254 NEW STARS IN ANDROMEDA AND ORION. 
among themselves in their drawings; furthermore, that 
the appearance of a bright body in its midst would of 
itself obscure minor changes. 
Other astronomers hold that a physical connection be- 
tween star and nebula is the more reasonable conclu- 
sion; and I incline toward this theory in spite of the 
fact that many of the best authorities are not disposed 
to favor it. The appearance of a new star in 1860 with- 
in the Scorpio nebula, to which I have already alluded, 
gives greater import to this supposition. Is it probable 
that so rare a phenomenon as a new star should present 
itself twice in twenty-five years within a nebula, and in 
a nebula known in one case to be resolvable, and in the 
other supposed to be so upon spectroscopic evidence, — 
is it probable that this should twice occur and yet that 
the connection should be accidental only ? Herr Seeli- 
ger, of Munich, who believes the connection is a real 
one, advances the hypothesis that the phenomenon may 
be attributed to a collision between two of the supposed 
constituent members of the close group forming the 
nebula. He takes up the problem mathematically and 
constructs a formula which gives the rate at which a 
body brought suddenly to a great heat by collision 
would lose light ; and he finds for himself a satisfactory 
agreement between the theoretical values obtained by 
this process and the values obtained by observation. 
But the broad assumption of data necessary to the de- 
duction of these theoretical values detracts from the im- 
portance of such evidence. Still, in a close group col- 
lision is not improbable, and perhaps quite as probable 
as that there should occur in this long observed nebula, 
unchanged within recording epochs, so sudden a trans- 
formation from causes lying within the bodies them- 
selves. | 
It is, of course, possible that other temporary stars 
may have been caused by collision. Sir John Herschel 
1388s 
