438 Accepted by S. Kambhampati: 13 Feb. 2004; published: 23 Feb. 2004 1 ZOOTAXA ISSN 1175-5326 (print edition) ISSN 1175-5334 (online edition) Copyright © 2004 Magnolia Press Zootaxa 438: 1–15 (2004) www.mapress.com/zootaxa/ Andesipolis, a puzzling new genus of cyclostome Braconidae (Hymenoptera) from the Chilean Andes, with descriptions of three new species JAMES B. WHITFIELD, WON-YOUNG CHOI & KYONG-IN SUH Dept. Entomology, University of Illinois, Urbana, IL 61801 USA ABSTRACT A new genus of braconid wasp, Andesipolis Whitfield & Choi, n. g., is described from the Andean region of Chile. It clearly belongs to the cyclostome lineage of Braconidae but otherwise is diffi- cult to assign definitively to tribe or subfamily, as it has some morphological features typical of Rhysipolini, others typical of Rhyssalini, and a few unique features. Three species, A. whartoni n. sp., A. masoni, n. sp. and A. framea n. sp. are described to represent the range of morphological variation encompassed by the new generic concept; other species are known to exist in collections but await a full-scale revision of this locally abundant genus. Key words: parasitoids, Rhyipolini, Rhyssalini, Hormiini INTRODUCTION The cyclostome braconid wasps loosely known as “Hormiinae” and Exothecinae” have long posed problems for classification (Hedquist , 1963; Shaw, 1983; Belokobylskij, 1984; Whitfield & Achterberg, 1987; Whitfield, 1988, 1992; Shaw and Huddleston, 1991; Belokobylskij 1992, 1993a, b; Quicke, 1993; Wharton, 1993; Achterberg, 1995; Whit- field & Wharton, 1997; Quicke and Belshaw, 1999; Spencer & Whitfield, 1999; Scatolini et al., 2002). Three large subfamilies (Braconinae, Doryctinae and Rogadinae), as well as several smaller ones, are clearly closely related to the hormiine and exothecine genera, yet the subfamilies are currently defined in such a way as to exclude these genera. “Hormi- ines” and “exothecines” are thus often assigned to small subfamilies or tribes of their own with varying composition. It is likely that a complete overhaul of the cyclostome groups will be required to produce a more stable and accurate long-term subfamily- and tribal- level classification, most likely with fewer subfamilies than are currently recognized, and