34 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS dissolve i• a small quantity of water and make up to 1 litre in a volumetric flask. Pipette 25 ml of the well-mixed solution into a 100 ml glass, stoppered, measuring cylinder, add 25 ml methylene blue indicator and 15 ml chloro- form, stopper the cylinder and shake vigorously. Allow to stand for 1 minute, when the dye will be seen to be concentrated in the lower layer. Add, from a burette, the standard cetrimide solution, 2 ml at a time, shaking vigorously after each addition and allowing the two layers to separate before continuing. When the layers begin to separate more quickly, decrease the amount of titrant added before shaking, and watch carefully the return of the blue colour to the aqueous phase. When the colours are nearing the same depth in the two layers, decrease further the volume of titrant added between shakings, and examine the colours of the layers by standing with back to the window and holding the cylinder up with a white tile immediately behind it. Finally, the addition of 0-1 ml titrant will cause the colour in the two layers to appear identical and this is taken as the endpoint. The first titration is used as a guide carry out two further titrations in the same way. In these two titrations 80 per cent of the titrant may be added initially, and the titration completed after vigorous shaking, by gradual additions until the end point is reached. Note. The titrations should be carried out in good daylight and, during the winter months, should always be undertaken in the morning. For the purpose of this study the active anionic matter is normally calculated as "ml 0.002M cetrimide per gram of sample" ml 0.002• cetrimide i.e. Wt. sample in 25 ml aliquot Comments. This method, originally proposed by Epton, is based on the titration of the anionic material with a standard solution of a cationic surface-active agent, using a dye of cationic type as an indicator. The titration is conducted in a two-phase system, the aqueous phase containing sodium sulphate and sulphuric acid, and the second phase consisting of chloroform. Initially, some of the artionic detergent reacts with the whole of the indicator, forming a chloroform-soluble complex, so that the chloroform layer is blue and the aqueous layer colourless: C•6H•sNaSC1 q- R.SO30M = C•6H•sNaS.SOaOR q- MC1 The addition of the cationic solution causes the formation of a similar complex with the remaining anionic and this is also extracted by the chloroform: R•NX + RSOaO• -• RSOaOR•N + •X When all the free anionic detergent has reacted with the cationic titrant,

PRELIMINARY NOTES ON THE DETERMINATION OF ANIONIC SUL?HATES 35 the latter displaces the small amount combined with the indicator and frees the methylene blue, which is soluble in water and migrates back to the aqueous layer. The end-point is usually taken as the stage when the two layers, viewed under standard conditions of illumination, appear of equal colour intensity. There are a number of difficulties associated with this method and some of the most important are outlined below: (1) The end-point is not easy to see, since the colour shade is usually slightly different in the two phases. Consequently, careful standardi- zation of lighting conditions is required and there is considerable risk of operator to operator variation. (2) Since the end-point depends upon partition between two solvents, standardization of the final volumes is required. (3) The exact technique of addition of the titrant, and the rigour of shaking between additions may be critical. (4) The visual end-point is not necessarily identical with the stoichi- metric end-point, although this is not of great importance if stan- dardization and determination are taken to identical end-points. However, the chain length of the hydrophobic portion of the mole- cule affects the relationship between the visual and stoichimetric end- points. Chinnick and Lincoln (4) have shown that alkyl sulphates with a carbon chain of 10 atoms or less do not give an end-point at all, and that the visual and stoichimetric end-points appear to coincide at about C•4. It is therefore necessary to standardize the cationic titrant against a standard anionic of the same chain-length as that being determined. (5) Quaternary ammonium salts are generally used as cationic titrants, pyridinium salts giving a different end-point. Here, again, there is some evidence that the chain length of the hydrophobic portion of the cationic molecule has an effect on the result. 1.3. The pToluidine Hydrochloride Method (5) 6.8% ptoluidine hydrochloride in distilled water. Carbon tetrachloride. A.R. 0.1N alcoholic potassium hydroxide. 0.1% mcresol purple in 95% alcohol. Method Weigh accurately about 8 g sample into a 100 ml beaker and dissolve in about 50 ml distilled water. Transfer the solution carefully to a 100 ml