ECOLOGY OF PRIMARY TERRESTRIAL CONSUMERS 257
sheep, cattle and deer down to 40% for moose and elephant
(Graham, 1964; Dinesman, 1967; Davis and Golley, op. cit.).
Interestingly, insect herbivores appear to be considerably less
efficient feeders, reaching assimilation levels of 29% for the
caterpillar Hyphantria, 27% for the grasshopper Orchelimum,
36% for the grasshopper Melanoplus, and 33% for the spittle-
bug Philaenus (Gere, 1956; Smalley, 1960; Wiegert, 1964,
1965). Detritus feeders seem to have even lower efficiencies,
as witness values of 20% for oribatid mites and 10% for the
millipede Glomeris (Engelmann, 1961; Bocock, 1963). Thus,
grazing food-chains and detritus food-chains appear to be
characterized by quite different assimilation rates.
Part of the energy assimilated by herbivores is stored as
growth or production, while the rest is respired, dissipated
as heat in the oxidation of organic matter. These respiratory
losses may amount to as much as 98% of the energy assimi-
lated (Petrusewicz and Macfadyen, 1970) but vary greatly,
depending on such factors as environmental temperature,
level of activity, and age of the individual. In mammals,
there is a tendency for the weight-specific respiration rate to
rise as body weight decreases, and this, in combination with
the fact that growth and reproduction rates tend to be greater
in small species than in large ones, apparently results in a
rather constant low production/assimilation ratio of 1–2%
for mammalian populations; values of this magnitude have
been calculated for both elephants and mice (Wiegert and
Evans, 1967). In contrast, herbivorous insect populations
seem to show much higher assimilation efficiencies, on the
order of 35–45% and, when only young life stages (larvae,
nymphs) are involved, even of 50–60% (Macfadyen, 1967).
Data are still too few to permit satisfactory generalizations,
but the importance of further studies is evident when it is
remembered that energy lost in respiration is irrevocably
removed from the ecosystem and cannot be recycled.
Synthesis of these process ratios leads to an evaluation
of the overall efficiency with which herbivores convert the
energy of primary (plant) production into their own tissue.
The value of the production/consumption ratio for a fairly
broad spectrum of vertebrate and invertebrate herbivores
seems to range from less than 1% to 15–20% as a maximum
(Petrusewicz and Macfadyen, 1970), with some evidence
that the insects are generally more efficient than the non-
domesticated mammals; this difference may be due in large
part to the necessity for the latter to maintain a steady, high
body temperature. When these values are considered along
with the proportions of available food ingested (see above),
the efficiency of energy transfer of herbivores is rarely found
to exceed 15%, and this has been suggested as a likely maxi-
mum level for natural ecosystems (Slobodkin, 1962).REGULATORY MECHANISMS OF
PRIMARY CONSUMERSAlthough terrestrial herbivores are occasionally so abundant
as to deplete the local supply of plant food, as sometimes
happens when an introduced species like the Japanese beetle
( Popilia japonica ) or the Europena gypsy moth ( Porthetriadispar ) enters a new biotic community, such cases seem to
be the exception rather than the rule, and, in contrast to many
aquatic ecosystems, the bulk of net annual primary production
on land is not eaten in the living state by primary consumers
but dies and is acted on by decomposer organisms. Thus it
appears that, on the whole, land herbivores usually occur at
densities well below the level of their available plant food, and
the reason or reasons for this are of great ecological interest.
Despite the fact that sound generalizations about abstract
concepts such as “trophic levels” are not easily arrived at
(Murdoch, 1966), several hypotheses about the regulation of
herbivore numbers have been suggested. The apparent rarity
of obvious depletion of vegetation by herbivores, or of its
destruction by meteorological catastrophes, has led Hairston,
Smith and Slobodkin (1960) to the belief that herbivores are
seldom limited by their food supply; after rejecting weather
as an effective control agent, they conclude that herbivores
are most often controlled by their predators and/or para-
sites, interacting in the classical density-dependent manner.
These views have been questioned on such grounds as
(1) that much green material may often be inedible, unpalatable
or even unreachable by the herbivores present, so that food
limitation might occur without actual depletion (Murdoch,
1966) or (2) that native herbivores such as forest Lepidoptera
and grasshoppers will often increase and cause serious defo-
liation even in the presence of their predators (Ehrlich and
Birch, 1967). Despite these and other criticisms, Slobodkin,
Smith and Hairston (1967) find it unnecessary to modify
their views in any essential respect.
The possibility that herbivore (and other animal) popu-
lations have some capacity for self-regulation has not been
overlooked. Pimentel (1961, 1968) has suggested the concept
of “ genetic feedback ,” according to which population density
influences the intensity of selective pressure, selection influ-
ences the genetic composition of the surviving individuals,
and genetic composition influences the subsequent popula-
tion density. The models thus far proposed to explain how this
system works appear to be untenable (Lomnicki, 1971), but
the general validity of the concept seems to gain support from
the numerous instances of the co-evolution of plants and her-
bivorous animals, e.g., the association between certain orchids
and their insect pollinators (Van der Pijl and Dodson, 1966),
or that between certain species of Acacia and ants (Janzen,
1966), which have been interpreted as involving reciprocal
selective interaction.
Another suggested mechanism for self-regulation involves
the elaboration of social behavior patterns, e.g., territoriality,
social hierarchies, and warning displays, which tend to main-
tain animal populations at relatively low densities, thereby
reducing the possibility of depletion of food supplies or other
resources (Wynne Edwards, 1962, 1965). The evolution of
such a mechanism seems to require that natural selection act
on groups rather than on individual organisms, and for this
reason Wynne Edwards’ hypothesis has been heavily criti-
cized (Williams, 1966; Maynard Smith, 1964).
The basic importance of self-regulatory mechanisms
and other density-dependent interactions in limiting popula-
tions of animals has also been questioned. Ehrlich and BirchC005_002_r03.indd 257C005_002_r03.indd 257 11/18/2005 10:20:40 AM11/18/2005 10:20:40 AM