Since it is difficult to study the dynamic biological aspects of brain tumour invasion using histological sections of surgical biopsy and autopsy tissues, various laboratory systems have been devised. Animal models are less than ideal as chemically-induced brain tumours suffer from the fact that they have a low incidence and a long latency, while transplanted tumours grow predominantly by expansion, due to high proliferative activity, and not by diffuse local invasion as in human brain tumours. Various in vitro assays have, therefore, been established for both migration and invasion. These include the simple scratch technique in a confluent cell monolayer, the use of cloning rings and the "Transwell" modified Boyden chamber technique. More complex, three-dimensional culture model systems have also been developed, using chick heart, optic nerve or reaggregated fetal brain as "targets" for the invasion of neoplastic glia. Each method has yielded important information on the mechanisms which underlie brain tumour invasion. Moreover, individual microenvironmental factors may be modulated in these laboratory systems to determine their influence on the migration/invasion process.