SINGLE SEED SELECTION. 



5 



result given is the average of a number of seedlings, the experiments 

 seem to be fairly conclusive on the point. 



The results are quoted in his classical paper Table V., p. 116, and 

 are as follows : — 



The average height above ground of the four series of large seeds of 

 the same species was 7'9 cm. and the weight 87 grams; seedlings of 

 medium seed were 7'1 cm. in height and weighed 46 grams. The off- 

 spring of the small seeds averaged 6'4 cm. in height and 43*1 grams in 

 weight. 



The difference in individual series was even more pronounced, but 

 these average results are most instructive. 



Professor Zavitz, of the Guelph College, Ontario, has also achieved 

 extraordinary success by the method of single seed selection. 



In one well-known experiment he selected one single grain of 

 Joanette oat. This seed was sown in 1903. The whole of the grain 

 from it was again sown, and with the result that in 1905 this single 

 seed's descendants produced a harvest of 100 bushels of grain and 

 8,748 pounds of straw! 



There is also one point with regard to this method of single seed 

 selection which should appeal to everyone. Most researches in plant- 

 breeding involve enormous expense, elaborate apparatus, and specially 

 trained experimenters. 



But single seed selection is within the reach of every gardener or 

 farmer; the greatest possible care must, of course, be exercised, but 

 that is really all that is necessary. 



Some small experiments of my own were undertaken in order to 

 test whether there could be, as one seemed to gather from the evidence, 

 any marked result produced by the selection of seed. 



The theory had not really much chance in this case, for the seeds 

 with which I worked were taken from ordinary commercial packets 

 supplied by one of the very best seedsmen in Britain. 



All these seeds were therefore good and selected and all of the same 

 variety. 



I simply picked over a packet and selected the ten best and largest 

 seeds, the ten smallest, and ten which seemed to be of about average 

 size. The differences in size were not really very marked, and I was 

 not at all hopeful as to the results being of any interest. In the 

 experiment with Broad Beans, I got from A (the ten largest seeds) 77 

 beans, from the small seed 0 only 48. The weight of the whole crop 

 A was also- 80 per cent, heavier. With Cauliflowers the A plants 

 were unfortunately destroyed by an accident, but the difference between 

 the B plants and C plants was very marked indeed. The size of the 

 heads from medium seed was obviously much greater and they were 

 much more symmetrical than those from C. With Beetroot the average 

 weight of 7 B plants (medium) was 2' 6 oz. ; that of 5 C plants onl}* 

 1*7 oz. W^ith Turnips I could distinguish no particular difference be 

 tween B and C rows. 



These small experiments convinced me that even with selected 



