Reviews — E. 8. Simpson — Rocks, etc., W. Australia. 327 



remarkable structure, and has been described by Cohen, Kynaston, 

 and Hatch & Corstorphine. It presents the appearance of a crater 

 lake, dry in winter, but until disturbed by working it contained 

 a brown brine in the rainy season. It is surrounded by a rim of 

 granite some 200 feet high, with a gentle slope on the outside and 

 a steep slope on the inside. The rim of granite is a most remarkable 

 feature. The floor of the pan has been pierced by borings to a depth 

 of 1,100 feet, and it is found to be composed of alternate layers of salt 

 and mud. The salts have on the average very nearly the theoretical 

 composition of trona, Na 2 C0 3 . NaHC0 3 . 2H2 0, with some sodium 

 chloride but very little sulphate. The evidence as to the origin of 

 these salt deposits is scanty and not very satisfactory ; they may be 

 derived from the decomposition of the perthitic felspar of the granite 

 or from underground waters of either volcanic or meteoric origin. 

 Opinions differ as to the origin of the pan itself. It may be due to 

 the impact of a large meteorite, like Coon Butte in Arizona, or it 

 may be a volcanic pipe like the Maare of the Eifel or Geitsi Gubib in 

 South- West Africa ; the latter appears the more probable, although 

 the upward bulging of solid granite in the form of a dome or even of 

 a ring seems rather difficult of belief. It is evident that the last 

 word has not yet been said on the origin of structures of this and 

 related types. 



R. H. Rastall. 



II. — Analyses of Western Australian Rocks, Meteorites, and 

 Natural Watebs. By Edward S. Simpson, B.E., B.Sc, F.C.S. 

 Bull. !Nb. 67, Western Australia Geological Survey. 



THIS bulletin is a collection of the records of all the analyses 

 performed in the Government laboratory between the years 

 1897 and 1916. It deals with rocks, meteorites, and natural waters. 



The rocks are divided into igneous,metamorphic, and sedimentary, 

 and these divisions are subdivided according to silica percentage to 

 facilitate reference to the lists. The compiler states that these 

 analyses are not to be taken as a quantitative representation of the 

 rock-types of Western Australia, but that they deal, as is natural, 

 with the types of greater interest to the miner and the agriculturist. 

 In addition to the analyses of the igneous rocks, Brogger-Hobbs 

 diagrams are given of many types, and in most cases the " mode" 

 and "norm" are given, and the rock is classified according to the 

 quantitative system. 



Most of the rocks fall into the described groups, but two of them, 

 an amphibolite and a gabbro, fall inside no known group, and have 

 a sufficiently definite composition to warrant placing them in a new 

 group (IV, i, i, 3, 2), for which the name Murchisonose is proposed. 



In the analyses of the sedimentary rocks two points of interest 

 present themselves. The first is the composition of the dune sands 

 from Dongara, which consist principally of foraminifera, calcareous 

 algae, and powdered mollusca, and have at times as little as 3-8 per 

 cent of silica, being almost entirely made up of calcareous matter. 

 The other point of interest is the excessive humidity of the 



