We here report on observations of 86 GHz SiO (v=1, J=2-1) maser emission from the Mira variables R Leo and o Ceti and from Orion A, made between December 1977 and June 1979 with the new Onsala 20 m millimeter wave telescope equipped with a room temperature mixer. The SiO fluxes from R Leo and o Ceti appear to be correlated with their near infrared intensities, and to have a distinct phase lag with respect to the visual light curve. An irregular behaviour of the R Leo maser, at a time when the star approached an unusually bright maximum, is very probably a manifestation of the extreme sensitivity of the maser process to any disturbances. Definite intensity variations on a time scale of a day have been observed. The total integrated flux of the Orion A SiO maser is relatively stable but there are considerable relative intensity variations between the two main components, and weak emission appears to be present in the entire interval between the two strong features. Pumping considerations indicate that a very efficient, probably radiative pump is needed.

Linearly polarized SiO emission spread over 12 km/s has been detected from the star R Leo. The position angle of polarized emission varies systematically with respect to the spectral line center. Interpreted in terms of radiative transfer theory, this change in position angle may be due to magnetorotation, which allows the determination of the magnetic field (9×10−3/cos θ Gauss), and the SiO systemic velocity (−1 ± 2 km/s).

SiO emission in both a strong maser transition in the first excited vibrational state and in a weaker transition in the ground vibrational state both arise from the same small region in the Kleinmann-Low Nebula in Orion. Within the errors, the source position coincides with that of the infrared source IRc2.

A recent investigation based upon the OH luminosity distribution (Nguyen-Q-Rieu et al. 1979) has shown that Type I OH sources associated with Mira variables (OH Miras) are weak OH sources and are therefore only detected within ∼1 kpc from the Sun. Type II OH-Miras, which are more intense and rarer than Type I OH-Miras, are probably more distant objects. The group of unidentified Type II OH-IR sources probably consists of Type II OH-Miras of high OH luminosity. The IR colour index, which is usually higher for Type II sources, suggests that they have a colder and denser dust shell.

In a short review paper one is often forced to survey the subject in a broad way and to ignore many, perhaps even important, details. Such is the case here. My remarks are intended to indicate how I view the current status of the field. In general, I shall avoid qualifying my statements, even when I am clearly guilty of oversimplification. If someone finds my paper provocative, so much the better.

We analyze in detail the pumping of 18 cm OH masers by the overlap of far infra-red lines in HII/OH regions and in circumstellar shells. We present some results of a model of pumping of HII/OH masers by thermal overlap. Pumping by overlap appears to be quite general but very complicated. It can account for the different observed main-line masers. However, other types of pumping, mainly by collisions with H or by near infra-red radiation, are possible if certain conditions are met for the physical and astrophysical parameters.

This review covers recent progress in observations of masers in regions of star formation, especially in the years since the reviews by Burke (1975) and Lequeux (1977). Four regions of special interest are described in more detail: Orion, W51, W49 and W3.

We have measured the absolute positions of five OH maser sources associated with HII regions, using the Jodrell Bank phase compensated radio linked interferometer (Norris, Booth & Davis, 1980); these are reproduced in Table 1. In addition, we have mapped the relative positions of the features in the OH spectra at 1665 MHz for all of the sources and at 1667 MHz for 3 of them (W30H, NGC 7538 and W49). We are now able to compare the distribution of the OH masers in a given source with its radio continuum map and to investigate the proximity of masers in the 2 main line transitions and the H2O masers.

Hyperfine components of the far-infrared lines of OH are separated by velocity intervals ranging from about .6 to 8.0 km s-1 (fig. 1). As early as 1969 Litvak had suggested that overlapping of these components due to either thermal or systematic motions could be the source of OH main line inversion. We investigate here the effects of overlapping due to systematic motions.

The results of a new approach to the problem of the collisional step in the pumping cycle for OH and CH masers are reported. Rotationally inelastic collisions of OH and CH with both open and closed shell collision partners are considered using an expression derived from scattering theory. It is shown how Λ-doublet population inversion may arise in OH and CH. H-atoms and H2 may show opposite behaviour with CH, one partner cooling where collisions with the other lead to inversion. Implications for maser action are discussed and reveal excellent qualitative agreement with observation. Λ-doublet population inversion in OD, CD and NH+ is also considered.

Astronomical studies of the CH and OH masers have been based on the hypotheses of both radiative and collisional pumping. The mechanism of collisional inversion of the ground state Λ-doublets by neutral perturbers was first described by Bertojo et al. (1976) who made a semi-quantitative analysis of the collision process, based on calculations of potential energy curves. More recent work (Kaplan and Shapiro, 1979; Dixon and Field, 1979; Flower, 1980) has, however, been critical of some of the assumptions made by Bertojo et al. Consideration of the symmetry and the form of the interaction of CH and OH with the spherically symmetric perturbers para-H2 and He shows that current knowledge of the interaction potential is insufficient for quantitative results to be obtained.

All the existing models of H2O masers fail to explain such a strong source as W49 N. Observed and theoretical quantities are related by: nH2OWPℓ3 ≳ 1046 S, where S is the maser flux density (in Janskys), nH2O is the H2O number density (cm−3), Wp is pump rate (s−1), and ℓ is the length of amplification region on the line of sight (cm). Models involving vibrational activation (or deactivation) of H2O by H2 (Goldreich and Kwan, 1974; Norman and Silk, 1979), with the usual cross-section σν ≲ 10−19cm2, require ℓ > 1016 cm for the strongest H2O features (∼104 Jy), which is unacceptable in view of the VLBI results. Besides, because σν is so small, it is questionable if vibrational pumping could control rotational level populations at all. Depending on the energy source and sink there are four possible schemes of rotational pumping: CR, RC, RR, and CC (C - collisional, R - radiative). The first was modelled by de Jong (1973) and by Shmeld et al. (1976). Though difficulties with the sink (Goldreich and Kwan, 1974; 1979) are avoidable in the model by Shmeld et al. (Strelnitsky, 1979), ℓ ≳ 1015 − 1016 cm is still required for the strongest features. Therefore other possibilities of rotational pumping are being investigated. One CC-model is presented below.

Previous studies of time variations in water vapour sources associated with HII regions by Sullivan (1971, 1973) and Gammon (1976) have revealed variability on time-scales of 106-107s. Sullivan monitored nine sources over a period of 14 months and Gammon observed W49 (H2O) for 13 months. In an attempt to gather more complete information on the variability of interstellar water masers we have monitored 18 sources for the 32-month period 1974 September-1977 May. The sources selected are listed in Table 1. They were the strongest known at the beginning of the programme above declination −20°.

We wish to report briefly here the results of new observations of the H2O maser emission from the galaxy NGC 4945 made in 1979 July with the Itapetinga radio telescope. The H2O emission that we detected in 1978 September, shown in fig. 1, represented an intrinsic intensity about ten times that of W 49, the strongest source in our Galaxy, and raised the question of its nature (Marques dos Santos and Lepine, 1979). The main purpose of the new observations was to obtain a more accurate position for the source and to look for variability.

Kinematic and physical models of high velocity H2O masers are briefly discussed.

A far-ultraviolet, wide-field camera/spectrograph investigation has been proposed for Spacelab missions. By use of an objective grating, this experiment could survey large areas of the sky, obtaining spectra of early-type stars as faint as 12th visual magnitude in the 950-2000 Å wavelength range, with 1 to 2 Å spectral resolution.

Ultraviolet observations of interstellar molecules are currently being obtained primarily with two astronomical satellites, Copernicus (OAO-3) and International Ultraviolet Explorer (IUE). The former covers the wavelength range down to 900 Å with 0.05 or 0.2 Å resolution. IUE can observe fainter or more highly reddened stars than Copernicus, with 0.1 Å resolution, but its wavelength range does not include the resonance absorptions of the important interstellar molecules H2, HD, and N2.

When the lowest possible noise is required the maser amplifier remains the unquestionable amplifier choice for moderate bandwidth microwave and long wavelength millimeterwave receivers. The reason for the outstanding low noise properties of the maser is partly that it is cooled to a few degrees K (≲ 4 K) but also that the amplification process is the most fundamental one, i.e. amplification takes place when quanta (photons) are added directly to the signal field by stimulated emission of radiation from energetically excited particles.

Millimetre-wave spectral lines show a wide range of linewidths, from 0.1 km s-1 in dark clouds to 30 km s-1 in some massive molecular clouds; at 100 GHz the corresponding Doppler widths range from 33 kHz to 10 MHz. SiO masers at 86 GHz have two lines typically 1 MHz wide spaced by 6 MHz. For most applications a resolution of 100 or 250 kHz is suitable.

In mm-wave spectroscopy, the acousto-optic spectrometer (AOS hereafter) compares favourably with conventional spectrometers such as filter-banks or autocorrelators. Although its frequency resolution is limited to 20 kHz and its instantaneous bandwidth is limited to several hundred MHz, it has 103 resolvable points. Because of the simplicity and stability of the AOS, it is not difficult to construct and maintain a system of 104 resolvable points by the parallel operation of spectrometers. At Tokyo Astronomical Observatory (TAO) a new AOS has been constructed which is equipped with two different types of TeO2 AO deflectors and provides high resolution (38 kHz) or wide band (220 MHz) spectra. It has 1728 frequency channels (figure 1). Another AOS has been built at Kisarazu Technical College (KTC) for a CO survey at 115 GHz. It has 256 frequency channels and a bandwidth of 230 MHz (figure 2).