November 23, 1905J 



NA TURE 



9i 



Prof. A. Young described a remarkable case of an 

 artesian well in the Karroo which shows a daily fluctu- 

 ant m in its discharge. The curve obtained during some 

 week's by a self-recording apparatus is very regular, and 

 has a period of almost exactly nh hours. The ampli- 

 tude shows a marked variation, corresponding in time 

 with the phases of the moon, and analogous to marine 

 spring and map tides. The outlet is more than 2700 feet 

 above sea-level, and the author suggests that the water, 

 which contains a large amount of inflammable gas, is 

 It, reed up from a great depth through fissures by the 

 pressure of natural gas, and that the observed fluctuation 

 is a minor effect, due to the moon, superimposed on the 

 effect of the constant gas-pressure. The phenomenon is 

 scarcely affected by barometric changes. 



At a joint meeting with the Geographical Section on 

 the second day, Mr. H. C. Schunke-Hollway gave an 

 account of the physical geography of Cape Colony. Mr. 

 Rogers read a paper on Glacial periods in South Africa, 

 in which he described the glacial deposits til Table Moun- 

 lain Sandstone- (Silurian?) and of Dwyka- (Carboniferous) 

 age, each formed of materials derived from the north. 

 There is no satisfactory evidence of glacial action in 

 later times, the glaciated forms of certain hills in Griqua- 

 land West, cited by Stow, being now known to have been 

 produced at any rate not later than Dwyka times, since 

 similar forms may be traced underneath the surrounding 

 Dwyka conglomerate. They have been preserved by a 

 thick covering of Dwyka and other beds, which have only 

 recently been removed. 



Prof. A. Penck (Vienna) contributed a paper on changes 

 i f climate as shown by variations of the snow-line and 

 upper tree-limit since Tertiary times, in which, from a 

 consideration of the geological evidence as to the relative 

 height of the snow-line and tree-line in Glacial times, he 

 drew conclusions as to the cause of the glacial con- 

 ditions. The facts pointed to a lowering of temperature 

 as the cause of the glaciation rather than to an increase 

 of precipitation. Prof. Penck suggested that an examin- 

 ation of the higher parts of the Drakensberg might prob- 

 ably reveal traces of a Pleistocene Ice age in South Africa, 

 though hitherto satisfactory evidence of this has been 

 wanting. 



Prof. YV. M. Davis, of Harvard, brought forward 

 evidence for the sculpture of mountains by glaciers. 1 He 

 based his arguments principally upon the marked differ- 

 ence in form between valleys proved in other ways (e.g. by 

 the presence of striations) to have been once glaciated, 

 and those which have not been glaciated, the differences 

 being in nature and distribution such as glaciers would 

 cause on the assumption that they could erode. 



Papers were also read by Prof. Sollas, on the continent 

 t i Anita in relation to the physical history of the earth; 

 by Prof. J. Milne, on recent advances in seismology; by 

 Mr. E. H. L. Schwarz, on " Baviaan's Kloof, a Contribu- 

 tion to the Study of Mountain Folds "; and by Mr. 11. T. 

 Ferrar, on the geology of South Victoria Land, giving 

 the results of his observations on Antarctic rocks and 

 glaciers made during the voyage of the Discovery. 



Prof. Sollas sketched a possible way in which the present 

 distribution of oceans and continents on the globe may 

 have arisen. The earth is not strictly a spheroid, but 

 resembles an ellipsoid, of which the shortest axis passes 

 through the poles, while the longest lies in the plane of 

 the equator and emerges in Central Africa. The distribu- 

 tion of land and water is such as would obtain if the 

 earth had the form of a pear which had been somewhat 

 compressed in the direction of its core, and thereby caused 

 to bulge laterally. Africa would be situated on the broad 

 end of the pear, and would represent the remains of the 

 primeval continent — a supposition consistent with the 

 known absence of marine sediments over the greater part 

 of the interior, notwithstanding the thick accumulations 

 of flat-bedded strata existing there. 



Mr. Schwarz 's paper contained an account of a remark- 

 able piece of geological structure observed in the valley of 

 the Baviaan's River, a tributary of the Gamtoos River, in 

 the neighbourhood of Port Elizabeth. On the Baviaan's 

 River occur certain outliers of Enon conglomerate (Creta- 



x The papers by Prof. Penck and Prof. Davis will be published shortly in 

 the Geographical Journal. 



ceous) which have been found by bore-holes to occupy steep- 

 sided, basin-shaped depressions with no outlet, in Palaeozoic 

 rocks of the Bokkeveld and Witteberg series (Cape 

 system). The basins are bounded by faults or steep dip- 

 slopes, and are explained as having been formed by two 

 series of cross-foldings trending E.S.E. to W.N.W. and 

 N.E. to S.W., which took place while the country was 

 covered with the Enon conglomerate, the latter being 

 faulted down upwards of 1000 feet. The author objects 

 to the usual explanation of rock-folding as produced by 

 a direct tangential thrust against an obstacle, caused by 

 shrinkage of the earth's crust, and suggests that it may 

 in fact be gradually produced by earthquake-waves 

 travelling through one kind of rock (say sedimentary beds 

 resting on granite) and encountering a mass of rock having 

 a different modulus of elasticity (as, for example, a boss 

 of the underlying granite). The effect of this would be 

 to heap up the strata in folds against the obstacle, some- 

 what as when waves break on the shore. 



At Johannesburg a considerable number of the papers 

 were, appropriately, of mineralogical and petrographical 

 interest. 



The proceedings opened with the delivery of the presi- 

 dential address by Prof. Miers (Nature, August 24, vol. 

 lxxii. p. 4051. 



Prof. J. \V. Gregory followed with two papers of special 

 interest to gold miners. In one of these, on the Rhodesian 

 Banket, he stated that he had found during a recent ex- 

 amination of the district that the name had been applied 

 to several different rocks which are locally auriferous — 

 not only to an undoubtedly sedimentary conglomerate form- 

 ing the main mass of the material, but also to crush- 

 conglomerates and breccias, and to a diorite dyke with 

 segregations of amphibolite. The Rhodesian conglomerate 

 may probably be rightly called Banket, but differs con- 

 siderably from the Banket of the Rand in its fluviatile 

 origin, the greater variety in size and composition of the 

 pebbles, and its probably greater age. The question as 

 to the right of the Rhodesian deposit to the name of 

 " Banket " aroused considerable discussion. 



In his second paper (on the Indicators of the goldfield 

 of Ballarat — a study in the formation of gold pockets) 

 Prof. Gregory showed the secondary origin of the so-called 

 "indicators," or thin iron-stained bands, which traverse 

 the slaty country-rock of Ballarat and lead to rich 

 pockets of gold at the points where they intersect the 

 otherwise barren quartz reefs. The indicators are shown 

 by microscopic and field evidence to be narrow seams of 

 chlorite or rutile needles, which are not quite, though, as 

 a rule, nearly, parallel to the bedding, and cannot there- 

 fore be of sedimentary origin. 



Prof. R. Beck, of Freiberg, gave a summary of recent 

 investigations on the origin of pegmatites as products of 

 the crystallisation of the residual mother liquors of a 

 solidified plutonic magma. Certain ore-veins have been 

 formed thus as metalliferous pegmatites, for example the 

 tin veins of Zinnwald and Embabaan, the copper ores of 

 Telemarken and the auriferous quartz-reefs of Berezowsk, 

 the Yukon district and Passagem, and other places in 

 Brazil. The presence of tourmaline in certain gold-quartzes 

 bears out this view of their origin. 



Prof. A. P. Coleman, of Toronto, dealt with the mag- 

 matic segregation of sulphide ores. The recent complete 

 mapping of the eruptive sheet with which the nickel-ore 

 deposits of Sudbury (Ontario) are all connected, shows 

 that the Sudbury ore is, like the pyrrhotite nickel ores of 

 Norway described by Vogt, really a product of segregation 

 from the rock, of which it forms an integral part with 

 every gradation between ore and rock. Gravitation has 

 probably played a large part in the segregation process. 



Prof. Grenville A. J. Cole read a paper on marginal 

 phenomena of granite domes, in which he upheld the view 

 that banded gneissic rocks are due rather to the incorpor- 

 ation of trie surrounding rocks with the materials of an in- 

 vading granite than to simple dynamic metamorphism ; 

 the banding is produced by igneous flow, and is especially 

 marked in cases where the absorbed rocks were sedi- 

 mentary or already foliated. 



On the second day Mr. G. W. Lamplugh gave his 

 report of a journey, made under the auspices of the 

 association to examine the zigzag gorge of the Zambesi 



no. 1882, VOL. 73} 



