| |
To receive a reprint of this article, please request from Stephen Peterson:
steve@pollination.com
Abstract
Postdiapause developmental rates of three
groups of Alfalfa Leafcutting Bee, Megachile rotundata (F.), from a common
source were compared. In 1990, bees from Manitoba, Canada, were raised
in Idaho, California and Manitoba. Diapausing prepupae from these locations
were incubated during 1991 in observation cells and development time and
weight of adults were recorded. Maturity of the bee pupae was rated from
one to seven, defining six distinct development periods. Overall, bees
from Manitoba emerged 2.8 d earlier and weighed 3.8 mg more than the bees
from Idaho and California. Most of the difference among groups and between
the sexes occurred between the prepupal and teneral pupal stages (Period
1). After Period 1, an average of 147 degree-days were required for adult
emergence.
Key words: Alfalfa Leafcutting Bee, Megachile rotundata, development,
incubation, weight, latitude
Introduction
The Alfalfa Leafcutting Bee, Megachile
rotundata (F.), is the primary pollinator for alfalfa seed in the northwestern
United States and western Canada. The bee is a Eurasian species thought
to have been introduced to North America several times (Stephen and Osgood
1965). In the Northwest, yields of 2200 kg/ha of seed (2000 lb./A) are
possible when the leafcutting bee is used as the pollinator (Johansen
et al. 1979). A major industry has developed in Canada to provide disease-free
bee larvae to seed producers in the United States.
In the management of leafcutting bees, an
understanding of bee development during incubation is important for two
reasons. First, the timing of insecticide use for parasitoid control is
only recommended during the eighth through thirteenth day (Murrell and
Gayton 1987). Second, the timing of adult bee release in the field to
coincide with alfalfa bloom is critical to obtain maximum yields. Incubation
of leafcutting bee prepupae usually begins 3 weeks prior to alfalfa blossoming.
Adult emergence can be delayed by cooling the incubator to 15 or 20°C
during the final days of incubation (Rank and Goerzen 1982), if blossoming
is later than normal because of cool temperatures or if stormy conditions
occur when bees are to be released.
Delaying bee emergence occasionally leads
to confusion among growers as to when adult emergence will actually occur.
Also, there is concern among bee producers that bees from some locations
develop at different rates than bees from other locations. The purpose
of this study was to determine whether rearing bees from a common source
in different locations could modify bee developmental rates and to determine
the thermal requirements for progression of bee pupae through the development
stages.
Materials and Methods
During 1990, groups of bees from Fisher
Branch, Manitoba, Canada, were reared in Fresno, California, U.S.A. (36°45'N,
119°75'W), Midvale, Idaho, U.S.A. (44°30'N, 116°45'W), and
Fisher Branch, Manitoba (51°0'N, 97°30'W). Diapausing offspring
from these bees were used in this study. The bees from Idaho and Manitoba
had one complete and a partial second generation in 1990 while the California
bees had two complete and a partial third generation in 1990.
The diapausing prepupal bees from the three
locations were held at 4°C for 9 months. Bees from Manitoba and Cailfornia
were subsequently shipped to Idaho. Cocoons were then removed from the
wood laminate nest material. Cocoons were opened with a razor blade and
the prepupae removed with forceps. The prepupae were then placed in compartmented
trays, similar to those used by Hobbs and Richards (1977). Trays consisted
of a plastic divider of 100 cells (10 by 10) with plate glass taped to
the top and bottom. Each cell was 14 by 14 by 9 mm. One-hundred-fifty
prepupae from each location were placed in individual cells. The trays
were then placed in a common incubator in darkness at a constant 29.0°C
and relative humidity between 50 and 60%.
Each insect was checked daily and rated
with respect to maturity. A scale was developed to rate morphologically
distinct stages in the development sequence. Bees were rated as follows:
stage 1 = prepupa (white, legless prepupa)(Figure 1), stage 2 = teneral
pupa (white, recently molted pupa)(Figure 2), stage 3 = light-eye pupa
(color of compound eye ranges from light pink to tan)(Figure 3), stage
4 = dark-eye pupa (compound eyes are dark red or brown, as are ocelli)(Figure
4), stage 5 = blackening pupa (body with some pigmentation, beginning
with dorsal thorax and mandibles)(Figure 5), stage 6 = black pupa (entire
body pigmented)(Figure 6), stage 7 = adult (pupal integument shed, wings
expanded, light stripes visible on abdomen)(Figure 7).
When each bee reached the adult stage
it was placed in a labeled petri dish and frozen. Adult bees were sexed
and weighed with a precision of ±0.1 mg after death.
Six postdiapause periods were analyzed statistically,
i.e., Period 1 was the number of days from stage 1 to stage 2, Period
2 was the number of days from stage 2 to stage 3, etc. If a period lasted
less than 24 hr, the length of the period was scored as 0 d. Analysis
of variance was used to test for significant differences (a = 0.05) among
locations and between sexes for the postdiapause periods and the weight
of the bees. The protected least significant difference method was used
to separate the means.
Results
The proportion of pupae surviving to adulthood
differed among locations (Chi-square = 13.78; df = 2; P < 0.01). Survival
(% ± 95% C.I.) was 90.0 ± 5.0%, 87.3 ± 5.7%, and
75.3 ± 8.0% for the Manitoba, Idaho and California locations, respectively.
Among the surviving bees, no interaction
between sex and location occurred for any of the periods (F = 0.33 to
2.85; df = 2,361; P > 0.05). The first period ranged from 7 to 20 d
in length. For Period 1, males (mean = 10.4 d) developed significantly
faster than females (12.0 d)(F = 62.17; df = 1,354; P < 0.0001), and
a significant difference among locations occurred (F = 78.16; df = 2,354;
P < 0.0001) (Figure 8). Bees reared in Idaho and California required
2.8 d longer than bees from Manitoba in Period 1.
Period 2 ranged from 0 to 3 d. No significant
difference between males (1.2 d) and females (1.2 d) was observed (F =
0.24; df = 1,354; P > 0.50)(Figure 8), but a significant difference
among locations occurred (F = 4.69; df = 2,354; P < 0.01). Bees reared
in Manitoba required 0.2 d longer than the Idaho and California bees to
progress through Period 2.
The third period ranged from 0 to 4 d. Males
(1.8 d) developed significantly faster than females (2.0 d) in Period
3 (F = 6.97; df = 1,354; P < 0.01), and a significant difference among
locations was observed (F = 7.95; df = 2,354; P < 0.001)(Figure 8).
The bees reared in Idaho and California required 0.3 d longer than the
bees from Manitoba in this period.
Period 4 ranged from 1 to 5 d in length.
In Period 4, no significant difference between males (3.5 d) and females
(3.4 d) occurred (F = 1.55; df = 1,354; P > 0.20)(Figure 8). A significant
difference among treatments for location was observed (F = 15.10; df =
2,354; P < 0.0001), with bees from Manitoba requiring 0.3 d longer
than the bees reared in Idaho and California.
The fifth period ranged from 0 to 4 d. No
difference between males (1.5 d) and females (1.5 d) occurred in Period
5 (F = 1.32; df = 1,354; P > 0.20)(Figure 8). A significant difference
among locations was observed (F = 6.21; df = 2,354; P < 0.01), with
the bees reared in California requiring 0.3 d longer than the other locations.
Period 6 ranged from 1 to 5 d in length. Males (2.7 d) developed significantly
faster than females (3.2 d) (F = 51.54; df = 1,354; P < 0.0001)(Figure
8). A significant difference among locations occurred (F = 4.06; df =
1,354; P < 0.05), with Idaho bees requiring 0.3 d longer than bees
from Manitoba, while California bees were intermediate.
For complete postdiapause development, i.e.
prepupal to adult stage, no interaction between sex and location occurred
(F = 1.39; df = 2,354; P > 0.20). Females took 2.2 d longer than males
(F = 105.65; df = 1,354; P < 0.0001). Overall, bees reared in California
and Idaho developed significantly slower than the bees from Manitoba (F
= 73.32; df = 2,354; P < 0.0001). Manitoba bees emerged 2.8 d earlier
than the other locations. It is clear that Period 1 had the most influence
on the overall postdiapause development of the bees.
No significant interaction between sex and
location for bee weight was observed (F = 1.39; df = 2,353; P > 0.05).
As expected, female bees weighed significantly more than males (F = 145.34;
df = 1,353; P < 0.0001). The females (36.5 ± 0.4 mg, mean ±
sem) weighed 7.6 mg more than the males (28.9 ± 0.4 mg). A significant
difference among locations occurred for weight (F = 18.81; df = 2,353;
P < 0.0001), with bees from Manitoba (36.0 ± 0.7 mg) weighing
3.8 mg more than bees reared in Idaho (32.1 ± 0.5 mg) and California
(32.2 ± 0.7 mg).
Sex ratios did not differ significantly
among locations (Chi-square = 0.49; df = 2; P > 0.50). Overall, females
accounted for 60.0 ± 6.5% (95% C.I.) of the adult bees.
Discussion
Richards and Whitfield (1988) reported
that 295 degree-days, with a temperature threshold of 15.7°C, were
required for 50% adult emergence of the Alfalfa Leafcutting Bee. They
found no differences in postdiapause development among four different
locations within Canada. In contrast, we found significant differences
among three more widely separated locations. Bees from Manitoba required
20.6 d at 29°C to reach 50% adult emergence, or 274 degree-days. Bees
reared one season in Idaho or California took 23.3 and 23.4 d, respectively,
for 50% emergence. These bees averaged 311 degree-days for postdiapause
development.
Pankiw et al. (1980) reported differences
among latitudes of Alfalfa Leafcutting Bees; however, as Richards and
Whitfield (1988) point out, storage conditions varied considerably in
their study which prevents valid comparisons. According to our study,
Alfalfa Leafcutting Bee development can be modified by changes in latitude.
We observed differences among locations of up to 12% of the postdiapause
developmental period.
Why the leafcutting bees reared in California
and Idaho developed slower than bees from Manitoba is not clear. It is
noteworthy that these bees performed "best" in their home environment,
with regard to weight and development time. In general though, wild bees
tend to emerge earlier in northern latitudes to fully exploit the intense
bloom in the relatively short season. We speculate that a short postdiapause
period might be advantageous for the Alfalfa Leafcutting Bee in northern
latitudes. Our study indicates that Alfalfa Leafcutting Bees lengthened
their postdiapause development with only one season of exposure to more
southern latitudes.
Adult bee weight has been shown to vary
with nest hole size (Klostermeyer et al. 1973), emergence day (Rothschild
1979), rearing temperature (Tepedino and Parker 1986), and diapausing
versus non-diapausing bees (Tepedino and Parker 1988). We found that bee
size can also vary with location. Klostermeyer et al. (1973) demonstrated
that provision size affected the size of the adult bee and Bohart and
Nye (1971 unpublished data) found indications that the amount of nectar
added to the pollen mass affected bee size. Thus, it is possible that
smaller provisions or less nectar was provided to the Idaho and California
bees. Because all locations were from alfalfa seed fields, lack of adequate
pollen would not seem to be a factor. Perhaps optimal foraging conditions
were longer lasting or more nectar was available in Manitoba compared
to California and Idaho.
Differences between the sexes in developmental
time occurred during the first, third and sixth postdiapause periods.
Period 1 accounted for 68% of the overall difference between the sexes.
Development time in the first period also accounted for virtually all
(99%) of the difference among the locations. Thus, the time required to
terminate diapause and initiate pupation is subject to the most variation
among populations and between the sexes. To
aid bee producers who are uncertain of when adult bees will emerge, some
guidelines have been developed from this data set to determine when adults
will emerge based on pupal morphology (Table 1). Because most variation
occurs in the first postdiapause period, the data for periods two through
six were combined across locations to obtain means. From the table, for
example, if the average stage of a sample of bees is dark-eye (4), then
an additional 7.9 d are still required at 29°C for 50% adult emergence
(Table 1), which equates to 105 degree-days. Likewise, if most bees are
in the dark-pupa stage (6), then the average adult emergence should be
in 3 d at 29°C or 40 degree-days.
Acknowledgments
We thank Carolyn Nyberg for technical assistance.
Thanks are also due to Daniel Mayer and Keith Dorschner for reviewing
the manuscript and offering suggestions. This research was funded by grants
from USDA-ARS, USDA-APHIS, FMC Corporation, and the Idaho Alfalfa Seed
Commission. Scientific paper no. 91769, University of Idaho Experiment
Station.
References
HOBBS, G.A., AND K.W. RICHARDS. 1977. An examination of methods used
in western Canada to estimate populations of alfalfa leafcutter bees.
Bee World 59:67-70.
JOHANSEN, C., C. BAIRD, R. BITNER, G. FISHER, J. UNDERRAGA AND R. LAUDERDALE.
1979. Alfalfa seed insect pest management. Western Regional Extension
Publication 0012.
KLOSTERMEYER, E.C., S.J. MECH, JR. AND W.B. RASMUSSEN. 1973. Sex and
weight of Megachile rotundata (Hymenoptera: Meagachilidae) progeny associated
with provision weights. J. Kansas Entomol. Soc. 46:536-548.
MURRELL, D. AND D. GAYTON. 1987. Alfalfa seed and leafcutter bee production
in Saskatchewan. Saskatchewan Agriculture, Soils and Crops Branch. 10
pp.
PANKIW, P., J.A.C. LIEVERSE, AND B. SIEMENS. 1980. The relationship between
latitude and the emergence of alfalfa leafcutter bees, Megachile rotundata
(Hymenoptera: Megachilidae). Can. Ent. 112:555-558.
RANK, G.H. AND D.W. GOERZEN. 1982. Effect of incubation temperatures
on emergence of Megachile rotundata (Hymenoptera: Megachilidae). J. Econ.
Entomol. 75:467-471.
RICHARDS, K.W., AND G.H. WHITFIELD. 1988. Emergence and survival of leafcutter
bees, Megachile rotundata, held at constant incubation temperatures (Hymenoptera:
Megachilidae). J. Apicultural Res. 27:197-204.
ROTHSCHILD, M. 1979. Factors influencing size and sex ratio in Megachile
rotundata (Hymenoptera: Megachilidae). J. Kansas Entomol. Soc. 53:392-401.
STEPHEN, W.P. AND C.E. OSGOOD. 1965. The induction of emergence in the
leaf-cutter bee Megachile rotundata, an important pollinator of alfalfa.
J. Econ. Entomol. 58:284-286.
TEPEDINO, V.J. AND F.D. PARKER. 1986. Effect of rearing temperature on
mortality, second-generation emergence, and size of adult in Megachile
rotundata (Hymenoptera: Megachilidae). J. Econ. Entomol. 79:974-977.
TEPEDINO, V.J. AND F.D. PARKER. 1988. Alternation of sex ratio in a partially
bivoltine bee, Megachile rotundata (Hymenoptera: Megachilidae). Ann. Entomol.
Soc. Am. 81:467-476.
Table 1. Requirements, in days and degree-days, for average adult
emergence of Alfalfa Leafcutting Bees from pupal stages two through six.
Parma, Idaho, 1991.
| Pupal Stage |
Days to 50% Emergence |
Degree-Days to 50% Emergence |
| 2 |
11.0 |
147 |
| 3 |
9.8 |
130 |
| 4 |
7.9 |
105 |
| 5 |
4.5 |
60 |
| 6 |
3.0 |
40 |
| |
Incubation temperature = 29°C.
Figure Captions
Figure 1-7. Stages 1-7 of Alfalfa Leafcutting Bee postdiapause development.
Figure 8. Days required for Periods 1-6 of Alfalfa Leafcutting Bee males
and females from Manitoba (MAN), Idaho (IDA), and California (CAL), subsequently
reared in Idaho. Period 1 = prepupa to teneral pupa, Period 2 = teneral
to light-eye pupa, Period 3 = light-eye to dark-eye pupa, Period 4 = dark-eye
to blackening pupa, Period 5 = blackening to black pupa and Period 6 =
black pupa to adult. Vertical lines represent SEM. Locations with different
letters are significantly different by the protected LSD method.
Figure 9. Compartmented trays with immature leafcutting bees. Stages
of bees in row 1 (left to right): 4, 4, 4, 4; row 2: 4, 3, 4, 2; row 3:
1, 3, 4, 2.
Figure 10. Stages of bees in row 1 (left to right): 6, 6, 6, 6; row 2:
6, 5, 5, 4; row 3: 4, 5, 6, 4.
|