116 Angular Ajierture of Ohjed-glasses. 



object-glass the margin is so feeble as almost to forbid the measure- 

 ment of its boundary. 



I first tried the plan described by myself many years ago. I 

 got a block of flint glass of the same refractive power as hard Canada 

 ■P,^ o balsam. This I had no diffi- 



culty m nndmg (the specinc 

 gravity was 3'1). Fig. 3, 

 is the glass, 2^ inches wide ; 

 the side h was polished, and 

 the opposite one, c, greyed 

 and divided into degrees. As 

 the divisions were on a line, 

 they were of course irregular ; 

 but this caused no dilBBculty, 

 as they were lined off from 

 a dividing circle. Half-way down the exact centre of polished 

 edge there was a fine diamond cut, from which point the divisions 

 were lined, and on which the object-glass to be tested for immersed 

 aperture was focussed. As the refraction of the glass is the same, it 

 must represent the balsam angle, which can be read off by the extent 

 of the light on the division scale of degrees. It made no difference 

 in the angle whether water is admitted in front lens or not (as in 

 theory it should not), because the medium is in the form of a 

 parallel plate, but in each case the object-glass must be focussed on 

 to the glass surface. The measurements obtained this way were on 

 too minute a scale to be very reliable. I therefore adapted a semi- 

 cylinder of glass, of density 3 • 1 and 1^ diameter, with polished 

 edges, such as has been used in the instrument for demonstrating 

 the law of Descartes. The centre of radius was exactly in the 

 polished plane, and this point indicated by a line ruled by a 

 diamond, upon which the object-glass was focussed. A minute stop 

 of leaf metal was placed in the centre, so as to cut off extraneous 

 rays. The semi-cylinder was fixed in the stage of a microscope 

 which had a rotating base divided into degrees. The body was set 

 horizontally, and a lamp placed some two feet away. The index 

 was set with a half-luminous field, and the object-glass focussed 

 carefully on to the glass surface. Unless this is done there will be 

 considerable error in the measurement. The microscope is next 

 rotated, and when the light again bisects the field, the degrees are 

 read off. By these means the balsam aperture ivith closed lenses 

 was only 68'', instead of 98° as stated ; and this angle of 68° was 

 the same whether water was introduced or not between the plane 

 surface of front lens and semi-cylinder, taking care to focus in either 

 case. The principle of operation was identical with that of the 

 ordinary sector, to the moving index aim of which the hemisphere 

 may be supposed to be fixed, and the focal front of object-glass a, 



