A N Y fundamentally The indicator solutions formerly used gave a atid (2) the size of the reaction important a d v a n c e s p~ in unbu#ered solu~ions more than tiE vessels. Their values for pI1 h a v e b e e n m a d e in were determined by the coloritrue p H as determined with properly adjusted lnetric water purification practice in the past thirty years, not so much (kohydrie) indicators. At l e d 2 liters of water used were exceedingly dilute, b e c a u s e of the introduction of should be used for all jar tests for plant control varying from 0.0002 . $ I to 0.002 new m e t h o d s as h e c a u s e of in order to eliminate effects due to the size of the M , and were in most cases very a better understanding of the cessel. slightly buffered. Recent work various principles involved in by Xolthoff and Kameda (/I), the methods o r i g i n a l l y emSulfate in amounts front 25 lo 250 p . p . ni. and Fudge (8), Sclllegel ployed. ezlend the zone of rapid floc formation for to the and Steuher ( l o ) , F~~~~~~ and The r e s e a r c h work o f t h e acid s&. Chloride ion ezerts relatiliely little Acree ( 5 ) , and others has shown colloid chemist explains the proceffect below 6.5 p H but does eslend the zone that serious errors may result in ess of coaguliation ( f o r m a t i o n toward the si&. ~~b~~~ the determination of the pH of of floc) for the removal of color results are.found to hold on a serniplant scale. unbuffered solutions by the color turbidity. Colloid chemistry orimetric method unless certain has showii t h a t t h e o r i g i n a l Conlin~llous stirrirt!f is absolutely rwcessary precautions are taken. The pH ideas concerning the action of the conduct of jar tests which will check euch of tile indicator solution itself floc were largely incorrect, but other and give accurate data for plant operation.must not he very different from that a more or less fortuitous that of the unbuffered soluti~n choice of coagulants was a happy treated o r it will p r o d u c e a one. Even with full knowledge of tlie phenomena, no better correspnding change in the pH of the solution. A coloricoagulants have been found. Saville (9) shows that the color mctric method for adjusting indicator solutions described by present in natural waters is colloidal.Acree and Fswcett (1) permits adjustment to any desired Miller (7) seems to have been the first to p i n t out the ef-pH value or series of values within the useful range of the ficient role of the trivalent aluminum and ferric ions in neu-particular indicator. It is thus possihio to employ an inditralioiiig and precipitating the negatively charged color col-cator solution termed "isohydric" of approximately the same loids, and to conclude that the absorptive property of the pII value as that of the solution being tested. Isohydric indigehtinous floc plays, as a rule, a relatively unimportant part cator solut.ions are recommended wherever accurate results in the process. on unbuffered solutions are desired. This p i n t becomes of Miller summarizes his own work together with t...