In this work examined were pollen grains of 6 species of Nomocharis and 4 species of Lilium, all under LM and SEM, but Nomocharis saluenensis Balf. f. under TEM. The pollen grains are monocolpate (distal) in Nomocharis, monocolpate or 2-3-porate in the species of Lilium examined. Two types of pollen grains in Nomocharis are recognized based upon the exine stratification and sculpture. The pollen characters and gross morphology, and their evolutionary trends in both Nomocharis and Lilium are discussed. Our conclusions are as follows: 1. Lilium henrici Franch. 、L. souliei (Franch.) Sealy 、L. lophophorum (Bur. et Franch.) Franch. and L. nanum Klotz. et Garcke differ from Nomocharis in the aperture and sculpturing elements. Their pollen morphology supports their being placed in Lilium. 2. The pollen grains of both Lilium lophophorum (Bur. et Franch.) Franch. and L. nanum aggregate are monocolpate or 2-3-porate, which shows an evolutionary trend of aperture from monocolpate to porate, as suggested by Walker and others [Walker et al., (1975), and Chanda (1979)]. 3. In Balfour’s (1918) classification system of Nomocharis, N. pardanthina Franch. and N. meleagrina Franch. belong to Sect. Nomocharis, while N. forrestii and N. biluoensis are placed in Sect. Ecristata, but their pollen grains are all reticulate. They refer to type I (see Table 2). Thus, it does not seem reasonable placing in a single section, Sect. Ecristata, N. forrestii and N. biluoensis with the reticulate exine and N. saluenensis and N. aperta with the two-row-ritipilate exine. 4. Nomocharis saluenensis Balf. f. differs from Lilium in the exine sculpture, which seems to support placing it in Nomocharis. 5. Muri of N. forrestii Balf f. and N. biluoensis s.y. Liang show a tendency of connecting each other by heads of pili, and these two species are intergraded with the other species of the type I with respect to gross morphorlogical characters. It is thus considered that the reticulate pollen grains in Nomocharis is derived from ritipilate ones. This opinion is also held by Nair (1965). 6. Nomocharis differs from Lilium in the exine ornamentation, but some species of Nomocharis are very similar to Lilium in the gross morphology. It shows the close affinity between Nomocharis and Lilium.

The Gongga Mountain Region, located on the eastern fringe of QinghaiXizang Plateau and at the north-eastern end of Hengduan Mountains, is one of the well-known large mountain areas in Sichuan Province. There are forty five high peaks with the elavation of 6000 m or over in the area, among which the Gongga Mountain is the highest one, with its summit being at the altitude of 7556 m, whereas the Dadu River Valley in the eastern part of the area is only 1150 m above sea level; The relative height in the region is thus about 6400 m. As we know so far the orchids in Gongga Mountain Region comprise 34 genera and 75 species with 1 variety, of which 12 species are epiphytes, 59 species with 1 variety terrestrials and 4 species saprophytes (Table 1.). I. The geographical distribution in the Gongga Mountain Region. The vertical distribution of the orchids in the area. Eastern flank: There are 39 species with I variety of orchids at 1150-2300 (-2400) m alt. in the subtropical evergreen broadleaf forest zone, of which 12 species, such as Bulbophyllum andersonii, Dendrobium hancockii, Otochilus porrecta, etc., are epiphytic (including a semiepiphyte, Pleione bulbocodioides), 25 species with 1 variety, such as Bletilla formosana, Calanthe davidii, Cypripedium henryi, etc., are terrestrial, and 2 species are saprophytic (i.e. Gastrodia elata and Neottia listeroides), the upper limit of the real epiphytic orchids is 1800 m alt. At the altitude 2300(-2400)-3600 (-3800) m, in the coniferous forest zone, found are 23 species, including 20 terrestrial species with 1 variety, such as Amitostigma gonggashanicum, Calanthe fimbriata, Coeloglossum viride, etc., 2 saprophytic species (Neottia acuminata and Risleya atropurpurea) and one semiepiphytic species. There are only 2 terrestrial species (i.e. Cypripedium tibeticum and Orchis chusua) at 3600(-3800)-4000m alt., in alpine shrub zone, the upper limit of the terrestrial orchids being 4000 m alt. At 4000-4400 m alt., in alpine meadow zone no orchids have so far been found. Western flank: There are 14 species of orchids, such as Habenaria limprichtii, Hemipilia flabellata, Satyrium ciliatum, etc., at 2300-2800 m alt., in the river valley shrub zone, and they all belong to the terrestrial orchids. At the altitude of 28003800m, in the coniferous forest zone found are 19 species of orchids, among which 18 species are terrestrial (such as Amitostigma monanthum, Cephalanthera longifolia, Peristylus coeloceras, etc.) and one is saprophytic. There are only 4 species of terrestrial orchids (i.e. Cypripedium tibeticum, Gymnadenia orchidis, Orchis chusua and O. diantha) at 3800-4800 m alt., in the alpine shrub-meadow zone, the upper limit of vertical distribution of terrestrial orchids being 4400 m alt. Twelve species here are common to the eastern flank. II. The floristic composition of orchids in Gongga Mountain Region. 1. Twenty three species, belonging to 13 genera, are widespread in Whole EastAsian Region. 2. Eight species, belonging to 8 genera, are the elements of the Sino-Japanese Subregion. 3. Forty one species with 1 variety, belonging to 24 genera, are the elements of the Sino-Himalayan Subregion, more than five times the elements of the Sino-Japanese Subregion. 4. The floristic features of the orchids in the area. (1) The floristic elements are mainly temperate and subtropical ones. (2) The life form is mostly terrestrial. (3) The species endemic to China are prolific (35 species with 1 variety, belonging to 18 genera, are endemic to China, and 26 species with 1 variety are distributed in south-western China and its adjacent region; Amitostigma gonggashanicum is endemic to the area). The floristic composition of orchids in the area is characterized by the dominance of terrestrial species and temperate and subtropical East-Asian elements, though with a few Indo-African tropical elements (such as genus Satyrium (1 species) ). In conclusion, it may be considered that the species of orchids are abundant and floristic elements are comparatively complex in the Gongga Mountain Region.

Some species of Hemerocallis were cultivated in the campus of Futan University, including the evergreen H. aurantiaca Baker. The original plant of H. aurantiaca was introduced from Zhangzhou, southeastern Fukien Province. The karyotype formula of H. aurantiaca is 2n=33=12m+9sm+3st+3T+6m(sat). The karyotype differs from those of summer-green or evergreen H. fulva (k(2n)=33=3M+21m+6sm +3T) and H. fulva var. kwanso (K(2n)=33=3M+l8m+6sm+3st+3T). The vouchers are kept in FUS.

The Astilbe Buch.-Ham. ex D. Don was founded in 1912. There are now 18 species throughout the world. USSR, Thailand, Indonesia, Bhutan, Nepal d Kashanmir each has only 1 species; Korea, Philippines and USA each has 2; India 3; Japan 6; and China 7 (including 3 endemics and 1 new variety). And northeast China, north China and northwest China each has 2; central China and southwest China each 4; eastern China 5. Thus the distribution centre of this genus seems to be in the region covering Japan and eastern, central, and southwest China. This genus is divided into two sections: Sect. Simplicifoliae Engl. and Sect. Astilbe. Sect. Simplicifoliae may be considered as the primitive one because it has 5 ordinary petals. This section consists of about 10 species: 5 in China (east China 4; southwest China 3; central China, north China and northeast China each 2; Northwest China 1), 5 in Japan, 2 in Korea, 1 in Philippines and India each. According to the distribution of this section, the author suggests that the centre of origin of this genus be in the forested parts from Japonthrough east China to southwest China.

A new subtribe of the family Urticaceae, Sphaerotylinae C. J. Chen, a new genus, Sphaerotylos C. J. Chen and its type species S. medogensis C. J. Chen from south-eastern Xizang (Tibet) in China are described and illustrated. The subtribe is regarded as a primitive group in the tribe Boehmerieae. The ring-shaped stigma and the ball-like connective are the first discovery in the family Urticaceae. The types of the stigma of the family, their taxonomic significance and their evolution are also discussed in the present paper.

Artemisia intricata Franch. was described in 1884 by H. Franchet, but its exact systematic position, which genus in the Anthemideae does it belong to, is still not clear. After having examined all the morphological characters, I have found that it is best transferred to the genus Stilpnolepis Krasch. Its annual habit, heads arranged into corymbs, similar hermaphrodite and tubular florets, yellow corollas, naked receptacles and achenes without pappus all fall into the circumscription of Stilpnolepis Krasch. Thus, a new combination, Stilpnolepis intricata (Franch.) Shih, is made here in the paper. The monotypic genus Elacanthemum published in 1978 by Ling and Ling Yueou-ruenn, based on Artemisia intricata Franch., therefore, should be rejected andregarded as a synonym of the genus Stilpnolepis Krasch., and Elacanthemum intricatum (Franch.) Ling et Y. R. Ling should correspondingly be treated as another synonym of Stilpnolepis intricata (Franch.) Shih.

A new diatom species, Cyclotella asterocostata is described in the paper. The species was collected from Miyan Reservoir in Beijing, Tangshan in Hebei, Shanyang in Liaoning, Suzhou in Jiangsu, Guilin in Guangxi. It is characterized by a wide marginal region, with the width of the marginal region about half of the valve radius. There are radiate rows of costa alternate with alveola on the surface of valve central region and marginal one. The species has not an isolated punctum, but has some scattered puncta, on the surface of valve central region.

“Quan Fang Bei Zu” is a compiled work mainly for folklores, poems and other literary works concerning some common plants with some botanical information in it. It is certainly not a pure botanical work, covering no more than 240 species of plants, and thus has little use as a reference book in indexing names even in a primitive sense. Therefore “Quan Fang Bei Zu” could not be considered as a botanical dictionary. Xu Wen-xuan and his co-workers’ argument that “Quan Fang Bei Zu” was the most perfect ancient botanical codes and records till then is not convincing. Actually “Tu Jing Ben Cao” is of higher value than the book under discussion from botanical point of view.