September 17, 1908] 



NA TURE 



487 



METEOROLOGY OF THE INDIAN OCEAN.' 

 "T^HE present volumes form a noteworthy contribution to 

 •*■ tlie systematic knowledge of the meteorology of the 

 sea on the lines laid down by the International Congress 

 held in London in 1S74. According to that plan the surface 

 of the earth is divided by parallels of latitude and 

 meridians, each 10° from the next, into sections numbered 

 consecutively, starting from the square 0-10° N. and 

 0-10° W. of Greenwich, and proceeding west as indicated 

 in the accompanying figure. 



The observations from the meteorological log-books are 

 then sorted according to the squares in which they are 

 taken, and after a sufficient number has been obtained a 

 definite scientific record of the meteorology of the region 

 is made possible. If we look at it from another point of 

 view, we might say that a knowledge of the " climate " 

 of the different parts of the ocean may be obtained by this 

 method. 



The publications give the result of fifty years' observa- 

 tions, and some idea of the magnitude of the labour in- 

 volved in reducing them to a form in which they are avail- 

 able for the sailor and the meteorologist may be gained 

 from the fact that for each month about half a million 

 observations had to be treated. 



The tables give the values of the various elements for a 

 large number of single degree squares in the following 

 groups of 10° squares : — Nos. 63-60, 24-32, 323-332, 350- 

 368, 396-406, 430-442, 466-478. These squares cover the 

 whole of the Indian Ocean from .Africa to Australia down 

 to 50"' S. lat. 



The charts are not merely diagrammatic representations 

 of the information contained in the tables. The guiding 



Jo' 



Jo" 



/o' 



Icy (V 



Jo'S 



principle in their construction has been to exhibit such 

 information as a sailor will find useful, without overloading 

 the map with unnecessary details or results of doubtful 

 utility. On this account, also, the charts of cloud dis- 

 tribution, of rain and of fog, have been omitted from the 

 present issue. The charts show the ocean current, the 

 wind, the general circulation of sea and air, the mean 

 pressure distribution, and the isotherms for the surface of 

 the sea and for the atmosphere for each of the months 

 September, October, and November. There are, in addi- 

 tion, three charts giving for each month the best routes 

 for steam and sailing ships, at the same time showing 

 the trajectories of cyclones and the northern limits of fog 

 and floating ice. 



An interesting result immediately seen on comparison 

 of the wind and current charts is the fact that in regions 

 where the wind is fairly steady in direction the ocean 

 current flows to the left of the direction of the prevailing 

 wind. Thus in the region of the trade winds, which blow 

 from the south-east during these months, the ocean current 

 flows steadily due west at an angle of 45° with the wind, 

 while further south, where the prevailing wind is VV. to 

 W.N.W., the prevailing current comes from W. to 

 W.S.W., although the regularity of the deviation is not 

 so marked. This may be compared with Nansen's observa- 

 tions during the drifting of the Frain in northern latitudes. 



1 Koninklijk Nederlandsch Meteorologisch Instituut, No. 104. Oceano- 

 eraphische en Meteorologische Waarnemingen in den Indischen Oceaan, 

 September, October, Novetntier. 1856-1904. Tabellen en Kaarten. (Utrecht: 

 Kemink & Zoon, igo8.) Prijs (met atlas) 5.50 fl. 



NO. 2029, VOL. 78] 



He found the direction of drift deviated 20°-40° to the 

 right of the wind direction. The theory has been developed 

 by Ekman {Arkiv for Matemalik, vol. ii.. No. 11), who 

 found that in the open ocean, while the depth of the wind- 

 produced current varied with the latitude, the deviation, 

 an effect of the earth's rotation, was independent of the 

 latitude and equal to 45°, except in the immediate neigh- 

 bourhood of the equator. The current caused by the trade 

 winds, on reaching the African coast, is deflected towards 

 the south, and produces a marked effect on the tempera- 

 ture of the sea off the coast of Natal, where it is fully 

 5° C. warmer than in the same latitude in the open ocean 

 or on the Australian coast. 



.•\ point especially worthy of commendation is the inser- 

 tion in the tables of the number of observations on which 

 each result is based. In the charts, too, the number of 

 observations utilised for each 5° square is given. It is 

 thus easy to see at a glance if an apparently anomalous 

 result is doubtful owing to insufficiency of observations or 

 really represents a feature worthy of investigation. In the 

 pressure charts not only are the mean isobars drawn, but 

 the mean pressure for each 2° square is inserted, one of 

 the principal reasons for this being that small departures 

 from the normal value in the barometric height are fre- 

 quently in tropical regions a valuable indication of 

 approaching cyclonic disturbances. 



.An important feature in the tables is the representation 

 of the stability of the ocean currents and the wind. The 

 value of this is taken to be the ratio of the resultant 

 velocitv to the mean velocity taken without regard to 

 direction. It furnishes the sailor with an estimate of the 

 probabilitv of his meeting with the current or wind 

 indicated by the resultant. It is an attempt at scientific 

 statistical forecasting which will probably be further 

 developed. 



The number of observations of rain is given as a per- 

 centage of the total number of observations for each 

 square, but no information is given regarding the amount 

 of rain. No observations of humiditv are published, an 

 omission to be regretted in view of the importance of water- 

 vapour both climatologically and in the thermodynamics 

 of the atmosphere. E. G. 



THE EVOLUTION OF DECORATIVE DESIGN. 

 'T'HREE explanations of the development of decorative 

 design occupy the field at present : — first, that 

 originally suggested by Prof. F. W. Putnam in connection 

 with the ornamentation found on the pottery of the 

 Chiriqui Indians, and independently advocated in this 

 country by Mr. H. Balfour and Dr. A. C. Haddon, that 

 conventional designs are developed from attempts at 

 realistic representations, which gradually degenerate into 

 a purely conventional representation, in which, at its later 

 stages, the realistic origin can hardly be recognised ; 

 secondly, that of Semper, which emphasises the influence 

 of material upon the development of the design ; thirdly, 

 the theory that the explanations of conventional motives are 

 essentially secondary in character, and are due to the later 

 association of the existing decorative forms with realistic 

 designs. This last, the view propounded by Mr. A. L. 

 Kroeber, Clark VVissler, and Dr. Franz Boaz, has now 

 been re-stated and supplemented by fresh arguments and 

 material by the last of these writers in a paper on the 

 " Decorative Designs of .Maskan Needle-cases," published 

 in vol. xxxiv. of the Proceedings of the United States 

 National Museum. 



This group of objects in their most generalised type dis- 

 plays the following features : — a tube slightly bulging in 

 the middle ; flanges at the upper end ; small knobs under 

 the flanges ; a long concave face at the upper end of the 

 tube ; long parallel lines with small forks at their lower 

 ends setting off the concave face ; border designs consist- 

 ing of lines at the upper and lower ends of the flanges 

 and on the concave face ; and an alternate-spur band at 

 the lower end of the tube. The case itself is formed out 

 of a strip of skin pulled into a tube, which protects the 

 needle against breakage ; and its most peculiar feature is 

 a pair of wings or flanges at the upper end, below which 

 are two small knobs on opposite sides of the tube. 



