Ophiopogon caulescens (Blume) Backer, O. prolifer Lindl., O. malayanus Ridl., O. dracaenoides (Baker) Hook.f. and O. pierrei L.Rodr. are taxonomically reviewed. Both O. prolifer and O. malayanus are treated as conspecific with O. caulescens. The filaments of O. caulescens are united laterally along their entire length or about halfway from the base up. The ovaries of O. dracaenoides and O. caulescens are concave at the apex. This consistency in the ovary structure implies a close relationship between the two species. Ophiopogon dracaenoides is recorded as new to Thailand. Ophiopogon pierrei is distinguishable from O. caulescens by the free filaments and the ovary which is nearly hemispheric at the apex.

The chromosome numbers of the following twelve taxa of Deutzia have been counted: D. bungoensis (2n = 52), D. crenata var. crenata f. crenata (2n = 130), D. crenata var. crenata f. pubescens (2n = 78) and var. floribunda (2n = 78), D. gracilis var. gracilis (2n = 26), var. microcarpa (2n = 26), and var. zentaroana (2n = 52), D. maximowicziana (2n = 26), D. scabra var. scabra (2n = 26) and var. sieboldiana (2n = 26), D. uniflora (2n = 26) and D. yaeyamensis (2n = 26). Among them seven are diploid (2n = 26), two are tetraploid (2n = 52), two are hexaploid (2n = 78) and one is decaploid (2n = 130).

The previously assumed uniformity of the basic chromosome number (x = 13) for all taxa of Deutzia is supported by the data. Chromosome counts from five taxa, D. bungoensis, D. crenata var. crenata f. pubescens and var. floribunda, D. gracilis var. microcarpa and var. zentaroana, are new reports, and the counts from the other seven taxa are confirmation of previous counts. It is suggested that ploidy level differentiation is a major factor of speciation in Deutzia.

Canna plurituberosa T.Koyama & Nb.Tanaka from northern Argentina is described here with a detailed illustration. Chromosome number of 2n = 18 was also reported.

An East Asian Vicia bungei Ohwi and a North American V. americana Willd. are phenotypically and ecologically similar to each other. Therefore, V. bungei has been treated as V. americana var. sinensis Gunn. However, the former species has root tubers but the latter has not. Vicia bungei should be treated as a distinct species.

×Toisochosenia kamikotica (Kimura) Kimura was recognized as a presumed hybrid between Toisusu urbaniana (Seem.) Kimura and Chosenia arbutifolia (Pall.) Skvortsov. It was described in Salix based on only one male tree found by A. Kimura in Kamikochi, Azumi-mura, Nagano Prefecture in 1931. This tree, however, died and its basal trunk is still remaining in July 1999. No living plants of the presumed hybrid are believed to be known in natural habitat. Fortunately, one female tree similar to ×Toisochosenia kamikotica was found by Imai in 1997. Based on a comparison between specimens collected by Imai from this female tree and those of type and authentic specimens of ×Toisochosenia kamikotica, the Imai's tree is identified by Ohashi and Azuma as a female plant of ×Toisochosenia kamikotica (Kimura) Kimura. ×Toisochosenia kamikotica (Kimura) Kimura is important for further studies in systematics of Salicaceae on the intergeneric hybrid. Because of the extremely scarcity of the living plant in the natural habitat, however, the possibility of extinction of this species is much feared.

Galarhoeus sendaicus (Makino) H.Hara var. musashiensis Hurus. (= Euphorbia sendaica Makino var. musashiensis (Hurus.) Hurus., nom. illeg.), an extinct plant in the wild or critically endangered plant in Japan, was reexamined taxonomically. Var. musashiensis was not distinguishable from sendaica by stem length as an index of gracility as well as its creeping rhizomes and degenerative female flowers in cyathia. It suggests that musashiensis should not be distinguished from sendaica as any infraspecific taxa. E. sendaica is endemic to southern Tohoku District and Kanto District of Honshu, and inhabiting in floor or on margin of Cryptomeria japonica plantation, broad-leaved deciduous forest, or mixed forest on hills.

Acer nipponicum Hara is divided into two subspecies by the form of inflorescences; subsp. nipponicum distributed in Kinki and Chugoku districts, Shikoku and Kyushu, and subsp. orientale Yamaz. distributed in Tohoku and Chubu districts. By the indumentum of leaves, each subspecies is divided into two varieties; subsp. nipponicum var. nipponicum and subsp. nipponicum var. australe Yamaz., subsp. orientale var. orientale and subsp. orientale var. koshinense Yamaz.

Pollen grain size of 200 anemophilous angiosperms was analyzed to clarify its ecological significance in pollination syndromes. They are summarized as three types: explosive type, strong-wind type and long-filament type. Flowers with the explosive type correspond to the group of the smallest pollen size. They grow in moderately windy environment like gaps in woods. Their pollen grains can float much longer in the air after discharge from anthers to reach to stigmas. Flowers with the strong-wind type are equivalent to the group of medium pollen size. They, including many woody and weedy species occurring in wastelands, are able to make better use of stronger wind than the explosive type flowers for pollen dispersal. Flowers with the long-filament type correspond to the group of the largest pollen size. Many of them, belonging to Gramineae and Cyperaceae, make simple communities, and pollen grains can reach stigmas after short flight.