APICULTURE ANDSOCIALINSECfS
Field Assays for Hygienic Behavior in Honey Bees (Hymenoptera: Apidae) MARLA SPIVAK ANDDANIELLE LAURA DOWNEyl Department
of Entomology, 219 Hodson Hall, University of Minnesota, St. Paul, MN 55108
J. Econ. Entomol. 91(1): 64-70 (1998) ABSTRACT Honey bee, Apis mellifera L., hygienic behavior is a mechanism of disease resistance and a mode of defense against the parasitic mite Vm-roajacobsoni Oudemans. Hygienic bees uncap and remove diseased and parasitized brood from the nest. The propagation of colonies that demonstrate resistance to chalkbrood and American foulbrood and that remove pupae infested by Varroa mites is becoming increasingly important in apiculture. This study evaluates 2 commonly used field assays used to screen colonies for hygienic behavior: the freeze-killed brood and the pierced brood assays. Both involve determining the time required for worker bees to remove dead capped brood from a section of comb. Colonies in the experiment displayed a wide range of removal rates and were grouped as hygienic, nonhygienic, or intermediate. The results of experiments 1 and 2 indicated that neither the age nor the source of the frozen brood had a significant effect on the removal rate by hygienic colonies (i.e., those colonies that consistently uncapped and removed freeze-killed brood within 48 h). In experiment 3, only a weak correlation was found between the removal of young freeze-killed and pierced pupae, but a significant correlation existed between the removal of pre-eclosion freezekilled and pierced pupae. Experiment 4 examined cues that elicit removal behavior by hygienic and non hygienic colonies. When pupae were pierced with an insect pin through the base of the cell (without piercing the wax cell capping), there was no difference in the number of pupae removed by the hygienic and nonhygienic colonies. On average, 30% of all pierced pupae survived the treatment, which considerably diminished the accuracy and reproducibility of the test. When pupae were treated with hemolymph extracted from either a live or freeze-killed pupa, there was also no difference in the rate of removal by hygienic and nonhygienic colonies. These results indicate that bees from nonhygienic lines can be induced to express hygienic behavior only if a sufficiently strong stimulus is present. Both hygienic and nonhygienic colonies removed significantly more pupae treated with hemolymph from a dead pupa than hemolymph from a live pupa, indicating that the cue that stimulates removal behavior is stronger in dead pupae. It is concluded that the freeze-killed brood assay is the most conservative and reliable screening procedure for hygienic behavior. The following procedures are recommended: Randomly selected comb sections (5 by 6 cm each) of capped brood should be cut from 1 healthy colony, frozen, and introduced into the test colonies. The assay should be repeated at least twice. Only colonies that remove >95% of freeze-killed brood within 48 h in both tests should be considered hygienic. When developing hygienic breeder stock, the hygienic colonies should be challenged with the American foulbrood or chalkbrood pathogen to ensure resistance. KEY WORDS
honey bee hygienic behavior, chalkbrood, Varroa
HYGIENICHONEYBEES,Apis mellifera L., have the ability to detect, uncap, and remove diseased brood from their nest before the causative organism reaches the infectious stage (Rothenbuhler 1964). Hygienic behavior is a mechanism of resistance to at least 2 diseases of honey bees: American foulbrood, caused by the bacterium Paenibacillus larvae subsp. larvae Heyndrickx et al. (Woodrow and Holst 1942, Rothenbuhler 1964); and chalkbrood caused by the fungus, Ascosphaera apis (Maassen ex Claussen) Olive & Spiltoir (Gilliam et al. 1983, 1988). Because honey bees reuse brood cells, diseased brood must I Department of BiologicalSciences, Simon Fraser University, Burnaby, BC, Canada V5AIS6.
eventually be removed from the nest. However, hygienic behavior occurs at a relatively low frequency (=10%) in honey bee colonies thus far surveyed (Spivak and Gilliam 1993, Oldroyd 1996). Hygienic behavior also is one mechanism of defense against the parasitic mite Van-oa jacobsoni Oudemans. The Asian bee Apis cerana F., the original host of Varma, removes mite-infested pupae from the nest, thus interrupting the life cycle of the mite (Peng et al. 1987, Rath and Drescher 1990). In European or western honey bees, some colonies of A. mellifera carnica Pollmann detected, uncapped, and removed mite-infested pupae (Boecking and Drescher 1991, 1992); colonies of A. m. ligustica Spinola specifically bred for hygienic behavior re-
0022-0493/98/0064-0070$02.00/0 © 1998EntomologicalSociety of America
February 1998
SPIVAK AND DOWNEY:
HYCIENIC
moved significantly more infested pupae than nonhygienic colonies (Spivak 1996). Because of the need to find lines of A. mellifera that tolerate Varroa infestation and that have reduced susceptibility to chalkbrood and American foulbrood, there has been a renewed interest in hygienic behavior by researchers and apiculturists. The most commonly used field assay for hygienic behavior involves recording the amount of time for bee colonies to detect, uncap, and remove brood from a comb section 5 by 6 cm containing =100 larvae and pupae per side of the comb that has been cut from a frame within the brood nest of a colony, frozen at -20°C for 24 h, and placed in the nest of the test colony (Taber 1982, Taber and Gilliam 1987). Colonies that remove the freeze-killed brood from the comb section within 48 h are considered hygienic; colonies that take longer than 6 -7 d to remove the brood are considered nonhygienic (Taber and Gilliam 1987, Spivak and Gilliam 1993). There are several difficulties inherent in the freeze-killed brood assay. It is relatively labor intensive and damages the comb and brood nest. More importantly, the rate of removal of the freeze-killed brood within a particular colony is not always consistent between assays (Rodrigues et al. 1996, Oldroyd 1996). It is known that the expression of hygienic behavior is influenced by environmental conditions; for example, incoming nectar increases the rate of removal of dead and diseased brood (Thompson 1964, Momot and Rothenbuhler 1971). However, the experimental assay itself may also contribute to the variability. Rodrigues et aI. (1996) found that colonies removed freeze-killed brood that were recently sealed with a wax capping more quickly than pupae that had been sealed for 5 d. Another factor contributing to the variation in the assay may stem from the source of the frozen brood. For example, it is often more convenient to cut multiple comb sections from one colony and introduce the frozen brood sections into other colonies. Because wax contains cues used in nestmate recognition (Breed et al. 1995), the bees may perceive that the introduced comb section has a foreign odor and remove the brood more rapidly from it than from comb sections from their own colony. A 2nd assay for hygienic behavior was developed by Newton and Ostasiewski (1986). This assay involves inserting an insect pin through larvae or pupae covered by wax cappings and recording the time required for colonies to remove the pierced brood. This assay is not as labor intensive as the freeze-killed brood test and does not cause as much damage to the combs in the colony. However, holes left in the wax cappings and the exposed hemolymph from the pierced larvae or pupae may increase the rate of removal of the brood by the bees. Newton and Ostasiewski (1986) determined there was a significant correlation between the freezekilled brood assay and the pierced brood assay (r = 0.956). This correlation was based solely on 5 col-
BEHAVIOR
IN HONEY
BEES
65
onies, all of which took >6 d to remove freeze-killed brood and thus were nonhygienic. To date, the correlation between hygienic behavior and disease resistance has been based on colonies selected for their response to freeze-killed brood (Gilliam et al. 1983, Spivak and Gilliam 1993), cyanide-killed brood (Jones and Rothenbuhler 1964), or to a pathogen (Rothenbuhler 1964). The only study that compared colony resistance to American foulbrood with the rate of removal of pierced brood yielded unexpected results because colonies that were slower to remove pierced brood were determined to be more resistant to B. larvae (Danka and Villa 1994). The goal of the field assay is to screen colonies for those that display the most rapid hygienic behavior and then to test the hygienic colonies for their ability to remove diseased or Varroa-infested brood from the nest. It is important to determine whether procedural variations in the assay have the same effect on all colonies, particularly those that remove freeze-killed brood consistently within 48 h. The objectives of the current study were to determine if procedural variations in the freeze-killed brood assay affect the results, and to reexamine the correlation between the freeze-killed and the pierced brood assays. Materials and Methods Four experiments were conducted. Experiment 1 tested whether the age of the brood within the section of frozen comb had an effect on the rate of its removal by the bees. Experiment 2 tested whether the source of the brood had an effect on the removal rate. In experiment 3, the results of the freeze-killed assay and the pierced brood assay were compared using colonies that displayed a range of removal rates of freeze-killed brood. Experiment 4 was designed to determine cues that elicit rapid removal behavior by bees. All tests were conducted in summer months of 1994 through 1996 at the University of Minnesota in an apiary on the St. Paul campus. Some of the colonies used in the experiment were part of a bidirectional selection program for hygienic behavior (Spivak 1996). These queens were raised from colonies that consistently removed at least 95% of the freeze-killed brood within 48 h, or from colonies that took >6 d to remove frozen brood. The queens used in the experiments were instrumentally inseminated with 4-6 j.Llsemen from drones from other colonies with similar removal rates. Additional colonies included in the experiment were not part of the breeding program and contained naturally mated queens. All colonies were derived from an Italian line of honey bees, except for 3 colonies used in 1994 that were Buckfast in origin. Experiment 1: Does the Age of the Frozen Brood Affect Its Rate of Removal? Two sections of comb (5 by 6 cm each) containing =100 cells per side were cut from 11 colonies in 1994 and from 10 colonies in
66
JOURNAL
OF ECONOMIC
1995. In 1994, to obtain pupae of known ages, the queens in each of the 11 colonies were caged on a frame and allowed to lay eggs for at least 24 h before they were released. Nine days later, the queens were caged again on a different frame in the same colonies for 24 h. Two comb sections were cut from each colony 20 d after the queens were caged on the 1st frame. On day 20, the pupae on the 1st frame were 1 d from eclosing as adults, and on the 2nd frame they were larvae or prepupae, having been sealed with a wax capping by nurse bees in the colony within the previous 48 h. In 1995, the ages of the brood were estimated visually according to Jay (1962). To obtain brood (larvae and prepupae) in the early stages of development, combs were selected that contained 5th instars in the process of being sealed. Older pupae were obtained by selecting combs containing pupae that were within 2-3 d of ec\osion (noted by uncapping some cells and determining if the head and thorax were medium to dark brown), and another comb section was cut from this frame. The 2 comb sections (5 by 6 cm each) were cut from each colony and placed in the freezer at -20°C for 24 h. The number of intact sealed cells (the cells in which the wax capping was undamaged) in each of the frozen sections was recorded, and the sections were inserted in the same frames from which they were removed. The number of pupae that were completely removed from the cells was counted on days 1 and 2 after the frozen brood was introduced and every 2 d after that until the frozen brood was removed. Brood was not counted as removed if the cell was only uncapped or if any part of the dead brood remained. In this way, the most conservative estimate of removal was recorded. The colonies were hygienic if they completely removed an average of95% of the frozen brood from both comb sections within 48 h. They were nonhygienic if, on average, >5% of the frozen brood remained within both combs on day 6. The remaining colonies were intermediate. The data were analyzed using a split-split plot design for each year. The whole plot was a I-way analysis of variance (ANOVA), with colony type as the main effect tested against the colony within type error term. The split-plot divided the colonies into different treatment groups, and the split-split plot was a repeated measures in time on the treatment groups from the split-plot (SAS Institute 1995). The repeated measures was used to analyze only days 1 and 2 when the assay has the most biological relevance in terms of selecting colonies that displayed the most rapid hygienic behavior. Experiment 2: Does the Source of the Frozen Brood Affect Its Rate of Removal? Nineteen colonies were tested in 1995. Two comb sections (5 by 6 cm each) containing capped brood of unknown age were cut from each colony and were frozen for 24 h. One section was inserted into the same colony from which it was cut; the other was inserted into a different test colony. The number of pupae com-
ENTOMOLOGY
Vol. 91, no. 1
pletely removed from each comb section was counted on days 1, 2, 4, and 6 after the frozen brood was introduced. The percentage of brood that was removed from each of the comb sections was calculated. The colonies were grouped as being hygienic, nonhygienic, or intermediate based on the same criteria as in experiment 1. The data was analyzed using a split-split plot design as in experiment 1. Experiment 3: Is There a Correlation Between Removal of Freeze-Killed and Pin-Killed Brood? Eleven colonies were tested in 1994 (the same colonies from experiment 1), and 19 colonies were tested in 1995 (from experiment 2). In 1994, the freeze-killed and pierced brood were of known age (recently sealed or pre-eclosion) and were derived from the same colonies from which the combs were cut. In 1995, the freeze-killed and pierced brood were of unknown age derived from different source colonies. The freeze-killed brood sections were introduced as in experiment 1. A number 5 insect pin (following Newton and Ostasiewski 1986) was used to pierce 49 sealed brood cells (7 groups of 7 cells in hexagonal arrangements) through the center of the cell cap, penetrating the body of the pupa until the pin reached the base of the cell. In 1994, the frozen inserts and pierced cells were inspected on days 1, 2, 4, 6, 8, and 13. In 1995, they were inspected on days 1, 2, 4, and 6. The number of pupae completely removed from the comb was counted in both cases. A Pearson correlation coefficient was calculated to compare the time of removal of the pierced and freeze-killed brood. Experiment 4: What Cues Elicit Rapid Removal Behavior? This series of tests was conducted in 1996 using only colonies from the bidirectional selection program for hygienic behavior. A Jenter box (Brushy Mountain Bee Farm, Moravian Falls, NC) was used to test the removal response of 8 hygienic and 5 nonhygienic colonies to treated brood (following methods of Boecking and Drescher 1992, Spivak 1996). A Jenter box contains =300 plastic worker cells. Ninety of the cells within the box have false bottoms fitted with removable plugs that allow access to individual pupae through the bottom of the cell. Brood of known age was obtained by caging a laying queen over cells within the Jenter box. Three treatments were given to brood that had been sealed within the previous 48 h within the Jenter box. In the 1st treatment, the plugs were removed and 10 pupae per colony were pierced with a number 00 insect pin. In the 2nd and 3rd treatments, a 5-l-Lldrop of hemolymph extracted from a live pupa (treatment 2) or from a freeze-killed pupa in the same stage of development (treatment 3) was applied with a pipetman on the surface of the posterior end of 10 different pupae per treatment. Another 10 cells per colony serving as controls had the plugs removed and replaced with no treatment. The treated and control cells were marked on a transparent sheet of plastic (following Infantidis 1983), and were inspected on days 2, 4, 7, and 10 to de-
FE'bruary 1998
SPIVAK
AND DOWNEY:
HYGIENIC
BEHAVIOR
IN HONEY
1994
67
BEES
1995 Intermediate
hygienic
100 100
80
80
"i; Q>
"> Q>
60
E
.
E
.
l!!
l!!
~
60
0
~ 40
o prepupae
20
•
non hygienic
40
20
preeclosion
0
o
prepupae
•
preecloslon
0 day 1
day2
day 4
day 6
Fig.!. Mean percent:!: SE of freeze-killed brood in 2 devdopmental stages removed by 4 hygienic, 4 intermediate, and 3 nonhygienic colonies by days 1, 2, 4, and 6 in 1994. Comb sections containing larvae or prepupae and pupae 1-2 d from eclosion (determined by caging a laying qUE'enover empty cells) were introduced into each colony, and the number of cells from which brood was removed was recorded on days 1, 2, 4, and 6. The colonies rellloved significantly lllore prepupae than pre-eelosion pupae by day 2 (day 1, F = 3.92; df = 1, 8; P = 0.08; and day 2, F = 5.57; df = 1, 8; P = 0.05. termine if the bees had detected and removed the treated and control brood. The results were analyzed using a split-plot design (no repeated measures) from day 2 only because the percentage of removal did not change appreciably in most colonies after the 2nd d. Results Experiment 1. In 1994, the colonies removed significantly more young brood than preeclosion pupae by day 2 (Fig. 1). The hygienic colonies removed the pupae of both ages at the same rate; thereforE', the differences in removal rate were indicative only of the responses of the intermediate and non hygienic colonies. In 1995, the hygienic and intermediate colonies tended to remove older pupae more quickly than prepupae, whereas the reverse was true of the nonhygienic colonies (Fig. 2). However, the ANOVA revealed no significant effect of age of brood on the removal rate on day 1 or day 2. Experiment 2. The colony source of the frozen brood did not have a significant effect on how quickly the bees removed the dead brood from the nest (day 1, F = 0.11; df = 1, 16; P = 0.74; and day 2, F = 0.79; df = 1,16; P = 0.39). Experiment 3. Some of the pupae were not killed when pierced, but eclosed normally as adults. Other pierced pupae showed signs of development but did not eclose, or died from chalkbrood infection. In all,
day 1
day2
day4
day 6
Fig. 2. Percent:!: SE of freeze-killed brood in 2 developmental stagesremoved by 3 hygienic, 5 intermediate, and 2 nonhygienic colonies in 1995. Comb sections containing older larvae or pupae in early stages of development and pupae in late stages of development (estimated visually) were introduced into each colony, and the number of cells from which brood was removed was recorded on days 1, 2, 4, and 6. The colonies did not remove significantly more prepupae than pre-eelosion pupae (P > 0.05 all days). few colonies removed 100% of the pierced pupae. In 1994, the Pearson correlation coefficient between the number of days it took the 11 colonies to remove 95% of both the pierced and thc freeze-killed prepupae was r = 0.687 (P = 0.06). The correlation coefficient between removal of pierced and freezekilled pupae near ec1osion in the same colonies in 1994 was r = 0.8905 (P = 0.001). In 1995, four of the 19 colonies had not removed 95% of the frozen brood by day 6 when inspections were terminated and were not included in the analysis. The correlation coefficient between the number of days it took the colonies to remove 95% of the pierced and freeze-killed brood of unknown age was 0.4752 (P = 0.07).
Figures 3 and 4 compare the results of the two assays 24 h after the experiment began in 1995. Fig. 3 shows that all the pierced brood was removed in contrast to only a portion of the freeze-killed brood that was removed within the same period. If a determination of the degree of hygienic behavior was made on day 1, two colonies would be considered hygienic based on the freeze-killed brood assay, but 5 colonies might be considered hygienic based on the removal of pierced pupae (Fig. 4). By day 2, the number of hygienic colonies increased to 5 and 13 for the freeze-killed brood and pierced brood assays, respectively (Fig. 4). Experiment 4. There was no significant difference between the hygienic and nonhygienic colonies in the amount of treated and control pupae removed from the Jenter Box on day 2 (F = 3.39; df = 1, 12;
68
JOURNAL
OF ECONOMIC
Vol. 91, no. 1
ENTOMOLOGY
100 90
Chyglenlc Ii3nonhygienic
80
b
70
11>
60
~
50
l!! 40
*
30 20 10
o control
Fig. 3. An example of the discrepancy between the results of the freeze-killed and pierced brood assaysby 1 colony 24 h after treatment. Approximately 40% of the freeze-killed brood remained sealed (right side of photo) . However, all of the pierced brood was completely removed (6 of the 7 groups of 7 cells in hexagonal arrangements can be seen on left side of photo). P = 0.144) (Fig. 5). By day 10 (1 d before eclosion), the hygienic colonies had removed an average of only 69 ± 28.7% of the pierced pupae. The nonhygienic colonies had removed 68 ± 31.1%. However, there was a significant treatment effect (F = 15.95; df = 3,29; P = 0.001); more pupae were removed that had been pierced or treated with hemolymph from a freeze-killed pupa than pupae that were treated with hemolymph from a live pupa or the untreated controls (Tukey honestly significant difference [HSD] test). Discussion The results of the first 2 experiments indicate that neither the age nor the source of the frozen capped
• freeze-killed 100
">o CD
E
e ~ o
hemolymph deadpupa
pierced
Fig. 5. Mean percent ± SE of brood removed by 8 hygienic and 5 nonhygienic colonies by day 2 after the brood was treated. Different letters above bars indicate significant treatment effects (Tukey test for mean separation).
brood had a significant effect on the removal rate of dead brood by highly hygienic colonies (those colonies that consistently removed freeze-killed brood within 48 h). In 1994, colonies that demonstrated slower rates of removal of freeze-killed brood (the intermediate and nonhygienic groupings in this study) tended to remove young brood faster than pupae near eclosion. The 1994 results were consistent with those of Rodrigues et al. (1996). It is unclear why the difference in the rate of removal of young brood and pre-eclosion pupae by the intermediate and nonhygienic colonies was not observed in 1995. It is likely that there was greater variation in the stage of development of the brood within comb sections in 1995 than in 1994 because the queens were not caged over the frames at specific intervals. If the development stages over-
+ + +
+++++
+++++
o pierced
90 80 70 60 50 40 30 20 10
o
hemolymph live pupa
~I
I
Nonhygienic
J
I
I
Intermediate
Hygienic
Fig. 4. Percent of freeze-killed and pierced brood removed by individual nonhygienic, intermediate, and hygienic colonies 24 h (day 1) after treatment in 1995 (solid and open bars). Stage in development of the pupae was unknown. Cross symbol over bars indicate the 13 colonies that removed all of the pierced brood within 48 h, and might be considered hygienic based on that assay.
February 1998
SPIVAK AND DOWNEY:
HYGIENIC
lapped, the difference in removal rates may have been obscured. The aim of the field assay is to screen colonies for those that display the most rapid hygienic behavior. Although the response of colonies in the intermediate and non hygienic groupings may vary depending on the age of the frozen brood, highly hygienic colonies will remove all brood from any source colony at the same rate. In addition, our unpublished results indicate that highly hygienic colonies demonstrate more consistent rates of removal between consecutive trials. There was a weak correlation between the removal of young freeze-killed and pierced brood, and a significant correlation between the removal of pre-eclosion freeze-killed and pierced pupae in 1994. However, the freeze-killed brood assay is a more conservative test of hygienic behavior, as illustrated in Figs. 3 and 4. To find colonies that display very rapid removal of pierced pupae, the colonies should be inspected within 24 h of treatment because of the tendency of the bees to remove pierced brood more rapidly than frozen brood. However, it may not always be feasible to inspect a colony within 24 h of treatment. When pupae were pierced with a very fine (number 00) insect pin through the base of the Jenter box (i.e., without poking a hole in the wax cell capping), there was no difference in the amount of pupae removed by the hygienic and nonhygienic colonies by day 2. On average, 30% of all pierced pupae survived the treatment, which considerably diminished the accuracy and reproducibility of the test. When pupae were treated with hemolymph extracted from either a live or freeze-killed pupa, there was again no difference in the rate of removal by hygienic and nonhygienic colonies. These results indicate that bees from nonhygienic lines can be induced to express hygienic behavior only if a sufficiently strong stimulus is present. The stimulus, in this case, was the hemolymph of live or dead pupae, and most likely the cues were olfactory. Although hygienic behavior is heritable, its expression may be dependent on the perception of appropriate cues as well as on the "willingness" of bees to remove those agents releasing such cues. Both hygienic and nonhygienic colonies removed significantly more pupae treated with hemolymph from a dead pupae than hemolymph from a live pupa within 2 d, indicating that the cue that stimulates removal behavior is stronger in dead pupae. Because the wax cell capping was not opened during the experimental procedure, the cue must have been detectable to the bees through the wax. After day 2, there was only a very slight (at most 10%) increase in the number of treated pupae that were removed by hygienic and nonhygienic colonies. When the remaining treated pupae were inspected on day 10, all had developed normally. We speculate that the hemolymph was shed from the pupa with the last molt, or when the pupa spun its cocoon, eliminating its effect as a stimulus for removal behavior.
BEHAVIOR
IN HONEY
BEES
69
Newton and Ostasiewski (1986) claimed that the speed with which cells were uncapped was not caused by the pin hole in the cell cap or by the loss of hemolymph from the dying pupa. In 1 set of their experiments, the cell cappings were removed without damaging the live pupae, which resulted in the bees recapping the cells. Spivak and Gilliam (1993) also showed that hygienic and non hygienic colonies replaced the cell cappings over uncapped brood. In addition, experiments by Newton and Michl (1974) and by Free and Winder (1983) revealed that the tendency of the bees to replace the capping increased with the age of the pupa. The latter results may explain the tendency of some colonies to remove dead young pupae more quickly than older pupae. The propagation of hygienic colonies that demonstrate resistance to chalkbrood and American foulbrood and that are able to remove pupae infested by Varroa mites is becoming increasingly important in apiculture. At this time, the freezekilled brood assay is the most conservative and reliable screening procedure for hygienic behavior. We recommend the following procedure to researchers and apiculturists interested in screening colonies for hygienic behavior. Sections of randomly selected capped brood should be cut from a healthy colony, frozen, and introduced into the colonies to be tested. The assay should be repeated at least twice. Only colonies that remove >95% of freeze-killed brood within 48 h in both tests should be considered hygienic. An important consideration is that most but not all hygienic colonies will be resistant to chalkbrood after 1 generation of selection (Gilliam et al. 1988, Spivak and Gilliam, 1993). Therefore, when developing hygienic breeder stock (through instrumental insemination) for research or commercial purposes, we also recommend that the hygienic colonies be challenged with the chalkbrood pathogen and possibly with the American foul brood pathogen to ensure that the breeder stock is resistant. Acknowledgments We particularly thank Gary Reuter for his assistance in all stages of this study and manuscript preparation. Michael Lamb, Rebecca Melton, and John Breyfogle also contributed to data collection. Carl Schwarz (Simon Fraser University) provided statistical advice, and Martha Gilliam, Ben Oldroyd, and Mark Winston made helpful comments on the manuscript.This research wasfunded by a grant from Minnesota Agricultural Utilization Research Institute (AURI-PRO 130) with matching funds from the Minnesota Honey Producers and Hobby Beekeeping Associations. This is publication No. 97-1170104 from the Minnesota Agricultural Experiment Station. References Cited Boecking, 0., and W. Drescher. 1991. Response of Apis mellifera L. colonies infested with Varroa jacobsoni Oud. Apidologie 22: 237-241.
70
JOURNAL OF ECONOMIC ENTOMOLOGY
1992. The removal responses of Apis mellifera L. colonies to brood in wax and plastic cells after artificial and natural infestation with Varma jacobsoni Oud. and to freeze-killed brood. Exp. App\. Acaro\. 16: 321-329. Breed, M. D., M. F. Garry, A. N. Pearce, B. E. Hibbard, L. B. Bjostad, and R. E. Page, Jr. 1995. The role of wax comb in honey bee nestmate recognition. Anim. Behav. 50: 489 - 496. Danka, R. G., and J. D. Villa. 1994. Preliminary observations on the susceptibility of Africanized honey bees to American fOlllbrood. J. Apic. Res. 33: 243-245. Free, J. B., and M. E. Winder. 1983. Brood recognition by the honeybee (Apis mellifera) workers. Anim. Behav. 31: 539-545. Gilliam, M., S. Taber III, and G. V. Richardson. 1983. Hygienic behavior of honey bees in relation to chalkbrood disease. Apidologie 14: 29-39. Gilliam, M., S. Taber III, B. J. Lorenz, and D. B. Prest. 1988. Factors affecting development of chalkbrood disease in colonies of honey bees, Apis mellifera, fed pollen contaminated with Ascosphaera apis. J. Invertebr. Patho\. 52: 314 -325. Infantidis, M. D. 1983. Ontogenesis of the mite Van-oa jacobsoni in worker and drone honeybee brood cells. J Apic. Res. 22: 200-206. Jay, S. C. 1962. Colour changes of honeybee pupae. Bee World 43: 119-122. Jones, R. L., and W. C. Rothenbuhler. 1964. Behaviour genetics of nest cleaning in honey bees. II. Responses of two inbred lines to various amounts of cyanide-killed brood. Anim. Behav. 12: 584-588. Momot, J. P., and W. C. Rothenbuhler. 1971. Behaviour genetics of nest cleaning in honeybees. VI. Interactions of age and genotype of bees, and nectar flow. J. Apic. Res. 10: 11-21. Newton, D. C., and D. J. Michl. 1974. Cannibalism as an indication of pollen insufficiency in honeybees: ingestion or recapping of manually exposed pupae. J. Apic. Res. 13: 235-241. Newton, D. C., and N. L. Ostasiewski, Jr. 1986. A simplified bioassay for behavioral resistance to American foulbrood in honey bees (Apis mellifera L.). Am. Bee J. 126: 278 -281.
Vol. 91, no. 1
Oldroyd, B. P. 1996. Evaluation of Australian commercial honey bees for hygienic behaviour, a critical character for tolerance to chalkbrood. Australian J. Exp. Agric. 36: 625-629. Peng, Y. S., Y. Fang, S. Xu, L. Ge, and M. E. Nasr. 1987. Response of foster Asian honey bee (Apis cerona Fabr.) colonies to the brood of European honey bee (Apis mellifera L.) infested with parasitic mite Van-oojacobsoni Oudemans. J. Invertebr. Patho\. 49: 259-264. Rath, W., and W. Drescher. 1990. Response of Apis ceruna Fabr. towards brood infested with Van-oo jacobsoni Oud. and infestation rate of colonies in Thailand. Apidologie 21: 311-321. Rodrigues, I., J. Beetsma, W. J. Boot, and J. Calis. 1996. Testing hygienic behavior in four different honeybee strains. Proc. Sec. Exp. App\. Entomo\. Neth. Entomo\. Soc. 7: 83-88. Rothenbuhler, W. C. 1964. Behavior genetics of nest cleaning in honey bees. IV. Responses of Fl and backcross generations to disease-killed brood. Am. Zoo\. 4: 111-123. SAS Institute.
1995. GLM procedure. user's manual. re-
lease 6.10. SAS Institute, Cary, NG Spivak, M. 1996. Hygienic behavior and defense against Van-oa jacobsoni. Apidologie 27: 245-260. Spivak, M., and M. Gilliam. 1993. Facultative expression of hygienic behaviour of honey bees in relation to disease resistance. J. Apic. Res. 32: 147-157. Taber, S., III. 1982. Determining resistance to brood diseases. Am. Bee J. 122: 422-425. Taber, S., III., and M. Gilliam. 1987. Breeding honey bees for resistance to diseases. Korean J. Apic. 2: 15-20. Thompson, V. C. 1964. Behaviour genetics of nest cleaning in honeybees. III. Effect of age of bees of a resistant line on their response to disease-killed brood. J. Apic. Res. 3: 25-30. Woodrow, A. W., and E. C. Holst. 1942. The mechanism of colony resistance to American foulbrood. J. Econ. Entomo\. 35: 327-330. Received for publication 30 December 1996; accepted 22 September 1997.