Page 14 - Volume 6, Issue 2, July 2013

© Benaki Phytopathological Institute
Predation of P. quatuordecimpunctata on A. fabae 61
where N e is the number of prey consumed,
N0 the initial prey number available and P0,
P1, P2 and P3 are the intercept, linear, quadrat-
ic and cubic coefficients estimated using
the method of maximum likelihood. Regres-
sion outcome suggested a type II function-
al response. The visual inspection of the ob-
served vs expected probabilities showed
good agreement, as the proportion of prey
eaten decreased as prey density increased.
Data fitted to the random predator
equation proposed by Rogers (1972). In or-
der to derive more stable estimates of calcu-
lated parameters, the Lambert W function
was used, which provides a numerical solu-
tion of the equation. In terms of Lambert W,
the random predator equation can be writ-
ten as (Bolker, 2008):
where a is the attack rate, i.e. the per cap-
ita prey mortality at low prey densities; b
the handling time, T the total time that prey
was exposed to predator; T/b the maximum
number of prey which can be attacked by
the predator during the time interval (maxi-
mum attack rate). Fitting was performed us-
ing the method of maximum likelihood (R
Development Core Team, 2010). The mortal-
ity of the prey was assumed negligible in the
absence of predator. Significant differences
between estimated parameters were evalu-
ated using 95% confidence intervals.
Results
The estimated parameters of the logistic
regression analysis on the proportion of
prey eaten as a function of initial prey den-
sity are presented in Table 1. Fitting vs ob-
served probabilities of the proportion of A.
fabae prey eaten by each of the larval instars
of P. quatuordecimpunctata suggest a con-
tinuous decrease in prey consumption with
increasing prey density (Fig. 1). Thus, an in-
verse density-dependent prey mortality ex-
hibited by P. quatuordecimpunctata larvae
(type II functional response).
The random predator equation was used
to describe the predation rates of P. quat-
uordecimpunctata larval instars. The mod-
el fitted the observed data reasonably well,
as the fitted probabilities of the number of
prey consumed lie within the main bulk of
the data (Fig. 2). Although the estimated
attack rates did not differ among larval in-
stars, maximum attack rates increased at
the fourth instar larvae. Thus, the maximum
numbers of prey which can be attacked
were 2.3, 5.0, 14.5 and 34.5 for the first, sec-
ond, third and fourth instar larvae, respec-
tively (Table 2).
Discussion
Our results show that predation ability for
each larval instar of P. quatuordecimpunc-
tata increases with a deceleration rate,
as A. fabae density increases, indicating a
type II functional response. Papanikolaou
et al. (2011) showed that a type II function-
al response was also exhibited by P. quat-
uordecimpunctata larval instars to A. fabae
when exposure time was 24 h. It is expect-
ed that type II functional responses affect
negatively the population dynamics of prey
and predator (Oaten and Murdoch, 1975). As
the intrinsic rate of increase and population
doubling time of P. quatuordecimpunctata
support the potential of a successful biolog-
ical control agent (Kontodimas et al., 2008;
Pervez and Omkar, 2011), these data should
also be taken into account for final conclu-
sions on the performance of the predator in
biological control practice.
Our study supported that the handling
times of P. quatuordecimpunctata of the old-
er larval instars were shorter than those of
the younger ones, hence the predation abil-
ity increased at the fourth instar. This is due
to the higher energy requirements of the
fourth instar larvae and their ability to pur-
sue, capture and consume faster the prey
items. Papanikolaou et al. (2011) reported the
same trend in handling times of larval instars
of P. quatuordecimpunctata, where the mean