CONVEKGENCE AND CHANGE OF FUNCTION 255 



of the foot passes down the middle of the third digit. If we 

 compare the foot of Thoatherium (Fig. 109, D, E) with that of a 

 horse (Fig. 97) it is hard indeed to imagine that the animals 

 possessing such closely similar and highly specialized limbs are 

 not nearly related to one another. Expert palaeontologists, 

 however, tell us that they are not, and we must believe that the 

 resemblance is due entirely to adaptation to a similar . mode of 

 life in the two cases. We shall have something to say as to how 

 this adaptation was brought about in the. case of the horse when 

 we come to discuss the ancestral history of that animal in 

 Chapter XIX. 



It will be readily understood from the examples which we have 

 been considering that the phenomena of Convergence provide 

 many pitfalls for the unwary biologist, and have led to many 

 other mistakes in classification besides the popular error of 

 placing the whales amongst the fishes. " We have already noticed 

 how the limbs of arthropods have come to bear a superficial 

 resemblance to those of vertebrates, though so absolutely different 

 in their essential structure that no anatomist would dream of 

 regarding them as homologous organs. Many aquatic arthropods, 

 belonging to the class Crustacea, Jijse the lobsters, cray-fishes 

 and crabs, breathe by means of gills which bear a superficial 

 resemblance to those^of fishes, but are again by no means 

 homologous structures, and there are other resemblances between 

 arthropods and vertebrates, due probably to convergence, which 

 have led more than one observer to conclude that vertebrates are 

 descended from arthropod ancestors, a conclusion which is by no 

 means justified by the facts. 



We may now further consider the process known as 



of^ function/' in which an organ primarily adapted and used for 

 one purpose takes on a new and altogether -different duty and 

 becomes modified accordingly. The lungs of air-breathing 

 vertebrates, for example, are generally believed to be homologous 

 with the swim-bladder of fishes, for both arise in the same way as 

 outgrowths of the front part of the alimentary canal, in the 

 region of the throat. The swim-bladder of a fish is a hydrostatic 

 organ ; it is filled with gas, the amount of which can be regulated 

 by suitable means, and assists the animal in maintaining its 

 proper position in the watejc-J In the dipnoan fishes, such as 

 the Australian mud-fish (Neoceratodus, Fig. 110), the South 

 American Lepidosiren and the African Protopterus, the walls of 



