The effect of requeening in late July on honey bee colony development on the Northern Great Plains of North America after removal from an indoor winter storage facility

Abstract

The effect of late summer requeening on the subsequent development of honey bee colonies during autumn (Harris, 2008b), and when confined in an indoor wintering facility (Harris, 2009) was extended with observations on sealed brood production, colony size and colony demographics every twelve days from 11 March until 14 August after they were removed from their winter quarters. Average adult populations declined for the first 48 days, and then recovered over the next 24 to 36 days once adult emergence consistently exceeded worker mortality. Rates of mortality for wintered workers were similar to those recorded for bees emerging in April, May, June, July and most of August. The last surviving bees from worker cohorts marked in September and October 1976 died between 3 June and 15 June 1977. Requeening treatment effects were quite variable and not statistically different. Requeened colonies were, however, generally larger than those headed by older queens when the experiment was terminated on 14 August and these colonies were killed and counted. The nine largest colonies belonged to the requeened treatments and contained on average 8,637 more bees (range 85 to 17,735) than the largest colony that had not been requeened. One of the requeened colonies was estimated to have contained slightly more than 80,000 adult bees at its peak population on 9 July.

Full text

Journal of Apicultural Research and Bee World 47(4): 271–280 (2008) © IBRA 2008
ORIGINAL RESEARCH ARTICLE
Effect of requeening on fall populations
of honey bees on the northern Great Plains
of North America.
J. Lloyd Harris
1,2*
1
Department of Entomology, University of Manitoba, Winnipeg, Manitoba, R3C 2N2, Canada.
2
Present address: 2839 Lakeview Ave., Regina, Saskatchewan, S4S 1G6, Canada.
Received 24 January 2008, accepted subject to revision 13 April 2008, accepted for publication 24 June 2008.
*
Corresponding author. Email: Lharris@agview.sasktelmail.com
Summary
The effects of requeening honey bee colonies in the last week of July with newly mated queens, mature queen cells, or supersedure
cells, on sealed brood, adult worker bee populations and colony population demographics were assessed at twelve day intervals until
early December. Requeening altered brood rearing patterns, adult worker bee populations and colony demographics. Requeened
colonies contained populations with higher proportions of young bees. By early December, colony population sizes converged
amongst treatments and were not statistically different.
Efecto del reemplazamiento de reinas en poblaciones débiles
de abejas melíferas al norte de las Grandes Llanuras de
Norteamérica.
Resumen
Se evaluaron los efectos del reemplazamiento de reinas en colonias de abejas melíferas con reinas recién apareadas, celdas reales ó
sustitución de celdas, sobre la cría operculada, la población de obrera adulta y otros parámetros poblacionales de la colonia a
intervalos de 12 días a partir de la última semana del mes de julio hasta principios de diciembre. El reemplazamiento alteró los
patrones de la cría, de la población de abeja adulta y los parámetros poblacionales de la colonia. Las colonias con reinas
reemplazadas presentaron poblaciones con altas proporciones de abejas jóvenes. A principios de diciembre, el tamaño poblacional
de la colonia coincidió entre tratamientos y no presentaron diferencias estadísticas.
Introduction
Providing colonies with large fall populations of young bees in
temperate climates has been considered to be an essential part
of preparing colonies for winter. For many years, it has been
suggested that the easiest way of achieving this is to introduce
young queens into colonies in the fall (Bliss, 1936) since these
colonies will continue to rear brood late into the fall, long after
colonies with older queens cease brood rearing (Free and Racey,
1968; Moeller, 1977, personal observation, 1972-1980).
There are several ways of providing colonies with a young
queen in late summer. The method traditionally used involves
removing the “old” queen and then replacing her with a newly
mated queen. In recent years, however, some large commercial
beekeeping businesses have viewed requeening queenright
colonies with mature queen cells as being more economically
efficient, but this can result in variable success. Peer (1977)
reported acceptance of greater than 80 %, while Jay (1981), Boch
and Avitabile (1979) and Szabo (1982) obtained 13 to 31 %
acceptance. Since requeening success with queen cells has been
shown to be influenced by how long a colony has been queenless
(Baribeau, 1976), inconsistent results should be anticipated when
DOI: 10.3896/IBRA.1.47.4.07

272 Harris
the resident queen is not removed before attempting to requeen
with mature queen cells. Acceptance rates are considerable
better (37.9 – 84.6 %) when cells are introduced into queenless
colonies (Invernizzi et al., 2006). Occasionally, beekeepers will also
requeen colonies by removing the queens from the colonies
allowing supersedure queens to be reared from eggs or very
young larvae. A similar situation occurs if colonies accidentally
become queenless.
Although there is much speculation on the positive effects of
requeening colonies on fall honey bee populations, there have
been few studies. The recommendation for requeening colonies
in late summer or fall has been that colonies should be
requeened at least two months before they are packed for winter
or two months before the first killing frost (Phillips and Demuth,
1918a; 1918b; Richmond, 1924; Gooderham, 1926; Dyce and
Morse, 1974). The first day of August has been viewed by many
beekeepers as an appropriate time to requeen colonies (Farrar,
1944).
This study was undertaken to provide some information
regarding the merit of late summer or fall requeening and the
relative benefits of various requeening methods. Of primary
concern was the effect on brood rearing during the fall, the
ultimate size of the overwintering colony, and the age of the bees
forming the wintering colonies. Three different requeening
methods timed to coincide with the period when commercial
beekeepers in Manitoba, Saskatchewan, Alberta and North
Dakota typically remove honey were investigated.
Materials and Methods
The experiment was conducted on 24 colonies in a commercial
apiary located in south central Manitoba, established in spring
with commercial package bees. Colonies were initiated with 0.9
kg of an Italian strain of honey bee supplied by Wenner Apiaries,
California, USA. Each colony was provided with a newly mated
sister queen to minimize the influence of genetics on colony
development. Each hive initially contained 4 combs of honey, 2
frames containing the equivalent of at least 1 full frame of pollen,
1 frame of sugar syrup and 2 frames of drawn comb. Within the
apiary, the hives were arranged in a circular pattern 1 m apart
with hive entrances facing outwards to minimize drifting between
colonies (Free, 1958; Jay, 1966). All colonies were hived on the
evening of 24 April after sunset to allow the bees to become
accustomed to the hives before conditions were conducive for
flight. Queens were placed in their respective colonies still in their
shipping cages and then released after three days. Colonies were
divided into four groups of six colonies and randomly assigned to
their respective treatment groups in April. The first group
retained their original queens and served as a reference group for
evaluating the requeening treatments. The untreated group of
colonies was reduced from six to five colonies (n=5) after one
these colonies became queenless prior to the application of the
treatments and was removed from the experiment. The
remaining eighteen colonies were made queenless during the last
week of July (26 July or Day 96 after hiving) and requeened with
newly mated queens (n=6), mature queen cells (n=6), or were
allowed to rear new queens from supersedure cells (n=6).
All colonies received similar management during the spring
and summer, being managed for honey production with additional
hive bodies being added as required. Honey was removed from
colonies on 26 July (Day 96 after hiving), and again on 19 August
(Day 120). Sugar syrup was fed to colonies commencing on 31
August (Day 132) and insulated inner covers were placed on
colonies on 11 November (Day 204).
Colonies were sampled at twelve day intervals according to
the methodology outlined by Harris (1985). Sealed brood area
was measured using a grid of 2.54 cm squares (Nolan, 1925;
Moeller, 1961). Fully occupied squares were counted directly and
partially filled squares were estimated. Sealed brood area
measurements were then converted to cell estimates based on
the number of cells per unit area. Worker bee survivorship was
determined from the survival rates of successive cohorts of 100
newly emerged workers marked with paint (Harris, 1979) as
determined from marked bee census data collected at regular
twelve day intervals. Adult worker bee populations were
estimated by summing worker bees emerging from sealed brood
and correcting for their age related mortality (Harris, 1985).
Worker bee mortality was not influenced by Nosema apis,
Acarapis woodi or Varroa destructor.
Statistical tests
Requeening treatments were evaluated in terms of their effects
on brood rearing, adult worker bee populations, and colony
demographics. Treatment effects were subjected to a one-way
analysis of variance and treatment means compared with a
Tukey’s HSD test. Means followed by the same superscripted
designation were not statistically different (P>0.05) on that
observation date. Sealed brood data were also analysed for
treatment effects, date effects and treatment-date interactions
using SAS multivariate analysis for repeated measures.
Results
Sealed brood
Seasonal changes in sealed brood production are shown in Fig. 1
and Table 1. Pretreatment colony development is also shown in
Fig. 1, commencing at hiving and every twelve days thereafter until
the requeening treatments were applied on 26 July (Day 96 after
hiving). Sealed brood production was similar in colonies in the
four treatments groups prior to treatment application (P>0.05).
Twelve days after colony were requeened (7 August; Day
108), sealed brood production had declined to 62.7 to 64.2 % of
that which it had been on 26 July (Day 96). After this initial
reduction in brood rearing, requeening treatment effects became
obvious as the brood rearing patterns deviated from those
observed in the untreated colonies. The overall requeening
treatments temporarily interrupted brood production in the
colonies and then skewed the “normal” late summer and fall
brood rearing curves to the right (Fig. 1).
Statistical differences in sealed brood production between
treatments using a one-way analysis of variance (P< 0.05) were
only observed on 19 August and 31 August (Days 24 and 36)
(Table 1). After 31 August, sealed brood production in requeened
colonies was greater than that in colonies that had retained the

Requeening honey bees on the Great Plains of North America 273
original package queens, as was cumulative sealed brood
production (P=0.04; Table 1). Multivariate analysis of the sealed
brood data indicated significant differences between treatments
(P=0.0034), date (P<0.0001) and treatment-date interaction
(P<0.0001). These differences combined with age related
mortality ultimately defined colony demography and the
overwintering colony’s size, since most of the winter colony is
produced after the end of August in Manitoba (Harris, 1980;
2008; Mattila et al., 2001).
In November, brood rearing resumed as shown by sealed
brood measurements taken on 5 December but at levels slightly
higher than those reported by Avitabile (1978) in November and
December. Sixty percent (3 of 5) of the colonies with the original
package queens and 83 % (15 of 18) of the requeened colonies
contained sealed brood, but the amount was similar in all
treatments (P>0.05). The small amount of brood reared in
November was enough to produce a modest increase in adult
populations by mid December in 12 of the 18-requeened
colonies and 3 of the 5 original package queen colonies.
Adult bee populations
Seasonal changes in adult bee populations are shown in Fig. 2 and
Table 2. Populations in all treatment groups were similar prior to
requeening on 26 July (P>0.05), but the normal/untreated
colonies contained more bees. For the first twelve days after
colonies were requeened, all colony populations continued to
increase and were relatively unaffected by the requeening
treatments because workers emerging for twelve days following
the requeening treatments would have been produced by the
original queens. Worker bee populations subsequently declined
rapidly in all requeened colonies until 30 October after which
population decline was less dramatic (Fig. 2). In broodless colonies
which retained the original package queen, this transition from
short lived to long lived bees began on or about 18 October
(Mattila et al., 2001).
Once colonies ceased to rear large amounts of brood, their
adult populations converged towards a common size. Between 7
August (Day 108) and 30 October (Day 168) adult populations
declined from 50,000 – 60,000 bees to less than 17,000, a decline
of approximately 70%. Only those bees emerging after the last
week of August became part of the wintering colony, especially
those that emerged between mid September and the end of
October. By 5 December, only 44.3 to 53.6 % of workers
emerging from sealed brood between 31 August and 5
December had survived long enough to be part of the winter
colony. Hence, the wintering colonies were defined by the
amount of brood reared after 31 August (Day132), its frequency
distribution with respect to time and individual cohort
survivorship.
Statistical differences between treatments by date are shown
in Table 2. Statistical differences (P<0.05) between treatments did
not occur until 36 days after colonies were requeened (31
August) and were no longer detectable after 6 October. Colony
populations continued to decline until 5 December when the
colonies were moved to a wintering facility. Colonies requeened
with newly mated queens or requeened with mature queen cells
contained approximately 3000 more adult bees than colonies
allowed to requeen themselves from supersedure cell or colonies
which retained the original package queens.
Age structure
Changes in colony age structures for each treatment group are
shown in Figs 3 to 5. Colony age structures were determined for
each time period by estimating worker production from sealed
brood estimates and worker mortality using survival estimates
from marked worker bees according to the Harris Population
Methodology (Harris, 1985). The cumulative effects of colony
natality and mortality rates are shown for each treatment group
(Figs 3 to 5) and provided colony demography estimates based
upon the proportion of the colony in various age classes.
Colony demography was similar before treatments were
applied. Changes in colony demography were most evident in
colonies requeened with mature queen cells or colonies
requeened with queens reared from supersedure cells than in
colonies requeened with newly mated queens. The development
of the overwintering population is illustrated by the dark grey
shading on the population age structures and its production
begins with worker cohorts emerging between 31 August and 12
September (Fig. 4).
Fig. 1. Seasonal trends in sealed brood production in colonies started with 0.9 kg of worker bees and subsequently left untreated
or requeened with newly mated queens, mature queen cells or colony reared queens from supersedure cells.

274 Harris
Fig. 2. Seasonal trends in adult worker bee populations started with 0.9 kg of worker bees and subsequently left untreated or
requeened with newly mated queens, queens from mature queen cells or queens from supersedure cells.
Table 1. Sealed brood production from 7 August (Day108 after hiving) to 5 December (Day 228) as affected by requeening
colonies on 26 July (Day 96) with newly mated queens, queens from mature queen cells or queens from supersedure cells.
Treatments followed by different letters are statistically different (ANOVA) at the stated probability levels.
Table 2. Adult bee populations from 26 July (Day 96) to 5 December (Day 228) as affected by requeening colonies on 26 July
(Day 96) with newly mated queens, queens from mature queen cells or queens from supersedure cells. Treatments followed by
different letters are statistically different at the stated probability levels.

Requeening honey bees on the Great Plains of North America 275
Fig. 3. Seasonal changes in the age structures of honey bee colonies at 12 day intervals between 26 July and 31 August in response
to being requeened on 26 July with newly mated queens, queens from mature queen cells or queens from supersedure cells.

276 Harris
Fig. 4. Seasonal changes in the age structures of honey bee colonies at 12 day intervals between 12 September and 18 October in
response to being requeened on 26 July with newly mated queens, queens from mature queen cells or queens from supersedure
cells.

Fig. 5. Seasonal changes in the age structures of honey bee colonies at 12 day intervals between 30 October and 05 December in
response to being requeened on 26 July with newly mated queens, queens from mature queen cells or queens from supersedure
cells.
Requeening honey bees on the Great Plains of North America 277

278 Harris
Discussion
Colony development is controlled by “birth”, death, emigration
(i.e. swarming and drifting) and immigration (i.e. drifting).
Management of these four variables is critical to effective colony
management. In this study, colonies were managed to promote
“birth” and to minimise the adverse effects of the other three
variables. Drifting of bees between colonies was minimized by the
physical arrangement of colonies within the apiary. Swarming was
prevented by providing colonies with adequate hive space and
the timely removal of stored honey. Colonies were treated with
antibiotic to control brood diseases, and were not infested by
tracheal or varroa mites since this experiment took place before
their introduction into North America. Insecticides were not
applied to the surrounding crops. As a consequence, colony
development was primarily controlled by changes in colony
“birth” rates induced by the various requeening methods,
although there was also an inflection in worker mortality in
October, which significantly affected the ultimate colony
development just prior to winter.
In Manitoba, colonies are typically requeened either in May or
June prior to the beginning of the honey flow or after mid-to late
August immediately after the honey flow. When requeening takes
place is a matter of economics, preference, opportunity,
convenience and geographic location. These variables define a
“window-of-opportunity” that will vary for every beekeeper and
be progressively wider at more southerly latitudes.
In this experiment, dequeening followed by requeening
treatments occurred during the honey flow and coincided with
the removal of honey supers on 26 July and within 21 days of the
anticipated end of any significant honey flow. This ensured that
there would be a supply of newly mated, local queens and mature
queen cells, that requeening would have minimal effects on
colony populations and that additional labour did not have to be
expended to handle full honey supers more than once. A second
“window of opportunity” was available for requeening colonies
on 19 August when honey was once again removed from the
colonies at the end of the honey flow. Although this timing would
have been acceptable, it was arbitrarily rejected in favour of the
earlier data because of the different methods used to requeen
these colonies.
Brood production
Pretreatment brood rearing appeared to be “normal” in all
colonies prior to implementation of the requeening treatments.
However, a closer examination of the sealed brood data in Fig. 1
suggests otherwise. Sealed brood production peaked the third
week of June just as the honey flow began and steadily declined
even before colonies were requeened on the 26 July. This
occurred despite the addition of two Langstroth supers of empty
comb to colonies on 2 July and a third super on 15 July. These
three honey supers should have been adequate to prevent any
overcrowding. Bees were not excessively dense on the combs
nor were they clustered at the hive entrance or on the exterior
face of brood nest supers. There should have also been adequate
storage space for the 69.85 kg of honey collected in July.
Apparently, there was adequate empty comb space in the hive
for brood rearing and honey storage, but its placement within the
hive may have affected cell utilization by the queen. When empty
comb is not contiguous with the existing brood, it is almost never
selected for brood rearing, so the empty combs placed above the
existing supers containing honey were essentially unavailable to
the queen. When there was a strong honey flow, honey and
nectar was stored in the brood areas as quickly as the new
workers emerged from them. Once this honey was stored within
the brood nest, its storage became permanent when it was in
excess of the immediate needs for brood rearing within the
brood nest, effectively reducing cell availability for egg laying. In
feral colonies, a similar condition may occur in cavities with
limited brood space. Provided there is adequate space within the
hive, however, workers continually add comb to the bottom of
the existing combs. These new cells near the bottom of the
combs are contiguous with the existing brood nest and will
immediately have eggs deposited into them rather than honey.
Apparently, cells need to be extended to a certain depth before
workers will deposit honey into them while the queen is less
particular (personal observation).
Since brood rearing declined prior to treatment, one is left to
conclude that the queen and workers were competing for cell
usage. Initially, the competition for space was very subtle with
honey storage encroaching on the brood area along its periphery.
It became more evident on 26 July when there was a strong
honey flow and workers were storing nectar and honey in cells in
the brood area as soon as they were vacated. This competition
between the queen and workers for cell utilization probably
obscured any immediate treatment effects on 7 August. Insertion
of empty combs or frames of foundation into the brood nest in
mid June could perhaps have alleviated this situation.
The net effect of the requeening treatments was to alter
normal brood rearing patterns with respect to time and intensity.
After 31 August, requeened colonies reared more brood than
original package queen colonies. The altered brood rearing
patterns directly affected colony demography and colony size.
Only those bees emerging from sealed brood after the last
week of August became part of the wintering colony, especially
those that emerged between mid September and the end of
October. Because of this, the overwintering populations were
defined by the amount of brood reared after 31 August (Day
132), its frequency distribution with respect to time and individual
cohort survivorship.
Colony populations
Adult populations declined between 7 August and early October.
The decline was rapid, and primarily related to the rapid decline
in brood rearing (Fig. 1) associated with honey storage within the
brood nest, requeening treatments, and the cessation of flowering
of nearby canola (Brassica napus) after 19 August plus the
absence of other major nectar or pollen producing plants.
Worker survivorship was relatively constant throughout this
period until October when workers began to live longer.
Following this inflection in the survivorship curves, the pace of
population decline slowed significantly. The difference between
colony birth rates and worker death rates defined the population
decline; the larger the difference, the faster the colony’s
population declined.
Because colony population size converged between
treatments in November and December, it would appear that

Requeening honey bees on the Great Plains of North America 279
colony populations could be reduced immediately after the end
of the honey flow in August to about 20,000 to 25,000 adult
workers without seriously affecting the colony population in
December. It might even result in saving much honey that might
be better utilized as winter feed rather than as feed for bees that
are no longer needed for honey production.
Although colony populations in the four treatments on 5
December were not statistically different (P>0.05), the trend to
higher populations in colonies requeened with newly mated
queens and with mature queen cells is noteworthy. Requeening
colonies, regardless of the method used, did not reduce the size
of the wintering colony on 5 December, but the age composition
differed considerably.
Colony demographics
Colony demographic patterns observed in these colonies were
substantially different from those proposed by Bodenheimer
(1937), but were very similar to those proposed by Fukuda
(1983) after he applied life table data from Sapporo, Japan to
Bodenheimer’s sealed brood data and made some modifications
to Bodenheimer’s methodology (Fukuda, 1971).
The colony age structures reported here incorporated egg
and larval estimates and adjusted them for the duration of the
respective stages rather than the number of each stage required
to produce the observed pupal estimates (Fukuda, 1983). The
most striking feature of these age distributions is the speed at
which the various cohorts diminish as they aged beyond 36 days
of age. Apparently, bees do not die of old age or colony age
distributions would have resemble those proposed by
Bodenheimer (1937), which show large numbers of bees
progressing to the older age classes and then dying abruptly and
more or less in unison.
Requeening colonies in late summer or early fall changed
colony demographics. These changes were primarily related to
seasonal mortality and survival rates and seasonal variation in
brood production as influenced by the queen’s age and the
method used to requeen the colonies. At the end of the season,
all requeening methods produced colonies with higher
proportions of younger bees than in original package queen
colonies. This may, however, prove to be irrelevant to population
build up in the spring, since Nickel and Armbruster (1937), Free
and Spencer-Booth (1959), and Fukuda and Sekiguchi (1966) have
all noted that bees die at about the same rate in the spring
regardless of their chronological age.
Conclusions
Fall or late summer colony requeening alters seasonal brood
rearing patterns, colony population size and colony demographics.
The way it is changes depends upon timing and requeening
method. Colonies requeened with newly mated queens, queen
cells or supersedure cells reared more brood later in the season
than did those colonies headed by queens that had been in
colonies since spring. As a consequence, requeened colonies
were composed of a much higher proportion of younger bees,
the significance of which has yet to be ascertained. Colonies
requeened with newly mated queens and queen cells had slightly
larger average populations than colonies requeened with
supersedure cells or those that retained their original package
queens. The differences were not statically different (P>0.05) but
are probably indicative of a strong trend towards higher
populations.
The most significant benefit from requeening colonies in the
mid summer or very early fall may not just be the higher
proportion of young bees reared by these colonies in the fall.
Requeening also allows colonies that were requeened with newly
mated queens and queens from mature queen cells to have a
slightly larger adult populations for winter and young queens that
can be obtained from local sources, which are less likely to die
during the winter or to be superseded the following season.
Acknowledgements
I wish to express my appreciation to Dr. Rob Curry and Dr. Don
Nelson and two unknown reviewers for helpful suggestions and
criticism of the manuscript; Dr. S. C. Jay for the opportunity to
have conducted the research; the National Research Council of
Canada and the Harris Honey Company for their financial
support; Ms. J. Casey, Mr. J. DePape, Mr. B. Fingler, Ms. G. Gerbrandt,
Mr. P. Gregory, Mr. J. Gruszka, Mr. L. Hrenchuk, Mr. D. Melia, and Ms.
L. Taylor for their untiring and diligent assistance in collecting data;
and to Mr. R. Graham and Mr. D. Dixon for providing the queens
and queen cells used in this study.
References
AVITABILE, A (1978) Brood rearing in honey bee colonies from late autumn to
early spring. Journal of Apicultural Research 17: 69–73.
BARIBEAU, M (1976) Timing of introducing queen cells when requeening colonies.
American Bee Journal 116: 109.
BLISS, R K (1936) Wintering bees in Iowa. Iowa State College of Agriculture and
Mechanic Arts Extension Circular 141 (revised).
BOCH, R; AVITABILE, A (1979) Requeening honey bee colonies without
dequeening. Journal of Apicultural Research 18: 47–51.
BODENHEIMER, F S (1937) Studies in animal populations. II. Seasonal population
trends of the honey bee. Quarterly Review of Biology 12: 406–425.
DYCE, E; MORSE, R A (1974) Wintering honey bees in New York State. New York
State College of Agriculture and Life Sciences Extension Bulletin 1054.
FARRAR, C L (1944) Productive management of honey bee colonies in the
northern states. United States Department of Agriculture Circular No. 702.
FREE, J B (1958) The drifting of honey bees. Journal of Agricultural Science, Cambridge
51: 294–306.
FREE, J B; SPENCER-BOOTH, Y (1959) The longevity of worker honey bee (Apis
mellifera). Proceedings of the Royal Society, London A34: 140–150.
FREE, J B; RACEY, P A (1968) The effect of the size of honey bee colonies on food
consumption, brood rearing and the longevity of bees during the winter.
Entomologia Experimentalis et Applicata 11: 242–249.
FUKUDA, H (1971) Improvement of Bodenheimer's method for estimating
individual number in honey bee colonies. Journal of the Faculty of Science,
Hokkaido University Series VI. Zoology 18: 128–143.
FUKUDA, H (1983) The relationship between work efficiency and population size
in a honey bee colony. Researches on Population Ecology 25: 249–263.
FUKUDA, H; SEKIGUCHI, K (1966) Seasonal change of the honey bee worker
longevity in Sapporo, North Japan, with notes on some factors affecting the
life-span. Japanese Journal of Ecology 16(5): 206–212.
GOODERHAM, C B (1926) Wintering bees in Canada, Dominion of Canada,
Department of Agriculture Bulletin No. 74-new series.

280 Harris
JAY, S C (1966) Drifting of honey bees in commercial apiaries. III. Affect of apiary
layout. Journal of Apicultural Research 5: 137–148.
JAY, S C (1981) Requeening queenright honey bee colonies with queen cells or
virgin queens. Journal of Apicultural Research 20: 79–83.
HARRIS, J L (1979) A rapid method for colour-marking single honey bees with
fluorescent paint. Journal of Apicultural Research 18: 201–203.
HARRIS, J L (1980) A population model and its application to the study of honey
bee colonies. MSc.Thesis, University of Manitoba, Winnipeg, Manitoba, Canada.
HARRIS, J L (1985) A model of honey bee colony population dynamics. Journal of
Apicultural Research 24: 228–236.
HARRIS, J L (2008) Development of honey bee colonies initiated from package
bees on the Northern Great Plains of North America. Journal of Apicultural
Research 47: 141–150. DOI:10.3896/IBRA.1.47.2.10
INVERNIZZI, C; HARRIET, J; CARVALHO, S (2006) Evaluation of different queen
introduction methods in honey bee colonies in Uruguay. Apiacta 41: 1–20.
MATTILA, H R; HARRIS, J L; OTIS GW (2001) Timing of production of winter bees
in honey bee (Apis mellifera) colonies. Insectes Sociaux 48: 88–93.
MOELLER, F E (1961) The relationship between colony population and honey
production as affected by honey bee stock lines. United States Dept.Agric.
Production Research Report No. 5.
MOELLER, F E (1977) Overwintering of honey bee colonies. United States
Department of Agriculture Bulletin No. 169.
NICKEL, K H; ARMBRUSTER, L (1937) Vom Ledensablauf der Arbeitsbienen
besonders auch bei Nosemaschäden. 1. Teil. Archiv für Bienenkunde
18: 257–287.
NOLAN, W J (1925) Brood rearing cycle of the honeybee colonies. United States
Department of Agriculture Bulletin No. 1349.
PEER, D (1977) Requeening with queen cells. Canadian Beekeeper 6: 89.
PHILLIPS, E F; DEMUTH, G S (1918a) The preparation of bees for outdoor
wintering. United States Department of Agriculture Farmers’ Bulletin 1014.
PHILLIPS, E F; DEMUTH, G S (1918b) Wintering bees in cellars. United States
Department of Agriculture Farmers’ Bulletin 1014.
RICHMOND, R G (1924) Wintering bees in Colorado. Colorado Agriculture College
Circular 45.
SZABO, T I (1982) Requeening honey bee colonies with queen cells. Journal of
Apicultural Research 21: 208–211.