Bike-sharing offers significant convenience for short-distance travellers and is regarded as an effective solution to the first-and-last-mile problem. However, existing studies have largely overlooked cross-city travel within the bike-sharing context and its associated cycling spatial structures. Cross-city travel behaviour refers to the movement of a traveller from one city to another. This behaviour can be categorised into two types: a single trip from one city to another (Type I), and two consecutive trips in different cities, indicating a cross-city journey between the two (Type II). Using extensive dockless bike-sharing data from Zhejiang Province, China, this chapter employs a network-based approach to identify the cycling spatial structures arising from cross-city travel. The results reveal that cross-city cycling is less common compared to intra-city travel, and it is more challenging to travel between prefecture-level cities than between county-level cities. The cycling spatial structures differ considerably across city levels and cross-city travel types. Type II patterns tend to be more intricate and interconnected, forming well-linked city networks, while Type I patterns are generally simpler. At the county level, cross-city patterns reveal approximately ten distinct spatial communities, some of which exhibit a misalignment between community-based and administrative boundaries. This study enhances our understanding of cross-city travel behaviour and cycling spatial structures. It also provides insights for policymaking and practical operations, such as the design of subscription packages and the reconfiguration of operational areas.