Published in 2010 in No. 2 of the Journal of the Research Institute for World Languages, Osaka University, this Japanese-language article by Shimizu Masaaki (清水政明) is titled “Discovery of the Lost Portion of Hộ Thành Mountain Inscription (1342): Taboo Characters and the Irregular Sino-Vietnamese Readings in the Yu Rhyme Group” (護城山碑文（1342）欠落部の発見―所収避諱文字と虞韻所属例外字音―). The article reports the discovery, during field research in March 2006, of the previously lost first twelve lines of the Hộ Thành Mountain Inscription (1342), which Henri Maspero had described in 1912 as the oldest surviving evidence of Chữ Nôm material in Vietnamese historical phonology. The author reconstructs the contents of the newly identified portion and uses this material to investigate the historical pronunciation of characters recorded in the inscription. Particular attention is given to taboo characters (避諱字) and their relationship to irregular Sino-Vietnamese readings. The study identifies a taboo character occurring after the personal name Mạc and examines its relationship to the character 球, whose Sino-Vietnamese reading cầu is irregular within the Yu (虞) rhyme group. Through comparison with other historical sources, the author argues that this irregular reading resulted from the practice of taboo-name avoidance under the Trần dynasty, which led to the substitution of one vowel for another. The article broadens this analysis by examining other examples of irregular Sino-Vietnamese readings preserved in historical texts. It considers how phonological substitution associated with name taboos could affect the development and transmission of Sino-Vietnamese pronunciations, including alternations involving vowels and diphthongs. Tables in the study compare regular and irregular readings and relate them to historical Chinese rhyme categories and Vietnamese phonological developments. More broadly, the study demonstrates the importance of epigraphic evidence, taboo-name practices, and historical inscriptions for reconstructing the history of Sino-Vietnamese pronunciation. The rediscovery of the lost portion of the Hộ Thành Mountain Inscription provides new textual evidence for understanding the relationship between Chinese characters, Vietnamese historical phonology, and the social conventions that shaped reading practices in medieval Vietnam. The author concludes that Vietnamese taboo-reading practices followed principles that were fundamentally different from those documented in Chinese sources, highlighting the value of Vietnamese epigraphic materials for research on historical linguistics, Sino-Vietnamese readings, chữ Nôm, and the history of Vietnamese writing systems.