On some Meteorological Conditions controlling Nocturnal Badiation. 393 
Excepting that this sequence is somewhat more irregular, perhaps on 
account of the small number of observations dealt with, it shows no such 
striking contrast to Table 4 as might have been anticipated. Of the 
27 nights of incipient dew, twelve showed a diminution of the gradient 
from one hour to the next when dew had begun to form in the interval ; 
ten showed an increase, while the remaining five showed no change at 
all. Both the diminution and the increase of the gradient averaged about 
0-5 degree F. 
But a partial if not complete explanation of this curious result is 
furnished by a comparison between the radiation-temperature gradients 
and the corresponding velocities of the wind. In the respective hours 
during which dew began to form on 26 evenings (the anemometer was out 
of action on one night of the 27), ten showed an increase of wind velocity 
as compared with the preceding hour, 15 showed a decrease, and in one 
case there was no change. The important point, however, is that the 
general tendency on these nights of incipient dew is for the wind velocity 
to decrease, i.e. : — 
XIX. XX. XXI. XXII. XXIII. Mid't. 
3-3 3-1 3 0 2-6 2-7 miles per hour. 
Whereas, as we saw from Table 4, the tendency on clear nights generally 
is for the velocity to increase. Since, then, the velocity decreases instead 
of increasing, the radiation-temperature gradient should increase on this 
account (see Table 9). As it does not increase, but on the contrary 
decreases, it follows that the formation of dew does cause, or rather 
accompanies, a decrease of the gradient. Of course the slackening of the 
wind is responsible in many cases for the circumstance that any dew 
forms at all. 
Table 10 gives details of the observations under this head. The 
respective quantities in heavier type show : — 
1. The value of the temperature difference between the two ther- 
mometers at the hour when dew was first seen ; 
2. The velocity of the wind in the hour during which condensation 
began on the lower thermometer. 
