Cornell University, 

 Ithaca, October 12, 1895. 



The Honorable Commissioner of Agriculture^ Albany : 



Sir : The one subject which is uppermost in the minds of the 

 fruit-growers of western New York is the cause of the failures of 

 the apple orchards to bear. There are two methods of investigating 

 the subject. One method aims to collect data from the orchards 

 themselves, from every condition and location in which they are 

 grown, and to reflect upon the mass of observation and experience 

 which is thus acquired. This method is essentially one of general- 

 ization, and it is safe only when the student brings to. his aid an 

 extended series of facts, and when he considers them with judicial 

 deliberation. Its chief fault is tlie danger that the student may 

 overlook certain minor facts, and that his generalizations may be 

 applied to too many diverse conditions. Yet, in most subjects 

 touching the general economy of agriculture, such as the manage- 

 ment of land and crops and business, it is a most promising method 

 of research. 



The other method consists of a minute examination of a few facts 

 or a small field, and the drawing of such conclusions from them as 

 seem to apply to broader areas. It is essentially a specialization, 

 and it is safe only when the facts under observation are positively 

 understood, and when applications are made by the student with 

 the greatest caution. Its fundamental details are so exact, and it 

 presents such an array of figures and facts, that it at once enlists the 

 sympathies of the reader and convinces him ; therefore, it is gen- 

 erally considered to be the proper type of scientific inquiry. Its chief 

 fault is the danger that conclusions which are undoubtedly true for 

 a narrow field may be held to be equally true for a wide one. 



When the Nixon bill was passed a year and a half ago, and we 

 were asked to study the horticulture of western New York, it was 

 conceived that an inquiry into the lessening productiveness of 

 orchards was more needed than any other single investigation. 

 From that time until now the subject has been constantly under 

 consideration, and two bulletins (72 and 84) have already been 

 devoted to some aspects of it. The question has been divided into 



38 



