196 University of California Puhlications in Botanij [Vol. 5 



other "blends" of this same hybrid. Evidences of "segrega- 

 tion" are thus apparent in these F^ plants and this segregation 

 seems in certain cases to be correlated with increased seed pro- 

 duction. With these facts in mind it seems possible that the 

 X. Tahnmm parent of Mr. Bellair's crosses was of hybrid origin 

 and that F^ individuals involving much the same cross have been 

 produced in our cultures. 



It seems reasonable, thus, to assume that the F^ hybrids pro- 

 duced as a result of crossing N. sylvestris and .V. Tahacum- 

 varieties' back and forth can be taken out of the absolutely 

 sterile category and included among those hybrids that are only 

 partially fertile (cf. Lock, 1909). "With the reference to the 

 cytological evidence on the question of the partial fertility of 

 these hybrids a considerable investigation is under way, but the 

 following facts previously noted seem pertinent in this connec- 

 tion. A separating or absciss layer is formed in the case of 

 at least 50 per cent of the hybrid flowers before any fertilized 

 ovules could be normalW matured, and in the majority of cases 

 where such a separating layer is not outwardly visible, a very 

 small amount of viable seed is formed. Obviously the formation 

 of this separating layer cuts off all po.ssibilities of the transfer 

 of food materials to the maturing ovules and it appears to be 

 plain that in the parental species a very considerable demand 

 upon available food supplies is made by such maturing seed 

 (Goodspeed, 1913 (1)), Again it has been found that, given 

 five or six flowers just past anthesis on the primary, terminal 

 inflorescence of a lateral shoot, if the remaining buds, smaller 

 leaves and partially opened flowers present on this lateral are 

 allowed to remain and proceed to their full development, all or 

 all but one of these five or six flowers past anthesis will fall in 

 a week or ten days. On the other hand if all the terminal buds 

 about these five or six flowers and all buds below in the axils of 

 young leaves are removed, all or all but one of these five or six 

 flowers past anthesis will ripen seed capsules which will contain 

 on an average five viable seeds each. On the one hand, then, 

 there definitely appear to be certain combined physiological and 

 structural hindrances to any seed production at all in these F^ 

 hvbrids and on the other hand cvtological investigation mav dis- 



