CF THE CHARACTERISTICS OF THE DIATOMACE^. 45 



long, the markings lose their sharpness to a great extent, 

 owing to a kind of glazing operation, which follows on the 

 melting of the alkalies contained in the water. On this 

 account distilled water should be employed, so as to avoid 

 the presence of calcareous salts as far as possible. 



After this preparation, the frustules may be placed under 

 the microscope, where a most astonishing sight awaits the 

 eye. So wonderfully minute, however, are the markings 

 brought out by the incineration of the vegetable matter, 

 that, except in a few rare instances, they escape detection 

 altogether when the transmitted light is direct ; that is to 

 say, when the rays falling upon the object are parallel to 

 the axis of the instrument. In most kinds of Diatomacese 

 the light must be directed obliquely to the lines of striae, 

 before their minute structure becomes visible even with 

 the highest powers. Oblique light, it may be observed, is 

 obtained by thrusting the mirror to one side of the axis 

 of the microscope. The diaphragm should be removed, to 

 allow free passage to the rays, when the latter will fall, so 

 to speak, upon the edges of the lines, and render them 

 visible. If this proceeding is skilfully managed, the whole 

 system of lines, of which some species possess several, 

 running in different directions, is brought under the eye of 

 the observer. 



It is well here to remind the student, that, in order to 

 sharpen his experience and to accustom himself to the 

 detection of these minute points, he should provide himself 

 with what are commonly termed ' test objects,' of which the 

 Diatomaceous family offer some excellent examples. For 

 the lower powers, those magnifying from 200 to 220 times 

 linear,* Pleurosigma attenuatum forms a good test; for 



* 'When one dimension only of an object is taken into account, 

 viz. the breadth or diameter — and this is the ordinary manner in 

 which the magnifying power is taken — objects are then said to be 

 magnified so many diameters, or so many times linear. But objects 

 are really as much magnified in the other dimension, or in their 

 entire surface; so that the true expression of their amplification 



