JO PHILOSOPHICAL TRANSACTIONS. [aNNO 1770. 



in the same time, it is obvious, that the density of the medium, through which 

 they fall, bears no sensible proportion to the density even of the lighter substance. 

 The medium through which such substances fall, in a transparent glass receiver, 

 is composed of some small portion of rarefied air, and, to all appearance, of the 

 sarhe quantity of light, as the receiver contains before exhaustion. For the 

 quantity of light, in a transparent receiver, can by no means suffer any diminu- 

 tion, by the action of the air-pump. The density of the light therefore, in our 

 air, is certainly too small to bear any sensible proportion to that of gold or even 

 of cork. And the density of cork bears, though a great, yet a finite and a 

 sensible proportion to that of the atmosphere; because in air gold and cork do 

 not descend with equal velocities. Hence, I think, I may conclude with the 

 greatest certainty, that the density of the sun's light at the earth, or on the 

 surface of the orbis magnus, is too small to bear any sensible proportion to the 

 density of common air; and I shall hardly underrate the density of the light, if 

 I reckon "it only | ^^ „„ part of the density of the air, or , o ^ ; „ „ „ part of the 

 density of common water. Suppose this to be the density of light upon the 

 surface of the orbis magnus, and it will be found by computation, that its 

 density on the sun's surface, must be less than -,-1-^ of the density of common 

 water. From these considerations, 1 think it may be concluded with the greatest 

 certainty, that the quantity of matter that issues from the sun in light, has been 

 greatly overrated in the foregoing computations. 



I apprehend, however, that the density of each separate particle cannot be less 

 than has been supposed: but that the magnitude of each is less, and the 

 arrangement less compact.* Let the density of each, and the number that 

 issues at any one time from the sun, remain as before; and let us consider, in 

 what proportion the magnitude of each particle, must be diminished so that 

 they may altogether form a fluid, on the sun's surface, 173 times less dense than 

 water. Let abcd represent a section of the sun's 

 sphere, through the centre s. In the spherical 

 surface abcd, take any point b. Join sb, and take 

 Be, b/I each equal to the semidiameter of a particle 

 of light. On the centre s, at the distances se, sf, 

 imagine two other spheres, egh,Jhl, one inclosing, 

 the other inclosed within the sphere abcd. The 

 solid space efklgh, is the space that contains all the 

 particles of light, with their interstices, that issue 

 together from the spherical surfaces abcd; and 

 because ef bears an exceedingly small proportion to sb, therefore the spherical 

 surfaces egh, fhl, are very nearly equal each to the other, and to the spherical 



•* , • Vide Instant. Product, p, 37. 



