1942: Astronomy

Astrophysical Observatory at Tonanzintla.

Many an astronomer has wished that he might have an observatory so located that from it the entire Milky Way might be observed. Mexico's new astrophysical observatory, located at Tonanzintla, is at latitude 19 degrees north and alone among important modern observatories commands a view of the whole Milky Way. This observatory was formally dedicated on Feb. 17, by President Manuel Avila Camacho of Mexico. The dedication ceremony was indeed a notable occasion and was attended by many distinguished astronomers of Mexico, the United States, and Canada. Following the dedication ceremonies an Inter-American Scientific Conference was held, the appropriate central theme of which was the structure and constitution of the Galaxy and its relationship to other galactic systems. The new observatory is equipped with a 27-31-inch Schmidt telescope and several smaller instruments. A very ambitious program of research has been outlined for this new observatory and its able staff of Mexican astronomers.

Solar Corona.

Many observational methods and types of instruments have been applied in recent years to the study of the solar corona, that intriguing outer gaseous envelope of our sun. During the year two pieces of work have added further data to our knowledge of the corona. Both are the result of observations made of total eclipses of the sun. Menzel and Petrie, from a study of spectra made in Russia at the 1936 total eclipse, have determined the wave-lengths of seven new lines in the spectrum of the corona. The determination of accurate wave-lengths of these new lines is the first step in their identification. When these new lines are identified we shall know more about the physical conditions in the corona under which these lines are produced.

The other piece of research, carried out by two Chinese astronomers at the total eclipse of Sept. 21, 1941, is of particular interest because of the very simple equipment used. The total light emitted by the corona is an important datum and very difficult to measure. Furthermore, there is some evidence to indicate that the total amount of light emitted may vary. Any reasonably reliable measure of the total emitted light is therefore valuable. Chang and Li used an ordinary Weston exposure meter, a 9 cm. camera lens, and a sensitive galvanometer to make their measurements. After carefully calibrating the exposure meter for the range of intensities involved, they compared the total light received from the corona during total eclipse with the amount of light received from the full moon. They found that at this eclipse the total light received from the corona was 0.39 that received from the full moon.

Asteroid Eros.

The asteroid Eros, the most observed of all the asteroids because of its usefulness in determinations of the solar parallax, has long been an interesting object for another reason. Its light varies in a period of approximately 5^h 17^m. This has been attributed to its irregular, elongated shape as it rotates about its shortest axis. Recently Roach has made very accurate photoelectric observations of this asteroid on six nights. From these he has obtained a composite light curve which is the most accurate yet obtained. Roach then has examined the type of theoretical light curve which will best represent the observed curve. He finds that if he adopts Lambert's law of reflection, and assumes a geometrical figure with its three dimensions in the ratio of 1:4:13, he obtains a theoretical curve which is a good fit for the observed curve. If he assumes an idealized case of a three-axis ellipsoid, the ellipsoid would have the dimensions 35 km., 15.6 km., and 7.2 km. This theoretical curve fits the general run of the observations remarkably well. There still remain, however, small departures of the observed light from this theoretical curve. These departures can best be explained by assuming that the geometrical figure of the asteroid is irregular. This is a conclusion previously arrived at, but from much less accurate data than presented in Roach's extremely accurate light curve.

Comets.

Seven comets were discovered during the past year. After preliminary orbits had been computed it was found that five of the seven 'discoveries' were actually rediscoveries of previously observed comets. One of these comets was first discovered in 1867. The two comets which proved to be actually new to observers were both discovered photographically at the Harvard Observatory by Whipple.

Novae.

One of the most interesting discoveries of the year was made on Nov. 10, when a nova was discovered by Dawson in the constellation of Puppis. With rare good fortune, this nova was discovered before it reached maximum brightness. This made possible the detailed study of its spectrum before maximum light, which will greatly aid in the interpretation of this star's 'life history' during the time that expanding shells of gas were blown off. Nova Puppis reached maximum brightness on Nov. 12, at which time it was about 0.5 magnitude and a conspicuous object low in the southern sky for observers in mid-northern latitudes. By Nov. 30 it had faded to 5.5 magnitude and was near the limit of naked-eye brightness.

Records of Chinese and Japanese astronomers of the eleventh century describe the appearance of a 'guest star' in the constellation of Taurus in the year 1054 ad. Recently published translations of these records by Duyvendak aid in locating the position of this object in the sky. Furthermore, a study of these translations seems to clearly indicate that this 'guest star' was no ordinary nova but a supernova — as a matter of fact one of the brightest supernovae ever recorded. This new evidence is of particular interest because about twenty years ago Lundmark and Duncan suggested that this 'guest star' and the well known Crab Nebula in Taurus were probably the same object. The position is as nearly the same as can well be judged from the early records. It is known that bright novae do frequently show nebulosity around them some years after the outburst — Nova Aquilae 1918 and Nova Herculis 1934 are two such examples. Furthermore, from a series of photographs of the Crab Nebula made over a period of years, the rate of expansion can be approximately determined. Using this rate of expansion, and the present dimensions of the nebula, the approximate time of outburst which produced the nebula can be computed. This turns out to be about 1100 ad — not too great a discrepancy since the rate of expansion need not have been constant. Mayall and Oort have recently made a careful study of all of the available evidence. They conclude that the Crab Nebula was indeed produced by the outburst of the supernova observed by the Chinese in 1054 ad.

Nebulae.

Baade has recently photographed the Crab Nebula in several regions of wave-length. He finds that the nebulosity is composed of two parts. Photographs made in the Ha (red) region of the spectrum show an outer system of filaments of a most beautiful and varied structure, whereas photographs made in the blue show an inner mass of amorphous structure. He finds that it is these outer filaments which give rise to the line spectrum, the inner amorphous mass producing a continuous spectrum. It is clearly apparent that photographs of other nebulae in several wave-length bands will aid greatly in the interpretation of these objects.

It has long been known that hydrogen is the most abundant element in the atmospheres of most stars. Recently a very surprising discovery about the abundance of elements in a stellar atmosphere has been announced by Unsold. He has made a thorough study of line widths in spectra of the hot star Tau Scorpii. From this study he finds that hydrogen is the most abundant, and neon the second most abundant element in the atmosphere of this star. To find that neon, the gas used in the tubes of many advertising signs, is so abundant in a stellar atmosphere, though so rare on earth, is indeed surprising. This discovery will certainly have an effect on our interpretation of the conditions existing in stellar atmospheres. It will be most interesting to see if tests of the abundance of neon in the atmospheres of other stars bear out Unsold's findings in the case of Tau Scorpii.

It is known that many of the early type, hotter, stars are surrounded by gaseous shells. Struve has made a thorough study of the available observational data on these stars. He finds that in an ideal case there are three layers which are observable. The lowest layer is a stationary reversing layer which produces the ordinary absorption line spectrum. There is an intermediate layer which is stationary and produces an a Cygni type of spectrum with exceedingly sharp lines. Above these two layers is an expanding layer of varying optical thickness which produces a P Cygni type of spectrum with bright lines. Struve suggests that the two outer layers may originate from the rapid rotation of the lower reversing layer, and that the structure of these layers is a function of the radiation pressure due to the Lyman α flux of the shell. This investigation will serve as a guiding stimulus for much further work on the many stars with gaseous shells.

Wyse has made a detailed study of the spectra of ten gaseous nebulae. Nine of these are planetary nebulae, the other the great Orion nebula. This study involved the accurate measurement and identification of many spectral lines — about 270 spectral lines were measured and some sixty per cent of these were identified — all of the measured lines being produced in the nebulosity and not in the exciting stars. From this thorough study Wyse concluded that the chemical composition of all of the planetary nebulae is essentially the same. The relative abundance of elements is likewise similar and there appears to be no important difference in chemical composition between these nebulae and the diffuse nebulae such as the one in Orion. Such differences as appear in the lines measured can be attributed solely to differences in the physical conditions existing in the various nebulae. There is no evidence of a division of these nebulae into two sequences, carbon and nitrogen, as had previously been suggested.

An extensive investigation of the rotation of Messier 33, the spiral nebula in Triangulum, has been carried out by Mayall and Aller. This is the second great spiral to be so investigated, Babcock having made a similar study of the Andromeda Nebula, Messier 31, a few years ago. Mayall and Aller made radial velocity observations of twenty-five condensations which show bright-line spectra. These condensations were well distributed over the spiral. In addition, several long exposures were made which record the spectrum of the central unresolved nebulosity. As a preliminary to the determination of the velocity of rotation, the velocity of the system was determined to be —167 km/sec, and the inclination of the principal plane of the nebula to the line of sight was found to be about 33 degrees. Using these results, and making the assumption of simple circular motion in the plane of the spiral, rotational velocities for the individual condensations were obtained. These results were then combined to give a curve of rotational velocities. This curve shows that the main body of the spiral, which extends to 18 ft. from the center, apparently rotates very much like a solid body. In other words, in this part of the spiral the rotational velocity increases fairly uniformly with distance. On the other hand, the outer part of the spiral extending between 18 ft. and 30 ft. from the center shows the rotational velocity decreasing with distance from the center as would be characteristic of a planetary system. The transition from one type of rotation to the other occurs at a distance of approximately 1,000 parsecs from the center, the rotational velocity in this region reaching its maximum value of 120 km/sec. When these results are compared with similar ones for the neighborhood of the sun in our own Galaxy, we find encouragement for the current hypothesis that the sun is actually located at a very considerable distance from the center or main body of our own stellar system.

1941: Astronomy

Solar Discoveries.

Studies of the sun are of especial importance for we can hope to learn much more about it than we can about distant stars. Whatever details we learn about the sun aid us in interpreting the 'cues' and 'hints' from the far more distant stars. Any detailed study of the solar atmosphere requires, among other data, a knowledge of the intensities of absorption lines in the solar spectrum. For many years spectroscopists have had to depend on estimated intensities of these lines, primarily on the estimates of Jewell and Rowland. The Rowland scale of intensities has been carefully calibrated, and the intensities of a few hundred of the stronger lines have been measured, but there has remained an urgent need for a precise photometric atlas of the Fraunhofer lines. This need has now been met by a prodigious volume prepared by Minnaert and his colleagues at Utrecht. This work, Photometric Atlas of the Solar Spectrum, contains 174 pages (over 300 feet) of microphotometer tracings of the solar spectrum on a direct-intensity scale, covering the spectral region λ3332 to λ8771. The scale of the tracings is approximately 2 centimeters per Ångstrom so that wave-lengths, based on the Revised Rowland Table, may be read from the curves with an accuracy of approximately 0.01 Å. Measures of central intensities taken from the Atlas are correct to about 5 per cent.

The great majority of the lines in the spectrum of the solar disk have now been identified. There has, however, remained one group of lines which defied identification. These are 22 lines in the spectrum of the solar corona. That these could not be produced by any unknown chemical element has been accepted, but it remained for Edlen, from his investigation of the spectra of highly ionized atoms, to determine the identity of the brightest of these lines. He found that two of them are produced by highly ionized calcium atoms, seven by highly ionized iron atoms, and six by highly ionized nickel. Since these fifteen lines which are now identified by Edlen produce 97 per cent of the corona's total radiation, his identifications are of particular importance.

Facilities for the detailed study of solar prominences at the McMath-Hulbert Observatory of the University of Michigan have been greatly improved during the year. A new spectroheliograph, which can be operated in conjunction with the tower telescope, now gives the rate of motion of prominence material in the line of sight. Since this instrument and the spectroheliokine-matograph on the tower telescope can be operated simultaneously, the McMath-Hulbert observers can now obtain a simultaneous record of the motion of the prominence material in three dimensions and thus can obtain a remarkably accurate picture of these puzzling solar outbursts. To test the reliability of the methods used in analyzing these data, Mohler has constructed models of prominences based on his analysis and has then photographed the models and compared these photographs with those of the actual prominence. The similarity is so striking as to leave little doubt of the correctness of the methods used.

One of the important factors which must be considered in the analysis of light curves of eclipsing variables is the degree of darkening at the limb for these stars. This is difficult to determine for very distant stars, but should be relatively easy to determine for the sun. This has been done with high accuracy except for the region very near the limb. The solar image in this region is so distorted by the effects of the earth's atmosphere that only during the crescent phases of a total solar eclipse can accurate measures be made. Three Dutch astronomers, Messrs. Ferwerda, Uitterdijk, and Wesselink, have made such observations by means of a very ingenious photographic device, using aluminum-coated spectacle lenses and amateur movie cameras. From hundreds of observations they found that, whereas over most of the solar disk the brightness varies as the cube root of the distance from the limb, near the limb the brightness varies as the tenth root of the distance. They also found that the color remained constant, which suggests that the layer of the sun very near the apparent surface must be a nearly isothermal one.

The distance from the earth to the sun is a standard to which all other measurements in both the solar and stellar systems are referred. An accurate determination of this distance is therefore of interest and importance. One of the most accurate methods of determining this distance is the simultaneous observation from different stations of an asteroid which comes close to the earth. Eros, at the opposition of 1931, came within 16,000,000 miles of the earth. Since it was at that time a bright telescopic object it offered an excellent opportunity for a new determination of the solar parallax—a measure of the sun's mean distance in terms of the equatorial radius of the earth. Some thirty observatories throughout the world made series of observations of Eros at that time. Recently Jones, who has analyzed these thousands of observations, has announced the results of this great cooperative program. He finds, from all right ascension observations, the value 8'.7875 ± 0'.0009 for the solar parallax, and from all declination observations, the value 8'.7907 ± 0'.0011. The mean value is thus 8'.790 ± 0'.001, which is somewhat smaller than the mean of previous values and corresponds to a distance of 93,003,000 miles. Using this new mean value for the solar parallax Jones has obtained from the observations a new value of the ratio of the mass of the earth to that of the moon, namely, 81,271 ± 0.021.

Planets.

New determinations of the masses of Pluto and of Neptune's satellite have been made during the year. The new value for the mass of Pluto resulted from a comprehensive comparison of Newcomb's theory for the motion of Neptune with the observations of this planet from 1795 to 1938, made by Wylie at the United States Naval Observatory. He found the mass of Pluto to be (0.300 ± 0.028) 10^-5 sun's mass. Since the mass of the earth is 0.301 x 10^-5 sun's mass, it appears that the two planets have practically identical masses.

The mass of Neptune's satellite, Triton, has been determined by Alden from observations made at Johannesburg. He found that the ratio of the mass of Triton to the mass of the system is 0.0013 ± 0.0003. The mass of Neptune is roughly seventeen times the mass of the earth, so that the mass of Triton is approximately 0.022 the mass of the earth, or 1.8 times the mass of the moon.

Comets.

There were six comets discovered during the year. Two of these discoveries were actually rediscoveries of previously known objects, whereas the other four were new comets, one being easily visible to the naked eye for observers in the southern hemisphere.

Variable Stars.

The number of discoveries of variable stars in recent years has reached so large a figure that the need for a catalogue of these stars giving essential data and references to the literature has become urgent. This need has now been met by Prager in a supplementary volume to his History and Bibliography of the Light Variations of Variable Stars which has just been published. This volume, which contains data for 3,592 variable stars discovered between 1931 and 1938, will be of inestimable value as a reference volume for all observers of these stars.

Among the interesting and important variable stars are the so-called cluster variables. The light variations of these stars are somewhat similar to those of the Cepheids, though the periods are shorter. Recent observations of these stars by Fath and by Schwarzschild show that there is a secondary period of light variation superimposed on the primary period. Schwarzschild has analyzed this material from the standpoint of pulsation theory and finds that these secondary periods may be higher modes or overtones of the fundamental periods. He finds that the periods of the cluster variables in Messier 3 can be divided into two groups and that the group with shorter periods comprises stars which pulsate in the first overtone of the periods of stars in the other group. In other words, for one group the fundamental period predominates while for the other group the first overtone of the fundamental period is predominant in the pulsation. This result is of much importance in studying the pulsation mechanism.

The determination of the distances of these clusters is fraught with uncertainty because of the absorption of their light by interstellar material. In order to avoid this difficulty it is important to investigate clusters in regions where the obscuration by interstellar material is either absent or relatively small. Cuffey has done this for three clusters in the direction of the constellations Monoceros and Canis Major, where the Milky Way is fairly uniform and richly populated. In this direction he finds relatively little interstellar obscuration to a distance of approximately 3,300 parsees, and he also finds a marked absence of extra-galactic nebulae from these fields. This absence of spiral nebulae would seem to indicate that there is considerable absorption beyond these clusters. This must mean that either there is a concentration of material near the boundary of our galaxy, or that there is a spiral arm containing clouds of absorbing material and extending to a distance greater than the supposed average radius of our galactic system.

Probably no other star in the sky has received so much attention from astronomers as has the third magnitude eclipsing binary, Beta Lyrae. It has proved to be one of the most puzzling of these interesting stellar systems. Irregularities in the light curve and variations in the spectra have both defied adequate explanation. During the past year several astronomers have tackled this problem in search of a solution, the most profitable of these studies being the investigations of Struve and Kuiper. The latter has suggested a model for this system, based on dynamical calculations and on the spectrographic results of Struve, which seems to explain many of the puzzling features of the system. The two component stars of this binary are so close to each other, as they revolve about their common center of gravity, that their atmospheres partially merge. Since the two components are of quite different density, though of nearly the same mass, atoms of gas lying between the two will be attracted toward the denser star. There will thus be set up a current in the atmospheres. According to Kuiper's model, most of this circulating gas will continue to move about the two stars, but some of it having higher velocity will escape from behind the denser star and form a great spiral of gas about the pair. This escaping material will give a disc of gaseous matter which lies in the plane of the orbit, and if this orbit plane lies almost in our line of sight, this gas will come between us and the stars and thus account for most of the observed spectral peculiarities.

Nebulae.

The distribution of mass in extra-galactic nebulae is one of the important data needed for the study of the dynamics of these huge stellar systems. Of interest, therefore, is the recent investigation of the spiral nebulae Messier 31 and 33 by Wyse and Mayall. These astronomers have made numerous radial velocity measures at varying distances from the center of each of these nebulae, and find that in each case the mass is widely distributed throughout the nebula. The marked contrast between this result and the obviously strong concentration of light near the center of the nebula suggests that there is no apparent correlation between the distribution of mass and of luminosity. Possibly through this investigation a reinterpretation of observations of the rotation of our galaxy may lead to a similar result concerning the distribution of mass in our own stellar system.

1940: Astronomy

Observatory Equipment.

Two observatories have added to their equipment for solar investigations during the year. The Harvard College Observatory has established a new high-altitude station in the Colorado mountains. This observing station, under Menzel's direction, has been equipped with a coronagraph of design similar to that devised by Lyot. The location of the station at an elevation of over 11,000 feet should provide the Harvard observers with exceedingly fine atmospheric conditions.

The McMath-Hulbert Observatory has placed in operation two new instruments for solar observation. A 24-inch reflector replaces the original 10-inch and a new 70-foot tower telescope with attached laboratory has been built. The addition of these new instruments makes possible the simultaneous observation of solar prominences in three dimensions and of the attendant energy output. The 70-foot tower telescope is to be used primarily for physical measurements, the attached laboratory being especially equipped for this purpose, leaving the 50-foot tower telescope and the new 24-inch reflector for studies of prominence motion. This new equipment and the novel observational techniques already developed at the McMath-Hulbert Observatory place this institution in the front rank of the world's solar observatories.

One important datum needed for a true understanding of the physical conditions which exist in our sun is an accurate determination of the amount of energy radiated by the sun at various wave lengths throughout the observable range of spectrum. Pettit at the Mt. Wilson Observatory, has made a study of this spectral-energy curve in the ultraviolet region of the spectrum, using a double quartz monochromator for his observations. After correcting for the effects of the earth's atmosphere, his observations give energy curves both for integrated sunlight and for the light from the center of the solar disk. These curves must then be corrected for the amount of energy cut off by the normal absorption lines of the solar spectrum, the so-called Fraunhofer lines. This is exceedingly difficult, for the determination of the true intensity of the continuous spectrum on which these absorption lines are superimposed requires the highest possible spectroscopic resolution. After correcting for these lines as far as possible, Pettit finds that the resulting energy curve in the ultraviolet is still quite different from that of an ideal black body. It is possible, however, that much of this discrepancy will be removed when more reliable corrections for the absorption of the Fraunhofer lines is possible.

Comets.

Five comets were discovered in 1940. Of these, two were previously known periodic comets which were rediscovered, whereas the remaining three were new objects. The comet which attracted most attention was that found on Harvard Observatory plates by Cunningham. This comet was approaching the earth when discovered, though still at a considerable distance. During the last month of the year it was visible to the naked eye in the very early evening, but soon disappeared from the view of northern observers because of its rapid southward motion.

The earth's atmosphere is an imperfectly transparent medium whose behavior shows considerable variation. It is therefore of much importance to the interpretation of astronomical radiometric observations that the amount and extent of the absorption produced by the earth's atmosphere should be determined with all possible accuracy. Adel and Lampland, at the Lowell Observatory, have made an extensive and careful investigation of the nature and amount of absorption produced by our atmosphere at various wave lengths in the infrared solar spectrum, their study covering the region from 5.5µ to 14.0µ. Their results will be of especial importance to planetary observers.

Meteors.

One means of determining the height above the earth's surface to which the atmosphere extends is the determination of the heights at which meteors appear. Numerous observers have noted apparent seasonal variations in these observed heights. McIntosh has made a study of all available data from the published real paths of sporadic meteors and finds a marked seasonal variation in the height of these meteors. The amplitude of this seasonal variation proves to be greatest in the lower levels of the atmosphere. McIntosh also finds that there is a similar seasonal variation in the width of the meteor zone. The cause of this variation lies apparently in the atmosphere itself, the greater height occurring in the summer due to heating and expansion of the atmosphere.

An important and interesting problem connected with the study of meteors is the question whether they are regular members of our solar system or come in from interstellar space. This question is of particular interest in the case of those very bright meteors frequently called fireballs. In 1868 Galle, the discoverer of Neptune, made a rather thorough study of the Pultusk fall of meteorites which occurred in that year. The orbit which Galle determined for this meteorite indicated that it entered the earth's atmosphere with a hyperbolic velocity; in other words that it had come in from interstellar space. Recently Wylie has made a study of this meteor and concludes that with one exception the data used by Galle were correct. This exception is the height of appearance of the meteor. Investigations of numerous fireballs by Wylie and by Watson have shown that the high velocities obtained by astronomers who use reports of the general public are frequently not real. The reason for this is that the adopted heights of initial appearance of spectacular fireballs, determined in more recent years, is less than that for the faint meteors, and furthermore there is very little variation in the well determined heights of initial appearance of fireballs which vary greatly in size and brightness. Wylie finds that if he adopts for the height of initial appearance of the Pultusk meteor a value commensurate with well determined heights for recent bright fireballs, this meteor turns out to have been a regular member of our solar system, with a 2.37 years period of revolution about the sun. It is possible that a number of meteors previously thought to have come in from interstellar space may actually have been regular members of the sun's family before being swept up by the earth.

Studies of the Moon.

A number of determinations of the temperature of the surface of the moon have been made during the past few years. Of particular interest are measurements of the change in the temperature of the lunar surface during a total lunar eclipse. Recent measurements of this sort have been reported by Pettit. His observations were made at the Mt. Wilson Observatory with a 20-inch reflecting telescope of short focal length and a sensitive thermocouple. Both a cover glass and a water cell were used to determine the temperature. A point near the center of the lunar disk was chosen for measurement. Pettit found that for this point the temperature dropped from 371° K to 200° K during the partial phases and then slowly dropped to 175° K during totality. From these measurements he found that the rate of radiation from the moon's surface was nearly proportional to the energy received by it from the sun. Using familiar radiation data and the known physical constants of lava, Pettit was able to determine the equivalent thickness of the radiating layer. An equivalent thickness of approximately one inch satisfied the conditions. Since solid rock is a relatively efficient radiator, Pettit's result indicates that the surface of the moon cannot be solid rock, but must be covered with some agglomerate such as volcanic ash or pumice.

Planet Venus.

Speculation concerning the possible existence of life on the planet Venus has continued over many years, encouraged to a great extent by the similarity of the size of Venus to that of the earth. Some years ago detailed investigations of the spectrum of Venus by Adams and Dunham showed that the planet's atmosphere contained a large percentage of carbon dioxide and that there was no trace of the presence of water vapor. One important datum necessary for a complete picture of the physical conditions existing on Venus is the surface temperature. Recently reliable laboratory measurements of the total radiation of gaseous carbon dioxide have been made. Wildt has made use of these measurements to determine an upper limit to the 'greenhouse effect' which is produced by the carbon dioxide in the planet's atmosphere. He finds from his study that the temperature of the sub-solar point on Venus is somewhere between 366° K and 408°K, the true temperature being probably closer to the higher value. Wildt concludes from this that the surface temperature is at least somewhat higher than the terrestrial boiling point of water.

When a mixture of water vapor and gaseous carbon dioxide is exposed to sunlight, a photochemical process takes place, with the synthesis of formaldehyde as a result. Although the atmosphere of Venus is now nearly, if not entirely, devoid of water vapor it is possible that Venus during its earlier history may have had a tenuous atmosphere in which both carbon dioxide and water vapor were present. Wildt finds no trace of formaldehyde bands on recent ultraviolet spectrograms of Venus. However, he points out that in a moist atmosphere formaldehyde would rapidly polymerize into a mixture of polyoxymethylene hydrates. This mixture is a white crystalline solid, and Wildt believes that the dense clouds in the atmosphere of Venus, since they cannot be water vapor, may consist of these solid polyoxymethylene hydrates. Astronomers at the Lowell Observatory have suggested that the exceptional brightness of the halo observed about Venus when it is between us and the sun may be due to 'dust' particles. Wildt's hypothesis is therefore of particular interest.

Publications.

Two new astrophysical monographs, sponsored by the Astrophysical Journal, have been published during the year. One, Spectra of Long-Period Variable Stars by Merrill, presents in concise form the results of the author's long study of these interesting stars. From a detailed discussion of the various changes observed in the spectra of these stars during their period of variation, and a discussion of measurements of total radiation and of light variation in different wave length regions, Merrill concludes that pulsation, such as observed in the case of Cepheid variables, cannot account for the observed light variations. He points out that at the low surface temperatures of these stars a relatively small change of temperature will produce a considerable change in the observable light, though the observed temperature changes are not sufficient to account for all the light variation. Another source of light variation is found in the considerable absorption of light in the spectra of these stars by bands of titanium oxide, and this band absorption will account for some though not all of the observed variation. A third possible source of light variation, suggested by Merrill, is the formation of a cloud or veil of either solid or liquid particles which are condensed from the gases of the star's outer atmosphere during certain phases of the light variation. It has been shown by Wildt that such condensation of compounds of low volatility is probable in the atmospheres of the coolest stars. It seems likely, from Merrill's discussion, that all three possible causes may contribute to the large range of light variation observed.

The other astrophysical monograph, The Masses of the Stars by Russell and Miss Moore, presents the results of the most extensive study of stellar masses ever made. All of the available data on visual, spectroscopic and eclipsing double stars which have been adequately observed have been included in this investigation. Of particular interest among the results obtained is the authors' conclusion that the relation between the masses and the luminosities of the stars may be represented by a simple linear function. The determination of this simple relation between mass and luminosity from so large an amount of data is of very great importance for the solution of many astrophysical problems. The monograph also includes a catalogue of the newly determined dynamical parallaxes of 2,529 stars.

Recent Investigations.

It has long been known that the component stars of a close binary system have a more or less ellipsoidal figure. The light-curves of short-period eclipsing stars show this very clearly. Several investigators have pointed out that the force of gravity at different points on the surface of such a star is different due to the unequal distances of these points from the center of gravity of the star, and that this gravity-effect will cause an unequal distribution of light over the surface of the star. Recently Sitterly has made a preliminary investigation of the effect of this unequal distribution of light on the observed light-curves of these stars. He finds that for the very close pairs this effect can be easily observed and that it must be allowed for in the determination of the photometric elements. Sterne has investigated the effect of this unequal light-distribution on spectroscopic observations and finds that when the gravity-effect is not allowed for, the orbital eccentricity for close pairs determined from these observations is too high. However, he finds that if proper allowance is made the corrected eccentricities agree with photometric results for the same stars, thus solving a problem which has greatly puzzled astronomers for some time.

During recent years a large number of stellar parallaxes, both those determined from trigonometric measurements and those determined from relative intensities of carefully chosen pairs of lines in stellar spectra, have accumulated. Russell and Miss Moore have made a detailed study of the errors of spectroscopic parallaxes, both systematic and accidental, based on the parallaxes of 1,140 stars for which reliable determinations by both methods were available. Certain deviations from a linear correlation between the mean reduced spectroscopic and trigonometric parallaxes are apparent but these, they find, are no larger than can be accounted for by ordinary accidental and sampling errors. It is suggested that most of the apparent errors of spectroscopic parallaxes may be accounted for by differences in the relative abundance of hydrogen and the heavy elements, and by differences in the magnitude of the contribution made by negative hydrogen ions to the general opacity of stellar atmospheres. These investigators find that after appropriate corrections have been applied, the probable error of a corrected spectroscopic parallax is ±21 per cent for main-sequence stars and ±16 per cent for giants.

Numerous studies of the magnitudes, distances, velocities, and structure of extra-galactic nebulae have been made during recent years. Little attention, however, has been paid to the colors of these objects. Of considerable interest, therefore, is a recent study of the colors of seven of these galaxies by Seyfert. Using red- and blue-sensitive photographic plates he determined both red and blue magnitudes of various features in each of these nebulae. The observed differences between red and blue magnitudes determines so-called red indices which are effectively indications of color. Seyfert finds that these nebulae exhibit a general background intensity of light of the same color as the nucleus, which decreases uniformly outward. The spiral arms, however, and the nebulous condensations observed in them are found to be considerably bluer than the nucleus.

1939: Astronomy

New Equipment.

One of the outstanding events of the year in the astronomical field was the dedication, May 5-8, of the W. J. McDonald Observatory. Located on Mount Locke, near Fort Davis, Texas, it is to be operated cooperatively by the Universities of Texas and Chicago. In the words of its president-elect, Dr. Homer P. Rainey, 'the University of Texas had the funds for an observatory but did not possess a faculty in astronomy . . . while the University of Chicago had an outstanding faculty, but was in need of more and better equipment for observation.' The dedication was made the occasion of a symposium in which leading experts in this country and abroad were invited to take part. The field under discussion was that of the galactic and extra-galactic structure in relation to the universe as a whole.

The main instrument, an 82-inch reflector, second in size only to the 100-inch reflector of the Mt. Wilson Observatory (and third to the Mt. Palomar 200-inch reflector), has been designed, according to Director Otto Struve, to be as efficient as possible for the study of the spectra and brightness of individual stars and nebulae. The telescope is made very short (f/4) and hence will gain in power by the smallness of the star images. A special spectrograph converts it into an f/2 instrument. In Director Struve's words: 'There are a billion stars or more which have never even been looked at and which this telescope will be capable of analyzing and classifying.'

The telescope was in full operation for several weeks before the dedication. Kuiper had already taken separate spectrograms of 300 stars which he knew to be relatively near because of their large proper motion. Two of them turned out to be of the type he was hoping to identify — 'white dwarfs.' This discovery raises the known number of these enormously dense stars of stripped atoms to 18.

The favorable latitude of the new observatory, making accessible regions of the sky which cannot be successfully reached from the more northern observatories, coupled with the large amount of clear weather, the good vision, and the enthusiasm of the staff should bring rich results.

The Buhl Planetarium in Pittsburgh is the fifth to be installed in the United States. The building contains a public observatory, lecture room and exhibit hall. A special feature is the room for amateur telescope makers. James Stokley, formerly at the Fels Planetarium in Philadelphia, is the director.

A Schmidt camera with the speed of an f/9.62 ratio has been constructed at the Mt. Wilson Observatory without excessive steepness of the curves of the correcting plate. The mirror is solid and silvered on the back. The photographic plate, one-half inch square, is pressed against the upper surface of the mirror at a distance from the reflecting surface of one-half the radius of curvature. (Christic and Hendrix contributed a very instructive article on the Schmidt camera to the August 1939 number of the Scientific American.)

Experiments in Solar Radiation.

With the gift of Godfrey L. Cabot 'for research on the utilization of solar radiation for the tasks of man,' experimentation is well under way at the Massachusetts Institute of Technology on efficient ways of trapping the heat of the sun on the roof of a house and storing it in the basement. An experimental house has been built which will be heated in winter and air conditioned during the summer with solar energy. Water, heated in tubes on the roof, will be piped to a large well-insulated water storage tank in the basement where it is expected to keep the water hot for several months, thus smoothing over the cloudy and sunny weather.

A 'solar-silo,' to collect and store heat from a solar heater, has been suggested by F. G. Cottrell. It would consist of a deep, cylindrical, concrete-lined pit, filled to near the top with loose, dry sand and closed with about ten feet of glass wool. Heated air would be brought to the top of the sand and drawn off as needed from the bottom.

Photography As an Aid to Astronomy.

A complete ring of light around the dark disk of Venus when it is nearly between us and the sun, which had previously been observed visually, was successfully photographed at the Lowell Observatory. Its brightness suggests that there are fine dust particles scattered through the gaseous atmosphere.

In order to take full advantage of the unusually close approach of Mars in 1939, by going where it would reach a high altitude, E. C. Slipher, who has been photographing Mars in different colors at every opposition, took his cameras and plates to Bloomfontein, Orange Free State.

At the Lick Observatory Wright has made a careful observational study by the spectroscopic method of the rotation period of Saturn. He finds at the equator a period of 10 hours and 2 minutes, which is about 2 per cent shorter than that found from the few observations of spots. He confirms the hitherto scanty evidence of an increase in period up to latitude 36° and shows that there is a further increase in higher latitudes.

McMath and Sawyer have given a summary (Publ. Observatory Michigan, vol. 7, No. 9) of their motion picture work on solar prominences with the tower telescope at Lake Angelus. A nearly continuous watch is kept for eruptions on the sun and as soon as the beginnings of a prominence are sighted operation of the complicated and ingenious photographing mechanism, even to photo-electric 'guiding,' is started and continued as long as the prominence performs. 'Frames' are secured at the rate of four or more per minute. In the three years that the tower telescope has been in operation, over 150,000 exposures have been made. Improvement in the speed and precision of projection and measurement of the films has also been carried forward, so that the ease and accuracy of the study of this great wealth of material is now comparable with that attained in the taking of the pictures.

The pictures, showing the astonishing development of a prominence and giving an almost continuous record of the motion of knots and streamers furnish a rapidly accumulating store of material of great scientific value.

The authors present arguments for believing that the apparent motion in prominences is really a motion of material rather than a travelling excitation. They also outline a new classification of prominences and summarize the conclusions they have drawn from their observations.

The 'coronaviser,' devised by A. M. Skellett, is another promising instrument for observing solar prominences and corona without an eclipse. It is a special television apparatus which scans the sky around the sun, separating the component of the photo-electric current arising from the glare (mainly a d-c component) from that arising from the coronal features (mainly a-c) and amplifying the latter component. The reproduction appears on the screen of a cathode ray tube. Numerous images of prominences have been obtained and a number of the images showed features that apparently were of coronal origin. (See also PHOTOGRAPHY.)

Comets.

This was a very rich year for comet seekers. No less than twelve comets were found. One more discovery was announced, but several observers with large telescopes looked in vain for it and so it was regarded as unverified and was not given a designation. Of the twelve, seven were known periodic comets returning to the neighborhood of the sun and were detected by professional astronomers with large telescopes close to their calculated positions. The other five were new comets discovered in systematic search with small telescopes. Comet 1939d was the brightest in several years. On April 19 it was of the third magnitude. Brightness, however, goes with nearness to the sun, and it was necessary to look at just the right time after sunset and have a northwest horizon that was free of trees, buildings, and clouds. It was the most discovered comet of the year. Hassel first reported it from Oslo, but it was soon learned that two Russian astronomers had discovered it just ahead of him. So it has been labeled the Jurlof-Achmarof-Hassel comet. Reports of other independent discoveries kept coming in for several days but no further additions were made to the comet's name. The new comet 1939b has now been observed long enough to show that it is moving in an elliptic orbit so it is expected that this comet, to be known after the discoverer as Comet Vaisala, will be back in about ten years. Comet Rigollet, another of the new ones, is found by Cunningham to have practically the same orbital elements as a comet discovered in 1788 by Caroline Herschel and apparently never seen since.

Watchers were all ready on the nights of October 9 and 10 for a repetition of the brief but brilliant meteoric shower which had occurred six years ago. The earth was again crossing the orbit of the Giacobini-Zinner comet and, while the comet itself was not yet at the rendezvous, it was anticipated that enough debris left behind by the comet might have become scattered along its path to give a display. No reports of any sizable shower have been made. Chances may be better next year when the earth crosses the orbit behind the comet.

Sun's Energy.

Bethe's picture (Physical Review, 55, 434) of the maintenance of the sun's output of energy, for which he received the A. Cressy Morrison prize, is essentially the consumption of hydrogen with the production of carbon and helium and the release of energy. The complicated atomic reaction which he finds would produce the observed output of energy may interest some readers. The first step is the combination of hydrogen and carbon producing an isotope of nitrogen of atomic weight 13. Being unstable this breaks down into an isotope of carbon of the same atomic weight and releases a positive charge of electricity. Another hydrogen nucleus combines with the carbon 13 producing nitrogen. This unites with a third hydrogen nucleus and an isotope of oxygen of atomic weight 15 is formed. This breaks down into nitrogen 15 and releases another charge of electricity. The nitrogen takes on another hydrogen nucleus and breaks up into carbon and a helium nucleus. Following out this theory one concludes that in the 'white dwarfs,' like the companion of Sirius, hydrogen has been practically all used up.

The Milky Way.

Stebbins and his associates have made a big contribution toward the true picture of our galaxy by their measurement, with the photo-electric photometer, of the color excesses of 1332 of the bluer stars. They have determined with much increased accuracy how much yellower some of these stars are than they should be according to their spectral type. Those most affected are in or near the plane of the Milky Way, but the effect is larger in some patches than in others. This leads to a determination of the total absorption of starlight as it comes through different regions and different distances. There is much corroborative evidence of this obscuring matter. Color excess is one of the most potent means for outlining its extent and hence making allowance for its effect on the apparent brightness of stars. The brightest patches of the Milky Way are only half as bright as they would be if there were no obscuring matter. In some areas only as little as 1/20 to 1/40 of a star's light gets through a distance of one thousand parsecs. One can realize how badly deceived astronomers were in judging distance by apparent brightness before they knew that a lot of the dimming was produced by obscuring matter and not by distance.

Schilt thinks he has evidence from his statistical studies that the measured parallaxes of stars are considerably too large and hence the accepted distances of the nearer stars too small, owing to systematic errors in the measurements.

Shapely announces that our galaxy is surrounded by faint stars of very high velocities of recession which unquestionably belong to our galaxy but are now far beyond the accepted limits of the galaxy. He points to similar star-halos around other galaxies, such as the Andromeda nebula.

Supernovae.

Three more 'supernovae' have been discovered by Zwicky with the 18-inch Schmidt telescope on Mt. Palomar. Another one was found at Harvard on a plate taken in 1937. Zwicky's discoveries in his systematic search since the Schmidt telescope was put in operation number just about one-half those reported in all astronomical history. He still thinks his earlier estimate of the frequency of appearance of supernovae as about one per nebula per 600 years is about right. The great number of the spiral nebulae, in which they appear, is the thing that makes the systematic watch worth while.

Honors Awarded.

Among this year's honors is the award of the Bruce Gold Medal of the Astronomical Society of the Pacific to Harlow Shapley. A letter is sent each year to the directors of six observatories (named in the deed of gift) asking each to nominate not more than three persons 'for distinguished services to astronomy.' The directors of the society then select the medalist from those nominated.

Several astronomers from America attended a congress on astrophysics held in Paris, July 17-23, under the chairmanship of H. N. Russell. Novae, white dwarfs, and the life history of a star were especially considered.

Necrology.

Sir Frank Dyson, Astronomer Royal of Scotland from 1905 to 1910, and Astronomer Royal of England from 1910 to 1933, died on May 25 while on a voyage from Australia to South Africa. He was 71 years old.

Suggested Reading.

In May 1937 the International Council of Scientific Unions established The Committee on Science and its Social Relations 'to promote thought upon the development of the scientific world picture and upon the social significance of the applications of science.' The report on astronomy, prepared by Bart J. Bok is printed in Popular Astronomy 47, 356. Among its topics it includes the popularization of astronomy (books and periodicals, planetaria and museums, adult education and public nights at the observatory, the press and the radio); the activities of the amateur astronomical societies of the world and the lines of research that are open to amateurs; where to find abstracts of papers and reports of research in progress; and what has been accomplished by international cooperation. It is an extremely valuable contribution.

Those interested in the many pioneering problems met with and solved in the construction of the world's greatest telescope on Mt. Palomar will be entertained with David O. Woodbury's book The Glass Giant of Palomar. Dr. Adams gives a very readable account of the optical figuring of the 200-inch mirror, the design and construction of the mounting and the work done on Mt. Palomar to make it habitable in Nature, Feb. 25, 1939, and reprinted in Journal Royal Astronomical Society Canada, July. Aug. 1939.

Astronomy, by W. T. Skilling and E. S. Richardson, New York, Henry Holt and Co., 1939, pp. 579, $3,00, is described by one reviewer as '. . . not too difficult for a student with very little scientific background, yet fairly complete and up to date in its treatment of the latest advances in the subject.'

1938: Astronomy

Stockholm Convention.

The International Astronomical Union, which ordinarily convenes every three years, met Aug. 3-10, 1938, in Stockholm under the presidency of Esclangon, director of the Paris Observatory. Eddington was elected to succeed Esclangon and the invitation to hold the next meeting in Switzerland in 1941 was accepted.

New Astronomical Equipment.

The 82-inch mirror for the McDonald Observatory on Mount Locke, Texas, will soon be shipped to Texas where the building and mounting are ready to receive it. A formal dedication with distinguished visiting speakers is scheduled for May of the coming year. This telescope will rank as second largest in the world until the completion of the 200-inch telescope at Mt. Palomar, California. In the latter, the final grinding of the mirror is nearly completed. Then there remains the polishing.

The University of Michigan has recently received a disk which should give a mirror over ninety inches in diameter. The grinding will not be started, however, until the project, including the moving of the Observatory to a more favorable site, is adequately financed.

The skeleton spectrograph for the study of diffuse nebulae, constructed at the McDonald Observatory, consists of an equatorial mounting carrying a rectangular plane mirror and adjustable wide slit, from which light is reflected to a stationary plane mirror located in the continuation of the polar axis, and thence back to a point on the opposite side of the mounting from the rectangular mirror, where it passes through quartz prisms into a Schmidt camera. The length of the camera is thus effectively about 150 feet. It works very fast and with it, large areas can be quickly examined for the existence of emission lines.

Those present at the meeting of the International Astronomical Union last summer were astounded at the results obtained by Lyot in the clear air of Pic du Midi with telescopes designed to eliminate all stray light and with very clean optical parts. Without the advantage of an eclipse he has been able to photograph the bright lines of the corona and, with the introduction of a filter transmitting only the red line of hydrogen, has secured motion pictures showing the rapidly changing eruptive prominences in wonderful detail. The spectro-heliokinematograph has been developed by McMath for a similar purpose.

In the slit spectrograph very little of the light from a star gets through the narrow slit and the spectrum builds up on the plate to observable intensity only with long exposures. Bowen has arranged a set of narrow mirrors in such a way that they cut the image of a star into narrow strips, place them end to end, and send them all through the slit. He calls the device an 'image slicer.'

A great step forward in the efficiency and accuracy of the photoelectric photometer has been made by Kron at the Lick Observatory. He mounts two prisms in such a way that they can be moved at right angles to the optical axis until they are in position to receive light from two stars, such as a variable star to be studied and a constant comparison star. A third prism, in the optical axis, can be rotated to reflect first the light of one star and then the light of the other to the photoelectric cell. The necessity of repeatedly moving the whole cumbersome telescope from one star to the other and back with the attendant adjustments is entirely avoided, and the danger of changes in atmospheric transmission is minimized by making the observations on the two stars nearly simultaneous. With this 'photo-comparator,' as Kron calls it, he secures observations with probable errors as small as ±.0015 magnitude in six minutes.

Massachusetts Institute of Technology has received a gift of $647,700 from Godfrey L. Cabot the income of which is to be devoted to search for direct methods of converting the sun's radiant energy into useful power or storing such power for future use.

Planets and Satellites.

The news of the addition to the solar family of a planet or moon is exciting news. The last addition was the planet Pluto. Twenty-four years ago Nicholson discovered the ninth satellite of Jupiter. Last July by careful search he added Jupiter X and Jupiter XI. This kind of search is very difficult, for these objects are of the nineteenth magnitude — about one twenty millionth as bright as a first magnitude star — and, of course, they are only to be found very close to the very brilliant disk of Jupiter. Assuming that they are about as good reflectors as other satellites, we conclude, that they are only about twenty-five miles in diameter. The calculated orbits show that Jupiter X revolves about Jupiter in about 260 days and also in other respects is remarkably similar to VI and VII. Jupiter XI, on the other hand, falls into a group with VIII and IX revolving on the outskirts of the Jupiter system with a retrograde motion — opposite to the direction of nearly all other motions of revolution or rotation in the solar system — in about 700 days. These two sets of triplets offer very interesting problems.

Asteroids.

The Reinmuth 'Object,' classed as an asteroid and designated 1937UB, took on such an access of velocity right after its discovery that no one knew where to look for it and it was not seen again. With the pre-discovery positions from photographic plates on which it was subsequently found it was possible to compute an orbit of probably fair accuracy. Its point of closest approach to the sun is well within the earth's orbit, the perihelion distance being something like six-tenths of an Astronomical Unit; the eccentricity has the extremely high value of about 0.6; and its period is a little over two years. Calculations differ as to how far from the earth it was when nearest, but apparently it was not far outside the moon's orbit. In all these respects it strongly resembles Apollo, discovered in 1932, and Adonis, discovered in 1936. Their orbits are all strongly perturbed in their near approaches to the earth. Apollo, although observed over an interval of 17 days, has never been seen since and is pretty certainly lost. Conditions were more favorable for Adonis and it was observed over a considerably longer interval so there is reason to hope that it may be kept track of. There is very little hope for Hermes (as Reinmuth calls his 'Object'). One difficulty is its tiny size, perhaps a mile in diameter, which makes it exceedingly faint in reflected sunlight unless it is very close.

Comets; Aurora Borealis; Eruptive Prominence on Sun.

Hope long deferred must sometimes be very chastening to those who search the sky with telescopes night after night, month after month, in the hope of picking up a comet. Last year eight were found, this year one. And this one had been found once before (Gale 1927VI), so that there was some idea of when and where to look for it. It was very faint and was picked up photographically by Cunningham at Harvard.

The display of the Aurora Borealis widely seen over the earth on Jan. 25, 1938, of the coronal type and beautifully tinged with color, is said to have been the most brilliant for over two centuries.

On March 20, 1938, at the start of the routine daily observation of the sun with the spectro-heliograph on Mt. Wilson, a conspicuous eruptive prominence was seen over the north pole of the sun and about 296,000 km. above the chromosphere. It was then photographed repeatedly over a period of 2 hrs. 34 min., when observations were stopped by clouds. In that time it reached a height of 1,550,000 km., or 1.12 solar diameters, above the chromosphere — the greatest height ever observed. The velocity was observed to increase from 65 km/sec in two rather sudden changes in 200 km/sec.

Stellar Motions and Positions.

Boss's General Catalogue of proper motions and stellar positions, planned and started thirty-three years ago by Lewis Boss and carried to completion by Benjamin Boss and his colleagues at Albany, has just appeared. There are five volumes with entries of 33,342 stars. Its forerunner, the Preliminary General Catalogue, published in 1910, has hitherto been the standard source of material for all studies of positions and motions of the stars.

The published notes about the star Wolf 424 this year form an interesting bit of astronomical history. Kuiper made a tentative prediction in May that this star would be found to have even a larger parallax and be even nearer than Wolf 359. The argument was that it was also a dwarf red star (hence of about the same low intrinsic brightness) but appeared considerably brighter than Wolf 359. Then Reuyl announced that he had measured the parallax and found it to be much smaller than Kuiper had predicted, adding that the photographic image of the star appeared elongated. Kuiper then examined it carefully with the 40-inch lens and found it to be a double star. This solved the puzzle: two stars were contributing to the brightness and they were therefore much farther away than was at first supposed.

Variable Stars.

A generous grant has made possible the establishment of a Bureau of Variable Star Research at the Harvard Observatory. An intensive study is planned of nearly all the known variable stars brighter than the eleventh magnitude, some 2,000 in number, on the half million photographic plates that have accumulated in the last fifty years.

Binaries.

The star e Aurigae has been a puzzle ever since the suggestion was first made that it is an eclipsing binary with the unusually long period of 27 years. Eclipse lasts for nearly two years and the star stays at practically constant minimum brightness for about 11 months. Only one spectrum is visible and this persists right through the minimum although with some variations in the lines. Apparently the eclipsing star is practically dark and is semi-transparent to the light of the bright star. Kuiper, Struve and Stromgren have joined mental forces and picture to us a system of two stars, the bright one with 190 times the sun's diameter, the dark one 2,700 times. The orbit is inclined so that the bright star passes behind only the outer part of the dark star. This outer part is practically transparent except for an ionized shell, or 'Heaviside layer' produced by the bright star. This layer causes a weakening of the light of the bright star without greatly affecting the spectrum. There are a few other eclipsing binaries, such as z Aurigae, and the recently recognized VV Cephei, which appear to be comparable systems.

By photographing the spectrum of one star, then moving the telescope and photographing the spectrum of another star of a different spectral type, Hynek has produced artificial composite spectra duplicating natural composite spectra and confirming the opinion that duplicity is the most frequent cause of this characteristic. By photographing suitable pairs it was possible to judge what difference in brightness of stars of different spectral type one can have and still be able to see that the spectrum is composite. This is a very important question for eclipsing binaries where the certainty of the solution for the orbital elements often depends on the estimate of relative brightness of the two components from the intensity of the lines of their combined or composite spectrum. Various estimates have been given as to how bright the fainter component has to be in order to make its lines show on the photograph. Hynek finds that under very favorable conditions the difference may be as great as 5 magnitudes. This is a surprisingly large figure. Certain spectral combinations are very hard to detect.

Supernova.

Zwicky pictures a 'supernova' as a star in which a sudden enormous outpouring of energy is rapidly followed by a settling into a 'collapsed neutron star.' The tremendously increased gravitational force would slow down the light from the interior so that it would take infinitely long to reach the surface. And the light — even if it did arrive — would be of zero energy and hence incapable of affecting our instruments of observation.

Necrology.

During the year 1938 Astronomy has suffered severe losses in the death of George Ellery Hale, William Wallace Campbell, one-time director of the Lick Observatory, Francis G. Pease, William H. Pickering and Ernest W. Brown.