1008 



SCIENCE 



[N. S. Vol. XXV. No. 



However, I had previously noted and recorded' 

 this fact of the reversion of the offspring of 

 hybrids to the grandparent types. 



Later Montgomery (1901) and Sutton 

 (1902), working on non-hybrid forms, which 

 have chromosomes of varying size, presented 

 striking evidence of the fact that in early 

 germ cells there are pairs of homologous chro- 

 mosomes, one of paternal and one of maternal 

 origin, and that these homologous chromo- 

 somes unite in synapsis. The ensuing reduc- 

 tion division simply brings about their sepa- 

 ration and segregation in different cells. 



Although I had elaborated the idea of the 

 relation between the chromosomal phenomena 

 of hybrids and the reversion of their progeny, 

 in my doctoral thesis of 1900, the published 

 statement of the details did not come out until 

 November, 1902,' but these details show that 

 I had hit ujkju practically the Mendelian idea. 

 Very soon after the appearance of my paper 

 in 1902, Cannon, Wilson and Sutton published 

 conclusions of a very similar nature, endeavor- 

 ing to offer a cytological interpretation of the 

 Mendelian principles, which had had their 

 renaissance in the meantime. Only Can- 

 non's conclusions, however, were based upon 

 the actual study of hybrid material. In this 

 way it became established that there is a mech- 

 anism in germ cells which, assuming that the 

 chromosomes are concerned in heredity, could 

 possibly lead to a segregation of paternal and 

 maternal qualities. 



But if we accept the idea of Boveri and of 

 Sutton that each chromosome contains only a 

 certain pro rata of adult characters, and that 

 reduction is simply the separation of homol- 

 ogous mates, then besides facing the unproven 

 though necessary implication that whole blocks 

 of grandparental characters are barred from 

 the third generation, we must also recognize, 

 as pointed out by Davenport and others, that 

 this theory is not in harmony with certain 

 facts of non-alternative (non-Mendelian) in- 

 heritance. How, for example, are we to ac- 

 count for the well-authenticated instances of 

 (1) blended inheritance, (2) of pronounced 



'Zoological Bulletin, Vol. 11., No. 5, 1899. 



' ' Hybridism and the Germ-cell,' University of 

 Cincinnati Bulletin, No. 21, November, 1902. 



atavism, such as is seen in mongrelized breeds 

 of pigeons, which tend to return to the primi- 

 tive ancestral blue-rock type with apparently 

 none of the components missing, (3) of ex- 

 clusive inheritance or prepotency, (4) of the 

 later outcropping of dominant characters in a 

 recessive individual, which has sprung from 

 a gamete that, according to this chromosomal 

 theory, has been purged of such dominants, or, 

 in other words, the contamination of a given 

 character by its allelomorph, and (5) of the 

 ' fixation ' of a hybrid mosaic gamete, of which 

 the qualities do not segregate in subsequent 

 generations ? 



Even though we have visible evidence of the 

 separation of homologous chromosomes, I fail 

 to see why this necessarily means that each 

 has retained all of its original qualities, or 

 has taken on no new ones. It is easily possible 

 that there can be persistence of form in these 

 chromosomes without persistence of qualities. 

 Supposing that the chromosomes differ from 

 one another in their physical properties (and 

 they doubtless do, more or less, since we can 

 recognize constant types), the form might 

 easily be dependent upon the physical con- 

 sistency of each of the particular chromo- 

 somes, which, in fact, is made up of at least 

 two substances, viz., chromatin and linin. It 

 is evident that in pro-synaptic stages when the 

 chromosomes are characteristically present as 

 hazy granular masses, numerous conjugations 

 or exchanges of individual granules could oc- 

 cur, and yet the matrix (linin) bearing the 

 qualities, persist and appear finally in its 

 characteristic form. 



In my ovtm study on the spermatogenesis 

 of hybrids, I was finally led to consider the 

 strict individuality of the chromosome as per- 

 sisting only in hybrids derived from widely 

 divergent forms. In the case of fertile hy- 

 brids my idea was expressed as follows : " The 

 cases in which the entire plasmas are segre- 

 gated are then probably but magnified images 

 of what occurs among the specific qualities 

 of the milder crosses." ' Furthermore, I have 

 suggested that the occasional inequalities in 

 the divisions of individual chromosomes might 



' Cincinnati Lancet-Clinic, May 9, 1903. 



