tions and cleanings necessanl}\jieglected during the War have all 

 been completed. 



The most important ])art of llic reconstruction has been the 

 reorganising of the Avork of the Station so as to bring it into touch 

 with modern conditions of agriculture on the one side and of science 

 on the other. The purpose of the Station is to gain precise know- 

 ledge of soils, fertilisers, and the growing plant in health and 

 disease, and then to put this knowledge into such a form that 

 experts can use it. The work of the Station falls into two great 

 divisions — the soil and the healthy plant ; and the insects, fungi and 

 other agencies disturbing the healthy relationships and causing 

 disease. The two divisions are linked up in many ways, and every 

 effort is made to find fresh relations between them. If farmers are 

 ever to avoid the very serious losses they now suffer from plant 

 diseases and ])ests, it will be by prevention rather than by cure. 



The method adopted is to start from the farm and work to the 

 laboratory, or vice versa. There are four great divisions in the 

 laboratory — the biological, chemical, physical and statistical — 

 which may be regarded as the ])illars on which the whole structure 

 rests. But the method of investigation differs from that of an 

 ordinary scientific laboratory where the problem is usually narrowed 

 down so closely that only one factor is concerned. On the farm 

 such narrowing is impossible ; many factors may operate and elimi- 

 nation results in conditions so artificial as to render the enquiry 

 meaningless. In place, therefore, of the ordinary single factor 

 method of the scientific laboratory, liberal use is made of statistical 

 methods which allow the investigation of cases Avhere several 

 factors vary simultaneously. Thus in the crop investigations a 

 large number of field observations are made ; these are then treated 

 statistically to ascertain the varying degrees to which they are 

 related to other factors — such as rainfall, temperature, etc. — and 

 to indicate the probable nature of the relationships. Thus the 

 complex problem becomes reduced to a number of simpler ones 

 susceptible of laboratory investigation. 



It is confidently anticipated that this method will prove effec- 

 tive in bringing the full help of science to bear on the farmers' 

 problems. 



REPORT ON THE WORK DONE 



DURING 1918, 1919, and 1920. 



THE funclit>n of the Rotliamsted Exi>enmental Slatit^n is to 

 gain exact information about soils and the growth of crops 

 in health and disease. This information is indispensable 

 to the teacher, indeed without a basis of precise knowledge no 

 system of agricultural education could possibly stand: it is needed 

 also by the advisory experts and by the expert farmer \\ho wishes 

 to improve on current good practice and secure better results than 

 his predecessors. It is, however, essential that the information 

 gained should be as correct as possible, and consequently every 

 precaution must be taken to guard against wrong results. Wrong 

 information has been responsible for many costly errors in the 

 past : the deep drainage <^f llic 'fifties and 'sixties, the burying of 



