M.-P. VOL. L] CRAWFORD CONSTANT OF REFRACTION. 107 



are two pairs of threads in every micrometer, but one of 

 which is generally used. 



There is a separate broken telescope for setting. This 

 is supported on wyes attached to either pier and is at the 

 level of the lower rim of the circle. By means of this the 

 circle can be seen either from the north or from the south, 

 so that the settings can be made very conveniently. 



The illumination for both the field of view and for the 

 circles under the microscopes is furnished by a Rochester 

 lamp placed in a cylindrical case 9 feet from the axis of the 

 instrument. This light also illuminates the heads of the 

 microscope micrometers. Most of the heat from this lamp 

 is carried out of the room by a pipe which extends from 

 directly over the lamp through the roof to the outside air. 



A simple mechanism enables the observer to change the 

 system of illumination from a bright field with dark wires to 

 a dark field with bright wires and vice versa; he can also 

 reduce the amount of illumination at will. 



The brick piers supporting the instrument are 34 inches 

 by 44 inches at the floor of the room and 22 inches square 

 at the top. The sides next to the telescope are vertical. 

 They are cased in wood with a layer of felt between the 

 surfaces. The platforms for the microscope reader are 

 entirely disconnected from the casing of the piers. 



The microscope bearers are 23 inches in diameter and 

 17 inches long. The wyes for the pivots of the instrument 

 are attached to the inner faces of these frames. 



The weights of the counterpoises hang from levers 

 26 inches long. The fulcra are in the centers of the levers 

 and are 6 inches from the inner faces of the microscope 

 bearers. 



Two collimators, of same aperture and focal length as the 

 Meridian Circle, are suitably mounted. The collimator 

 micrometers are 35 feet 6 inches apart. 



2. The Room. The Meridian Circle house on Mount 

 Hamilton has been most admirably designed. Its efficiency 

 will become apparent from the meteorological data to be 

 given later. 



(2) May 5, 1903- 



