THE FORMATION OF LYMPH. 417 



on this subject. His experiments made on horses led him to 

 three main conclusions in support of Heidenhain's views :■ — ■ 



1. That the amount of lymph may rise when the arterial 

 pressure is diminished. But, as already pointed out, arterial 

 pressure alone is no measure of intra-capillary pressure. 



2. The proportional composition of blood-serum is not 

 the same as that of the lymph produced at the same time. 



3. The osmotic pressure of the lymph is higher than 

 that of the blood-plasma. 



Starling points out that these two latter arguments lose 

 a good deal of their weight if it is remembered that the 

 amount of lymph obtained is extremely small, and that we 

 do not know what changes it had undergone on its way 

 through the tissues ; it is quite possible it may have taken 

 up excess of salts from the tissue cells ; certainly the meta- 

 bolic changes in the tissues would increase the osmotic 

 pressure of the lymph that bathes them. 



In his last article Starling tackles the important subject 

 of lymphagogues, and here again he considers that just as 

 in the experiments involving mechanical interferences with 

 the circulation, the increase of lymph can be explained by 

 the filtration hypothesis, and that recourse to a theory of 

 lymph secretion is superfluous. 



Take the second class of lymphagogues (sugar, etc.) first ; 

 the mechanical explanation of their action is as follows : On 

 their injection into the blood, the osmotic pressure of the 

 circulating fluid is largely increased, and so water is attracted 

 from lymph and tissues into the blood by a process of osmosis 

 until the osmotic pressure of the blood is restored to the 

 normal. A condition ofhydrsemic plethora is thereby pro- 

 duced, attended with a rise of pressure in the capillaries 

 generally, especially in those of the abdominal viscera, 

 which causes a great increase in the transudation of fluid 

 from the capillaries, and therefore in the lymph flow from the 

 thoracic duct. Heidenhain pointed out that the lympha- 

 gogue effect of these substances is proportional to their 

 osmotic pressures, and therefore to the figure obtained by 

 dividing their osmotic co-efficient by their molecular weight. 

 Since the hydrsemic plethora must be also proportional to 



