178 Report of the State Geologist. 



In this case the regression may be simply explained by the action 

 of natural selection. As a striking example, we may cite the 

 case of birds. The wing and the foot of the bird are in a state 

 of regression in relation to the normal type of the limbs of Ver- 

 tebrates; various bones are found in a rudimentary state, the teeth 

 do not exist in these animals in an adult condition. But phylogenic 

 evolution demonstrates clearly how this regressive adaptation 

 gradually established itself. The most ancient bird known at the 

 present day, the ArchcBojpteryx, possesses limbs much nearer 

 these ancestral type, and teeth of the same kind as do the birds of 

 the Cretaceous, which are even more specialized than the Archoe- 

 opteryx. If now we take into consideration the results furnished 

 by embryogenic development, we see that the limbs of the 

 ostrich in the young state show characters resembling those of 

 these ancient forms ; we see that very young parrots have in the 

 alveoles teeth which do not develop and which among other 

 birds seem never to have existed at any period. In this case the 

 parallelism of Ontogeny and Phylogeny is striking, and furnishes 

 a clear conception of the mechanism of regression. 



After the discovery of the law of parallelism it appeared 

 as though the definitive method of phylogenic classification had 

 been found, and numerous systems have been, proposed, founded 

 exclusively on the characters of development, for example, on 

 the position or nature of the vitellus and of the coverings of 

 the ^^g. If one of these systems rested on a basis strictly exact 

 in theory, we ought to put aside the inconvenience it might 

 present of being always difficult or impossible of application in 

 Palaeontology, and do our best in our endeavors to conform to it. 



But this is not the case, and the bearing of the law of Serre_s 

 and Haeckel is restricted by other phenomena, which we will 

 now consider. 



Embryogenic acceleration. — Yery often two proximate forms? 

 for instance, two species of the same genus [pY more frequently 

 two proximate genera) develop in very different modes. No per- 

 son would for a moment conclude from this that the ancestral 

 series of these two forms were distinct, the more so as the differ- 

 ences appear most generally in the earlier stages. It must 

 indeed be admitted that in these cases the normal development, 



