136 C. M. CHILD. 



permanganate method makes directly visible gradients such as 

 the gradient in the unfertilized sea-urchin egg, which was not 

 very satisfactorily demonstrated by the direct susceptibility 

 methods (Child, '160), but the existence of of which was inferred 

 from the differential inhibition of development resulting from 

 exposure of the unfertilized egg to inhibiting agents (Child, 

 'i6c). In various cleavage stages, hydroid planulse, and echi- 

 noderm blastulae and gastrulse the color gradient is very distinct 

 and uniform, even within the limits of a single blastomere in the 

 earlier stages. In short the method is one of great beauty and 

 delicacy and so entirely simple that it can readily be used for 

 classes or lecture demonstration. 



Practically the only precaution to be observed is to provide 

 for uniform distribution and supply of permanganate to all part 

 of the surface of the organism. The permanganate of course 

 gradually disappears from the solution and if the volume of 

 solution is not very large as compared with the protoplasmic 

 surfaces it must be renewed from time to time. Moreover, in 

 the case of organisms lying undisturbed for any considerable 

 time on the bottom of the dish, the surface next the glass always 

 stams less rapidly or less deeply than other regions. These 

 difficulties are readily avoided by slight agitation of the solution 

 at short intervals. Ciliary movement soon ceases in all except 

 very low concentrations of permanganate so that even with 

 motile developmental stages and free swimming protozoa as well 

 as with non-motile forms or stages, frequent agitation of the 

 solution is desirable. 



Besides being useful as a staining agent for demonstrating the 

 gradients as color gradients, permanganate can also be used, in 

 at least many cases like KNC and other agents to demonstrate 

 differences in susceptibility as gradients in death and disinte- 

 gration. Thus far it has been less used in this way than as a 

 staining agent, but it has been found for example that in various 

 ciliate infusoria and early developmental stages of certain forms 

 very low concentrations such as m/ 50,000 brjng about death and 

 disintegration before staining occurs. In all such cases observed 

 the disintegration gradient is similar to the staining gradient. 

 In the higher concentrations there is apparently more or less 



