182 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1926 
14. Hitherto we have considered the constitution of individual 
stars. We now turn to the birth of starry systems. Before the 
modern aids to exact telescopic work were available, it was found 
that nebulous patches of very definite forms are in the heavens. 
These are now resolved by photography into such beautiful spirals 
as that in the constellation of the Hunting Dogs. Others of these 
spirals are more or less inclined to our line of vision, and some are 
even seen directly on edge. Seen under these various angles of 
presentation, we find that all of these so-called spiral nebulae are of 
flattened, watchlike shape. 
Curiously enough, too, our own galaxy, which contains some 30,- 
000,000,000 stars or more, is of this flattened type, and extends out 
quite five times as far along the Milky Way as towards its poles. 
Sir William Herschel suggested more than a century ago that the 
spiral nebule were, as he picturesquely termed them, “island uni- 
verses.” In other words, he suggested that they are galaxies like 
ours, but so distant that the largest telescopes of his time could not 
separately distinguish their stars. 
This has now proved true. Hubble with the 100-inch telescope 
on Mount Wilson has succeeded in resolving some of the spiral 
nebulae in part into stars of exceeding faintness. By means of cer- 
tain of these stars, called Cepheids, which have a peculiar type of 
variable brightness, he has shown that the great nebula of Andro- 
meda and others are nearly 1,000,000 light-years distant. 
Considering the similarity of shape, of general spectrum, and 
other points of resemblance, there is no little doubt that we, our- 
selves, reside not excessively far from the center of a galaxy which, 
to the intelligent dwellers on the Andromeda galaxy, if such there 
be, must seem like a great spiral nebula. 
15. In all these spirals, of which there are some hundreds of 
thousands in the heavens, two arms depart from opposite sides of 
the central condensation. ‘There are, however, other nebulous objects 
which present gradations of form back from the spiral to the 
spindle, to the ellipsoid, and finally to the sphere without hint of 
structure. Professor Jeans has computed the relations of masses 
and motions which would cause a globe of gas to elongate its equa- 
torial radii unde” rotation, and pass on to the spindle shape. At a 
certain degree of extension, the gravitation toward the center would 
be nearly balanced by the centrifugal force of rotation. Then the 
trifling differences of attraction from different directions exercised 
by the other masses of the universe would cause tidal extensions to 
leave the parent mass at opposite ends of a diameter. Jeans goes on 
to show how these extensions would draw away more and more 
material into ropelike arms, which, as they extended, would neces- 
sarily wind into the spiral forms. He even succeeded in proving 
