Page Banner

United States Department of Agriculture

Agricultural Research Service

1 - Index Page (scroll down for more information)
2 - A USDA-ARS Project to Evaluate Resistance to
3 - An Importation of Potentially Varroa
4 - Evaluations of the Varroa-resistance of
5 - Resistance to the Parasitic Mite Varroa
6 - Multi-State Field Trials: Varroa Response
7 - Multi-State Field Trials: Honey Production
8 - Multi-State Field Trials: Acarapis Response
9 - The Release of ARS Russian Honey Bees
10 - Hygienic Behavior by Honey Bees from
11 - Well Groomed Bees Resist Tracheal Mites
12 - Well Groomed Bees Resist Tracheal Mites (1998)
13 - Suppression of Mite Reproduction (SMR Trait)
14 - Varroa jacobsoni Reproduction
15 - Population Measurements
16 - The SMR/VSH trait explained by hygienic behavior of adult bees
Suppression of Mite Reproduction (SMR Trait)

Breeding Honey Bees that Suppress Mite Reproduction

Although acaricides control varroa mites in colonies of honey bees, use of chemicals endangers bees and hive products.  Bee breeders strive to reduce exposure of bees to chemicals by developing stocks of bees that resist the mites.  Towards that goal we began breeding bees for resistance to varroa mites more than five years ago.  Our project focused on finding varroa-resistance in honey bees from the U.S.

Initially, we found no bees that could survive varroa infestation without chemical control.  Short field tests (Figure 1) were used to carefully measure growth of bee and mite populations in colonies that had genetically different queens.  We defined resistance as the ability of a colony of bees to significantly limit growth of mite populations below the average colony.  In any group of colonies, there is considerable variation in the rate of growth of mite populations.  We hoped that small genetic differences between colonies of bees mediated differences in growth of mite populations. 

We needed lines of bees that consistently and predictably limited the growth of varroa mite populations before identifying genetic traits related to resistance.  Our strategy was to use queens from colonies of bees that significantly limited mite growth as breeder queens.  Virgin queens and drones were raised from several different breeder queens.  Then various combinations of drones and queens were made using instrumental insemination to control the matings (Figure 2).  The newly inseminated queens were tested for varroa resistance in short field tests during the following season.  The best queens were again chosen as breeders.  The entire process was repeated through several generations until the ability to limit growth of varroa mite populations had been enhanced. 

Because selection for varroa-resistance was based on overall mite growth, we knew little about the mechanism of resistance.  All colonies started a field trial with the same mix of bees and mites.  The only known differences were the test queens.  We measured characteristics known to be associated with varroa-resistance (e.g. hygiene, grooming, reduced postcapping period, etc.) from all colonies during field trials.  Then we searched for those traits that correlated best with the mite populations at the end of a test.    

Collecting and mixing bees in bulk package

Step 1:  A field test begins by collecting and mixing 50-70 lbs of bees from colonies that are not resistant to varroa mites.  We choose colonies that have substantial populations of the varroa mites.

Adding bees to innoculation package

Step 2:  The large mass of bees and mites is subdivided into smaller 500 gram units.  Scoops of bees are added to pre-weighed cages.  The cages are weighed again after bees are added to get an accurate estimate of the weight of the bees.

Innoculation cages placed in colony

Step 3:  Each test colony is given two cages of bees, a test queen, 4 combs and a feeder.  The cages are paired so that each colony receives about 1 kg of mite-infested bees.  The colonies remain closed (screen over entrances) for two days to minimize drift between colonies.

Full frame of bees

Step 4:  Bee and varroa mite populations grow during an 80-115 day period when populations of bees and mites are estimated from each colony (see Measuring Mite Populations).  We select colonies with the lowest mite growth as breeder queens.

Figure 1Our typical set up of a field test.  We try to ensure that all colonies are similar at start of the test.  Typically, we use 25 colonies in a short test, and each colony begins the trial with ca. 1 kg of worker bees, 400 mites, and a queen bee.  We measure mite and bee populations at the beginning and at the end of the test.

Although several characters predicted mite growth, the percentage of non-reproducing mites (%NR) correlated the most strongly with mite population growth.  Female varroa mites reproduce within the capped brood cells of the honey bee (see Varroa Reproduction).  Non-reproducing mites are those that enter brood cells to reproduce and either do not lay eggs, or if they do lay eggs, none of the daughters can mature before the adult bee leaves the brood cell.   We define several types of non-reproducing mites:  (a) mites that die before laying eggs, (b) live mites that do not lay eggs, (c) mites that produce only a son, (d) mites that produce progeny that die before reaching adulthood, and (e) mites that produce progeny so late in the development cycle of the bee that they do not have enough time to reach adulthood. 

Figure 2A queen being instrumentally inseminated.  This technique allows the bee breeder to control the genetics of colonies of bees.  It is very difficult to control the natural mating process of bees, which occurs high in the air. Instrumental insemination of a queen bee


We open capped brood cells containing tan-colored pupae (Figure 3) to measure %NR.  Usually, a varroa mite lays all of her eggs by this stage of bee development (see Varroa Reproduction).  We evaluate 30 singly-infested brood cells from a colony to determine %NR.  We decide a mite's reproductive success by identifying the sex and maturation of her offspring.  If the numbers of daughters and their development are considered normal, the mother mite is normally reproductive.  However, if her oldest female daughters are under-developed and will not reach adulthood in the remaining time of the host pupa's development, or if there are no progeny, the mother mite is non-reproductive. 

Brood cells containing tan-colored pupae

Figure 3Brood cells containing tan-colored pupae are examined to measure the percentage of non-reproducing mites.  Non-reproducing mites do not produce mature daughters before the host bee leaves the brood cell.  We examine mite families from 30 singly-infested cells per colony.  Mite families are evaluated for the numbers and maturity of the female offspring to decide if the mother mite will succeed in producing at least 1 mature daughter during the time remaining in the metamorphic development of the bee pupa.  

Nearly all colonies of bees have some non-reproducing mites.  In several tests, mite growth was lowest in colonies with the highest %NR.  We changed our selection strategy to concentrate on the %NR rather than on overall growth of the mite population.  We knew that a genetic character in bees was somehow causing mites to become non-reproductive in some colonies.  We call the trait 'suppression of mite reproduction' (SMR).  Although we classify non-reproductive mites into 5 categories, only 2 of them are consistently associated with resistance to varroa in our resistant bees.  These two categories are (1) living mites that do not lay eggs, and (2) dead mites that had laid no eggs.  The dead mites are unusual because most of them are entrapped by the cocoon (a condition rarely seen in control, or susceptible colonies of bees).

We do not know exactly what causes mites to become entrapped by the cocoon or to simply not lay eggs upon entering a brood cell.  These symptoms of abnormal mite reproduction become apparent only after 4-6 weeks of placing a queen with the SMR trait into a colony of bees.  This delayed suppression of mite reproduction is called SMRD.  A second type of mite suppression occurs within the first brood cycle of placing a queen into a colony of bees.  The acronym for this immediate suppression of mite reproduction is SMRI.  Although we have seen both types of suppression of mite reproduction, our work has focused on SMRD.

Experiments suggest that daughter mites raised in colonies with the SMRD trait are the mites affected by the bees (and not the original mites that start an experiment).  The mites that do not lay eggs had low numbers of stored sperm in their spermathecae when compared to mites that reproduced normally (Figure 4).  More than half of these mites had no sperm at all.  Currently, we do not know if the low sperm counts result from lack of mating between mites (see Varroa Reproduction), or if the sperm transferred by males are non-viable and do not reach the spermatheca within the female.

An intact spermatheca removed from an adult female varroa mite (as seen under a compound microscope).

a.  An intact spermatheca removed from an adult female varroa mite (as seen under a compound microscope).

Ruptured spermatheca and released sperm for counting.

b.  Gentle pressure will rupture the spermatheca and release the sperm for counting.

A single spermatozoon separated from the group of sperm

c.  A single spermatozoon separated from the group of sperm.

Living mites that lay no eggs (blue bar) had lower numbers of stored sperm when compared to mites that produced normal families (green bar) (15-28 mites were examined per bar)

Figure 4Living mites that lay no eggs (blue bar) had lower numbers of stored sperm when compared to mites that produced normal families (green bar) (15-28 mites were examined per bar).    

In the early years, the ?best? or most resistant colony had 35--40% non-reproducing mites, which was only slightly better than the 10-25% non-reproducing mites found in colonies of unselected or susceptible bees.  Now, we routinely produce inbred resistant colonies that contain 60--100% non-reproducing mites (Figure 5).  

Unselected or control bees

Unselected or control bees

 Bees with the SMR trait

Bees with the SMR trait

Figure 5 Reproductive status of mites from susceptible and varroa-resistant colonies of bees after 50 days in a short field test.  Resistant bees were produced from queens selected for the suppression of mite reproduction delayed (SMRD).  Each pie graph represents 230-300 mites sampled from 10 different colonies.  Key: entrapped (red) = mites that died by being entrapped by the cocoon; no eggs (blue) = living mites that laid no eggs; other (yellow) = other infertile mites; and normal (green) = reproductive mites that will likely produce at least 1 mature daughter. 

We now have varroa-resistant stocks of bees inbred for the SMR trait, and these colonies greatly limit mite growth.  The U.S. queen rearing industry is geared toward the production of naturally mated queens, which makes the production of commercial inbred resistant queens very unlikely (unless queens are mated in an isolated area such as an island).  However,  queen producers can readily produce hybrid queens.  We found mite growth to be intermediate between resistant bees and susceptible bees when resistant queens are free-mated with susceptible drones (Figure 6).  Although colonies with hybrid queens (resistant x control) had intermediate populations of mites, they had half the mites found in the susceptible controls. Hence, even hybrid queens should provide beekeepers a tangible level of resistance.


Final mite populations

Figure 6Final mite populations for 57 colonies having 3 types of queens after a 115 day field test (mean above each bar).  Resistant queens were inbred for the SMRD trait (red).  Resistant x Control were queens having the SMRD trait free-mated to drones that did not have the trait (yellow).  Control queens were queens that lacked the SMRD trait mated to drones that also lacked the trait (blue).  All colonies began the test with 0.9 kg of bees and about 650 mites.


Reference to full articles

Harris J. W. and J. R. Harbo (2000) Changes in reproduction of Varroa destructor after honey bee queens were exchanged between resistant and susceptible colonies.  Apidologie 31: 689-699.

Harbo J. R. and J. W. Harris (1999) Selecting honey bees for resistance to Varroa.  Apidologie 30: 183-196.

Harbo J. R. and J. W. Harris (1999) Heritability in honey bees (Hymenoptera: Apidae) of characteristics associated with resistance to Varroa jacobsoni (Mesostigmata: Varroidae). Journal of Economic Entomology 92 (2): 261-265.

Harris J. W. and J. R. Harbo (1999)  Low sperm counts and reduced fecundity of mites in colonies of honey bees (Hymenoptera: Apidae) that are resistant to Varroa jacobsoni (Mesostigmata: Varroidae). Journal of Economic Entomology 92 (1): 83-90.

Harbo J. R. and R. A. Hoopingarner (1997) Honey bees (Hymenoptera: Apidae) in the United States that express resistance to Varroa jacobsoni (Mesostigmata: Varroidae).  Journal of Economic Entomology 90: 893-898.

<< Previous    1     2     3     4     5     6     7     8     9     10     11     12     [13]     14     15     16     Next >>

Last Modified: 3/26/2014
Footer Content Back to Top of Page