40 A.GRICULTUBAL ALvxUAL 



Special note is made of the dates of the last killing frost in 

 spring- and the first in fall. If for any reason the observer fails 

 to note the dates of frosts, the date of the last freezing tempera- 

 ture, 32 degrees, in spring, and the date of the first freezing 

 temperature in fall, are regarded as the dates of the last and first 

 killing frosts for the respective seasons. 



The accompanying charts (Figs 10, 11, and 12) showing the 

 average date of the last killing frost in spring, the first in fall, ami 

 the average number of days between frosts, are based on records 

 made at aboul let) places in the state. These charts, for the most 

 part are self-explanatory. It may be pointed out, however that 

 the avt rage dates of frosts, as well as the average number of days 

 between frosts, shown by these charts, are simply arithmetical 

 averages. For example, the average date of the last killing frost 

 in spring for the region along the shore of Lake Ontario falls on 

 May 1. which means that in half the years killing frost occurred 

 before May 1, and in half the years after May 1. It is obvious, 

 therefore, that at the time of the average date as shown by the 

 chart the risk of frost is 50 per cent — too great for tender crops. 

 In spring the risk of frost decreases rapidly after the average date, 

 falling to perhaps 20 per cent at the end of five days and practi- 

 cally disappearing at the end of ten days after the average date. 

 In the fall the frost risk increases rapidly as the average date of 

 the first fall frost approaches, becoming again 50 per cent at the 

 time of the average date. 



With respect to the average length of the season between frosts 

 it should be noted also that there are perhaps ten days at either 

 end of the season when frosts are likely to occur, and that the 

 season of practical immunity from frost is about twenty days 

 -diorter that the average as shown by the chart. 



PRECIPITATION 

 PRECIPITATIC/X RECORDS 



The accompanying charts, showing the average annual and sea- 

 sonal precipitation, are based on observations made daily at about 

 200 places in the state. The records- vary in length and also in 

 the years covered, but no records covering a period of less than 



