Revietvs — Watei' and Volcanic Activity. 231 



and the Grainger Shale (Carboniferous). The deposits in the Grainger 

 Shale are of very little importance. The ores of the Tellico Sandstone 

 are 'replacement ores', and are due to the leaching out by surface 

 agencies of the calcium carbonate from beds of ferruginous limestone. 

 The deposits of the Kockwood formation belong, however, to the 

 ' Clinton type ' of iron-ore ; they occur as lenticular beds of hematite 

 analogous to the strata of sandstone, shale, and limestone with which 

 they are associated, and consist of a mixture of ' fossil ore ' and 

 ' granular ore '. 



Fossil ore consists of aggregates of fragments of crinoids, corals, 

 brachiopods, and trilobites, many of which still consist of calcite, but 

 in places tliey are now wholly composed of ferric oxide. Granular 

 ore is a variety of oolitic ore, consisting of flattened grains lying with 

 their flat sides parallel to the bedding and cemented by ferric oxide ; 

 the grains often contain a nucleus of quartz. 



These hematite beds were formed by the deposition in the waters 

 of shallow bays or lagoons of sediments containing iron together with 

 calcium carbonate, silica, and alumina. Much of the calcium carbonate 

 was deposited as fragments of fossils. There is no evidence that the 

 iron was introduced from other beds of the Rockwood or any other 

 formation. It is held that the replacement of calcium carbonate by 

 ferric oxide may have taken place quite early in the history of the 

 deposits, and that the unweathered ores were formed in essentially 

 their present condition, contemporaneously with the associated 

 sandstone and shale. 



The distinction between the mode of formation of ' replacement 

 ores ' and of ores of the Clinton type is one of great economic 

 importance, for those of the Clinton type will not be expected to 

 siiow any regular decrease of iron content in depth, which is usually 

 the case with ' replacement ores '. 



The bulletin contains detailed descriptions and sections of the 

 mining prospects, and concludes with some useful notes on raining 

 nnd on the iron industry. 



VI. — Watku and Volcanic Activity. By A. L. Dav & E. S. Shepherd. 

 Bull. Geol. Soc. America, vol. xxiv, pp. 573-606, pis. xvii-xxvii. 



IN a previous issue of this journal there appeared a review of 

 a very striking work entitled Recherches sur V exhalaiso7i volcanique, 

 by Albert Brun, a chemist of Geneva. Brun was impressed by the 

 frequent absence of evidence for the widely accepted aqueous theory 

 of volcanoes ; he had accordingly devoted his spare time to the study 

 of various active craters, and had been led to the conclusion that in 

 paroxysmal eruptions water does not exist among the gases emanated. 

 Much of the evidence for this conclusion was derived from the study 

 of Kilauea, where Lowthian Green had been led to the same conclusion 

 as Brun as early as 1887. The American geologists have taken up the 

 challenge and have soTight to make a careful examination of the emana- 

 tions from Kilauea in order to test the validity of ]}run's conclusions. 



The gases examined by Brun were collected by means of tubes, 

 the end of which was 250 feet above the surface of the lava ; these 



