Discrete fracture network representations of discontinuities in rock masses have been shown to be useful in capturing heterogeneity in rock mass properties. Providing computational efficiency in the resulting simulations and analyses is attained, these fracture representations can be combined with structural modelling and sampling algorithms. Multiple fracture network realisations can be generated and the resulting rock mass properties interrogated. Statistical analyses based on fracture connectivity, block size distribution and slope stability can be performed and provide results defined in terms of confidence intervals. For sedimentary geology consisting of dense bedding, equivalent medium continuum methods have traditionally been used in preference to discrete fracture representations due to the large numbers of structures involved and resulting computational complexity. In this paper, it is shown that stochastic representation of these layers can be employed. An analytical solution to accommodate bedding given an assumed block size distribution has been derived. Using this formulation, polyhedral modelling has been used to investigate the influence of bedding on block formation and block size distributions using field data. It is shown that the analysis is both computationally efficient and can capture truncation of size distribution by such layers without numerical methods.