This talk presents recent advances Munda historical phonology. Based on the work on proto-Munda in Sidwell & Rau (2014), the talk will give an update on the current understanding of the consonantism and vocalism as well as our understanding of development of the prosodic structure in the Munda branch. The picture that arises from the current developments show proto-Munda to be a fairly typical Austroasiatic language – a language that was however in the process of being Indianized. In particular, the prosodic structure was already strongly influenced by the splitting of initial consonant clusters and a considerable number of non-Austroasiatic lexemes in the proto-Munda lexicon hint at the influence of mostly unknown South Asian languages. The consonant correspondences are by now well understood, the remaining questions relate to how the proto-Munda system of consonants looked like and to phonological details of the posited proto-phonemes. I will in particular address issues concerning glottal phonemes as well as the issue of Pinnow‘s *q. While Shorto‘s assertion that [t]he vocalism, indeed, is the crux of the problem of reconstructing PMK (Shorto 1976) rings true for proto-Munda as well, the situation is better than generally believed. It is now possible to reconstruct the proto-vowels for a large part of cognate sets. The remaining sets, for which no reconstruction is currently possible, can be classified into well defined groups with common complicating conditions. These include for example words with epenthetic schwa or the Rapu‘d words already identified by Pinnow (1959). Taking the initial clusters posited by Shorto in his Mon-Khmer Comparative Dictionary (Shorto 2006) as the starting point, the talk will examine the patterns of cluster-splitting that occurred in the development of proto-Munda and the resulting difficulties for the vocalism. The aim is to give a comprehensive overview of proto-Munda phonology, to specify open questions and to address the implications for future research.