i;2 AXXAL!^ XEW YORK ACADEMY OF SCJEXCES 



it will be j^ossible to reverse the problem and search in the variable solar 

 phenomena for the periodicities which might correspond to the changes 

 of atmospheric temperature. But the problem is extremely complicated. 

 All we know about climatic variations is derived from observations taken 

 at the earth's surface. Xow. Charles J. Kidlmer'* "lias shown that in 

 the belt of the northern United States and southern Canada where storms 

 on the average are most numerous, the number of storms varies almost 

 directly in harmony with the number of sun-spots, just as is the case 

 with tropical hurricanes. In other areas, however, the reverse appears 

 to be true, and there is a decrease in storminess. The general conclusion 

 seems to be that when sun-spots are few in number cyclonic storms move 

 in a great variety' of tracks, but when spots are numerous the storms tend 

 to confine themselves to a few well-defined tracks, so that storminess is 

 more or less restricted to certain areas within which it is highly concen- 

 trated.** '^ If we bear in mind the fact that storms are centered by 

 columns of ascending air. we must presume that on the average differ- 

 ences in storm frequency (and intensity) must produce corresponding 

 differences in the vertical distribution of temperature. Ellsworth Hunt- 

 ington supposes that *"ia this way increased activity of the sun might 

 cause increased activity upon the earth, and the earth's surface might 

 become cooler while the upper air above the reacli of convection might 

 become warmer.'" "'^ This most interesting suggestion could be verified. 

 I suppose. 



Utilizing the table of Leo Kelley. giving the annual frequencies of 

 eruptions/" as well as Koeppen's worlds mean temperature table. I made 

 for both overlapping ten-yearly totals. The following diagram ( Fig. T ) 

 represents the figures graphically and shows plainly that some sort of a 

 relation between the numbers of volcanic eruptions and atmospheric tem- 

 perature is undeniable. 



Temperature seems to be higher when the eruptions are more frequent 

 and when volcanic activity is attenuated; atmospheric temperature seems 

 also to be lower than the average. This result is not in favor of the 

 volcanic dust explanation of climatic variations and if taken in con- 

 nection with the remarkable coincidences Ijetween the well pronounced 



'* Charles J. Kcllmer : --The Shift of the Storm Track." In : E. Huntington : "Tho 

 CUmatic Factor..." p. 19.3. Carnegie Inst, Pub. No. 192. 1904. 



^Ellsworth Huxtixgtox : Op. cit., p. 253. 



"« Manuscript communication. Huntington discusses the question at length in his 

 recent paper, "The Solar Hypothesis of Climatic Changes." Bull. Geol. Soc. Amer.. vol. 

 25, p. 477. 1914. 



"^ JOHX MiLXE : "'Eighteenth Report of the Committee on Seismological Investiga- 

 tions," p. 65. Report of the >i.3rd meeting of the Brit. Assoc. Birmingham. 191.S. 



