Discoxylaria myrmecophila J.C. Lindq. & J.E. Wright



According to personnel at (B) Hennings' specimen of H. sanguineum was probably destroyed during World War II.

Discoxylaria is a monotypic genus. As far as we know Hypocreodendron is known only as the anamorph of Discoxylaria and is produced only on the discoid stromatal processes. Hypocreodendron thus is likewise to be considered as monotypic. It might be considered to be a synnema with intricate stipe and divergent capitulum sensu Seifert and Okada, 1990. We have not examined material of Poroniopsis, but accept the opinion of Lindquist and Wright (1964) that it is synonymous with Hypocreodendron.

Discoxylaria was well-described by Lindquist and Wright (1964), except that the ascospore germ slit was overlooked and a germ pore described. It is easy to see how the germ slit was missed. Very little holotype material is mature and the germ slit is fairly inconspicuous even on the most favorable material. We cannot explain the description of a germ pore; there is no suggestion of such in holotype material. In any case, Discoxylaria is readily assignable to family Xylariaceae on the perithecia associated with a stroma, on the cylindrical asci with amyloid apical ring, and on the brown inequilateral ascospores with a germ slit. Discoxylaria is distinct from Xylaria in its distinctive anamorph with percurrently proliferating conidiogenous cells located in or on a distinctive receptacle that apparently never bears perithecia. Percurrently proliferating conidiogenous cells are known in X. longipes Nitschke, but are apparently uncommon (Rogers, 1983). It has a superficial resemblance to Poronia, but that genus has an entirely different kind of anamorph, Lindquistia Subram. & Chandrash., that is characterized by disarticulating elements (Stiers et al., 1973). The Lindquistia is replaced on the disc by perithecia (Stiers et al., 1973). In many, if not most, Xylaria species the anamorph also is replaced by perithecia.

It is not certain why Hennings (1897) described the anamorph as being pycnidial, a structure accepted by Lindquist and Wright (1964) rather tentatively. No sign of pycnidia are seen in naturally-produced or culturally-produced material. However, old discs become somewhat pitted as localized parts of the hymenium deteriorate and these might have been interpreted as conidiomata. It is our impression that conidial production slows or ceases as perithecia mature.

We were able to study Hypocreodendron sanguineum in detail owing to the abundant material produced in culture. Conidia are produced at the apices of mostly unbranched conidiogenous cells. The single nucleus of the conidiogenous cell divides and a daughter nucleus squeezes into the conidial fundament. This process is repeated as subsequent conidia are produced. Conidial nuclei are usually seen to occupy most of the spore volume, as described for some xylariaceous and diatrypaceous fungi (see Glawe and Rogers, 1986 and refs. therein). In some cases, however, nuclei seem compact and occupy a lesser part of the conidium volume.

By FM conidiophores have a distinctive digitate morphology that we call "bony witch's fingers". By DIC a swelling in the secession zone is evident, giving a clue to conidiogenesis and secession. By SEM the swelling is seen to be a series of rings and a part of each ring is left on the conidiogenous cell and the other part forms the swelling or flare on the conidial base. Examination of thin sections by TEM confirms that the rings on the conidiogenous cell apex are indeed annellations, albeit appearing somewhat amorphous and not forming the neat tapering apices encountered in some other fungi.