io8 



NATURE 



[March 30^ 1916 



with a trough fewer than 700 miles away. Further, 

 the author contends that his results are of value in fore- 

 casting rain. Taking the normal cirrus direction as 

 west, a departure of 8° to the north was associated 

 with general rains, 4° to the south with partial rains, 

 and 12° to the south with a failure in rain. Mr. 

 Quayle contends his results show that cirrus move- 

 ments can be used as guides in forecasting- the weather, 

 and gives some general rules in application to the 

 weather of Melbourne. 



The usual method of cartographical representation 

 of density of population based on the consideration of 

 each census district as a whole has many drawbacks. 

 Not only does it entail frequent sudden breaks in con- 

 tinuity when a district with a high degree of density 

 adjoins one with a low, but it has the fatal objection 

 of being founded on purely arbitrary political divisions. 

 A new and far more scientific method has been worked 

 out by Mr. B. C. Wallis, and described, with specimen 

 maps of Hungary, in the Geographical Journal for 

 March, 19 16 (vol. xlvii., No. 3). Mr. Wallis has taken 

 the average density for each commune, the smallest 

 area for which there are returns, and, using these figures 

 as " spot heights," applied the principles of contour 

 lines. The result is a justification of the method. In 

 like manner, Mr. Wallis has applied this method to 

 illustrate the distribution of nationalities in Hungary, 

 and has produced an instructive map, wjiich is of far 

 greater usefulness than the old-fashioned and rather 

 meaningless chart in which the percentage of each 

 nationality is given in figures of different colours in 

 each commune. The paper goes on to deal with some 

 of the results of the distribution which the map illus- 

 trates. 



Calculating machines form the principal subject of 

 a paper by M. Leonardo Torres y Quevedo in the 

 Revue ginirale des Sciences (xxvi., 21), under the 

 title, " Essais sur I'Automatique." It deals with the 

 construction and principle of devices, mainly electrical, 

 for performing arithmetical and other operations with- 

 out human intervention. A possible cause of error, 

 namely, the production of electric sparks, is considered, 

 and a method of obviating this risk is suggested. 



Results of magnetical, meteorological, and seismo- 

 logical observations for the month of August, 1915, 

 and the annual report for 19 14, of the Royal Alfred 

 Observatory, Mauritius, show that an exceedingly 

 valuable series of results is being secured under the 

 directorship of Mr. A. Walter. A table is given of 

 the means and extremes of the principal meteoro- 

 logical elements for the year compared with previous 

 results from about 1875. Other tables give the 

 monthly departures from average of the various 

 meteorological elements, also the' mean hourly velocity 

 of the wind for the eight five-year periods from 1876 

 to 1914, and other information of a meteorological, 

 magnetical, and seismological character. During the 

 visit of the German cruiser Emden to the station at 

 Cocos Island the meteorological instruments were 

 destroyed. Much valuable information is secured by 

 the observaton,- staff from the logs of ships traversing 

 the Indian Ocean, and by this means details are 

 obtained relative to the formation and movement of 

 cyclones over the Indian Ocean. 



On July 15, 1896, the Pacific coast of Japan was 

 attacked by a tremendous ocean wave, the sea off the 

 coast of Miyato rising and falling alternately. The 

 second crest reached the maximum height, and the 

 oscillation then decaved rapidly. The hypothesis that 

 this disturbance was due to a sudden depression of the 

 sea-bottom over a limited area forms the subject of a 



NO. 2422, VOL. 971 



I hydrodynamical investigation of the wave motion 

 [ theoretically produced by such a disturbing cause. 

 This theory, which assumes the sea-bottom to be of 

 uniform depth and the depression to be circular, 

 appears to give results according with those of ob- 

 servation to a reasonable degree of closeness. The 

 times from the beginning of the earthquake to the first 

 wave, the interval between the first and second, and 

 the fact that the second is the highest, are results in 

 which theoretical calculations accord fairly well with 

 results of observation. The paper, which is by Keiz6 

 Sano and Kea Hasegama, is published in the Pro- 

 ceedings of the Tokyo Mathematico-Physical Society, 

 viii., 7. 



In pamphlet No. 20 of Mededeelingen en Ver- 

 handeiingen of the Koninklyk Nederlandsch Meteoro- 

 logisch Instituut (Utrecht, 1915, pp. 24), P. H. Gall6 

 discusses steamer routes from Durban and Cape Agul- 

 has to various parts of the Dutch East Indies. This 

 subject has recently attracted a certain amount of 

 attention on account of the increase of shipping be- 

 tween these points, and since the paper was written 

 Dutch East Indian passenger liners have started going 

 round by the Cape instead of through the Suez Canal. 

 The factors affecting a choice of routes are : — (i) The 

 equatorial counter-current and other easterly currents 

 near the equator; (2) the south equatorial current and 

 the easterly drift south of lat. 35° S. ; (3) the West 

 Australian current; (4) the N.E. and S.W. monsoons; 

 (5) the S.E. trade winds; (6) easterly winds south of 

 lat. 30° S. ; and (7) westerly winds S. of lat. 35° S. 

 Various charts indicate a region round the Cocos 

 Islands, between lat. 10° and 20° S. and between 

 long. 90° and 105° E. as being more or less free from 

 cyclones, but according to the author the existence of 

 this region is doubtful, and requires further investiga- 

 tion. The publication is chiefly of nautical interest. 



The annual report of the director of the Department 

 of Terrestrial Magnetism of the Carnegie Institution 

 of Washington for 19 15 shows the large amount of 

 work which has been done by the department both on 

 land and at sea. We have referred as occasion has 

 offered to the magnetic survey work at sea carried 

 out on board the Carnegie, which last year devotf 

 herself to the Pacific. The present report announct 

 that the following land surveys -have been con^ 

 pleted : — Through Central Brazif from Rio de Janeir 

 to Para; interior of southern China and Mongolia 

 general magnetic survey of Australia; Australasia 

 and West Pacific Islands; the Belgian Congo an 

 Angola and the south-west coast of Africa. At Wash 

 ington itself the new buildings have been brough 

 into use, and a considerable amount of reconstruction 

 and improvement of instruments has been accom 

 plished, especially in relation to measurements o 

 atmospheric electricity. Abstracts of thirteen paper 

 which have been published by the staff during th 

 year are appended to the report, which contains^ ■ 

 record of work for the good of the world_ on whici 

 the Carnegie Institution may justly pride itself. 



A PAPER on the Rangoon River Training Works, b 

 Sir George Cunningham Buchanan, read before th 

 Institution of Civil Engineers on March 21, contain 

 some interesting details of an engineering undei 

 taking of considerable importance. The Port^ f- 

 Rangoon, to-day the third port of the Indian Empire 

 is situated on the left bank of the river of the san- 

 name, which constitutes one of the deltaic mouths < 

 the Irrawaddv, at a distance of about 28 miles frO!' 

 the sea. At this point, which also marks its junctic! 

 with two other effluents, the stream assumes^ a veii 

 sinuous course, and, swinging round a bend in froi 



