I. Dans le domaine de l'astrométrie avec ses applications pratiques, ainsi que dans les problèmes de l'astronomie stellaire concernant la structure et éevolution de la Galaxie, l'étude des mouvements propres des étoiles tient une place importante.

La méthode photographique permet de déterminer avec une grande exactitude les mouvements propres d'un nombre enorme d'étoiles, y compris les plus faibles. L erreur systématique essentielle, dont on doit tenir compte en ce cas, est l'erreur dépendant de la magnitude de l'étoile.

The programme of proper motions with the 20-inch astrograph of the Lick Observatory was conceived and planned by Dr W H. Wright, formerly Director of the Lick Observatory. Not only did he formulate the work in considerable detail but he obtained the grant to purchase the telescope and planned it as a precision instrument admirably adapted to its purpose. A paper published in the Proceedings of the American Philosophica Society, Vol. 94, no. 1, 1950, describes the programme as formulated by Dr Wright.

In the spectrum of the fifth-magnitude peculiar A star HD 124224 the absorption lines of He 1 vary in strength with a period of only 0-52 day. The lines of Si 11 are always much stronger than in normal stars. They vary out of phase with the helium lines, but in phase with a group of unidentified lines that seem generally to occur in the so-called ' silicon stars

The lines of this spectrum are shallow and wide, and radial velocity determinations are inaccurate on this account. The hydrogen lines indicate no velocity changes, but He I 4026 is displaced to the violet before maximum strength, and to the red after. The silicon lines, on the other hand, are probably displaced in the opposite sense. These displacements, together with the diffuseness of the lines, strongly suggest that we have to do here with a star that rotates in half a day about an axis perpendicular to the line of sight. If this is so, the helium lines are produced preferentially near the approaching limb before maximum strength, and near the receding limb after maximum strength.

In Heidelberg photo-electric measurements have been made on the spectrum variables ι Cassiopeiae, 73 Draconis, γ Equulei, χ Serpentis, є Ursae Majoris, BD + 33° 1008 and others. According to our results all of these stars are variable with an amplitude of 0.01 magnitude and period equal to that of the spectral variability. Mr Bahner has carried out some further spectrophotometrical measurements of the line intensities in the spectra of ι Cassiopeiae, χ Serpentis and є Ursae Majoris. These measurements have enabled him to determine the phase relationship between the total light and line variations. The magnetic-field variable HD 133129 is likewise variable in magnitude with a small amplitude. The period, which might be several days long, could not as yet be determined with certainty.

The RW Aurigae stars form a class of variables that has recently been recognized, notably through the work of C. Hoffmeister. Members are assigned to this class solely on the basis of their type of light variation; specifically, through the possession of rapid, non-periodic fluctuations in brightness. The spectral type and luminosity play no part in the classification. Since the prototype, RW Aurigae, is itself a 't Tauri star', some misunderstanding has arisen as to the relationship between the RW Aurigae variables on one hand, and the T Tauri stars on the other. It is important to note that assignment to the T Tauri class is based entirely upon spectral criteria, in particular, the possession of a characteristic emission-line spectrum; the nature of the light variation (because the great majority, if not all, of the T Tauri stars are variable) is not considered in making the assignment.

1. Le but des recherches actuellement en cours est l’interprétation, aussi complèt que possible, du spectre des céphéides. A chaque phase de sa pulsation, l'atmosphèr d'une céphéide peut être représentée par l'atmosphère d'une certaine étoile, de températur effective et de gravité à déterminer de façon que son spectre soit identique à celui de 1; céphéide. Pour déterminer, dans chaque phase, les valeurs de Te et g, nous avons calcul un certain nombre de modeles d'étoiles de types F et G, géantes et supergéantes, corre spondant aux valeurs suivantes de Te et g:

Certain absorption lines in the spectrum of W Virginis have been found to be double during a short phase interval centred approximately at light maximum. They behave as if a sequence of velocity variation starts abruptly with the appearance of lines showing high velocity of approach, just before light-maximum and while the lines of the preceding sequence showing high velocity of recession are still present. As the sequence continues, the lines, which at first showed high velocity of approach, move longward until they in turn show high velocity of recession. They do not disappear until the next light-maximum, by which time another velocity sequence is well advanced. The duration of a velocity sequence is thus from shortly before one light-maximum until shortly after the next.

The profiles of the absorption lines in cepheids (their width, asymmetry, and doubling) have to be interpreted in terms of the motions throughout the reversing layer. To cover the entire volume of the reversing layer one has to integrate both over the surface of the visible hemisphere and throughout the depth of the reversing layer. The effect of the surface integration has already been discussed in this symposium by Dr Savedoff. Here will be discussed the effects of the depth integration.

L'étude de deux spectres à grande dispersion de ηAquilae (Phases: 0,013 et 0,556) et de ζ Geminorum (Phases: 0,017 et 0,55) mis grâcieusement à la disposition de l'un de nous par l'Observatoire du Mont Wilson a permis de mettre en évidence une relation entre les vitesses radiales et les hauteurs chromosphériques de Mitchell des raies correspondantes ou leurs intensités résiduelles ou, dans le cas des raies de Fe 1, leurs potentiels d'excitation.

Line profiles were obtained from spectra of RR Lyrae taken with the 100” coudé with a dispersion of 10 A./mm, Profiles from two plates taken at maximum radial velocity and from two plates taken at minimum radial velocity were oppositely asymmetric as was predicted by the hypothesis of radial pulsation. A numerical measure of the change of the asymmetry was obtained which yielded Δa= +0·22 ± 0·017 for these profiles.

A modified Baade procedure, applied to Struve and Blaauw's determinations of radial velocity and of spectral type and to Walraven's light curve yielded for this star a median radius

The cluster-type variables deserve particular attention since they are connected with two problems of fundamental importance: the phenomenon of pulsating stars and the existence of different subsystems or populations of stars in our own and other galaxies. On the other hand there are some obscure points in our understanding of the physical and dynamical properties of this group of stars as, for instance, their failure to give a period-spectrum relation or to obey the proper period-density relation. With this in view Dr Struve suggested to me during my stay at the Yerkes and McDonald Observatories to undertake a comparative spectrophotometric investigation of a series of these stars.

AI Velorum was discovered by E. Hertzsprung(1) in 1931 on plates taken at Johannes-burg; he found it to be an RR Lyrae variable with a period of od·1116, but later on F. Zagar(z) and A. van Hoof (3) showed that the light curve is very irregular. [G. Herbig(4) found that more or less the same kind of irregularities are present in the radial-velocity curve as well.

It is remarkable that bright hydrogen lines, requiring 13 volts for their production should appear in the spectra of objects as cool as long-period variables. The reversing layer and the photosphere, with temperatures of about 2000°, have nothing like the necessary amount of energy to offer. The curious irregularities in the Balmer series do not yield the interpretation, it is now clear that they are due to absorption by overlying gases. This fact indicates one important item, namely, that the origin of the bright line is not in an extended outer atmosphere, but in a level lying near, and even partly below the photosphere.

The presence and behaviour of the emission lines in the spectra of the long-period variable stars pose one of the most interesting problems yet unsolved in astronomical spectroscopy.

There is no time in a ten-minute paper to go into detail. Let me say, though, that the primary question is ‘Where and how, in the atmospheres of the cool late-type giants of spectral types M, N and S which make up the long-period variables, can there exist the source of energy for production of the observed emission lines?’

The measurement of recent high-dispersion spectrograms of Mira Ceti makes it possible to review and revise some of the conclusions published in 1926 in regard to the atmospheric motions of this well-known long-period Me variable star.

Spectrograms of χ Cygni have been obtained with the Mills Spectrograph of the Lick Observatory. For the purpose of spectrophotometry, the intensity spots have been given to all the plates using the Mills sensitometer. These spectrograms have been traced by the photo-electric microphotometer at the Yerkes Observatory, and equivalent widths of nearly 215 atomic and molecular lines have been derived.

The curve of growth was constructed by two methods. One is to make use of the observed values of the transition probabilities. The other is to make use of the theoretical values assuming LS coupling. Although Wright has used both values for constructing a single curve of growth, we have used them separately.

Element number 43, now called technetium, has had a curious history. As a hypothetical element, when the implications of the periodic table were first realized, it was called eka-manganese after its chemical homologue. In 1925 three German chemists announced its detection by means of lines in the X-ray spectrum, and gave it the name masurium; but their discovery has not been confirmed and is now considered erroneous, although their parallel discovery of rhenium, element number 75, was valid.

Wave-lengths measured on coudé spectrograms of three long-period variables near maximum light allow an extreme S-type star (R Gem) to be compared with two M-type stars (R Hya and R Leo) in the AA5600-6500 region. The known bands of TiO, VO, and ScO, which are strong in this part of the spectra of the M stars, are too weak to be measured in R Gem. Conversely the bands of ZrO, which dominate the spectrum of R Gem, are extremely weak in the M stars. The only bands which are prominent in all three spectra are those of YO, and the intensity of the ƛ6132 band of this molecule appears to be a sensitive means of interpolating between types M and S.