HISTORY OP THE INFLAMMATORY PROCESS. 1J5 



are withdrawn, and the whole cell^ losing its normal shape, 

 becomes very actively amoeboid. As a result of my observa- 

 tions I am led to believe that no sharp line of demarcation 

 exists between the so-called fixed or stellate and the wandering 

 cells. After ingesting as much as possible of the foreign 

 suljstance, the connective tissue cells withdraw their pseudo- 

 podia and appear as rounded masses. 



These observations show that the cells of the connective 

 tissue in the larval Tritou^s tail must distinctly be regarded as 

 phagocytes, and that they act as such during inflammation. I 

 have several times observed a multiplication of these elements 

 in an inflamed Triton larva : but the process, which can easily 

 be followed in the living subject, with all its associated nuclear 

 changes, happens so seldom that it can hardly play an important 

 part in normal inflammation. 



In the frog tadpole the extravasation of white blood-corpus- 

 cles is much more important than in Triton. My observations 

 confirm those of several investigators, who assume, as Von 

 Ileckliiighausen has recently done, an active wandering on the 

 part of the corpuscles themselves, eftected by the protrusion 

 of numerous pseudopodia, similar to those extruded by the 

 resting corpuscles of many Invertebrates. 



In the frog larva, also, the phagocytes collect round the point 

 of stimulation, and eat up as much of the exciting object as 

 they can (I used in stimulating tadpoles a fine glass tube, 

 filled witb cinnibar or carmine). In this state they often 

 remain for days or even weeks. When a fully gorged phago- 

 cyte dies, it is immediately devoured by another, so that one 

 can often see a single large cell, containing one or two dead 

 phagocytes, whose nuclei have already disappeared : there is, 

 however, no formation of true multinuclear plasmodia, which 

 indeed I have very seldom seen in any amphibian larva. 



I must now point out two results of ray observations. First 

 of all, that the most copious diapiedesis in inflamed tadpoles 

 was not in the immediate vicinity of the glass tube, but at 

 some distance from it ; and secondly, that I never saw any 

 tendency towards a definite aggregation of transudation pro- 



