In this paper a revision is given of the Malesian species of the Crabgrasses, or Digitaria Haller ( Gramineae). The research was done at the Rijksherbarium, Leyden, while many other Herbaria were shortly visited; some field work was done in Indonesia, Australia, and Papua-New Guinea. The foundation for the study in this large and cosmopolitan genus must be Henrard’s monumental work ‘Monograph of the genus Digitaria’ (1950), which is therefore extensively cited and discussed. Henrard based his division in sections, 32 in the subgenus Digitaria, with an emphasis on the amount of spikelets per grouplet and the various types of hairs, but such a subdivision appears difficult to maintain. As only part of the species of Digitaria occurs in Malesia, not representing all sections, a new infra-generic system can not be given. As far as the sections present in Malesia are concerned, it appeared that the Biformes, Horizontales, and Parviglumae had to be united with the section Digitaria, the Remotae and Subeffusae had to be merged into one, the Remotae, while the Atrofuscae had to be included, at least partly, in the Clavipilae, here renamed Filiformes. The subgenus Solitaria is better regarded as a section of the subgenus Digitaria. The distinction between annual and perennial in the everwet tropics poses a problem; these conceptions have been maintained, but on a more clearly defined basis. In the region studied 27 species occur, 3 of which are new to science; 25 species are indigenous, 2 are introduced and established; in the key 4 more species have been taken up, which were introduced only once, but have probably vanished. One subspecies and one variety are distinguished, both under a new name. All taxa were also studied for their extra-Malcsian distribution; from the resulting synonymy it appeared that about 52 species recognized by Henrard could be reduced to 24. The genus is estimated to contain about 170 species instead of about 325. Finally, some species are discussed which were said either to occur in Malesia or to belong to Digitaria, but for which one or the other of these suggestions proved to be false. Of these taxa two species and one variety were given a new name, while one new variety could be described.