Key points are not available for this paper at this time.
From deep optical images of three clusters selected by virtue of their X-ray luminosity and/or optical richness (1455+22; z=0. 26, 0016+16; z=0. 55 and 1603+43; z=0. 89), we construct statistically-complete samples of faint field galaxies (I \ 25) suitable for probing the effects of gravitational lensing. By selecting clusters across a wide redshift range we separate the effects of the mean redshift distribution of the faint field population well beyond spectroscopic limits and the distribution of dark matter in the lensing clusters. A significant lensing signature is seen in the two lower redshift clusters whose X-ray properties are well-constrained. Based on these and dynamical data, it is straightforward to rule out field redshift distributions for I \ 25 which have a significant low redshift excess compared to the no evolution prediction, such as would be expected if the number counts at faint limits were dominated by low-z dwarf systems. The degree to which we can constrain any high redshift tail to the no evolution redshift distribution depends on the distribution of dark matter in the most distant lensing cluster. In the second paper in this series, we use the lensing signal to reconstruct the full two-dimensional mass distribution in the clusters and, together with high resolution X-ray images, demonstrate that their structural properties are well-understood. The principal result is therefore the absence of a dominant low-z dwarf population to I \25.
Smail et al. (Thu,) studied this question.