For 75 women with squamous cell carcinoma of the vulva who underwent radical vulvectomy and inguinofemoral lymphadenectomy, the authors assessed the efficacy of four models for selecting patients who could have been treated adequately with local excision of the tumor. Each of the three models proposed by Andreasson and Nyboe covered 25% of the patients, none of whom had groin metastases or died of cancer. Local recurrence in the vulva occurred in 10%. A model suggested by the International Society for the Study of Vulvar Disease covered almost 10% of the patients. One of seven patients had groin metastases, none died of cancer, and one of seven developed local recurrence in the vulva. The criteria of the clinically best suited model are tumor not situated on the clitoris and less than 4 cm in diameter, with only slight hyperchromasia. This model ought to be tested in a randomized study.