172 



NATURE 



[February 9, 1922 



Letters to the Editor. 



[The Editor does not hold himself responsible for opinions 

 expressed by his correspondents. Neither can he undertake 

 to return, or to correspond with the writers of, rejected 

 manuscripts intended for this or any other part of Nature. 

 No notice is taken of anonymous communications.} 



The Antitrades. 



The long series of pilot-balloon ascents made at 

 and near Batavia (lat. 6° ii' S., long. io6° 50' E.) 

 during the years 1909-17 has given a fair knowledge 

 of the system of air-currents over West Java up 

 to great heights. The general outcome of this inves- 

 tigation has been communicated to the Royal Acaderny 

 of Science of Amsterdam.' My endeavour to explain 

 that system led to a controversy between Dr. Braak 

 and .myself and Prof, van Everdingen." After re- 

 newed consideration of the problem I have come to 

 new results which I propose to set out provisionally 

 here. 



In the memoir presented to the Amsterdam 

 Academy is inserted a synoptical table containing the 

 mean directions and velocities of the wind for each 

 month and for height-intervals of i km. up to a height 

 of 24 km. In it the principal air-currents have been 

 made conspicuous by letter colouring and framing. 

 They are : — 



First, the west monsoon prevailing during the 

 southern summer in the bottom layers up to 5-6 km. 



Above it, up to 10-13 km., blow easterly winds 

 with southern components, which I would call trade- 

 like winds. In the winter season such winds blow 

 in the bottom layers up to 3 km. 



Above these tradelike winds blow antitradelike 

 winds, i.e. easterly winds with a northern component. 

 Their upper limit reaches to 18 km. from December 

 until March ; it goes down to 12 km. in June, and 

 again rises to the maximum height of 21 km. in 

 October. The velocities show two maxima : in 

 February at a height of 15 km, (12 m./sec.) and in 

 August at 14 km. (22 m./sec): in April they are very 

 weak. Not only is their velocity a maximum, but 

 also the transport of air-mass. 



Over the antitradelike current appear again currents 

 of tradelike character ; however, from March until 

 September an eastward moving air-mass is embedded 

 in them, reaching heights of 24 km. in ,'maximo.' 

 Very high balloon flights in March and September 

 revealed the existence of strong (30-40 m./sec.) 

 easterly winds up to 30 km. 



Considering these results, three principal questions 

 arise : — (i) Are the tradelike winds real trades? 

 (2) Is the antitradelike current a true deflux from the 

 equator towards the sub-tropics? (3) Whence do the 

 great velocities of the high antitradelike and upper 

 tradelike w^Inds originate? 



The currents mostly possess a stationary character, 

 and consequently their directions will be In close 

 agreement with the trend of the isobars In their level. 

 For Java the latter will be conditioned by the neigh- 

 bourhood of the Australian continent. As in the 

 southern winter over Australia Is settled a circular 

 High, we mav expect over Java the trend of the 

 Isobars to be E.N.E.-W.S.W. and the gradient to be 

 towards the equator. However, by friction with the 

 earth surface the air blows across the isobars and 

 takes an E.S.E.-W.N.W. direction. This means real 



1 Proceedines, April i6, iqt8. 



S Tijdschrift v. h. K. Aardrijkskundig Gen., vo'. 35, igiS, No. i, and 

 vol. 36, igig, No. 4- 



S Owing to a typographical error in the synoptical table the velocities at 

 the levels 18, 19, and 20 km. for June have wrongly been given as 

 I m./sec. instead of lom./sec. 



NO. 2728, VOL. 109] 



outflow to the equator ; thus the tradelike wind men- 

 tioned above is a trade. 



In the southern summer over Australia lies a Low, 

 causing the west monsoon, but above this Low the 

 gradient is reversed and a High prevails. This causes 

 in the same manner as mentioned above a tradelike 

 wind. The friction required for it, I presume,, is 

 caused by the streaming one over another of the two 

 currents with contrary directions (the west-east 

 below, the east-west above). Thus, I think, the 

 first of the three questions put forward has been 

 answered in the affirmative : the tradelike winds are 

 trades. 



As to the second question, we may consider first the 

 southern winter season. In it the gradient Australia- 

 Java is reversed at the level of ±5 km. But 

 does it change too in the other season at 3 km.? 

 Apparently not, because, going upwards, the easterly 

 winds do" not then change to westerly ones; they 

 back only from E.S.E. to E.N.E., while the velocity 

 does not vanish. Now, admitting the absence of 

 friction in these layers, and consequently assuming 

 the current to follow the course of the Isobars, we 

 come to the conclusion that this course remains 

 mainly the same when going upwards, or the Aus- 

 tralian High subsists in these higher layers, though 

 perhaps shifting somewhat to the eastward. 



Accepting this, we may ask: Might it be that the 

 antitradelike current flows around the Australian High, 

 bringing about thus the deflux towards the sub- 

 tropics ? 



In that case the antitradelike current should be a 

 true antitrade, although of local character. But then 

 we are obliged to admit that a flux towards the 

 equator will also occur at the opposite side of the oval 

 system of the Australian High ; only the deflux 

 should surpass it by the mass of air (or part of It) 

 which ascends from the surface in the equatorial belt. 

 This influx, too, may give us an answer to our 

 third question : What Is the cause of the great east- 

 west velocities of antitradelike and upper tradelike 

 winds ? Exner * points to the fact that ascension of 

 air at the equator Is able to Increase its east-west 

 velocity only by a fraction, and, therefore, tries to 

 explain the great velocities of high equatorial east 

 winds by shifting of air from higher latitudes towards 

 the equator with preservation of rotational moment. 

 A meridional shift from latitude + 15° causes velocities 

 from 30-40 m./sec. 



My result for the antitradelike current over Java 

 Is the same as that obtained by Sir Napier Shaw 

 when calculating isobars for the level at 8000 metres.* 

 He, too, finds long-stretched Highs, and he speaks 

 of the flowing of air around these Highs, by which 

 the east-west wind velocities of the equator act on 

 the opposite currents of the sub-tropics as by chain- 

 drive pulling. 



However, through lack of data Sir Napier Shaw 

 had to calculate his isobars by means of one and the 

 same set of vertical temperature-gradients for the 

 whole hemisphere, which, of course, makes the results 

 somewhat doubtful for the equatorial belt, because 

 there the critical pressure-differences at the 8000- 

 metre level are small only. 



For that reason I have sought for another inde- 

 pendent way to solve the antitrade problem, and T 

 think I have found It by mapping the average direc- 

 tions of cirrus drift as observed in the equatorial 

 belt. 



Cirrus floats there at levels of about 11 km., and 



4 "Dynamische Meteoro'ogie." 1917, p. 182. 



•'■' Rede Lecture. Natupe, Inly 21, 1921. p. 653. Sir Napier Shaw mo<t 

 kindly provided me recently with a copy of the unpublished isobaric charts 

 which he constructed for the northern hemisphere 



