290 DOI: 10.1111/j.1466-822x.2006.00224.x © 2006 Blackwell Publishing Ltd www.blackwellpublishing.com/geb Global Ecology and Biogeography, (Global Ecol. Biogeogr.) (2006) 15, 290–302 RESEARCH PAPER Blackwell Publishing Ltd Energy availability, abundance, energy-use and species richness in forest bird communities: a test of the species–energy theory Mikko Mönkkönen 1 *†, Jukka T. Forsman 2 and Folmer Bokma 3 ‡ ABSTRACT Aim To test the ‘more individuals hypothesis’ as a mechanism for the positive association between energy availability and species richness. This hypothesis predicts that total density and energy use in communities is linearly related to energy availability, and that species richness is a positive function of increased density. We also evaluate whether similar energy–density patterns apply to different migratory groups (residents, short-distance migrants and tropical migrants) separately. Location European and North American forest bird communities. Methods We collected published breeding bird census data from Europe and North America (n = 187). From each census data we calculated bird density (pairs 10 ha -1 ), energy use by the community (the sum of metabolic needs of individuals, Watts 10 ha -1 ) and geographical location with an accuracy of 0.5°. For each bird census data coordinate we extracted the corresponding monthly values of actual evapotran- spiration (AET). From these values we calculated corresponding AET values that we expected to explain the density energy use of forest birds: total annual, breeding season (June) and winter AET. We used general linear modelling to analyse these data controlling for the area of census plots, forest type and census method. Results Total density and energy use in European and North American forest bird communities were linear functions of annual productivity, and increased density and energy use then translated into more species. Also resident bird density and energy consumption were positive functions of annual productivity, but the relationship between productivity and density as well as between productivity and energy use was weaker for migrants. Main conclusions Our results are consistent with the more individuals hypothesis that density and energy use in breeding forest bird communities is coupled tightly with the productivity of the environment, and that increased density and energy consumption results in more species. However, not all community members (migratory groups) are limited by productivity on the breeding grounds. Keywords Body size, Europe, migrant birds, more individuals hypothesis, North America, productivity, resident birds, spatial autocorrelation, species interactions. *Correspondence: Mikko Mönkkönen, Department of Biological and Environmental Sciences, University of Jyväskylä, POB 35, FIN- 40014 Finland. E-mail: vemonkko@bytl.jyu.fi 1 Department of Biology, University of Oulu, POB 3000, FIN-90014, Finland, 2 Department of Ecology and Evolution, Evolutionary Biology Centre, Uppsala University, Norbyvägen 18 d, SE-75236 Uppsala, Sweden and 3 Department of Biology, University of Oulu, POB 3000, FIN-90014, Finland †Present address: Department of Biological and Environmental Sciences, University of Jyväskylä, POB 35, FIN-40014 Finland. ‡Present address: Department of Ecology and Environmental Science, Umeå University, SE-901 87 Umeå, Sweden. INTRODUCTION Geographical variation in species diversity has been studied intensively for almost two centuries (e.g. von Humboldt, 1808; Wallace, 1878). The best-known pattern is a gradient of increasing species richness toward the tropics, which is observed at several spatial scales, from local communities to continental species pools, and in many animal and plant taxa (Hillebrand, 2004). Longitudinal variation among regions at the same latitude has also been well described (Currie & Paquin, 1987; Latham & Ricklefs, 1993; Huntley, 1994; Mönkkönen & Viro, 1997). A plethora of hypotheses exists to explain this variation (e.g. reviewed by Pianka, 1966; Rohde, 1992; Rosenzweig, 1995; Willig et al., 2003) and new ones con- tinue to appear (Ritchie & Olff, 1999; Colwell & Lees, 2000).