15 



are kept under close observation by a team of three workers, 

 Messrs. Garner, Eden and Maskell, who view them respectively 

 from the standpoints of the agriculturist, the ecologist and the 

 plant physiologist. This gives information about the climatic 

 factors but not much about the soil. Certain of the Rothamsted 

 experiments are therefore repeated as precisely as is practicable 

 on a number of farms chosen in different parts of the country to 

 represent important soil and climatic conditions. 



The analytical method has so far been applied only to the 

 wheat and barley data. 



Of the various climatic factors affecting wheat at Rotham- 

 sted, rainfall is the most important, accounting for some 

 30 to 40 per cent, of the whole climatic effect. Broadly 

 speaking, winter rainfall is harmful, especially on land in good 

 condition, spring rain is less harmful or even beneficial, and 

 summer rain is harmful, but the detailed effects depend on the 

 kind of manuring. No less than five different types of action 

 are found on the Broadbalk field. When fertility is low, winter 

 rain has but little bad effect ; when, however, the land is in good 

 condition, each inch of rain above the average in January reduces 

 the harvest by one or two bushels of grain per acre. May rain- 

 fall, on the other hand, may do good, especially on land where 

 the nitrogen supply is large relative to the potash. It further 

 appears that wheat receiving farmyard manure is happier in a 

 dry climate that in a wet one, while certain schemes of artificial 

 manures work out better under wet conditions than others do. 



The investigations of Broadbalk and other Rothamsted data 

 have brought to light several interesting ways in which farmyard 

 manure behaves quite differently from artificial manures. The 

 variation in crop yield from year to year is less with farmyard 

 manure than with artificials ; so also is the variation in the 

 effect of rain at different times of growth ; while the land deterio- 

 rates less under the heavy strain of continuous cropping. On 

 the rotation land clover residues are found to have a steadying 

 effect on the yield of wheat similar to that shown by farmyard 

 manure. 



The advantage of the statistical treatment is that it enables 

 definite expression to be given to these facts, so that the complex 

 field phenomena become reduced to a series of single factor in- 

 vestigations which can be dealt with by the methods of plant 

 physiology. 



FERTILISER INVESTIGATIONS. 



The fertiliser investigations are in the main reported under 

 the various crop headings; reference must, however, be made 

 here to certain general results of considerable interest that have 

 been obtained. 



Notwithstanding the wetness and general bad character of 

 the seasons, especially of 1924, nitrogenous manures exerted their 

 full effect. The average gains from the use of 1 cwt. sulphate 



