350 



THE FOKMS OF TISSUES 



[CH. 



one-quarter of the original cell*. But while both the cyhndrical 

 wall and the spherical wall would be of less area than the plane 

 transverse partition after that limit (of one-quarter volume) was 

 passed, the cyhndrical would still be the better of the two up to 

 a further limit. It is only when the volume to be partitioned ofE 



* The principle is well illustrated in an experiment of Sir David Brewster's 

 {Trans. R.8.E. xxv, p. Ill, 1869). A soap-film is drawn over the rim of a wine- 

 glass, and then covered by a watch-glass. The film is inclined or shaken tiU it 

 becomes attached to the glass covering, and it then immediately changes place, 

 leaving its transverse position to take up that of a spherical segment extending 

 from one side of the wine-glass to its cover, and so enclosing the same volume of 

 air as formerly but with a great economy of surface, precisely as in the case of our 

 spherical partition cutting off one corner of a cube. 



