Page:Sir William Herschel, his life and works (1881).djvu/198

176 can deny, with any show of reason, that, on the whole, his suppositions must be valid. On the whole, the stars which we call faint are farther from us than the brighter ones; and, on the whole, the brilliancy of our brightest and nearest stars is not very far from the brilliancy of the average star in space. We cannot yet define the word very by a numerical ratio.

The method struck out by was correct; it is for his successors to look for the special cases and limitations, to answer the question, At a certain distance from us, what are the variations which actually take place in the brilliancy and the sizes of stars? The answer to this question is to be found in the study of the clusters of regular forms, where we know the stars to be all at the same distance from us.

Frequently in the course of his astronomical work, found himself confronted by questions of physics which could