Adopting Prof. Newcomb's parallax of 0"·14 for the famous star 1830 Groombridge, the velocity perpendicular to the line of sight is about 150 miles a second. The velocity _in_ the line of sight--as shown by the spectroscope--is 59 miles a second approaching the earth. Compounding these two velocities we find a velocity through space of about 161 miles a second! An eminent American writer puts into the mouth of one of his characters, a young astronomer, the following:-- "I read the page Where every letter is a glittering sun." From an examination of the heat radiated by some bright stars, made by Dr. E. F. Nicholls in America with a very sensitive radiometer of his own construction, he finds that "we do not receive from Arcturus more heat than we should from a candle at a distance of 5 or 6 miles." With reference to the progressive motion of light, and the different times taken by light to reach the earth from different stars, Humboldt says, "The aspect of the starry heavens presents to us objects of _unequal date_. Much has long ceased to exist before the knowledge of its presence reaches us; much has been otherwise arranged."[302] The photographic method of charting the stars, although a great improvement on the old system, seems to have its disadvantages. One of these is that the star images are liable to disappear from the plates in the course of time. The reduction of stellar photograph plates should, therefore, be carried out as soon as possible after they are taken. The late Dr. Roberts found that on a plate originally containing 364 stars, no less than 130 had completely disappeared in 9¼ years! It has been assumed by some writers on astronomy that the faint stars visible on photographs of the Pleiades are at practically the same distance from the earth as the brighter stars of the cluster, and that consequently there must be an enormous difference in actual size between the brighter and fainter stars. But there is really no warrant for any such assumption. Photographs of the vicinity show that the sky all round the Pleiades is equally rich in faint stars. It seems, therefore, more reasonable to suppose that most of the faint stars visible in the Pleiades are really far behind the cluster in space. For if _all_ the faint stars visible on photographs belonged to the cluster, then if we imagine the cluster removed, a "hole" would be left in the sky, which is of course utterly improbable, and indeed absurd. An examination of the proper motions tends to confirm this view of the matter, and indicates that the Pleiades cluster is a comparatively small one and simply projected on a background of fainter stars.