128 Dr. G. C. Wallich on the Markings of the Diatomacea 



impossible to succeed. If we look upon an ordinary plano- 

 convex condensing lens, placed at a distance from the eye, in the 

 direct line of the source of light, provided no objects present 

 themselves in the background, the convexity is inappreciable. 

 The moment, however, that an angle is formed by the rays, the 

 convex character strikes us. And so with objects exhibiting 

 angular facets, although in a less easily demonstrable degree, 

 owing to the impossibility of receiving several sets of rays, each 

 parallel to one series of facets. Under the microscope, the same 

 principle holds good : facets are never seen by strictly direct 

 light. The mere placing the axis of the instrument in a line 

 with the source of light by no means fulfils this condition. In 

 viewing transparent objects, and more especially objects of such 

 high refractive power as the valves of the Diatomacese, oblique 

 light must be engendered. We must in this instance court 

 shado\\ s as our only means of grasping the reality. To exhibit 

 these much coveted " dots," shadows are still indispensable, but 

 here they are imperfect ones*. The greater the obliquity of 

 the illuminating rays, so long as definition i-emains perfectly clear 

 and free from coloured spectra, the greater will be the distinct- 

 ness with which the true character of markings will be seen. 

 Unless they are thus distinctly seen, appearances must prove 

 deceptive, and even the photographic picture may err in modo. 



In a valuable paper '^ On Species of Diatomacese,'^ under the 

 signature of one of our first authorities (Professor Walker- Arnott, 

 Journal Microscop. Science, vol. vi. p. 200), the following- 

 passage occurs : — 



" But as all Diatoms, with strise composed of dots, have four 

 rows of striffi (two diagonal, one horizontal, and one longitudinal), 

 and as the visibility of each depends, when delicate, on the posi- 

 tion the valve presents to the illuminating pencil of oblique 

 light, the closer or more difficult strise are sometimes seen when 

 the others are not, and thus may be occasionally mistaken for 

 the predominating or coarser ones, which alone are made use of 

 in specific characters. I may here remark that, when the dots 

 are so placed as to form rectangles, the transverse and longitu- 

 dinal striai are always the most remote, and therefore predomi- 

 nate; and it is generally supposed that when the dots are 

 quincuncial, the diagonal lines are always most apparent ; but 



* In a paper by Hugo von Mohl (in the December Number of the An- 

 nals and Magazine of Nat. History, p. 444) the author is at pains to j)rove 

 tliat polarized light has been neglected heretofore in microscopic examina- 

 tions ; and, in j)ro()f of the advantages to be derived from its employment 

 under scientific adjustment and management, he states tliat the markings 

 on Pleurnsigma angulatum may actually be sliowu to consist of " six-nkled 

 dots." 



