
We have used the 2dF instrument on the AAT to obtain redshifts of a sample of
z<3, 18.0<g<21.85 quasars selected from SDSS imaging. These data are part of a
larger joint programme: the 2dFSDSS LRG and QSO Survey (2SLAQ). We describe
the quasar selection algorithm and present the resulting luminosity function of
5645 quasars in 105.7 deg^2. The bright end number counts and luminosity
function agree well with determinations from the 2dF QSO Redshift Survey (2QZ)
data to g\sim20.2. However, at the faint end the 2SLAQ number counts and
luminosity function are steeper than the final 2QZ results from Croom et al.
(2004), but are consistent with the preliminary 2QZ results from Boyle et al.
(2000). Using the functional form adopted for the 2QZ analysis, we find a faint
end slope of beta=1.78+/0.03 if we allow all of the parameters to vary and
beta=1.45+/0.03 if we allow only the faint end slope and normalization to
vary. Our maximum likelihood fit to the data yields 32% more quasars than the
final 2QZ parameterization, but is not inconsistent with other g>21 deep
surveys. The 2SLAQ data exhibit no well defined ``break'' but do clearly
flatten with increasing magnitude. The shape of the quasar luminosity function
derived from 2SLAQ is in good agreement with that derived from type I quasars
found in hard Xray surveys. [Abridged]

We have crosscorrelated the WMAP data with several surveys of extragalactic
sources and find evidence for temperature decrements associated with galaxy
clusters and groups detected in the APM Galaxy Survey survey and the ACO
catalogue. We interpret this as evidence for the thermal SunyaevZeldovich (SZ)
effect from the clusters. Most interestingly, the signal may extend to ~1 deg
around both groups and clusters and we suggest that this may be due to hot
`supercluster' gas. We have further crosscorrelated the WMAP data with
clusters identified in the 2MASS galaxy catalogue and also find evidence for
temperature decrements there. From the APM group data we estimate the mean
Compton parameter as y(z<0.2)=7x10^(7). We have further estimated the gas mass
associated with the galaxy group and cluster haloes. Assuming temperatures of 5
keV for ACO clusters and 1 keV for APM groups and clusters, we derive average
gas masses. Using the space density of APM groups we then estimate Omega_gas.
For an SZ extent of theta_max=20', kT=1 keV and h=0.7, this value of
Omega_gas=0.04 is consistent with the standard value of Omega_baryon=0.044 but
if the indications we have found for a more extended SZ effect out to theta=60'
are confirmed, then higher values of Omega_gas will be implied. Finally, the
contribution to the WMAP temperature power spectrum from the extended SZ effect
around the z<0.2 APM+ACO groups and clusters is 12 orders of magnitude lower
than the l=220 first acoustic peak. But if a similar SZ effect arises from more
distant clusters then this contribution could increase by a factor >10 and then
could seriously affect the WMAP cosmological fits.

We report on measurements of the cosmological constant, Lambda, and the
redshift space distortion parameter beta=Omega_m^0.6/b, based on an analysis of
the QSO power spectrum parallel and perpendicular to the observer's line of
sight, from the final catalogue of the 2dF QSO Redshift Survey. We derive a
joint Lambda  beta constraint from the geometric and redshiftspace
distortions in the power spectrum. By combining this result with a second
constraint based on mass clustering evolution, we break this degeneracy and
obtain strong constraints on both parameters. Assuming a flat cosmology and a
Lambda cosmology r(z) function to convert from redshift into comoving distance,
we find best fit values of Omega_Lambda=0.71^{+0.09}_{0.17} and
beta(z~1.4)=0.45^{+0.09}_{0.11}. Assuming instead an EdS cosmology r(z) we
find that the best fit model obtained, with Omega_Lambda=0.64^{+0.11}_{0.16}
and beta(z~1.4)=0.40^{+0.09}_{0.09}, is consistent with the Lambda r(z)
results, and inconsistent with a Lambda=0 flat cosmology at over 95 per cent
confidence.

We present a new CCD survey of galaxies within the N and S strips of the
2dFGRS areas. We use the new CCD data to check the photographic photometry
scales of the 2dFGRS, APMBGC, APMStromlo Redshift Survey, DurhamUKST (DUKST)
survey, Millenium Galaxy Catalogue (MGC) and Sloan Digital Sky Survey (SDSS).
We find evidence for scale and zeropoint errors in the 2dFGRS northern field,
DUKST and APM data of 0.10, 0.24 and 0.31 mag. respectively; we find excellent
agreement with the MGC and SDSS photometry. We find conclusive evidence that
the S counts with B<17 mag are down by ~30% relative to both the N counts and
to the models of Metcalfe et al. We further compare the n(z) distributions from
the B<17 mag. DUKST and B<19.5 2dFGRS redshift surveys using the corrected
photometry. While the N n(z) from 2dFGRS appears relatively homogeneous over
its whole range, the S n(z) shows a 30% deficiency out to z=0.1; at higher
redshifts it agrees much better with the N n(z) and the homogeneous model n(z).
The DUKST n(z) shows that the S `hole' extends over a 20 by 75 degree squared
area. The troughs with z<0.1 in the DUKST n(z) appear deeper than for the
fainter 2dFGRS data. This may be evidence that the local galaxy distribution is
biased on >50 h1 Mpc scales which is unexpected in a LambdaCDM cosmology.
Finally, since the Southern local void persists over the full area of the APM
and APMBGC with a ~25% deficiency in the counts below B~17, this means that its
extent is ~300 h1 Mpc by 300 h1 Mpc on the sky as well as ~300 h1 Mpc in the
redshift direction. Such a 25% deficiency extending over ~10^7 h3 Mpc^3 may
imply that the galaxy correlation function's powerlaw behaviour extends to
\~150 h1 Mpc with no break and show more excess largescale power than
detected in the 2dFGRS correlation function or expected in the LambdaCDM
cosmology.

We have optically identified a sample of 56 featureless continuum objects
without significant proper motion from the 2dF QSO Redshift Survey (2QZ). The
steep numbermagnitude relation of the sample, $n(\bj) \propto 10^{0.7\bj}$,
is similar to that derived for QSOs in the 2QZ and inconsistent with any
population of Galactic objects. Follow up high resolution, high
signaltonoise, spectroscopy of five randomly selected objects confirms the
featureless nature of these sources. Assuming the objects in the sample to be
largely featureless AGN, and using the QSO evolution model derived for the 2QZ,
we predict the median redshift of the sample to be $z=1.1$. This model also
reproduces the observed numbermagnitude relation of the sample using a
renormalisation of the QSO luminosity function, $\Phi^* = \Phi^*_{\rm \sc
qso}/66 \simeq 1.65 \times 10^{8} $mag$^{1}$Mpc$^{3}$. Only $\sim$20 per
cent of the objects have a radio flux density of $S_{1.4}>3 $mJy, and further
VLA observations at 8.4 GHz place a $5\sigma$ limit of $S_{8.4} < 0.2$mJy on
the bulk of the sample. We postulate that these objects could form a population
of radioweak AGN with weak or absent emission lines, whose optical spectra are
indistinguishable from those of BL Lac objects.

(ABRIDGED) We present a power spectrum analysis of the 10K catalogue from the
2dF QSO Redshift Survey. We compare the redshiftspace power spectra of QSOs to
those measured for galaxies and Abell clusters at low redshift and find that
they show similar shapes in their overlap range, 50150h^{1}Mpc, with
P_QSO(k)\propto k^{1.4}. The amplitude of the QSO power spectrum at z~1.4 is
almost comparable to that of galaxies at the present day if Omega_m=0.3 and
Omega_Lambda=0.7 (the Lambda cosmology), and a factor of ~3 lower if Omega_m=1
(the EdS cosmology) is assumed. The amplitude of the QSO power spectrum is a
factor of ~10 lower than that measured for Abell clusters at the present day.
At larger scales, the QSO power spectra continue to rise robustly to ~400
h^{1}Mpc, implying more power at large scales than in the APM galaxy power
spectrum measured by Baugh & Efstathiou.
We split the QSO sample into two redshift bins and find little evolution in
the amplitude of the power spectrum. The QSO power spectrum may show a spike
feature at ~90h^{1}Mpc assuming the Lambda cosmology or ~65 h^{1}Mpc assuming
an EdS cosmology. Although the spike appears to reproduce in both the North and
South strips and in two independent redshift ranges, its statistical
significance is still marginal and more data is needed to test further its
reality.
We compare the QSO power spectra to CDM models to obtain a constraint on the
shape parameter, Gamma. For two choices of cosmology (Omega_m=1, Omega_Lambda=0
and Omega_m=0.3, Omega_Lambda=0.7), we find the best fit model has Gamma~0.1
+0.1.

When the 2dF QSO Redshift Survey (2QZ) is complete, a powerful geometric test
for the cosmological constant will be available. By comparing the clustering
along and across the line of sight and modelling the effects of peculiar
velocities and bulk motions in redshift space, geometric distortions, which
occur if the wrong cosmology is assumed, can be detected.
In this paper we investigate the effect of geometric and redshiftspace
distortions in the power spectrum parallel and perpendicular to the observer's
line of sight. Ballinger et al. developed a model to estimate the cosmological
constant, $\Lambda$, and the important parameter $\beta \approx
\Omega_m^{0.6}/b$ from these distortions. We apply this model to a detailed
simulation of the final 25k 2QZ, produced using the Virgo Consortium's huge
{\it Hubble Volume} Nbody $\Lambda$CDM light cone simulation. We confirm the
conclusions of Ballinger et al.; the shape of the redshiftspace and geometric
distortions are very similar. When all the uncertainties are taken into account
we find that only a joint $\Lambda  \beta$ constraint is possible.
By combining this result with a second constraint based on mass clustering
evolution, however, we can make significant progress. We predict that this
method should allow us to constrain $\beta$ to approximately $\pm0.1$, and
$\Omega_{m}$ to $\pm0.25$ using the final catalogue. We apply the method to the
2QZ catalogue of 10000 QSOs and find that this incomplete catalogue marginally
favours a $\Lambda$ cosmology.

We examine the highest S/N spectra from the 2QZ 10k release and identify over
100 new lowionisation heavy element absorbers; DLA candidates suitable for
higher resolution followup observations. These absorption systems map the
spatial distribution of highz metals in exactly the same volumes that the
foreground 2QZ QSOs themselves sample and hence the 2QZ gives us the unique
opportunity to directly compare the two tracers of large scale structure. We
examine the crosscorrelation of the two populations to see how they are
relatively clustered, and, by considering the colour of the QSOs, detect a
small amount of dust in these metal systems.

We describe a method from which cosmology may be constrained from the 2QZ
Survey. By comparing clustering properties parallel and perpendicular to the
line of sight and by modeling the effects of redshift space distortions, we are
able to study geometric distortions in the clustering pattern which occur if a
wrong cosmology is assumed when translating redshifts into comoving distances.
Using mock 2QZ catalogues, drawn from the Hubble Volume simulation, we find,
that there is a degeneracy between the geometric and the redshiftspace
distortions that makes it difficult to obtain an unambiguous estimate of
Omega_{m}(0) from the geometric tests alone. However, we demonstrate a new
method to determine the cosmology which works by combining the above geometric
test with a test based on the evolution of the QSO clustering amplitude. We
find that we are able to break the degeneracy and that independent constraints
to +20% (1 sigma) accuracy on Omega_{m}(0) and +10% (1 sigma) accuracy on
beta_{QSO}(z) should be possible in the full 2QZ survey.
Finally we apply the method to the 10k catalogue of 2QZ QSOs. The smaller
number of QSOs and the current status of the Survey mean that a strong result
on cosmology is not possible but we do constrain beta_{QSO}(z) to 0.35+0.2. By
combining this constraint with the further constraint available from the
amplitude of QSO clustering, we find tentative evidence favouring a model with
nonzero Omega_{Lambda}(0), although an Omega_{m}(0)=1 model provides only a
marginally less good fit. A model with Omega_{Lambda}(0)=1 is ruled out. The
results are in agreement with those found by Outram et al. using a similar
analysis in Fourier space. (Abridged)