80 SYNOPSIS OF AMERICAN FOSSIL BRACIIIOPODA. (uull.st. 



Greatest representation in both genera and S])ecies was during' the 

 Ordovk-ian. after wliicli jjeneric differentiation was i)ractically restricted 

 to the DiscinidiL' and Craniidic. Crania persisted tlironghont the post- 

 Ordovician, and for longevity equals the atrematous genus Lingula. 



The percentage of widely dispersed S))ecies is about the same as in 

 the Atremata, and likewise is greatest in those families with the longest 

 jdiylogenetic history, as Acrotretidie, Discinida;, and Craniida^ 



Development was along two lines. In one a broad fissure (the most 

 primitive condition of the pedicle oi)ening in this order) is retained as 

 a mature character (Treimitida'). Later geologically, and at the matu- 

 rity of the individual in derived forms, the fissure is gradually closed 

 posteriorly, leaving a long, narrow slit, at one end of which the pedicle 

 emerges (Discinida'). The other line ( Acrotretacea) probably develoi)ed 

 and inherited holoperipheral growth in the ventral valve, very rapidly 

 producing a small subceutral circular foramen, since this feature is 

 already well developed in tlie Lower Cambrian Acrotretidte, and in 

 advance of the greatest development of the l)iscinid;e. It is probably 

 this second branch that gave origin to the degraded family Crauiidte. 

 The protegulum in the dorsal valve of Acrotretacea is probably always 

 marginal, whereas in the Discinacea it is always more or less central. 



It is remarkable that Crania, so unlike other living brachiopods 

 and occurring abundantly in the seas of to-day, has never been com- 

 pletely studied developmen tally or ontogenetically. The taxonomic 

 position of the Craniida^ is therefore not actually determined, and Hall 

 and Clarke incline to follow Waagen in regarding the Craniacea as 

 equivalent in rank to the Atremata and Neotremata. These authors 

 'write : ' 



It is nevertheless to be observed tliat uo trace of a former pedirle-slit incision or 

 perforation is found on mature or iuimaturo shells, and it would be difficult to co7n- 

 prchend in what manner such an essential modification of the shell could be wholly 

 concealed by later growtli. Were the pedicle marginal in primitive growth stages, 

 and subsequently atrophied, the obliteration of the marginal opening by later 

 resorption and growth would be a readily intelligible process. There is, hence, in 

 this default of evidence, a good reason to doubt the close affinities of Crania and 

 Pholidops to the Diacaulia [= Neotremata]. Present knowledge would seem to 

 indicate that they Avere primarily of the type of the Mesocaulia [= Atremata], and 

 that their resemblance to the Diacaulia is wholly of secondary growth. Waagen's 

 term for this group, Gastrojiegmata (or Craniacea), may therefore prove to be 

 equivalent to each of these other two divisions. 



Brachiopod embryolog*y demands a pedicle in the early stages of 

 Crania. The ventral valve carries the pedicle, and it is always this 

 valve Avhi('h is attached by cementation or otherwise. The writer has 

 observed in Yale University Museum a specimen of PJioUdopsovata with 

 a cicatrix of attachment, around which point growth is holoperipheral, 

 as in all Keotremata. Specimens of Pholidops are sometimes pre- 

 served with both valves in position and delicately attached to Bryozoa, 



1 Palieontology of New York, Vol. VIII, Part II, 1895, p. 325. 



