High levels of survivorship of all

High levels of survivorship of all stages of C. capitata
were observed over the range 15–30°C consistent with previous studies (Messenger & Flitters, 1958; Crovetti et al.,
1986; Delrio et al., 1986; Vargas et al., 1996). Developmental
time of C. capitata from egg to adult ranged from 16 to 64
days, between 30 and 15°C. These results for the immature
stages are in general agreement with data from previous
studies (Messenger & Flitters, 1958; Tassan et al., 1983;
Crovetti et al., 1986; Delrio et al., 1986; Vargas et al., 1996).
Values for the temperature threshold and thermal
constant were also consistent with previous studies, except
for the period of larval development. In Hawaii, Vargas et al.
(1996) found a lower temperature threshold of 5.2°C and a
thermal constant of 139 DD for the larval stages of C.
capitata, when using a linear model and working over a
range of constant temperatures from 16 to 32°C, similar to
the range used in our study. These important differences
could result from the utilization of different rearing diets, the
stock used, or from rearing conditions (e.g. larval density).
These authors, however, recognized that the temperature
threshold for larval development was probably
underestimated in their study. The low value of this
threshold in their study may have resulted from the
behaviour of cohorts: larvae were observed crowding at the
bottom of diet cups, probably to keep warm. Indeed, high
densities of larvae in diet produce high levels of metabolic
heat which can also affect developmental rate (Tanaka et al.,
1972; Hooper, 1978). This behaviour was not observed in our
study because of the low density of larvae used. Moreover,
the threshold for egg plus larval stage calculated by Tassan
et al. (1983) (t = 9.7°C) match our results. As the linear
temperature summation model is less valid at extremes of
temperature, it would be interesting to substantiate our data
with further studies at lower temperatures, close to the
developmental threshold, using non-linear models
(Schoolfield et al., 1981; Wagner et al., 1984).
The lower developmental threshold for ovarian
maturation of C. capitata has been estimated as 16.6°C by
Tassan et al. (1983) compared to 8.1°C in our study. This
difference may be due to laboratory adaptation of the strains
tested or to different biological characteristics of C. capitata
strains, as has been observed in other groups of insects
(Lopez-Edwards et al., 1999).
No previous work has been published on the
development of C. rosa at different temperatures. No
immature stages of C. rosa were able to develop at 35°C and
survivorship was low at 30°C. Mean larval development
time and duration of ovarian maturation were significantly
longer in C. rosa than C. capitata at all temperatures.
The biology of C. catoirii has not been studied before. The
survivorship of this species was generally lower than that of
C. rosa and C. capitata at the temperatures studied. The
rearing diet used for C. catoirii in this study was developed
originally for C. rosa and may not be optimal for this species.
Lower survivorship rates of C. catoirii may be linked more to
the dietary constraints factors than to intrinsic differences in
survivorship abilities, though this rate is higher than that of
C. rosa and that of C. capitata at certain given temperatures. It
would be necessary to check the results obtained with C.
catoirii on other artificial diets or on host fruits.
Linear regressions of developmental rate against
temperature for all three species show that most of the
correlation coefficients are close to one, indicating a strong
linearity of the model between 15 and 30°C. The temperature
summation model is thus a convenient means for estimating
development times of these species over the range of
temperatures studied. As the upper developmental
thresholds have not been precisely investigated, it would be
interesting to study development at temperatures ranging
from 30 to 35°C in more detail to establish whether the
species exhibit distinct temperature preferences.
In our study, the duration of ovarian maturation was
used to assess the complete life cycle from egg-laying of one
generation to egg-laying of the next one. However, Kasana &
Aliniazee (1994), in their study on the effect of temperature
on the pre-oviposition period of Rhagoletis completa Cresson
(Diptera: Tephritidae), showed that numerous females never
laid eggs despite the presence of mature eggs in their
ovaries. If such a phenomenon should exist in the studied
species, then the duration of their whole life-cycle would be
underestimated in our study.
Comparisons of developmental times among the
different species show that they differ mostly during the
larval stages. As far as pre-imaginal development is
concerned, C. capitata has a shorter life-cycle than the two
other species within the range of temperatures studied. At
25°C for instance, the life-cycle of C. capitata (18 days) is
three days shorter than that of C. catoirii and five days