We present metal abundances in 15 damped Ly \alpha systems ( DLAs ) from the Complete Optical and Radio Absorption Line System ( CORALS ) survey , designed to be free from any biasing effects due to extinction of QSOs by dust in intervening absorbers .
It has long been suggested that such biasing may explain differences in metallicity between damped Ly \alpha systems and coeval luminous galaxies , and between model predictions and observations .
We use our measured zinc and chromium abundances ( combined with those for five more CORALS DLAs from the literature , giving us a very nearly complete sample ) to test whether the metallicity and degree of dust depletion in CORALS DLAs are significantly different from those of existing , larger , samples of DLAs drawn from magnitude limited , optical surveys .
We find that the column density weighted metallicity of CORALS DLAs , { [ } \langle { ( Zn / H ) } _ { DLA } \rangle { ] } = -0.88 \pm 0.21 in the redshift interval 1.86 < z _ { abs } < 3.45 , is only marginally higher than that of a control sample from the recent compilation by Kulkarni et al. , { [ } \langle { ( Zn / H ) } _ { DLA } \rangle { ] } = -1.09 \pm 0.10 .
With the present limited statistics this difference is not highly significant .
Furthermore , we find no evidence for increased dust depletions in CORALS DLAs—their [ Cr/Zn ] ratios conform to the known trend of increasing depletion ( decreasing [ Cr/Zn ] ) with increasing metallicity , and we have encountered no cases where Cr is as depleted as in local cold interstellar clouds .
These results , when combined with the earlier findings of the CORALS survey reported by Ellison et al .
in 2001 , make it difficult to invoke a dust-induced bias to explain the generally low level of chemical evolution exhibited by most DLAs .
Rather , they indicate that large scale optical QSO surveys give a fair census of the population of high redshift absorbers .