Theoretical models for the evolution of inducible anti-predator defens
es predict that there should be fitness costs associated with the defe
nse, otherwise it should be permanent. Thus, in the absence of predato
rs and with limited resources, defended individuals are expected to be
outcompeted by individuals lacking the costly defense. Crucian carp,
Carassius carassius, increase in body depth in response to chemical st
imuli from piscivorous fish, and the deeper body constitutes a morphol
ogical defense against gape-limited piscivores. Here, we have performe
d a field experiment investigating effects of population density on th
e relative competitive performance of shallow-bodied and deep-bodied c
rucian carp. We assigned groups of six individually marked fish (the '
'focal'' individuals) of either morph into enclosures with either low
or high crucian carp density, and monitored changes in body mass, leng
th, and body depth of these fish for 4 mo. In addition, we determined
sex, and final liver and gonadal mass, of all focal fish. At high popu
lation density, there was a reduction in food resources (zooplankters)
, which resulted in reduced gonad mass, reduced gonadosomatic indices,
and reduced growth in fish of both morphs. Unexpectedly, the relative
allocation to livers was higher at high density. There was no differe
nce in body mass gain between morphs at the low density, whereas at th
e high density shallow-bodied fish gained twice as much body mass as d
eep-bodied fish. Hence, when predators were absent and food resources
were limited, deep-bodied individuals suffered a substantial cost when
competing with shallow-bodied conspecifics. This study verifies the t
heoretical prediction that an inducible defense results in fitness cos
ts in the absence of predators. Further, as we found no measurable dis
advantage of the deep-bodied morphology at high food availability but
a strong disadvantage under intense competition for food, our results
also demonstrate the importance of relating defense costs to levels of
intraspecific competition.