288 HABIT AND INTELLIGENCE. [cHAP. 



first impulse will probably be to conclude that it is an 

 inextricably complex network. Such a conclusion would 

 no doubt be incomparably nearer the truth than the 

 notion which it has superseded, of a chain ; and I think it 

 likely that the majority of naturalists are at present rest- 

 ing in that conclusion. Nevertheless, I believe it will be 

 found, that as the notion of a single chain of affinities 

 rested on a total misconception of the facts of comparative 

 morphology, so that of a network, with affinities in aU 

 directions, rests on an incomplete knowledge of them. As 

 I believe already stated, I believe the true form of classification is 

 in form ^ that of a tree, the branches of which, after diverging, never 



like a tree, reunite. 



The analogy of a tree may help us further. Were it 

 possible for an intelligent man to see for the first time a 

 large and many-branching tree without knowing anything 

 of its mode of growth, he would at first fancy that its 

 branches formed a network ; and it would require a good 

 deal of careful examination before he could be quite con- 

 vinced that such an inference was a purely visual illusion, 

 and that in not a single case do they reunite after diverg- 

 ing. Just so, I believe the natural and plausible notion of 

 a network of affinities is due to our imperfect knowledge. 

 Groups The organic tree has two main trunks, the vegetable and 

 unite after ^^^ animal : these, at least, do not reunite, but diverge and 

 diverging, re-di verge into classes, orders, genera, and species, though 

 not with the regularity that appears to be implied in those 

 somewhat technical expressions. I do not say it is proved 

 that groups never reunite after diverging ; I only say that 

 I believe every advance that is made in systematic biology 

 and in true classification tends to prove it. 

 Affinities It is, however, scarcely to be hoped that the outline of 

 probably *^^ Organic genealogical tree will ever be traced in all its 

 never be parts with even approximate fulness and accuracy. Great 



T)Grf6ctlv 



traced. progress, no doubt, has been made, and more remains to 

 make. Many connecting links of affinity have been 

 found : some, by the more careful anatomical or micro- 

 scopic study of organisms already known ; others, by the 

 discovery of new forms, fossil as well as living ; and 



