T. H. Holland — Constitution of Laterite. 69 



alumina in laterite exists, as it does in bauxite, in the form of 

 hydrous oxides. Kaolin must thus be removed finally from the 

 list of weathering products ; it is formed generally, perhaps 

 exclusively, by the action of subteiTanean vapours on aluminous 

 silicates. This conclusion necessitates a re-examination of many 

 so-called argillaceous substances ; many doubtless include the 

 kaolin already existing in the kaolinized aluminous silicates before 

 they are exposed to the weather. But it is probable that some 

 of the red clays of past geological ages, formed under subaerial 

 conditions, contain free hydrous oxides of alumina ; and for those 

 that are shown to contain hydrous silicate of alumina, it would be 

 well to test the possibility of a secondary reunion of aluminic 

 hydrate and free silicic acid. 



(2) To account for the fact that an aluminous silicate undergoes 

 a more complete disintegration under tropical conditions than under 

 the deep-seated and presumably high-temperature conditions of 

 kaolinization, the writer suggests that laterite is due to the agency 

 of lowly organisms, possibly akin to the so-called nitrifying bacteria. 

 With these there are probably forms akin to the bacteria which 

 oxidize and fix ferrous compounds, and which, precipitating the 

 silica in the colloid form, permit its removal by the dilute alkaline 

 solutions simultaneously formed. This would account for the facts, 

 (ffl) that laterite is confined to the tropics, or at least is more con- 

 spicuously developed under tropical conditions ; (6) that although 

 the laterite cover is 100 feet or more in thickness, there is a sharp 

 change from the soft decomposition-product to the absolutely fresh 

 rock below ; (c) that though laterite can form at temperate altitudes, 

 it is not observed in temperate latitudes, where, with a similar 

 average annual temperature, there is a prolonged winter; and 

 {d) that laterite is a superficial pi'oduct. 



(3) The development of concretionary structures in laterite being 

 accompanied by the loss of water, it is suggested that, in compounds 

 where a constituent is loosely held, the 'crystalline affinity' by 

 which physical molecules tend to unite and form crystals may 

 be more energetic than the chemical affinity ; that, in other words, 

 the heat of formation of a crystal of Fej O3 may be greater than the 

 thermal value of the compound [Fe- 0\ 3 H^ 0]. Hence crystalline 

 hematite may form spontaneously at the expense of amorphous 

 limonite. Similarly, the two molecules of water having a high 

 specific volume in gibbsite (Alj Hj O4 . 2 Hj 0) are expelled by the 

 formation of a crystal of diaspore (AI2 Hj 0^). Assuming that the 

 change of entropy can be expressed in thermal formulae, the value 

 [Al H 0'-, Al H -] must be regarded as greater than [Al H 0^ H^ 0], 

 and both limonite and gibbsite must be regarded as unstable at 

 tropical temperatures when both exist together. As the ferric 

 hydrate shows a greater tendency towards dehydration than the 

 corresponding hydrate of alumina, it possibly acts as a catalytic 

 agent ; and it would be unsafe, therefore, to assume, from the 

 evidence of laterite, that 'pure gibbsite possesses the power of 

 spontaneous dehydration. 



