



ZOOLOGY AND BOTANY, MICROSCOPY, ETC. 713 



2. In the Bar ml plate, action begins about the same place and 

 extends somewhat further each way. 



3. In the Isolar, action begins in the blue and indigo, where it is 

 sharply sensitive, but is feeble in the yellow. With a prolonged ex- 

 posure its sensitiveness reaches to about 596 in the orange. 



4. In the Isochrom, a very rapid plate, fairly sensitive in the yellow, 

 blue and indigo with a short exposure, action reaches to nearly 610 in 

 the orange. 



5. In the Lumiere Pantachromatic, action starts away strongly and 

 extensively in the blue and indigo, though faintly in the orange and 

 yellow, and very weakly in the green. With exposure it reaches to 

 nearly 650. 



6. The Mawson plate is not unlike the Barnet, but it is a trifle 

 quicker and a little more evenly sensitive. 



7. In the Cadett plate we seem to have the greatest evenness of any 

 isochromatic on the market. Action commences over nearly the entire 

 range of the spectrum at one and the same moment, even with the 

 exposure of half a second. It extends with a long exposure far into 

 the ultra-violet, and to 650 at the red end. 



8. The Flashlight plate is merely given as an example of the limited 

 sphere of sensitiveness in the ordinary unstained plate. Action com- 

 mences in the blue and violet nearly evenly, but no amount of reasonable 

 exposure will produce effect much further than about 550. 



Suppose that it is desired to photograph a blue-stained bacillus on 

 a white ground. Then in order to increase contrast between the two 

 colours in the photographic image, the brilliancy of one or the other 

 must be weakened so as to affect the emulsion less. This can be done 

 by staining the screen with such a colour that the blue is obliterated, 

 i.e. that it becomes black to the " eye " of the plate — but the plate 

 must be sensitive to the colour itself. 



The three leading dyes for which we have to find contrasting screens 

 •are : Loffler's blue, gentian violet, and carbolfuchsin. Plate XIII. gives 

 a spectrograph of Loffler's blue ; and figs. 4, 5, 6 show the absorption 

 bands peculiar to it, with short, medium, and long exposures. It will 

 be seen that this dye transmits light as far as 500. If we now try and 

 use a green pot-glass screen (much recommended for general use by 

 the author, three exposures of which are shown in figs. 1, 2, 3, 

 Plate XIII.), it will be seen that the glass transmits light from 474 to 

 5S0 or 590. .There is, therefore, an ^overlap, through which a con- 

 siderable amount of light passes (figs. 7 and 8). Snch a combination of 

 dye and screen is therefore useless. 



Amongst the many dyes tried, Aurantia was found to be the best. 

 Its absorption bands are shown in figs. 9, 10, and 11. In the longest 

 exposure (No. 11) only light between 510 and 632 is transmitted, 

 so that if this screen be used with the blue (fig. 15) there is just a 

 margin of safety. Orange F is almost as good, but with a fairly long 

 exposure there is a suspicious leakage at the position of the arrow. 



Plate XIV. deals with gentian violet, figs. 1 and 2 showing its two 

 spectrographs. A single thickness of green pot-glass (fig. 4) is not 

 dense enough to cut off all the light ; but two thicknesses make it a 



