1926 SYSTEMATIC 21 



June 



there is, in most cases, quick return to venter rounded. In order of 

 increasing persistence of runcination the genera may be placed as follows : 

 Cerericeras, Galilaeiceras, Gowericeras, Galilaeites, Galilaeamis, Kepplerites, 

 Catasigaloceras, Sigaloceras. In Cerericeras the runcination is feeble and 

 quite transitory ; in Sigaloceras the runcination is marked and persistent. 

 A broken example of Sigaloceras, S.B. Coll. 4122, shows, in the 

 interior, the impression of the periphery of a whorl about 7 mm. in 

 diameter : this has the Parkinsonian, or rather Bigotitean venter (T.A. 

 m. 29, 53) — ribs not quite opposite, with a median interruption. Later 

 the venter becomes rounded, with crossing ribs ; later still, feebly 

 runcinate, with the runcination becoming more and more definite, while 

 the ribs pass over with a slight sag. At about 75 mm. diameter there 

 are signs of rib-failure on the distinctly runcinate venter — the sign of 

 an oncoming smooth stage. 



Two genera, Toricelliceras and Toricellites, have to be placed 

 provisionally with those mentioned above, yet they seem almost to be 

 more like forerunners of Kosmoceratidse — at any rate, Toricellites does. 

 Both differ from Gowericeratidae in having attained more than runcinate 

 venter. Both, however, resemble Gowericeratidje in rapid decline to 

 rounded venter — differing in this respect from Kosmoceratidas. 



The point now comes for consideration — do these two genera 

 represent, in the attainment of certain characters, stages through which 

 the other Gowericeratids have passed, but have in the main lost by 

 ' skipping,' or do they represent stages to which the other Gowericeratids 

 tried to attain, but in the main failed to achieve, and so dropped back 

 again to lower stages ? — some of them making feeble efforts to remain, 

 for instance, runcinate, others making a prolonged effort in runcination, 

 presumably- on the road to subsulcation. In this view Cerericeras has 

 only very transitory attainment of runcination, while Sigaloceras shows 

 a very prolonged stage, but has no sign of attaining subsulcation. 



Gulielmiceratidae : The order of development appears to be as 

 follows : — Runcinate venter, Gulielmina ; subsulcate venter with border- 

 ing tubercles gaining in strength as the ventral ribs fail, Gulielmiceras, 

 Gttlieltnites, Anakosmokeras : these three genera end by becoming smooth, 

 except for retaining, or perhaps rather, it might be said, re-developing, 

 inner marginal nodes. A smooth, or nearly smooth, venter and well- 

 develojjed lateral bituberculation are other features. 



There is one peculiarity in the date of these genera : Gulielmina 

 and Gulielmiceras belong to the middle ProplanuUtan Age : then there 

 is a time, the Reineckeian Age, when no Kosmocerataceae seem to be 

 found ; later, in early Kosmoceratan, are two genera remarkably like 

 Gulielmiceras, namely Gulielmites and Anakosmokeras. 



Parapatoceratidae : There are, as yet, only two genera — criocone 

 with curved whorls just out of contact, Crioconites — and toxocone, with 

 whorls curved, just short of straight, Parapatoceras. 



There seems no reason to suppose that forms of this family are 

 " an independent development perhaps of the later family Reineckeidae " 

 (Spath, Blake Coll. ; Pal. Ind. ix (i), 1924, 12) ; but, rather, that the 

 origin is to be sought in penecontemporaneous forms allied to Toricellites. 

 In Toricellites the cadicone coronate stage of the inner whorls is but 

 feebly involute. Economic disturbance to such a form might produce 

 rapid acceleration in two ways — in shape, from serpenticone to criocone, 

 in ornament, from tuberculate to costate, that is, costate post-tuberculate. 



