Clarke — The Naples Fauna, 



127 



metrical, attached to the conch above with a rather broad and gradual con- 

 striction. We have no clue to its actual size, the author having given no 

 dimensions, but the figure indicates that with relation to the size of 

 the conch it is considerably larger than the protoconch in B. 

 gracilior. This specimen, further, shows little evidence of the 

 nepionic expansion so well marked in B. gracilior and B. aciculum. 



With regard to the shells of Bad rites gracilior we know not 

 only their protoconch and early stages but their mature characters 

 and life history. There can be no doubt that these represent a 

 typical example of the genus Bactrites, as described by Saxd- 

 berger, and a species very closely allied to his own B. gracilis. Figure 102 



.., , „ , A reduced 



1 he apparent discrepancies 111 these protoconchs referred to Bac- copy of 



1 ICl 1 11 !>1 Branco's 



trites may 111 some measure be accounted for by our knowledge of the figure, 

 goniatites. We have observed that a difference in the size of the protoconch 

 is possible within the limits of a given genus ; though this is not great in any 

 progressed genus such as Maxticocekas, nor is it accompanied by any noticable 

 variation in form. In the genus Tokxoceeas, notable for its long duration, wide 

 diffusion and abundant development through the Devonian, a genus which is 

 simple in its septal development but shows evidence of acceleration in some 

 other respects, there is a marked difference in the size of the protoconch in 

 examples of the early Torn, urvingulare and the late 'Torn, hicostctam. 



We have shown in the foregoing and following pages the possibilities of 

 variation in actual size of the protoconchs in some of the goniatitine genera 

 and it will be observed that the protoconchs in the early and simple genera 

 Agoniatites and Axarcestes are of colossal size when compared with such 

 later and progressed genera as Manticoceras, Torxoceras aud Probeloceras. 

 At the same time the progressed genus Saxdbergeroceras also possesses so 

 large a protoconch that we dare not assume unusual size in this body to be an 

 unfailing index of simplicity in the species, though it be undoubtedly true in 

 general, and this fact of the presence of larger protoconchs in earlier and simpler 

 species of goniatites may afford an explanation of the conspicuous difference 

 in size in the Wissenbach protoconch of Bactrites and that of the later B. 

 gracilior. 



To account for the apparent difference in form, no explanation suggests 

 itself. Our figures of B. gracilior show some variations in the dimensions of 

 this structure which are not readily explained unless they be ascribed to the 

 action of resorption, but none of them presents an ovoid form like that of the 

 Wissenbach shell. 



