377 



without both these condensers. He did not advise the use of chromatic con- 

 densers, except with oblique light. Paraboloids and spot lenses might be 

 regarded as altogether out of date, being quite superseded by condensers. 

 Diffused daylight was suitable for low powers up to a § , a concave mirror 

 ground glass, and a diaphragm being used. A great falling off in sharp- 

 ness would be experienced with powers higher than a f, unless a con- 

 denser were used ; even then daylight was so uncertain that it was im- 

 possible to test objectives by it, that is to say, the best picture an 

 objective is capable of producing cannot be obtained by daylight. Sunlight 

 from a heliostat, used obliquely, gives the strongest resolutions known. For 

 general work he recommended the light from a common paraffin lamp with 

 a ^ inch wick. In the case of transmitted light the picture is that of the 

 source of light interrupted by the objjct ; it is, therefore, of the first 

 importance that the light should be as pure as possible. Pure light, 

 with the ordinary lamp chimney, is hardly practicable. The flame being 

 more or less distorted by the specks, striae, and irregularities in the curva- 

 ture of the glass, as well as by the reflections from the inner surface of the 

 chimney. To obviate this he had a metal chimney made by Swift, blackened 

 inside, the glass part being an ordinary 3x1 slip. He then insisted that if 

 the finest results were wanted, it was absolutely necessary to focus, by 

 means of the condenser, the image of the source of light on to the object, 

 keeping the condenser accurately centred to, and the direction of the 

 illuminating beam in, the optic axis. It was of no importance whether the 

 light came from the lamp direct, or was reflected from the plain mirror, pro- 

 vided that there were no secondary reflections. Secondary reflections in a 

 plain mirror could be got rid of by rotating the mirror in its cell. For 

 direct transmitted light he considered that from a paraffin lamp with a tj- 

 inch wick the best, and it would give enough intensity for all ordinary re- 

 solutions with oblique light ; but for special resolutions a lamp with a 

 larger wick, lime light, or electric arc lamp might be used. The incan- 

 descent lamp for microscopical purposes was quite useless. It was not 

 intense enough for oblique light, and for direct light the image of an in- 

 candescent filament was not suitable to view an object in. For low power 

 w r ork a larger surface would be required feo be illuminated than that covered 

 by the edge of the flame. The best way, he thought, of doing this was to 

 expand the image of the flame into a disc by means of a bull's eye. The 

 bull's eye, like the condenser, must be accurately focussed and centred on 

 the edge of the flame. Those blue patches and irregularities of illumina- 

 tion which so often troubled microscopists were due to not taking sufficient 

 care in focussing and centring the bull's eye. The intensity of the light 

 was regulated by the angle of the bull's eye. Mr. Swift made one for him 

 on Herscbel's formula, viz., a meniscus and a biconvex, a form long used in 

 first-class magic-lanterns. It has the advantage that by taking away the 

 meniscus the focal length of the bull's eye was increased, and its intensity 

 greatly diminished. Great intensity would be required in crossing difficult 

 diatoms with a double slot, whereas an object such as a heliopelta, with a 

 f, would be better seen with less intensity. The lamp w r ith the bull's eye is 



