本文闡述植群多樣性之意義、測計方法及其與生物多樣性之關系，植群多樣性可用數學方法評估一地區最大環境梯度上之物種轉換率（β多樣性），以代表該地區之植群復雜性及分化程度，然具體之表現方法則為植物社會之分類與命名。受到世界各地生態學者不同觀點之影響，植群分類之理論與方法極具多樣化，以分類準則之選擇而言，可分為植群形相、環境差異、生態系特性與植相組成等研究法，其中採用植相組成為準則者，能充分顯示一地區之生物多樣性。植相分類法又因植物社會組織觀念以及分類技術之差異，遂產生多樣化之分類系統。本文除探討植群分類紛歧性之原因外，特別強調分類系統整合之必要性，并推薦歐洲大陸法瑞學派之植相社會學研究法，作為植相分類與植群系統整合之方法，對其劃分植群型之關鍵加以說明。本文最後則略述台灣植群研究歷史，就植群取樣方法、命名方式、資料分析技術等項目加以檢討，提供未來台灣植群整合研究之參考。

The present paper elucidates the content of vegetation diversity and its relationship to biodiversity. Although vegetation diversity can be evaluated mathematically by species turnover (13diversity) along the greatest environmental gradient, it is better realized through the classification and nomenclature of plant community. The theory and approach of vegetation classification vary considerably due to the diversified viewpoints of ecologists. Major criteria used in classification include physiognomy, environment, ecosystem and floristic composition. The floristic approach is the best expression of biodiversity, and its result of classification is further complicated by different views concerning the nature of plant community and the technical details of algorithm. This article deals with the factors leading to the diversified results of vegetation classification, and the need for developing a standardized synthetic system of classification. Reviewing comment is also made with respect to the history of vegetation study in Taiwan and the resultant diversified units, with special reference to sampling techniques, data analysis and nomenclature. The theory and synthetic procedures of floristic phytosociological approach (according to Zurich-Montpellier school) are recommended for the future synthesis of vegetation types in Taiwan.