[ 253 ] 



VI. The Vertical Temperature ])ixti-il>titi<>n in the Atmosphere over England, 

 and some Remarks on the General and Local Circulation. 



W. H. DINKS, F.K.S. 



Received March 14, Head May 11, 1911. 



DURING the past four years a considerable number of small free balloons carrying self- 

 recording instruments have been sent up in the British Isles, and sufficient observa- 

 tions have now accumulated to give some idea of the conditions which prevail over 

 England, to a height of about 10 miles, in summer and winter, in cyclonic and anti- 

 cyclonic weather. 



Material Available. 



The method of obtaining observations is fully described in a publication of the 

 Meteorological Office, M.O. 202. It wiU suffice here to state that a small self- 

 recording instrument, weighing 1 oz. (35 gr.), is attached by about 30 ft. (9 metres) 

 of strong thread to a small rubber balloon. The balloon is 1 ft. diameter when 

 unstretched. It is filled with hydrogen until it is expanded to about 1 m. diameter, 

 securely tied up, and then let go. Tn"e balloons generally rise until they burst, and 

 carry the instrument on the average to a height of 10 miles (16 km). A label 

 offering a reward of S.s 1 . is attached to the instrument, and the reward is claimed and 

 the instrument returned in two cases out of three. 



The trace when recovered consists of a pressure-temperature diagram, from which 

 the temperature at any height may be obtained within about 1 C. It is compara- 

 tively seldom that a balloon fails to reach 10 km. height, or that the record is 

 undecipherable ; and more than 95 per cent, of the instruments recovered are able 

 to yield a more or less satisfactory record. The heights reached have been very 

 uniform, and mostly lie between 14 and 18 km., but in order that there may be no 

 doubt as to the effect of extrapolation the tables are not carried beyond 14 km. 



In all about 400 balloons have been sent up, and rather over 250 records have been 

 obtained, but in forming the averages given in the following tables all these have not 

 been used. The ascents of the first six months of the series are excluded, partly that 

 the period dealt with may be an exact number of years, viz., the years 1908, 1909, 

 and 1910, and partly because owing to want of experience the earlier results cannot 

 be as reliable as the later. Thirty-two ascents are thus cut out. Of the remainder, 

 nine either failed to reach 10 km. or had a defective trace. But the 200 odd ascents 



VOL. ccxi. A 476. 12.6.11 



