318 
curve is to be noted. At 600 mb, within one or two 
kilometres of the leading edge of a sheet of frontal 
altocumulus, the air was supersaturated with respect 
to ice, the observed supersaturation being confirmed 
by the formation of persistent condensation trails by 
the aircraft. At the same level, but 10 or 20 km from 
the cloud sheet, the air was dry with a frost-point de- 
pression of about 22C and a relative humidity of about 
13 per cent, substantially on the smooth curve of the 
environment. The supersaturation with respect to ice 
at 450 mb is also to be noted. 
Very Low Relative Humidities. Air of relative humid- 
ity as low as 1 per cent is occasionally found over south- 
ern England in ordinary subsiding polar maritime air, 
but the lowest relative humidity we have observed, 
0.65 per cent, was at 800 mb on January 17, 1946, in 
air which was an outflow from a strong continental 
winter anticyclone (Figs. 9a and 9b). On that day the 
general structure of the overrunning moist air was 
shown by all the radiosonde ascents in southern Eng- 
land, though the details vary. The moist air is of Med- 
iterranean origin. 
5 2 es 
YZ = 
S80 205, 1 2, Y 35 
x ala 
o—o AIR TEMPERATURE SZ / 7 I) 4 
400},0——o-==-© FROST-POINT / oN / \e0l, 
“80 VSN Z JS 
Y 
: / ; X a { ! 
600 \ 
VS SSNS 
= = 7S x Tr 
700 Pes 7 1 /¥ 
0 / LIN le 
PK 6 Z 5 SY 
800 |= ) S ji 
= : P— 3/10 Gu., MAINLY TONE 4 | 
3009 7 Ne-- NJ 4 J NeazE TOP <j + 
D us X= 
1000 eo ef Ae SH S SY oto 
MB TEPHIGRAM I7-JANUARY-46 1400 GMT FT KM 
Fig. 9a.—Very dry air in a continental winter anticyclone, 
1400 GMT, January 17, 1946. Ascent made over southern 
England at approximately 51°N 2°W. 
WEST 20° to° 
Fie. 9b —Sea-level synoptic chart for approximately 0600 GMT, 
January 17, 1946. 
THE UPPER ATMOSPHERE 
The dry air could have originated from saturated air 
by adiabatic compression from a level of 350 mb at 
212K. If water vapour had diffused into the layer then 
its origi would be higher and colder. It is possible that 
the air had originated from within the stratosphere and 
that the additional water vapour required, 18 mg m=, 
had diffused into it. It would then be necessary for the 
air to have cooled by radiation to its present potential 
temperature. 
Small-Scale Temperature and Humidity Patterns in 
Free Avr. Frith [7], using a frost-pomt hygrometer, has 
made careful investigations of the detailed temperature 
and humidity structure in the free air both in a hori- 
zontal and in a vertical plane. In a horizontal plane he 
finds evidence for the existence of closed temperature 
and humidity patterns on a scale of tens of miles and 
some similarity between the temperature and humidity 
isopleths; low frost pomts tend to be associated with 
high air temperature and vice versa. While the temper- 
ature variations are normally only of the order of one 
degree centigrade, the frost-point variations may be up 
to 15C. 
Similarly, in the vertical plane, when strong strati- 
= 
55 7 S37 
a heows| ENG 
O-=-0 FROST-POINT 
400 0a 
AO xt | TEMPERATURE 
Mo, ste FROST-POINT } pescent 
0. 
MB TEPHIGRAM 30-JULY-47 1400 GMT 
Fie. 10a.—Variation of observed conditions with position. 
Observations during ascent and descent showing changes, 
1400 GMT, July 30, 1947. Ascent over southern England at 
approximately 51°N 1°W. (By R. Frith.) 
Fie. 10b.—Sea-level synoptic chart for approximately 0600 
GMT, July 30, 1947. 
