MENDEL'S PRINCIPLES APPLIED TO WHEAT HYBRIDS. 891 



" long " + 50 % " intermediate " + 25 % " short." The average percentage 

 of the ninety-five plats works out approximately 26'5 "long" + 48*2 

 intermediate -I- 24*5 " short," being a close approximation to the Mendelian 

 ratio of 25 : 50 : 25. 



This case is peculiarly interesting from the practical point of view, as 

 it shows that normal intermediate hybrids of the blended type of inherit- 

 ance also separate their characters in accordance with the Mendelian 

 Principles, thus confirming what the writer has already shown in orchid 

 hybrids (Journal R.H.S. xxvii. pp. 614-624), viz. that Mendel's Principles 

 do not depend at all upon the question of Dominance in the first generation, 

 which latter is merely a phase of inheritance, the actual causes of which 

 are at present not quite clear, though it seems probable that there is some 

 connection between inbred races and Dominance. 



Formerly it was thought that the Mendelian Principles were only 

 applicable to Dominant hybrids and crosses, but now that intermediate 

 hybrids are being brought into line, and as they appear to be generally 

 more numerous than Dominant hybrids, it seems likely that the Mendelian 

 Principles will ultimately be extended to cover all the phenomena of 

 hybridisation and cross-breeding. At the same time it should be clearly 

 recognised that there are undoubtedly many complications existing which 

 were not fully known to Mendel, and which future experiments alone can 

 unravel. 



The remaining' two intermediate hybrids, when self -fertilised, proved 

 slightly exceptional inasmuch, as they produced no " short " forms at all, 

 but simply 75 % intermediate and 25 % " long " (see Table XII., Plats G22 

 and G19). It will be noted, again, that the exceptions only occur in one 

 cross, Table XII., and that Plat G22 is exceptional in all three characters, 

 while Plat G19 is exceptional in two characters out of three, all of which 

 lends colour to the suggestion made under Table C, that one of the varieties 

 of this cross is not truly pure and constant in its characters. 



Having dealt with the behaviour of all the "Dominant" hybrids, we 

 now come to the consideration of those few forms which in the original 

 cross showed the " Recessive " form rather than the usual "Dominant." 

 In other words, the " Recessive " becomes Dominant, and the " Dominant " 

 Recessive. This may at first sight appear to be a small matter, and one 

 which Mendel does not seem to have experienced, but it appears to me to 

 be of great importance, as it touches the vital question of individual 

 versus ancestral inheritance. 



In Prof. Spillman's experiments, there appear to have been fourteen 

 cases in the four characters, out of a total of 249, and these when self- 

 fertilised seem to have given two different results : viz. in six cases the 

 "Recessive" character was again reproduced almost pure, and in the 

 remaining eight cases in the approximate proportion of 75 % " Recessive " 

 to 25 % " Dominant." (For details see Table I., Plats A3, A4 ; III., C7 ; 

 VI., El, Ell ; XL, 115 ; XII., G7, G9, G15, HI, Gil, G21.) 



It will again be noted that all the six cases in which the Recessive 

 remained almost pure occurred in the same cross (Tables XL and XII.) 

 which gave the exceptions to the Mendelian cases. Disregarding these, 

 therefore, and looking at the remainder of Prof. Spillman's facts, it 

 would appear that the Mendelian Principles are followed in every case, 



