i GENERAL CONSIDERATIONS. 529 
We come, then, to a comparison of the relative effects of the cranial and 
sympathetic nerves as the final part of the evidence for the existence of 
t wo classes of nerve-fibres. It is said that the difference in the percentage 
composition of sympathetic saliva, and of that produced by stimulating 
the cranial nerve, can only be satisfactorily explained by supposing that 
secretory and trophic fibres are present in both, and that the number 
of trophic til lies relatively to the secretory is greater in the sympathetic 
than in the cranial nerves. 
This conclusion seems to me to be legitimate and unavoidable, if a 
diminution in the blood supply to the glands brought about by vaso- 
constrictor nerves does not markedly increase the percentage of organic 
substance in the saliva secreted. But this, so far, has not been shown 
to be the case (cf. p. 508). The question can hardly be settled until 
means are found of stimulating the sympathetic vaso-constrictor fibres 
of the salivary glands without stimulating the sympathetic secretory 
fibres. 
We find, then, that the hypothesis of a separate class of trophic 
fibres, although affording a convenient explanation of a certain number 
of facts, can hardly be considered proved at any point. It presents also 
certain difficulties of its own which we need not insist on here. 
On the whole, I think the most probable view is, that only one 
kind of nerve-fibre runs to the gland-cells, and that this causes all the 
changes in the gland-cells which are capable of being caused by nerve 
stimulation. These changes include the taking up proteid material 
from the lymph, some katabolic action — shown by the setting free of 
carbonic acid — and changes leading to the passage of water and salts 
through the cell. It is not improbable that the nervous impulses hasten 
the conversion of absorbed proteid to secretory substances, and it is 
perhaps possible that they increase the solubility of the secretory suit- 
stance already formed. The effect of the secretory fibres, as regards 
the amount and percentage composition of the saliva obtained, would 
naturally vary with the strength of the stimulus, the condition of the 
gland at the time, the quality and quantity of the blood flowing through 
the gland. 
The exact processes which take place in gland-cells and wdhch lead 
to secretion is at present outside the range of our knowledge. The high 
secretory pressure naturally suggests osmosis as the cause of the passage 
of water and of salts. And, about five and twenty years ago, the view 
that secretion is due to the formation in the cells of a substance of high 
endosmotic pressure was put forward by Hering and others. Much 
more is known now of the phenomena of osmosis than was known then ; 
but the nature of the process is still so obscure, that to attempt to 
explain secretion on the lines of osmosis is to venture on little better 
than conjecture. 
It may, however, be worth while to state briefly some points regarding the 
relation, or possible relation, of osmosis to secretion. 
We will consider, first, what facts of secretion we could in some sort 
account for, on the theory that osmotic pressure is of the same nature as 
gaseous pressure, and assuming that osmosis does take place in the gland-cells. 
The facts which it seems most feasible to offer an explanation of are, the 
occurrence of secretion -when the cells are stimulated and not at other times, 
the increase in the rate of flow daring stimulation, the increase in the per- 
centage of salts hi saliva with increase in the rate of flow. 
vol. 1.— 34 
