1172 
some connection between the speed of C7 and the speed 
of cyclones (British Isles). 
Cirrus moving at high speed, even if associated with 
a fast-moving storm, do not necessarily mean prompt 
arrival of a storm center. Usually the faster the C7, 
Taste VIII. Cirrus SPEED AND STORM SPEED 
(Adapted from Clayton [23, p. 463]) 
Cite gaastl Storm speed Ratio 
(mph) Ci/storm 
Range (mph) Average (mph) 
34— 56 49 20 2.4 
56- 79 70 25 2.8 
79-101 90 29 3.1 
101-123 106 32 3.4 
146-168 152 34 4.5 
(Adapted from Pick and Bowering [77]) 
Storm speed 
Number of Ci speed 
cases (mph) Range Per cent of Per cent 
(mph) total cases | probability 
<25 85 35 
20 *<26 
<45 100 81 
25-45 52 46 
52 26-50 = 
<45 96 81 
>25 82 65 
74 51-100 
>45 34 19 
>25 75 65 
16 >100 
>45 25 19 
the greater the ratio of their speed to that of the storm, 
and therefore, the farther they will have advanced 
ahead of the storm center. 
Another of Clayton’s tabulations [23] relates Cz mo- 
tions to temperature changes (Table IX). Observations 
Tapite [X. Crrrus Motions anp TEMPERATURE CHANGES, 
Buus HILu 
n 
ate Temguavaime BON nes al auscoeee a gn eae 
Ci moved qhange (per cent) (per cent) (per cent) 
NW Rise in 12 hr 60 64 76 
W Rise in 12 hr 58 69 74 
SW Fall in 12 hr 57 66 63 
NW Rise in 24 hr 43 48 35 
W Rise in 24 hr 34 34 32 
NW Fall in 24 hr 48 46 59 
W Fall in 24 hr 57 61 61 
SW Fall in 24 hr 70 83 75 
of Cz, totaling 410, all made at 8 a.m., were used and 
frequency of rises and falls in the subsequent twelve and 
twenty-four hours noted. Cirrus from the west and 
northwest were usually followed by a rise in twelve 
hours and then a fall. The rise could be partly dis- 
counted as normal diurnal variation, 8 p.m. beg some- 
what warmer than 8 a.m. In other cases, falls pre- 
dominated. 
If indications of this type were strengthened with 
CLOUDS, FOG, AND AIRCRAFT ICING 
more cases and related to other features, they might 
be of more value. 
Brooks and Harwood [18] noted that fast Cz occurred 
“before twenty-four out of thirty-two heavy snowstorms 
(75 per cent) but before only fourteen of twenty-five 
very heavy ones (56 per cent). While a fast-moving 
storm favors precipitation in the form of snow, because 
of the lack of time for advection of warm air, it also 
passes so soon that the precipitation is more likely to 
be moderately heavy than very heavy. 
The relationship of Cz velocities to weather changes” 
is summarized in Table X (after Clayton [23]). 
TABLE X. CoMPARISON OF CIRRUS VELOCITIES AND WEATHER 
CHANGES 
5 . Mean daily Mean daily . 
Ci velocity Average . . Rain 
Civelocit change in change in ariability* 
ims] Stam” |e | PGR | “oeread 
34-56 49 0.11 3. 19 
56-79 70 0.12 4.2 37 
79-101 90 0.18 5.3 39 
101-123 106 0.19 6.4 31 
146-168 152 0.23 7.8 51 
* Rain every other day would be 100% variability. 
Showers. Cloud observations are particularly useful 
in forecasting the occurrence and locations of showers. 
Showers developing from pre-warmfrontal instability at 
middle levels are usually merely punctuations to gen- 
eral upglide precipitation. Their occurrence, however, 
can sometimes be anticipated from Acc or Ac densus 
mammatus, which reveal the presence of strong con- 
vection in middle levels. Longer in advance, C2 floccus 
and Ci densus, and especially Cz nothus, will indicate 
the approach of showery weather. If these are seen 
in the front zone of an open-wave cyclone, it is fair to 
surmise that they are related to pre-warmfrontal con- 
vection rather than simply to cold-frontal overturnings. 
Cloud Motions and Shapes as Indicators of Wind, 
Wind Shear, Vertical Currents, and Turbulence. The 
progressive motion of a cloud, turbulence on its periph- 
ery, and deviations of tall clouds from the vertical 
indicate actual and relative winds aloft. When cumuli- 
form clouds lean or Cz trails depart from the vertical, 
it is obvious that the wind at one height in the cloud is 
not the same as that at another height. In a rapidly 
growing cloud, the amount of deviation from the verti- 
cal is the resultant of the rate of fall of the cloud par- 
ticles and the rate of change of wind velocity with 
height, both of which may vary. Mrs. Malkus [63, 64] 
has recently described the process for Cu, and Ludlam 
[62], for Cz. Ludlam has presented evidence of a more 
rapid fall of crystals in Cz when they grow in passing 
through moist layers. When crystals begin to melt they 
will fall faster. In other words, the up-and-down shape of 
a cloud represents simple or complex streamlines of 
the rising or falling air columns the motion of which is 
rendered visible by the cloud material [13; 62, pp. 48— 
49]. Strong shear, adverse to thunderstorm develop- 
ment, may be recognized in advance on days otherwise 
favorable for thunderstorms and used accordingly in 
forecasting, as Brooks pointed out in 1922 [15, p. 283], 
