288 ZOOLOGY. 



conditions, and therefore not hereditary ; and a third that 

 differences of environment which are supposed to give rise 

 to selection are slight or " continuous," while species are 

 separated by distinct jumps or are "discontinuous." To 

 all these objections may be added that already mentioned, 

 namely, that specific differences are not always or in the 

 majority of cases differences of adaptation. 



12. Mutations. The Mendelians on the other hand 

 conclude from the results of experimental breeding (see 

 p. 218) that characters are transmitted as " units " or 

 " factors," which are not subdivided but are either present 

 or absent. Mutations consist in the addition or omission 

 of one of these units. The cause of the mutation is not 

 known with certainty, but it is blastogenic, and possibly 

 arises from unequal distribution of the factors in the cell 

 divisions of the gametocytes within the parent. There can 

 be little doubt that the differences between species in 

 nature are of the same kind as those which distinguish 

 breeds and varieties in cultivated animals and plants, and 

 that many of these characters have arisen suddenly and 

 not by a gradual process of selection. 



13. Mutations and Metamorphosis. On the other 

 hand there are many zoological phenomena which the 

 doctrine of mutation and blastogenic variation does not 

 explain or agree with, and which appear more in accord- 

 ance with so-called Lamarckian principles, i.e. the heredity 

 of modifications due to conditions. One of these pheno- 

 mena, which is not considered by the mutationists, is that 

 of metamorphosis. In the well-known case of the frog 

 and other Amphibia, the development of the lungs and 

 disappearance of the gills take place gradually as the 

 individual changes from an aquatic to a terrestrial exis- 

 tence, and this is only a type of numerous similar cases, 

 such as the flat-fishes, in which the change of structure 

 proceeds gradually and continuously in correspondence 

 with a gradual change of habit. 



Recapitulation, i.e. the repetition of ancestral stages in 

 development, is not universal, but it is verv frenuent in 



