JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

Grouse and vole numbers may peak after peaks in the seed crop of bilberry (Vaccinium myrtillus) because of reduced levels of feeding deterrents in bilberry plants. We predicted that grouse reproduction depends also on summer (June-September) temperatures in the 2 previous years, because bilberry plants will be less exhausted after a high seed crop in or after warm summers, and thus rebuild their chemical defence more quickly. After berry peak years, population indices of capercaillie (Tetrao urogallus) and bank vole (Myodes glareolus) in southern Norway were negatively related to summer temperatures in the previous year or previous 2 years. Willow grouse (Lagopus lagopus) chick production in five areas in Norway was negatively related to summer temperatures in the 2 previous years when controlling for vole density. A similar pattern was found for the bilberry-feeding moth (Eulithis populata), an important prey for grouse chicks. In eastern Norway, autumn densities of capercaillie and black grouse (Tetrao tetrix) were more likely to peak in vole peak years at high altitudes, where summer temperatures are low. We conclude that high summer temperatures may limit grouse reproduction through the effect on bilberry plants and that a warm climate thus adversely affects population levels of grouse.

In southern Norway, population¯uctuations of the herbivorous bank vole Clethrionomys glareolus and the granivorous wood mouse Apodemus sylvaticus were analysed in relation to mast seeding of bilberry Vaccinium myrtillus, sessile oak Quercus petraea and Norway spruce Picea abies. Bilberries ripen in August, acorns are dispersed in September, whereas spruce seeds are usually dispersed during winter or spring. The bilberry index of both the current and previous year contributed signi®cantly to explain the population growth indices of bank vole. The spruce seed index signi®cantly in¯uenced only the bank vole, whereas the acorn index signi®cantly in¯uenced only the wood mouse, apparently through increased winter survival. Because of the lack of signi®cant responses to bilberry masting by granivorous wood mice, we conclude that the most likely explanation for the observed peak populations of bank voles in post-mast years of bilberry is that high seed crops have a positive in¯uence on the chemical composition of bilberry plants, which are important as food for bank voles during winter.

Food selection by capercaillie (Tetrao urogallus) chicks was studied using human-imprinted chicks. Invertebrates made up more than 50% of the chicks' diet during the 1st week. Feeding on plants increased to approximately 100% in 7-week-old chicks. Berries of Vaccinium myrtillus made up 85% of the plant material eaten. Berries were eaten once they ripened. Other important plant foods were flowers of Erica tetralix and Melampyrum pratense. Ants (Formicidae) were the most utilized invertebrate food. The observed decrease in the proportion of invertebrates in the diet with increasing age was not a result of decreasing invertebrate abundance. Sweep net samples showed significantly fewer invertebrates when chicks were 1 – 3 weeks old than when chicks were 4 – 6 weeks old.

The plant stress hypothesis states that plant stress factors other than herbivory improve herbivore performance due to changes in the content of nutritive or defensive compounds in the plants. In Norway, the bilberry (Vaccinium myrtillus) is important forage for the bank vole (Myodes glareolus) in winter and for the moose (Alces alces) in summer and autumn. The observed peaks in bank vole numbers after years with high production of bilberries are suggested to be caused by increased winter survival of bank voles due to improved forage quality. High production of bilberries should also lead to higher recruitment rates in moose in the following year. We predict, however, that there is an increasing tendency for a 1-year delay of moose indices relative to vole indices with decreasing summer temperatures, because low temperatures prolong the period needed by plants to recover in the vole peak year, and thus positively affect moose reproduction also in the succeeding year. In eight out of nine counties in south-eastern Norway, there was a positive relationship between the number of calves observed per female moose during hunting and a bilberry seed production index or an autumn bank vole population index. When dividing the study area into regions, there was a negative relationship between a moose-vole time-lag index and the mean summer temperature of the region. These patterns Communicated by C. Gortázar suggest that annual fluctuations in the production of bilberries affect forage quality, but that the effect on moose reproduction also depends on summer temperatures.