weight (Grenier and Bonnot, 1983). As a con-
sequence, differences in pupal weights
between in vitro- and in vivo-reared E. lar-
varumare reflected in the adult stage (see
also below).
In many cases, adults of predatory insects
from various orders reared on artificial food
are smaller than those obtained on insect
prey (e.g. Racioppi et al., 1981; Hussein and
Hagen, 1991; De Clercq and Degheele, 1992;
Hattingh and Samways, 1993; Ogura and
Hosada, 1995; Zanuncio et al., 1996; De
Clercq et al., 1998a; Cohen, 2000a). However,
Rojas et al.(2000) and Arijs and De Clercq
(2001) obtained female adults of the preda-
tory bugs P. bioculatusand Orius laevigatus,
respectively, with equal weights to those of
predators maintained on insect prey.
AbnormalitiesAbnormalities observed in artificial rearing
are usually deformations of wings (expanded
or not) and of the abdomen. The percentage
of deformed females, i.e. mainly unexpanded
wings, is a criterion for measuring abnormal-
ities in Trichogramma (Cerutti and Bigler,
1995). Bloated larvae with unexpanded wings
are produced in vitrodue to too few eggs per
artificial host egg (Grenier et al., 1995). Wing
and abdominal deformations occurred in
Trichogrammareared on artificial diet, but no
differences in the morphology of their genital
apparatus were detected (Consoli and Parra,
1996). The same wing and body anomalies
were recorded after Trichogrammawas pro-
duced for ten generations in vitro(Nordlund
et al., 1997). Suboptimal diets have also been
reported to cause deformation of wings and
abnormal pigmentation in some predatory
coccinellids (Atallah and Newsom, 1966;
Racioppi et al., 1981).
Development and reproduction parametersDuration of different developmental stagesMany workers have assessed developmental
rates of immature stages to evaluate the
nutritional value of a diet. Usually the imma-
ture development of parasitoids (from egg or
first-instar larva to adult emergence) is
longer under artificial conditions than in nat-
ural hosts. This was observed in the tachinid
fly E. larvarum(Dindo et al., 1999) and in the
ichneumonid D. introita (Greany and
Carpenter, 1998; Ferkovich et al., 1999). The
development from egg to adult on artificial
diets was delayed for Trichogramma pretiosum
compared with natural hosts, but no differ-
ences were observed for Trichogramma galloi
(Consoli and Parra, 1996). The development
of the egg stage was not very often tested,
but for Trichogramma australicum it took
longer in artificial diet vs. natural-host eggs
(Dahlan and Gordh, 1998). In T. australicum
also the larval stages develop more slowly in
artificial diet vs. natural-host eggs (Dahlan
and Gordh, 1998). Total development time is
not different in Bracon mellitorand Catolaccus
grandisreared in vivovs. in vitro(Guerra et
al., 1993; Guerra and Martinez, 1994).
Usually, when the whole preimaginal devel-
opment is longer, only the larval develop-
ment is delayed, the embryonic and pupal
development being less dependent on the
diet composition.
Likewise, developmental times in preda-
tors are usually longer on artificial diets.
Some workers have even reported total
developmental periods two to three times
longer than those observed in controls on
natural food (Hagen and Tassan, 1965;
Vanderzant, 1969; Racioppi et al., 1981;
Hussein and Hagen, 1991). Whereas in a
number of studies differences in develop-
mental times were consistent across all larval
instars, differences in some species became
more important in later instars (Racioppi et
al., 1981; Zanuncio et al., 1996; Wittmeyer and
Coudron, 2001). Because Wittmeyer and
Coudron (2001) found differences in develop-
mental durations between in vitro- and in
vivo-reared P. maculiventristo be most marked
in the fifth (final) nymphal stage, they sug-
gested selecting this stage to evaluate the
suitability of a diet. Similar developmental
periods for artificially and naturally reared
predators have occasionally been reported,
e.g. by Grenier et al.(1989) for Macrolophus
caliginosus, by De Clercq et al.(1998a) for indi-
vidually reared P. maculiventrisand by Arijs
and De Clercq (2001) for O. laevigatus.120 S. Grenier and P. De Clercq