Mat 1, 1914] 



SCIENCE 



649 



SCIENTIFIC BOOKS 



Gruppenweise Arthildung. By Hugo De 



Vries, Professor of Botany in Amsterdam. 



Berlin, 1913. Verlag von Gebriider Born- 



traeger. Pp. viii + 365. Figs. 121. Colored 



plates 22. Price 24 marks. 



Somewhat more than ten years has passed 

 since the appearance of the first volume of 

 " Die Mutationstheorie " and we are most 

 fortunate to have from Professor De Vries an- 

 other book that is an extension of the former 

 discussion and that also brings forward a 

 remarkable body of new observations of very 

 great interest to the workers in genetics. It 

 is not often that an investigator is able to 

 follow an earlier work of the scope of " Die 

 Mutationstheorie " with a book of as great 

 import, yet we have in " Gruppenweise Art- 

 bildung " a volume that perhaps outranks the 

 first contribution in matter and in exposition. 

 It will surprise even the students of CEnothera 

 to note the remarkably wide range of crosses 

 among these forms that De Vries has made 

 and the extraordinary mass of observations 

 that he has accumulated. These are carefully 

 indexed and readily accessible for reference. 



The first part deals with the origin of spe- 

 cies by mutation. This is a summary of the 

 views developed in " Die Mutationstheorie " 

 with De Vries's answers to various criticisms 

 that have been expressed to about the year 

 1912. There has been no essential change in 

 his interpretation of the " mutations " from 

 CEnothera Lamarchiana, an explanation of 

 which is offered by a somewhat more detailed 

 statement of his theory of intracellular pan- 

 genesis. Pangens are assumed in any indi- 

 vidual to be in an active, inactive or labile 

 state, and mutations arise when they are in the 

 labile condition. 



As to the status of CEnothera Lamarch- 

 iana, De Vries stands by his original 

 position. It is to him representative of a wild 

 species of American origin. Papers of the 

 reviewer concerning the identity of Lamarck's 

 plant of 1796 and on the problem of the 

 origin of the Lamarchiana of De Vries's cul- 

 tures have appeared too recently to find a 



place in his discussion, and probably for the 

 same reason there is no discussion of the 

 studies of Heribert-Nilsson. De Vries takes 

 a clear position against the view that the 

 " mutants " of Lamarchiana represent the 

 splitting of a hybrid form. He believes that 

 hybrids between species of CEnothera breed 

 true as illustrated by his reported observations 

 on the cross between biennis and muricata. 

 When an CEnothera cross gives a wide range 

 of variants in an F, generation, as in the case 

 of the hybrids between grandiflora and Amer- 

 ican types of hiennis, De Vries apparently 

 assumes that a mutating habit has descended 

 from one or both of the parents. 



Part two, treating of reciprocal and double 

 reciprocal crosses, gives in detail the data upon 

 which the conclusions reported by De Vries in 

 1911 were based. CEnothera biennis and 0. 

 muricata of Holland are two well-defined spe- 

 cies, readily distinguishable, which breed true. 

 Their reciprocal hybrids exhibit constant and 

 marked differences and in the most striking 

 of their vegetative characters strongly re- 

 semble the pollen parent. So uniform is this 

 behavior that De Vries expresses the results 

 with respect to the characters concerned by 

 two formulae — by^m^m,, and m X ^ = ^• 

 The important peculiarity of these hybrids is 

 then the fact that they differ sharply from one 

 another to a degree very unusual in reciprocal 

 crosses. Furthermore, the reciprocals are re- 

 ported to breed true without exhibiting varia- 

 tion in the F, generation as might be expected. 



These reciprocal crosses may be crossed 

 with one another in two ways to give double 

 reciprocals — (biennis X muricata)y^{muricata 

 y^biennis), and (muricata 'X.biennis)y(^{bien- 

 nis yC, muricata) . When this is done the con- 

 trasted characters of the parent type which 

 occupies the center of the formula appear to 

 drop out and the resulting double reciprocal 

 hybrid presents the characters of the parent 

 which occupies the peripheral position. Ex- 

 pressed in simple formulae, which only apply 

 to the vegetative characters under considera- 

 tion— (fe X m)Xi'm Xb)=h and (m X &)X 

 (b yC m) ^ m. The products of the double 



