The expression of the law may be thus generalized:—the number of light-particles (or, if the phrase be preferred, the number of light-impressions) received upon the shifting plane, will be _inversely_ proportional with the squares of the distances of the plane. Generalizing yet again, we may say that the diffusion—the scattering—the irradiation, in a word—is _directly_ proportional with the squares of the distances. [Illustration] For example: at the distance B, from the luminous centre A, a certain number of particles are so diffused as to occupy the surface B. Then at double the distance—that is to say at C—they will be so much farther diffused as to occupy four such surfaces:—at treble the distance, or at D, they will be so much farther separated as to occupy nine such surfaces:—while, at quadruple the distance, or at E, they will have become so scattered as to spread themselves over sixteen such surfaces—and so on forever. In saying, generally, that the irradiation proceeds in direct proportion with the squares of the distances, we use the term irradiation to express _the degree of the diffusion_ as we proceed outwardly from the centre. Conversing the idea, and employing the word “concentralization” to express _the degree of the drawing together_ as we come back toward the centre from an outward position, we may say that concentralization proceeds _inversely_ as the squares of the distances. In other words, we have reached the conclusion that, on the hypothesis that matter was originally irradiated from a centre and is now returning to it, the concentralization, in the return, proceeds _exactly as we know the force of gravitation to proceed_. Now here, if we could be permitted to assume that concentralization exactly represented the _force of the tendency to the centre_—that the one was exactly proportional to the other, and that the two proceeded together—we should have shown all that is required. The sole difficulty existing, then, is to establish a direct proportion between “concentralization” and the _force_ of concentralization; and this is done, of course, if we establish such proportion between “irradiation” and the _force_ of irradiation. A very slight inspection of the Heavens assures us that the stars have a certain general uniformity, equability, or equidistance, of distribution through that region of space in which, collectively, and in a roughly globular form, they are situated:—this species of very general, rather than absolute, equability, being in full keeping with my deduction of inequidistance, within certain limits, among the originally diffused atoms, as a corollary from the evident design of infinite complexity of relation out of irrelation. I started, it will be remembered, with the idea of a generally uniform but particularly _un_uniform distribution of the atoms;—an idea, I repeat, which an inspection of the stars, as they exist, confirms.