That some of the ancients knew that the Milky Way is composed of stars is shown by the following lines translated from Ovid:—
“A way there is in heaven’s extended plain
Which when the skies are clear is seen below
And mortals, by the name of Milky, know;
The groundwork is of stars, through which the road
Lies open to great Jupiter’s abode.”[470]
From an examination of the distribution of the faint stars composing the Milky Way, and those shown in Argelander’s charts of stars down to the 9½ magnitude, Easton finds that there is “a real connection between the distribution of 9th and 10th magnitude stars, and that of the faint stars of the Milky Way, and that consequently the faint or very faint stars of the galactic zone are at a distance which does not greatly exceed that of the 9th and 10th magnitude stars.”[471] A similar conclusion was, I think, arrived at by Proctor many years ago. Now let us consider the meaning of this result. Taking stars of the 15th magnitude, if their faintness were merely due to greater distance, their actual brightness—if of the same size—would imply that they are at 10 times the distance of stars of the 10th magnitude. But if at the same distance from us, a 10th magnitude star would be 100 times brighter than a 15th magnitude star, and if of the same density and “intrinsic brightness” (or luminosity of surface) the 10th magnitude would have 10 times the diameter of the fainter star, and hence its volume would be 1000 times greater (103), and this great difference is not perhaps improbable.
The constitution of the Milky Way is not the same in all its parts. The bright spot between β and γ Cygni is due to relatively bright stars. Others equally dense but fainter regions in Auriga and Monoceros are only evident in stars of the 8th and 9th magnitude, and the light of the well-known luminous spot in “Sobieski’s Shield,” closely south of λ Aquilæ, is due to stars below magnitude 9½.
The correspondence in distribution between the stars of Argelander’s charts and the fainter stars of the Milky Way shows, as Easton points out, that Herschel’s hypothesis of a uniform distribution of stars of approximately equal size is quite untenable.
It has been suggested that the Milky Way may perhaps form a ring of stars with the sun placed nearly, but not exactly, in the centre of the ring. But were it really a ring of uniform width with the sun eccentrically placed within it, we should expect to find the Milky Way wider at its nearest part, and gradually narrowing towards the opposite point. Now, Herschel’s “gages” and Celoria’s counts show that the Galaxy is wider in Aquila than in Monoceros. This is confirmed by Easton, who says, “for the faint stars taken as a whole, the Milky Way is widest in its brightest part” (the italics are Easton’s). From this we should conclude that the Milky Way is nearer to us in the direction of Aquila than in that of Monoceros. Sir John Herschel suggested that the southern parts of the galactic zone are nearer to us on account of their greater brightness in those regions.[472] But greater width is a safer test of distance than relative brightness. For it may be easily shown than the intrinsic brightness of an area containing a large number of stars would be the same for all distances (neglecting the supposed absorption of light in space). For suppose any given area crowded with stars to be removed to a greater distance. The light of each star would be diminished inversely as the square of the distance. But the given area would also be diminished directly as the square of the distance, so we should have a diminished amount of light on an equally diminished area, and hence the intrinsic brightness, or luminosity of the area per unit of surface, would remain unaltered. The increased brightness of the Milky Way in Aquila is accounted for by the fact that Herschel’s “gages” show an increased number of stars, and hence the brightness in Aquila and Sagittarius does not necessarily imply that the Milky Way is nearer to us in those parts, but that it is richer in small stars than in other regions.
Easton is of opinion that the annular hypothesis of the Milky Way is inconsistent with our present knowledge of the galactic phenomena, and he suggests that its actual constitution resembles more that of a spiral nebula.[473] On this hypothesis the increase in the number of stars in the regions above referred to may be due to our seeing one branch of the supposed “two-branched spiral” projected on another branch of the same spiral. This seems supported by Sir John Herschel’s observations in the southern hemisphere, where he found in some places “a tissue as it were of large stars spread over another of very small ones, the immediate magnitudes being wanting.” Again, portions of the spiral branches may be richer than others, as photographs of spiral nebulæ seem to indicate. Celoria, rejecting the hypothesis of a single ring, suggests the existence of two galactic rings inclined to each other at an angle of about 20°, one of these including the brighter stars, and the other the fainter. But this seems to be a more artificial arrangement then the hypothesis of a spiral. Further, the complicated structure of the Milky Way cannot be well explained by Celoria’s hypothesis of two distinct rings one inside the other. From analogy the spiral hypothesis seems much more probable.
Considering the Milky Way to represent a colossal spiral nebula viewed from a point not far removed from the centre of the spiral branches, Easton suggests that the bright region between β and γ Cygni, which is very rich in comparatively bright stars, may possibly represent the “central accumulations of the Milky Way,” that is, the portion corresponding to the nucleus of a spiral nebula. If this be so, this portion of the Milky Way should be nearer to us than others. Easton also thinks that the so-called “solar cluster” of Gould, Kapteyn, and Schiaparelli may perhaps be “the expression of the central condensation of the galactic system itself, composed of the most part of suns comparable with our own, and which would thus embrace most of the bright stars to the 9th or 10th magnitude. The distance of the galactic streams and convolutions would thus be comparable with the distances of these stars.” He thinks that the sun lies within a gigantic spiral, “in a comparatively sparse region between the central nucleus and Orion.”
Scheiner thinks that “the irregularities of the Milky Way, especially in streams, can be quite well accounted for, as Easton has attempted to do, if they are regarded as a system of spirals, and not as a ring system.”
Evidence in favour of the spiral hypothesis of the Milky Way, as advocated by Easton and Scheiner, may be found in Kapteyn’s researches on the proper motions of the stars. This eminent astronomer finds that stars with measurable proper motions—and therefore in all probability relatively near the earth—have mostly spectra of the solar type, and seem to cluster round “a point adjacent to the sun, in total disregard to the position of the Milky Way,” and that stars with little or no proper motion collect round the galactic plain. He is also of opinion that the Milky Way resembles the Andromeda nebula, “the globular nucleus representing the solar cluster, and the far spreading wings or whorls the compressed layer of stars enclosed by the rings of the remote Galaxy.”