Plantlets of Poa alpina L. taken from a single viviparous mother plant in August 1991 were pot-grown in an experimental garden in Abisko, Swedish Lapland. Half of the pots were infected with the annual hemiparasitic angiosperm Rhinanthus minor L. A proportion of the plants were harvested in August 1992: the infected grasses were less than half the mass of uninfected controls and had a higher proportion of brown (dead) tissue, but a similar partitioning of biomass into roots, stems and leaves. The grasses remaining after this harvest were divided into four groups: control (infected neither in 1992 nor in 1993 (−−)), newly infected (1993 only (−+)), not re-infected (+–) and re-infected (++). Total biomass accumulation by August 1993 was greatest in control grasses (−−) and least in those plants which had been infected for two consecutive years (++). Plants infected for the first time in their second year of growth (−+) showed fewer visible effects than those infected in their first year of growth (+–), but the relative biomass accumulation in 1993, when expressed per gram of dry matter existing at the end of the 1992 growing season, was least in those plants infected for the first time in 1993 ((- +) 6·41 g g−1) and greatest in those which had previously been infected but were relieved of infection in the second year ((+−) 12·65 g g−1). Unlike that in plants harvested in the first year of the investigation, the pattern of biomass partitioning differed between the groups of 2-yr-old plants depending on the year of infection. Uninfected plants (−−), and those plants infected for the first time in 1993 (−+) produced many viviparous plantlets. Those plants infected in 1992 ((+−) and (++)) showed negligible plantlet production, but had more numerous and larger leaves, which senesced more slowly and had higher nitrogen concentrations than those of uninfected controls. The maximum rate of host photosynthesis, photosynthetic light compensation point and quantum efficiency of photosynthesis, measured on a unit green-leaf area basis in July 1993, did not differ significantly between the four groups.
Parasites which were grown on 6-month-old grass (1992 harvest) were only two thirds of the mass of those growing on older grasses (1993 harvest), and those on grass parasitized for two years (++, i.e. those without plantlets) accumulated twice as much biomass as those growing on grasses infected for the first time in 1993 (–+, i.e. those with plantlets). Parasites which were growing on previously parasitized hosts also had greater foliar nitrogen concentrations than those growing on hosts not parasitized until 1993 (−+).