Still further: if the sun be surrounded by a gaseous envelope, its density, as aforesaid, ought to diminish from the solar surface outward to its upper limits; but the fact is, the material of 1474 K line always appears in the spectrum of chromosphere, which would seem to indicate, by its place, that it is as much more dense than hydrogen as is magnesium vapor, or even the vapor of iron. But the evidence of the spectroscope makes this 1474 K material far less dense than that of hydrogen, and this is a contradiction that is very troublesome to the student of solar physics. In studying the polarization of the light of the corona, it is clear that the amount of polarized light reflected from a particle at the surface of the sun is nothing, "because the luminous source there is a surface with an angular subtense of 180°;" hence polarization of the corona near the limb of the moon ought to be small, farther away, larger. But observation shows that the contrary is true, _i. e._, the percent. of polarized light increases as the corona is observed nearer the limb of the moon during totality. These are a few of the difficult questions that stand in the way of accepting the foregoing theories as facts pertaining to, or well grounded knowledge of, the constitution of the sun. They are by no means all, or possibly the most important ones. They are certainly among those that are receiving very general attention at the hands of physicists at the present time.--_Sidereal Messenger._ CHANGES IN THE STELLAR HEAVENS. By J. E. GORE, F.R.A.S., Honorary Associate and Vice-President of the Liverpool Astronomical Society.