Objective: To determine the optimal intercostal space (ICS) to perform thoracoscopic-assisted lung lobectomy. Study design: Cadaveric study. Animals: Six mature, medium-sized canine cadavers. Methods: Cadavers were placed in right or left lateral recumbency. A 15-mm thoracoscopic cannula was inserted in the middle third of the 9th or 10th ICS. A wound retraction device was placed into a 7-cm minithoracotomy incision created in the middle third of the 4th-7th ICS on the left side and the 4th-8th ICS on the right side. The pulmonary ligaments were sectioned by using a combined intracorporeal and extracorporeal technique. Each lung lobe was sequentially withdrawn from the wound retraction device at the respective ICS and side. A thoracoabdominal stapler was positioned to simulate lung lobectomy, and the distance from the stapler anvil to the hilus was measured. Results: Simulated thoracoscopic-assisted lung lobectomy performed at left or right ICS 4 and 5, compared with other ICS evaluated, resulted in a significantly shorter median distance from the stapler anvil to the pulmonary hilus of the left cranial and caudal lung lobes and right cranial and middle lung lobes, respectively (all P < .05). Lobectomy at right ICS 5 or 6 resulted in a significantly shorter median distance from the stapler anvil to the pulmonary hilus of the right caudal and accessory lung lobes, respectively (both P < .05). Conclusion: These data may inform minithoracotomy positioning to optimize tumor margin excision during thoracoscopic-assisted lung lobectomy for treatment of pulmonary neoplasia in dogs. Clinical significance: Complete lung lobectomy is possible by using the described thoracoscopic-assisted technique in normal, cadaveric lungs.