The paper has been withdrawn by the authors due to some error in the manuscript.

Recent redshift-distance measurements of Type Ia supernovae (SNe Ia) at cosmological distances suggest that two-third of the energy density of the universe is dominated by dark energy component with an effective negative pressure. This dark energy component is described by the equation of state $p_{x} = w \rho_{x}$ $(w \geq -1)$. We use gravitational lensing statistics to constrain the equation of state of this dark energy. We use $n(\Delta\theta)$, image separation distribution function of lensed quasars, as a tool to probe $w$. We find that for the observed range of $\Omega_m \sim 0.2 - 0.4$, $w$ should lie between $-0.8 \leq w \leq -0.4$ in order to have five lensed quasars in a sample of 867 optical quasars. This limit is highly sensitive to lens and Schechter parameters and evolution of galaxies.

In this paper, observational constraints on the cosmic equation of state of dark energy ($p = w \rho$) have been investigated using gravitational lensing statistics. A likelihood analysis of the lens surveys has been carried out to constrain the cosmological parameters $\Omega_{m}$ and $w$. Constraints on $\Omega_{m}$ and $w$ are obtained in three different models of galaxy evolution: no evolution model (comoving number density of galaxies remain constant), Volmerange and Guiderdoni model and fast merging model. The last two models consider the number evolution of galaxies in addition to the luminosity evolution. The likelihood analysis shows that for the no-evolution case $w \leq -0.04$ and $\Omega_{m}\leq 0.90$ at $1\sigma$ (68% confidence level). Similarly for the Volmerange $&$ Guiderdoni Model the constraints are $w \leq -0.04$ and $\Omega_{m} \leq 0.91$ at $1 \sigma$. In fast merging model the constraint become weaker and it allows almost the entire range of parameters. For the case of constant $\Lambda$ ($w =-1$), all the models permit $\Omega_{m} = 0.3$ with 68% CL which is consistent with the value of $\Omega_{m} $ inferred from various other cosmological observations.

Explaining the formation and evolution of galaxies is one of the most challenging problems in observational cosmology. Many observations suggest that galaxies we see today could have evolved from the merging of smaller subsystems. Evolution of galaxies tells us how the mass or number density of the lens varies with cosmic time. Merging between the galaxies and the infall of surrounding mass into galaxies are two possible processes that can change the comoving number density of galaxies and/or their mass. We consider five different evolutionary models of galaxies .These models are: Non evolutionary model, Guiderdoni and Volmerange model, fast merging, slow merging and mass accretion model. We study the gravitational lens image separation distribution function for these models of evolving galaxies. A comparison with data for lensed quasars taken from the HST Snapshot Survey rules out the fast merging model completely as this model produces a large number of small-separation lenses. It is possible that the mass accretion model and the non evolutionary model of galaxies may be able to explain the small angle separations.

Identification of gravitationally lensed Gamma Ray Bursts (GRBs) in the BATSE 4B catalog can be used to constrain the average redshift $ < z >$ of the GRBs. In this paper we investigate the effect of evolving lenses on the $< z >$ of GRBs in different cosmological models of universe. The cosmological parameters $\Omega$ and $\Lambda$ have an effect on the $< z >$ of GRBs. The other factor which can change the $< z >$ is the evolution of galaxies. We consider three evolutionary model of galaxies. In particular, we find that the upper limit on $< z >$ of GRBs is higher in evolving model of galaxies as compared to non-evolving models of galaxies.

We study the effect of the cosmological constant on the statistical properties of gravitational lenses in flat cosmologies (Omega_{0}+lambda_{0} = 1). It is shown that some of the lens observables are strongly affected by the cosmological constant, especially in a low--density universe, and its existence might be inferred by a statistical study of the lenses. In particular, the optical depth of the lens distribution may be used best for this purpose without depending much on the lens model. We calculate the optical depth (probabilty of a beam encountering with a lens event) for a source in a new picture of galaxy evolution based on number evolution in addition to pure luminosity evolution. It seems that present day galaxies result from the merging of a large number of building blocks. We have tried to put limit on the cosmological constant in this new picture of galaxy evolution. This evolutionary model of galaxies permits larger value of cosmological constant.