ON SPECTRUM MICROSCOPES. 19 



is very convenient, and with some modifications suggested by the author 

 may probably be much improved. Each observer should construct a table 

 applicable to his own apparatus, and describe the results in wave-lengths. 



The chief objection to such a form of instrument is the absence of a 

 fixed datum, and the necessity of verifying the adjustment by reference 

 to the sodium D-line, and also the fact that the measurements vary 

 with the focal adjustment. In order to overcome these difficulties, and 

 also to be able to measure with greater accuracy than is possible with 

 the above-named quartz scale, in which the bands are fixed, the author 

 has contrived an instrument, exhibited by Messrs. R. and J. Beck, 

 fully described in a paper in the Monthly Microscopical Journal for 

 1875, vol x i v -> P- 269- This consists of a piece of quartz, i inch long, 

 cut and mounted so that the light passes in the line of the principal 

 axis ot the crystal, along which there is no double refraction, but 

 circular polarisation. This is mounted between two Nicol's prisms, 

 one of which can be rotated along with a graduated circle, and the 

 other turned for adjustment. This gives a spectrum with seven black 

 bands, each of which moves into exactly the position of the one 

 above or below for each half revolution of the circle, which is so 

 graduated that it can easily be read off to one one-hundredth of 

 this half revolution. The author has constructed a table of wave- 

 lengths, as ascertained by using a diffraction spectroscope. By means 

 of this new apparatus it is easy to measure the position of the centre 

 of well-marked absorption bands to within one-millionth of a mille- 

 metre. The chief objection to the apparatus is that it is somewhat 

 large, but by placing it under the stage in the fitting made for the con- 

 denser, using the binocular spectrum arrangement, and placing the 

 object in front of the reflecting prism, very excellent results can be 

 obtained. It must, however, be admitted that it will be of use more 

 for ascertaining the general laws of spectra and the value of minute 

 differences, than for carrying out the ordinary kind of qualitative 

 analysis of colouring matters for which the spectrum microscope is so 

 well adapted. For this kind of research various branches of biology 

 furnish an almost boundless field, and promise most valuable results. 



Professor CLIFTON, F.R.S., after giving a sketch of the history of the 

 discovery of phenomena of interference of light, and of the progress 

 made in the investigation of this branch of optics, drew attention ta 



