MICHIGAN STATE POMOLOGICAL SOCIETY. 85 



nurseryman, compelled by this demand for variety, must prop- 

 agate many kinds, — must keep a great variety of sorts, or 

 Bhut up shop. And so long as the nurserymen of Michigan 

 are obliged, in order to supply the popular demand, to grow 

 fifty varieties of apples, forty of which are of second grade, 

 just so long will the fruit-growers of Michigan plant fifty 

 varieties, forty of which are of second grade to say the least ; 

 for the nurserymen do not grow their trees to throw them 

 away. Somebody plants them. To strike, then, at the root 

 of this evil, the purchaser must know that such atid such are 

 the best and only kinds to jjlant. This knowledge must be 

 widespread, and accepted by unanimous consent. Such kinds 

 only will then be propagated to any extent. To make any 

 knowledge widespread and unanimous, it must be convincing. 

 To make people generally convinced that a list of the kinds 

 of fruit to be planted, furnished by this Society, for instance, 

 is correct in every particular, is not so easy a matter, where 

 that list is made up from members' opinions, for it is classed 

 by many as simply an opinion, which is not necessarily any 

 more nearly correct than their own, and consequently loses a 

 great degree of its force. But if, in giving to the world our 

 opinion as to the best and most prolific varieties to plant, that 

 opinion is backed by the figures obtained from our tests, which 

 show that such is the case, the matter assumes a difierant 

 asjject at once, and, although this Society possibly might 

 furnish a list to-day that the test of five years might not 

 change in any particular, yet if it was not unanimously 

 accepted it Avould lose a great share of its value, as I think 

 has been clearly shown. 



But to proceed. New varieties of all kinds of fruit are con- 

 stantly being produced by people all over the State, some of 

 them, without doubt, excelling in quality, productiveness, etc., 

 anything heretofore known, while many, very many, are worse 

 than useless, perhaps, while others, again, are of medium 

 worth. Now we must test them all, or a greater portion of 



