Figure 11.3 Operation of social effectors for social homeostasis of honeybee hives. a: Distributions of bees on a comb in three different environmental temperatures. Temperatures indicate the environmental temperature. [After Wilson (1971)] b: Functional distribution of bees in a typical huddle, showing the different locales within the huddle of bees engaging in the indicated behaviors.

Figure 11.3 Operation of social effectors for social homeostasis of honeybee hives. a: Distributions of bees on a comb in three different environmental temperatures. Temperatures indicate the environmental temperature. [After Wilson (1971)] b: Functional distribution of bees in a typical huddle, showing the different locales within the huddle of bees engaging in the indicated behaviors.

flight muscles, generating the bulk of the heat produced by the cluster. At the same time, another group of bees, located at the outer margin of the cluster, adopt postures that interweave the chitinous hairs of each bee with its neighbors, forming a kind of downy coat that helps insulate the cluster. This social differentiation into shivering and down-weaving bees cannot be explained as organismal homeostasis en masse. Social homeostasis, then, exists—objective and sensible criteria can be used to identify it, and it seems clearly to operate at a social level of organization, at least among honey bees.

The crux of the problem for social homeostasis then becomes the deeper question: does similarity in the outward signs of homeostasis imply similarity of function? Individual bees undoubtedly carry around within them the neural machinery for the negative feedback control of their own body temperatures. There must exist within each bee sensors for body temperature, some neural machinery to process this information, and effectors that will alter the bee's heat budget. Social homeostasis, however, implies that there are feedback loops that extend beyond the boundaries of the individual bees—meta-loops, if you will, that control and coordinate the individuals in the colony. While it is easy to accept the conventional view of neurally based negative feedback controllers operating within each bee, the idea of meta-loops governing social homeostasis is more difficult to comprehend.

In fact, social homeostasis often operates through novel mechanisms of feedback control, some of which do not involve negative feedback at all. Interestingly, many of these mechanisms involve structures built by the colony inhabitants, structures which harness and transform both metabolic energy and physical energy in the environment to power the external physiology that social homeostasis requires.

The following discussion of social homeostasis will focus on regulation of the nest atmosphere. The air inside most social insect nests is more humid, richer in carbon dioxide, and poorer in oxygen than the outside air. These concentration differences drive fluxes of oxygen into the nest and fluxes of carbon dioxide and water vapor out. The nest atmosphere appears to be under the colony's adaptive control, because its composition is frequently steady, despite substantial variation in the flux rates for these gases.

The Little Big Bang: Positive Feedback Meta-Loops Let us turn first to an example of social homeostasis that does not involve a negative feedback operation in the organisms. The mechanism I shall describe is, in fact, common among social insects. It involves an explosive (figuratively, of course) mobilization of workers in response to a perturbation of the colony—what I call the "Little Big Bang."

Anyone foolhardy or unfortunate enough to have broken into a nest of bees or wasps has experienced the Little Big Bang in one of its more dramatic forms: the defensive swarm that explodes in a furious frenzy from the disturbed nest. In a bee hive, the frenzy is set off when one or a few bees experience conditions that elicit an alarm reflex, which includes the emission of an alarm signal. Often, the alarm signal is a volatile chemical, an alarm pheromone, released from specialized glands in the worker's body.4 When another bee senses the alarm pheromone, this triggers an alarm reflex in it, which initiates the reflex in yet others, and so forth. Thus, an alarm signal from a single perturbed worker bee can spread rapidly through all the members of the colony, eliciting the frantic and aggressive behavior that soon turns into a defensive swarm. You will recognize this as positive feedback, which we have already encountered in the context of diffusion-limited accretion growth of sponges and corals.

4. In some cases, the alarm pheromone is a single chemical. Among bees, for example, the alarm pheromone is isoamyl acetate, released from the worker bee's sting pouch. In other cases, the alarm pheromone is a complicated chemical cocktail. Among ants, alarm pheromone is released from glands around the head, the mandibular glands, which emit a mixture of terpenes, hydrocarbons, and ketones, some or all of which serve as alarm pheromones.

Positive feedbacks turn up frequently in physiological systems, because they are very handy switching mechanisms, and organisms often need to switch their physiology from one function to another. Physiology dominated by negative feedback, which resists change, would be a stable but boring place. For example, at some point in an organism's life cycle, it must switch its use of energy from maintenance to reproduction. If the use of energy for maintenance is governed by negative feedback, something must either negate, oppose, or switch off these systems so that the energy can go elsewhere. Frequently, that something is activation of a positive feedback loop.

The Little Big Bang, being a social response, involves positive feedback meta-loops. The curious thing about the Little Big Bang is that it is often employed for social homeostasis, as exemplified by the defensive response of termites to a breach in the structure that houses their colony. In Africa and Asia, termites often build above-ground structures, or mounds, to house and protect their colonies. These termites must often deal with holes made in their mounds, either by predators, such as aardvark and aardwolf, or by inclement weather, such as torrential summer thundershowers. Breaching the mound wall is roughly the termite's equivalent of a broken window in an air-conditioned building. In the rooms near the broken window, the building environment is perturbed by the outside air, and people who sit closest to the broken window feel the greatest change. People down the hall or on another floor may not even be aware that a window has been broken somewhere in their building. The air in a termite colony is similarly affected by a hole in the mound wall. In the immediate vicinity of the breach, the air is perturbed by locally steep gradients in the partial pressures of oxygen, carbon dioxide, and water vapor.

One way to deal with a broken window in an air-conditioned building is to isolate the break somehow—close the door to the office containing the broken window, for example. Most likely, the person to do this would be someone sitting close to the broken window, since that person would feel the perturbation most strongly. If you add to this scenario an interaction between office workers, such that the sight of one person getting up to close the door to the office with the broken window elicits other office workers to begin closing off other nearby doors, or the halls, or the stairwell doors, then you have something akin to the response of worker termites to a breach of their mound wall. If a termite encounters the locally perturbed atmosphere around a breach—such as anomalously high pO2, low pCO2, low humidity, or breezy conditions—it responds with an alarm reflex. First, it will grab a handy grain of sand and cement it into place with a drop of gluey secretions from its mouth—this is the daubing reflex. Second, it releases a chemical alarm pheromone that wafts through the rest of the nest. Finally, disturbed workers tap their heads rapidly against the mound's walls, sending a sound vibration spreading through the mound like a telegraph.

For about the first ten minutes following the breach, the activity in the vicinity of the opening is fairly haphazard. Termites stumble into the perturbed environment around the breach almost by chance, and when they do, their alarm reflexes are triggered. There is little evidence of coordinated activity at this point: daubs of cement are placed seemingly at random around the breach. After a few minutes, the number of workers at the breach begins to increase slowly, and in about ten minutes the first termites recruited by the initial alarm signal start to arrive. These too deposit their grains of sand and emit their own alarm signals. Then the number of workers at the breach increases rapidly. Eventually, so many recruits appear that the initially haphazard building activity merges into a marvelous construction project, with teams of termites erecting pillars, walls, and galleries. After about an hour or two the breach is sealed off.

What I have described here is social homeostasis—a perturbation of the colony atmosphere elicits a response that returns the colony atmosphere to its state prior to the perturbation. It sounds like classical negative feedback, but it is not. Rather, it involves two positive feedback meta-loops: a "fast" loop that mediates a rapid response between individual workers;and a "slow" loop that involves the colony as a whole. Let us take each in turn.

The fast meta-loop involves the daubing reflex. When another worker termite comes along and detects the residue of a previous termite's daubing reflex, a daubing reflex is elicited in it. This has been dubbed stigmergy (from the Latin stigma, "sign" or "mark," + ergon, "work," literally, "driven by the mark"). It is a form of positive feedback, and it results after time and repetition in the construction of pillars and walls (Fig. 11.4). Stigmergy alone is not sufficient for the Little Big Bang, though. To set it off, the fast response— stigmergy—must be coupled to a slower positive feedback meta-loop that drives the recruitment of new workers to come to the site of the breach. The speed of the recruitment response is limited by how rapidly the alarm signal can be transmitted from the termites initially encountering the breach to new termites elsewhere in the nest and how quickly these new termites can come to the site. In termite colonies, this is a relatively slow process, because the spread of any chemical signal may require minutes to spread throughout the nest. Termites are not very fast runners, so they require some time to make their way to the site of the breach. Once set in motion, though, every termite that is recruited to the breach itself emits alarm signals, and the recruitment increases rapidly in the manner characteristic of the Little Big Bang. The result is a large mobilization of energy (in the form of the work done by the workers) that keeps on increasing until the perturbing influence is literally overwhelmed. Only then does the intensity of the "slow" loop begin to diminish and recruitment decline.

Pushmi-Pullyu Ventilation: Manipulating Colony-Scale Energy Gradients

Honeybees sometimes meet their colony's gas exchange needs by actively ventilating the hive. Workers positioned at the hive's entrance beat their wings to create a fanning effect. The mechanism, so seemingly

Figure 11.4 Stigmergyand positive feedback. When a termite encounters either a hole or a fresh daubing, it sends out an alarm signal (!) and contributes to the plug. At right, the fresh daubings, shown from the side, pile up to form a new column or wall.

simple, in fact involves a complicated interplay between multiple gradients of potential energy in the hive, some of which are metabolic and some of which are physical. The architecture of the hive plays an important role in this interplay.

This mechanism of social homeostasis takes its name from the famous pushmi-pullyu of Dr. Dolittle. The pushmi-pullyu, to remind you, was an extraordinary camel-like creature with two front halves, each facing in opposite directions, but with no hind quar ters. Normally, the pushmi-pullyu coordinated its two front halves very nicely: one watched while the other slept, one talked while the other ate, and so forth. But the pushmi-pullyu could get confused about which way was forward. If both heads were equally determined to go in the direction each thought was forward, the animal would go nowhere. Only if one head was more determined than the other could the pushmi-pullyu move.

Social homeostasis of the atmosphere of honeybee hives involves a process I call pushmi-pullyu ventilation. In short, pushmi-pullyu ventilation works like an on-off switch for gas exchange in the hive. Just as a heater can regulate the temperature of a room by being switched alternately on and off, so too can pushmi-pullyu ventilation of a hive regulate the hive's atmosphere. The phenomenon itself will require some explaining, but before I do so, let us first lay some background about bees and their hives.

Before beekeepers came along, honeybees commonly housed their colonies in hollow cavities in trees. These cavities are naturally isolated from the outside world, but if a bee colony takes up residence in one, the worker bees typically isolate it even more, sealing off all openings to the cavity, save one, with "bee-glue" or propolis, a very hard and resinous wax. Commonly, the one remaining opening is located below the colony, and it serves as a port of egress and entry for the worker bees shuttling back and forth between the hive and their sources of food. It also serves as an exchange port for the hive's respiratory gases: oxygen, carbon dioxide, and water vapor.

Oxygen and carbon dioxide must move through the hive's entrance hole at a rate adequate for the colony's needs. Each gas's rate of movement and rate of consumption must be balanced, as an organism's physiological state is balanced, according to equation 11.2. The flow of oxygen, for example, through the hive involves two processes: the consumption of oxygen by the bees (MOi, with units like grams per second or milliliters per minute) and an exchange of oxygen (QOi with similar units) across the hive entrance. Note that we are treating the hive environment now as functionally equivalent to the internal environment of an organism. In the terminology used in equation 11.2, MOl is a physiological flux (PF) while QOl is a thermo-dynamically favored flux (TFF). Homeostasis requires the two to match:

To some degree, oxygen will flow passively into the hive simply as a consequence of the colony's consumption of oxygen. As the colony consumes oxygen, the pO2 in the hive atmosphere will drop, and the partial pressure difference thereby established drives oxygen passively across the hive entrance. Thus, the TFF of oxygen can be represented by an equation similar to Fick's law:5

where pO2o, and pO2h are the partial pressures of oxygen outside and inside the hive, respectively, and Kp is the passive conductance for oxygen. For our purposes, we shall assume that passive conductance is determined by the size and shape of the hive's entrance hole. This is intuitively easy to grasp: oxygen moves through larger holes faster than through smaller holes.

A bee colony that relies on passive movements of gas alone cannot develop social homeostasis. Consider, for example, what must happen if the colony's demand for oxygen increases, as it might if the outside temperature cooled. Equation 11.5 tells us there are two ways this increased demand can be met.

First, increased oxygen consumption could be supported by an increase in the partial pressure difference driving oxygen across the hive entrance (pO2o -pO2h). Because the partial pressure of oxygen in the atmosphere is essentially fixed, an increased partial pressure difference can only come about through a drop of the hive's pO2. That result would be the antithesis of homeostasis. Alternatively, the bees could increase the hive's passive conductance (Kp), which could increase flux without any change of the partial pressure difference. This also is problematic. Any alteration of the hive's conductance would require structural modification of the hive entrance: widening it, or shortening it, or some combination of the two. Bees do this, obviously, but it is a lot of work, and it is suitable only for long-term or chronic alterations of metabolism, not for the minute-by-minute modulation of flux that social homeostasis would presumably require.

Social homeostasis of the hive atmosphere must therefore circumvent the constraints of equation 11.5. Bee colonies do so by a social behavior: stationing workers at the hive entrance where they fan their wings in place. The fanning adds a component of forced convection to the oxygen's flux, supplementing the passive exchange rate with an active ventilatory component, QV. Because the ventilatory flux depends upon the number of bees fanning (nb) and how energetically each individual fans (qV), the flux of oxygen can be modulated minute-by-minute simply by altering the number and activity of bees recruited to do the fanning:

Q02 = Q02TFF + QV = Kp(p°2o - p°2h) + nbqv(pO2o - ?°2h)

5. This equation is similar to Fick's law because Fick's law technically refers to a flux driven by the mechanism of diffusion. In the example, I am being deliberately agnostic about just what the mechanism of exchange is; my point is just that it occurs by some passive mechanism, which may partially involve diffusion but may be supplemented by other mechanisms, like convection induced by winds blowing past the hole. If such processes behave similarly to diffusion, we say the flux is governed by the Fick principle.

If fanning could be modulated by a meta-loop coupled to oxygen demand, then variation of the hive's oxygen consumption could be supported with no disruption in the gas composition of the hive atmosphere: in short, social homeostasis.

At least in honeybee colonies, we know that such a meta-loop exists. For bees, the signal for recruitment to fanning is the hive's pCO2. This can be demonstrated by the colony's response to imposed perturbations of the hive atmosphere. The hive's pCO2 can be increased artificially by adding CO2 to the hive from a gas cylinder, independently of any variation of the colony's oxygen consumption or metabolic rate. This is akin to the wet paper towel trick I described earlier: altering the local environment independently of the effectors intended to control it. As CO2 is added to the hive, workers begin to show up at the hive entrance and fan in place until the hive pCO2 declines, even though the "production rate" of carbon dioxide is artificially elevated. The greater the perturbation of hive pCO2, the more workers will be recruited to fan. The hive pCO2 will then return to its pre-disturbed value, even though the CO2 flux has been increased.

The Metabolic Meta-Loop

While the meta-loop linking fanning behavior to hive pCO2 is one of the components of pushmi-pullyu homeostasis, it is not the whole mechanism. The colony's metabolism, in addition to its effects on the hive atmosphere, also elicits other potential energy gradients within the hive. These gradients interact with the hive's architecture to promote still other feedback meta-loops. It is these that are at the heart of pushmi-pullyu ventilation, so let us understand what they are and how they arise.

Metabolism of glucose liberates a considerable amount of heat, which warms the hive air. In addition, the consumption of oxygen and its replacement with carbon dioxide and water vapor reduces the average molecular weight of the air in the hive.6 Both reduce the density of air in the hive, and gravity therefore im

6. Oxygen has a molecular weight of about 32 g mol- , and it is replaced by equal quantities of carbon dioxide and water vapor. Carbon dioxide, with a molecular weight of 44 g mol-1 is somewhat heavier than the oxygen it replaces, but its addition to the hive's atmosphere is more than offset by the addition of the much lighter water vapor (molecular weight 18 g mol-1). The overall change is a reduction in the average "molecular weight" of air inside the hive.

parts buoyant forces to it. As the colony metabolizes, therefore, spent air will be lofted upward to the top of the hive (Fig. 11.5a).

What happens next depends upon the configuration and number of holes that open into the hive. Let us look first at the arrangement of holes common to many commercial bee hives: two openings, one at the bottom, which serves as the normal entry and egress port for the hive, and one at the top, which can be opened or closed by the beekeeper. In a hive with this configuration, the buoyant energy imparted to the air will do work: it will drive a natural convection of air through the hive, lofting the hive air out through the top hole and drawing a steady flow of fresh air in through the hive entrance (Fig. 11.5b). This mechanism is known as thermosiphon ventilation.

A hive so constructed will be self-regulatory to a degree, because the physiological flux (the rate of oxygen consumption) is physically coupled to the flux rate of oxygen into the hive. The physical coupling arises, of course, because of the link between colony metabolism and the hive's air density. Increased levels of oxygen consumption will dissipate more heat from the bees' bodies to the hive air, magnifying the metabolism-induced change of air density, which will, in turn, drive faster rates of ventilation, ultimately increasing oxygen flux into the hive. Indeed, this is why commercial hives have two holes. The self-regulatory tendency of a two-hole hive, supplemented by the beekeeper's control over the upper hole, spares bees from the need to drive regulatory ventilation themselves. And if bees are relieved of the chore of fanning, they can spend their energy gathering pollen, making honey, or maintaining the hive—increasing the colony's productivity, in other words.

The picture is very different in a wild hive with the more common natural configuration of a single entrance hole near the bottom. Because there is no hole at the top of the hive, the buoyant forces imparted to the spent air cannot do ventilatory work, and the energy in the buoyant forces accumulates at the top. To ventilate the nest, fanning bees must do work against buoyant force diffusion gas exchange r bulk flow' in bulk flow out

Figure 11.5 Effects of hive architecture on respiratory gas exchange in a metabolizing honeybee colony. a: In a closed hive, the metabolism-induced buoyant forces (indicated by upward-pointing vectors) drive spent air up to the upper portions of the cavity, resulting in gradients of pCO2, pO2, and pH2O from the top to the bottom of the cavity (indicated by shading). b: In a hive with an opening at the top, the metabolism-induced buoyant forces drive a bulk flow of air through the hive.

this tendency of hive air to rise upward. This they do by orienting their bodies with their heads pointing inward so that they drive hive air outward through the hive entrance. What emerges from this balance of forces is a dynamic movement of air that is at the heart of pushmi-pullyu ventilation.

Pushmi-pullyu ventilation arises because the bees recruited to fanning duty do not fan continuously. Rather, the activity of the fanning bees is synchronized somehow so that they all fan for a time, then all stop fanning for a while. During the period of active fanning, the bees power a bulk flow of air outward. In so doing, they draw down the stratum of spent air to a level below where its buoyant energy alone would take it (Fig. 11.6). This is akin to storing energy in a spring—the spent air moves downward because the fanning bees do work against the buoyant forces driving it upward. When the fanning stops, this stored energy is now available to drive a bulk flow of air upward, which draws a bolus of fresh air inward through the hive entrance. The alternating phases of synchronized fanning (pushmi) and relaxation (pullyu) result in a tidal flow of air across the hive entrance, similar to the alternating cycle of exhalation and inhalation that characterizes our own breathing (Fig. 11.7). The combination of cycle time between fanning and not fanning and recruitment of workers to fanning results in very sensitive control of the hive atmosphere.

Social homeostasis in this case arises because of two meta-loops that govern the behavior of the bees in the colony. One we have already discussed—the metaloop responsible for recruitment of worker bees to fanning duty. An additional meta-loop operates just among the bees that are actually on fanning duty, the interaction that controls the synchronization of fan-

Figure 11.6 "Pushmi-pullyu" ventilation of a honeybee hive.

Was this article helpful?

0 0
Oplan Termites

Oplan Termites

You Might Start Missing Your Termites After Kickin'em Out. After All, They Have Been Your Roommates For Quite A While. Enraged With How The Termites Have Eaten Up Your Antique Furniture? Can't Wait To Have Them Exterminated Completely From The Face Of The Earth? Fret Not. We Will Tell You How To Get Rid Of Them From Your House At Least. If Not From The Face The Earth.

Get My Free Ebook

Post a comment