Some distant telescopic companions to red stars have been described as blue. This may be in some case due, partly at least, to the effect of contrast. In others the blue color seems to be real. This has been shown spectroscopically to be the case with the bluish companions of Beta Cygni. The physical cause of the difference of color is still more or less a matter of mystery. Although we can not consider it proved that the red stars are cooling and “dying out” suns, as has been suggested, we may, I think, conclude that their temperature, although doubtless very high, must be lower than that of the white stars. We know that a bar of iron when heated to redness is not so hot as when raised to “white heat,” and although the analogy between hot iron and stellar photospheres may not be a perfect one, it seems probable that the higher the temperature of a star, the whiter its color will be. Most of the white stars, as Sirius, Vega, and those only yellow or slightly colored, show spectra of Secchi’s first and second types, while the great majority of the red stars exhibit banded spectra of the third and fourth types. To this rule there are, however, like other rules, some notable exceptions. For instance, Aldebaran, Alpha Hydræ, Xi Cygni, and 31 Orionis, although distinctly reddish stars, show well-marked spectra of the second or solar type. On the other hand, Rho Ursæ Majoris and Omega Virginis, which, according to Dunér, are only slightly yellow, have well-marked spectra of the third type. An apparent change of color seems in some cases to be well established. The supposed red color of Sirius in ancient times is well known. A certain established change is found in the case of the famous variable star Algol, which is distinctly described as red by Al-Sûfi in the Tenth Century. It is now pure white, or nearly so, and this is probably the best attested instance on record of change of color in a bright star. Schmidt’s Nova Cygni of 1876 was noted as “golden yellow” on the night of its discovery. When it had faded to the eighth magnitude, Dr. Copeland called it “decided red,” but when examined at Lord Crawford’s observatory in September, 1877, its color was recorded as “faint blue”! The new star in the Andromeda nebula was considered to be yellowish or reddish by most observers when near its maximum, but about a month later its color was noted as “bluish.”