Annals of Applied Biology ISSN 0003-4746 RESEARCH ARTICLE Host influence on germination and reproduction of the facultative hemi-parasitic weed Rhamphicarpa fistulosa S. Kabiri 1 , A. van Ast 1 , J. Rodenburg 2 & L. Bastiaans 1 1 Centre for Crop Systems Analysis, Wageningen University, Wageningen, the Netherlands 2 Africa Rice Center (AfricaRice), Dar es Salaam, Tanzania Keywords Infestation level; Oryza sativa; rain-fed lowland rice; rice vampireweed; seed production; Striga; weed management. Correspondence S. Kabiri, Centre for Crop Systems Analysis, Wageningen University, Wageningen, the Netherlands. Email: Stella.Kabiri@wur.nl Received: 6 May 2015; revised version accepted: 26 January 2016. doi:10.1111/aab.12288 Abstract Rice Vampireweed, Rhamphicarpa fistulosa, was a minor parasitic weed until recently when rice cultivation in sub-Saharan Africa was expanded into marginal wetlands, that are the parasite’s natural habitat. Unlike most of the parasitic weeds, R. fistulosa is facultative, meaning that the parasite is able to complete its life cycle without a host. However, when not connected to a host plant, its biomass and seed production is lower. Because very little is known regarding the germination ecology of the parasite, the main objective of our study was to identify the cues that favour germination. We hypothesised that, first, being a wetland species, germination of R. fistulosa is stimulated by light and high soil moisture. Second, we hypothesised that if host plant presence increases its reproductive output then a germination stimulatory effect from host presence is likely to have developed. A Petri-dish and pot experiment showed that light and completely saturated soils were a requirement for germi- nation, demonstrating that germination requirements of R. fistulosa are typical of species that grow in environments with fluctuating water levels. A pot exper- iment in which five infestation levels of R. fistulosa were installed in the absence and presence of a rice plant, showed that host plant presence resulted in a 3.7 times higher seed production rate and a 15% larger average seed size. Despite this reproductive advantage, a pot experiment with three rice cultivars, selected because of their difference in strigolactone production, showed that host plant presence, regardless of the development stage, did not influence the emergence rate of R. fistulosa. In a follow-up study, the germination stimulation effect of root exudates collected from the same three rice cultivars and a treatment consisting of an artificial germination stimulant (GR24) was compared with a treatment consisting of plain water. In these treatments, seeds of R. fistulosa were compared with seeds of the obligate parasite Striga hermonthica. Germina- tion of S. hermonthica was strongly advanced by the presence of root exudates and GR24 but was completely absent in water, whereas germination of R. fistu- losa in all treatments was similar to that in plain water. The absence of a host recognition mechanism at the germination stage suggests that the regulation of germination through light and soil moisture is near optimal. Our finding might also indicate that for this facultative parasitic plant species, a more opportunistic germination strategy is superior. Implications of the findings for management of R. fistulosa in rice cultivation are discussed. Ann Appl Biol (2016) 1 © 2016 Association of Applied Biologists