348 PALEONTOLOGY OF NEW YORK. 



must long be a matter of legitimate speculation, and in view of this fact a few 

 arguments of such a nature in this place will be permissible. The living 

 representatives of Rhynchonella and Terebratula are animals in which a very 

 considerable part of the brachia does not become sufficiently spiculized to form 

 a continuous calcareous support. In R. {Hemithyris) psittacea, for example, the 

 brachia are as highly developed in the form of coiled spiral arms as they could 

 have been in most of the ancient spire-bearers, but their calcareous supports 

 are only the short lamella) known as the crural processes. All of the living 

 Anctlobbachia which possess a long curved loop like that of Cryptonella and 

 DiELASMA of the Palaeozoic, have an unsupported median unpaired spiral arm, 

 coiled in a direction which is the reverse of that prevailing among the spire- 

 bearers If, now, we are to interpret the condition of the brachia in their 

 nearest living representative, it becomes necessary to assume that on the one 

 hand, the palaeozoic rhynchonellids possessed long colled spiral arms, and, on 

 the other, that Dielasma and its palaeozoic allies and affines, when mature, were 

 provided with the unpaired coiled arm of Terebratella. This assumption, in 

 the first place, totally destroys the inference above made as to the primitive 

 relation of the rhynchonellids to the terebratuloids and spire- bearers ; and, 

 secondly, would seem to necessitate a novel and unexpected interpretation of 

 the brachial structure in all the spire-bearers. If Dielasma possessed the median 

 arm, supported at its base by the transverse band of the loop, which corre- 

 sponds to the jugum of the spire-bearers, then in the DiELASMA-stage of Zvao- 

 SPIRA and other spiriferous shells, where this stage was well defined, there must 

 also have been a median coiled arm of some extent. This median arjn, in 

 living forms, is due, as shown by Beecher, to the necessity of finding room for 

 the cilia or tentacles multiplying at the extremities of the brachia. The mere 

 presence of the transverse band in Dielasma and the DiELASMA-stage of Zyoo- 

 8PIRA, implies a similar extension of the brachia, and from the analogy, a median 

 arm. The subsequent growth of the brachia in Zygospira, carrying the calca- 

 reous ribbon forward, beyond the bases of the loop and into lateral spiral cones, 

 would not of itself afibrd sufficient reason for assuming that the growth of the 

 brachia at their extremities, which produced the median arm, was necessarily 



