AGRICULTURAL DEPARTMENTS AND RESEARCH 67 



soils, manures, etc., can be carried out by the advanced 

 students. In the Entomological Section, in a similar 

 manner, insect breeding, section cutting of insects for 

 microscopic examination in connection with diseases and 

 similar work can be done by the students of entomo- 

 logy. The plant breeder's work, which largely consists of 

 the examination of the economic value of the plants he has 

 produced by hybridization and selection, and the collec- 

 tion of detailed information regarding the behaviour of 

 different strains, must be assisted in the field experiments 

 by a practical agriculturist. It is a handicap to the 

 efficiency of his own work if he is obliged to arrange the 

 ordinary work of preparation of the soil and cultivation 

 of the crop which is being experimented with under field 

 conditions. It becomes necessary, therefore, that he 

 should have a farm superintendent working in close co- 

 operation with him, and in such a manner that the proper 

 utilization of his suggestions, etc., may be ensured. 

 Valuable results have doubtless been lost in the endeavour 

 to employ the scientific investigator for the practical 

 application of his theories. The students in the Botanical 

 Section should conduct all the germination and other 

 special tests and assist in the supervision of flower count- 

 ing and other field operations. 



In every country it is important that the occupation of 

 the research agriculturist should be dissociated as far as 

 possible from purely administrative work, although, in 

 order to ensure the best results and to comprehend the 

 requirements of agriculture, the administrative head of 

 the department should be acquainted with scientific 

 agriculture in all its branches. 



In conjunction with scientific agricultural work, it is 

 necessary that facilities for the publication of results and 

 recommendations should be supplied. An agricultural 

 journal is a necessity, and should not only act as a means 

 of distributing knowledge, but should serve as a record 

 of progress. It is unnecessary that such a publication 

 should appear at fixed periods; in fact, it is better that it 

 should not be designed to do so, as the limitation of time 

 in the preparation of a contribution may detract from 

 its completeness and consequent value. 



