99 
1913-14.] Principia Atmospherica. 
and again assuming a value of 0/p, we can compute Ad provided we have a 
value of 6 which can properly be substituted in the equation. 
Here again we must have recourse to the mean value, as we have no 
observation of actual temperature at the time ; but, again, the error made is 
not fatal to the practical success of the calculation, because 0 comes in as a 
factor which affects the scale of the variation ; it does not affect the sign. 
By taking the mean value for the month instead of the actual value the 
error is probably less than 10 per cent., and the whole error of employing 
mean values for actual values probably amounts to less than 20 per cent. ; 
and in considering the distribution of pressure and temperature in the 
upper air we are not yet in a position to reject observations and informa- 
tion which may be in error by as much as a fifth. 
Consequently we may properly use the calculation here indicated to 
give at least a rough but working idea of the distribution of pressure and 
temperature at successive levels in the atmosphere when we know the 
velocity and direction of the wind there. 
The errors in p/0 and 0 are less important in considering the nature of 
the distribution, because the same values, right or wrong, are used for both 
components at the same level. 
The following table of monthly averages gives values which may be 
used in the absence of any special information for the particular occasion : — 
Table III. — Average Temperature at Different Levels for Months. 
1. For British Isles. Taken from “Geophysical Memoirs,” No. 2 (W. H. Dines). 
Height, 
kilo- 
metres. 
Jan. 
Feb. 
Mar. 
April. 
May. 
June. 
July. 
Aug. 
Sept. 
Oct. 
Nov. 
Dec. 
14 
216° 
217° 
219° 
221° 
222° 
223° 
222° 
221° 
219° 
217° 
216° 
215° 
13 
216 
217 
219 
221 
222 
223 
223 
221 
219 
218 
217 
216 
12 
217 
218 
219 
220 
221 
222 
222 
221 
221 
219 
218 
217 
11 
217 
217 
217 
219 
220 
221 
222 
222 
221 
220 
219 
218 
10 
220 
220 
220 
222 
224 
225 
226 
226 
226 
224 
223 
221 
9 
224 
223 
224 
226 
229 
231 
234 
233 
233 
231 
228 
225 
8 
230 
229 
230 
232 
236 
238 
241 
241 
241 
238 
235 
332 
7 
237 
236 
237 
239 
242 
245 
247 
248 
247 
245 
241 
238 
6 
243 
243 
244 
246 
249 
252 
255 
255 
254 
251 
249 
245 
5 
250 
249 
250 
252 
256 
259 
261 
262 
261 
258 
255 
252 
4 
257 
256 
257 
259 
262 
265 
267 
268 
267 
264 
261 
258 
3 
263 
262 
263 
265 
268 
271 
273 
274 
273 
270 
267 
264 
2 
267 
266 
267 
270 
273 
276 
278 
279 
278. 
275 
272 
269 
1 
271 
271 
273 
276 
279 
282 
283 
283 
281 
279 
275 
272 
Gd. 
276 
276 
277 
282 
285 
288 
289 
289 
286 
283 
280 1 
277 
I give in Table IY. a specimen of the calculation as applied to the 
results of a sounding with a pilot balloon on April 29, 1908. 
