本研究檢討有關都市化之相關研究及統計資料藉以澄清台灣地區主自1900-1985年間都市化之情形及其與人口遷移之關係。台灣在日據時代都市化的程度很低，而且在開發初期工業不發達、交通不便阻礙了人口遷移。o自1953年開始一系列的經建計劃之後許多新興工業設於大都市及其週圍之市鎮，因而加速都市化之過程。本研究發現台灣都市成長的模式各階段並不一致，在1950-60年代是倒U字型，1960-70年代是向上波動型，1970-85年代是 J字型。假若觀察的時間較長(如1947-85年間)則呈現向上波動型。都市大小與都市成長並非單向的關係，其間之關係受社會發展的階段及觀察時間長短的影響。一般而言觀察時期越長，且越是晚近的資料則都市大小與都市成長呈現向上波動的關係。本研究資料顯示都市人口增加的速度遠比都市化比例改變的速度快。 在1920到1940年代期間，台灣地區的人口逐漸由早期墾殖地移至新開發地區，在都市化的速度中亦伴隨著大都會人口之分散。許多學者認為開發中國家都市人口之成最主要是因自然增加的結果，但是台灣地區都市人口之增加，鄉村人口之移入更具影響力。

This paper has reviewed previous research and statistical data concerning urbanization and its relationship to migration in Taiwan during 1900-1985 period. The level of urbanization was very low when Taiwan was under Japanese rule. The low level of industrialization and the difficulty of communication and transportation in the early stage of development may have impeded migration. Beginning in 1953, Taiwan began a series of economic construction p1ans. Modern industries have been burgeoning in many big cities and their peripherial townships, thus accelerating the urbanization process. The data clearly show that the growth rate of the urban population has been much faster than the rate of change in th.e proportion that is urban. During the 1920-40 period the population of Taiwan gradually moved from places of early settlement to new city centers or newly developed areas. It is quite clear that the process of metropolitan dispersion has increased with the development of urbanization in Taiwan. The pattern of city growth was not consistent from period to period. It was reverse U-shaped for the 1950-60 period, upward-wave-shaped for the 1960-70 period, and J-shaped for the 1970-85 period. If a longer period (e.g., 1947-85) was observed, then it became upward-wave-shaped again. It is quite clear that there is no simple relationship between city size and city growth. This relationship is affected by stage of development; it also appears to be affected by duration of observation. Generally, the longer the interval observed, and the later the period analyzed, the more the upward-wave-shape holds for the relationship. Many scholars maintain that urban growth in developing countries results primarily from the natural increase of urban populations. However, judging from the rapid urban growth in Taiwan during the 1953-85 period, and considering the natural increase contributed by migrants in urban areas, it is likely that rural-urban migration is a more important source of urban growth.