« PrécédentContinuer »
A few of the nebulae shown are very faint, but time has not permitted the securing of longer exposures in such cases. Some were unsuitable for photographic enlargement, because of small size or distortion due to distance from the center of the plate, and drawings of these are reproduced. These nebulae are 4282 (18), 678 (19), 169 (20), I 1029 (30), 5506 (31), 4312 (32), 828 (33), and 3607 (34). Throughout this paper nebulae will be referred to by their number in Dreyer's New General Catalogue with the omission, in general, of the letters "N.G.C." Following each N.G.C. number, in parentheses, is the number of the figure in the accompanying illustrations. It is scarcely necessary to call attention to the inevitable loss of the finer details, which is unfortunately inherent in all photo-engraving processes, and to state that the phenomena discussed are invariably much clearer on the original negatives than in the reproductions. It was found impossible to show the delicate lane of 4526 (17) in a photographic enlargement until the lane had been accentuated by retouching; this reproduction may then be regarded as a quasi-sketch; no others have been retouched in any way.
1 From left to right.
'False star images occur in some of the illustrations, as for example in the following:
2 The scale drawn on figure 18 should be numbered 0 —12", not 0 — 30".
Fig. 3; 7.1 mm. from right hand edge, 15.2 mm. from top of illustration.
Fig. 27; 37.1 3.9
Fig. 63; 20.6 11.7
The first group of data comprises those nebulae which are seen almost exactly edgewise and show indubitable evidence of dark lanes. These are:
891 ( 1) 3718 (13)
4013 ( 2) 4388 (15)
7814 ( 3) 2968 (16)
4565 ( 4) Finest example of the class. 4282 (18)
5866 ( 5) Lane inclined to major axis. 4710 (28)
4402 (11) 1888 (29) The narrow, somewhat irregular lane
4244 (12) is lost in the reproduction.
Merging imperceptibly into the group just given are a number of spirals seen almost but not exactly edgewise, and a few whose planes make a small but appreciable angle with the line of sight, and which show clear evidence of dark lanes.
4594 ( 6) Brightest and most remarkable object
of the class. The very indistinct
evidences of spiral character are
lost in the reproduction. 5746 ( 7)
3628 ( 8) Rather irregular.
From descriptions or sketches given by Herschel, Rosse, Shaw, and others, the following nebulae almost certainly belong to the class of edgewise spirals with dark lanes: 134, 520, 1316, 1365, 3189-90, 3521, 4217, 5128, 7172.
Several of these are inaccessible from our latitude, and the others have not yet been photographed with the Crossley reflector.
Considering the proportion of the spirals which may be expected, if oriented at random in space, to present themselves almost exactly edgewise to us, the fact that as many as thirtyeight of those observed to date show dark bands must be regarded as an indication that this phenomenon is a very common one, and is perhaps the rule, rather than the exception. There are a few large spindle-shaped nebulae which do not show the effect, at least on the scale of the Crossley photographs. An example is 3115, in Lick Observatory Publications, VIII, plate 22. In this nebula even very short exposures show no trace of a lane; a very narrow lane might escape detection because of the masking effect of the grain of the plate. A large number of uncatalogued, very small, spindle-shaped nebulae are found on the Crossley plates; could these be photographed on a considerably larger scale it is probable that many would show dark lanes; such effects are suspected in a number of cases not illustrated in this paper.
It is possible that several rather irregular, greatly elongated nebulae give less definite evidence on the same point.
4631 (22) Quite patchy and irregular.
7537 (55) Perhaps a "single-whorled" spiral.
7640 (65) Narrow lane across center at the junction of the two branches of the spiral.
3079 (67) Quite irregular; matter is all on one side of the bright, and presumably nuclear portion.
3034 (68) Quite irregular and patchy; no trace of spiral structure.
5195 (74) This is the companion nebula to the great spiral 5194 (Messier 51), and shows an interesting lane just below and parallel to the major axis of its nuclear portion. This lane is almost lost in the reproduction.
The effects in all these may be due solely to their patchy and irregular structure, though it is not improbable, in view of the frequent evidence of obstructing material found in the more regular edgewise spirals, that a part, at least, of the darker patches may be due to a similar cause.
Before proceeding to any discussion of the meaning of the phenomenon it will be advisable to examine, as a second division of the observational evidence, that large class of spiral nebulae whose principal planes make a small but appreciable angle with our line of sight, appearing as greatly elongated ellipses. In the approximately round spirals, whose planes lie almost at right angles to the line of sight, there are frequent irregularities of structure, with one whorl of the spiral longer or brighter than the others. But in both the nearly round and in the greatly elongated spirals the effects of irregularity of form or of difference of intensity should, when we consider a sufficiently large number of cases, give asymmetrical effects oriented at random with regard to the nucleus in the rounder forms, or with reference to the apparent major axis in the elongated forms. So far as actual irregularities of the form are concerned, we should then expect to see as many of the greatly elongated spirals showing an asymmetrical condition along the major as along the minor axes of their elliptical projections. This, however, is not found to be the case. In the relatively frequent instances where the greatly elongated spirals show any lack of symmetry, such asymmetry is almost invariably along the minor axis, and roughly parallel to the major axis of the ellipse. This asymmetry manifests itself frequently in "lanes" prominent on one side of the major axis and faint or invisible on the other; in a fan-shaped nuclear portion; in an apparent displacement of the nucleus along the minor axis; in a considerably greater brightness of the nebular material on one side of the major axis, or in various combinations of these effects.
Among the nebulae showing dark lanes on one side of the major axis are the following:
3623 (23) Very striking; at one end the band appears to cut across a bright whorl.
4429 (38) Fine narrow lane near nucleus; compare the similar nebula 4526 (17).
4826 (54) A remarkable case; lane seems absolutely black and cuts across a whorl at the right end.
4274 (61) Narrow lane just below nucleaus, scarcely visible in reproduction; longer one below.
5394-5 (73) The smaller, two-branched spiral at the right shows a narrow lane just above the nucleus.
I have endeavored to include in the Crossley Reflector programme all cases accessible from this latitude where two nebulae are described as being very close together, in the hope of securing similar evidence of occulting matter in other nebulae than those nearly edgewise. None of these thus far photographed shows an actual occulting effect except perhaps in the ease of 4567-8 (69). These fine spirals are about two and three minutes of arc in length, respectively, and there is apparently a slight amount of overlapping. While the evidence can not be regarded as conclusive, there is some indication of an occulting effect in the dark lane on the upper side of the lower of the two pebulae; this lane appears to cut across the end of the lower whorl of the upper nebula (4567). Additional probability is lent to this view by the asymmetrical structure of the lower of the two nebulae (4568), where it will be noticed that most of the nebular matter is on the opposite side of the major axis from the dark lane.
The giant spirals 3031 (Messier 81) and 224 (Andronuda) are shown in figures 76-78; figure 77 shows the nebula in Andromeda as taken with the Crossley reflector, and figure 78 the same nebula taken on the much smaller scale of the Willard portrait lens of the Crocker telescope. Though these objects make a greater angle with the line of sight than the majority of the elongated spirals shown in the illustrations, a study of these enormous spirals shows very interesting evidence in the lanes more prominent on one side of the major axis than on the other. Taken by themselves there would be little reason to consider the lanes shown in these two nebulae as anything else than actual vacant lanes between the nebular whorls. Viewed in connection with the other elongated spirals, however, there is good reason for the belief that a similar manifestation of occulting matter is strongly, though not certainly indicated.
Contributory evidence is afforded by the elongated spirals in which the nebular matter is considerably fainter on one side of the major axis of the projected ellipse, or where the nebular matter extends considerably further from the apparent center on one side of the major axis.
A few spirals show minute, irregular, vacant patches in their central portions, which may well be due to the presence of small clouds of obstructing material; among such are 147, 205, 2655, 3077, and 4449 (not illustrated).
It is not impossible that these effects may be due to the same general cause that produces certain occulting effects in our own galaxy, though there is manifestly a good deal of assumption involved in postulating the same character of occulting material in the vicinity of objects so different in spectrum, in space-distribution, and space-velocity as the spirals and, for instance, the great diffuse nebulosities. It will, however, be of interest to give a few illustrations of such intra-galactic manifestations of occulting material as contributing, though perhaps but remotely, to the subject under discussion.
1. Photographs and star-counts agree in showing a marked falling off in the number of stars in the vicinity of several of the great diffuse nebulosities, as, for example, the Orion and North America nebulae. One of the most striking exampl-s of this effect is afforded by the diffuse nebulosity II 5146, shown in figure 79. It seems very clear that the cutting off in the number of the stars around such a nebulosity mast be due to the presence of occulting and non-luminous matter outside the luminous portion of the nebula.
2. The "coal sacks" and starless regions in or near the Milky Way may possibly be mere "holes." A theory which postulates occulting matter is, however, inherently at least as probable; it is difficult to avoid the impression that something is blotting out everything beyond. in such regions as that southeast of 0 Ophiuchi, where an exposure of two hours with the Crossley shows an almost absolute stellar d<-sert. A similar remarkable region is that at \*h 55TM, —37% of which Shaw says (Helwan Bull., no. 9;, "There are as many as twenty reseau squares (each 5' X 5') in which not a single star appears on a two hours• exposure with the reflector. These dark spaces are especially noteworthy as the region can not be said to be in the Milky Way, being 19° from the galactic plane and well away from the visible star clouds."
3. Professor Barnard has photographed and described many small starless spots which he believes to be "dark nebulae," visible through their projection upon a background of stars. Several of these have been photographed with the Crossley reflector, and it seems impossible, on carefully examining these negatives, to believe that these are actual holes in the Milky Way. The most remarkable object of the class known to me is that at \lb 57m, —27° 50', which Professor Barnard discovered in 1883 with a five-inch telescope. The enlarged Crossley photograph is reproduced in figure 80. This object is about six minutes of arc in length, and attention is directed particularly to the roughly circular protuberance at the southwest corner, as clear cut as an ink drop and perfectly dark. As Dr. Campbell has well pointed out. such appearances, if of the character of holes in the Milky Way, must be of an age to be reckoned by hundreds of millions of years, and the random motions of the near-by stars should long since have obliterated the clearcut edges, if not the hole itself.
Another beautiful example of a dark nebula is that near N.G.C. I 434, 5»> 36.0TM —2° 27', which is reproduced elsewhere in this volume as figure 5, accompanying the paper, "Descriptions of 762 Nebulae and Clusters Photographed with the Crossley Reflector." A description of this remarkable object is given at the proper place in this list; cf. also Publ. Astr. Soc. Pac, 30, 65, 1918.
4. A similar favorable background for rendering dark nebulae visible is that afforded by the very large diffuse nebulosities. Dark spots and rifts are seen in many of these. Perhaps the most striking and convincing effects of this nature are those seen in the enormous diffuse nebulosity 6523 (Messier 8), reproduced in figure 81. It is impossible to conceive that these inky-black, clearcut spots are holes extending through a mass of nebular matter for a distance to be measured in light-years.
5. In a number of spectroscopic binaries the H and K lines of calcium give a constant radial velocity, and do not show the periodic shift of the other lines in the spectrum. In several others, these lines show a periodic shift, but of smaller amplitude than that shown by the other spectrum lines. A probable explanation of the phenomenon is that which postulates the interposition between us and the binary of a cloud of non-luminous calcium, though there are some difficulties in this hypothesis. The following list gives the binaries which show this phenomenon; those marked with an asterisk are the cases where the II and K lines give different velocities from the other lines (data taken from Campbell's Third Catalogue of Spectroscopic Binary Start, in press).
It is perhaps not especially remarkable that all the objects of this type thus far recorded should be spectroscopic binaries, in view of the fact that such a difference of velocity would be most easily discovered and investigated in this class of objects. The H and K lines are so far to the violet that they are not employed in most of the high dispersion radial velocity surveys which have been made to date, but it would appear to be a fruitful field of investigation