750 
Journal of Agricultural Research 
Vol. XXI, No. 10 
would be stronger than that originating elsewhere. If the attraction 
exerted is of more limited range, so that the proximity of farm buildings 
and small towns would exert the greater influence, we would probably 
find the greater number of specimens to the north, east, and south, 
which agrees fairly well with the facts. 
Tabus VII .—Climatological data relating to third dispersion test at Dallas, Tex. 
Date. 
Temperature. 
Wind. 
Rain. 
Humidity. 
Ac¬ 
tual 
sun. 
Max. 
Min. 
Mean. 
Direction during successive 
hours from 6 a. m. to 
8 p. m. 
Velocity 
during 
day. 
Max. 
Min. 
Mean. 
1918. 
°F. 
°F. 
°F. 
Miles. 
Inch. 
P.d. 
P. cl. 
P. it. 
P.cU 
Sept. 18... 
91 
66 
78.5 
2 E., 2 SB., 8 S., 2 SE. 
5 to io 
0 
78 
36 
57 
80 
19... 
79 
62 
70. 5 
i S., s NE., 6 E., 2 NW. 
5 to 21 
0.57 
65 
56 
60 
34 
20... 
68 
55 
61. 5 
14 N. 
7 to 13 
0 
72 
55 
63 
70 
21... 
70 
47 
58.5 
2 N..6E., 2 NE..4E. 
i to 9 
0 
83 
34 
58 
100 
22... 
76 
5 i 
63- 5 
2 E., ii SE., i S. 
6 to 9 
0 
67 
33 
50 
99 
23... 
80 
53 
66. s 
2 E., 12 SE. 
4 to 12 
0 
61 
25 
43 
99 
24... 
85 
55 
70. 0 
6 SE., i S., 4 SE., 3 S. 
7 to 17 
0 
68 
45 
56 
93 
25... 
87 
60 
73 - 5 
5 SE., 6 vS., 3 SE. 
IO to 20 
0 
85 
52 
68 
78 
26... 
72 
63 
67-5 
3 E.,3 N., i NE..7NW. 
4 to ii 
.86 
83 
69 
76 
25 
27... 
72 
55 
63-5 
4 N., i NW., o N. 
4 to 14 
0 
78 
42 
60 
75 
28... 
75 
48 
61.5 
12 NW., i W., i NW. 
4 to 9 
0 
87 
35 
61 
100 
29... 
85 
5 i 
68.0 
6 W .,3 NW., 2 NE .,3 E.... 
2 to 7 
0 
73 
29 
51 
99 
30... 
88 
60 
74.0 
SW., iSW., 4 S., 4 E.. 
3 to 7 
0 
61 
23 
42 
100 
Oct. 1... 
89 
62 
75-5 
4 E., 7 S., 3 SE. 
1 to 10 
0 
62 
27 
44 
93 
2... 
87 
61 
74.0 
3SE..4W., sSE., 2E. 
2 to 7 
0 
66 
33 
49 
93 
3... 
89 
63 
76.0 
3 E., I S., 3 SE., 7 s. 
2 to 7 
0 
61 
29 
45 
98 
4... 
89 
66 
77-5 
2 E., 12 SE. 
5 to 9 
0 
58 
32 
45 
100 
5 ... 
89 
67 
78. 0 
3 S., 2 SW., 7S., 2 SE. 
3 to 13 
0 
70 
38 
54 
62 
As will be seen by referring to Table VII, the period during which this 
test was conducted was extremely unfavorable for determining the 
possible influence of wind on the dissemination of flies. During the 
entire period of recovery there was no considerable number of consecutive 
days in which the wind was constant. In other words, the wind condi¬ 
tion would be expressed as “choppy,” and with its continual varying 
from one quarter to another it is almost impossible for one to weigh the 
wind influence on dissemination in this test. For 7 hours following the 
liberation of the first lot of flies there was blowing a 5- to 10-mile breeze 
from the south. It is conceivable that the flies when liberated drifted 
rapidly northward with the wind and that many were subsequently 
carried back and in other directions with the wind changes. The assump¬ 
tion that they traveled with the wind is hardly borne out by the recovery 
of several flies in the south traps on the next day (Table VIII) and by the 
recovery of four flies in the east traps on September 22, where they must 
have gone at right angles to or against the wind. Moreover, there is 
strong evidence, as indicated by the collections of marked house flies in 
the four traps to the west of the point of liberation, that there is no pro¬ 
nounced tendency for them to travel with the wind, or at least that 
tendency is easily overcome by other influencing factors. During the 
first 5 days of the test not a single marked house fly was recovered in 
the west traps, and during this period the prevailing wind during the hours 
