THE OCCURRENCE OF FROST 



351 



the lowest average minimum temperature, the surface averaging 1.7° higher than 

 the 5-inch level and 1.4° lower than the 36-inch level. In a garden on upland 

 the differences were less. Cox summarizes his observations on this point as 

 follows: "The temperature at 2.5 inches averaged lowest, 44.5°[F.], instead of at 

 5 inches, as on the bog, but the difference was very slight between these two 

 elevations — 0.1°. The surface thermometer averaged highest, 45.5° but there 

 was only 1° difference on an average between the two extremes while the average 

 surface reading was 0.6° higher than at 36 inches. The average for the entire 

 season fairly represents the conditions prevailing each month, the highest in 

 each case occurring at the surface and the lowest at 2.5 inches." Table 58, 

 compiled from Cox's report, shows minima for nights selected because of the 

 low temperatures and indicates no substantial variation from his averages.^® 



Table 58. 



-Minimum Temperatures in Open Over Sandy Loam 

 (Degrees Fahrenheit) 



It is evident that these differences are not constant. Some light is 

 thrown on the effect of radiation by data compiled from Greenwich 

 observations showing that a thermometer on grass fully exposed to the 

 sk}^ registered lower than a thermometer suspended 4 feet from the 

 ground :'^^ 



Degrees 

 fahrenheit 



On cloudless nights 9.3 



Half cloudy 7.3 



Principally cloudy 6.8 



Entirely cloudy 3.4 



Influence of Soil. — Reference is made again to Cox's work for data 

 concerning the minimum temperatures over two different soils. Table 

 59 shows minima for selected nights with the average for the month. 

 The difference, striking at the surface, becomes very slight at 3 feet. The 

 differences up to 5 inches are, however, of no little significance to the 

 strawberry grower. They are to be regarded as due to character of the 

 soils, since other conditions were uniform. Incidentally it may be stated 

 that Cox considers it possible for identical atmospheric conditions to 

 cause a light frost in the spring and not in the fall because of the difference 

 in soil temperatures at the two seasons. To the extent that a high day 

 temperature indicates considerable heat furnished the soil, it diminishes 



23 



