220 THE FORMS OF CELLS [ch. 



a little salt to the lower layers of water, the drop may be made 

 to float or rest upon the denser liquid. 



We have already seen that the soap-bubble, spherical to begin 

 with, is transformed into a plane when we relieve its internal 

 pressure and let the film shrink back upon the orifice of the pipe. 

 If we blow a small bubble and then catch it up on a second pipe, 

 so that it stretches between, we may gradually draw the two pipes 

 apart, with the result that the spheroidal surface will be gradually 

 flattened in a longitudinal direction, and the bubble w^U be trans- 

 formed into a cylinder. But if we draw the pipes yet farther 

 apart, the cylinder will narrow in the middle into a sort of hour- 

 glass form, the increasing curvature of its transverse section being 

 balanced by a gradually increasing negative curvature in the 

 longitudinal section. The cylinder has, in turn, been converted 

 into an unduloid. When we hold a portion of a soft glass tube in 

 the flame, and "draw it out," we are in the same identical fashion 



Fig. 62. 



converting a cylinder into an unduloid (Fig. 62 a) ; when on the 

 other hand we stop the end and blow, we again convert the 

 cylinder into an unduloid (b), but into one which is now positively, 

 while the former was negatively curved. The two figures are 

 essentially the same, save that the two halves of the one are 

 reversed in the other. 



That spheres, cylinders and unduloids are of the commonest 

 occurrence among the forms of small unicellular organism, or of 

 individual cells in the simpler aggregates, and that in the processes 

 of growth, reproduction and development transitions are frequent 

 from one of these forms to another, is obvious to the naturalist, 

 and we shall deal presently with a few illustrations of these 

 phenomena. 



But before we go further in this enquiry, it will be necessary 

 to consider, to some small extent at least, the curvatures of the 

 six different surfaces, that is to say, to determine what modification 



