ON THE " SALT WATER SELECTION " METHOD OF SEEDS. 437 



As may be clearly seen, these results fully confirm those 

 obtained by Ando and show that tlie heaviest and best grains are 

 mainly included in the lots with higher specific gravities. Ida's 

 carefull observations also show that the percentage of husk has more 

 or less influence upon the specific gravity of the grain ; and this 

 seems to confirm our view with regard to the question of the 

 relation of the weight and the specific gravity of a grain with its 

 husk. Ida's results may be tabulated as follows : — 



Sp. gr. 



Average weight 

 of a grain in 

 nic;r. 



Average weight 

 of husk of a grain 



Ratio of hu'sk to 

 grain taking 

 liusk as 



Per cent of husk. 



SORT, JAVA. 



Below I.oo 



17-50 



4.90 



I 



: 3-51 



28.4 



I.oo— 1.05 



24.30 



5.06 



I 



: 4.80 



20,5 



I 05— 1. 10 



25.90 



513 



I 



: 5.00 



20.0 



i.io — 1.15 



27.80 



5.20 



I 



: 5-35 



18.7 



1. 15— 1.20 



28.40 



5.20 



I 



: 546 



18.3 



Over 1.20 



29.00 



5.20 



I 



: 5-77 



17-3 



SORT, VAMADA. 



Below 1 .00 



17.40 



430 



I 



4.05 



24,6 



I.oo— 1.05 



22.40 



430 



I 



5-21 



19.2 



1.05 — I.IO 



24.90 



4.20 



I 



5-93 



16.S 



I.IO— 1. 15 



27.00 



4,20 



I 



6.43 



15-5 



1. 15— 1.20 



29.60 



4-30 



I 



6.88 



14-5 



Over 1.20 



30.90 



4.40 



I : 



7.00 



14.4 



SORT, SI^RATA^L\. 



Below 1.00 



18.30 



3.80 



I 



: 4.82 



20.7 



I.oo— 1.05 



23-30 



4.20 



I 



: 5-55 



18.6 



1.05 — I.IO 



25.80 



4.20 



I 



: 6.14 



16.2 



I.IO— 1. 15 



28.30 



4.60 



I 



6.15 



16.2 



1.15— I 20 



29.40 



4.80 



I 



6.13 



16.2 



Over 1.20 



31-50 



4.80 



I 



6.O5 



15.0 



