138 BOARD OF AGRICULTURE. 



It will be noticed that both starch and protein decrease quite rap- 

 idly in the seed tuber from the time of plaiitin;^ up to the time of 

 harvesting. The sugar increases for a time but finally almost dis- 

 appears. The inference to be drawn from these analyses is that the 

 starch and protein are used up in supporting the growth of the 

 young plants springing from the seed tuber. It appears also that at 

 least a portion of the starch is converted into sugar before being 

 transferred to the growing plants. 



Microscopic examination of the seed potato during the growing 

 season by the same investigators and others corroborates the evidence 

 given b}' chemical analysis. 



The experiment suggests the propriety of using large potatoes for 

 seed, in order that the 3'oung plants may be well supplied with nour- 

 ishment during the earl}' stages of growth. 



A large amount of experimental work has been carried out, both 

 b}' practical farmers and scientific men, to settle this point. And yet 

 the question comes up for discussion nearl}' every year in our agri- 

 cultural papers, together with the question of the advisability of 

 planting whole or cut potatoes. 



The most extensive and carefully conducted experiments bearing 

 on these points that have come to my notice are those b}' George 

 Maw reported in the Journal of the Royal Agricultural Society of 

 P^ngland for the year 1867. In a series of 39 trials, 25 were in favor 

 of large sets while 14 were in favor of small sets for seed. The 

 gains in the 25 comparisons in which the large sets gave the greater 

 3'ield were sufficient to counterbalance the gains of the 14 trials in 

 which the small sets gave the greater yield and leave a net balance 

 of 1 ton 10 cwt. per acre in favor of the large sets for the 39 trials. 



Mr. Maw sums up the results of his experiments as follows : 



Firstly. Every increase in the size of the sets from 1 oz. up to 8 

 ozs. in weight, produces an increase in crop much greater than the 

 additional weight of the sets planted. The net profit over and above 

 the extra weight of sets in planting 4 oz. sets in lieu of 1 oz. sets 

 amounted on the whole series of experiments to between 3 and 4 

 tons per acre ; and the further profit on the increase of the size of 

 the set from 4 ozs. to 8 ozs. averaged about 5 tons per acre ; all of 

 the intermediate steps partaking proportionally of the increase. 



Secondly. The advantage in favor of large sets is more marked 

 in late than early varieties. 



Thirdly. In the use of small sets of from 1 oz. to 3 ozs. in weight, 

 a larger balance over and above the sets was obtained by planting 



