'85 



average, from 5 to 10 per cent, less than that obtained 

 by planting whole tubers of medium weight. 



The experiments warrant M. Girard's deductions, 

 though we should not follow them entirely, ;as in 

 growing nearly 200 acres annually we have to use a 

 considerable quantity of cut sets, and our experience 

 does not agree with that founded on these experi- 

 ments. Occasionally on land in harsh condition in 

 dry seasons cut sets have shown missed plants to some 

 material extent, but in average seasons very few occur. 

 Our experience concurs with his in his first note.. In 

 notes 2 and 3 our experience is not similar in degree. 

 In the case of note 2, however, only half the seed was 

 planted. To make the experiment level an equal 

 weight should have been planted, which would have 

 occasioned closer planting on the part of the cut sets. 

 The practice of planting two or three small tubers in 

 one hole, as mentioned in Note 4> does not give so 

 good a result as planting ,them separately. We 

 planted twenty acres of Wonder chats nine inches 

 apart in the rows this year, in preference to planting 

 small ones together, as our experience has shown it is 

 more advantageous. The roots appear to ramify 

 through the whole of the soil more rapidly than when 

 two; or more are planted in one spot. A very similar 

 reason causes farmers to single their turnips and 

 mangels, rather than leave them in bunches of two 

 or three. But, as before stated, the .preparation and 

 condition of the soil have a great influence on the 

 j-elative value of cut and uncut sets and of large and 



