(e.g. Rapaport’s rule; Ricklefs, 2004 ), whereas in aquatic systems it has played a
far more prominent role in thinking about community structure. This can be
attributed to several important differences between terrestrial and aquatic
systems. First, body size is a concept that is applicable to many plants and
detrital particles in aquatic systems, particularly in pelagic systems, because
of the way many organisms feed, and the fact that material is carried in suspen-
sion. Second, predation, which is often clearly size dependent, has long been
recognized as a structuring force in aquatic systems, whereas a greater emphasis
in terrestrial systems has been placed on the role of lateral competitive inter-
actions within a single trophic level, or on plant–herbivore–parasitoid food
chains, which form a major component of terrestrial systems but where size
has a less obvious role (e.g. Lehman & Tilman,2000). Finally, the size distribu-
tion of particles (including organisms) is relatively easy to measure in aquatic
systems with automated counting systems: these include flow cytometry of
algae and ciliates using Coulter^1 counters (Pitta, Giannakourou & Christaki,
2001 ), continuous plankton recorders attached to ocean-going vessels (Beare
et al., 2003) and remote sensing of fish populations using passive integrated
transponder (PIT) tags (Achord, Levin & Zabel,2003). It is perhaps not surprising,
then, that there has been a greater focus on the role of body size in determining
the structure and dynamics of aquatic food webs (e.g. Kerr,1974; Cousins, 1980 ,
1985 ; Woodward & Hildrew,2001, 2002b; Schmid-Arayaet al., 2002a; Woodward
et al., 2005a–c) than in terrestrial systems, with the notable exception of soil food
webs which also include traditional predators and prey and size-delimited
trophic levels (e.g. de Ruiteret al., 1995).
Empirical data on body-size patterns and effects from aquatic systems have
stimulated thinking about community organization and, in parallel with this,
size has for some time been drawn into theoretical ideas about aquatic com-
munity dynamics (e.g. Kerr1974; Cousins, 1980 , 1985; Emmerson & Raffaelli,
2004 ; Reuman & Cohen, 2005 ). However, the demands of theory for data, both
for parameterization and for testing predictions, has increasingly outstripped
their availability, both in terms of numbers of studies and also the scale at which
data are collected (Brown & Gillooly,2003). For example, food-web studies – both
empirical and theoretical – usually deal with species, or higher taxonomic (or
other) aggregations, where the averaging of characteristics such as size obscures
many of the details of what individuals are actually doing. At the other end of
the scale, laboratory experiments often focus on size effects within limited size
ranges of individuals of one or two species, while community theory often seeks
to apply such results across species. Perhaps the greatest bridging of scales
comes with macroecological relationships, where the effects of size on funda-
mental processes such as metabolic rate, are linked to large-scale patterns
in species distributions and diversity (Brown, 1995 ; Gaston & Blackburn,
2000 ). What such analyses make explicit is that the patterns that arise at
BODY SIZE AND PREDATORY INTERACTIONS IN FRESHWATERS 99