The purpose of this paper is to analyze vowel quantity in several Czech dialects to determine the relationship between Czech quantity and Proto-Slavic accentology. Making use of what we know from existing studies (Hruška, Gebauer), together with a focused analysis of selected manuscripts dating from the thirteenth century, and further aided by the work that historical linguistic studies has provided in the reconstruction of Proto-Slavic (Timberlake, Dybo), the present work will analyze the development of quantitative vocalic relationships in several Southern Bohemian dialects. The comparative historical method is used combined with dialect geography. The paper concentrates on selected lexical items, mostly nouns, that have good Proto-Slavic pedigrees and that are relevant for answering the questions that relate to the expected reflexes of the canonical Proto-Slavic accentological paradigms: acute oxytonic, acute barytonic, and circumflex. The dialects cited often exhibit patterns different from the literary language and should be taken into consideration when applying Czech data to arguments regarding e.g. the distribution of the neo-acute and the lengthening of short vowels under retracted stress from the neo-acute, the validity of the neo-circumflex, the lexical items identified as acute oxytonic, acute barytonic, and circumflex, etc. Several dialects in southwestern Bohemia (the Chod dialects, the dialect of Stříbro, and the dialects of Doudlebsko) that preserve what appear to be the most ancient reflexes of the neo-acute in masculine monosyllabic nouns and which are most reliable in preserving acute length will be examined. Special attention will be given to the Chod dialects around Domažlice which have preserved older patterns in the feminine (j)o-stem nouns than the literary language and which also exhibit more reliable reflexes of the neo-acute.