154 JOHN N. LOWE 



cation of potassium, it is unqualifyingly modified by its anion or 

 the residual part of the undissociated molecules. 



Another interesting feature observed was that after a longer 

 or a shorter interval after the first contraction there followed a 

 peripheral expansion of the pigment cells (figs. 6, 7, 8, 9 and 10), 

 that is, the pigment cells put out processes which became 

 longer and longer as time went on but which never reached the 

 original size they had before treatment with the potassium salt 

 solutions. This expansion set in earlier in KI where the contrac- 

 tion took place first, evidently the secondary expansion or pa- 

 ralysis is reciprocal of the first contraction. The expansion is 

 in the order of the first contraction (figs. 6, 7, 8, 9 and 10). 



I> N03> Br> Cl> SO4 



This peripheral migration of the pigment is in the nature of 

 a paralysis. The paralytic state (depression) is soon followed by 

 death of the pigment cell. The walls of the pigment cell dis- 

 integrate and the pigment granules flow into the interspaces of 

 the body tissues. Death of the cells takes place often before 

 the expansion is complete, and then premature disintegration of 

 the pigment cells occurs. The condition or extent of the de- 

 generation is dependent upon the 'physiological state' of the 

 melanophores and the individual fish. 



The maintenance of the irritability of the melanophores fol- 

 lowed the same order, correlated with this was the longevity of 

 the fish. The fish lived the longest in K2SO4 and KCl. They 

 died very rapidly in KI. 



The reactions varied with the concentration of the solutions, 

 for in solutions of 0.1 M or less the changes were slightly slower. 

 Molecular solutions gave no results but killed the fish imme- 

 diately. 



2. Effects of sodium salts. Here as in the potassium salts the 

 embryos used had their melanophores expanded. It was ob- 

 served that the neutral salts of sodium produced a contraction 

 of the melanophores very slowly. In some instances the contrac- 

 tion did not take place in 92 to 116 hours, especially in the solu- 

 tions of Na2S04 and NaCl. The contraction in Nal was complete 



