76 



NATURE 



[May 23, 1895 



Thus spoke Wemer Siemens, a man who, by his long 

 and eventful life, was specially qualified to speak with 

 authority on this subject, and the results which have, 

 during the few years of its existence, already been 

 achieved at Charlottenburg are proving him a true 

 prophet. 



In conclusion, we may say that these volumes will be 

 found most interesting, not only on account of the insight 

 they give regarding the development of the electrical 

 industry, but also on account of the interesting personality 

 which pervades the whole. W. Watson. 



ATMOSPHERIC PRESSURE OF THE NORTH 



ATLANTIC OCEAN. 

 Repartition dc la Prcssion Atmospht'riquc stir lOcdan 

 Atlantiquf Septentrional, tfapres les Obsenuitions de 

 1870 a 1889, avec la Direction Moycnne du Vent sur 

 les Littoraux. Par le Capitainc G. Rung. (Copen- 

 hagen : 1894.) 

 THIS -Atlas, showing the monthly and annual atmo- 

 spheric pressure and prevailing winds over the 

 North .-Xtlantic and connected seas, is a fine example 

 of cartography and typography. The monographs for 

 this and the other oceans have generally dealt only with 

 February, May, August, and November; but this work 

 presents us with the results for each of the twelve months, 

 and for the year, on a mean of the twenty years from 

 1870 to 1889. 



The really heavy part of the work carried out by 

 Captain Runy has been the calculation of the monthly 

 means from the nine years' daily weather charts of the 

 Danish and (German meteorologists from December 1880 

 to November 1889, including the similar charts of the 

 Meteorological Council for the year ending August 1883. 

 This has been done for eighty points over the ocean 

 between lat. 10' and 77' 30' N. and between long. 25° E. 

 and 80- W. 



It being desirable that the discussion should cover a 

 longer period than nine years, the twenty years ending 

 with 1889 were juloptcd, these years being selected with 

 the view of utilising the fifteen years' means (1870-84) 

 for this part of the globe which have been published 

 in Buchan's " Challenger Report on Atmospheric Circula- 

 tion," thus greatly facilitating the inquiry. The means 

 for the subsequent five years were independently worked 

 out, and thereafter combined with liuchan's to make 

 up the twenty years' means. The next step was to 

 bring, by the usual method of differentiation, the nine 

 years' means of the ocean stations to approximate means 

 for the twenty years. Table iv. gives the means thus 

 calculated for ninety-two coast or land stations surrounding 

 the ocean, and Table v. for the eighty ocean stations. 

 The mean directions of the wind have been calculated 

 for the stations in Denmark and its colonies ; but for all 

 other stations the tlat.i have been taken siinptiiiler from 

 the ^^ Challenger Report." It might materially have aided 

 the inquiry in the north-western part of the ocean if 

 means for pressure and wind direction had been calcu- 

 lated and given for the I^ibrador stations at Moffenthal, 

 Zoar, Nain, Okak, Hebron, and Rama, the observations 

 at which have been published from 1882 to 1889. 

 NO. 1334, VOL. 52] 



The monthly and annual means for the eighty ocean 

 stations, and the charting of the results on the thirteen 

 maps, constitute the novel part of Captain Rung's work, 

 and must be regarded as a substantial addition to our 

 knowledge of the meteorology of the North .\tlantic. 

 This remark holds good emphatically as regards tlic 

 northern half of this ocean, and for the five months from 

 May to September. Thus, for these months, we have 

 now a more accurate knowleilgc of the distribution of 

 atmospheric pressure and of the prevailing winds north 

 of latitude 60' than could have been obtained from any 

 work previously published on the subject. 



But such well-merited praise cannot be extended to the 

 working out of the results for the five winter months from 

 November to March. .\n examination of the Danish 

 and German daily weather-maps of the .Atlantic of the 

 nine years for these months shows that over the whole 

 ocean to the north of a line drawn from St. John's, New- 

 foundland, to V'alentia, observations from a ship at sea 

 is an event of extrcTnely rare occurrence. The con- 

 sequence is that the monthly means for this important 

 region, from which fresh information is so desir- 

 able, ha\c been obtained wholly from the observations 

 made at the land stations of this part of the ocean. 

 Hence the results given in the .Atlas cannot be regarded as 

 a contribution to the meteorology of the ocean. In this 

 .Atlas, what strikes one at first sight as new fact is the 

 distribution of atmospheric pressure during the winter 

 months from the south-west of Greenland round by Ice- 

 land to north of Norway, particularly the three or four 

 distinct areas of pressure a little lower than prevails 

 generally over this region. But a close examination of 

 the daily weather-maps themselves suggests the idea th,;t 

 these three or four low-pressure systems may be no more 

 than the outcome of an interpretation, made in construct- 

 ing these daily maps, of the amount of pressure over the 

 ocean drawn from the pressure and winds observed at 

 the land stations, the interpretation being made in the 

 complete absence of observations at sea. Thus the 

 observations made at the Greenland stations since 1840 

 amply show that the winds on its coast are very greatly 

 deflected from their true direction, as that would be deter- 

 mined by the distribution of pressure, by the high ground 

 and valleys near the coast. It is in this connection tliat 

 a discussion of the Labrador obserxalions would have 

 come in so handy. 



Captain Rung has raised a side issue to his report in a 

 discussion of the distribution of atmospheric pressure in the 

 interior of .Southern .Scandinavia, where the .Atlas shows 

 a sinjjular local excess of pressure in the winter months, 

 which excess is also plainly shown by his monthly means 

 of the Norwegian, Swedish, and Danish stations. In 

 looking closely at this matter, it is necessary to leave out 

 of view the means for Dovrc, Tonset, and Roros, which 

 approach to, or exceed, 2000 feet above the sea, their 

 positions not being suitable in discussing small sea-lcvcl 

 differences of pressure such as are here dealt with. We 

 have calculated afresh the January means for all other 

 stations not exceeding 620 feet in height, for the same 

 twenty years, and obtain a set of figures differing widely 

 from those published in the Atl.as, whii.h give no coun- 

 tenance to the idea of a local excess of pressure in 

 winter over this region. To test the matter in another 



