110 The Structure of Eupodiscus Argus. 



theoretical with the experimental results is shown to be absolutely 

 perfect. 



If the glass tested should possess the extreme aperture (170°, or 

 upwards), the fluid balsam used in the experiment would give the 

 reduced aperture within a few minutes of 83° ; the reduced angle 

 changing much more slowly than the angle in air when near its 

 limits. 



It is to be observed that the investigation given above shows 

 — what before examination might not be suspected — that the 

 results are entirely independent of the kind of glass used for the 

 objective-front. The index of the glass occurring both directly and 

 inversely disappears from the compound ratio ; so that the reduced 

 aperture remains exactly the same, whatever may be the nature of 

 the glass, or the value of its refractive index. 



IV. — The Structure of Eupodiscus Argus. By Samuel Wells. 



Plate XI. (Lower portion). 



The elucidation of the Eupodiscus Argus given by Mr. Slack in 

 your December number did not agree with my previous observa- 

 tions ; but as I had a supply of specimens obtained from sea-weed 

 washings in Buzzard's Bay, on the south coast of this State, I made 

 further examinations to see if I could discover the appearance 

 represented by Mr. Slack. I looked at different valves dry and in 

 balsam, covered and uncovered, using Beck's parabolic reflector, as 

 well as Professor Smith's arrangement for opaque objects, the use 

 of which was recommended by Mr. Slack in his paper. 



The valve of this diatom is remarkable for its opacity, its thick- 

 ness being about t^W ; it presents, therefore, a beautiful appear- 

 ance as an opaque object with a binocular. The structure of the 

 outer or convex surface can be readily made out with a low power. 



It is dotted with depressions irregular in size, shape, and 

 arrangement ; between these depressions the surface rises in ridges, 

 which glisten and sparkle like fresh snow. No arrangement of 

 light (except transmitted) varies this appearance. The depressions 

 are unmistakable, and, as appears by the use of the binocular, and 

 the examinations of the edges of fragments, are pockets extending 

 nearly, but not quite, through the valve. In Fig. A I have 

 endeavoured to represent the upper surface of a fragment, and in 

 Fig. C a vertical section. 



The average diameter of these depressions is about ^Vr". 



The inner or concave surface is much more difficult of reso- 

 lution; its structure is quite different to that of the convex 

 surface. It is nearly smooth, has no ridges, and (probably) no 



