145 



soil, should work near the plants, while deeper cultivation may be had 

 at a distance from the plant which the roots have not reached. 



The form of shovel preferred in the experiments at this station is 

 known as the " Eagle's Claw." It consists of eight small shovels, which 

 are attached to the beams of a two-horse cultivator, four shovels work- 

 ing on each side of the row of cane. The form of these shovels is such 

 that they do not enter the soil deeply ; they thoroughly pulverize all the 

 surface soil and destroy weeds, and work close to the growing plants 

 with little injury to the roots. 



We have alluded to these points because we believe the yield of sugar 

 is often materially lessened by injuring the roots of the canes. Mutila- 

 tion of the cane plants above the surface of the soil is known to pro- 

 duce a lessened yield of sugar, and injuries to the cane plants below 

 the surface doubtless decrease it also. Many cane-growers, as they " lay 

 by " their cane crop, or finish the cultivation and see its deeply and 

 closely cultivated canes free from weeds, do not realize that while de- 

 stroying weeds they nearty destroyed their cane plants, and while 

 working for their canes they were working against them and against 

 the yield of sugar. 



DISTANCE APART AT WHICH CANES SHOULD STAND IN THE ROW. 



This is a subject which has attracted considerable interest among 

 sorghum-growers lately. Mr. Hughes obtained last year the highest 

 yield of sugar per acre ever reported for sorghum. According to his 

 statement, " this was occasioned by carefully planting the hills closer 

 and giving it good attention, together with favorable rains."* As a 

 contribution to the solution of this important question, the following 

 analyses may be recorded. 



Two experiments were made with different plots of cane, both of 

 which had been planted with drill. The planting had been rather 

 uneven, and some rows were much thicker than others. A thick and a 

 thin row in each plot were chosen, the canes counted and cut for a con- 

 siderable distance, which was measured, and the whole run through the 

 mill. The number of canes divided into the length of row cut gave 

 the average distance apart of the canes in the row, and from the weight 

 of the whole sample the yield per acre was calculated. The following 

 table gives the results of the analyses: 



*Bull. No. 17, diem. Div., U. S. Dept. Agriculture, p. 

 H056 Bull. 20 10 



