These long-exposure plates are intended especially for what may be called descriptive astronomy. With them, as we have seen, advantage is taken of cumulative light-effects on the sensitive plate, and the telescope's light-gathering and space-penetrating powers are vastly increased. We are enabled to carry our researches far beyond the confines of the old visible universe. Extremely faint objects can be recorded, even down to their minutest details, with a fidelity unknown to older visual methods. But at present we intend to consider principally applications of photography in the astronomy of measurement, rather than the descriptive branch of our subject. Instead of describing pictures made simply to see what certain objects look like in the sky, we shall consider negatives intended for precise measurement, with all that the word precision implies in celestial science. Taking up first the photography of stars, we must begin by mentioning the work of Rutherfurd at New York. More than thirty years ago he had so far perfected methods of stellar photography that he was able to secure excellent pictures of stars as faint as the ninth magnitude. In those days the modern process of dry-plate photography had not been invented. To-day, plates exposed in the photographic telescope are made of glass covered with a perfectly dry film of sensitized gelatine. But in the old wet-plate process the sensitive film was first wetted with a chemical solution; and this solution could not be allowed to dry during the exposure. Consequently, Rutherfurd was limited to exposures a few minutes in length, while nowadays, as we have said, their duration can be prolonged at will. When we add to this the fact that the old plates were far less sensitive to light than those now available, it is easy to see what were the difficulties in the way of photographing faint stars in Rutherfurd's time. Nor did he possess the modern ingenious device of a combined visual and photographic instrument. He had no electric controlling apparatus. In fact, the younger generation of astronomers can form no adequate idea of the patience and personal skill Rutherfurd must have had at his command. For he certainly did produce negatives that are but little inferior to the best that can be made to-day. His only limitation was that he could not obtain images of stars much below the ninth magnitude. To understand just what is meant here by the ninth magnitude, it is necessary to go back in imagination to the time of Hipparchus, the father of sidereal astronomy. (See page 39.) He adopted the convenient plan of dividing all the stars visible to the naked eye (of course, he had no telescope) into six classes, according to their brilliancy. The faintest visible stars were put in the sixth class, and all the others were assigned somewhat arbitrarily to one or the other of the brighter classes.