26 
POPULAR SCIEJSCE REVIEW. 
in their passage through it. There is good reason to suppose 
that the product of chemical union assumes generally the liquid 
rather than the gaseous form, since it has frequently been noted 
that a faint continuous spectrum may be discerned along with 
that of the bright gaseous lines in many nebulae, and which is 
not confined to any particular portion of them. 
We may now pretty clearly depict the condition of a nebula 
according to this hypothesis, when after the lapse of many ages 
a large number of comets and accompanying meteor-streams 
have been absorbed into its substance. It will consist of a 
greater or less residue of the original gaseous constituents, 
which, for reasons well known to spectroscopists, will still con- 
tinue to yield most conspicuously the characteristic bright line 
spectrum, and dispersed throughout the mass an immense 
number of liquid nuclei, all tending towards the common 
centre of gravity in spiral orbits, the centre being occupied by a 
brilliant white-hot liquid mass. The temperature of such a 
nebula will be much higher than of one less developed — a fact 
which will tend to keep the density of the gaseous constituents 
at nearly the same level, in spite of the increased gravitational 
tendency to condensation near the centre. We may even con- 
ceive that this intense heat might be sufficient to render 
gaseous the liquid nuclei When a numerously bright-lined spec- 
trum should make its appearance, but it would then be neces- 
sary to assume that the faintness of the spectrum would make 
it difficult to distinguish the difference between this and one 
truly, continuous. We may, on the contrary, imagine the well- 
developed nebula to become a compact liquid mass, in this case 
also yielding a continuous spectrum. Nor does it seem unlikely 
that a considerable number of bodies in this latter state of ex- 
istence may actually be discovered in the heavens. Many 
nebulae, after having resisted the utmost efforts of astronomers, 
armed with the largest telescopes, to class them among the 
clusters of stars, have equally refused to be entered in the list of 
true nebulae by yielding the discontinuous spectrum, and whose 
real condition remains, therefore, for the present an enigma. 
Of the two suppositions the second appears to me to be the 
most probable ; yet if it be accepted, it will be necessary to re- 
construct the nebular hypothesis, if it can still be maintained 
under the altered circumstances. There are, however, some facts 
which tend strongly to show that the first suggestion is the 
more correct one, or, in some cases at least, more closely ap- 
proximates to the true condition of things. It is well known 
that the great nebula in Andromeda — one of those which defy 
alike the powers of the telescope and spectroscope — presents a 
very anomalous spectrum. Though apparently continuous, 
some portions of the red and orange are altogether wanting, 
and the more refrangible portion consists of a mottled band 
