The Lives of Bees
Page 14
comb that is held in 13 of the full- size (deep) frames that fit in a Langstroth
hive. Most of each colony’s comb was, as expected, the smaller- celled
worker comb, but there was also a sizable portion of the larger- cell drone
comb, 17 percent, on average (range, 10%–24%). I regret to report that
we did not measure the sizes of the worker cells in these bee- tree nests.
But working from photographs of the combs in three of these wild colo-
nies’ nests (e.g., Fig. 3.7), I have determined the mean wall- to- wall dimen-
sion of worker cells in the brood combs of three of the wild colony’s nests:
5.12, 5.19, and 5.25 millimeters (0.201, 0.204, and 0.206 inch). So, on
average, their worker cell size was 5.19 millimeters. For comparison, in
my managed colonies, which have built their combs on standard beeswax
comb foundation purchased from various manufacturers, the worker cells
have a somewhat larger mean wall- to- wall dimension: 5.38 millimeters
(0.212 inch). Once I did a study that looked at small- cell comb as a means
of controlling infestations of Varroa mites (discussed in chapter 10), and in
the pilot- work stage of this study I induced colonies to build combs on
small- cell foundation. The cells of the worker comb built by these colonies
had an average wall- to- wall dimension of just 4.82 millimeters (0.190
inch). We can conclude, therefore, that the wild- colony nests studied in
the mid- 1970s contained worker comb whose cell size was between that
of the standard- cell worker comb and small- cell worker comb found in
beekeepers’ hives but much closer to the former than the latter.
The bees in these wild colonies had organized the use of their combs in
the way that is familiar to all beekeepers, storing honey in the upper region
of the nest, rearing brood below, and creating a band of cells holding pol-
Seeley.indb 108
2/21/2019 8:07:33 AM
The Nest 109
len in between, though often with additional cells of pollen scattered
within the colony’s brood- nest region. Most of the nests were collected
and dissected in late summer—late July and August—and except for one
colony that was queenless, all were thriving. The average sizes of the
worker bee and drone populations of the queenright colonies were 17,800
and 1,004, respectively. It was clear that the colonies’ production of more
workers and drones would soon be limited by the filling of their combs
with honey. On average, the colonies had 15.1 kilograms (33.2 pounds)
of honey stores and had already filled roughly 50 percent of the cells in
their nests with honey: 56 percent of the worker cells and 48 percent of
the drone cells. They had brood in 25 percent of their worker- comb cells
and 26 percent of their drone- comb cells, so 19 percent of their worker
cells and 26 percent of their drone cells were vacant. Overall, it appeared
that these colonies were making good progress toward having the 25- plus
kilograms (55 pounds) of honey stores and the large population of young
bees that each would need to survive the winter. We examined all the
combs closely for signs of brood diseases—American foulbrood, Euro-
pean foulbrood, sacbrood, and chalkbrood—but found none. Our study
was conducted some 10 years before the first detections of tracheal mites
( Acarapis woodi, in Florida, in 1984) and Varroa mites ( Varroa destructor, in Florida, in 1987) in North America, so naturally we did not find either of
these two parasites.
NEST- SITE SELECTION
The tree cavity or rock crevice that houses a wild colony’s nest is the cen-
ter of the universe for its inhabitants. It is the spot where these bees have
built their nest, the place they will defend with their lives, and the only
site on earth to which they return from miles around bearing loads of
nectar and pollen. Both the nesting site and the beeswax combs inside are
parts of the colony’s set of survival tools that extend beyond the bodies of
its members. It is obvious to anyone who has peered inside a wild colony’s
nest and admired its combs (Fig. 5.5) that these labyrinthine structures are
products of the bees living there. After all, the beeswax used to build each
Seeley.indb 109
2/21/2019 8:07:33 AM
Fig. 5.5. Combs in the nest of a honey bee colony living in a hollow tree.
Seeley.indb 110
2/21/2019 8:07:34 AM
The Nest 111
comb is a secretion of the bees’ bodies, and the marvelous hexagonal- cell
structure of each comb is a product of the bees’ behavior. What is less obvi-
ous, though, is that the hollow tree or rock pile that shelters this intricate
nest is also part of the colony’s extended tool kit for survival. As we shall
see, although honey bees do not build their nesting sites, they do carefully
choose them, so the cavity that a colony occupies is also a product of its
members’ behavior.
The honey bee’s process of choosing a dwelling place unfolds during
colony reproduction (swarming), which occurs mainly in late spring and
early summer (May–July) in the Ithaca area. The first step in this house-
hunting process begins even before a swarm has left the parent nest. A few
hundred of a colony’s oldest bees, its foragers, cease collecting food and
turn instead to scouting for new living quarters. This requires a radical
switch in behavior. These bees no longer visit brightly lit, sweet- scented
sources of nectar and pollen; instead they investigate dark places—knot-
holes, cracks in tree limbs, gaps among roots, and crevices in rocks—
always seeking a snug cavity suitable for housing a honey bee colony.
Upon discovering a potential homesite, a scout spends nearly an hour
examining it closely. Her inspection consists of a few dozen trips inside
the cavity, each one lasting about one minute, alternating with trips out-
side. While outside, the scout scurries over the nest structure around the
entrance opening and performs slow, hovering flights all around the nest
site, apparently conducting a detailed visual inspection of the structure and
surrounding objects. While inside, the bee scrambles over the interior
surfaces, at first not venturing far inside the cavity, but with increasing
experience pressing deeper and deeper into the remote corners of the
hollow. When her examination is complete, the scout will have walked
some 50 meters (about 150 feet) or more inside the cavity and so will have
crossed all its inner surfaces. Experiments that I conducted with a cylindri-
cal nest box whose walls could be rotated freely while its bright entrance
opening remained stationary—so, in effect, the scout stepped onto a
treadmill when she went inside the nest box—have shown that a scout bee
judges a potential nest cavity’s volume by sensing the amount of walking
Seeley.indb 111
2/21/2019 8:07:34 AM
112 Chapter 5
required to circumnavigate it. Exactly how a scout bee judges a cavity’s
roominess using the information gained from the walking (and occasional
flying) movements that she makes inside the cavity remains a mystery.
The long du
ration of the nest- site inspections made by scout bees sug-
gested that they assess multiple properties of a site to judge its suitability.
Moreover, the regularities in certain nest- site properties that Roger Morse
and I had found—in entrance area, entrance height, cavity volume, and so
forth—supported the idea that bees have strong preferences in their hous-
ing. It was also possible, however, that the consistencies we found merely
reflected what was generally available in tree cavities. At this point, I
turned to surveying the scientific and beekeeping literature for informa-
tion about the nest- site preferences of honey bees, but I found almost
nothing, just one article in a French beekeeping magazine on how to build
attractive bait hives for catching wild swarms. This situation surprised me,
because I knew that beekeepers had worked for centuries to design the
perfect hive, and I figured they might have looked to the natural living
quarters of honey bee colonies for guidance, but evidently they had not.
At the same time, finding this gap in our knowledge delighted me, for I
realized then that my curiosity about the bees’ natural homes had drawn
me to a region of uncharted territory in the biology of Apis mellifera.
The method I developed for asking the bees about their nest- site prefer-
ences was simple: set out nest boxes that differed in certain properties and
see which ones were occupied preferentially by wild swarms. More spe-
cifically, I set out nest boxes in groups of two, three, or four, with the
boxes in each group identical except for one property, such as entrance
area or cavity volume. The boxes within each group were spaced about 10
meters (33 feet) apart on similar- size trees (or a pair of power- line poles)
where they were matched in their visibility, wind exposure, and location
(Fig. 5.6). Each group of boxes served to test one (potential) nest- site
preference, and it did so by giving swarms a choice between one box
whose properties all matched those of a typical nest site in nature (e.g.,
average entrance size, average cavity volume, etc.) and another box (or
boxes) identical to the first box except in one property (e.g., entrance
Seeley.indb 112
2/21/2019 8:07:34 AM
The Nest 113
Fig. 5.6. Two nest boxes mounted on power- line poles. The two boxes offer
identical nesting sites (same cavity volume and shape, same entrance height and
direction, etc.), except that the one on the right has a smaller entrance opening
(12.5 square centimeters/2 square inches) than the one on the left (75 square
centimeters/12 square inches).
size), the value of which was atypical. In this way, wild swarms were tested
for a preference in the one variable in which the boxes differed. For ex-
ample, to test whether the distribution of entrance areas shown in Figure
5.2 reflects a preference for small entrance openings, I set up pairs of cubi-
cal nest boxes that were identical except that one box had an entrance area
of 12.5 square centimeters (ca. 2 square inches), which was typical, and
the other box had a large entrance area of 75 square centimeters (ca. 12
square inches, the size found in a Langstroth hive), which was atypical.
Planning this investigation was easy, but executing it was hard. Alto-
gether, I built 252 nest boxes in the winter of 1975–1976, and I deployed
them in small groups over the countryside around Ithaca in the summers
of 1976 and 1977. Luckily, wild swarms were plentiful, so the experimen-
Seeley.indb 113
2/21/2019 8:07:35 AM
114 Chapter 5
Table 5.1. Nest- site properties for which honey bees do or do not show preferences,
based on nest- box occupations by swarms. A > B, denotes A preferred to B; A = B denotes no preference between A and B.
Property
Preference
Function(s)
Entrance size
12.5 cm2 > 75 cm2
Colony defense and thermoregulation
Entrance direction
south > north facing
Colony thermoregulation
Entrance height
5 m > 1 m
Colony defense
Entrance location
bottom > top of cavity
Colony thermoregulation
Entrance shape
None: circle = slit
(Both shapes work well)
Cavity volume
10 < 40 > 100 liters
Storage space for honey; colony
thermoregulation
Cavity shape
None: cubical = tall
(Both shapes work well)
Cavity dryness
None: wet = dry
(Bees can waterproof a leaky cavity)
Cavity draftiness
None: drafty = tight
(Bees can caulk cracks and holes)
Combs in cavity
with > without
Economy of comb construction
tal plan worked. My nest boxes attracted 124 swarms, enough to reveal
many of the bees’ secrets about what they seek in a homesite.
Table 5.1 summarizes the results of this study. We see that the bees in
these wild swarms revealed preferences in four aspects of the entrance
openings of their nest cavities. This was not surprising, given that this pas-
sageway is the interface between the colony and the rest of the world. All
of a colony’s food, water, resin, and fresh air comes in through this open-
ing; all its waste and debris goes out of it; and all attacks by predators will
focus on this point of greatest vulnerability. We also see that the wild
swarms expressed preferences about just two features of the cavity itself:
its volume and whether it was furnished with beeswax combs (from a
previous colony).
FUNCTIONS OF THE BEES’ HOUSING PREFERENCES
Entrance Size
I set up 14 test sites where swarms had a choice between two nest boxes
that differed only in entrance size, and at six of them a wild swarm occu-
pied one of the boxes, always the one with the smaller, 12.5- square- centimeter
Seeley.indb 114
2/21/2019 8:07:35 AM
The Nest 115
Fig. 5.7. The right half of a hive entrance (2 × 14 centimeters/0.8 × 5.5 inches)
that was reduced in late summer when the bees built a propolis wall over most
of the opening, leaving just the two small passageways shown. The one on the left
is barely wide enough for three bees to crawl through simultaneously.
(2- square- inch) entrance opening. This was unsurprising, because a small
entrance helps a colony defend itself against animals that want to steal its
honey. Most beekeepers around Ithaca, for example, know to reduce the
entrance openings of their hives in the autumn, especially once the frosts
have destroyed the flowers, and the yellow jacket wasps ( Vespula spp.), now
starving, try desperately to get to the bees’ stores of honey. A small en-
trance probably also helps a colony stay warm in winter by minimizing
draftiness in its nest cavity. This may also explain why some colonies, in
late autumn, will reduce their nest’s entrance by closing off most of the
opening with a propolis wall pierced by just a few openings, each one just
large enough to allow pass
age of a bee or two (Fig. 5.7).
Entrance Direction
Test sites where the paired nest boxes faced southeast, south, or southwest
had markedly higher probabilities of occupation than those where the
boxes faced northwest, north, or northeast. Evidently, the bees prefer
their nest entrances to have a southerly exposure. A study conducted by
Seeley.indb 115
2/21/2019 8:07:35 AM
116 Chapter 5
Tibor I. Szabo in Alberta, Canada, found that a hive with a south- facing
entrance, relative to one with a north- facing entrance, had a lower prob-
ability of becoming plugged by ice and snow in winter, and that having the
hive entrance open all winter improved nest ventilation and colony health.
A south- facing entrance probably also helps the bees by providing a solar-
heated site from which they can take off to conduct cleansing flights—to
eliminate accumulated body wastes—on mild days in midwinter.
Entrance Height
I established eight test sites with paired nest boxes that were either high
and low, and I caught six swarms at them, all in the high box of the pair.
This indicated a clear preference for nest entrances high off the ground,
consistent with the distribution of entrance heights for the nests of wild
colonies located by bee hunting (Fig. 5.2). In chapter 10, I will describe a
natural experiment in the Arnot Forest that shows one way that a wild
colony benefits from having its nest entrance high up: it lowers the risk of
detection by bears. It may also reduce the likelihood of nest damage in
winter by woodland rodents such as deer mice ( Peromyscus maniculatus).
Entrance Location
This variable was tested by setting out 12 pairs of nest boxes. Both boxes
in each pair provided a cavity that was 100 centimeters (ca. 40 inches) tall
and 20 centimeters (ca. 8 inches) wide and deep. But one box had its en-
trance opening at floor level, while the other had it flush with the ceiling.
Ten of these 12 pairs of nest boxes attracted a swarm: eight moved into
the box with a bottom entrance, and two into the box with a top entrance.
In chapter 9, we will look at a study by Derek M. Mitchell, an engineer
and physicist, that sheds light on the benefits to the bees of not having an
opening near the top of their nest cavity.
Cavity Volume