The genus Apatema Walsingham, 1900 in the Canary Islands and Madeira is revised based on external morphology, genitalia and DNA barcodes. We recognize 18 species, 13 of which are described as new: Apatema confluellum Falck & Karsholt, sp. n. (Spain: Tenerife), Apatema minimum Falck & Karsholt, sp. n. (Spain: Lanzarote and Fuerteventura), Apatema lapalmae Falck & Karsholt, sp. n. (Spain: La Palma), Apatema sallyae Falck & Karsholt, sp. n. (Spain: Gran Canaria), Apatema lanzarotae Falck & Karsholt, sp. n. (Spain: Lanzarote), Apatema mixtum Falck & Karsholt, sp. n. (Spain: Tenerife), Apatema pseudolucidum Falck & Karsholt, sp. n. (Spain: Gran Canaria), Apatema griseum Falck & Karsholt, sp. n. (Spain: Gran Canaria), Apatema grancanariae Falck & Karsholt, sp. n. (Spain: Gran Canaria), Apatema brunneum Falck & Karsholt, sp. n. (Spain: Tenerife), Apatema transversum Falck & Karsholt, sp. n. (Spain: Gran Canaria), Apatema stadeli Falck & Karsholt, sp. n. (Spain: Tenerife) and Apatema skulei Falck & Karsholt, sp. n. (Spain: Gran Canaria). The monotypic Dysallomima Gozmány, 2008, syn. n. is treated as a new synonym of Apatema and its type species Lampros coarctella Rebel, 1896 is transferred as Apatema coarctella (Rebel, 1896) comb. n. Apatema helleri (Rebel, 1910), comb. n. is transferred from the Blastobasidae. A. mediopallidum Walsingham, 1900 is removed from the list of Lepidoptera of the Canary Island. Morphology of all species is described and figured in detail. All species occurs in the Canary Islands and only one, A. fasciata (Stainton, 1859) occurs both in these islands and in the Madeiran archipelago. Records of A. mediopallidum Walsingham, 1900 from the Canary Islands are based on misidentifications. Analyses of DNA barcodes support the identification distinctiveness of the species as each species appear monophyletic, well-supported and genetically isolated in the analyses. The only exception is A. coarctella, which could not be rendered monophyletic with respect to a single A. mediopallidum specimen, when sequences from non-Canarian Apatema were included in the analysis. We argue that this result is an artefact.