1014 PHYSIOLOGY 



endothelial cells forming the capillary wall, and that lymph is therefore 

 to be regarded as a true secretion. A careful investigation of the known 

 experimental facts has failed to show that the endothelial cells act otherwise 

 than passively, as filtering membranes of variable permeability. The factors 

 which are responsible for the transudation of lymph may be divided into 

 two classes mechanical and chemical, the former depending largely on the 

 pressure of the blood in the vessels, and the latter chiefly on the metabolism 

 of the cells outside the vessels. 



According to the views here laid down, the formation of lymph may be 

 compared to a process of filtration. If this be correct the amount of lymph 

 formed in any given capillary area must be dependent on the difference 

 of pressure between the blood in the vessels and the fluid in the extra vascular 

 tissue spaces. This latter pressure is normally extremely low, so that in 

 attempting to test the truth of this view we must try the effects of altering 

 the pressure inside the vessels, in the expectation of finding that the lymph 

 production will rise and fall as the capillary pressure is increased or dimin- 

 ished. On attempting to carry out such experiments in different parts of 

 the body, we have to recognise another factor besides the capillary pressure, 

 viz. the permeability of the vessel- wall. Whereas the capillary walls in the 

 limbs and connective tissues generally present a very considerable resistance 

 to the filtration of lymph through them, and keep back the larger portion of 

 the proteins of the blood-plasma, the intestinal capillaries are much more 

 permeable, giving at moderate capillary pressures a continual flow of lymph 

 and separating off only a small proportion of the proteins. It is in the 

 liver, however, that we find the greatest permeability. Here a very small 

 pressure suffices to produce a great transudation of lymph, containing 

 practically the same amount of protein as the blood- plasma from which it is 

 formed. 



The ease with which fluid passes out from the capillaries of the liver is probably due 

 to the fact that these vessels, unlike most other capillaries of the body, have not a com- 

 plete endothelial lining. Thus it is impossible to display a continuous endothelial lining 

 by means of silver nitrate. The cells surrounding the capillaries are large and branched, 

 and possess marked phagocytic powers, so that after an injection of carmine granules 

 or bacteria into the blood-stream these bodies are found in quantity within the cells. 

 Owing to the incompleteness of this investment the liver-cells in many places abut 

 on the lumen of the capillary. On injecting the blood-system of the liver the injection 

 is found to run with ease into channels situated within the cells themselves, and it is 

 reasonable to conclude that the blood-plasma takes the same course through these 

 intracellular channels, by which it passes into the lymphatics which lie at the periphery 

 of the lobules. 



In experiments on the lymph production in the limbs alterations of 

 capillary pressure have but slight effect. The lymph-flow from a limb 

 lymphatic is practically unaltered by changes in its arterial supply, although 

 a definite increase may be obtained by ligaturing all the veins of the limb 

 .so as to cause a very great rise of capillary pressure. The lymph-flow from 

 the intestines can be measured by collecting the lymph from the thoracic 

 duct. If the lymphatics which leave the liver in the portal fissure be 



