524 SCIENCE PROGRESS. 



action current is far shorter than that for the flow of fluid in 

 the ducts even when a high pressure is established in these 

 by artificial means. Bayliss and Bradford measured the 

 former upon excitation of the dog's chorda at '^ sec, while 

 Hering measured the latter, distending the ducts at a pres- 

 sure of 150 to 200 mm. of mercury and using a Pick's 

 spring manometer, at the shortest as t*2 sees. 1 If then we 

 are forced to exclude electro-capillary current, as at any rate 

 the principal factor in the phenomena, we are forced back 

 upon the theory of Biedermann, and it must be admitted 

 that, though the explanation is based upon the obscure 

 " chemismus " of protoplasm, it has much in its favour. 



It is, however, difficult to see clearly what right Bieder- 

 mann and Bohlen have for considering water secretion as 

 the cell process connected with the production of the ingoing 

 current, for the fact that increment of the water in the cells 

 swells the ingoing current might surely also receive ex- 

 planation on the more general ground of increased proto- 

 plasmic activity, since it is generally accepted that there is 

 a relation between the activity of a tissue and the amount 

 of water it contains. In the skin of the eel water has a 

 distinctly deleterious effect, reducing or reversing the normal 

 current, though the effect is done away with by substituting 

 normal saline solution. 



Whether or no water secretion has any direct connection 

 with gland currents, might perhaps be decided by using for 

 experiment cases in which there is no secretion of water, as, 

 for instance, the uropygial gland of the bird or the Harderian 

 of the mammal, or perhaps the crop of the breeding pigeon, 

 which in its lateral lobes becomes practically a sebaceous 

 gland during the time the birds are giving " milk " to their 

 young. 



In conclusion, it may conduce to clearness to put in 

 tabular form the explanations offered by the two most 

 important theories for some of the ordinarily observed 

 experimental facts. 



1 Ludwig originally measured this latency with a mercurial manometer 

 at from four to twenty-four seconds. 



