Stage III, locally advanced NSCLC, represents an incredibly heterogeneous group of patients. Optimal therapy for this group is controversial and the role of surgery is not clearly defined. There have been several randomized trials over the past three decades that have helped to guide our decision-making. In patients with T3N1 tumors, surgery is the primary treatment and there is now evidence for the use of adjuvant chemotherapy. Consensus has shown that the majority of IIIB tumors are not amenable to resection. Exceptions to this are selective T4 tumors by virtue of a satellite nodule or those with isolated invasion of the spine, superior sulcus, carina or vena cava. Such tumors require technically difficult resections and are reserved for patients with excellent performance status and no evidence of N2 disease. Patients with N2 disease represent the largest proportion of patients with stage III disease. There is an increasing understanding of the importance of multi-modality therapy for N2 disease, but the exact role and timing of chemotherapy, radiation and surgery remains unclear. The role of surgery is determined by the bulk of the mediastinal node involvement. Clearly, not all N2 disease is the same. Patients with micometastatic disease and single station nodal involvement have the greatest chance for cure and surgery has a significant role in their treatment. In addition, the ability to sterilize mediastinal lymph nodes with induction therapy correlates strongly with survival. However, the ideal form and timing of induction therapy has yet to be determined. Bulky multi-station disease is frequently not amenable to surgery and is best approached with definitive chemotherapy and radiation.