STRUCTURE OF NERVE CELL MEMBRANES 135 



no other isomer can be oriented in the membrane interspace in such a way as 

 to provide the Cl-membrane interaction present in the case of the gamma 

 isomer. The narcotic effects of the 8 and other isomers can be explained on the 

 basis of solubility and non-specific interaction. It would appear that the orienta- 

 tion of a molecule in a membrane interspace may be of decisive importance for 

 chemical excitation and that an interspace without any assumed chemical 

 specificity is able to discriminate between closely related compounds and is 

 able, thereby, to be variously affected. 



Another molecule, DDT (2,2 bis (p-chlorphenyl) 1 , 1 , 1 , trichloroethane) 

 can be shown not to penetrate except in one particular orientation. The five 

 chlorine atoms in the molecule are not considered to bind at specific points, 

 but as in BHC they serve to intensify the London-van der Waals forces between 

 the molecule and the groupings in the interspace. Since the interspace is pre- 



o 



sumed to be of the order of 100 A. long (the membrane thickness) somewhat 

 more than 20 DDT molecules are required to fill it. There are a rather large 

 number of homologs of DDT only a very few of which have any action at all 

 comparable as convulsants; these compounds, as well as the BHC isomers, 

 provide a rather critical test of the lattice defect mechanism of action already 

 suggested because the membrane model must clearly reject inactive compounds 

 and accept the active one. In Fig. 5 is shown DDT and some homologs oriented 

 in the membrane interspace. This is the contracted size of the membrane 

 interspace and the actual size of the unoccupied interspace would be larger. 

 The membrane phase change suggested as an explanation for the mechanism of 

 action of DDT also affords an understanding of the anomalous negative tem- 

 perature coefficient of DDT action (Lindquist et al, 1945; Potter and Gillham, 

 1946; Hafliger, 1948). With macromolecules of the sort suggested as components 

 of the membrane, a sharp phase transition with temperature can hardly be 

 expected for the membrane-DDT complex; melting will, however, become more 

 and more probable as the temperature is raised. Increased excretion or detoxi- 

 fication appears ruled out as an explanation for the effect (Vinson and Kearns, 

 1952) as does the suggestion of Munson (1954) that lipid depots in insects 

 compete more effectively with receptors for DDT at elevated temperatures. 

 Such an explanation is hardly compatible with the observations (Welsh and 

 Gordon, 1947) on the essentially irreversible binding of DDT that is to be 

 expected if it forms a membrane-DDT complex. 



The excitatory action of DDT can be antagonized by certain anesthetic 

 agents but an hypothesis that the anesthetic blocks those enlarged interspaces 

 around a DDT site cannot be entertained because there is no way to compel 

 the anesthetic molecules to select such sites from all others offered by the 

 membrane. Two types of anesthetic action are distinguishable from the model; 

 in the first, small molecules occupy interspaces of all sizes equally well so that 



