12 MASS. EXPERIMENT STATION BULLETIN 195. 



Excessive drought, combined with low or high temperatures, will also 

 reduce the yield, but not usually to so great an extent as in the case of 

 excessive rainfall. Good examples of the effects of drought on yield are to 

 be found in 1907, and to a lesser extent in 1913. 



Subnormal rainfall with subnormal temperatures does not reduce the 

 yield to any great extent, providing they are not excessive (see 1909, 

 1910, 1912). The quahty of the leaf is apt to suffer somewhat, however, 

 in such cases. 



A careful examination of the data leads us to the conclusion that for 

 the development of our best crops we must have a season with normal or 

 slightly subnormal rainfall, fairly well distributed, together with practi- 

 cally normal or supernormal temperatures. 



Another factor of great importance is the distribution of the rainfall. 

 For the best results the rainfall should be more or less evenly distributed 

 over the growing period, not all in any one month or on a very few days 

 of the month. The season of 1917 shows very well the effects of an unequal 

 distribution of rainfall which, while averaging little above normal, reduced 

 the yield. Here a very wet June, followed by a very dry July, brought 

 the crop along to the danger point, and then the excessive rainfall of 

 August, while helping growth, did not permit of a normal maturity, and 

 as a result we get a subnormal yield. 



As a result of the examination of these plates we may state that, in 

 general, rainfall is the major limiting factor of growth {and this necessarily 

 includes soil moisture), together with temperature. 



Excessive seasonal rainfall is invariably followed by a reduction in yield 

 independent of temperature. 



Subnormal rainfall, wheyi accompanied by temperatures excessively above 

 normal, reduces the yield. 



Subnormal rainfall, when accompanied by subnormal temperatures, does 

 not apparently reduce the yield to any extent unless the rainfall is very much 

 beloiv normal. 



In other words, there are apparently certain well-defined limits between 

 which we may' expect to get a normal yield, or better. A total rainfall for 

 the season of from 6| to 12 inches, if fairly well distributed, will give a 

 good crop. Less than this or greater will give a yield below normal. 



These conclusions are general and seasonal, and do not take into con- 

 sideration other factors, such as effect of precipitation on soil temperature, 

 fertilizer applied, parasitism of disease-producing organisms, type of soil, 

 and method of culture. All these factors do play a more or less important 

 role concurrently, but fundamentally rainfall and temperature are to be 

 considered the limiting factors in production. The other factors men- 

 tioned are decidedly more local in their action, and are often possible 

 of correction. They are what might be termed "individualistic," while 

 the others are "communistic." 



The effects, in any one season, of the rainfall and temperature as related 

 to growth also make a very interesting study, and any one interested 



