26 GENESIS OF THE ARIETTDyE. 



equivalents, however close, have been derived or carried across the genetic lines 

 of descent from an equivalent representative species of one branch to that of 

 another. Nor could the similarities of such forms have been derived in any 

 series from the radical species, because involved whorls, keels, channels, etc. did 

 not exist in the discoidal stock forms. Parallel series and equivalent forms, also, 

 occur often in such zoological and geological relations that any sequence or 

 descent of one from the other is improbable ; as, for example, Aturia of the 

 Tertiary, and Clymeninae of the Devonian ; or Centroceras of the Devonian, 

 or Subclymenia of the Carboniferous, and Agoniatites which began in the 

 Silurian. 



These facts speak with great force for the continuity in descent of the dis- 

 coidal shells, and for the existence of a primitive trunk line of generalized 

 radicals, beginning with the earliest times and lasting into the Jura. The uni- 

 versality of the phenomena leads at first to the supposition that we can account 

 for morphological equivalence of species in different series by some invariable 

 law of growth, such as is evidently the cause of the more exact parallelisms 

 which occur between different individuals of the same species. We might con- 

 sider each species as representing a hereditary grade of structure in the develop- 

 ment of a series, just as any period in the life of an individual would represent a 

 stage of development inherited from some ancestral form. 



We were led into this error at first, but it is an inadmissible supposition in 

 the light of the facts given above. These show, that the representative forms 

 are absolutely new forms in their respective genera or groups, possessing char- 

 acters not found in the stock or chronological trunk of discoidal radicals, and 

 their resemblances are therefore homoplastic, and not homogenous. 



There are also man)'' kinds of series among fossil Cephalopoda, and in some 

 of these forms similar to those of the Ammonoids and Nautiloids are not pro- 

 duced, as in the Sepioids and Belemnoids. In these orders entirely new modi- 

 fications accompanied equally complete changes in habits and habitat. The 

 crawling and shell-covered, littoral, radical Orthoceras has in these orders be- 

 come changed into a swimming and predatory mollusk, the shell having- become 

 internal. It seems evident in these cases, that the forces of the surroundings 

 and new habits deflected the Sepioids and Belemnoids from the more normal 

 course taken by the Nautiloids and Ammonoids, and thus made the repetition of 

 form or equivalence in the shells impossible, except very rarely, and then only in 

 a very limited sense. Such, for example, are the similarities which exist between 

 the internal shell of Spirula and the external shell of Lituites, or between the 

 pseudo shell of the female Argonauta, 1 and the true external shell of one of the 

 compressed Ammonitinas, like Cosmoceras or Hoplites. 



The disappearance of the siphon in the Sepioids, and the naked young of the 

 existing forms of this order, show that too much weight can hardly be given to 

 the modifying and eventually controlling influence of changes of habit, or, what 

 is the same thing, the effects of the surroundings in any new habitat, whether 



1 See Evolution of Cephalopoda, Science, III., No. 52, 53, 1SS1; Foss. Ceph. of Mus. Comp. Zool., Bul- 

 letin, I., No. 1; Proc. Am. Ass. Adv. Sci., XXIII., 1SS3, p. 341. 



