The present paper was intended to describe further details on the genus Comastoma established by the writer in 1961. W_<ETTSTEIN> (1896) separated this group from the section Amarella of Gentiana as a distinct section, giving the name Comastoma. This treatment had been taken up by many authors either under Gentiana as W_<ETTSTEIN> (1896) proposed originally or under Gentianella as S_<MITH> did (1945, in H_<YLANDER>) until 1956, when A. and D. L_<OVE> transferred it to Lomatogonium, giving emphasis on its chromosome number. As has been stated by A. and D. L_<OVE> (1956 & 1961), the haploid number of chromosomes for both Lomatogonium rotatum and so-called Gentianella tenella is actually 5, but the form of stigma, the number of vascular bundles in each corollalobe and the position and shape of fimbriated scales at corolla-orifice draw a clear cut line of demarcation between both genera. G_<ILLETT> (1957), on the other hand, considered the present genus to be a subgenus of Gentianella ; in the genus Gentianella, however, the fimbriated scales at corolla-orifice are run through by vascular bundles and the epipetalous gland is always only one, not in pairs as in Comastoma. The section Comastoma of Gentiana established by W_<ETTSTEIN> thus forms a distinct group differing from Gentiana, Gentianella and Lomatogonium. In the light of the above findings, the present writer established the new genus Comastoma, to which belong about 7 species from the circumboreal or alpine region of the Northern Hemisphere. From the point of view of phylogeny, the present genus seems to have sprung out from the common ancestor, Pregentiana in a line quite separate from that of Gentiana (Fig. 5).
抄録全体を表示