230 Castle. 



open question whether such a thing as a race pure genetically fur au}' 

 particular seed width is obtainable. 



The explanation which has been adopted by most of those who 

 have given attention to size inheritance is substantially that outHned 

 by Bateson in 1902 and elaborated by Nilssox-Ehle in 1909, and is 

 regarded by those who adhere to it as an "extension" of Mendelism. 



On this view: (1) Size differences not of environmental origin 

 are due to Mendelizing unit-factors or "genes". (2) These lack domi- 

 nance, so that crosses produce intermediates, but they segregate when 

 gametes are produced precisely as ordinary Jlendehan unit characters 

 do. (3) The unit factors which are responsible for size differences 

 clearly distinguishable are numerous so that segregation does not occur 

 in simple 1:2:1 ratios but in those which are so complex that they 

 produce seemingly smooth continuous variation curves. (4) The more 

 numerous such factors are, the more nearly will Fs resemble Fi in its 

 variability. If their number were unlimited, it is possible that we 

 should be unable to distinguish the variability of Fa from that of Fi, 

 in a hmited number of observations. 



This explanation, it seems to me, has arisen from the idea that 

 Mendelian gametes are pure, the idea advanced but not adopted by 

 Bateson in 1902, but accepted without question by most of those who 

 have since concerned themselves with the study of Mendelism. The idea 

 of gametic purity, however, has not received any adequate support from 

 the few observations and experiments which have been made with a 

 view to test its validity, and if we discard this idea the several assump- 

 tions involved in the so-called Mendelian explanation of size inheritance 

 become quite superfluous. 



By "gametic purity" in Mendelian crosses, we understand the idea 

 that a particular unit-character is ever and always the same and can 

 not he modified by crosses; that it emerges fi-om a cross in the gametes 

 formed by a cross-bred individual exactly the same as it existed in the 

 pure-bred ancestor of the cross-bred. 



In justice to Bateson, it should l)e said that when in 19o2 he 

 discussed the subject of gametic purity he scarcely had in mind ab- 

 solute identity of a unit character at all times, as the foregoing defi- 

 nition implies. He expressly disclaims this idea, as the following passage 

 (p. 128) shows: 



•'From analogy — an unsafe guide in these fields — and from 

 what is known of discontinuous variation in general, we incline to the 

 view that even though the figures point to a sharp discontinuity between 



