274 



REPORTS ON THE STATE OP SCIENCE. 



titrating solution is enough to make a sharp change in the colour. For 

 methyl red about a drop of alkali is enough to decide its end-point ; but 

 the result so obtained (49-95 c.c.) may differ by about 1 part in 1,000 

 from the true value. Phenolphthalein gives an equally sharp end-point, 

 the error of which is a little over 1 part in 1,000 in the other direction. 

 We could therefore titrate with both in the solution, and take as the most 

 correct result the mean of the two values obtained. For most purposes, 

 however, an accuracy of 1 part in 1,000 is ample. Methyl orange, however, 

 comes on the steep part of the curve, and its colour changes comparatively 

 slowly. Also the final result, which will be about 49-8 c.c, is considerably 

 less accurate than those given by the other indicators used. Litmus 

 should give the exact point ; many workers find it, however, an incon- 



CONC^ H ION 



Fig. 2. 



venient indicator to use ; perhaps if its constitution could be determined, 

 and a pure product prepared, it would again come into extensive use. 



Still more striking is the difference between the results obtained by 

 using various indicators when one of the titrating liquids is a weak electro- 

 lyte. In this case, as is well known, the concentration of H° ions in the 

 solution when exact equivalents of acid and base are present is not the 

 same as in pure water, owing to the hydrolysis of the salt, or, in other words, 

 incomplete combination of the acid and base. Further, excess of acid or 

 base does not alter the concentration of H° ions to a large extent, since it is 

 partly used up in destroying hydrolysis ; also, if the weak electrolyte is 

 in excess, the degree of its dissociation, small in any case, is further reduced 

 by the presence of the neutral salt. Even in the titration of ammonia by 

 HC1, and acetic acid by soda, where hydrolysis is as a matter of fact 

 extremely small, and easily destroyed altogether by a slight excess of 



