INTRODUCTION 31 



example, sediments which we find deposited in rivers are found to contain 

 chiefly the forest-living and slow-moving, browsing types, while deposits 

 which were made on flood plains contain the grazing, swift-moving types. 

 Another fact of great interest is that this separation of the quadrupeds 

 or Herbivora naturally brought about a separation of their carnivorous 

 enemies into powerful, short-limbed types with massive jaws, and swifter, 

 light-limbed types, with more slender parts. 



Adaptation to alternations of habitat. — In the long vicissitudes of time 

 and processions of continental changes, animals have been subjected to 

 alternations of ha])itat either through their own migrations or through the 

 migrations of the environment itself; that is, a habitat to which an animal 

 has become adapted may be abandoned for a long period of time and 

 atlaptations are acquired for a second habitat. Following this again, life 

 in the first habitat may be resumed. Dr. Louis Dollo has contributed most 

 brilliant discussions ^ to this alternation of habitat theory as applied to 

 the interpretation of the anatomy of the marsupial mammals of Australia 

 and of several of the orders of reptiles and fishes. Although often obscure, 

 the anatomical proofs of adaptations corresponding to alternations of 

 tabitat are recorded both in the skeleton and in tlie soft parts of animals. 



In his brilliant essay of 1880 Huxley ^ suggested that the primitive forms 

 of marsupials were all arboreal, or tree-living, an hypothesis which has 

 been abundantly confirmed by the careful studies of Dollo' and Bensley,'' 

 according to which we may imagine that the marsupials passed through: 

 (1) a very primitive land-living, or terrestrial phase, in which the limbs 

 would be normally developed; (2) an arboreal, or tree-living phase, in 

 which some modifications of the limbs for grasping of the boughs would 

 be acquired, as illustrated in the tree phalangers of Australia; (3) a sec- 

 ondary land-living, or terrestrial phase, in which the arboreal adaptation 

 of the limbs is checked and a new adaptation to swift-moving, or cursorial 

 habits is acquired, as in the kangaroos, in which the hind limbs especially 

 are modified for leaping and rapid progression; (4) a return to arboreal 

 life, Avith further adaptations for tree-living habits in limbs which have 

 already been extremely modified in course of the earlier phases, as in the 

 tree-kangaroos. 



Improbable as such a theory of alternation of habitats appears at first 

 to be, it is none the less supported by the strongest anatomical evidence 

 in the study of the feet of the marsupials, in which the record of one adap- 



• 



* Dollo, Louis, Los ancStres des Marsupiaux ctaiont-ils arboricolcs? Trnv. Stat, zoiil. 

 Wimcreux, Tome VII, 1899, pp. 188-600, pi. XII. 



- Huxley, T. H., On the Application of the Law.s of Evolution to the Arrangement of 

 the Vertebrata and more Particularly of the Mammalia. Proc. Zodl. Soc, 1880, pp. 649-662; 

 Sci. Mem., Vol. IV, pp. 457-472. 



■' Bensley, R. .\rthur. On the Evolution of the Australian Marsupialia; with remarks on 

 the Relationships of the xMarsupials in General. Trans. Linn. Soc, London (2) Vol. IX, Pt. 3, 

 190.3, pp. 83-214. 



