GENERA OF THE SUBORDER PENTAMEROIDEA 



161 



than additional strengthening of the palintropc. The 

 brachiophore supporting plates are well developed, dis- 

 crete, and convergent toward the floor of the valve, 

 defining a rather deep notothyrial cavity. The brach- 

 iophores, judging by internal molds, were long, slender 

 processes extending from the point of union of the 

 accessory septal plate and the palintrope. 



In the ventral valve a feature of particular interest 



is the shallow pit at the front end of the pseudospon- 

 dylium. Such a pit also occurs in Syntrophina, and is in 

 exactly the same position as the anterior adductor scar 

 in Billingsella. Behind this pit can be seen the central 

 adductor track and on each side of this, impressions 

 which are regarded as diductor scars. This antero- 

 central pit may represent the final place of the adductor 

 scar in its forward migration during growth. 



Superfamily PENTAMERACEA Schuchert 1896 



Specializing Protremata probably originating in the 

 Syntrophiacea (family Syntrophiidse ) , characterized by 

 spondylia that in the primitive families are nearly 

 always supported by single septa (^^ spondylium sim- 

 plex) and in the derived families by double septa 

 (coalesced septal plates = spondylium duplex), but in 

 rare instances in any of the families may hang free 



TabU 17 



Pentameridx 



-^Stricklandidae 



Camerellidae 



Syntrophiidx 



or posteriorly supported by a remnant of the septum. 

 The more important family characters occur in the 

 dorsal cardinalia, which may be discrete or medially 

 coalesced into a cruralium that bears the adductor mus- 

 cles. Cardinal process usually absent, or rudimentary 

 as a linear ridge or slight boss. Rudimentary or modi- 

 fied remnants of the deltidium occur rarely, and chilidia 

 never. Shells narrow-hinged, with small interareas, 

 or decidedly rostrate with plane areas; exterior smooth, 

 costate or multicostate. Test fibrous, and, so far as 

 known, without endopunctje. 



Geologic range. — Middle Ordovician to close of 

 Devonian. 



Embraces the following families: 



Camerellidae Hall and Clarke 



Pentamerida; McCoy 



? Stricklandidjc Hall and Clarke 



The genetic relationships appear to be as shown in 

 Table 17. 



After completing their survey of the great group of 

 orthids, the writers turned naturally and logically to a 

 study of the Pentameracea. They had embarked on 

 this course and carried on their research for about six 

 weeks when it was interrupted by the removal of the 

 junior author to Washington. This had at once dis- 

 advantages and advantages. At Washington it was 

 possible to add observations on the strnnge genera 

 Brooksina and Harpidium and on the even more 

 aberrant Cymbidium. In the National Museum col- 

 lections also are some fine specimens of Orthotropia 

 which show the necessity for removal of this genus to 

 the Dalmanellacea near Linoporella. 



As the presentation of the pentamerids now stands, 

 it represents a survey of two of the finest collections 

 of these shells in this country. However, the work 

 can not be considered as final. In any study such as 

 this, in which the elucidation of internal details depends 

 on the destruction of materials, it would have been 

 desirable to have still more specimens. As an example 

 of this need it might be stated that, in all of the speci- 

 mens of Pentamerus studied, we did not find a single 

 example which retained a deltidium, yet Hall and 

 Clarke reported such a structure. Furthermore, 

 where there was abundance of common species, the 

 rarer forms were represented in each collection by 

 one or two specimens which could not be sacrificed to 

 serial sectioning. Therefore much remains still to be 

 done. We have, however, seen, and in the majority 

 of cases sectioned, specimens representing the genotype 

 of nearly all the pentamerid genera. If nothing else, 

 the study strives to make clear exactly what each genus 

 is, and, as far as possible, to eliminate from each the 

 non-typical species. It has therefore been necessary to 

 create a few new genera. Besides Orthotropia, the 

 probability is that Str'icklandla will eventually have to 

 be removed from its honored place among the pentam- 

 erids. On the other hand, as we have shown on earlier 

 pages, the Porambonitid.'c belong with the Orthacea. 

 The old family Pentamerid.-c is, however, now divided 

 into three, the Camerellida;, Stricklandidas, and the 

 very varied Pentameridae, on the basis of certain in- 

 ternal characteristics. Recent studies by Kozlowski 

 have shown that the Camerophoriidae are spondylia- 

 bearing Rhynchonellacea. 



