32 C. M. CHILD. 



returned to water after a day or two and then brought into per- 

 manganate no traces of the reduction gradient have thus far been 

 found. The persistence of traces of the gradients for a short 

 time after action of a fixing agent can scarcely be due to the per- 

 sistence of a permeability gradient, even if such were present in 

 the living animal. Another interpretation appears more probable, 

 viz., that complete inactivation of the oxidizing enzymes does not 

 always occur at once in the fixing agents, consequently traces of 

 the original differential may persist for a short time. Individuals 

 killed before exposure to permanganate never become opaque 

 black but remain yellow or brownish indefinitely in excess of the 

 agent. Evidently killing before exposure decreases greatly the 

 reducing power of the protoplasm. 



The gradients of differential staining with neutral red and 

 the higher concentrations of methylene blue are visible only during 

 the early stages of staining in living material. As the staining 

 of living material progresses the regional differences in depth of 

 color gradually disappear and, so far as the eye can distinguish, 

 the staining becomes uniform throughout long before the depth 

 of color is sufficient to obscure differences. But if staining is 

 continued until cytolysis occurs the cytolytic gradients are the 

 same for any particular developmental stages as those observed 

 with other agents. 



The differential staining gradients may represent primarily the 

 differences in permeability of different regions to the dyes, but if 

 appearances can be trusted these differences in permeability dis- 

 appear as the toxic action of the dyes progresses and in spite of 

 the fact that -all regions seem to be stained uniformly, the regions 

 which are more susceptible to other agents and which show 

 greater reducing power are more susceptible to these dyes. Here 

 as in other forms, such differential permeability as may exist ap- 

 pears to be merely one aspect of the differences in physiological 

 condition which constitute the gradients. 



In the case of methylene blue the ability of the living proto- 

 plasm to reduce and decolorize the dye rather rapidly introduces 

 another factor and makes possible demonstration of the gradients 

 as gradients in rate of reduction and decoloration, particularly 

 in the blastula, planula and later stages, in which the rate of metab- 



