52 THE DIRECTOR, KXPERIMESTAL FARMS 



as to the supply laid up for it in the seed, the rounp plant is able through its rootlets 

 to begin to feed at once on the abundant food with which it is surrounded. Hence 

 it does not matter so much under these circumstances whether the seed kernel i3 

 plump or shrunken ijrovided the germ is strong and vigorous. We have had m the 

 past several instances where farmers from Manitoba and the Xorth-west have sent 

 samples of small wrinkled, thrivelli-d seed, but with good germinating power, 

 asking advice a:^ to whether such grain should be used for seed. I have invariably 

 advised farmers not to sow such grain. In some instances it has been sown, and I 

 have received samples after harvest of good grain produced from such seed and heavy 

 crops reported. This can be done in other parts of tlie Dominion, but I mention these 

 facts because I think they go to show that we need not expect such good results from 

 the carefid selecting of seeds in the Xorth-west country as we may look for where 

 plant-food is less abundant and other conditions are less favourable. It stands to 

 reason that under average conditions plump seed, whether hand-selected or well 

 screened, is necessary to give the plant a good start, so that it may have its roots well 

 grown from nourishment stored in the seed itself before it is thrown on its own 

 resources — with sue-h a good start its chances of maturing a good crop are much in- 

 creased. 



By Mr. Ileyd : 



Q. Suppose there is a continuous selection from year to year, would not the 

 qualities of the selected seed be a factor you did not experience in the first year ? 



A. AVe have been carrying that on for several years, but, perhaps, we have not 

 pursued it long enough to be able to answer such a question in a decided manner, but 

 my impressiiin is that the selecting of grain for seed or having the seed well cleaned 

 and screened so that the sample may be plump, is a most important thing for the 

 farmer in the eastern parts of Canada, and one which will in the long run well repay 

 him for any extra trouble he may take in this way. 



By Mr. \Yllson : 



Q. I tliink that the choice of plump seed for sowing would recommend itself to the 

 jcomnion sense of everybody ? 

 A. Y,s. 



By Mr. Farquliarson : 



Q. In unselee-ted seed you ha^•e more pickles ? 



A: I beg your pardon. 



Q. With unselected seed you have more pickles or kernels of grain, probably 50 

 per cent. In selecting your seed you simply throw away the small wheat. You would 

 have many more jirains in the bushel of unselected wheat. That may have something 

 to do with the yield? 



A. That may possibly have something to do with it. although I think the reasons 

 I have advanced are sufficient to explain the results. The evidence submitted points 

 to the importance and value of the selecting of grain for seed, it also shows that such 

 selected seed dues not always produce the results expected. Further experiments will 

 be conducted along this line. 



The next point I wish to refer to is in connection with additional work which has 

 been done in the cross fertilization of grain. 



It has been several-times ^ngi;e>ted by members of this committee in years past 

 that it would be a very interesting and important matter to endeavour to obtain 

 crosses between (joose wheat, which i;- a very vigcrrous and productive wheat in On- 

 tario, and the Red Fife. Three years ago. we succeeded in making such crosses, and 

 some additional ones were made last year, so we now liave a series of crosses between the 

 Red Fife and Goose, and Red Fife and Roumanian, a hard, ricy wheat similar to the 

 Goos-e, which has on the average been more productive. We have also succeeded in cross- 



