282 



THE EFFECT OF FOOD 



from a simple recognition of the relation between cause 

 and effect. Growth can only take place at the expense 

 of food material, and unless this is always more than 

 sufficient for the needs of the organism, the rate of 

 growth must be dependent upon it. 



In spite of the importance of changes in feeding as a 

 source of variation, the number of direct and une- 

 quivocal experiments made upon the subject is compara- 

 tively small, for most of them are complicated by simul- 

 taneous changes in other conditions as well. Upon 

 members of the vegetable kingdom, the experiments 

 made by Lawes and Gilbert * at Rothampsted during 

 the last fifty years afford most valuable evidence. 

 These concern the effect of various manures on the 

 growth of barley, wheat, and various leguminous plants. 

 In the accompanying table are given the average 



amounts of barley grain (in bushels per acre) obtained 

 each year from soils treated in various ways. These 

 observations were carried on for forty years in succes- 

 sion (1852-91) upon the same land, and so represent 

 strictly average results, from which errors due to varia- 



* " The Rothampsted Experiments," p. 78, Edinburgh and Lon- 

 don, 1895. 



