26 



The Journal of Heredity 



butterfly. Analogies are interesting but 

 they do not constitute scientific argu- 

 ment, however much they may aj^peal to 

 the sociaHstic and half educated mind. 

 Much of the present-day mechanism has 

 a foundation not more substantial than 

 the resemblance between a butterfly and 

 an angel. 



NATURE OF SPECIES 



Our author makes a serious beginning 

 with his subject in connection with the 

 discussion of the perennial subject of the 

 nature of a species. He points out that 

 the Linnean conception of a species was 

 a collection of like individuals, influenced 

 more or less in unimportant details by 

 conditions of environment. For some 

 reason, which he does not make clear, 

 he also conceives this to be the mor- 

 phological definition of a species. An 

 unfortunate defect of the work through- 

 out is a strong bias against morphology, 

 which Darwin strongly states in his 

 "Origin of Species" to be the soul of 

 biology. To the Linnean definition of 

 a species the author applies the name 

 Linneon. Jordan in the nineteenth 

 century pointed out, in a controversy 

 with De Candolle, that it was necessary 

 that a species should not only be defined 

 as an aggregation of like individuals, 

 but that there should be added to this 

 conception the quality of coming true 

 to seed. To this conception the author 

 applies the name of Jordanon. His own 

 conception of a species is an assemblage 

 of like individuals, which not only breed 

 true to seed but likewise show themselves 

 genetically pure when back-crossed. He 

 naively admits, however, that this 

 definition does not work in every case 

 and that the morphological criterion of 

 pollen sterility must be employed in 

 doubtful instances. Following Lotsy's 

 own terminology the genetically deter- 

 mined species may perhaps be desig- 

 nated a Mcndelon. It seems clear that 

 genetical analysis, in addition to the 

 older criteria of Linnaeus and Jordan, 

 cannot be considered as infallible criteria 

 of species, however much the tenden- 

 cies of the moment may seem to justify 

 such a conclusion. There are many 

 cases of known hybrids which are quite 

 constant under the most severe genetical 



analysis. If a definition of a species is 

 something really attainable, it must be 

 arrived at by the use of all possible data, 

 and above all those supplied by internal 

 morphology. To the conception of a 

 species thus broadly founded, we may 

 perhaps, in harmony with Lotsy's ter- 

 minology, apply the designation, Dar- 

 winon, after the greatest of all biologists. 

 Perhaps the most suggestive chapter 

 in the work is that which deals with the 

 possibility of Linneons, or assemblages 

 of like individuals, taking their origin 

 from self-fertilized hybrids on the one 

 hand, or from a hybrid community in 

 which free intercrossing obtains on the 

 other. In the former instance heterozy- 

 gosis in a few generations is reduced to an 

 extremely small percentage, on the basis 

 of formulae put forsvard by Jennings. 

 In the case of the free intercrossing of 

 heterozygotes, dominance and partial 

 dominance in a short time make for an 

 apparent uniformity which would well 

 accord with the ordinary systematic 

 conception of a species, that is the 

 Linneon as defined by our author. It 

 is perhaps well to point out in the pres- 

 ent connection the interesting results 

 obtained by Professor Tower of the 

 University of Chicago in his interbreed- 

 ing in nature of two species of potato 

 beetle, namely the Colorado beetle, 

 D. decimlineata, and the Mexican, D. 

 oblongata. A uniform community was 

 obtained in a few generations, which 

 quite accords with the theoretical con- 

 clusions reached by Reimers. and cited 

 by Lotsy in the case of the freely crossing 

 heterozygous population. 



A FUNDAMENTAL DEFECT 



The fundamental defect in the work 

 under review seems to be an excessive 

 confidence in experimental results. Its 

 author assumes that, since variability 

 in ofi"spring is a wcll-kn(^wn conse- 

 quence of crossing, hence all variability is 

 an indication of heterozygosis. This 

 will be admitted by all but jug-handled 

 geneticists to be a conclusion without 

 solid foundation. The morphological 

 criteria of hybrids, geograpliical dis- 

 tribution, an(l dcvcloi:)ment in geologi- 

 cal time should nil be taken into con- 

 sideration in arriving at conclusions 



