Physical origins of the cosmic ultraviolet background

The author describes two experiments carried out to determine the origins of the far ultraviolet diffuse background. He designed the first experiment to make high sensitivity (minimum detectable continuum flux ∼ 5-10 phcm-2s-1A-1si-1), moderate resolution (λ/δλ ∼ 100) spectroscopic measurements in the 600-2000 A band. All aspects of the instrument design, including the choice of the optical configuration, diffraction grating, detectors, filters, spectrograph design, and baffling systems were determined by five key considerations: maximization of sensitivity and spectral resolution, and minimization of biases and noise introduced by stellar contamination, airglow scattering, and internal background. The analysis was governed by a concern for all the systematic errors which can bias a measurement of small amplitude fluctuations. Fluctuations were observed that were correlated with the instrument scan direction, and are therefore instrumental in origin. Fluctuations at a lower level (Δl/l ∼ 4%) were also detected along the scan direction, which may be instrumental, or may arise from the integrated light from galaxies. The power spectrum of these fluctuations was used as an upper limit to derive a constraint on the amplitude of correlated fluctuations that can be produced by galaxies