OH 
 
L I B RAHY 
 
 OF THE 
 
 U N IVLRSITY 
 
 OF ILLINOIS 
 
 5T0.5 
 
 ILL 
 v. 25-30 
 
 CO 
 
The person charging this material is re- 
 sponsible for its return to the library from 
 which it was withdrawn on or before the 
 Latest Date stamped below. 
 
 Theft, mutilation, and underlining of books 
 are reasons for disciplinary action and may 
 result in dismissal from the University. 
 
 UNIVERSITY OF ILLINOIS LIBRARY AT URBANA-CHAMPAIGN 
 
 ^J'/LOlNGUS£ 
 
 SEP l 197, 
 
 BUILDING ir3^tJtftf< 
 
 JUN Ui37<i 
 SEP * A 
 
 WILDING U 
 SEr- . 
 
 Vtfl 
 
 <;£ ONI 1 * 
 
 L161— O-1096 
 
Digitized by the Internet Archive 
 
 in 2011 with funding from 
 
 University of Illinois Urbana-Champaign 
 
 http://www.archive.org/details/morphologyanatom27morg 
 
.7 
 
 ». 2 
 
 The Morphology and 
 
 Anatomy of American Species 
 
 of the Genus Psaronius 
 
 JEANNE MORGAN 
 
 Illinois biological monographs: Number 27 
 
 THE UNIVERSITY OF ILLINOIS PRESS 
 URBANA, 1959 
 
 $3.00 
 
ILLINOIS BIOLOGICAL MONOGRAPHS is the general title for a series of mono- 
 graphs in botany, entomology, zoology, and allied fields. Volumes 1 through 24 con- 
 tained four issues each and were available through subscription. Beginning with number 
 25 (issued in 1957), each publication is numbered consecutively. No subscriptions 
 are available, but standing orders will be accepted for forthcoming numbers. Prices 
 of previous issues still in print are listed below, and these may be purchased from the 
 University of Illinois Press, Urbana, Illinois. Requests for exchange arrangements should 
 be addressed to the Exchange Department, University Library, Urbana, Illinois. 
 
 BAKER, FRANK COLLINS (1922): The Mol- 
 
 luscan Fauna of the Big Vermilion River, 
 Illinois, with Special Reference to Its Mod- 
 ification as the Result of Pollution by Sew- 
 age and Manufacturing Wastes. 15 pis. 
 Vol. 7, No. 2. $1.25. 
 
 balduf, w. v. (1959): Obligatory and 
 Facultative Insects in Rose Hips. 12 pis. 
 No. 26. $3.50. 
 
 beaver, paul Chester (1937): Experi- 
 mental Studies on Echinostoma revolutum 
 (Froelich), a Fluke from Birds and Mam- 
 mals. 3 pis. 20 charts. Vol. 15, No. 1. $1.00. 
 
 BENNETT, HARRY JACKSON (1936): The 
 
 Life History of Cotylophoron cotylopho- 
 rum, a Trematode from Ruminants. 9 pis. 
 Vol. 14, No. 4. $1.50. 
 
 cahn, alvin robert (1937): The Turtles 
 of Illinois. 31 pis. 20 maps. 15 figs. Vol. 16, 
 Nos. 1-2. $3.00. 
 
 chen, hsin kuo (1935): Development of 
 the Pectoral Limb of Necturus maculosus. 
 11 pis. Vol. 14, No. 1. $1.00. 
 
 COOPER, ARTHUR REUBEN (1918): North 
 
 American Pseudophyllidean Cestodes from 
 Fishes. 13 pis. Vol. 4, No. 4. $2.00. 
 
 CREGAN, SISTER MARY BERTHA (1941): 
 
 Generic Relationships of the Dolichopo- 
 didae (Diptera) Based on a Study of the 
 Mouth Parts. 30 pis. Vol. 18, No. 1. $1.00. 
 
 DOAK, CLIFTON CHILDRESS (1935): Evolu- 
 
 tion of Foliar Types, Dwarf Shoots, and 
 Cone Scales of Pinus with Remarks Con- 
 cerning Similar Structures in Related 
 Forms. 32 figs. Vol. 13, No. 3. $1.50. 
 
 eddy, samuel (1934): A Study of Fresh- 
 Water Plankton Communities. 9 figs. Vol. 
 12, No. 4. $1.00. 
 
 essex, Hiram eli (1928): The Structure 
 and Development of Corallobothrium, with 
 Descriptions of Two New Fish Tape- 
 worms. 5 pis. Vol. 11, No. 3. $1.00. 
 
 faust, ernest Carroll (1918): Life His- 
 tory Studies on Montana Trematodes. 9 
 pis. 1 fig. Vol. 4, No. 1. $2.00. 
 
 FISHER, HARVEY I., and GOODMAN, DONALD 
 
 c ( 1955) : The Myology of the Whooping 
 Crane, Grus americana. 40 figs. Vol. 24, 
 No. 2. $2.50. 
 
 GAMBHJL, WILLIAM G., jr. (1953): The 
 Leguminosae of Illinois. Vol. 22, No. 4. 
 $3.00. 
 
 GOODNIGHT, CLARENCE JAMES (1940): The 
 
 Branchiobdellidae (Oligochaeta) of North 
 American Crayfishes. 3 pis. Vol. 17, No. 3, 
 $1.00. 
 
 cutberlet, john earl (1915): On the 
 Osteology of Some of the Loricati. 5 pis. 
 Vol. 2, No. 2. $.50. 
 
 Hayes, william Patrick (1930): Mor- 
 phology, Taxonomy and Biology of Larval 
 Scarabaeoidea. 15 pis. Vol. 12, No. 2. 
 
 $1.25. 
 
 heiss, Elizabeth M. (1938): A Classifica- 
 tion of the Larvae and Puparia of the 
 Syrphidae of Illinois, Exclusive of Aquatic 
 Forms. 17 pis. Vol. 16, No. 4. $1.50. 
 
 HIGGINS, GEORGE MARSH (1921): The 
 
 Nasal Organ in Amphibia. 10 pis. Vol. 6, 
 No. 1, $1.00. 
 
 higley, ruth (1918): Morphology and 
 Biology of Some Turbellaria from the 
 Mississippi Basin. 3 pis. Vol. 4, No. 3. 
 $1.25. 
 
 hoff, c. clayton (1942): The Ostracods 
 of Illinois— Their Biology and Taxonomy. 
 9 pis. Vol. 19, Nos. 1-2. $2.50. 
 
The Morphology and Anatomy of 
 American Species of the Genus Psaronius 
 
Reconstruction of a psaroniaceous tree fern showing Caulopteris features in leaf scar 
 arrangement (spiral arrangement). This reconstruction is based on a plant approxi- 
 mately 20 to 25 feet tall. 
 
The Morphology and 
 
 Anatomy of American Species 
 
 of the Genus Psaronius 
 
 JEANNE MORGAN 
 
 Illinois biological monographs: Number 27 
 
 THE UNIVERSITY OF ILLINOIS PRESS 
 URRANA, 1959 
 
Board of Editors: william r. horsfall, h. orin halvorson, francis j. 
 
 KRUIDENIER, WILSON N. STEWART, AND AUDREY D. TAYLOR. 
 
 THIS MONOGRAPH IS A CONTRIDUTION FROM THE DEPARTMENT OF DOTANY, 
 UNIVERSITY OF ILLINOIS. DISTRIDUTED: DECEMRER 15, 1959. 
 
 © 1959 by the board of trustees of the university of illinois. manu- 
 factured in the united states of america. library of congress catalog 
 card no. 59-15336. 
 
S7i 
 
 ACKNOWLEDGMENT 
 
 The author wishes to extend to Professors W. N. Stewart of the 
 University of Illinois and Theodore Delevoryas of Yale University her 
 sincere appreciation for their valuable assistance with this manuscript. 
 Appreciation is also expressed to the following persons and institutions 
 for the loan or gift of many specimens used in this research: Dr. H. N. 
 Andrews, Washington University; Dr. Virginius Chase, Peoria, Illinois; 
 Dr. J. W. Hall, University of Minnesota; Dr. R. M. Kosanke, Illinois 
 State Geological Survey; Dr. A. H. Brickie, Ohio University; Dr. P. D. 
 Voth, University of Chicago; Dr. R. W. Baxter, University of Kansas; 
 Mr. F. B. Cookson, Emporia, Kansas; and The Rio de Janeiro Museum. 
 
 Preparation of this manuscript for publication was financed in part by 
 the National Science Foundation. 
 
CONTENTS 
 
 I. INTRODUCTION 1 
 
 II. OBSERVATIONS 4 
 
 III. GENERAL DISCUSSION 46 
 
 IV. TAXONOMIC TREATMENT OF THE PSARONII OF NORTH AND SOUTH 
 AMERICA 62 
 
 V. SUMMARY 68 
 
 LITERATURE CITED 70 
 
 GLOSSARY 72 
 
 PLATES 75 
 
 INDEX 107 
 
I. INTRODUCTION 
 
 Psaronius is an organ genus first proposed by Cotta ( 1832 ) for the 
 fossil stems of certain tree ferns belonging to floras of the Carboniferous 
 and Permian periods of geological time. 
 
 In gross structure (see frontispiece) Psaronius was an erect, un- 
 branched stem up to 50 feet high with an apical crown of large com- 
 pound leaves. The leaves were arranged in vertical rows or in a spiral. 
 As the leaves aged they abscised, leaving large elliptical scars on the 
 surface of the stem. Older parts of the stem toward the base were 
 covered with a thick mantle of adventitious roots, causing the base of 
 the stem to appear much enlarged. 
 
 The general appearance of these ancient ferns compares favorably 
 with modern tropical tree ferns belonging to the families Cyatheaceae 
 and Marattiaceae of the Filicineae. However, the structure of their 
 stem, roots, and sporangia show that the affinities of these extinct ferns 
 are with the eusporangiate Marattiales. For reasons explained later in 
 the text, Psaronius is placed in the family of Psaroniaceae of the 
 Marattiales. 
 
 Judging from the structure of its roots, Psaronius grew in a swamp 
 environment along with seed ferns, lepidodendrids, coenopterid ferns, 
 calamites, sphenophylls and other plants typical of the Carboniferous 
 landscape. As in many existing tropical rain forests, the Paleozoic en- 
 vironment in which these plants existed was uniform in temperature 
 and available water. Not until the Permian period is there any evidence 
 of important environmental changes in the upper part of the Paleozoic 
 era. These changes apparently brought about the extinction of the 
 Psaroniaceae as evidenced by the absence of their stem remains in 
 younger strata. 
 
 In Europe during the eighteenth and nineteenth centuries, specimens 
 of Psaronius were prized as semiprecious stones and were called Staar- 
 steine, meaning "starling stones," because the cross-section of the root 
 mantle resembles the markings on the breast of the European starling. 
 Stenzel ( 1854 ) suggested that this was a corruption of Stern Steine ( star 
 stones), referring to the stellate arrangement of the actinosteles of the 
 roots. The descriptive term Wurmsteine has also been applied to speci- 
 mens of Psaronius, referring to the rather wormlike appearance of the 
 vascular bundles in cross-section. 
 
 Sprengel (1828) was the first to recognize that these fossil specimens 
 were remains of fern stems for which he established the genus Endo- 
 
 1 
 
2 AMERICAN SPECIES OF THE GENUS PSARONIUS 
 
 genites. Cotta in 1832 removed them from that genus when he established 
 the genus Psaronius. Since that time, hundreds of Psaronius stem frag- 
 ments have been collected from the Carboniferous of central France, 
 Bohemia, England, and the United States. Until the present, however, 
 little has been learned about the plant as a whole. 
 
 Since the discovery of Psaronius the literature has been filled with 
 conflicting ideas concerning the morphology of these stems. Some papers 
 (Zeiller, 1890; Stenzel, 1906) reveal a knowledge of the genus advanced 
 for the time, but many of the morphological interpretations of these 
 authors are now known to be incorrect. In these papers, and in certain 
 other treatments of the genus (Hirmer, 1927; Seward, 1910; Scott, 1920) 
 the more internal parts of the vascular system were described, without 
 critical examination, as a series of very complex anastomosing strands. 
 Diagrams of the cycles of vascular tissue seen in longitudinal section have 
 been prepared by Farr (1914), Hirmer (1927), and Schoute (1938), in 
 an attempt to show its organization. Shoute's simple diagram approxi- 
 mates the generalized condition of a Psaronius vascular system but leaves 
 much to be desired in morphological interpretation. Many stem speci- 
 mens have remained undescribed because of their complex and often 
 confusing vascular systems. With new information gained from the 
 present study, these specimens may now be correctly interpreted and 
 named. 
 
 One of the primary objects of this research was to obtain some idea 
 of ontogenetic variability present in a stem of Psaronius. It was necessary 
 to study a large number of stem fragments and in this way determine 
 the spectrum of variability occurring from base to apex. After the extent 
 of ontogenetic variability within a species had been esablished, it was 
 then possible to evaluate that variation which remains constant among 
 different stems and characterizes species. Enough stem material was 
 available to observe the morphology and ontogeny of two species of 
 Psaronius from a level near the base of the stem to a point near the apex. 
 With this knowledge of ontogenetic variability for two species it was 
 then possible to interpret variability in species represented by only a 
 small number of stem fragments. 
 
 For the sake of simplicity and clarity most specimens mentioned in 
 this article have been referred to by capital letter. Although a large 
 number of additional stem fragments were investigated, they are not 
 referred to specifically. A list of coal-ball and ironstone specimens is 
 presented below. The reference letter, accession number, and source is 
 given for each specimen. Unless otherwise indicated, the specimens are 
 now in the paleobotanical collection, Department of Botany, University 
 of Illinois, Urbana, Illinois. Stratigraphic position of various collecting 
 sites are given at the end of this list. 
 
INTRODUCTION 3 
 
 Specimen A Washington University Coal Ball 108 (now at the Uni- 
 versity of Illinois) — Berryville, Illinois. 
 
 Specimen B C.B. 2841 — Berryville, Illinois. 
 
 Specimen C C.B. 189, 196, 266, 267 (this specimen was found as 
 several isolated fragments which have since been ce- 
 mented together) — Berryville, Illinois. 
 
 Specimen D C.B. 2844— Berryville, Illinois. 
 
 Specimen E C.B. 2778, 2779, 2789 (these three fragments are all part 
 of a single stem) — Dix, Illinois. 
 
 Specimen F C.B. 2842 — Berryville, Illinois. 
 
 Specimen G C.B. 2680 — Berryville, Illinois. 
 
 Specimen H C.B. 2681 — Berryville, Illinois. 
 
 Specimen I C.B. 2682 — Berryville, Illinois. 
 
 Specimen J University of Minnesota Coal Ball 35 (now at the Uni- 
 versity of Illinois) — West Mineral, Kansas. 
 
 Specimen K Washington University Coal Ball 698 (now at the Uni- 
 versity of Illinois) — West Mineral, Kansas. 
 
 Specimen L C.B. 1539 — West Mineral, Kansas. 
 
 Specimen M Washington University Coal Ball 649 (now at the Uni- 
 versity of Illinois) — West Mineral, Kansas. 
 
 Specimen N Illinois State Geological Survey Coal Ball 803 (at the 
 Illinois State Geological Survey, Urbana, Illinois) — West 
 Mineral, Kansas. 
 
 Specimen O Illinois State Geological Survey Coal Ball 1254 (at the 
 Illinois State Geological Survey, Urbana, Illinois) — West 
 Mineral, Kansas. 
 
 Specimen P C.B. 2111 — West Mineral, Kansas. 
 
 Specimen Q C.B. 3066 — West Mineral, Kansas. 
 
 Specimen R C.B. 1221 — Berryville, Illinois. 
 
 Specimen S C.B. 1700 — St. Wendel, Indiana. 
 
 Specimen T Illinois State Geological Survey Coal Ball 51 (now at the 
 University of Illinois) — New Delta, Illinois. 
 
 Specimen U Ironstone concretion Specimen 2608 — Wilmington, Illi- 
 nois. 
 
 Specimen V Ironstone concretion Specimen 2609 — Wilmington, Illi- 
 nois. 
 
 Specimen W Ironstone concretion Specimen 728 — Coal City, Illinois. 
 
 Stratigraphic Position of Collecting Sites 
 
 Berryville, Illinois — Upper McLeansboro group, upper Pennsylvanian 
 
 System. 
 Coal City, Illinois — Lower Carbondale group above No. 2 coal, middle 
 
 Pennsylvanian System. 
 
4 AMERICAN SPECIES OF THE GENUS PSARONIUS 
 
 Dix, Illinois — Upper McLeansboro group, upper Pennsylvanian System. 
 New Delta, Illinois — Top of Carbondale group, middle Pennsylvanian 
 
 System. 
 St. Wendel, Indiana — Parker coal, lower McLeansboro group, upper 
 
 Pennsylvanian System. 
 West Mineral, Kansas — Cherokee shale, Des Moines series, middle 
 
 Pennsylvanian System. 
 Wilmington, Illinois — Francis Creek Shale above No. 2 coal, middle 
 
 Pennsylvanian System. 
 
 II. OBSERVATIONS 
 
 A. Psaronius blicklei sp. nov. 
 
 1. Description 
 
 A well-preserved fragment of Psaronius (figs. 1, 2, 25, 26 specimen A) 
 was recovered from a stream bed near Berryville, Illinois. With the 
 possible exception of one stem of another species described later, this 
 stem has the smallest cross-sectional dimensions in its lower portions of 
 any known American specimen. At the lowest portion of the three-foot 
 length of stem preserved in the coal ball, cross-sectional dimensions are 
 2.3 X 1-5 cm. in its slightly crushed condition. At the upper extremity 
 the diameter approaches 3 cm. The vascular bundles are arranged con- 
 centrically to produce a polycyclic condition and each bundle possesses 
 groups of protoxylem along its internal face. The metaxylem cells, which 
 make up the bulk of the bundle, average 75 microns in diameter and are 
 interspersed with xylem parenchyma cells occurring singly or in groups. 
 The tracheids themselves are long and tapering, with scalariform thick- 
 enings. Secondary thickenings on the walls of protoxylem are spiral and 
 scalariform. The region of the phloem is not well preserved, but a band 
 of poorly preserved cells completely encircles each vascular bundle. The 
 phloem zone is in a better state of preservation in some of the larger 
 stems and will be discussed under the section on histological details. No 
 endodermis is present around the bundles. The steles are embedded in 
 a rather compact parenchymatous ground or fundamental tissue which is 
 limited at the periphery of the stem by a thin band of sclerenchyma. Also 
 embedded in the ground tissue at various places, in association with the 
 steles, are groups of sclerenchyma fibers. 
 
 Interspersed throughout the fundamental tissue are numerous cells 
 with dark contents very closely resembling the tannin cells of some of 
 the living Marattiaceae (West, 1915). Lysigenous lacunae in the funda- 
 mental tissue, characteristic of this species, are sparse in the more basal 
 
OBSERVATIONS 5 
 
 portions and average 1 mm. in diameter. They resemble, at least super- 
 ficially, the mucilage canals of extant marattiaceous ferns (West, 1915). 
 These lacunae do not appear to be filled with a deposit of any kind and 
 there is no evidence of an epithelial layer. They are believed to be 
 lysigenous in origin because of the lack of any epithelium and the pres- 
 ence of remains of disintegrated cell walls which are found at the pe- 
 riphery of the cavity. 
 
 At the time of leaf trace separation, a "bulge" appears in the scleren- 
 chyma sheath, and a leaf trace which has separated from the stem stele 
 bends up and out into it. The sheath then inrolls upward and inward 
 until it completely surrounds the trace on its inner face. The break in 
 the leaf base and the included trace resulting from abscission of the 
 frond is responsible for the apparent presence of two separate vascular 
 strands in a leaf base near the periphery of the stem. The free trace is 
 actually a curved strand which at successively higher levels has progres- 
 sively larger amounts of the abaxial portion broken off (figs. 3, 27). 
 
 Immediately surrounding the limiting layer of the stem is a zone 
 composed of adventitious roots. The vascular supply of these roots 
 originates from the peripheral cauline bundles. Surrounding the roots is 
 a compact mass of sinuously radiating rows of parenchyma cells which 
 arise by proliferation of parenchyma cells of the ground tissue of the 
 stems and roots, and also from parenchyma cells interspersed among the 
 sclerenchyma which surrounds both stem and adventitious roots. This 
 will be discussed later in the detailed description of the root mantle. 
 Still farther to the outside is a zone of roots, not encased in the sec- 
 ondary parenchyma, called the free root zone. These roots have for the 
 most part arisen from more distal portions of the stem, and in their down- 
 ward and outward course have either grown free of the compacted mass 
 of the inner root zone or have not yet been incorporated in it. The roots 
 of the inner layer are, as a rule, smaller in diameter than the free roots 
 and are surrounded by a layer of sclerenchyma intermixed with paren- 
 chyma cells. The parenchyma cells are somewhat sparse and not readily 
 visible in cross-section, but their presence is attested by their prolifera- 
 tion to form the secondary parenchyma in which these inner roots are 
 embedded. Within the sclerenchyma sheath surrounding roots of the 
 inner root zone there is a cortex composed of relatively large compact 
 cells. Tannin-like deposits similar to those found in the fundamental 
 tissue of the stem are present in these roots and in the secondary paren- 
 chyma cells surrounding them. Embedded in the ground parenchyma is 
 an actinostele bearing four to seven arms. Protoxylem is found at the tips 
 of the arms. Scattered xylem parenchyma cells are found interspersed 
 among the tracheids; pockets of phloem alternate with the xylem arms 
 of the actinostele. The free roots are also surrounded by a layer com- 
 
6 AMERICAN SPECIES OF THE GENUS PSARONIUS 
 
 posed of sclerenchyma cells. Although not directly observed, parenchyma 
 cells are probably intermixed with the sclerenchyma cells. Parenchyma 
 cells, if present, do not proliferate while the roots are in the free con- 
 dition. Proliferation most likely does take place later in development as 
 the plant grows taller, encasing the roots in a mass of secondary paren- 
 chyma cells. Development of this kind would be necessary to allow for 
 increased diameter of the root mantle in larger stems. The root zones of 
 some basal specimens have a somewhat layered appearance which might 
 correspond with periods of secondary parenchyma production. The more 
 peripheral region of the cortex of a free root is composed of very loosely 
 arranged parenchyma, giving it a "netted" appearance (fig. 77). An 
 analogous condition is found in the stems of certain extant aquatic ferns, 
 e.g., Marsilea. The more internal cells of the cortex are compact and 
 surround a stele similar in structure to roots of the inner zone. Tannin-like 
 deposits are also present in the compact inner cortical cells of free roots. 
 Psaronius roots have been observed to branch ( Reed, 1949 ) , and branch- 
 ing is common in roots obtained from the Berryville locality (fig. 77). 
 
 At the lowest level of the stem (specimen A, figs. 1, 25) with the 
 smallest diameter, the vascular system is characterized by a dicyclic 
 condition. In this region, while the plant was small, there were three 
 orthostichies of leaves. This is revealed internally by the presence of 
 three vertical rows of leaf gaps in the outer stelar cycle. Leaf gaps are 
 situated above the point of separation of the leaf traces, and the over- 
 lapping of these gaps produces a dictyostelic structure. Cross sections 
 at certain levels of the stem give the appearance of a siphonostele with 
 only a single leaf gap, but levels slightly above or below show the dic- 
 tyostelic condition again (fig. 11). Leaf traces always arise from discrete 
 parts of the outer cycle, and alternate in specimen A, with three vertical 
 vascular strands which are cauline. 
 
 In all of the stems to be discussed, the cycles are numbered suc- 
 cessively from the outside toward the center. Leaf traces have their origin 
 in the innermost cycle, a relatively simple open cylinder at the lower 
 stem levels. As this cylinder passes upward and across the ground tissue 
 it fuses with two peripheral cauline bundles, filling the gaps in the 
 outer cycle in a spiral sequence corresponding to the separation of leaf 
 traces (fig. 10A, B). The portion of the open cylinder which fills the 
 gap is a leaf trace originating in the central cycle. The two free arms 
 (fig. IOC) of the inner cycle separate and turn back toward the center 
 of the stem where they fuse (fig. 10D), closing the first gap in the 
 inner cycle. This cylinder in its upward course then passes across the 
 ground tissue, filling the gap of the next higher leaf trace in the phyllo- 
 tactic sequence with a new leaf trace. It should be noted that the next 
 
OBSERVATIONS 7 
 
 inner cylinder, like the outer cycle, is composed of leaf traces alternating 
 with cauline bundles. 
 
 Above the point of separation of the inrolled edges (fig. 10E, F), the 
 two central vascular bundles fuse — not at the edges, but along their 
 convex portions — thus leaving two short edges projecting for distance 
 toward the center of the stem (fig. 10G). These projections separate as 
 a single curved strand, thus initiating the third cycle (fig. 10H). The 
 second cycle continues in its upward course, filling, in sequence, the 
 spirally arranged gaps in the outer cycle. Each time the second cycle 
 gives off a leaf trace to the outer cycle, it passes between two peripheral 
 bundles. Only rarely does the second cycle fuse at the same level to both 
 edges of the gap of the outer cycle; instead, one of the two sides fuses 
 with a peripheral bundle at a lower level than the other (fig. 101). The 
 third cycle, which originated from the cycle immediately to its outside, 
 functions in the same way as the second, filling the leaf gaps in the 
 latter in a manner identical with gap closure in the outer cycle by the 
 second vascular cylinder. At its lowermost level the third cycle is repre- 
 sented by a slender strand (figs. 2, 26) which fuses to one edge of the 
 gap in the second cycle (fig. 101, J). The closing of the gap is com- 
 pleted by an inrolling of the cauline portion. Where the innermost cycle 
 is fused to one side of a gap in the second cylinder, it is represented by 
 a curved edge which is met by an inrolling of the cauline bundle on the 
 opposite side of the gap. A pinching off of the two recurved ends (fig. 
 10J, K) reconstitutes the third cycle. 
 
 At this level, the third cycle is a very small strand (fig. 10K, L) be- 
 cause a part of it has been contributed to the formation of a leaf trace. 
 At higher levels this strand increases in volume through the additional 
 differentiation of tracheids. A similar increase in volume is also brought 
 about in each of the investing steles. Because of this volumetric increase, 
 accompanied by an increase in stelar number, the dimensions of the 
 vascular system keep pace with over-all increase in stem volume. 
 
 An idealized, three-dimensional reconstruction of the stelar system of 
 specimen A is portrayed in fig. 11 (lowest portion at bottom left; highest 
 at top right). In the length represented by this figure, three leaf traces 
 arise and the addition of a new cycle in the center of the stem can be 
 followed. The diagram also shows the course of the second cycle, which 
 contributes traces to the outer one. Beginning at the top of the sixth 
 section from the bottom, the method of initiation of the third cycle is 
 portrayed. Above that section is shown the manner in which both the 
 most internal and next outer cycles contribute traces to their respective 
 outer cycles. 
 
 Specimen A just described is probably the most nearly basal portion 
 
AMERICAN SPECIES OF THE GENUS PSARONIUS 
 
 of a Psaronius stem thus far described. That it is a basal portion is shown 
 by the relative simplicity of the stem with only three rows of leaf traces, 
 its rather small diameter, and the correspondingly large ensheathing 
 growth of adventitious roots extending approximately 12 inches around 
 the stem in all directions. Based on the evidence presented, it is obvious 
 that as large and complex a stem as Psaronius infarctus, which has as 
 many as fourteen rows of leaf traces and an extremely polycyclic vascular 
 system, could not be a basal fragment of Psaronius. 
 
 Fragments of large and complex stems have been found in America 
 during recent years. By establishing an intergrading series composed of 
 many of these fragments, (specimens A through H; figs. 25-33), it is 
 possible to recognize progressive stages of an ontogenetic series from the 
 simple base to the complex distal portions of the stem. Because frag- 
 ments representing this series were not found organically connected, 
 certain anatomical characters must be relied upon for verification of the 
 fact that these fragments belong to one species. The specimens used in 
 the ontogenetic series to be described retain the same stelar morphology 
 throughout. They are also similar in the presence of tannin cells, ly- 
 sigenous lacunae, and the characters of the sclerenchyma. Very basal 
 portions of these stems may lack certain of these characters; however, 
 higher levels in the same fragment show them. 
 
 Specimen B is a stem fragment that has a four-cycled stelar condition 
 (figs. 3, 27). The level of the stem depicted by these figures appears to 
 have only three cycles; however, the fourth and innermost cycle is fused 
 with the next outer cycle before contributing to the formation of a leaf 
 trace which will be given off at a higher level. That specimen B is a 
 portion of a stem taken at a slightly higher level is evidenced by the 
 larger number of cycles and also by the five well-developed orthostichies 
 of leaves present at this level. The manner in which a cycle may be added 
 to the stelar system has been shown in specimen A, and the method by 
 which this stem was able to add a new row of leaf traces will be dis- 
 cussed in connection with specimen E, which is one of the still more 
 distal portions of P. blicklei. In addition to increased complexity of the 
 vascular cylinder, the lacunae characteristic of this species become in- 
 creasingly larger and more numerous. These are found in abundance 
 scattered throughout the ground parenchyma, and several are usually 
 found associated with the concave face of a departing leaf trace. Often 
 two or more lacunae coalesce to form a large, irregular cylindrical cavity 
 in the cortex. 
 
 Specimen C (figs. 4, 28) shows the structure in a P. blicklei stem but 
 at a somewhat higher level than in specimen B. The fragment actually 
 possesses five concentric cycles, only four of which are visible in this 
 representative cross-section. There has been no increase in the number 
 
OBSERVATIONS 9 
 
 of leaf traces in this specimen; it exhibits five rows as does specimen B. 
 Thus, it becomes apparent that there is no exact correlation between in- 
 creasing size, number of cycles, and numbers of rows of leaf traces 
 through short distances. A generalized pattern is apparent, however, as 
 evidenced by a correlation between a very large number of orthostichies 
 in a stem and a correspondingly large number of stelar cycles. This 
 generality holds true for specimens with the spiral and whorled leaf bases 
 but not in distichous, and possibly not in the tetrastichous and hexa- 
 stichous species to be discussed later. 
 
 Specimen C also shows a larger number of lacunae than were present at 
 lower levels. In addition to the isolated patches of sclerenchyma de- 
 scribed for specimen A there are, at this higher level, bands of scleren- 
 chyma fibers which roughly follow the configuration of the steles. These 
 will be discussed in greater detail in the description of specimen F. 
 
 A diagram of the stelar pattern present in a specimen having a spiral 
 phyllotaxy with few orthostichies is shown in fig. 12, which represents 
 an idealized cross-section of specimen C. Leaf traces and "potential" 
 leaf traces are represented in fig. 12 by solid black, while the cauline 
 portions of the steles are unshaded. The interpretation of the stippled 
 areas will be explained in connection with description of fig. 14. In stem 
 cross section (fig. 12) there are five concentric cycles although only 
 four appear to be present. The fifth or innermost cycle is fused to the 
 fourth cycle ( fig. 13A, B ) at the left side of the lower gap in the fourth 
 cycle. When that gap is closed the two strands combining the fourth 
 and fifth cycles do not fuse along the edges but the part of the strand 
 composed of the fifth cycle bends toward the center of the stem (fig. 
 13C ) . The gap is closed as a result of fusion of the two flanking strands 
 along their convex faces leaving two projections directed toward the 
 center of the stem (fig. 13D). These inner projections, which actually 
 represent the fifth cycle, separate as a single strand which migrates 
 toward the other gap in the fourth cycle (fig. 13E). Similar to specimen 
 A the fusion of the innermost cycle to the next outer one results in the 
 transfer of a portion of the central cycle (leaf trace) to the one just out- 
 side. At the same time there is a compensation for the diminution in 
 size of the innermost cycle by an addition of vascular elements originat- 
 ing from one of the incurved edges of the next outer cycle at the point 
 of closing of the gap. 
 
 The fourth cycle in the stem diagrammed in fig. 12 behaves in the 
 same way as the second cycle in specimen A. The convex portion in the 
 lower left part of the fourth cycle, which consists in part of a leaf trace, 
 fuses with the inrolled edges of the third cycle filling the gap in the 
 third cycle immediately to outside of the fourth cycle. In this manner 
 a trace is passed from the fourth to the third cycle and when the trace 
 
10 AMERICAN SPECIES OF THE GENUS PSARONIUS 
 
 becomes separated a gap is left in the fourth cycle. This gap is again 
 partly filled by the fifth cycle at a higher level. The third transfers a trace 
 to the second in a similar way, and with slight modifications the second 
 cycle passes traces to the outermost one. All of the stelar changes just 
 described proceed simultaneously in the stem; for example, the fourth 
 cycle passes off a trace to the third while the gap on the opposite side 
 of the third cycle is being filled in part by the fifth cycle. These various 
 fusions and gap formations often make it difficult on cursory examination 
 to distinguish the cycles and occasionally (as in fig. 11) there may appear 
 to be fewer cycles than is actually the case. However, a section a short 
 distance above or below the level in question readily shows the true 
 number. 
 
 The free edges of the peripheral cauline strands are usually con- 
 spicuously inrolled (figs. 12, 14A). The fusion of a trace in the second 
 cycle next to the outermost cycle takes place at positions along the 
 convex portion of the peripheral bundles a short distance from the 
 edges of those bundles ( fig. 14B ) . When the second cycle separates from 
 the outer one, a part of each peripheral bundle flanking the trace (the 
 free edges) is added to either side of the gap in the second or more in- 
 ternal cycle (fig. 14C). Before the trace becomes free from the outer- 
 most cycle, the parts of the adjacent cauline bundles represented by 
 stippled areas in the diagrams are contributed to the leaf trace (fig. 
 14C). In a diagram (fig. 12) it is difficult to specify the exact portions 
 which function as leaf traces, cauline bundles, or portions of bundles 
 which are contributed either to leaf traces or the next inner cycles. 
 These have been designated only arbitrarily in the diagram and do not 
 show that with increasing vertical distance there are pronounced changes 
 in tangential dimensions and in configurations of the vascular bundles. 
 For example, a peripheral cauline bundle which contributes segments 
 to a leaf trace and the next inner cycle must necessarily become en- 
 larged at higher levels before the next higher leaf trace of the same 
 orthostichy fuses to it. Because of this, the exact amount of vascular 
 tissue which the peripheral bundle contributed to a leaf trace or to the 
 second cycle can be only an approximation in a diagram of a cross- 
 section. Similarly, the tangential extent of the leaf traces or their exact 
 position within a stele are only approximations. 
 
 At its point of separation a leaf trace is slightly curved with the con- 
 vexity outward. At successively higher levels the convexity increases and 
 eventually the free edges become inrolled and approach each other. Not 
 infrequently, a trace in any of the cycles may separate in a position to 
 one side of its point of fusion from the cycle immediately within. This 
 "shift" in position of the trace often obscures the pathway of a trace 
 
OBSERVATIONS 11 
 
 from the center of the stem to the periphery. Figure 12 shows a "shift" 
 in which the position of the gap on the right side of the third cycle does 
 no coincide with the position of the leaf trace in the second cycle. 
 
 Specimen D (figs. 5, 29) is a stem that has six cycles and six rows of 
 leaf traces. The sixth and most internal cycle is shown in fig. 29 by a 
 small strand now fused with the next outer (fifth) cycle. 
 
 The lowest portion of the stem of specimen E (figs. 6, 30) is char- 
 acterized by seven orthostichies of leaf traces. Because of the increased 
 number of rows of traces, the circular outline of sclerenchyma tissue 
 common in smaller stems is replaced by a more sinuous configuration 
 of the sclerenchyma which cuts in on the adaxial faces of leaf bases. 
 The encircling band of sclerenchyma is also much thicker than that of 
 specimen A. The stem at this level has reached such a degree of com- 
 plexity that it is difficult to ascertain with accuracy the exact number 
 of cycles present in the stem. As the number of rows of leaf traces in- 
 creases, the number of leaf traces which must originate from the inner- 
 most cycle likewise increases. This accounts for the much dissected 
 appearance of central portions of larger stems caused by overlapping 
 of numerous leaf gaps. Further, because of fusions of strands of one 
 cycle with those of the next outer cycle, the internal stelar situation 
 becomes much more complex and impossible to follow unless many serial 
 peels are closely observed. There is actually no necessity for such de- 
 tailed observations because the morphology in such complex specimens 
 is very likely nothing more than a compounding of the simple condition 
 described for specimen A. Although the actual anastomoses of the 
 individual strands vary with the stem and even within different levels 
 in the same stem, the overall pattern of leaf trace initiation appears to 
 be the same. 
 
 Near the bottom of specimen E the stem within the root mantle 
 measures 8 X 6.5 cm. in cross-sectional dimensions, while near the upper 
 end of the well-preserved portion corresponding dimensions average 10 
 X 7.5 cm. Only a small portion of the root mantle is preserved around 
 the stem. 
 
 These observations made of fragments forming the ontogenetic series 
 A-E show that the number of rows of leaf traces range from three in 
 the most basal specimen A to seven for specimen E. The fact that ortho- 
 stichies are added as the stem grows was recognized by Zeiller (1890) 
 and others. In spite of these observations, confusion remained concern- 
 ing the interpretation of stem structure because workers before and 
 after Zeiller (1890) persisted in indiscriminately naming species of 
 spiral forms based on the number of orthostichies present. Specimen E 
 just described clearly shows a series of steps in the addition of an eighth 
 
12 AMERICAN SPECIES OF THE GENUS PSARONIUS 
 
 row of leaf traces to a previously seven-rowed condition, and provides 
 additional evidence that the number of orthostichies is the product of 
 ontogenetic variability. 
 
 In fig. 15A (specimen E), three of the seven peripheral cauline 
 bundles have been labeled a, b, and c. There is one well-established path- 
 way of leaf trace divergence situated between cauline bundles a and b. 
 At 15B a "bulge" on the bundles of the next inner cycle indicates that 
 a leaf trace will join with bundles a and b at a higher level. The leaf 
 trace connecting peripheral bundles b and c has separated below the 
 level of 15B, but has separated farther to the right than would be 
 expected, considering the usual pathway of divergence. This leaves 
 peripheral bundle b unusually large in tangential extent. There is a 
 slight indentation in the bundle emphasized by a band of sclerenchyma 
 (15B). In fig. 15C it may be observed that a large portion (left side) of 
 the bundle formerly labeled b has been set free from the right side and 
 has assumed the nature of an additional leaf trace at the periphery of a 
 stem before its departure into a petiole base. The portion labeled b' 
 (fig. 15C) is actually a small remaining segment of b to which the leaf 
 trace is still attached. To the left edge of peripheral bundle b is attached 
 a leaf trace which will be given off at a somewhat higher level, and 
 which is not yet fused to bundle a. A slight outward projection at the 
 point of attachment of the two is caused by a partial section through 
 an adventitious root departing from the peripheral bundle. A slight 
 inward projection is a part of the stelar contribution to the next inner 
 cycle. 
 
 Thus, it is clear that a single, unusually large peripheral bundle divides 
 to give off a leaf trace from its middle portion, leaving its two ends with 
 inrolled edges; each of these ends forms a single peripheral cauline 
 bundle ( b' and b ) . As a result there has been added to the stem one new 
 leaf trace, representing the beginning of a new orthostichy, and one new 
 peripheral bundle (b'). The new peripheral bundle V attains a normal 
 size because the leaf trace between bundles a and b' is set free farther 
 to the left than would ordinarily be expected for that channel, and also 
 because additional tracheids were differentiated by the apical meristem 
 to give the bundle a longer tangential extent. The separation of the leaf 
 trace fused to peripheral bundles b and c (fig. 15A) slightly to the right 
 of its usual place of separation forms a gap in the outer cycle which 
 does not correspond to the path of divergence of traces of the same 
 orthostichy evidenced in the inner cycles. There is actually no "shift" 
 of the leaf trace, but rather one part of the trace is contributed to pe- 
 ripheral bundle b while the opposite edge of the trace consists of a 
 portion of peripheral bundle c. The leaf trace supplying the next higher 
 leaf of the same vertical row originates in the center of the stem and 
 
OBSERVATIONS 13 
 
 bends upward and outward through the previously established path- 
 way. When it has fused to the second cycle it "shifts" in a manner similar 
 to that described for the leaf trace at fig. 15A. The separation of the 
 trace from the second cycle (fig. 15C-F) forms a gap in that cycle 
 directly opposing the gap in the first cycle. By similar "shifts" of traces in 
 the cycles nearer the center of the stem there is eventually created a 
 straight pathway for leaf traces supplying leaves in the orthostichy re- 
 located by the departure of the trace (fig. 15A). Thus, the shift in 
 position of the leaf trace starting in the outermost cycle is repeated all 
 the way toward the center of the stem at successively higher levels by the 
 cycles within. A strikingly analogous situation prevails in the establish- 
 ment of a new (eighth) pathway for traces to leaves in the orthostichy 
 initiated by the trace arising from peripheral bundle b (fig. 15B). The 
 new leaf trace has originated entirely from a cauline strand. The trace 
 directly above originates in an identical manner from a cauline strand 
 in cycle two (fig. 15E-I), thus extending the pathway of leaf trace 
 divergence one cycle toward the interior. Successively higher traces owe 
 their origin to cauline bundles of cycles closer to the center. 
 
 In a higher portion of the stem fragment an identical method of origin 
 of another orthostichy has been observed. As before, one of the pe- 
 ripheral bundles becomes larger as a result of the "shift" in position of 
 separation of a leaf trace attached to it. The first trace of the new row 
 (ninth) arises from the middle portion of the large peripheral cauline 
 bundle leaving two peripheral cauline strands after it separates. Although 
 only a small portion of the specimen is well-preserved above this region, 
 it may be assumed that the establishment of the ninth pathway of leaf 
 trace divergence into the center of the stem is the same as that for 
 the new orthostichy added at a lower level. The separation of the first 
 trace in the ninth row initiates the new pathway in the outermost cycle, 
 but the gap formed there is opposed only by cauline material in the 
 cycle just within. Since cauline strands gave rise to leaf traces in the 
 eighth row, it seems likely that a similar condition prevails for the ninth 
 orthostichy. 
 
 It is apparent that with the addition of one or more orthostichies of 
 leaves on the stem a change in phyllotaxy results. While the lower part 
 of the specimen showed seven rows of leaves in a 2/7 spiral sequence, 
 this succession of leaves must be replaced by another series in the eight- 
 rowed condition, and still another when nine rows are present. In the 
 eight-rowed spiral condition the only regular phyllotaxy that can 
 prevail is a 1/8 or 3/8 spiral. The 1/8 spiral probably never occurs in 
 the plant kingdom. Because the lower levels of the stem of this species 
 have already passed through a 2/5 sequence, a 3/8 spiral phyllotaxy 
 might be expected for progressively higher levels on the stem. In de- 
 
14 AMERICAN SPECIES OF THE GENUS PSARONIUS 
 
 termining the phyllotaxy from a single cross-section, traces are counted 
 beginning with those closest to the periphery of the stem and progressing 
 inward in order. In a 3/8 phyllotaxy two traces would be by-passed be- 
 tween the lowest trace and the one next in sequence. Two more traces 
 would be passed before the next higher trace is reached. Thus, with 
 three turns about the circumference of the stem, eight traces are en- 
 countered before reaching the one in the same orthostichy as the first. 
 A stem fragment described by Blickle (1940, fig. 1) represents such a 
 spiral condition with a regular 3/8 phyllotaxy. 
 
 In specimen E it is not possible to by-pass two rows of traces between 
 successive leaf traces in the eight-rowed condition, nor is there a 1/8 
 progression. Instead, the nearest phyllotactic spiral that can be de- 
 termined is one which passes through a leaf trace in every other row. It 
 would be impossible to arrive at the eighth leaf of the series (the leaf 
 directly above the starting point) by stopping at intervals separated by 
 arcs of equal magnitude (in this case, arcs of 90° or the distance on the 
 circumference between two rows of leaves). This would result in the 
 spiral passing through the same four rows at all times. Thus, in certain 
 places, the phyllotactic spiral must necessarily pass through adjacent 
 rows or through an interval of three rows. Since nine rows become 
 established on stem fragment E at a higher level, the irregular sequence 
 of trace departure prevalent in the eight-rowed condition does not persist 
 for too great a distance. The new sequence which becomes established 
 appears to be one of 2/9, although it is not possible to fit every trace in 
 this strict sequence. The explanation probably lies in the fact that the 
 leaves at this level of the stem are approaching a whorled condition, each 
 whorl consisting of leaves in alternate rows. In a whorled condition 
 which is well illustrated in the three succeeding stems of this series there 
 may be a slight vertical deviation among members of one whorl. How- 
 ever, in all cases the members of a single whorl arise much closer to a 
 given level than the leaves of either the next lower or next higher whorl. 
 It is obvious that in a nine-rowed state a whorled condition cannot exist 
 unless leaves of two adjacent rows arise at the same level. Although the 
 leaves in specimen E are still spirally arranged even in the nine-rowed 
 state, the retention of the 2/9 phyllotaxy seems to suggest that it is 
 approaching a whorled condition. Leaves encountered in the first turn 
 around the stem may be considered the ones belonging to one whorl, 
 while those passed through in the second turn indicated by the phyllo- 
 tactic fraction probably constitute the other whorl. This, of course, de- 
 pends upon the leaf chosen as the starting point. 
 
 Additional evidence indicating that the upper portion of specimen E 
 is approaching a whorled condition is shown by a specimen from Berry- 
 ville ( specimen F, figs. 7, 31 ) in which eight rows of leaves had become 
 
OBSERVATIONS 15 
 
 whorled by the time the eight-rowed condition was attained. Still another 
 specimen (specimen I, fig. 34) from Berry ville changes from a spiral to 
 a whorled condition in what appears to be the eight-rowed stage. Thus, 
 although all stems may not have reached the condition with whorled 
 leaves at exactly the same stage of development, it seems likely that this 
 condition in the Berryville stems was attained somewhere between seven 
 to ten orthostichies of leaves. If a whorled specimen is cut obliquely it 
 may appear to be a spiral form, but there is no phyllotactic fraction to 
 fit such a condition and it can readily be detected as a poor cross-sectional 
 cut. 
 
 Specimen F with eight rows of traces represents a well-established 
 whorled condition. In such a stem the peripheral bundles have the 
 appearance of being paired, a configuration brought about by the out- 
 line of the band of sclerenchyma around them, and by their association 
 with the trace between them. At a slightly higher level the alternate 
 pairs of peripheral bundles appear to be coupled. The external aspects 
 of this condition will be discussed in a later section. 
 
 In stem specimen F there are approximately eight concentric cycles. 
 Prominent bands composed of sclerenchyma cells are conspicuous 
 throughout the fundamental tissue. Sclerenchyma tissue also often lines 
 the concave faces of the peripheral bundles. In some areas of the cross- 
 sections it develops hooklike extensions around the free edges of some 
 of the vascular bundles ( fig. 31 ) while in other parts, spur-like projections 
 are formed. These projections are most frequently associated with the 
 peripheral bundles and leaf traces, and at higher levels there is an ex- 
 tension of inwardly directed spurs from the sheath of sclerenchyma to 
 form the lateral flanks of the leaf base (fig. 30). There is often the 
 tendency of sclerenchyma between two cycles to fuse through a leaf 
 gap ( fig. 31 ) . The abundance of sclerenchyma varies in amount in 
 corresponding levels of different stems and with different levels in the 
 same stem. 
 
 The next well-preserved specimen in the series (specimen G, figs. 8, 
 32) is one with 12 rows of leaves through most of the length preserved. 
 This stem fragment has an average diameter of approximately 11 X 10 
 cm. exclusive of the root mantle. The rather heavy band of sclerenchyma 
 surrounding the stem emphasizes the paired condition of the peripheral 
 bundles typical of the larger whorled specimens. Certain of the vascular 
 bundles are often distorted or even partly obliterated by a proliferation 
 of parenchyma cells of the ground tissue. In places the proliferation 
 appears to have been rapid, with divisions in the same plane, since a 
 seriation of cells can often be discerned. In other parts, added paren- 
 chyma cells appear to have been formed more slowly, dividing in various 
 planes, resulting in a parenchyma tissue differing very little from the 
 
16 AMERICAN SPECIES OF THE GENUS PSARONIUS 
 
 original ground tissue; the distortion of some of the original structures 
 suggests, however, that proliferation had taken place. As a general rule 
 most of the proliferated parenchyma tissue is found near the periphery 
 of the stem; however, in parts of specimen G proliferation occurred in 
 the center of the stem where seriated parenchyma cells can be observed 
 surrounding the steles. This phenomenon was responsible for the sug- 
 gestion that secondary development took place in Psaronius steles 
 (Moon, 1939). In many parts of the stem, proliferation of parenchyma 
 has partly filled some of the lacunae, an observation which proves that 
 these cavities are normal features of the stem and not the products of 
 decomposition. The meristematic potentialities of parenchyma cells in 
 the stem and in roots will be discussed later in an attempt to explain 
 the nature of the tissue surrounding the inner roots. 
 
 Specimen H (figs. 9, 33) is a stem fragment with more rows of leaves 
 than any other specimen in our collection. Although it is somewhat 
 crushed, 14 rows of traces can be observed arranged in alternating 
 whorls of seven. In the most complete section this stem measures 13.5 X 
 7.5 cm. 
 
 An idealized cross-section of a stem with leaves in whorls is depicted 
 in fig. 16. At the level of this section there are 12 rows of leaf traces 
 (two whorls of six), and seven concentric stelar cycles. As in fig. 12, 
 cauline bundles have been left unshaded while leaf traces are drawn in 
 black. Stippled areas in the peripheral cauline bundles represent the 
 parts of those bundles which are contributed to the outgoing leaf traces. 
 
 Because of the absence of the usual innermost cycle, and because it 
 is almost impossible to determine the configuration of the inner cycle 
 in a stem with seven or more rows of leaves, it is represented in the 
 diagram (fig. 16) by a ring of unshaded bundles. In this stem, the leaf 
 traces originating from the center proceed upward and outward, filling 
 the gap between each successive adjacent pair of cauline bundles. As in 
 smaller stems a portion of a cycle to which a leaf trace is attached moves 
 upward and outward and fills the gap in the next outer cycle before 
 actually passing off the trace. It is not unusual for part of a cycle to be- 
 come attached to the next outer cycle on one side of the gap first, or for 
 one edge of a leaf trace to become separated from the inner cycle before 
 the other. In the diagram, a trace which was transferred from the fourth 
 to the third cycle is attached on the right side only to a cauline bundle 
 of the third cycle, while at its left edge it is attached to cauline bundles 
 of both the third and fourth cycles. Apparently, then, it was set free 
 first at its right edge from the cauline bundle of the fourth cycle. 
 
 P. blicklei, sp. nov. Diagnosis: Stem having three to fourteen or more 
 orthostichies of leaves, the increase occurring at successively higher 
 levels; leaves spirally arranged in lower portions, becoming whorled 
 
OBSERVATIONS 17 
 
 higher in the stem; up to 10 or more concentric stelar cycles in the stem; 
 cells of leaf traces smaller in diameter than cells of other vascular 
 bundles; leaf traces near periphery of stem thinner than other vascular 
 bundles; traces assuming the Stipitopteris configuration; stem surrounded 
 by a definite sclerenchymatous layer; sclerenchyma cells present in small 
 patches and surrounding leaf bases in lower portions of stem, becoming 
 more abundant at progressively higher levels; parenchymatous ground 
 tissue compact with no or few lysigenous cavities, the lacunae becoming 
 more abundant and larger higher in the stem; tannin-like deposits in 
 parenchyma cells of ground tissue of stem and roots, and in secondary 
 parenchyma cells surrounding roots; proliferation of parenchyma cells 
 occasional especially in upper portions of stem; dense mantle of roots 
 below, becoming less extensively developed above; four to six protoxylem 
 groups on the arms of the actinosteles of roots. 
 
 2. Discussion of P. blicklei 
 
 The supposition that the series of Psaronius stem fragments ( A-H ) de- 
 scribed in the preceding section represents the ontogenetic develop- 
 ment of a single plant is perhaps a new and slightly unusual approach in 
 paleobotanical studies. It could be argued that the presumed ontogenetic 
 stages represented by the different stem fragments form a series of closely 
 intergrading species. Based on other reports (Hirmer, 1927; Gillette, 
 1937; Baxter, 1953 ) , six or more species might be named from this group 
 of fragments. A multiplicity of species could be anticipated if earlier 
 classifications (Zeiller, 1890; Stenzel, 1906; Hirmer, 1927) were used, in 
 which the number of orthostichies, shape of leaf traces, number of in- 
 ternal strands, and certain invalid characters of sclerenchyma tissue 
 were relied upon. It is obvious that if two such stems as shown by figs. 
 25 and 33 were to be compared without the knowledge of the series de- 
 scribed above, they would undoubtedly be considered distinct species. It 
 is understandable that previous workers, because of incomplete knowl- 
 edge of the subject, set up different portions of the same plant as separate 
 species. If, with the information obtained from studying stem fragments 
 in the sequence A to H, however, one considers any two adjacent speci- 
 mens of the series, it becomes impossible to make distinctions among 
 the members. If this group of stems is a series of closely related inter- 
 grading species, then their characters overlap to the point where it is 
 impossible to separate them. 
 
 A thorough study of approximately 24 specimens from the Berryville 
 and one from the Dix localities has been made with reference to possible 
 points of species differentiation. It seems appropriate to consider each of 
 these characters with relation to segments from the base of the apex 
 of P. blicklei. 
 
18 AMERICAN SPECIES OF THE GENUS PSARONIUS 
 
 One of the more obvious features of P. blicklei is the great number of 
 lacunae found in the ground tissue at higher levels of the stem. These 
 are much more sparse at lower levels, gradually increasing in number 
 and in size as higher levels. Even lower in the stem than the level shown 
 in fig. 1, the ground tissue appears very homogeneous. At a slightly 
 higher level in the same coal ball (fig. 1), lacunae definitely appear in 
 association with the concavity of the leaf trace, as well as other places 
 in the ground parenchyma. This is not an unusual occurrence, since these 
 cylindrical cavities are lysigenous in origin. Thus these cavities may 
 develop at a later stage in ontogeny, with the exact time of initiation 
 varying in different stems. Because of this, relative numbers of lacunae 
 should be avoided as a key character and caution should be used in 
 assuming that a stem is completely lacking in lacunae if only a small 
 fragment is examined. A consideration of the age of the stem and the 
 level (basal or distal portion) is also necessary. Certain types of preser- 
 vation, such as a particular type of silicification or preservation in iron- 
 stone, may obscure details of lacunae distribution. In all of the speci- 
 mens assigned to P. blicklei ( including all of the fragments used in the 
 series shown by figs. 1-9 and 25-33) these lacunae are present in the 
 stem. 
 
 Small cells containing black or brown tannin-like remains are also pres- 
 ent in all fragments of these stems and in the secondary parenchyma 
 tissue surrounding the roots. All characters of the root zones in all frag- 
 ments of P. blicklei studied seem very similar. 
 
 Sclerenchyma tissue, forming a thin band around the stem and de- 
 parting leaf traces at the lowermost level of the stem, shows a great 
 development throughout the series of specimens comprising this se- 
 quence. At higher levels it is found in association with many of the 
 internal cycles and in cross-sectional aspect assumes many diverse con- 
 figurations. In addition to the surrounding layer of sclerenchyma, every 
 stem shows varying numbers of sclerenchyma cells in the ground tissue, 
 depending on the stage of development. 
 
 Extreme care must be taken when considering phyllotactic characters 
 as a source of specific differentiation. With the aid of specimen E of the 
 series, we see that stems of this species add new rows of leaf traces in 
 the upper levels of the stem. They progress from a 1/3 spiral (specimen 
 A ) to a 14-rowed whorled condition ( specimen H ) in our collection, and 
 it is conceivable that in certain stems even more than seven members 
 per whorl might be attained. Both clockwise and counterclockwise 
 spirals were observed in specimens of P. blicklei, but this difference can- 
 not be used to differentiate two species because it is well known that 
 even within a single plant phyllotactic patterns vary (Esau, 1953). 
 Since spiral forms can change the phyllotactic fraction by the initiation 
 
OBSERVATIONS 19 
 
 of additional orthostichies, and since specimens can become whorled at 
 different levels, the characters of frond arrangement are worthless ex- 
 cept in special instances, e.g., the distichous, tetrastichous, and certain 
 other specimens which will be discussed in detail. Modern tree ferns 
 such as Cyathea also follow the sequence from a spiral to a tightly 
 whorled condition (Ogura, 1927). 
 
 It has also been observed in specimen A that increase in complexity 
 and size of the vascular system may be caused by the addition of a new 
 cycle and by the differentiation of more tracheids within the same ob- 
 conical vascular cylinder. Also, more highly developed polycyclic stems 
 are not necessarily of a different species. The tendency of certain authors 
 (Baxter, 1953) to use numbers of "internal strands" as specific characters 
 is likewise to be avoided, since it should be obvious that these strands 
 are complete cycles of a highly polycyclic stem. Further, leaf traces are 
 given off in the same manner in all fragments and the basic pattern of 
 stelar behavior is similar in all cases. 
 
 Because of intergrading characters within a single specimen, it has 
 been impossible to delimit more than one species in this group of stems. 
 
 P. blicklei could easily have attained a height of 30 feet, a figure sug- 
 gested by previous workers for similar stems. The basal portion is a 
 massive structure consisting of a root mantle which surrounds a small 
 stem base approximately 2.3 X 1.5 cm. in diameter at the lowest pre- 
 served level. Specimen A approximates the length of stem (more than 
 20 cm. ) necessary for the addition of a well-developed cycle, a change 
 which probably required an even greater length at higher levels where 
 the stem becomes more complex. Specimen E required approximately 
 40 cm. for the addition of one orthostichy and 20 cm. for the addition 
 of another. This indicates that even in the same stem the distance may 
 vary as would most probably be the case in different stems of the species. 
 The stem began as a small obconical axis, later becoming encased in a 
 huge mass of roots. By the addition of new cycles, each one more in- 
 ternal than the previous one, it increased in complexity. Each stelar 
 cycle, along with the cortex and sclerenchyma sheath, increases in 
 volume at higher levels by addition of primary tissues derived from the 
 apical meristem. Increased complexity may also be attributed to the 
 addition of new orthostichies of leaf traces and to the effect their path- 
 ways of divergence have on the more internal cycles. Increase in the 
 over-all size of the stem as well as increase in diameter of each con- 
 centric cycle is well illustrated by specimen A, which varies in diameter 
 from 2.3 Xl-5 cm. in the more basal portions to approximately 3 cm. 
 in the upper, well-preserved limits of the coal ball. A similar increase 
 may be observed in specimen E, which at the lower end has a stem 
 diameter of 8 X 6.5 cm., increasing to 10 X 7.5 cm. through a distance 
 
20 AMERICAN SPECIES OF THE GENUS PSARONIUS 
 
 of approximately 50 cm. That many of these stems attained a whorled 
 state has been adequately proved. However, it is not inconceivable that 
 some, for various reasons (e.g., ecological), never survived long enough 
 to become whorled. As higher levels of both spiral and whorled condi- 
 tions are reached, the vertical distance between fronds decreases as well 
 as the horizontal distance. The latter decrease is brought about in part 
 by the addition of new orthostichies. 
 
 All specimens of this group show evidence of a root mantle about the 
 periphery of the stem. Logically, stem fragments with progressively more 
 internal complexity indicate a higher level in the plant and possess 
 fewer and fewer roots at the periphery. It might be argued that this is 
 a phenomenon of preservation; however, the remarkable constancy of 
 finding fewer roots in the distal regions and great masses of them at 
 the base would suggest that preservation is not the only factor involved. 
 Because of this mantle of adventitious roots, the leaf scar arrangement 
 on more basal portions of the plant is often obscured. Comparison with 
 living members of the Cyatheaceae suggests that more distal regions of 
 the stem lacked the conspicuous sheath of roots (see frontispiece). 
 However, this does not rule out the probable occurrence of these roots 
 in fewer numbers growing in sinuous courses down the stem among leaf 
 bases and over newly formed leaf scars. It is obvious that fronds 
 did not extend through an extensive covering of roots, a fact which 
 supports the idea that the upper portions of the stem lacked a very 
 extensive root mantle. That some roots were present between the fronds, 
 while they were still attached to the stem, is supported by the fact that 
 these roots grew down over the leaf scar very soon after the frond 
 dropped away. If these positions of frond abscission were left exposed 
 to the drying conditions of the atmosphere for a great length of time, it 
 is most probable that the outer cortical cells would not have been 
 capable of secondary proliferation which helps interlock the sheath of 
 roots to the stem. 
 
 At higher levels of the stem where leaf scars are exposed, and at 
 lower levels where the root mantle has been removed, the external 
 appearance of the stem varies with the conditions of its spiral or 
 whorled leaf arrangement and with the angle of divergence of the leaf. 
 In some lower portions of the stem the angle of divergence is slight in 
 comparison with that of other stems; however, a greater angle is of ne- 
 cessity substituted at higher levels since the closely crowded condition 
 of the whorls does not allow for even slightly decurrent leaf bases. As 
 the closely arranged whorled condition is attained, the positions of 
 the peripheral cauline bundles become externally prominent and they 
 appear as sinuous strands on the surface of the stem ( fig. 71 ) . Internally, 
 it was observed that at the level where a leaf trace is fused to two pe- 
 
OBSERVATIONS 21 
 
 ripheral cauline bundles the bundles are closer together than they are 
 when an adjacent trace at the next higher node becomes fused with one 
 of those peripheral bundles and another farther around the stem. This 
 explains the apparent paired arrangement of the peripheral bundles in 
 cross-section and is manifested externally by their regular sinuous colum- 
 nar appearance. 
 
 No fronds have yet been found attached to these stems, so it can only 
 be postulated that at the apex of the large whorled forms, two whorls of 
 fully developed fronds may have been retained on the stem at the same 
 time. To have more persisting would seemingly present a serious prob- 
 lem of arrangement because of the closeness of the leaf bases and an 
 entangling of the fronds themselves. It is conceivable, however, that still 
 another whorl of young fronds could have begun the unfolding process. 
 On the basis of compressions of Spiropteris found in shale and certain 
 well-preserved circinately coiled frond tips in coal balls from Berryville, 
 it is suggested that these fronds could have undergone circinate vernation 
 during their development similar to living tree ferns. 
 
 B. Psaronius melanedrus sp. nov. 
 
 1. Description 
 
 Well-preserved specimens of Psaronius stems from the Fleming coal 
 near West Mineral, Kansas, have been found, but not in as great abun- 
 dance as the Berryville representatives. In most of the larger stems, only 
 the centers were preserved or the whole stem was somewhat crushed, 
 making a reconstruction of their vascular system difficult. The quality of 
 preservation in some of the smaller stems, however, was remarkably 
 good. A study of closely spaced sections from a well-preserved basal 
 portion was made in an effort to show (1) the manner in which the 
 small, most internal cycle fuses with the next outer one when passing off 
 a leaf trace, and (2) the way in which this small strand reconstitutes 
 itself before its contribution to the next higher leaf trace in the phyllo- 
 tactic sequence in the cycle immediately outside. The results of these 
 observations were found to be in agreement with those made on the 
 most nearly basal portion of P. blicklei ( fig. 10, specimen A of the Berry- 
 ville series). The series of diagrammatic cross-sections (specimen J, 
 figs. 17, 35) represent the most basal specimen of Psaronius described 
 from the Fleming coal. It shows the beginning of a supposed ontogenetic 
 series of stem fragments (J through M) from Kansas. The stem at the 
 lowest level preserved is characterized by a three-cycled condition (fig. 
 17). There are three orthostichies of leaves: one represented by the leaf 
 base completely surrounded by sclerenchyma in fig. 17A, one represented 
 by the slight "bulge" in the large bundle on the right side, and a third 
 
22 AMERICAN SPECIES OF THE GENUS PSARONIUS 
 
 orthostichy at the position of the gap between the cauline bundles at the 
 lower left corner of fig. 17A. The third or most internal cycle is identi- 
 fied at this level by the small open cylinder in the center of the stem. It 
 is oriented in such a position that, at a higher level, it will fill the gap 
 in the next outer cycle ( cycle 2 ) . Cycle 2, at the same level, bends toward 
 the gap on the lower left side of the first or outermost cycle. At B of fig. 17 
 and in fig. 35, the gap just mentioned in the outermost cycle has been 
 filled by the leaf trace from the second cycle. At this level in the first 
 or outermost cycle, the usual overlapping of gaps to form a typical 
 dictyostelic arrangement is not present. 
 
 At lower and higher levels of the stem, however, such a situation is 
 apparent (fig. 17A, C). It appears that at these levels the outer vascular 
 cylinder is interrupted at the position of two pathways of divergence 
 by leaf gaps. Also, at B (fig. 17) the third cycle has fused with a cauline 
 bundle on either side of the gap in the second cycle, toward which it 
 was migrating at a lower level (fig. 17A). The two strands projecting 
 inward toward the center of the stem at this level (B) are the free 
 edges of the small open cylinder representing the third cycle, with the 
 possible addition of a few vascular elements contributed from the 
 inrolled edges of the next outer cycle. The sharp bend in the right edge 
 of the second cycle in fig. 17B and in fig. 35 is mechanical and should 
 not be confused with the free edges of the central cycle. At C (fig. 17) 
 slightly higher in the stem, the two inwardly projecting strands of the 
 third cycle have separated from the second cycle and fused along their 
 opposite edges. This results once again in a small open cylinder with 
 the convex portion of this strand directed toward the gap between two 
 bundles of the second cycle. At the same level (diagram C) the leaf 
 trace at the lower right side of diagram C, represented in diagrams A 
 and B as a slight bulge between two cauline bundles, has been set free 
 and is particularly enclosed in a sclerenchyma sheath. Also at this level 
 the leaf trace, depicted at the top of the diagrams A and B, has com- 
 pletely departed from the stem. A comparison of this specimen with the 
 early stages of development of P. blicklei (specimen A, fig. 10) makes 
 it clear that the manner of both leaf trace origin and reconstitution of 
 the most internal cycle is the same in these two species in specimens 
 where the internal cycle is very small. 
 
 Specimen J, which is slightly crushed, has cross-sectional measure- 
 ments of 2.3 X 1-8 cm. At the point of leaf trace departure from between 
 two peripheral cauline bundles the vascular strand becomes much smaller 
 in radial thickness (fig. 35), with a corresponding decrease in diameter 
 of tracheids; the leaf traces of leaf bases at a higher level retain this 
 relative thinness. In specimen J the leaf base is a very large structure in 
 relation to the size of the stem, measuring 9 mm. in tangential direction. 
 
OBSERVATIONS 23 
 
 The ground tissue as well as the secondary parenchyma tissue surround- 
 ing the roots have tannin-like deposits in some of their cells. No lacunae, 
 like those described for P. blicklei, are present in this or any of the stems 
 of this series. The stem is surrounded by a mantle of roots which are not 
 well preserved; however, the actinosteles of many individual roots were 
 preserved and bore five or six arms of primary xylem. A very heavy 
 sclerenchyma sheath separates the stem and root zone except on the 
 outer side of a departing leaf base, where it was torn off at the time of 
 abscission of a leaf. Specimen D (fig. 28) shows that certain cells sur- 
 rounding roots of the inner root mantle are produced by proliferation 
 of stem parenchyma cells in the region of leaf bases after leaf abscission. 
 
 Specimen K (fig. 36), considered to be a fragment of a higher level 
 of a stem of this species measures 5 X 2.3 cm. in cross-section. It is 
 characterized by five orthostichies of leaf traces and probably five con- 
 centric cycles of vascular tissue within the stem. A mantle of roots, one 
 to three roots wide, surrounds the stem, but it is likely that more roots 
 were sloughed off the outside of the mantle during preservation. At 
 various places around the stem, adventitious roots connected with 
 the peripheral cauline bundles are seen in oblique cross-section passing 
 through the sclerenchyma sheath of the stem. The portion of the sheath 
 immediately outside of the peripheral cauline bundle at the top of 
 fig. 36 is incomplete, and cortical parenchyma cells inside the stem have 
 proliferated profusely, making a homogeneous connective tissue between 
 the fundamental tissue of the stem and the radiating rows of secondary 
 parenchyma cells surrounding roots of the inner root zone. In all respects, 
 features of this stem indicate that it belongs to the same species as 
 specimen J. Such characters as an increased number of stelar cycles and 
 a larger number of leaf orthostichies suggest that specimen K is from a 
 higher level of the plant than the preceding stem fragment. 
 
 Specimen L (fig. 37) is a fragment of P. melanedrus which, judging 
 from complexity, has reached about the same level of development as 
 the preceding stem (specimen K). Five orthostichies of leaves alternate 
 with five concentric cycles of vascular tissue. This stem also possesses a 
 very thick, well-preserved sclerenchyma sheath. The leaf traces in leaf 
 bases of specimens of K and L (figs. 36, 37) are narrower and are com- 
 posed of tracheids smaller in diameter than those of the vascular bundles 
 of the stem proper. 
 
 Through the length of specimen L (more than 10 cm.), there is no 
 root mantle enclosing the stem. One possible explanation is that a root 
 mantle had once been present and had been shorn away during the 
 processes of preservation. The specimen was examined for evidence to 
 substantiate such a suggestion, but it could not be confirmed. At various 
 levels two to four adventitious roots were observed passing through the 
 
24 AMERICAN SPECIES OF THE GENUS PSARONIUS 
 
 sclerenchyma sheath in their downward and outward course to the pe- 
 riphery of the stem. However, there remained no evidence of any of 
 these roots outside the sheath. Further, there was no evidence of the 
 mechanical separation between the stem and roots with their surrounding 
 secondary parenchyma tissue. 
 
 It may be postulated that in the absence of evidence that a root 
 mantle had been broken away from around this stem, it is a distal portion 
 of a young stem which had not developed a thick zone of roots. The 
 few roots passing through the sclerenchyma sheath might easily have 
 been severed at their points of emergence from the limiting layer of the 
 stem, so that they are not visible in external view. At a later stage of 
 development (at the level represented by this specimen), more adventi- 
 tious roots could have been formed from the peripheral cauline bundles 
 and at their points of exit from the stem the secondary proliferation of 
 parenchyma cells might have started, the combination of these events 
 producing the root mantle. 
 
 Specimen M (fig. 38) which averages 10.5 X 3.8 cm. in diameter and 
 measures over 6 cm. in length, represents a still higher level of a stem 
 of P. melanedrus. It is characterized by five orthostichies of leaf traces 
 which are relatively thin in comparison to the other vascular bundles of 
 the stem (fig. 39, specimen M). The vascular system of this stem frag- 
 ment is composed of six concentric cycles. In this specimen the structure 
 of the ground parenchyma has been preserved better than in any other 
 stems of this species studied. It is apparent that the fundamental tissue 
 consists solely of parenchyma cells except in portions where the limits 
 of a leaf base have been marked off by sclerenchyma tissue. Tannin-like 
 remains are abundant in the fundamental tissue of the stem and leaf 
 bases, and in the secondary parenchyma cells surrounding the roots. The 
 stem is limited by a typical thick band of sclerenchyma tissue, but no 
 sclerenchyma bands are to be found in association with steles as in P. 
 blicklei. 
 
 A conspicuous feature of these stems is the configuration of scleren- 
 chyma tissue around leaf bases at the periphery of the stem. At the 
 lowest level of differentiation of a leaf base, sclerenchyma tissue is first 
 visible as two inwardly projecting strands, one on either side of the 
 trace. Often, at this level, these two strands are not connected to the 
 limiting layer of the stem (fig. 38, bottom). As the leaf base increases 
 in differentiation, small patches of sclerenchyma cells become fused into 
 a flat band on the adaxial face of the leaf base, and at higher levels this 
 becomes continuous with the two inwardly directed projections of 
 sclerenchyma. Still higher, the sclerenchyma tissue around the trace be- 
 comes attached to the layer surrounding the stem, and in cross-section 
 the leaf base appears to be delimited by a thick, nearly rectangular band 
 
OBSERVATIONS 25 
 
 of sclerenchyma. As the leaf base separates from the stem, the inner face 
 retains its square configuration, but when the leaf base is completely 
 separated the outline of the stem at that position is smooth once again. 
 This unusual development of sclerenchyma tissue is characteristic of this 
 species. 
 
 Although the fundamental tissue and outer cycles of certain of the 
 larger stems of P. melanedrus were not well preserved, the vascular 
 bundles of the central portion of specimen N ( fig. 40 ) are f usainized with 
 almost diagrammatic clearness. The morphology of the most internal 
 cycle is of particular interest, since in the length of stem preserved, the 
 most internal cycle is relatively large and well-developed in comparison 
 with the very small central strands of specimens A and J. Although 
 basically this larger cycle functions in the same way as the smaller 
 strands of specimens A-J in giving off leaf traces, there are, because of 
 the greater size of the strand and greater complexity due to the presence 
 of more orthostichies, certain peculiarities which are difficult to interpret 
 unless the morphology of such a type has been studied in detail. A 
 thorough knowledge of the morphology of the stem in question must be 
 had before attempting to distinguish species on the basis of character- 
 istics of stelar morphology. 
 
 The diagrams A-H in fig. 18 represent the three most internal cycles 
 of specimen N at progressively higher levels. In the coal ball a more 
 external cycle was preserved throughout most of the length. This cycle 
 and any others remaining after fossilization have not been diagrammed 
 with the exception of level C, in which part of this cycle has been shown. 
 
 The outermost cycle of the three in the diagram (fig. 18), which will 
 be arbitrarily referred to as cycle 1, is not shaded and is outlined by a 
 solid black line. The second or next inner cycle has been stippled and is 
 outlined with a solid line, and the third or most internal cycle is repre- 
 sented as a solid black band. The position of a small cauline bundle be- 
 longing to the second cycle which has not been well preserved is desig- 
 nated by the stippled area just inside the outermost cycle at the lower 
 left of diagram A. This strand is not outlined. Numbers have been placed 
 to indicate the position of orthostichies in a 2/5 spiral, which represents 
 the phyllotactic sequence of this stem. In all diagrams of this series the 
 same orthostichy is designated by the same number. 
 
 At the orthostichy labeled 1 in fig. ISA, the leaf trace in the outer cycle 
 is the first of those in this series of cross-sections to be given off. In the 
 next inner cycle at this same position, there is a leaf gap between the 
 small, stippled, unoutlined strand, and the tangentially expanded por- 
 tion of that cycle (cycle 2). This gap has been partially closed by the 
 uniting of one edge of the adjacent cauline bundle with one free edge of 
 the leaf trace which is being given off from the third or most internal 
 
26 AMERICAN SPECIES OF THE GENUS PSARONIUS 
 
 cycle. The portion of the third cycle, which at this level is closing the 
 gap in the second cycle, is actually a leaf trace, and unlike the central 
 cycles of many other stems investigated, fusion of the central cycle to the 
 next outer one occurs directly on the free edge of the leaf trace. As a 
 result, no incurved cauline bundle projects inward on that side after 
 fusion. The other side of the trace will fuse with the cauline bundle on 
 the opposite side of the gap at a higher level. At the position of the 
 second orthostichy of leaves in the sequence, a leaf trace is joined at 
 both edges to a cauline bundle in the outer cycle. Here also a leaf gap 
 in the second or next inner cycle is partially closed by a trace being 
 given off from the third cycle. This leaf trace has fused with the cauline 
 bundle at the right side of the gap, and will fuse on the left side and 
 completely close the gap at a higher level. At the position of the third 
 orthostichy, a leaf trace from the second cycle is in the process of filling 
 the gap in the outer cycle and has already fused with the cauline 
 bundle at the left of the gap. The cauline bundle of the second cycle, 
 previously attached to the leaf trace which joined the outer cycle, has 
 been set free, and at a higher level the cauline bundle on the right 
 side will separate from the trace as the trace joins the cauline bundle on 
 the first cycle. At the fourth orthostichy a leaf trace in the second cycle 
 turns upward and outward before filling the leaf gap in the outer cycle. 
 At the position of the fifth orthostichy there is also a leaf gap in the 
 first cycle which will be filled by a leaf trace from the second cycle soon 
 after the gap in the first cycle at the fourth orthostichy is filled. 
 
 Level B represents a somewhat higher level in which the leaf traces 
 shown at level A are somewhat farther out, and although most traces 
 differ only slightly in appearance from those at level A, two striking 
 changes have occurred in the length of the stem between levels A and 
 B. At the position of the first orthostichy, the gap in cycle 2 has been 
 completely closed by a leaf trace from the innermost cycle. At row 3 
 the trace from the second cycle has been completely fused to the outer 
 one leaving a gap in cycle 2. 
 
 At level C the leaf trace attached to the outer cycle of orthostichy 1 
 is being given off to fill the gap in a cycle still more external than cycle 1. 
 Up to this point cycle 1 has been designated as the outermost cycle 
 throughout this series. With this exception, the still more external cycle 
 diagrammed at this level by dashed lines does not enter into the dis- 
 cussion and has only been included at this level for the sake of 
 clarity since one of the other cycles under consideration is, for a time, 
 in close proximity to it while passing off a leaf trace. At the position of 
 the second row of leaves, a leaf trace remains attached to the first cycle 
 between the cauline bundles. The gap in die second cycle which was 
 only partially closed at level B has now been completely closed, and the 
 
OBSERVATIONS 27 
 
 third cycle has released the trace. The slight constriction at the left side 
 of the trace indicates that the leaf trace which has just been contributed 
 to the second cycle has not completely fused with the cauline bundle of 
 the second cycle. At the position of the third orthostichy, a leaf trace is 
 still attached to the outermost cycle and the leaf gap in the second cycle 
 is closed on the left side at this level by a trace contributed from the 
 most internal cycle. At orthostichy 4 a leaf trace passes off to the outer- 
 most cycle. This is evidenced by the slight indentation in the outer 
 cycle and by the position of the cauline bundles on either side of the 
 gap that has been left in the second cycle. Even at this level it is apparent 
 that the convex arm of the most internal strand is bending outward (as 
 well as upward) to close the gap in the second cycle. 
 
 At level D there is a gap in the outer cycle at orthostichy 1 left by the 
 trace seen departing at level C. At orthostichy 2 there is also a gap 
 between the cauline bundles of the outer cycle left by a departing leaf 
 trace. At the fourth orthostichy a leaf trace is attached to the cauline 
 bundles of the outer cycle. At orthostichy 5 a trace from cycle 2 has 
 passed to the first cycle leaving a gap in the second one. The small 
 cauline bundle in the second cycle left after separation of the trace 
 appears isolated in this section because of the presence of a gap on either 
 side of it. 
 
 At level E in the third orthostichy, leaf traces are attached to both the 
 outer and next inner cycles so that no gap is present at this level. A gap 
 in the second cycle of the fourth orthostichy is also partially closed by 
 the fusion of the edge of the most internal cycle with the cauline bundle 
 on the right of the gap. At the point of fusion of these two cycles, there 
 is an increased amount of inrolling of the adjoining edges of the two 
 cycles involved, resulting in an inwardly directed projection. At the 
 position of orthostichy 5, a leaf trace is still attached to the outer cycle 
 and the gap in the second cycle is being closed by a trace given off by 
 the most internal cycle. The small cauline bundle of the second cycle 
 at the lower right of diagram D is not conspicuous at E, since the gap 
 to the left of it has been filled. 
 
 At level F it may be seen that the leaf gap in the first cycle at ortho- 
 stichy 1 present at level E has now been filled, and a gap remains be- 
 tween the cauline bundles of the second cycle. A leaf trace has separated 
 from the outer cycle at orthostichy 3, leaving a gap. The gap in the second 
 cycle at orthostichy 4, which was only partly closed at level E, is now 
 filled with a trace. The inwardly directed projection in that position 
 referred to in the description of level E has separated, and still higher 
 (level G) fuses with the internal strand. No particular significance is 
 attached to this fusion, and at an even higher level (level H) the small 
 strand has become completely "absorbed" by the internal strand. 
 
28 AMERICAN SPECIES OF THE GENUS PSARONIUS 
 
 At level H, in the position of the second orthostichy, the leaf trace 
 from cycle 2 has been given off to the first cycle. At orthostichy 4 a leaf 
 trace has separated from the outermost cycle and at the fifth row the 
 gap in cycle 2 which was only partially closed at level G is now com- 
 pletely filled with a leaf trace. 
 
 At various places along the length of the innermost cycle a very 
 narrow gap appears, thus explaining the apparently two-stranded struc- 
 ture of the central cycle as suggested by figures 18E, F, G. These gaps 
 are in no way connected with leaf traces and are closed simply by the 
 coming together of the vascular tissue on either side. This phenomenon 
 should not be confused with the condition in other stems in which a 
 gap is formed in the innermost cycle after a leaf trace is passed off to the 
 next outer cycle. 
 
 Specimen O (figs. 19, 41) is also preserved only in the center, and 
 probably represents a portion of a stem corresponding in level to speci- 
 men N. It also has five orthostichies, and the paths of divergence of leaf 
 traces are discernable even in the central cycles. As in specimen N, the 
 most internal cycle (fig. 19, diagrammed in solid black) fuses to the 
 next outer one ( stippled ) first along the free edge. The leaf trace of the 
 innermost cycle is present on the edge of that cycle, and no gap is formed 
 when the trace is passed off to the next outer cycle. 
 
 It appears then that the configurations of the inner cycles of specimen 
 N and O are typical for P. melanedrus, but only for certain levels in the 
 stem where a relatively small number of rows of leaves (e.g., 5) is 
 present. In the three-rowed condition the central cycle is constructed 
 like that of similar stages of P. blicklei. Furthermore, in stems with more 
 than five orthostichies the innermost cycle is not a single strand, but is 
 dissected with two or more gaps. It seems likely that the cross-sectional 
 configuration assumed by the central cycle in specimens N and O is a 
 result partly of the number of orthostichies and partly of the cross- 
 sectional length of the strand at those levels. When more orthostichies 
 are present, leaf traces are passed off from the innermost cycle to several 
 gaps in the next outer cycle almost simultaneously. This results in the 
 formation of two or more gaps in the central cycle, which would there- 
 fore no longer be represented by a single strand. In this region of the 
 stem, the innermost cycle is very probably approaching the upper limit 
 of size for an undivided strand of the central cycle, and it is quite prob- 
 able that at only a slight distance above the levels represented by speci- 
 mens N and O a new cycle was added in the center. 
 
 Because of the manner in which traces are contributed from the 
 central cycle to the next outer one in specimens N and O, it is suggested 
 that the central cycle consists of a cauline strand with leaf trace or 
 potential leaf trace material at either end, since in almost all cases 
 
OBSERVATIONS 29 
 
 observed the leaf traces originate at the two edges of the central cycle. 
 The original size of the central cycle is reconstituted after traces separate 
 from it, by differentiation of a larger number of vascular elements. 
 
 The different configurations assumed by the innermost cycle and the 
 different ways in which it contributes leaf traces to outer cycles in 
 various specimens of P. melanedrus might be considered bases for 
 specific distinction. It should be emphasized, however, that the structure 
 of the central cycle depends on its position in the stem. Undoubtedly 
 closer to the base of the plant, this cycle was considerably smaller and 
 supplied traces for fewer orthostichies. Higher in the stem it seems 
 likely that cycles more internal to this one were added after this once 
 most internal cycle assumed a cylindrical configuration with alternating 
 cauline strands and leaf traces. These changes in morphology are there- 
 fore ontogenetic and do not, in this case, account for variability among 
 species. 
 
 Evidence that the structure of the innermost cycle in specimens N and 
 O is only temporary can be found in another stem fragment (specimen 
 P, fig. 42) from the West Mineral locality in which again the inner 
 cycles are the only ones preserved. Pathways of six rows of leaf traces 
 can be observed in these cycles. Thus, this stem fragment probably repre- 
 sents a relatively higher level than is suggested by specimens N and O. 
 The innermost cycle in specimen P is divided by gaps so that at no part 
 of its length in the specimen is it a single strand. In specimens N and O 
 it was seen that the central cycle supplied two traces to the next outer 
 cycle in rather close succession. Specimen P, however, has an additional 
 orthostichy, and in this stem fragment the innermost cycle gives off three 
 traces in a very short vertical distance, thus accounting for its dissected 
 nature. It seems certain that specimen P with a larger number of rows 
 and increased complexity of the central cycle is simply a portion of a 
 stem from a higher level than that of N and O. The differences it exhibits 
 are ontogenetic and are to be expected at upper levels of such a stem. 
 
 Still another stem fragment of P. melanedrus (specimen Q, fig. 43) 
 from the West Mineral locality is a rather crushed specimen 26.8 cm. 
 long. Although the vascular system is considerably distorted, this frag- 
 ment is significant in having retained certain of the features near the 
 periphery of the stem. Six rows of leaves were borne on this stem frag- 
 ment, the largest number observed in P. melanedrus, although certain 
 other large stems which are badly crushed very likely had more ortho- 
 stichies. Characters of the sclerenchyma tissue at the outer limits of the 
 stem and around leaf bases are identical with those of specimens K, L, 
 and M, all of which possess a smaller number of rows of leaves. 
 
 Psaronius melanedrus, sp. nov. Diagnosis: Stem with three to many 
 orthostichies of leaves; leaves spirally arranged in lower portions, prob- 
 
30 AMERICAN SPECIES OF THE GENUS PSARONIUS 
 
 ably becoming whorled higher in the stem; two to many concentric 
 stelar cycles in the stem; radial thickness of xylem in stem vascular 
 bundles ranging from .4 mm. in the basal portions to at least 1.1 mm. 
 higher in the stem; stem surrounded by a heavy sclerenchymatous layer, 
 ranging from .6 mm. in thickness in basal portions to 1.2 or more milli- 
 meters higher in the stem; sclerenchyma cells entirely lacking in funda- 
 mental tissue except that delimiting leaf bases which are enclosed by a 
 heavy band of fibers and are roughly rectangular in cross-section; paren- 
 chymatous ground tissue lacking lysigenous lacunae; tannin-like deposits 
 in parenchyma cells of ground tissue and root mantle; root mantle of 
 varying thickness in different portions of the stem; roots arising from 
 basal portion of stem larger in diameter than roots arising at upper levels; 
 protoxylem at the tip of each of four to six arms of the actinostele of 
 each root; inner roots in radial files. 
 
 2. Discussion of P. melanedrus 
 
 As a result of this study of eleven well-preserved specimens and others 
 with a poorer quality of preservation from the Fleming coal near West 
 Mineral, Kansas, it has become evident that there are no sound criteria 
 for establishing more than one species based on the fragments in this 
 collection. All stems possessed the characteristic heavy sclerenchyma 
 sheath, the tannin-like deposits in the fundamental tissue of the stem 
 and in the secondary parenchyma tissue surrounding the roots, similar 
 characters with regard to the morphology of the stele, and completely 
 lacked sclerenchyma bands among the cycles such as are present in 
 P. blicklei. This species is readily differentiated from P. blicklei because 
 of the complete absence of lysigenous lacunae, the absence of scleren- 
 chyma bands associated with the cycles of the stem, and the typical 
 heavy sclerenchyma sheath forming the almost "square" leaf bases in 
 cross-section. 
 
 Since only a single species is represented, the series shows ontogenetic 
 development through increased size and increased complexity by the 
 addition of more vascular cycles and new orthostichies of leaves. 
 
 In its sequence of development P. melanedrus follows the general 
 pattern found in P. blicklei. Near the base of a plant of P. melanedrus 
 the stem has a small diameter with a small number of orthostichies (as 
 few as three were observed), and with very few stelar cycles. Although 
 only a small number of roots was observed, it is almost certain that a 
 root mantle of considerable radial extent surrounded the stem in this 
 region. As in P. blicklei, increasing size of the stem with increasing 
 vertical distance would necessitate the development of a basal supporting 
 buttress. In P. melanedrus, cylinders of vascular tissue are added in the 
 center of the stem in the higher portions and there is an increase in the 
 
OBSERVATIONS 31 
 
 number of leaf orthostichies. Six vertical rows is the largest number 
 observed among the fairly well-preserved specimens, although a very 
 crushed stem fragment from the West Mineral locality shows indication 
 of having possessed at least eight rows of leaves. 
 
 A very conspicuous feature of this species is the change in the ratio 
 of size of the leaf base with respect to that of the stem. In specimen J 
 (fig. 35), the cross-section dimensions of the leaf base while still attached 
 to the stem are, at the level measured, .9 X -5 cm. compared to the 
 dimensions of the stem at the same level which are 2.3 X 1-8 cm. At 
 progressively higher levels of the stem, leaf bases enlarge considerably 
 in diameter, but the corresponding increment of growth of the stem is 
 much greater. For example, in specimen M (fig. 38), leaf bases average 
 2.7 X 1-2 cm. while cross-sectional dimensions of the stem at the same 
 level are 10.5 X 3.8 cm. This phenomenon occurred also in P. blicklei, 
 but is much more striking in the West Mineral species where, a leaf 
 base at a basal portion has rather large cross-sectional dimensions in 
 relation to the stem at the same level. 
 
 At lower levels of P. blicklei, the sclerenchyma sheath at the stem 
 periphery is nearly cylindrical except where interrupted by the outward 
 passage of a leaf trace. In P. blicklei, as more leaf orthostichies were 
 added, the outline of the sclerenchyma tissue in cross-section assumed 
 a strongly sinuous configuration. Where leaf arrangement became 
 whorled, the sclerenchyma sheath is conspicuously lobed and exhibits a 
 definite radial symmetry. It extends farthest out around the peripheral 
 bundles and dips inward at the position of a leaf trace. Contrary to the 
 condition at lower levels, where a leaf base forms a "bulge" at the 
 periphery of the stem below the point of leaf abscission (fig. 35), in 
 higher regions the dropping off of a leaf resulted in the tearing away 
 of some of the limiting sclerenchyma tissue between two peripheral 
 bundles (figs. 16-18). As a result the outline of the sclerenchyma tissue 
 in higher regions of the stem of P. blicklei is never smoothly cylindrical. 
 
 In P. blicklei the "paired" arrangement of the peripheral bundles, 
 emphasized by the configuration of the sclerenchyma tissue, is a result 
 of the presence of a rather large number (eight or more) of rows of 
 leaves. This arrangement begins to become apparent, however, even 
 with a smaller number of rows. On the other hand, for various reasons, 
 P. melanedrus retains a considerably smoother outline at the periphery 
 of the stem with relatively shallow indentations in cross-sectional out- 
 line of the sclerenchyma sheath. As a general rule, portions of stems of 
 P. melanedrus have a larger stem diameter than sections of P. blicklei, 
 which have the same number of orthostichies of leaves. Similarly, a 
 portion of a stem of P. melanedrus equal in diameter to a given section 
 of P. blicklei usually has fewer rows of leaves and a smaller number of 
 
32 AMERICAN SPECIES OF THE GENUS PSARONIUS 
 
 stelar cycles. Consequently there is less "crowding" of leaf traces and 
 peripheral bundles in the outermost cycle. The leaf traces, therefore, 
 generally possess a larger ratio of tangential to radial diameter when 
 compared to P. blicklei. Admittedly, leaf traces of the latter species 
 appear flat in cross-section just above the point of release from the pe- 
 ripheral bundles. Higher in the stem, however, just below the region 
 where parts of the trace are lost due to leaf abscission, there is a tan- 
 gential to radial diameter ratio of 1 or even less. Thus, the leaf base 
 delimited by sclerenchyma tissue surrounding a leaf trace of P. blicklei 
 has a relatively smaller tangential diameter compared to leaf bases of 
 P. melanedrus. The sclerenchyma tissue on the adaxial face and flanks 
 of a leaf base of P. blicklei makes a rather sharp interruption in the cross- 
 sectional outline of the sclerenchyma surrounding the stem. The indenta- 
 tions of the peripheral sclerenchyma tissue become more shallow as the 
 leaf base of P. blicklei progresses upward and outward. In P. melanedrus, 
 because of the relatively greater tangential dimension, the leaf base 
 appears to have a smoother limiting layer of sclerenchyma because the 
 indentations in the peripheral sclerenchyma caused by the departing 
 leaf base are much shallower. Furthermore, because of the greater ratio 
 of stem circumference to number of rows of leaves in P. melanedrus 
 compared with that in P. blicklei, there is a relatively larger portion of 
 the peripheral sclerenchyma tissue which is uninterrupted by leaf bases. 
 The arrangement of adventitious roots in P. melanedrus is of par- 
 ticular interest because of the conspicuous radial alignment of the roots 
 embedded in parenchyma (figs. 38, 43, specimens M and Q). The places 
 of origin of adventitious root traces may possibly have been along 
 nearly vertical rows on the peripheral cauline bundles, or on more in- 
 ternal cauline bundles. This would result in radiating files of roots as 
 those from above grew downward and outward, outside of those arising 
 at lower levels. The radial alignment, however, would have become 
 distorted if the roots were suspended in the air. It seems certain that 
 the direction of proliferation of parenchyma from the surface of the 
 stem and from the periphery of each of the roots themselves must have 
 played a large part in confining the roots to so regular a pattern, 
 especially if they did not originate in vertical rows on the cauline bundles. 
 Thus the roots were embedded in the secondary parenchyma soon after 
 their emergence from the stem if the secondary parenchyma aids in 
 their alignment. 
 
 C. Psaronius chasei sp. nov. 
 
 1. Description 
 Specimen R ( fig. 44 ) is a portion of a stem from the Berryville locality 
 
OBSERVATIONS 33 
 
 which is approximately 28 cm. long. The stem is a large one with many 
 rows of leaves. Only one half of the specimen was preserved in the coal 
 ball, but this measures 9x5 cm. in cross-section in its slightly crushed 
 condition. The positions of at least six rows of leaves may be observed 
 about the periphery of the stem, but because of the incompleteness of 
 the specimen the exact number for the entire stem could not be de- 
 termined. As demonstrated for P. blicklei, however, the number of rows 
 of leaf traces cannot be used for determination of species. 
 
 The stem in specimen R is surrounded by a definite sclerenchyma 
 sheath more similar in thickness and configuration to that of P. blicklei 
 than to P. melanedrus. Numerous adventitious roots, connected with the 
 peripheral cauline bundles and vascular bundles of the second cycle 
 (fig. 45), can be seen traversing the layer of sclerenchyma tissue in a 
 downward and outward course. From their frequency and ease of 
 discernment in the sclerenchyma sheath, it might appear that these roots 
 continue through the sheath in a vertical direction for some distance 
 rather than passing through it at the same angle assumed in the funda- 
 mental tissue. Such is not the case, however, since a closer observation of 
 well-preserved ground tissue, especially on the outside of peripheral 
 cauline bundles, shows these roots in equal abundance. A very small 
 quantity of sclerenchyma tissue is usually located between the cauline 
 bundles of the second and third cycles. Tannin cells are present in 
 abundance in the fundamental tissue of the stem, in the roots, and in 
 secondary parenchyma tissue surrounding these roots. No lysigenous 
 lacunae like those in P. blicklei are found in the fundamental tissue. The 
 vascular tissue of the stem is composed of eight or more concentric 
 cycles, an exact determination being impossible because of the incom- 
 pleteness of the specimen. The xylem cells of the leaf traces at the pe- 
 riphery of the stem are relatively smaller than those of the cauline 
 bundles, and in over-all appearance these leaf traces are thinner than 
 the other vascular bundles. 
 
 The stem is enclosed in a mantle of adventitious roots and secondary 
 parenchyma which surrounds the roots of the inner root zone. Although 
 many large free roots are found adjacent to this specimen in the coal 
 ball, at no place along its length can they be observed emerging from 
 the inner root zone and it is not known where, or even if, they were con- 
 nected to this stem. It is interesting to note that in the peripheral regions 
 of the inner root zone, the secondary parenchyma cells seem to exceed 
 the inner roots in radial extent and the parenchyma cells may be observed 
 in clusters forming lobed and often slightly crushed masses of tissue 
 (fig. 46). In this particular specimen the secondary parenchyma often 
 extends as much as 5 mm. beyond the most externally placed root. 
 
 Another specimen (fig. 47), which is apparently the same species, 
 
34 AMERICAN SPECIES OF THE GENUS PSARONIUS 
 
 was discovered at the Calhoun collecting locality in southeastern Illinois. 
 Calhoun is approximately the same stratigraphic level as the Berryville 
 collecting site. This stem was a large whorled form with nine rows of 
 leaves on only a little over one-half of the cross-section. The preserved 
 half of the stem fragment measures 9.5 X 3.7 cm. in cross-section and is 
 approximately 16 cm. long. Tannin-like deposits are present in cells of 
 the fundamental tissues of the stem and roots and in the secondary 
 parenchyma surrounding the roots. The fundamental tissue of the stem 
 lacks lysigenous lacunae. A band of sclerenchyma tissue surrounds the 
 stem and there is apparently some of this tissue between the bundles of 
 the second and third cycles. 
 
 Psaronius chasei, sp. nov. Diagnosis: Stem probably having three to 
 many orthostichies of leaves; leaves probably spirally arranged in lower 
 portions, becoming whorled higher in the stem; up to ten or more con- 
 centric stelar cycles in the stem; stem surrounded by a definite scleren- 
 chymatous layer; sclerenchyma bands between second and third vascular 
 cycles; parenchymatous ground tissue with no lacunae; tannin-like de- 
 posits in parenchyma cells of the ground tissue of stem concentrated 
 near vascular bundles and in roots and secondary parenchyma sur- 
 rounding roots. 
 
 A stem from the St. Wendel locality in southwestern Indiana shows 
 characters very similar to P. chasei. One specimen showing about one- 
 half of the stem (specimen S, fig. 48), and another fragment showing 
 two peripheral bundles (fig. 49) were used along with a more poorly 
 preserved specimen for this study. It has been determined that no 
 lysigenous cavities characteristic of P. blicklei are present in this stem. 
 Tannin-like deposits are quite abundant in the cells of the fundamental 
 tissue of the stem and roots and in the secondary parenchyma sur- 
 rounding the roots. In stems from this locality the tannin-like deposits 
 appear to be found in even greater abundance in the cells of the funda- 
 mental tissue adjacent to the vascular bundles of the stem ( fig. 49 ) . The 
 presence of tannin is emphasized more in figure 49 because a different 
 type of preservation of that substance gives it a very black appearance. 
 The stem is surrounded by a zone of sclerenchyma tissue at the pe- 
 riphery, and external to this is a mantle of adventitious roots. The struc- 
 ture of individual roots was not well preserved. 
 
 2. Discussion of P. chasei 
 
 Although the specimens (figs. 48, 49) from St. Wendel were not 
 preserved in their entirety, certain features of these stems indicate 
 affinities with P. chasei from the Berryville locality (specimen R, fig. 
 44). Lysigenous lacunae characteristic of P. blicklei are not present in 
 either P. chasei or the St. Wendel stems. Tannin-like deposits in the 
 
OBSERVATIONS 35 
 
 cells of the fundamental tissue are very abundant about the vascular 
 bundles of the stem. The type of preservation merely emphasizes these 
 deposits in the St. Wendel specimen (fig. 49), however, the same rela- 
 tive abundance of such cells in this position may be demonstrated for 
 specimens of P. chosei from Berryville. P. blicklei does not exhibit a 
 concentration of such cells containing tannin-like materials in associa- 
 tion with the vascular tissue. The sclerenchyma sheaths of P. chosei and 
 specimen S from St. Wendel are very similar in appearance with no such 
 strikingly different features in either as might be found for example in 
 P. melanedrus from West Mineral. 
 
 Although P. chasei and the specimens from St. Wendell both exhibit 
 similar characters, it should be recalled that the St. Wendel stems con- 
 sidered were not well-preserved ones and thus it is possible that when 
 better material is available specific differences may become apparent. 
 
 With the introduction of P. chasei as a new species, two separate and 
 distinct species of Psaronius are known from the Berryville locality. It 
 is further known that at least two anatomical organ genera of psaroni- 
 aceous petioles (Stipitopteris and Stewartiopteris) exist both at the 
 Berryville and St. Wendel localities. Although not reported in any 
 publication, specimens of Stipitopteris have been found at Berryville 
 and numerous specimens of the stems of P. blicklei show the typical 
 stipitopterid configuration of the leaf base on the surface of the stem. 
 
 Above its point of separation from the peripheral bundles, a leaf trace 
 assumes a horseshoe-shaped cross-sectional configuration, the concavity 
 and inrolled ends of which are directed toward the center of the stem. 
 At higher levels it may retain this configuration of the petiole ( Stewarti- 
 opteris) or by further tangential inrolling and fusion of the free ends 
 it may develop into the stipitopterid type. Specimens of P. blicklei may 
 be directly associated with the stipitopteroid type of petiole, since all but 
 two of the series of stems in figs. 25-33 (Berryville series) show at least 
 one stipitopterid configuration on their surface. The absence of this type 
 of petiolar configuration in the two smallest stems can probably be ex- 
 plained by the variation in configuration of leaf bases at various levels 
 in a plant. In more nearly apical portions of taller stems, a leaf base is 
 situated in a depression flanked by portions of the stem containing 
 peripheral bundles, while in more nearly basal portions of stems a leaf 
 base "bulges" out before its separation. As a result, abscission of leaves 
 at lower levels on the stem tore off a trace closer to the level of its 
 separation from the peripheral bundles and the stipitopterid configura- 
 tion was probably lost. Other specimens of P. blicklei not figured also 
 show at various places along their length the stipitopterid configuration. 
 
 Thus far the stipitopterid type of leaf trace has never been observed 
 on specimens of P. chasei. Further, at levels much higher than the leaf 
 
36 AMERICAN SPECIES OF THE GENUS PSARONIUS 
 
 trace seen at the upper left of fig. 44, the two remaining "j-shaped" 
 strands have still not fused along their inrolled adaxial edges. It is 
 possible, then, that the stipitopterid type of petiole trace, which is more 
 commonly found, is associated with the stems of P. blicklei. Also, P. 
 blicklei is by far the more abundant of the two Psaronius stem species 
 from Berryville. The less frequently discovered Stewartiopteris petiole 
 could perhaps be the petiole of the less common P. chasei stem. 
 
 D. Description of Other Previously Undescribed Psaronii 
 
 A stem fragment (specimen T), approximately 13 cm. long, from the 
 New Delta, Illinois, locality, is shown in fig. 50. This stem measures 
 approximately 14 X 6.5 cm. in cross-section. The fundamental tissue of 
 the stem contains no lysigenous cavities, but cells bearing dark tannin- 
 like deposits are present. These deposits are not concentrated in the 
 cells adjacent to the vascular tissue as are those of P. chasei. This speci- 
 men is thought to be a whorled form, but the arrangement of the leaves 
 at the levels preserved is slightly distorted due to the addition of a new 
 orthostichy to the specimen. The length of the specimen is not sufficient 
 to allow for a readjustment necessitated by the addition of an ortho- 
 stichy, and thus presents a regular phyllotaxy. The method of addition 
 of this new orthostichy is identical with the addition of new rows of 
 leaves described for P. blicklei. Six or seven concentric cycles of vascular 
 tissue comprise the stem stele and the whole stem is surrounded by a 
 thick root mantle, the roots of which are often in fairly even radiating 
 rows. The sclerenchyma sheath which completely surrounds the stem, 
 except at points of leaf trace departure, is, for the length of the stem 
 preserved, very similar in character to that of the West Mineral species, 
 P. melanedrus. 
 
 Since characters of the sclerenchyma tissue are in agreement with 
 those of P. melanedrus, and since the cells with tannin-like deposits are 
 not clustered about the vascular bundles as in P. chasei, this specimen 
 has been tentatively assigned to P. melanedrus rather than to P. chasei. 
 The fact that the stratigraphic level of the New Delta locality is lower 
 than that of Berryville, St. Wendel, and Dix localities (nearer the strati- 
 graphic level of the West Mineral locality) may be of some significance. 
 
 One of the more unusual specimens of Psaronius described from 
 America is specimen U ( figs. 25, 51 ) . This stem fragment represents the 
 only known member of the distichous psaronii discovered in America. 
 It is extremely unfortunate that the stem is not of the coal-ball type of 
 preservation. Instead of being preserved, the fundamental tissue of the 
 stem has been replaced by a fine grained claystone. The stem, which is 
 from a collecting locality near Wilmington, Illinois, measures approxi- 
 mately 15 cm. in length and averages 7.6 X 6.8 cm. in diameter. 
 
OBSERVATIONS 37 
 
 Outlines of a few roots may be observed about the periphery of the 
 stem, but it is not clear whether these roots were in the process of 
 traversing the cortex in their downward and outward course or whether 
 this is actually the most internal limit of the root mantle. Because of 
 the poorly preserved nature of the specimen, only fragments of the 
 sclerenchyma sheath remain on the sides of departing leaf bases (figs. 
 20, 51 ) . It is probable that the stem was completely enclosed by such 
 a sclerenchyma sheath as is usually the case. In figs. 20F and 51, three 
 cavities of possible lysigenous origin, as were common in P. blicklei, are 
 located close to the vascular bundle and in the concavity of the depart- 
 ing leaf trace. The positions of the vascular bundles are easily discernible 
 because of a different type of preservation in this part of the stem. 
 Calcite and other minerals are found in these regions along with some 
 carbonaceous remains of tracheids, so that in certain instances, poorly 
 preserved xylem cells may be observed. When peels of the stem are 
 made, such cells are evident in the vascular bundles; otherwise only the 
 rough outlines of the bundles are present. 
 
 The vascular system of the stem is composed of five concentric stelar 
 cycles, each of which is composed of two tangentially elongated cauline 
 bundles which alternate with two pathways of leaf trace divergence. 
 Leaf traces originate in the most internal cycle and are passed off in 
 their upward and outward course to each successively more external 
 cycle. At fig. 20A, the most internal cycle (cycle 5) is represented by 
 the small vascular strand within cycle 4. At level B the most internal 
 cycle has fused with the cauline bundle on the right side of the leaf 
 gap in the fourth cycle, and at level C, at a still higher position in the 
 stem, the inrolled end of the right side which had fused, has now broken 
 free. The left side of the gap is not closed until level D is reached. At 
 level E, the trace formed just below and on the other side of the stem 
 from the one just discussed is being given off from the fourth to the third 
 cycle. At level F, the fourth cycle has almost set free the trace just 
 mentioned. Throughout the length of the stem preserved two traces, one 
 from either side, have reached the periphery of the stem. Had the stem 
 fragment been preserved in sufficient length it would have been possible 
 to observe the origin of these traces from the most internal cycle. 
 
 With respect to the morphology of leaf trace formation and the 
 manner in which the most internal vascular cycle is reconstituted after 
 contributing to a leaf trace, this stem agrees in its ontogenetic develop- 
 ment with spiral forms of Psaronius blicklei, P. melanedrus, and a 
 whorled form to be discussed next. 
 
 The configurations of the free edges of the cauline bundles adjacent to 
 the leaf gap oriented at the bottom of cycle 2 (fig. 20F) are the result 
 of contribution of vascular tissue by the peripheral cauline bundles to 
 the second cycle. 
 
38 AMERICAN SPECIES OF THE GENUS PSARONIUS 
 
 The discovery of this specimen is of particular interest because it 
 represents the first distichous specimen of the genus to be described 
 from America. Although Stenzel (1906) and others have figured distich- 
 ous specimens, none could be compared in detail with this specimen 
 because none could be obtained from Europe for examination and 
 partially because the rather unusual type of preservation of this specimen 
 obscures histological details. Stenzel (1906) shows a specimen of P. 
 gutbieri, the leaf trace of which supposedly divides into two halves as 
 it reaches the leaf base. Hirmer (1927) shows a photograph of P. simplex 
 with the same configuration of the leaf trace. Other European specimens, 
 e.g., P. simplex f. integer figured by Stenzel, show as many as three leaf 
 traces on one side of the stem already broken free from the peripheral 
 cauline bundles, but none completely set free from the stem in its own 
 leaf base. None of these has divided into two strands. As far as general 
 morphology of the leaf trace is concerned, the American specimen re- 
 sembles the European species P. musaeformis. 
 
 This stem does not exhibit good histological details so often necessary 
 in delimiting species of Psaronius, and it is possible that better preserved 
 specimens will be discovered. As a result there will once again be 
 difficulties of comparison between European and American specimens. 
 For this reason, no new specific epithet has been designated for this 
 distichous member. 
 
 Specimen V (figs. 21, 52) is a stem also preserved in claystone and 
 also from the Wilmington, Illinois, locality. Specimen V measures 24 
 cm. long. At the lowermost levels preserved, the stem averages 9.5 X 
 8.6 cm. in diameter and approximately 7.5 X 6.5 cm. at higher levels. 
 This decrease in circumference is probably the result of a combination 
 of causes: (1) more crushing of the stem has occurred at higher levels; 
 (2) some of the specimen is missing at these levels so that an accurate 
 estimation of the actual diameter is made difficult; and (3) the stem 
 appears to vary slightly in circumference because of the various con- 
 figurations of the vascular bundles within the stem, depending on the 
 stage of departure of the leaf traces. The maximum diameter is ap- 
 parently reached as the leaf traces are set free from cauline bundles. 
 Measurements of the higher levels were all made either above or below 
 the points of separation of leaf traces, and at those levels the stem 
 appears almost hexagonal in cross-section. 
 
 The vascular system of the stem is composed of five or six concentric 
 cycles, and six orthostichies of leaves are arranged in alternating whorls 
 of three each. No sclerenchyma sheath could be demonstrated to com- 
 pletely surround the stem. Since sclerenchyma tissue does cut in from 
 the periphery of the stem to enclose leaf bases, however, there is reason 
 to assume that a typical sheath was present. Because of the type of 
 
OBSERVATIONS 39 
 
 preservation none of the fundamental tissue was preserved, and it is 
 not known whether this stem possessed the lysigenous lacunae previously 
 mentioned for P. blicklei. In fig. 21 the series of cross-sections of the 
 stem from A (lowest level) through I (highest level) shows the method 
 of leaf trace departure and the orientation of the vascular cylinders. 
 Preservation of the outline of vascular bundles was not so distinct in 
 this stem as in the distichous specimen, making it difficult to determine 
 in cross-section which is a fissure in the claystone and which is actually 
 part of a stelar cycle. 
 
 One region which might be confusing in considering the anatomy of 
 specimen V is exemplified by level G ( fig. 21 ) . In the pathway corre- 
 sponding with the orthostichy at the upper left portion of cross-section 
 G, the fifth cycle fills the gap in the fourth cycle and two free edges 
 project toward the center of the stem. In the pathways at either side, 
 the fourth cycle "bulges" outward before passing off traces to cycle 3. 
 A similar stage is not apparent in other pathways at this level but higher 
 in the stem the same configuration is assumed. 
 
 Figure 22C is a series of diagrams representing progressive changes 
 in the fourth ( stippled ) and fifth ( black ) cycles in an idealized situation 
 where leaf traces in alternating pathways are always at the same stage 
 of development. In fig. 22A the innermost cycle has not yet filled the 
 gaps in the next outer one. The cycle corresponding to the fifth cycle of 
 fig. 22B has filled the gaps in the fourth cycle, with the free edges of 
 the fifth cycle projecting toward the center of the stem. In fig. 22C the 
 free arms have separated and adjacent strands are shown migrating 
 toward each other before fusing. A similar series of steps then takes 
 place in the alternate set of pathways. 
 
 Peripheral cauline bundles contribute to the next inner cycle in the 
 manner described for other specimens of Psaronius. 
 
 It is quite possible that this specimen represents a form of Psaronius 
 which, after becoming whorled at the six-rowed condition, never added 
 further orthostichies of leaves. This would place it artificially with such 
 forms as the distichous and tetrastichous psaronii, which, as far as is 
 known at the present time, never added more orthostichies after reaching 
 the typical number for the respective group. The whorled condition in 
 this specimen occurs slightly earlier in ontogeny than any form yet de- 
 scribed from America. As was suggested for P. blicklei, the whorled 
 condition occurs only after a certain number of rows is attained and 
 when that number of orthostichies is divisible by two. Members of 
 each whorl then include alternate leaf traces as seen in cross-section. 
 
 It is possible, however, that the stem represented by specimen V, like 
 P. blicklei, continued to add orthostichies after becoming whorled. 
 Specimen V is not of sufficient length to determine this. If it is one of 
 
40 AMERICAN SPECIES OF THE GENUS PSARONIUS 
 
 the latter group, then it merely became whorled earlier in ontogeny with 
 two fewer orthostichies of leaves than is the usual condition for P. 
 blicklei. 
 
 Another stem (specimen W, fig. 53) from the Coal City, Illinois, 
 locality, also is preserved in claystone. The tracheids of the cycles of this 
 stem are very clearly defined, and slightly oblique longitudinal sections 
 of adventitious roots are visible in connection with the peripheral cauline 
 bundles. A portion of the sclerenchyma sheath is also in evidence, cutting 
 in about the leaf bases, but no histological details of the fundamental 
 tissue were preserved. The stem measures 10.5 cm. in length and averages 
 8.3 X 8.2 cm. in cross-section. The stele is composed of six concentric 
 cycles of vascular tissue and the phyllotaxy is a 2/7 spiral. An examina- 
 tion of close cross-sections of this specimen has given evidence that the 
 sixth or most internal cycle at the levels preserved in this specimen is 
 a very small strand, but that morphologically it resembles the internal 
 strand of the P. melanedrus as depicted in fig. 17. In fig. 53 the con- 
 figuration of two of the cauline bundles at the top of cycle 2 suggests 
 that the peripheral cauline bundles of the outermost cycle have con- 
 tributed vascular tissue to this cycle. 
 
 Concerning the general morphology of leaf trace origin and reconsti- 
 tution of the most internal cycle, this stem exhibits characters similar to 
 the general patterns previously described for the genus (e.g., those of 
 P. blicklei and P. melanedrus) . It is interesting to note that specimen W 
 from the same general locality as the hexastichous specimen just de- 
 scribed ( specimen V ) has seven rows of leaves but has not, at this level, 
 become whorled, nor could it become whorled until an even number of 
 orthostichies is reached. This could support the supposition that the 
 hexastichous specimen is a different type of stem which never attained 
 more than a six-rowed condition. 
 
 A point that needs further consideration when better material is avail- 
 able is the striking contrast of the distances between cycles of these 
 specimens (specimens U, V, W) with those of, for example, P. blicklei. 
 The much greater distance between cycles is typical of all three speci- 
 mens described from the Wilmington and Coal City area, however, not 
 enough is known yet of stems of these groups to postulate a possible 
 significance of this feature. 
 
 E. P. brasUiensis Brongniart 
 
 1. Additional observations 
 
 One section approximately 1 cm. in thickness of the specimen used by 
 Brongniart ( 1872 ) in his description of P. brasUiensis was obtained from 
 a museum at Rio de Janeiro, Brazil. This silicified specimen was col- 
 
OBSERVATIONS 41 
 
 lected from the Chapado do Jaboti, Estado do Maranhao, and is thought 
 to be lower Permian in age. The cross-section of the stem exclusive of 
 the root zone measures 9.5 X 8.7 cm. A study of the changes in leaf 
 trace development through one centimeter of the stem gives some in- 
 formation concerning the morphology of the outermost vascular bundles. 
 
 At the periphery of the stem, four peripheral cauline bundles alternate 
 with four pathways of leaf trace divergence. The leaves are arranged 
 in alternating whorls of two, with one member of each whorl being 
 given off slightly below the one directly opposite it (figs. 54, 55). The 
 second cycle is also composed of four smaller cauline bundles alternating 
 with the four pathways of leaf trace divergence. Within the center there 
 are many other vascular strands which, in this specimen, appear to be 
 organized in a manner similar to the concentric cycles common in 
 Psoronius. 
 
 Leaf traces are recognizable in the pathways of divergence as far in 
 as the third cycle. In the pathways of divergence on the left and right 
 of the cross-section (fig. 54), leaf traces are attached on either side of 
 the gaps to the cauline bundles of the second cycle. Immediately out- 
 side these traces, on either side of the pathway of divergence, are strands 
 of vascular tissue which are contributions to the leaf trace from the 
 peripheral cauline bundles. It is suggested that at a somewhat higher 
 level, the pathway of divergence widened between these two strands, 
 allowing the leaf trace to pass upward between them. These strands, at 
 this level, fuse on either side of the leaf trace before it departs and is 
 enclosed in the sclerenchyma tissue of the leaf base. At the pathway of 
 divergence directed toward the bottom of the figure, a leaf trace at the 
 periphery of the stem is partially enclosed in a leaf base by a scle- 
 renchyma sheath cutting in diagonally on either side of the trace. The 
 manner of contribution of peripheral cauline bundles to leaf trace for- 
 mation is best observed in the pathway at the top of figs. 54, 55. A leaf 
 trace was set free from the cauline bundles of cycle 2 at lower levels. 
 The edges of the peripheral cauline bundles away from the gap in the 
 outer cycle become much elongated. The inrolled portion of each one 
 of these separates, and after a rotation of 180° the free edges formerly 
 attached to the peripheral cauline bundles fuse with the free edges of 
 the outgoing trace. Cauline bundles of more internal cycles were 
 observed to contribute to leaf traces in the same manner (fig. 54, cycle 
 3, lower orthostichy ) . Thus, in this stem, the leaf traces separate from 
 the cauline bundles, receive a contribution from the next outer pair of 
 cauline bundles, and fuse with the next outer cycle in that sequence. 
 After separating from the second cycle, the leaf traces receive a contribu- 
 tion from the peripheral cauline bundles but do not fuse with them. 
 
 It appears, then, that the stem consists essentially of four series of 
 
42 AMERICAN SPECIES OF THE GENUS PSARONIUS 
 
 cauline bundles alternating with four persistent pathways of leaf trace di- 
 vergence. The gaps are filled only at points where leaf traces are attached 
 to the cycles. The number of strands composing the more internal cycle 
 is greatly increased because each time a trace is given off to the next 
 outer cycle, the cauline bundles of that cycle contribute a small strand 
 of vascular tissue to each side of the leaf trace. 
 
 Large leaf bases, almost square in outline, enclose the leaf trace. The 
 sclerenchyma sheath completely surrounding the stem is relatively thick 
 and is broken only at points of leaf trace departure. Internal bands of 
 sclerenchyma may be observed between the leaf trace from the second 
 cycle and the peripheral cauline bundles. 
 
 Strands of vascular tissue are found accompanying the leaf trace on the 
 adaxial side in the concavity. Some of these strands appear to accompany 
 the leaf trace in its upward and outward course from the center of the 
 stem ( fig. 55, c.f . leaf traces on right and left side of the stem ) . A close 
 observation of the leaf traces at the top and on the left side of fig. 54 
 seems to indicate the possibility that some of these vascular strands may 
 have separated from the free edges of the horseshoe-shaped leaf trace 
 in the leaf base, as is common in many modern fern genera (e.g., 
 Cibotium ) . 
 
 Adventitious roots which are connected with the peripheral cauline 
 bundles may be observed in figs. 54 and 55, passing through the cortex 
 and sclerenchyma sheath. The stem is surrounded by a root mantle 
 averaging approximately 5 cm. in thickness. The steles of the roots are 
 polyarch but further details were not sufficiently preserved to allow a 
 more thorough investigation. The roots are enclosed in a mass of sec- 
 ondary parenchyma. Evidences of proliferation from the leaf bases may 
 be observed in figs. 54 and 55, and farther out in the root zone these 
 secondary parenchyma cells are visible among the roots. 
 
 2. Discussion of P. brasiliensis 
 
 Although the stem and root mantle of P. brasiliensis have been pre- 
 viously described (Brongniart, 1872) and figured by Zeiller (1840), 
 Pelourde (1908), Scott (1920), and Hirmer (1927), the new sections 
 obtained from the museum at Rio de Janeiro along with a re-examination 
 of the cross section of this stem (fig. 56), as figured in Zeiller (1890), 
 has yielded new information concerning the morphology of P. brasiliensis. 
 According to Zeiller (1890), the strands in the concavity of the leaf 
 trace were pits formed during the process of preservation. For the speci- 
 men just described they are known to be vascular tissue, part of which 
 comes from the stem as the leaf trace departs into the petiole and, 
 possibly, part from strands which separate from the inrolled edges of 
 the leaf trace. This configuration of the trace appears to resemble some- 
 
OBSERVATIONS 43 
 
 what the dissected traces of certain modern marattiaceous ferns (e.g., 
 Angiopteris) , and might be considered a phylogenetic advancement 
 within the genus. 
 
 The section (fig. 56) used by Zeiller (1890) shows in detail the exact 
 method of contribution of vascular elements from the peripheral cauline 
 bundles to leaf traces above their points of separation from the second 
 cycle. The section investigated by the author is not of sufficient length 
 to substantiate the method of contribution, although there is evidence 
 that it was similar (note the peripheral cauline bundle in the lower right 
 of fig. 55). When one edge of the peripheral cauline bundle elongates, it 
 also inrolls so far that the free, incurved edge fuses with the middle of 
 the peripheral cauline bundle ( fig. 56P a ) . A separation is formed in the 
 resulting loop at a point adjacent to the leaf gap (fig. 56P 4 ), and the 
 adaxial portion "swings" inward with the point of fusion to the peripheral 
 cauline bundle as the "hinge" (fig. 56P 3 ). After a rotation of almost 
 180°, the strand separates at the "hinge" and fuses to the edge of the 
 leaf trace (fig. 56P 3 ) which has been released from the second cycle. 
 
 Although the method of contribution from the peripheral cauline 
 bundles to leaf traces is unique in this species, some specimens of P. 
 blicklei show in part a similar situation in which the leaf trace and the 
 second cycle do not directly fuse with the peripheral cauline bundles 
 (described in section on vascular system). In other characters (e.g., 
 persistent leaf gaps, cauline bundles, root zone) P. brasiliensis corre- 
 sponds to the usual structure of the genus. 
 
 F. Histological Details 
 
 In most of the stems observed, the fundamental tissue of the stem is 
 composed of large, thin-walled parenchyma cells (figs. 39, 57, 58). This 
 tissue is relatively compact and the cells appear to be of uniform size. 
 A variation of the usual condition of the ground parenchyma is found in 
 P. pertusus (called P. lacunosus by Blickle, 1940). In this species, the 
 cells are arranged in a very loose network with large intercellular spaces 
 between them (fig. 82). This type of cortex is very similar in appearance 
 to the so-called "aerenchyma" of the cortex of Psaronius roots in their 
 more distal portions. In such stems the net-like arrangement of the 
 cortical cells becomes more compact near the vascular strands and at 
 the periphery of the stem near the surrounding sclerenchyma sheath. 
 
 Certain of these cells (figs. 39, 57, 58) contain black or dark brown 
 deposits which may possibly be tannin-like substances collected in the 
 cells when the plant was living rather than artifacts accumulated during 
 preservation. These cellular contents superficially resemble tannin de- 
 posits in the cortical cells of members of the Marattiaceae (West, 1915), 
 
44 AMERICAN SPECIES OF THE GENUS PSARONIUS 
 
 and may also be found in the cortex of roots and in the secondary 
 parenchyma surrounding roots. The tannin-filled cells are generally thin- 
 walled and do not differ in size or shape from surrounding parenchyma 
 cells. They are fewer in number at very low levels of specimen A, which 
 is probably very near the bottom of the stem, but increase in number 
 higher in the stem. 
 
 It has been suggested by West (1915) that tannin cells of the 
 Marattiaceae hold waste products of the plant and that numbers of 
 them increase with age at a particular level of the stem. This could also 
 be true of such cells in Psaronius. However, since age determination 
 of particular levels of fossil plants having no secondary wood is very 
 difficult, it can only be suggested that a similar condition could have 
 prevailed in these fossil forms. 
 
 In P. blicklei (figs. 57, 59, 73) large lysigenous lacunae extend verti- 
 cally through varying vertical distances, depending upon the numbers 
 that have become coalesced throughout the length of the stem. These 
 cavities are found scattered through the ground parenchyma of the stem 
 and petiole bases ( fig. 73 ) . They are fewer in number at lower levels of 
 stems of this species (specimen A), but become much more abundant 
 at higher levels and probably increase in number at a particular level 
 as the stem becomes older. They are considered to be of lysigenous origin 
 because there is no evidence of an epithelial layer of secretory cells 
 lining the cavities, and because there is evidence of partially broken- 
 down cell walls within them. 
 
 The cavities superficially resemble mucilage canals described for mem- 
 bers of the Marattiaceae (Kuhn, 1889; West, 1915). West states that most 
 of these canals of the Marattiaceae are protogenetic, lysigenous mucilage 
 canals; however, in the extant ferns some protogenetic schizo-lysigenous 
 canals are found, for example, in petioles of Angiopteris evecta. Whether 
 the cavities in the cortex of Psaronius ever contained mucilage is not 
 definitely known, as most of these cavities are filled only with calcite in 
 the petrifactions studied. Occasionally a light brown mucilage-like de- 
 posit is found. The cells between cavities apparently break down to allow 
 as many as five or more such cavities to coalesce either horizontally, 
 vertically, or in both directions. The function of such structures or their 
 contents, if any, is not known, nor is it known for members of the 
 Marattiaceae. 
 
 Cortical cells at the periphery of the stem grade more or less gradually 
 into apparently thicker-walled cells with large lumens, and finally into 
 fibrous cells of the sheath (fig. 39). A sclerenchyma sheath of very 
 thick-walled fibers completely encircles the outside of the stems except 
 at points of leaf trace departure (fig. 31), where it cuts in around the 
 adaxial face of leaf bases. Isolated groups of fibers are found scattered 
 
OBSERVATIONS 45 
 
 throughout the fundamental tissue of most stems (fig. 59), and well- 
 defined bands of sclerenchyma tissue are found in various configurations 
 in association with the steles of certain species (e.g., P. blicklei) and 
 with all leaf bases (figs. 31, 59). Further evidence that the large cavities 
 in the ground parenchyma of stems of P. blicklei were formed during 
 the life of the plant, and are not due simply to disintegration of the 
 plant before or during fossilization, is the internal sclerenchyma present 
 around these cavities which appears in cross-section as finger-like pro- 
 jections between them. The fact that these fibers are found in such con- 
 figurations would seem to indicate that they differentiated later in 
 ontogeny from cortical parenchyma cells. Thus, at a particular level, a 
 stem could have developed more of these sclerenchyma cells with in- 
 creasing age. It has also been determined that the radial thickness of 
 this limiting layer of sclerenchyma tissue increases at higher levels of 
 stems. In P. melanedrus it ranges from .6 mm. in more nearly basal 
 portions to 1.2 or more millimeters at higher levels. Intermixed with 
 these fibers were parenchyma cells which proliferated to form the sec- 
 ondary parenchyma tissue of the root mantle. The cells of the funda- 
 mental tissue of the stem were also capable of further divisions, resulting 
 in radiating rows of cells. In certain instances the production of these 
 secondary cells might be a reaction to wounding (figs. 60, 61). Often 
 very large areas are filled with such cells to the complete obliteration 
 of vascular tissue, etc. As shown by fig. 61, a leaf trace has been pushed 
 completely to one side by this secondary growth. At other places in the 
 stem of P. blicklei these proliferating cells grow into and often com- 
 pletely close, at least for a certain distance, the vertically extending 
 lacunae (fig. 62). No useful function has yet been ascribed to this type 
 of secondary growth, and the reason that these cavities should be closed 
 again after their formation is not accounted for. It seems that once di- 
 vision was initiated in these cells, it continued until mechanical com- 
 paction caused the process to cease. Although the majority of the 
 divisions are in the same plane causing the formation of radiating rows 
 of cells, other divisions did take place resulting in some instances in a 
 large mass of irregularly arranged cells. Certain parenchyma cells of 
 some stems have proliferated about the vascular bundles of some stems 
 (fig. 60). The function of this tissue is likewise unknown. Its presence in 
 some stems has caused some erroneous interpretations because certain 
 authors (Pohlig, 1910; Schuster, 1911; Sterzel, 1912; and Moon, 1939) 
 have interpreted the radial rows of cells in this position as secondary 
 xylem. The name "Xylopsaronius" was applied by European authors to 
 forms supposedly having secondary wood. 
 
 The vascular bundles consist of primary xylem with xylem parenchyma 
 cells intermixed among the tracheids (figs. 57, 58, 63, 64). Protoxylem 
 
46 AMERICAN SPECIES OF THE GENUS PSARONIUS 
 
 cells are situated in groups along the internal surface of these vascular 
 strands (fig. 64) and thickenings on the walls of metaxylem elements are 
 scalariform (fig. 65). Tracheids of leaf traces are smaller in diameter 
 than those of the remainder of the vascular bundles of the stem ( fig. 39 ) . 
 A zone of thin-walled cells composing the phloem completely surrounds 
 the xylem strands (figs. 57, 64). In cross section, certain cells of the 
 phloem zone can be seen as a discontinuous layer, usually one cell in 
 width. The cells are large and cylindrical and have been referred to as 
 sieve cells (Bower, 1926) (figs. 57, 63). These appear to have definite 
 walls although no sieve areas were preserved in the longitudinal sections 
 observed (fig. 58). These cells appear to be in layers two or three cells 
 in thickness at the free edges of the vascular bundles. They are prominent 
 in this region of the phloem in all stems where preservation allowed 
 study of the phloem zone, with the exception of lower levels of specimen 
 A. Probably sieve cells of so small a stem did not develop as great di- 
 ameters and are not so readily discernible. No layer which could be 
 referred to as endodermis has been observed. 
 
 III. GENERAL DISCUSSION 
 
 A. Vascular System 
 
 Although earlier workers have described the stelar anatomy of 
 Psaronius (Zeiller, 1890; Rudolph, 1906; Stenzel, 1906; Hirmer, 1927; 
 Schoute, 1938; and Gillette, 1937), much was left undone because their 
 specimens were too fragmentary and not well preserved, or because a 
 thorough study of the material had not been made. This resulted in an 
 incomplete and often mistaken understanding of the morphology of the 
 plant. Zeiller ( 1890 ) , for example, showed that stems may increase in 
 complexity at higher levels; nevertheless, Stenzel (1906) published a 
 key to species of the genus using numbers of inner cauline bundles as a 
 prominent criterion of specific differentiation. It was further recognized 
 by Hirmer ( 1927 ) that living free ferns increased in size, and the ortho- 
 stichies of leaves could increase in number as occurs in members of the 
 Cyatheaceae (Christ, 1910). Yet, Hirmer (1927) continued to distin- 
 guish, in his classification, between psaronii polystichi spirales and 
 psaronii polystichi verticillati and did not use the knowledge of extant 
 forms in the interpretation of the morphology of fossil ferns. 
 
 Specimen A (P. blicklei, fig. 25) from Berryville, Illinois, is the most 
 nearly basal level of Psaronius discovered in America. The vascular 
 system in this specimen is composed of two concentric cycles of vascular 
 tissue. The outer cycle at the levels depicted (fig. 10A-L) usually 
 
GENERAL DISCUSSION 47 
 
 exhibits two leaf gaps, and, because at higher levels more gaps are 
 formed by the addition of more orthostichies, it is referred to as a 
 dictyostele. The most internal cycle presents more of a problem in in- 
 terpretation. At levels C, E, F, G, J, and L two leaf gaps are present in 
 the cross-section. At levels interspersed among those of the ascending 
 series just mentioned, the strand of the central cycle bears a closer re- 
 semblance to a simple siphonostele. In fact, at levels K and L the third 
 cycle is so small it might be interpreted as a protostele. Variations in 
 the definitions of stelar types and a consideration of the ontogeny of 
 many fern steles from protostele to siphonostele to dictyostele would 
 allow several interpretations of these vascular cylinders. In the discussion 
 to follow, describing the ontogeny of each newly added cycle, the 
 cycles will be referred to as going through stages of development from 
 protostele to dictyostele. 
 
 At the lowest preserved level of specimen A (figs. 1, 25) the typical 
 pattern of development for a dictyostele is well established. It is modi- 
 fied only by a greater number of orthostichies of leaves at higher levels, 
 but even this increase does not alter the established pattern of older 
 orthostichies. 
 
 The gaps in such a dichtyostele are, at least in the outer cycles, pri- 
 marily leaf gaps, and contrary to the condition in many modern ferns 
 are closed only when the next leaf trace to be given off above fuses at 
 a higher level with the cauline bundles on either side of the gap. The 
 first leaf trace originating at a certain position in the innermost cycle is 
 of considerable significance, since it determines the position of one of 
 the gaps in that cycle. Similarly, with the addition of a new orthostichy 
 the trace supplying the first leaf in that row determines the position of 
 the gap in the first cycle corresponding to the added orthostichy. Gaps 
 in the second, third, and even more internal cycles are formed in the 
 same way. Thus, all persistent gaps in all of the cycles were formed 
 originally by a leaf trace arising from certain positions; and once formed, 
 these gaps are closed for only short vertical distances by other leaf 
 traces originating from farther within. Some inrolling of the cauline 
 material into the gap to meet an approaching leaf trace does occur. 
 Actually then, cauline material does help fill the gap. This partial gap 
 closure would not occur, however, had not a leaf trace been approaching 
 the gap from below and farther in toward the center of the stem. 
 
 At levels close to the base of a stem such as those diagrammed in fig. 
 10, the addition of a new cycle can be clearly followed and has been 
 described in detail for P. blicklei. For P. blicklei and for P. melanedrus 
 (fig. 17) at the lowest levels described, the newest and most internal 
 cycle is always reconstituted, after contributing to a leaf trace, by bundles 
 of cauline origin which separate from the next outer cycle after a leaf 
 
48 AMERICAN SPECIES OF THE GENUS PSARONIUS 
 
 trace from the most internal cycle has fused with it (fig. 17A-C). At 
 higher levels, as the stem attains a greater diameter, the internal cycle 
 also expands by further differentiation of vascular elements and becomes 
 a well-developed cylindrical dictyostele (fig. 17A). 
 
 A slight variation of this pattern was observed in parts of P. melane- 
 drus and in specimen W (fig. 53). In P. melanedrus (fig. 18) the most 
 internal cycle is not reconstituted by contributions from cauline bundles 
 on either side of the trace to the center of the stem. Rather, the traces 
 arise from the edges of the most internal strand, making this strand 
 theoretically all potential leaf trace material. Theoretically this cycle 
 would cease to exist if the apical meristem had not continued to form 
 additional vascular elements, resulting in continual expansion of the 
 innermost cycle. It has been suggested in the description of P. melanedrus 
 that at higher levels the most internal cycle sends off so many traces at 
 approximately the same level that its S-shaped configuration, seen in 
 cross-section, is no longer efficient and is replaced by a cylindrical 
 dictyostele. Further, if additional cycles were to be added to such a 
 stelar system their addition internally would necessitate a change from 
 the S-shaped configuration of this strand. 
 
 That the pattern set by P. blicklei (fig. 10) and P. melanedrus (fig. 
 17) for more basal levels may occur at higher levels and even in whorled 
 forms is confirmed by specimen U (figs. 21, 22, 52). Where three traces 
 are given off almost simultaneously from the most internal cycle, six 
 cauline bundles migrate upward and inward to the center of the stem 
 as the three traces are passed off to the next outer cycle. In reality, six 
 cauline bundles may not be obvious at one level because leaf traces, in 
 departing from the most internal cycle, may lag slightly one behind the 
 other. Thus, it is apparent that in spiral forms with perhaps a 2/7 
 phyllotaxy or in some of the larger whorled forms where, for example, 
 seven members per whorl are given off at approximately the same 
 level, there is considerable dissection of the more internal cycles and the 
 anastomoses of these strands present diverse cross-sectional configura- 
 tions which are very difficult to interpret. 
 
 When the second cycle containing a leaf trace joins the outermost 
 cycle, it fuses along the convex faces of the peripheral cauline bundles. 
 At the level where a leaf trace is released from the second cycle, that 
 cycle separates taking with it on either side of the resulting gap, the 
 free edge of each of the two peripheral cauline bundles (fig. 14A-C, 
 fig. 66A-D ) . In addition to fusion between adjacent cycles at the path- 
 ways of leaf trace divergence each cycle, at its point of origin, is joined 
 to the one immediately outside of it. 
 
 Once a leaf gap is established in a cycle (except the most internal 
 cycle) it persists in that position throughout the length of the stem. The 
 
GENERAL DISCUSSION 49 
 
 basic pattern of each cycle is a cylinder of isolated strands which are 
 temporarily joined at various places by leaf traces passed on from inner 
 cycles. Since leaf traces are not released from a cycle before they fuse 
 to the next outer one, two cycles may be joined for a time at the position 
 of a pathway of trace divergence (fig. 67). Unlike other ferns where 
 leaf gaps have a limited vertical extent, the presence of gaps of indefinite 
 length is the normal condition in Psaronkis. 
 
 In addition to the gaps in cycles at the pathways of leaf trace de- 
 parture, gaps may occur at other positions in more internal vascular 
 bundles. These are closed at higher levels without having performed any 
 apparent function (fig. 18E-H). Occasionally a strand of vascular tissue 
 may become separated from the inner or outer face of a cauline bundle 
 (fig. 68) and at a higher level fuse with either the next inner or outer 
 cycle. This characteristic is of more common occurrence in some stems 
 (e.g., C.B. 1221, P. chosei) than in others, but not in so striking a 
 manner that it could serve as a basis for specific differentiation. 
 
 The radial thickness of vascular bundles increases at higher levels in 
 the stem. This is true of P. blicklei and P. melanedrus, and when other 
 species are more completely known from base to apex it is likely that 
 they will exhibit the same characteristic. 
 
 Although the simplest steles observed for species of Psaronkis are 
 composed of two cycles, the method of addition of other more internal 
 cycles suggest a possible, still more simple stelar, construction at more 
 nearly basal levels. In specimens observed having two cycles, the inner 
 one, when considered by itself, has a structure identical with that of 
 solenosteles in many modern fern genera (e.g., Adiantum and Denn- 
 staedtia), although throughout most of its preserved length it is dis- 
 sected by two gaps. Like steles of the ferns mentioned above, the leaf 
 gap in the second cycle of Psaronius is closed simply by the arching over 
 and fusion of portions of the cycle from either side of the gap. This is 
 different from the closure of the gap in all cycles except the central one. 
 In other cycles, gaps are mainly closed by leaf traces and are reformed 
 as the trace leaves the cycle. The method of initiation of the third cycle 
 from the second in specimen A (fig. 10) suggests that, still lower in the 
 stem, the second cycle had a similar origin from the first or outermost 
 cycle. If this occurred, then the gaps in the single cycle which would be 
 present must have been closed simply by a fusion of the edges of the 
 stele on either side of the gap. 
 
 The configuration of the stele at a still lower level can only be postu- 
 lated, although some evidence suggests that it was protostelic. The pre- 
 dominance of a protostelic sporeling stage in ferns would indicate a 
 similar basal structure in Psaronius. The structure and "behavior" of the 
 most internal cycle in basal portions may be interpreted as showing 
 
50 AMERICAN SPECIES OF THE GENUS PSARONIUS 
 
 changes from an essentially protostelic strand to a dictyostele with many 
 gaps. The third cycle in specimen A (fig. 10) is only a slightly curved 
 strand as seen in cross-section and in the process of contributing a leaf 
 trace to the second cycle, no gap is left in it because the trace originates 
 from the edge of the strand ( see also fig. 13 ) . It is not until higher levels 
 are attained that this internal cycle expands and gaps are formed by the 
 separation of traces along various parts of its cylindrical structure. 
 Theoretically, then, if the outermost cycle were traced far enough down 
 in the stem its solenostelic structure would be expected to grade into 
 a protostele closely resembling the third cycle in specimen A. 
 
 Ideally each cycle in Psaronius, from its point of origin to higher 
 levels, recapitulates the sequence of development of each of the outer 
 cycles. Actually, however, the situation is complicated at higher levels 
 by the increased number of orthostichies. For example, should a new 
 cycle be added in a portion of the stem in which leaves are arranged in 
 alternating whorls of four, it is necessary for the central cylinder to 
 supply four traces to the next outer cycle almost simultaneously. For 
 that reason the protostelic nature of the innermost cycle, at its point of 
 origin, is modified very quickly. In addition to stems having spirally 
 arranged leaves at very low levels (e.g., such as a portion with three 
 rows of traces), the distichous forms of the genus show such a recapitu- 
 lation throughout their entire length. 
 
 The supposed ontogenetic development of the vascular system of 
 Psaronius from a protostele aids considerably in attempts to trace the 
 ancestry of those forms. A stem segment with 14 or possibly more rows 
 of leaves possesses an extremely complex vascular system, yet it has been 
 shown that such a stelar system developed ontogenetically from a simple 
 solenostele and perhaps even a protostele near the base of the plant. A 
 similar sequence of increasing complexity from a protostele to a poly- 
 cyclic dictyostele could represent the derivation of psaroniaceous forms 
 from a presumed protostelic ancestor. 
 
 A leaf trace is first observed in the outermost cycles of lower levels 
 of stems as a "bulge" of vascular tissue which becomes progressively 
 more pronounced at higher levels until it is set free from the outer cycle, 
 first on one side and then on the other. Where it is free from the cycle 
 the edges of the leaf trace become slightly inrolled on the adaxial side. 
 At higher levels of some stems, e.g., P. blicklei, where there are perhaps 
 seven leaf traces per whorl, the crowding together of the leaf traces 
 departing at the periphery of the stem results in a very noticeable change 
 in configuration of the trace in early stages of departure. When a trace 
 is being given off from the second cycle to the peripheral cauline 
 bundles, it has the appearance of an almost flat tangentially elongated 
 strand of vascular tissue (fig. 66A, B). At a higher level (fig. 66C), 
 
GENERAL DISCUSSION 51 
 
 when it has joined with the peripheral cauline bundles and become free 
 of the second cycle, the edges of the trace may precede the remainder 
 of the trace outward for a short distance up the stem, giving the strand 
 a concave configuration, the concavity of which is oriented in an adaxial 
 position. This configuration is retained after the trace is released, but 
 at still higher levels the middle of the trace migrates outward and up- 
 ward more quickly, causing the free edges of the trace to lag slightly 
 behind when seen in cross-section. At this level, the trace then assumes 
 the typical horseshoe-shaped cross-sectional configuration. A feature 
 exhibited by this leaf trace which is supposedly a primitive character 
 in fern anatomy is the undissected nature of the leaf trace in most 
 species of the genus. 
 
 A variation of the manner of contribution of the peripheral cauline 
 bundles to leaf traces and inner cauline bundles of the second cycle is 
 diagrammed in fig. 23A, B. At A it can be seen that the portion of the 
 peripheral cauline bundle (diagrammed in solid black), which will be 
 contributed to the trace and cauline bundles of the next inner cycle, has 
 separated before the leaf trace from the inner cycle fuses with the 
 peripheral cauline bundle. The sclerenchyma tissue around the peripheral 
 cauline bundles has extended between this strand (fig. 23A) and the 
 rest of the peripheral cauline bundle. The strand diagrammed in solid 
 black separates in the middle and the outer portion next to the pathway 
 of leaf trace divergence fuses with the leaf trace; the other portion fuses 
 to the next internal cycle and contributes to it. This method of con- 
 tribution occurs on both sides of the pathway of leaf trace divergence as 
 suggested by fig. 23B. The leaf trace then passes out between the 
 peripheral cauline bundles without fusing directly with either of them. 
 The contribution of vascular tissue by the peripheral cauline bundles 
 to the leaf traces, and to the cauline bundles of the next more internal 
 cycle, connects all three of these at lower levels. In a sense, then, this is 
 a double contribution (to leaf traces and to more internal cauline 
 bundles) by the peripheral cauline bundles, but it is slightly modified 
 from the usual condition in Psaronius. 
 
 This series, showing the method of contribution of peripheral bundles, 
 was observed in three stem specimens which in all other internal features 
 resemble P. blicklei. It is suggested that this method of leaf trace de- 
 parture is not a specific character to be used in delimiting new members 
 of the genus, because it is not uniform for all leaf traces around the 
 stem. Rather, it represents a difference in "timing" of contribution of 
 vascular tissue to departing traces and more internal cycles. 
 
 Zeiller (1890) figures a series of seven diagrams at successively higher 
 levels of a leaf trace departing from P. infarctus. According to Zeiller, 
 two small vascular strands separate from the peripheral cauline bundles 
 
 4NWHWTY V IbLINOli 
 MMWK 
 
52 AMERICAN SPECIES OF THE GENUS PSARONIUS 
 
 on either side of the leaf gap and migrate upward and into the pathway 
 of leaf trace divergence. These strands then supposedly fuse along the 
 free edges adjacent to one another and are sent off from the stem as a 
 typical horseshoe-shaped trace. The series of diagrams figured by Zeiller 
 to accompany the explanation does not clearly show this kind of trace 
 formation. As the two strands migrate away from the peripheral cauline 
 bundles on either side of the gap toward the pathway of leaf trace 
 divergence, they pass in close proximity to the cauline bundles of cycle 
 2. Although not shown by Zeiller's figure, it is possible that these small 
 strands which become much elongated at this level fused with a trace 
 given off from the next inner cycle before their departure into a leaf 
 base. If this were the case, then the method of leaf trace departure de- 
 scribed by Zeiller (1890) would be identical to that just described for 
 P. blicklei. 
 
 It has been observed that leaf traces generally originate in the most 
 internal cycle and migrate upward and outward, fusing in turn with each 
 more external cycle on either side of the leaf gap. In specimen F, a stem 
 having approximately eight cycles, this condition is slightly modified. 
 Fig. 24 is a series of diagrams of the third through the sixth cycles. At 
 level A, a leaf trace from a still more internal cycle than those mentioned 
 (3-6) is fused at the right side of the pathway of divergence with a 
 cauline bundle of cycle 5. The left side of the trace still attached to an 
 inner cycle is lagging behind the right side in its upward and outward 
 course, so that it has not yet joined cycles 6 or 5. Also, at level A it 
 should be noted that the inrolled edge of the cauline bundle of cycle 4 
 on the left side of the gap is much elongated and at the next higher 
 level (level B) it has incurved so much that it has touched and fused 
 with the cauline bundle of cycle 5. At level C this inwardly projecting 
 strand on the left side of the pathway of divergence separates from 
 cycle 5, taking with it as a contribution the inrolled end of a strand 
 belonging to cycle 5. Also, between levels B and C the inwardly pro- 
 jecting leaf trace attached at level B to the right side of the gap in 
 cycle 5 has migrated up and across the leaf gap so that it now fuses on 
 the convex edge of the vascular segment contributed by cycle 5 on the 
 left side of the gap. This strand then separates at level D, taking with 
 it the inrolled end contributed by cycle 5. This strand at higher levels 
 migrates back to the center of the stem and has no further part in the 
 development of this leaf trace. The inrolled edge of the cauline bundle 
 on the right side of the gap in the fourth cycle has become separated, 
 and at C the leaf trace and the cauline bundle of the fourth cycle to 
 which it is attached on the right edge fuse to the small strand along 
 the convex face. Vascular elements have been contributed to the left 
 part of the leaf trace by the inrolled portions. At level D the trace has 
 
GENERAL DISCUSSION 53 
 
 assumed a more normal configuration, although the concavity is still 
 toward the outside of the stem and cycle 5 has not yet released this trace 
 on the right side of the gap. The small strand diagrammed in solid black 
 which originated from the fourth cycle at the right side of the gap 
 contributes some vascular elements to the leaf trace and some to the 
 cauline bundle in the fifth cycle. These elements are added before the 
 fifth cycle and the leaf trace fuse to the fourth cycle on the right side of 
 the gap. Although the configuration of leaf trace and cauline bundles in 
 specimen F is identical to the normal condition, the method of contribu- 
 tion of material from the cauline bundles to the leaf trace and to inner 
 cycles differs. The end result is exactly the same as in the visual sequence 
 of trace departure, except that addition of vascular tissue to the leaf 
 trace and inner cycles occurs in a different manner. Leaf trace divergence 
 of this type is present in several specimens observed, differing only in 
 details from the type of departure described above. It occurs quite often 
 (at least four times) in the length of specimen F preserved in the coal 
 ball, while at other points on the circumference traces are given off 
 simultaneously in the usual manner. For this reason, and because these 
 stems exhibiting the modified type of trace formation resemble P. blicklei 
 in all other respects, this type of trace formation should not be con- 
 sidered as basis for specific differentiation. 
 
 At still higher levels of the petioles the inrolled edges of the adaxial 
 side may inroll toward one another and actually fuse along their convex 
 edges (fig. 69). The two projecting free edges then separate from the 
 outer cylinder of vascular tissue, resulting in a complete ring of vascular 
 tissue with a small strand in the upper portion of the center of this ring 
 (Lenz, 1942). This configuration, which is that assumed by petioles of 
 the organ genus Stipitopteris, is visible on the outside of many stems 
 where the root zone has been removed (fig. 70). At the most basal level 
 of P. blicklei, for example, the departing leaf bases are not protected at 
 the point of emergence from the stem by projecting cauline bundles as 
 is the case at higher levels. This petiole base was shipped off closer to 
 the stem than those at higher levels. Because of this, more of the petiole 
 was removed to a position at a level lower in the petiole base than the 
 position where the stipitopterid configuration is normally assumed. Thus, 
 the stipitopterid configuration could not be observed on the stem surface 
 as is possible at higher levels. Petiole traces of other species of Psaronius 
 never fuse along their inrolled edges, and thus in face view of the stem 
 they exhibit an open horseshoe-shaped bundle scar, the concavity of 
 which is oriented upward (fig. 71). 
 
 Each leaf trace in its upward course also bends outward at various 
 angles, depending upon the nearness of leaf traces below it and those to 
 be given off at higher levels. Figure 72 shows the angle of leaf divergence 
 
54 AMERICAN SPECIES OF THE GENUS PSARONIUS 
 
 of a specimen of P. blicklei having fourteen rows of leaves. Even with 
 this large number of orthostichies, the angle of divergence is less than 
 that depicted by Gillette ( 1937 ) for P. septan gulatus. It is possible that 
 the traces of such a stem as specimen H may have had a somewhat 
 greater angle of divergence, while smaller stems send off traces at a 
 much smaller angle (e.g., specimen A). The most obvious conclusion 
 and one which is supported by actual measurements is that there is a 
 correlation between levels of the stem and angles of leaf trace diver- 
 gence. This would scarcely justify indiscriminate use of the angle of 
 leaf trace divergence as a specific criterion in naming new species of 
 Psaronius. 
 
 Figures 70 and 73 give some idea of the variation in size and con- 
 figuration of the leaf trace scars, and can serve to point out some of the 
 hazards of considering shapes and measurements of these scars for use 
 as specific characters. Figures 70 and 73 are both of P. blicklei, the leaf 
 trace of fig. 70 measuring 1.75 X 1-2 cm. being at a much lower level 
 of the stem than fig. 73 measuring 4 X 1-4 cm. The measurements given 
 are for specific specimens and can vary in other stems and also with the 
 level in the petiole at which abscission took place. Higher up on the 
 stem, where the leaves become whorled, the surface configurations often 
 depend upon the proximity of the whorls of leaves one above the other. 
 Several specimens of silicified stem fragments of Psaronius stripped of 
 their root mantles and showing external features of the stem have been 
 examined with regard to leaf scar configurations. In the specimen in 
 fig. 74 the vertical distance between leaves in the same orthostichy is 
 greater than in the specimen in fig. 75. When a leaf abscised from the 
 stem such as that in fig. 74, the stripping-off of the limiting layer re- 
 sulted in a somewhat pointed configuration below each trace. In such a 
 stem as is shown in fig. 75, where there is a much shorter distance 
 between leaves one above the other in the orthostichy, the vertical 
 dimension of the petiole is restricted and leaf abscission formed a 
 different type of scar. The configuration of leaf scars depends not only 
 on the actural position of leaf abscission, the level on the stem at which 
 abscission occurs, and the effects of proliferation of parenchyma at the 
 points of abscission, but also to a great extent on the degree of weather- 
 ing of the fossil. Where there has been a considerable amount of 
 weathering the sclerenchyma bands around the individual leaf scars 
 becomes confluent in two vertical lines flanking the orthostichy. The 
 prominence of the positions of peripheral cauline bundles also varies 
 in stems of the same species, at lower levels presenting a smooth, flat 
 surface between leaf bases and at higher levels developing externally 
 into bulging, sinuous bands among orthostichies ( fig. 71 ) . 
 
 Many species of Caulopteris and Megaphyton have been named on 
 
GENERAL DISCUSSION 55 
 
 superficial leaf scar characters and on size distinctions. Since anatomi- 
 cally it is not possible to substantiate these numbers of species, it is 
 suggested that naming on the questionable characters just mentioned 
 would, in part, account for this discrepancy. 
 
 At higher levels when the leaf bases are free from the stem they are 
 known to assume at least two configurations, Stipitopteris (Lenz, 1942; 
 Morgan and Delevoryas, 1952a) or Stewarfiopteris (Morgan and Dele- 
 voryas, 1952b). Most Psaronins leaf trace scars which have been re- 
 ported from surface features either assumed the open horseshoe-shaped 
 configuration or the closed cylinder with a small W- or inverted V-shaped 
 strand in the center. Baxter ( 1953 ) has attributed the horseshoe-shaped 
 configurations visible in the leaf bases of his specimen of P. cooksonii to 
 a preservational fissure between the sclerenchyma sheath of the petiole 
 and the more internal cortex. He states that the petiole trace at this 
 point "consists of a straight to slightly curved strand. ... Its upward 
 passage into the petiole is almost vertical for approximately 1.6 cm., at 
 which level it reaches the surface of the petiole scar just within the lip 
 of the cup-shaped petiole base. Consequently, the true vascular bundle 
 scar is not visible in the face view of the Caulopteris surface but can be 
 seen if the protruding petiole bases are viewed from the top." 
 
 The type specimen of P. cooksonii was examined with Baxter's in- 
 terpretation in mind and it is the conclusion of the present writer that 
 the horseshoe-shaped configuration on the stem surface represents the 
 vascular strand of the petiole as in all other known psaronii. The leaf 
 trace figured by Baxter (1953, 2F11) has been broken off at a level 
 approximately 1.5 cm. below the rest of the cross-section of the stem. In 
 face view, the broken surface is at the bottom of the leaf scar below the 
 point where the bottom of the horseshoe configuration would have 
 extended. The best interpretation of this condition is that because of 
 crushing of the stem, and since this trace has not yet reached the surface 
 of the leaf base, it has not yet attained the broadly expanded and curved 
 Caulopteris trace. Other scars on the stem exhibiting the typical U-shaped 
 configuration on the surface also show evidences of tracheids at the 
 surface of the bundle scar. 
 
 B. Root Zone 
 
 Surrounding the stem except in the more apical portions is a zone of 
 adventitious roots. Roots of the more internal regions of this zone are 
 surrounded by radially aligned parenchyma which proliferated from 
 various parenchyma cells of the stem and roots. This tissue in which 
 the roots are embedded is not composed of compacted hairs as suggested 
 by earlier workers (e.g., Farmer and Hill, 1902; Solms-Laubach, 1911), 
 
56 AMERICAN SPECIES OF THE GENUS PSARONIUS 
 
 but is a solid mass of parenchyma cells which were capable of under- 
 going frequent divisions. Adventitious roots arise at various points along 
 the length of the peripheral cauline bundles. Figure 57 shows the bul- 
 bous base of a root trace departing from a peripheral cauline bundle of 
 P. blicklei. The roots grew downward and outward to the periphery of 
 the stem, where, at least for a certain length of the root, they are com- 
 ponents of the inner root mantle. While still within the stem, root traces 
 are surrounded by a layer of cortex and sclerenchyma. Roots near their 
 points of origin from the stem have a relatively smaller diameter which 
 gradually increases toward the distal extremity of each root. Roots from 
 more nearly apical regions of a plant are also relatively smaller in 
 diameter at their points of origin than roots arising from more nearly 
 basal portions (figs. 25, 32). Close to the point of exit from the stem, a 
 root consists of a sclerenchyma sheath enclosing an actinostele and a 
 very compact cortical zone. Certain cells of the cortex usually contain 
 tannin-like deposits, and often these cells appear to form a complete ring 
 about the stele (fig. 76). The steles are composed entirely of primary 
 xylem and phloem, and have been observed to possess from three to nine 
 xylem arms (fig. 77). Small, thin-walled phloem cells are situated in 
 the furrows between these arms. Tracheids of much smaller diameter 
 are situated in the ends of the arms of the actinostele and are presum- 
 ably protoxylem. In many roots proliferation of parenchyma cells had 
 occurred about their steles, giving the superficial appearance of sec- 
 ondary xylem. Branching of roots (fig. 77) which had been previously 
 figured by Reed (1949) appears much more common in roots of the 
 genus than was suggested. As the roots grew longer they increased in 
 size and parenchyma cells of the cortex assumed the netted appearance 
 often described as characteristic of roots of the free root zone. Study of 
 specimen A, which is surrounded by a massive mantle of well-preserved 
 roots, shows that roots need not necessarily be free of the secondary 
 parenchyma at levels where the netted cortex is present. At distances 
 removed from their points of origin, roots develop this type of cortex, 
 and at lower levels in their downward course they may or may not be- 
 come encased in secondary parenchyma. 
 
 The cortex of a free root is pictured in fig. 77. Large intercellular 
 spaces become more sparse near the stele of the root. The cortex is more 
 compact and is composed of smaller cells in this region. Tannin-like 
 remains may be observed widely scattered through the entire cortex 
 (fig. 77). Near the periphery of the root, outside of the netted paren- 
 chyma, the cortex is more compact and large intercellular spaces are 
 absent. The cells are much smaller in diameter in this region and 
 grade into a surrounding layer of thick-walled sclerenchyma cells. 
 
 It is believed that in addition to increase in size, roots also differenti- 
 
GENERAL DISCUSSION 57 
 
 ated more xylem arms in the distal extremities as the stele increased in 
 size. The larger roots have an average of one or two more xylem arms 
 on the actinostele, and it seems reasonable that more arms could have 
 been differentiated as the exarch protostele increased in size while the 
 root grew in length. 
 
 None of the roots observed showed evidences of pericycle, endo- 
 dermis, or phellogen as were described by Reed ( 1949 ) . She suggests 
 that the so-called "lacunar" cortex is of secondary origin, being derived 
 in more distal regions of the root by the activity of a phellogen. She 
 further suggests that the sclerenchyma layer at the periphery of the 
 cortex is of a periderm-like nature, this also having been derived from 
 the phellogen. In all specimens observed by the present author the cells 
 of the inner cortex grade gradually into the outer sclerotic layer without 
 evidence of a phellogen, and therefore it is assumed that, with the 
 exception of certain parenchyma cells, all tissues of these roots were 
 normally of primary origin. They cannot be favorably compared with 
 stems of certain modern water plants (e.g., Decodon verticillatus) which 
 have an aerenchymatous "periderm" laid down by a phellogen. Rather, 
 the situation in these roots is analogous to that in stems of Nelumbo and 
 Marsilea where the cortex is primary in origin. 
 
 Most cells of the compact inner cortex, more external netted cortex, 
 and certain parenchyma cells of the limiting layer are capable of sec- 
 ondary divisions, but these cells, because of their arrangement, could 
 by no means be considered to be phellogen. They underwent cell di- 
 visions and the resulting radiating parenchyma (fig. 78) served to form 
 part of the mass of secondary parenchyma of the inner root zone. Pro- 
 liferation was most abundant on the surfaces of the roots away from the 
 center of the stem, much less along the sides, and none was observed 
 arising from internal faces of roots. 
 
 Additional parenchyma tissue of the same type was derived secondarily 
 from cortical cells of leaf bases (fig. 76) and from parenchyma cells 
 intermixed with the fibers of the stem sheath (fig. 36). The nature of 
 the tissue surrounding the more internal roots then seems fairly clear. 
 It was formed by proliferation of various parenchyma cells (although 
 not in the form of elongated hairs) as suggested by Farmer and Hill 
 (1902) and Solms-Laubach (1911) rather than being part of the 
 original stem cortex through which the roots grew (Stenzel, 1906; Sahni, 
 1935). It seems definite that the long-debated question as to its intra- 
 cortical or extracortical nature is settled. 
 
 Cells of this region also contain the tannin-like deposits (figs. 46, 76). 
 At higher levels, lobed masses of parenchyma project beyond the root 
 zone (fig. 46). 
 
 Adventitious roots have also been observed to arise from the second 
 
58 AMERICAN SPECIES OF THE GENUS PSARONIUS 
 
 vascular cycle of the stem. In fig. 45 one such root is seen in connection 
 with the second cycle, and the cross-sections of four roots may be ob- 
 served in the sclerenchyma band on the adaxial side of the leaf trace. 
 If at lower levels these roots grew downward and outward into the con- 
 cavity of the leaf trace, there is no path open by which they could have 
 reached the periphery of the stem other than by growth through the 
 vascular tissue of the trace. Unless they followed such a course, which 
 has not been observed, or grew around the outside of the trace, these 
 roots have neither functions of support or absorption — since they must 
 have ceased growth while still within the stem. The majority of func- 
 tioning roots originated in the peripheral cauline bundles. 
 
 Baxter ( 1953 ) made the incorrect deduction that petioles of fronds 
 pass outward for some distance through the root mantle. It is possible 
 that this misconception is due to the interpretation of the relationship 
 of leaf bases and root mantle as presented by Zeiller (1890), whom 
 Baxter cited for the authority. Although none of Zeiller's figures show 
 leaf bases among the roots of the inner root mantle he explains their 
 absence by supposing that the parenchyma increased in amount behind 
 the apical meristem and that the roots passed down through it. The 
 external surface of the stem together with portions of the petioles, be- 
 came detached leaving the cortex traversed by roots exposed. As a result 
 of evidence presented in this paper about the derivation of the secondary 
 parenchyma surrounding the roots, the erroneous nature of Zeiller's 
 explanation becomes apparent. Further, leaf bases have never been 
 found attached and extending out through the root mantle. Based on his 
 mistaken ideas concerning the relationships of roots to leaf bases, Zeiller 
 (1900) prepared a reconstruction of the stem and root mantle of 
 Psaronius. At the surface of the root mantle some distance from the stem 
 he figured Caulopteris scars which were supposedly the ends of petiole 
 bases which traversed the mantle. At a lower level of the reconstruction 
 the root mantle was removed ( since he considered this part of the stem, 
 its removal amounted to decortication of the stem) and the actual stem 
 surface with its scars had been designated Ptychopteris. This conception 
 of Caulopteris is obviously not that which Lindley and Hutton ( 1831-33 ) 
 originally described for the form genus, nor is Ptychopteris to be re- 
 garded as stem surface features without excessive weathering. Zeiller 
 (1890) did, however, show some surfaces of Psaronius stems which 
 were obviously much weathered and which he correctly designated 
 Ptychopteris. Little is gained by the continued use of the form genus 
 Ptychopteris in anatomical descriptions of Psaronius, since its formation 
 is only a matter of degree of weathering and, if necessary, this condition 
 may be adequately expressed without benefit of a generic name. 
 
 Baxter (1953) suggests that his specimen may be referred to Ptychop- 
 
GENERAL DISCUSSION 59 
 
 teris except that in the case of P. cooksonii the roots were not supposed 
 to have been present on the stem, since the upper three-fourths of the 
 specimen completely lacked roots. This specimen has been incorrectly 
 compared by Baxter to the form genus Ptychopteris because of Zeiller's 
 erroneous description of that genus, and further because the author of 
 P. cooksonii misinterpreted the stem fragment as an apical portion. A 
 re-examination of the type material has shown roots projecting at 
 numerous points from the surface of the stem just outside the peripheral 
 cauline bundles. It is probable that a complete root mantle had been 
 broken off the stem, since very numerous roots and grooves indicating 
 former positions of roots cover the leaf scars. Several silicifiecl specimens 
 in the University of Illinois collections show an almost identical con- 
 figuration of the stem surface after the root mantle was broken off in the 
 laboratory. 
 
 As previously suggested in connection with P. blicklei, many roots 
 near the apex of the stem begin their course as free roots appressed 
 to the surface of the stem between leaf bases. At lower levels they be- 
 came enclosed in the proliferating parenchyma cells as did the roots 
 produced at lower levels. These roots continued their downward and 
 outward course through the inner root zone while gradually increasing 
 in diameter, until their outward rate of growth exceeded that of the 
 secondary parenchyma tissue. From this level to the base of the stem 
 (at least until subsequent growth of parenchyma enclosed them) they 
 are considered to be free roots. The mantle of roots about larger stems 
 may exceed 2V2 feet in diameter. Thus, by addition of roots from higher 
 levels the circumference of the plant increased as the roots grew out 
 and down over those formed at lower levels. 
 
 The twofold function of the root mantle is obvious when the structure 
 of this tree fern is considered. The stem itself is an obconical structure 
 which reached heights of 10 meters or more, but at its base may be as 
 small as 2.3 X 1.5 cm. in diameter. This is the smallest stem cross-section 
 observed, but there must have been still smaller portions of these stems. 
 Although some stems at higher levels, in addition to the external scle- 
 renchyma sheath, developed bands of fibers in association with the 
 vascular bundles, many never developed this additional means of sup- 
 port. Obviously, then, the root mantle was one of the main means of 
 support of these plants. 
 
 One very essential function of the mantle has been almost entirely 
 overlooked by previous workers. Since at the lowest levels the amount 
 of vascular tissue of the upright stem is very small in comparison with 
 the enormous increase at higher levels, it becomes apparent that in 
 more nearly basal regions this small stele was not mechanically efficient 
 enough to transport the volume of water carried by the vascular cylinders 
 
60 AMERICAN SPECIES OF THE GENUS PSARONIUS 
 
 at higher levels, nor by the leaf traces in the huge fronds borne by such 
 plants. The mantle of roots then performs the very necessary function 
 of directly supplying moisture to the vascular tissue of the stem at higher 
 levels. A fact that should be emphasized in connection with conduction 
 is that if there were no contributions of more external cycles to inner 
 ones, and if there were no fusion of cycles of the stem one to another at 
 the points where traces are passed from one cycle to another, then these 
 roots would function largely in supplying moisture to the peripheral 
 cauline bundles — since there would be no efficient means of its reaching 
 more internal cycles. 
 
 C. Comparisons with Other Families of Ferns and Taxonomic Position 
 
 For many years investigators have attempted to determine the rela- 
 tionships of the Psaroniaceae with recent ferns. The two extant families 
 which have received the most attention are the Marattiaceae and Cy- 
 atheaceae (including the Dicksoniaceae ) . The characters used have 
 been interpreted differently in existing literature. The parenchyma tissue 
 of the root zone, for example, was compared by Stenzel ( 1906 ) with that 
 of Marattiaceae, but it was shown before by Farmer and Hill ( 1902 ) in 
 their work on Angiopteris that roots of this extant genus were intrusions 
 into the cortical tissue. A point of similarity is that Marattia roots have 
 an actinostelic vascular strand similar to that of Psaronius roots. In the 
 Marattiaceae, roots arise mainly from more internal portions of the 
 stele, while in Psaronius a few originate from this region but the great 
 majority of them are in connection with the peripheral cauline bundles. 
 The presence of secondary parenchyma tissue around them is a feature 
 peculiar to Psaronius. 
 
 With regard to the stem, it was thought to resemble, for example, 
 that of Angiopteris (Mettenius, 1864). Angiopteris, however, does not 
 exhibit the expanded cauline bundles which play an important part in 
 contributing to leaf trace formation, although Mettenius (1864) figured 
 a stem of Angiopteris grown under abnormal conditions in which the 
 strands of the more external cycles had fused into tangentially expanded 
 bundles, four to six of which form a cycle. Certain authors (e.g., Bower, 
 1926 ) suggested that the modern much dissected stele of Angiopteris is 
 derived by a reduction or dissection of the less interrupted, more primi- 
 tive cycles in Psaronius. Others (e.g., Seward, 1910) believed that the 
 peripheral cauline bundles of Psaronius perform a function similar to 
 the accessory strands in members of the Cyatheaceae. The occurrence of 
 large sieve cells in the phloem is also a feature similar to living Maratti- 
 aceae. 
 
 The psaroniaceous leaf trace, which is generally undissected, has been 
 
GENERAL DISCUSSION 61 
 
 compared by Seward ( 1910 ) with those of such ferns of the Cyatheaceae 
 ( Dicksoniaceae ) as Dicksonia antarctica. It seems to the present author 
 that if such superficial characters as shape of trace are to be cited as 
 evidence of affinities, the vascular traces of Matonia should be taken 
 into consideration. More similarity exists between petiole traces of certain 
 of the Marattiaceae and those of Psaronius, in spite of the usually dis- 
 sected nature of marattiaceous forms. Jeffrey (1910) and others have 
 described the gradual change of arrangement of the separate strands in 
 the leaf base, from an open horseshoe to an arrangement in which the 
 strands toward the adaxial side bend inward, fuse, and become isolated 
 in the center of a ring of bundles. Except for the fact that the trace is 
 dissected, the sequence follows exactly that assumed by Stipitopteris. 
 
 Finally, a consideration of the synangia borne on fronds assignable to 
 Psaronius stems seems to indicate affinities with the Marattiaceae. 
 Strasburger (1874) has pointed out that the marattiaceous character of 
 Scolecopteris is unmistakable since S. elegana, in its sporangial form, 
 resembles Marattia and Christensenia in the circular disposition of the 
 sorus, while the outline of their distal free portions resembles Angiopteris; 
 in dehiscence it compares with all three, but especially with Marattia. 
 The minutiae of these sporangia (e.g., Scolecopteris, Ptychocarpus, and 
 Asterotheca) cannot be fully discussed here. The only point to be con- 
 sidered is that these sporangia are of the eusporangiate type which would 
 logically link them with the Marattiaceae. 
 
 Most authors seem to regard the much-dissected vascular system of 
 the stem and leaf traces of marattiaceous ferns as reduced forms of the 
 more primitive psaroniaceous types. It has been pointed out, however 
 (Seward, 1910), that other families of ferns exhibit characters which 
 are primitive in nature and could themselves be derived from a psaroni- 
 aceous ancestor. It is possible that more than one modern family of 
 ferns (e.g., Cyatheaceae including the Dicksoniaceae and Marattiaceae) 
 was derived through reduction from a psaroniaceous ancestor. If this 
 were the case one could assume that the line of development leading to 
 modern marattiaceous ferns has not in some characters ( sporangia, etc. ) 
 proceeded as far phylogenetically as certain members of the Dicksoni- 
 aceae, which themselves retain other primitive features (e.g., undis- 
 sected leaf traces ) . Unlike most members of the Cyatheaceae, which are 
 usually typified by a dissected stelar system, Thyrsopteris elegans and 
 Saccoloma elegans (Bower, 1926) appear to have remained relatively 
 unchanged and are strikingly similar anatomically to Psaronius. In these 
 forms, leaf traces originate in the center of the stem and in their upward 
 and outward course fill the gaps for short vertical distances in the outer 
 cycles. It is further suggested that such features (pronounced outer 
 cycle and much dissected inner cycle) as are found in Psaronius brasili- 
 
62 AMERICAN SPECIES OF THE GENUS PSARONIUS 
 
 ensis and Tietea singularis may represent transitional types leading to 
 modern families since they show great dissection of the internal cycles 
 and the beginning of dissection of leaf traces. 
 
 It is believed, because of uncertain evidence and the possibility that 
 more than one modern family of ferns may have had its origin in a 
 psaronioid complex, that the position of Psaronius is best expressed by 
 placing it in a separate family as advocated by Stenzel (1854, 1906). 
 
 IV. TAXONOMIC TREATMENT OF THE PSARONII OF 
 NORTH AND SOUTH AMERICA 
 
 Some authors (e.g., Rudolph, 1906) maintain that the psaronii are 
 closely allied to the modern Marattiaceae, and on this basis there has 
 been a tendency to group the psaronii with this family as its fossil repre- 
 sentatives. The present author is inclined to agree with Stenzel (1906) 
 and Seward ( 1910) in assigning the genus Fsaronius to a different family. 
 The spelling of the family name Psaronieae should be modified to 
 Psaroniaceae in keeping with the modern spelling of taxa of that rank. 
 
 Because of the accumulation of new evidence presented in this paper, 
 a more inclusive generic description of Psaronius has been compiled as 
 follows : 
 
 Arborescent fern stems occasionally attaining a height of 10 meters or 
 more, fronds with petioles of the Stipitopteris or Stewartiopteris type 
 borne at the crown; frond arrangement in some stems ranging from a 
 spiral having three orthostichies in lower portions, to a whorled condition 
 of 14 or more rows of leaves in alternating whorls of seven, or more 
 orthostichies at higher levels of the stem; other stems having a distichous, 
 tetrastichous (two leaves per whorl), or possibly a hexastichous (three 
 leaves per whorl alternately arranged) arrangement throughout the 
 length of the stem; stem obconical having from two to twelve or more 
 concentric stelar cycles, the number increasing at successively higher 
 levels; each new cycle formed within the central cycle and attached to 
 it at its point of origin; vascular bundles endarch, amphiphloic; per- 
 sistent leaf gaps between cauline bundles filled only at certain positions 
 by leaf traces; stem usually surrounded by a definite sclerenchymatous 
 zone; certain cells of fundamental tissue and secondary parenchyma 
 around roots often containing tannin-like deposits in many species; 
 sclerenchyma bands, and often groups of fibers, occasionally associated 
 with the vascular bundles; adventitious roots arising from the peripheral 
 cauline bundles and often from more internal cycles; root traces polyarch; 
 cortex of roots becoming netted in more distal regions; roots near the 
 stem compacted by a mass of secondary parenchyma tissue proliferated 
 
TAXONOMIC TREATMENT 63 
 
 from stem and root cortices, and from parenchyma cells intermixed with 
 fibers in the sclerenchyma zones surrounding the stem and individual 
 roots; free roots generally at the periphery of this region. 
 
 It is obvious that an artificial classification of the psaronii such as that 
 followed by Hirmer ( 1927 ) should be discarded. Psaronius blicklei, for 
 example, could be assigned to more than one of Hirmer's categories 
 since the leaves are spirally arranged near the base and whorled at higher 
 levels. 
 
 As a result of this study it is necessary to reduce certain previously 
 named species of Psaronius to synonymy, to add several new members, 
 and to suggest that certain specific names be discarded since they are 
 based on inadequately known, poorly preserved specimens. Specimens 
 in the latter category include P. junceus Herzer (1900), P. borealis Mac- 
 bride ( 1904-6 ) , and P. illinoensis Hoskins ( 1928 ) . Nomenclatural changes 
 in existing species and descriptions of new ones are as follows: 
 
 Psaronius giffordii Gillette, Bot. Gaz. 99:98-100, 
 figs. 13, 17-20. 1937. 1 
 
 1937 Psaronius septangulatus Gillette, Bot. Gaz. 99:82-95, figs. 1-12. 
 1937 Psaronius peoriensis Gillette, Bot. Gaz. 99:95-98, figs. 14-16. 
 1953 Psaronius cooksonii Baxter, Bot. Gaz. 115:35-40, figs. 1-4, 6. 
 
 Orthostichies of leaves of available specimens numbering 6-12, prob- 
 ably throughout the length of the plant, ranging from three at the lower- 
 most levels to 12 or more higher in the stem; phyllotaxy in available 
 specimens ranging from a 2/7 spiral to alternating whorls of six each; 
 vascular cycles numbering from six to ten, probably ranging from one 
 to ten or more throughout the length of the plant; definite sclerenchyma 
 sheath surrounding the stem except at points of leaf trace divergence; 
 ground parenchyma dense; ground parenchyma with scattered gum cavi- 
 ties; lacunae absent; tannin cells apparently absent. 
 Localities: Princeville Twp., Peoria Co., Illinois; Stark Co., Illinois; 
 Hamilton, Kansas. 
 
 Age: Upper Pennsylvanian for Illinois specimens; Upper Pennsylvanian 
 for Kansas specimen. 
 
 Horizon: Unknown for Illinois specimens; Wabaunsee group, Pennsyl- 
 vanian System, for the Kansas specimen. 
 
 Holotype: Specimen 2613 (formerly no. 32 of the Peoria Academy of 
 Science), now in the paleobotanical collections, Department of Botany, 
 University of Illinois, Urbana, Illinois. Pieces of specimen 32 in the De- 
 partment of Botany, University of Chicago, Chicago, Illinois. 
 
 It has been necessary to group P. septangulatus (fig. 79) and P. peori- 
 
 1 The ending ii is used here in accordance with Recommendation C, Article 82 of 
 the 1952 International Code of Botanical Nomenclature. 
 
64 AMERICAN SPECIES OF THE GENUS PSARONIUS 
 
 ensis ( fig. 80 ) described by Gillette ( 1937 ) with P. giffordii of the same 
 author (Gillette, 1937, fig. 17), since a re-examination by the present 
 author showed no distinguishing features and according to the descrip- 
 tions of the specimens they vary only in numbers of orthostichies (phyl- 
 lotaxy), numbers of internal strands (stelar cycles), and characters of 
 size and shape of the leaf scars. On the basis of evidence presented 
 previously in this paper, the characters described for each of these 
 species may vary greatly in one species and even throughout the length 
 of a single stem, leaving no valid means of distinguishing them. No 
 definite sclerenchyma sheath was described for the specimen formerly 
 known as P. peoriensis, although a re-examination of the specimen then 
 serving as the type has shown that a sheath was present. P. giffordii, 
 according to Gillette (1937), possessed six orthostichies of leaves in 
 alternating whorls of three each. One orthostichy, although missing, was 
 postulated as being present on the stem, based on his estimation of the 
 position of the center of the stem and the arrangement of leaves about 
 the circumference. An examination of the original specimen has led the 
 present author to believe that originally the stem fragment bore eight 
 orthostichies of leaves in alternating whorls of four. Other characters 
 ascribed to this stem place it in P. giffordii. 
 
 The specimen described by Baxter (1953) as P. cooksonii differs only 
 superficially from P. giffordii in number of orthostichies, size of "secre- 
 tory cells" (Baxter, 1953) or "gum canals" (Gillette, 1937), and inter- 
 grades with P. giffordii in numbers of internal strands (vascular cycles) 
 and size and shape of leaf bases. None of these characters is valid for 
 specific differentiation in the case cited. 
 
 The name P. giffordii has been retained over P. septangulatus and P. 
 peoriensis, since P. septangulatus implies a condition not constant for 
 the species while P. peoriensis suggests only a restricted locality for the 
 species. 
 
 Psaronius brasiliensis Brongniart, Bull. Soc. Bot. 
 France 19:3-10, 1872. 
 
 Figured by Unger De palmis fossilibus, in Martius, Historia Naturalis 
 Palmarum, p. 70, pi. 1, fig. 4. 1850. 
 
 Stem bearing four orthostichies of leaves in a decussate arrangement; 
 at least four concentric vascular cycles with a varying number of more 
 internal vascular strands probably representing additional much-dis- 
 sected cycles; innermost strands often appearing as closed or open cylin- 
 ders; four cauline bundles alternating with four pathways of leaf trace 
 divergence; leaf traces receiving contributions of vascular tissue from 
 the cauline bundles below the point of fusion to them, although not 
 fusing with peripheral cauline bundles; stem surrounded by definite 
 
TAXONOMIC TREATMENT 65 
 
 sclerenchyma sheath broken only at points of leaf trace departure; leaf 
 traces in leaf bases becoming dissected with a number of small vascular 
 strands accompanying the trace in its adaxial concavity; vascular strands 
 accompanying the trace in its adaxial concavity; vascular strands accom- 
 panying the leaf trace originating in part from the center of the stem 
 and in part from dissection of the free edges of the trace; internal bands 
 of sclerenchyma occasional between leaf trace and cauline bundles; 
 mantle of polyarch adventitious roots; secondary parenchyma between 
 roots of root mantle. 
 
 Locality: Chapado do Jaboti, Estado do Maranhao, Brazil. 
 Age: Thought to be Lower Permian. 
 Horizon: Unknown. 
 
 Holotype: Specimen no. 1445 in the Paris Museum, other pieces of which 
 are at the Rio de Janeiro Museum and the British Museum. 
 
 The history of this specimen is somewhat unusual and caused con- 
 siderable difficulty in determining the author and original citation. A 
 fragment of a stem of this species was figured by Unger in Martius' 
 (1850) Historia Naturalis Palmarum 22 years before Brongniart's (1872) 
 description of P. brasiliensis (Pelourde, 1912). The specimen serving as 
 the basis for Brongniart's description is considerably better preserved 
 and more complete than that figured by Unger, and since no description 
 accompanied Unger's illustration Pelourde (1912) suggested that the 
 more complete fragment described by Brongniart in 1872 should be con- 
 sidered the type specimen. The better preserved specimen was not 
 figured with Brongniart's (1872) description of the specimen. However, 
 1872 is considered to be the valid publication date of P. brasiliensis, since 
 this date precedes December 31, 1911, the date after which the descrip- 
 tion and illustration must be simultaneously published. 
 
 An unusual kind of fern stem thought to have originated from the 
 same locality as P. brasiliensis was described as Tietea singularis by 
 Solms-Laubach (1913). This specimen, although less complete than P. 
 brasiliensis, shows evidence of a larger number of orthostichies. Internal 
 cycles are much less conspicuous than in P. brasiliensis and are evident 
 only as a large number of hollow cylindrical strands of vascular tissue, 
 some of which are completely closed. Further, the leaf traces in Tietea 
 are much more dissected than those in P. brasiliensis, although in the 
 latter there is some indication of dissection of the traces. There is con- 
 siderable evidence that Tietea has affinities with P. brasiliensis, but the 
 two forms must remain separate until additional material becomes avail- 
 able and further investigations are undertaken. 
 
 Psaronius pertusus A. H. Brickie, sp. nov. (fig. 81). 
 In the type specimen, eight peripheral cauline bundles alternate with 
 
66 AMERICAN SPECIES OF THE GENUS PSARONIUS 
 
 eight pathways of leaf trace divergence; leaves arrangd in a 3/8 spiral 
 phyllotaxy (probably fewer rows at more nearly basal levels, becoming 
 whorled with more orthostichies above); five concentric stelar cycles 
 in the length of the holotype preserved (probably having fewer below 
 and a greater number above); ground parenchyma cells having a net- 
 like arrangement; tannin cells scattered in ground tissue; lysigenous 
 lacunae and gum cavities lacking; definite sclerenchyma sheath sur- 
 rounding the stem. 
 
 Localities: Shade and Lodi Twp., Athens Co., Ohio; Morgan Co., Ohio; 
 Calhoun, Richland Co., Illinois. 
 Age: Upper Pennsylvanian. 
 
 Horizons: In Ohio: Pittsburgh coal or No. 8 seam, lower Monongahela 
 formation; Arnoldsburg sandstone, upper Monongahela formation; Little 
 Waynesboro coal, Monongahela formation. Upper McLeansboro group 
 of Illinois, Pennsylvanian System. 
 
 Holotype: Specimen no. B-202 in the Ohio University Museum Collec- 
 tion, Athens, Ohio. 
 
 The excellently preserved specimen showing morphological details of 
 the vascular tissue in fig. 81 was collected from the Monongahela forma- 
 tion of Lodi Twp., Athens Co., Ohio, by Dr. A. H. Blickle. The diagnosis 
 of P. pertusus was written by him for this specimen. Fig. 82 of the 
 ground tissue was photographed from a specimen obtained from the 
 Calhoun collecting locality of Illinois. 
 
 P. pertusus Blickle was formerly designated as P. lacunosus Blickle 
 ( 1940 — see Arnold, 1947 ) although never validly published as such. The 
 name has been changed to P. pertusus to avoid confusion, since the ly- 
 sigenous cavities of P. blicklei can more readily be termed lacunae than 
 the large intercellular spaces of the ground parenchyma of P. pertusus. 
 
 Psaronius blicklei Morgan, sp. nov. (fig. 1-12, 14, 15, 25-34). 
 
 Stem having three to fourteen or more orthostichies of leaves, the in- 
 crease occurring at successively higher levels: leaves spirally arranged 
 in lower portions, becoming whorled higher in the stem; up to 10 or 
 more concentric stelar cycles in the stem; cells of leaf traces smaller in 
 diameter than cells of other vascular bundles; leaf traces near periphery 
 of stem thinner than other vascular bundles; traces assuming the 
 Stipitopteris configuration; stem surrounded by a definite sclerenchyma- 
 tous layer, sclerenchyma cells present in small patches and surrounding 
 leaf bases in lower portions of stem, becoming more abundant in the 
 form of bands associated with stele at progressively higher levels; at 
 lower levels, parenchymatous ground tissue compact with no or few 
 lysigenous cavities, the lacunae becoming more abundant and larger 
 higher in the stem; tannin-like deposits in parenchyma cells of ground 
 
TAXONOMIC TREATMENT 67 
 
 tissue of stem and roots, and in secondary parenchyma cells surrounding 
 
 roots; proliferation of parenchyma cells occasional especially in upper 
 
 portions of stem; four to six protoxylem groups on the arms of the 
 
 actinosteles of roots. 
 
 Localities: Berryville, Lawrence Co.; near Dix, Jefferson Co.; Calhoun, 
 
 Richland Co., Illinois. 
 
 Age: Upper Pennsylvanian. 
 
 Horizon: Upper McLeansboro group. Pennsylvanian System. 
 
 Types: Specimens A-I in the paleobotanical collections, Department of 
 
 Botany, University of Illinois, Urbana, Illinois. 
 
 This species in all probability represents the most completely known 
 of the psaronii. A series of specimens from close to the base to fragments 
 high in the stem was used in the description of the morphology of the 
 vacular system. A wealth of histological detail is also available for this 
 species. 
 
 The species was named for Professor A. H. Blickle, who did most of 
 the research on Ohio psaronii and at whose suggestion this project was 
 undertaken. The author is also indebted to him for much aid and assist- 
 ance in obtaining representative Ohio specimens. 
 
 Psaronius chasei Morgan, sp. nov. (figs. 44-47). 
 
 Stem probably having three to many orthostichies of leaves; leaves 
 probably spirally arranged in lower portions, becoming whorled higher 
 in the stem; up to 10 or more concentric stelar cycles in the stem; stem 
 surrounded by a definite sclerenchymatous layer; sclerenchyma bands 
 between second and third vascular cycles; parenchymatous ground tissue 
 with no lacunae; tannin-like deposits in parenchyma cells of the ground 
 tissue of stem (concentrated near vascular bundles), in roots, and sec- 
 ondary parenchyma surrounding roots. 
 
 Localities: Berryville, Lawrence Co.; Calhoun, Richland Co., Illinois; 
 St. Wendel, Posey Co., Indiana. 
 Age: Upper Pennsylvanian. 
 
 Horizon: Upper McLeansboro group in Illinois; Parker coal a short dis- 
 tance above the Lonsdale limestone in Indiana. Pennsylvanian System. 
 Type: C. B. 1221 in the paleobotanical collections, Department of Botany, 
 University of Illinois, Urbana, Illinois. 
 
 This species is named for Dr. Virginius Chase of Peoria, Illinois, who 
 presented, to the University of Illinois, a gift of many fine silicified 
 Psaronius specimens which were of considerable aid in this study. 
 
 Psaronius melanedrus Morgan, sp. nov. (figs. 17, 19, 35-43). 
 
 Stem with three to many orthostichies of leaves; leaves spirally ar- 
 ranged in lower portions, probably becoming whorled higher in the 
 
68 AMERICAN SPECIES OF THE GENUS PSARONIUS 
 
 stem; two to many concentric stelar cycles in the stem; radial thickness 
 of xylem in stem vascular bundles ranging from .4 mm. in the basal 
 portions to at least 1.1 mm. higher in the stem; sclerenchyma cells en- 
 tirely lacking in fundamental tissue except that delimiting leaf bases 
 which are enclosed by a heavy band of fibers and are roughly rectangular 
 in cross-section; parenchymatous ground tissue lacking lysigenous la- 
 cunae; tannin-like deposits in parenchyma of ground tissue and root 
 mantle; root mantle of varying thickness in different portions of the stem; 
 roots arising from basal portion of stem larger in diameter at points of 
 origin than roots arising at upper levels; protoxylem at the tip of each 
 of four to six arms of the actinostele of each root; inner roots in radial 
 files. 
 
 Locality: Four miles south of West Mineral, Kansas. 
 Age: Middle Pennsylvanian. 
 
 Horizon: Fleming coal, Cherokee shale, Des Moines series, Pennsyl- 
 vanian System. 
 
 Types: Specimens J-Q in the paleobotanical collections, Department of 
 Botany, University of Illinois, Urbana, Illinois. 
 
 This species is well represented by stem fragments from near the base 
 to levels high on the stem. It has been named P. melanedrus because of 
 the heavy dark sclerenchyma sheath which completely surrounds the 
 stem. 
 
 V. SUMMARY 
 
 During the course of the research described in the preceding pages, 
 several approaches to the clearer understanding of the morphological 
 charactristics of the genus Psaronius were employed. It has been estab- 
 lished, through examination of a large number of Psaronius specimens 
 from various localities, that it is possible to reconstruct with accuracy 
 the vascular system of Psaronius blicklei sp. nov. From a detailed ex- 
 amination of specimens representing basal portions of this arborescent 
 genus, it has been determined that in early ontogeny stems of Psaronius 
 were exceedingly simple, consisting of one or two cylinders of vascular 
 tissue surrounding a central strand of the same tissue. It becomes evi- 
 dent, at successively higher levels in the stem, that this solid inner strand 
 of tissue, by lateral growth, gives rise to added inner cycles of vascular 
 tissue. This observation lends support to the argument that numbers 
 of investing steles are not in themselves good taxonomic characters. 
 
 It has also been established that the phyllotaxy undergoes several 
 revisions throughout the length of stems of P. blicklei. At the base of 
 such stems leaves produced early in development are arranged in a 
 
SUMMARY 69 
 
 loose spiral. Higher on the stem this spiral arrangement becomes more 
 compact, until many old stems reach a tightly whorled arrangement con- 
 sisting of as many as 14 orthostichies. The exact manner in which new 
 orthostichies of fronds were added, and the way in which new and 
 more internal stelar cycles were added to produce complex, polycyclic 
 dictyosteles, has been discussed in detail. 
 
 It has also been observed that there is considerable increase in stem 
 diameter at higher levels, giving the stem an obconical configuration. 
 The unbalanced structure thus produced is compensated for by an 
 enveloping mantle of roots which aided tremendously in support and in 
 absorption. The situation is somewhat analogous to that of a corn plant 
 which depends upon innumerable adventitious prop roots for anchorage, 
 support, and the bulk of its absorption. 
 
 A number of erroneous ideas concerning the root mantle and frond 
 origin (i.e., Zeiller, 1890; Farmer and Hill, 1902; Stenzel, 1906; Solms- 
 Laubach, 1911; Sahni, 1935; Reed, 1949; Baxter, 1953) have been cor- 
 rected in the light of knowledge gained in this study. 
 
 Morphological evidence obtained from the examination of the vascular 
 systems of many Psaronius stems makes it clear that variation within the 
 genus must be divided into: 
 
 (1) Ontogenetic variation known to occur between base and apex 
 of a single plant, and also at the same levels of the plant with increasing 
 age. 
 
 (2) Variation which remains constant among different species. 
 With a thorough understanding of the first type of variation, it has 
 
 been suggested that many of the previously named American and 
 European species are not valid because they probably represent various 
 ontogenetic stages of stems of the same species. 
 
 Considering variation which is known to remain unaffected by growth 
 stages, the valid American species are: 
 
 (1) P. brasiliensis Brongniart, 1872. 
 
 (2) P. giffordii Gillette, 1937. 
 
 (3) P. pertusus Blickle, 1940. 
 
 (4) P. blicklei Morgan, sp. nov. 
 
 (5) P. melanedrus Morgan, sp. nov. 
 
 (6) P. chasei Morgan, sp. nov. 
 
 Suggested relationships of Psaronius with members of such extant 
 families as Cyatheaceae, including the Dicksoniaceae, and with the 
 Marattiaceae were discussed. It is proposed, however, that the genus 
 Psaronius be retained in its own family instead of being placed in any 
 presently existing taxonomic group. 
 
LITERATURE CITED 
 
 Arnold, C. A. 1947. (See Psaronius.) An introduction to paleobotany. Mc- 
 Graw-Hill, New York. 194-196. 
 
 Baxter, R. W. 1953. The coal age flora of Kansas III. Psaronius cooksonii: a 
 new species showing Caulopteris features. Bot. Gaz. 115:35-44. 
 
 Blickle, A. H. 1940. Ohio psaronii. Ph.D. thesis, University of Cincinnati. 
 62 pp. 
 
 Bower, F. O. 1926. The ferns. Vol. II. Cambridge Univ. Press, London. 344 pp. 
 
 Brongniart, A. 1872. Notice sur le Psaronius brasiliensis. Bull. Soc. Bot. France 
 19:3-10. 
 
 Christ, H. 1910. Die Geographie der Fame. G. Fischer, Jena. 355 pp. 
 
 Cotta, B. 1832. (See Psaronius.) Die Dendrolithen in Beziehung auf ihren 
 inneren Bau. Dresden and Leipzig. 27-36. 
 
 Esau, Katherine. 1953. (See Phyllotaxy.) Plant anatomy. Wiley, New York. 
 339-343. 
 
 Farmer, J. B., and T. G. Hill. 1902. On the arrangement and structure of the 
 vascular strands in Angiopteris evecta, and some other Marattiaceae. Ann. 
 Bot. 16:371-402. 
 
 Farr, C. H. 1914. Notes on a fossil tree fern of Iowa. Proc. Iowa Acad. Sci. 
 21:59-65. 
 
 Gillette, N. J. 1937. Morphology of some American species of Psaronius. Bot. 
 Gaz. 99:80-102. 
 
 Herzer, H. 1900. Six new species, including two new genera, of fossil plants. 
 Ohio State Acad. Sci., 9th Ann. Report. 22-29. 
 
 Hirmer, M. 1927. (See Psaronius.) Handbuch der Paleobotanik. R. Olden- 
 bourg, Munich and Berlin. 545-566. 
 
 Hoskins, J. H. 1928. Notes on the structure of Pennsylvania plants from Illi- 
 nois. II. Bot. Gaz. 85:74-82. 
 
 Jeffrey, E. C. 1910. The Pteropsida. Bot. Gaz. 50:401-414. 
 
 Kiihn, R. 1889. Untersuchungen iiber die Anatomie der Marattiaceen und 
 anderer Gefasskryptogamen. Flora 72:457-521. 
 
 Lenz, L. W. 1942. Contributions to our knowledge of American Carboniferous 
 floras III. Stipitopteris. Ann. Missouri Bot. Gard. 29:59-68. 
 
 Lindley, J., and W. Hutton. 1831-33. Fossil flora of Great Britain. Vol. I. 
 J. Ridgway, London. 121-122. 
 
 Macbride, T. H. 1904-06. On certain fossil plant remains in the Iowa her- 
 barium. Proc. Davenport Acad. Sci. 10:153-162. 
 
 Mettenius, G. 1864. Uber den Bau von Angiopteris. Abh. K. Sachs. Ges. 
 Wiss. (Math.-Phys. CI) 6:499-570. 
 
 Moon, Geraldine. 1939. Notes on the histology of an Illinois Psaronius. Trans. 
 Illinois State Acad. Sci. 32:95-97. 
 
 Morgan, Jeanne, and T. Delevoryas. 1952a. An anatomical study of Stipi- 
 topteris. Amer. Jour. Bot. 39:474-478. 
 
 . 1952b. Stewartiopteris singularis: a new psaroniaceous fern rachis. 
 
 Amer. Jour. Bot. 39:479-484. 
 
 70 
 
LITERATURE CITED 71 
 
 Ogura, Y. 1927. Comparative anatomy of Japanese Cyatheaceae. Jour. Fac. 
 
 Sci. Imp. Univ. Tokyo. Sect. 3 Botany 1:141-350. 
 Pelourde, F. 1908. Recherches comparatives sur la structure de la racine chez 
 
 un certain nombre de Psaronius. Bull. Soc. Bot. France 35:352-359; 377-382. 
 . 1912. Observations sun le Psaronius brasiliensis. Ann. Sci. Nat. Bot. 
 
 ser. 9, 16:337-352. 
 Pohlig, H. 1910. Xijlopsaronius. Les premieres Filicinees characterises par la 
 
 formation du bois. Bull. Soc. Beige Geol. 24:335-339. 
 Reed, Fredda D. 1949. Notes on the anatomy of two carboniferous plants, 
 
 Sphenophyttum and Psaronius. Bot. Gaz. 110:501-510. 
 Rudolph, K. 1906. Psaronien und Marattiaceen. Vergleichend anatomische 
 
 Untersuchung. Denkschr. K. Akad. Wiss. Wien ( Math.-Nat. Kl.) 78:165- 
 
 201. 
 Sahni, E. 1935. The roots of Psaronius, intra-cortical or extra-cortical? A dis- 
 cussion. Current Science 3:555-559. 
 Schoute, J. C. 1938. (See Psaronius.) In Verdoorn, ed., Manual of pteridology. 
 
 Martinus Nijhoff, The Hague. 81-83. 
 Schuster, J. 1911. Xijlopsaronius — der erste Farn mit sekundarem Holz? Ber. 
 
 Deutsch. Bot. Ges. 29:545-548. 
 Scott, D. H. 1920. (See Psaronius.) Studies in fossil botany. 3rd ed. Vol. I. A. 
 
 & C. Black, London. 268-278. 
 Seward, A. C. 1910. (See Psaronius.) Fossil plants. Vol. II. Cambridge Univ. 
 
 Press, London. 412-426. 
 Solms-Laubach, H. Graf zu. 1911. Der tiefschwarze Psaronius Haidingeri von 
 
 Manebach in Thuringen. Zeitschr. Bot. 3:721-757. 
 . 1913. Tietes singularis. Ein neuer fossiler Pteridenstamm aus Brasilien. 
 
 Zeitschr. Bot. 5:673-700. 
 Sprengel, A. 1828. Commentatio de Psarolithus. Halle. P. 37. 
 Stenzel, G. 1854. Uber die Staarsteine. Nova Acta Akad. Leop. Carol. 24:753. 
 . 1906. Die Psaronien, Beobachtungen und Betrachtungen. Beitr. 
 
 Palaontologie und Geologie von Osterreich-Ungams und des Orients 19:85- 
 
 123. 
 Sterzel, J. T. 1912. Uber Xylopsaronius Pohlig. Naturwiss. Ges. Chemnitz, Ber. 
 
 18:65-69. 
 Strasburger, E. 1874. Ueber Scolecopteris elegans Zenk., einen fossilen Farn 
 
 aus der Gruppe Marattiaceen. Jen. Zeitschr. Naturwiss. 8:81-95. 
 Unger, F. 1850. De palmis fossilibus. In Martius, Historia naturalis palmarum, 
 
 p. 70. 
 West, C. 1915. On the structure and development of the secretory tissues of 
 
 the Marattiaceae. Ann. Bot. 29:409-422. 
 Zeiller, R. 1890. Bassin Houiller et permien d'Autun et d'Epinac. Flore Fossile. 
 
 Etudes des Gites Miner. France. Fasc. II, part 1:178-271. 
 . 1900. (See Psaronius.) Elements de paleobotanique. Carre and Naud, 
 
 Paris. 118-123. 
 
GLOSSARY 
 
 abaxial — oriented away from the axil of a leaf. 
 
 actinostele — an arrangement of vascular tissue commonly occurring in roots 
 
 of plants where the primary xylem forms a core with radiating ridges. In 
 
 cross section the xylem has a star-shaped configuration. 
 aerenchyma — parenchyma tissue forming an anastomosing network of cells 
 
 separated by large intercellular spaces. 
 amphiphloic — a condition where phloem is external and internal to the 
 
 xylem. 
 axil — the angle between the stem and the upper face of the attached petiole 
 
 of a leaf. 
 calamites — Paleozoic ancestors of scouring rushes. These were arborescent 
 
 plants. 
 Carboniferous — recognized in some geological classifications as a period 
 
 (system) in the upper Paleozoic. 
 cauline — pertaining to stems. 
 coal ball — calcareous, dolomitic or pyritic rocks containing fossilized remains 
 
 of plants. Usually found associated with upper Paleozoic coals. 
 coenopterids — an artificial assemblage of extinct herbaceous plants with fem- 
 
 like sporangia and stems. 
 dictyostele — a stele with its vascular tissue dissected into smaller units by 
 
 leaf and branch gaps. 
 distichous — two-ranked, with leaves on opposite sides of a stem and in the 
 
 same vertical plane. 
 endodermis — a layer of cells delimiting a stele from surrounding tissues of the 
 
 cortex. 
 epithelium — a layer of cells lining a cavity. 
 
 eusporangiate — a condition in which spores are formed inside a thick- 
 walled sporangium usually originating from several subepidermal initials. 
 
 The eusporangium is typical of most vascular plants except for Filicales. 
 fundamental tissue (ground tissue) — extensively developed tissue composed 
 
 primarily of parenchyma cells. Usually forms a matrix in which vascular 
 
 bundles or steles are embedded. 
 hexastichous — six-ranked, with leaves on six sides of the stem and in the 
 
 same vertical plane. 
 ironstone — a rock composed of fine clay particles and impregnated with iron 
 
 compounds. In some areas ironstone forms around a nucleus of plant or 
 
 animal remains. 
 leaf gap — an interruption in the vascular system of a stem produced by the 
 
 bending out of a segment of vascular tissue (leaf trace) supplying a leaf. 
 
 Leaf gaps are axillary to leaf traces. 
 leaf trace — a strand of vascular tissue connecting the stem stele and the 
 
 vascular system of a leaf. 
 lepidodendrid — any one of a number of heterosporous, arborescent lycopsids 
 
 (club mosses) common to the upper Paleozoic. 
 
 72 
 
GLOSSARY 73 
 
 lysigenous — formed by the break down or dissolution of cells or tissues. 
 
 metaxylem — that primary xylem differentiated after protoxylem from tissues 
 derived from an apical meristem. 
 
 orthostichy — in phyllotaxy, an imaginary vertical line along a stem axis 
 connecting leaf bases directly above one another. 
 
 Paleozoic — an era of geological time between the Proterozoic and Mesozoic. 
 
 parenchyma — a tissue composed of thin walled cells. These are living cells in 
 extant plants and usually function in food storage. 
 
 Permian — a subdivision (period) marking the upper limits of the Paleozoic 
 era. 
 
 protostele — a core of vascular tissue composed of centrally located primary 
 xylem surrounded by a continuous layer of or patches of primary phloem. 
 
 protoxylem — that primary xylem first differentiated from tissues derived from 
 an apical meristem. 
 
 root mantle (zone) — a continuous layer of adventitious roots around the 
 lower parts of the stem. This layer varies from a few roots to a mantle several 
 inches thick. 
 
 scalariform thickenings — horizontal, bar-like (ladder-like) thickenings on 
 the walls of tracheids. 
 
 sclerenchyma — a tissue derived from parenchyma cells which develop thick- 
 ened walls. Usually dead at maturity and give support. 
 
 seed fern — an extinct vascular plant bearing naked seeds on large, compound, 
 fern-like leaves. 
 
 siphonostele — a stele with vascular tissue surrounding a pith and with over- 
 lapping leaf and branch gaps. 
 
 solenostele — a stele with a pith but lacking overlapping branch and leaf 
 gaps. 
 
 sphenophylls — small protostelic ancestors of scouring rushes apparently be- 
 coming extinct by the end of Paleozoic era. 
 
 stele — the conducting tissues of root or stem composed of xylem, phloem and 
 usually delimited externally from other tissues by an endodermis. 
 
 tetrastichous — four-ranked, with leaves on four sides of the stem and in the 
 same vertical plane. 
 
 tracheeds — conducting and supporting cells of primary and secondary xylem. 
 Dead cells at maturity with distinctive pits or apertures in their walls. 
 
 vascular system — all parts of the conducting system composed of xylem and 
 phloem. 
 
Plates 
 

 sV, iA- 
 
 
 \e 
 
 ij/ 
 
 
 , v ^> / 
 
 
 Ss» 
 
 
 1 r-r^^?n\ wr 
 
 C\ 
 
 ■fi\jQb&} -- 
 
 ^'fevr^^X^ 
 
 /•re jV^ j ^>vo//-.y w 
 
 9 ' 
 
 Figs. 1-7. Psaronius blicklei sp. nov. Diagrammatic cross-sections of specimens A-F. 
 Steles are diagrammed in solid black, sclerenchyma is stippled. Fracture lines are 
 diagrammed in fig. 15. Compare with figs. 25-31. 
 
 Figs. 8, 9. Psaronius blicklei sp. nov. Diagrammatic cross-sections of specimens G, H. 
 Steles are diagrammed in solid black, sclerenchyma is stippled. Compare with figs. 
 32, 33. 
 
 76 
 

 v ^) 
 
 &A 
 
 ^/7?j 
 
 f ^ x 
 
 f.-i 3 
 
 J /i 
 
 
 
 ^"•B™**^ 
 
 
 %^JJ 
 
 % ■ J :i 
 
 ■t'r 
 
 10. 
 
 Fig. 10. Psaronius blicklei sp. nov. Sections A-l represent diagrammatic cross-sections 
 of successively higher levels, showing the origin and reconstitution of the most central 
 cycle of specimen A. Steles are diagrammed in solid black, sclerenchyma is stippled. 
 
 77 
 
V: 
 
 Fig. 11. Psaronius hlicklei sp. nov. Diagrammatic reconstruction of the stele of 
 specimen A. The section on the bottom at the left is the lowest level depicted and 
 that on top at the upper right of the diagram is the highest. 
 
6> O U 
 
 B C 
 
 14. 
 
 Fig. 12. Psaronius blicklei sp. nov. Diagrammatic cross-section of a stem portion with 
 spiral phyllotaxy. Based on specimen C (compare fig. 28). Leaf trace and potential 
 leaf trace material is diagrammed in heavy, solid black lines; the sclerenchyma is 
 designated by a thin line at the periphery of the stem. Stippling represents cauline 
 material which will at higher levels be contributed to departing leaf traces. 
 
 Fig. 13. Psaronius blicklei. Diagrammatic cross-sections showing successively higher 
 levels (A-E) of the two most internal cycles of a stem at about the level exhibited 
 by specimen C. The manner in which the innermost cycle (unshaded) fills a gap in 
 the next outer cycle (solid black), and the sequence for its reconstruction, are shown. 
 See text for full explanation. 
 
 Fig. 14. Psaronius blicklei. Diagrams A-C showing contribution of vascular tissue of 
 peripheral cauline bundles to the next internal cycle and to the leaf trace of specimen 
 C. Leaf traces, solid black; cauline bundles, unshaded; contribution of peripheral 
 cauline bundles to leaf trace, stippled. See text for full explanation. 
 
 79 
 
Fig. 15. Psaronius blicklei. Diagrams A-I showing the origin of a new orthostichy and 
 the beginning of a new pathway of divergence for leaf traces of this orthostichy. 
 Specimen E. See text for full explanation. 
 
 80 
 
I o> 
 
 17. 
 
 Fig. 16. Diagrammatic reconstruction of the cross-section of a large, whorled 
 Psaronius stem based on specimen G. Leaf traces and potential leaf trace material 
 are diagrammed in black, cauline bundles are unshaded, and potential contributions 
 of the peripheral cauline bundles to the leaf traces are stippled. See text for full 
 explanation. 
 
 Fig. 17. Psaronius melanedrus sp. nov. Diagrammatic cross-section (A-C) of suc- 
 cessively higher levels, showing leaf trace contribution by and reconstruction of the 
 most internal cycle of specimen J. Steles are diagrammed in black, sclerenchyma is 
 stippled. See text for full explanation. 
 
 81 
 
o 
 
 GO 
 
 CD 
 
 "V. co w -* 
 35 to ca 
 
 "■ -* g iS 2 
 
 
 ctf O 
 
 03 QJ 
 
 Eh a) 
 
 M *-" -5 
 
 < a 
 
 s 
 
 o 
 
 ' ~ - S "a 
 
 -.3 
 
 S3 H ^ 
 
 i-£h en 
 
 o 2 
 
 O J3 
 
 5 S 
 
 c -s 
 
 •43 O _ 
 
 O i-rt 
 Q, 2 
 
 c 1) 
 
 <D 
 
 o <u 
 
 w -o ^3 2 
 J. .& - 1 a «> 
 
 g Oq >>J2 « 
 
 O J5 C 55 -r< 
 0) +5 03 T3 >% 
 « 03 ^ •- ^ 
 
 » C ■" ,i fl 
 
 n S O H)\ 
 
 J> & 03<» 
 
 ■.fl .S3 •— ' M °* 
 
 0) 
 
 3 -c 
 
 
 
 Q I 
 
 2 3T3 
 
 g o <u 
 „ h 10 h 
 
 D _, 03 <U 
 
 G ^ > -Q 
 
 s 
 
 •43 
 
 w 0) co 
 a 4-i 03 
 
 COX 
 
 4-> U 
 
 ^ 
 
 <» oj 
 
 S 0) 1) rt 
 
 § x- s> 
 
 -c : -5 
 
 55 OJ 
 
 v. J3 co 
 
 Oh b 
 
 P. 4> 
 
 UJ 
 
 00 co 
 
 fa o «< ■§ o 
 
Fig. 19. Psaronius melanedrus sp. nov. Diagrammatic cross-section of specimen O, 
 C.B. 1254e bot. #1. Compare fig. 41. Outer cycle unshaded and diagrammed with 
 dashed fines, next inner cycle unshaded and limited by solid fines, next inner one 
 stippled, and most internal cycle is diagrammed as solid black strands. X 2.62. 
 
 Fig. 20. Specimen U. Diagrammatic cross-sections (A-F) of specimen U at suc- 
 cessively higher levels showing method of leaf trace departure. Steles are diagrammed 
 as heavy black fines, the edge of the stem is shown as a thin solid fine, and the 
 peripheral sclerenchyma cutting in about leaf bases is stippled. See text for full ex- 
 planation. Compare F with fig. 51. Specimen 2608. 
 
 83 
 

 V \l^{ & ; 
 
 
 / 
 
 i\ y 
 
 
 f\ rcs\ A 
 
 /CV^ 
 
 <W£ 
 
 <r> ^S"x 
 
 \\&y j j 
 
 Fig. 21. Specimen V. Diagrammatic cross-section (A-I) of specimen V at succes- 
 sively higher levels. Compare diagram G with fig. 52. See text for full explanation. 
 Specimen 2609. 
 
 84 
 
& % 
 
 V=r 
 
 23. b 
 
 2 2. b 
 
 IX. 
 
 4 
 
 
 2 4. 
 
 Fig. 22. Specimen V. Diagrammatic cross-section (A-C), reconstructing the manner 
 in which leaf traces are given off from an internal cycle of specimen V. White and 
 black lines on appropriate cycles designate cauline and potential leaf trace material. 
 The most internal cycle is diagrammed in solid black, the next outer one is stippled, 
 and the most external one depicted here is unshaded. Arrows indicate the directions 
 in which strands will migrate at higher levels. See text for full explanation. Specimen 
 2609. 
 
 Fig. 23. Psaronius blicklei. Diagrammatic cross-sections (A-B) of a portion of a stem. 
 B is at a higher level than A. See text for full explanation. Contributions from the 
 peripheral cauline bundle to the leaf trace and cauline bundles of next inner cycle 
 are diagrammed in solid black, all other vascular tissue is unshaded. Sclerenchyma is 
 stippled. A— C.B. 2413b bot. #1; B— C.B. 2413c top. #1. 
 
 Fig. 24. Psaronius blicklei. Transverse section of a portion of a stem at successively 
 higher levels (A-D). See text for full explanation. Contribution from cycle 4 to the 
 next inner cycle and to a leaf trace is diagrammed in solid black. All other vascular 
 tissue is unshaded. Cycles are consecutively numbered from the outside toward the 
 center. C.B. 2842e bot, A— #1, B— #14, C— #16, D— #17. 
 
 85 
 
Figs. 25-29. Psaronius blicklei sp. nov. 
 
 Fig. 25. Transverse section of specimen A. Washington University C.B. 108a top #6. 
 
 Fig. 26. Transverse section of specimen A at a higher level. Note most central strand 
 representing cycle 3. Washington University C.B. 108g top #1. 
 
 Fig. 27. Transverse section of specimen B. C.B. 2841a top #4. 
 
 Fig. 28. Transverse section of specimen C. C.B. 196c top #3. 
 
 Fig. 29. Transverse section of specimen D. C.B. 2844e bot. #1. 
 

 * 
 
 * 
 
 * 
 
 
 a 
 
 a 
 
 n 
 
 
 o 
 
 o 
 
 a 
 
 
 •^ 
 
 
 •*-* 
 
 
 cC 
 
 o 
 
 -= 
 
 
 Ol 
 
 Ol 
 
 o 
 
 
 00 
 
 TF 
 
 00 
 
 
 1> 
 
 qo 
 
 CD 
 
 > 
 
 CM 
 
 CM 
 
 cq 
 
 o 
 
 
 
 
 fl 
 
 pq 
 
 pq 
 
 pq 
 
 a 
 
 U 
 
 U 
 
 U 
 
 c/3 
 
 
 
 
 S 
 
 w 
 
 fa 
 
 d 
 
 2 <U D 
 
 Ol 
 
 jj 
 
 CO 
 
 u 
 
 
 a) 
 
 c; 
 
 <z> 
 
 CO 
 
 
 
 <a 
 
 CO 
 
 
 O 
 
 CO 
 
 fa 
 
 > 
 
 GO 
 
 
 C 
 
 
 cS 
 
 
 Sh 
 
 
 H 
 
 
 a 
 
 
 CO 
 
 
 s 
 
 
 
 
 fa 
 
 H H 
 
 fa 
 

 Fig. 33. Psaronins blicklei sp. nov. Transverse section of specimen H. C.B. 2681e 
 bot. #1. 
 
 Fig. 34. Stem cross-sections of Psaronius blicklei, showing a lower level of the 
 specimen (A — C.B. 2682b top. #1) with spiral phyllotaxy and a higher level (B — 
 C.B. 2682g top #1) with whorled frond arrangement. Specimen I. 
 
 Fig. 35. Psaronius melanedrus sp. nov. Transverse section of specimen J. Note pro- 
 liferation from leaf base ground parenchyma forming tissue about inner roots. Uni- 
 versity of Minnesota C.B. 35c (1) bot. #1. 
 
Figs. 36-38. Psaronius melanedrus sp. nov. 
 
 Fig. 36. Transverse section of specimen K. Note cells of fundamental tissue of stem 
 have proliferated through the slightly discontinuous sclerenchyma sheath shown at 
 the top of the photograph. Washington University C.B. 698c bot. #4. 
 
 Fig. 37. Transverse section of specimen L. C.B. 1539f top #1. 
 
 Fig. 38. Transverse section of specimen M. Note configuration of leaf bases. Wash- 
 ington University C.B. 697c top #2. 
 
Figs. 39-43. Psaronius melanedrus sp. nov. 
 
 Fig. 39. Transverse section of a portion of specimen M, showing relative thicknesses 
 and tracheid diameters of a leaf trace ( right ) and a peripheral cauline bundle ( left ) . 
 Note heavy sclerenchyma sheath and presence of tannin-like deposits in certain of 
 the parenchyma cells. C.B. 697f top #1. 
 
 Fig. 40. Transverse section of the central portion of the stele of specimen N. Illinois 
 Geological Survey C.B. 803a bot. #1. 
 
 Fig. 41. Transverse section of stele of specimen O. C.B. 1254e bot. #1. 
 
 Fig. 42. Transverse section of stele of specimen P. C.B. 2111a (2) top. #1. 
 
 Fig. 43. Transverse section of specimen Q. C.B. 3066d bot. #1. 
 
 91 
 
m —II 
 
 45 B b 
 
 Figs. 44-46. Psaronius chasei sp. nov. 
 Fig. 44. Transverse section of specimen R. C.B. 1221g bot. #64. 
 Fig 45 Portion of a transverse section of specimen R showing an adventitious root 
 trace in connection with the second cycle. Four other roots may alsc> be obsenred 
 in the sclerenchyma sheath on the adaxial side of the leaf base. C.B. 1221f top #33. 
 
 Fig. 46. Portion of a transverse section of specimen R, showing ^secondary 
 parenchyma exceeding in radial extent the most externally placed root. C.B. 1221g 
 bot. #64. 
 
Figs. 47-49. Psaronius Chasei (?) sp. nov. 
 
 Fig. 47. Transverse section of specimen in Illinois Geological Survey C.B. 140a #2. 
 
 Fig. 48. Transverse section of specimen S. C.B. 1700d top #1. 
 
 Fig. 49. Transverse section of a portion of a specimen in C.B. 1607a top #1. 
 
1 2 3 4 5 
 
 Fig. 50. Transverse section of specimen T. Illinois Geological Survey C.B. 5ld bot. 
 #1. 
 
 94 
 
m 
 
 /? 
 
 ^ H 
 
 
 
 <u 
 
 A* 
 
 -If) 
 
 ro 
 <\j 
 
 BO 
 
 "*V* 
 
 KJ *^" -■ ^SanpHi^fffe -■ 
 
 "" l *N, 
 
 ^ 
 
 
 *. ' 
 
 52 
 
 ro 
 
 OsJ 
 
 a. 
 
 Fig. 51. Transverse section of stem of specimen U. Specimen 2608a bot. #1. 
 Fig. 52. Transverse section of specimen V. Specimen 2609g top #1. 
 Fig. 53. Transverse section of specimen W. Specimen 728a top #1. 
 
Fig. 54. Transverse section of Psaronius brasiliensis. Specimen at the Divisao de 
 Gealogia e Mineralogia, 404 Aveaida Pasteur, Praia Vermelha, Rio de Janeiro, 
 Brazil. Since Fig. 54 is of the lower face of the piece, the negative was reversed in 
 printing to make the picture correspond with Fig. 55, the upper face. 
 
 Fig. 55. Transverse section cf Psaronius brasiliensis at a higher level. 
 

 
 ro 
 
 -M^*""' 
 
 .* .'-'••'j 
 
 
 
 
 
 Stt""*"- '" 
 
 *t - 
 
 
 "sTt 'if * ' 
 
 K*.^ - 
 
 (\J 
 
 tfi 
 
 v V 
 
 
 m' 
 
 J 
 
 
 MX,'* 
 
 .'►. 
 
 
 8*$^' 
 
 ffl 
 
 
 SP**r 
 
 ->»o — 
 
 — 
 
 
 £# 
 
 
 
 
 
 St.*"' '-* \- 
 
 "-*!»« • 
 
 
 
 
 o 
 
 Fig. 56. Transverse section of Psaronius brasiliensis. After Zeiller (1890). 
 
 Figs. 57-59 Psaronius blicklei 
 
 Fig. 57. Transverse section of a portion of Psaronius blicklei, showing an adventitious 
 root trace in connection with a peripheral cauline bundle. Note sieve cells about 
 
 7*Sl f" an J' acunae > and tannin-like deposits in cells of fundamental tissue. CB 
 zoolk bot. #1. 
 
 Fig. 58. Longitudinal section showing vascular and fundamental tissue s— sieve 
 cells; t — tracheids; p — parenchyma. CB. 2413a edge #3. 
 
 Fig. 59. Transverse section of a portion of a stem. Note lacunae and configuration of 
 sclerenchyma. CB. 2681a top #2. 
 
Figs. 60-62 Psaronius blicklei sp. nov. 
 
 Fig. 60. Transverse section of a portion of a stem showing secondary parenchyma 
 around a peripheral cauline bundle and in the fundamental tissue. C.B. 2680h top 
 
 #2. 
 
 Fig. 61. Portion of a transverse section showing leaf trace distorted by growth of 
 secondary parenchyma. It — leaf trace; p — parenchyma; pob — peripheral cauline 
 bundle. C.B. 2371c bot. #2. 
 
 Fig. 62. Transverse section of a portion of Psaronius blicklei, showing growth of 
 secondary parenchyma into lacunae of the stem. C.B. 2371a bot. #1. 
 
Figs. 63-65 Psaronius chasei 
 
 Fig. 63. Transverse section of a portion of Psaronius chasei. Note sieve cells around 
 xylem strands. C.B. 1221f top #19. 
 
 Fig. 64. Portion of a vascular bundle greatly enlarged. C.B. 1221g bot. #22. 
 
 Fig. 65. Longitudinal section of a portion of a metaxylem tracheid, showing scalari- 
 form secondary thickenings. C.B. 1221g edge #1. 
 

 2 ii ft 
 
 •5 -^ c 
 
 2 OT) 
 
 <3 n ■ _l 
 
 r? S »> 
 
 ^ C CO 
 
 ^ .2 N 
 
 o . 
 
 len 
 
 next 
 
 C.B 
 
 
 .5 <u 
 
 8£ u 
 
 Dj 
 
 « Ti =tt= 
 
 <-H 
 
 o «j ft 
 
 ^ o 2 
 
 9 g 3 
 
 < r s 
 
 
 rO —^ co 
 
 
 ^ jj c*l 
 
 
 > rt PQ 
 
 
 ^ o (J 
 
 
 >h w 
 
 
 D CD J. 
 
 
 CM A = M 
 
 
 
 
 >,^S: 
 
 
 T! CD 
 
 
 1^ & 
 
 
 - 8 S^ 
 
 
 id h W 
 
 
 CD Ol 
 
 
 rt -5 • 
 
 E 
 
 » - ft ffl 
 
 £ 
 
 ° 1 &<? 
 
 
 a) o <! 
 
 D§ C'W 
 
 £ £ .2 * 
 
 msve 
 tribu 
 lanat 
 top 
 
 B § &2 
 
 our -1 
 
 H-l t> 
 
 
 CD '> * . 
 
 "S o b « 
 
 wa ~S ^ V 
 
 • ., +> 
 
 CD -~ X J. 
 
 CD ^ D Q 
 
 . 2 * J .„ 
 
 U .9 « «tf 
 
 
 M hS (U -J. 
 
 [ii ~c en =fp 
 

 J«££ & - > 
 
 *^j£.J~ ~*&*2? «Jo 
 
 Fig. 67. Transverse section of a portion of Psaronius hlicklei, showing a leaf trace 
 attached to two cycles at the same level. Note lacunae, sclerenchyma bands, and 
 isolated sclerenchyma strands. C.B. 2842e bot. #71. 
 
 Fig. 68. Transverse section of a portion of Psaronius chasei. At a a small strand of 
 vascular tissue is projecting from the surface of a cauline bundle. See text for full 
 information. C.B. 1221h bot. #24. 
 
 Fig. 69. Psaronius blicklei. Transverse section of a portion of a stem showing leaf 
 trace characteristics. Illinois Geological Survey C.B. 371g top #1. 
 
 101 
 
»2tl af PS T7 f CkleL SurfaCe ° f a St6m Sh ° wing leaf trace characteristics. Note 
 points of exit of adventitious roots from stem above leaf scar and roots over lower 
 portion of leaf scar. C.B. 196d (3). 
 
 Fig. 71. Psaronius sp. Weathered surface of a stem. Specimen 2625. 
 
 mTside NoTeT ""w Y^ °i * *"' Sh ° Wing leaf baSe ^acteristics from 
 me side. Note lacunae in fundamental tissue. C.B. 2777b. 
 
 Fig. 73. Psaronius blicklei. Surface of a slightly weathered stem showing leaf scar 
 characterises and lacunae. Note roots over leaf scar at right. C.B. 2777b 
 
74 
 
 icmj i i 
 
 01 234 5678 
 
 75 
 
 
 
 Fig. 74. Psaronius giffordii. Surface of stem showing leaf scar characteristics. Speci- 
 men 2612. 
 
 Fig. 75. Psaronius sp. Surface of stem showing leaf scar characteristics. Specimen 
 2622. 
 
 Fig. 76. Psaronius blicklei. Transverse section of a portion of stem and inner root 
 zone at the position of a leaf base. Note proliferation of fundamental cells of the 
 leaf base and tannin-like deposits in certain of the cells of the roots. C.B. 2681k 
 bot. #1. 
 
Fig. 77. Psaronius blicklei. Transverse section of a free root which is branching. 
 Washington University C.B. 108c bot. #3. 
 
 Fig. 78. Psaronius chasei. Transverse section of the inner root zone showing pro- 
 liferation of parenchyma cells. C.B. 1221 top #47. 
 
 104 
 
Figs. 79, 80. Psaronius giffordii 
 
 Fig. 79. Transverse section of stem. Photograph of slide of specimen 5, University of 
 Chicago Collection. 
 
 Fig. 80. Transverse section of stem. Photograph of slide of specimen 3, University of 
 Chicago Collection. ' 
 
 105 
 
r*e- 
 
 Figs. 81, 82. Psaronius pertusus Blickle, sp. nov. 
 
 Fig. 81. Transverse section of a stem. Specimen B — 202 in the Ohio University 
 Museum Collection. 
 
 Fig. 82. Portion of a transverse section of a stem showing fundamental tissue in 
 detail. Illinois Geological Survey C.B. 147b top #1. 
 
INDEX 
 
 Adiantum, 49 
 Angiopteris, 43, 60, 61 
 Angiopteris evecta, 44 
 Arnold, 66 
 Aster otheca, 61 
 
 Baxter, 17, 19, 55, 58, 63, 64, 69 
 Blickle, 14, 43 
 Bower, 46, 60, 61 
 Brogniart, 40, 42, 64, 65 
 
 Caulopteris, frontispiece, 54, 55, 58 
 
 Christ, 46 
 
 Christensenia, 61 
 
 Cibotium, 42 
 
 Cotta, 1, 2 
 
 Cyathea, 19 
 
 Cyatheaceae, 1, 20, 46, 60 
 
 Decodon verticillatus, 57 
 Dicksonia antarctica, 61 
 Dicksoniaceae, 60, 61 
 
 Endogenites, 1, 2 
 Esau, 18 
 
 Farmer and Hill, 55, 57, 60, 69 
 Farr, 2 
 
 Gillette, 17, 46, 63, 64 
 
 Herzer, 63 
 
 Hirmer, 2, 17, 38, 42, 46, 63 
 
 Hoskins, 63 
 
 Jeffrey, 61 
 
 Kuhn, 44 
 
 Lenz, 55 
 
 Lindley and Hutton, 58 
 
 Macbride, 63 
 
 Marattia, 60, 61 
 
 Marattiaceae, 1, 43, 44, 60, 61, 62 
 
 Marattiales, 1 
 
 Marsilea, 6, 57 
 
 Matonia, 61 
 
 Megaphijton, 54 
 
 Mettenius, 60 
 
 Moon, 16, 45 
 
 Morgan and Delevoryas, 55 
 
 Nehimbo, 57 
 
 Ogura, 19 
 
 Pelourde, 42, 65 
 
 Pohlig, 45 
 
 Psaroniaceae, 1 
 
 Psaronius, 2-3; comparisons with other 
 ferns, 60-62; distichous form, 36- 
 38, fig. 51; fundamental tissue, 43, 
 figs. 58, 81-82; general discussion, 
 46-55; generic description, 62; his- 
 tology, figs. 57-62; historical back- 
 ground, 1-2; lacunae, 43-44, fig. 
 59; listing of valid species, 63-68; 
 phloem, 46, figs. 63-64; reduction 
 of certain species to synonymy, 62- 
 65; root zone, 55-60, figs. 45-46, 
 57, 77-78; sclerenchyma, 44-45, 
 fig. 39; secondary proliferation of 
 parenchyma, figs. 46, 60, 62; 
 sources of specimens, 2-3; strati- 
 graphic localities, 3-4; taxonomic 
 position, 62; taxonomic treatment 
 of all species, 62-68; vascular sys- 
 tem, 46-55; xylem, 45; figs. 1-24 
 25-44, 51-56, 58 
 
 Psaronius blicklei: general discussion, 
 17-21; general morphology, 4-17; 
 specific diagnosis, 16-17, 66-67; 
 figs. 1-15, 23-24, 57-62, 66-67, 
 69-70, 72-74, 76-77 
 
 Psaronius borealis, 63 
 
 Psaronius brasiliensis: additional ob- 
 servations, 40-42; general discus- 
 sion, 42-43, 61; specific description, 
 64-65; figs. 54-56 
 
 107 
 
108 
 
 AMERICAN SPECIES OF THE GENUS PSARONIUS 
 
 Psaronius chasei: general discussion, 
 34-36; general morphology, 32-34; 
 specific diagnosis, 34-67; jigs. 44- 
 49, 63-65, 68, 78 
 
 Psaronius cooksonii, 59, 63, 64 
 
 Psaronius giffordii, 63, 64; figs. 74, 
 79-80 
 
 Psaronius gutbieri, 38 
 
 Psaronius illinoensis, 63 
 
 Psaronius infarctus, 8, 51 
 
 Psaronius junceus, 63 
 
 Psaronius lacunosus, 43, 66 
 
 Psaronius melanedrus: general discus- 
 sion, 30-32; general morphology, 
 21-30; specific diagnosis, 29, 30, 
 67, 68; figs. 17-19, 36-43 
 
 Psaronius musaeformis, 38 
 
 Psaronius peoriensis, 63, 64 
 
 Psaronius pertusus, 43; specific de- 
 scription, 65-66; figs. 81-82 
 
 Psaronius septangulatus, 54, 63 
 
 Psaronius simplex, 38 
 
 Psaronius simplex F. integer, 38 
 
 Psaronius sp. morphology of previ- 
 ously undescribed specimens lack- 
 ing specific designations, 36-40 
 
 Ptychocarpus, 61 
 
 Ptychopteris, 58 
 
 Reed, 6, 56, 57, 69 
 Rudolph, 46, 62 
 
 Saccoloma elegans, 61 
 
 Sahi, 57, 69 
 
 Schoute, 2, 46 
 
 Schuster, 45 
 
 Scolecopteris, 61 
 
 Scolecopteris elegans, 61 
 
 Scott, 2, 42 
 
 Seward, 2, 60, 61, 62 
 
 Solms-Laubach, 55, 57, 69 
 
 Spiropteris, 21 
 
 Sprengel, 1 
 
 Staarsteine, 1 
 
 Stenzel, 1, 2, 17, 38, 46, 57, 60, 62, 
 
 69 
 Sternsteine, 1 
 Sterzel, 45 
 
 Stewartiopteris, 35, 36, 55, 62 
 Stipitopteris, 1, 19, 35, 53, 55, 61, 62 
 Strasburger, 61 
 
 Thyrsopteris elegans, 61 
 Tietea singularis, 62, 65 
 
 Unger, 65 
 
 West, 4, 5, 43, 44 
 Wurmsteine, 1 
 
 Xylopsaronius, 45 
 
 Zeiller, 2, 11, 17, 42, 43, 46, 51, 52, 
 58, 69 
 

 HOFFMEISTER, DONALD F. (1951) : A TaXO- 
 
 nomic and Evolutionary Study of the 
 Pinon Mouse, Peromyscus truei. 5 pis. 24 
 figs. Vol. 21, No. 4. Cloth only, $3.50. 
 
 HOFFMEISTER, DONALD F., and GOODPAS- 
 TURE, woodrow w. (1954): The Mam- 
 mals of the Huachuca Mountains, South- 
 eastern Arizona. 27 figs. Vol. 24, No. 1. 
 $3.00. 
 
 HOPKINS, SEWELL HEPBURN (1934): The 
 Papillose Allocreadiidae— A Study of Their 
 Morphology, Life Histories, and Relation- 
 ships. 4 pis. 6 figs. Vol. 13, No. 2. $1.00. 
 HUMES, ARTHUR grover (1942): The 
 Morphology, Taxonomy, and Bionomics of 
 the Nemertean Genus Carcinonemertes. 4 
 pis. 1 map. Vol. 18, No. 4. $1.50. 
 
 HUNTER, GEORGE WILLIAM, III (1930): 
 
 Studies on the Caryophyllaeidae of North 
 America. 7 pis. Vol. 11, No. 4. $1.25. 
 jones, george NEVILLE (1946) : American 
 Species of Amelanchier. 23 pis. 14 maps. 
 Vol. 20, No. 2. Cloth only, $2.00. 
 
 kendeigh, s. Charles (1941): Territorial 
 and Mating Behavior of the House Wren. 
 32 figs. Vol. 18, No. 3. $1.50. 
 
 ( 1952) : Parental Care and Its Evo- 
 lution in Birds. 35 figs. Vol. 22, Nos. 1-3. 
 $4.00. 
 
 kindred, james e. (1919): The Skull of 
 Amiurus. 8 pis. Vol. 5, No. 1. $1.25. 
 
 kramer, sol (1950): The Morphology 
 and Phylogeny of Auchenorhynchous 
 Homoptera (Insecta). 6 charts. 16 pis. Vol. 
 20, No. 4. $2.00. 
 
 KUDO, RICHARD ROKSABRO (1934): Studies 
 
 on Some Protozoan Parasites of Fishes of 
 Illinois. 8 pis. Vol. 13, No. 1. $.75. 
 
 (1944): Morphology and Develop- 
 ment of Nosema notabilis Kudo; Parasitic 
 in Sphaerospora polymorpha Davis, a Para- 
 site of Opsanus tau and O. beta. 12 pis. 7 
 figs. Vol. 20, No. 1. $1.25. 
 
 mackin, john GiLMAN (1936): Studies on 
 the Morphology and Life History of Nem- 
 atodes in the Genus Spironoura. 6 pis. 2 
 figs. Vol. 14, No. 3. $1.00. 
 
 MANTER, HAROLD WINFRED (1926): Some 
 
 North American Fish Trematodes. 6 pis. 2 
 charts. 1 fig. Vol. 10, No. 2. $1.50. 
 
 may, henry gustav (1920): Contribu- 
 tions to the Life Histories of Gordius to- 
 bustus Leidy and Paragordius varius 
 (Leidy). 21 pis. Vol. 5, No. 2. $1.50. 
 
 meglitsch, paul a. (1940): Cytological 
 Observations on Endamoeba blattae. 8 pis. 
 Vol. 17, No. 4. $1.00. 
 
 miller, edwin lynn (1936): Studies on 
 North American Cercariae. 8 pis. Vol. 14, 
 No. 2. $1.50. 
 
 mizelle, john dary (1938): Compara- 
 tive Studies on Trematodes (Gyrodacty- 
 loidea) from the Gills of North American 
 Fresh- Water Fishes. 5 pis. Vol. 17, No. 1. 
 $1.00. 
 
 nichols, ray janney (1937): Taxonomic 
 
 Studies on the Mouth Parts of Larval 
 
 Anura. 3 figs. 8 charts. Vol. 15, No. 4. 
 
 $1.00. 
 
 ray, james davis, jr. (1956): The Genus 
 
 Lysimachia in the New World. 20 pis. 11 
 
 maps. Vol. 24, Nos. 3-4. $2.50. 
 
 roback, selwyn s. (1954) : The Evolution 
 and Taxonomy of the Sarcophaginae (Dip- 
 tera, Sarcophagidae) . 9 charts. 34 pis. Vol. 
 23, Nos. 3-4. $4.00. 
 
 ross, Herbert h. (1931): Sawflies of the 
 Sub-Family Dolerinae of America North 
 of Mexico. 6 pis. Vol. 12, No. 3. $1.25. 
 
 ( 1937) : A Generic Classification of 
 
 the Nearctic Sawflies (Hymenoptera, 
 Symphyta). 17 pis. Vol. 15, No. 2. $2.00. 
 
 SCHOOF, HERBERT FREDERICK (1942): The 
 
 Genus Conotrachelus Dejean (Coleoptera, 
 Curculionidae) in the North Central 
 United States. 9 pis. Vol. 19, No. 3. $1.50. 
 
 schopf, james morton (1943): The Em- 
 bryology of Larix. 6 pis. 86 figs. Vol. 19, 
 No. 4. $1.50. 
 
 smith, frank (1923): The Calciferous 
 Glands of Lumbricidae and Diplocardia. 
 12 pis. Vol. 9, No. 1. $1.25. 
 
 SOLHEIM, WILHELM GERHARD (1930): 
 
 Morphological Studies of the Genus 
 Cercospora. 4 pis. Vol. 12, No. 1. $1.00. 
 
 spooner, charles stockman (1938) : The 
 Phylogeny of the Hemiptera, Based on a 
 Study of the Head Capsule. 24 pis. 2 figs. 
 Vol. 16, No. 3. $1.00. 
 

 <|K#iilH^ll 
 
 sprague, victor (1941): Studies on 
 Gregarina blattarum with Particular Ref- 
 erence to the Chromosome Cycle. 6 pis. 
 2 figs. Vol. 18, No. 2. $.75. 
 stannard, lewis j., jr. (1957) : The Phy- 
 togeny and Classification of the North 
 American Genera of the Suborder Tubu- 
 lifera (Thysanoptera). 14 pis. No. 25. 
 $2.50. 
 
 Stevens, frank Lincoln (1924): Para- 
 sitic Fungi from British Guiana and Trini- 
 dad. 19 pis. 1 map. Vol. 8, No. 3, $1.25. 
 
 (1927) : Fungi from Costa Rica and 
 
 Panama. 18 pis. 1 map. Vol. 11, No. 2. 
 $1.25. 
 
 and RYAN, SISTER MARY HILAIRE 
 
 (1939): The Microthyriaceae. Vol. 17, 
 No. 2. $1.50. 
 
 STICKNEY, FENNER SATTERTHWAITE 
 
 (1923) : The Head-Capsule of Coleoptera. 
 26 pis. Vol. 8, No. 1. $2.00. 
 tehon, leo roy (1935): A Monographic 
 Rearrangement of Lophodermium. 5 pis. 
 Vol. 13, No. 4. $2.00. 
 
 van cleave, harley j. (1953): Acantho- 
 cephala of North American Mammals. 13 
 pis. 10 figs. Vol. 23, Nos. 1-2. $4.00. 
 
 VAN DEVENTER, WILLIAM CARL (1937): 
 
 Studies on the Biology of the Crayfish 
 Cambarus propinquus Girard. 46 graphs. 
 Vol. 15, No. 3. $1.00. 
 
 vestal, Arthur G. (1949): Minimum 
 Areas for Different Vegetations; Their De- 
 termination from Species-Area Curves. 6 
 figs. Vol. 20, No. 3. $2.00. 
 
 WATSON, MINNIE ELIZABETH (1916): 
 
 Studies on Gregarines I— Including De- 
 scriptions of Twenty-One New Species 
 and a Synopsis of the Eugregarine Rec- 
 ords from the Myriapoda, Coleoptera, and 
 Orthoptera of the World. 15 pis. Vol. 2, 
 No. 3. $2.00. 
 
 yuncker, truman George (1921); Re- 
 vision of the North American and West 
 Indian Species of Cuscuta. 13 pis. Vol. 6, 
 Nos. 2-3. $2.00. 
 

 BBBhSS 
 
 KHHe 
 
 BBS 
 
 I - .'■.*'.*;..'!tS 
 
 SB 
 
 t 
 
 ■ 
 
 MHH