388 PROCEEDINGS OF THE GEOLOGICAL SOCIETY. [April 22 



Analysis, No. 87, by Dr. Yoelcker. Purple portion, Horderley 

 sandstone, Horderley, Shropshu'e. 



Water of combination 0*42 per cent. 



Soluble (P^'otoxide of iron 1-11 



[ feesquioxide of iron 3"51 



Alumina and other constituents 



soluble in hydrochloric acid . . . 2-47 

 Insoluble siliceous matter con-"| 



taining oxides of iron, weighed I 92*49 

 as sesquioxide 1-055 J 



100-00 

 Analysis jS"©. 88, by Br. Yoelcker. Green portion. 



Water of combination 1 -29 per cent. 



Soluble r Protoxide of iron 2-48 



i Sesqui oxide of iron 3-51 



Alumina and other constituents 



soluble in hydrochloric acid ..,. 5*37 

 Insoluble siliceous matter con- 1 



taining oxides of iron, weighed V 87 "35 

 as sesquioxide 0"668 J 



100-00 



In the green part of this example tbere is an excess of protoxide, 

 compared with that in the purple, without a corresponding diminu- 

 tion in the soluble sesquioxide, though the total amount of iron is 

 nearly equahzed by the excess of insoluble oxides of iron in the purple. 



There are, therefore, two distinct causes operating together in 

 determining the colour of the pale blotches in red beds and slates — 

 first, the exposure of the pre-existing green protosilicate by the par- 

 tial or entire removal of the red sesquioxide, and, secondly, the 

 conversion of any remaining anhydrous sesquioxide into the hydrous 

 form. 



The higher silicates, which tend to a yellow colour, may also be 

 in part the colouring-matter of the depleted blotches. Their sepa- 

 rate determination in the analyses has not been attempted ; and it is 

 sufficient for my present purpose to observe that they are numerous, 

 that their constitution varies with their occurrence, as double silicates 

 with lime, magnesia, &c., and that their colours range from green 

 to yellow. 



12. On the Ferruginous Banding of Yellow Sandstone. — Another 

 form of variegation, distinct from those before referred to, is seen in 

 the arrangement of hydrous sesquioxide of iron in yellow strata, as 

 bands and fields of colour disposed unconformably to the stratifi- 

 cation. 



That this ferruginous banding is independent of the original me- 

 chanical arrangement is obvious ; and the question it suggests is, 

 whether the variegation results from the infiltration of the iron on 

 to a light ground, or from a rearrangement of the yellow sesquioxide 

 originally pervading it. 



In some cases, as in fig. 53, the disposition of the ferruginous 

 lilies bears a resemblance to the way in which a coloured liquid 

 g tains and pervades an absorbent body (for example, stains of coff'ee 



