associated with the use of unnatural food
implies a greater risk that unadapted geno-
types (i.e. those that will be inefficient in con-
trolling a pest in the field) will eventually
dominate the laboratory population.
Unfortunately, little effort has been made to
quantify genetic variability in entomophages
bred on artificial media, e.g. by allozymic
analyses or by direct analysis of DNA poly-
morphisms via PCR-based techniques.
Further, inbreeding can often increase
homozygosity and may thus increase the risk
that recessive deleterious alleles are
expressed. In species where there is no
inbreeding depression, crosses among differ-
ent inbred strains of the same species could be
a useful tool for minimizing the effects of lab-
oratory adaptation, as in Trichogrammanr.
brassicae. The persistence of genetic variation
for several traits within this species indicates
heritable variation, which could be manipu-
lated, when necessary, to optimize the effi-
ciency of the strain used in biological control.
Thus, inbreeding might provide a way to
increase the useful life of a strain in mass-
release programmes (Sorati et al., 1996). Given
the lack of inbreeding depression even after
many generations of inbreeding, the same
was also believed to be true for the predator P.
maculiventrisproduced in vivo(De Clercq et
al., 1998b). Conversely, outbreeding (or het-
erosis), as in the crossing of Cotesia flavipes
strains from different origins, may improve
the performance of a parasitoid, including
female sex ratio, rate of development and size
of females (Gu and Dorn, 2000).
Conclusions
Tests for quality comparisons between nat-
ural enemies that were reared artificially or
on their natural host were mainly conducted
on the first generations after in vitroculture,
but on rare occasions effects of continuous
culture for several generations have been
tested (e.g. Hassan and Hagen, 1978; Gao et
al., 1982; De Clercq and Degheele, 1992, 1993;
Nordlund et al., 1997; Cohen, 2000a). We sug-
gest that it may not be advisable to maintain
entomophagous insects on synthetic diets for
many generations, because they may suffer
from non-intentional selection, inducing a
reduction in genetic variability and finally a
deterioration in performance. On the other
hand, the frequent introduction of new
strains to initiate in vitromass production
could generate inconveniences, such as the
necessity for a few generations of laboratory
adaptation, the risk of misidentification of
the introduced strain or species and the dan-
ger of introducing pathogens or hyperpara-
sitoids (see Chapters 1 and 10).
The ultimate test for quality of ento-
mophagous insects is the assessment of their
field efficiency measured as the rate of para-
sitism or predation. However, besides being
expensive and time-consuming, the com-
plexity of a field setting may obscure the
actual causes for the failure or success of nat-
ural-enemy releases. Therefore, the first
assessment of the quality of an in vitro- or in
vivo-produced beneficial will preferably be
done in a laboratory setting.
Currently, quality control of in vitro-reared
entomophagous insects has been done for the
major part only by comparing selected char-
acteristics between in vitro- and in vivo-grown
insects in the laboratory. Obviously, such
comparisons should be done in a fair way,
with artificial diets being compared with the
best natural-rearing protocols. Further, it is
important to try to define which parameters
should be considered as key criteria to be
tested in a first quality assessment of ento-
mophages. Fecundity and the rate of parasiti-
zation in parasitoids and the predation
capacity in predators are probably the most
relevant criteria for estimating the ultimate
quality of a natural enemy (Table 9.1).
At the laboratory level, however, such bio-
logical parameters could be associated with
biochemical parameters, as we demonstrated
above. We believe that it is worthwhile
assessing these biochemical parameters
because, contrary to biological traits, they can
be used to suggest modifications of the in
vitro-rearing system, eventually leading to an
improvement of the insects produced. Excess
or deficiencies of some elements could be bal-
anced by deletion or supplementation of
nutritional components in the diet, based on
a better understanding of the nutritional
physiology of an insect. One could say that126 S. Grenier and P. De Clercq