be thus determined; but if the first experiment gives a satisfactory result, no other is required.

A star of the first magnitude should next be brought into the field of view. If an irradiation from one side is perceived, part of the object-glass has not the same refractive power as the rest; and the part which is defective can be determined by applying in different positions a cap which hides half the object-glass. If the irradiation is double, it will probably be found that the object-glass has been too tightly screwed, and the defect will disappear when the glass is freed from such undue pressure.

If the object-glass is not quite at right angles to the axis of the tube, or if the eye-tube is at all inclined, a like irradiation will appear when a bright star is in the field. The former defect is not easily detected or remedied; nor is it commonly met with in the work of a careful optician. The latter defect may be detected by cutting out three circular cards of suitable size with a small aperture at the centre of each, and inserting one at each end of the eye-tube, and one over the object-glass. If the tube is rightly placed the apertures will of course lie in a right line, so that it will be possible to look through all three at once. If not, it will be easy to determine towards what part of the object-glass the eye-tube is directed, and to correct the position of the tube accordingly.

The best tests for determining the defining power of a telescope are close double or multiple stars, the components of which are not very unequal. The illuminating power should be tested by directing the telescope towards double or multiple stars having one or more minute components. Many of the nebulæ serve as tests both for illumination and defining power. As we proceed we shall meet with proper objects for testing different telescopes. For the present, let the following list suffice. It is selected from Admiral Smyth's tests, obtained by diminishing the aperture of a 6-in. telescope having a focal length of 8½ feet:

A two-inch aperture, with powers of from 60 to 100, should exhibit

A four-inch, powers 80 to 120, should exhibit

The tests in the first column are for definition, those in the second for illumination. It will be noticed that, though in the case of Polaris the smaller aperture may be expected to show the small star of less than the 9th magnitude, a larger aperture is required to show the 8th magnitude component of σ Cassiopeiæ, on account of the greater closeness of this double.

In favourable weather the following is a good general test of the performance of a telescope:—A star of the 3rd or 4th magnitude at a considerable elevation above the horizon should exhibit a small well defined disc, surrounded by two or three fine rings of light.

A telescope should not be mounted within doors, if it can be conveniently erected on solid ground, as every movement in the house will cause the instrument to vibrate unpleasantly. Further, if the telescope is placed in a warm room, currents of cold air from without will render observed objects hazy and indistinct. In fact, Sir W. Herschel considered that a telescope should not even be erected near a house or elevation of any kind round which currents of air are likely to be produced. If a telescope is used in a room, the temperature of the room should be made as nearly equal as possible to that of the outer air.

When a telescope is used out of doors a 'dew-cap,' that is, a tube of tin or pasteboard, some ten or twelve inches long, should be placed on the end of the instrument, so as to project beyond the object-glass. For glass is a good radiator of heat, so that dew falls heavily upon it, unless the radiation is in some way checked. The dew-cap does this effectually. It should be blackened within, especially if made of metal. "After use," says old Kitchener, "the telescope should be kept in a warm place long enough for any moisture on the object-glass to evaporate." If damp gets between the glasses it produces a fog (which opticians call a sweat) or even a seaweed-like vegetation, by which a valuable glass may be completely ruined.

The observer should not leave to the precious hours of the night the study of the bearing and position of the objects he proposes to examine. This should be done by day—an arrangement which has a twofold advantage,—the time available for observation is lengthened, and the eyes are spared sudden changes from darkness to light, and vice versâ. Besides, the eye is ill-fitted to examine difficult objects, after searching by candle-light amongst the minute details recorded in maps or globes. Of the effect of rest to the eye we have an instance in Sir J. Herschel's rediscovery of the satellites of Uranus, which he effected after keeping his eyes in darkness for a quarter of an hour. Kitchener, indeed, goes so far as to recommend (with a crede experto) an interval of sleep in the darkness of the observing-room before commencing operations. I have never tried the experiment, but I should expect it to have a bad rather than a good effect on the eyesight, as one commonly sees the eyes of a person who has been sleeping in his day-clothes look heavy and bloodshot.

The object or the part of an object to be observed should be brought as nearly as possible to the centre of the field of view. When there is no apparatus for keeping the telescope pointed upon an object, the best plan is so to direct the telescope by means of the finder, that the object shall be just out of the field of view, and be brought (by the earth's motion) across the centre of the field. Thus the vibrations which always follow the adjustment of the tube will have subsided before the object appears. The object should then be intently watched during the whole interval of its passage across the field of view.

It is important that the student should recognise the fact that the highest powers do not necessarily give the best views of celestial objects. High powers in all cases increase the difficulty of observation, since they diminish the field of view and the illumination of the object, increase the motion with which (owing to the earth's motion) the image moves across the field, and magnify all defects due to instability of the stand, imperfection of the object-glass, or undulation of the atmosphere. A good object-glass of three inches aperture will in very favourable weather bear a power of about 300, when applied to the observation of close double or multiple stars, but for all other observations much lower powers should be used. Nothing but failure and annoyance can follow the attempt to employ the highest powers on unsuitable objects or in unfavourable weather.

The greatest care should be taken in focussing the telescope. When high powers are used this is a matter of some delicacy. It would be well if the eye-pieces intended for a telescope were so constructed that when the telescope is focussed for one, this might be replaced by any other without necessitating any use of the focussing