the River Severn at Worcester. 



261 



Precipitates produced by concentration of 1000 cub. centims. 

 of the Waters to about one-sixth. 





April. 



July. 



October. 



February. 



Above Haw- 

 ford Brook. 



At Eailey's 

 Ferry. 



Total precipitate... 

 containing — 

 Lime 



grm. 

 •0980 



•0515 

 •0049 

 trace 

 •0058 



grm. 

 •0833 



•0450 



•0050 



trace 



undetermined 



grm. 

 •0924 



•0483 

 •0038 

 trace 

 •0030 



grm. 

 •0479 



•0245 



•00098 

 trace 



grm. 

 •0480 



•0246 

 •0020 

 trace 



Magnesia 



Alumina &c. . 

 Silica 











In the Table which follows I have given the results which I 

 have already mentioned as having been obtained in 1857-58; 

 although imperfect, they are of interest as affording points of 

 comparison with the waters of the past year. So far as I can 

 learn, periods of drought preceded the taking of all the waters, 

 with the exception of a little rain which occurred high up the 

 river towards the end of May 1858 ; and the height of the river, 

 the concentrated condition of the three first waters, and the ab- 

 sence of sediment agree with these observations. These cir- 

 cumstances preclude a strict comparison with the waters of the 

 past year. It may, however, be roughly remarked that the 

 chloride and sulphate appear to have diminished, whilst the 

 silica has increased since 1857 ; but even these changes may be 

 entirely due to the altered conditions of the rainfall. The esti- 

 mations of loss on ignition are undoubtedly too high; for the 

 treatment with diammonic carbonate, as then adopted, was of too 

 short duration fully to replace the lost carbonic anhydride. 



With regard to the much debated question of the mode of 

 combination in which the saline constituents should be assumed 

 to exist in spring- and river- waters, it appears to be most desi- 

 rable, in the absence of a definite knowledge of the chemical 

 changes which occur among salts in solution, to endeavour to 

 trace the source whence the different salts have been derived, 

 ♦and to consider that in their intermixture they preserve their 

 original forms of combination. It is frequently taken for 

 granted, for instance, that when solutions of calcic or magnesic 

 salts meet with solutions of alkaline carbonates, a complete in- 

 terchange of constituents takes place, however dilute the solu- 

 tions may be ; and the calcic or magnesic carbonate assumed to 

 be formed is said to be dissolved either by the water, or by more 

 or less carbonic acid, or by the agency of some solvent salt. 

 Experiment, however, shows that, when carbonic acid and sol- 



