报春花科植物的地理分布

ON THE GEOGRAPHICAL DISTRIBUTION OFTHE PRIMULACEAE

根据Ｔａｋｈｔａｊａｎ世界植物区系分区对报春花科２２属在世界各地以及在中国各省区的分布作了较详细的统计，在此基础上，将报春花科各属归纳为１０个分布型，认为中国西部横断山区和东西马拉雅为报春花科的现代分布中心和多样化中心；高加索—阿尔卑斯山脉为第二分布中心；中国云南、贵州南部，广西西部至越南、泰国北部和缅甸西北部山地是报春花科植物最可能的起源中心；报春花科的起源时间应在早第三纪或晚白垩纪．

Primulaceae,comprising 22 genera and about 1000 species, is found in almost alltemperate and subtropical parts of the world as well as in some tropical montane regions. Although the total range of the family is worldwide, most of the genera are muchmore restricted. The 22 genera may be divided into the following distribution patterns.1. Subcosmopolitan: Lysimaehia, Anagallis, Somolus.2. North temperate: Primula, Androsace, Cortusa, Trientalis, Glaux.3. West & Central Asia: Dionysia, Sredinskya.4. East Asia: Stimpsonia.5. Sino-Himalaya: Bryocarpum, Pomatosace, Omphalogramma.6. Northeastern Asia-Western North America disjunct: Dodecatheon.7. Eurasian-Eastern North America disjunct' Hottonia.8. South Europe-Mediterranean: Cyclamen, Soldanella. Asterolinon, Cons.9. Tropical Africa mountains: Ardisiandra.10. Temperate South America: Pellatiacesa.Following Takhtajan's floristic regions of the world, the number of genera and speciesin each region are counted. The statistics show that the regions in richness of number ofgenera and species are successively: Eastern Asiatic Region (561/12), Irano -Turanian Region(150/12), Circumboreal Region (118/13), Mediterranean Region (63/9), North AmericanAtlantic Region (29/8), Sudano-Zambezian Region (31/6), Indochinese Region (30/4),Madrean Region (26/8), Rocky Mountain Region (27/6), Hawaiian Region(11/1), MalesianRegion (9/4), Chile-Patagonian Region (6/3), Northeast Australian Region (6/3),Brazilian Region (5/3). But the genera and species arc not evenly distributed ina region. The family is, however, an outstanding example of widely distributed, but exhibiting endemism, with the bulk of itS species (with 70% genera and 60% species) concentratedin two distribution beltS: one stretches from western China along Himalayas, covering avery narrow (c. 240 km wide) strip of territory, to Kashimir, but at the eastern end, thebelt widens, including Yunnan, western Sichuan, Guizhou, upper Myanma and northernVietnam; the other stretches from Caucasus along Alps to Pyrenees, includingmontane regions of south Europe in the north and extending to the cost regionof Mediterranean in the South. There are 12 genera and more than 500 species in China.The greatest concentration of species and diversity appears in western China and EastHimalaya, while Caucasus-Alps-Pyrenees is the secondary present distribution centre. Because the Himalaya and Alps-Pysenees are, geologically speaking, quite new and lack ofprimitive taxa, these areas are not considered as the centre of origin. On the contrary, themost primitive subgenus Idiophyton of Lysimachia, which is often cited as an archaicmember of the Primulaceae that suggests a connection on the Myrsinaceae, and theprimitive section Samuelia of Androsace, section Carolinella and section Monocarpicae ofPrimula are all confined to southern Yunnan, Guizhou, Guangxi, northern Vietnam, Myanmaand Thailand. Although there were transgression and regression of the sea, this area neverentirely submerged again since it raised in paleozoic period. It seems reasonable to presumethat the Primulaceae was origionally evolved in the mountane region of SE China includingnorthern part of Thailand, Myanma and Vietnam. In Tertiary, some species of the threelarge genera,viz.Lysimachia, Androsaee and Primula, might be widely distributed in Eurasia.During the Pleistocene ice age, they retreated to and survived in the refuges in westernChina,Caucasus and mountains of South Europe.Later they were well developed in theseareas and spread rapidly along the mountain ranges,and the present distribution patternswere formed.In the complete absence of any fossil evidence, an accurate account of the time of originis impossible. In the three families of Primulales(Myrsinaceae,Theophrastaceae,Primulaceae)only a few fossil leaves were found in Tertiary stratum from Europe, NAmerica and Greenland. Most taxonomists agree that the three families are closely allied,and A.Cronquist pointed out that'No one family of the Primulales is likely to bedirectly ancestral to either of the others'.If this point of vie