This paper presents a rigorous, unified framework based on Ordinary Differential Equations (ODEs) to study epidemic routing and its variations in sparse and highly mobile networks. The authors derive ODEs as limits of Markovian models under a natural scaling as the number of nodes increases, providing closed-form expressions for several performance metrics and numerical solutions that do not scale with the number of nodes. The framework is used to investigate the trade-offs between delivery delay and resource consumption, such as buffer space and power, and to study different buffer management schemes under buffer constraints. The paper also explores the recovery process, including packet deletion at infected nodes, and compares the performance of various epidemic routing schemes, such as $K$-hop forwarding, probabilistic forwarding, and spray-and-wait. The results show that buffer-constrained epidemic routing can achieve good performance with appropriate buffer management, suggesting that sizing node buffers to limit packet loss is not always necessary. The paper concludes with a discussion on future work, including the investigation of anti-packet deletion schemes and their overhead.This paper presents a rigorous, unified framework based on Ordinary Differential Equations (ODEs) to study epidemic routing and its variations in sparse and highly mobile networks. The authors derive ODEs as limits of Markovian models under a natural scaling as the number of nodes increases, providing closed-form expressions for several performance metrics and numerical solutions that do not scale with the number of nodes. The framework is used to investigate the trade-offs between delivery delay and resource consumption, such as buffer space and power, and to study different buffer management schemes under buffer constraints. The paper also explores the recovery process, including packet deletion at infected nodes, and compares the performance of various epidemic routing schemes, such as $K$-hop forwarding, probabilistic forwarding, and spray-and-wait. The results show that buffer-constrained epidemic routing can achieve good performance with appropriate buffer management, suggesting that sizing node buffers to limit packet loss is not always necessary. The paper concludes with a discussion on future work, including the investigation of anti-packet deletion schemes and their overhead.