EXPLANATORY HYPOTHESES 485 



top in figure 1 shows Wolfer's sun-spot numbers inverted. A comparison 

 of this with the curve of temperature departures shows considerable 

 agreement. Nevertheless there are also pronounced discrepancies. Rec- 

 ognizing this and also feeling convinced of the importance of solar radia- 

 tion, our two sets of authorities have both adopted the same method. 

 They have combined the p3^rheliometric curve with that for sun-spots. 

 The result is given in the third curve, P -|- S. This agrees with the tem- 

 perature departures much more closel}^ than does either of the other 

 curves. Apparentl}-, if we judge from the curves of figure 1 alone, the 

 earth's temperature depends on at least two primary factors : first, the type 

 of solar activity indicated by sun-spots, and, second, the amount of solar 

 heat which can reach the earth's surface through an atmosphere which is 

 sometimes relatively clear and at others more or less filled with dust. 

 Between the two factors, as appears in figure 1, there seems to be a great 

 difl^erence, for changes of temperature seem to agree with variations in 

 solar spottedness much more closely than with volcanic eruptions. 



Disagreement hettueen solar and terrestrial tern, per ature changes. — At 

 this point attention should be directed to a somewhat striking phenom- 

 enon. Few people would question that if the sun should become notably 

 warmer or cooler the earth's temperature would also change. This con- 

 clusion, apparently inevitable, is the basis of the ordinary form of the 

 solar hypothesis of climatic changes. Yet in figure 1 we see that the 

 actual measurements of the radiation received from the sun do not agree 

 with the temperatures recorded in the parts of the earth's surface where 

 the sun is able to act most effectively. The changes in the amount of 

 solar radiation which gets through the atmosphere and reaches the earth's 

 surface appear to have a distinct efl^ect, as is evident from the fact that 

 the P + S curve is more like the temperature curve than is the simple S 

 curve which represents sun-spots alone. Yet apparently there is some 

 other solar factor whose influence on changes of terrestrial temperature 

 is much greater than that of variations in the amount of solar heat. 



This conclusion is emphasized by another fact. Abbott^ states that the 

 net result of observations made by the Smithsonian Institution and by 

 other agencies during a long series of years is to show that during years 

 of maximum sun-spots the solar constant is higher than during years of 

 minimum spots. This is exactly contrary to the conditions of terrestrial 



^ C. G. Abbott: Science, vol. 39, 1914, p, 347. "An increase of a hundred sun-spot 

 numbers corresponds to an increase of about 0.07 per square centimeter per minute in 

 the solar radiation outside the earth's atmosphere." Abbott gives the mean value of 

 the solar constant as 1.933 calories. Since sun-spot maxima at present range from 

 sun-spot numbers of about 45 to 155, this means that during the sun-spot cycle the 

 solar constant may be from 1.0 per cent to 5.G per cent gr-^ater at sun-spot maxima 

 than at minima. 



