426 NATURAL SCIENCE. dec. 



Figures 6, 7, and 8. This specimen shows no definite protoconch 

 and hardly a trace of constriction. There is an abrasion of the surface, 

 as shown in Fig. 7, but the shrivelhng is greatly exaggerated. The 

 outline of Fig. 6 is correct, and shows a sHght swelling a little distance 

 from the apex. 



In short, we see no essential differences between these specimens 

 and that represented in Figures 18 and ig,in which the protoconch is 

 said by Hyatt to be removed and the opening to be plugged by 

 secondary calcite, and which is actually figured for the sake of contrast 

 with the specimens said to retain the protoconch. In making these 

 criticisms, which are based on independent observation, we have the 

 advantage of following the opinion of the great palaeontologist, 

 Barrande. 



It does not follow from the preceding remarks that the species of 

 Ovthoceras in question (assuming that they do belong to that genus) 

 were unprovided with protoconchs at some period of their lives ; nor 

 does it follow that the little lump seen at the ends of the specimens 

 in the British Museum may not be some representative of a proto- 

 conch. It may very well be the calcareous infilling of a thin- walled 

 protoconch ; but there is no reason for regarding it as the shrivelled 

 remains of a conchiolinous shell. 



If therefore the general conclusions of Hyatt were to be accepted, 

 it became desirable that some concrete instances should be found, on 

 the one hand, of undoubted Ammonoidea that should be straight like 

 an Ovthoceras, and on the other hand of undoubted Nautiloidea that 

 should possess a definite protoconch. When such a form was found, 

 it would obviously be a matter of some difficulty to decide whether it 

 belonged to the Ammonoid or Nautiloid branch ; and if it really were 

 the case that no criteria could be discovered, then such a form might 

 claim to be a true connecting link. Still, there are certain minute 

 features which are generally held, in a rough way, to distinguish 

 these two orders ; such are the position of the siphuncular passage 

 through the septa, the nature of the external ornament, the contour of 

 the septum, usually called the suture-line, and the proportions of 

 length and breadth in the conch itself. These are characters well 

 known to all who have ever worked on Cephalopoda, they are the 

 ready means by which we tell an ammonite from a nautilus, and, as 

 they are given with more or less accuracy in all text-books of 

 palaeontology, we need not further consider them. Enough to say 

 that any straight, protoconch-bearing cephalopod shell can, by their 

 means, be referred with a high degree of probability to its proper 

 order. 



The first step in advance was made by Branco, who showed that 

 the straight BactriUs, from the Devonian rocks of Germany, possessed 

 a well defined protoconch of ovoid shape (5). The affinities of 

 Badrites had before this been a subject of some dispute : while some 

 placed it near the goniatites, others regarded it as merely a close ally 

 of Ovthoceras, differentiated therefrom only by the marginal position of 

 the siphuncle. It was natural that the discovery of a protoconch 



