88 



The Journal of Heredity 



tion in the expression of a sinj^'le j^'enelic 

 factor are found in Antirrhinum and 

 Drosophila. In the former Baur* dis- 

 covered a factor which in inire races 

 conditions the production on the same 

 stem of zyKomor])hic and i)eloric flowers. 

 Mor^an^ has rejKjrted two factors in 

 the pomace fly, viz, for abnormal 

 abdomen and supernumerar\' lej^s, which 

 vary in their exjjression in direct relation 

 to certain elements in the environment. 

 Examples of variability in simple Fi 

 hybrids also occur in Drosophila. Mor- 

 gan^ describes a dominant mutant 

 character, bar eye, in which the normal, 

 roundish eye of the wild fly is replaced 

 by a narrow, bar-sha])ed orjj;an. In the 

 heterozygous condition the sha])e varies 

 from almost round to nearly typical bar, 

 but the F2 ratios prove that bar eye is 

 conditioned by a single factor. Again 

 in Nicotiana tabacum there is a variety' 

 calycina, which produces abnormal 

 flowers. When crossed with another 

 variety having normal flowers the great 

 majority of the flowers in Fi, according 

 to Clausen and Goodspeed, ^^ are normal, 

 but some abnormal ones occur and these 

 fill in a continuous series from the 

 normal type to flowers almost as 

 abnormal as calycina. None of these 

 variations in character expression is of 

 genetic significance. 



EFFECT OF HYBRIDIZATION 



But there is another as]5ect of 

 \-ariation which may well be considered 

 with reference to the case in hand. It 

 is well known that complex hyVmdiza- 

 tion may sometimes increase the varia- 

 bility of characters in Fi individuals. 

 A single illustration must suffice . When 

 Nicotiana tabacum var. calycina is 

 crossed with A^ sylvestris the Fi plants 

 exhibit a much higher develo])ment of 

 the calycine character than a]j])ears in 

 Fi hybrids of calycina with other 

 tabacum varieties. Also all degrees of 

 the calycine character apjjear on the 

 same jjlant, from flowers having only 

 a very slight jjetaloid tendency in one 

 calyx lobe to those exhibiting the full 

 calycine character. 



Now the deciduousness and persist- 



' Baur, E. ICinfuhrung in dit- i-xptTimcntellc VcrtThungsk'hre, 1914, y. 106. 

 "Clausen, R. E., and Goodsi)ccd, T. H. Hereditary Reaction-system Relatiims — An 

 Extension of Mendelian Concepts. Proc. Nat. Acad. Sci. 2, April, 1916, p. 240. 



ency of the calyx lobes in the Le Conte 

 pear and Transcendant crab might be 

 referable to either of the two aspects of 

 \'ariation described above. The varia- 

 bility in the ]jarents and the varialjility 

 in Fi indi\'iduals may be a normal fea- 

 ture of calyx lobe development. Hence 

 persistency and deciduousness of calyx 

 lobes are not necessarily to be considered 

 as absolute, alternative characters, as 

 Tufts assumed, since it is ])ossible that 

 they are merely fortuitous variations in 

 degree of development. 



On the other hand, pomologists 

 generally consider these varieties to be 

 interspecific hybrids. Moreover, in each 

 case one of the supposed parents is 

 known to have deciduous or mostly 

 deciduous lobes, and in the case of the 

 pear the other supposed parent is 

 characterized by persistent lobes. It 

 is not unlikely, then, that each parent 

 contributed one or more factors for its 

 characteristic condition of the cal^Tc 

 lobes. But since these varieties are 

 com])lex hybrids it would be expected 

 that a somatic condition already vari- 

 able in the ]jarents would exhibit even 

 greater variability in Fi oft"s]jring. 

 Other crabs of the Transcendant group 

 also exhibit great variability as to 

 adherence of the calyx lobes, as does 

 also the Kieffer pear, which is considered 

 to have the same hybrid origin as the 

 Le Conte. If the original crosses or 

 similar ones were reiieated, quite difi"er- 

 ent ratios might be observed. If other 

 trees of these varieties should be 

 examined, wide variability in the ratios 

 might be found, although all the trees 

 of a given variety belong to the same 

 clone. The difference found would be 

 attributable to environmental effects, 

 including effect of stock on scion, or to 

 vegetative factor mutations, but not to 

 "somatic segregation." The authors 

 believe that increased variability due 

 to com])lex heterozygosis explains the 

 phenomena in question. 



Sl'.MiMARV 



1. The term "somatic segregation" is 

 considered to be misleading, it since 



