684 TRANSACTIONS OF SECTION 1. 



phenomena shown by sensitised plain muscle can most reasonably be explained 

 by colloidal interaction on the surface of the fibres. The result of this is 

 increased permeability and excitation resulting therefrom. 



I referred previously to the electrical change in excitable tissues and its 

 relation to the cell menibrane. It was, I believe, first pointed out by Ostwald 

 and confirmed by many subsequent investigators, that in order that a membrane 

 may be impermeable to a salt it is not a necessary condition that it shall be 

 impermeable to both of the ions into which this salt is electrolytically diesociated. 

 If impermeable to one only of these ions, the other, diffusible, ion cannot pass 

 out beyond the jioint at which the osmotic pressure due to its kinetic energy 

 balances the electrostatic attraction of the oppositely charged ion, which is 

 imprisoned. There is a Helmholtz double layer formed at the membrane, the 

 outside having a charge of the sign of the diifusible ions, the inside that of the 

 other ions. Now, suppose that we lead off from two places on the surface of a 

 cell having a membrane with such properties to some instrument capable of 

 detecting differences of electrical potential. It will be clear that we shall obtain 

 no indication of the presence of the electrical charge, because the two points are 

 equipotential, and we cannot get at the interior of the cell without destroying its 

 structure. But if excitation means increased permeability, the double layer 

 will disappear at an excited spot owing to indiscriminate mixing of both kinds of 

 ions and we are then practically leading off from the interior of the cell, that is, 

 from the internal component of the double layer, while the unexcited spot is 

 still led off from the outer component. The two contacts are no longer equi- 

 potential. Since we find experimentally that a point at rest is electrically 

 positive to an excited one, the outer component must be positive, or the membrane 

 is permeable to certain cations, impermeable to the corresponding anions. Any 

 action on the cell such as would make the membrane permeable — injury, certain 

 chemical agents, and so on — would have the same effect as the state of excitation. 

 If we may assume the possibility of degrees of pei'meability, the state of inhi- 

 bition might be j)roduced by decrease of permeability of the membrane of a 

 cell, wnich was previously in a state of excitation owing to some influence 

 inherent in the cell itself or coming from the outside. This manner of account- 

 ing for the electromotive changes in cells is practically the same as that given 

 by Bernstein. 



It will be found of interest to apply to secretoi-y cells the facts to which I 

 have directed your attention. If we suppose that the setting into play of 

 such cells is associated with the production of some osmotically active substance, 

 together with abolition of the state of semi-permeability of the membrane 

 covering the ends of the cells in relation with the lumen of the alveolus of the 

 gland, it is plain that water would be taken up from the lymph spaces and 

 capillaries and escape to the duct, carrying with it the secretory products of the 

 cells. This process would be continuous as long as osmotically active substances 

 were formed. Such a process has been shown by Lepeshkin to occur in plants and 

 we have also evidence of increased permeability during secretory activity in the 

 gland cells of animals. From what has been said previously, it is evident that 

 electrical differences would show themselves between the permeable and semi- 

 permeable ends of such cells, as has been found to be the case. 



As a modifiable structure, we see the importance of such a membrane as 

 that described if it takes part in the formation of the synapse between neurones. 

 The manifold possibilities of allowing passage to states of excitation or inhibi- 

 tion and of being affected by drugs will be obvious without further elaboration 

 on my part. 



Enough has already been said, I think, to show the innmnerable ways in which 

 phenomena at phase boundaries intervene in physiological events. Indeed, there 

 ai-e very few of these, if any, in which some component or other is not controlled 

 by the action of surfaces of contact. But there is one especially important 

 case to which I may be allowed to devote a few words in conclusion. I refer 

 to the contractile process of muscle. It has become clear, chiefly through the 

 work of Fletcher, Hopkins, and A. V. Hill, that what is usually called muscular 

 contraction consists of two parts. Starting from the resting muscle, we find that 

 it must have a store of potential energy, since we can make it do work when 

 stimulated. After being use:! in this way, the store must be replenished, since 



