1916] on The Movements of the Earth's Pole 679 



for solving the riddle of the Kimurti variation, and other places should 

 join in the attack. Unfortunately the observations are very 

 laborious and require the almost exclusive attention of an observer. 

 There is therefore a very real want of an instrument which shall 

 demand something short of the whole time of a skilled astronomer. 

 With this object, and also with the intention of eliminating certain 

 sources of error, instruments of new form have been devised. A 

 short account of these will be of interest. 



1 shall not here attempt any description of the methods of 

 observation used. It will be sufficient to say that, as what we want to 

 find is the direction of the zenith at the place, all methods ultimately 

 depend either upon a level, giving us the horizontal plane, or upon 

 a plumb line, giving us the vertical, and that of these two the level 

 is the one that has almost exclusively been employed by the 

 astronomer. The level is an instrument capable of a high degree 

 of precision but it has the disadvantage of being very susceptible to 

 temperature changes and, as both the glass tube of the level and the 

 spirit with which it is filled are bad conductors of heat, it is 

 impossible to ensure that it is at an even temperature throughout. 

 Irregularities are thus produced which the reading of both ends of 

 the bubble only partially eliminates. The mere fact of an observer 

 standing near a sensitive level to read it may seriously vitiate its 

 accuracy. 



Some of these errors may be avoided, and such errors as are due 

 to faulty reading of the level graduations by the observer entirely 

 eliminated, by making the level an integral part of the instrument 

 by floating the whole in liquid. The first application of this principle 

 to an astronomical instrument was by Chandler, who carried out his 

 series of latitude observations, already mentioned, with an almucantar, 

 being a transit telescope floated in a trough of mercury. The name 

 almucantar means a small circle of the heavens parallel to the horizon, 

 and it will be sufficiently obvious that if the telescope can be set at 

 any angle with the float, then as the instrument is rotated in the 

 trough, or the whole trough itself is turned, the line of sight of the 

 telescope will move round such a circle. With this instrument 

 the stars are observed, not as in a transit circle crossing a vertical 

 line, but crossing a horizontal circle of constant altitude. For con- 

 venience of calculation this horizontal circle would generally be 

 selected as that through the celestial pole at the place. Chandler's 

 instrument was purposely designed so as to differ as little as possible 

 from the ordinary visual type, and must have been a most difficult- 

 instrument to use. The fact that he got such excellent results from 

 it is no small tribute to his manipulative skill. The use of this form 

 of instrument cannot be said to have found great favour among 

 astronomers ; there is only one example of it in this country, and, as 

 far as I know, none on the Continent. The one we have is at the 

 Durham University Observatory, and was designed by the present 



