The validity of the species Absidia ramosa (sporangiospores regularly elliptical to ovoid) or its synonymity with A. corymbifera (sporangiospores spherical to irregularly ovoid) was studied in 56 pertinent strains including the 2 neotype strains described by Ellis & Hesseltine (1966). For each strain, the length and breadth of 200 unselected sporangiospores were measured microphotographically and the length expressed in mean percentage of the breadth. The minimum mean was 104'9% for the most spherical-spored strain and the maximum t67-1% for the most elliptical-spored one, but the means of the remaining 54 strains were continuously distributed between the two extremes rendering any demarcation impossible or arbitrary. Independently of the mean form, morphological variation and mean size of the sporangiospores were randomly distributed among the strains. The 56 strains were crossed with each other (1540 crossings). Zygospores were produced with all strains, but there were marked differences in mating capacity, the "strongest" strain reacting positively with as many as 47 (85.5%), the "weakest" strain with only 3 of the 55 partners (3"6%). The mating capacity was reciprocally correlated with the asexual sporulation (production of sporangia and sporangiospores), but no correlation was found to exist with the similarity or diversity of the sporangiospore form of the mating partners. Accordingly A. ramosa cannot be retained but must be reduced to synonymity with A. corymbifera.Since, in too many instances, a strain C also formed zygospores in crossings with both partners A and B of other positive crossings, it was impossible to classify the strains -or +.Such "paradoxical" mating does not seem to have been observed previously in Mucorales.
