160 L. J. MULLENS 



Also, I am not sure that it is seriously embarrassed by Dr. Davies' idea that 

 the membrane may be more complicated than Dr. Mullins' sketch, because I 

 think the gist of Dr. Mullins' model, really, is that there is a two-dimensional 

 lattice that can be affected by inclusions, and I don't imagine he would cling too 

 tenaciously to the fact that these are protein cylinders. 



I do wonder if there is a discriminating experiment between whether the 

 narcotic agents have their effect on the membrane or on metabolic entities in- 

 side. If one accepts his model in the simplest terms it seems to me one should 

 get the same curve for threshold activity versus number of carbon 

 atoms, whether the agent is delivered on one side of the membrane or the other. 

 I would like to ask him if he feels this is technically a sensible experiment, if he 

 feels it is technically feasible, say, by micro-injection? 



Dr. Mullins: I would like to say something to Dr. Davies first. I don't dis- 

 agree particularly with what he has said. The only objection I would have is 

 that it does not get down to this level of explanation that is satisfying to me, 

 namely, the type of molecular alteration which is to bring about this electrical 

 disturbance that we regard as a physiological function of the cell. As I gath- 

 ered it, he would have some kind of an interfacial change at one of the sur- 

 faces take place which would alter the structure in some unspecified way so 

 that it would become permeable to ions. I proposed the type of structure that 

 I did because it seemed to me a way of having this alteration occur from appli- 

 cation of physical principles and with a minimum of accessory hypotheses or 

 special mechanisms. I tried to point out as I went along that it did involve as- 

 sumptions, you have to size the thing somehow or other, and you have to agree 

 that if things fit closely and exert a strong van der Waals type of attraction 

 that you have — this could give rise to phenomena analogous to crystallization 

 or to a lattice distortion or crystal defect, and that in the process of creating 

 this local order you pay for it by an increase in the disorder surrounding this 

 region for many tens of molcules away and therefore many tens of interspaces 

 away. This gives at least a plausible mechanism for how excitation can be both 

 initiated and amplified because I think we all agree that we need an amplifier 

 action. 



Now, the cylinders are very long. The ones I proposed were 100 A simply 



o 



because people have settled on this kind of figure. I believe there is also a 50 A 

 membrane thickness school which I certainly do not disagree with. But along 

 this length it would be inconceivable to me that it is a solid lipid all the way 

 through, the surface of these cylinders may be primarily CH 2 groups, with gaps 

 in which the polar groupings of the underlying structure, protein namely, would 

 project through and these areas would suffice to explain many of the kinds of 

 properties that we presently assign to purely a protein part of the membrane. 

 Also, there are the ends of the cylinders which I expect are protein, the largest 



