多年以前，筆者曾撰文探討明代軍戶中的幫丁，以軍政條例及遼東檔案為主，探討「原籍軍戶」與「衛所軍戶」中餘丁角色的分化。本文繼舊稿之後，一方面對歷來學者相關論述加以檢討，為「舍人」、「餘丁」、「幫丁」等名詞正名；另方面配合文集、碑刻、方志中所見資料，除對遼檔中出現的幫丁存在形態有更深入的分析，也藉由對天津三衛以及廣東沿海衛所軍役的介紹，說明衛所軍餘角色分擔受區域影響呈現差異性。對「原籍軍戶」餘丁角色之分化也提出補充，此即軍冊中「繼丁」的出現。繼丁免本身差役，五年一度專職負責向戶丁收取盤纏，解送衛所；營中故絕無人時，繼丁需即時補役。也因此，軍冊中繼丁的記載常被用為軍役糾紛時審判的依據。唯軍冊編審多因循不實，本文即以族譜資料為例加以說明。

Some years ago, the author wrote a paper about the so-called bangding (幫丁) of Ming dynasty military households, men designated to assist their relatives by providing military service. Drawing on military ordinances and a compilation of records from Liaodong (the Ming Dai Liaodong Dang'an Huibian明代遼東檔案匯編), that paper explored the diversity of roles performed by non-conscript members of both ”original” (原籍) and ”outpost” (wei-so衛所) military households. Taking up where previous work has left off, this paper discusses and clarifies definitions of terms such as sheren (舍人), yuding (餘丁), and bangding. This research also draws on anthologies, stele inscriptions, and local history gazetteers. This allows for a more in-depth analysis of the forms of bangding status and service that appear in the Liaodong archives. Furthermore, by explaining aspects of military service at outposts in Tianjin and coastal Guangdong, the paper will indicate how the roles performed by non-conscript men at outposts varied from region to region.The paper also elaborates on the diversity of roles performed by members of ”original” military households, focusing namely on jiding (繼丁) in military household records (junce軍冊). A jiding was exempt from conscript labor to his local government because he was instead responsible for collecting money from the members of his household and escorting it to a wei-so for use there once every five years. When the previous head of an outpost died without descendents, a jiding may be required to provide military service. For these reasons, records involving jiding in military household records were often used as the basis for adjudication in disputes involving military service. The paper also cross references clan genealogies and military records to show that the latter were particularly prone to contain inaccuracies.