182 ME. JUKES-BROWNE AND TEOE. HARRISON- 



agglutinations partly argillaceous, with some opaque granules of 

 iron or manganese oxide. Some chips of hornblende or augite can 

 be identified, and occasionally a fragment which may be a greenish 

 glass or possibly a -flake of biotite. 



The other slide (kindly lent by Prof. Bonney) was from a depth of 

 2700 fathoms near St. Helena. This material is equally fine, with 

 the same aggregations and the same very minute, angular mineral 

 chips, but coccoliths are certainly present and in large numbers. 



It is evident that there is much similarity between these Atlantic 

 ' red clays ' and the argillaceous earths of Barbados, the special 

 points of resemblance being the minuteness of the constituents, the 

 angularity of the mineral particles, the presence of many fragments 

 of felspar or of a felspar-like mineral, and the rarity of particles 

 which can be recognized as quartz. 



When a marine clay formed in water of less depth, such as the 

 Gault, is compared with these a certain amount of similarity is 

 seen. There are many angular fragments, some very minute, in 

 an argillaceous base ; but most of the grains are more rounded and 

 at least subangular, quartz is more abundant, and many of the 

 felspar-fragments are larger and show indications of decomposition.^ 



§ y. The Chemical Composition oe the Eocks. 



The samples taken for analysis were in all cases drawn from 

 large quantities of the rocks selected in the field to represent as 

 closely as possible their average composition, so that the analyses 

 indicate the general composition of the local deposits, and not that 

 of small samples which may or may not be truly representative 

 of the beds under examination. We regard this as a matter of 

 great importance, and consequently have spared no pains to effect 

 our object. 



In the material of the Oceanic Deposits the main interest lies in 

 the existing proportions of calcium carbonate, of colloid silica, of 

 clay (silicates of alumina and iron not decomposable by cold dilute 

 hydrochloric acid), and of quartz. The following is the method 

 adopted wdth the view of determining as accurately as possible the 

 colloid silica present as such in the rocks, and of avoiding that de- 

 composition of some of the silicates present which necessarily 

 ensues w^hen clays or marls are treated with hot concentrated 

 mineral acids, a result which leads to an over-estimate of the 

 amount of colloid silica and to an under-estimate of the clay. 



The samples were all analysed in the air-dried state, but the 

 results have been calculated so as to show the composition of the 

 rocks free from hygroscopic water. They were reduced to very 

 fine powders in an agate mortar, and two quantities of from four 

 to five grammes each were separately treated with dilute (2| Y^) 

 hydrochloric acid in the cold (30° C). After the acid in the first 



'^ These remarks are confirmed by the descriptions of the mineral particles 

 of ' red clay ' and ' bine mud ' in Dr. John Murray's ' Eeport on the Deep-sea 

 Deposits' (see pp. 196 and 231), 1891. 



