Withanaarachchi,W.A.K.D.H.Almeida, S.V.A.Wijesiri, G.S.2026-06-082026-06-082023-11-03Proceedings of the Postgraduate Institute of Science Research Congress (RESCON) -2023, University of Peradeniya, P 52978-955-8787-09-0https://ir.lib.pdn.ac.lk/handle/20.500.14444/7742One of the most prominent problems of topological graph theory is to determine the type of surface a nonplanar graph can be embedded. Almost complete results have been obtained for 4-edge connected graphs. The methods that were used to obtain specific results (finding the maximum and minimum genus embedding) for 4-edge connected graphs do not generalise for 3-edge connected graphs. Graph embedding is an important representational technique that aims to maintain the structure of a graph while learning low-dimensional representations of its vertices. The aim of this research project was to study the embedding of 3-edge connected Harary graphs H₃,n. Specifically to complete the problem of maximal embeddings of 3-edge connected Harary graphs. The result is proved using Jungerman’s study, which showed that for any graph 𝐺, 𝐺 is upper-embeddable if and only if it has a spanning tree T such that 𝐺 ∖ 𝑇 has at most one component with an odd number of edges. More specifically, a spanning tree for each graph was observed by dividing all 3-edge connected Harary graphs into two groups: odd number of vertices and even number of vertices. The pattern of a set of deleting edges and corresponding spanning trees was generalised in both cases. It was proved that H3,n is upper-embeddable, and the maximum genus of H₃,n is given by 𝛾𝑀 (𝐻3,𝑛) = ⌊ (2+𝑛) /4 ⌋ for each n, by analysing the odd components of the complement of the corresponding spanning trees.en-US3-Edge connected graphsHarary graphSpanning treeUpper-embeddabilityArticle