THE FORMATION OF LYMPH 351 



curare, extracts of crab's muscle, leech extract, dilute solutions of egg albumin 

 and peptone, nuclein and metabolic products of Bacteria, water extract of straw- 

 berries, etc. The seat of this increased formation is almost exclusively in the 

 liver and the pressure in the liver capillaries shows only a temporary rise. The 

 cause in this case cannot be sought in any sort of filtration, and the osmotic 

 processes cannot play any part, since the quantity of injected substance was 

 always very small. It is most natural therefore to conceive of the process as 

 secretory in nature, unless one supposes with Starling that the liver capillaries 

 are injured by the substances used and thus permit a freer passage of fluid 

 which however is not yet proved. 



The flow of lymph from the glandular organs at least always increases when 

 the glands are active. Stimulation of the salivary glands through their secretory 

 nerves for example raises the quantity of lymph in the vessels of the neck. 

 Injection of sodium taurocholate produces a copious secretion of bile and the 

 lymph stream in the lymphatics of the liver swells in size. The same is true 

 when the formation of urea in the liver is intensified by the injection of am- 

 monium tartrate; and we should probably include here also the increase in the 

 quantity of lymph flowing through the thoracic duct during digestion of proteid. 



According to Asher and Barbera, the activity of the gland cells is the pri- 

 mary phenomenon in these processes and the increased production of lymph is 

 only secondary. It is clear of course that a secreting gland must receive more 

 water, the more active is its production. It can be easily understood also that 

 a certain part of this water should be carried away by the lymphatics ; the only 

 question is, What are the forces which cause the increased output of lymph? 



Here, again, one may conceive of an active participation of the capillary 

 endothelium, and yet the possibility of a change in the osmotic tension of the 

 lymph by the process of secretion of such a nature that new quantities of lymph 

 would pass out of the capillaries by pure osmosis is not excluded. A definite 

 decision between these two explanations is not possible at present, because exact 

 quantitative determinations of the osmotic pressure of the lymph and of its 

 changes during secretion are wanting. 



4. Not only water, however, but salts and organic foodstuffs also pass from 

 the blood into the lymph. Here again theoretical explanation of the phenomena 

 meets with certain difficulties. The metabolism of the different organs of the 

 body differs greatly with respect both to quantity and kind; they require dif- 

 ferent substances in very different quantities. A milking cow, for example, 

 secretes daily from the milk glands 25 1. of milk containing 42.5 g. of calcium; 

 which means that from the capillaries of the milk glands there passes a much 

 larger quantity of calcium than from all the other capillary regions of the body 

 put together. 



This holds true with regard to combinations of iodine by the thyroid gland. 

 The thyroid takes up the iodine occurring only in excessively small quantities 

 throughout the body (the blood of the dog contains according to Gley and Bour- 

 cet 0.01-0.11 mg. of iodine per liter) and stores it up in a compound very rich 

 in iodine (Baumann). 



The different organs, therefore, must possess a specific power of selection, 

 in virtue of which each levies upon the blood for the constituents necessary to 

 its activity. Since however the gland cells are not attached to the capillary cells, 

 but are separated from them by lymph spaces, they cannot themselves exercise 

 this power of choice but must delegate it to the capillary cells. 



Cohnstein rejoins with the supposition that after the parenchyma cells of 

 the glands, etc., have removed a certain constituent from the lymph, the latter 

 receives its replenishment from the blood by a process of diffusion, so that no 



