6 THE AMERICAN MONTHLY [Jan 



proceeding from the central red portion of the disc only, 

 and as already pointed out, when there is a large excess 

 of light of one color over that of another color,the stronger 

 swamps out the weaker. We now gradually close the iris 

 diaphragm, thereby shutting out more and more of the 

 green light, the color of the background will then be seen 

 to change to a fainter and fainter green and a point will 

 be reached where it appears neutral tinted or almost 

 white. Fig. 4. 



Not so, however, the object ; on this white ground the 

 diatom shines forth resplendent in the hues of red and 

 green more or less undiminished in intensity, the ridges 

 and higher structures appear green, the other parts red. 

 For it is clear tnough that the different parts of the dia- 

 tom will not have the red and green light falling upon 

 them or be throwing the light up into the objective in the 

 precise ratio to form white light. The ridges catch a deal 

 more oblique light, which happens to be green, and much 

 of the fiuer and more transparent structure catches and 

 passes up an excess of red light in the objective. Again, 

 wherever there are very fine perforations or holes in the 

 shells of the diatoms these appear pure red, because so 

 very little of the obliquely falling green light passes 

 through the holes into the objective, and thus to the eye. 

 This enables us to say at a glance and with certainty,that 

 such and such spots are perforations in a membrane, whilst 

 others are small raised prominences, etc., things which in 

 the ordinary way it is exceedingly difficult to determine, 

 and upon which many controversies have been held. 



For the best effects with the kind of illumination just 

 described (which fig. 5 illustrates diagrammatically), the 

 total utilized cone of light from the condenser should fill 

 from two-thirds to three-quarters the aperture of the ob- 

 jective. The way to see whether this is the case is to take 

 the eyepiece out of the microscope and look down the 

 tube, then we can see the back lens of the objective, and 



