158 STORRS AGRICULTURAL EXPERIMENT STATION. 



upon the total yield of the crop. The particular importance of 

 the results is the effect of the nitrogen of the fertilizer upon 

 the percentage of nitrogen, and hence of protein in the crop. 

 This is indicated by a comparison of the averages of the results 

 from the pots with the different quantities of nitrogen, which 

 are given in the last column of each table. 



According to the usual custom, the percentage of protein is 

 estimated by multiplying the percentage of nitrogen by the 

 factor 6.25, on the assumption that all the nitrogen determined 

 represents protein containing 16 per cent, of nitrogen. This 

 involves at least two sources of error. The different forms of 

 proteid vary in the proportions of nitrogen they contain, so 

 that the factor that should be used may range between 5.8 and 

 6.4. Furthermore, a part of the nitrogen in the crop may be 

 present not as protein but in the form of nitrates, which have 

 no food value, yet any nitrogen that would come from them in 

 the determination would be estimated as protein by this method. 

 In the experiments here reported, however, the error due to 

 nitrates would probably be inappreciable. In the experiments 

 of I goo the amounts of nitrogen in the fertilizer were much 

 larger than those in 1901, but in tests for nitratcs'^^ in the crops 

 of 1900 only very small quantities were found, and in only a 

 part of the samples. The amount of nitrogen present in the 

 form of nitrates in the crops of 1901 would therefore probably 

 be negligible if there were any. In any case, the error due to 

 the presence of the nitrates would be much less than that due 

 to the use of wrong factors. 



For practical purposes, however, the usual method of esti- 

 mating protein is perhaps sufficiently accurate; and in order 

 that the results of these experiments may be directly compara- 

 ble with those previously reported, it seems best to follow the 

 same method here that has been used before. 



DISCUSSION OF THE EXPERIMENTS OF 19OI. 



The experiments of 1901 include one series of pots with 

 meadow fescue grass, one with Hungarian grass, one with 

 barnyard millet, and one with soy beans. Each experiment 

 comprised four groups of pots; in the experiment with meadow 



* Rept. Storrs Station 1900, p. S3. 



