II.] ON HEREDITY. 89 



the same degree in each successive generation, although the 

 reahzation would be less perfect. The complete disappearance 

 of a rudimentary organ can only take place by the operation 

 of natural selection ; this principle will lead to its elimination, 

 inasmuch as the disappearing structure takes the place and 

 the nutriment of other useful and important organs. Hence 

 the process of natural selection tends to entirely remove the 

 former. The predisposition towards a weaker development 

 of the organ is thus advantageous, and there is every reason 

 for the belief that the advantages would continue to be gained, 

 and that therefore the processes of natural selection would 

 remain in operation, until the germ had entirely lost all 

 tendency towards the development of the organ in question. 

 The extreme slowness with which this process takes place, 

 and the extraordinary persistence of rudimentary organs, at 

 any rate in the embyro, together with their gradual but finally 

 complete disappearance, can be clearly seen in the limbs 

 of certain vertebrates and arthropods. The blind-w^orms have 

 no limbs, but a rudimentary shoulder-girdle is present close 

 under the skin, and the interesting fact has been quite re- 

 cently established^ that the fore-limbs are present in the 

 embyro in the form of short stumps, which entirely disappear 

 at a later stage. In most snakes all traces of limbs have been 

 lost in the adult, but we do not yet know for certain whether 

 they are also wanting in the embryo. I might further mention 

 the very different stages of degeneration witnessed in the 

 limbs of various salamanders ; and the anterior limbs of 

 Hesperornis—thG^ remarkable toothed bird from the cretaceous 

 rocks — which, according to Marsh ^, consists only of a very 

 thin and relatively small humerus, which was probably con- 

 cealed beneath the skin. The water-fleas [DapJuiidae) possess 

 in the embryonic state three complete^and almost equal pairs 

 of jaws, but two of these entirely disappear, and do not 

 develope into jaws in any species. In the same way, the 

 embyro of the maggot-like legless larva of bees and wasps 

 possesses three pairs of ancestral limbs. 



There are, however, cases in which, apparently, acquired 



^ Compare Born in 'Zoolog. Anzeiger,' 1883. No. 150, p. 537. 

 * O. C. Marsh, ' Odontornithes, a Monograph on the extinct toothed 

 Birds of North America,' Washington, 1880. 



