JANUARY 14, 1897 | 
I wish to point out that from the centre of the conden- 
sation the luminosity gradually gets less and less until at 
last we have no Juminosity greater than that of the sur- 
rounding sky. In the nebula itself we find exquisite 
and 
spirals, starting apparently from different points, 
Fic. 4.—Messier 74 Piscium, 1893 
gradually coming towards the centre, and if we look along 
these spirals we see that the star-like masses, which may 
not be stars, are in many cases located on the spirals, 
representing apparently minor condensations, each itself 
Fic. 5.—Spiral nebula, Messier ror Urs Majoris. 
being probably brighter than the other parts because it 
is more disturbed. 
So much, then, for the autobiographical records we 
now possess of some of the most perfect spiral nebulie 
in the heavens. 
NO. 1420, VOL. 55 | 
NATURE 251 
We see that they all resemble perfect eddies in appear- 
ance; the question arises, are they perfect eddies in fact ? 
On the meteoritic hypothesis they may well be so, for if 
moving streams of meteorites encounter resistance to 
their motion due to disturbances by other masses, the 
sheets of meteorites are bound to behave like sheets of 
water ; in any case, the ozus proband lies with those 
who hold the contrary view. But in these celestial 
maélstroms there are bound to be smaller eddies ; and, 
if Swift had had the opportunity of studying Dr. Roberts’ 
photographs, a more grandiose image might have re- 
placed that in his well-known lines— 
** So naturalists observe, a flea 
Has smaller fleas that on him prey, 
And these have smaller still to bite ’em, 
And so proceed ad znfinttum.” 
The question is well worth asking, not only to enable us 
to explain the photographic and spectroscopic phenomena, 
but also because we seem to be in presence of forces 
which must ultimately result in a true star with rotation, 
a concomitant of star life which is not easy to explain, 
Fic. 6.—M 33 Trianguli, 
and which Lord Kelvin has shown would certainly of be 
produced by collisions of two finished cosmical bodies.! 
There is another spiral nebula, however, which may 
carry us a little further along the same line. 
In M 33 Tnrianguli we have something apparently 
different from those that have preceded ; so much so, that 
Dr. Roberts, who till quite recently has remained silent 
with regard to the physical origin of the more regular 
spirals, has suggested that we may here be in presence of 
meteoritic collisions.* He writes :— 
“Tt will be observed that there are two large, very 
prominent spiral arms, with their respective externz il 
curvatures facing north and south, and that the curves 
are approximately symmetrical from their extremities to 
their point of junction at the centre of rev olution, where 
there is a nebulous star of about tenth magnitude with 
dense nebulosity surrounding it, and elongated i in north 
and south directions. Involved in this nebulosity are 
three bright stars and several faint nebulous stars ; the 
two arms also are crowded with well-defined stars and 
faint nebulous stars with nebulosity between them ; and 
1 Proc. R.T., 1. Ivi. p. 7o. 
Xii. p. 15- 
