This article understands abductive inference as encompassing several special patterns of inference to the best explanation whose structure determines a promising explanatory conjecture (an abductive conclusion) for phenomena that are in need of explanation (Sect. 7.1 ). A classification of different patterns of abduction is given in Sect. 7.2 , which is intended to be as complete as possible. A central distinction is that between selective abductions, which choose an optimal candidate from a given multitude of possible explanations (Sects. 7.3 and 7.4 ), and creative abductions, which introduce new theoretical models or concepts (Sects. 7.5–7.7). While the discussion of selective abduction has dominated the literature, creative abductions are rarely discussed, although they are essential in science. This paper introduces several kinds of creative abduction, such as theoretical model abduction, common-cause abduction, and statistical factor analysis. A demarcation between scientifically fruitful abductions and speculative abductions is proposed, by appeal to two interrelated criteria: independent testability and explanatory unification. Section 7.8 presents applications of abductive inference in the domains of belief revision and instrumental/technological reasoning.