652 



SCIENCE 



[N. S. Vol. XXVI. No. 672 



there, the prospect of successfully tracing 

 them among the specific types does not look 

 very hopeful. 



Again, no correspondence between the 

 chromosome numbers and complexity of 

 structure has ever been asserted to exist. 

 Low forms may have many; highly com- 

 plex types may have few. 



Then, on the contrary, very closely allied 

 types may show great differences in these 

 respects. As you are aware, Rosenberg 

 has shown that one species of Drosera has 

 20, while another has 10.^ Again, Miss 

 Lutz has found a similar state of things in 

 CEnothera gigas, which has 28, while (Eno- 

 thera lata has 14. Obviously this doubling 

 means something definite, but it is not sug- 

 gestive of the determination of specific dif- 

 ference. In Aphis Miss Stevens, on the 

 other hand, has shown how wide a diversity 

 may be presented by the chromosomes of 

 forms so alike as to have passed for one 

 species. These difi'erences prove both too 

 little and too much. I can not but believe 

 that all this evidence points to the conclu- 

 sion that we are about to find among the 

 chromosomes one more illustration of the 

 paradoxical incidence of specific difi'erence, 

 not the fundamental phenomena on which 

 that difference depends. Among coleop- 

 terists punctulation is sometimes a feature 

 of great systematic importance. To dip- 

 terists neuration and chsstotaxy sometimes 

 give useful critical data. In certain orders 

 of Lepidoptera, the Hesperidse, for ex- 

 ample, the structure of the gonapophyses 

 sharply distinguishes the species where all 

 outward tests fail. But proceeding farther 

 with each of these criteria, we are sure to 

 come upon other groups where for a long 

 series of diverse types the critical feature, 

 so important elsewhere, may show no dif- 

 ferences, or, on the contrary, may show 



^ Important evidence as to these chromosome 

 numbers has been published by E. R. Gates, Bo- 

 tanical Gazette, February and July, 1907. 



hardly any stability. I have digressed out- 

 side my province in these remarks. My 

 excuse must be that I have a rare oppor- 

 tunity of speaking to a great school of 

 cytologists, who must, sooner or later, be- 

 come the colleagues of us breeders in the 

 attack on genetic problems, and I can not 

 resist saying how the facts strike an ob- 

 server who is highly intei'ested, and I may 

 truly say unprejudiced. I suspect, then, 

 that the specificity of the chromosomes may 

 conform in general to these other phenom- 

 ena of specificity. 



There remains the suggestive fact that 

 all that has been witnessed regarding the 

 behavior of the chromosomes is in, fair har- 

 mony with the expectations which our 

 Mendelian experience would lead us to 

 form respecting the hypothetical "bear- 

 ers" of varietal differences. On the other 

 hand, with one striking exception, nobody 

 has been able to connect a cytological dif- 

 ference with a character-difference in any 

 instance. The exception, of course, is the 

 case of the accessory chromosome which 

 Professor Wilson so admirably demonstra- 

 ted to us yesterday. Of that I shall speak 

 again hereafter. 



But though, in regard to these profound- 

 er questions, our knowledge is so defective, 

 the results of experimental breeding are 

 beginning to limit the problem in very defi- 

 nite ways. We know first the fact deduced 

 from Mendel's original experiments with 

 peas, that the bodily characters may result 

 from the transmission of distinct unit-fac- 

 tors. According to Mendel's own concep- 

 tion these factors existed in alternative or 

 allelomorphic pairs, of such a nature that 

 only one member of any one pair can be 

 carried by a gamete. Now though we can 

 not quite prove this first account to be 

 wrong, it is nevertheless possible to express 

 all Mendelian phenomena in terms of a 

 simpler system, according to which the al- 

 lelomorphism may be represented as con- 



