172 



were with no manure, and with nitrate of soda, acid phosphate, and 

 kainit combined. 



Wheat, test of varieties (pp. 11-13). — Tabulated notes on thirty-six 

 varieties, tested on ninetieth-acre plats, on clay-loam soil, which was 

 *' rather poor." The best results were with Tasmanian lied, l^igger, 

 New Golden, German Emperor, and Michigan Amber. 



From several years' experience with Tasmauiaii Red and Nigger we can especially 

 recommend them for this locality. Both are bearded varieties, very hardy and pro- 

 ductive. Nigger wheat is especially adapted to this region, having originated in Ken- 

 tucky, so far as is known. Diehl-Mediterranean, Mediterranean Hybrid, and Michigan 

 Bronze are the same variety under different names. It is well worth a trial. 



Fulcaster, one of the uewer varieties, Avhile uot doing well with us in this test, 

 can nevertheless be recommended for hardiness and strong growth. As a rule, it is 

 a good yielder comjiared with other varieties. 



Wheat, methods of culture (pp. 13, 14). — Four methods, (1) planting 

 in rows by hand and cultivating with the hoe ; (2) sowing in drills, with 

 no cultivation; (3) sowing in drills with mulching; and (4) sowing 

 broadcast, with harrowing, were compared on 8 fourth-acre plats. The 

 results are stated in a table. The average yield per acre of the two plats 

 in each class was as follows: hoed jdats, 12 bushels; drilled plats, 19^ 

 bushels; mulched i^lats, 18J bushels; harrowed plats, 16^ bushels. 



Sorghum, test of varieties (pp. 14, 15). — Tabulated notes on forty-four 

 varieties. The following varieties were the earliest to ripen, and were 

 fit to cut at about the same time, September 3 : Chinese, Georgia, No. 5, 

 No. 30, Sorghum Saccharatum, White Amber, Dutchess Hybrid, Kan- 

 sas Eed, No. 18, Price's New Hybrid, Waubausee, Whiting's Early. A 

 change of chemist at the station prevented the making of sugar de- 

 terminations, as was intended. 



Clover, experiment with fertilizers (pp. 15, IG). — Various fertilizers 

 (nearly the same as those used in the experiment with oats above re- 

 ferred to) were applied to 8 half-acre plats of light clay-loam land which 

 had been in red clover for two years, and the results compared with 

 those from two unfertilized plats. The amounts of fertilizers and the 

 yield of the first and second crops are given in a table. Eains seriously 

 affected the curing of the present crop. 



If we were to accept the results in this table as bearing on the first crop, we should 

 see that : 



(1) The largest yield comes from unfertilized land. 



(2) The next largest yield comes from a mixture of two thirds phosphate and our* 

 third potash. 



(3) The third best results come from cotton-seed meal. 

 However, experience teaches us two things: 



(1) That unfertililized lands give us the poorest returns. 



(2) That the cottonseed meal could not have been much used as a i)lant food by 

 the first growth, as the plants were already too advanced in growth to have felt the 

 effects of the decomposition of the nieal. 



This is a good example of the general uncertainty of the results of plat tests. Such 

 experiments, to have a signiticaut value, should be permanent, extending over many 

 yeais, under as nearly identical conditions as possible. 



