But Herschel's theory concerning the Milky-way rested on the erroneous assumption that the stars are uniformly distributed in space, and also that his telescopes penetrated through the entire depth of the Galaxy. Further study showed him that his telescope of twenty feet, and even his great forty-foot telescope, which was estimated to penetrate to a distance 2,300 times that of stars of the first magnitude, failed to resolve some parts of the Galaxy into stars. Meanwhile, the structure of the Milky-way being better known, the irregular condensation of its stars became apparent, while the mutual relation existing between binary and multiple systems of stars, as also between the stars which form clusters, was recognized, as showing evidence of closer association between certain groups of stars than between the stars in general. Herschel's system, which rested on the assumption of the uniform distribution of the stars in space, and on the supposition that the telescopes used for his gauges penetrated through the greater depths of the Galaxy, being thus found to contradict the facts, was gradually abandoned by its author, who adopted another method of estimating the relative distances of the stars observed in his gaugings. This method, founded on photometric principles, consisted in judging the penetrating power of his telescope by the brightness of the stars, and not, as formerly, by the number which they brought into view. He then studied by this new method the structure of the Milky-way and the probable distance of the clustering masses of which it is formed, concluding that the portion of the Galaxy traversing the constellation Orion is the nearest to us. This last result seems indicated by the fact that this portion of the Milky-way is the faintest and the most uniform of all the galactic belt.