18 



More detailed figures are given in the following table: 



Table III. — Showing treatment of and results from plats on holhvorm experiment farm 

 of Department of Agriculture at Wills Point, Tex., 1904- 



Aside from the 3'ields of cotton, the influence of the several treat- 

 ments on the fruiting of the plant was determined by actual counts 

 of all the fruit on 20 plants for each plat, on several different dates 

 during the season. 



Vigorous, well-branched plants were selected in various parts of 

 each plat. These plants were not marked, and consequently a differ- 

 ent set of plants was used each time the counts were made. The fruit- 

 age was determined in this Ava3^five times during the summer, at inter- 

 vals of about sixteen days. The first count was made on June 9, 

 square production having just begun; the last on August 15, when 

 the bolls were beginning to open. 



The prime object in making these records was to determine the rel- 

 ative earliness of fruit production, (1) with King seed, earlv planting, 

 fertilizers, and thorough cultivation; (2) with King seed, early plant- 

 ing, no fertilizer, and thorough cultivation, and (3) with Eowden 

 (native) seed, late planting, no fertilizer, and ordinary cultivation. 

 The records made on plants grown under the conditions named in the 

 last group are used as a basis of comparison. 

 212 



