Images de page







Before attempting to describe the observations which are the subject of this paper I may be permitted to give some account of the manner in which, from time to time, direction has been given to the work to which they relate; for the aim with which that work was begun has been so altered by factors developed by the investigation itself that the purpose toward the end was quite different from that with which it was originally undertaken.

In a former paper1 I published the wave-lengths of some of the brighter nebular lines. These wave-lengths had been measured for the purpose of investigating the numerical relations between the positions of certain features of the bands in Nova Persci No. 2 and the positions of the nebular lines. At that time it was planned to extend the measures to the fainter nebular lines, but other work claimed my attention for a very considerable period, and it was not until about ten years later that the observations were taken up as a sort of stop gap in another investigation, to be fitted in as occasion offered, when one or another of the necessary instruments was available. The observations naturally centered on a few of the bright nebulae which supplied in their spectra all of the known lines, the object being not primarily discovery but rather added accuracy in the measurement of wave-length. This fact accounts for the comparatively large number of observations of three or four of the brightest objects. High and low dispersion spectrographs were used as occasion demanded, one series of plates being secured with a grating. The observations were believed to be complete in the summer of 1914, but the last plate of the series, a long exposure on the nebula N.G.C. 6572, revealed the fact that the center of that object emits the radiation 4686A as a band having a width of about fifteen Angstroms. This line is known to be common to both the nebulae and the Wolf-Rayet stars. In the nebulae it is normally narrow like the other nebular lines and in the Wolf-Rayet stars it is a broad band. Its occurrence as a broad band in a nebula therefore indicated a relationship between the nebulae and this group of stars. The relationship referred to is an important one and is discussed at some length in Chapter III. Further observation showed the presence of four additional bands characteristic of the Wolf-Rayet spectra. The observation appeared to be of sufficient importance to warrant the examination of other nebular nuclei, a work which was undertaken and which constitutes what I may term the second division of the investigation. A detailed account of this phase of the work will be found in later pages. At present I should like to draw attention to the necessity imposed by observations of this sort of accurately following the object under observation, in order to insure that the spectrum of the nucleus will be properly isolated from that of the rest of the nebula. In ordinary observing with an astronomical spectrograph the object is allowed to "drift" so as to furnish long and easily measurable spectral lines, and this practice has in the past undoubtedly been responsible for the concealment of valuable information with respect to the spectra of nebulae. When accurate following was resorted to the nuclei of quite a number of planetaries were satisfactorily photographed, and in addition a number of interesting facts were developed with regard to the localization of lines in some of the nebulae. One of the most striking of these is the peculiar restriction of the radiation 4686A to the inner areas of certain objects, resulting in a shortening of the spectral line. The phenomenon had, unknown at the time to me, previously been observed by Max Wolf in the ring nebula in Lyra, The shortening is more pronounced in some nebulae than in others, and the behavior of 4686A is shared by other lines, singularly enough by the members of a group in the immediate vicinity. In an attempt to systematize the data resulting from these observations a number of spectra were arranged in an order determined by the length of 4686A2 and other lines, and a sequence was obtained in which spectra lying in juxtaposition closely resemble each other while between the extremes there is little in common other than the hydrogen lines and 3727A.3 The scries might have been completed without a considerable hiatus by adding the spectrum of a Wolf-Rayet star without surrounding nebulosity.

i L. 0. Bull., 1, 153, 1902.


Whatever significance this exhibit may have, it seemed at the time to be suggestive of an order of nebular development, and directed attention to the necessity of a study of the distribution of line intensity throughout the nebulae if we are to understand more fully the relationship between these objects and the stars which appear to be forming within them. With respect to such an investigation the method of long and short lines has the obvious limitation of measuring the distribution only along the diameter or other line traversed by the slit, and some form of slitless spectrograph or prismatic camera seemed more suited to the requirements of the proposed observations. It chanced that at about this time I had in hand the re-design of a quartz spectrograph for the Crossley reflector, and this instrument was therefore so constructed that it could be used either with or without a slit. Its performance as a slitless instrument has been most satisfactory. Monochromatic images are recorded from the ultra-violet to H/3 with practically the same faithfulness to detail that characterizes a direct photograph. Figure 2 of plate L illustrates to advantage the definition obtainable with it under the best conditions. The diameters of the slightly elliptical hydrogen rings recorded on this plate are about 2" by 3". The instrument has proved most effective in the observation of the distribution of the various lines throughout the nebulae. This phase of the work I have termed the third division. It occupied, almost exclusively, the observing time during the years 1916-17.

To recapitulate, the parts into which the work has divided itself are three in number:

1. The measurement of wave-lengths and the intensities of nebular lines.

2. The study of the nebular nuclei.

3. The investigation of the distribution of nebular radiations throughout the nebulae. The first chapter is devoted to a description of the observations, the second, third, and fourth

to the aspects of the work indicated by the above divisions, while the fifth concerns itself with a number of miscellaneous results.

Acknowledgments are due to Director Campbell for placing at the disposal of this research the necessary instrumental equipment and for the generous provision, by purchase, of optical accessories as the need for them developed. The labor of preparing the observing list was facilitated through the privilege of consulting the extensive series of photographs of planetary nebulae secured by Dr. H. D. Curtis. In the computation of the effect of atmospheric absorption on the nebular lines assistance has been rendered by Miss Mary Heger of the Berkeley Astronomical Department of the University of California. Miss Adelaide Hobe and Magister Holger Thiele of this Observatory have aided in the preparation of the manuscript for publication. The writer desires to express his gratitude for this encouragement and assistance.

2 Froc. Nat. Acad. Set. Am., 1, 590, 1915.

3 The sequence referred to is that indicated in plate XLV with the addition at the lower end of the spectrum of N.G.C. 40. Better reproductions of individual spectra are found in some of the other plates.



The following descriptions of the observations of the spectra will, it is believed, be found self-explanatory. Included with the account of each object is a table in which are entered such details of observation as are likely to be required. It has seemed best not to burden the pages with data which will not be used, but if the process of abbreviation has been too thorough supplementary information can be secured by communicating with the Lick Observatory. The letter under the head of Instruments refers to the description, to be found in the appendix, of the particular spectrograph used in the observation.

A majority of the observations were made with the quartz slitless spectrograph (g), and the frequent references to the "images" refer to the monochromatic images given by each line in the spectrum of the nebula.

The dates given are in Greenwich Mean Time.


[ocr errors][ocr errors][merged small][merged small][merged small][ocr errors][ocr errors][ocr errors]

















































[ocr errors][ocr errors][ocr errors][ocr errors][ocr errors][merged small][merged small][merged small][merged small]

The system used in assigning intensities to the nebular lines is explained on page 242. There are two scales, depending on whether the optical parts of the instrument used are of glass or of quartz, Ig corresponding to glass and Iq to quartz.

Two systems of classification are given hi the observing list: the well known one developed by Miss Cannon4 and a tentative system, described in Chapter V, which the writer has found to be useful.

N.G.C. 40. a = 0" 7m6; 8 = +71° 32'

Expoisure Region of

Date Plate No. Instrument time Spectrum

1914 Dec. 29.76 685 i 6" 0m blue-violet

1915 Mar. 3.66 687 i 1 25 blue-violet
Mar. 18.71 689 i 3 0 blue-violet

1916 Aug. 31.92 753 g (slitless) 4 0 ultra-violet

A large dim nebula, nearly circular in outline, brightest at the outer edge, and showing considerable structure. There is a comparatively bright central star, or nucleus.

The character of the spectrum is very well indicated in plates XLIV, figures 3, 4, 5, and L, figure 3. 3727A is the brightest line and forms the most conspicuous of the monochromatic images in plate L, figure 3. Five of the hydrogen images are also shown. These and the bright image are alike in size and structure. N, and N2 are not recorded on this plate, but appear as exceedingly faint lines on the strongest of the slit spectrograms. They seem to be shorter than the other lines, but they are so faint this is uncertain. The slit spectrograms were made with a 5-inch camera, and are on a very small scale.

The relative intensities of the images are:

line Iq line Iq

H/3 51 He (15)

H7 49 Hf (10)

US 25 3727 97

The nucleus is exceedingly interesting. It furnishes a Wolf-Rayet spectrum of the most unmistakable type. It was first observed by Paddock, who found the broad band 4650A to be present. This is the brightest radiation in the entire spectrum of the nebula. Prolonged exposure brings out numerous other nucleus bands. The lumpy appearance of the nucleus spectrum is due to the presence of these. Figure 5 of plate XLIV is a cylindrical enlargement which exhibits the details to better advantage. Along the top of the spectrum are indicated the positions of the principal Wolf-Rayet bands given by Campbell. The correspondence is more definitely shown in table 2, where the second column gives, with a single exception, the wavelengths of bright bands in B.D. -4-30° 3639, as measured by the writer. The noted exception is 4334A, which is a band found in other stars of this class. The reason for selecting B.D. -|-30° 3639 for this comparison is that the wave-lengths are better determined in this star than in any of the others, and include the ultra-violet.

Paddock has photographed some of the brighter bands, using several times the linear dispersion of the present observations. He finds that the narrow nebular hydrogen line H/? broadens in the nucleus to a wide band, and the same is doubtless true of the other hydrogen lines. This behavior is analogous to that of 4686A in N.G.C. 2392, which furnishes another of the few cases that have come under my observation of the occurrence in the same object of a nebular line and nucleus band of identical origin. The phenomenon would appear to be of comparatively rare occurrence. *

*Ann. H. C. O., 76, 19, 1914-16.

N.G.C. 40. Wave-lengths


* These wave-lengths are determined by extrapolation and are therefore uncertain.

f These lines are so faint that it is impossible to describe them accurately. They are probably shorter than the hydrogen lines, and NL impresses one as being strengthened in the nucleus. A longer exposure would be required to settle these points.

t The astronomers of Harvard College Observatory have called attention to the spectrum of this nebula (H.C.O. Circ., No. 98, 1905; Ann. H.C.O., 76, 40, 1916) as possibly being intermediate between a nebula and a fifth type (Class O) star. "Its spectrum resembles the fifth type, its appearance is that of a nebula." The spectrum is reproduced in plate 1 of the second of the references just given.

The relationship of the nebulae to the stars is such an important one that it seems appropriate to examine this statement somewhat critically, particularly in view of the vagueness of the context from which it is taken. The spectrum is shown in the

Sublication referred to in juxtaposition with the spectrum of N.G.C. 7662, and, aside from the weakness or absence of Ni and 2, there is little to differentiate it from the spectrum of a nebula in which 4686A (which is the wave length assumed, incorrectly as Paddock has pointed out, for the brightest spectrum band) is unduly strong. Furthermore, this critical band, which is actually broad, appears narrower than the really monochromatic nebular lines. As indicated in the present text the spectrum has stellar characteristics, but they are due to the involved star and not to the nebula, which is quite a large one. An objective prism spectrogram of the Orion nebula would present a combination of stellar and nebular spectrum, analogous in certain respects to the one under discussion, yet we should hardly venture to say of that object that its spectrum resembles one of a Class B stttr, while its appearance is that of a nebula.. In any event attention should be directed to the fact that the stellar characteristics pertain to the stellar nucleus, and not to the comparatively large nebula as the context of the reference appears to indicate.

A valuable account of certain features of the spectrum of N.G. C. 40 is given by Paddock (Lick Oba., Butt., 9, 30, 1915).

« PrécédentContinuer »