300 MATHEWS. 



ter methods. For some of the facts, also, errors of method 

 greatly diminish the value of the testimony they offer, and 

 some of that evidence depends upon the assumption that all se- 

 cretions are probably due to the same cause. Hence, whether 

 the theory of secretory nerves is true or not, it must be admit- 

 ted, I believe, that little of the evidence which has hitherto been 

 presented in support of that hypothesis can be accepted as it 

 stands. 



While fully aware, therefore, of the strong a pidori probability 

 that nerves may act on gland cells so as to affect osmosis through 

 them, and while appreciating the strength of the evidence that 

 they do so act, I feel myself compelled, for the reasons presented 

 in the following criticisms of that evidence, to question whether 

 secretion is really controlled in this manner. 



But not only is the evidence upon which the secretory nerve 

 theory rests inconclusive; there are also certain weaknesses in the 

 theory itself which deserve more attention than they have 

 hitherto received. It is by no means easy to understand how 

 the nerve can affect the cell in such a way as to cause a secre- 

 tion. The mere discharge of liquid from the cells into the gland 

 lumen would, as pointed out elsewhere, lead to no secretion 

 from the gland ducts. To obviate this difficulty Heidenhain 

 supposed that, while the secretory nerve diminished the resist- 

 ance of the inner end of the cell, the outer zone imbibed water 

 from the lymph and capillary. The outer zone exerted an at- 

 tractive pull upon the lymph. By the imbibition of this lymph 

 the secretion was forced along the ducts. This explanation leads 

 at once to difficulties. Not only is the explanation exceedingly 

 hypothetical, but it is difficult to see why, if the pull on the lymph 

 comes from the outer zone, secretion should be slowest after 

 long stimulation, or during paralytic secretion, w^hen the outer 

 zone is at its greatest development, and how secretion can take 

 plate at all, or with any rapidity, in glands in which the outer 

 zone has almost, or completely, disappeared, as in mucous sal- 

 ivary glands, the stomach or pancreas, after a long rest. It is 

 also difficult to understand sympathetic secretion, which takes 

 place during a period of vascular constriction. Nor can we ig- 



