THE VISUAL PIGMENTS 



COLLINS and morton (1950c) supposed the conversion of visual 

 purple to indicator yellow to be an oxidative process involving the 

 loss of two electrons. Accordingly, they proposed the following 

 structural formula for the visual purple chromophore, 



one of at least twelve resonance forms, the most probable being those 

 with the negative change on the carbon atoms adjacent to the nitrogen 

 atom. 



In 1953, however, Collins, referring to unpubUshed work by 

 COLLINS, MORTON, PITT and STOK, reported that the compound 

 formed by interaction between retinene and methylamine was, in 

 fact, a *Schiff 's base' with the formula, 



and not the product of one amino group with two retinene molecules 



as at first had been thought. This substance, retinene methylimine, 



was obtained in the crystaUine state, m.p. 73°C. It had an absorption 



maximum at 365 mju in ethyl alcohol, a molecular weight (Rast's 



method using camphene) of 290 and contained 4-82 per cent 



nitrogen (C19H27CH = N— CH3 requires M.W. = 297, 4-71 per cent 



N). On acidification a substance with Amax = 440 m/u in ethyl 



alcohol was obtained. This analogue of acid indicator yellow could 



not be crystallized. The best preparation had m.p. = 140°C and 



contained 3-9 per cent nitrogen. The hydrochloride of retinene 



+ - 



methyhmine, C19H27CH = NHCHgjCl, requires 4*2 per cent N, in 



fair agreement with the experimental figure. However, collins 



(1953) considered that a substance of this formula would have an 



absorption maximum at a wavelength similar to that for the free 



imine (365 mju) and that the observed position (440 m^u) implied a 



doubled molecule for the acid form. This seems to be an unnecessary 



hypothesis for the Amax position might well shift as a result of the 



positive charge carried by the nitrogen atom in the hydrochloride 



formula. A molecular weight determination would help to settle the 



matter, but in the meantime it would seem that the most likely 



108 



