resumed after about November 1, so that most of the larvae found were over half grown. In lots 4 and 5, when exam- ined November 20 and 22, from 10 to 12 per cent of the eggs were still unhatched. It seems probable that at this time some of the eggs, but no large percentage, failed to hatch. Most of the larvae died during the first stage and but few lived to be over half grown. On Novem- ber 12, in lot 2, 25 eggs were found in 98 squares, which were col- lected November 7, showing that the egg period had been doubled. Mr. Conradi notes in this connection that the life-history period was first noticed to be decidedly lengthened after November 9, eggs laid November 10 requiring 5 days to hatch. From 24 eggs hatched between November 24 and 28, 15 larvae died within a day after hatch- ing, 7 by th£ time they were half grown, and but 2 developed into weevils, which emerged December 4. On December 12 an occasional larva was found in the field and sev- eral were brought into the laboratory, where 3 transformed to pupee January 7, but then died. They would undoubtedly have failed to pupate in the field. MORTALITY OF HIBERNATING WEEVILS. The mortality of weevils during hibernation is shown in Tables I and III. Table I gives 4 the mortality of weevils confined in cages at dif- ferent dates from October 7 to November 17, as previously described. In but two lots did weevils survive the winter. In these lots (5 and 7), of 500 weevils an average of 6.2 per cent survived the winter and were active between March 22 and March 24. On March 21, however, 10.5 per cent were alive, and it was noted that these weevils had been active since March 18, so that practically 7.5 per cent of the two lots survived the winter under the supposedly favorable conditions furnished them in the hibernation cages. 9223— No. 63, pt 1—07 3 12 PAPERS ON THE COTTON BOLL WEEVIL, ETC. Table III. — Percentage of weevils surviving hibernation infields. Year. Locality. Number of weevils to a stalk in fall. | Weather: December- March. Date first emerged. Dates counted, inclusive. Weevils counted. Area counted. Number of weevils to 100 stalks. Percentage surviv- ing hibernation. Rainfall. Tempera- ture. Win- ter. Nor- mal. Win- ter. Nor- mal. 1900-1 Vienna « ft +5. 54 8. 32 + 2.7 55.8 Mar. 10 10 acres 1902-3 do 1.5 + 6.80 8.32 + .5 55.8 £ acre c 1903-4 do (?) -6. 08 8. 32 + 14.1 55.8 Mar. *i5* MaVi7d::: 2*400 1 acre 31.0 d 30 1900 Booth e-4-5. 71 14.47 -5.0 55.5 May 22-31. 2,000 2 acres 13.0 13 1903 Hunter / +9. 60 7.02 - .8 54.7 May 30 6 140 stalks . . 4.0 1902 Wellborn . . (/— 8. 84 13. 40 -9.1 55.7 May 22-29. 397 2 acres 2.5 2.5 1903 do h2 . 00 <7+2. 91 13. 40 -11.6 55.7 May "8* June 26- 1 2, 125 stalks. .04 .08 1 July 17. 1903-4 do 1.00 {7-2.45 13.40 + 2. 1 55. 7 Apr. 17 May 20- 26 1,600 stalks. 1.6 . 1.6 June 17. 1903 College Sta- -1.38 12.72 -3.5 53.6 May 15 June 18... 5 940 stalks . . .5 : tion. i 1903 doA -1.38 12.72 - 3.5 1 53.6 do 2 40 stalks . . . 5.0 5.0 1904 doJ ".*75 -6. 12 12. 72 +14.1 53.6 Mar. *29* May 19- 49 2,930 stalks. 1.66 1.66 Junel4. 1904 do.»* ... 1.00 - 6.12 12. 72 +14.1 53.6 Apr. 1 May 22- 20 1,840 stalks. 1.0 1.0’ June 9. 1904 Rosprim.M.. (?) -6.12 12.72 +14.1 53.6 Apr. 8 June 7 10 1,000 stalks. 1.0 a Observations by Louis Teltschick. ft Record for Hallettsville. c Three-fourths-acre trap rows. d Estimate of several fields, May 17. e Records of Houston, as far as given. /Record for New Braunfels. a Records for Brenham. h August 17, 2 weevils pei;100 stalks. i In sorghum in 1902. fc Cotton, 1902, a crossroad by barn. l Cotton not destroyed in fall of 1903. Cotton destroyed in fall of 1903. n In corn in 1903. In the field we have never found the percentage surviving at the college to approach this number, except in one instance. As shown in Table III, in the spring of 1903 cotton was planted near the college barn, where it had been grown the previous year and left standing. The weevils had the most favorable conditions possible for hibernat- ing around the barn. They appeared so numerously on the cotton in the following spring that on June 18 there were 5 hibernated weevils to every 100 stalks. More weevils were found in this cotton thruout the year than in any other locality. If it be fair to assume that about one weevil to a stalk occurred in this field at the time of hibernation in the fall of 1902, then about 5 per cent survived. It is quite possi- ble that there were more than this number in the fall, and the per- centage surviving would therefore be somewhat less. Elsewhere, in the neighborhood of the college, not over 2 per cent have ever been found to survive the winter in the field, tho several fields within a distance of 4 miles have been carefully observed. In the Brazos bottom, on Smith & Carson’s plantation, Professor Mally (report, 0 p. 16) records picking 397 weevils from 2 acres May 22-29, 1902, or 2.5 weevils to 100 stalks, which would give not over 2.5 per cent surviving hibernation. In 1903 but 1 weevil was found to 2,125 stalks, or 0.04 to 100 stalks, which would give a percentage of hibernation of not over 0.04 and probably 0.08, as there were 2 weevils Report on the Boll Weevil, Austin, Tex., August, 1902. HIBERNATION AND DEVELOPMENT OF BOLL WEEVIL. 13 to a stalk the previous fall. During the present year (1904) but 1.6 per cent survived hibernation on the plots counted. This figure is open to slight error, as upon this plantation there were possibly 20 acres of cotton about a mile away, which were planted much earlier than was most of the plantation. This early cotton was infested at the rate of 20 weevils to 100 stalks on May 31. It is evident that the weevils which hibernated on the plantation as a whole had concentrated on this early planting, but if the total number of weevils upon this early cotton had been distributed over the whole acreage there would have been not over 2-J- weevils per 100 stalks or 2.5 per cent surviving hibernation at the most. Two per cent would probably be very nearly correct. At Victoria, Hunter and Hinds have shown that approximately 15 per Fig. 3. — Rainfall and temperature records of College Station and Victoria, Tex., for 1902-3, compared with normals. cent of late-developed weevils survived the winter of 1902-3 in hiber- nation cages. The very careful records made by Mr. Teltschick by hand-picking the weevils during the spring (see p. 59 of that report) shows that this very closely approximates the number of weevils nor- mall} 7 surviving under field conditions in Lavaca County, but that dur- ing the last winter (1903-4) fully 30 per cent of the weevils, or twice as many as usual, hibernated successfully. In 1900 Professor Mally’s records (1. c., p. 32) show that Mr. Kramer picked 2,000 weevils at Booth, Fort Bend County, or 13 to 100 stalks, which would give at least 13 per cent surviving the winter, in a latitude practically the same as that of Vienna. It will be seen, therefore, that whereas 15 per cent may be considered an average of the number surviving 14 PAPERS ON THE COTTON BOLL WEEVIL, ETC. hibernation in southern Texas, normally onty about 2 per cent, and never over 5 per cent, survive under held conditions at College Station, and about 2£ per cent in the Brazos bottom at Wellborn. In other words, only about one-fifth as many weevils survive the winter here as do in southern Texas. Whether a corresponding difference will be found in the mortality between central Texas and northern Texas remains to be determined. This will depend on the weather condi- tions discust below, but it seems doubtful that there will be so great a difference. TIME OF GREATEST MORTALITY DURING HIBERNATION. In Table I it may be seen that in lots 4, 5, and 6, including 600 weevils, there was a mortality of at least 51 per cent during the month of December, and that in lot 5 there was a mortality of 9.6 per cent during January and but 2.8 per cent during the remainder of the year. Referring to figure 2 it may be seen that at College- Station in the winter of 1903-4 the rainfall was greatest in December, while during the balance of the winter it was below normal. The maximum in December approached the normal, whereas usually the maximum for winter occurs in January. The temperature was very close to normal thru January, but considerably above normal during the remainder of the winter. In general, an excessive rainfall seems to be a most important factor in increasing the mortality of the weevil. The temperature does not usually depart sufficiently from the normal to be an important factor, except that when occurring with a deficient rainfall an excess of tem- perature appears to be decidedly beneficial to hibernation. By refer- ring to figures 4 and 5, showing the normal rainfall and temperature for Hallettsville, Tex., and the monthly mean temperature and rainfall dur- ing the winters of 1899-1900, 1900-1901, 1902-3, and 1903-4, it may be seen that in 1900-1901 there was a decided deficiency in rainfall and a slight excess in temperature, whereas in 1902-3 there was an abnor- mally large rainfall and a nearly normal temperature. In 1901 and 1904 the weevils survived in unusally large numbers, while in 1903 the number hibernating successfully was considerably less than nor- mal. The same general conclusions are shown when the rainfall and temperature records as given for College and Brenham in Table III are compared with the percentage of weevils surviving at College and in the Brazos bottom below Wellborn. The rainfall and temperature records are shown diagrammatically in figures 2 and 3, giving also the normal rainfall and departure from normal for College Station and Victoria in 1902-3 and 1903-4. If it be true that the rainfall is the most important climatic factor in the mortality of the hibernating weevils, it is interesting to com- pare the rainfall of different sections of the State as given by the HIBERNATION AND DEVELOPMENT OF BOLL WEEVIL. 15 Fig. 4.— Rainfall and temperature records, Hallettsville, Tex., 1899-1901, compared with normals for each year. 16 PAPERS ON THE COTTON BOLL WEEVIL, ETC. Flo. 6.— Rainfall and temperature records, Hallettsville, Tex., 1902-1904, compared with normals for each year. HIBERNATION AND DEVELOPMENT OF BOLL WEEVIL. 17 Texas section of the U. S. Weather Bureau. The normal total rain- fall during* the winter months of December to March for southwestern Texas is 6.46 inches, for the coast country 8.26 inches, and for central Texas 9.83 inches, while the total normal effective winter tempera- ture a during December to March, inclusive, is, for southwestern Texas 1,631.4° F., for the coast country 1,190.6° F., for eastern Texas 1,483.1° F., for northeastern Texas 934.9° F., and for central Texas 1,294.3° F. The average effective temperature for the eastern and southwestern countries is therefore from 200° to 300° higher than that of central Texas for this period and from 450° to 700° F. higher than that of the coast and northeastern Texas. If we consider the more crit- ical period of December to February, inclusive, the southwestern and eastern countries have 300° to 200° F. more effective temperature than central Texas and from 600° to 500° F. more than northeastern Texas. It would therefore seem that both the temperature and rain- fall are far more favorable to the successful hibernation of the weevil in southwestern Texas than in central, coast, or northeastern Texas, and that the conditions as to temperature and rainfall in the coast country during the critical period are very similar. In northeastern Texas, however, as the weevils do not emerge from hibernation until con- siderably later in the spring, the larger rainfall occurring during April may have some effect on them. As before mentioned, floods during the period of hibernation seem to have a marked effect upon the mortality of the weevil, as there were notably fewer weevils thruout the Brazos bottom in this section in the spring following the flood of February, 1903. PLACES OF HIBERNATION. Considerable time has been spent in attempting to find the places in which the weevils spend the winter, but the results have been rather unsatisfactory. This has also been the experience of many observing practical farmers. Numerous cotton fields were examined in the neighborhood of the college during January and February, 1903, and tho Mr. Newell and the writer made diligent search in all conceivable places over a considerable acreage, embracing an area of 2 miles, ex- amining cornstalks, grass, bark, fence posts, gins, etc., we found but 1 live boll weevil during the winter and this one was under a small pile of dried manure. On -November 29, 1903, Mr. A. F. Conradi found 2 weevils in a pile of old cotton brush in our experimental field, 3 under the bark of a log near the field, and 2 in the cotton field under a bunch of dead cotton leaves. On December 15 a thoro examination a Total normal effective temperature is reckoned for each month by subtracting 43° F. (the point at which animal activity is supposed to begin) from the normal mean temperature for that month, multiplying the difference by the number of days in that month, and adding together the four products thus obtained. 18 PAPERS ON THE COTTON BOLL WEEVIL, ETC. was made of logs, leaf rubbish, etc., in a ravine adjoining our cotton field, but no weevils were found. On December 28 a similarly futile search was made along Carters Creek, in which neighborhood a good deal of cotton had been grown. On January 5, 1904, a 30-acre cotton field was thoroly examined, but no weevils were found. Corn stubble standing near a very badly infested field was also examined without finding any weevils. Our information regarding the hibernating places of the weevil is therefore more of a deduction based upon obser- vations of the places in which they appear first in the spring than a conclusion from actual observations. The observations by Mr. Con- radi in the early winter probably indicate the normal places for hiber- nation — that is, under dead leaves, in old cotton brush, and under loose bark. In the hibernation cages, where the weevils were furnished an abundance of rubbish, it was found that many of them which were hibernating successfully had crawled into the cavities made by borers in dead wood and in similar positions where they were well protected. It has been often noticed that in a wooded country the weevils appear first in spring along the borders of fields next to the woods and gradu- ally work inward from the edges, so that it seems probable that in a wooded country most of them hibernate in woodland. Around out- buildings and barns also are found favorable places, as there is always more or less rubbish and protection in such situations. In 1903 more than five times as many weevils were found in a piece of cotton near the college barn, where cotton had been grown the previous year, than were found in any other locality in that neighborhood. It is also noticeable that weevils are always more numerous near gins than at a distance from them. Undoubtedly, where much rubbish and grass are present and where the soil remains loose and is not packed by rains, large numbers of the weevils winter in the cotton fields. The fact that in 1903 an exceedingly small number of weevils survived on Smith & Carson’s plantation on the Brazos River must have been due to their having hibernated in the field, large numbers having been killed otf by the excessive rains and doubtless more by the flood of February, 1903, which covered the fields to a depth of several inches over a large part of the plantation. It was quite noticeable that few weevils were found on Brazos bottom plantations in this section in 1903. In all probability the flood would have had but little effect later in the season after the weevils had emerged. It is noticeable that weevils are much more abundant where cotton is planted in fields where sorghum stubble has been allowed to remain all winter adjoining a last year’s cotton field. Thus, in the spring of 1904, the first weevils found on Smith & Carson’s plantation were on a cut of cotton along the Brazos River which had been in sorghum and surrounded by cotton in 1902 and where several large sorghum stacks stood thru the winter. Tho planted early and kept well cultivated, HIBERNATION AND DEVELOPMENT OF BOLL WEEVIL. 19 so little cotton was made on this piece that it was not worth picking, while cotton planted at the same time and cultivated in the same man- ner on other parts of the plantation made a good yield. Professor Mally (1. c., p. 57) has given the observations of Mr. Teltschick upon finding weevils hibernating in the crevices of the soil around the cotton stalks and roots, at a depth of 3 inches. On March 7, 1901, a raw, windy day, upon 35 stalks, he found 7 live and 2 dead weevils from 1 to 3 inches below the surface. In September, 1902, he stated that he had again found weevils in a similar situation during the previous spring, but not as many of them as in 1901. Mr. Telt- schick recently writes as follows: I found but few weevils in crevices around stalks during the last two winters, partly because there were no crevices (frequent rains filling them up as soon as formed) and partly because freezes were severe enough to keep cotton from coming out during any part of the last two winters; whereas in 1900 we had neither rain enough to till up crevices nor frost enough to keep cotton from budding out at inter- vals at the base of the stalk, which latter fact accounts, no doubt, for the relatively large number of weevils found within the crevices. TIME OF EMERGENCE FROM HIBERNATION. Hunter and Hinds a state that the first weevils emerge when the mean average temperature has been for some time above 60°. Our observations at the college upon the earliest emergence from hiberna- tion are as follows: In 1903 the first weevil was found on May 15 at the college after carefully searching about 100 feet of unchopped row. Repeated examinations during the next two weeks showed but a very small number of weevils. On May 8 a weevil was found in the Brazos bottom after hunting about 20 minutes, and after May 5 planters in the bottom occasionally reported finding weevils. In 1904, on cotton planted March 17, the first weevils were found, on March 29. On April 17 and 18 only an occasional weevil could be found at College Station or in the Brazos bottom after a considerable search. In 1903 Mr. Teltschick wrote me on March 1 from Lavaca County as follows: Up to a severe freeze of two weeks ago (February 16-17) I could at any time find some — not many — in my neighbors’ fields apparently feeding on buds or small leaves at the base of the stalk, which had never been entirely frozen. However, he did not find many on his own cotton until April, when he picked some from trap rows which were planted on March 13. In 1904 he noticed the first weevils about the middle of March, but, as in previous years in other localities in southern Texas, we have records of their feeding on shoots from stumpage cotton much earlier than this. In general, it seems that when the normal mean temperature «Bul. 45, Div. Ent., U. S. Dept. Agric., p. 82. 20 PAPERS ON THE COTTON BOLL WEEVIL, ETC. passes 68° F. the first weevils commence to emerge from hibernation, and this conclusion is confirmed by all of our observations. It will be seen from the temperature curves in figure 2 that at Victoria the normal mean temperature usually reaches 68° about April 1 and at College about April 15. Figure 2 shows that in 1903 the mean average temperature reached 68 about one week later at Victoria. In 1904 (fig. 2) the mean temperature reached 68° about March 10 at Victoria and the first weevils were found about the middle of March, I which is also true of Lavaca County, while at College Station 68° was reached a week later than usual, but the t a mperature was above 66° for the month preceding April 15. If the mean rise in temperature ; from February 15 to May 15 be plotted, the line would reach 68° about April 5, which very closely approximates the date of the first Fig. 6. — Comparison of normal temperatures, October 15 to June 15, for various districts'in Texas. appearance of the weevils at College Station for that year. From all the data available, therefore, it seems very safe to assert that when the temperature reaches 68° or has been above 65° for several weeks i the first weevils will emerge. By consulting figure 6 this time nmj' ; be seen for any section of the State. The curve shows that weevils l would normally emerge from hibernation about 10 da\ r s earlier in the I southwestern and coast countries than in central and northeastern Texas, i and about 20 days earlier than in northwestern or western Texas. The greatest number of weevils emerge considerably later and usu- I ally about the time cotton commences to square. The time of the principal emergence and the beginning of general oviposition by the hibernating brood is practically the same. A few belated weevils appear after this time, as is shown by the field records of Mr. Telt-I schick for 1903 and 1904. For if the bulk of the weevils emerges at the HIBERNATION AND DEVELOPMENT OF BOLL WEEVIL. 21 time the cotton commences to square freely and frequent counts be made, the number will be found to decrease for a few days, and there is then a sudden rise due to the maturing of the first brood early in June, and it is practically impossible to determine in the field whether any hibernated weevils subsequently emerge. However, in 1904, it is certain that no weevils emerged later than June 29, for at that date practically no weevils could be found in the field, tho the bulk of the hibernating weevils appeared from June 5 to June 10. Consequently this year (1904) certainly no weevils emerged later than 22 days after the maximum number. It has previously been shown by Professor Mally (1. c., p. 16) and by Hunter and Hinds, and noted above, that the weevils emerge grad- ually over a period of from 4 to 6 weeks, normally more nearty the latter time. This range covers practically the same length of time as from planting to the formation of the first squares on the bulk of the cotton in any given community. The gradual emergence from hiber- nation is shown by the number of weevils to 100 stalks in Table III for 1904 at College Station. In 1903 Mr. Teltschick picked the wee- vils from three-fourths of an acre of trap rows, planted March 13, on which the first bloom appeared May 28, and secured weevils as follows: April, about 40 weevils. May 4, 36 weevils picked in 1 hour’s time. May 11, 43 weevils picked in 1 hour’s time. May 18, 68 weevils picked in 2 hours’ time. May 25, 54 weevils picked in 4 hours’ time. In 1904 Mr. Teltschick made observations especially to determine this point. His notes are as follows: Two plats of one-fourth of an acre each w T ere selected in that part of the field which was farthest away from my own as also from any other cotton patch. The two plats themselves were separated by 25 rows of corn and surrounded by sugar cane on two sides and by timbered land on the other two. No cotton had been grown on or near these plats in 3 years. Plat 1 was planted in King cotton March 4, w r as up to a good stand March 14, and chopped to a stand April 11. From plat 1 the weevils were picked as follows: April 1, 14 weevils. May 2, 96 weevils. April 11, 27 weevils. May 12, 140 weevils. April 21, 26 weevils. May 23, 150 weevils . a Plat 2 was planted in Shine cotton April 25 (7 weeks later than plat 1), was up to a good stand April 30, and was chopped to a stand May 12 (one month later than plat 1 ). On plat 1 the first squares appeared May 2 and on plat 2 May 28. All fallen and flared squares were picked up on both plats at intervals of 8 days up to July 1. From plat 2 there were picked, on May 12, 3 weevils; May 23, 35 weevils. These records show very clearly the gradual emergence of the wee- vils from winter quarters. In 1903 the maximum number of weevils ^Weevils were picked also during June, but these doubtless belonged to the first summer brood. 22 PAPERS ON THE COTTON ROLL WEEVIL, ETC. appeared on May 18, in 1904 on Ma}^ 23, and in 1901 on Ma} 7 10-15; but as the intervals between pickings were a week or 10 days it would seem safe to judge that the normal time for the maximum number to emerge at Vienna is about May 15 to 20. It will also be noticed that the weevils appear much more rapidly and early on early planted cotton, but the maximum number does not appear at the time of squaring of the earliest planted cotton, but at about the time of squar- ing of the main crop, thus giving a decided advantage to the early planted cotton. The determination of the exact time of the emergence of the maximum number of weevils will be practically the same as that for the time of maximum oviposition for the hibernating brood, concerning which see page 25, and which in Lavaca County will ror mally occur about May 20, at College Station about June 1, and in northern Texas probably about June 10. PREVENTION OF HIBERNATION. It may be readily seen from the foregoing discussion that any methods which will reduce the number of weevils going into hiberna- tion in the fall are of the utmost importance in the control of the pest, inasmuch as nature aids in further reducing their numbers dur- ing the winter. All students of the problem of boll-weevil control have agreed that in this prevention of successful hibernation lies the most effectual means of reducing the numbers of the weevil and pre- venting injury the next season. Two methods are advocated: (1) The destruction of the stalks in the fall as long before frost as possible, and (2) rotation of the cotton crop, planting on land not in cotton the previous year and as far from such land as is possible. FALL DESTRUCTION OF STALKS. By the destruction of the stalks as early as possible in the fall the number of weevils entering hibernation is reduced in several wa 3 7 s:| (1) All immature stages in the squares and bolls, which might develop into weevils if the stalks remained standing, will be destined. (2)1 Probably three-fourths of the adult weevils in the field at the time of burning may be destroyed if the work be properly managed. (3) By I clearing the field of cotton some time before the normal time when weevils will enter hibernation their food is destroyed and the adults! either starve or migrate to other fields. Were this work done uni-jti formly over a considerable area, the weevils would surely starve,: as they can not be forced into hibernation before the normal time. That a large mortality follows if this is attempted is shown in our records (see pages 2-3). These advantages are apparent, tho but few actual records of the benefit of such fall destruction of stalks have been published. From the conditions under which experiments must be made it is difficult to demonstrate in a tangible way the actual HIBERNATION AND DEVELOPMENT OF BOLL WEEVIL. 23 benefit to be thus derived, but in the case of plats burned over in the fall of 1903 we found there were far fewer weevils in 1904 than in fields where the stalks had been left. On October 23 and 25 the stalks on our experimental plats, covering about 30 acres, were cut with a single cutter made of a V-shaped sled, with knives projecting out and backward on either side about a foot beyond the rows. A team draws the cutter with a man on it, cutting .2 rows at once and about 15 acres per day. The stalks .were then raked and piled, and were dry enough to burn in a week. On one- half the field 2 rows in every 20 were left standing as trap rows. No weevils were found later on the dried piles, but they had assembled in large numbers on the standing trap rows. Thirty-six were found on four plants. These trap row's were then cut, left over night on the ground, raked the next morning, and placed on the dried piles. That afternoon examination by the writer showed that on the green stalks on the dried piles there was an average of 1 weevil to every square and boll, often 5 or 6 being found on a square. The torch was then applied and the piles quickly burned. After this many weevils were found on the shoots growing out from the stumpage and an occasional stalk that was left. (This might be obviated more completely by plowing out the stalks.) A herd of cattle was therefore turned into the field and all herbage was devoured in a very few days. A better destruction could hardly be desired, and but little extra labor and expense was involved. At the apiary a small plat containing two-thirds of an acre was left with the stalks standing all winter. This plat was surrounded by a grove and was half a mile from other cotton. Observations by the writer show that in 1904 upon the 30 acres upon which the stalks were burned there appeared of the hibernated brood during the spring but 1.2 weevils to 100 stalks, while on the apiaiy plat where the stalks had been left there were about 4 weevils to 100 stalks, or over three times as many hibernated successfully where stalks remained standing as where they were destroyed. Again, on July 13 the field where stalks were burned averaged 3.3 weevils to 100 stalks, while fields of Messrs. Boyett and Gray, just over the railroad, not 100 yards distant, where the stalks had not been burned, tho other conditions were practically similar, had 6 and 12 weevils to 100 stalks, respectively, or an average of three times as many weevils as on the field where stalks were burned. Were all conditions exactly the same, as regards culture, land, etc., and were it possible to prevent the migration of the weevils from neighboring fields to those where the stalks had been burned, the lat- ter would undoubted^ show a much greater yield at the end of the season, but as it is very rarely possible to eliminate these factors the 24 PAPERS ON THE COTTON BOLL WEEVIL, ETC. above records show in a striking way, based upon extensive counts, the difference in amount of infestation during early summer where stalks have been destroyed the previous fall. CROP ROTATION. The value of the rotation of crops was strikingly shown by the dif- ference between the number of weevils occurring on the experiment station variety plats at the college barn, where cotton had been planted the previous year (1902), with corn adjoining, and our plats, situated a mile distant on land which had never been in cotton, having grown sorghum in 1902, tho cotton was grown not 100 yards distant. The cotton was planted April 5 at the college barn, and on June 17 there were 5 weevils to 100 stalks, 2 being found on the 40 stalks examined, whereas on the other field there were but 0.5 to 100 stalks, only 5 being found on 940 stalks examined. Again, on August 13 the varie- ties at the barn, planted May 3, had 233 weevils to 100 stalks, 163 being found on 70 stalks, while only 5.5 to 100 stalks occurred in the field where there had been no cotton the previous year, only 10 being found on 180 stalks from all parts of the field. On the same day 48 weevils to 100 stalks were found on Mr. James Boyett’s cotton, just across the railroad from our field, his land having been in cotton the previous year, while on Mr. Suber’s cotton, a mile distant, 53 to 60 per 100 stalks were found, this crop also following cotton . a WHERE THE WEEVILS FIRST APPEAR. It has been frequently observed in all parts of the State that the weevils first appear on the cotton field along the side adjoining wood- land, if there be any timber adjoining. From the extensive reports of observers thruout the State there seems to be no doubt that cotton in timbered sections is much more seriously injured by the weevil than that on prairie land, owing to the more successful hibernation in the timber. It has also been observed by many that the weevils first appear and commence to be injurious in that part of the field where they were most abundant during the previous year. In southern Texas the weevils may be found on stubble cotton very earty in the season; in fact practically all winter. They hardly seem to hibernate in the strict sense of the word, as under such conditions they are found feeding on warm days practically thruout the winter. Upon emerging from hibernation the majority of the early weevils seek the earliest cotton, which may be either stubble cotton or planted, and they will be found more abundant^ upon it than elsewhere, even tho the early cotton forms but a very small part of the acreage on a a The same value of rotation has been noted by Mr. Louis Teltschick at Vienna. (See Mally, Report on Boll Weevil, 1902, p. 66, paragraph 4.) HIBERNATION AND DEVELOPMENT OF BOLL WEEVIL. 25 large plantation, and altho later cotton may be up, but still small. Having assembled upon the early cotton, most of these weevils seem to stay upon it until it squares, and they do not spread generally to the later cotton until it also commences to square. This habit must be taken into consideration in estimating the number of weevils upon a given area in the spring, for if the cotton be not planted at the same time over the entire place, neither the earliest cotton nor that planted later will give a fair idea of the number of weevils which have hiber- nated successfully. The total area must be taken into consideration and the average for the whole place estimated by duly weighting the number found upon the acreage of both early and late planting. By the time the majority of hibernated weevils emerge, however, the medium and late-planted cotton will be so far advanced as to hold the weevils near where they emerge, and no further concentration may be expected. SUMMER BROODS OF THE WEEVIL. The first summer brood of weevils from eggs deposited by the hibernated adults begins to emerge from squares during the last ten days in May in southern Texas, and farther north, in central and northern Texas, continues emerging during J une and the first ten days in July. Thus in Lavaca County there is a marked increase in the number of weevils found in the field after June 1 or sometimes during the last week of May. In 1901 Mr. Teltschick secured 2,136 weevils up to the third week of May on 10 acres* the maximum appearing during the second week of May; but during the last week of May he secured on the same area 2,114 weevils, or very nearly as many as during the whole spring, showing that the first new brood was then emerging. In 1903, up to June 1, on three-fourths of an acre of trap rows he had picked 241, with a maximum on May 18; but on June 1 he secured 368, and during the month of June 1,759, with the maxi- mum number on the 15th. In 1904, on the one-fourth-acre plot pre- viously mentioned he secured 453 up to June 1, the maximum number of 150 on May 23; but on June 4 250 were secured and 146 on June 18. Mr. Teltschick writes as follows: In my opinion very few weevils, if any, matured from punctured squares before June either in this or previous years in this section. Cotton planted here March 1 (this is as early as any is planted, the greater bulk never being planted before April 1) as a rule begins to form squares on April 25. It requires the squares at that time from two to three weeks to grow large enough to provide the food necessary for the development of the weevil. The squares are, of course, being punctured as soon as they form, and I doubt whether any eggs are deposited in them so long as they are too small to permit the weevil secreting itself within the involucre. At College Station the first fallen squares were found on June 17, 1904. The first weevils from these squares emerged J une 23 and the 26 PAPERS ON THE COTTON BOLL WEEVIL, ETC. last on J une 29. At Terrell, Tex. , the first brood commenced to emerge on June 30 and the last of these was secured on July 12. Altho the hibernating weevils may continue to oviposit for some time there seems to be but little overlapping in the appearance of the first and second generations of weevils. There is a period from the first to the middle of July thruout the greater part of the infested area when but little damage is reported from the weevil and when but few weevils are found. This is the period during which the second brood is developing within the squares. About the middle of July and during the next ten days the second brood of weevils appears and is generally reported. In some instances the injury by the second brood is serious, especially where large numbers have hibernated suc- cessfully, and after the second brood appears practically all the squares become punctured and no more cotton is made. This condition is more common in southern Texas, but usually the injury by the second brood is no more serious than that due to the first in late J une, and in a great many cases it is reported as being decidedly less. In general, unless there be exceptional rains, the hot suns of July seem to check the development and injury of the weevil. The distinction betw een the second and third broods is not so marked. The period of oviposition of the second brood doubtless being longer than that of the first, the third brood commences to appear before the last of the second brood has emerged. However, there is still a very decided and notable increase in the number of weevils from the first to the middle of August,, depending upon the latitude and season. This is generally noticed, and commonly reported as the third brood. Usually with the emergence of this brood the weevils have become sufficiently numerous to destro}^ all the squares as they form so that no more blooms are seen. From this time the weevils attack the maturing bolls much more frequently. After the appearance of the third brood the different generations become so involved that it is impossible to distinguish between them. At Terrell, Tex., five distinct broods were reared in 1904 up to October 3, and these weevils then commenced to oviposit, so that there would doubtless have been a maximum of six complete broods in that localit}^. The records at Terrell are as follows: From the first fallen squares injured by the hibernated brood, weevils emerged on June 30, Jul}- 5, and July 12. On July 5 these weevils were placed in a cage upon fresh squares, and the weevils developing from these I squares emerged on July 22, or in from 17 to 22 days. These were 1 placed on squares on July 23, and the weevils from them, or the third 0 brood, emerged August 17 to 20, or in 25 to 28 days. These weevils | were placed on squares August 20, and the weevils developing from H them, or the fourth brood, emerged September 8 and 9, or in 19 to 20 j' days. These were placed on squares September 10, and the weevils I HIBERNATION AND DEVELOPMENT OF BOLL WEEVIL. 27 developing from them, or the fifth brood, emerged October 3, or in 23 days. In all cases this period of development includes to a greater or less extent the feeding period, which must precede oviposition. Practically the same number of generations seems to occur at Col- lege Station. Altho we have not determined this by rearing weevils as rapidly as possible thruout the season, the field records show the first three broods very distinctly, and during the fall there seem to be three more broods between September 1 and the time of hibernation. The number of broods after September 1 is very largely controlled by the temperature and rainfall, which combine to affect the food sup- ply and the mortality of the developing weevils. It is probable, how- ever, that in a late season seven broods might develop at College Station, as the field records (see page 10, Table II, for 1903) show the weevils (probabl 3 T of the sixth maximum brood) to have been freely ovipositing early in November, and if frosts did not occur until late in that month a large number of these would probably emerge. In southern Texas, Hunter and Hinds have shown that with the average season of reproductive activity, extending from May 1 to November 15, eight generations may be produced during the year. Thus there is probably a difference of two broods in the maximum number of generations between southern Texas and northern Texas, but this difference will not materially affect either the injury by the weevil or the number of weevils going into hibernation, as the weevils seem to reproduce in practically the same length of time in northern Texas as in southern Texas, the summer temperature being nearly the same, and after the third brood they are sufficiently numerous in either section to prevent the further fruiting of the crop. Nevertheless, factors aside from a difference in the number of generations will prob- ably cause less damage in northern Texas. By September 1 to 15 the weevils usually reach the maximum num- ber and during the fall, as before observed, their number depends primarily on the food supply and weather conditions, so that the num- ber going into hibernation depends on local conditions. The difference in the amount of injury which the weevil may do in central or northern and in southern Texas will depend therefore, first, upon the smaller number of weevils hibernating successfully and, second, upon their appearing later in the spring so that the cotton may be able to make a more rapid growth in comparison with the development of the early broods of the weevil under the high temperatures prevailing during early summer in northern Texas than will be the case farther south in late May. But it is not probable that the mere number of broods will make any material difference in the amount of injury. The first three broods seem to be well defined, as is clearly shown by the numerous reports of our voluntary observers and by the reports of the observers of the crop service of the U. S. Weather Bureau. The amount of 28 PAPERS ON THE COTTON BOLL WEEVIL, ETC. injury done depends upon the abundance of these broods, which depends largely upon the time of appearance and also upon the relative abundance of the hibernating brood. MORTALITY OF SUMMER BROODS. The mortality occurring in summer and fall broods may properly be distinguished from that among weevils which hibernate, for in the summer the greatest mortality occurs in the immature stages. The number of weevils developing from squares in which eggs have been laid is shown in Table IV. Table IV . — Percentage of weevils from infested, squares. Date. Where picked and how kept. Number of squares. Per cent emerged. Weighted per cent emerged. 11,250 23.5 23.5 1,522 39 675 30 350 34 377 23 2,924 34.3 14, 174 25.7 162 40 40 200 39.5 43 30 243 37.8 405 38.7 400 9.0 9.0 210 2.3 68 2.9 354 0 91 1.1 1.1 1,123 3.9 May —1902 July 8, 1904 Sept. 10, 1903 Sept. — , 1903 Sept. 21, 1903 Total of Total of Sept. — , 1903 Sept. 21,1903 do Total of Total of July 15, 1904 Sept. 11,1903 Sept. 22, 1903 Sept. 11,1903 Sept. 21,1903 Picked from ground; placed in cloth bags and kept in laboratory under dry condi- tions Picked from ground; kept under fairly normal conditions do ,... .do do lots 2, 3, 4, and 5 ... lots 1, 2, 3, 4, and 5 . Picked from plants; « kept on moist soil Picked from ground; kept between shaded rows do lots 7 and 8 lots 6, 7, and 8. Picked from ground, under screen; placed in part sun Picked from ground; placed in sun Picked from plants; n placed in sun Picked from ground; placed in sun Picked from plants; a placed in sun Total of lots 9, 10, 11, 12,13. a Picked from plants when squares would drop off by touching them. Lot 1 is based upon the following note made by Mr. Wilmon Newell. During May, 1902, Professor Mally gathered infested squares i in fields along the Brazos River below Wellborn. About 1^ bushels || of these squares were brought to College Station, sewn up in cheese : cloth sacks, and weevils allowed to hatch. June 15 the weevils were picked out and by careful measurement there were 115 c. c. of them, and 15 c. c. contained 315 weevils, or 1,763 weevils to a bushel. Subsequently ;! we have found about 7,500 squares in a bushel, thus indicating that only about 23.5 per cent produced weevils. The squares were kept in the laboratory in a dry place, and the lack of moisture may proba- bly account for the low percentage producing adults. In another test 2,921 squares gave a weighted average of 31 per cent producing weevils. These were picked up and kept under fairly normal condi- HIBERNATION AND DEVELOPMENT OF BOLL WEEVIL. 29 tions. From 405 squares which were kept on moist soil or between shaded rows weevils were developed from 38.6 percent, but where the squares were left in the hot sun, among 1,123 squares but 1 per cent produced weevils. This shows strikingly the value of having rows wide apart so that the sun may strike the fallen squares. Table V. — Percentage of mortality of weevils at different stages of development. Lot. Date. Picked from— Condition. Number of squares. Dead. Alive. Weevils | emerged. | Possible to emerge. Squares dry. 1 8* Y c3 ft 8 ft ft j Weevils. Total. Larvse. 8 0 ft | Weevils. o H 1 Aug. 12 133 36.3 1.5 2.2 28 4 24 56 2 July 14 do 572 44 2 3 23 11 17 51 1904. 3 July 8 do . . . 1,522 57.6 3.4 ... 39 39 4 Sept. 11 do In shaded 200 60 0.5 39.5 39.5 rows. 5 Sept. 22 do ... do 43 69 7 2 21 30 6 Sept. 21 do ... In sun 210 64 32 1 0.6 1.4 0.5 2. 5 7 do do . . . do 68 92.8 1.4 4.4 1.4 7.2 8 ... .do . . . do . . . Squares 100 61 18 3 2 1 3 1 10 15 brown. 9 do ... do ... Squares 268 65 5.61.5 0.7 11 9 1.8 21.8 4.8 26.6 green. 10 Sept. 11 Plants . . do. a .. 400 59.5 1.8,0. 5 «6. 8 1 30.5 31.5 Lots 6-10 , weighted 1 average... 1,046 63 10 i 3 77 5.4 3.1 0.5 9 14.8 Average of sou ares in whieh 26 3 8 37 14 10 1 25 39 some si ;age is fou nd. 1904. 11 Oct. 5 Plants a . Squares 100 44 44 55 1 56 0 green. Remarks. Counted just from field. Do. Allowed to emerge. Do. Partly al- lowed to emerge. Do. Do. Counted just from field. Do. Allowed to emerge. Seven died in emerging. Counted just from field. a Squares picked from plants when just ready to drop. The count made on October 5, 1904, fully confirms this. These squares were picked from the plants, taking those that would drop into the hand when touched, as in previous similar work. An imme- diate examination showed that 55 per cent contained live larvse of various sizes, but mostly about half grown. Tho a few were ready to pupate, but one pupa was found, while there were 14 in which the work of young larvae was clearly recognizable but in which no larva could be found. All squares showed egg punctures, but no eggs could be found in them. The time at which the greatest mortality in the development of the weevil occurs is a matter of some practical importance and the data given in Table V throw considerable light upon this point. We have but few observations as to the mortality in the egg stage, but Hunter and Hinds state that but few eggs fail to hatch, which we have found to be the case in the laboratory. As the square receiving an egg does 30 PAPERS ON THE COTTON BOLL WEEVIL, ETC. not fall for about 10 days and remains green long after the hatching of the egg, which takes place in about 2.5 days, there seems no reason to suspect any considerable mortality at this period. After the infested squares have dried or decayed it becomes very difficult to find any trace of larvae which die before they are half or two-thirds grown. The table shows that in about 63 per cent of the infested squares no trace of weevil stage could be found. In this percentage are the squares which had received feeding punctures only and a large num- ber showing external signs of egg punctures, thus indicating that a great mortality occurs while the larvae are less than half grown. Fur- thermore, it will be observed that of those individuals which pass the larval stage succesfully 1.6 per cent died in the squares or while emerg- ing. It has previously been shown that only about 1 per cent of squares placed where fully exposed to sunshine developed weevils, j It would therefore appear that the mortality was largely due to the 1 squares becoming so dried that the food supply was cut off from the young larva. Hence, it would seem that the sooner the square falls ! after being punctured, the higher will be the mortality, provided the ground is not shaded by close rows, and that any means for knocking off the squares before they would naturally fall, or a variety or strain of cotton which sheds its squares more quickly after being punctured, will be of decided advantage in increasing the mortality of the weevil. This seems to be even more marked where the squares are covered by ; the soil as is shown in Table VI. Table VI. — Percentage of mortality of weevils in squares buried in soil. 3 Date. Depth buried. Condition. Number of squares. Per cent dead. Per cent alive. a c s- a P- Squares dry or de- cayed. 3 A j Pupae. Weevils in square. | Weevils in soil. Total. [ Larvae., a l p Ph Weevils in squares. From squares. 1 Aug 15 5 inch Kept moist in laboratory. a 25 2 do 1 inch do << 25 3 do Is inches. . do a 25 4 Aug. 29 1 inch do a 25 72 4 4 24 5 do 1*- inches.. do a 25 28 4 12 44 38 6 Sept. 21 11 inches.. In field 50 74 2 76 8 7 do 2 j inches.. Rather dry 78 75 4 9 b 88 4 1 8 4 inches do 200 clOO A ...!.do 1 ( )n surface In sun 180 85 15 0 B Sept. 18 do 174 62 26 12 0 c July 15, 1901. do .... Under screen 400 91 9 1 .... .do 1 inch Rather wet 400 94. 5 3 10 do 2 inches . . do ion 81 10 11 do . 3 inehes . . .do 400 92. 5 12 do 4 inches .. do 400 95 3 0.5 a Contained larvae or pupae when buried, ft Dead weevils eaten by ants; squares green. o Ants’ nest just beneath. HIBERNATION AND DEVELOPMENT OF BOLL WEEVIL. 31 It is frequently asserted that frequent cultivation is of value against the weevil in that the immature stages are killed covering the squares with soil. The data given in Table VI hardly confirm this. Rarely will squares be buried over 1 or 2 inches. Lots 1, 2, 3, 4, 5, 9, and 10 show that the mortality in squares buried at this depth is about the same as under normal conditions. Lots 6 and T seem to show a benefit from burial, but these were placed in rather dry soil in the field in open sun, and when compared with lots A and B on the sur- face it is seen that the burying insured a greater per cent developing. It seems safe to say that in no case will burial for not over 2 inches result in any great mortality, and that in all probability the percent- age of weevils developing will be greater than if left exposed to the sun on the surface. It is more frequently claimed that by plowing squares under deeply after a good rain the moist soil will rot them and the weevils will be unable to escape. Unfortunately we have no conclusive tests upon this point, but it would seem that a suffi- ciently deep plowing to bury squares 4 inches just after a rain would be such a decidedly bad agricultural practise as to possibly offset any benefit that might be derived from destroying the weevils in the squares buried. RATE OF INCREASE OF THE WEEVIL. Hunter and Hinds have shown that a female weevil normally lays some 150 eggs during an average of about 54 days, and that nearly one-half of the eggs are deposited during the first third of the period of oviposition. Allowing 24 days for the development of the average adult and 18 days for the oviposition of one-half of the eggs, they also estimate that the average length of a generation is about 42 days. The sexes of the weevil are produced in about equal numbers. With these factors as a basis we may easily compute that if on June 1 there were an average of 2 weevils to 100 stalks of cotton, on July 15, at the end of the first theoretical brood, there would be 50 weevils; and that, if one-half of these were females and the usual percentage survived, on September 1 there would be 1,250 weevils to 100 stalks. But these computations make no allowance for the* mortality in the immature stages, and extended observations in the field show that such rapid multi- plication does not really occur. It has been ascertained by the writer that the first brood of weevils rarely numbers more than five times the hibernated brood, and at College Station often but two or three times. At Vienna, in 1903, Mr. Teltschick picked 241 weevils from his trap rows up to June 1, while on June 1 to 7, 15, and 22 he picked 759 weevils on the same area. This doubtless represents very accurately the increase of the first brood over the hibernated weevils, showing an increase of 315 per cent. It is of course probable that some weevils had left the trap rows for the planted cotton, which was barely commencing to square 32 PAPERS ON THE COTTON BOLL WEEVIL, ETC. on June 22, but the. number would have been comparatively small and the end of the generation was probably considerably later, so that an increase of ten times, or 1,000 per cent, would be a conserva- tive estimate. The largest number of weevils at College Station early in July was found on EckoPs cotton, which was planted very early, on July 14, 1904, when there were 27 weevils to 100 stalks. This field was adjoining a gin and 1 mile away, on Ayer’s cotton, there were but 15 weevils to 100 stalks. On the same date, on the Brazos River, there were but 10 weevils to 100 stalks, and on our experimental plats, where the stalks were destroyed the previous fall, the hibernated brood had numbered barely 1 weevil to 100 stalks. There were but 3.3 weevils of the first brood to 100 stalks. Accurate figures for the rate of increase from the first to the second brood are not available, but from general observations and from reports of the injury done by the second brood, as compared with the first, it is safe to assert that the rate of increase is not materially greater for the second brood than for the first. The rate of increase of the third brood seems to be decidedly greater, but, as the late-developing individuals of the second brood and the first of the third brood more or less overlap, it is diffi- cult to determine this accurately in the field. However, the total increase at the appearance of the third brood over the number of hibernated individuals can be given in one or two cases. Along the ! Brazos River, in 1903, there were but 2 weevils to 100 stalks on June 17, and on August 25 47 weevils to 100 stalks, or an increase of twenty-three times between these dates, the collection at the latter j date probably representing part of the fourth brood. In 1903, at the > college on our experimental plats, there were 5.5 weevils to 100 stalks ; on August 13 where there were 1.5 on June 18, representing the hibernated brood. The greatest increase we have observed from the hibernated brood until the middle of August was on the plat of the experiment station at the college barn in 1903, where there were 5 hibernated weevils to 100 stalks on June 18 and 233 on August 13, or an increase of forty-seven times. From a careful study of the data it seems safe to assert that the usual rate of increase from the second to the third brood is not over ten or fifteen times in this section, as there ! is usually not over 1 weevil to the stalk of cotton until after mid- August. It will be seen, therefore, that tho the theoretical increase of the weevil, based upon a careful study of its habits and the mor- tality as far as can be ascertained, would show an increase of twenty- five times from between the first and second and second and third theoretical broods of 42 days each from June 1 to September 1, as a matter of fact, during the first period mentioned, the increase up to and including the second brood, emerging about the third week in July, is usually considerably less than twenty-five times the hibernated brood probably not over fifteen times — and that the total increase HIBERNATION AND DEVELOPMENT OF BOLL WEEVIL. 33 from the 1st of June to the 1st of September is only about fifty times, certainly not over sixty-five times, where theoretically it would be six hundred and twenty-five times. The reason for this discrepancy is unknown to the writer, but is a fact for which the cotton planter should be decidedly thankful. It may be that the mortality of the immature stages of the weevil in the squares is greater than the count of thousands of squares has shown, that many of the adult weevils die or are destroyed before reproducing, or that the number of eggs laid and the length of the period of oviposition in the field are much less than those observed in the laboratory. That this is true of the hiber- nated brood is shown on page 12. It would seem that the discrepancy is mostly due to the two latter factors. This discrepancy but empha- sizes the necessity for accurate!}" observing the field conditions on a very large scale in different fields in different sections of the State constantly thruout the season if we are to obtain accurate knowledge of the true habits, rate of increase, and injury of the boll weevil, as it occurs in the field. To summarize briefly, it would seem that if there be 2 weevils to 100 stalks of the hibernated brood, then by early in August there will be a sufficient number of weevils to stop the blooming of the cotton. Furthermore, with the normal increase it seems that in about a month after 4 or 5 weevils are found on 100 stalks, there will have been suffi- cient increase to prevent further fruiting of the cotton, and there will then be from 75 to 100 weevils to 100 stalks. The utter impossibility of making a cotton crop where a large percentage of the weevils sur- vive the winter will therefore readily be seen unless exceptional meas- ures can be taken to check their multiplication. Weather conditions unfavorable to the development of the weevils are decidedly the most important controlling factors, and that such conditions not infrequently occur is shown by the reports of voluntary observers in southern Texas in 1904. INJURY TO SQUARES IN RELATION TO THE NATURAL INCREASE OF SQUARES. The formation of squares upon the cotton plant depends upon so many different factors that it is exceedingly difficult to make any gen- eral statements regarding the increase in their numbers. Thus the character of the soil, the nature of the variety of cotton, and the cli- matic conditions prevailing all influence the formation of squares. It may be clearly seen from Tables VII to X that some varieties form squares much faster than do others, and that between the same dates the same variety will not produce the same number of squares in dif- ferent places. However, from an examination of the tables we may form some idea of the natural increase of the squares early in the season during the critical period of fruit formation. During the 2 34 PAPERS ON THE COTTON BOLL WEEVIL, ETC. weeks following the formation of the first 2 squares the number of squares will increase ten or twelve times; in other words, there will be from 20 to 25 squares; and during the next 2 weeks this number will increase two times, or at the end of the month there will be about 40 squares, or about twenty times as manias at the beginning of the month, when there were but 2 squares. At the end of the second month there will be an increase of about two times over the first month, or there will be approximately 100 squares to the plant. During the first third of the third month the number of squares usually remains about constant, but after that, tho squares are con- stantly forming, there is a tendency to decrease for a time. Table VII. — Progress of infestation by boll weevil and production of fruit upon cotton plats at College Station , Tex., in 1903. Variety, Jones Im- proved . Allen Excelsior Parker Griffith... King June 18 July 11 July 27 Aug. 12 Sept. 5 June 18 July 11 Aug. 12 June 18 July 11 Aug. 12 Sept. 5 June 18 July 11 Aug. 12 June 18 July 11 Aug. 12 June 18 July 11 July 27 Aug. 12 Sept. 5 Squares. 20 25 30 10 5 20 25 10 20 25 10 5 20 25 10 20 25 10 20 25 50 1047. 5,57 125 0.9 .7 9.1 56 35 15 6.3 6 21 50 97 Blooms. 13 12.7 51 Bolls. Wee- vils. 4.6 12 6 .5 .. 12 8 6 .24 8.5 30 Yield per acre. Pounds of seed cotton. Sept. 17 Sept. i7 Sept. 16 Sept, i 6 Sept. 17 Sept. 16 178 290 Oct. 20 Oct. 20 Oct. 20 Oct.’ *20 Oct. 20 358 HIBERNATION AND DEVELOPMENT OF BOLL WEEVIL, 35 Table VIII. — Comparison of 'progress of infestation by boll weevil and production of fruit upon variety plats on Carson & Smith’s plantation , Brazos River , Wellborn , Tex . , in 1903. Plat. Variety. 1. .. Native 2 Territory 3 Dickson. 4 Schley.. 5 Bohemian 3,4,5. King 8 Shine . 9 do 8-9... 8 8-9... ....do. 10.... Shine . 11.... do. 12.... do 13.... do 12-13. do. 10-13. do 14.... do 15 do 15-16 do 17 Wellborn . Date of examin- ation. June 26. July 17. . Aug. 17. Aug. 25. June 26. July 17. . Aug. 25. June 26. July 20.. Aug. 17. Aug. 25. June 26. July20. . Aug. 17. Aug. 25. June 26. July20. . Aug. 17. Aug. 25. June 26. July 20.. Aug. 25. June 26 — do . . . July 20 Aug. 25 June 26 do . . . do ... July 17 June 26 July 17 Aug. 17 June 26 ■July 17 -Aug. 25 July 20 Aug. 25 June 26 July 20 Aug. 25 a Squares per plant. 1 Bolls. Yield per acre. X 0) CO 'd 1 , d 00 M 1 Pounds seed cot- ton. I Number of stall ined. Number. | Punctured. o a d OP o Fh <0 Ph Blooms per plan Per plant. o 0) O <0 CL, 1 u P4 c n > 0) & a £ Number. 1 Punctured. Per cent punctured. OJ o. CO a o o s Per plant. Per cent I punctured. Fh P< 02 G) G> Date. First pick- ing. Date. Second pick- ing. Total. Bales. Mar 1. May 30. . July 3.. Aug. 5.. Aug. 28. July 3.. Aug. 6.. Aug. 28. May 30. . 140 3.0 15 40.0 4.0 10.0 195 195.0 84.2 43.1 2.2 9.6 37.5 140 Native Mar .. Av. 15 i9.0 3.06 i6. i 24.0 Sept. 26. 171 Dec. 1... 142 313 .21 5 80.0 33.0 41.2 2.0 8.0 20.0 20 King Apr. 1. Av. 5.6 6.0 Sept. 26. 210 Dec.l... 148 358 .24 July 3.. Aug. 5.. Aug. 28. July 3.. Aug. 5.. Aug. 28. July 3.. Aug. 5.. Aug. 28. July 3.. Aug. 5.. Aug. 28. July 3.. 50 40.0 3.8 9.5 5 57.0 31.6 56.4 1.8 12.0 20.0 140 Av. 19.0 Sept. 26. 181 Oct. 15. . 278 460 .30 Shinp Apr . . . 50 20.0 . 7 3.5 5 97.0 42.0 43.3 2.0 . 10.2 33.3 120 Native Apr. . . Av. 50 i2.0 i. 62 lL3.5 |13. 0 Sept. 26. 150 Dec.l... 120 270 .18 5 53.0 24.6 46.4 .6 5.8 27.5 80 Native Apr . . . Av. 50 23.6 3.34 1.4.5 5.0 Sept. 26. 173 Dec.l... 57 230 .15 5 42.0 19.6 46.6 1.0 12.4 24.1 140 Shine May .. Av. 100 . 9 11.0 Sept. 26. 366 Nov. 1.. 57 423 .28 Aug. 5. . Aug. 28. July 3.. Aug. 5.. Aug. 28. 5 62.0 18.2 29.3 .2 2.4 8.3 160 Av. 3.0 Dec. 1... 80 80 .05 Native May 1. 100 . 7 5 36.0 16.2 45.0 .2 .6 33.3 40 2.0 HIBERNATION AND DEVELOPMENT OF BOLL WEEVIL. 37 Table X. — Comparison of progress of infestation by boll weevil and production of fruit upon plats of cotton at College Station , Tex., in 1904. Yield per acre. d V* cc ; v ii o. 1 Pounds seed cotton. Variety. Date of examinatii Number of stalks. | Total per plant. | Total punctured. Per cent punc- tured. Dropped. ”3 o H Per plant. Perfect. Open. Total per plant. • Punctured. Total. Per 100 stalks. Date. j First picking. Date. Second picking. Total. Bales. May 22 May 30 June 17 2,500 2,500 400 2 0.08 O . 2 .08 5 5.5 36 1.6 5 S-S 320 3 . 9 m r< ' 400 400 2 .5 8* June 27 15 12 .2 0 174 0.43 55 0. 13 0 0 c ft July 13 Sept. 12 May 22 May 30 J une 9 50 39 .35 34 166 3.3 307 6.1 1 2 25 4.5 98 86 (a) 0 455 99 108 36 144 Aug.28 258 Sept. 14 ‘252 510 .32 bio Ph-2 2,500 2,500 320 1 . .04 10 .4 0 c >— 1 o3 3 .9 June 15 2,500 400 5 toft June 17 4.8 21 1.2 1 («) 0 p . June 27 380 22 14 .2 92 164 .43 91 .24! 0 July 13 Sept. 12 May 22 May 30 June 17 50 43 18 .83 38 178 3.8 309 6.2 2 4 25 6.5 142 86 <"> 491 23 *83 46 184 Aug.27 313 Sept. 10 348 661 • 4 js to 1,200 4 .3 1,200 50 14 1. 1 9.7 25 5.1. 7 0 (a) 2 0 ® §J June 9 250 .8 fift June 27 400 25 152 1.5 8 53 .13 0 0 3, Part V. C. B. W. I., February 0. 1007. PAPERS ON THE COTTON BOLL WEEVIL AND RELATED AND ASSOCIATED INSECTS. NOTES ON THE PEPPER WEEVIL. ( Anthonomus seneotinctus Champ.) By F. C. Pratt, Assistant. OCCURRENCE IN TEXAS. In a previous bulletin a Mr. C. M. Walker gives an account of the pepper weevil, which had been injuring peppers in Texas. Investiga- tion during the fall of 1905 resulted in proving the disappearance of this insect at Boerne, Tex., where it was first reported. Diligent search in October and November by the writer, who was assisted by Mr. L. Lamm, failed to show the existence of the weevils at that place, altho peppers had been grown as extensively as before. At San Antonio, Tex., many truckers had given up the growing of peppers on account of their experience the year previous, and on one patch at Collins Gardens fully 80 per cent of the pods were attacked, and several bushels of chilli and sweet peppers were shipped from there to Dallas for observation. Several places, Floresville, Seguin, and New Braunfels, in the vicinity of San Antonio, were visited to ascertain the extent of the infested territory, but without success. Corpus Christ! also was vis- ited, but no injury was noticed and no reports were obtained concern- ing the weevils. Mr. J. C. Crawford reported the presence of the weevil about 8 miles south of San Antonio, the grower having claimed to have noticed this insect for several years. At Cotulla, Tex., Mr. Henry Caley reported slight injury at one of his truck farms near that place, altho the insect was not seen by the writer. ORIGIN. That this insect (adult, fig. 10) has been introduced into the United States from Mexico in recent years, there is no doubt, the exact date at this time being undeterminable. Inquiries at San Antonio com- mission houses elicited the information that peppers were shipped from Laredo, Artesia, and Cotulla. As a matter of fact, few peppers are raised at these places, the principal crop being onions. Mr. Cale} 7 , at Cotulla, denied ever shipping peppers and was very skeptical about 55 a Bui. 54, Bur. Ent., U. S. Dept. Agrie., pp. 43-48, 1905. 56 PAPERS ON THE COTTON BOLL WEEVIL, ETC. any having- been shipped from there at any^ time. Further information showed that peppers were bought up in carload lots in Mexico and shipped to the United States and without much doubt rebilled from the localities mentioned to give an impression that these peppers were home grown. Dr. A. W. Morrill reported that a large number of peppers were shipped annually to the United States from the dis- tricts of Zacatecas and San Luis Fotosi, Mexico, and he had collected the weevil at Tlahualilo, Durango, Mexico. FOOD PLANTS. fig. io.— repper weevil (Anthou- j s evident that the weevil does not breed omus seneotinctus ): Adult, . Much enlarged (after Hunter in the wild or bird pepper, for these plants and Hinds). were examined thoroly at all the points visited as well as at Victoria, Tex. The bell or sweet pepper (PI. II, figs. 4-6), chilli (PI. II, figs. 1-3), and tabasco peppers of several varieties are always affected wherever the weevil occurs. NATURAL ENEMIES. As in the case of the boll weevil ( Anthonomus grandis Boh.), altho a few parasites were reared, these did not occur in sufficient numbers to keep the insect in check. Mr. J. C. Crawford has identified the parasitic species as Braconmellitor Say, and Catolaccus incertus Ashm. A small ant i Solenopsis geminata Fab., fl was noticed to be actively engaged in removing larvae and pupae from the pods which had weevil exit holes in them, but in no case did the ants make an entrance themselves. In the parasite breeding cages thousands of small mites were observed by Mr. W. W. Yothers, who reported that their presence in no way incommoded the weevils. To determine the average number of weevils that may infest the chilli peppers, several experiments were tried, with the following results: Experiment No. Number of pods. Number of weevils Period covered, emerged. 1 1 . 80 86 Oct. 18-Nov. 4. ; 2 230 156 Do. , 3 200 143 Oct. 18-Nov. 11. 4 100 123 Oct. 18-Oct. 31. Total . . 010 508 Thus it will be noticed that an average of nearly one adult to each pod was bred. "See Bui. 63, Pt. III, Bur. Ent., U. S. Dept. Agric. An Ant Enemy of the Cot- ton Boll Weevil. By W. E. Hinds. Bui. 63, Part V, Bureau of Entomology, U. S. Dept, of Agriculture. Plate II. Work of the Pepper Weevil 'Anthonomus /eneotinctus Champ). Fig. 1 — Emergence holes of adult in chilli pepper. Fig. 2. — Full-grown larva in situ in chilli pepper. Fig. 3.-— Pupse in cells in chilli pepper. Fig. 4.— Undeveloped bell pepper pod, showing numerous egg and feeding punctures. Figs. 5, 6. — Malformed bell pepper pods, caused by egg punctures. Figs. 1-3 three times natural size; fig. 4 twice natural size: figs. 5, 6 one-half natural size. (Original.) LIBRARY UNlVtRSlTY Of ILLINOIS URBAN* NOTES ON THE PEPPER WEEVIL. 57 The maximum numbers of individuals found in two pods were as follows: Pod. Adults. Pupae. Larvae. Total. 1 ... 2 5 2 9 2 ... 5 6 11 Many pods showed six egg punctures, from which live larvae, in one instance, developed. It was also noticed that in nearly every case where egg insertion was made beyond the middle of the pod the larva was to be found slightly back of the puncture, but where the egg was deposited between the stem and middle of the pod the larva was located immediately underneath. The average number of egg punctures to a pod proved to be about two and one-third. PROLIFERATION. « Proliferation was noticed in pods purchased in market at San Antonio, Tex., October 6, 1905, and the following two series of experi- ments were conducted by Dr. W. E. Hinds and Mr. W. W. Yothers: Experiments to determine the effects on the, pepper weevil of proliferation in the pepper pods. FIRSTr SEIZES. Oct. 14.. 12 11 91.7 0 13 10 9 6 1 0 1 5.0 1 8.3 0 0 0 0 0 0 3 45 Do... 120 111 92.5 0 125 102 19 34 1 1 0 0.7 9 7.51 2 o! o 0 0 0 ( c ) (c) Oct. 31.. 21 21 100.0 1 55 34 10 1 1 0 0 0.0 0 0 0 0 0 1 0 0 0 155' 9 153 143 93.5 1 193 146 38 4! 3 1 1 1.2 10 6. 5 2 0 0 “ o' °i 0 a See also Bui. 59, Bur. Ent., U. S. Dept. Agric., p. 38, 1906. 6 This is a record of the proliferated areas on inside of pods and is a minimum.— W. E. H. f Not recorded. 58 PAPERS ON THE COTTON BOLL WEEVIL, ETC. REMEDIES. In addition to the usual gathering and destrojdng of fallen pepper pods, cultural experiments were conducted in the'following manner: Some pods were buried in boxes under 1, 2, and 3 inches of soil respectively, keeping one series dry and the other moist — conditions which would result necessarily from irrigation. In the former series the pods dried up and in the latter the pods rotted. Experiments to determine the effect upon the pepper weevil of covering fallen inf ested pods with soil. Num- ber of pods. Number of punc- tures. Depth of soil. Condi- tion of soil. Period covered. Number of weevils emerging Oct. 19-Nov. 4. Number of weevils found dead in soil Nov. 17. 25 62 Inches. 1 Wet .... Oct. 19-Nov. 14 . . . i 1 adult; 1 6 adults. 25 61 1 ' Dry 1 do parasite. 9 adults _ 12 adults. 25 56 2 Wet .... do 4 adults« . . . 1 adult. 25 60 9 Dry do 1 3 adults. . . 5 adults. 50 115 3 do . . . j 15 adults 1 14 adults. a Three of these weevils died as soon as they came thru the soil. Thus it will be readily seen that the covering of the fallen pods every two to three weeks with from 1 to 3 inches of soil would be advan- tageous. This could be arranged by bedding high and then lowering the soil. Where irrigation is practised the decay of the pods is greatly hastened, thus depriving the larvte of this food supply. It is only fair to state that at the time the pods in these experiments were buried most of the larvae were nearly full fed, consequently a larger number came to maturity than would have been the case had burial occurred while the larvae were small. Two other experiments were conducted by Mr. W. W. Yothers in large cages in the open, with the following results: Experiments to determine the effect upon the pepper weevil of covering fallen infested pods with soil. Number of pods. Depth of soil. Condition of soil, i Date of exami- nation. Nu “, bcr Number of weevils weevils dead enmr^red.j n 80 Nov ‘ »*• Inches. 500 3 Dry« Oct. 20 64 ! None found. 450 2 Drv« 1 Oct. 28 • 50 1 Do. " Heavy rains October 24 and November 10. October 31 about 200 of the last-mentioned lot of pods w r ere examined and found to contain 73 living and 0 dead stages; of this number 47 weir pupae and 15 were larva*. The latter necessarily would have starved, owing to tin* rapid' decay of the buried pods. O U. S. DEPARTMENT OF AGRICULTURE BUREAU OF ENTOMOLOGY— BULLETIN NO. 63, PART VI. L. O. HOWARD, Entomologist and Chief of Bureau. PAPERS ON THE COTTON BOLL WEEVIL AND RELATED AND ASSOCIATED INSECTS. THE STRAWBERRY WEEVIL IN THE SOUTH-CENTRAL STATES IN 1905. BY A. W. MORRILL, Special Field Agent. Issued January 22 , 1007 . WASHINGTON: GOVERNMENT PRINTING OFFICE. 1907. LETTER OF TRANSMITTAL U. S. Department of Agriculture, Bureau of Entomology, Washington , D. (7., December i, 1906. Sir: I have the honor to transmit herewith a brief report on the status of the strawberry weevil in the south-central States in 1905, by Dr. A. W. Morrill, special field agent in this Bureau. As this straw- berry pest is related to the boll weevil and is extensively parasitized there is a possibility that some one or more of its parasites ma}^ be induced to attack the boll weevil. Its greater or less abundance in or near the region infested by the boll weevil is therefore a matter of importance. I recommend the publication of the report as Bulletin No. 63, Part VI, of this Bureau. Respectful I}-, L. O. Howard, Entomologist and Chief of Bureau. lion, el ames Wilson, Secretary of Agriculture. ii CONTENTS. Page. The strawberry weevil in Texas 59 The strawberry weevil in Louisiana 60 The strawberry weevil in Arkansas 61 Varieties of strawberries grown 62 in U. S. D. A., B. E. Bui. 63, Part VI. C. B. W. I., January 22, 1907. PAPERS ON THE COTTON BOLL WEEVIL AND RELATED AND ASSOCIATED INSECTS. THE STRAWBERRY WEEVIL IN THE SOUTH-CENTRAL STATES IN 1905. ( Anthonornus signatus Say. ) By A. W. Morrill, Special Field Agent. For some time the Bureau of Entomology has been experimenting along the line of inducing parasites of related native species of weevils to prey upon the cotton boll weevil ( Antlionomus grandis Boh.). One of the commonest and most parasitized of such weevils is the strawberry weevil (. Antlionomus signatus Say). It was conse- quently important to ascertain whether this insect occurred in or near the infested regions of Texas and Louisiana ,and to what extent it was controlled by parasites. In connection with this work, during the month of April, 1905, advantage was taken of an opportunity to observe the occurrence and injury of the strawberry weevil in the leading strawberry -producing sections of Texas, Louisiana, and Arkansas. Examinations were made in all cases in the blooming season when the insects would, if present, occur in their greatest abundance. In consideration of the destructiveness of this species in the strawberry fields of Maryland and North Carolina during the past few years, the data concerning its occurrence and nonoccurrence in the important strawberry-growing sections of other States are deemed of sufficient importance to be recorded. THE STRAWBERRY WEEVIL IN TEXAS. According to the latest available statistics (1899), a more than four- fifths of the strawberries produced in the State of Texas are grown in two limited areas comprizing three counties. Of these sections the « Twelfth Census of the United States, 1900. 59 60 PAPERS ON THE COTTON BOLL WEEVIL, ETC. one ranking first in importance includes Brazoria and Galveston coun- ties — two adjoining coast counties near the southeastern corner of the State. In this section examinations were made on April 7 and 8 at Alvin and Galvestofi, respectively, but without finding any indications of the presence of the strawberry weevil. The second largest straw- berry-growing section of Texas is included in Smith County, near the northeastern corner of the State. An examination was made at Tyler, near the center of this section, on April 13, and again on April 28. On the first of these dates a single adult specimen was taken, and on the second occasion two adult specimens. In a total of about three hours spent near Tyler in fields of strawberries, blackberries, and dewberries, searching for these weevils and for evidence of their work, only the three adults and less than fifty injured buds were found. The only published record known to the writer of the strawberry weevil’s occurrence in Texas was by Dr. F. H. Chittenden, a who reported injury by this species, in 1897, to blackberries and dewberries near Denison, in Grayson County, in the north-central section of the State. The fruit farms of the two correspondents who reported serious losses in that year were visited by the writer on April 21, 1905, and examina- tions made in fields of strawberries, blackberries, and dewberries. One of the owners reported that the weevils had been scarcely noticeable since April, 1897, when they were very destructive and so abundant that on sunshiny days they seemed to fairly swarm on the foliage of the food plants. This encouraging report concerning its scarcity in recent years gives ground for the hope that conditions in northern Texas will as a rule be unfavorable for this pest, altho a certain degree of inter- mittence in abundance during a series of years has been generally observed in other localities. THE STRAWBERRY WEEVIL IN LOUISIANA. In 1899, as shown by the Census report , b 90 per cent of the straw- berries produced in this State were grown in Tangipahoa Parish. This parish is located a short distance northwest of New Orleans. Hammond, one of the two leading shipping points in the parish, was visited on April 11. Examinations in strawberry fields and of wild blackberry bushes along the roadsides, in a downpour of rain, resulted in finding no evidence of the presence of the strawberry weevil neai* this place. As the day was so far from an ideal one for tinding adults of this species, the only significant point is the failure to find a single flower bud injured in a manner to cause their presence to be suspected. a Bui. 10, n. s., Div. Ent., U. S. Dept. Agric., pp 82-83. & Twelfth Census of the United States, 1900. STRAWBERRY WEEVIL IN SOUTH-CENTRAL STATES. 61 THE STRAWBERRY WEEVIL IN ARKANSAS. The strawberry-growing industry in Arkansas is a very important one, in 1899 the production of this State being nearty four times that of Texas and Louisiana combined; or, in round numbers, nearly 13,000,000 quarts. Three sections of the State were visited, including the two leading strawberry-growing districts which together produce more than four-fifths of all the strawberries grown in the State." The first in importance of these sections comprizes three adjoining counties in the northwest corner of the State-M^enton, Washington, and Crawford. Van Buren, located in Crawford Count}^, the southernmost of the three named, was visited on April 27. A single adult specimen of the strawberry weevil was taken on a wild blackberry growing by the roadside, but in the strawberry fields the only indications of its presence which were found were a few scattering severed flower buds. The second section in importance in strawberry production in Arkan- sas is located in White County, which is situated a short distance northeast of the geographical center of the State. This county in 1899" produced one and a half million quarts of strawberries, or a quantit} 7 a little short of equaling the total production of the State of Louisiana. Bald Knob and Judsonia, two leading strawberry shipping points in Smith Count} 7 ', were visited on April 25. Many strawberry fields were visited, and the weevil was found to be moderately abun- dant near these localities. The destruction in the many fields exam- ined was estimated to range from 5 to 25 per cent of the flower buds, averaging between 10 and 15 per cent. Along the roadsides a small percentage of the flower buds of wild blackberries had been wholly or partially severed. The weevil and its destructive habits were found to be fairly well known among the strawberry growers in this section, by whom the weevil was reported to have varied in abundance from year to } T ear, and in occasional years in the past to have accomplished con- siderable damage. The destruction of the flower buds, as mentioned above, did not necessarily represent a proportionate loss to the grow- ers, as picking had begun about ten days earlier, and berries produced from flowers which set at a later date than the writer’s examination very likely would remain unpicked owing to the expected falling off in the market demands. The third locality in Arkansas where an examination was made is near the center of the State, 1 mile from Little Rock. An hour’s search in a strawberry field on April 26 revealed a few severed buds, while not a hundred yards away wild blackberries, which were very common along the roadsides, had from 5 to 10 per cent of their flower buds destroyed. The only adult specimen of the strawberry weevil taken in this vicinity was on a blackberry bush. “Twelfth Census of the United States, 1900. 62 PAPERS ON THE COTTON BOLL WEEVIL, ETC. VARIETIES OF STRAWBERRIES GROWN. At Alvin, Tex., the favorite variety among the strawberry growers is the Klondike, while at the other localities mentioned individual preference seems to be divided between the Lady Thompson, Michel, and Excelsior varieties. In the section of the country to which these notes refer, the writer has never met any strawberry growers who use any but bisexual varieties. Arkansas growers, who probably suffer more or less loss every year from the strawberry weevil, should be encouraged to introduce suitable early pistillate varieties, as the insect in that State threatens to cause, occasionally at least, considerable destruction. O U. S. DEPARTMENT OF AGRICULTURE, BUREAU OF ENTOMOLOGY BULLETIN NO. 63, Part VII. L. O. HOWARD, Entomologist and Chief of Bureau. PAPERS ON THE COTTON BOLL WEEVIL AND RELATED AND ASSOCIATED INSECTS. THE COTTON STALE-BORER. BY A. C. MORGAN, Special Field Agent. Issued February 9 , 1907 . WASHINGTON: GOVERNMENT PRINTING OFFICE. 1907. LETTER OF TRANSMITTAL. U. S. Department of Agriculture, Bureau of Entomology, Washington , D. C., December i, 1906. Sir: I have the honor to transmit herewith a short account of the cotton stalk- borer, prepared by Mr. A. C. Morgan, special field agent in this Bureau. Altho this insect appears, as a rule, to attack cotton stalks which are already injured or diseased, thus causing onl\ slight damage to the crop, Mr. Morgan finds evidence that healthy stalks may be occasionally attacked. I recommend the publication of the paper as Part VII of Bulletin No. 63 of the Bureau of Entomology Respectfully, L. O. Howard, Entomologist and Chief of Bureau. Hon. James Wilson, Secretary of Agriculture. n CONTI: NTS. Page Distribution - 63 Host plants , 63 Life history . 64 Injury and economic importance 64 Natural enemies 65 Remedy 65 Bibliography 66 ILLUSTRATIONS. Page. Plate III. Fig. 1. — Adult or beetle. Fig. 2. — a, galleries; b, work. Fig. 3. — Emergence holes of adults. Fig. 4. — a, larva in situ; b, pupa in situ. Fig. 5. — a, gallery running into root of plant; b , castings behind larva ... 64 in U. S. D. A., B. E. Bui. 68, Part YU. C. P». W. I.. February 0, 1907. PAPERS ON THE COTTON BOLL WEEVIL RELATED AND ASSOCIATED INSECTS. THE COTTON STALK-BORER. ( Ataxia crypta, Say. ) By A. C. Morgan, Special Field Agent. During October, 1905, while inspecting cotton fields in the vicinity of Yoakum, Tex., the writer’s attention was attracted by the great number of dead stalks. Examination showed that these stalks con- tained larvae, pup*e, and adults of the cotton stalk-borer (. Ataxia crypta Say). DISTRIBUTION. The specimen which Say described in 1832 as Lamia crypta came from Louisiana. Haldeman’s specimen, described as A. sordida , came from Alabama. Leng and Hamilton recorded the species from Penn- sylvania, Alabama, Louisiana, Texas, and New Mexico. In Biologia Centrali-Americana its habitat is extended to Almolonga, Mexico. It has been reported to the Department from Tucson, Ariz. ; New Orleans, La.; Savannah, Ga. ; Grant, Fla.; and Round Mountain, San Diego, Beeville, Bexar, Edgar, Flatonia, and Dallas, Tex. During 1905 the writer found it in Texas at Yoakum, Sublime, Corpus Christi, Cotulla, Alice, and Victoria. During February and March of 1906, Mr. W. W. Yothers collected hibernating adults from cotton bolls at Athens, Brenham, Navasota, and Calvert. Tex., and on March 1, 1906. the writer also obtained specimens from bolls at Flatonia, same State. HOST PLANTS. The cotton stalk-borer attacks a number of plants. At Tucson, Ariz., it was reported on Helianthus and Xanthium. At Savannah. Ga. ; New Orleans, La., and Round Mountain, Tex., larvae have been found boring in the twigs and trunk of the fig. Mr. E. A. Schwarz informs the writer that it has been bred frequently from the stem of Ambrosia, and one report of the borer in this plant was received by 63 64 PAPERS ON THE COTTON BOLL WEEVIL, ETC. the Department from Edgar, Tex. On March 1, 1906, the writer found larvae in the stems of Ambrosia and Helianthus at Flatonia, Tex. It has been reported breeding in cotton from Bexar and San Diego, Tex., and the writer found it in October, 1905, breeding in this plant at Yoakum, Sublime, and Victoria, and in November at Alice, Tex. The adult has been reported from peach and apricot, but there are no records to show that it injures either of these plants. Leng and Hamilton state that specimens were found boring in dry twigs of box elder (Acer negundo) and hackberry. LIFE HISTORY. Very little work has been done upon the life history of this species. The egg is laid upon the host plant and the larva upon hatching bores into the bark or wood of the plant, where it feeds until full grown, pupates, and finally, when adult, cuts its way out. The winter is very probably past in any of the three stages of larva, pupa, or adult, for it has been reported by Mr. E. A. Schwarz as emerging in May, 1897, from a stalk of cotton cut in December, 1896. In October, 1905, the writer found the insect in cotton stalks in all stages of development from the very small larva to the adult ready to emerge from the stalk. In southwest Texas, where the winters are ver} 7 mild, it is very likely 7 that development proceeds slowly all winter long, except in the most severe winters, when it would probably cease for a few weeks. INJURY AND ECONOMIC IMPORTANCE. Rarely has the cotton stalk-borer been reported, and in no case has serious damage been ascribed to it. Altho it was described by Say as earty as 1832, no mention is made of it in economic literature until 1896, and since that time only live short notes have been made, all of which are apparently based upon Dr. L. O. Howard’s first published notice in 1896, which reads as follows: There is but one borer in the stalks of cotton, and that is the long-horned beetle known as Ataxia crypta (Say). It is occasionally mistaken for an enemy of the plant, but investigation fias shown that it lays its eggs upon, and its larvae bore into, only such stalks as have been damaged by some other cause, such as rust. It follows injury to the plant rather than causes it. The rather general occurrence of this insect at Yoakum, Tex., in 1905 has made it seem advisable to publish this more extended note. Twenty-six fields were examined in this locality, and showed that from 5 to 10 per cent of the stalks were infested by this insect. One Held had 20 to 25 per cent of its stalks injured by the borer. It is to be regretted that the discovery of this infested locality was not made earlier in the season, so that a more conclusive study could have been made. Bui. 63, Part VII, Bureau of Entomology, U. S. Dept, of Agriculture Plate III. The Cotton Stalk-borer (Ataxia crypta). Fig. 1. — Adult or beetle. Fig. 2. — a, Galleries; b, work. Fig. 3. — Emergence holes of adults. Fig. 4. — a, Larva in situ; b , pupa in situ. Fig. 5. — a, Gallery running into root of plant; b, castings behind larva. Fig. 1 twice natural size; figs. 2-5 two-thirds natural size. (Original. ) LIBRARY UNIVERSITY Of ILLINOIS URBANA THE COTTON STALK-BORER. 65 As has been stated by Doctor Howard, it is the larva of the borer that works in the cotton stalk. It confines its attack to the pith of the plant. The amount of injury depends upon the size of the plant and the number of borers. The writer found one plant about 3 feet high at Yoakum which contained eight larvae, and which showed four emergence holes. The twelve borers had eaten out all the pith from the smallest branches to 3 inches below the surface of the ground in the tap-root. The small twigs were only thin shells and crumpled like paper between the fingers. Nearly all infested plants showed unmistakable evidence of previous injury, yet the comparatively fresh and healthy condition of a few stalks points to the probability of an occasional attack upon perfectly healthy stalks. In the field showing an infestation of 20 to 25 per cent of its stalks it was estimated that the injured stalks had borne at least 80 per cent as much cotton to a stalk as had the uninjured ones. If it is true that the borer attacks only such stalks as are dis- eased or otherwise injured, then only a little damage can be attributed to it; yet the borer’s attack, added to the injury due to disease, must hasten the death of the plant and thereby lessen the number of bolls to mature. NATURAL ENEMIES. From material collected at Yoakum, Tex.; on- October 28 two para- sites of Ataxia crypta were bred. One. Ileterospilus websteri Ashm., had entirely covered the pupa of A. crypta with its cocoons. In the majority of cases, however, this parasite had spun its cocoon after crawling away from the body of its host. Several masses of these cocoons were found with no trace of either the larva or pupa of A. crypta visible, and in one instance the larvae of this parasite were found in a hollow cotton stalk and no form of A. crypta near. The second parasite belongs to the genus Etroxys and is a new species. There is some doubt that this species is a parasite upon Ataxia, but since it is parasitic upon beetles and its pupae were found in a stalk bored out by A. crypta , it is here placed on record as a parasite of the latter. REMEDY. If at any time the injury by the insect should become serious the numbers of the borer can be easily reduced b} r destroying the infested stalks early' in the fall. The time of destruction must, of course, vary in different localities, but it should be as early in October as possible for the latitude of Yoakum. By October 28 not more than 10 per cent of the borers had emerged. If the stalks had been destroyed early in October very few borers would have escaped. To obtain the best results, stalks should not be cut but plowed out. for many of the 66 PAPERS ON THE COTTON BOLL WEEVIL, ETC. borers go 2 or 3 inches below the surface of the ground in the tap-root and pupate there. In regions infested by the boll weevil, if the farmers would follow the directions given by this Bureau for the destruction of stalks, the cotton stalk-borer would not appear. In regions uninfested by the boll weevil but infested by the cotton stalk- borer, an occasional examination of infested areas for the emergence holes of the borer or an examination of the stages in the stalk— which can be made by splitting the stalk with a knife — will show the planter when to begin plowing and burning. BIBLIOGRAPHY. 1832. Say, Thomas. — Insects of Louisiana, pp. 5 and 6, and Entomology of North America (Le Conte, ed. ), vol. 1, p. 302. Original description under the name of Lamia crypia. 1847. Hai.deman, S. S. — Trans. Amer. Phil. Soc., vol. 10, p. 56. A description of the insect under the name Ataxia sordida. 1878. Schwarz, E. A. — Coleoptera of Florida. Proc. Am. Phil. Soc., Vol. XVII, p. 457. 1883. Le Conte, J. L., and Horn, G. H. — Classification of Coleoptera of North America, p. 329. Authority for present nomenclature. 1885. Bates, H. W. — Biologia Centrali- Americana, vol. 5, p. 347. A short note upon synonymy. 1896. Howard, L. O. — Bui. 33, Office of Experiment Stations, p. 347. First economic note. 1896. Leng, Chas. W., and Hamilton, John. — The Lamiinae of North America.