One main difficulty in such observations arises from the diminution of starlight, caused by its passage through a prism. If a rope of a dozen strands is untwisted, each of those strands is far weaker than the whole rope was; and each strand of color in the twisted rope of light is of necessity much more feeble, seen by itself, than the whole white ray seen in one. Another difficulty in our country generally is the climate, which gives so few days or hours in the year for effective work. Yet the full amount accomplished by seizing upon every possible opportunity is, in the aggregate, astonishing. A third and very pressing difficulty attendant upon examination of star-spectra is caused by the incessant motion of our air, through which, as through a veil, all observations have to be made. The astronomer can never get away from the atmosphere; and unless the air be very still--that is to say, as still as it ever can remain--the spectrum-lines are so uncertainly seen as to make satisfactory results impossible. Dr. Huggins has sometimes passed hours in the examination of a single line, unable to determine whether or no it precisely coincided with the comparison-line of some earthly substance. In _this_ matter, no leaping at conclusions is admissible. The photography of stars would be easy enough if one could just expose a plate to the shining of the star, and there leave it to be impressed--there leave the star-ray to sketch slowly its own image. But this is hardly possible. The unceasing motion of the earth, causing the star perpetually to pass away from the telescopic field, and the exceeding narrowness of the slit opposite to which the star has to be kept in a stationary position, make the most accurate adjustment needful. Clockwork alone can not be trusted. If it could, the star and the photographic apparatus might be comfortably “fixed,” and left to do their own work. Instead of which, while a photograph is proceeding, it is desirable that an observer should sit gazing patiently at the telescope-tube, where the image of the star is seen, ready at any moment to correct by a touch the slightest irregularity in the clockwork motion of the telescope, and so to prevent a blurred and spoilt reproduction of the star, or of its spectrum. One hour, two hours, three hours, at a stretch, this unceasing watchfulness may have to be kept up, and no small amount of enthusiasm in the cause of science is requisite for so monotonous and wearying a vigil. As noted earlier, it has been found possible, if the photograph of a star or nebula is not completed in one night, to renew the work and carry it on the next night, or even for many nights in succession. This is especially practicable in spectroscopic photography.