ZOOLOGY AND BOTANY, MICROSCOPY, ETC. 765 



consists of a small pulley with a V-groove being forced by a spring to 

 bind on a round steel rod, which is screwed to the main pillar. This 

 results in a far nicer movement than the rack-and-pinion gives, since a 

 geometric fit is obtained, and there can be no play or backlash. The 

 movable carriage carries through a ball-joint an arm, at the far end of 

 which the lens-holder is mounted, a ball-joint being used here also. 

 The total length of the arm from the centre of the pillar to the centre 

 of the lens is 2:! : (j cm. 



Koristka's Large Model I.*— This stand (fig. 119) is specially con- 

 structed for photomicrography. The upper part is inclinable to 90°, and 

 can be fixed in any position by means of a clamp. There is a mechanical 

 stage with ample vertical and horizontal movements, and the substage 

 arrangements are very complete. 



Reichel, C. — Das Mikroskop als Hifsmittel in der Werkstatt. 



[The author points out how a Microscope of small magnifying power (10 

 diam.) may be made very useful in checking the accuracy of fine machine 

 work.] Deutsch. Median. Zeitschr., 1909, p. 1. 



(2) Eye-pieces and Objectives. 



Improved Triple Object-glass.f — Although J. W. Gilford designed 

 this object-glass for a telescope, yet the method of its construction will 

 be interesting to microscopists. The principle, briefly put, is the con- 

 struction of a single compound lens out of three varieties of glass so 

 selected that three rays of the solar spectrum (B, D and C) are com- 

 bined, and that therefore perfect achromatism as regards those rays is 

 obtained. An essential preliminary was the determination of the 

 refractive indices of suitable glasses to the utmost accuracy. It was 

 found that indices extending to five places of decimals were insufficient, 

 and the author carried his observations to seven places. 



The author had already previously described^ a method by which 

 the optical constants of glasses can be very accurately measured. To 

 obtain the refractive index it is necessary not only to measure the 

 deviation of the ray from the normal due to the action of the angle of 

 the prism through which it has passed, but also to measure the angle 

 itself, an operation far more liable to error than measuring the devia- 

 tion. To avoid this the prism is polished on all three sides, the 

 deviations are measured at all the angles, are added together, and the 

 mean deviation, on the principle that the three angles of every triangle 

 are always equal to two right angles (180°), may be used for the 

 purpose of calculating the refractive index as being the deviation for an 

 angle of 60°, although no angle has actually been measured. 



The method is not absolutely correct, but the average error due to 

 the method is less than unity in the seventh decimal place. This error 

 is caused by the three angles not being quite equal to one another, but 

 as, with measurements by the ordinary method, it is difficult to avoid 

 error even in the fourth decimal place, it will be readily seen that for 

 all practical purposes considerably greater accuracy is obtained. The 



* Koristka's Catalogue, xiii. (1908) pp. 8 and 9, fig. 3. 



+ Monthly Notice. Hoy. Astro. .Soc, lxix. (1908) pp. 118-25 (1 pi. and 2 figs.). 

 See also The Observatory, Jan. 1909, pp. 41-2. 

 % Proc. Roy. Soc, Feb. 13, 1902. 



