650 



SCIENCE 



[N. S. Vol. XLII. No. 1088 



somes of which the lata and semi-laia types 

 with 15, the semi-gigas with 21 and the gigas 

 with 28 chromosomes, respectively, are the best 

 known. Gates gives a clear account of his 

 important studies on lata and comprehensive 

 discussions of triploidy and tetraploidy, the 

 latter with a long list of genera with species in 

 which the chromosome number has been 

 doubled. It is not likely that there will be 

 disagreement with Gates's chief conclusion 

 that the characters of these plants are corre- 

 lated to some extent with the peculiarities of 

 their chromosome counts, which are above the 

 normal 14, and that some of the variants from 

 Lamarchiana and other species, called " muta- 

 tions " are due to irregularities in the distribu- 

 tion of their chromosomes. 



However, certain features of the chromosome 

 behavior at the time of reduction in the species 

 studied by Gates are not discussed from all 

 points of view. Gates emphasizes the fact that 

 in the Oenotheras studied by him the pairing 

 of the chromosomes previous to the reduction 

 division is very loose, so that mechanical ar- 

 rangements favor the irregularities of distribu- 

 tion which actually occur. In these observa- 

 tions other students of Oenothera cytology 

 agree for Lamarchiana and some of its " mu- 

 tants " and for certain other forms. But Gates 

 assumes that this peculiarity, i. e., a loose pair- 

 ing of chromosomes, is a condition prevalent 

 among cenotheras in general, including those 

 which he believes to be pure species, and that 

 the forms of this genus depart from the rule, 

 supported by an overwhelming mass of evi- 

 dence, that in pure species there is a very exact 

 pairing of chromosomes previous to the reduc- 

 tion divisions. Studies among such animals 

 and plants as have a series of chromosomes of 

 diilerent sizes and forms have shown a most 

 remarkable association of these in strictly 

 homologous pairs, and the presumption is justi- 

 fied that a pairing of homologous chromosomes 

 at the time of reduction is the rule in pure 

 species. 



One CEnothera is known, namely, a line of 

 grandifloraj in which the chromosomes follow- 

 ing synapsis become associated very definitely 

 into ring-shaped pairs which are assembled in 



a very orderly manner at the equatorial plate 

 of the reduction division and are subsequently 

 distributed in equal numbers by the separation 

 of the members of each pair. This is a history 

 entirely in accord with that generally reported 

 for animals and plants and the conditions allow 

 of little or no opportunity for irregularities of 

 chromosome distribution. Now it also happens 

 that the seeds of this line as recently tested are 

 almost wholly fertile and that there is scarcely 

 more than a trace of pollen sterility. In short, 

 this line with respect to its fertility and to the 

 regularity of its reduction process presents the 

 behavior to be expected of a pure species. 

 Since one line of CEnothera is regular in its 

 behavior, there are probably other lines equally 

 so, and these may prove to be the species purest 

 of all the cenotheras in their genetical consti- 

 tutions. They have not yet been studied either 

 cytologically or through breeding experiments. 

 On the other hand, Lamarchiana and certain of 

 its derivatives, types of very low seed and pollen 

 fertility, irregular in their reduction divisions, 

 and extraordinary in their breeding habits, 

 have been given much attention and assumed 

 by the mutationists to be representative of pure 

 species. In the face of these exceptional pecul- 

 iarities the natural assumption until other- 

 wise disproved should be that Lamarchiana 

 and these derivatives are not pure species and 

 that the irregularities of their eytological and 

 breeding habits are the result of a hybrid con- 

 stitution. This is a phase of the matter which 

 Gates does not consider. 



A long chapter on " Hybridization " and 

 one on the " Relation between Hybridization 

 and Mutation " discuss the nature of muta- 

 tion crosses, Mendelian splitting, blending and 

 modification of characters, twin hybrids, double 

 reciprocal crosses, etc. There is summarized 

 here the most important of Gates's breeding 

 studies, including those between grandiflora 

 and ruhricalyx. 'So far as the statements of 

 ratios and percentages are concerned, the criti- 

 cism of course applies here as to all past 

 CEnothera work that we have no assurance of 

 complete results since there is no means of 

 knowing what proportion of the seeds sown 

 was fertile or how imperfectly representative 



