ON THE SENSITIVENESS OF INDICATORS. 209 



have a large amount of XO' ions in the solution, and a very small amount of 

 HXO and XOH. The colour of such a solution will be entirely due to the 

 ions. Further, the law of mass action requires that the proportion of 

 HXO to XOH molecules in the solution must always be constant, so that 

 from a physico-chemical standpoint it is quite legitimate to speak in this 

 case of the undissociatcd molecule being red in solution, provided we 

 always understand by the ' undissociated molecule ' the equilibrium 

 mixture of tautomeric forms. For simplicity's sake then, we keep to 

 Ostwald's original theory in discussing the conditions under which the 

 colour change of an indicator takes place, always remembering to allow 

 due weight to the modern conception of the ' pseudo-acid ' when necessity 

 for this arises. One condition, however, must be noticed. It is essential 

 that the actual chemical change of an indicator from one coloured form to 

 another should be rapid : now, as we have seen, the ionisation of an 

 indicator is accompanied by a simultaneous tautomeric change, and it is on 

 this that the change in colour really depends. The formation and com- 

 bination of ions is instantaneous ; a tautomeric reaction not necessarily so. 

 Hence we arrive at a necessary condition for a good indicator : that the 

 tautomeric change from the pseudo to the true acid form should be so 

 rapid as to be practically instantaneous. Cyanine is an example of a bad 

 indicator from this point of view, since its colour change takes an appre- 

 ciable time. 



The question of the chemical constitution of the different possible forms 

 of an indicator is of great interest, especially when the indicator is ' am- 

 photeric,' that is to say, has both basic and acidic groupings in the molecule, 

 as, for example, methyl orange. But this question is impossible to treat 

 from a general point of view ; its discussion is best confined to individual 

 cases. For this reason I shall not go into it at present, but merely mention 

 it in passing as a subject whose intrinsic interest and importance make it 

 especially valuable for discussion in a meeting like the present. In 

 particular there is the disputed question whether the colour change of 

 an indicator such as methyl orange is primarily due to its basic or acidic 

 functions, a question upon which opinion appears to be sharply divided. 



In volumetric analysis we use indicators, broadly speaking, to discover 

 when exactly equivalent quantities of solutions of bases and acids have 

 been added to each other. Experience has shown, however, that not only 

 do different indicators often give widely different results in one particular 

 experiment, but also that an indicator which appears to give satisfactory 

 results in one case will not do so in another. It becomes therefore 

 essential to discover precisely under what conditions the colour change of 

 an indicator takes place, and to define exactly the steps we must take to 

 select an indicator which will give the best results in any particular 

 volumetric operation. Since, as we have already seen, the change in 

 colour is produced, directly or indirectly, by a change in ionisation, this 

 problem reduces to one of defining the conditions under which an indicator 

 dissociates. 



Now let us consider, for simplicity's sake, an indicator l which is a 



1 The following method of treatment applies equally to basic indicators. Some 

 indicators have more than one colour change, and hence cannot be treated in so simple 

 a manner, but these are not in common use. Methyl red, an exception to the last 

 statement, has two distinct colour changes, but one of these is negligible in ordinary 

 circumstances, although it becomes of importance when this indicator is used for 

 the colorimetric determination of hydrion concentrations. 



