ZOOLOGY AND BOTANY, MICROSCOPY, ETC, 95 



fusion of two views renders the superposed projections absolutely inde- 

 pendent of each qther, because they appear in relief. The transforma- 

 tion of the superposed orthogonal projections into stereoscopic views is 

 very simple, because in constructing each contour curve, it is only 

 necessary to concern oneself with the scale of reduction, and with the 

 position of the centre of the sheet of paper containing the projections. • 

 Each contour curve being, in regard to a surface immediately under it, 

 supposed brought near to the observer by a quantity proportional to the 

 graphic equi-distance (a quantity chosen arbitrarily once for all), the 

 scale of reduction is given by the formula a//, in which a is a variable 

 equal to the distance of the sheet of paper from the observer's eyes, 

 and / a constant equal to the distance between the observer's eyes and 

 the stereoscopic views. As to the centre M of the sheet of paper, it will 

 always lie at that normal to the stereoscopic views which passes through 

 the middle of the line joining the optic centres 0, 0' of the two eyes. 

 'Consequently its projection in is always in the plane xy passing through 

 the centres S, S' of the stercvoscopic views. Moreover, it is as much to 

 the left (for the right eye) or more to the right (for the left eye) of 

 S and of S' as the variable a is proportionally smaller. Its position may 

 be obtained either by construction (intersection of D M and of 0' M 

 ■with xy) or by the calculation m^ = cif/a, where 6? is a constant equal 



0' 



to . The construction of the two views is so much the more 



2 



simplified, inasmuch as the projection of each curve intended for the 

 left eye is rigorously superposable on the corresponding projection 

 intended for the right eye. The only difference between two corre- 

 sponding curves is that the projection of the centre M, instead of having 

 the same relation with regard to S and to S', is placed symmetrically 

 with regard to the middle of the straight line joining these two pdints. 

 In practice there is advantage in drawing much enlarged the stereoscopic 

 images and in reducing them photographically. There is also a gain in 

 not putting the two corresponding views on the same mount : partly, 

 because each observer can then place them at the most favourable 

 separation, and, partly, because the position of the two views can be also 

 arranged for inversion, which will sometimes be more convenient for the 

 •observation of certain details. 



(5) Microscopical Optics and Manipulation. 



Spherometer of Precision.* — The chief feature of this instrument, 

 •says J. Guild, is the method employed for detecting the exact contact 

 between the micrometer screw and the surface under test. The micro- 

 meter terminates in a small sphere of about 1 • 5 mm. diameter. A 

 microscope with a suitable illuminating apparatus is mounted above, 

 and the Newton's rings surrounding'the point of contact are observed. 

 By watching the behaviour of the rings when the screw is brought up 

 the exact point of contact is determined. The sensitivity is about one- 

 ten-thousandth of a millimetre. 



* Trans. Optical Sec, January 10, 1918. 



