New York Weather Bureau. 469 



The Great Lake region is very fairly- represented in table 13 

 by five stations well distributed over the regio'n. It will be seen 

 that the conditions are here very favorable as regards the occur- 

 rence of froist, and especially so in the section of the Chatauqua 

 grape belt represented approximately by the record of Erie, Pa. 

 The following information concerning this region is given by 

 Mr. S. S. Crissey of Fredonia: 



" The first killing frost of autumn for the past twenty-five 

 years has not occurred until the last week in October or after 

 October 25th. The one notable exception was the autumn of 

 1891, when it occurred on October 12th. In several years it has 

 noit occurred until in November. I now speak of a frost or freeze 

 of sufficient severity to injure ripe grapes exposed on the vines. 

 I have had a Concioird vineyard in bearing since 1869, and have 

 never (until last year) had a pound injured in autumn." 



"As to the danger from spring frosts: The vine begins to leaf 

 out here May 10th, and we rarely have a severe frost after that 

 time. I have never suffered in my locality, which has a high 

 airy exposure; but some vineyards unfavorably located were 

 injured in 1889 and 1891 by frosts about May 17th. Where lands 

 lie facing the lake or with a northern slope they rarely, or almost 

 never, suffer from frosts after May 15th. Of course, froists before 

 May 10th would d'oi no injury." » ♦ * 



" The most serious injury to vineyards in this section during 

 the past twenty-five j^eans occurred in two cases when winter 

 came on with the ground in unusually dry condition, giving a 

 ' dry freeze.' * * * To guard against this condition, which 

 occurs about one year in ten, vineyardists in the dryer localities 

 have abandoned fall plowing of their ground." 



Mr. J. Kyckman of Brockton furnishes statements substan- 

 tially similar to the above, adding that the total extent of grape 

 vine3^^rds in the Chautauqua belt is now about 30,000 acres. 



The Central Plateau. — The early dates of frosts at Coopers- 

 town and Humphrey, as compared with those of other stations 

 of the table, are explained by the high elevation of the general 

 land surface in their vicinity, which is also intersected by deep 



