20 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS home user market have relied on oxidation dyes which were at one time exclusive to hairdressing salons. A typical product will contain para- phenylene diamine or paratoluylene diamine as the prime ingredient among perhaps twenty separate constituents. Such a complicated product still has to be mixed with hydrogen peroxide immediately before use. Among the advantages are a short processing time, good coverage of white hair, good intensity and levelling and an absence of scalp staining. The principal disadvantages associated with the use of oxidation dyes are complications of the process and the toxicity. Paraphenylene diamine can cause derma- titis through sensitisation. Much of the risk of "para dyes" can be avoided if a patch test, carried out prior to the main application, causes no reaction. Paratoluylene derivatives are believed to be less harmful than para- phenylene compounds (4). It is of interest to note that when a recent meeting of dermatologists were asked about the incidence of hair colourant dermatitis, they were of the opinion that fewer cases were now seen and that there was a greater incidence among men than women. Considerable interest has been aroused by the introduction of hair colourants based on acid dyes containing small amounts of organic solvent. A dilute aqueous solution of benzyl alcohol (2-4% v/v) is found to increase the rate of dyeing with certain acid dyes. The solvent-assisted dyeing process, discovered by Peters and Stevens (5), has been the subject of much discussion and investigation. The mechanism is still not completely understood, but it is thought that the dye dissolves in the solvent, which, by virtue of its small molecular size, and nonionic character, penetrates the hair rapidly. It has been established that those dyes whose rates of dyeing are increased all partition preferentially into benzyl alcohol in simple partition experiments when the dye, benzyl alcohol and water are shaken and equilibrated. It has been the authors' experience that, while solvent dye systems dye the hair more rapidly, they also dye the skin of the scalp. So pronounced is this scalp staining that the overall effect is less acceptable than with the oxidation dyes. The changing market situation has caused manufacturers to increase research efforts in the belief that increased research effort will lead to technical innovation and ul{imately to product innovation, based either on known or novel dye systems. However irrespective of the dyeing system used, there are two common and fundamental aspects of hair colourant technology, namely measure- ment of reflected colour and dyestuff/substrate interaction which give valuable aid to the formulator. This paper will review the recent progress

HAIR COLOURANTS -- A PRACTICAL APPROACH 21 in applied colour physics, and will discuss the theories of dyeing and give an example of the application of rate measurement. REFLECTED COLOUR Applied colour physics The particular area of colour physics of most importance is that of reflected colour. Although both the theories and measurement of colour have been established for some time, the interpretation of the data obtained has proved complex. Recent developments in instrumentation and the increasing availability of computers has now made the exploitation of this information feasible as evidenced by developments in the textile industry which are applicable to hair colourants. These developments have taken place in three main areas :-- (a) Establishment of colour atlases. (b) Matching of colours by mixtures of dyestuffs. (c) Establishment of colour tolerances. The theory of diffuse light reflection most often quoted is that due to Kubelka and Munk (6). The original rigorous formula assumed a homogeneous substrate and was extremely complex. An approximate simplified formula is often used and in many cases gives good agreement with practical results. Despite the approximations and the non-fulfilment of the original assumptions, some workers have endeavoured to improve the agreement by adding arbitrary terms which recomplicate the simplified formula. Recently, a new approach has been used to develop a relation between the amount of colouring material and the light sc: ttered. This used equations due to Chandrasekhar (7). Many theories of colour vision have been proposed [a list is given by Judd and Wyszecki (8)] and a number of these theories have been used to derive re!•tionships between instrumental and visual observation. The most commonly accepted systems of relationship are t?,e Commission Internationale de l'Eclairage (C.I.E.) standard observer (9), the Adams transform into uniform colour space, and the associated equations tc calculate National Bureau of Standards (N.B.S.) colour differences (10). Measurement of light reflection and colour The colour of hair can be measured either by a skilled colourist, a tristimulus colorimeter, or a spectrophotometer. The first published data on the colour of hair was from N.B.S. in 1934 (11). Each method of