CELESTIAL SPECTROSCOPY. 89 
to make the nebulae as bright as they appear to us. On such an 
assumption the average temperature may be low, but the individual 
particles, which by their encounters are luminous, must have motions 
corresponding to a very high temperature, and in this sense be 
extremely hot. 
In such diffuse masses, from the great mean length of free path, the 
encounters would be rare but correspondingly violent, and tend to bring 
about vibrations of comparatively short period, as appears to be the 
case if we may judge by the great relative brightness of the more refran- 
gible lines of the nebular spectrum. 
Such a view may, perhaps, reconcile the high temperature, which the 
nebular spectrum undoubtedly suggests, with the much lower mean 
temperature of the gaseous mass which we should expect at so early a 
stage of condensation, unless we assume a very enormous mass, or that 
the matter coming together had previously considerable motion or con- 
siderable molecular agitation. 
The inquisitiveness of the human mind does not allow us to remain 
content with the interpretation of the present state of the cosmical 
masses, but suggests the question— 
What see’st thou else 
In the dark backward and abysm of time? 
What was the original state of things? how has it come about that by 
the side of aging worlds we have nebule in a relatively younger stage? 
Have any of them received their birth from dark suns, which have col- 
lided into new life, and so belong to a second or later generation of the 
heavenly bodies? 
During the short historic period, indeed, there is no record of such an 
event; still it would seem to be only through the collision of dark suns, 
of which the number must be increasing, that a temporary rejuve- 
nescence of the heavens is possible, and by such ebbings and flowings 
of stellar life that the inevitable end to which evolution in its appar- 
ently uncompensated progress is carrying us can, even for a little, be 
delayed. 
We can not refuse to admit as possible such an origin for nebule. 
In considering, however, the formation of the existing nebulwe we 
must bear in mind that, in the part of the heavens within our ken, the 
stars still in the early and middle stages of evolution exceed greatly in 
number those which appear to be in an advanced condition of condens- 
ation. Indeed, we find some stars which may be regarded as not far 
advanced beyond the nebular condition. 
It may be that the cosmical bodies which are still nebulous owe their 
later development to some conditions of the part of space where they 
occur, such as, conceivably, a greater original homogeneity, in conse- 
quence of which condensation began lessearly. In other parts of space 
condensation may have been still further delayed, or even havenot yet 
