ANNUAL REPORT, 1932 9 



Soy Beans. In tlie test of 1932, several of the varieties which gave low yields 

 in previous years were discarded as being unsuitable for forage under conditions 

 here. Some new varieties were included in the test. The highest-yielding 10 

 varieties were: Mukden (new 1932) (2.59 tons hay), Wea, F.P.I. 50536 (new 

 1932), F.P.I. 63242-2 (new 1932), Dunfield, Harbinsoy, Manchu, F. C. 03654-A, 

 Mansoy, F.P.I. 53933-3B (2.33 tons). Rows were 36 inches apart in 1932, 30 

 inches in previous years. 



Field Experiments with Onions. (M. E. Snell and A. B. Beaumont.) 



Fertilizer ratios. The field experiments with lime and fertilizer ratios were 

 continued with no radical changes. This completes three years' records with 

 onion sets, which have proved better adapted to the work than seed onions. The 

 past season, the 4-12-8 (organic) and the 4-12-4 ratios (these figures refer to 

 percentage of NH3, P2O5, and K2O) were the only ones which yielded higher than 

 their 4-8-4 check plots. The 6-8-8, 4-8-8, 4-12-8 (inorganic), and the 2-12-8 ratios 

 were again poorer than their check plots. 



Over the three-year period the 4-12-4, 4-12-8 (organic), and the 4-8-4 (with 

 2-0-0 nitrate top-dressing) gave slightly higher yields than their 4-8-4 check plots. 

 The 6-8-8, 2-12-8, 4-8-8, and 4-12-8 (inorganic) were all inferior to their check 

 plots. The 6-8-4, 8-16-8, and 4-7-4 (with 2-1-0 fish top-dressing) were about 

 equal to their check plots. Application of lime, in excess of three tons, gave an 

 increase in yield of 5 to 10 per cent for the three-year period. When the lime 

 was withheld and the phosphorus doubled with a 4-8-4 and a 4-12-4, the yield 

 was increased about 10 per cent. When 600 pounds per acre or 1200 pounds per 

 acre of superphosphate were supplied in addition to the 4-8-4, and lime was 

 applied annually for five years at the rate of 1000 pounds per acre instead of in 

 three 1-ton applications over a period of seven years, the increase was 21 and 27 

 per cent respectively. However, the same applications of superphosphate applieri 

 with 500 pounds of lime annually, gave but slight increase over the three 1-ton 

 applications without the extra superphosphate. 



Onion set experiment. Last year some preliminary work was done in an attempt 

 to determine methods of getting maximum yields of a desired size of set for plant- 

 ing. Also, the effect of the conditions under which they were grown on the keep- 

 ing quality of the sets was studied. In these tests 50 pounds of seed per acre, 

 500 pounds per acre of a 4-12-8 fertilizer, and seetl distributed over a two-inch- 

 wide row proved the most satisfactory of the rates tried. The most desirable 

 size of set, 1/2 to 3/4 inch, showed least loss in storage from sprouting, decay, and 

 moisture loss. This size sprouted less than the largest size and decayed less 

 than the smallest two sizes. The lighter seedings and the wider rows showed less 

 loss than the heavier seedings and narrower rows. There was little decay where 

 500 pounds per acre of a 4-12-S fertilizer were used. 



During the past season more extensive work was done. Four fertilizer levels 

 were used; namely, 500 pounds per acre of a 0-12-8, 2-12-8, and 4-12-8, and 1000 

 pountls of a 4-12-8. The rates of seeding were 30, 40, and 50 pounds per acre, and 

 the seed was distributed over one-, two-, and three-inch-wide rows with the 500 

 pounds per acre of the 4-12-8 fertilizer; but two-inch rows were used on all other 

 fertilizer treatments. It was again clearly shown that there is a balance between 

 the food supply and the population, which must be maintained in order to produce 

 maximum yields having a high percentage of a desired size. Forty pounds of seed 

 per acre were adequate with 500 pounds of the 2-12-8, but 50 pounds were needed 

 with 500 pounds of the 4-12-8; 1000 pounds per acre of the 4-12-8, even with a high 

 rate of seeding, supplied too much plant food for a good crop of sets. Five hun- 



