52 GENESIS OF THE AKIETID.E. 



or ephebolic stages of a Nautiloid, than in the young of the same individual. 

 The ancient primitive radicals like Piloceras and Endoceras had huge siphons, and 

 were macrosiphonulate, whereas in all the remaining forms it was quite small or 

 microsiphonulate. The siphon was also far less important and smaller among the 

 Ammonoidea as a whole than among Nantiloidea, since there were no macrosi- 

 phonulate forms in this order. It was also, as has been stated above, much larger 

 and more important in the typembryos, and in the earlier nsepionic stage among 

 Ammonoids than subsequently in the growth of the same individuals. 



This organ was also far less important, and smaller and less perfectly formed, 

 among the Belemnoids than among Ammonoids, and finally among the Sepioids 

 it became reduced to a mere rudiment, being distinguished with difficulty in the 

 internal shell. 



We can therefore say, with some confidence, that the siphon became reduced 

 in size and importance during the progressive period of evolution or epacme 

 of each group of the Cephalopoda, and also followed a parallel course during 

 the development of the individual. When one reviews the various positions, 

 decreasing size, and lessening importance of the siphon in the various groups 

 of Cephalopoda, he becomes aware that these characteristics correlate with each 

 other, but they do not seem to be dependent upon any other character. They 

 are, however, correlative with higher specialization. Thus, in Nautilus, the posi- 

 tion of the siphon is variable ; in Ammonoids, a more highly specialized type, it 

 becomes more invariable, and always ventral in nealogic stages ; in IJelemnites 

 which remain straight, and in those more or less coiled, like Spirilla, the siphon 

 is also constant, but on the ventral side of the shell. It seems, therefore, that 

 fixity of position is not dependent upon close coiling, but is purely a condition 

 of specialization, and is an accompaniment of the decrease in size and impor- 

 tance of the organ. The fixity and preservation of this differential character, 

 one of the most important in distinguishing the Ammonoids, could not have 

 been due to natural selection, since an organ invariably tending to become of 

 less importance in every order could neither have been advantageous, nor have 

 offered a favorable lever for this law to work with. 



Specialization has in all cases appeared to us to be due, not to natural selection, 

 but to physical selection, or the production of suitable modifications by the action 

 of forces which changed in a similar way large numbers of the same species, 

 perhaps nearly all the individuals in the same locality or same habitat, within 

 a comparatively limited period of time. 1 



1 See in this connection the interesting researches of Dr. A. S. Packard, in the Memoirs of the National 

 Academy of Sciences (IV., Ft. I., p. 137 el seq.), in which this untiring investigator deduces similar conclu- 

 sions \\ ith regard to the action of physical surroundings upon cave animals. Up also repeats after these new 

 and profound studies the assertion made in former publications, that natural selection does not appear to him 

 to be a cause of modification or of the preservation of variations, but the result of the action of other factors. 



Dr. Packard's own words are as follows: " Such a phrase as ■ natural selection,' we repeat, does not to 

 our mind definitely bring before us the actual working causes of the evolution of these cave organisms, aid 

 no one cause can apparently account for the result." The causes are "change in the environment." " dis- 

 use of certain organs," "adaptation," "isolation," and "heredity operating to secure for the future the 

 permanence of the newly originated forms as long as the physical conditions remain the same." 



" Natural selection, perhaps, expresses the total result of the working of these five factors, rather than 



