Large scale patterns in animal assemblages are increasingly being used as a method of understanding how species abundances in such assemblages are determined (see Lawton 1989, 1991, for reviews). In particular, the relationship between species abundance, measured as density or number of individuals, and species body size has received considerable interest (Brown and Maurer 1987, Gaston 1988, Morse et al. 1988, Tokeshi 1990, Basset and Kitching 1991, Nee et al. 1991, Griffiths 1992, Novotny 1992, Blackburn et al. 1993, Currie 1993). Much less attention has been given to other measures of abundance across animal assemblages, such as the distributions of biomass and energy use across species. One notable exception is a study of a bird assemblage by Maurer and Brown (1988). Maurer and Brown showed how biomass and energy use were distributed across 380 species of North American terrestrial birds, using literature data on body size, population density and individual energy use. When biomass and population energy use are plotted against mean body weight for individual species on log-log axes, values for individual species fall within well defined tetrahedra (Figs 3 and 4 in Maurer and Brown 1988, reproduced as Fig. la,b here; see also discussion on p. 1927 of Maurer and Brown 1988). The upper boundaries of these tetrahedra, representing maximum species biomass and energy use for each body size, increase linearly up to a weight of about 60 g, before levelling off. The lower boundaries increase linearly across the whole body size range, with slopes of approximately 1.0 and 0.67 for the biomass and energy use graphs respectively. The left hand boundary of each tetrahedron is determined by the variation in abundances of the very smallest species in the assemblage, but what constrains the upper and lower boundaries is not obvious, and Maurer and Brown (1988: 1930) suggest that defining these constraints should be a focus of future macro-ecological research. We agree wholeheartedly with the aim of using macroecological patterns in assemblages to understand species abundances. We would, however, draw attention to explanations for both the lower and upper boundaries of the two tetrahedra which suggest that the biomass and energy use distributions are artefacts of the abundance data from which the species biomass and energy use values were calculated. Some account of these artefacts will be needed in future considerations of these distributions. The explanations presented are partly extensions of previously published observations on possible artefacts in assemblage patterns (Blackburn et al. 1990), and apply to the patterns in biomass and energy use in identical fashion. We deal with the lower boundary first. Maurer and Brown calculated species biomass, B, for each bird species as B = M-D, where M is the mean body weight of individuals of the species, and D is the species density. The density data were taken from the Breeding Bird Surveys (BBS) conducted by the US and Canadian Fish and Wildlife Services, for which birds are counted over a large number of standard transect routes. They were calculated as the average species density per route on which the species was recorded. Since the minimum density a bird species can have in the BBS data is 1.0 individual/route, the minimum value for B is M, when D = 1.0. Thus for the plot of log species biomass against log body size, species points cannot lie below the line with slope 1.0 and intercept at the origin (when B = M. 1.0, or logB = logM+ logl.0 = logM). Since the bird assemblage used by Maurer and Brown includes species of density 1.0 across a wide range of body sizes (Brown and Maurer 1987), this slope indeed approximates 1.0 (Maurer and Brown 1988: Fig. 3).