IN THE tropics: III. 133 



more from G. W. S. With the single exception already men- 

 tioned (e. 7 of Table 30), the shape of the grains, whether 

 sugary or half-dent (3 of the great-grand parents were of the 

 indent form and 5 of the sugary ). worked out quite simply 

 according to Mendelian expectation. 



The results as regards colour are indicated in the following 

 table, in which Y denotes strong yellow, y very pale yellow. 



Table 29. 



Cob produced the following grains: — 



Bl. & 

 W. \V. & Y.. Bl. & Bl.&W. W.&Y. Total. 

 & y. (& y).* VV. & y. (& y). 



18 . . — . . 5 . . — . . — . . 4/> 



— . . 37 . . . . — . . 7 . . 44 



7 . . — .. — .. 1 .. — .. 8 



—..—.. 11 .. 9 .. 18 .. 38 



* (& y) := and y in some cases. 



It will be clear from the above that the result as regards 

 yellow and white may be considered to be Mendelian if the 

 pale yellow grains are ranked with the " whites " instead of 

 with the " yellows." A more or less similar state of things is 

 now known to occur in several groups of plants, having been 

 first recorded by Tschermak in the case of kidney beans. In 

 the present case the observations are not sufficient to allow 

 of a complete account of the phenomena. It is not clear, 

 moreover, from which of the various parental strains this 

 character took its origin. 



The actual proportion of pale yellow grains was difficult to 

 determine, but they made up approximately 10 per cent, of the 

 total number in the cobs in which they occurred. In Table 

 30 the pale yellow grains were counted as far as possible 

 with the whites, but they were probably not always truly 

 distinguished from the full yellows, so that the recorded 

 proportion of the latter to whites is only approximately 

 accurate. 



