| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Behavioral Ecology Vol. 10 No. 5: 552-556
© 1999 International Society for Behavioral Ecology
Influence of resource level on maternal investment in a leaf-cutter bee (Hymenoptera: Megachilidae)
Department of Entomology, University of California, Davis, CA 95616, USA
Address correspondence to J-y. Kim. E-mail: jkim{at}ss.nises.affrc.go.jp .
Received 20 May 1998; revised 11 February 1999; accepted 11 February 1999.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Fisher's (1930
) prediction
of equal investment for each sex in a panmictic population is influenced by a
number of ecological factors, among which resource availability plays a major
role, particularly when the population exists under changing resource
availability. Rosenheim et al. proposed a multifaceted parental investment
model based on the underlying assumption that individual females determine
their sex investment according to resource availability and oocyte
availability to maximize reproductive success. The model predicts that greater
availability of resources used for provisions will lead to (1) an increase in
the proportion of females produced (when the female is the larger sex) and (2)
an increase in the amount of provisions per offspring and thus an increase in
offspring size. I tested these predictions by a controlled experiment using a
leaf-cutter bee, Megachile apicalis. I presented two levels of food
resources to the nesting females, which were allowed to forage and nest in
cages. The experimental results supported these parental investment model's
predictions. Key words: Hymenoptera, investment ratios, leaf-cutter bee, Megachile apicalis, Megachilidae, parental investment, resource levels, sex allocation.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Fisher's (1930
) theory of
parental investment and sex allocation predicts equal investment for each sex
in a panmictic population. In other words, if the production of a daughter
requires twice the amount of parental resources as the production of a son,
the number of sons is predicted to be twice the number of daughters in the
population (Charnov, 1982;
Fisher, 1930). Aculeate
hymenopteran insects such as bees, wasps, and ants not only have maternal
control of sex allocation, but provide all food resources to their developing
offspring and therefore provide an ideal system to test Fisher's theory.
Studies of sex-investment patterns have revealed a number of factors that
potentially cause a deviation from Fisher's prediction, including conflicts
within eusocial colonies (Alexander and
Sherman, 1977; Noonan,
1978; Trivers and Hare,
1976); conflicts between mates
(Brockmann and Grafen, 1989);
local resource enhancement (Schwarz,
1988; Stark,
1992); local resource competition
(Visscher and Danforth, 1993);
local mate competition (Cowan,
1991); partially overlapping generations
(Brockmann and Grafen, 1992;
Tepedino and Parker, 1988);
and maternal size effect (Sugiura and
Maeta, 1989). In addition to these factors, resource availability
has been suggested as another important factor that would explain deviations
from Fisher's prediction. Tepedino and Torchio
(1982) reported increased
proportions of female offspring from maternal bees that were provided with
super-abundant resources throughout the season, given that the female is the
sex with the larger body size. Since this landmark experiment, surprisingly
little experimental work has been carried out to study this phenomenon in
bees. This current paper reports sex allocation data for bees studied under an
experimental manipulation of resource levels by using a caged leaf-cutter bee,
Megachile apicalis.
Recently, a few authors have proposed theories to explain a shift in sex
allocation in response to changing resource availability
(Frank, 1995;
Rosenheim et al., 1996).
Rosenheim et al. (1996)
proposed a "multifaceted parental investment" model based on the
assumption that individual females determine their sex investment according to
resource availability and oocyte availability to maximize reproductive
success. Under this model, greater availability of resources used for
provisions will lead to (1) an increase in the proportion of the larger sex
produced and (2) an increase in the amount of provisions per offspring and
thus an increase in offspring size. In this paper, I attempt to analyze the
experimental data in light of these predictions.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Sources of bees
Megachile apicalis Spinola is an adventive leaf-cutter bee introduced into the western United States from Europe (Barthell and Thorp, 1995
;
Cooper, 1984;
Stephen, 1987;
Thorp et al., 1992). Like most
other Megachile, females of this species construct linear nests of
leaf-covered brood cells in natural cavities such as tree holes and hollow
twigs. However, in this study, I successfully reared individual females in
small cages using artificial nesting tunnels. The female is the sex with the
larger body size in this species. Approximately 50 cocoons of M. apicalis broods obtained in Davis, California, USA, were kept under outdoor conditions shortly before winter 1994. In the following spring, about 2 weeks before emergence, the cocoons were moved into a laboratory room (22-25°C; 46-73% relative humidity). Groups of 10-20 emerging adults were kept in mesh-walled cages (30 cm x 20 cm x 15 cm; mesh size: 1.3 mm) provided with 20-30 freshly cut flower heads of yellow star thistle, Centaurea solstitialis L., and a feeder containing 50% sucrose solution. The relatively low room temperature kept the activity of the adult bees minimal except for feeding, thus avoiding mortality due to random flight and collisions within the cages. The caged bees were placed outdoors once a day for 10-20 min, during which they fed on the cut flower heads and sucrose solution.
Obtaining mated bees
Females were mated with males of different maternal lines using two
methods. First, I introduced virgin females singly into mesh-walled cages
holding males. This resulted in 2 copulations out of more than 20 cumulative
mating attempts among 10 individuals of both sexes. Two females from these
copulations were used in the experiment. These two females produced only sons,
indicating that there was no insemination. Thus, data from these two females
were not included in the analysis of sex-investment ratio. However, I used
these data in calculations of brood cell completion rate because their nesting
behavior appeared to be normal.
Second, virgin females were introduced singly into a nest cage (described later) holding four to seven wild-caught males. All females were successfully inseminated with this method, as they all produced both sons and daughters.
Nest cages
Seven cylindrical cages (height: 50 cm; diam 50 cm) were constructed with
iron fence frames and nylon nets (mesh size: 1 mm). To standardize the amount
of nest space provided for each tested female, I used artificial nest tunnels
of a uniform size. For each cage, I provided a stick made of pine wood (15 cm
x 2 cm x 2 cm), into which a longitudinal tunnel 14 cm deep was
drilled. A paper straw (SweetheartTM, length: 14 cm; internal diameter:
6.4 mm) was inserted to the tunnel. To provide leaves for the nesting female,
three to five young shoots of Wisteria sp., each 10-20 cm long with
about 5-15 leaflets were placed in the cage in a water bottle. The lengths of
the median veins of the leaflets were about 1.5-5 cm.
To provide pollen and nectar to the female, freshly cut flower heads of yellow star thistle were collected near the University of California at Davis campus and provided in water bottles. Only flower heads in which < 50% of the pins (the pistil ensheathed with a pollen sac) had exuded pollen were harvested. Flower stalks were cut at 10 cm below the heads and the cut ends immediately placed in water. Thirty flower heads were bundled to make a flower bouquet. I arranged the flower heads to form a convex platform surface of about 10 cm in diameter. Thorns were cut off with a pair of scissors because they disrupt the florets and feeding by the females. The bouquet was kept fresh in a water bottle. Flowers were found to exude pollen continuously for at least 2 days after harvesting. Each of six cages received one mated bee, one nest stick, one bottle of Wisteria leaf shoots, and two or four bouquets of yellow star thistle depending on the experimental resource level treatment (see below). The cages were kept in a glass-house that was kept at 28-31°C during the day and 19-25°C during the night (65-90% relative humidity).
There were two treatments: low and high resource levels. The low resource level had one new bouquet and one 2-day-old bouquet; the high resource level had two new bouquets and two 2-day-old bouquets. Every 2 days new bouquets were introduced to the cages, and the oldest bouquets were discarded to maintain a relatively constant level of pollen availability. Five drops (about 0.3 ml in total) of 50% sucrose solution were placed daily on all new bouquets (one bouquet in the low resource level and two in the high resource level).
Each individual females provided with either the high or low resource level during the time required to complete an entire nest. When the females plugged the first nest, the resource level was either changed from low to high, from high to low, or not changed. The two treatments were balanced, so that equal numbers of females would start nesting under either treatment. However, there were two instances where females completed a nest and initiated another nest before the resource level was changed. I allowed these females to continue nesting under the same resource level. Due to limited time and cage availability, females completing five nests were removed from the cage unless they voluntarily terminated nesting. There was only one 5th nest with 13 brood cells made under the high resource level, which was pooled with the 4th nest. At the end of the experiment, I measured the head width of each female to the nearest 0.1 mm using an ocular micrometer.
I removed paper straws containing brood cells ("cells" hereafter) from each nest within 2 days of the nest's completion (i.e., females capping a tunnel entrance). The cells (leaves plus provisions) were weighed to the nearest 0.1 mg using a Mettler balance and incubated in a petri dish at 30°C for 12 days and 22°C for 12 days (55-90% relative humidty). Non-overwintering offspring emerged without undergoing chilling and were sexed upon emergence. Overwintering cocoons were chilled at 6°C for 4 weeks to terminate hibernation and incubated at 30°C for 12 days and 22°C for 12-20 days (35-65% relative humidity) to obtain emerging adults. Of the total of 426 cells collected from the caged females, there were 49 dead cells, which I excluded from the analysis for sex-investment ratios.
Calculations of cell completion and relative investment in
daughters
I calculated cell completion rates by dividing the total weight of all
cells for each nest by total number of days spent producing these cells. If a
cell was started or finished in the morning, that day was included into or
discounted from the total number of days, respectively. If a cell was left
incomplete but finished in the following morning, this second day was
discounted. Therefore, measurement of days for cell completion had an error of
less than 1 day.
Three related measures of relative investment in daughters were applied for each of 10 females that produced progeny of both sexes. First, I calculated the investment in daughters based on cell weight by dividing the total weight of daughter cells by the total weight of all cells. Second, I calculated the investment in daughters based on adult progeny weight by dividing the total adult weight of daughters by the total adult weight of all progeny. Third, numerical proportion of daughters was obtained as the number of daughters divided by the total number of progeny. In addition, I calculated average cell weight and adult progeny weight for each female and compared these between high and low resource levels.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Nesting behavior and resource levels
The first female started nesting in a cage on July 11, and the last female finished nesting on October 19 in 1995. Each female took 7-15 days to adjust to the cage environment and begin nesting behavior, and all 12 females tested in the experiment successfully constructed 2 or more nests. Ten females completed four nests, and the other two completed two and five nests, respectively. Nests contained an average of 11.1±3.0 (mean ±SD, n = 47) cells. The number of cells per nest did not differ between the high and low resource level treatments (Wilcoxon signed-rank test: z = 0.4, p >.5). There was a significant decrease in the number of cells per nest in later nests produced by individual females, but the resource treatments had no effect on the number of cells (two-way ANOVA of number of cells for each nest, for nest sequence number: F = 9.74, df = 3, p <.0001; for resource level: F = 0.09, df = 1, p >.7).
The two resource treatments successfully created environments of differing resource availability. The average cell completion rate was significantly higher in the high resource level treatment (Wilcoxon signed-rank test: z = 2.8, n = 12, p <.005). The average rates over all females under the high resource level was on average 1.3 times greater than those of the same 12 females tested under the low resource level (Figure 1A). Similarly, the difference in cell weight/day between the treatments analyzed for each individual female was also significant (Wilcoxon signed-rank test: z = 2.8, n = 12, p <.005). The average cell weight/day over all females under the high resource level was again on average 1.3 times greater than those of the same 12 females tested under the low resource (Figure 1B). For the females that were in the process of rapid cell completion, it was often observed that the cells were completely provisioned but eggs were not laid. In more than three occasions, the females were found feeding or simply resting on or near the entrance of the nest for fairly lengthy periods without provisioning or laying an egg in the fully provisioned cell. These informal observations suggest that the rate of cell completion may have outstripped the rate of egg maturation in some cases.
|
Effects of resource levels on investment in daughters
Investment in daughters was greater under the high resource level treatment
than the low resource level treatment. The three related measures of
investment in daughters supported this. The investment based on cell weight
was significantly higher under the high resource level than under the low
resource level (Wilcoxon signed-rank test: z = 2.29, p
=.022; Figure 2A). The
investment based on adult weight was also higher under the high resource level
than under the low resource level (Wilcoxon signed-rank test: z =
2.29, p =.022; Figure
2B). The numerical proportion of daughters was also higher under
the high resource level (Wilcoxon signed-rank test: z = 2.32,
p =.020, Figure 2C).
Three of the 10 females consistently produced nests entirely composed of sons
under the low resource level.
|
Effects of resource levels on investment in individual cells and
adult progeny
Average cell weight was greater for daughters produced under the high
resource level than those under the low resource level, and the subtle
difference was significant under a pairwise test (Wilcoxon signed-rank test:
z = 2.12, n = 7, p =.0343;
Figure 3A). However, the
difference was marginally nonsignificant for sons (z = 1.78,
n = 10, p >.07; Figure
3A). Similarly, average weight of adult daughters produced under
the high resource level was also greater (Wilcoxon signed-rank test:
z = 2.38, n = 7, p <.0176;
Figure 3B). There was no
significant difference for adult sons (z = 0.255, p >.7;
Figure 3B).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
The experiment showed that a greater availability of resources used for provisions increased maternal investment in daughters and increased total investment per offspring (daughters). These results agree with the predictions of the multifaceted parental investment model (Rosenheim et al., 1996
).
Sex-investment ratio and resource levels
There is apparently a causal relationship between sex-investment ratio and
resources. When parental investment is measured in terms of the amount of food
provided to each offspring, a greater proportion of the less expensive sex
(male in these studies) is expected to be produced
(Frohlich and Tepedino, 1986;
Maeta, 1978;
Sugiura and Maeta, 1989;
Tepedino and Torchio, 1982;
Visscher and Danforth, 1993).
The sex-investment ratio was measured using three variables: total cell
weight, total adult progeny weight, and number of progeny, and all supported
the prediction that females invest more in daughters when resources for
provisioning are more abundant (Figure
2).
Although the current data are congruent with the predictions of the dynamic
model (Rosenheim et al.,
1996), it should be emphasized that a key assumption of the
model—namely, that egg availability may limit female
reproduction—cannot be evaluated, because no egg production data are
available in this study. According to the model by Rosenheim et al.
(1996), under a high food
resource level, oocyte development will be relatively more limiting than food
used in offspring provisioning. This suggests that the cost of foraging and
provisioning relative to egg production decreases under the high resource
level. Thus, the investments for males and females will become more nearly
equal, and therefore the investment ratio approaches 1:1, given that there is
no significant difference in size between male and female eggs
(Kim, 1996). During the
current study, it was observed that individual females in the process of
provisioning with obviously rapid rates under the high resource level were
either feeding or resting, but not ovipositing in a fully provisioned cell.
Although this observation does not constitute definitive evidence of egg
limitation, it does suggest that the relative speed of egg maturation may be
lagging behind that of provisioning.
The data also match the predictions of the nonlinear fitness returns model
(Frank, 1995), which predicts
increased investment in female offspring under an increased resource level.
According to this model, the female/male allocation increases by shifting the
switching point from female to male production as the resource level
increases. This model also predicts that there would be a point at which no
female is produced, which in fact was observed in the current experiment where
three females produced all-male nests under the low resource level
(Frank, 1995).
Field data have demonstrated that sex-investment ratio is correlated with
seasonal changes in floral resource levels. Megachile apicalis
females of the spring-emerging generation produce more daughters as
flower-head counts of yellow star thistle increase in the field
(Kim, 1996). The current
experimental results suggest that a causal relationship underlies the temporal
correlation between sex ratio and resource availability observed in the
field.
Investment per offspring and resource levels
Current data indicate that the mean quantity of food (estimated by average
cell weight) changed in response to changing resource levels
(Figure 3A). Torchio and
Tepedino (1980) observed the
same response in the field for the bee Osmia lignaria propinqua,
which allocated less provisions for offspring under conditions of decreasing
floral resources. Although similar observations were made on other univoltine
bees under greenhouse conditions (Sugiura
and Maeta, 1989; Tepedino and
Torchio, 1982), the current experiment is more controlled because
it provides set amounts of food resources to individual females separately
confined in a standardized environment.
A careful analysis of both cell and adult weight produced under the high
and low resource levels revealed that there was a slight difference between
cells for daughters and sons (Figure 3
A, B). Females gave
significantly more provisions to cells of daughters under the high resource
level, but there was a marginally nonsignificant difference for cells of sons.
Why does investment per offspring not decrease proportionately for both sexes?
A potential answer to this question may lie in the shapes of the curves
relating fitness returns to size, which may differ between the sexes where
females increase the fitness return with size faster than males (Charnov,
1987; Frank, 1995).
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
I thank J. A. Rosenheim for his advice, suggestions, and encouragement for this research. I also thank J. A. Rosenheim, R. W. Thorp, and L. S. Kimsey for their reviews of earlier drafts of this article.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Alexander RD, Sherman PW, 1977. Local mate competition and parental investment in social insects. Science 196:494-500.
Barthell JF, Thorp RW, 1995. Nest usurpation among females of an introduced leaf-cutter bee, Megachile apicalis.Southwest Entomol 20:117-124.
Brockmann HJ, Grafen A, 1989. Mate conflict and male behaviour in a solitary wasp, Trypoxylon (Trypargilum) politum (Hymenoptera: Sphecidae). Anim Behav 37:232-255.
Brockmann HJ, Grafen A, 1992. Sex ratios and life-history patterns of a solitary wasp, Trypoxylon (Trypargilum) politum (Hymenoptera: Sphecidae). Behav Ecol Sociobiol 30:7-27.
Charnov EL, 1982. The theory of sex allocation. Princeton, New Jersey: Princeton University Press.
Cooper KW, 1984. Discovery of the first resident population of the European bee, Megachile apicalis, in the United States (Hymenoptera: Megachilidae). Entomol News 95:225-229.
Cowan DP, 1991. The solitary and presocial Vespidae. In: The social biology of wasps (Ross KG, Matthews RW, eds). Ithaca, New York: Comstock; 33-73.
Fisher RA, 1930. The genetical theory of natural selection. Oxford: Oxford University Press.
Frank SA, 1995. Sex allocation in solitary bees and wasps. Am Nat 146:316-323.
Frohlich DR, Tepedino VJ, 1986. Sex ratio, parental investment, and interparent variability in nesting success in a solitary bee.Evolution 40:142-151.
Kim JY, 1996. Maternal investment and reproductive strategies of an adventive leaf-cutter bee, Megachile apicalis (Hymenoptera: Megachilidae) (Ph.D. dissertation). Davis, California: University of California.
Maeta Y, 1978. Comparative studies on the biology of the bees of the genus Osmia of Japan, with special reference to their management for pollination of crops (Hymenoptera: Megachilidae). Bull Tohoku Natl Agric Exp Station 57:1-221.
Noonan KM, 1978. Sex ratio of parental investment in
colonies of the social wasp Polistes fuscatus. Science
199:1354-1356.
Rosenheim JA, Nonacs P, Mangel M, 1996. Sex ratios and multifaceted parental investment. Am Nat 148:501-535.
Schwarz MP, 1988. Local resource enhancement and sex ratios in a primitively social bee. Nature 331:346-348.
Stark RE, 1992. Sex ratio and maternal investment in the multivoltine large carpenter bee Xylocopa sulcatipes (Apoidea: Anthophoridae). Ecol Entomol 17:160-166.
Stephen WP, 1987. Megachile (Eutricharea) apicalis, an introduced bee with potential as a domesticable alfalfa pollinator. J Kans Entomol Soc 60:583-584.
Sugiura N, Maeta Y, 1989. Parental investment and offspring sex ratio in a solitary Mason bee, Osmia cornifrons (Radoszkowski) (Hymenoptera, Megachilidae). Jpn J Entomol 57:861-875.
Tepedino VJ, Parker FD, 1988. Alternation of sex ratio in a partially bivoltine bee, Megachile rotundata (Hymenoptera: Megachilidae). Ann Entomol Soc Am 81:467-476.
Tepedino VJ, Torchio PF, 1982. Phenotypic variability in nesting success among Osmia lignaria propinqua females in a glasshouse environment (Hymenoptera: Megachilidae). Ecol Entomol 7:453-462.
Thorp RW, Frankie GW, Barthell J, Gordon D, Newstrom L, Griswold T, Schmidt J, Thoenes S, 1992. Ecological research: long-term studies to gauge effects of invading bees. Calif Agric 46:20-23.
Torchio PF, Tepedino VJ, 1980. Sex ratio, body size and seasonality in a solitary bee, Osmia lignaria propinqua Cresson (Hymenoptera: Megachilidae). Evolution 34:993-1003.
Trivers RT, Hare H, 1976. Haplodiploidy and the
evolution of the social insects. Science
191:249-263.
Visscher PK, Danforth BN, 1993. Biology of Calliopsis pugionis (Hymenoptera: Andrenidae): nesting, foraging, and investment sex ratio. Ann Entomol Soc Am 86:822-832.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||