1: \begin{abstract}

2: We study gravitational lensing with a multiple lens plane approach, proposing a simple analytical model for the probability distribution function (PDF) of the dark matter convergence, $\kappa$, for the different lens planes in a given cosmology as a function of redshift and smoothing angle, $\theta$. The model is fixed solely by the variance of $\kappa$, which in turn is fixed by the amplitude of the power spectrum, $\sigma_8$.  We test the PDF against a high resolution Tree-Particle-Mesh  simulation and find that it is far superior to the Gaussian or the lognormal, especially for small values of $\theta \ll 1'$ and at large values of $\kappa$ relevant to strong lensing. With this model, we predict the probabilities of strong lensing by a single plane or by multiple planes. We find that for $\theta \sim 10''$, a single plane accounts for almost all ($\sim 98\%$) of the strong lensing cases for source redshift unity. However, for a more typical source redshift of $4$, about $12\%$ of the strong lensing cases will result from the contribution of a secondary clump of matter along the line of sight, introducing a systematic error in the determination of the surface density of clusters, typically overestimating it by about $2-5\%$.  We also find that matter inhomogenieties introduce a dispersion in the value of the angular diameter distance about its cosmological mean. The probable error relative to the mean increases with redshift to a value of about $8\%$ for $z\simeq 6$ and $\theta \sim 10''$. 

3: \end{abstract}

4: