132 L. J. MULLINS 



lions. Such substances ought, in theory, to be ideal for determining interspace 

 size but certain practical considerations make this not so. The first difficulty 

 is the quantized values of the diameter; in the series CH 4 , CCI4 , CBr 4 , and 

 CI4 the diameters increase in large steps and so does the intermolecular at- 

 tractive force of the compounds. This is equivalent to saying that the affinity 

 of the halogens of high electron density may compensate for the diminution in 

 the number of interspaces into which the substances may fit. 



The general features of the model for the membrane permit an understanding 

 of the mechanism of narcosis by molecules that are small in comparison with 

 the mean interspace size. The increase in Na + permeability taking place during 

 excitation requires that a large number of channels be unoccupied while the 

 narcotic will tend to occupy channels at random and thus to diminish ion fluxes 

 taking place during activity. The resting ion fluxes need not be proportionately 

 affected; indeed, as will be shown subsequently, narcosis by molecules that 

 are greater in size than the mean of the interspaces is capable of taking place 

 with various effects on ion transfers. 



Specific and Non-Specific Receptors 



While it is generally agreed that a variety of molecules of different sorts can 

 cause narcosis, that the concentrations necessary are predictable from the 

 physical parameters of the molecule, and that narcosis appears to be an equilib- 

 rium phenomenon, no such agreement is to be found when the capacity of a 

 molecule to initiate nervous excitation is to be considered. Rather, it is con- 

 ventional to postulate the existence of receptors on the cell surface that are 

 responsible for the observed excitatory (and sometimes inhibitory) specificity 

 of particular chemical compounds. Because receptors are situated very close 

 to the membrane, it is reasonable to inquire how one distinguishes between these 

 two structures; the difference lies mainly in the fact that the membrane is not 

 supposed to detect chemical specificity. If, therefore, it can be shown that the 

 membrane model can detect small differences in molecular structure, we can 

 dispense with the concept of receptors and incorporate their features into the 

 membrane model. 



One of the most versatile sets of receptors is the olfactory epithelium of man 

 and a study of the threshold for the detection of various substances may serve 

 as an introduction to what might be called non-specific receptors. Results of 

 threshold studies for homologous series of paraffins, chloroparaffins and alcohols 

 show that (a) the threshold is about the same for any of these three compounds 

 at the same number of carbon atoms, and (b) stimulation by these substances 

 is strongly dependent upon the size of the molecule (C 4 being about optimal 

 and 50 times better than Ci or C s (Mullins, 1955a)). As, in the paraffins, there 

 is no reactive chemical grouping, it appears necessary to postulate that it is 

 the size of the molecule that limits its stimulating effectiveness. It is also an 



