VOL. LX.] PHILOSOPHICAL TRANSACTIONS. " 6Q 



But the waste of the sun's substance must lessen the attraction between the 

 earth and sun. As the attraction lessens, ,lhe earth will recede to greater 

 distances. And hence there will arise a further diminution of the sun's apparent 

 diameter, and a prolongation of the anomalistic year. The density of each 

 particle of light has been supposed 3 times that of iron, or 23 times the mean 

 density of tlie earth.* Therefore, as often as the sun's loss by light amounts 

 to an earth in size, it will amount to 23 earths in matter. The matter of 23 

 earths is , ^ j ^ ^ d of the sun's matter, if the sun's parallax be Q". Therefore in 

 38513000O Egyptian years, the sun loses lajjs d of his mattter; and the 

 gravitation towards the sun, at any given distance, diminishes in the same 

 proportion. But this alteration is much too small to discover itself in the 

 motions of the earth or any of the planets. I will not at present consider, by 

 what law the distance of the planets from the sun would increase, because the 

 inquiry could not be reduced to a small compass; but it is obvious that, what- 

 ever that law may be, it must arise solely from the diminution of gravitation: 

 and the like is to be said of the anomalistic periods. And therefore, while the 

 diminution of gravitiition is insensible, the changes in these circumstances must 

 be insensible too. Of all the changes to which our system may be obnoxious, 

 those which should arise from the waste of the sun's substance in light, on the 

 supposition that light is an actual emanation of matter from the sun, reckoning 

 that waste at the utmost, are perhaps the least considerable. 



In the foregoing computations, the instantaneous emission of light has been 

 greatly over-rated. For if the particles of light were of the magnitude and 

 density which has been assigned to each, and were to issue from the sun in the 

 close arrangement that has been supposed, they would form a sort of crust on 

 the sun's surface, at least 12 times more dense than water, i. e. 9600 times more 

 dense than our atmosphere in the parts next the earth's surface, if the density of 

 common water compared to that of air be reckoned only as 800 to 1 .^ But if 

 the density of light on the sun's surface be 960O times that of our air, its 

 density when it arrives at the earth, or its density on the surface of the orbis 

 magnus,J should be more than -i-th part of the density of our air. When 

 substances of different specific gravities, as a piece of gold and a piece of cork, 

 descend, in the exhausted receiver, with equal velocities, and fall equal heights 



• I reckon the mean density of the earth no greater than that of common water. It is certain 

 that it cannot be less. Sir Isaac Newton reckons it 5 or 6 times greater ; but I confess that I am 

 not satisfied with his reasons for making it so great. If I have underrated it, I have, in so doing, 

 given the advantage to Dr. Franklin's objection. 



t It is well known that the density of water to that of air, is as 850 to I at least. 



X By the surface of the orbis magnus, I mean to denote a particular place in absolute space, 

 namely the surface of that sphere which is concentric with the sun, and hath the earth's mean 

 distance from tke sun for its scmidiameter. 



