LYMPH FORMATION 593 



But is increased cell metabolism the only physiological con- 

 dition associated with increased lymph flow ? Asher considers 

 that it is. So far as this association is concerned we have still 

 to discuss the way in which lymph formation is altered during 

 tissue activity. We shall deal with this point later. 



Previous observers, Ludwig, Heidenhain, Hamburger, Starling, 

 &c., had demonstrated a number of experimental means by 

 which lymph flow could be increased. It is true that in many 

 cases it would be hard to maintain that the experiments 

 reproduced conditions which are known to occur in the normal 

 body. Nevertheless these experiments are of great importance in 

 the history of the subject. Besides detailing these older experi- 

 ments and considering the views held on their mode of action, it 

 is necessary to consider their relation to the results obtained by 

 Asher. There are three possible relations between the factors 

 concerned in lymph formation as demonstrated by these experi- 

 ments and tissue activity : they might alter lymph flow by affect- 

 ing tissue activity, or they might be the means by which tissue 

 activity brings about an altered lymph flow, or they might 

 constitute an independent experimental means of altering lymph 

 flow which might or might not play a part in lymph formation 

 in the normal body. 



Before proceeding to the actual experiments it is necessary 

 to consider briefly two factors, filtration and diffusion, which play 

 a considerable part in these views and experiments. 



By filtration is meant the passage of water and dissolved 

 substances through a membrane owing to differences of hydrostatic 

 pressure on its two sides, which difference constitutes the filtering 

 force. 



The filtrate obtained from any given fluid varies both in 

 quantity and composition with changes in the filtering force and 

 the permeability of the membrane. 



A dead animal membrane is of such a nature that while it 

 lets through all crystalloids in solution, it keeps back all solid 

 undissolved substances and a proportion of dissolved colloids. 



If such a fluid as blood serum is filtered through an animal 

 membrane it is found that the total quantity of filtrate varies 

 directly as the filtering force, and for different membranes directly 

 as the permeability of the membrane. 



As regards the composition of the filtrate, a distinction has to 



