The diameter of stellar images produced by Earth-based telescopes often exceeds the width of the entrance slit of the associated spectrograph. It is not uncommon for more starlight to be wasted on the jaws of the slit than is transmitted by the slit. The purpose of an image-slicer is to redirect the light from the slit jaws into the slit; to do so, the length of the slit illuminated by starlight must be increased, either by stacking slices of the image along the slit, or by elongating the image and superimposing full-length slices of it. Thus, in order to work, an image-slicer of whatever type must lose spatial resolution of the sky along the slit.

Four Reasons are advocated to stress the interest of Fabry-Perot Spec-trographs (F.P.S.) as potentially useful instruments for the future Very Large Telescopes :

1. - Reasonably wide spectrum of Astronomical uses

2. - Good use of Telescope area

3. - Relative cheapness

4. - Versatility

The major advantages of the FTS technique are (1) multiplexing, (2) throughput, (3) instrumental profile, (4) stability of frequency calibration, and (5) spectrophotometry accuracy. The multiplex advantage is realized only if one is detector noise limited for the signal within an individual spectral-resolution element. At optical and thermal infrared wavelengths, this is only the case at high spectral resolution (≥ 50000) for modern detectors. By the time the VLT is operating one expects this to also be the case in the 1- to 2.5-micron region. At resolutions ≥ 50000 there are severe problems matching dispersive spectrographs to the VLT aperture, whereas existing FTS instruments already have adequate through-put to match to fields of a few arcsec with a VLT. When the other advantages are considered, the FTS is the instrument of choice for high-resolution (≥ 50000) spectroscopy of absorption features with a VLT. Foreseeable astrophysical applications include observations of interstellar and circumstellar features and of fully resolved profiles of photospheric and planetary lines.

During 1983, the advisory committee for instrumentation at the Anglo Australian Telescope (AAT) recommended that, as a matter of high priority, an échelle spectrograph should be procured for the f/36 coudé focus. The aim was to satisfy the high spectral-resolution needs of the community which, hereto, have been neglected. We at University College London (UCL) were invited to consider the design and our proposal was subsequently accepted by the Board. We are now at an advanced stage in optimizing the optical system, and mechanical design is proceeding. We plan to test the complete system in the laboratory at UCL prior to installation and commissioning at the AAT which we anticipate will be during 1987.

Imaging Spectroscopy is a technique in which a spectrum is obtained for each spatial resolution element across a wide field. The data is essentially 3-D, and may be viewed as a series of monochromatic images, or as a two dimensional array of spectra. A device generating such data may be called an imaging spectrometer. In a previous paper (Atherton, 1983 SPIE 445, 535) three different imaging spectrometers - based on grating, Fabry-Perot and Fourier Transform devices - were compared in terms of their ability to obtain spectral and spatial information over a wide field and broad band, to the same spectral resolution and S/N ratio, using the same detector array. From such a study it is clear that interferometer based devices are significantly faster than conventional grating spectrographs.

A preliminary 30-arcmin prime focus (f/2.0) refracting field corrector system for the University of California Ten-Meter Telescope (UC TMT) is presented which features 1/4-arcsec images containing more than 80% of the energy, over limited passbands within the wavelength range λ3300Å to λ1.0µ. Provision has been made in this system for an atmospheric dispersion corrector (ADC) but same has not yet been realized. Optical elements herein are small enough that this design could be scaled up to a Fifteen-Meter NNTT/SMT.

A compact 40-arcmin internal Cassegrain (f/1.75 hyperbola to f/5.0) broad-passband (λ3300Å to λ1.0µ) corrector, suitable for imaging and multi-object spectroscopy at the UC TMT, is presented which features 1/4-arcsec images containing more than 90% of the energy when averaged over field angle and color.

Three 60-arcmin external Cassegrain correctors for 300-inch f/1.8 and f/2.0 parabolic primary mirrors are presented which are suitable for a Fifteen-Meter NNTT/MMT. Image quality is comparable to the UC TMT Cassegrain corrector and it exceeds that of the UC TMT preliminary prime focus corrector system by a substantial margin. Each of these correctors contains an ADC which has been implemented in one example, eliminating 4.0 arcsec of differential atmospheric refraction with an rms residual of +/-0.10 arcsec over the broad passband (λ3300Å to λ1.0µ). A 60-arcmin external Cassegrain (f/1.8 extreme hyperbola to f/4.5) corrector with ADC yields yet a factor two in image quality but said hyperbolic primary mirror would be incompatible with angular field requirements in the thermal infrared.

A (300-inch) 40-arcmin external Cassegrain (f/1.0 parabola to f/4.0) broad-passband (λ3300Å to λ1.0µ) corrector with ADC is presented. Image quality is comparable to the previous Cassegrain correctors. The practicality of this design, together with recent advances in optical manufacturing capability of large, fast, nonspherical optics, suggests that relatively inexpensive compact telescopes of very large collecting area may be possible in the near future.

Improved designs of refractive correctors produce excellent images with fast telescopes such as those with an F/1.5 prime focus and F/3.5 secondary focus. The fields are flat and there is compensation for the chromatic effect caused by windows. Disadvantages of such correctors are that stray light is produced at the optical surfaces, the elements must be supported at their edges, prerequisite high quality glass is available in only limited sizes, and all wavelengths are not transmitted.

Reflective correctors, on the other hand, can produce diffraction limited images at all wavelengths and the mirrors can be supported across their backs as well as at their edges. Disadvantages are that the images are degraded by any substantial window (such as a detector faceplate), there is more central obstruction, and the correctors are sometimes very large and heavy.

Except, perhaps, for a specialized telescope, such as one devoted to multi-object slit spectroscopy using fibres, the refractive corrector is preferable at fast foci.

A good combination is a Ritchey-Chretien (R-C) telescope with refractive correctors at the fast prime and secondary foci, and a reflective corrector-magnifier for the slow infrared focus.

The purpose of this paper is to point out that

1) proper treatment of polarization by a telescope is not only important for polarimetry, but also for other observations requiring high signal-to-noise ratio, and

2) this does put a constraint on telescope design, which, however, is not unduly restrictive.

Polarimetry is the observational technique which can detect anisotropics in point sources, their environment, or the medium between them and us. Modern optical polarimetry can be linear or circular, and is making progress towards spectropolarimetry and imaging polarimetry using panoramic detectors (e.g. McLean, 1984). Given sufficient photons, the precision obtained is as high as 1 part in 50 000 (see Odell, 1981 for a project requiring this precision), but more often a precision between 1 part in 1 000 to 10 000 is sufficient. (Spectro)polarimetry has been applied to planets, the Sun and other stars, stellar systems and galaxy nuclei; for a few modern investigations, see Baur, 1981; Jones et al, 1981; Schmidt and Miller, 1980. Experience in radio-astronomy has shown that when facilities for polarimetry are offered, many applications emerge from the astronomical community, yielding data that cannot be obtained by other techniques. It is essential that at least some of the large optical telescopes are capable of ob-serving polarization cleanly.

To exploit the full effectiveness of large ground based telescopes, the focal reducer technique can be applied very adequately. We review recent Hoher List observatory developments of observing techniques for direct imaging, field spectroscopy with multi-slit and masking methods for background supression and double-grism arrangements (heliometer principle)for absolute radial velocity determinations with a focal reducer field spectrograph.

The technology and terminology of integrated infrared detector arrays are discussed. Specific infrared array multiplexer designs include the charge-coupled device (CCD) and charge-injection device (CID). Laboratory data and telescope imagery obtained with three arrays (inSb CCD, Si:In CCD, and Si:Bi CID) are used to illustrate the performance and promise of integrated arrays in astronomical applications.

A 8x8 InSb CID array, developed in France by SAT (1) (Société Anonyme de Télécommunications) for terrestrial applications, has been tested at Meudon Observatory in order to evaluate if its performances at LHe temperature, meet the requirements of infrared astronomy. Next sections give : 1) Description of the array and the experimental conditions; 2) The test results; 3) Work under development.

The requirements on gratings and coatings for astronomical use differ from the general industrial requirements primarily in the scale of the components to be fabricated. Telescopes have large primary mirrors which require large coating plants to handle the components. Dispersive elements are driven by the requirement to be efficient in the presence of large working apertures, and usually optimize to large size in order to efficiently use the incoming radiation. Beyond this, there is a “new” technology of direct electronic sensors that places specific limits upon the image scale that can be used at the output of a telescope system, whether direct imagery or spectrally divided imagery is to be examined. This paper will examine the state of the art in these areas and suggest some actions and decisions that will be required in order to apply current technology to the predicted range of large new telescopes.

The importance and applied use of fibreoptics in astronomy has received rapidly growing attention in the past 5 years, particularly for instrumentation where the lightness, flexibility and simplicity of fibres, compared with classical optical systems, can be exploited to full advantage.

Angel and Angel et al, who seem to be the first to have used an optical fibre to link a telescope to an instrument, also made the first proposal for the construction of a VLT (FLOAT) consisting of 40 independent mirrors linked to a single instrument via optical fibres. Since that time many authors, including Connes, Serkowski et al, Hubbard et al, Heacox, Hill et al, Vanderriest, Courtes, Tubbs et al, Gray, Lund et al, Schiffer, Watson et al, Vanderriest et al, and Felenbock et al, have proposed or reported various applications involving fibres with astronomical instrumentation.

Current practice with four-meter reflectors is reviewed and the good match with fine-grained photographic plates of high DQE is examined. The present designs of 15-meter-class telescopes appear to point to instruments with a large flat focal plane and an image scale in the 5 arc sec/mm range. This requires an emulsion of considerably larger microcrystal size than currently used. Extension of the present emulsion type to the required grain size is not likely to be efficient; however, the use of new emulsion technologies should offer the opportunity to devise appropriate photographic detectors for use with an f/6, 5 arc sec/mm telescope. A gain in detective quantum efficiency appears possible.

the large field Lallemand electronographic camera has proved to be an ideal receptor for bidimensional photometry with the C.F.H. 3.6 m. Telescope. It permits to measure faint stars up to about B = 25. Plates of a large variety of objects have been obtained ; for example : crowded fields such as the nearby galaxy Messier 33 ; fields where one must recognize faint galaxies from stars; objects for which a high spatial resolution is needed, such as the optical jet of Messier 87 or gravitational lenses. It is possible to predict the limits of the receptor with a very large telescope.

A photon-counting system with a 512-channel parallel output digital image tube is presented. Electronics developped separately for each detector channel as well as data aquisition are optimized for low power consumption and high count rates. This detector, characterized by wide dynamic range, very low noise and high photo-metric accuracy, is especially suitable for spectrophotometry and calibrations.

Accurate photometry of fast variable objects requires simultaneous measurements of object, sky and comparison star with high time resolution. Additionally, if reliable colours of short time variations are of interest each light source has also to be measured simultaneously in several wavelength regions.

The prototyp of a 3 channel 5 colour photometer fitting these requirements was constructed at the Universitats-Sternwarte Miinchen in 1983 and shall be described in the following. The development of this instrument was carried out in two steps: At first a 3 channel photometer had been designed for observations in integral light. It was successfully tested at Calar Alto Observatory (Mitt. Astron. Ges. 60, p. 474, 1983)

Interstellar matter is certainly one of the fields where a very large telescope (VLT) will prove to be most fruitful. This includes (somewhat paradoxically, but this will be explained later) the study of extended emissions. I will now examine in turn the different domains of interest for a VLT.

I. Neutral diffuse matter

Optical and near IR observations will mainly contribute to this domain through high-resolution spectroscopy of interstellar absorption lines in the spectra of stars. These lines are resonant lines of atoms (NaI, KI, etc.) or ions (CaII, TiII, etc.) as well as of some molecules (CH+, CH, CN, CS+, C2 in the near IR). Clearly this kind of study is always photon - limited; a VLT will collect more photons than present telescopes, thus increase the possibilities considerably.

Star formation is an extremely active area of current research with studies ranging from the detailed properties of forming stars in the Galaxy to the evolution of galaxies at cosmological distances. Some of the problems being tackled are of long standing but many are related to unexpected discoveries which have followed from the application of new and improved observing capabilities throughout the electromagnetic spectrum during the last few years. Amongst others, these include the nature of y and X-ray sources in molecular clouds, the origin of the molecular mass outflows and various related phenomena (e.g. H2O masers, hot H2 emission...) observed in the vicinity of forming massive stars, the mechanism responsible for generating bursts of star formation in galaxy nuclei and the nature of star formation in cooling flows of intracluster gas. At the present time therefore there would be no shortage of good optical/infrared observing proposals which could take full advantage of the order of magnitude gain in sensitivity and, at least in the infrared, the improved angular resolution promised by most of the proposed VLT projects. The need for improved capabilities in the near and mid-infrared is also being pressed by developments in related observational techniques. In particular, sub-mm and mm wave molecular astronomy is set to advance rapidly during the next few years with the completion of several dedicated large dishes. Far infrared astronomy has also recently come of age with the completion of the IRAS all sky satellite survey and the adoption by ESA of ISO (infrared Space Observatory) as its next major astronomy mission.

Let me start by reminding you in a very general frame what kind of information we expect to be able to derive from the observation of stars.

First of all, there are observations - of photometric nature - which allow us to detect the location of stars in the sky. Modern astrophysics enables us, in the general case, to add more or less stringent information on the star distances. On the whole, one reaches in this way a picture of the galactic distribution of stars, i.e. information on the space location.

A second kind of information we can derive from photometric measurements is that concerned with the age of stellar objects. As is well known, this is done - mainly - by constructing HR diagrams for stars which are members of stellar clusters.