OPTICAL INSTRUMENTS AND SILVERING GLASS SURFACES. 419 
ordinary equatorial. If, however, in place of having the principal axis 
parallel to the axis of the earth we place it horizontally H. and W. and 
swing a telescope at right angles so as to describe a great circle 
through the pole, we get the ordinary transit circle, by which, up to 
quite recently, all the exact measurements which have given us the 
absolute places of the stars in the heavens have been taken. ‘These 
instruments, one of which is supposed to be a necessary equipment of 
every large observatory, have been made heavier and heavier and 
stiffer and stiffer by way of eliminating flexure and some other 
troubles ; but they have all a more or less inherent fault, the heavier 
you make an instrument the greater is the flexure and the more subject 
it is to fluctuations of temperature and the movements of the piers on 
which it is supported. The instrument, which Dr. Chandler has 
invented, differs most remarkably from the transit circle, when we 
consider that the object is the same, that is, to get the position of stars 
accurately defined—but the methods of working are widely different— 
instead of using a great circle determined by a horizontal axis on the 
earth, Dr. Chandler uses horizontal circles. He has a telescope not 
connected mechanically with the earth, but floating in a bath of 
mercury in such a way that if the telescope is pointed to any part of 
the sky and the whole turned round on the vertical axis, the telescope 
will sweep round on a small circle, whose centre is the zenith and thus 
any star can at the points of its intersection of this small circle be 
observed. 
The mechanical arrangements were then explained. There is no 
absolute contact between the telescope floating and this lower part, 
nothing beyond a couple of pins to hold it in its position. Now any 
mechanically-minded man would think, of course, that this is a very 
unsteady arrangement and the slightest thing would upset it, and when 
it turns round there seems to be a chance that it will not come to the 
same position, but strangely enough, Dr. Chandler, who does not seem 
to have been labouring under the same fear of preconceived notions 
that most.of us possibly would have been doing, put the thing to the 
proof, and he found as the result of experiment that the telescope 
absolutely pointed to some part of the small circle it originally pointed 
to when rotated in azimuth so accurately that there could not have been 
a deviation of one hundred thousandth of an inch between the ends of 
the float. ‘he accuracy with which the thing comes back to its 
original position with regard to the centre of gravity seems incredible. 
I know that some years ago when I was devising a gravitational 
instrument (I wish Professor Boys was not here) for I hardly like to 
talk of his own subject, and thought of a means of measuring a slight 
attraction by having a floating ball with a circumferential motion, I 
gave it up because I thought that surface tension would affect it. I 
should rather like to ask Professor Boys what would happen if one hag 
a ball floating in mercury, and on turning it the curves of the surface 
of the mercury when the ball cuts the surface vary. I have asked one 
or two people, but have not yet received a satisfactory answer. This 
is one form of instrument which, as I say, has been practically buried 
for the last 14 years. The first instrument which Dr. Chandler made 
