526 



NATURE 



\Oct. 24, 1872 



bright field illumination, and dark field illumination of micrometer, 

 are shown. 



The position circle illumination is very simple (see Fig. i), a 

 single rellector R, attached to the inside of the tube directs a 

 constant beam of light on the back of the glass circle at r. 



The bright field illumination is effected by a very small cen- 

 tral rellector, r', which sends the light directly into the field of the 

 micrometer. 



This method is, I believe, now generally considered to give 

 the best results, and has, as far as I am aware, but one disad- 

 vantage, viz., that the arm which supports the small mirror 

 produces a little diffraction, and consequently deterioration of 

 definition. 



This objection I have in some measure reduced by making the 

 arm and mirror removeable at pleasure by pulling or releasing a 

 string, so that while actually observing, it can be removed and 

 replaced instantaneously. 



In devising the dark field illumination, I started on the hypo- 

 thesis that there were two essential points to keep in view, viz., 

 that the lines should be illuminated on both sides (not one), and 

 that the angle at which the light sliould be thrown upon the 

 wires should be very great, so that the blackness of the field as 

 seen through the eye-piece should not be injured. 



I found that the best result is obtained by placing four prisms 

 of total rellcction round the field of the micrometer, just behind 

 the wires, and of such an angle that the light thrown upon them 

 should be refiected upon the wires at an angle such as is shown 



in the diagram Fig. 2, where \v is the position of the wires in 

 the focus of the objective. 



In order that this scheme of illumination should be carried out 

 effectually from the light ofa single lamp hanging on the declina- 

 tion axis, it is necessary that a certain annular portion of the micro- 

 meter which embraces these prisms should be constantly illumi- 

 nated from this lamp, and this is effected in the following way : 

 a portion of the sliglitly divergent beam of light, shown in Fig. 2, 

 proceeding from the lamp on thedeclination axis, is passed through 

 a very low power convex lens, /, which renders the beam slightly 

 convergent. 



This is not necessary, but a mere matter 01 convenience, as it 

 reduces the necessary size of the reflector and lens afterwards re- 



quired. The light is now taken up by a reflector, R, within the 

 tube, and directed towards the eye-end at such an angle that it 

 crosses the axis of the telescope just at the inner end of the eye- 

 piece tubes, X ; hence it is passed through a piece of glass of a 

 peculiar shape, r r, which I call, for want of a better name, an 

 annular prism lens. This piece of glass has a hole cut in 

 it large enough to admit the whole pencil of light from the object 

 glass. 



The use of this annular prism lens is twofold: — 

 1st. It has to alter the direction of the beam of light be- 

 fore diagonally thro%vn across the tube, RX, to that parallel to the 

 axis of the telescope : and 



2nd. It is necessary that it should have a slightly converging 

 effect to reduce the size of the illuminated circle it produces. 



