A. The genusCucumis and its close relatives in the family Cucurbitaceae have long supported a disorganized tangle of misinformation resulting from incorrect identification of plant cultures. Several disease resistance surveys and other studies on exotic species of cucurbits have reported information on unverified and often misnamed cultures, but during the past decade many identity errors inCucumis have been corrected by critical inquiry. B. The data from cross-fertility tests are compatible with data from studies on morphology, cytogenetics, phytogeography, and physiology (cucurbitacins). They indicate: 1. C. anguria var.longipes is closely related toC. anguria var.anguria and may be its progenitor. 2. C. leptodermis andC. myriocarpus are related as varieties or subspecies rather than as distinct species. 3. C. zeyheri is no closer than a subspecies toC. prophetarum, and it may be a distinct species. 4. C. dinteri andC. sagittatus are related as varieties or subspecies rather than as distinct species. 5. C.hardwickii andC. sativus are related as varieties rather than as distinct species. C. The cross-fertility studies indicate thatCucumis species can be tentatively organized into four cross-sterile groups: 1. Eight spiny-fruited, African species or varieties ofCucumis are closely related toC. anguria var.anguria, yielding partly fertile hybrids with it. 2. C. metuliferus is not closely related to any other species used in these studies. 3. C. sativus, including many cultivated forms and the spiny-fruited, seven-chromosomeC. hardwickii from India, is not closely related to any of the spiny-fruited, 12-chromosome species from Africa or their cultivated relative,C. anguria var.anguria. 4. The three African species with young fruits hairy and lacking operculate spines (C. humifructus, C. melo and its cultivated forms, and C.sagittatus includingC. dinteri) are more closely related to one another than they are to any of the other species studied. D. Crosses of C. melo with C.sagittatus yield partly developed seeds. Work with additional collections of those species and their close relatives may yield a “bridging species” or form that will permit transfer of genes from other species intoC. melo. Embryo culture techniques may be useful in culturing the minute embryos observed in these crosses. E. The abundance and specific structure of soft spines on fruits in the anguria group appear to be unrelated to cross-fertility barriers. Spine differentiation occurs within as well as between interfertile entities.