PHYSIOLOGICAL ECOLOGY
Effects of Temperature on Fecundity and Longevity of Phoracantha
recurva and P. semipunctata (Coleoptera: Cerambycidae)
LINDA F. BYBEE, JOCELYN G. MILLAR, TIMOTHY D. PAINE,1 KATHLEEN CAMPBELL,
AND CHRISTOPHER C. HANLON
Department of Entomology, University of California, Riverside, CA 92521
Environ. Entomol. 33(2): 138Ð146 (2004)
ABSTRACT The cerambycid beetles Phoracantha recurva Newman and Phoracantha semipunctata
F. were accidentally introduced into southern California where they have caused signiÞcant mortality
to eucalyptus trees. In recent years, P. recurva populations have increased rapidly with a concomitant
decrease in P. semipunctata numbers in their shared habitat in southern California. P. recurva has been
collected in Þeld surveys earlier in the season, suggesting that this species may be active over a broader
range of temperatures and consequently have a longer seasonal activity period. We conducted lifetime
fecundity studies to examine egg production, egg hatch, and adult longevity of beetles of both species
subjected to temperatures typical of late fall and late winter/early spring in southern California.
Overall, there were no differences in lifetime fecundity between P. recurva and P. semipunctata. P.
recurva did not seem to reproduce over a larger range of temperatures than P. semipunctata, and there
did not seem to be any substantial differences in their reproductive biologies. P. recurva eclosed slightly
faster and lived slightly longer at reproductive temperatures than P. semipunctata, suggesting that P.
recurva may have a slight reproductive advantage over P. semipunctata. However, it seems unlikely that
such an advantage can account for the rapid increase in P. recurva populations and the simultaneous
decrease in P. semipunctata populations in southern California.
KEY WORDS Phoracantha recurva, Phoracantha semipunctata, Cerambycidae, fecundity, longevity
IN ITS COUNTRY OF ORIGIN, Australia, the eucalyptus
longhorned borer, Phoracantha semipunctata F. (Coleoptera: Cerambycidae), is a minor pest of Eucalyptus
L’Hér (Browne 1968, Moore 1963), primarily infesting
dead or fallen limbs or unhealthy trees (Tooke 1935;
Browne 1968; Chararas 1969a,b). Recently, P. semipunctata has become a severe pest of eucalyptus in
countries where both the trees and the beetle have
been introduced (Bytinski-Salz and Newmark 1952,
Ivory 1977, Gonzalez-Tirado 1986). It was Þrst detected in California in 1984 (Scriven et al. 1986), and
since then has spread throughout most of the state,
causing signiÞcant tree mortality (Paine et al. 1995,
Hanks et al. 1997).
In southern California, P. semipunctata adult emergence is not tightly synchronized, and beetles are
present continuously from late spring until fall (Hanks
et al. 1993a). Adults of both sexes are attracted to
volatile compounds emitted by eucalyptus trees and
logs, where they mate and oviposit (Hanks et al.
1996b). Egg masses are laid under loose bark and in
bark cracks (Hanks et al. 1993a,b; Paine et al. 1995).
Eggs hatch in 3Ð5 d, and neonate larvae penetrate the
bark and feed along the cambium, rapidly destroying
the vascular tissues and killing the tree (Hanks et al.
1
Corresponding author.
1990, Paine et al. 1995). Late-stage larvae burrow into
the sapwood where they construct a pupation chamber. Adults emerge from these pupal chambers several
weeks to many months later by chewing their way out
through the plug of frass and excelsior blocking the
entrance tunnel. The life cycle from egg to adult has
been reported to be 2 to 3 months in warm conditions
and as long as 9 months in cool conditions (Mendel
1985, Ali et al. 1986).
The congeneric species, Phoracantha recurva Newman, is similar to P. semipunctata in size, appearance,
and general biology. It is also widely distributed in its
native habitat, Australia, where it is not of great economic importance (Froggatt 1916, Lounsbury 1918,
Wang 1995). P. recurva has been introduced and established into many of the same countries as P. semipunctata, but overall, it does not seem to have become
as serious a pest as P. semipunctata. For example, in
Zambia, forest hygiene efforts initiated to reduce populations of P. semipunctata resulted in a more drastic
reduction of P. recurva (Loyttyniemi 1980). Rearing
experiments in Zambia using fallen eucalyptus conÞrmed that P. semipunctata was the dominant species
attacking the logs (Loyttyniemi 1983).
P. recurva was Þrst detected in southern California
in 1995 (Hanks et al. 1997), and in contrast to many
areas of the world, it has become associated with
0046-225X/04/0138Ð0146$04.00/0 䉷 2004 Entomological Society of America
April 2004
BYBEE ET AL.: EFFECTS OF TEMPERATURE ON Phoracantha
signiÞcant tree mortality. In areas of southern California where both species are present, P. recurva populations have dramatically increased and P. semipunctata populations have simultaneously decreased in
their shared habitat. For example, 1.4% of the beetles
emerging from naturally infested eucalyptus during
summer 1996 were P. recurva, whereas 74% of the
beetles emerging from naturally infested logs in 1997
were P. recurva (Hanks et al. 1997). The following
year, ⬎95% of the beetles emerging from naturally
infested logs were P. recurva, and that trend has continued to the present. Furthermore, the total number
of adult beetles emerging each year does not seem to
have changed dramatically. This indicates that the
host resources are continuing to be fully used by
Phoracantha beetles, but primarily by P. recurva rather
than P. semipunctata.
There are several possible reasons for this rapid
turnover of one species by another. First, lifetime
fecundity of P. recurva females may be greater than
that of P. semipunctata females. Second, P. recurva may
be reproductively active across a larger range of temperatures. For example, P. recurva adults have been
found earlier in the season in southern California than
P. semipunctata (L.F.B., unpublished data), suggesting
that the latter species may tolerate a broader range of
temperatures and consequently may have a longer
seasonal activity cycle. If this is the case, P. recurva may
be active and reproducing during cooler times of the
year, and capable of colonizing available host material
when adult P. semipunctata are not active.
In California, P. semipunctata females lay eggs in
batches of up to 40 eggs under loose bark (Hanks et al.
1993a, Paine et al. 1995). In summer, P. semipunctata
live for a month or more under Þeld conditions, with
adults feeding on nectar and pollen of eucalyptus
ßowers (Chararas 1969a,b; Scriven et al. 1986; Hanks
et al. 1990), and females laying up to 300 eggs (Scriven
et al. 1986). In a laboratory study, P. semipunctata
adults survived for up to 84 d on a diet of dilute malt
extract and eucalyptus ßowers, and readily laid eggs in
bark crevices in clusters of up to 50 eggs, which
hatched in 6 Ð15 d, depending on temperature (Ivory
1977). On the same diet, P. recurva adults survived for
up to 96 d, also laying eggs in bark crevices in clusters
of up to 40 eggs, and P. recurva eggs hatched more
quickly than P. semipunctata eggs (Ivory 1977). In a
subsequent study in California, diets of sucrose and
pollen resulted in signiÞcantly higher fecundity and
longevity for P. semipunctata in the laboratory compared with beetles given only distilled water (Hanks
et al. 1993b). Under laboratory conditions at 25⬚C, on
a diet of 30% sucrose solution and honey bee-collected
pollen the average lifetime fecundity of P. semipunctata was 180 eggs per female (Hanks et al. 1993b).
Previous studies have demonstrated that temperature affects the fecundity and longevity of P. semipunctata. In Israel, Mendel (1985) found egg masses of
P. semipunctata in very small numbers during the winter but readily found large numbers in the warm season. In Israel, adults were found to live ⬇40 d in
summer and up to 180 d in winter (Mendel 1985). In
139
related studies, P. semipunctata oviposition did not
occur at temperatures below 15⬚C (Avidov and Harpaz 1969), and ßight activity of adult P. semipunctata
beetles also was restricted to nights when the temperature was above ⬇15⬚C (Chararas 1969a, Loyttyniemi 1983).
The ecology and reproductive biologies of P. semipunctata and P. recurva have not been well studied in
their native range (Froggatt 1916), primarily because
the beetles rarely cause economic injury to the native
eucalyptus forests of Australia. However, in regions of
the world where they have become established, one or
both species have caused signiÞcant mortality to eucalyptus plantations and landscape trees. A sound
knowledge of their reproductive biologies may be
critical to understanding the temporal and spatial patterns of interspeciÞc interactions that have been observed in the novel sites. One of the critical factors in
determining the relative ability of each species to
exploit available resources may be their seasonal activity cycles. That is, if one species becomes reproductively active earlier at the beginning of the season,
and remains active longer at the end of the season, it
may gain an advantage over its congener by exploiting
their limited, shared host resource more effectively.
However, ßight activity and reproductive activity are
not necessarily synonymous. Therefore, our objective
was to determine the effect of temperatures typical of
late fall and early spring in southern California on egg
production, egg hatch, and adult longevity of P. semipunctata and P. recurva.
Materials and Methods
Adult beetles of both species were obtained from
established laboratory colonies that are continually
infused with Þeld-collected beetles. Voucher specimens are contained in the Entomology Research Museum of the Department of Entomology at the University of California in Riverside, CA. Freshly cut
eucalyptus trees were cut into logs ⬇0.5 m in length
for ease of handling. Logs were dried for 10 d at ⬇27⬚C,
the log ends were then waxed to slow further desiccation, and the logs were hand infested with P. recurva
and P. semipunctata neonate larvae. Larvae were introduced into each log by cutting a thin slit into the
bark, placing the requisite number of neonates in the
slit with a Þne paintbrush, covering the slit with paper,
and stapling the paper to the log. Logs were held in the
laboratory for 24 h at ⬇27⬚C after infestation to allow
the neonates to bore into them and to prevent the
neonate larvae from being dislodged during handling
of the logs. Logs were then stored in a greenhouse for
10 Ð12 wk at 32Ð35⬚C, followed by transfer to cold
room storage for 1Ð 6 mo at ⬇8⬚C until enough had
been infested to begin the experiment. Logs were then
moved from cold storage to incubators at ⬇32⬚C, and
adult beetles began to emerge 4 Ð 6 wk later. As adult
beetles emerged they were separated daily by species
and sex and held in cages for ⬍7 d until used in
experiments. Additional information regarding laboratory rearing can be found in Hanks et al. (1993b).
140
ENVIRONMENTAL ENTOMOLOGY
Male and female laboratory-reared P. semipunctata
and P. recurva ⬍ 7 d old were caged in pairs in cylinders (12 cm in height, 12 cm in diameter) made from
aluminum window screen with 14-cm-diameter plastic
petri dishes on the top and bottom. Whereas body
weight is positively correlated with the number of eggs
produced by the females in some Cerambycidae (Shibata 1987, Wang et al. 1998), this does not seem to be
the case with P. semipunctata (Hanks et al. 1996a), and
so various sizes of females were used. A 12.5-cmdiameter disc of P8 Þlter paper (Fisher, Pittsburgh,
PA) was placed on the bottom of each cage. A second
12.5-cm-diameter circle of P8 Þlter paper was wrapped
around an 8.5-cm-diameter petri dish that was placed
on top of the Þrst sheet of Þlter paper. Female beetles
preferred to oviposit between the two layers of Þlter
paper, and the eggs were harvested easily by cutting
out the paper discs with adhering eggs without damaging them. A 6 by 6-cm square of poster board hung
from the side of the cage with a paperclip was provided as a refuge for the beetles. Because adult beetles
naturally feed on nectar and pollen of eucalyptus
ßowers (Chararas 1969a,b; Scriven et al. 1986; Hanks
et al. 1990), each cage was provided with one 8-ml vial
of 5% sucrose solution plugged with a 4-cm long #2
cotton roll (Patterson Dental Supply, Inc., Saint Paul,
MN) and 0.2 g of eucalyptus pollen (Great Health,
Brea, CA). The sucrose solution and eucalyptus pollen
were replaced every 48 h when eggs from each cage
were collected and counted. All collecting and sampling was carried out in the morning during the early
part of the photophase, when adult beetles are quiescent.
Average daily temperatures in southern California,
taken from the California Irrigation Management Information System in Riverside (CIMIS Weather Collection 2003), were used to choose environmental
chamber temperatures. P. semipunctata and P. recurva
pairs were held in screen cages at four temperatures
(10⬚C, 15⬚C, 20⬚C, and 25⬚C; 11 replicates per treatment) in environmental chambers at ambient relative
humidity and lit with six ßuorescent (Sylvania
F24T12/CW/HO, 24-inch-long, cool-white high-output, 35-Watt) and two incandescent (ABCO 25T10,
25-Watt, tubular, clear) lights. Higher temperatures
representative of Riverside were not included because
the objective was to determine the effects of cooler
temperatures associated with late fall and early spring
on oviposition, egg viability, and adult longevity. A
photoperiod of 12:12 (L:D) h was used to provide
ample dark hours when the beetles are most active and
prevent the possibility of reproductive diapause. Each
of the four temperature treatments was conducted
with 11 replicates for each species. To ensure that
females were not sperm limited, males were replaced
every 3 wk with newly emerged males. Any males that
died at less than three weeks of age were replaced with
a newly emerged male, ⬍ 7 d old. The study continued
until all the females in each treatment had died.
The period of time before each female began laying
eggs was recorded as the preoviposition period, and
the time between when the last eggs were laid until
Vol. 33, no. 2
death was recorded as the postoviposition period.
During the oviposition period, the two layers of Þlter
paper between which females oviposited were removed every 48 h and examined for newly laid eggs.
After eggs were counted, they were returned to the
laboratory colony as part of the general rearing operation. Every 2 wk, two egg masses from each cage were
subsampled to determine egg viability. Subsamples
consisted of ⬇10 eggs removed from each mass and
placed in separate 50 by 9-mm petri dishes (Falcon,
Franklin Lakes, NJ) in a temperature chamber.
To test the effect of temperature on viability or
hatching success, a dish containing one of the groups
of eggs was returned to the original temperature conditions (oviposition temperature), whereas the other
was placed in a chamber with a selected standard
temperature (20⬚C). The subsamples were examined
for 4 wk or until all eggs hatched. After 4 wk, unhatched eggs were dried and shriveled, and clearly not
viable.
Statistical Analysis. We used analysis of variance
(ANOVA) to test for differences between treatment
means. The Fmax test (Zar 1999) was used to conÞrm
that data satisÞed the assumptions of ANOVA (sample
variances not signiÞcantly different). Where data
failed to meet ANOVA assumptions, we used the nonparametric KruskalÐWallis method of ANOVA by
ranks (Zar 1999) to test differences between means
(SigmaStat 1997). The TukeyÐKramer method was
used to identify signiÞcant differences among multiple
means (SigmaStat 1997). The inßuence of temperature on time to eclosion and egg viability was tested
using a General Linear Models procedure for nested
data with female as the replicate and subsampled egg
mass as the sample (SAS Institute 1996). Arcsin
square-root transformation was performed on all percent data (Zar 1999). To normalize P. semipunctata
longevity data and P. recurva preoviposition period,
square-root transformation was performed, but nontransformed data are presented in Þgures and table.
Results
Lifetime Fecundity. The replacement of one beetle
species with another in the new geographic range
could be a result of differences between species in
either total fecundity or in the ability of P. recurva to
successfully reproduce across a broader range of temperature than P. semipunctata. Females of both species
laid more eggs at 25⬚C than at 20⬚C (P. recurva F ⫽ 5.70;
df ⫽ 1, 20; P ⫽ 0.027; and P. semipunctata, F ⫽ 11.51;
df ⫽ 1, 20; P ⬍ 0.01; Fig. 1). Because only a single P.
recurva female and three P. semipunctata females laid
eggs at 15⬚C, and no eggs were laid by any female held
at 10⬚C, these temperatures were not included in the
statistical analyses. The single P. recurva that laid eggs
at 15⬚C laid a batch of 49 eggs and never laid eggs again.
The three P. semipunctata that laid eggs at 15⬚C averaged 57 eggs per female. Two of the three P. semipunctata females laid a single batch of eggs, whereas
the other female laid two egg masses. At 20 and 25⬚C,
beetles laid several batches of eggs throughout the
April 2004
BYBEE ET AL.: EFFECTS OF TEMPERATURE ON Phoracantha
141
Fig. 1. Lifetime fecundity of P. recurva and P. semipunctata (mean ⫾ SE) at four different temperature. At temperatures
indicated by the asterisk (*), only one P. recurva and three P. semipunctata laid eggs. Comparisons are made within species
and signiÞcant differences are indicated by different letters.
oviposition period. There were no differences in lifetime fecundity between P. recurva and P. semipunctata
at 25⬚C (F ⫽ 0.14; df ⫽ 1, 20; P ⫽ 0.71) or at 20⬚C (F ⫽
1.69; df ⫽ 1, 20; P ⫽ 0.21).
Egg Viability and Time to Eclosion. Although eggs
may be laid across a range of temperatures when the
beetles are active, differences in the ability of eggs of
each species to hatch and the length of time that it
takes eggs to hatch at temperatures characteristic of
early and late seasonal conditions could contribute to
the relative success of each species. Temperature had
a signiÞcant effect on egg viability and the rate of
embryonic development for both beetle species. P.
recurva eggs that were both oviposited and held at
25⬚C hatched sooner than eggs oviposited at 25⬚C and
held at 20⬚C (F ⫽ 8.01; df ⫽ 19, 67; P ⬍ 0.0001; Table
1), which in turn hatched sooner than eggs that were
both oviposited and held until hatch at 20⬚C (F ⫽ 6.95;
df ⫽ 16, 76; P ⬍ 0.0001). Egg viability decreased when
eggs were both oviposited and held at 20⬚C compared
with eggs oviposited at 25⬚C and held at 20⬚C (F ⫽ 5.40;
df ⫽ 17, 87; P ⬍ 0.0001) or oviposited and held at 25⬚C
(F ⫽ 6.06; df ⫽ 18, 86; P ⬍ 0.0001). There were no
differences in embryonic development rates of eggs
oviposited and held at 25⬚C and eggs oviposited at 25⬚C
and held at 20⬚C (F ⫽ 1.43; df ⫽ 20, 73; P ⫽ 0.14).
Similarly, P. semipunctata eggs that were oviposited
and held for development at 25⬚C hatched sooner than
eggs oviposited at 25⬚C and held at 20⬚C (F ⫽ 4.84; df ⫽
19, 45; P ⬍ 0.001), and eggs oviposited at 25⬚C and held
at 20⬚C hatched more quickly than eggs oviposited and
held at 20⬚C (F ⫽ 1.98; df ⫽ 15, 43; P ⫽ 0.041). Viability
of eggs that were both oviposited and held at 25⬚C was
not signiÞcantly different than that of eggs oviposited
at 25⬚C and held at 20⬚C (F ⫽ 1.78; df ⫽ 19, 48; P ⫽
Table 1.
0.054), and eggs both oviposited and held at 20⬚C (F ⫽
1.67; df ⫽ 15, 46; P ⫽ 0.092).
When chamber temperature and developmental
temperature were both 25⬚C, there was no signiÞcant
difference in the average number of days it took for P.
recurva and P. semipunctata eggs to hatch (F ⫽ 1.37;
df ⫽ 20, 57; P ⬍ 0.18), but hatch rates were signiÞcantly
higher for P. semipunctata than P. recurva (F ⫽ 1.78;
df ⫽ 20, 60; P ⫽ 0.045). For eggs oviposited at 25⬚C and
developed at 20⬚C, embryonic development of P. recurva eggs was faster than that of P. semipunctata eggs
(F ⫽ 4.22; df ⫽ 18, 55; P ⬍ 0.0001), but there was no
signiÞcant difference in the percentage of viable eggs
(F ⫽ 1.23; df ⫽ 19, 61; P ⫽ 0.26). When both oviposition and development temperatures were 20⬚C, P.
recurva embryonic development was faster than that
of P. semipunctata eggs (F ⫽ 4.47; df ⫽ 13, 64; P ⬍
0.0001), but P. semipunctata had signiÞcantly higher
egg viability than P. recurva at this temperature (F ⫽
6.99; df ⫽ 13, 72; P ⬍ 0.0001).
The egg mass from the one P. recurva female that
oviposited at 15⬚C was not tested for egg viability, but
two of the three egg masses deposited by P. semipunctata females at 15⬚C were subsampled. The P. semipunctata eggs that were both oviposited and held at
15⬚C had Þve of the 20 eggs hatch, 26 Ð28 d after
collection. For the P. semipunctata eggs that were laid
at 15⬚C and developed at 20⬚C, 10 of the 20 eggs
hatched, 10 Ð12 d after collection. These data were not
included in the statistical analysis because so few females laid eggs (Table 1).
Adult Longevity. As expected, temperature had a
signiÞcant effect on longevity of adult beetles of both
species. P. recurva lived signiÞcantly longer at 20⬚C
than at 10⬚C or 25⬚C (F ⫽ 14.20; df ⫽ 3, 39; P ⬍ 0.001,
Effect of temperature on days to hatch and percentage of viable eggs for P. recurva and P. semipunctata (mean ⴞ SE)
Oviposition
temperature
(⬚C)
Egg
developmental
temperature
(⬚C)
Phoracantha
species
No. of
replicates for
days to hatch
Days to
hatch
No. of
replicates for
% egg hatch
Mean %
egg hatch
15
15
20
20
25
25
25
25
15
20
20
20
20
20
25
25
P. semipunctata
P. semipunctata
P. recurva
P. semipunctata
P. recurva
P. semipunctata
P. recurva
P. semipunctata
2
2
51
27
42
32
45
33
26.4 ⫾ 0.39
10.4 ⫾ 0.26
10.3 ⫾ 0.24
11.7 ⫾ 0.33
8.5 ⫾ 0.26
10.4 ⫾ 0.29
5.5 ⫾ 0.2
6.5 ⫾ 0.34
2
2
58
28
47
34
47
34
0.25 ⫾ 0.05
0.5 ⫾ 0
62.3 ⫾ 4.3
69.5 ⫾ 5.7
72.8 ⫾ 4.4
65.6 ⫾ 4.3
73 ⫾ 3.9
76.6 ⫾ 4.7
142
ENVIRONMENTAL ENTOMOLOGY
Vol. 33, no. 2
Fig. 2. Adult longevity of P. recurva and P. semipunctata (mean ⫾ SE) at four different temperature. Comparisons are
made within species and signiÞcant differences are indicated by different letters.
Fig. 2), whereas longevity at 15⬚C was not different
from that at either 20⬚C or 25⬚C. P. semipunctata adults
lived signiÞcantly longer at 15⬚C and 20⬚C than at 10⬚C
and 25⬚C (KruskalÐWallis statistic ⫽ 18.83; P ⬍ 0.001,
Fig. 2). P. recurva lived signiÞcantly longer than P.
semipunctata at 25⬚C (KruskalÐWallis statistic ⫽ 6.42;
P ⫽ 0.01) and 20⬚C (KruskalÐWallis statistic ⫽ 8.18;
P ⫽ 0.004), whereas there was no difference between
adult longevity of the two species at 15⬚C (F ⫽ 0.37;
df ⫽ 1, 16; P ⫽ 0.55) or 10⬚C (F ⫽ 0.09; df ⫽ 1, 20; P ⫽
0.77).
Oviposition Periods. The amount of time the females of each species are reproductively active at
cooler temperatures could also contribute to a relative
reproductive advantage between the beetles. As temperature decreased from 25 to 20⬚C, P. recurva preoviposition period increased (F ⫽ 8.66; df ⫽ 1, 20; P ⫽
0.008; Fig. 3), but there were no signiÞcant differences
in the duration of P. recurva oviposition (F ⫽ 0.47; df ⫽
1, 20; P ⫽ 0.50) and postoviposition periods (F ⫽ 3.58;
df ⫽ 1, 20; P ⫽ 0.073) at the two temperatures. However, because the total longevity of P. recurva was
greater at 20⬚C than at 25⬚C, the number of days P.
recurva spent in one or more ovipositional stages
would have to be greater. To correct for greater total
longevity at 20⬚C, the number of days in any ovipositional stage was calculated as a percentage of the total
life span. By this measure, at 20⬚C the P. recurva preoviposition (KruskalÐWallis statistic ⫽ 3.89; P ⫽ 0.049)
and postoviposition periods (KruskalÐWallis statistic ⫽ 3.94; P ⫽ 0.047) were longer than at 25⬚C,
whereas there was no signiÞcant difference in the
percentage of total adult lifetime that P. recurva spent
in oviposition (KruskalÐWallis statistic ⫽ 3.504; P ⫽
0.061) at the two temperatures (Fig. 4).
In contrast, for P. semipunctata, there were no signiÞcant differences in the durations of preoviposition
(KruskalÐWallis statistic ⫽ 3.19; P ⫽ 0.074), oviposition (F ⫽ 2.49; df ⫽ 1, 20; P ⫽ 0.13), or postoviposition
periods (F ⫽ 3.58; df ⫽ 1, 20; P ⫽ 0.073) at 25⬚C and
20⬚C. However, when comparing these periods as a
percentage of total lifetime, P. semipunctata spent a
larger percentage of total lifetime in preoviposition
behavior at 20⬚C than at 25⬚C (F ⫽ 5.06; df ⫽ 1, 20; P ⫽
0.036). It also spent a smaller percentage of total lifetime in the oviposition period (KruskalÐWallis statistic ⫽ 6.074; P ⫽ 0.014), whereas there was no difference in the percentage of total lifetime spent in the
postoviposition period at 20⬚C versus 25⬚C (KruskalÐ
Wallis statistic ⫽ 0.68; P ⫽ 0.41).
At 20⬚C, preoviposition (F ⫽ 5.56; df ⫽ 1, 20; P ⫽
0.029) and oviposition periods (F ⫽ 7.28; df ⫽ 1, 20; P ⫽
0.014) were signiÞcantly longer for P. recurva than for
P. semipunctata, but there was no difference in duration of postoviposition periods (F ⫽ 0.71; df ⫽ 1, 20;
P ⫽ 0.41). The preoviposition period was also significantly longer for P. recurva at 25⬚ (F ⫽ 8.38; df ⫽ 1,
20; P ⫽ 0.009), but there were no differences between
species in the duration of the ovipositional (F ⫽ 1.80;
df ⫽ 1, 20; P ⫽ 0.20) and postovipositional periods (F ⫽
3.58; df ⫽ 1, 20; P ⫽ 0.073) at this temperature.
When comparing the ovipositional periods as a percentage of adult lifetime, the preoviposition (F ⫽ 0.67;
df ⫽ 1, 20; P ⫽ 0.42), oviposition (F ⫽ 0.77; df ⫽ 1, 20;
P ⫽ 0.39), and postoviposition periods (KruskalÐWallis statistic ⫽ 3.42, P ⫽ 0.064) for P. recurva and P.
Fig. 3. Average days spent in oviposition periods for P. recurva and P. semipunctata.
April 2004
BYBEE ET AL.: EFFECTS OF TEMPERATURE ON Phoracantha
143
Fig. 4. Average percentage of adult lifetime spent in oviposition periods for P. recurva and P. semipunctata.
semipunctata at 20⬚C were not signiÞcantly different.
Similarly, at 25⬚C, the fractions of the total adult lifetime spent in the preoviposition (KruskalÐWallis statistic ⫽ 1.73; P ⫽ 0.19) and oviposition periods were
not signiÞcantly different (KruskalÐWallis statistic ⫽
0.48; P ⫽ 0.49), but the postoviposition period was
signiÞcantly longer for P. semipunctata (KruskalÐWallis statistic ⫽ 8.99; P ⫽ 0.003).
Discussion
Although Þeld observations had shown that P. recurva is active earlier and for a longer period of the
year in California than P. semipunctata (L.F.B., unpublished data), there do not seem to be any substantial differences in their reproductive biologies at any
of the temperatures used in this study. Consequently,
it would not seem that P. recurva has a signiÞcant
reproductive advantage relative to its congener and
that the change in the relative proportions of the
species in the community is probably not due to differences in fecundity. Both P. recurva and P. semipunctata laid more eggs at 25⬚C than at 20⬚C, and the
lifetime fecundities of the two species at each temperature were similar. Furthermore, P. recurva did not
oviposit over a larger range of temperatures than P.
semipunctata, suggesting that even though P. recurva
may be active during the cooler times of the year when
P. semipunctata is not ßying, the females are probably
not ovipositing on available host resources. One P.
recurva and three P. semipunctata oviposited eggs at
15⬚C and no eggs were laid by either species at 10⬚C.
These data are consistent with previously reported
temperature thresholds for P. semipunctata oviposition, indicating that oviposition ceased below ⬇15⬚C
(Avidov and Harpaz 1969). Lifetime fecundity of P.
semipunctata at 25⬚C was greater than reported previously (Scriven et al. 1986), but was lower at 20⬚C
than described in these earlier reports. This latter
result is possibly due to differences in adult diets
between the studies.
The speed of embryonic development and the viability of P. recurva eggs increased with temperature.
Although P. semipunctata eggs also developed more
quickly as temperature increased, viability was not
greater at the higher temperatures. For both species,
eggs completed development within 4 Ð20 d, with P.
recurva eggs developing more quickly in warmer conditions than P. semipunctata eggs.
P. recurva adults lived longer than P. semipunctata at
both 25⬚C and 20⬚C. This difference could be biologically important, because these temperatures were
also the only temperatures at which signiÞcant oviposition occurred. There were no differences in longevity of P. recurva and P. semipunctata at 15⬚C and
10⬚C.
The only possible biological advantages in phenological development for P. recurva found in this study
were faster development of eggs and increased adult
longevity at reproductive temperatures compared
with P. semipunctata. It seems unlikely that these relatively small differences could account for the rapid
replacement of P. semipunctata by P. recurva under
southern California conditions. Although P. recurva is
active for a longer period in the year than P. semipunctata, our results suggest that this may not provide
a reproductive advantage in phenological development, because under the cooler conditions prevailing
at the beginning and end of the growing season, oviposition by both species is minimal. However, P. recurva might be better able to exploit intermittent periods of warmer temperatures during the winter,
laying eggs that hatch relatively quickly. Consequently, early and late-season adult activity may provide a selective advantage during short periods of
favorable conditions throughout the cooler months.
Furthermore, early colonization confers a deÞnite
advantage to developing larvae (Paine et al. 2001).
The larvae of both beetle species develop in the cambium during the same time of the year, so interspeciÞc
competition is inevitable. Heavy infestations of P.
semipunctata larvae result in the destruction of virtually the entire cambium layer of host trees, suggesting
that the cambium is a limited resource (Chararas
1969a,b; Drinkwater 1975, Scriven et al. 1986). IntraspeciÞc competition in P. semipunctata has been
observed from overcrowding in the larval stage, with
clear evidence of cannibalism when feeding channels
144
ENVIRONMENTAL ENTOMOLOGY
intersected (Powell 1982). Thus, older, established
larvae have a competitive advantage over younger,
later developing larvae in both inter- and intraspeciÞc
encounters (Paine et al. 2001).
Overall, the relatively minor differences in reproductive biology reported here do not seem to provide
a satisfactory explanation for the dramatic increase in
P. recurva populations and the simultaneous decrease
in P. semipunctata populations in California. The interactions between the two species and a selective egg
parasitoid may provide an alternative explanation. The
success of an invasive herbivore in colonizing a new
environment in the presence of potential competitors
may be a function of the action of shared natural
enemies (Lawton 1986). Beginning in 1993, the egg
parasitoid Avetianella longoi (Siscaro) (Hymenoptera:
Encyrtidae) was released for control of P. semipunctata, the only Phoracantha species in California at the
time. A. longoi quickly became established in southern
and central California, resulting in high rates of parasitism of naturally laid P. semipunctata egg masses
(Hanks et al. 1996c, Paine et al. 1997). A braconid
larval parasitoid, Syngaster lepidus (Brullé), also has
been released intermittently since 1993, but there is no
conclusive evidence of its permanent establishment
(Paine et al. 1997). It is clear from examination of Þeld
collected egg masses that A. longoi prefers P. semipunctata over P. recurva as a host, and laboratory
studies have demonstrated that P. semipunctata eggs
are also more suitable hosts for successful parasitoid
development (Luhring et al. 2000). If the populations
of P. semipunctata are reduced by parasitism to a
greater extent than those of P. recurva, then the host
tree resources would be available for exploitation by
P. recurva. Thus, differential parasitism may be a signiÞcant and possibly the major factor contributing to
the rapid increase in P. recurva populations relative to
those of P. semipunctata.
Although the subject of much speculation in the
literature, well documented examples of parasitoidmediated species replacement are rare (Denno et al.
1995). Settle and Wilson (1990) were able to show that
differential parasitism by an egg parasitoid virtually
eliminated a native grape leafhopper from California
vineyards, allowing an invading leafhopper to become
the dominant species. The presence of one herbivore
may increase the abundance of a shared natural enemy
and result in a signiÞcant reduction in the population
size of a second herbivore through apparent competition (Holt 1977, 1984; Settle and Wilson 1990; Karban
et al., 1994; Hamback and Bjorkman 2002; Pope et al.,
2002), particularly if there is an amensal relationship
between competing herbivores (Bonsall and Hassell
1997). However, apparent competition does not seem
to be a critical factor in maintaining the relationships
between the Phoracantha species in California. The
parasitoid is only marginally successful in P. recurva
and the preference for P. semipuctata remains strong
in both the Þeld and in culture despite intense selection in laboratory colonies (Luhring et al. 2000),
which functionally limits the extent to which the as-
Vol. 33, no. 2
sociation of these herbivores can be characterized as
sharing a natural enemy (Luhring et al. in press).
Differences in the reproductive biology of the two
beetle species do not explain the changes in their
relative densities. Introduction and establishment of
an egg parasitoid that prefers P. semipunctata to P.
recurva provides an alternative explanation. However,
it is also possible that there are differences in the
competitive ability between the two species in securing larval food resources. Denno et al. (1995) predicted that insects with life history traits like these
beetles that are closely related, introduced, aggregative, and sessile species feeding on discrete resources
would be subject to signiÞcant interference competition. We are completing the appropriate controlled
competition experiments to test this prediction.
The interactions among invasive species in novel
environments are critical to understand within the
context of the interactions in their native range. The
relative importance of the factors that regulate populations may be very different in the different geographic locations. To date, there is no further detailed
information on oviposition by natural populations of P.
semipunctata and P. recurva in native habitats or in
other parts of the world where both species have been
introduced. Similarly, information on changes in relative population sizes after recent reassociations of
both beetle species and selected natural enemies in
novel locations (e.g., Mediterranean Europe or North
Africa) is not yet available. This information, and additional information on other aspects of P. recurva and
P. semipunctata life cycle and biology, may be necessary before the critical factor or combination of factors
responsible for the rapid rise in densities of P. recurva
populations and the concomitant decline in populations of P. semipunctata in California and elsewhere
can be determined.
Acknowledgments
We thank Jose Arredondo for assistance with this study.
We gratefully acknowledge Þnancial support from USDAÐ
CSREES award #99-35302Ð 8188.
References Cited
Ali, A. D., J. S. Hartin, and T. D. Paine. 1986. The eucalyptus
longhorned borer. Weeds, Trees Turf. 12: 44 Ð 45.
Avidov, Z., and I. Harpaz. 1969. Plant pests of Israel. Israel
University Press, Jerusalem. Israel.
Bonsall, M. B., and M. P. Hassell. 1997. Apparent competition
structures ecological assemblages. Nature 388:371Ð373.
Browne, F. G. 1968. Pests and diseases of forest plantation
trees. Clarendon Press, Oxford, UK.
Bytinski-Salz, H., and S. Newmark. 1952. The eucalyptus
borer (Phoracantha semipunctata) in Israel. Trans. IX Int.
Congr. Entomol., Amsterdam 1951. 1: 696 Ð 699.
Chararas, C. 1969a. Biologie et ecologie de Phoracantha
semipunctata F. (Coleoptere Cerambycidae xylophage)
ravageur des Eucalyptus en Tunisie et methods de protection des peuplements. Ann. Inst. Nat. Rech. For. Tunie
2: 1Ð37.
Chararas, C. 1969b. Etude biologique de Phoracantha
semip. F. (Coleoptere: Cerambycidae xylophage)
April 2004
BYBEE ET AL.: EFFECTS OF TEMPERATURE ON Phoracantha
speciÞque des Eucalyptus en Tunisie et recherches sur
la vitalite et lÕadaptation de ces essences. Acad.
DÕAgriculture de France, C. R. 55: 47Ð57.
CIMIS Weather Collection. 2003. California Irrigation
Management Information System, California Department of Water Resources. Sacramento, CA. www.cimis.
water.ca.gov
Denno, R. F., M. S. McClure, and J. R. Ott. 1995. InterspeciÞc interactions in phytophagous insects: competition
reexamined and resurrected. Annu. Rev. Entomol. 40:
297Ð331.
Drinkwater, T. W. 1975. The present pest status of eucalyptus borers Phoracantha spp. in South Africa, pp. 119 Ð
129. In Proc. 1st Congr. Entomol. Soc. S. Africa 1975.
Entomological Society of Southern Africa, Pretoria,
South Africa.
Froggatt, W. W. 1916. Forest longicorn beetles and their
parasites. Agric. Gaz. New South Wales, Sydney, xxvii. 8:
561Ð567.
Gonzalez-Tirado, L. 1986. Phoracantha semipunctata F.:
damages caused in the province of Huelva during 1983
and 1984. Bol. San. Veg. Plagas 12: 147Ð162.
Hamback, P. A., and C. Bjorkman. 2002. Estimating the consequences of apparent competition: a method for hostparasitoid interactions. Ecology 83: 1591Ð1596.
Hanks, L. M., J. G. Millar, and T. D. Paine. 1990. Biology and
ecology of the eucalyptus longhorned borer (Phoracantha
semipunctata F.) in southern California, pp. 12Ð16. In D.
Adams and J. Rios [eds.], Proceedings, 39th Meeting of
the California Forest Pest Council, 14 Ð15 November
1990. California Department of Forest and Fire Protection, Sacramento, CA.
Hanks, L. M., T. D. Paine, and J. G. Millar. 1993a. Host
species preference and larval performance in the woodboring beetle Phoracantha semipunctata F. Oecologia
(Berl.) 95: 22Ð29.
Hanks, L. M., McElfresh, J. S., J. G. Millar, and T. D. Paine.
1993b. Phoracantha semipunctata (Coleoptera: Cerambycidae), a serious pest of eucalyptus in California: biology and laboratory-rearing procedures. Ann. Entomol.
Soc. Am. 86: 96 Ð102.
Hanks, L. M., J. G. Millar, and T. D. Paine. 1996a. Body size
inßuences mating success of the eucalyptus longhorned
borer (Coleoptera: Cerambycidae). J. Insect Behav. 9:
369 Ð382.
Hanks, L. M., J. G. Millar, and T. D. Paine. 1996b. Mating
behavior of the eucalyptus longhorned borer (Coleoptera: Cerambycidae) and the adaptive signiÞcance of
long “horns.” J. Insect Behav. 9: 383Ð393.
Hanks, L. M., T. D. Paine, and J. G. Millar. 1996c. Tiny wasp
helps protect eucalyptus from eucalyptus longhorned
borer. Calif. Agric. 50: 14 Ð16.
Hanks, L. M., Paine, T. D., J. G. Millar, and C. Campbell.
1997. Another tree killing pest of eucalyptus invades California. Calif. Plant Pest Dis. Rep. 16: 19 Ð21.
Holt, R. D. 1977. Predation, apparent competition and the
structure of prey communities. Theor. Popul. Biol. 12:
197Ð229.
Holt, R. D. 1984. Spatial heterogeneity, indirect interactions, and the coexistence of prey species. Am. Nat. 124:
377Ð 406.
Ivory, M. H. 1977. Preliminary investigations of the pests of
exotic forest trees in Zambia. Commonw. For. Rev. 56:
47Ð56.
Karban, R., D. Hougen-Eitzmann, and G. English-Loeb.
1994. Predator-mediated apparent competition between
two herbivores that feed on grapevines. Oecologia (Berl.)
97: 508 Ð511.
145
Lawton, J. H. 1986. The effect of parasitoids on phytophagous insect communities, pp. 265Ð287. In J. Waage and D.
Greathead [eds.], Insect parasitoids. Academic, New
York.
Lounsbury, C. P. 1918. The Phoracantha beetle. A borer pest
of eucalyptus trees. Union S. Africa Dep. Agric., Div.
Entomol. (sine loco), Local Series 24: 10.
Loyttyniemi, K. 1980. Life history and harmfulness of the
eucalyptus longhorn beetle, Phoracantha semipunctata
(Fabricius) in Zambia. Proceedings, XI Commonwealth
Forestry Conference, Trinidad, 1980. Forestry Department, Ndota, Zambia.
Loyttyniemi, K. 1983. Flight pattern and voltinism of Phoracantha beetles (Coleoptera, Cerambycidae) in a semihumid tropical climate in Zambia. Ann. Entomol. Fenn.
49: 49 Ð53.
Luhring, K. A., T. D. Paine, J. G. Millar, and L. M. Hanks.
2000. Suitability of the eggs of two species of eucalyptus
longhorned borers (Phoracantha recurva and P. semipunctata) as hosts for the parasitoid Avetianella longoi Siscaro.
Biol. Control 19: 95Ð104.
Luhring, K. A., J. G. Millar, T. D. Paine, D. Reed, and H.
Christiansen. Oviposition preferences and progeny development of the egg parasitoid Avetianella longoi: factors
mediating replacement of one species by a congener in a
shared habitat. Biol. Control (in press).
Mendel, Z. 1985. Seasonal development of the eucalyptus
borer Phoracantha semipunctata, in Israel. Phyoparasitica
13: 85Ð93.
Moore, K. M. 1963. Observation on some Australian forest
insects. 15. Some mortality factors of Phoracantha semipunctata F. (Coleoptera: Cerambycidae). Proc. Linn. Soc.
New South Wales 88: 221Ð229.
Paine, T. D., J. G. Millar, and L. M. Hanks. 1995. Integrated
program protects trees from eucalyptus longhorned
borer. Calif. Agric. 49: 34 Ð37.
Paine, T. D., J. G. Millar, T. S. Bellows Jr., and L. M. Hanks.
1997. Enlisting an underappreciated clientele: public
participation in distribution and evaluation of natural
enemies in urban landscapes. Am. Entomol. 43: 163Ð168.
Paine, T. D., J. G. Millar, E. O. Paine, and L. M. Hanks. 2001.
Inßuence of host log age and refuge from natural enemies
on colonization and survival of Phoracantha semipunctata.
Entomol. Exp. Appl. 98: 157Ð163.
Pope, T., E. Croxson, J. K. Pell, H.C.J. Godfray, and C. B.
Muller. 2002. Apparent competition between two species of aphid via the fungal pathogen Erynia neoaphidis
and its interaction with the aphid parasitoid Aphidius ervi.
Ecol. Entomol. 27: 196 Ð203.
Powell, W. 1982. Age-speciÞc life table for the eucalyptus
boring beetle Phoracantha semipunctata (F.) (Coleoptera: Cerambycidae) in Malawi. Bull. Entomol. Res.
72: 645Ð 653.
SAS Institute. 1996. SAS userÕs guide for statistics, release
6.13 ed. SAS Institute, Cary, NC.
Scriven, G. T., E. L. Reeves, and R. F. Luck. 1986. Beetle
from Australia threatens eucalyptus. Calif. Agric. 40: 4 Ð 6.
Settle, W. H., and L. T. Wilson. 1990. Invasion by the variegated leafhopper and biotic interactions: parasitism,
competition, and apparent competition. Ecology 71:
1461Ð1470.
SigmaStat. 1997. SigmaStat statistical software userÕs manual, release 2.03 ed. SPSS Science, Chicago, IL.
Shibata, E. 1987. Oviposition schedules, survivorship
curves, and mortality factors within trees of two cerambycid beetles (Coleoptera: Cerambycidae), the japanese
146
ENVIRONMENTAL ENTOMOLOGY
pine sawyer, Monochamus alternatus Hope, and sugi bark
borer Semanotus japonicus Lacordaire. Res. Popul. Ecol.
29: 347Ð367.
Tooke, F.G.C. 1935. Insects injurious to forest and shade
trees. Bull. Dep. Agric. For. South Afr. Pretoria 142: 33Ð39.
Wang, Q. E. 1995. A taxonomic revision of the Australian
Genus Phoracantha Newman (Coleoptera: Cerambycidae). Invertebr. Taxon. 9: 865Ð958.
Wang, Q. E., G. Shi, and L. K. Davis. 1998. Reproductive
Vol. 33, no. 2
potential and daily reproductive rhythms of Oemona hirta
(Coleoptera: Cerambycidae). J. Econ. Entomol. 91: 1360 Ð
1365.
Zar, J. H. 1999. Biostatistical analysis, 4th ed. Prentice Hall,
Upper Saddle River, NJ.
Received for publication 9 July 2003; accepted 11 October
2003.