STERILITIES OF WILD AND CULTIVATED POTATOES. 29 



able to produce fruits in abundance. There appear to be no direct 

 correlations with competition for food between vegetative organs of 

 storage and fruits with seeds which determine or even strongly 

 influence fruit production. 



There remains, however, the question whether correlative con- 

 ditions directly determine or influence either the abscission of flowers 

 or the relative development of the sex organs in flowers, and par-' 

 ticularly of the stamens and pollen. 



The abscission of flower buds and flowers is highly variable and 

 is obviously responsive to environmental influences. Special studies 

 of this condition in the potato by Young {27, p. 17) "suggest rather 

 definitely that moderately cool weather, especially at night, favors 

 the setting of seed and that a gradually falling temperature with a 

 moderate amount of moisture is especially favorable." Young 

 notes that "it is not unusual for a wave of warm weather in early 

 summer to be followed by the nearly or quite complete shedding of 

 the buds and blossoms of the potato." That their environmental 

 conditions operate through internal conditions and correlations is 

 quite obvious. The studies which Kendall (.9) has made with a 

 related genus (Nicotiana) indicate that temperature is an important 

 factor influencing abscission. 



When flowers of healthy potato plants remain attached until they 

 are fully open and pollen is dehiscing, the general condition of stamens 

 and the quantity oi viable pollen appears to be very constant, at least 

 for the varieties grouped in class 1. This statement is based on 

 comparative studies of varieties grown at the New York Botanical 

 Garden and at Presque Isle of early and late plantings of the same 

 variety and of plantings grown with different fertilizers. There 

 may be variations giving different grades of either maleness or 

 femaleness or of both, and such variations may be rather irregular 

 or decidedly cyclic. It would seem that the comparatively short 

 blooming period of a potato plant limits and largely precludes the 

 possibility of such noteworthy changes in sex as are seen in the 

 successive flowers on a plant of Cleome spinosa {22) . 



The influence of various infectious diseases on the quality of pollen 

 has not been carefuUy studied by the writers. In the varietal test 

 plats grown at Presque Isle diseased plants are rogued earh^, and 

 material for adequate comparative tests of pollen from diseased and 

 healthy plants of varieties or seedling strains normally having con- 

 siderable viable pollen (in class 1) was not available. 



It is noteworthy that under conditions of abundant flowering (and 

 under the same conditions) varieties like McCormick, Green Mountain, 

 Australian Blue, and Berrick exhibit characteristic differences in the 

 degree of pollen sterility which are remarkably constant for each 

 and which enable the general grouping into classes 1, 2, 3, and 4 to be 

 made, as indicated in preceding pages. There is also evidence that 

 there is no immediate and direct correlative compensation that 

 causes pollen sterility. Pollen sterility is decidedly an inherent char- 

 acteristic. There is obviously some hereditary basis for the con- 

 dition. 



In an effort to determine the hereditary values of different grades 

 of pollen sterility in potatoes Salaman {18) considers pollen fertility 

 and pollen stenlity rather sharply as contrasted characters and 

 suggests that male sterility is here "a dominant hereditary char- 



