160 M. Agassiz on the Relations between Animals' 



rect experiment upon Polypi, Medusae and Echinoderms, some of 

 which are struck with almost instantaneous death when brought 

 into fresh water, and decompose with astonishing rapidity. I have 

 seen, on dropping an Ophiura into fresh water, all the articulations 

 dismembered and entirely separated within a few minutes. 



No one of the three other great types of the animal kingdom 

 is either so exclusively marine, or even so exclusively aquatic as 

 that of Radiata ; for among Mollusca we have quite a number of 

 terrestrial genera, and even a large number of freshwater genera 

 and families. 



Among Articulata we notice also large numbers of freshwater 

 species, and a still larger number of terrestrial forms. Finally, 

 among Vertebrata we find the most promiscuous occurrence of 

 marine, freshwater and terrestrial forms. It is now important 

 to ascertain whether we may trace, beyond the Radiata, a direct 

 relation between structure and the element in which animals live, 

 and whether the gradation of this structure has any reference to 

 the surrounding media, as it unquestionably has among the Ra- 

 diata. 



Let us first consider the Mollusca, and perhaps revise their 

 classes in a zoological point of view before undertaking the in- 

 vestigation of their relations to the media in which they dwell, 

 allowing in this revision a "due influence to embryology as far as 

 it can influence this question at present. 



The number of classes which should be admitted among Mol- 

 lusca is the first point of importance we have to consider. Since 

 the Barnacles or Cirripedia, which Cuvier still considered as a 

 class among Mollusca, are now known to belong to the type of 

 Articulata, and to be most conveniently combined with Crusta- 

 cea, we have five classes of Mollusca left, if we follow Cuvier's 

 arrangement of these animals, as he distinguishes Cephalopoda, 

 Pteropoda, Gasteropoda, Acephala and Brachiopoda, as so many 

 distinct classes of the type of Mollusca in the order of gradation 

 just mentioned. It will hardly be necessary at present to insist 

 upon the close relation which exists between Brachiopoda and the 

 other bivalve shells. Indeed, anatomical investigations of these 

 animals have shown that they are not only constructed upon the 

 same plan, but that the differences between Brachiopoda and or- 

 dinary Acephala are scarcely as great as the differences which 

 exist between Ascidia and Lamellibranchiate Acephala, which 

 Cuvier nevertheless placed in one and the same class. We shall 

 therefore consider Tunicata, Brachiopoda and Diphyra, as one 

 great natural class under the name of Acephala, to which we also 

 refer, as mentioned above, the type of Bryozoa, which has been 

 so long combined A^dth Polypi. As to the Pteropoda and Gaste- 

 ropoda, though they are still generally considered as two classes. 



