158 G. K STEWART 



not a few denying that the suprarenals are essential. Thus, .Mayer (a) 

 stated thai; after excision of the suprarenals in rabbits, piqure caused no 

 glycosuria in the 8 hours following the operation. Borberg(a) obtained the 

 same result, but only in one rabbit. Kahn(&), in two rabbits which had 

 survived removal of the suprarenals at two operations, saw no glycosuria 

 after piqure, and this was confirmed by Kahn and Starkenstein, and by 

 Biberfeld(fr), who, however, admits that observations on sugar in the urine 

 are unsatisfactory in the absence of blood sugar estimations. On the other 

 hand, Wertheimer and Battez(a) (&) obtained distinct glycosuria in the cat 

 (14 times out of 43) and in the dog (9 times out of 35) on performing 

 piqure after removal of both suprarenals, necessarily in acute experiments. 

 As regards investigations in which the blood sugar was estimated, the 

 results are also contradictory. Thus, Jarisch concluded that, after dener- 

 vation of the suprarenals in rabbits, piqure did not cause any increase in 

 the blood sugar. But his experiments were vitiated by the fact that the 

 glycogen store of the liver was in every case so low that there could not 

 have been any certainty that in normal animals a distinct hyperglycemia 

 would have resulted from piqure. The conclusion of Freund and 

 Marchand that piqure causes an increase in blood sugar after removal of 

 the suprarenals in rabbits, although certainly correct, could not be deduced 

 from their results, as the piqure was made only two or three hours after 

 removal of the suprarenals and under anesthesia, which itself causes 

 hyperglycemia. Confusing as these results were, the question is neverthe- 

 less not a difficult one to settle and the astonishing lack of uniformity in 

 the results is due almost entirely to the neglect of essential conditions. 

 Two main ways of approaching the question are theoretically open : (a) the 

 estimation of the output of epinephrin under the action of the factors 

 inducing the experimental hyperglycemia (or glycosuria), in order to show 

 whether this can be sufficiently great to permit the introduction into the 

 blood of the necessary excess of epinephrin as determined by artificial 

 injection; (>) study of the sugar content of the blood (or urine) in the 

 absence of the suprarenals or after interference with their epinephrin 

 secretion. The first method of approach might seem the most direct, but 

 it has been little used. Such results as have been obtained on the rate of 

 output of epinephrin in ether narcosis and asphyxia are quite unfavorable 

 to the view that epinephrin plays any sensible role in the production of the 

 hyperglycemia. Thus in etherized cats Stewart and Rogoff(e) (s) found 

 an average output of 0.0002 mgm. epinephrin per kgm. of bodyweight per 

 minute, and this was not increased in asphyxia. Now Underbill was 

 unable to produce glycosuria in non-narcotized rabbits even when he ad- 

 ministered epinephrin at approximately 20 times this rate for more than 

 2!/2 hours by continuous infusion into a vein, although he found, in 

 confirmation of Ritzmanh, that these large doses caused some glycosuria in 

 urethanized rabbits. The statement of Waterman and Smit, repeated 



