30 



gelatine by reason of their close adhesion to the surrounding cells. 

 But in many parts of these preparations the gelatine has shrunk away 

 from the wall as a whole leaving no external line of colouring matter 

 such as is represented by Dr. Holmgren, and in these parts the 

 intracellular canals are just as fully injected as elsewhere. Nor is 

 there in the liver so far as my own observations go (and in this 

 matter I entirely agree with Professor Browicz) any space between 

 blood- capillaries and liver-cells, the cells being in the closest possible 

 apposition to the walls of the vessels 1 ). Moreover extravasations due 

 to escape of the injection-mass from the blood-vessels do not take 

 the form described by Holmgren but appear as irregular masses 

 excavating their way between the liver-cells and sending no offsets 

 whatever into the interior of the cells. 



That the canaliculi shown in the preparations described by me 

 are artefacts is a supposition which I should have supposed could not 

 be entertained for a moment by any experienced histologist. They 

 are well defined anastomosing passages which form a network within 

 nearly every cell throughout the whole liver. The in- 

 jection material which fills them is of the same intense colour as that 

 within the blood-capillaries and appears at places to be connected 

 with this by fine threads. That it has not passed into the cells by 

 way of the lymphatics appears evident from the fact that such in- 

 jection material as is found in the lymphatics, e. g. in those of the 

 portal canals, is of a far paler tint than the material within the blood- 

 vessels and intracellular canaliculi ; the result, a I suppose, of dilution 

 with lymph. From these evidences the most probable explanation 

 seemed and still seems to me that the injection has passed directly 

 from the interior of the capillaries into the intracellular canaliculi by 

 means of pre-existent communications 2 ). I do not deny the possi- 

 bility that the injection might first pass into perivascular lymphatic 

 clefts (assuming that such clefts really exist; an assumption which is, 

 to say the least, open to considerable doubt) ; there is no clear evidence 

 of its having done so and, moreover, on this hypothesis it would be 

 still more difficult to understand why it should have passed into the 

 interior of the cells in so extraordinarily complete a manner throughout 

 the whole organ as is seen in these preparations, rather than having 

 made its escape by way of the efferent lymphatics. 



Remarkable as this injection from the blood-vessels of intracellular 

 canaliculi at first sight appears it must be born in mind that the 



1) This is nowhere more clearly shown than in a figure given by 

 Dr. Holmgeen himself (see this Journal, Bd. 22, p. 10). 



2) I regret that in my first communication (loc. cit. p. 20) I repre- 

 sented Professor Browicz as agreeing with me that the preparations in- 

 dicate a direct communication between the vessels and liver-cells, 

 whereas Professor Browicz speaks of it as indirect. But the actual 

 difference between our views seems to me far less than is indicated by 

 these termes. 



