THE MUTATION THEORY. 



177 



present ; so that the flowers of the wreaths found in Egyptian tombs are 

 like those of to-day, because the chmate of Egypt has not altered. But 

 when he introduced his Oenotheras from very sandy soil to a richly 

 manured garden, they at once began to vary, or '* break," as gardeners 

 say. He alludes to geological periods ; but the frequent gaps by the loss 

 of strata and our ignorance of the organisms missing, invalidate any 

 conclusions of geological periodicity in all kinds of organisms. In the 

 short section on " The Explanation of Adaptations " referred to he 

 says : This problem lies outside the scope of the present book, which is 

 only concerned with the empirical foundations of the theory of descent " ; 

 and he holds to the ' ' operation of natural selection " ; so it would seem 

 that he is unaware of Darwin's alternative, in which " no selection of 

 any kind is required . " * 



We now come to the most important part of the book, viz., his 

 elaborate experiments, with their tabulated results, on the degrees of 

 fixation by heredity, given in percentages, of a number of teratological 

 cases. 



The total results may be concentrated in the fact that when the 

 " monstrosity " has once appeared nourishment tends to increase and 

 estabhsh the malformation. It then possesses an inherent tendency 

 to be hereditary. Lastly, the percentage varies from zero to nearly a 

 hundred. Professor de Vries attributes the first appearance to latent 

 causes which are unknown. 



In the third chapter the author states that new species arise — 



A. By the formation of new characters. 



B. Without the formation of new characters. (1) existing charac- 

 ters becoming latent; (2) latent characters become active {^.g- 

 atavism); (3) from hybrids. 



A might perhaps be illustrated by kohl-rabi or species with dis- 

 sected types of foliage, if a single character be allowed to be specific. 



B (1) Might be illustrated by a land plant becoming aquatic, with 

 dissected foliage and various losses of anatomical details. 



B (2) Would apply to spinescent species losing their spines by 

 developing them into branches when supplied with water. 



B (3) Hybrids are not regarded as species, if known to be such. 



Alluding to " atavism by bud- variation, " the author refers to " one 

 of the best instances," as he considers it, Cephalotaxus pedunculata 

 fastigiata, resembling the Irish yew. The erect stem and erect branches 

 bear leaves scattered all round on the f or, perhaps, f plan, according 

 to Oarriere's figuresf ; but horizontal shoots have the leaves also hori- 

 zontal, exactly as m the common yew, the young shoots of which, if 

 at all erect, have the leaves scattered. There is, however, no change in 

 the insertion of the leaves, which are merely twisted on their short 

 petioles, so that this " scattered-leaved " form really represents the 

 primeval type, though it may be the parent form of the so-called 



* Animals and Plants under Domestication, ii., p. 271, &c., and in several 

 places in the 6th edition of the Origin of Species. 



t Production et Fixation des Varietes dans les Vegetaux, pp. 44, 45 (1865). 



VOL. XXXVII. N 



