The value of cytology to the systematist in phylogenetic studies has been pointed out by many workers (Stebbins 1950) and the specific aid that the Commelinaceae might derive from such a study was suggested by Anderson and Sax (1936) in their cytological monograph of the family. Studies in the cytology and the taxonomy of the family have not been abundant in the past fifteen years but some have been made. The most recent taxonomic treatment was that of Woodson (1942) in which some generic changes were made and the tribe Tradescantieae was delimited as it is used in this study. The tropical members of this tribe are poorly understood, both cytologically and taxonomically, and monographs in both phases are badly needed. It is hoped that this study may be of some value toward that end and will bring the cytological literature up to date. Cytological studies were made using the smear technique and the chromosomes were stained in aceto-carmine or orcein. Herbarium specimens with photomicrographs have been made for most of the species studied and these are deposited in the Missouri Botanic Garden Herbarium. Acknowledgment is due R. E. Woodson, U. T. Waterfall, and H. J. Kidd for assistance in identifying these materials. Tripogandra. A large number of genera have been proposed in the past to exclude different elements from Tradescantia because of the dissimilarity of the stamens or the absence of foliaceous bracts. Woodson (1942) places them all in this genus. Two basic chromosome numbers, 6 and 8, are known. Eight appears to be widespread but at present only two cases of six are known. Among those with a basic number of eight, two have been previously recorded, T. disgrega by Darlington (1937), and T. grandiflora by Wu (1941). T. disgrega is a diploid with chromosomes of medium size, of which one is sub-median and seven are sub-terminal in their primary constrictions. T. grandiflora is a tetraploid also with medium sized chromosomes of which 14 have terminal or sub-terminal constrictions and two are median. T. elongata has been reported as 2n = ca. 50 (Simmonds 1954) and in all probability has eight as a basic number. In this study T. pflanziil and T. amplexicaulis are found to be diploids I The identity of this species aiid its inclusion in this genus is questionable. The material was received as Descantaria pflanzii but it has not been checked against the original description. It has many characteristics of Tripogandra cutmanensis and also some of T. elongata but is certainly not typical of either. To my knowledge the species has not been previously treated as Tripogandra, nevertheless it has been included here in this study beause of its cytological and morphological similarities. 30