ASTRONOMY: A. VAN MAANEN 
389 
density of the center of the nebula has been reduced to show the motions 
more clearly. 
If the results as illustrated on the plate could be taken at their face 
value, they would certainly indicate a motion of rotation, or possibly 
motion along the arms of the spiral. Without expressing a final opin- 
ion as to the character of the motion, which must be determined by 
future work, it may be of interest to examine the evidence afforded by 
the existing material. Comparing the motions with the directions of 
the branches of the spirals, we find from 52 points in which this direction 
can be specified with fair accuracy, that the mean divergence of the 
motions is 7° =*= 4° toward the concave side of the spirals. 
To discuss the internal motions from the standpoint of rotation, they 
were analyzed into two components, along and perpendicular to the 
radius, the latter for convenience being spoken of as the rotational 
component. 
The results are as follows: 
78 points have a left-hand motion (N. W. S. E.), only 9 moving right- 
handedly; 58 points appear to be moving outward, while 28 show motion 
inward. The rotational motion is the larger in the majority of cases, 
viz., 63 points. The mean rotational motion is 01022 left-handed; 
the mean radial motion is 0!007 outward. The probable reality of the 
result is indicated by the satisfactory agreement of the pairs of plates, 
as shown by the following summary. 
''rot /*rad. 
+0T021 +0r004 Ritchey 1910 and 1915 
-1-0 . 032 +0.012 Ritchey 1910 and 1915 
-1-0.012 -K0.007 Keeler and Curtis 1899 and 1914 
-1-0.017 -f0.006 Keeler and Perrine 1899 and 1908 
The measures indicate a small but scarcely reliable decrease of rotational 
motion with increasing distance from the center, as shown by the follow- 
ing table : 
Distance from 
Center /'rot. No. of Points 
<3'.l 0*024 19 
3.1 to 5.0 , 0.028 29 
5.1 to 7.0 0.014 18 
>7.0 0.019 21 
The annual rotational component of 0.022 at the mean distance 
from the center of 5' corresponds to a rotational period of about 85,000 
years. If we knew the parallax of the nebula, and if we couid assume 
that the motions and the distances of the points from the center are 
