Bee Culture - September, 1986


"This paper is timely and is needed. It is refreshing to read a discourse on the Africanized bee that thoughtfully examines the bee as it is rather than observing superficially and parroting the usual dire predictions of calamity when the bee arrives in the United States.
"The paper is unusually well written. Conclusions are based on discriminating observations and thorough analyses, and recommendations are positive and practical.
"It is an excellent paper."
- Harry H. Laidlaw

From September 18 to October 2, 1985, it was our good fortune to participate in multifaceted studies of the pollination of cacao (cocoa and chocolate) in Costa Rica. One element of these studies involved the movement of honey bee colonies into a cacao plantation to determine the incidence of honey bee foraging among numerous cacao varieties (genotypes). The ten colonies we used were moved from a mid-elevation (approx. 900 m.) locality to a humid lowland locality (50 m.) within the Caribbean coastal floodplain, of eastern Costa Rica. This and other activities gave us the opportunity to gain experience with "Africanized" bees. "Africanized" bees (AB) have been in Costa Rica for approximately two years and the country has been declared "Africanized". Nevertheless, numerous colonies of "European" bees (EB) there remain intact, particularly where they are well managed.

The following are our observations, impressions and conclusions regarding the AB in Costa Rica. These are based on our experiences with the ten colonies used in our studies provided by Dr. William Ramirez, Department of Entomology, University of Costa Rica, observation of other colonies in Dr. Ramirez's apiary on the University of Costa Rica campus, observations of several colonies in each of three apiaries maintained by Marla Spivak, a student conducting research on AB and EB in Costa Rica, several colonies in the apiary of a Costa Rican beekeeper and observation of a single feral colony. All except the feral colony were in one, two or three (deep) storied Langstroth hives. Our ten test colonies, from an apiary with both AB and EB colonies, were initially headed by "Africanized" queens that were believed (by Dr. Ramirez) to have been twice superseded before we arrived: The supersedure queens probably mated with both "Africanized" and "European" drones in his campus apiary where EB colonies predominated. These colonies were as gentle as, or more gentle than, many of our coloines now at Madison. Their behavior in no way affected our pollination study.

To be sure, our exposure to AB was brief. However, because of our collective experience, we believe that we are in a position to provide unique interpretation of our observations. Particularly noteworthy in this regard we point out that one of us (A.M.Y.) is a population biologist while another (B.J.E.) has extensive experience in the biochemistry of insect behavior. The third (E.H.E.) is a honey bee behaviorist, well acquainted with the behavior of many of the honey bee phenotypes first bred by Dr. C. L. Farrar and subsequently by Dr. F. E. Moeller and Mr. E. R. Harp at Madison.

We do not pretend to be experts on "Africanized" bees nor do we believe that all of our observations are unique. We do base our comments on our knowledge of bee behavior and on our practical and research experience. We believe that some of our observations will contribute to a better understanding of the problem of "Africanized" bees. Finally, we reluctantly use the terms "Africanized" and "European" and do so only because they are so widely used that to do otherwise would lessen our ability to communicate herein (see Conclusions).

Lest we be misunderstood, we point out at the outset that bees clearly identifiable as "Africanized" can be irascible, difficult to work with and certainly undesirable from the standpoint of the commercial beekeeper as well as that of the average citizen. However, some desirable behavioral traits were noted even in the most irascible colonies and some AB colonies were quite gentle. The AB we observed had no behavioral traits that haven't been seen over the past 15 years in the honey bee breeding program at Madison (see below). Moreover, the behavior of AB is highly variable between colonies of "pure" AB as well as those of AB/EB crosses. It is obvious that AB interbreed with EB thus producing colonies with intermediate behavioral traits: Some colonies, identified as AB/EB crosses appeared more irascible than colonies identified as fully AB. (Note: The term "Africanized" implies crossing between ABlEB populations.) Finally, it occurs to us that most apiculturists have been preoccupied with demonstrating differences between AB and EB, but with limited success. Far more obvious and readily documentable are the similarities between AB and EB because, after all, they are all honey bees.

Behavior of Managed Colonies

- Foraging Behavior

"Africanized" worker bees enter and leave the hive much more rapidly than do their European sisters; hence, we presume, their reputation for fast flight. AB appear to fly directly to and often into the hive entrance, entering without utilizing a landing board, while returning EB fly somewhat lazily about the hive, then alight at the entrance and walk in. Bees in our test colonies flew to and from the hive entrance in a more direct fashion than EB but less so than AB thus indicating that these colonies were indeed the result of an AB/EB cross.

Pollen gathering behavior among our test colonies was assessed by: 1. rate of adaptation to foraging in an unfamiliar environment; 2. net quantities of pollen collected; 3. diversity of pollen types gathered; and 4. variability in gathered pollen diversity between colonies. All test colonies appeared comparable to EB colonies at Madison for each criterion. The results of these studies will be summarized elsewhere.

- Nesting Behavior

Cell size in combs constructed by AB colonies is said to be smaller than that of EB. Costa Rican officials measure cell size to determine whether the colony is "Africanized" (4.7-4.9 mm for 10 linear cells = AB; 5.0mm and above for 10 linear cells = EB). Cells from our test colonies (all drawn older comb) measured 5.1 mm for 10 cells. We note that traditionally, beekeepers have selected strains for increased size of their queens - a characteristic reflected also in their worker progeny.

It was pointed out to us by Dr. Ramirez and Marla Spivak that, unlike some (but not all) EB, AB tend to fill entire frames with brood more so than do EB. Honey is stored primarily above the brood nest. AB pollen stores in brood comb are minimal but AB's fill frames with pollen when it is available. One AB colony had 9 deep frames of brood and 10 frames of pollen. Unlike EB, communication spaces (about 2 cm between wax and frame) are evident in AB colonies at the sides and bottom of each brood frame. The extensive AB brood rearing results in rapid expansion of the population.

In this latter regard, it is important to note that over the last 40-50 years certain queen breeders in the United States have been successful in selecting strains of bees specifically adapted to the needs of the northern beekeeper who kills colonies in the fall. This beekeeper desires a strain of bees that builds up rapidly in the Spring and stores honey above the brood nest, This enables the beekeeper to extract 100% of the season's honey production easily and without contamination of the product with larval remains, the empty equipment is subsequently stored until the following spring. These specially bred strains mimic AB in this behavior.

We noted that the excessive use of smoke is undesirable with AB as it is with EB.

- Defensive Behavior

We were most impressed with the complexity of the defensive behavior of AB, not for the reasons one might suppose but because all except two of the discrete defensive behavior subtraits described below were collectively evident in the AB colonies observed. We have seen each of these subtraits at Madison on more than one occasson - usually independent of one another in a given colony or occasionally in various combinations within a colony. Other beekeepers have undoubtedly made similar observations. We and others at Madison have classified these subtraits of defensive behavior as follows:

A. "Defenseless". It is possible to select honey bees that are almost entirely defenseless. At Madison we selected and maintained, for several years, a strain of bees which, when disturbed, would not defend the colony in any way. Upon opening the hive and even after extreme and repeated disturbance of the colony, these bees would retreat into the hive and remain motionless. All that would be visible were bees, lined up at the top edges of the frames, looking out. We often demonstrated this behavior to visitors to the amazement of most. (Note: we did not see this trait in Costa Rica.)

B. "Biters". Workers of certain genotypes use their mandibles to bite at and pull hairs on exposed areas such as the head, arms and hands. When seen as an independent trait these bees rarely sting unless overtly incited.

C. "Looking for Just the Right Place". This may be an extension of the biting syndrome; However, it is identifiably different and may or may not be accompanied by biting. Here, worker bees land and buzz fiercely for an extended period while crawling over the skin. If given enough time these bees will sting, but the experienced beekeeper knows that these bees are easily brushed away long before they sting.

D. "Rear End First". Defenders of select colonies may dart out to meet their victim and sting so quickly that one gets the incorrect impression that these bees approach rear end first and sting before their feet touch down. This behavior is clearly the opposite of that described in C above. These bees are generally deemed undesirable as such behavior reduces beekeeper working efficiency.

E. "Jumpy". Worker bees are described as jumpy when disturbed, as upon opening of the colony. These bees dart up and out, usually landing upon the beekeeper or a nearby object. Such bees rarely sting unless overtly incited.

F. "Intimidation". Here, we refer to those bees that dart out from the colony entrance or from an opened colony and repeatedly bump into subjects that they apparently perceive as threatening, such as a beekeeper. Bees exhibiting this behavior do not normally sting without further provocation. Anthropomorphically, one would describe their behavior as that of attempting to provoke a response. Certainly, if an unprotected person blinks or swats at these bees they will sting. Bees from some colonies may follow a retreating beekeeper for a considerable distance. This may be an extension of jumpiness.

G. "Runny". Runny bees are relatively well-known to experienced beekeepers. Colonies are described as runny when the bees run excitedly over the combs and hive interior after the colony is opened. Excess smoke aggravates this behavior. Queens are usually difficult to find in such colonies, hence this behavior has an undesirable economic impact. Sometimes this condition is so pronounced that the bees will run around inside the colony in waves with directional flow resembling a race. We have seen colonies of runny bees that seldom sting unless the colony's disturbed for prolonged periods of time. Other such colonies will sting readily, this latter behavior may reflect the presence of other elements such as D above.

We found it noteworthy that the most intractable AB colonies that we observed in Costa Rica possess all of the above defensive traits except A and C. While AB almost simultaneously jump, bite and intimidate extensively, the careful beekeeper can avoid AB stings. However, once stung, multiple stings can be expected shortly thereafter: Successive stinging at or near the site of an initial sting is characteristic of most bees but pronounced in AB. Here both environmental and genetic components to defensive behavior are suggested.

We were permitted to open one AB colony and search for the queen so that the colony could be requeened. The bees were "runny". Once the queen was caged (all but two frames were examined before she was found) the colony was closed. We noted that the majority of disturbed/airborne bees settled down and quickly reentered the colony (as do EB) even though that typical, ever present cadre of AB intimidators followed us several hundred yards back to our vehicle. We were forced to drive away in order to lose the last of these bees.

Feral Colony Observations

We made brief observations on a feral colony of AB within an abandoned section of the cacao plantation. This colony, in a fallen hollow tree, was no more than 100 yards from several homes that lacked screened windows and doors. Costa Ricans from these houses have harvested honey from this colony with minimal protection and reported that the bees had a nasty sting. Five of us walked about the clearing where this exposed colony was located: One of us moved to within 2-3 feet of it, photographed it and retrieved a piece of comb from beneath it without arousing the bees.


Since the beginning, when primative man first attempted to keep primitive honey bees, beekeepers have been selecting and breeding strains of bees with particular desirable attributes such as size, gentleness, manageability, productivity, nonswarming, wintering ability, disease resistance, etc. This process of selection and breeding over the ages, continues even today. The result is that existing domestic strains of honey bees, as kept by beekeepers, are the products of extensive artificial selection much as are dairy and beef cattle, poultry, horses and even domestic dogs and cats. On the other hand, feral (wild) populations in many parts of the world probably resemble their ancestral progenitors. Presumably, honey bees in many parts of Africa have undergone far less artificial selection and therefore are probably more akin to naturally selected (wild) bees. It is possible, as some have suggested, that the "tapping" of wild bees (a practice which destroys the colony) may have selected towards greater aggressiveness as the more docile colonies are most frequently robbed (destroyed). It seems obvious that there exists a continuum among honey bee populations wherein domesticated (artificially selected and largely European) bees are found at one extreme, while wild (primarly naturally selected African and other) bees are at the other.

It is clear to us, from our observations, the observations of others as well as our knowledge of honey bee behavior and genetics, that the so-called "Africanized" bee is simply a mixture of races and behavioral traits. The term "Africanized" is an unfortunate choice in that it represents typological thinking - it causes one to think that this "Killer Bee" (an even more ludicrous synonym) is a discrete genetic entity. Hence, we perceive a widespread misinterpretation of the "Africanized" bee problem and how to solve it. The "Africanized" bee appears to be a predominantly (but not entirely) wild, naturally selected, phenotype while the "European" bee is predominantly the domestic, artificially selected, phenotype to which we have become accustomed. Phenotype is, by definition, the result of the effects of environment and heritability. The combined effects of these two factors are still confounded in ongoing studies.

"European" bees, those kept by beekeepers in the Americas, are usually
described as a mixture of races (and genes). It is further accepted by most experts that "African" bees in South America outcross with "European" bees there. Interestingly however, the progeny of this crossbreeding are then called "Africanized" - no longer a variable mixture of races (albiet somewhat expanded) even though only a single race (African) with some admittedly less desirable traits has been added to the genetic base. We think it equally or more appropriate to say that the African bees became Europeanized. The popular concept of what constitutes an "Africanized" bee appears to be way out of line with reality. Even the accepted terminology (e.g. "African" "Africanized") is misleading and should, if possible, be abandoned. Our observations lead us to conclude that the AB phenomenon is little more than an extended, over-popularized, case of gene introgression and selection - an evolutionary process well-documented in many other organisms under natural conditions. The introduction of African bees into South America represents the sudden breakdown of an isolating barrier (the Atlantic ocean), providing the impetus for a unique glimpse of how gene introgression works in honey' bees.

The "migration" of "Africanized" bees northward through the Americas may be a somewhat unique event of biological interest. However, in our view, the process of "Africanization" is not the invasion of a single race of bees, rather it is the intrusion of a naturally selected, heterogeneous, highly adapted feral (wild) gene pool into homogenous populations of domestic honey bees that are the product of extensive artificial selection and therefore vulnerable to the pressures of natural selection leading to reversion to the wild type. Typological thinking (in terms of an "Africanized" type) has lead to a hasty acceptance of the universality of AB behavior and therefore discourages additional research directed at finding variability within and among populations of honey bees in South and Central America. An understanding of the process of "Africanization" may be eluding us because of this philosophical handicap.

In Part II we will present our ideas for the development of management strategies for "Africanized" bees.


(1) U.S. Department of Agriculture, Agricoltural Research Service, Bee Research Unit, Department of Entomology, University of Wisconsin, Madison, Wisconsin 53706.
(2) Department of Entomology, University of Wisconsin. Madison, Wisconsin 53706.
(3) Invertebrate Zoology Section, Milwaukee Public Museum, Milwaukee, Wisconsin 53233.