WEATHER GOVERNED BY THE SUN — ABBOT 115 



B. Precipitation. — Continiied. 



Fair, 11: New Haven, Albany, Philadelphia, Washington, Charleston, 



Peoria, Galveston, Santa Fe, Denver, San Francisco, Spokane. 

 Half and half, 8: Key West, Cincinnati, Omaha, Helena, Salt Lake City, 



San Diego, Red Bluff, Portland. 

 Bad, 5: Hatteras, Mobile, Nashville, Abilene, Cheyenne. 



Doubtless readers will wish to inquire whether it is reasonable that 

 periodicities of small amplitude in solar variation, such as those 

 shown in figure 2, are competent to produce temperature changes 

 as large as those found. In the following table, I give a percentage 

 analysis of this question. Unconsciously, when we think of the 

 temperature, we base it approximately on the zero of the ordinary 

 thermometer, so that a change of 5° F. seems to us as if it were a 

 change of 5 percent or more in temperature. Really we should con- 

 sider the temperature as measured from the absolute zero. This is 

 -273° of the Centigrade scale and -523°.4 F. Thus a Fahrenheit 

 temperature of 70° is really 593°.4 Absolute, and a change of 5° is 

 only about 1 percent in temperature. 



The table shows that though the ranges of the periodic changes 

 in solar radiation are all less than 1 percent, so too are almost all the 

 corresponding changes in temperature. Indeed, if measured in per- 

 centages, the temperature changes average but from one-third to 

 nine-tenths as great as the changes in solar radiation. On the whole, 

 the relationship does not seem unreasonable and leads us to the 

 remarkable conclusion that an important and perhaps a major part 

 of the departures from normal, which make up weather as distin- 

 guished from climate, originate in these newly discovered variations 

 in the radiation of the sun. If so it is clear that long range weather 

 prediction is impossible if based solely on the earth's conditions, 

 excluding solar variation as a factor. 



36923—36- 



