ORIGIN OF DIFFERENTIALS. 49 



explanation of the observed phenomena ; but that the characters, as in Weiss- 

 mann's theory, singly or in part could have affected the ovum when they first 

 appeared, does not seem to be sustained by the facts. The examples cited above 

 of the transmission of the characteristics of the asiphonula and csecosiphonula 

 to the typembryo of the Ammonoidea give similar evidence with regard to the 

 origin of embryonic characters, and are directly against Weissmann's position. 



Barrande and Munier-Chalmas have tried unsuccessfully to prove the abso- 

 lute invariability of differentials among fossil Cephalopoda by means of the great 

 apparent differences between the embryos of Nautiloids and Ammonoids, but 

 the discovery of a protoconch in Orthoceratidse has demonstrated their error, 1 

 and we confidently anticipate the discovery of some form in which the proto- 

 conch will exhibit intermediate characters. 



The rostrum in the Ammonoids contrasts decidedly with the central sinus of 

 the same region of the aperture in Nautiloidea, and is a differential of importance, 

 which ought to be mentioned here. The rostrum of the mesozoic forms indicates 

 that the Ammonoids did not possess the ambulatory pipe or hyponome 2 of Nau- 

 tiloids, which causes the ventral sinus in the aperture and striae of growth in that 

 order. As noted above, they could not have been swimmers in the same sense as 

 the existing Nautilus, and they must have been for the most part strictly littoral 

 crawlers. The habit of crawling as a slower mode of progression combined with 

 varied habitats of shallower waters, may have been the cause of the higher spe- 

 cialization and greater variety of forms and structures which they exhibit. The 

 change from a ventral sinus to a rostrum in the aperture began among the higher 

 Goniatites during paleozoic time, and is shown by the narrower ventral sinus 

 of the aperture, and in the lines of growth on the shell. This sinus also is more 

 distinctly marked, and is present oftener, in the devonian forms of Goniatitinae 

 than in the carboniferous species. 



The Clymeninae of the Devonian have very small ventral sinuses in many 

 forms, and in others the hyponome may have been absent. The specimens ob- 

 served by the author have not as a rule exhibited the lines of growth with great 

 clearness, but many of Gumbel's figures give the striations, and in some species 

 they pass straight across the venter. The Ammonoids of the Trias are appar- 

 ently completely transformed, the rostrum and the ventral saddle in the lines 

 of growth indicating the constant absence of a hyponome. 



As the ambulatory hyponome disappeared, all the sutures became more 

 complicated or ammonoidal, and, in correlation with the greater lengthening 

 of ventral and dorsal lobes, the central zone of the septum changed from con- 

 cave to convex. When one tries to attribute the origin of convexity in the 

 septa, or the multiplication and lengthening of lobes, or the marginal digita- 

 tions of the sutures among ammonoids to fortuitous variations, he finds at once 

 that their history in the groups of the order is correlative with the phenomena 

 of morphological equivalence. As was long ago observed by Von Buch, and 



1 Science, III., No. 52, 1884, p. 126; and also above pp. 10, 11. 



2 See Foss. Ceph. Mus. Comp. Zool., Proc. Am. Ass. Adv. Science, XXXII., 1883, p. 340; also Science, 

 III., Xo 52, 1S84, p. 123; and above, p. 29. 



7 



