As an illustration of the degree of refine ment attained in astrophotography with such an instrument as the 60-inch reflector, the at tachment used in direct photography, and illus trated in Fig. 1, may be briefly described. It is the improved double-slide plate-carrier, de veloped by Ritchey from the simple form devised by Common. The telescope tube of course moves by clock-work to follow the ap parent motion of the heavenly bodies due to the earth's daily rotation. This motion of the telescope is marvelously smooth and accurate, but is nevertheless far from being accurate enough to suffice, alone. To secure the sharpest attainable photographs the observer must watch, throughout the entire time of exposure of the sensitive plate, a suitable guide-star as close as possible to the object being photographed; this is done by means of a higFi-power eye-piece or microscope with illuminated cross-lines of spider-web at its focus. Minute corrections of position are constantly introduced by the ob server by means of two screws, the large milled heads of which are held in the observer's fin gers; these screws move the two slides, at right angles to each other, which carry the plate holder. A second guiding eye-piece, on the opposite side of the plate-holder, is also used to permit the detection and correction of any slight rotation of the field; this rotation is certain to occur in an instrument as powerful and sensitive as the 60-inch, and long-exposure photographs would be ruined by it unless such corrections were introduced. Furthermore, as the length of the steel tube of the telescope changes slightly with temperature throughout the night, the focal plane of the mirror changes slightly with reference to the plane of the pho tographic plate. Hence an accurate method of refocussing is used, by which this relative change is corrected at frequent intervals. With these refinements very long exposures on ex cessively faint objects can be continued hour after hour and night after night with the cer tainty that no errors can occur which could injure the sharpness of the photographs. Two
concrete examples will illustrate the results secured by these methods. A photograph of the spiral nebula Messier 33 Trianguli, taken with an exposure of hours, shows over 26,000 nebulous stars undoubtedly belonging to the nebula itself, besides about 3,500 other stars not nebulous, which lie between us and the nebula; all of these are within an area of the sky 30 minutes of arc square. A photograph of the globular star-cluster Messier 13 Herod's, with 11 hours' exposure, shows by actual count over 30,000 stars in the cluster, outside of the dense central region in which the star-images to some extent superpose, and where counting is therefore impossible; photometric measures show, however, that at least an equal number of stars are present in this dense central region. In the latter photograph the faintest stars shown are of 21/2 magnitude; this is five mag nitudes, or approximately 100 times, fainter than the faintest stars which can be seen (visually) in the largest visual refractors. In the best of these photographs the fainter star images are less than one second of arc in diameter.
Another illustration of the extraordinary results in astrophotography made possible by the great modern reflecting telescopes is the recent discovery of numbers of very faint nova or °temporary stars" in the spiral nebula. These stars apparently correspond in all re spects to the nova which occur in our own stellar system or galaxy; they form another link in the chain of evidence tending to show that some of the spiral nebulae, at least, are other stellar systems similar to our own. A comparison of the average brightness of all of the nova which have occurred in our own galaxy (about 5/2 magnitude) with the aver age brightness of all of those which have re cently been discovered in the photographs of spiral nebula (about 14 magnitude), affords what is at present the most reliable method of estimating the average distance from us of the spiral nebula, as compared with that of the nova in our galaxy; this relative distance is thus shown to be of the order of 50 to 1.