We have established that the analgesically active isomers '(-)-methadone' ((-)-6-dimethylamino-4: 4-diphenylheptan-3-one), (-)-ethyl-l: l-diphenyl-3-di-methylaminobutyl sulphone, and (+)-3-dimethyl-amino-1: l-(2'-thienyl)-but-l-ene, possess identical configurations which are related to D-(-)-alanine. These relationships were established in the following way. The hydrolysis of (-)-ethyl-1: l-diphenyl-3-dimethylaminobutyl sulphone gives (-)-3-dimethyl-amino-1: 1-diphenylbutan-l-ol4. This alcohol, after dehydration followed by reduction, gave (+)-3-di-methylamino-1: 1 -diphenylbutane, which was also obtained from (-)-4-dimethylamino-2: 2-diphenyl-pentane nitrile, the precursor of (-)-methadone. L-Ethyl-3-dimethylaminobutyrate, derived from phthaloyl-L-alanine via an Axndt-Eistert reaction followed by reductive methylation, was treated with phenyl magnesium bromide to yield (+)-3-dimethyl-amino-1: 1-diphenylbutan-l-ol. Reaction of this same ester with thienyl lithium gave (-)-3-dimethyl-amino-1: l-(2/-thienyl)-butan-l-ol, which gave (-)-3-dimethylamino-1: l-(2'-thienyl)-but-l-ene on dehydration. In a recent communication, Bick5 has advanced evidence from which he concludes that morphine is coixfigurationally related to L-(+)-alanine. His argument relates the two substances via apomorphine and laudanosine and depends on the validity of a method of assigning configuration first employed by Leithe6'7. Leithe maintained that the specific rotation of bases possessing analogous asymmetric centres and of like configuration are displaced in the same direction when examined in a series of solvents of increasing polarity.Because the validity of Leithe's method has not been rigidly established, we have determined the specific rotations, in a series of solvents, of the compounds shown in Table 1. The configurations of these compounds are identical and have been established as described above. It is evident that in this series no relationship exists between configuration and direction of specific rotational change brought about by an increase in the polarity of the solvent. We are therefore of the opinion that the method used by Bick to establish the relationship of morphine and apomorphine to the L-amino-acids is of doubtful value. Furthermore, we consider that the configurational assignments given by Leithe to alkaloids of the laudanosine and berberine type are not conclusive.The identical configurations of the analgesically active isomers reported in this communication ; the antagonistic action exhibited by N-allylnormorphine ('Nalorphine') towards morphine and many of the synthetic analgesics ; the antagonistic action towards morphine shown by (-)-N-allyl-3-hydroxymorph-inan but not by the corresponding (+) isomer8 ; the identical configurations of (-)-N-allyl-3-hydroxy-morphinan and (-)-3-hydr oxy-N-methyl-morphinan9 (the isomer which is analgesically active whereas the (+)-isomer is inactive) give further support to the statement10 made some time ago by one of us (A. H. B.) that "before analgesic action can be mediated directly or indirectly, it is possible that the stereochemical configuration of the drug must be complementary to that of a certain tissue surface or enzyme system". The presence of specific receptor sites for analgesic action is now a probability rather than a possibility. A detailed account of the investigation of certain structural aspects of analgesics will be published elsewhere.Table 1 wfMe.CH(NMe2). CH2.CPh2.iJ Base in hexane Base in benzene Base in ethanol Hydro-chloride in water R = -CO.Et -34-4 (0 = 1-2) -24-0 (0 = 1-3) -26-0 (0 = 1-5) -125 (ref. 11) (0 = 1-5)R = -SO2.Et - + 12-0 (in acetone (ref. 12) 0 = 2-0) 0 (ref. 12) (0 = 2-0) -31-6 (ref. 12) (0 = 5-0) R = -CN -65 1 (0=0-9) -59-5 (0 = 0-8) -50-9 (0 = 0-8) + 6-0 (ref. 11) (0 = 5-0)R =- H -66-2 (0 = 4-8) -50-3 (0 = 4-7) + 12-0 (0=3-4) + 52-7 (0 = 1-0) 1= -CO2.Et -85-5 (0=2-0) -73-8 (0 = 1-8) -50 0 (0 = 1-8) -37-3 (0 = 1-0) iJ = -OH + 36-8 (0 = 0-8) +32-6 (0 = 1-9) -27-3 (0 = 0-8) -45-1 (0 = 1-0)Me.CH(NMe2). CH: CPh3 + 128 (0 = 1-0) + 123 (0 = 1-0) + 170 (0 = 1-0) +213 (0 = 0-9) The effect of concentration upon the specific rotation is not significant in comparison with the effect of the solvent.