UPPER TRIASSIC SPECIES. Ill 



cJiyceras. The development generally of a keel, or, in some varieties, of a 

 raised abdomen, over which the pilse do not pass, shows that this is a differ- 

 ent genus, characterized by a different mode of development. The septa 

 are qnite similar to those of Tracliyceras ; but it is very evident that in the 

 Tracliycemtidce the septa cannot be looked to for generic differences. Great 

 differences also occur in the amount of involution of the different species, 

 and in the development of their external characters. This is shown by con- 

 trasting the species Blakei, Tracliyceras hrevidorsatum, or T. BrotJieus, with the 

 type of the genus T. aon. 



"The forms and characteristics of the young in these three species could 

 hardly be more different, and yet their septa are very similar. Possibly a 

 closer study of the lobes will bring out corresponding differences; but at 

 present it is safer to rely upon the development of external features in this 

 family."— (A. H.) 



Gymnotoceras rotellifokme. Meek. 



Plate 10, figs. 9 and 9 a. 



Shell discoid-lenticular, with periphery subangular, or very narrowly 

 rounded; convexity only about one-fourth the greatest breadth; umbilicus 

 very small, or scarcely more' than two-fifths the breadth of the outer volu- 

 tion, with its nearly vertical walls meeting the lateral surface of the volution 

 so as to form a subangular margin; whorls laterally compressed, with greatest 

 convexity within the middle of the sides, thence converging outward with 

 gentle convexity toward the periphery, all increasing gradually in convexity 

 and more rapidly in breadth; each inner turn almost completely embraced 

 in a profound sinus or concavity of the inner side of the succeeding larger 

 one. Surface in the young ornamented with small, slightly flexuous costae 

 scarcely distinct from the lines of growth, but becoming apparently most 

 defined in young shells about one inch in diameter, after which further in- 

 crease in size rendered them very obscure, broader, and more distant, until 

 they gradually died out, leaving the sides smooth, or nearly so, in a speci- 

 men two inches broad; costse and lines of growth curving strongly forward 

 as they approach the periphery, which the former do not cross or reach. 

 Septa with fom* lateral lobes on each side, decreasing rapidly in size from 

 the largest or first one (which is oblong, and, like the second and third, 



