Natural neighbor can adaptively identify clusters of arbitrary shape. However, it is often difficult to obtain satisfactory clustering results when dealing with complex datasets. To solve this issue, this paper proposes a Hierarchical clustering algorithm based on fusion rule of competitive neighborhood graph and genetic distance (HCCG-GD). Firstly, HCCG-GD uses the competitive neighborhood graph to divide datasets into sub-clusters, effectively identifying the core regions of the data and optimizing the performance of clustering. Then, it employs a novel fusion rule based on the concept of genetic distance from biology to allocate remaining points and dynamically merge sub-clusters. The fusion rule ensures that sparse region data are correctly allocated by using the minimum genetic distance to guide the merging process. Numerical experiments based on synthetic and real datasets demonstrate the effectiveness and superiority of HCCG-GD.