Control with Arthropods - Biological and Environmental Control of Disease Vectors - page 20

al ., 2007; Chandra et al ., 2008; Mandal et al .,
2008; Kweka et al ., 2011), and therefore even
though their predatory capacity has been
validated, it is not known whether they will be
able to be mass reared and deployed with as
much ease as copepods have been. Even
Toxorhynchites , which have operationally been
used for dengue vector control, are not easy to
mass produce, with individual rearing required
in some circumstances to prevent cannibalism
(Focks, 2007). On the other hand, copepods can
be easily mass produced and transported (Nam
et al ., 2000), and this may be one of the reasons
why they have been the most successfully
deployed arthropod predator.
The second reason for problems with the
deployment of arthropods in Anopheles and
Culex vector control is ovipositional avoidance
(Shaalan and Canyon, 2009). As discussed
above, Stegomyia control can be compromised if
Toxorhynchites and Stegomyia species do not
share ovipositional preferences. Furthermore, if
mosquitoes are repelled from potential
oviposition sites by a control agent, then that
control agent cannot of er sustainable control
because successive generations of the vector
mosquito will not be brought into contact with it
(Howard et al ., 2011). Therefore, it is important
to test whether proposed larval control tools
negatively af ect mosquito oviposition. Wild
Culex mosquitoes exhibited a strong ovipositional
avoidance of backswimmer ( Notonecta irrorata )
infested pools in America; during predator
infestation, 83% of egg rafts were found in the
control pools whereas there had been no
signifi cant dif erence in preference before
predator introduction. This avoidance was
hypothesized to be driven by predator-released
chemicals, since the increased oviposition in the
control habitat remained for 2 days after the
removal of the backswimmers (Blaustein et al .,
2005). This ovipositional avoidance has also
been shown in the laboratory. Munga et al .
(2006) placed Notonecta sp. into rainwater for 3
days in order to condition the water. They then
set up an oviposition choice experiment where
gravid An . gambiae mosquitoes could choose
between the conditioned backswimmer water
and unconditioned rainwater. Signifi cantly
more eggs were laid in the unconditioned when
compared to the Notonecta- conditioned habitat
(Munga et al ., 2006). Recently, Warburg et al .

(2011) confi rmed these results by also showing
that An . gambiae avoided ovipositing in
Notonecta- conditioned water. Specifi c Notonecta
maculata- released hydrocarbons known to repel
oviposition of the mosquito Culiseta longiareolata
were used to test for oviposition avoidance.
However, these hydrocarbons did not repel An .
gambiae oviposition, suggesting that the two
mosquito genera use dif erent chemical cues to
detect whether predators are present in a
potential oviposition site. On the other hand, the
presence of copepods has been shown to attract
ovipositing Stegomyia mosquitoes (Torres-

Estrada et al ., 2001).
The third reason that further work looking
at fi eld deployments of arthropod predators for
anopheline and culicine mosquito control may
not have proceeded is that, at present, there are
more ef ective and more viable biological
alternatives currently used or under investi-
gation. Larvivorous fi sh (see Chandra et al .,
Chapter 3, this volume) can reduce malaria
vector numbers by 96% (Howard et al ., 2007).
Environmental management has also been
shown to be ef ective at reducing disease
transmission (Keiser et al ., 2005) (see Ensink et
al ., Chapter 8, this volume and Konradsen et al .,
Chapter 9, this volume), as has house
modifi cation (Kirby et al ., 2009; see Kirby,
Chapter 7, this volume). The biopesticides Bti
and Bacillus sphaericus (Bs) are commercially
available and are being used to control
mosquitoes all over the world (Lacey, 2007).
Entomopathogenic fungi are another promising
tool that could be used against dengue (de Paula
et al ., 2008), malaria (Howard et al ., 2010) and
fi lariasis (Scholte et al ., 2003) vectors in the
future (see Stevenson et al ., Chapter 5, this
volume).
For all disease control strategies, com-
ponents of an IVM programme should be
chosen on merit. Factors relating to cost-
ef ectiveness, ei ciency, environmental impact
and the specifi c local situation need to be
considered (Lacey and Orr, 1994; WHO, 2004).
At present, the only arthropods that fulfi l the
necessary requirements to be intentionally
integrated into an IVM programme are copepod
predators (Kay and Nam, 2005). While
Toxorhynchites have shown some promise, there
are still problems associated with their
deployment. Currently, the use of arthropods to