〔1〕Experimental Brain Tumors

　Experiments in gliogenesis with adult　inbred strains of mice have eliminated the　necessity to explain certain types of ana—plasia on the basis of dedifferentiation of　tumor cells．　The single cell theory of　origin of gliomas is false；　"mixed"　gliomas　develop from the simultaneous multiplica—tion of cells of different type．　The　"pure"　or single cell type of glioma is that which　arises from the proliferation of but one　single variety of cell, or it occurs when a　malignant cell type overgrows and replaces　a benign type．

　so-called microgliomas have not been　produced experimentally．

　Cells which reach a certain stage of mor—phologic and functional differentiation no　longer can be stimulated to proliferative　activity．　This is probably the reason why　most neurones are incapable of neoplastic　transformation．　The experimental tumors　which resemble the　"gangliogliomas"　are，　in reality, gliomas which have entrapped　normal neurones in their midst．

　Tumors in the cerebellum which are　indistinguishable from human medullobla—stomas have been readily produced in adult　mice．　It is reasonably certain that these　tumors do not arise from embryonic cell　rests or bipotential　"medulloblasts"．　They　are rather neuroblastomas, derived from　the neurones in the granular cell layer of　the cerebellar cortex．　Occasionally, tumors　of this variety also contain proliferating　astrocytes and are therefore　"mixed"　tu—mors．

　Abundant experimental evidence is avail—able which indicates that gliogenous tumor　cells are incapable of invading blood ves—sels．　This seems to be the reason why　this class of neoplasm dose not metastasize　spontaneously extracranially．　Transplants　of gliomas, however, grow readily in all　viscera and subcutaneously in homologous　animal species．　Intravenous transplants are　often destroyed by what appears to be a　local immune tissue response on the part　of the host, or even a humoral cytotoxic　effect on the injected tumor cells．