4 MASS. EXPERIMENT STATION BULLETIN 195. 



any case the factor of error would be identical from year to year, and the 

 general comparative results may be considered trustworthy. 



The yield has fallen off seriously only in 1915, and to a lesser extent in 

 1917; but in spite of these low yields the average jaeld of tobacco in the 

 period 1900-18 has been above the average of the forty-year period 

 1870-1910. 



There is no justification for the statement that the yield of tobacco 

 on Massachusetts fields has been decreasing graduaUy for the past ten 

 years; but, on the contrary, in spite of the exceedinglj^ low yield of 1915 



Fig. 1. — -Average yield per acre of Massachusetts tobacco from 1£01 to 191.S. inclusive. 

 The horizontal dash line indicates the average yield from 1870 to 1910. 



the calculated average is being maintained. If, however, we consider 

 only the yields of 1915, 1916, 1917, and 1918, it is true that a yield below 

 the average will be found ; but some of these seasons have been admittedly 

 unfavorable for tobacco, from the meteorological standpoint, and high 

 yields could not be expected. Also it is incorrect, or at least misleading, 

 to base a statement of general yield on so few j^ears' data. The same con- 

 ditions meteorologically have in the past produced almost identical results, 

 as will later be pointed out. 



Weather Factor ix Tobacco Growtxg. 



In general, it may be stated that the first half or more of the growing 

 season of 1916 was decidedly unfavorable to the growth of tobacco. Con- 

 ditions improved from shortly after mid-season until the crop was har- 

 vested, and a rapid and apparently satisfactory growth was made. The 

 leaf was of good size and color, but although seemingly in good condition, 

 was inclined — as later developments proved — to run rather light in 



