LOEB, ATCHLEY, AND PALMER 



593 



or chest fluid was received under oil, to avoid any marked change irt 

 hydrogen ion concentration through loss of CO2. The following physi- 

 cal and chemical determinations were made on the blood sera and 

 the edema fluids; freezing point depression, specific conductivity, 

 CI, HCO3, Na, K, glucose, non-protein nitrogen, protein per cent 

 (by Kjeldahl and ref ractivity) , and, in certain cases, urea and Ca. 

 A description of the methods employed, and other details, will be 

 published later. The results of these observations are collected in 

 Table I. 



In four cases, serum was placed in a thin collodion sac with a capac- 

 ity of about 6 cc. and immersed in a bottle containing 250 cc. of edema 

 fluid. This bottle was kept at 25° C. A manometer was placed in 



TABLE II. 



the collodion sac, and the serum level was so adjusted that there would 

 be little or no change in level with the establishment of equilibrium. 

 After 18 hours, the contents of the sac were analyzed for protein per 

 cent, K, and Cl. The results of these analyses were compared with 

 the original concentrations and with those of the edema fluid as 

 shown in Table 11. The bottle containing the edema fluid and collo- 

 dion sac was kept closed with a rubber stopper and a soda-lime tube 

 to prevent change in hydrogen ion concentration. 



DISCUSSION. 



From the data in Table I, the following relationships between blood 

 serum and edema fluids are apparently constant, (a) The freezing 



