20 



KNOWLEDGE. 



January, 1915. 



Various theories have been put forward to account for 

 their origin. They cannot be uphfted by subterranean 

 gas, as borings ha\-e failed to reveal any gas reser\'oirs, 

 while their uniform height and occurrence only at the ends 

 of the " passes " also militate against such an origin. The 

 most probable explanation is that they are due to the 

 seaward flow of a semi-fluid clay under the shallow water 

 near the ends of the passes, and that this flow is caused 

 by the pressure of the sediments deposited by the river, 

 the upward buckling being due to the resistance offered 

 by the so-called " foreset " beds, which are comparatively 

 thick deposits of coarse sandy material just off the shore. 



CLARE ISLAND. — In one of the recent reports on the 

 Clare Island Survey, T. Hallissy gives an account of the 

 geology of the island. The effects of differential weathering 

 are well shown by the rugged nature of the surface, the 

 knolls and ridges generally consisting of hard, gritty rocks, 

 while the valleys and depressions are formed in soft shales. 

 There is abundant evidence to show that the island was 

 covered by the Central Irish glacier at the period of 

 maximum glaciation, roches inoutonnies and glacial striae 

 being found at considerable heights above sea level, while 

 the lowest superficial deposit on the island is a fine-grained, 

 compact boulder-clay containing many striated blocks of 

 Umestone. The presence of the limestone shows that the 

 direction of the invading ice was a httle south of west, and 

 confirmation of this is obtained from the facts that the 

 axis of a large drumlin on the south side of the island is in 

 the same direction, and that the polished sides of the roches 

 moutonnees face eastwards. The presence of a coarser 

 boulder-clay above the compact deposit indicates a later 

 glaciation, which seems to have been of local origin, as the 

 boulders are mainly granite and serpentine, which can be 

 matched by rocks occurring close by on the mainland. 

 There are also a number of Post-Glacial deposits, such as 

 blown sand, peat, and alluvium, the peat containing 

 numerous trunks of Scotch fir. 



The island may be divided into two parts : a low-lying 

 tract, which seems to be part of the great Carboniferous 

 peneplain of Central Ireland, and the highlands, wliich are 

 probably remnants of the Pre-Carboniferous plateau. 

 Although the island was separated from the mainland until 

 the Glacial period, a Post-Glacial connection with the main- 

 land is most probable, as the inter\'ening strait is very 

 shallow, the bottom being about the same horizon as the 

 submerged forests which are common in the district. 



GEOLOGY. 



By G. \V. Tyrrell, A.K.C.Sc, F.G.S. 



CHIASTOLITE IN SOUTH AUSTRALIA.— A great 

 find of chiastohte crystals at Bimbowrie, South Austraha, 

 recently made by Mr. G. R. Howden, is fuUy described by 

 Dr. (now Sir) Douglas Mawson in the Memoirs of the Royal 

 Society of South Australia, Volume II (1911). The crystals 

 occvu- in a schistose rock, and in two localities thej- have 

 weathered out from the matrix in such quantities that the 

 ground is thickly stiewed with clriastolite as a residual 

 surface mantle. " The properties of hardness and resistance 

 to decay of chiastohte, the comparative softness of the 

 matri.x, and the arid conditions fostering wind erosion, and 

 thereby removing the finer products of disintegration, all 

 promote the accumulation of such a chiastohte mantle." 

 The crj'stals ha\-e been produced by the thermal meta- 

 morphism of a Cambrian slate in contact with both acid and 

 basic intrusive igneous rocks. The crystals are of huge size. 

 Some fragments picked up were seven centimetres in 

 diameter ; others were eighteen centimetres long. Many 

 of the crystals are distorted owing to the crushing of the 

 matrix. The cross-shaped zones of inclusions characteristic 

 of chiastohte are well shown in these crystals An interest- 



ing mode of alteration to an intimate mixture of musco\dte 

 and corundum is also recorded. 



ROCKALLITE.— The lonely island of Rockall, situated 

 in the North .\tlantic, about one hundred and seventy miles 

 west of the Hebrides, and about half-way between Ireland 

 and Iceland, is largely composed of a unique igneous rock, 

 which has been termed " rockallite." This rock was 

 described and named by Professor J. \V. Judd in 1897. 

 Only two landings had previously been made on the island, 

 and in these two cases it was only possible for sailors to 

 jump from a boat on to a projecting point of rock and to 

 knock off pieces, which they threw into the boat. An 

 expedition of 1896 was unable to land, but a landing was 

 effected a few years ago, when a Uner from North America 

 ran on to it, and was wrecked with loss of life. Had a 

 geologist been present amongst the survivors the oppor- 

 tunity would doubtless have been taken to secure more 

 specimens. 



Professor Judd described the rock as a holocrystalline 

 admixture of soda-pyroxenes (aegirine and acmite), albite 

 felspar, and quartz, named in the order of crystallisation. 

 As the original chemical analysis of this rock was incom- 

 plete, and as it is of unusual systematic importance. Dr. 

 H. S. Washington has made a new and complete chemical 

 analj^sis [Quarterly Journal Geological Society, June, 1914), 

 which has disclosed some interesting features. Both 

 analyses show high silica, ferric oxides, and soda, combined 

 with low alumina, ferrous oxide, magnesia, lime, and potash ; 

 but Dr. Washington's analysis, too, also demonstrated the 

 presence of unexpectedly large amounts of the rare minor 

 constituents, zirconia and ceria, which occur to the extent 

 of 1-17 and -37 per cent, respectively. Since minerals 

 noted for the abundance of these constituents are absene, 

 it is concluded that they must be contained in the pyroxent. 



The rock is a highlv sodic variety of the alkali-granites, 

 and is comparable to aegirine-granites from Madagascar 

 and Corsica. In the American Quantitative Classification 

 it falls into the sub-rang " rockallose " ; and an indication 

 of the exceptional character of the rock maj- be found in 

 the fact that the two analyses of rockallite are the only 

 representatives of this sub-rang amongst eight thousand 

 analyses of igneous rocks collected by Dr. Washington. 



AEGIRINE AND .\CMITE.— The presence of zirconia 

 and ceria in rockallite and the ascription of these con- 

 stituents to the pyroxene present in the rock lead to the 

 interesting speculation that they may be the cause of the 

 differences between aegirine and acmite. Both of these 

 pyroxenes occur in rockallite, and they are unquestionably 

 closely related chemically, crystallographically, and optically; 

 but there are marked differences in colour and in character 

 and strength of pleochroism, indicating a certain chemical 

 difference between the two. Dr. Washington suggests that 

 the presence of zirconia and ceria may be characteristic 

 of acmite, and the cause of its difference from aegirine. Acmite 

 is a brown mineral, whereas aegirine is green. It is pointed 

 out that all cerium-rich silicate minerals, such as melano- 

 cerite, allanite, cerite, mosandrite, and rinkite, are yellow 

 or yellowish brown, with weak pleochroism. Similarly the 

 zirconia-bearing pvroxenes, such as rosenbuscliite, lavenite, 

 wohlerite, and hiortdahte, are all yellowish, and have 

 weak pleochroism. It is possible, therefore, that ceria and 

 zirconia form an essential part of the molecule of the brown 

 acmite, whilst they are wanting in the green aegirine. This 

 is, however, merely a suggested hypothesis, and needs to 

 be tested by chemical study of aegirine and acmite, which 

 is being undertaken by Dr. Washington. 



METEOROLOGY. 



By WiLLi.AM Marriott, F.R.Met.Soc. 



THE WEATHER OF JANUARY.— Dr. C. Leeson 

 Prince, who was for many years a very keen observer in 



