BACKGROUND : The Mediterranean fruit fly, Ceratitis capitata (Wiedemann) (Diptera: Tephritidae) is a highly invasive
species now with an almost cosmopolitan distribution. Two other damaging, polyphagous and closely-related
species, the marula fruit fly, Ceratitis cosyra (Walker), and the Natal fly, Ceratitis rosa Karsch, are not established
outside of sub-Saharan Africa. In this study, adult water balance traits and nutritional body composition were
measured in all three species at different temperatures and levels of relative humidity to determine whether
tolerance of water stress may partially explain their distribution.
RESULTS : Adult C. capitata exhibited higher desiccation resistance than C. rosa but not C. cosyra. Desiccation
resistance of C. capitata was associated with lower rates of water loss under hot and dry conditions, higher
dehydration tolerance, and higher lipid reserves that were catabolised during water stress. In comparison with
C. capitata, C. cosyra and C. rosa lost water at significantly higher rates under hot, dry conditions, and did not
catabolise lipids or other sources of metabolic water during water stress.
CONCLUSIONS : These results suggest that adult physiological traits permitting higher tolerance of water stress
play a role in the success of C. capitata, particularly relative to C. rosa. The distribution of C. cosyra is likely
determined by the interaction of temperature with water stress, as well as the availability of suitable hosts for
Additional file 1: Table S1. Linear regression for the relationship
between body mass (mg) and body water content (mg) for cohorts of
three Ceratitis species that were subsequently tested for desiccation resistance and water loss rate at two temperatures. The equation for each
relationship was used to estimate initial body water content from initial
body mass for flies subjected to desiccation and water loss rate assays.
Additional file 2: Table S2. Analysis of deviance table for the final fitted
parametric survival model that describes desiccation resistance of three
Ceratitis species with respect to species, sex, temperature (Temp) and
relative humidity (RH). Initial body mass was included as a covariate in the
model. Data were fitted to a Weibull hazard function. Type III likelihood ratio
tests were used to construct the analysis of deviance table. Significant
effects (P < 0.05) are indicated by bold type.
Additional file 3: Table S3. General linear model for the relationship
between species, sex, temperature (Temp) and relative humidity (RH) on
the dehydration tolerance of three Ceratitis species. Estimated body water
(determined from initial body mass using the equations in Table S1) was
included as a covariate. Significant effects (P < 0.05) are indicated by bold
Additional file 4: Table S4. General linear model for the relationship
between species, sex, temperature (Temp) and relative humidity (RH)
on water loss rate of three Ceratitis species. Initial mass was included
as a covariate. Significant effects (P < 0.05) are indicated by bold type.