Beekeeping in Cuba

Text of the article

Cuba is known for having the world’s largest population of European Apis mellifera that are resistant to the varroa mite.1 This is often lauded as the perfect example of treatment free beekeeping leading to the problem solving itself. Let us examine how they actually arrived where they are.

First let us picture this place. Cuba is an oblong island about half as big as Victoria, with about 160% the population. That’s 3.5 x our state population density altogether, but as you know Victoria has a lot of relatively empty space. The population density of Cuba is half that of the Mornington Peninsula which I think gives a better sense of it.

For comparison, here is Victoria in true scale comparison

to Cuba, the number of houses represent the distribution of the population, giving a sense both of the comparative

population and distribution. In this scenario western

Victoria would have much better beaches.

The climate of Cuba matters particularly when talking about comparative effects of varroosis. Climatically it most resembles coastal Queensland between Mackay and Rockhampton – subtropical, wet/dry seasons, high humidity. The coldest parts of the year would only get to around 12°C overnight and there’d probably never be a seasonal brood break.

European honey bees were first imported to Cuba from Florida on May 2nd, 1763,2, 3, * and, critically, the Africanized honey bees that have become pervasive in South, Central and southern North America have never arrived in Cuba.4 Honey is now (or was in 2009 anyway) Cuba’s second most valuable agricultural export.5

* Though there are references to now-lost memoirs that they may have been brought over as early as 1750 (Bande 2001).

Only the Cuban Beekeeping Enterprise (Empresa Apicola Cubana) AKA Apicuba is legally permitted to buy honey from beekeepers in Cuba.6 The communist economy of Cuba also results in interesting passages like

“Under the current conditions of Cuban beekeeping, visiting and treating on two successive occasions with an interval of 7–10 days the 500–700 colonies of a Basic Unit of Cooperative Production (UBPC) requires a fuel consumption that could be beyond the means of the enterprise and the allocations determined by MINAGRI and the country for this activity, even if it is economically viable.”7

MINAGRI in this context is the Ministry of Agriculture (going by a particularly 1984-esque contraction). As of 2015 there were 2000 beekeepers in Cuba running 220,000 hives,1 organized into 60 UBPCs, 20 Agricultural Production Cooperatives, and 1,100 Credit and Service Cooperatives. Producers were paid the equivalent of AUD$645-1095 per ton by Apicuba, the honey then being exported to Europe for as much as AUD$4,200 per ton (all figures 2015 conversion rates).8

Varroa was detected in Cuba in April 1996,9 in three apiaries in Matanzas province, though, in a now familiar pattern, subsequent delineation found it already widespread in the surrounding area10 and it is believed to have arrived a few years earlier, possibly from illegal queen imports.1 Movement of bees between major regions was prohibited,1 but it quickly spread throughout the country causing national annual losses directly attributed to varroa of approximately 4%.11

Between December 1996 and February 1997 another Cuban researcher found 27 Alcohol Wash Equivalent (AWE) (9%), as well as 36% of worker brood infested, and 74% of drone brood.12 In 1998, 88 hives studied had an average infestation of adult bees of 21 AWE.13

The chief veterinary officer of ApiCuba lists the management strategy in 2001 in a manner that sounds much like our own integrated pest management pyramid: “control of the epizootiological structure of the species [ie “Cultural Methods”]; training of producers and technical personnel; genetic improvement in search of tolerant bees; biotechnical measures with the use of trap comb and chemical treatments with synthetic and organic products”

I’ll quote extensively from said chief veterinary officer’s 2001 report because it gives a real good picture of what they were looking at 5 years after varroa arrived:

“Controlling drone breeding by different methods has proven too laborious when trying to avoid chemical treatments (Calis et al., 1999b), while chemicals, including those applied on plastic supports, are so far the most effective and easiest to apply means, despite their known drawbacks.

(Continued on page 19)

Beekeeping In Cuba, cont’d

(Continued from page 18)

Since 1996, the chemical product Bayvarol has been used in Cuba, maintaining a control efficacy of over 98% to date, with no residues detected in 134 honey samples analyzed by the Chemical Residue Monitoring Plan in 2000 and 2001 (Cuba, 2001b,c). The possibility of alternating chemical treatments with essential oils, organic acids or hormonal substances, of more variable effectiveness but less contaminating, they are necessary options while waiting for varroa-tolerant bees.

Cuba's planned Control Program bases its strategy on integrated pest management, with the shortterm goal of using only organic substances. In countries such as the United States, the sole registration of Apistan made the development of chemical resistance foreseeable (Sanford, 1993), which was found to be widespread years later (Eischen et al. 1998) and resulted in the lost opportunity to rationally alternate pyrethroids with organic treatments.

The above is from the introduction for a study on using the thymol product Apilife Var. The study was performed in 1998, at which point the Varroa-positive hives along the coast had all been treated previously with Bayvarol and the inland hives had only just recently gotten Varroa and not previously been treated. In the first field roll-out of Apilife Var, nearly 80% of the hives were killed outright, and it’s unclear what went wrong (an initial average infestation rate of 22% (equivalent to alc wash of 66), may have already been too high), but a second roll-out was more successful with no colony losses and good control achieved, and the study ended up recommending the use of the product either as the sole treatment or in rotation with synthetics.13

In another paper published in 2001, researchers at the Apicultural Experimental Station in Havana looked at mats from 41 hives that had been collected daily for one week, during which 2099 mites had fallen (7.3

(Continued on page 20)

Beekeeping In Cuba, cont’d

(Continued from page 19)

mites/colony/day), and found 79.6% of them showed biting damage from bees.14 This, if true, would be a really interesting finding as it would indicate the European bees of Cuba had very very quickly developed this hygienic practice.

A separate 2003 study (7 years post Varroa arrival) looked at ten hives at an apiary of the Agrarian University in Havana that had last been treated with Flumethrin 10-12 months prior, with the queens of poorly performing hives having been regularly replaced. They found 92-92.5% of the dropped mites to show damage from bees, and removal of dead brood via the freezekilled-brood-assay averaged 82% (12 of the 20 samples being over the 95% removal recommended for hygienic behavior).15

In 2007, Dr JL Sanabria of the Agrarian University of Havana, Cuba, writes “In Cuba, a treatment with Bayvarol (flumethrin) is applied annually to control parasite populations. However, Cobey (2000) states that routine treatment in colonies minimizes the association of pests and diseases by delaying the bees' ability to develop their own survival defense mechanisms,” which both foreshadows the resistance idea Cuba would later become famous for, and I can’t resist pointing out that both Tocal’s Liz Frost and I studied under that same referenced Sue Cobey around the time this 2007 paper was written.

This experiment looked at 13 hives which had survived without treatment for already three years at that time, since August 2002 (six years after arrival of varroa). These hives were monitored for a year. During this time the number of hives dropped from 13 to 11 and back to 12, which as we know such fluctuations are normal. The average mites per adult bee was 3.61% (appx 11 Alc Wash Equivalent (AWE), and only ranging from 6.3-16.7 AWE). They also looked at percentage of worker and drone brood pupae that were infested, and found 4.89% of the worker brood and 34.94% of the drone brood examined to be infested on average. Infestation rates of hives receiving regular synthetic treatments in Cuba “more recently” (writing in 2007 but appears to reference 2004 studies) were at similar levels.16

Demedio again (author of some of the earlier cited papers), writing in 2009 updates us on the situation now 13 years on: Chemical products are no longer the primary means of controlling honeybee varroosis in Cuba, but they remain a key component of an integrated pest management (IPM) system that has proven effective after 12 years of combating the Varroa destructor mite. One of the factors that has allowed for the rational use of these products, without any reported chemoresistance or detected residues in Cu-

* although Varroa had been in the States for 22 years at that point so I’m not sure what the five years refers to.

+ And the citation for this ill-fitting fact is rather oddly the 2020 paper “Nosema ceranae and RNA viruses in honey bee populations of Cuba,” which does not mention acaracides. (Rod 2020)

ban honey to date, has been the strict adherence to regulations governing the import and use of such substances. The primary link in ensuring this compliance is the Central Pesticide Registry, a state entity that governs the procedures for authorized institutions to conduct the necessary evaluations required to issue permits for their use. One such institution is the Agrarian University of Havana, which, in coordination with other entities, has carried out field efficacy trials of several of these products, including Bayvarol (flumethrin), Apivar and Cuban Apizel strips (amitraz), Apilife VAR [Thymol], and CheckMite+ (coumaphos), the latter still in development. This paper presents the results of these tests and the technical requirements to be met in order to obtain the results, which are submitted to an Expert Committee and either accepted or rejected by the Regulatory Authority.

[…] Although initially, and despite its high cost, Cuba had to resort to applying chemical treatments indiscriminately, the need for an integrated management system to make beekeeping sustainable was always considered a key strategy.

And he draws a timely comparison to how the US was faring at that same time:

A recent [to 2009] and cautionary example, though not the only one, is that after five years of fighting varroosis,* the only legally approved treatment in the US was Apistan® (fluvalinate), which had been administered once a year with good results. However, in Florida, this was later increased to two treatments (spring and fall), combined with diagnosis every 30-90 days. Sanford (1993) asserted that it was a mistake to have only one product registered for the treatment, and its consequence was the emergence of chemoresistance in numerous states after 10-12 years (Eischen et al., 1998; Ellis, 2001).5

The above Demedio 2009 “The Registration of Products for Control of Varroosis of the Honey Bee in Cuba” lists the quoted products (Bayvarol, Apivar, Apizel, Apilife VAR, and Chickmite+), and recommends their use --not without the cautions, but nonetheless—“The Cuban experience has allowed for the timely and orderly evaluation and registration of synthetic and organic chemical means that today constitute legalized options to be integrated into the IPM system for varroosis and that has allowed for a significant reduction in dependence on these means.” Another paper17 notes that in the previous three years 156,000 treatments valued at €462,000.00 had been imported to Cuba. And so, when we turn to the next paper we’ll consider, the 2022 “Recapping and Mite Removal Behaviour in Cuba: Home to the World’s Largest Population of Varroa Resistant European Honeybees” that put Cuba on the map, so to speak, for its notable resistance to varroa, and it says “no mite treatments have been administered for over 20 years”.+1 I simply don’t see how that statement

(Continued on page 21)

Beekeeping In Cuba, cont’d

(Continued from page 20)

can be reconciled with the 2009 paper.

The mentioned 2022 recapping paper (of which our recent conference speaker Stephen Martin is the 5th author), looked at 37 hives in December 2021, and found 13% worker brood infestation and 40% drone brood infestation. In March 2022, looking at only the ten hives at the CIAPI apiary (more on this acronym below), the worker brood infestation was found to be 23%. Recapping behaviour was found in only 33% of noninfested cells but in 72% of infested cells (ie the bees are successfully targeting varroa) (comparably, they found 2% recapping in 15 varroa naïve colonies from Australia, UK and Hawaii). Another mystery of life lies in this quote: “Although, in this study no significant difference was found in the reproduction of varroa in recapped or non-recapped cells supporting the findings of two previous studies. Therefore, recapping may be playing a minor role in resistance.”

They say extraordinary claims require extraordinary evidence, and it seems 37 hives isn’t a lot to base assertions on about Cuba’s 220,000 hives which have made them all now the topic throughout the world. Especially when one considers these 37 hives may well have been taken from the best apiaries of the state breeding

program we’ll look at next. The Cuban state-organized breeding program actually began before varroa arrived, in 1982, identifying the hives that produced the most honey while being tolerant to diseases. This achieved hives producing an average of 73-94 kg of honey with low incidence of bacterial diseases.6

Once varroa arrived in 1996, the breeding program focused on prioritization of the “selection and reproduction of the most productive colonies despite Varroa infestation” while managing hives with an integrated pest management strategy that this paper maintains did not use acaricides. Seven years after the detection of Varroa, the number of hives in Cuba finally began to increase again in 2003.6

In 2013 the Cuban Bee Selection Improvement and Breeding Program (BEE-CSIB) was created by the Centro de Investigaciones Apıcolas (CIAPI) (prior to that there was no structured plan beyond a culture of doing so among the beekeepers).18 Bee-CSIB evaluated 4039 colonies, finding an average of 74.3% dead-brood -removal, with 39.5% scoring over 80%.6

The breeding program focused on only reproducing from the most productive hives (this, and not varroa hygienic traits, appears to have been the primary criteria), with hives to be evaluated on performance across

(Continued on page 22)

Beekeeping In Cuba, cont’d

(Continued from page 21)

at least one full harvest season (six months), each participating province (11 of Cuba’s 15 provinces) would run a self contained program without human-mediated genetic exchange between provinces. 52 breeding stations were operated and in addition to directly breeding bees they engaged to train beekeepers in queen breeding.6

The first generation of selection began with each province’s top beekeepers’ best hives, which were relocated to the program apiaries. They were then carefully monitored for six months for hygienic behavior. Varroa levels were monitored in hives selected for breeding but was not part of the selection criteria. Hives that exhibited excessive swarming, queen mortality, disease, weakness or high aggressivity were excluded. The selection program aimed to conform to the recommendations of Buckler et al 2013 “Standard methods for breeding and selection of Apis mellifera queens.”6

Essentially hives had to meet hygiene benchmarks and from all those that met them further selection was based on greatest honey production. 20-40 colonies per province were selected to be bred from. The best colony was used to produce 100-200 drone mother colonies at the breeding centers while eggs were grafted from the other top hives. Subsequent generations were selected from this stock.6

Between 2014 and 2023 1,191 breeding colonies were selected from 4,857 candidates, creating six generations of queens. Though as mentioned varroa resistance wasn’t an emphasized trait, the median infestation rate fell from 13.2 AWE (4.4%) to 2.7 AWE (0.9%) by the fifth generation. Interestingly, while the text of Perez 2025 says they achieved an upward trend of honey production, the provided figures don’t actually support that assessment, at best it fluctuated. They do note that hives of provinces involved in the breeding program produced more honey on average (45.9 kg / hive) than hives in provinces not participating (32.5kg / hive), but don’t give us comparative production prior to the programme commencing so it’s entirely plausible that the most productive provinces participated and these differences already existed. It is noted that the apparent reduction of honey production during the latter part of the study may be explained by Covid-era movement restrictions.6

Average honey production of hives headed by queens

So we’re still wondering when they banned the use of acaricides. Let’s go back to earlier papers. As of 2012 we’re still seeing statements by Yoandra Valle, director of the state honey company, Apicuba, that keeping varroa levels low “could be helped by the use of Integrated Pest Management, which includes the use of … and varroicide treatments if the average [AWE] reaches or exceeds 15.”19

So I just asked them. Jorge Demedio, author of the much cited 2001 paper was kind enough to respond to me. He does say (translated from Spanish like most of the papers here) “I can assure you that while there were some initial applications of Bayvarol, and to a lesser extent, Apivar, none have been applied for over 20 years.” (Demedio pers. corr.)20

“The fact that our beekeeping industry was integrated into a large national company fostered uniformity and mutual support throughout this work. On a more specific technical level, the high degree of professionalism among beekeepers and the existence of some 80 queen bee breeding centers, supported by a comprehensive training system, facilitated several key actions, such as: epidemiological control, with the identification of outbreaks, surveillance areas, and threatened zones; clinical and laboratory diagnosis; traceability of hive components and products; and selection of the most productive hives, with lower infestation rates and higher levels of hygienic practices.”20

He also added: “Like all the beekeepers I've met, we form a global "guild" of good people, sensitive and passion-

of the first six generations of the Cuban Bee-CSIB queen breeding program.

ate about bees and, consequently, about nature. When, in 2000, I sent photos of the mites I had collected for my thesis and requested the assistance of Dr. Dennis Anderson at CSIRO, I will never forget the speed and kindness with which he responded and the courtesy of sending me a copy of his publication, which, among other things, recognized the new mite species, Varroa destructor Anderson and Trueman, as the cause of the major problem worldwide. In my response to you, I am trying to reciprocate the kindness shown to me by your compatriot 26 years ago.”20 So what is to be concluded in the end? Well the endlessly repeated statement that “they never used treatments in Cuba” is plainly not true. There appears to have been wide-

(Continued on page 23)

This is the plain-text version, extracted from the printed pages for readers who use screen readers, text-only browsers, or prefer reflowable text. Captions and sidebars appear where they fell in the page layout. The typeset pages are available in the reader.