Urban system is the component of specific urban or regional planning in China as well as in other countries. This paper is concerned with the theoretical studies and practical tasks for urban structure and allocation in China's modernizations. First of all, the auther distinguishes urban groups into three types, centripetal concentration, centrifugal decentration and centripetal centrolization.Then the definition of urban system is regarded as a group of cities and towns which exists spatial interaction and composes an organized hierarchy.The author divides China's urban systems into three levels, system of city hinterland, system of nodal economic region and system of national territory or big region. M. Beckmann's urban hierarchy and C.Zipf's ranksize rule including B. Berry's empirical research are introduced in detail. The applicability and shortcomings of these models for China's planning are fully discussed.

This Article discusses the following problems.1. When was the Chong-ming Town established? The former books of the general records of the Chong-ming county were all confused. The county records after Yong-Zheng (雍正) Dynasty assumed that the Chong-ming Town was established in 705 A.D..According to the ancient books of historical geography "Yude-jisen, 兴地纪胜" (Record of the Territory) and "Dushi-fangyu-jiyao,读史方兴纪要 (The Summary of Historical Territory), this town was established in the 10th century, in the period of Wn Kingdom of the Five Dynasties (五代时吴国).2. On the problem of five migrations, the county records after the Yong-zheng Dynasty were also confused. These records all considered that the place of county government had been migrated from the west to the east. The atlas of the "Records of the Zheng-de Dynasty 正德崇明县志 "denoted that the Island of Yao-liu Sha (姚刘沙岛) was situated at the east part of the Zhong-ming Island, and the place of the county government had been migrated first to the north, then to the west, finally from west to east. The goverment was named Chong-ming Zhow (崇明州) in 1277; and established at Yao-liu Sha Island, but later changed to Chong-ming County in 1369. The first migration was 7.5 km. norh ward in 1352. The second migration was north ward to a small town named Qin-jia-fu (秦家符) in 1420. The third migration was to the San-sha Island (三沙岛), 25 km. northwest of Yao-liu sha, in 1529. The fourth migration was to ping-yang Sha Island (平洋沙岛), 15 km. southwest of San-sha Island, in 1550. The fifth migration was to the chang-sha Island (长沙岛), southeast of San-sha Island, present Chong-ming county government since in 1583.3. The evolution of the Island may be divided into three periods:(1)The formation period of the Island, from 7-11 century. (2) The period of increasing and great changing of the numeral Islands, from 12-16 century, and (3) The period of stablity and enlargement of the Island, from 17-20 Century. By the dyke building of the Island, the Island began to be stable, and enlarged toward the northern coast at present. The northern passage of the Yang-tze River is becoming narrower gradually.

In this paper, by using the data of seven city climate stations in Beijing which were set up by ourselves from 1982-1985, and the meteorological data of the suburbs in the same period, we discussed the situation of (mean, maximum and minimum) temperature and the causes of heat island, as the rate of temperature rises and drops and the level of obstruction affects the vaaririation of temperature in the city and its suburbs. Generally, most articles investigated the phenomena of heat island with some days data, maybe three or four days, that obtained from moving observation.So it is difficult to analyse the rate of rising and dropping of temperature in the city and its suburbs, because of the shortage of long period data.On the base of the city climate data from 1982-1985, we discussed the above-mentioned problems. The main results are obtained as follows.1. The phenomena of heat island exist in four seasons of a year, but it is the strongest in the winter, the weakest in the summer, and spring and autumn is in the medium.2. The mean, maximum and minimum temperatures in the city are higher than those in the suburb. The difference between the city and suburb is about 1.2-2.4℃ from Jan.-Dec.for mean temperature.Their difference is a maximum in January of the winter, a minimum in August of the summer. For maximum temperature, city temperature is only 0.4-1.6℃ higher than the suburb from Jan.-Dec. The maximum difference between city and suburb temperature appears in the monthes of spring and autumn, May and September. In the opposite with mean temperature, the difference is a minimum in January of the winter.Although urban temperatures, regardless of mean temperature, maximum minimum temperature, are higher than that of suburb, the difference between the city and suburb minimum temperature is the greatest. Their difference is about 1.4-3.6℃ from Jan.-Dec. As the same as the mean temperature, the difference is a maximum in winter, a minimum in summer.3. The rate of temperature rising and dropping also has a difference between the city and its suburbs. The rate of rising temperature in the city is slower than that in the suburb from March-April. It is opposite from April to May, the rate of temperature in the city is faster than that in the suburb, and the rate of temperature drops faster in the city than in the suburb from September to October and slower in the city than in the suburb from October to November. We found that different rates of temperature rising and dropping between the city and suburb is a reason which causes the formation of urban heat island of Beijing.4. According to the research on the temperature of different distances from the city center, we found that the temperature is decreasing as the distance increases from the city center outwardly. The decreasing amount of temperature is not equal with distance outwardly in different direct ions. The decreasing amount with distance increasing from city center outwardly is smaller eastward or westward than southward or northward, because Beijing city is larger in width in east-west direction than that in south-north direction. This fact indicates that the level of city obstruction is an important factor in causing the formation of urban heat island.

The regionalizetion of comprehensive physical geography may be divided into two types in nature. One is knowability, and the other is applicability. The former possesses a strict system, which reflects correctly the regional variation. The aim of it is to systematize and theorize the regional data, and to promote the development of the science, The latter, according to the principle of regioal vaiation, reflects the relations between agricultural production and natural complex on the earth surface. Especially, it mainly discusses the agricultural harmful degree of flood, aridity, low temperature, strong wind and soil degeneration factors etc, and finds out the way of solving the factors. It provides basis of science for the development of agriculture production. The former has better scientific basis, and the latter has greater econmic significance.

During the years of Hong Wu and Yongle in the Ming Dynasty, there appeared several largescaled, organized migrations along the middle and lower reaches of the Yellow River. This article discusses the social background of them, the process and the social functions of such migrations. It focuses on the following points.Each migration in the early years of Ming Dynasty is the result of one of the measures taken by the rulers to prevent the economy depressed by war from decline. The removal of the country's political center (i.e. the removal of capital from Nanjing to Beijing) greatly diverts the migrations. Such changes stabilized the society then, quickeming the step of the economic blooming and flurishing, and developing the cultural intercourse among the people of the several provinces along the middle and lower reaches of the Yellow River. Especially the large scaled migrations from the area around Shanxi province into Beijing greatly balanced the population, and solidified Beijing's position in politics, economy and military superiority.

In this paper, a unmerical model is presented to calculate the reflected radiation from the surrounding terrains in mountainous areas, on the basis of our developing models for calculating the terrain parameters and global radiation over rugged sites.The major input parameters of the model are altitude and albedo data at each grid,time interval, grid-space and the long-term average of thecloud amount and the sunshine percentage at the nearest adjacent weather station to the sudied area. The output one of it is in the formof the map of its distribution.The tests, made in Dabie Mountain, in the south of China, manifest that the reflected radiation from the surrounding terrains is in general much smaller than other components of radiat ion ballance.However, it becomes so important that it must not be neglected ifthe mountainous areas in question are entirely covered with snow, which is of some representation in the north of China in winter.The comparison of the results given by the present model and the single-slope one makes it clear that they are noticeably different in magnitude and distribution, the variation of the former being more intense from place to place than that of the latter, especially if snow-melting degree is not the same in the whole studied mountainous areas.

This paper first illutrates by examples that many researches of heat-island intensity lack comparability because there arc different sources of information and different methods of research. Next it discusses city influences on climate. and explains, from urban climatic point of view, differences among built-up area suburbs and rural area, and how to choese the representive observation sites of temperature in order to guarantee that temperature of observation can reflect the influence of city on climate.On this basis, the writer puts forward and recommend the following several methods of expressing urban heat-island tintensities.1. The observation sites should be well distrubited. to ues the difference between mean temperature of built-up area and rural area to express urban heat-island intensity.2. The observation sites should be distributed on different urban funcional areas: to use the difference between weighted average of built-up area and rural area to express urban heat-island intensity,3. Determining urban heat-island intensity with regression analysis.4. Analysing urban heat-island intensity with statistical or theoretial analysis.5. Analysing urban heat-island intensity with remote sensing method.In the case of metropolis, if there are considerable differences between topography and natural landscape of built-up and rural area, it is neccessary to make temperature corrections with synthetic-seperate method according to the difference between topography and natural landscope of built-up and rural area before urban heat-island intensity is computed.

The Qinghai-Xizang Plateau, includs mainly the Qinghai province and the Xizang region. It is the highest and greatest plateau on the Earth. But no unanimous idea has been reachad about its exact extent. Based on the characteristics of the major structural landforms of this plateau and the distribution of the plateau surface higher than 4000m above sea level, the author has delimited its extent. More attention has been paid to the eastern boundary that has not been clarified for a long time.The eastern boundary of Qinghai-Xizang Plateau runs along in the middle and northern part of the Hengduan Mts, i, e, along the east and south side of Min Shan-Chaping Shan-Qionglai Shan-Jiajin Shan-Daxue Shan-Jinping Shan-Yulong Shan, It's northeast corner pesses into the east end or Qilian Mts. and western part of Qinling Mts.and its southeast corner mingles with the Hengduan Mts, In these two aveas no obvious structural and ge-omorphological boundaries exist. But according to the distribution of an almost continuous peak plane above 4000m, we can set the northeast boundary of the plateau as lying to the west of the line of Minhe-Kangle-Minxian-Wudu-Wenxian and the southeast boundary as running to the north of Lijiang-Lushui. The southern boundary is situated on the south foot of the Himalayas and the northern boundary on the north foot of western Kun-]un Mts.-Altun Mts. and Qilian Mts, The western boundary of the Plateau merges with Pamir.

The paper dealing with water quality analysis is based on long series of data which consists of 69 representative hydrochemistry stations in the Huanghe Basin.These statistical data in eludes 500 station-years.The component and pattern of natural hydrochemistry in the Huanghe Basin are rather complex. The mineralization degree of river water varies from 200 to 5000 mg/L, the mineralization degree of river water in upper and lower reaches is adequate, but the characterastics of water quality (mineralization degree, total hardness and main ions) have obvious zonal distribution in its middle reaches, and they basically conform with natural conditions and distribution trend of water and heat.

This paper studies the formation and distribution of the modern volcanoes in the depopulated areas of the northern part of Tibet by analysing acria] pictures and maps. It is important to research on the relationships between volcanic activities in the northern part of Tibet, the uplift of Tibet and plate tactonies.The distribution of the modern volcanoes in the northern part of Tibet is typified by zones on the sorthern foot of Kunlun and Aerge mountains. It extends 1050 km from west to east-most of which lie in the fault basin between 35°20'N and 36°20'N.The eruption in this region is probably resulted from the new tectonic boundary which was caused by the compression of the Tibet onto the Talimu massif by southern Asia Continent.These volcanoes are from west to east as follows.

Landscape geography as a field of comprehensive geographical research occupies an important place in the system of geographical sciences. Because of the limitations of phenomenology, close and static studies and the backwardness of study methods and techniques, traditional landscape geographical research could not profoundly reveal and explain the essential aspects of landscape features structures, especially dynamics.The research level was relatively low. Since 60's, the theoretical and methodological ideas of modern systematic sciences and ecological sciences have been adopted by landscape geographers, computer, remote sensing and mathematical modelling techniques have been applied in landscape research. These greatly promoted the development of theories and methodology of systematical Synthesis and the improvement of investigation techniques in landscape geography.In recent periods, many landscape geographers pay more attention to landscape dynamics, the experiment investigations at stationary measuring points, in representative model areas and in laboratories were strengthened. Applied landscape research were further developed. Landscape ecology becomes a very active field of landscape research. All the trends mark a significant advance of landscape geography.

自然地理研究室1984年以来分别参加了“黄淮海平原的治理和开发”、“京津生态系统特征”、“黄土高原国土整治”等攻关及重点项目;全国1:100万土地类型制图等连续工作继续在进行,几年来,在同志们的努力下,克服了人少任务重的困难,取得了一批成果。

Slopeland accounts for some 70% of the total land area of South China. With minor exceptions, its productivity is rather low and is declining as a result of soil depletion and degradation and various processes of denudation. And the off-site damages of denudation are no less serious. It is possible to maximize the productivity and maintain or improve the soil as the natural medium for plant growth with certain fast growing plants. As fast growing plants fix more carbon from the atmosphere, the corresponding requirement of nutrients from the soil is higher. For this reason, plants that fix nitrogen and plants that send roots deep to draw nutrients from the subsoil should be included in the production system. In the choice of plants, it is imperative to strike a good balance between the demands of the local inhabitants and cash product markets far away. Since capital is in general a limiting factor in China, elements allowing quick returns merit special attention. To reducerisks from natural hazards and changes in markets, diversification in the system is an essential feature, multipurpose plants in particular may serve as a device of spreading risks,Awise practice is to take a small watershed as a basic unit for planning, in which plots and belts of different functions are laid out" so as to maximize productivity without des truction, With all these points in view, the geographic background of the slopeland of South China is discussed in conjunction with Acacia auriculiformis, A, decurrens, A.mangium, A, mearnsii, A, saligna, Ailanthus altissima, Albizia falcataria, Alnus acu-minata, A, rubra, Amaranthus tricolor, Calliandra calothyrsus, Canavalia ensiformis, C, gladiata, Cassia siamea, Casuarina glauca, Crotalria juncea, Dalbergia sisso, Derris indica, Desmodium spp, Eucalyptus pellitta, Gledtsia triacanthos, Hibiscus cannabicus, Lespedeza spp, Leucaena leucocephala, Muucana spp, Psophocarpurs palustris,P, tetragonolobus, Robinia pseudoaca-cia, Sesbania grandiflora, Sorghum bicolor,S, sudnense and Trenia orientalis as examples. The exercise has led to the conclusion that there is every possibility to ameliorate the environment and enhance the overall productivity of the land. This seems beyond the shadow of a doubt, but efforts to this end may not always meet with success, unless they are supported by appropriate experimental research and areal investigations, Some priority themes for study are recommended and explained.If slopeland utilization and amelioration as recommended is successfully executed, its importance cannot be exaggerated. Recent years have seen accelerated economic development in South China, The demands for agricultural and forest products, both at home and abroad, have increased rapidly. The pressure is further accentuated by the growth of population and the shift of interest or people in the rural areas to activities other than agriculture and forestry.On the plains,the standard of living is much higher than earlier, and the man/land ratio is exceedingly high. The slopeland regions are much Jess developed and the inhabitants there are under a living condition essentilly the same as tea years ago. The problems thus arising are more and more acute, especially when the slopeland deterioration, is taken into consideration, Adop-t.ing the proposed strategy, steps to solve the problems will be directed to the right quarter.

On the basis of the principle of dam-break, the possible flooded area has been simulated under the condition of the rare flood of the Lower Yellow River. It turns out that if the breaking levee takes place at the section of Yuanyang or Caogang, the flooded area borders on the northern levee of the Yellow River in the south upstream from Jinan, the northern levee of the Tuhai River downstream from Jinan; Majai River in the north downstream from Nanle, the upland of palaeochannel upstream from Nanle. The maximum flooded areas of different flood hydrographs are 22082 and 21190km² respectively. The reasons why the result differs from that of other workers are the different analysis between the landforms and landscapes.Subsequently, it is suggested;1) Present flood protection project should further be perfected and fully used so that the stream patterns will be controlled and the channel of the main flow will be fixed.2) The investments to the possible flooded area should be decreased, first, to estimate the effect of the economy, secondly, to present the decision for effect resulted from possible levee-breaking.3) The older channel should be given up and the new one taken up. Digging a new river course will be the strategic decision to solve the levee-breaking of the lower Yellow River. It is mainlined that the course of the new river should be designed along the depression belt in the possible flooded area. There are four advantages for man-made new river: 1) The stream course is shorter.and the slope of the channel is biger; 2)The losses and the investments is less; 3) The new stream course has the capability for draining the maximum flood and 4) The construction of the preject is more convenient.

Some core samples have been collected from nearby the Dianshan Lake's bridge and round the Tai Lake. By qualitative and quantitative micropaleon-tological analyses of them, abundant foramini ferids, ostracods, oospore membranes of Chara. molluscs, echinoid spines, pollen of pinus and shieldform fossils were found. Based on the ecological characteristics combining with sed imentological, paly nological as well as 14C data, this paper suggests that two episodes of transgression have occurred in the Dianshan Lake and Tai Lake since mid-Late Pleistocene (Q₃² and Q₄). However, both present lakes were formed in Late Holocene and five stages of geological development of them are recognized as follows;1. Mid-Late Pleistocene A large-scale transgression occurred in thisarea, Dianshan Lake was located in littoral zone and Tai Lake, in lagoon-bayenvironment with warm water forams. such as Asterorolalia and Globorotalia.2. Latest Pleistocene A large-scale sea-level subsidence caused the occurrence of regression. The two lakes changed into fresh-water lakes. No fossil has been found.3. Early Holocene With the beginning of Holocene-Transgression, Dian-shan Lake and the eastern Tai Lake turned into lagoons while the other part into shallow-water environment on shore and only with euryhalline forams.4. Mid-Holocene When the Holocene high sea-level occurred (0500-7500 years ago) the two lakes became estuary, after then the estuaries disappeared. The Dianshan Lake had been separated from and changed gradually into a fresh-water lake again, and lastly, emerged to a land. But the transgression remained its influence in southeastern and northwestern sides of Tai Lake to form the transitional facies sediments. The blackish forams and open sea small-sized forams were found there.5. Late Holocene Because the lake outlets were narrowed or silted up, the surface runoff converged into depressions to form a group of lakes. As a result, the extent of the two lakes enlarged constantly. Forams were seldom found in sediments.

Shui Jing Zhu"(Notes For the "Water Scripture")is a famous ancient geographical dook written by Li Daoyuan in the beginning of the 6th century A.D..This book recorded the existence of "the South river" in Tarim Basin. Howeverthere are many different viewpoints to the "South river". Some resea-archers consider that the "South river" didn't exist, and others consider that the South river runs across the center of Taklimakan desert from west to east adn so on.According to the textual research on "Shui Jing Zhu" and comparing the image taken from the satellite with the results of archaeological studies, the author of this paper considers that the "South river" was not fabricated, it did exist in ancient time.The "South river" had two sources.One was the Ya-rkand river and the other was the Karakhash river. But in ancient time the Ka-rakhash river didn't flow north-eastward and join the Yulongkhash river as it does today.In ancient time, the karakhash river flowed north-west ward had joined the Yarcand river.Then the joined river flowed north-east ward and got the name "South river", decause it was roughly parallel to the Kashgar riverthe "North river" named by Li Dauyuan-keeping quite a distance. And then when the Yulongkhash river and Keriya river run across the Taklimakan desert north ward.it roughly paralleled the "North River" keeping quite a distance So it also got the name "South river". Final ly, a river branched from the "North river" at the north-Past extremity of Taklimakan dosert and flowed south-westward, it was another fragment of the "South river.

Land series mapping is a new field in land research and geographical mapping This paper discusses some basic problems in the mapping work of the loess plateau in Northern Shanxi,1. The objects and features of the series of land type maps.This land unit in different levels can be taken as the cartographical objects of the maps in different scale in Northern Shanxi,as follows: 1:1000000 scale may be regarded as compound land type 1:500000 scale as complicated land type, 1:100000 scale as simple land type, 1:50000 scale as land unit, and 1:10000 scale as land element.There are totally six kinds in mapping of a series of land type, each scale has a special feature of its own.2. The procedure and methods of compiling of the series of land type maps.The procedure of the mapping consists of four stages; namely, the preparatory stage, the stage of comliling the draft maps, the stage of co-ordinating the series of maps, and the final completing stage,The compiling of land type series follows the general criterion of land type maps, and in addition, in compiling large, middle, and small scaled land type maps, each scale has special methods of its own. But any map of land type should be made by combining the conventional methods with new technical methods.3. The feature, the theoretical basis and the principles of the series of land type maps.There are chiefly four features of the series of laud type maps, namly. independent character, connection, subordinate and entirety,The theoretical basis of the series mapping is that the land have different levels in a hierarchy and the close connection, This land unit in different levels can be taken as the cartogrphical objects of the maps in different scale.Three basic principles can be followed in the mapping. 1.Taking the different maps into a series. 2.Keeping the viewpoint as one in the series mapp ing, 3. Consider ing the cartographical generalization.

In this paper the generalized spore-pollen scheme and the temperature curve of the studied region since the Last Pleistocene are constructed based on analysis of the spore-pollen from ten drill holes, According to the climate variation, the pas t 1200011 yea rs are divided in to three s tages, seven clima tic periods, with wh-ich ten more climatic fluctuations are contained, and the overall amplitude of te-mperature changes is about 10 degrees een tigrade. The com parison of this curve with otlier tvpical climate curves in china and abroad illustrates that the I'alaco climate variability is glohal.

The mothod for computing the longtcrm land evaporation in China is classified into three kinds in this peper. The first one belongs to a constant relationship. The second one gives a formula in which some parameters represent the effects of geography and climate.The thrid one can only be used in some special regions. Thus, the formula E/r = thD (E₀/r) is suggested to compute the longterm average evaporation in this paper.The suggested formula has been analysed and demonstrated in mathematics and in the concepts of climate and physics, and it is considered to be rational and feasible.This it, checked by a large amount of data in China and other countries meeting the heed of the precision. An analysis on some special climalical and geographical conditions and hydrolog ical effects is made for the parameter D which represents the affecting factor of various climates and underlying surfaces.

The accumulated temperature steadily above 10℃ in the central and east era temperate area of the United States of America has been calculated by using an empirical formula revised on the basis of the Conrad's formula. It has been found that, in the studied area of the United States, the accumulated temperature is generally higher than that in the similar latitude of the eastern temperate area of China.It is beyond a shadow of doubt that the ac-cunmulated. temperature in the former is at least not lower than that in the latter, This new discovery corrects a traditional viewpoint, which has been in fashion for a long lime among the geographers of China, that the heaf resource of the Eastern Asia is the richest in the same latitude zone of the world due to the summer monsoon. The higher accumulated temperature in the central and eastern temperate area of the U.S. is associated with the slight ly longer period with temperature steadily above 10℃ and the warmer mean monthly temperatures within this period except July. The hottest month in the year. The northern states west of the Croat Lakes have approximately equal accumulated temperatures with the Northeast of China in the same latitude.

The evolution of land types includes two sides: one is space and the other time. For land syslem is an opening matter system, not only matter but also energy exchanges with its surroundings continually. The theory of dissipative structure was used to analyse the evolution process in time and the formation of dissipative structure of land types. The entropy DS of land system includes negative entropy deS and inner entropy diS. That is dS= deS+diS. If diS/dt=-deS/dt>0, the land types show in stable state, if diS/dt/≠-deS/dt, the land types begin to evolve. The evolution of land types is completed through its phase change. Increase of deS of land in the evolution process of land types shows the evolution of land types in good condition, the contrary is in bad condition.Two kinds of the process of land types evolution in the JinYun Mountain were discussed in this paper. One is in natural condition and the other under cultivating. For the latter, the human action is a very important factor in the evolution process of land types, because both diS and deS are affected by human action. The direction and speed are determined by the way of the use of land.The evolution and the dissipative structure of the land system are formed by the, fluctuation in the system. In the use and protection of land, any kind of fluctuation phenomenon must be noticed, especially for the land in unstability.

This report is dealt with all together 48 ¹⁴C dating datas,which have been dated with chemical sampling of carburet, synthesized by Li-means from Sep. 1984 to Dec 1986 by ¹⁴C Dating Lab.Institute of Geography, Chinese Academy of Sciences.Those datas include 27 geological-geographical samples, 10 paleao-geographical samples of China, and 2 geological samples of Yugo-lasvia.

本文综合介绍了近年来关于大气中CO₂浓度增加对气候影响的研究。通过对CO₂浓度、全球气温及海平面高度变化的分析,和对火山活动、太阳常数变化的估计,讨论了在过去一百年气候变化中CO₂的作用,并介绍了对未来气候变化预测的研究结果。

为了交流学术经验,讨论应用问题,吸取国外先进经验,探讨国际合作研究模式与途经,在国际地理学会、国际摄影测量与遥感学会、国际制图协会、国际发展研究中心(加拿大)等许多国际学术团体和中国地理学会、国家自然科学基金委员会、国家遥感中心的倡议和支持下,由中国科学院地理研究所主办的“国际地理信息系统学术讨论会(北京,1987)”于5月24至28日在京召开。 来自五个国际学术团体、七个国家和地区、以及中国北京、上海、南京等13个城市的70名中外科学家、教授、研究人员出席了会议。

The kernel of theoretical geography is derived from the interaction and interconnect between physical factors and human activity occuring or occured in geographical environment. It should be to interpret or discorve some more general law on the integral foundation consisting "of "geographical phenomena", "geographical events", "geographical structure", "geographical process", "geographical function", "geographical effect", and "geographical quanlity". The theoretical geography has also explored the philosophical connotation and methodology of geography. And, it should reflect whole hard core of geography in the integral scale and the advanced level.The discussions of theoretical geography tend to fall into connection of two categories, the unity and the reality of natural world. Scholars who praise the theories and regulations do so in the belief that it is the essential manifestation of all superficial phenomena. Indeed, the dynamic and the transition of matter, energy, and information in geographical systems seem to be as the theme of theoretical geography.Besides the argumentation dealing with definition,history, and basic principles of theoretical geography, in the paper we have already concluded its contents into seven aspects;1. "Chaos and order" of geographical environment,2. Geographical systems, 3. Spatjal structure,4. Analysis of geographical process, 5. Organization effect and geographical gradient. 6. Geographical coupling of temporal and spatial research, 7. Behavious and perception of geographical body.

The theory of spherical leaven and plane earth (天圆地平说) in remote antiquity was a kind of primitive cosmologieal hypotheses, of which the Huntian Theory (浑天说) in ancient China was an outstanding model. This theory held that the heaven is a sphere with the place of odservation as its center and the earth a concentric circular plane. Furthermore the earth consists of a land mass in its central area and a ring-shaped ocean in its margin and disects the heaven into a visible upper hemisphere and an invisible lower hemisphere.This paper pays special attention to its universalism and localism.This theory appeared not only in ancien.t China hut also in the western worldjsuch as Babylonia and ancient Greece. Our country placed stress on the spherical heaven and expressed it with a celestial globe (浑天象). On the contrary,the western world put emphasis on the circular plane earth and showed it with a wneel-ahaped world map.This is its universalism.In different areas,this theory had different centers of the earth surface. Eor example, ancient Chinese scholars took Yang Cheng(阳城), the first capital of Xia dynasty, as the center of the world, whereas ancient Greek philosophers considered their home-land as the common center of the spherical heaven and the circular earth This is the localism of this theory.

The Hexi desert and oases region of Gansu Province belong to internal drainage area where all rivers originate from Qilian Mountains. From east to west, the rivers can be divided into four drainage systems; Shiyang river system, Hei river system, Shule river system and Harteng river system. Har-teng river flows into Suganhu Basin which lies in the northern margin of Caidamu Basin, while all theot her three drainage systems flow into Hexi Corridor, Hei river is the greatest and longest of all these rivers, stretching north to Alashan Plateau. This paper mainly discusses the distribution and change features of Hei river system based on the ancient and modern maps and salt content in ancient rivers and lakes and deduces how it has changed from exterior river to interior river, from integrate drainage system to many seperate drainage systems, from a river with its middle and lower reaches in the same course to a special river which changes alternately between surface water and undergronud water in the middle and lower reaches, and forming many oases of varying sizes along the bank of the river in certain regions.

This paper makes an approach to ecological classification of mire in the Sanjing Plain according to the view of ecology, selecting nine indexes of mire ecosystem, and dy means of principal component analysis, cluster analysis and discriminant analysis.The author considers that the aim of mire classification is to reveal essential properties of mire ecosystem and differences of mire formation and development. We not only try hard to reflect essential dif ferences among mire types, but also make great efforts to explore quantity scale measuring difference among mire types. Mire classification must be accor ding to essential properties of mire ecosystem, which are important elements for flows of energy and material in mire ecosystem.On the basis of classifi-catioh principle and method mentioned above, the mire in the Sanjiang Plain can de grouped into two types (seasonal water-logged mire and perennial water-logged mire), four subtypes and eight dodies.This method is favourable for translating interrelationship of multidimensional space among mires which can't be illustrated in plane, into illustration in twodimen-sional figure, and measuring qualitatively differentiations in mire types and setting up discrimination mathmatics models of classification for unknown types of mire.In addition,the author suggests some indexes of mire eco-elassification. namely microgeomorphological index, vegetation type, clay content of sediment, origanic content, pH of plant residual laver by quantity. The eco-classi-fication has a certain significance in guiding for exploiting and utilizing mire scientifieally.

This paper defined the concept of tourism industry growth point and reinterpreted tourism industry growth path by applying the"path dependence"theory.With the example of Zhangjiajie tourism industry growth point,it put forward a universal SGGT analysis mode and researched its spatial evolution process and dynamic mechanism.The results showed:(1) specific historical conditions and social economic environment was a foundation for the formation of Zhangjiajie tourism growth point;demand and cultivation was the key factor of tourism growth point germination;innovation and development was the guarantee of tourism growth point while the tourism growth point was promoted to be growth pole by disintegration;(2)the three tourism growth points—Zhangjiajie National Forest Park,Tianzi Mountain and Suoxiyu—became the core growth points after they realized unification of the administrative jurisdiction in Zhangjiajie scenic spot by path creation and listed as a world heritage site;(3)driven by the generating power of tourism industry growth point and world heritage site management,Tianmen Mountain becomes a new growth point with its high quality tourism resources;(4)the two growth points of Wulingyuan scenic spot and Tianmen Mountain were formed as a result of joint action by tourism industry agglomeration and path creation in Zhangjiajie.The results provide a great benefit for the local government and enterprises to understand the tourism industry growth path from the perspective of tourism industry growth point and further offer scientific guidance for the sustainable development of tourism industry in the world heritage site.