410 
direction, and none were seen to moye in the 
opposite direction. The majority were less than 
one-half second in crossing the Moon’s disk 
(diameter = 29’.5). Quite a considerable num- 
ber traversed the disk in 0.1 or 0.2 second. 
Few were as slow as 1 second and only one re- 
quired 2 to 8 seconds. In this case the trajec- 
tory may have approached the line of sight. 
A majority appeared very small, not larger 
than Aristarchus (or 0’.3) ; few equaled Coper- 
nicus (or 0’.6) ; one only had a spread of wings 
equal to the longer diameter of the Mare Cris- 
ium (or 3/.0). In this case the wings were 
sharply seen when the focus was 0/.21 longer 
than the principal focal length. For most of 
the birds the focus scarcely required changing 
from that for the Moon’s surface. Some of the 
swiftest flights were made by birds of the 
smallest apparent size. 
After 8:15, as the moon rose higher (altitude 
25° to 35°), the number of birds diminished ; 
and at 9:15 intervals of 2 to 5 minutes elapsed 
between successive birds. The travelers no 
longer came in groups. Evidently the angular 
area of the Moon no longer filled the distance 
between the mean trajectories of the members 
of a group. The average size was larger, 
more nearly that of Copernicus; one bird, badly 
out of focus, equaled the Mare Crisium in size. 
At 9:15 the observations were discontinued, as 
the intervals were continually getting longer. 
Second night, September 24th. The watch 
was commenced at 7:45 with the Moon at a 
lower altitude than on the previous evening. 
The birds were less numerous, and after a half 
hour at 8:15, clouds began to gather, which 
soon completely covered the sky. In this half 
hour not over a dozen birds were seen. Two 
of these moved from south to north, traversing 
the diameter of the lunar disk in about 3 
seconds. These were the only birds seen fly- 
ing north on either evening, and the slowness 
of their speed indicates that they were prob- 
ably moying more nearly in the line of sight 
(or east and west). Every flutter of the wings 
was plainly visible with the telescope at its lu- 
nar focus. One bird, of an expanse equal to 
that of the Mare Serenitatis, was blurred be- 
yond recognition, and was evidently close at 
hand. One had the wavering flight of the 
SCIENCE. 
[N. S. Vou. VI. No. 141. 
goldfinch, and a diameter equaling that of 
Avristarchus (or 0./3). 
It seems possible that the prospect of cloudy 
weather may have deterred the migrating birds 
from starting on this evening. ‘The interval 
from sunset to moonrise was also longer (48 
minutes as against 24 minutes on the previous 
evening). 
I now proceed to the reduction of these es- 
timates, taking first the case of the single bird 
whose motion was slow enough to permit an 
exact adjustment of the focus, the focal length 
having to be increased for the bird from 57.59 
inches to 57.80 inches. The bird was nearly 
end on. Assuming its real spread to have been 
12 inches, and comparing its estimated angular 
diameter with the radius of a circle we have its 
3437.7 
distance = = 1146 feet. 
Reversing the process, we have from the law of 
lenses, 
Conjugate focal length =1321 feet, 
which would make the spread 
1321 
1146 
The spread of a robin is 16 inches, and the 
bird may have been of the size of a thrush. 
The majority of the flights were at right 
angles to the line of sight, and the lengths of 
our smaller birds (warblers, flycatchers, etc.) 
being from 5 to 7 inches, I shall assume an 
average true size of 6 inches, and an apparent 
(angular) dimension of 0’.3 in periods a and e, 
with a mean altitude of 20° ; while towards the 
close of period 6, the altitude had increased to 
35°, and the mean apparent angular diameter 
began to approach 0/.6. This gives for dis- 
+ 1213.8 inches. 
tances 
(a) and (c) peat X< $ = 5729.5 feet, 
3437.7 
(b) 7 >< } = 2864.8 feet, 
and for the heights of the birds above the ob- 
server’s level (235 feet above sea-level, 
(a) and (c) 5729.5 X sin 20° = 1959.6 feet, 
(b) 2864.8 X sin 35° = 1643.2 feet. 
These measurements indicate an altitude of 
about 200 feet above sea-level as the average 
