824 



NATURE 



[August 25, 192 1 



gneissic axis of Tyrone, the Caledonian area of 

 Armagh and Down, with the Newry granite in its 

 strike, the down-faulted Carboniferous series of Coal 

 Island, and the Cainozoic granite of the Mournes. 

 The Mesozoic beds, protected by the great plateaus 

 of basaltic lavas, are well seen encircling Lough 

 Neagh. Sheet i6 offers less variety, and shows the 

 rapid succession of Armorican anticlines and synclines 

 in the Devonian and Carboniferous systems round 

 Cork city. 



In a short paper to the International Congress of 

 Mathematicians, Strasburg, 1920, entitled " Une 

 application des polyn6mes d'Hermite h. qn probl^me 

 de statistique," Prof. Alfred Guldberg, of Christiania, 

 reaches the series recommended for the representation 

 of frequency curves and surfaces by Edgeworth in this 

 country, and by Charlier, Thiele, Bruns, and others 

 on the Continent. The large amount of inathematical 

 work on such subjects that is being done in Scan- 

 dinavia is noteworthy, but the application of the re- 

 sults of the mathematical work on a large scale to a 

 great variety of statistics seems to be required if we 

 are to estimate the ji«efulness of the work in practice. 



The report of the Royal Observatory, Hong-Kong, 

 for the year 1920 by Mr. Claxton, the director, shows 

 that the usual meteorological and magnetic results 

 have been continued. Automatic records of the tem- 

 perature of the air and evaporation were obtained with 

 a Richard dry- and wet-bulb thermograph, and the 

 direction and velocity of the wind with a Beckley and 

 a Dines-Baxendell anemograph. The amount of rain 

 is recorded automatically by a pluviograph, and the 

 amount of sunshine is registered by a Campbell-Stokes 

 recorder. Other observations are recorded by eye. 

 The mean barometric pressure and mean temperature 

 for tlie year were in fair agreement with the normals. 

 The total rainfall for the year was 10788 in., which 

 is about 24 in. above the normal. The fall in an hour 

 measured 144 in. on September 12, and 12-70 in. fell 

 in forty-nine hours on July 18-21. Tracks of sixteen 

 typhoons and four of the principal depressions which oc- 

 curred in the Far East in 1920 are given in the Monthly 

 Meteorological Bulletin for December. Observations 

 from the Philippines are now received in time for 

 insertion in the daily weather map. Wireless weather 

 telegrams were received from 140 ships in the course 

 of the year, and meteorological registers from 170 

 ships operating in the Far East. Upper-air research 

 is being considered, as is also the installation of a 

 seismograph. 



The Meteorological Magazine for July has an 

 article on the design of rain-gauges, which affects 

 largely the accuracy of rainfall measurements, now 

 being considered with greater assiduity than in the 

 past. The 5000 observers for " British Rainfall " 

 show the necessity for uniformity and precision in 

 the style of gauge. Universal adoption of the now 

 recognised standard patterns of rain-gauge is advo- 

 cated, and the rejection of certain obsolete patterns. 

 The forms approved are the Snowdon gauge and 

 patterns based on it, such as the Bradford gauge, the 



NO. 2704, VOL. 107] 



Meteorological Office pattern gauge, and the Sea- 

 thwaite gauge. Some of the essential features given 

 of an approved gauge are : The stout brass turned 

 ring terminating upwards in a knife-edge, exactly 

 5 or 8 in. in diameter, which forms the rim of 

 the gauge; the vertical cylinder, 4 to 6 in. deep, ex- 

 tending from the rim to the upper edge of the funnel, 

 which is intended to retain snow and hail, to prevent 

 the outsplashing of rain which has fallen upon the 

 funnel, and to reduce to a minimum the risk of loss 

 due to wind eddies; an inner collecting vessel, which 

 can be removed for measuring the fall without dis- 

 turbing the body of the gauge, the latter being slightly 

 sunk in the ground ; and a capacity of not less than 

 10 in. of rain for a daily gauge. Hints relative to 

 self-recording gauges are also given. Makers of rain- 

 gauges are asked to assist in the elimination of un- 

 desirable types of rain-gauge. 



Since its introduction in 1829 the Trevelyan rocker 

 has formed the subject of many scientific papers, but 

 they have all led to the conclusion that Faraday's 

 explanation of the motion of the rocker was sub- 

 stantially correct. According to Faraday the motion 

 is due to the expansion of the material of the sup- 

 port under one of the two ridges on the under- 

 surface of the rocker by heat communicated to the 

 material from the rocker. This expansion throws 

 the rocker on to its other ridge, allowing the first 

 portion of heated material to cool until it is again the 

 support for the rocker. This theory was put into 

 dynamical form by Davis in 1873, and has been 

 accepted as satisfactory. A recent study of the actual 

 motions of a rocker carried out by Prof. Chuckerbutti, 

 of Calcutta University, and given in vol. vi. of the 

 Proceedings of the Indian Association for the Cultiva- 

 tion of Science, shows, however, that the theory is 

 quite unsatisfactory. The tones produced are those 

 of the elastic vibrations of the system composed of 

 the rocker and its handle, and the pitch of each is 

 determined by these vibrations under the constraints 

 imposed on the rocker by the method of support. 



The third paper on the physical properties of clay, 

 read by Mr. A. S. E. .^ckermann before the Society of 

 Engineers, contains a record of forty-nine more ex- 

 periments, which carry our knowledge of this subject 

 considerably further. By boHing the clay and allow- 

 ing it to settle, some of the colloidal matter was got 

 rid of, and as a result the pressure of fluidity was 

 decreased bv about 25 per cent. When a disc is 

 pressed into a mass of clay, the mean radial speed 

 of flow of the clay underneath the disc is about one- 

 eighth the speed of penetration of the disc, and 

 the mean speed of penetration of the disc when the 

 load on it is just sufficient to produce the pressure 

 of fluidity is about i cm. per minute. Some in- 

 teresting experiments were made with the view of 

 ascertaining the behaviour of the clay immediately 

 below the disc; there appears to be a stagnant cap 

 of clay which remains in contact with the lower 

 side of the disc and travels with it. Even, under 

 considerable tangential stress there is no progressive 

 strain in clay containing. 25 per cent, of water, which 



