8G SYNOPSIS OF AMERICAN FOSSIL BKACHIOPODA. [bull. 87. 



If the percentage of widely distributed species witliiii a siii)erfiimily 

 is a criterion of its vitality, it will be seen that the liyiichonellacea 

 begin in the Ordovician Avith 50 per cent and decline to 23 per cent in 

 the Carboniferous. The JSpiriferacea, also beginning in the Ordovician, 

 have 50 per cent of their species widely distributed, becoming reduced 

 to 20 per cent in the Carboniferous. On the other hand, the Tere- 

 bratulacea were not widely dispersed in the Silurian, whereas in the 

 Devonian their distribution reached nearly 30, increasing to 34 percent 

 in the Carboniferous. Since no statistics of the European Mesozoic 

 and Cenozoic species of this nature are available, the writer can not 

 determine whether or not the Ehynchonellacea continue to decline with 

 such rapidity. It is known, however, that this superfamily has declined 

 considerably in the Cenozoic and late Mesozoic. After the Triassic the 

 Spiriferacea are essentially represented by Spiriferina, yet it too died 

 out with the Jurassic, while the Terebratulacea, which manifested pro- 

 gressively greater vitality during the Paleozoic, are believed to have 

 continued so nearly throughout the Mesozoic into late Cretaceous time. 

 Since then, however, they have also declined. 



In the ontogeny of Dielasma and Zygospira — loop-bearing and spire- 

 bearing genera respectively — Dr. Beecher and the writer have shown 

 that the Terebratulacea may not have been the last superfamily to 

 dev^elop, as was formerly supposed, and that it may have given rise, 

 during early Ordovician times, to the spire-bearing sux)erfamily Spirifer- 

 acea. The Terebratulacea probably originated in the Ehynchonellacea, 

 though no loop-bearing si)ecies are known until the spire-bearing forms 

 are well advanced, or until early in the Devonian system. While some 

 of the largest species of Terebratulacea are found in the Devonian of 

 America and Europe, yet throughout the Paleozoic this superfamily is 

 not a conspicuous one. In the Jurassic and Cretaceous systems of 

 Euroj)e, however, great specific differentiation and abundant indi- 

 vidual development took place. There is but 1 species of this super- 

 family in the American Silurian, while the Devonian has 50 species in 

 15 genera, an increase fifty times greater than that of the Silurian. In 

 the Carboniferous a sharp decline set in, and the superfamily is reduced 

 to 30 species and 8 genera. 



These facts suggest that either the superfamily Terebratulacea did 

 not originate in American seas or — which seems less probable — that 

 diminutive species occur whose interior characters have escaped detec- 

 tion. Further, since the earliest American i)rimitive genera, Kens- 

 seljeria and Trigeria of the Lower Devonian, have very large species, 

 neither these nor Centronella can be the earliest adult representatives 

 of this superfamily. When quite young, Zygospira, also, has a "cen- 

 troiiella-like loop," and it is possible that the primitive Terebratulacea 

 had their origin before the earliest appearance of Zygospira, or during 

 the earliest part of the middle Ordovician era. 



The great majority of telotrematous genera are rostrate in form, but 



