ZOOLOGY AND BOTANY, MICROSCOPY, ETC. 
813 
possible to insure tlie presence of rays near the red end of the spectrum 
which are serviceable in taking coloured preparations. Newcomb was 
amongst the first to undertake experiments of this kind. He mixed 
1 part of magnesium powder with 5-7 parts of pure nitrate of soda, 
and obtained thus an intensely yellow light. Rohrnann and Galewsky 
made many experiments with a number of different mixtures and obtained 
good results with the following receipt : — Mixture A. Magnesium, finely 
powdered, 9‘6 grm. ; potassium perchlorate, free from water, 13*8 grm. 
Mixture B. Barium tartrate, free from water, 5 • 7 grm. ; potassium 
perchlorate, free from water, 2*7 grm. 10 parts of A mixed with 1 part 
of B and 0*5 grm. of salt, free from water, added. From 1 to 3 grm. 
of this powder are used. Rohrnann and Galewsky also recommended 
other mixtures, in one of which acetate of copper was employed. 
As the result of a number of spectrograph ic investigations, the 
author comes to the conclusion that all complicated mixtures of salts 
of barium, copper, &c., such as these, must give place to the so-called 
smokeless flash-powder of Gaedicke. This powder consists of a mixture 
of magnesium and permanganate of potash which burns quickly, giving 
an intense light with little smoke. If the flash-light is taken with the 
spectrograph on an ordinary plate, not the slightest effect is shown in 
the red, yellow, and green, but some bright lines are produced on the 
border between the green and blue, joining on to the bright zone in the 
blue and violet. The effect is quite different on the erythrosin plate. 
In this case the bright zone begins already in the yellow by the 
Fraunhofer line D. In the centre between the lines D and E the 
silver deposit on the negative is very thick, and gives the impression 
that here there was more light effective than in the whole of the blue 
and violet together. Between the lines E and F the light effect 
gradually diminishes. In the blue and violet the effect is the same 
as on the ordinary plate. By using the erythrosin plate and inter- 
posing the yellow-green Zettnow filter, the blue and violet light is 
completely absorbed, and there remains only the strong maximum in 
the yellow-green between the lines D and E. These are exactly the 
relations which are wanted in photomicrography, and as they are found 
in sunlight. The maxima and minima of the light effect of this flash- 
light are distributed on the erythrosin plate exactly as with sunlight, 
only the maximum in the yellow green is much more intense. 
Coloured Photomicrograms * — MM. Lumiere, of Lyon, are the 
authors of a proc< ss for mechanically colouring photomicrograms. The 
best results have been obtained by the following methods. A carbon 
paper poor in colouring matter is chosen and sensitized in a solution 
of bichromate of potassium containing — water, 650 grm. ; bichromate of 
potassium, 25 grm. ; alcohol, 350 grm. After five minutes’ immersion 
the paper is dried and then exposed in the press. The duration of im- 
pression is determined by means of a photometer. The image is then 
developed on a thin ground glass by the usual methods. The positive is 
washed in cold water, immersed in alcohol for ten minutes, and finally 
dried. If properly done the proof is faint, sometimes scarcely visible. 
In order to colour it, solutions of the colours used in micrography, such 
* Bull. Soc. Belg. Micr., xvii. (1891) pp. 121-6. 
