Smith.] ^5^ LOct. 2, 



plioiial lol)e, and a pair of pointed lateral lobes. The saddles, both 

 external and internal are usually rounded, although even thej' may be- 

 come angular, as in old specimens of GlypMoceras sphcericum Martin. 

 The surface in most of the older members of the group is ornamented only 

 with strife, but in many, especially the later members, umbilical ribs are 

 developed, which in Pericyclus cross the abdomen. Periodic constric- 

 tions, or varices, representing temporary cessations of growth, are 

 found on most of the genera. 



Hyatt * says that the GlypJiioceratidai are derived directly from the 

 group Magnosellaridm, as represented by Pavodkeras of the Devonian. 

 And, in fact, the development of Glyphoceras diadema Goldfuss, as 

 worked out by Branco.f shows at 2.25 millimetres diameter a decided 

 resemblance to the adult sutures of Tornoceras. The younger larval 

 sutures of this form show derivation from a radicle like Anarcestes. 

 PI. xix. Fig. 5, shows the development of Tornoceras (Parodiceras) re- 

 trorsum Buch, after Branco, in PalmontograpMca, Vol. xxvii, PI. v. Fig. 

 7. We thus have probably the complete genealogy of the Glyphiocera- 

 tidce in the larval stages of the two genera, GlypMoceras and Torno- 

 ceras. PI. xix. Fig. 4, shows the development of GlypMoceras diadema 

 Goldfuss, after Branco, in PaliBontogrupMea, Vol. xxvii, PI. iv. Fig. 1. 



Oenus Gastrioceras, Hyatt. 



This genus was originally established by Hyatt (Proc. Boston Soc, Nat. 

 Hist., Vol. xxii, 1883, p. 327) to include evolute species with open um- 

 bilicus, trapezoidal or semilunular cross-section, and usually ribs or tuber- 

 cles on the sides ; the species included by Hyatt in this genus all have 

 prominent siphonal saddles, first lateral saddle broadly rounded, second 

 lateral saddle broad but inclined to be pointed ; the siphonal lobes are 

 long, narrow and pointed, and the lateral lobes broad and pointed. In 

 all the species cited by Hyatt (loc. cit.) as belonging to Gastrioceras 

 there is but a single pair of lateral lobes visible, that is, on the sides of 

 the shell ; and in the Second Annual Report Geol. Survey of Texas, 1891, 

 p. 355, Hyatt limits Gastrioceras to forms with a single pair of lateral 

 lobes and with the second pair on the umbilical shoulders. Hyatt (loc. 

 cit.) refers G. russiense Zwetajew to liis genus Paralegoceras, because 

 that species has the second pair of lateral lobes on the sides of the shell 

 and not on the umbilical shoulders. But Gastrioceras russiense has just 

 the same number of lobes as all other known species of Gastrioceras, 

 namely nine in all, and lacks the lobe on the umbilical border, which is 

 characteristic of Paralegoceras. Another species, Gastrioceras baylor- 

 ense White (Bull. 77, U. S. Geol Survey, p. 19, PI. ii, Figs. 1-3), also 

 has two pairs of lateral lobes. White's figures and description do not 

 show whether the umbilical lobe is present or not ; if it is, G. baylorense 

 rightfully belongs with Paralegoceras, but it most probablj^ belongs in 



*Proc.. Boston Soc. Nat. Hist., Vol. xxii, 1883, p. 322. 

 t Palxontographica, Vol. xxvii, PI. iv, Fig. 1. 



