Résumé :
The first part of the talk concerns molecular oxygen studied with the
Submillimeter Wave Astronomy Satellite (SWAS), which has carried out a
sensitive search for O2 in a wide variety of molecular clouds. We have
not detected this species at a level of a few times $10^{-7}$ relative to H2.
This is several orders of magnitude below the prediction of steady--state
gas phase chemistry, and suggests that significant other factors must be
at work. The most plausible, as it also explains low abundance of gaseous
water, is depletion of gas--phase oxygen onto grain surfaces.
The second portion of the talk concerns Arecibo observations of narrow
line HI self-absorption (HINSA) in cold quiescent clouds. In clouds we have
studied in detail, the fractional abundance of HI is only slightly greater
than the predicted steady state value. This indicates that the minimum time
that has evolved since these dark clouds started their evolution to
molecular form is $>$ 3 million yr. Atomic hydrogen in molecular
clouds thus sets a very interesting lower limit to the time scale
for the evolution of these regions and for star formation to take place.
