British distribution: Widespread, and generally common, but most records made as "Ustilago violacea".
Microbotryum lychnidis-dioicae is the familiar anther-smut on White Campion (Silene latifolia) and on Red Campion (S. dioica). In Europe it occurs too on the closely related Night-flowering Catchfly (Silene noctiflora) (Vánky, 1994), but S. noctiflora, an annual, arable weed species, is now rare in Britain and there seem to be no British records on this host.
M. lychnidis-dioicae is part of the Microbotryum violaceum complex (q.v.) of species that parasitise members of the Pink Family (Caryophyllaceae). These were formerly believed to be true smuts (Ustilagionomycetes) and were included in the large genus Ustilago (as U. violacea). However, far from being 'text-book' smuts, the campion smuts are actually allied to the rust fungi (Pucciniomycetes), and molecular evidence supports the recognition of M. lychnidis-dioicae as separate from M. violaceum.
As in the case of the true anther smuts, the fungus completely takes over the anthers of the host, which burst open to release the purple, powdery spores instead of pollen. Infected flowers have been recorded as lasting considerably longer han uninfected flowers, the fungus also inducing strengthening of the flower-base and filaments of the stamens (Uchida et al., 2003). It is noticable in the photographs above that while some flowers are crumpled with age, the centre part and infected stamens remain in good condition.
Remarkable about this species is that it is able to over-ride sex determination in the host plant. Both S. dioica and S. latifolia are dioecious, populations containing separate male and female plants, but M. lychnidis-dioicae causes anther-development in the female host-plants.
Uchida et al. (op. cit.) provide an excellent, illustrated summary of the development of anthers in both male and female plants. Sex expression in these Silene species is determined by an X-Y chromosome system, as in higher animals, XX coding for female and XY coding for male, the Y-chromosome being morphologically distinguishable. The Y-chromosome appears to carry genes that suppress female development, and in infected XX plants the fungus evidently partially substitutes for the Y-chromosome genes absent in female plants. It can be seen in the photographs in their paper that this replacement of female characters by male is not complete; female plants retain their characteristically different calyx shape and characters.
Morphologically, M. lychnidis-dioicae is little differentiated from closely related taxa. The spore surfaces carry a high, reticulate mesh (see photo above) and according to Vánky (1994, 1998, 2004), based on previous work by Scholz and Scholz, the meshes are irregularly polygonal and up to 1.4 µm in diameter, whereas the meshes are smaller and/or more rounded to regularly polygonal in M. violaceum, sensu stricto, and in other taxa of the complex. Vánky himself (2004) considers these differences to be small and variable.
There is now a very substantial body of scientific literature comparing strains on different host species, using molecular techniques to demonstrate further genetic differentiation and isolation (see, e.g. Le Gac et al., 2007; Devier et al., 2010), and a number of such strains within the M. violaceum complex have also been given names as species, adopting a very narrow species-concept. Genetic differences are small in comparison with those accepted by Kemler et al., and if certain of these additional taxa are worth recognition in formal taxonomy, they would seem better treated as infra-specific variants. One such entity is M. silenes-dioicae, based on the strain on Silene dioica, whereas M. lychnidis-dioicae is treated, despite its name, as the strain on S. latifolia. [I presume the typification of M. lychnidis-dioicae was considered.] However, apart from the close molecular relationship of these two taxa, Biere & Honders (1996) showed that cross transmission of strains between S. latifolia and S. dioica in either direction is possible (though much more readily from S. latifolia to S. dioica), and also to hybrids between the host species. Gene flow between strains on the two host plants is evidently limited, but overall it seems best to consider M. silenes-dioicae as a synonym of M. lychnidis-dioicae. Outside specialist laboratories, no other taxonomy is practical.