February 9, 1922] 



NATURE 



173 



at that height over Java the antitradelike winds blow 

 from May until October, while during the rest of the 

 y^ax winds with tradelike character prevail. 



The mean directions of cirrus drift which were at 

 my disposal (mostly borrowed from H. Hilderbrand- 

 «on ') I plotted separately for winter and summer, 

 and although they are very sparse I made an en- 

 deavour to construct lines of flow. The result is in- 

 corporated in the accompanying maps (Fig. i). They 

 should be regarded as a first trial ; e.g. no attention 

 <was given to the density of the lines of flow, only 

 to their direction. For three stations (Hawaii, 

 Ascension, and Congo) only annual means were given, 

 and they have been used for both summer and winter. 



Trying to design the lines of flow, it was apparent 

 this could be done only by assuming oblong systems 

 to exist at both sides of the equator, together with 

 a zonal stream winding about the equator. Of these 

 ovals those lying over Central America, Northern 

 Africa, and Southern Asia correspond fairly well with 

 the isobaric Highs found by Sir Napier Shaw at the 

 8-km. level. 



o Winter 



The mean amount of seasonal shifting found above, 

 i.e. 10°, fairly agrees with the corresponding shifting 

 of these high-pressure belts at the surface : — 



Nonhern IJelt 



January ... 32*^ 

 July 39*^ 



Shift ... 7° 



Southern Belt. 



January ... 37° 

 July 28° 



Shift ... Q° 



Resummg, it seems probable that in high levels 

 above the equator and winding about it flows a zonal 

 east-west current of stationary character, which is fed 

 by ascending surface-air and locally by air streaming 

 in from higher latitudes, which, moreover, maintains 

 its east-west motion. Also, that from it flows off in 

 other places air to the sub-tropical belts ; these cur- 

 rents of deflux bend from an east-west to a west-east 

 direction. 



This communication may prove anew that the know- 

 ledge of the direction of the cirrus drift in the 

 equatorial belt is important for the investigation of 

 atmospheric circulation between the tropics, but that 



Fig. I. — Lines of flow of cirrus drift. 



Estimating roughly the latitudes of the centres of 

 the current-ovals I find r-^- 



Latitude of centre 



Mean ... 10 

 The mean latitude of the northern ovals is about 

 ^"°, that of the southern about 15°. At the 

 face of the earth pressure is highest in latitudes 

 N. and 30° S. ; accordingly, when identifying the 

 current-ovals with pressure Highs, the latter are 15° 

 nearer to the equator at the ii-km. level than at sea- 

 ievel. This shifting is in agreement with the con- 

 siderations of Teisserenc de Bort and Exner Q.oc. cit., 

 p. 177), according to which the high-pressure belts 

 •with increasing height move towards the equator. 



* " Les Rases de 'a M^tiorologie dynamique," II. Also Nova Acta R. 

 Soc. Sc. Upsaliensis, ser. 4, vol. 5, No. i. 



the observations at our disposal are few and rather 

 insufficient. For that reason I appeal to those who 

 are in the position to make these observations to 

 supply this need. 



To observe in what direction cirrus floats is easy 

 and requires simple means only ; moreover, observa- 

 tions are not confined to fixed hours or days. Thus 

 they are particularly adapted to be made by amateurs 

 living in the tropical regions. W. van Bemmelen. 



Emmastraat 28, Haarlem, Holland, 

 December. 



Some Problems in Evolution. 



The controversy between Sir Archdall Reid and the 

 ■biologists is partly concerned with the meaning of 

 terms and partly with the understanding or misunder- 

 standing of physiological processes. Sir Archdall 

 insists on certain interpretations or definitions of the 

 terms "inherited" and "acquired." These terms 

 were first used by the biologists, and Sir Archdall 

 should not give them meanings of his own different 

 from those which they originally bore. He insists 

 that there are two kinds of variation, but only one 

 kind of character. The word " variation " has been 

 used to mean the small differences always found 



NO. 273.8, VOL. 109] 



