362 RAYMOND PEARL. 



other words, do 2 kernels which are intermediate somatically 

 give rise to any different sort of progeny when planted than do 

 kernels which belong clearly and indubitably to one or another 

 of the well defined gametic classes in F?" To answer this ques- 

 tion carefully controlled plantings of somatically intermediate 

 kernels were made in 1910. Series of starchy and of sweet 

 kernels were formed ranging in each case from pure white -at 

 one end to pure deep yellow at the other end. Then rows were 

 planted as follows: (i) pure white, (2) deep yellow, (3) the lightest 

 yellow to be found (== somatic intermediates), (4) the yellowest 

 whites to be found ( : = somatic intermediates). The kernels 

 in classes (3) and (4) were such as would be classified with 

 the yellows by some observers and with the whites by others. 

 The rows included about twenty plants each and were made 

 in duplicate, and in some instances triplicate for both starchy 

 and sweet series. In each row a varying number of ears were 

 self-fertilized (i. e., pollinated by hand with pollen borne on the 

 same plant). Owing to the numerous vicissitudes incident 

 to hand-pollination, together with pressure of other work, as 

 large a number of good ears as would be desirable was not ob- 

 tained. Some of the possible gametic combinations were not 

 represented at all in the progeny ears. This part of the investi- 

 gation is, in consequence, not complete. It seems desirable, 

 however, to present briefly the general result shown by the fifty 

 odd ears at hand. 



This result was that there was no discernible difference what- 

 ever between the progeny of groups (i) and (2) as a class, and 

 that of groups (3) and (4) as a class. In (3) and (4) some of the 

 kernels planted were of course heterozygotes and some were 

 homozygotes. The same was true, however, of the kernels of 

 (i) and (2). In each case a typical Mendelian result was ob- 

 tained, and ihis result could have been predicted in every case 

 (with the exception to be noted presently) had the gametic con- 

 stitution of the kernel been known when it was planted. It 

 could not have been predicted from the somatic appearance ol 

 the kernel. 



The only behavior of an exceptional character observed in 

 these selfed ears was that in certain of the white sweet kerm K, 



