BRITISH ASSOCIATION : ZOOLOGICAL SECTION. 395 



" Wings " of Pterosaurs and Bats. 



Long flexible bill of Apteryx and Snipes. 



Proteroglyph dentition of Cobras and Solenoglyph dentition of Vipers. 



Loss of the shell of Limax and Aplysia. 



Complex molar pattern of Horse and Cow. 

 Parately : Bivalve shell of Brachiopods and Lamellibranchs. 



Stretcher- sesamoid bone of Pterodactyls (radial carpal) ; of Plying 

 Squirrels (on pisiform) ; of Anomalurus (on olecranon). 



Bulla auris of Pythonomorph (quadrate) and Whale (tympanic) ; 

 is incus = quadrate. 



" Wings" of Pterosaurs, or Bats, and Birds. 

 The distinction between these three categories must be vague 

 because that between homology and analogy is also arbitrary, depend- 

 ing upon the standpoint of comparison. As lateral outgrowths of 

 vertebrae all ribs are homogenes, but if there are at least haemal and 

 pleural ribs then those organs are not homologous even within the 

 class of fishes. If we trace a common origin far enough back we 

 arrive near bedrock with the germinal layers. So there are specific, 

 generic, ordinal, &c, homoplasies. The potentiality of resemblance 

 increases with the kinship of the material. 



Bateson, in his study of Homoeosis, has rightly made the solemn 

 quotation : " There is the flesh of fishes . . birds . . . beasts, &c." 

 Their flesh will not and cannot react in exactly the same way under 

 otherwise precisely the same conditions, since each kind of flesh is 

 already biased, encumbered by inheritances. If a certain resemblance 

 between a reptile and mammal dates from Permian times, it may be 

 homogenous, like the pentadactyle limb which as such has persisted : 

 but if that resemblance has first appeared in the Cretaceous period it 

 is Homoplastic, because it was brought about long after the class 

 division. To cases within the same order we give the benefit of the 

 doubt more readily than if the resemblance concerned members of 

 two orders, and between the phyla we rightly seek no connection. 

 However, so strongly is our mode of thinking influenced by the prin- 

 ciple of descent that, if the same feature happen to crop up in more 

 than two orders, we are biased against Homoplasy. 



The readiness with which certain Homoplasies appear in related 

 groups seems to be responsible for the confounding of the poten- 

 tiality of convergent adaptation with a latent disposition, as if such 

 cases of Homoplasy were a kind of temporarily deferred repetition, 

 i. e. after all due to inheritance. This view instances certain recurring 

 tooth patterns, which, developing in the embryonic teeth, are said 

 not to be due to active adaptation or acquisition but to selection of 

 accomplished variations, because it is held inconceivable that use, 

 food, &c, should act upon a finished tooth. It is not so very difficult 

 to approach the solution of this apparently contradictory problem. 

 Teeth, like feathers, can be influenced long before they are ready by 

 the life experiences of their predecessors. A very potent factor in the 

 evolution of Homoplasies is correlation, which is sympathy, just as 

 inheritance is reminiscence. The introduction of a single new feature 

 may affect the whole organism profoundly, and one serious case of 



