junei6, J919 Relation of Weather to Fruitfulness in Plum 123 



greater germination after five days, when pollen was kept in saturated 

 air in a refrigerator (the temperature is not given), than under the same 

 conditions at room temperature. This being the case, the cooler tem- 

 peratures usually accompanying prolonged rains would be more favor- 

 able to a higher percentage of germination than higher temperatures. 

 From Plate 15 it will be seen that each season the minimum tempera- 

 ture falls below the lower limit of tube growth a number of times and 

 occasionally the lower limit of tube growth is even approached by the 

 maximum temperature. It is probable that the temperature influence 

 on tube growth would be similar to that on tube formation. 



The slow pollen-tube extension found under greenhouse conditions 

 serves as a basis for estimating what can be expected during periods of 

 low spring temperatures. That low temperatures have a much greater 

 influence some seasons than others is clearly shown by the extent the 

 minimum-temperature curve extends below the line of no tube growth 

 (5) drawn through each graph (Plate 15) at 40° F. The tem.perature 

 factor, therefore, has an important bearing upon the extent to which 

 fertilization fails to take place some seasons. While cool weather re- 

 tards tube growth, it does not appear to change rhaterially the time of 

 abscission of the style. 



RELATION BETWEEN THE WEATHER AT BLOOMING AND THE 



SETTING OF FRUIT 



With the foregoing analysis of weather in mind, it now remains to be 

 seen whether there is any correlation between the v/eather conditions 

 prevalent at bloom and the setting of fruit. While an ample set of fruit 

 does not certainly insure a full crop, a full crop can not be obtained 

 unless there is a set up to a certain point. In this way the weather 

 determines the possibility of a crop. 



During the years 1915, 191 6, and 191 7 there was a light set and a 

 light crop of plums at the Fruit-Breeding Farm. An inspection of 

 Plate 15 shows that different weather combinations occurred during 

 each of the three years. In 191 5, the outstanding features are the 

 frequent rains during bloom and the low-temperature period for one 

 week following. This single factor, according to the work of Goff (5) 

 on the temperature limits of tube growth, would make fertilization un- 

 certain, but it will be noted that following the cloudy, rainy weather of 

 the first four days of bloom there were two days of unusually windy 

 weather which interfered with bee flight at a critical time, and hence 

 rendered ample pollination uncertain. The following year, 191 6, bloom 

 was nearly a month later and was accompanied by a period of unusually 

 high temperature which extended to the period of tube growth. This 

 alone would have been very favorable to pollination, but during early 

 bloom there were two unusually heavy rains and five lighter ones. 

 Moreover, aside from actual injury to the bloom during such rains as 



