Vector Control Using Semiochemicals - Biological and Environmental Control of Disease Vectors - page 106

pit with the odour ventilated to a trap could trap
tsetse fl ies without the presence of any visual
cues (Vale, 1974). Also, by increasing the
number of hosts in a pit, and thereby increasing
the amount of host-derived semiochemicals
being produced, the number of tsetse fl ies
attracted increased signifi cantly (Hargrove and
Vale, 1978; Hargrove et al. , 1995). Further
experiments have demonstrated that odour is a
signifi cant stimulant for long-range attraction
of tsetse fl ies, but the fi nal landing response is
mediated by a combination of visual cues
(predominantly) and host odours (Vale, 1974).
For this reason, tsetse fl y trap targets usually
combine specifi c trap shapes and colours (e.g.
blue and black) with host odours to provide the
suite of host stimuli required for the full
attraction and landing or entry response.
The semiochemicals used in tsetse
control include CO 2 , 1-octen-3-ol, acetone,
4-methylphenol, 3-propylphenol and butanone
(Vale, 1980; Hassanali et al. , 1986; Torr et al. ,
1995). These semiochemicals were identifi ed by
examining the odour profi les of natural hosts, ox
in particular. The role of each of these chemicals
has been studied in laboratory and fi eld settings,
and each is involved in dif erent aspects of the
host location process, demonstrating the
importance of understanding the behavioural
action of semiochemicals when exploiting them
in traps appropriately. For example, 'fl ight
activation' is likely to only involve CO 2 . Carbon
dioxide, 1-octen-3-ol and acetone may be
involved in long-range responses, while CO 2 and
other unidentifi ed host odours af ect landing
behaviour (Torr, 1988, 1989, 1990). The most
ef ective blend of these chemicals, when released
at a dose that mimics a natural host, does not
attract as many insects as a natural host, which
suggests that other semiochemicals are yet to be
identifi ed that may increase trap catches further
(Torr et al. , 2006).
The fi rst successful demonstration of host
odour-baited traps and insecticide-treated
targets against tsetse was in the 1980s in
Zimbabwe (Vale et al. , 1986). In this study, an
island (5 km 2 ) population of Glossina morsitans
morsitans and G. pallidipes tsetse fl ies (population
estimate approximately 5000 males) was
reduced by 90% and 99%, respectively, through
the deployment of six traps containing CO 2 and
acetone. The traps were able to capture between

0.1% and 4% of the population over the course
of around 2 years. Once this reduction in
population had been achieved, the traps were
then replaced by 20 insecticide-treated targets
with acetone and 1-octen-3-ol baits, killing a
greater number of tsetse fl ies and resulting in
complete elimination of both species within 9
months (Vale, 1993). A second study over a
larger area (1000 km 2 ) successfully reduced G.
pallidipes by more than 99.9% (Vale et al. ,
1988). After the success of these two key trials,
this semiochemical-baited technology was
deployed in several countries. Baits for G.
pallidipes tsetse fl ies were later enhanced by the
identifi cation and addition of 4-methylphenol
and 3-propylphenol (Torr et al. , 1997) and now
form the 'POCA' blend, which is used to this day
in control programmes.
Although these semiochemicals have been
used successfully for the control of morsitans
(savannah) group tsetse fl ies, they have been far
less successful with the palpalis (riverine) group.
This lack of ei cacy is likely to be partially due to
the fact that their host range is far more variable
than the morsitans group tsetse, meaning that
the mixtures of ox-derived semiochemicals may
not be relevant to the more opportunistic
palpalis group fl ies. The host range of some
species has been shown to vary by host
availability and season, e.g. G. palpalis will feed
on pigs if they are abundant but in their absence
will readily feed on human beings and in the cold
season, they switch hosts to reptiles (Torr and
Solano, 2010). This highlights the need to
understand the relationship between vector,
host and environment before semiochemicals
can be exploited appropriately. This lack of
ei cacy is also because it has been generally
considered by previous research that palpalis
tsetse fl ies do not respond strongly to
semiochemicals unlike their morsitans group
counterparts. However, recent research has
demonstrated that host odours can mediate
signifi cant responses in palpalis group tsetse fl ies
(Rayaisse et al. , 2010), and researchers have
managed to tease apart the responses to dif erent
hosts by particular species. For example, G. f.
fuscipes responds to odours from monitor lizards,
but not cattle or humans, G. f. quanzensis
responds to pig odour, while G. gambiensis
responds to ox odour (Omolo et al. , 2009;
Rayaisse et al. , 2010). However, the responses to