Vol. I. C. C. Yii — Ordovician Cephalopoda of Central China . (ii) 31 



Shell large, somewhat slowly enlarging, with a large endoconch inside. It at- 

 tains a size of 305 mm. preserving a part of the living chamber. Apical extremity 

 unknown. Transverse section circular. Its rate of growth is computed about i in 

 XI. 5. Character of surface not observed. 



The ectosepta are gently concave, their convexity being about two thirds the 

 depth of the camera of the ectoconch or a Httle more. The distance between the 

 ectosepta varies in an irregular manner. It measures 11 mm. at a point about 

 90 mm. from the apex, where the diameter of the ectoconch is 31 mm; and 

 16 mm., where the diameter is 40 mm. When the ectoconch increases its diameter 

 to 50 mm., the ectoseptal distance is suddenly reduced to 13 mm. Nevertheless, the 

 ectoseptal distances continue on the whole to increase upwards until close to the living 

 chamber, where it retains the distance characteristic of the earlier stage (Five camerte 

 are contained in the space of 65 mm. including the last camera). 



The endoconch is very large, marginal and circular in cross section. It is 

 continued upwards in a large endocylinder, which is empty and has a diameter of 12 

 mm, where the ectoconch measures 33 mm. in diameter, this being about one third that 

 of the ectoconch. The organic stereoplasmic lining is clearly seen along the margins of 

 the ectosepta. The camerjB are partly filled up with the crystalline deposit, but the 

 upper-most ones close to the living chamber are entirely empty. Near the apical portion 

 of the preserved part of the shall the final conical endoseptum (endocone) is shown. It 

 has an apical angle of 10''. The rest of the endoconch has been converted into crystalline 

 calcite with much mixing of black material. The invaginated tubi bend inwards, and 

 then slightly outwards. When they reach the geniculations of the preceding ectosepta, 

 they turn again inwards and backward. 



Comparison: Hyatt proposed the generic name Vaginoceras for the form 

 which differs from Endoceras proper in having the longer invaginated tubi and more 

 numerous endosepta^ From the text figure of Endoceras wahlenbergi Foord"'' we find 



1. G. T. Troedsson considered, after the suggestion made by Foerste, the number of 

 endosepta as the only distinguishing character between the Endoceras and Vaginoceras. (See Troeds- 

 son: On the Middle and Upper Ordovician Faunas of northern Greenland, p. 24). Of course, the 

 length of the tubi can not be correlated with the number of the endosepta as Hyatt assumed. But 

 in this paper the length of the tubi is taken as the distinguishing feature rather than the number of 

 the endosepta. The reasons are: (i) The number of endosepta is sometimes unreliable, for they 

 may not be wholly preserved. (2) The spaces between the endosepta are often completely filled 

 with the crystalline calcite and the endosepta are not distinctly enough shown to determine their 

 number. (3) In many forms the endoconch as well as endosepta, limited to the apical portion of the 

 ectoconch, are often not preserved; and the empty, long endocylinder only remains. So far the length 

 of the tubi is always clearly exposed, no matter whether the specimen is well preserved or very 

 framentary. Therefore, the length of the tubi is chosen here as the important factor for distingui- 

 shing these two genera. 



2. loc. cit. 



