534 Bulletin 316 



the 22 years; if planted five days later, in i out of the 22 years; but if 

 planted so as to be up on June i it would have escaped frost in all of the 

 22 seasons and the risk of frost in the spring would have been eliminated 

 entirely. But fall frosts must be considered. Allowing eight days for 

 germination, about 112 days are required to mature com in this climate 

 after it is up. From June i to September 15 is only 107 days, yet with 

 so short a season as this the crop would have been frosted in the fall in 

 2 out of 23 years. Therefore it is evident in this case that the frost risk 

 could not be eliminated wholly, and must be carried at one end of the 

 season or the other. If, during the 22 years, the crop had been planted 

 so as to be up on May 20, thus eliminating the fall risk, the spring risk 

 would have been 5 out of 22, a rather heavy risk; if planted five days 

 later, thus dividing the risk between spring and fall, the crop would have 

 been frosted once in the spring and twice in the fall, a total of 3 times 

 in the 22 years; but if planted late enough to escape the spring frosts, 

 carrying the entire risk at the fall end of the season, it would have 

 been possible to have made 20 crops in the 22 years without any 

 injury from frost whatever. Moreover, since there were no frosts 

 between September 15 and 20, a season of 117 days could have been 

 obtained instead of 112 days without increasing the frost risk at either 

 end of the season. 



For any crop subject to injury from frost in either spring or fall it is 

 possible in this way to determine the schedule that would have produced 

 the best results in the past; and it is reasonable to assume that the crop 

 scheme that was best in the past will prove best for the future also. 



The frost record of a given locality should be of assistance as well in the 

 selection of varieties that will be likely to mature in that locality. For 

 example, the record for Bald wins ville shows that no frost occurred after 

 June I in 18 years and in only 2 years before October i. Thus, counting 

 from the date of planting. May 22, a variety requiring 130 days to mature 

 might be grown with a risk from frost in only 2 out of 18 seasons. At 

 Wedgwood the risk for the same variety would be 5 seasons out of 23. 

 At Shortsville, where no frosts have occurred in the fall before October 

 5 in 12 years, it would seem that a variety requiring 135 days to mature 

 might be grown successfully. 



It is understood that exigencies of soil or weather may at times interfere 

 with any plan that may be devised; but the nearer the approach to the 

 ideal in agricultural practice, which is the adaptation of the crop scheme 

 to the climate, the better will be the results obtained. 



In using the records given to determine the risk of injury by frost to tree 

 fruits, it should be remembered that the records are based on temperatures 

 taken mostly at a height of about four feet above ground'; and for the 



