TOBACCO INVESTIGATIONS. 13 



would do well to read Bulletin No. 39 of the Bureau of Soils, United 

 States Department of Agriculture, in which is reported a season's study 

 of these factors at Tarriffville, Conn. It would be out of place to detail 

 them here. 



In conclusion, it might be well to emphasize again that the general 

 reduction of crop yield per acre is invariably associated with the seasonal 

 rainfall and temperature, and not to any "running out" of the land (which 

 was amply shown by the fact that the 1918 crop, according to returns, 

 was much improved over those of 1915 and 1916, and was practically 

 normal in spite of the July drought), due, primarily, to a specific wide- 

 spread soil trouble. 



This conclusion does not in the least minimize the fact that in many locali- 

 ties there is undoubtedly trouble due to improper fertilization, methods of 

 culture, and disease-producing organisms; but these are specific problems, 

 and not susceptible to general analytical consideration except in the group 

 to which they belong. 



They 7nay assume major importance in seasons presenting abnormal 

 meteorological conditions, and in such seasons are often held responsible 

 for all reductions in yield. This should not be done, however. 



Soil Relations. 



During the season of 1916 many fields were observed where the uht 

 favorable condition of the crop could not be accounted for by the presence 

 of disease. On these fields the leaf was apparently normal in size or, in 

 some areas, small and undeveloped; but it was thin and papery, and did 

 not show the weight and quality of normal leaf grown on the same type 

 of soil. In these cases not attributable to parasitic organisms we must 

 look to ah unbalanced physiological relationship between the plant and 

 the soil, and this, naturally, first involves a study of the soil composition, 

 reaction, and methods of fertilization. 



Many theories have been advanced to explain this particular type of 

 non-productiveness of tobacco soils, such as overfertihzation, under fer- 

 tilization, excess of soluble salines, toxic substances formed in the soil 

 by the interaction of certain fertilizing constituents, injurious fertilizer 

 constituents, the accumulation in the soil of toxic excretions from the 

 roots of the tobacco plant, lack of potash, and a variety of other causes 

 any one or more of which may possibly, under certain conditions, furnish 

 the correct explanation. In general, however, it has been found difficult 

 to ascribe the condition of the crop on such fields to any one factor with 

 any degree of certainty, and it would appear that careful investigation is 

 necessary to clear up some of these questions. In aU probability no single 

 factor is responsible. 



It has generally been supposed that tobacco thrives best on a soil 

 approaching neutrality, but our observations lead us to believe that this 

 may not be entirely true, and that, possibly, some of our soils are too 



