126 E. R. HOSKINS AND M. M. HOSKINS 



Particulate matter is not taken into the tubules of the digiti- 

 form gland, except for an occasional granule, even though the 

 particles are very small. In this the gland resembles the kid- 

 ney, so the failure to take in granules does not argue against 

 possible excretory function. 



Excretory toxins caused congestion in all cases, but in only 

 one animal receiving potassium chromate was the injury to the 

 tubules very severe. Morgera ('16) obtained cytolysis of the 

 tubules by cutting off the blood supply, and the congestion noted 

 in our animals may account in this way for some of the degen- 

 eration noticed. One finds occasional cytolysed cells in the 

 digitiform gland of normal animals. The question of cellular 

 inanition must be considered in all experimental degeneration 

 of tissues produced by administered toxins. The toxins we used 

 certainly were injurious to erythrocytes, and this together with 

 interference with blood supply in congested organs must deprive 

 the cells of needed substances. However, we believe that a 

 part of the injury seen was caused by the toxins which the 

 gland was trying to excrete. 



2. Kidney 



The kidney (mesonephros) of the dogfish is of far less impor- 

 tance to the animal than the kidney (metanephros) of mammals, 

 as has been noted previously. Denis ('12) found that adults 

 produce an average of only 21.7 cc. of urine a day, and in our 

 experiments the greatest amount of urine obtained was 24 cc. 

 in twenty four hours, from an animal weighing more than 1.5 

 kg. The urea in the urine amounts to 0.6 to 1.4 per cent, or 

 less than half the concentration of that in the blood (Baghoni, 

 '06) and less than half that in mammahan urine. This differ- 

 ence is made up partly at least by a considerable amount of urea 

 excreted in the bile. The great concentration of urea in the 

 blood seems to be necessary to preserve fife, for Baglioni ('05) 

 found that isotonic (3.5 per cent) salt solution without urea was 

 unsatisfactory for perfusion experiments. This requirement 

 does not account for the small amount of urea excreted, as has 



