We presented here the results of a survey for quasars based on the ultraviolet-excess selection technique and
performed in a 20.6-square-degree field located in the vicinity of NGC 450. The quasar candidates were
selected by visual inspection of a U/B dual-exposure Schmidt plate; the survey was photometrically
calibrated a posteriori using the MAMA measuring machine. The derived limiting magnitude is B = 19.6
and the threshold value in 
is
-0.30. We presented a list of 140 quasar candidates (95 primary + 45 secondary) with accurate positions
and U/B magnitudes. The survey technique has been designed to favour the homogeneity of the
selection all over the plate, whereas the photometric reduction was necessary to derive the limits of the
survey. However, the photometric calibration suffers from several problems such as the fact that the faint end
of the photometric sequence is restricted to a small area of the plate. Therefore, we consider that the
homogeneous output of the survey is our list of candidates despite the given photometric characteristics of
the objects.
The primary candidates have been spectroscopically identified, leading to the detection of 59 quasars (out of
which only 6 were previously known). This suggests an efficiency of at least 60%. We did not
spectroscopically identify the secondary candidates. Instead, we initiated, in the same fields (see Sect. 1),
quasar surveys based on other selection criteria and we expect that the merging of all the different candidate
lists will allow to point out candidates with a higher efficiency. The photometric characteristics of the
secondary candidates seem very similar to those of the primary ones, although they are slightly redder and
fainter. This indicates that the exact completeness of the present survey cannot be properly estimated without
full identification of the secondary candidates. Comparison with other surveys suggests that the present one
should potentially be fairly complete (for z < 2.2) down to B = 19.0 (
= -0.30)
but it is definitely not complete in the range B 
[19.0, 19.6]. This could be due to a bluer selection
threshold in 
at these magnitudes.
We presented, for the different identified objects, information such as the nature of the object, its possible redshift, the line positions and equivalent widths, etc. Some of the objects have characteristics that deserve further investigation: besides the already studied associations Q0107-025A and B (Surdej et al. 1986) and Q0118-031A, B and C (Robertson et al. 1986), several quasars exhibit prominent line-of-sight absorption lines; one of the quasars is of BAL type.
We also investigated the spatial distribution of both the candidates and the confirmed quasars. We detected,
for the quasars of our sample, a significant deviation from randomness in the form of a propensity to cluster
in pairs on the celestial sphere with a typical scale of about 10 arcmin. We also noticed
a deviation towards a three-dimensional clustering attributable to the occurrence of only one very interesting
pair of quasars having similar redshifts:
Q0107-025A and B. The redshift distribution of the quasars in our sample exhibits one statistically
marginal anomaly that manifests itself as an excess of quasars in the bin z 
[0.9, 1.0]. This effect, if
real, certainly deserves further investigation.
A survey, very similar to the one presented here, has been performed in a neighbouring field labelled NGC 520 (see Surdej et al. in preparation). This type of U/B survey is, in fact, the oldest one and the precursor of our multitechnique quasar surveys.
Acknowledgements
This work was supported in part by contracts ARC 90/94-140 and 94/99-178 "Actions de Recherche Concertées de la Communauté Française'', SC-005 "Services Fédéraux des Affaires Scientifiques, Techniques et Culturelles'' (Belgium) and HCM network CHRX-CT92-0044 (European Union). O.M. was a European Space Agency fellow during his stay in Liège and he later received a financial support from the Société de Secours des Amis des Sciences (France). J.S. would like to acknowledge financial support from the FNRS (Belgium) for travel grants as well as PRODEX.