GRANULAR APPEARANCES 29 



this phenomenon, as has been said, is only to be observed 

 by focussiag somewhat below the middle line. An explana- 

 tion of it may be found in the following considerations. If 

 small, closely packed air-bubbles in a thick solution of gum 

 be studied under the microscope with a low power, by focussing 

 as sharply as possible on the equator of an air-bubble it is 

 seen that in the dark marginal zone, where each bubble 

 touches a neighbouring one, a clear spot of light makes its 

 appearance, so that in two neighbouring bubbles these spots 

 always stand exactly opposite. Nageli and Schwendener 

 have already (1865) made a theoretical investigation into 

 this phenomenon, and have referred it to the reflection of 

 light at tl^e underside of the air-bubbles. If now the tube 

 be lowered slightly from the middle focus, the clear spots 

 decrease iu brightness, but become, on the other hand, 

 broader, so that they traverse the whole dark marginal zone 

 as radial light bands ; simultaneously the bands of 

 neighbouring air-bubbles ws-A-m's to one another become 

 confluent, so that one obtains a kind of clear net, which is 

 stretched out at the point of contact between the air- 

 bubbles. Now since in our froth lamella we have before us, 

 in the same way, very closely crowded droplets of a less 

 refractQe fluid in one more strongly refractile, something 

 similar must take place. The clear, radial bands produced 

 by reflection of the light will be relatively very broad here, 

 since the diameter of the vesicles is very small. These 

 clear bands naturally fall on the middle portions of the 

 edges of the alveoli and make them lighter, while the nodal 

 points become no lighter, and hence appear relatively dark 

 by the effect of contrast. 



I take this opportunity to state that in successfully 

 prepared good froths I could not demonstrate any granular 

 structures whatsoever, which by their deposition had in any 

 way added to the actual distinctness of the nodal points. 

 It was rarely that I obser^-ed in some froths rounded bodies, 

 more or less scattered, of about the size of a small alveo- 

 lus, and hence having nothing to do with the nodal 

 points. 



Since the problem with regard to the true structure of 



