SUMMARIES OF THE HISTOLOGIC CHARACTERS, ETC. 



311 



and 4. This hybrid should perha] s be a igned to the 

 first or second class. In several other instances there 

 is evident tendency to dominance in one phase especially, 

 as in Eippeastrum titan-cleonia, II. dceones-zephyr, and 



l.ilium marhan. 



Apropos of intermediateness as a criterion of hybrids, 

 it is of interest to note that 4 of the hybrids (Xarcissus 

 poetaz triumph, N. j. t. bennett poe, X. cresset, and 

 Cymbidium ebumeo-lovrianum) do not in a single reac- 

 tion exhibit intermediateness. Two of these belong to 

 the first class, both being conspicuous because four- 

 fifths of the reactions of each hybrid are higher than 

 those of the parents. One belongs to the fourth class, 

 and there are no very definite parental leanings. One is 

 found in the third class, with very definite inclinations 

 to activities that are the lowest or the same as those of 

 both parents, especially the first and in the order given 

 (13, 9, and 4, respectively). 



In recapitulating the totals exhibited by these tables 

 several very interesting points of comparison are elicited 

 (summary of Table F ) . All together 1,018 reactions were 

 recorded, which are distributed as follows : Same as seed 

 parent 137 (13.4 per cent) ; same as pollen parent 94 

 (9.2 per cent); same as both parents 138 (13.6 per 

 cent) ; intermediate 236 (23.2 per cent) ; highest 187 

 (18.4 per cent) ; and lowest 22G (22.2 per cent). It is 

 very obvious that there are much more marked tenden- 

 cies to intermediateness, highness, and lowness than to 

 sameness of development in relation to one or the other 

 parent or both parents, there being somewhat less than 

 two-thirds of the reactions (63.8 per cent) that fall 

 within the first, and 36.2 per cent within the second 

 category. There is about an equal tendency to inter- 

 mediateness (23.2 per cent) as to lowest development 

 (22.2 per cent) and distinctly less tendency to highest 

 development (18.2 per cent) than to either of the for- 

 mer; and there is on an average approximately only 

 about one-half the tendency to sameness to the seed 

 parent (13.4 per cent) and to both parents (13.6 per 

 cent) as there is to intermediateness, the least tendency 

 being shown in sameness to the pollen parent (9.2 per 

 cent). Comparing the tendency to intermediateness 

 with the tendencies to highest plus the lowest reactivi- 

 ties, it is found that the latter predominate in the pro- 

 portion of 23.2 to 40.6 per cent, o'r approximating 1 : 2 ; 

 in other words, there is only a little more than one-half 

 the tendency to an intermediate reaction as there is to 

 one that is above or below parental extremes ; and there 

 is an equal tendency to sameness as one or the other 

 parent as there is to intermediateness. Tf a comparison 

 is made the number of intermediate reactions with the 

 total of other reactions the proportion is found to be 

 23.2 to 76.8 per cent or approximately 1:3, that is, 

 there is in general a likelihood of only 1 reaction in 4 

 being intermediate. When these intermediate reactions 

 are analyzed only 54 of 236, or somewhat more than 

 one-fifth and less than one-fourth (23 per cent), are 

 mid-intermediate, the larger proportion being closer to 

 one or the other parent than to mid-intermediateness. 



Tabu I 



