The fruit fly, Drosophila melanogaster with its well known genetics is used to detect newly induced mutations in germ cells. Assay systems are available to determine point mutations, chromosome aberrations and nondisjunction. The sex-linked recessive lethal test is of primary practical importance, because it can detect mutagens with and without chromosome-breaking ability, which induce point mutations, small deletions or aberrations having a recessive lethal effect. For routine testing X-chromosomes are exposed to a potentially mutagenic treatment while the chromosome is in a male germ cell. In two successive crosses the treated X-chromosome is transferred to males of the F2-generation. Here, males carrying the treated chromosome can be detected based on easily scorable phenotypic characteristics (eye shape and color). The X-chromosome in F2-males is hemizygous, and therefore any recessive mutation is expressed: The presence of a recessive lethal kills all these males before they have developed into a fly, and this is easily scored as the absence of one mendelian class among the progeny. The predictability of the test for established mutagens and carcinogens is quite good. In the order of 200 compounds are known for which positive results in the sex-linked recessive lethal assay were reported. Mono- and polyfunctional alkylating agents are easily detected even if they do not break chromosomes. Frameshift mutagens can be detected, but a systematic validation of the system for this type of damage would be very desirable. Of outstanding importance is the fact that a large variety of indirect acting carcinogens are easily detected in the Drosophila in vivo assay. It is assumed that the metabolic transformation takes place within the cell subsequently tested for the presence of induced mutations (e.g., spermatids). In addition to the more indirect evidence from mutagenicity tests, information is now accumulating on the biochemical characterization of xenobiotics metabolizing systems in Drosophila and on their inducibility. An important technical problem is the selection of the route of application. Depending on the aim of the experiment and on the chemical properties of the compound under test, the route of application has to be selected with care, in order to assure that a significant exposure of the germ cells to the chemical or to its metabolites is achieved. Occasionally problems are encountered with water-insoluble compounds and the use of DMSO as solvent. These problems are currently under investigation. Possible means to increase the sensitivity of the test system for polycyclic hydrocarbons and aromatic amines are also studied. The use of Drosophila as an alternative test provides information on: (i) mutations in germ cells, (ii) in vivo activation, (iii) broad mutation spectrum (base pair substitutions, frameshifts, deletions, etc.) in one single experiment, (iv) relevant dose level which can be expected to lead to a doubling of the human spontaneous mutation frequency. The Drosophila tests have to be used in all cases in which point mutations in germ cells are considered a possible risk.