NOMENCLATURE OF STAGES OF GROWTH AND DECLINE. 13 



functional siphon in these shells. Gerard Holm 1 first called attention to the 

 interesting character of the young stages of the siphon in Endoceras, and has 

 shown this organ to have been very large even in the young, having not only 

 a csecal beginning, as in other forms, but in several species having a swollen 

 or macrosiphonulate form which endured throughout several septa. In speci- 

 mens now in the Museum of Comparative Zoology, at least six septa were built 

 before any signs of contraction began to appear. In other cases figured by Holm 

 the siphonal caecum, though very large as compared with that of Orthoceras, was 

 attached to the first septum, as in all the shells so far known from that group, 

 and occupied only the first air-chamber. We should suggest to those having 

 materials for study, that the shells having this last character are very likely not 

 true Endoceratites, but perhaps the young of species of the genus Sannionites, 

 which, according to the classification followed by the author, is a genns distinct 

 from Endoceras, because the species possess a much slenderer siphon. Whatever 

 the fate of this suggestion, it is plain that transitional series exist in this group 

 between Sannionites and Cyrtocerina or Piloceras, and that gradations occurred 

 also in Piloceras, which show that contraction of the siphon began first in adnlts, 

 and then, according to the law of acceleration, was inherited in the nealogic 

 stages of immediate descendants, and finally became naapionic in the smaller 

 siphoned species of the genera Endoceras and Sannionites. This tendency to 

 contraction in the diameter of the siphon indicated the beginning of a series of 

 transformations which accompanied a decrease in size of the fleshy siphon, and 

 other correlative transformations, such as the decrease in length of the funnels, and 

 the contraction and straightening out of the calcareous endocones, so as to form 

 the walls of a tubular siphon. In other words, as the siphon contracted, the func- 

 tional endosiphon formed by the open and extended tips of the endocones was 

 finally brought into line with the funnels, and together with them formed the 

 microsiphon, which is consequently a degraded modification derived from the 

 funnels, endosiphon, and endocones of the Endoceratidse. The Orthoceratidae and 

 all the remaining forms, with some notable exceptions which we shall take up 

 and describe in future papers, had a microsiphon. The whole microsiphon formed 

 a continuous open tube of narrow diameter, reaching from the last septum to the 

 nsepionic septa, which represented the macrosiphonula. Doubtless the duration 

 of naspionic stages will be found to vary somewhat in ancient forms, but the 

 indications, so far as known, are in favor of the theory that the vast majority of 

 even ancient forms had a microsiphon, which was developed comparatively early 

 in the life of the animal. 



The nealogic stages of succeeding groups would be very interesting if 

 there were space to describe them, but we shall have to illustrate this part of 

 our work among the Ammonitinas. The protoconchs of Ammonoidea, including 

 the genus Bactrites, had, as remarked above, globose forms with calcareous 

 shells, and these shells were continuous with the apex of the conch, but the 

 aspect of the junctions was quite distinct from those of Nautiloids. The con- 

 striction between them and the apex was very slight in the uncoiled young of the 



1 Dames et Kayser, Paleontol. Abhandl., III., Part I. 



