n8 Animal Life and Intelligence. 



diverse descent and ancestry have, through similarity of 

 surrounding conditions or of habits of life, become, in 

 certain respects, assimilated. But some zoologists go 

 further than this. They maintain that the same genus 

 or species may, through adaptation to similar circum- 

 -stances, be derived from dissimilar ancestors. Some 

 palaeontologists, for example, believe that the horse has 

 been independently evolved along parallel lines in Europe 

 and in America. Professor Cope considers that in the one 

 continent Protohippus, and in the other Hipparion, was the 

 immediate ancestor of Equus. The probabilities are, how- 

 ever, so strongly against such a view, that it cannot be 

 accepted until substantiated by stronger evidence than is 

 yet forthcoming. 



A special and particular form of convergence, at any 

 rate in certain obvious, if superficial, characters, has already 

 been noticed in our brief consideration of mimicry. In 

 the first place, among a number of closely related species 

 of inedible butterflies, the tendency to divergence is checked, 

 so far as external markings and coloration are concerned, 

 that all may continue to profit by the resemblance, and 

 that the numbers tasted by young birds in gaining their 

 experience (for the avoidance seems to be at most incom- 

 pletely instinctive) may be divided amongst all the species, 

 thus lessening the loss to each. Secondly, there may be a 

 convergence of certain genera of distantly related inedible 

 groups (e.g. among the Heliconidae and the Danaidae), which 

 gain by being apparently one species, since the loss from 

 young birds is shared between them. And lastly, there is 

 the true mimicry of quite distinct families of butterflies, 

 not themselves inedible, but sheltering themselves under 



Siphonophora include two groups, closely resembling each other, but of 

 different ancestry: (a) The Disconanthro, traceable to trachomedusoid 

 ancestors ; (6) the Siphonanthae, traceable to anthomedusoid ancestors like 

 Sarsia. (2) M. Paul Pelseneer has beeH led to the conclusion that the 

 pteropod molluscs also include two groups resembling each other, but of 

 different ancestry : (a) The Thecosomes, traceable to tornatellid ancestors ; 

 (fc) the Gymnosomes, traceable to aphysiid ancestors. In each case, the 

 ancestral sea-slug has been modified for a free-swimming life. 



