The present paper deals with the following three aspects: 1. It attempts to discuss the problems on primitive forms of the family Araliaceae. The genus Tupidanthus Hook. f. & Thoms. was considered by H. Harms (1894) and H. L. Li (1942) as primitive, whilst another genus Plerandra A. Gray was regarded as primitive by R. H. Eyde & C. C. Tseng in 1971. Having made a detailed comparison of the taxonomical characters of these two genera, the present authors believe that both genera are not the most primitive in the Araliaceae. Their affinit yis not close enough and they possibly evolved in parallel lines from a common ancestor which is so far unknown yet. 2. By studying the systems of the past, the present authors believe that none of them is entirely satisfactory. Bentham (1867) recognized five ‘series’ (in fact, equivalent to ‘tribe’ with the ending-eae of names) based on the petaline arrangement in the bud, the numbers of stamen and the types of endospem. This is a plausible fundamental treatment for the Araliaceae, but choosing the endosperm as a criteria in dividing tribe is artifical. As we know today, both ruminate and uniform endosperm are usually presente in the same genus. Seemann’s system (1868) divided the Hederaceae (excl. Trib. Aralieae) into five tribes, in addition to the locules of ovary. The criteria are essentially the same as Bentham’s. The system of Hams (1894) divided the family into three tribes. Two tribes, Aralieae and Mackinlayeae, of Bentham are retained, but other groups were combined in the Trib. Schefflereae. However, Harms did not retain one of those three oldest legitimate names which had named by Bentham, that is contrary to the law of priority in the International Code of Botanical Nomenelature. Hutchinson (1967) adopted seven tribes for the family. The criteria essentially follow those of Bentham, but the inflorescence is overstressed. The inflorescence is an artifical taxonomical character in dividing tribes, because of some dioecious plants, such as Meryta sinclairii (Hook. f.) Seem., have two types of inflorescence in male and female plants. According to Hutchinson’s arrangement, the male and female plants would be put in separate tribes. 3. The present authors are of the opinion that in the study of a natural classification of plant groups emphasis should be laid not only on the characters of the reproductive organs, but on those of vegetative organs as well. The present revised system is based principally upon the characters of both flowers and leaves of the five tribes as follows: Trib. 1. Plerandreae Benth. emend. Hoo & Tseng Trib. 2. Tetraplasandreae Hoo & Tseng Trib. 3. Mackinlayeae Benth. Trib. 4. Aralieae Benth. Trib. 5. Panaceae Benth. emend. Hoo & Tseng

1. Based upon the analyses in the floristic elements of the three genera (Suaeda, Salsola and Zygophyllum) in different regions we can see that the genesis of our desert floras in these regions is very much diversified. The flora of Songaria is similar to that of the Middle Asia, while the Hosi Corridor seems to be a transitional area very close to Alashan and also related to the Tarim Basin in floristic elements. Thus, we may classify the desert floras into three parts: the flora of Songaria, of Alashan including the Hosi Corridor and of the Tarim Basin including the Tsaidam Basin. The ages and approaches in their formation are different. 2. There are plenty species but no or rare endemics in Songaria. In spring there are a number of ephemeral plants. The variation of aspect is evident. The vegetation cover is abundant. The floristic elements are developed from the flora of Middle Asia and it was formed in Quaternary period. 3. The floristic elements of the Tarim Basin are poor, but there are not few endemics and the distribution of the endemics is much limited. They are of the characteristics of relic species. Therefore it was formed in the Tertiary period and developed in Quaternary period. The elements are related to the Mediterranean flora. 4. There are a large number of endemics and many endemic monotypie genera in Alashan. They represent the flora formed in Tertiary period. Although it is of a special style, it relates both to the Middle-Asian and the Mediterranean flora. 5. The historic causes for the formation of the different floras lie chiefly on: (1) The rise of the Tibetan plateau and mountains strongly changed the climatic and edaphic conditions and in the long course of evolution some species survived or even developed, while the others deteriorated or even died out from the flora. (2) Because the circumstances of transgression or regression of the Tethys were different in these regions. (3) The mountain-making movement, the transgression and regression and the fluence of glaciation, all the mutation of these associated factors modified the climatic zonation and then the plant species changes followed, new species formed and migration of floristie elements occurred. (4) Songaria is the nearest region to the then Sibirian glacier, so the frozen injury to the flora might be the greatest. (5) In the Glacial period the descension of snow line in Songaria was greater than that of the Tarim Basin, so the frozen injurymight be greater.

Xizang (Tibet) is rich in Leguminosae flora, comprising 41 genera and 254 species so far known, exclusive of the commonly cultivated taxa (including 11 genera and 16 species). There are 4 endemic genera (with 8 species), 10 temperate genera (with 175 species) and 19 tropical genera (with 46 species) as well as the representatives of those genera whose distribution centers are in East Asia-North America, Mediterranean and Central Asia. 1. There are altogether 4 endemic genera of Leguminosae in this region. According to their morphological characters, systematic position and geographical distribution, it would appear that Salweenia and Piptanthus are Tertiary paleo-endemics, while Straceya and Cochlianths are neo-endemics. Salweenia and Piptanthus may be some of more primitive members in the subfamily Papilionasae and their allies are largely distributed in the southern Hemisphere. The other two genera might have been derived from the northern temperate genus Hedysarum and the East Asian-North American genus Apios respectively, because of their morphological resemblance. They probably came into existanc during the uplifting of the Himalayas. 2. An analysis of temperate genera There are twelve temperate genera of Leguminosae in the region, of which the more important elements in composition of flora, is Astragalus, Oxytropis and Caragana. Astragalus is a cosmopolitan genus comprising 2000 species, with its center distribution in Central Asia. 250 species, are from China so far known, in alpine zone of Southwest and Northwest, with 70 species extending farther to the Himalayas and Xizang Plateau. Among them, there are 7 species (10%) common to Central Asia, 12 species (15.7%) to Southwest China and 40 species (60%) are endemic, it indicates that the differentiation of the species of the genus in the region is very active, especially in the subgenus Pogonophace with beards in stigma. 27 species amounting to 78.5% of the total species of the subgenus, are distributed in this region. The species in the region mainly occur in alpine zone between altitude of 3500—300 m. above sea-level. They have developed into a member of representative of arid and cold alpine regions. The endemic species of Astragalus in Xizang might be formed by specialization of the alien and native elements. It will be proved by a series of horizontal and vertical vicarism of endemic species. For example, Astragalus bomiensis and A. englerianus are horizontal and vertical vicarism species, the former being distributed in southeast part of Xizang and the latter in Yunnan; also A. arnoldii and A. chomutovii, the former being an endemic on Xizang Plateau and latter in Central Asia. The genus Oxytropis comprises 300 species which are mainly distributed in the north temperate zone. About 100 species are from China so far known, with 40 species extending to Himalayas and Xizang Plateau. The distribution, formation and differentiation of the genus in this region are resembled to Astragalus. These two genera are usually growing together, composing the main accompanying elements of alpine meadow and steppe. Caragana is an endemic genus in Eurasian temperate zone and one of constructive elements of alpine bush-wood. About 100 species are from China, with 16 species in Xizang. According to the elements of composition, 4 species are common to Inner Mongolia and Kausu, 4 species to Southwest of China, the others are endemic. This not only indicates that the species of Caragana in Xizang is closely related to those species of above mentioned regions, but the differentiation of the genus in the region is obviously effected by the uplifting of Himalayas, thus leading to the formations of endemic species reaching up to 50%. 3. An Analysis of Tropical Genera There are 19 tropical genera in the region. They concentrate in southeast of Xizang and southern flank of the Himalayas. All of them but Indigofera and Desmodium are represented by a few species, especially the endemic species. Thus, it can be seen that they are less differentiated than the temperate genera. However, the genus Desmodium which extends from tropical southeast and northeast Asia to Mexio is more active in differentiation than the other genera. According to OhaShi_,s system about the genus in 1973, the species of Desmodium distributed in Sino-Himalaya region mostly belong to the subgenus Dollinera and subgenus Podocarpium. The subgenus Dollinera concentrates in both Sino-Himalaya region and Indo-China with 14 species, of which 7 species are endemic in Sino-Himalaya. They are closely related to species of Indo-China, southern Yunnan and Assam and shows tha tthey have close connections in origin and that the former might be derived from the latter. Another subgenus extending from subtropical to temperate zone is Podocarpium. Five out of the total eight species belonging to the subgenus are distributed in Sino-Himalaya and three of them are endemic. An investigation on interspecific evolutionary relationship and geographic distribution of the subgenus shows that the primary center of differentiation of Podocarpium is in the Sino-Himalaya region. Finally, our survey shows that owing to the uplifting of the Himalayas which has brought about complicated geographic and climatic situations, the favorable conditions have been provided not only for the formation of the species but also for the genus in cer-tain degree.

1) The Compositae in Tibet so far known comprise 508 species and 88 genera, which nearly amounts to one fourth of the total number of genera and one third of the total number of species of Compositae in all China, if the number of 2290 species and 220 genera have respectively been counted in all China. In Tibet there are all tribes of Compositae known in China, and surprisingly, the large tribes in Tibetan Compositae are also large ones in all China and the small tribes in Tibet are also small ones in all China. Generally speaking, the large genera in Tibet are also large ones in all China and the small genera in Tibet are likewise small ones in all China. In this sense it is reasonable to say that the Compositae flora of Tibet is an epitome of the Compositae flora of all China. In the Compositae flora of Tibet, there are only 5 large genera each containing 30 species or more. They are Aster, Artemisia, Senecio, Saussurea and Cremanthodium. And 5 genera each containing 10—29 species. They are Erigeron, Anaphalis, Leontopodium, Ajania, Ligularia and Taraxacum. In addition, there are 77 small genera, namely 87% of the total of Compositae genera in Tibet, each comprising 1—9 species, such as Aja-niopsis, Cavea and Vernonia, etc. 2) The constituents of Compositae flora in Tibet is very closely related to those of Sichuan-Yunnan provinces with 59 genera and 250 species in common. Such a situation is evidently brought about by the geographycal proximity in which the Hengtuang Shan Range links southeastern and eastern Tibet with northern and northwestern SichuanYnnnan. With India the Tibetan Compositae have 59 genera and 132 species in common, also showing close floristic relationships between the two regions. Apparently the floristic exchange of Compositae between Tibet and India is realized by way of the mountain range of the Himalayas. The mountain range of the Himalayas, including the parallel ranges, plays a important role as a bridge hereby some members of the Compositae of western or northern Central Asia and of the northern Africa or of western Asia have migrated eastwards or southeastwards as far as the southern part of Fibet and northern part of India, or hereby some Compositae plants of eastern and southeastern Asia or Asia Media have migrated northwestwards as the northern part of Central Asia. Some of the species and genera in common to both Tibet and Sinjiang indicate that this weak floristical relationship between these regions is principally realized through two migration routes: one migration route is by way of the Himalayas including the parallel ranges to Pamir Plataeu and Tien Shan, or vice versa. The other migration route is by way of northern Sinjiang to Mongolia, eastern Inner Mongolia, southwards to Gansu, Qinghai (or western Sichuan), eastern Tibet up to the Himalayas, or vice versa. However, Tibet is not entirely situated at a migration crossroad of the floral elements. An ample amount of the data shows that Compositae flora have a particular capability of development in Tibet. of the total number of species of Tibetan Compositae, 102 species and 1 genus (Ajaniopsis Shih) are endemic. Besides, 8 genera are regional endemics with their range extending to its neighbourhood. The higher percentage of endemics at specific level than at generic in Tibetan Compositae may be a result of active speciation in response to the new enviromental conditions created by the uplifting of the Himalayas. The flora in Tibetan Plateau as a whole appears to be of a younger age. 3) The uprising of the Himalayas and of the Tibetan Plateau accompanied by the ultraviolet ray radiation, the microthermal climate and the high wind pressure has, no doubt, played a profound influence upon the speciation of the native elements of Tibetan Compositae. The recent speciation is the main trend in the development of the Com-positae flora native in Tibet in the wake of upheaval of the plateau.

According to the information including type material with regard to Raphiolepis major Card. and R. indica (L.) Lindl., the morphological differences between these two closely related taxa are largely quantitative, i.e., the former has larger leaves, flowers and fruits, and longer petioles and pedicels than the latter. But since these characteristics are rather variable, they can hardly be proven to be sound in separating these two taxa as distinct. By using methods of quantitative analysis with scatter diagrams on herbarium materials, we have found that the characteristics used to discriminate the two taxa either tend to display continuous variational patterns, or vary at random. Both eases suggest that the morphological variations of individual sample plants are probably due to the different modes of growth or development of plants in response to varied environments. Since these two taxa are taxonomically inseparable from each other as a whole, it seems advisable to treat them as one single species. Thus, the lately published R. major Card.should be reduced to a synonym of R. indica (L.) Lindl.

The genus Lethariella subgen. Chlorea from Qin Ling Mountain Shaanxi province has been studied according to morphological characters and reactions of chemicals. The problem of grouping the species under the subgenus Chlorea has been discussed in thispaper.

Hooker (1882) has reported three varieties in Bidens pilosa species i. e., pilosa Proper, Linn. bipinnata Linn. and decomposita Wall. Clarke (1876) has included var. decomposita into B. pilosa proper. Taxonomic position of varieties pilosa proper and bipinnata is much confusing in literature. Babu (1977)has raised varieties pilosa proper and bipinnata as distinct species level into B. pilosa and B. bipinnata respectively. The present taximetric study also justified the distinction of these varieties as specific level. Hooker has reported three varieties of species B. pilosa i.e., pilosa proper Linn., bipinnata Linn., and decomposita Wall. from the Himalayan region of India. Clarke (1876) has included var. decomposita into B. pilosa proper. During taxonomic work on the family Compositae author has found that taxonomic position of var. pilosa proper and bipinnata is much confusing in literature. Babu (1977) has raised varieties pilosa proper and bipinnata as distinct species level. Looking to this discretion, it was de-eided to use taximetrics in interpreting taxonomic status of these taxa.