PARNELL, AYYANGAR, RAMIAH, AND AYYANGAR 199 
A further point of interest, and one of considerable economic importance, 
is that the weight of the grain varies with its shape, the coarser types being 
heavier. This has been recorded in several families of which one, No. 2584, 
for which measurements are given in Table X, gave the following figures :— 
WEIGHT OF 100 GRAINS IN GRAMMES 
Dark gold Medium gold Ripening gold 
Dark gold parent .. 2°145 
Ripening gold parent . 56 Hho 1°755 
F, os re 2-050 
F, ud 2°144 2-075 1:910 : 
The figures for F, represent the averages of 100 plants in each group ; for 
F, and the parents only the single plants concerned were weighed. 
It is not possible at present to give a definite interpretation of the con- 
nection between grain shape and the factor G. That a very close connection 
exists is obvious from the results given. Coarseness may be due to a separate 
factor closely linked with G, or, on the other hand, the factor G itself may be 
the determining factor. Further work, including a search for cross-over 
types, is necessary before a definite statement can be made. 
3. A DWARF HABIT. 
Reference has already been made to asingle-factor variation in grain-shape 
that is connected with a dwarf habit. The variety that introduced this habit 
is strikingly different from ordinary varieties in many important points and 
is the only one of its kind that has been seen. It was obtained, through the 
courtesy of Mr. McKerral, Deputy Director of Agriculture, Lower Burma, 
from the Government Agricultural Station, Hmawbi, Burma. 
The plant is shortened and thickened in all its parts and forms a very stifi 
erect clump. The leaf is coarse and rough, much broader than that of any 
ordinary variety and quite erect. The panicle is very compact, cigar-shaped 
and erect, due to its short stiff branches being closely pressed together. The 
grain is short and rounded. 
This dwarf variety, E. B. 304, has been grown for several seasons at 
Coimbatore and in one of its generations a natural cross appeared. This cross 
was normal in every respect and showed no trace of the dwarf characters 
described above. In F2 the whole of the dwarf characters segregated together 
