520 
ASTRONOMY: F. G. PEASE 
of the lines measured lie in the more refrangible portion of this region. 
A valuable check has been furnished by a photograph of the spectrum 
of the sky and spark taken in the same way as the nebular plate. The 
results of the measures by Mr. Adams and Miss Burwell are given in 
Table I. The correction for the earth's motion amounts to —1 km. 
for the mean epoch of the photograph and has been neglected. 
Adams 
Bunvell 
Mean 
No. lines Mean 
No. lines 
Mean 
V elocity 
km. 
km. 
km. 
5 th space 
2 
+ 830 
4 
+ 775 
+ 800 
4th space 
4 
960 
6 
905 
930 
West Side < 
3rd space 
4 
980 
6 
930 
955 
2d space 
6 
970 
6 
970 
970 
^Ist space 
6 
1155 
8 
1105 
1130 
5 
1210 
7 
1190 
1200 
6 
1250 
7 
1275 
1260 
1st space 
7 
1275 
7 
1275 
1275 
2d space 
7 
1330 
9 
1350 
1340 
East Side <| 
3d space 
5 
1390 
5 
1395 
1390 
4th space 
4 
+ 1380 
4 
1600 
1490 
5 th space 
2 
+ 1630 
+ 1630 
These velocities are shown in figure 5, the ordinates being kilometers 
per second and the abscissae seconds of arc. The velocity curve 
y = - 2.78 X + 1180 
was determined by a least squares solution. From this equation we 
find: 
1. The radial velocity of the nebula is + 1180 km., a value in good 
agreement with that of + 1100 km. found by Slipher. 
2. The linear velocity of rotation at a point 2 minutes of arc from 
the nucleus is over 330 km. 
3. Within the limits of accuracy of the measures the change of rota- 
tional velocity is linear, although there may be some variation in indi- 
vidual parts of the nebula. 
The third result may be interpreted either as indicating that the 
nebula is rotating as a solid body, or that the material is moving in 
accordance with a law which will give a linear velocity curve. The 
latter seems the more probable explanation, since the high velocity is 
opposed to our conceptions of stability in the case of a rotating solid 
body. In any event the results seem to be inconsistent with a system 
involving planetary motion about a central nucleus, since this would 
require an increase of linear velocity toward the center of the nebula. 
If we assume that the rotational component of the internal motion 
in this nebula is of the same order as that found by van Maanen for 
Messier 101, we can at once derive a value for the parallax. Van 
