Colonies grown from single conidia on Difco PDA 25–35 mm diam after 10 d, aerial mycelium sparse, cottony, white, surface of colony with numerous acervuli, some with dark bases, with orange conidial ooze; in reverse more or less colourless apart from the dark acervuli and orange conidial masses showing through the agar. Conidia (14–)17–18.5(–20) × 4(–5) μm (av. 17.6 × 4.1 μm, n = 30), cylindric, straight, tapering slightly near both ends. Appressoria mostly elliptic to subfusoid, deeply lobed. Perithecia not seen.

Geographic distribution and host range: Reported only from USA, pathogenic to Aeschynomeme.

Genetic identification: ITS sequences do not distinguish C. aeschynomenes from C. fructicola. These taxa are best distinguished using TUB2, GAPDH, or GS. Notes: Colletotrichum gloeosporioides “f. sp. aeschynomenes” has been used to refer to isolates pathogenic to Aeschynomene virginica, later developed as the weed biocontrol agent Collego (references in Ditmore et al. 2008). It has also been reported from a range of other hosts (TeBeest 1988). Our analyses, based on a single, authentic strain of C. gloeosporioides “f. sp. aeschynomenes” (TeBeest 3.1.3, apparently the source of the single spore isolate originally used in the development of Collego, Ditmore et al. (2008)) show it to be genetically distinct within the Musae clade of the C. gloeosporioides complex. Genetically close to the geographically and biologically diverse C. siamense, it differs morphologically from this species in having slightly longer and narrower conidia which taper slightly toward the ends, and in having larger, strongly lobed appressoria. An isolate deposited as C. gloeosporioides f. sp. aeschynomenes in CBS (CBS 796.72) by G.E. Templeton, one of the early C. gloeosporioides f. sp. aeschynomenes researchers (Daniel et al. 1973), is genetically distinct to TeBeest 3.1.3 and has been identified by Damm et al. (2012a, this issue) as C. godetiae, a member of the C. acutatum complex. The strain that we examined (Te Beest 3.1.3) matches genetically another strain often cited in the C. gloeosporioides f. sp. aeschynomenes literature (Clar- 5a = ATCC 96723) (GenBank JX131331). It is possible that two distinct species, both highly pathogenic to Aeschynomene in Arkansas, have been confused. A survey of additional isolates of Colletotrichum highly virulent to Aeschynomene in Arkansas would clarify the interpretation of the past literature on this pathogen. For example, C. gloeosporioides “f. sp. aeschynomenes” was initially reported as specific to Aeschynomene virginica (Daniel et al. 1973), while later studies reported isolates putatively of the same taxon, to have a wider host range (TeBeest 1988). Cisar et al. (1994) reported fertile ascospores from crosses between isolates identified as C. gloeosporioides “f. sp. aeschynomenes” and isolates of C. gloeosporioides “f. sp. jussiaeae”, a pathogen of Jussiaea decurrens. The position of C. gloeosporioides “f. sp. jussiaeae” within our phylogeny is not known, but these taxa could prove useful for better understanding of the biological differences between phylogenetically defined species of Colletotrichum.

Holotype: USA, Arkansas, on Aeschynomene virginica stem lesion, coll. D. TeBeest 3-1-3, PDD 101995; ex-type culture ICMP 17673 = ATCC 201874.