contrivance. Possibly this apparatus in its present form may be capable of improvement, but it is at least a step, and a very important step, in the direction of solving one of the most troublesome and difficult problems to be met with in the attempt to obtain a really satisfactory mounting for reflecting telescopes.

That the reflecting telescope is capable of doing excellent work in the hands of those who take sufficient care and trouble with the adjustment and in the working, is sufficiently evidenced in the results obtained by Draper, De la Rue, Common and Roberts, but the fact is that to obtain good results with the reflector in its present imperfect state of development, more labour and patience is required than most observers care to bestow on the work, and there is much to be said in excuse for this, for if an astronomer's time be taken up with the necessary attention to the details of his instrument he will not be able to pay that undivided attention otherwise possible, and at all times desirable, for his more legitimate work in the obtaining of results with that instrument. The fact that the reflector brings all rays of light to a common focus, irrespective of their wave-lengths, while the refractor is at best but a compromise, tells strongly in favour of the ultimate success of the reflector over its rival.

True, it may be said that the experiments in glass-making that have been carried on for some years at the Jena glass manufactory may yet eventuate in producing qualities of glass which will remove this reproach from the refractor, and enable us to perfectly balance the chromatic error, and at the same time be of a sufficiently permanent character to justify its use in the case of large objectives. No one would be foolish enough to attempt to make a large objective of any material which was not known from previous experience to be capable of preserving its perfection of surface for at least 20 or 25

years.

Unfortunately there is no test of permanency except the lapse of years; even if therefore some such glass were in existence at the present time, no maker who had any desire that his name should live after him in his work, would care to use this untried material until actual experience proved its character for permanence. This, and the fact that it has not yet been found possible to produce perfect pieces of this Jena glass of one-tenth of the weight of those already produced of the more ordinary varieties of optical glass, cuts off any hope we might otherwise have of being able for the present to produce large objectives with perfect correction for the chromatic aberration, and so long as this is the case, the reflector, which treats rays of all refrangibilities alike, has in this respect the advantage.

When we consider that the largest optical discs ever yet produced, and which were rightly considered a perfect triumph of art, are only 40 inches in diameter, and that on the other hand Lord Rosse's reflector of 72 inches diameter is now half a century old, it is tolerably evident that for the present, at least, we must look to reflectors if we want to increase to any large extent the power of our telescopes.

With this view, and understanding that it is likely an attempt

will be made to build an 8 or 10-foot reflector for the great exhibition to be held in Paris in 1900, it may be interesting to consider the conditions desirable to be fulfilled for such an instrument, and the most promising construction to satisfy those conditions. If a monster telescope, such as this, is to be mounted only in such a manner as will satisfy the ordinary conditions of star gazing, I fear the results will be disappointing, but let it be mounted in such a manner as to render it useable for the more delicate and refined work of the modern astronomy, and a grand and productive field of work is open to it.

But the problem of mounting an enormous instrument such as this, whose weight would probably amount to from 50 to 100 tons, so perfectly poised and so accurately driven by clockwork as never to vary from its true position by a quantity greater than the apparent motion of a star in one-twentieth of a second of time, is

B

sufficiently difficult to justify almost a doubt of its possibility, and this difficulty has been appreciated by others; for Dr. Common, who, as the maker of the largest equatorially mounted reflector ever completed, must be considered as the first authority, proposed some short time since to resort to the alt-azimuth form of mounting, with which it would, of course, be impossible to satisfy the. above condition.

Dr. Common himself has made a splendid advance in adopting the system of flotation of the polar axis; this principle of flotation appears to me to be capable of further development. It is perfectly possible to make a tube for a Newtonian reflecting telescope (which is necessarily closed at the lower end) of such a weight, and with its weight so distributed that it will not only float in water submerged to a certain point (preferably near the upper end), but will be in a state of equilibrium when placed at any or in every position down to a certain angle, which angle depends on the exact outside form of the tube. For instance, if AB (Fig. 1) be a tube closed at B and perfectly symmetrical round the axis AB, and the total weight of the tube be equal to the weight of water which is displaced when the tube is sunk to C, the weight of the different

sections along the axis A B can be so distributed that the tube will equally well remain in any other position, except it be so far turned over that the cylindrical part of the tube is lifted out of the water at one end and dipped at the other,

By making the spherical part of about the proportions of the figure, the tube can be depressed to within 25° of the horizon, and still remain in perfect equilibrium.

Now, suppose the tube to have a pair of trunnions attached at the water line, and these carried on a polar axis of, say, the English type (see Fig. 2), we have an equatorially-mounted telescope of any size, without any weight whatever on the bearings of the Dec axis, or, the tube may be lightened by an amount nearly equal to the weight of the polar axis, and there will then be practically no weight whatever on the bearings of that axis. So here we have a case of, say, an 80ton telescope mounted and carried by an equatorial, but without throwing any weight whatever on that equatorial; and the force necessary to drive the instrument is independent of the weight of the telescope, and dependent only on the friction necessary to be overcome in carrying the tube at an exceedingly slow rate through the

water.

Let us inquire into any possible disadvantages that may be urged against this form of mounting:

1. That the temperature of the water will often be different from that of the air; and consequently that there will be a detrimental mixture, at the mouth of the tube, of air from inside the tube, which will partake of the temperature of the water, with the outside air. This I would propose to avoid by making the tube double, with a space of some 3 inches between inside and outside tubes, hermetically closed except at the lower end, where there would be apertures in the inside envelope. The space between the two tubes would be connected through the trunnions with an air pump, worked by a gas or other motor, which would continually exhaust the air from between the two tubes, and thus cause a current of the outside air to pass continually down the tube and back to the pump by the space between the two tubes. This would keep the temperature of the inside tube and the air in the tube constant with that of the outside air.

2. The limited range of the equatorial. I have stated that the instrument would be in perfect balance down to 25° from the horizon. If desired, though no longer perfectly balanced, it can be used lower by employing a chain or wire rope connected between the lower end of the tube and the upper end of the polar axis, and the amount which the instrument would be out of balance, between 25° and 20°, would be very trifling.

Again, it will not be convenient to use the instrument within some 15° of the pole. It could be planned to go somewhat closer, but when it is considered that nine-tenths of the work required to be done can be commanded by this instrument, it is clearly better to design it to do that nine-tenths well than to strain it to doing another 5° that would only be useful on very rare occasion"

3. It may be urged that the friction of the water will prevent the rapid setting of the instrument. In a telescope of this size all the motions would be effected by motors of some description, guided by the observer from a commutator-board at the eye end, and there would be no difficulty in setting the telescope quite as quickly as could be expected considering its great size.

4. It may be objected that currents will be set up in the water by the moving of the telescope, which currents will affect the steadiness. No doubt this will be the case to some extent, but these will soon subside, and the motion necessary for following the stars will be so slow that no perceptible effect of this kind will be felt from it.

As to convenience in getting at the eye end, there need be no difficulty whatever in this form. As the eye-piece is only about 15 feet from the centre of motion, the movement of the observer is never more than 3 feet per hour. By means of a platform such as that shown in Fig. 2, running on rails, and quite independent of the instrument, the eye end is readily accessible at all times. To overcome the rotation of the tube as the instrument moves in right ascension, I would pierce the tube for eye-pieces every 30° round its circumference, and mount the flat mirror and cell in a collar so as to enable it to be readily rotated through intervals of 30°. By these means the image of the celestial object to be observed could be sent through either or any of the perforations of the tube, and the observer always observe in the direction most convenient to himself.

There are various difficulties about this construction which may naturally suggest themselves, but there are none, I believe, which cannot be overcome.

This is hardly the place to discuss details, but if there be any here who are sufficiently interested in this new form of equatorial to desire further information, I would refer them to an article in the present month's issue of Knowledge' which deals with most of the difficulties.

·

Putting aside now the question of reflectors versus refractors, there are some directions, applicable equally to reflectors and refractors, in which it is evidently possible to improve our designs for large equatorial instruments. It is not the first time that I have urged similar developments, but the advantages of what I recommended were not so apparent then as they are now in the present advanced state of astronomical research.

It may be remembered that when I lectured here in the year 1886 I strongly urged the desirability of employing some form of motive power for carrying out the various motions required in manipulating the large equatorials, domes, &c., and exhibited a model design I had made for the Lick Observatory, which illustrated the manner in which these various motors could be controlled by the observer, as well as the then newly devised lifting floor arrangement. The latter, that is the lifting floor, was, as you are aware, adopted by the trustees of the Lick Observatory, and I suppose we may assume that it was considered a success, as it has been copied in the case of