PRESENT STATE OF OUR KNOWLEDGE OF THE UPPER ATMOSPHERE. 123 
The irregularities probably arise through the tendency of the observers 
to estimate in even ‘ points ’ when the direction is nearly half-way between 
two points. The most frequent deviation for E. and W. winds is then 
1 point (11°), for N. winds nothing and for 8. winds 3 points (34°). In 
addition to showing the smallest deviation, N. winds show greater regu- 
larity than the others, while E. and S. winds show the least. 
Berson’s results lead to the following values for the total rotation ip 
cyclones and anticyclones :— 
5 
Heights 'O5km. 10km. 15km. 2:0km. | 2:5km. 3:0 km. | 4:0 km. 5°0 km. | 
| | 
erry 25 Steed | es 
° ° ° ° ° ‘ons ° ° 
Cyclones. . 4 Q |} -—2 =2 | 2 3 | Sf let 
Anticyclones . _—10 27 | 38 43 | 50 | 57 | 58 | 63 
The horizontal motion of the balloon would tend to increase the 
rotation in anticyclones and to diminish it in cyclones, and this probably 
accounts for the large difference found. Berson concluded that only part 
of the difference could be accounted for in this way, and that there was a 
considerable difference after correcting for this effect. The results of kite 
observations do not agree with this conclusion. The mean rotations, 
deduced from the Berlin observations in 1905, are as follows :— 
. | | Number of Cases 
Heights | 1 km. 2 km. SSG = | 
| 1 km. 2 km. 
a | = 
Anticyclonic . i # 25° 33° 54 26 
Cyclonic . : : | 27° Sith 110 52 
or including only those cases which reached 2 km. :— 
Anticyclonic : . ‘ 4 aoe 33° 
Cyclonic : : ; ‘ : ey orl 37° 
These values agree very nearly with the general mean values found 
above, 7.e., 29°, 32°. Two cases, in which the wind was very weak and 
completely changed its direction, have been excluded. It may be said 
therefore that the mean rotation of the wind in the first 2 km. is prac- 
tically independent of the direction of curvature of the isobars. 
___ The increase in velocity, expressed in terms of the ratio of the observed 
wind to the mean surface wind, under different pressure conditions, 
is, according to Berson, as follows in different layers (A anticyclonic, 
C cyclonic) :— 
0-95! -1-0| -15| —20 | —a5 | -s0 | —4+0| —50| >50 
Heights | km. | km. | km. | km. / km. | km kn. m km. 
— aca / fas : Owe 
Ratio A. . {| H6L| 167 | 166 | 1-86 | 1-96 | 203 | 240 | 3:15 | 4-07 
eno — | — | 3-00 | 4-11 | 1-17 | 1-22 | 1-44 | 1-89 | ota 
Ratio. C 1:96 2-00 2-09 | 217 | 245 | 249 | 266 | 3:57 | 5-03 
eae | | 1-05 | 1-09 | 123 | 1-25 | 1-38 | 1-79 | 2-52 | 
