OIKOS 98: 37–46, 2002 Infection by the systemic fungus Epichloe ¨ glyceriae and clonal growth of its host grass Glyceria striata Jean J. Pan and Keith Clay Pan, J. J. and Clay, K. 2002. Infection by the systemic fungus Epichloe ¨ glyceriae and clonal growth of its host grass Glyceria striata. – Oikos 98: 37–46. Many plant species have the ability to expand laterally through space by clonal growth. Plant pathogens can affect clonal growth characteristics thereby altering the success of host plants within populations and of clonal species within communities. We conducted a greenhouse experiment to determine the effects of the systemic fungal pathogen, Epichloe ¨ glyceriae, on clonal growth patterns of its host grass, Glyceria striata. We found that infected and uninfected plants produced similar total biomass and numbers of tillers plus primary stolons per mother ramet. However, biomass allocation to tillering (vegetative growth) vs stolon production (clonal growth) was significantly affected by pathogen infection. Infected plants produced more stolons and clonal growth biomass than uninfected plants while mother ramets of uninfected plants produced more tillers and biomass than infected plants. Stolon architecture of infected and uninfected plants also differed. In two of three popula- tions, infected plants produced stolons with greater biomass and shorter spacer lengths, even though mean stolon lengths were similar for infected and uninfected plants. These results contrast strongly with most other clonal plant-pathogen systems where infected plants are less vigorous and have reduced clonal growth compared to uninfected plants. Greater clonal growth may be an effective mechanism for host genotypes to persist and spread when seed production is prevented, as is the case with castrating pathogens like Epichloe ¨ glyceriae. J. J. Pan and K. Clay, Dept of Biology, Jordan Hall 142, 1001 E. Third St., Indiana Uni., Bloomington, IN 47405, USA ( jepan@indiana.edu). Clonal plant species are ubiquitous and important com- ponents of many plant communities (Oborny and Bartha 1995, Klimes et al. 1997), especially those dom- inated by monocots, as approximately 70% of extant monocot families are rhizomatous (Tiffney and Niklas 1985). Clonal plants commonly spread through both seed production and clonal growth via rhizomes, stolons, etc. (see Klimes et al. 1997 for a detailed categorization of clonal growth). The ability to grow clonally plays an important role in the persistence and spread of clonal plant species (Janzen 1977, Cook 1979, Eriksson and Jerling 1990, Oborny and Cain 1997), e.g. the success of many weeds has been attributed to clonal growth (Thompson 1991, Pys ˇek 1997). Clonal growth also plays an important role at the population level. Differences in clonal growth can affect competitive interactions and lead to dominance by a few genotypes (McLellan et al. 1997). Harberd (1961) found that one genotype of the rhizomatous Festuca rubra had spread about 220 meters laterally and estimated its age to be about 400 years. Thus, changes to clonal growth char- acteristics may alter the success of different plant geno- types within populations. Pathogens often modify host plant growth and repro- duction (Burdon 1987, Isaac 1992). Infected host plants generally have reduced fitness as a result of changes in physiology, destruction of host tissue, infection of re- productive structures, and are typically stunted and less vigorous than their uninfected counterparts (Burdon 1987, Jarosz and Davelos 1995). Grasses infected by systemic fungal endophytes commonly experience two types of modifications to their growth: reproductive Accepted 21 December 2001 Copyright © OIKOS 2002 ISSN 0030-1299 OIKOS 98:1 (2002) 37