THE UNIVERSITY OF KANSAS PALEONTOLOGICAL CONTRIBUTIONS November 30, 1967 Paper 24 REVISION OF SUBORDER CYATHOCRININA (CLASS CRINOIDEA) N. GARY LANE Department of Geology, University of California, Los Angeles ABSTRACT The suborder Cyathocrinina is divided into three superfamilies named Palaeocrinacea, Cyathocrinitacea and C,odiacrinacea. Each of the first two includes families judged by MOORE & LAUDON (1943) to be closely related, one superfamily differing from the other in the presence or absence of an anal sac. The Codiacrinacea is a new superfamily containing crinoids which earlier were grouped in the family Codiacrinidae, but now are thought to represent three separate lineages that merit separation at the family level. The Codiacrini- dae, as restricted, includes Codiacrinacea with five arm-bearing radials, and such crinoids range from the Devonian into the Permian. The new family Sycocrinitidae includes Lower Carboniferous to Permian genera that have only the D radial bearing an arm, the other radials being either present or absent. The new family Streblocrinidae consists of abrachiate (armless) crinoids in which the radials atrophied and disappeared during the Devonian, and pseudomonocyclic genera without radials which continue into the Permian. INTRODUCTION During preparation of materials on the sub- of grouping these three families into a separate order Cyathocrinina for Part T (Class Crinoidea) superfamily became evident. many years superfamilies were not used in for the Treatise on Invertebrate Paleontology, it For fossil crinoids, which were became evident that revision of the family Codia- the classification of ranked into subclasses and orders with their con- crinidae BATHER, 1899, was called for before that tained families. Relatively few suborders were group of crinoids would be in a taxonomic frame- recognized by BATHER (1899) and by MOORE & work suitable for inclusion in the Treatise. The LAUDON (1943), but the classification was thor- family included three phyletically distinct assem- oughly revised in 1953 by UBAGHS who established that were thought to warrant blages of crinoids several new suborders and superfamilies, especially of the three separation at the family level, two among the camerate and inadunate crinoids. Con- groups having separate evolutionary history from sequently, ample precedent exists for division of the Devonian through the Permian. Nevertheless, the Cyathocrinina at the superfamily level. these three lineages are thought to have had a MOORE & LAUDON (1943, p. 37) suggested that common, pre-Middle Devonian ancestry and to families which they classed in the Cyathocrinina stand closer to each other than to other families of fell into two relatively distinct groups, one consist- the Cyathocrinina. Consequently, the desirability ing of earliest known cyathocrinoids as well as 2 The University of Kansas Paleontological Contributions—Paper 24 specialized Silurian and Devonian families, all of the suborder, exclusive of the codiacrinids, share which have a low tegmen and an anal opening several common features, the most important of directly through the tegmen or the side of the cup. which is that all have an elongate anal sac. These These families are grouped here into the super- families are assembled as the superfamily Cyatho- family Palaeocrinacea. The remaining families in crinitacea (Fig. 1). SUPERFAMILY CODIACRINACEA The crinoids that have been grouped, until almost exclusively of small crinoids known from now, in the family Codiacrinidae are a diverse thecae with arms and columns rarely preserved. assemblage that includes 31 nominal genera. Pale- Some individuals are small, only 1 or 2 mm. in ontologists have disagreed concerning evolutionary maximum dimension, and can be classed as micro- trends within this group and their differing opin- crinoids, whereas others are several times larger, ions about the phylogeny of its components never but rarely more than 1 cm. in height. The dorsal have been resolved satisfactorily, mainly because cup typically is higher than wide, with all plates there has been no comprehensive review of all visible in side view, and consists of either two or genera that attempts to fit them into a coherent three circlets of plates. Infrabasals may number evolutionary pattern. In recent years important five, three, or be fused into a single plate; basais new studies on oldest known codiacrinaceans from are the five largest plates in the cup, although ex- Devonian and Mississippian rocks have shed light ceptionally only three basais may be present. The on the evolutionary history of these crinoids. These radials are usually small, may or may not have factors need to be taken into account in a revision arm-bearing articular facets developed on their such as this, and necessitate discussion too lengthy distal edges, and may number five, three, or be for appropriate inclusion in the Treatise. absent. Those genera characterized by absence of The superfamily Codiacrinacea is composed radials have a two-circlet pseudomonocyclic cup, the remaining circlets being the infrabasals and basais, rather than the basais and radials, as in the true monocyclic condition. One important aspect of study o of codiacrinaceans is the necessity for dif- -o ferentiating diminutive pseudomonocyclic mem- bers included o in the superfamily from small, 0 o o truly monocyclic crinoids 0 -o of the disparid family Q) Allagecrinidae. (.) o (.5 o 0 CRITERIA FOR DISTINGUISHING DIMINUTIVE CODIACRINACEANS AND ALLAGECRINIDS .cEn Small monocyclic allagecrinids, like Kallimor- Q) phocrinus and Hybochilocrinus, bear close resem- o blance to microscopic codiacrinaceans such as Coenocystis or Hemistreptacron. In these two unrelated groups the dorsal cup has only two cir- (./5 clets of plates, basais and radials in the allagecri- nids and infrabasals and basais in the codiacrina- ceans. In the latter assemblage the radials are postulated to have atrophied and disappeared, re- o sulting in a pseudomonocyclic condition. The fol- 6_ lowing morphologic criteria are available for dis- tinguishing these two groups of microcrinoids. 1) Allagecrinids, even small specimens less Flo. 1. Evolution of suborder Cyathocrinina. than 1 mm. in maximum dimension, have arm- Lane—Revision of Suborder Cyathocrinina (Class Crinoidea) 3 0 0 Pennsylvanian 0000 0 C7 O Clistocrinus innnan 6 0 6.00 uuuuu 9 ç_i ODU Lageniocrinus Dichostreblocrinus Mississippian of10,,oronon OVoroV&9 L ampadosocrinus AAA() 0 0 0 0U 0U Pentececrinus Ann,nn Devonian oanon 0 0\1 0 Q Streblocrinus Ç17. ‹R z Tytthocrmus Silurian Thalamocrinus FIG. 2. Devonian-through-Pennsylvanian genera of Streblocriniciae, n.fam. [The A ray and AE interray are the farthest right plates in each diagram. Radials are black and anal plates have vertical ruling.] 4 The University of Kansas Paleontological Contributions—Paper 24 bearing articular facets on the upper edges of one NIEWOEHNER, from Devonian-Mississippian transi- or more of the second circlet of cup plates. These tion beds, which is the oldest known abrachiate plates are correctly designated as radials and the (armless) codiacrinacean, it seems clear that two proximal circlet as basais. Codiacrinaceans with morphologically dissimilar groups of these crinoids only two circlets of plates in the cup do not display persisted from Devonian through Permian time. articular facets on plates of the second circlet. One of the groups lacked radial plates from the These plates are judged to be basais and the prox- beginning, so were pseudomonocyclic (Fig. 2, 3). imal circlet infrabasals, the radials having atro- The tendency for radials to atrophy and disappear phied and disappeared. The difference cannot be is foreshadowed by the Middle Devonian genus ascribed to ontogenetic change—that two-circlet Streblocrinus KoENIG & MEYER (1965), which has cups without facets are immature monocyclic only three small radial plates. Codiacrinaceans forms that have not yet developed facets on the with five arm-bearing radials begin with Codia- second (radial) circlet—because many specimens crinus in the Devonian and continue through to of two-circlet codiacrinaceans are several times the Permian (Fig. 4, 5). A third group, typified larger than much smaller arm-bearing allagecri- by Sycocrinites, has five radials, only one of which nids, yet do not display facets. Conclusive evidence is arm-bearing, the others being reduced in size for this argument is found in codiacrinaceans and lacking radial facets (Fig. 6). This group is with three-circlet cups like that of Embryocrinus first found in Lower Carboniferous rocks, and (Perm.) or Streblocrinus (Dey.), in which the continues into the Permian. radials are much reduced in size, lack facets, and At least two possible phyletic schemes may are at the point of disappearing. account for the temporal distribution of genera in 2) Many allagecrinids have radial plates un- the three morphologic groups distinguished on equally developed in size, and some have differ- the basis of number of arm-bearing radials. ences in the number of arms borne by radials Informally we may designate the three groups within the same cup. These allagrecrinids consis- as abrachiate, monobrachiate, and pentabrachiate tently display a bilateral symmetry plane in an codiacrinaceans. E-BC direction. No reported codiacrinacean has Each of these three groups could represent a this symmetry direction, even though asymmetric separate phyletic lineage. The abrachiate genera radials are known in several genera. developed in