38 



PEOF. H. H. SWIIS^NEETOX AND MR. A. E. TEUEMAIf [vol. Ixxiii, 



to Branco,! the siphon lies near the internal margin of the early 

 septum, as in most latisellate ammonoids ; while in certain forms, 

 as, for example, Goniiitites lamed var. calcuIifoDnis, it occupies 

 a ventral position from the first. Despite these variations in 

 jjosition of the siphuncle, the development of the septum pursues a 

 definite course. Assuming that there is no actual migration of the 

 si])lmneular cord through the substance of the mantle, it may be 

 taken as furnishing a datum-point. It then seems as though a 

 portion of the mantle ventral to this point is withdrawn from 

 taking part in the seci'etion of the septum, and that the mantle 

 dorsal to it increases, especiall}'- on the umbilical Avings. Through- 

 out development, and also despite all the distortions associated 

 with asvmmetrv (PI. IV), the siphuncular cord and ventral lobe 

 remain intimately related. 



The close similarity in form between the early septum and 

 the central part of the adult septum suggests that the study of 

 the latter may give some clue to the development. In actual 

 practice the contouring of a cleaned septum was found to be un- 

 satisfactory for this purpose, inasmuch as the instrument could not 

 be made to deal efficiently with undercut or even vertical portions 

 of the surface. On the other hand, septal sections proved highly 

 satisfactoi'v. Before considering these, it will be necessary to 

 provide a* basis for comparison by giving an account of the suture- 

 line of Dactylloceras and of its development. 



VIII. The Adult Sutuee of the Dactyloid^. • 



Before proceeding to the detailed description of the adult suture 

 in the specimen from which both developmental details and septal 

 sections were obtained, it will be helpful to refer brielly to the 

 general features of the suture-lines in the family Dactyloidse and 

 in the Cceloceran forms from which it has been derived. Unfor- 

 tunately, the material at our disposal was neither abundant enough 

 nor sufficiently complete to permit the elucidation of major and 

 minor lines of descent Avithin these families. Nevertheless, there 

 Avas sufficient to lead to the detection of certain tendencies which 

 throAv much light upon the problem now being considered. 



The material examined included the following species : — 



Dactylloceras cotmnune (Sowerby). 

 B. crassum (Young & Bird). 

 D. annulatum (Sowerby). 

 D. turriculatum (Simpson). 



Peronoceras fibxilaUim (J. de C. Sowerby). 

 Porpoceras andrsei (Simpson). 

 Coeloceras pettos (Quenstedt). 

 C. fonticulus (Simpson). 



The general relationships of some of the variations of the 

 suture-lines within this family may be illustrated from one 

 sj)ecimen of Dacfj/lioceras by a comparison of its last suture with 

 that of a slightly' earlier one, namely, the tenth from the last 

 (fig. 8, ii & iii). 



During the interval between the deposition of these two septa 



Op. cit. p. 02. 



