March 27, 1914] 



SCIENCE 



449 



ical state of the dye solutions. Our atten- 

 tion was rather dramatically called to this 

 phase by experiments with benzo-purpurine 

 B: 



NaOsS. 



A fresh, cold one-per-cent. solution of this 

 dye can be injected rapidly into the ear vein 

 of the living rabbit and will produce with- 

 in a few minutes a beautiful diffuse color- 

 ing of the whole animal, a coloring which 

 though at first affecting merely the body 

 fluid, can in a few hours be seen to have 

 established itself in the form of granules 

 in that distinct class of cells which are af- 

 fected by dyes of this group. If, on the 

 other hand, we inject a similar cold solution 

 of this dye, but one which has been allowed 

 to stand a few days before using, the ani- 

 mal invariably dies before our eyes with 

 typical symptoms of cerebral embolism. 

 If we boil the dye for a little while with 

 Ringer's solution instead of with water, the 

 effect is even more marked, for a single 

 cubic centimeter of such a solution after it 

 has been allowed to cool, kills the animal in- 

 stantly on injection. Why are the dye solu- 

 tions so different in their behavior? A few 

 test-tube experiments sufficiently answered 

 our question, for we were dealing with 

 an electrolytic precipitation of a colloid, 

 whose coagula were sufficient to plug the 

 cerebral vessels. By testing in this way 

 all of the dyes which we have reported as 

 negative, we were able to show that no one 

 of them failed to produce embolism on 

 injection. 



But while thus dangerous in the living 

 blood stream there might still be thought 

 to be no reason why these dyes should fail 

 to act when placed under the skin or in the 

 body cavity. In these situations, however, 



the dyes remain without invading much 

 of the remainder of the body, and we were 

 led to test the diffusion power of posi- 

 tive and negative dyes in a medium where 

 it was slow enough to be measured, i. e., in 

 2 per cent, gelatin, and without present- 

 ing here our tabular results we found no 

 exception to the rule that positive dyes pos- 

 sessed a rapid diffusion rate, while negative 

 dyes little if any diffusion at all. 



These facts harmonize, of course, per- 

 fectly with our knowledge of the behavior 

 of colloids, and make it certain that we are 

 dealing with phenomena which depend on 

 the size of the particles or aggregates in 

 our solution. A negative dye could not 

 reach many cells in the body, for its par- 

 ticles are all large and rapidly aggluti- 

 nated in the body fluids, and such large par- 

 ticles have slight, if any, power of diffu- 

 sion. Did, however, the cells in their 

 neighborhood accept these negative dyes? 



The investigation of the cells of the sub- 

 cutaneous tissues at the injection site 

 showed that even with the most negative 

 dyes a vital staining in this limited zone 

 had always taken place. Our failure to 

 stain more cells then was due solely to our 

 inability to reach them. We proved this 

 contention, we believe, conclusively by se- 

 lecting a dye D-14. 



SOaNa NaOsS 



which when placed under the skin has prac- 

 tically no powers of diffusion and which 

 when injected at a normal rate into the 

 blood stream always kills with the typical 

 picture of embolism. We found that the 

 very slow injection of a perfectly fresh, 

 cold one half per cent, solution in distilled 

 water obviated this accident; and although 

 for a few days the continual injection of 



