Royal Astronomical Society. 143 



The construction of the equatoreal enabled me to keep the same 

 points of the disc steadily fixed on the pile, as far as it was neces- 

 sary to obtain an invariable deflection of the needle of the galvano- 

 meter, and the position of the points of the sun, under examination, 

 was readily deduced from the reading of the declination circle. 

 There is no need to observe that all extraneous radiation was pre- 

 vented by large screens, supported by the telescope itself, and the 

 turning dome was only opened as far as it was necessary to intro- 

 duce solar rays on the object-glass. The telescope has 1™'25 focal 

 length, and 77mm aperture. 



" Now these are the results obtained in sundry experiments : — 

 1. The heat of the solar image is at the centre almost twice as 

 great as at the borders. This is found to be true, examining the 

 diameters both in right ascension and declination. 2. The maxi- 

 mum of temperature did not appear to be at the centre, but above 

 it, in a point distant from it about 3' of geocentric declination. 

 Constructing graphically the curve of the intensity of heat, taking 

 as abscisscB the parts of the sun's diameter, and as ordinatcB the 

 intensities themselves, it appears that this curve (a kind of inverted 

 parabola) is not symmetrically disposed about the axis of the ordi- 

 nates, but a good deal inclined towards the upper edge. I subjoin 

 some numbers which represent the intensity of heat in the parts of 

 the diameter of the sun taken in minutes, + above and — below 

 the centre of the image. 



Positions on the diameter"! _^j^'.gg ^^'.32 _,. ^\qq _^ 1^.32 .10^9 ^ilss 

 of the sun in decimation J ' ' ' ' 



Relative intensity of heat 57-39 88-81 100-00 99-48 81-32 54-34 



These are the results of eight series of experiments, none of which 

 is found in contradiction with the others, and their separate num- 

 bers are very nearly the same, so that the fact seems to me com- 

 pletely ascertained. It is certainly curious that the maximum of 

 heat corresponds with the position of the solar equator, as visible 

 from the earth at the epoch of the experiment (20th, 21st, 22nd 

 March). This leads naturally to the conclusion that the solar 

 equatoreal regions must be hotter than the polar regions, as was 

 suspected already from the more frequent appearance of the spots 

 there. The conclusion seems perfectly accurate, even admitting a 

 solar atmosphere, since the effect of this last should be to diminish 

 symmetrically the radiation around the centre of the image ; on 

 the contrary, if the polar regions are less hot than the equatoreal, 

 the intensity of heat should have been less in the lower part of the 

 image, where the south pole of the sun was visible ; and, conse- 

 quently, the parts having equal distance from the centre of the 

 image had a very different heliographical latitude, on account of 

 the inclination of the solar axis to the ecliptic. From these prin- 

 ciples only, the non-symmetry of the curve is accounted for. If 

 this alone is the cause, the curve will be found symmetrical in the 

 months of June and December, and reversed in September, since 

 in the two former the equator passes through the centre of the 



