Vll 



Approximate phytogeny of the families of Cephalaspidea. 



The group of families on the left side are the most primitive of 

 recent Tectibranchiata ; the median and right hand groups being 

 far more specialized, and more remote from the Notaspidea and 

 Anaspidea. 



A phylogenetic table of the shell-bearing Opisthobranchs has been 

 given by M. Cossmann in a work of great merit, "Essais de Paleo- 

 conchyliologie Comparee " (1895), derived mainly from his studies of 

 the fossil forms. The great discrepancy between the results obtained 

 by the distinguished French author and myself, are in part trace- 

 able to the widely different material studied, and in part to the fact 

 that Tectibranchs, like Pulmonates, are singularly non-committal in 

 the characters of the shell. In fact, I feel that it is not extreme to 

 state that the shells alone, in either group, are totally inadequate to 

 express the affinities of families and genera. In so many groups are 

 the shells more or less degenerate, so many are the cases of parallel 

 or converging development of the shells, that conclusions based upon 

 them alone, without a knowledge of the soft anatomy for a primary 

 guide, are practically valueless for the appreciation of the affinities 

 of genera and families, either in Tectibranchiata or Pulmonata. 

 There can be no doubt, however, that palaeontology will prove of 

 great value in supplementing the evidence of comparative anatomy; 

 and the best results can only be obtained by a union of the two 

 methods. 



