With a large penetration of low carbon technologies (LCTs) at medium voltage and low voltage levels, electricity distribution networks are undergoing rapid changes. Much research has been carried out to analyse the impact of employing LCTs in distribution networks based on either real or synthetic network samples. Results of such studies are usually case specific and of limited applicability to other networks. Topological properties of a distribution networks describe how different network components are located and connected; which are critical for the investigation of network performance. However, the number of network modelling and simulation platforms are limited in the open literature which can provide random realistic representations of electricity distribution networks. Thus, it is difficult to arrive to generalized and robust conclusions on impact studies of LCTs. As the initial step to bridge this gap, this paper studies the topological properties of real-world electricity distribution networks at the medium voltage level by employing the techniques from complex networks analysis and graph theory. The networks have been modelled as graphs with nodes representing electrical components of the network and links standing for the connections between the nodes through distribution lines. The key topological properties that characterize different types (urban and sub-urban) of distribution networks have been identified and quantified. A novel approach to obtain depth-dependent topological properties has also been developed. Results show that the node degree and edge length related graph properties are a key to characterize different types of electricity distribution networks and depth dependent network properties are able to better characterize the topological properties of urban and sub-urban networks. (C) 2017 The Authors. Published by Elsevier Ltd.