178 



The former of these two genera is a type characteristic of the Dex'onian for- 

 mation ; while the latter is confined to the Silurian and the Devonian formations 

 of North America. Thus, in point of geological age, too, the above conclu- 

 sion seem most probable, or even almost decisive : for at Ning-chiang this 

 fossil was discovered in company with other Brochiopods that represent the 

 Devonian age, nameh', - Itrrpa i iticulcirii var. dcsqiiaiiiata , var. richthofoii 

 and Kf/ yiic/ioiifl/d pa) allcli/^ipcda. 



According to Scuuchert, however, in the famih' Rh\'nchospiridae the 

 shell .structure is " abundantly punctate," while it is impunctate in the 

 family Athyridae. On examining the shell structure of the Chinese fossil 

 it was found to be entirely impunctate. The impunctate structure of the shell 

 constitutesthe sole essential difference between the present fossil and the genera 

 above quoted. In all outward characteristics the new fossil is allied to the 

 Rhynchospirid genera, but in point of shell structure it is an Athx'roid genus. 



Thus the question is transferred to the determination of the relative import- 

 ance of the two structural features, i. e., the shell structure and the presence 

 or absence of radial striae on the shell surface. The meaning of the punctae 

 of the shell is at preseiit Jiot clear. But riiRCJVAi.^' in his stud}' on Ta fbratiila 

 has determined (i) that " tlie shapes of the individual punctae depend to a 

 great extent on the state of preservation " ; (2) that in both the valves " there 

 is in general a progressive increase in density from the umbo outwards '" ; and 

 (3) that " the amount of variat'on i-n a species is so great as to make the den- 

 sity almost valueless as a specific character." In any case, the presence or ab- 

 sence of such punctae seems to be of great importance, for, in certain families 

 they are entireh' absent while they are very abundant in certain others — this 

 is true as far at least as our present knowledge of Brachiopods is concerned. 

 In every textbook much stress is laid upon shell structure as one of the 

 most important characteristics. 



On the contrar}' there are forms either striate, smooth or concentrically 

 costate in one and the same family or even in the same genus. The most 



1) F. G. Percivai,: — On the PLinctuation of the Shells of Terebratula. Geol. Magaz., Feb., 

 igi6. p. 51. 



