SPECIES CROSSES IN THE GENUS NICOTIANA. 225 



forbids our accepting this conclusion as a true desciij)tion of 

 the phenomena. It is legitimate to hope that future experi- 

 ment will reduce some parts at least of this apparent chaos to 

 some kind of order. 



In certain cases it appeared that a more or less intermediate 

 form or character persisted in the second generation from the 

 cross. Apparent blended inheritance of this kind was only 

 seen in the case of characters of general habit or shape, which 

 are very difficult of exact definition, although in other cases 

 signs of segregation in shape or habit were not wanting. The 

 phenomenon is one which has to be reckoned with , and it will 

 be useful to enumerate the reasons for hesitation in assuming 

 that such apparent blended inheritance must be due to the 

 production by the hybrid of germ cells containing a true and 

 permanent blend of character-determinants. 



The delayed segregation described in the case of certain seed 

 characters in other plants is hardly likely to help us here. Two 

 important considerations remain : — 



1. Tlie possibly very complicated nature of the characters 

 in question has to be considered. It is reasonable to suppose 

 that such a " character " as habit or the general shape of a 

 flower may depend upon the co-operation of quite a number 

 of aUelomorphic factors, even though we may never be able to 

 distinguish between the individual factors. 



Where two such factors are concerned, in cases where the 

 heterozygote form is intermediate, we should expect 1 out 

 of 16 F 2 individuals to resemble each parent form, and the 

 remaining 14 plants to be more or less intermediate. 



With three factors we should expect only 1 of each parent 

 type out of 64 individuals. 



There seems to be no reason for doubting that the total 

 number of visible factors even which go to make up the total 

 shape of a flower may be considerably greater than this. 



2. The second consideration lies in the possibility of a 

 differential fertility among the possible aUelomorphic combina- 

 tions. The possibility has to be borne in mmd tliat among the 

 offspring of a hybrid between two widely different species those 

 individuals which nearly resemble their immediate parent 

 may have a better chance of survival than the forms which 



