GENEALOGY. 65 



The Third Subseries of Coroniceras begins with Cor. latum} This species is 

 remarkable for retaining, until a late stage of growth, the characteristics of the 

 young of rotiforme, and for its exceptional form, the sides of the whorl being 

 exceedingly divergent, the pilae fold-like and heavily tuberculated, the abdomen 

 gibbous and slightly elevated. 



There is one variety of rotiforme with very stout and gibbous whorls in the 

 young, which cannot be distinguished from one variety of latum until a late 

 stage of growth, except by the singleness of the pilae. Single pilae occur, how- 

 ever, in many specimens of all varieties of latum, so that this is not a distinction 

 of constant value. Cor. latum must, therefore, be considered a direct descendant 

 of rotiforme. From this species the transition to Cor. BucklandP is accomplished 

 by numerous intermediate forms. These exchanged the form of latum in the 

 young for a sinemuriense-like stage, in which the abdomen became contracted 

 in breadth, the sides parallel, and the channels deep. This stage was retained 

 in some specimens for a long time, while in others it quickly gave place to 

 the huge pilae, parallel but gibbous sides, transversely broad whorls, and flattened 

 abdomen of the adults of the typical Bucldandi. 



In other forms, with similar young, the pilae assumed during their adult stage 

 the usual aspect, with either only a trace of tubercles, or none. I have not been 

 able to follow the transitions of this, or the sinemuriense variety, into the stout form 

 of Bucldandi. There can be but little doubt, however, that the large form found 

 at Lyme Regis differs only in having had a more accelerated development; i. e. in 

 skipping the double pila? and large tubercles of the sinemuriense stage. It evi- 

 dently acquired, at a very early period in the young, large untuberculated pilse, 

 and in old age was characterized by a very decided narrowing and rounding off 

 of the abdomen, obsolescence and bending forward of the pike, disappearance of 

 the channels, and a broader and less elevated keel. 



The evidence of transition from the sinemuriense variety to Cor. orbiculatum 

 rests upon similar grounds. The singleness and perfection of the pilae are the 

 only differences which separate the young of orbiculatum from the young of such 

 forms as variety No. 5 of sinemuriense. In the adults, however, the narrowness 

 of the abdomen, flatness of the sides, and their convergence outwardly, are 

 marked differences in aspect, which were greatly increased by advancing age. 

 The abdomen in some very large specimens became almost obtusely angular, as 

 in Vermiceras, and the pilae fold-like, much bent forwards, and the channels 

 obsolete. 



Agassiceran Series. 



This series obliges us to return once more to Psiloceras planorbe. It has two 

 subseries. 



First Subseries. — The young of Ac/as. Icevigatum 3 had a close resemblance to 

 the young of the compressed varieties of Cor. Tcridion, and to Cor. rotiforme in some 

 varieties, before the latter acquired tuberculated pilae. But this likeness was 



1 PI. iii. fig. 19-23; Summ. PI. sii. fig. 16. = PL iii. fig. 18; Summ. PI. xii fig. 17. 



3 PI. viii. fig. 9-14; Summ. PI. xiii. fig. 1. 



