172 A NEW SOLAR THEORY. 



follows a ininiinum quicker in proportion as the maximum is higher. 

 * * ■"' That the interval between minima and maximsi depends inti- 

 mately upon the intensity of development of the spots during that 

 period would seem, therefore, to ])0 quite certain, according to Lock- 

 yer's and Halm's investigations.''' 



In this connection another interesting consequence of the theory 

 may be mentioned. The above expression for v, as already pointed 

 out, shows that the theoretical spot-curve, after having passed 

 through a maximum, returns to zero when t=p. At this moment the 

 status quo ante appears to be reestablished, inasmuch as the photo- 

 spheric layers, by the assistance of atmospheric reflection, once more 

 receive an exact compensation for their loss of energy sustained by 

 outward radiation. If the oj^acity of the absorption la3^ers above tlie 

 photosphere could henceforth be supposed to remain constant, a repe- 

 tition of spot-activity would l)e impossible. But since the cooling of 

 these absorptive layers, l)eing the necessary consequence of deficient 

 contraction, is bound to proceed, it must eventuall}^ lead to a new 

 sequence of phenomena similar to those of the previous cycle. The 

 principal condition for the initiation of a new spot cycle must there- 

 fore be sought in the rate of cooling of the outer atmosphere at the 

 moment of the reestablishment of thermal equilibrium in the photo- 

 spheric layers. If the cooling proceeds rapidly the overheating of the 

 inner layers will be indicated by a comparatively early renascence of 

 eruptive activity. If, on the contrary, the cooling progresses slowly 

 a more considerable interval of time will elapse ])efore the state of 

 unstable equilibrium favorable to the formation of new eruptions 

 and spots can make its appearance. In the first case the mininuuu 

 will ])e of short duration and the cycles will follow each other in rapid 

 succession, whereas in the second case the minimum will appear pro- 

 tracted and the distance 1)etween consecutive cycles lengthened. In 

 other words, the length of the period must depend on the rate of cool- 

 ing of the atmospheric layers. Hence, from what has been said above, 

 it nuist also vary with the intensity of spot development. Cycles 

 exhibiting excessive spot activity must have less extended minima 

 and must therefore be of shorter duration than others distinguished 

 by a feel)le display of dynamical activity. This consequence of the 

 theory, too, appears to be contiraied by the observations. Thus the 

 mean length of period of the cycles 1 7(56. 5 — 1775.5 — 1781. 7 and 1833.9 — 

 1813.5, during which the display of spots was far above the average, 

 em])raccs not more than 9.3 years, while the feebl}^ developed cycles 

 1781.7 — 1798.3 — 1810.6 show a mean period of no less than 13.0 j^ears. 



The object of this brief abstract l>eing mereh" an exposition of the 

 main principles upon which I have ventured to build a new solar 

 theory, I shall not entcM' upon its various applications to the phenomena 

 coimected with the pei-iodic changes in the display of forces at the 



