Material examined. Sri Lanka: Central Province, on palm leaves, Dec. 1868, Thwaites 172 (K, holotype).
New Zealand (all collections on dead leaf-axes and stems of Rhopalostylis sapida H. Wendl. & Drude (Palmae)): Auckland, Waitakere Ra. (PDD 16427, 16428, 18515, 18326 & 20995); Cuvier Is. (PDD 23527); Little Barrier Is. (PDD 18326).

Fructifications stromatic scattered, sometimes gregarious, entirely superficial, on superficial, byssoid, vinaceous brown mycelium. Stroma pulvinate, almost circular, up to 5 mm diam., 2 mm thick, when mature brown-black, more or less carbonaceous, punctate with slightly sunken ostiolar openings; when immature pruinose and velvety; undersurface loculate, each loculus representing a single perithecium: stroma tissue united at level of ostiolar canal; uniform, pseudoparenchymatous, compacted in outer zone, hyphal walls pigmented brown and thickened. Perithecia few but usually over twenty in a stroma, monostichous, globose 200-350 ftm diam. never crowded, sunken in separate loculi, ostiolar canal 300-400 µm long, 15 fim diam. lined with paraphyses, opening not papillate. Perithecial wall poorly defined, hyaline, but subhymenial layer of hyaline, thin walled, pseudoparenchymatous hyphae, well developed. Asci and sterile hyphae form a hymenial layer lining perithecial cavity. Asci unitunicate, clavate, with small apical ring, staining blue with iodine, an ill defined apical plug also present; 42-50 X 7-8-5 µm, 8 spored, spores biseriately arranged. Pseudoparaphyses numerous, thin-walled, septate 4-5 µm wide and of indefinite length. Spores one celled, broadly elliptical, 7-9.5 X 3-3.5 µm, occasionally lightly pigmented. (Fig. 2A-B.)

Berkeley & Broome (1875) described the species as Hypoxylon deciduum from Central Province, Sri Lanka. Cooke (1883) included it in his list of Hypoxylon species under the section III 'Sphaeroxylon' Patouillard (1887) described this same fungus from New Caledonia as a new species, Bagnisiella palmarum and regarded it as a dothidiaceous rather than sphaeriaceous fungus. Theissen & Sydow (1915) commented on Patouillard's fungus, noting that perithecia were formed in separate loculi at the base of a stroma, and placed the species in a new genus, Catabotrys.
Petch (1924) noted that although he had collected Hypoxylon deciduum on several occasions in Sri Lanka it was never in a condition in which he could comment on the true nature of the fungus. In the Kew Herbarium there are two collections from Petch from Peradeniya (20 July 1909 and Nov. 1913), neither appears to be mature.
Petrak (1934) examined Patouillard's specimens of Catabotrys palmarum and amended the description, placing importance on both structure and form of stroma and perithecia; he suggested that the genus should be placed in the Hypocreaceae. Seaver & Waterston (1946) recorded the fungus from Bermuda on banana {Musa sp.). They also showed that Catabotrys palmarum was synonymous with Hypoxylon deciduum and agreed with Petrak that probably it should be disposed in the Hypocreaceae. Arx & Muller (1954) commented that Catabotrys decidua was typically sphaeriaceous. Petrak (1954) amended his earlier views and, noting the diaporthiaceous nature of fructifications, suggested that perhaps a new family Catabotrydaceae should be erected to contain the genus within the Sphaeriales.
Up to the present, observations on this fungus have been made only on a few collections in herbaria. Field studies show that stromata take several months to mature. Unfortunately the species is difficult to culture, but observations suggest that a conidial form similar to Nodulisporium Preuss could be associated with it. Perithecial structure certainly shows that the fungus belongs to the Hymenoascomycetes rather than Loculoascomycetes. Ascal pore apparatus and perithecial structure are in some ways characteristic of a diatrypaceous fungus, but the ascospores are distinctly different and perhaps more typical of Xylariaceae. No doubt studies in the perithecial development and life history will elucidate the true relationships of this species.

New Zealand. Material examined at K: Titocura, on Rhipogonum parvifiorum, Colenso s.n. (holotype); on R. scandens, Colenso b 129, b 132 & b 172; Dunedin, Berggren 244 & 245; Little River, Berggren 248; Waitaki, Berggren 253; Wellington, Ruamahanga, Kirk 11, 38 & 56. At PDD: on R. scandens, 15481, 18609, 21479, 23012, 26370 and other numbers.

Fructifications stromatic, scattered, discrete, sometimes gregarious, erum-pent, more or less globular, sometimes subcorneal with erumpent cortical tissues adhering to the exposed stroma; stroma becoming flattened, roughened and oval, sometimes circular, 0-5-1 mm diam., 1-2 mm long, outer wall brown-black, rugulose. Perithecia often fewer than ten per stroma, monostichous or nearly so, crowded and mutually compressed, flask shaped or ovate when mature, up to 260 µm diam. and 450 µm high; ostiolar canaL short, opening small, not papillate. Outer perithecial wall black, carbonaceous; inner hyaline and diffuse. Asci and sterile filaments forming hymenial layer lining perithecia. Asci slender, clavate 70-80 µm long, upper third widest, 6-7-5 mm diam., tapering quickly to long narrow stalk; thin-walled with distinct apical ring, spores eight, irregularly biseriate, in upper third of ascus. Ascospores one-celled, cylindrical, allantoid, pale olivaceous 8-12-5 X 2-3 µm. (Fig. 1A-B.)

Berkeley (1855) compared this species with Diatrype verrucaeformis Fr. and with Sphaeria smilacicola Schw. He illustrated asci and spores typical of a Diatrype. Cooke (1879) listed collections made by Berggren in his travels in the South Island of New Zealand during 1874-1875. In his account of 'Australian Fungi 3, Cooke (1892) illustrated this species, giving the spores as 14-15 µm long. On the type sheet (K), two asci and spores are illustrated, one with small allantoid spores typical of Diatrype and another with dark-coloured, broadly elliptical spores. It appears therefore that another fungus is also present on the type specimen. Similarly, two fungi are present on a number of the Kew collections of D. glomeraria. As Berkeley clearly specified a diatrypaceous fungus this name is adopted for this distinct endemic species common throughout New Zealand wherever Rhipogonum occurs.
As the host is a monocotyledonous liane, bark tissues are thin. This limits stromatic development and the usual black line which extends into the wood and segregates the fungus stroma is poorly defined.
On the type the host is given as Rhipogonum parviflorum, which according to Cheeseman (1925) is a synonym of R. scandens and the locality as 'Titocura' which no doubt refers to 'Titokura'.

The specimen at Kew referred to by Berkeley (1855), Colenso 5019, New Zealand, is not referable to this species, which has not been collected from New Zealand.

New Zealand collections at PDD include; on Crataegus monogyna Jacq. (PDD 20586); on Leptospermum scoparium J. R. & G. Forst. (PDD 18190, 18192, 32883); on Prunus armeniaca L. (PDD 18334); on Nothofagus fusca (Hook, f.) Oerst. (PDD 18197, 18200, 18201); on N. menziesii (Hook, f.) Oerst. (PDD 18198, 20821); on N. solandri (Hook, f.) Oerst. (PDD 18295, 18423); on N. solandri var. cliffortioides (Hook, f.) Poole (PDD 18199); on N. truncata (Colenso) Cockayne (PDD 18195, 18203, 21126).

This European species was recorded from Australia by Cooke (1892). It is common in New Zealand, occurring on dead branches of indigenous species (usually Leptospermum and Nothofagus) and a few introduced host plants.