The African (Brazilian) Bee Problem

ABJ, October, 1974 – Page 368-372

Plant Protection Institute, Agric. Res. Service
U.S. Department of Agriculture
Beltsville, Md. 20705


In June 1972, a final report (Michener, 1972) was made public on the findings of the Committee on the African Honey Bee. This Committee, which consisted of nine members, had traveled to Brazil for a firsthand study of the “African bee” (Brazilian bee). The committee operated under the Division of Biology and Agriculture, National Reasearch Council, National Academy of Sciences, Washington, D.C.

Because of this report and of influence by American beekeepers and other groups, there is an increasing demand for further investigations into the matter of the “Brazilian bee.”


According to the committee’s report, (p. 10) “Coloration is highly variable and is not considered by the committee to be a good characteristic for identifying the Brazilian bee.” Again, variation could be expected for the reason mentioned above.

In other words, since there are no known differences in the appearance between races, one cannot identify the Brazilian bee by sight.


It has been said that “Workers are usually slightly smaller than those of European bees. Queens also average slightly smaller (199 mg versus 208 mg for European bees).” Apparently past research groups did not weigh or measure any bees while in Brazil. One could then question these statements of size differences, since no statistical evidence for or against size difference exists.

Cell Size

Past study contends that natural worker cells for Brazilian bees average smaller than those of European bees. One research team made 33 measurements across 25 adjacent cells in combs from the northern states of Brazil and 22 combs from southern states. These ranges along with others are given below in Table 1.


From the data presented in this table, it is obvious that the cell size of worker cells measured on newly drawn comb varies greatly from region to region in this hemisphere. From the meager data in Table 1, one wonders if cell size may vary from small to large, progressing from a warmer to a cooler climate, as indicated. If this is true, the difference could be environmentally and not genetically related.

Measuring the diameter of a honey bee cell is not such a simple task as one might first suspect. For an insight into the problems involved, one should read the paper by Taber and Owens (1970) which delves into this subject in great detail. This matter of cell size is extremely important for it is one of the characteristics presented that may offer some means of identifying the “Brazilian bee” and certainly deserves further investigation.

Other Problems

Since no anatomical differences have yet been discovered, one must question many of the statements and observations made in the name of the Brazilian bee, many of which are shown in Table 2 below.


As one can observe, many of the claims are unsubstantiated and inconclusive. For example, in Item 4 concerning the acceptance of standard comb foundation, Wiese (1970) stated that the African bees do not accept standard foundation, whereas 17 of 18 beekeepers surveyed by the committee said they do accept standard foundation. Item 7 concerning honey production – there is no hard data to indicate either an increase or decrease in production. Various persons state one thing, while others state the contrary. Item 9, Wiese (1970) says the Brazilian bees are more susceptible to EFB. Other research in this area could find no evidence to support this statement, nor does Wiese present any data to substantiate the claim of increased susceptibility, this statement of “more susceptible to EFB” appears constantly in the literature, remains unchallenged, and is practically considered fact.

One of the main themes of past research is the aggressive characteristic of the Brazilian bee. In every instance, when claims and counterclaims are made to this point, one must ask himself, how the observer knew it was the Brazilian bee he was observing if he could not identify it by sight? It would appear that if bees are aggressive, they are Brazilian bees; if they are gentle, they are European.

There are a couple of items in the table that may give a clue concerning this point. Item 40 indicates that aggressiveness had decreased; however, it is the opinion of the committee that beekeepers are now experienced and protected adequately when they go into apiaries.

When one reads of Brazilian beekeepers working bees in their barefeet, in shorts, with neither veil nor gloves, one may wonder if perhaps this is why statements of aggressiveness have circulated.

The problem of reproductive swarming has been ascribed to the Brazilian bee as being common. However, the committee reports various beekeepers stating that bees rarely abscond or swarm if given adequate space, food, and water.

The committee also may have answered another question arising from reports of stings (see Item 31 in Table 2) resulting in death. The committee observed that this may occur because of “rash efforts to destroy hives in hysteria arising from over publicity about the danger.”

History of African Bee Importations

African bees (Apis mellifera adansonii) have in the past, been introduced into the United States and Europe previously without instituting a problem. Morse et al (1973) gives a good history of those early introductions. In the late 1960’s and early 70’s Woyke (1973) introduced A. m. adansonii into Poland where he built up a total of 30 colonies. He reported that they were more aggressive than the Italian bee, but that there was no significant difference in aggressiveness between the pure African and a race (A.m. silvarum) from northeast Poland. He further stated that the African bees did not swarm and their honey production did not differ from domestic colonies. One colony was left out-of-doors during winter and it survived. It is interesting to note that Woyke makes no mention of these bees spreading over Poland and becoming a problem throughout Europe, which they apparently did not.

In the early 1960’s Steve Taber, then of the Apiculture Research Branch of the USDA at Baton Rouge, La., imported and expermented with semen of the African bee, which was sent to him by Dr. Kerr from Brazil. In a personal communication from Taber to me he states, “The African bee semen sent me by W. Kerr was used to successfully inseminate queens; their daughter queens that were 50% African were then inseminated. Two more generations of impotation and insemination made the African parentage at Baton Rouge over 90% African.” Again, here as in Poland these African bees presented no special problems.

Critique of Committee Report

After carefully studying past evidence, certain questions arise in my mind. Is the Brazilian bee problem of such great magnitude that it deserves so much of our attention? Are not most of the characteristics attributed to this bee really the result of poor management? Have not many of the behavioral characteristics of Brazilian bees been reported as casual observations by non-scientists, their statements over publicized, unchallenged, and as a result, treated as fact?

Previously some scientists have tended to agree with or go along with these unsubstantiated statements.

These investigators have concluded that when the Brazilian bee “invades” the United States, the following will happen; (1) Regarding deaths to domestic animals, “There is no reason to believe that the same thing would not happen in North America. (2) The impact of Brazilian bees on the pollination services provided by beekeepers would probably be severe. (3) Hobby beekeepers would no doubt abandon their beekeeping operations. (4) The characteristics of the Brazilian bee would probably intensify the problem of apiary locations and seriously suppress beekeeping operations. (5) The number of available apiary locations would be reduced. (6) Rental costs for apiary sites would increase. (7) Greater numbers of hives might be needed to achieve pollination objectives. (8) Colonies may not be moved as frequently. (9) Honey production may drop. (10) Costs may rise. (11) The package bee industry might be seriously threatened. (12) Recruiting new persons in beekeeping could become more difficult. And finally, (13) “Any subbstantial change in the total pollination picture is likely to affect the overall ecology of the area. Major shifts in pollinator populations may occur if the Brazilian bee reached North America, and this could cause far reaching effects on plant and animal populations.”

After reviewing past inquiry into this subject, I cannot agree with many of these conclusions, or some of the recommendations as outlined below.

What would U.S. beekeepers do if Brazilian authorities proposed doing these same things to our bees?
1. Set up quarantines.
2. Destroy swarms and absconding colonies as they migrate across Central America by placing there a strip of trap hives containing attractive pheromones and toxicants.
3. Use aerial sprays of chemical insecticides.
4. Use poison bait stations.
5. Use bee diseases.
6. Flood the barrier area with desirable bees to change gene frequencies.
7. Kill wild colonies.
8. Requeen aggressive colonies and those preferring to swarm.
9. Set up a bee breeding program beginning with stocks of various origins.
10. Set up research sites in the far north such as Alaska, in temperate Florianopolis and a branch station at Mafra.
11. Survey the existing diseases in the Brazilian bee

The proposals as listed above would cost hundreds and hundreds of thousands of American dollars. Beekeepers must realize that if they push for and support such proposals, the money will come out of the funds of some research agency. These funds may not be additional funds from some source in the sky, but probably will be shifted from other projects related to agriculture. In light of the above, one must judge how severe the “Brazilian” bee problem is and whether or not it deserves any funding.

One of the committee’s recommendations was to develop methods to permit identification of the Brazilian bee. Recently the USDA has made a grant of $10,000 to the University of California to find new ways to distinguish the African bee from the Italian bees. I firmly agree with the objective of this research grant. Before we start spending hundreds of thousands of dollars setting up labs to study all facets of this bee problem, let us first determine if there is in fact a “Brazilian” bee problem.


Dr. Roger Morse et al., in a recent and very informative article in Bee World (1973) put forth a few interesting points concerning the introduction of African bees into this country. “(1) in many places other than Africa there have been, and there are, “vicious” bees; (2) some people prefer aggressive bees; (3) colonies with aggressive bees can be requeened so that the temperament of the bees is soon changed. This latter conclusion is most important for it is the obvious solution to the vicious bee problem. Not only does requeening change the temperament of the colony, but it does so very quickly, surely, and at relatively little cost.

The working individual with the most experience concerning this whole problem is the noted Brazilian scientist, Dr. Warwick Kerr. In the late sixties he also recommended on the basis of experimental evidence, that the aggressive queens be replaced by Italian queens.

(See Kerr 1968). Under his direction, 23,000 virgin A. m. liguista queens have been distributed to beekeepers in Brazil and this program has met with unqualified success.

Brazilian Scientists’ Comments on Committee Report


Camargo, J. M. F. and Goncalves, L. S. 1971. Manipulation procedures in the techniques of instrumental insemination of the queen honeybee Apis mellifera L. (Hymenoptera: Apidae). Apidologie 2, 239-246.

Cosenza, G. W. 1972. Estudo dos enrames de migracao de abelhas africanas. Proc. 1st Cong. Brasileiro de Apicultura. pp. 128-29.

Goncalves, L. S. 1970. Analise genetica do cruzamento e do torax. Tese de doutoramento apresentada a Fac. de Medicina de Ribeirao Preto, USP.

Guy, R. D. 1972. How does the Brazilian bee compare to South African bees. South African Bee Jour. 44, 9.

Kerr, W. E. 1968. Solacao e criar uma raca nova. O Apicultor 1, 7-10.

Kerr, W. E. 1971. Contribuicao a ecogenetica do algumas especies de abelhas. Ciencia e Cultura, 23, 89-90.

Kerr, W. E. and Bueno, D. 1970. Natural crossin between Apis mellifera adansonii and Apis mellifera ligustica. Evol. 24, 145-155.

Maa, T. 1953. An inquiry into the systematics of the tribus Apidini or honeybees. Treubia, 21, 525-640.

Melo, M. L. S. 1970. A qualitative analysis of the proteins in venoms from Apis mellifera (including Apis mellifera adansonii and Bombus atratus. J. Apic. Res. 9, 113-120.

Michener, C. D. [Chairman]. 1972. Final Report, Committee on the African Honey Bee. Div. of Biol. and Agric. N.R.C., National Acad. Sci., Wash., D.C., 95 pp.

Morse, D., Burgett, M., Ambrose, J. T., Conner, W. E. and Fell, R. D. 1973. Early introduction of African bees into Europe and the new world, Bee World 54, 57-60.

Taber, S. III and Owens, C. D. 1970. Colony founding and initial nest design of honey bees, Apis mellifera L. Anim. Behav. 18, 625-632.

Torchio, P. F. and Torchio, D. M. 1974. Larvae of the Apidae (Hymenoptera, Apoidea). Part I. Apini, Apis. Utah State Univ. Pub. (In press).

Wiese, H. 1970. Abelhas africanas, suas caracteristicas e tecnologia de manejo, In 1st Congresso Brasileiro de Apicultura, pp. 95-108.