358 Journal of Agricultural Research voi. v, no. 9 



The plants were watered only a few times, during an occasional dry 

 spell, and made an excellent growth. The growing period was from 

 June 15 to October 16, during which time the weather conditions were 

 fairly uniform, with high temperature and humidity. The average 

 monthly mean temperature ranged from 77.2° in June to 79° F. in 

 October. The monthly precipitations from June to September were, 

 respectively, 10.90, 11.98, 11.67, and 8.22 inches. There were, however, 

 some dry spells of a week or 10 days that apparently affected the plants; 

 note of this is made below. 



At 18 days the plants were thinned from 40 to 29 in each cylinder, at 

 which number they were kept during growth. The 11 plants removed 

 from each cylinder at this time served for the i8-day-old sample, while 

 for the 26-day-old sample 6 cylinders were cut; for the 48-day-old sam- 

 ple, 5 cylinders; and for the succeeding samples, 4 cylinders each. As it 

 was impossible to remove the roots completely from the heavy clay soil, 

 the weight of the roots is not recorded. The roots were removed, how- 

 ever, as completely as possible for analysis. 



In preparing the samples for analysis each leaf and stalk was washed 

 individually immediately after cutting to guard against loss of mineral 

 matter by leaching. Under such conditions there was probably a cer- 

 tain loss of mineral matter from withered leaves, but no appreciable loss 

 from the green leaves. However, this is practically of Uttle importance, 

 as the conditions of washing, while thorough, were no more severe than 

 those to which the plant would be subjected by rainfall. Even digesting 

 the leaves in cold water for 15 minutes extracted Httle mineral matter 

 from green leaves. Forty-five gm. of green rice leaves previously washed 

 on the plant were stirred up with i Hter of distilled water. The water on 

 evaporation yielded a residue of 0.008 gm. of mineral matter, part of 

 which was due to minute leaf hairs broken off in the stirring; 9 gm. of 

 withered leaves soaked for 15 minutes in 500 c. c. of water left a residue 

 of 0.057 §™- of mineral matter. 



The analytical methods employed were essentially those of the Asso- 

 ciation of Official Agricultural Chemists, ^ with a few exceptions. Prepa- 

 ration of the ash was by the optional method, igniting over a very low 

 flame without calcium acetate and leaching when necessary. Iron was 

 determined colorimetrically with potassium thiocyanate, this method 

 being preferable to titration with potassium permanganate for the small 



amounts present. 



ANALYTICAL RESULTS 



In Table I are given the data on the weight and composition of a single 

 plant wth respect to withered leaves, etc., at each period of sampling. 

 The weights of the plants were, of course, accurately determined, the 

 probable error of the weights and percentages of dry matter merely show- 



1 Wiley, H. W., et al. OflScial and provisional methods of analysis. Association of Official Agricultural 

 Chemists. U. S. Dept. Agr. Bur. Chem. Bui. 107 (rev.), 272 p., 13 fig. 1908. 



