T. H. Holland — Constitution of Laterite. 61 



showing that with the gibbsite there is a certain amount of a lower 

 hydrate, probably diaspore, Alg O3 . H2 0. 



Microscopic examination shows that the ferric hydrate occurs as 

 a casual staining between the fine scales of gibbsite, the two being 

 mechanically mixed ; and as the iron-staining varies in colour from 

 yellow, through brown to dark red, the hydrate varies in com- 

 position, including in different specimens, or at different stages of 

 dehydration in the same specimen, the various imperfectly defined 

 forms of ferric hydrate which have been given specific names — 

 limnite, Fcj O3 . 3 H, ; xanthosiderite, Fe, Og . 2 H, ; limonite, 

 2 Foa O3 . 3 H3 ; gothite, Fe2 O3 . H^ ; and'turgite, 2"Fe2 Oj . H^ 0. 



The results obtained by the analysis of a ' laterite sand ' — the 

 so-called ' low-level ' or detrital laterite of Indian geologists — show 

 a still smaller proportion of water, indicating a larger proportion 

 of diaspore and probably of the less hydrated forms of ferric oxide. 

 In this case, also, thei*e is an admixture of quartz-grains, some, as 

 they ought to be, partly rounded. 



Now then, there appears to be no doubt that during the weathering 

 of aluminous silicates in the tropics, the silica, alkalies, and alkaline 

 earths are removed in solution, whilst the alumina and ferric oxide 

 become hydrated and remain behind mechanically mixed, often as 

 pseudomorphs of the original mineral structures of the rock. In 

 microscopic structure and in chemical constitution Professor Bauer 

 has now shown that laterite reproduces the essential characters of 

 bauxite. As laterite is so abundant in tropical, moist climates, its 

 occurrence offers a simple explanation for the origin of bauxite, and 

 is merely another expression of the results of the prevalence of 

 tropical conditions in some higher latitudes during past geological 

 ages. With fair proof of the fact that the aluminous base of laterite 

 is a hydrate and not a hydrous silicate, several questions naturally 

 present themselves : to call attention to some of these is the object 

 of this paper. 



II. Possible Okganio Origin of Laterite. 



The Seychelle specimens do not tell the whole story ; for there 

 are laterites and laterites. In India we distinguish between ' high- 

 level ' and '■ low-level ' laterites : the former are lying on the rock- 

 masses from which they have been formed without removal from 

 the place of origin. Amongst these we have lateritic caps below 

 elevations of about 5,000 feet in the Peninsula of India, with marked 

 concretionary structures, whilst at higher levels the decomposition- 

 products show less pronounced signs of concretionary structures, 

 and often retain perfect casts of the original structures of the rock 

 from which they have been derived. ' Low-level ' laterites have 

 a detrital origin, and consist of sand-grains in which quartz as usual 

 predominates ; they merely differ from other detrital rocks in the 

 abundance of lateritic cement. In this form, which is our most 

 abundant type, and the one consequently most taken in the 

 regretfully few analyses which have been made, there is a much 

 smaller quantity of alumina and a higher percentage of iron. This 



