396 Intelligence and Miscellaneous Articles. 



be raised to 169980° by adopting the number of 27° as produced 

 by the solar radiation. 



I have already remarked, in my work on the Sun (p. 270), and 

 M. Hirn has lately repeated it, that the temperature of the radia- 

 tion may depend either solely on the superficial stratum of the sun, 

 or on a considerable thickness of its substance, according as this is 

 opaque or transparent. M. Hirn concludes that, if the transpa- 

 rence were nearly perfect, the temperature might well be only a 

 few thousand degrees ; but divers phenomena prove that, on the 

 contrary, the transparence is very imperfect. In one of my prece- 

 ding communications I reported the singular observation that the 

 currents of the penumbrae cross and pass one above another; 

 in that case the upper currents completely conceal the lower ; so 

 that the mass of the photosphere has no sensible transparency. 

 This observation of the crossing of the currents has lately been con- 

 firmed by Mr. Langley*. The defect of transparency can also be 

 established by observation of the thick jets of the metallic protube- 

 rances, in which one branch does not permit the other to be seen 

 through its thickness. Without admitting absolute opacity (for 

 the very strong light of the photosphere may well prevent the lower 

 strata from being distinguished), it is certain that the photosphere 

 is not completely transparent; for otherwise the margin of the 

 sun's disk would not appear sharply defined, but diffuse. 



The temperature above indicated is therefore not inadmissible. 

 It is doubtless very far from the number which would be given by 

 the direct application of Newton's law ; but it is also very far from 

 that which would result from the law of Dulong and Petit. It 

 seems to me that these comparative experiments almost completely 

 eliminate the considerations drawn from the theoretic law, and 

 that they give a lower limit of the temperature of the sun. 



Besides, if the solar temperature reached only a few thousand 

 degrees, its cooling would be sensible in a relatively brief interval 

 of time ; and this diminution of temperature would betray itself by 

 a notable acceleration of the rotation of the sun. Doubtless the 

 excess of velocity observed at the solar equator, which it has use- 

 lessly been attempted to account for by currents analogous to our 

 trade- winds, is due to this cooliug ; for, though very slight, it is by 

 no means nil. M. Faye's law, deduced from Carrington's, indeed 

 results from the simple comparison of the diminution which the 

 areas of the equatorial circles and the parallels must undergo by 

 the cooling of the mass estimated in their respective planes of ro- 

 tation. This progressive action must maintain constantly this 

 difference of velocity ; so that there is no need to have recourse, 

 with M. Roche, to a remote crisis ; for in that case the mass would 

 soon have arrived at uniformity in consequence of friction. 



The activity of the sun not being constant, it follows that neither 

 are so its losses by radiation, and consequently that the apparent 



* American Journal of Science, 3rd Series, vol. vii., February 18/4. 



