June 17, 1920] 



NATURE 



485 



with — on the outskirts of London in places where each 

 could be furnished with at least one hundred acres 

 of land at a total cost of no more than is asked for 

 the 11^ acres now proffered. 



The advantaj^es of such outlying centres would be : 

 (i) The students could live .in the neighbourhood of 

 the institution, either at home or in hostels, and would 

 not be compelled to take a long journey twice a day. 

 (2) There would be abundance of room for all pur- 

 poses, including recreation. (3) Each subject would 

 be able to have its own area of ground for the erec- 

 tion of a suitable institute, and for permitting 

 future extension. (4) Temporary buildings might be 

 put up until experience has shown what character of 

 permanent buildings ought to be erected. (5) The 

 classes would not be of an unwieldy size ; for we might 

 assume for each of the four centres two hundred and 

 fifty students, instead of the one thousand assumed 

 for the central institute, i.e. for each practical subject. 



American experience has shown tnat one hundred 

 acres is not too much land to provide for the buildings 

 of a modern university, and a scheme which assumes 

 that a single universitv for a city the size of London 

 can be accommodated in a space of ten, or even 

 twenty, acres is self-condemned at the outset. 



E. Sharpey Schafer. 



University New Buildings, Edinburgh, June 7. 



High Rates of Ascent of Pilot-Balloons. 



Abnormal rates of ascent shown by pilot-balloons 

 have in recent years aroused considerable interest 

 amongst aerologists. These digressions were mostly 

 ascribed to the occurrence of vertical air-currents, but 

 three years ago Wenger (Annalen der Hydrographie, 

 1917, Hamburg) suggested' that, for the greater part, 

 the observations did not permit of such an explana- 

 tion. He advanced the theory that these abnormal 

 rates were chiefly caused by increased turbulence of 

 the air, and he stated, making use of the Linden- 

 berg material, that the rates observed depended on 

 different conditions of the air, as wind, friction, etc., 

 which, no doubt, must have a notable influence on its 

 turbulence. 



In the same journal, however, it was shown that 

 a large number of observations made at Sofia (Bul- 

 garia) gave strong evidence that actually large ver- 

 tical air-currents had occurred, and that the high and 

 low rate should, without doubt, be ascribed to 

 upward and downward movements of the air. 



Between 19 12 and 19 17 much material regarding 

 these rates was gathered by the Batavia Ob- 

 servatory, and peculiar circumstances make this 

 material of critical value, for it consists of three series 

 taken in three localities differing in character, the 

 ascents being made at various hours during the day 

 and night. 



The first series was taken at Batavia during the dry, 

 sunny season, when land- and sea-breezes are 

 developed strongly; and the second at Bandung, a 

 town situated on a plateau 700 m. above sea-level, 

 surrounded by mountains. Insolation in the latter 

 case did not differ much from that at Batavia; the 

 mountain- and valley-breezes were only slight. 



The third series was taken by Dr. Boerema on a 

 small coral islet in the Java Sea at the end of the 

 west monsoon. There the influence of insolation and 

 of land- and sea-breezes is practically nothing. 



Most of the balloons were observed from two points ; 

 also, up to I or i^ km., half-minute readings were 

 taken between the usual observations made every 

 minute. Balloons of different weight and pattern were 

 used; those for the night carried a second balloon 



1 It came under my notice only a few months ago. 



filled with acetylene and a burner. For each kind 1 

 calculated the average rate observed at the level of 

 3 or 4 km., and derived from it the rate for the layers 

 beneath by applying the formula 



v = ad-i (d = air density). 



For the light balloons there remained, of course, a 

 change of rate with height, which, however, I was 

 unable to calculate, but I surmise was small. 



The accompanying diagram (Fig. i) gives for the 

 lowest layers up to 3 km. the departures of the mean 

 rates observed from those at 3 or 4 km. Evidently 

 it displays the contrast between the land and the sea 

 influence, i.e. on land the rates are increased by day, 

 but are normal by night, while at sea there is no 

 distinct increase in the average rate. 



At Batavia the positive digression runs parallel with 

 the strength of the sea-breeze and with the change of 



wind velocity caused by the Espy-Koppen effect. At 

 Bandung also this parallelism seems to exist. 



Consequently, at first sight, the explanation proposed 

 bv Wenger might be given : During the day tur- 

 bulence is enhanced on land by increase of wind velo- 

 city, by friction with the surface, and by insolation; 

 at sea, on the contrary, these causes are not present. 

 However, on more detailed examination this explana- 

 tion is not confirmed. Thus the rate of ascent did 

 not prove to depend on wind velocity, as the following 

 figures clearly show : 



Batavia. 



9 a. 



Noon-3 p.m. 



3-6 P- 



A Rate of 



ascent 



m. p. min. 



3 



63 



Wind 

 velocity 

 m. p. 'ec. 



3-8 

 40 



A Rate of 



ascent 



m. p. min. 



-4 



-64 



136 



Wind 

 velocity 

 m. p. sec. 



8-7 

 7-S 

 6-4 



^ Rate of 



ascent 



m. p. min. 



4 



56 



133 



Wind 



velocity 



m. p. sec. 



8-0 



5-6 

 6-9 



NO. 2642, VOL. 105] 



