Other specimens examined: USA, Florida, Sarasota, on Vitis sp. leaf, coll. S. MacKenzie SS-Grape-12, 2002 (ICMP 18706); Hawai’i, Aiea, on Clidemia hirta leaf spot, coll. S.A. Ferreira & K. Pitz, 14 May 2010 (ICMP 18659, ICMP 18660, ICMP 18661, ICMP 18662, ICMP 18663).

Colonies grown from single conidia on Difco PDA 25 mm diam after 10 d, aerial mycelium grey, cottony, sparse, surface of colony with numerous small, dark-based acervuli with deep orange conidial ooze and scattered setae, in reverse more or less colourless except for the acervuli and masses of conidial ooze showing through. After 18 d numerous globose, pale walled protoperithecia developing near centre of colony. Conidia (16−)18−20(−26.5) × (4.5−)5.5−6 μm (av. 19.3 × 5.5 μm, n = 48), broad-cylindric, ends broadly rounded, longer conidia sometimes tapering slightly towards the base. Appressoria variable in shape, some simple, subglobose, but often with a small number of broad, irregular lobes. Perithecia mature after about 21 d, dark-walled, about 200–250 μm diam with short ostiolar neck, perithecial wall of 3–4 layers of angular cells 10–15 μm diam with walls thin, pale brown to brown. Asci 8-spored 60–67 × 10–14 μm. Ascospores (14–)15.5–19(–21.5) × 4.5–5.5(– 6.5) μm (av. 17.2 × 5.0 μm, n = 46), oblong-elliptic, tapering to rounded ends, widest point toward one end, in side view flat on one side, rarely curved and if so, then slightly.

Geographic distribution and host range: First reported from Clidemia, native to Panama, and subsequently introduced to Hawai’i as a pathogen of that host. Genetically matching isolates occur on native Vitis and Quercus spp. in Florida (see notes below).

Genetic identification: ITS sequences do not separate C. clidemiae from C. aotearoa. The two species are best distinguished using ACT, GAPDH, or GS. Notes: Isolates referred to C. gloeosporioides “f. sp. clidemiae” by Trujillo et al. (1986) were highly pathogenic to Clidemia, but not to the other species of Melastomataceae tested. No voucher cultures of the original isolates collected from Panama were kept, but recent specimens isolated from naturalised Clidemia hirta plants in Hawai’i with typical disease symptoms are genetically uniform and distinct within the Kahawae clade. Phylogenetic, biological, and morphological evidence support this fungus being described as a new species within the C. gloeosporioides complex. A fungus isolated from a Vitis sp. in Florida and referred to as “Glomerella cingulata native host” by MacKenzie et al. (2007), is genetically close to our isolates from Clidemia and is here referred to the same species. Data in MacKenzie et al. (2007) shows the same fungus occurs on both Vitis and Quercus in Florida. Micromorphologically the isolates from Clidemia and from Vitis that we examined are similar with respect to the size and shape of appressoria, conidia, and ascospores. They are distinct in cultural appearance, the cultures of the Vitis-associated fungus having aerial mycelium darker and more dense, and a faster growth rate. Similar variation in cultural appearance is present in several of the phylogenetically defined species that we recognise. Whether or not the Clidemia-associated isolates are biologically distinct from the Vitis- and Quercus-associated isolates from Florida requires pathogenicity tests to determine.

Holotype: USA, Hawai’i, Aiea, on Clidemia hirta leaf spot, coll. S.A. Ferreira & K. Pitz, 14 May 2010, PDD 101997; ex-type culture ICMP 18658.