THE PHYSICAL CHEMISTRY OF VISUAL PURPLE 



putting p ^ I . . . makes it possible to predict the right order of 

 the molecular weights of several substances.' 



Measurements in the ultracentrifuge. The size and weight of 

 colloidal particles can sometimes be inferred from their behaviour in 

 the Svedburg ultracentrifuge. In this apparatus solutions can be 

 subjected to a force exceeding a miUion times that of gravity. Under 

 these conditions, colloidal particles and large molecular aggregates — 

 if they are heavy enough — slowly sediment out of solution. By 

 observing the movement of the upper boundary of the particles, one 

 can determine whether the system is monodisperse, i.e. contains 

 particles of uniform size, or is polydisperse. In the latter case the 

 boundary is diffuse or, sometimes, multiple. 



HECHT and PICKELS (1938) found that digitonin solutions of frog 

 visual purple sedimented as particles of uniform size with a molecular 

 weight of 270,000. For some time this was beheved to be the molecu- 

 lar weight of visual purple. In 1940, however, smith and pickels 

 reported that solutions of digitonin alone behaved in the ultra- 

 centrifuge as a monodisperse system having a micellar weight of at 

 least 75,000. Each digitonin micelle therefore contained about 

 60 molecules of digitonin (M.W. = 1,288). 



In a recent investigation, hubbard (1954) repeated and extended 

 the earlier measurements. She found that visual purple solutions 

 (from cattle retinae) sedimented as a stoichiometric complex ('RD-l') 

 of visual purple and digitonin. The composition of this complex was 

 not affected by the hydrogen ion concentration (between pH 6-3 and 

 9-6) nor by the presence of excess digitonin. From the sedimentation 

 constant, hubbard calculated that the molecular weight of RD-l 

 was 260,000-290,000, thus confirming the earUer observations of 

 HECHT and pickels. 



The relative proportions of digitonin and visual purple in the 

 RD-l complex were obtained by analysis of solutions which con- 

 tained only RD-l. The digitonin was estimated by its haemolytic 

 activity; the visual purple, by measuring the nitrogen content of the 

 solutions and assuming that visual purple contains 15 per cent N. It 

 was found that 14 per cent of the weight of the complex, i.e. 36,000- 

 41,000, was due to visual purple. 



hubbard deduced the molecular weight of visual purple in the 

 following way. From her own data it could be calculated that a 

 visual purple solution which had an optical density at 500 mfx of 

 1-0 per cm, would contain 2-4-2-7 x 10~^ moles of RD-l per litre. 



93 



