266 FUNDAMENTALS OF FRUIT PRODUCTION 



highest and the lowest rainfalls for August, September and October, the 

 highest mean temperatures for August, September and October and 

 the lowest mean temperatures from October to February are thus distin- 

 guished and the minimum temperatures for each month indicated. 

 The winters of 1895-1896, 1903-1904 and 1917-1918 may be considered 

 the seasons of greatest winter injury for this section in recent years. 

 It is evident at once that the rainfall preceding the winter of 1895-1896 

 was the lowest of any season reported; that preceding the 1903-1904 

 season was the highest and that preceding the 1917-1918 winter was very 

 close to an average. The three destructive winters were, then, preceded 

 by both extremes and an average rainfall. Though the rainfall for the 

 months considered was in 1897 only 0.02 inch more than that of 1896, no 

 serious damage was reported; though the rainfall for these months in 

 1915 was only 0.58 inch less than that of the same period in 1903-1904, 

 no widespread damage followed. 



The seasons with highest average temperatures for the last of the grow- 

 ing season, 1900 and 1906, were not followed by the greatest destruction. 

 The lowest monthly temperatures for October, November, January and 

 February came in years not distinguished for greatest winter-killing. 

 Only in December did the lowest monthly temperature occur in a winter 

 of extensive injury. The absolute minima for October, for November 

 and for January fall outside the years of greatest damage. 



Reviewing by seasons: the 1895-1896 winter shows extreme condi- 

 tions (heavy type) in low rainfall and low late winter temperatures; the 

 most noteworthy divergence of the 1903-1904 winter was the heavy 

 rainfall, while for 1917-1918 the high and low monthly precipitations 

 combined to make an average rainfall and the noteworthy features for 

 that winter were the low temperatures of November and December. Not 

 far from Geneva, at Ithaca, Bailey* wrote of the 1895-1896 winter, "the 

 phenomenal injury wrought by last winter was probably not wholly the 

 result of low temperature. The drought of the last summer and fall no 

 doubt augmented the injury." In reviewing the winter of 1903-1904, 

 for states east of the Mississippi, Stockman^** stated that the severity 

 "was not due to occurrence of very low minimum temperatures but to 

 the number and succession of days whose mean temperatures continued 

 below the normal ... At only two stations having 25 years or more of 

 records was the record of lowest temperatures broken. No record of 

 minimum temperatures at a regular Weather Bureau station was broken 

 during December and February." 



It seems, then, that an extreme of any one feature of the climate is 

 not of itself likely to cause widespread injury, but that injury depends 

 considerably on a combination of accentuated, rather than on isolated 

 extreme, conditions. Thus 1895-1896 may be described as in many ways 

 a characteristic Dakota winter in its drying out effects; in 1903-1904 



