> 



DEMONSTRATION OF AXIAL GRADIENTS. 135 



not affect the well-known fact that K 2 Mn 2 O 8 is readily reduced 

 by protoplasm to MnO 2 which produces a brown coloration of 

 the parts affected. The use of the permanganate for the purpose 

 of demonstrating the gradients in protoplasm is not dependent 

 in any way on Unna's methods and results and does not assume 

 their validity, but is based on the simple assumption that the 

 rate of reduction of permanganate and the consequent coloration 

 of the protoplasm may be expected to show some relation to the 

 physiological condition of the parts concerned, as regards their 

 metabolism, oxidative ability, reducing capacity or avidity or 

 demand for oxygen, in short, to some of the essential factors 

 concerned in the protoplasmic gradients. 



In most previous work with permanganate relatively high 

 concentrations have been used, e. g., I per cent, by Unna. In 

 such concentrations it is of course a powerful oxidizing agent 

 and extremely toxic, killing many of the lower animals almost 

 instantaneously. In my own work much lower concentrations 

 have been used ranging from m/i,ooo down to w/ioo,ooo, i. e., 

 0.0316 per cent, to 0.000316 per cent. 1 Even in a concentra- 

 tion of m/i,ooo, however, death usually occurs, at least in some 

 parts of the body, within a few minutes. Most of my observa- 

 tions have been made with concentrations of w/io,ooo or lower, 

 since it has been found that the regional differences in rate of 

 staining appear more clearly in the lower concentration, where 

 the reduction and death occur slowly and progressively, than in 

 the higher, where they are very rapid. In fact, when the or- 

 ganism is killed at once or very rapidly by the permanganate the 

 differences characteristic of the living animal do not appear at 

 all, or are very slight, and when it is killed by some other agent 

 before exposure to permanganate, the differences do not appear. 



The results thus far obtained with permanganate have not 

 only confirmed the results of the susceptibility methods, so far 

 as the same forms have been used, but in many cases the great 

 delicacy of the staining reaction has brought out very clearly 

 differences indistinctly or uncertainly shown by the cruder sus- 

 ceptibility methods with lethal concentrations. In fact the 



1 In making up solutions the formula K^MnjOs with molecular wt. 316 was used. 

 All concentrations are only approximate since solutions were not standardized. 



