Ultraviolet telescopes such as the International Ultraviolet
Explorer (IUE), which has performed outstandingly well in
orbit since 1978, and the Hubble Space Telescope (HST),
launched in 1990 and now clearly beginning to overcome its well-publicized
problems, have optical coatings of aluminum and magnesium fluoride and detector windows of magnesium fluoride or lithium fluoride whose
reflectivity and transmission make them insensitive to UV
wavelengths much shorter than 1150 Å. However, the interstellar medium (ISM)
remains fairly transparent down to the Lyman edge of hydrogen at
912 Å. At shorter wavelengths, the large photoelectric
absorption cross-section of hydrogen prevents observations of objects
more than about 100 pc (326 light-years) from the Sun.
(The ISM becomes transparent again in the soft X-ray region of the
spectrum at wavelengths less than 100 Å.) A number of important
spectral lines are found in the small portion of the spectrum extending
from 1216 Å (the wavelength of the hydrogen Lyman-
line) to
912 Å, including the entire Lyman series of atomic hydrogen, the
Lyman and Werner bands of molecular hydrogen, and the resonance
lines of several abundant elements in various ionization stages. One of
the most important features in this band is the resonance doublet of
OVI (five-times ionized oxygen) at 1032 and 1038 Å, which
arises in gas at temperature 
to 
K under conditions of collisional equilibrium. This
feature may be used to probe the highest temperatures accessible to
UV and optical astronomy. The study of gas at even higher
temperatures generally requires X-ray detectors.
Pioneering work in the wavelength region between the Lyman-
line and the
Lyman edge was performed with the Princeton experiment on Copernicus, one of the Orbiting
Astronomical Observatories (OAO) (Rogerson et al. 1973). Using very bright, hot stars as background sources of light, the Copernicus telescope and spectrometer were designed to perform absorption-line studies of the interstellar gas at high spectral resolution. However, only a small region of the spectrum could be covered with its scanning spectrometer in any single observation, and absolute flux measurements of the stars themselves were generally not possible.
The only other satellite instrument that has made extensive astronomical observations at wavelengths near the Lyman edge is the ultraviolet spectrometer, which was carried on both the Voyager 1 and 2 spacecraft (Broadfoot et al. 1977). This instrument was designed to study the outer planets during close encounters but, in spite of its small collecting area, it has also been used to detect a few other celestial objects in the extreme UV, 500-912 Å, and many more in the 912-1216 Å band (Holberg 1990, 1991). However, the Voyager instrument achieves only rather low resolution-about 18 Å for stars and 38 Å for diffuse objects such as nebulae. This has made identification of some spectral features with specific emission and absorption lines difficult.