A new species of Cyphomyrmex from Colombia, with further remarks on the genus (Hymenoptera, Formicidae). Author Kempf, W. W. text Revista Brasileira de Biologia 1968 28 35 41 http://antbase.org/ants/publications/4586/4586.pdf journal article 4586 1. Subgenus Cyphomannia Weber - This optional category was founded upon the lone Bolivian species laevigatus Weber . It is indeed a striking species. The completely smooth thorax, devoid of the costumary tubercles, teeth and spines is unique. Nevertheless, the overall appearance and a set of characters which to my mind are useful for infrageneric grouping (without going into subgenera) show that laevigatus falls nicely into the rimosus-group, in a stricter sense, i. e. excluding the somewhat aberrant forms such as longiscapus , costatus , wheeleri and possible allies. The head of laevigatus is practically identical with that of bicornis , and the latter also lacks a pronotal and epinotal armature, only the mesonotum having an anterior pair of conical low spines and a posterior pair of inconspicuous welts; in addition, the petiole and postpetiole of bicornis are practically smooth as in laevigatus (cf. Kempf, 1966: 177-9, figs. 4, 5, 23, 26, 32, 33). Moreover, Weber is wrong in assuming that when I first proposed this case of synonymy (Kempf, 1962) I had no direct knowledge of laevigatus-material. Already in 1961 (p. 518) I mentioned the specimens from Dutch Guiana which are perfectly identical with the types, available to me already by the end of 1962. Since Cyphomyrmex contains other striking and morphologically isolated species (v. gr. occultus , morschi , longiscapus ), there would be no end for decorative subgenera. Therefore, I consider Cyphomannia a useless burden which should stay buried for the sake of healthy taxonomy.