EDITORIAL. 705 



often neglected which would add to their information value and 

 assist in their interpretation. Too often the end sought seems to be 

 comparisons measured by ultimate production, rather than the 

 tracing of relations between causes and effects. 



Field experiments Avith agricultural plants, considered in their 

 fundamental aspects, are to large extent studies of the response of 

 these plants to definitely known environmental conditions; and yet 

 how rarely are these responses measured expressly or continuously at 

 successive stages. To determine the response effectually, the condi- 

 tions must be definitely known, and the plans must be so made and 

 the observations so taken that correlations can be attempted. The 

 life processes in the gi'owing plant need to be followed if the effects 

 of the special conditions imposed are to be determined. 



Consider the usual field experiment with fertilizers, or upon the 

 preparation of the land, or culture methods, or date or rate of seed- 

 ing, rotations, and the like. The land is selected and prepared with 

 care, divided into plats with mathematical accuracy, and the differ- 

 ent treatments carried out systematically. Notes are taken from 

 time to time on the general appearance of the plats, often supported 

 by photographs, the date of blooming or fruiting or other stages 

 noted, and when the crops are harvested pro\'ision is made to insure 

 against loss, and the weights or volume are accurately recorded. 

 Analyses may be made of the materials applied or sown, and of the 

 resulting crops. But the gi'owing crop is usually not studied in a 

 way to determine the manner in which it is responding to the special 

 conditions. Rarely, indeed, are any systematic measurements taken 

 of the plants, or attempts made to get at critical stages of growth 

 or the effects of the treatment at such times. The physiological 

 activities of the plants throughout their growth are not followed, and 

 often can not be followed because the field experiments have not been 

 supported by experiments which provide a larger measure of con- 

 trol. The growth under the different treatment can not be compared 

 with the normal growth at various stages, because provision for this 

 is lacking. 



If conditions of normal growth were maintained as a check, it 

 would be possible to judge of the accelerating or retarding effect of 

 each treatment at successive stages, and the adaptability of the plant 

 and its capacity to overcome adverse conditions or to minimize their 

 effects later could be arrived at with considerable accuracy. Fur- 

 thermore, something could be learned of the effects of changes in 

 temperature of the air or soil, or the periodicity of the rainfall and 

 other external phenomena which experiments in the field are subject 

 to. Instead of measuring these, the attempt is usually made to 

 eliminate them by averaging the results of several years. 



