162 



The Journal of Heredity 



There is here a large excess of hetero- 

 zygotes, increasing rapidly with the 

 intensity of the linkage, the results for 

 coupling being the same as those for 

 repulsion. 



11. SUBLETHAL FACTORS, OR FACTOR COM- 

 BINATIONS, AFFECTING THE ZYGOTE 



These factors usually cause the death 

 of a fraction only of the zygotes possess- 

 ing them, the mortality varying with 

 the en\-ironment from zero to perhaps 

 iolal. Selective elimination, or differen- 

 tial viability, are temis often applied to 

 this mortality, which may be found in 

 the second generation of plant crosses, 

 especially wide crosses. When vicin- 

 ism is absent, the presence of sublethal 

 factors may often be recognized by the 

 abnormal ratios in the second genera- 

 tion, which usually include a lessened 

 proportion of recessives. A back-cross 

 of Fi with a full recessive best shows 

 this differential viability. In Mat- 

 thiola, Saunders (40) found the hetero- 

 z^-gote 1 to 1 double-thrower to have a 

 sublethal factor combination. In flax, 

 I'ammes (46) ascertained that in a cross 

 of blue and white- flowered races, in 

 a progeny of 4,000 plants, there was 

 always a deficit in the recessive white- 

 flowered. The white-flowered plants 

 were found to have 13% fewer seeds 

 in a seed vessel than the blue-flowered 

 plants of the same family: and of the 

 seeds the white-flowered plants pro- 

 duced, a smaller percentage germi- 

 nated than of those from the blue- 

 flowered. Hence a^. was sublethal to 

 the embryos. {Aa plants, however, 

 were apparently just as viable as Ao 

 plants.) After reckoning in the ob- 

 served mortality, the ratios agreed 

 with expectation. 



If a recessive sublethal factor ao is 

 the cause of selective elimination in the 

 case of X of the individuals ]30ssessing 

 a-i in the second generation; then a 

 factor B being linked with A, there 

 will 1)e found, among the n surviving 

 ])lants of the second generation, 3.v 

 plants which will be with regard to B 

 in the proportions {}n'^-\-2mn)B<>: 2{m^-[- 

 mn-\-n')Bb:{n--\-2nin)}h; while the n-\- 

 3.V remaining plants of the second gen- 

 eration, in which a-i was not i)resent or 



was not eliminated, will be in the 

 ordinary proportion of IB^: 2Bb: I62. 

 It was shown above that the fornier of 

 of these two proportions gives ratios 

 greater than 3:1 for coupling, and less 

 for repulsion. Hence, in all cases 

 where a factor B is linked with a sub- 

 lethal factor, A or a, the second-genera- 

 tion ratios will differ from the normal, 

 not only for A and a, but also for all 

 other differential factors in the same 

 chromosome pair. 



HI. LETHAL FACTORS, OR FACTOR COM- 

 BINATIONS, ACTING ON THE POL- 

 LEN GRAINS AND EMBRYOSACS (hAP- 

 LOID GENERATION). 



These factors cause partial sterility, 

 where only a definite fraction of the 

 pollen grains and embryosacs remain 

 \'iable. They cause a selective elimina- 

 tion of pollen grains and embryosacs, 

 and their effect must be distinguished 

 from that of Z3^gotic factors which also 

 may cause a total abortion of pollen 

 grains (as in the recessive sweetpea 

 with aborted pollen, 18), or of em- 

 bryosacs (as in some double petunias), 

 or partial abortion (as I have found in 

 some F3 families from Stizolobium 

 crosses) ; but do not cause selective 

 elimination among members of the 

 haploid generation. 



A special case of partial sterility due 

 to lethal combinations of factors is 

 semi-sterility, where half of the pollen 

 grains and half of the embryosacs 

 perish because of their possession of 

 such lethal combinations. Semi-ster- 

 ility has been especially studied in three 

 Stizolobium crosses (5 and 6), where 

 a satisfactory hypothesis was that each 

 of the two combinations of factors, 

 KL and kl, not found in the original 

 ]3arents, was lethal; while either of the 

 two combinations, 7v7 and IcL, peculiar 

 to the ])arents, was not lethal. // and 

 P were factors (for lateness of flower- 

 ing, and ])igmentation of seedcoat) 

 linked with K. In /'u. both II and P 

 occurred in the usual 3 : 1 ratio, not- 

 withstanding the elimination of pol- 

 len grains and embryosacs. But the 

 second generation consisted of fertile 

 and semi-sterile plants in equal numbers. 

 Among these two classes, the factor //, 



