386 Examination of several Waters of Hartford, Conn. 



centration and evaporation highly colored, the color disappearing 

 upon ignition. This reaction was observed upon all the samples 

 from which this acid was separated, but a deficiency of material 

 precluded the quantitative estimation of it in all but No. 3, in 

 which the amount was 2-6 grains in one gallon, a quantity not 

 sufficient to saturate the excess of base. The large excess in 

 most of the waters may appear to exceed the bounds of proba- 

 bility, but repeated and concurring experiments show the results 

 to be correct. In the deficiency of knowledge in relation to the 

 modifications in which organic matter exists in combination with 

 bases in water, and the unsatisfactory methods which we possess 

 for its determination, it was preferred to state the results as ob- 

 tained without attempting to estimate the organic matter with 

 which the excess of bases is supposed to be combined. Its pres- 

 ence was abundantly proved by the action of nitrate of silver in 

 solution upon the specimen under examination. 



There were no indications of phosphates, apocrenic acid, or ot 

 potassa. The soap test, as might be inferred from a glance at the 

 table, indicates that they possess in an eminent degree the prop- 

 erty of hardness, becoming, with the exception of No. 3, imme- 

 diately and perfectly opaque upon the addition of the test to the 

 waters contained in a wine glass, accompanied in Nos. 1 ana 

 with the formation of a curdy precipitate occupying nearly halt 

 the bulk of the tested liquid. The order of succession, consid- 

 ering No. 3 the best, would be 3, 4, 2, 1, 5. T 



The following tables are the results of the analyses. Table L 

 shows the specific weights and the amount of solid contents as 

 found by evaporation and ignition in conjunction with a known 

 weight of pure anhydrous carbonate of soda, added to preven 

 loss by the decomposition of any chlorid of magnesium; the 

 weight of No. 5 is that found by analysis. The weights and meas- 



ures used were the French gramme and decimals, and the htie, 



hilt aro roi^MnaA in +Y\a toWac +n c-tnnAn^A TVnv orrainS. and. tile • 



but are reduced in the tables to standard Troy grains, ana ui» • 

 S. gallon of 231 cubic inches, the weight of one gallon distweu 

 water at 60° F. being taken as 58-328,886 grains. ^ 



Table II. gives the amount of the constituents as found by ' 

 tual analysis, without reconstruction. The discrepancy " etVi ?_ 

 the footings of this and the next table, arises from the loss o - 

 ygen which the bases undergo by the formation of haloid s 

 with chlorine. , • a 



Table III. In this table the results of the analyses are ^^^ 

 as they may be supposed to exist in the waters in their na '_ 

 state. The method recommended by Fresenius has been lo ^ 

 ed in reconstructing them, and the conclusions, though ia 

 criticism from the well known impossibility of accurately rP_ 

 ducing their original combinations, are supposed not to v 

 sentially from reality. 



