March 26, 1920] 



SCIENCE 



325 



reading of the great droughts, recently ei^^'tA^ 

 which were at their worst in west Texas and 

 the northern Great Plains. The flood of warm, 

 moist air from the Gulf is variable in size and 

 duration. These variations are felt most near 

 its western and northwestern limits, where 

 farmers have learned to look on partial crop 

 failures as normal. This variability, which is 

 the most important aspect of rainfall, aside 

 from the average amount, is clearly brought 

 out by Mr. Kincer in a number of graphs and 

 maps. In drought years as well as in years of 

 plenty, farmers are inclined to believe in stor- 

 ies of progressively decreasing or increasing 

 rainfall: comparisons of rainfall averages by 

 successive 20-year periods show, however, that 

 in this region there is no perceptible progres- 

 sive change in rainfall. 



In years of decreasing rainfall, real-estate 

 agents for the semi -arid lands of western Kan- 

 sas explain to prospective buyers that although 

 the total rainfall is decreasing, the decrease is 

 mostly confined to the washing and flooding 

 downpours, and that the proportion of rains of 

 beneficial amount is increasing. They are dis- 

 cussing another essential element which must 

 be considered in comprehensive rainfall dis- 

 cussions. Mr. Kincer presents maps showing 

 the average annual number of days with pre- 

 cipitation 0.01 to 0.25 inch, 0.26 to 1.00 inch, 

 and 2.00 inches or more. Further details of 

 rainfall intensity are given on maps showing 

 the average annual number of days with pre- 

 cipitation more than 1.00 inch in an hour, and 

 the maximum precipitation in 24 hours. Two 

 more maps which might be called " drouth 

 maps " show the percentage of years with 30 

 consecutive days or more without 0.25 inich of 

 rainfall in twenty-four hours from March 1 to 

 September 30, and the greatest number of 

 consecutive days without 0.25 inch of rainfall 

 from March 1 to September 30. These are all 

 based on the rainfall data for the 20-year 

 period, 1895-1914. 



There are three snow maps presented. A 

 large one shows the average annual snowfall of 

 the United States, 1895-1914, drawn on a 

 topographic base-map with close attention to 

 the effects of altitude and exposure. The other 



two maps show the average annual number of 

 days (1) with measurable snowfall, and (2) 

 with snow cover. In the eastern United States 

 (except near the Atlantic) the line of one day 

 with snow cover (the average of several days 

 in one winter, with no days in several years) is 

 near the 33d parallel of latitude; that of 30 

 days with snow cover lies close to the 39th 

 parallel; that of 60 days near the 42d; that of 

 90 days near the 43d, and that of 120 from 

 near the 44th in the East to the 47th in Min- 

 nesota. As a broad generalization, the number 

 of days with snowfall is about half the number 

 of days with snow cover. 



The publication of these interesting precipi- 

 tation maps with the discussion makes us hope 

 that still another year will not pass before the 

 issue of the long-expected precipitation sec- 

 tion of the Atlas of American Agriculture, 

 with its colored maps, carefully made graphs 

 and detailed discussion. Still later, the folio 

 on temperature and the other climatic elements 

 are to come. 



Charles F. Bkooks 



Washington, I>. C. 



SPECIAL ARTICLES 



INTERSEXES IN DROSOPHILA SIMULANS 



On the first day of January, 1920, a stock 

 of Drosophila simulans Sturtevant^ from 

 Rochester, Minn., was found to contain inter- 

 sexual individuals. Over 200 such inter- 

 sexual specimens from this stock and deriva- 

 tives of it have now been examined. About 

 a dozen of them have been dissected and about 

 the same number have been cleared in KOH 

 and examined in balsam. All these specimens 

 apparently belong to a single type. Male and 

 female parts are both present, as will appear 

 from, the following table. 



The intersexes are sterile, inasmuch as their 

 gonads are almost, if not quite, absent. Their 

 sexual behavior seems to agree best with that 

 of the normal females. They are courted by 

 males, but mating has not been seen. 



1 For a description of this species see Psyche 

 (1919), 26, p. 153. 



