Because intracranial germinomas are readily curable with radiation and chemotherapy or radiation therapy alone, the role of radical surgery has become debatable. This study assesses the optimum degree of surgical resection for intracranial germinomas. Twenty-nine patients who underwent surgery for germinoma were retrospectively analyzed (male/female ratio 27:2, median age 18 years). Among these 29 patients there were 10 solitary pineal, seven solitary neurohypophyseal/hypothalamic, and 12 multifocal or disseminated tumors. Biopsy samples were obtained in 16 patients (stereotactically in eight, transsphenoidally in four, and via frontotemporal craniotomy in four). Partial resection was attained in five patients (via a frontotemporal approach in three and occipitotranstentorially in two). Gross-total resection was achieved via an occipitotranstentorial route in eight patients with pineal masses. After surgery, 10 patients were treated with radiotherapy alone, and 19 received radiation and chemotherapy; complete remission was achieved in all 29 patients. The overall tumor-free survival rate was 100% at a median follow-up period of 42 months. There was no significant difference in outcome related to the extent of surgical resection. Postoperative neurological improvement was seen in only two patients, whereas transient postoperative complications, mainly upgaze palsy, were observed in six. One patient experienced a slight hemiparesis, bringing the surgical morbidity rate to 3% (one of 29). It is concluded that radical resection of intracranial germinomas offers no benefit over biopsy. The primary goal of surgery should be to obtain a sufficient volume of tumor tissue for histological examination. If there is strong evidence of germinoma on radiological studies, biopsy samples should be obtained. When a perioperative histological diagnosis of pure germinoma is made during craniotomy, no risk should be taken in continuing the resection.