本文檢視了王士禎等清人對五言律詩格律拗救的說法，就其中略而不詳者以唐詩實際 證例衍申其說，其中互相扞格者則辨其正訛，其中與唐詩實際聲調相出入者，則以統計方 式為之糾補。 其次，本文僅以五律為度者，實因五律聲調為近體之基礎，七律不過於五律之上添加 兩平或兩仄，五絕、七絕則聲調取於五律、七律之半，其聲調、拗救與律詩並無不同。 本文以「仄仄平平仄、平平仄仄平」及「平平平仄仄、仄仄仄平平」兩種律聯為準， 乃基於律詩除首句用韻之少數數例外，皆用此兩種律聯排列組合而成，實為律詩之基本單 位，兩種律聯之聲調及拗救明，則五律之聲調明矣。

This paper examines the scansion of the pentasyllable regulated poem of Ch’ing scholars and points out some of the mistakes of these scholars. First, we use the scansion of the Ch’ing scholars to measure the poems of T’ang dynasty and find that the scansion does not totally fit into T’ang poems. We therefore make some adjustment of the scansion of Ch’ing scholars. Secondly, our measurement of the T’ang poems limits to the penasyllable regulated poems based on the reason that the scansion of the penasyllable regulated poems can be used to the other poem categories too. Thirdly, our measurement of the T’ang poems limits to two sets of intonation of the penasyllable regulated poems based on the reason that these two set of intonation are the elements of all regulated poems or so-called recent style poems.