the River Severn at Worcester. 



253 



pcared to pass into the chlorhydric solution ; for the latter 

 blackened considerably by evaporation to dryness, and the un- 

 dissolved sediment, when dried at 140° C, presented the same 

 weight as it did after subsequent ignition. The separation of 

 the iron from the mixed ferric and aluminic oxides was effected by 

 the action of amnionic sulphydrate after addition to the chlor- 

 hydric solution of much citric acid and excess of ammonia. The 

 other determinations were made by the methods in ordinary use. 

 The sediment of April contained a trace of manganese, and that 

 of February small quantities of potassium, sodium, and carbo- 

 nates. 1000 cub. centims. of the water of April contained "0798 

 grm., and of the water of February '0511 grm. of suspended 

 matter, of which the following Table shows the percentage com- 

 position : — 



Constituents of the Sediment from the River Severn above "Wor- 

 cester, calculated in 100 parts of the dry substance. 



The portion soluble in 

 dilute hydrochloric 

 acid contains of 



Silica 



Alumina 



Ferric oxide.... 



Lime 



Magnesia 



Organic matter 

 [loss on ignition] 



The portion insoluble 

 contains of 



Silica 



Alumina 



Ferric oxide 



Lime 



Magnesia 



Potash and soda. 



April 12,1806. 



February 14, 1867. 



II. 



1-886 

 2-358 

 4717 

 1-584 

 •679 



81-603 



1-956 

 2-391 

 4-782 

 2191 

 -627 



SI -300 



III. 



Mean. 



49-043 



16848 



11717 



1-230 



2-635 





1-921 

 2374 

 4-749 



1-887 

 •653 



6-878 





2040 

 3061 

 4-285 

 1-371 



i -881 



11584 



6878 



81451 



99-913 



82-730 



55-918 



14-898 



8-775 



•913 



•881 



traces 



II. 



Total. 



1-638 

 4-838 4-838 



82-730 



99-206 



The suspended matter of the Severn bears a certain resem- 

 blance in composition to that of the Rhine at Bonn, as analyzed 

 by Bischof on the 24th of March, 1851, when the river was 

 swollen and turbid; and it does not differ greatly from that of 

 the mud of the Nile, near Cairo, which was analyzed by Johnson. 

 Bischof has remarked that these deposits have a general resem- 

 blance to the constitution of clay slates, which, although of various 



