546 R. A. SPAETH 



4 P.M., 5 P.M., and 6 p.m. Similar sets were killed. In every case the 

 scale having the most widely expanded melanophores was 

 chosen. These scales were all mounted. The series shows 

 very definitely the anion order 



Cl>Br>No3>I 



Finally the position of KSCN was determined. It fell 

 quite obviously beyond the iodide. A very few experi- 

 ments in which the action of iodide and sulphocyanate 

 were compared, served to determine this conclusively. 

 Thus the anion series, 



Cl>Br>No3>I>SCN 



was established for the preservation of irritabilty of the 

 melanophores in these five neutral salts of potassium. 



October 3, 1912 



12.30 P.M. ' After rinsing in tap- and distilled water scales were re- 

 moved from a 10 cm. female to a 0.1 M NaCl solution. 



12.58 P.M. After selecting, five sets of five scales each were immersed 

 in 8 cc. of 0.1 M KCl, KBr, KI^ and KSCN and 8 cc. 

 of 0.105 M KNO3. All the melanophores were contrac- 

 ted after four and one-half minutes. No d'stinct dif- 

 erences were observable in these contraction times as in 

 the previous experiments. (The selection in this case re- 

 quired less than three minutes and all the scales were 

 drained on filter paper before being placed on the glass 

 dipper. After all the scales were in place the surplus 

 NaCl solution was again drained off.) 

 1.40 P.M. All the scales in KSCN showed the beginning of the expan- 

 sion. A set of five scales (one from each salt solution) 

 was now killed in 95 per cent alcohol at 2, 3, 4, and 5 

 P.M. respectively. In making the selection at the end of 

 each hour the darkest scales were chosen in every case 

 and furthermore they were killed in their anion order, 

 beginning with the KSCN scale. This was done in order 

 to throw any possible advantage of the extra time (1-2 

 minutes) in favor of the less expanded chloride and bro- 

 mide scales. To ol)viate any possibility of confusing the 

 individual scales they were placed in separate vials and 

 subsequently were separately mounted under individual 

 cover-glasses, the set from each hour on a different slide. 

 These twenty scales were then photographed individually 

 (plates 2 and 3). 



« In this case the 0.1 M KI solution was made up fresh for this experiment 

 from a 'Highest Purity' sample of Merck's salt. 



