Treatment With Synthetic Brood Pheromone (SuperBoost) Enhances Honey Production and Improves Overwintering Survival of Package Honey Bee (Hymenoptera: Apidae) Colonies

Abstract

We evaluated a year-long treatment regime testing synthetic, 10-component, honey bee, Apis mellifera L. (Hymenoptera: Apidae), brood pheromone (SuperBoost; Contech Enterprises Inc., Delta, BC, Canada) on the productivity and vigor of package bee colonies in the lower Fraser Valley of British Columbia, Canada. Fifty-eight newlyestablished 1.3-kg (3-lb) colonies treated three times with SuperBoost at 5-wk intervals starting 30 April 2009 were compared with 52 untreated control colonies. Treated colonies produced 84.3% more honey than untreated control colonies. By 8 September 2009, SuperBoost-treated colonies had 35.4% more adults than untreated colonies. By 28 September, net survival of treated and control colonies was 72.4 and 67.3%, respectively. On 5 October, treated and control colonies were divided into two additional groups, making up four cohorts: SuperBoost-treated colonies treated again during fall and spring build-up feeding with pollen substitute diet (BeePro, Mann Lake Ltd., Hackensack, MN; TTT); controls that remained untreated throughout the year (CCC); colonies treated with SuperBoost in spring–summer 2009 but not treated thereafter (TCC); and original control colonies treated with SuperBoost during the fall and spring build-up feeding periods (CTT). There was no difference among cohorts in consumption of BeePro during fall feeding, but TTT colonies (including daughter colonies split off from parent colonies) consumed 50.8% more diet than CCC colonies during spring build-up feeding. By 21 April, the normalized percentages of the original number of colonies remaining (dead colonies partially offset by splits) were as follows: CCC, 31.4%; CTT, 43.8%; TCC, 53.59%; and TTT, 80.0%. The net benefit of placing 100 newly established package bee colonies on a year-long six-treatment regime with SuperBoost would be US$6,202 (US$62.02 per colony). We conclude that treatment with SuperBoost enhanced the productivity and survival of package bee colonies and hypothesize that similar results could be achieved with established colonies.

honey bee, brood pheromone, SuperBoost, honey, survival

In nature, the brood pheromone of the honey bee, Apis mellifera L. (Hymenoptera: Apidae), has both primer and releaser effects. The principal primer effect is stimulation of the mandibular and hypopharyngeal glands to produce enriched protein that is fed to the queen and larvae (Peters et al. 2010). Releaser effects include induction and enhancement of foraging behavior, resulting in more pollen being brought back to the hive (Pankiw 2004, 2007; Pankiw and Page 2001; Pankiw and Rubink 2002; Pankiw et al. 1998, 2004). Synthetic brood pheromone also induces increased consumption of pollen substitute diet by overwintered colonies (Pankiw et al. 2008).

SuperBoost (Contech Enterprises Inc., Delta, BC, Canada) is a commercial product based on the brood pheromone. It is a synthetic blend of 10 fatty acid esters (LeConte et al. 1990) formulated in a ratio that precisely mimics the natural composition (Pankiw et al. 2010). Tertiary-butylhydroquinone, a food-grade antioxidant, is added as 0.05% of the total composition to stabilize the methyl and ethyl esters of linolenic, linoleic, and oleic acid.

One SuperBoost treatment of 180 mg lasts at least 5 wk. Worker honey bees contact milligram amounts of pheromone that exudes daily from one side of a flat plastic pouch that is suspended in a holder between the frames of a hive.

In late summer 2007 in southeastern Texas treatment with SuperBoost caused higher ratios of pollen to nonpollen foragers in treated than control colonies for over 5 wk, and stimulated foragers to bring back heavier pollen loads to the hive (Pankiw et al. 2011). Moreover, colonies treated in August–September experienced 47 and 36% increases in brood comb area and adult population size, respectively, at the same time that untreated control colonies were respectively declining by 24 and 10.7%. In a study on spring build-up in British Columbia, Canada, SuperBoost-treated colonies consumed 50% more BeePro (Mann Lake Ltd., Hackensack, MN) pollen substitute diet over a 10-wk period than control colonies (Moeri et al. 2011). Treated colonies also had 2.4 and 2.0 times more brood comb area and adult population size, respectively, than control colonies, and produced over twice as many splits. Similarly, R. R. Sagili (personal communication) found that SuperBoost-treated colonies in two Oregon locations consumed more pollen substitute diet during fall feeding than untreated control colonies. Although both treated and control colonies declined in brood comb area and adult population size, the declines in SuperBoost-treated colonies were significantly less than in untreated control colonies.

The first shipment of package bees in the United States occurred in 1912 (Laidlaw 1992), and by 1932 it was described as a “rapidly developing” industry for beekeepers in the southern states (Anonymous 1932). Today, package bees for use in North America can be imported from Australia and New Zealand (USDA–APHIS 2008, British Columbia Ministry of Agriculture and Lands 2010). In the past, many beekeepers harvested honey and then killed off package bee colonies at the end of summer. However, the high cost and sometimes uncertain availability of package bee colonies have caused a shift to a practice of harvesting surplus honey and then maintaining the new package colonies over the winter (Tucker 1980). Early development of vigorous colonies in the spring after package bee colonies are established is critical to survival in the following winter.

We hypothesized that treatment with SuperBoost would enhance the productivity and survival of package bee colonies. We report the results of a year-long study in which package bee colonies in the lower Fraser Valley of British Columbia were put on a SuperBoost regime, beginning with their establishment in spring 2009.

Materials and Methods

Experimental Methods.

Table 1 provides a summary of treatments of 58 SuperBoost-treated and 52 untreated control package bee colonies from 30 April 2009 to 11 May 2010. All workers and queens were classified as New Zealand–Carniolan strain (Arataki Honey Ltd., Rotorua, New Zealand). Use of package bees ensured that all colonies were nearly identical in size at the start of the year-long experiment.

Table 1

Summary of experimental and apicultural activities carried out on newly established package bee colonies (Abbotsford, BC, Canada; 2009–2010)

SuperBoost pouches were loaded with 200 μl (180 mg) of stabilized brood pheromone blend (Pankiw et al. 2010) in the following ratio: methyl palmitate, 2.11%; ethyl palmitate, 4.19%; methyl stearate, 17.30%; ethyl stearate, 8.04%; methyl oleate, 21.66%; ethyl oleate, 7.50%; methyl linoleate, 8.09%; ethyl linoleate, 3.82%; methyl linolenate, 16.53%; ethyl linoenate, 10.77%; and tertiary-butylhydroquinone, 0.05%. For the first three 5-wk treatments in spring and early summer 2009, loaded pouches were mounted in Gepe 35-mm slide frames, and were suspended between the frames of a hive at the level of the brood comb by using 22-gauge galvanized steel wire (Pankiw et al. 2011). For fall-feeding in 2009 and spring build-up feeding in 2010, loaded pouches were mounted in a new two-piece plastic holder (Foster et al. 2011b).

Honey was harvested on eight occasions over ∼3 mo, beginning in early June 2009. Because the objective was to maintain vigorous colonies, honey harvesting was done in an extremely conservative manner, taking only frames that had only honey, or on occasion negligible areas of drone brood comb. Weight of honey was determined using a digital scale by subtracting the weight of a filled frame from the weight of an empty frame of the same dimensions. Most of the frames were 50.8 by 16.5 cm (85.5% for control colonies and 82.3% for SuperBoost-treated colonies). The remainder of the frames were 50.8 by 24.3 cm.

Evaluations of brood and honey comb area and adult population size were done at the end of summer 2009 before fall feeding and after spring build-up feeding in 2010 (Table 1). A grid consisting of a Langstroth deep frame divided into 6.45-cm² sections was used to estimate the area of brood and honey comb on each side of a frame by summing the number of sections overlying stored honey or brood cells. Populations of adult honey bees were estimated by measuring the area of both sides of each frame covered by adults and then converting this area to bee numbers, ≈1.5 bees per cm² (Pankiw et al. 2004).

Before fall feeding, SuperBoost-treated (T) and control colonies (C) were each divided into two additional groups. This was done by selecting paired colonies of similar size and vigor and randomly assigning one of them to be treated with SuperBoost during the fall and spring build-up feeding periods, and the other to be an untreated control. Some of the colonies selected on 28 September were found to be dead on 5 October 2009, contributing to uneven numbers in the four cohorts. Thus, 19 colonies were in the cohort receiving SuperBoost treatment at all three periods (TTT), 15 colonies remained as untreated controls throughout the year (CCC), 23 colonies that had received SuperBoost treatment three times in spring–summer 2009 were not treated thereafter (TCC), and 20 original control colonies were treated with SuperBoost during the fall and spring build-up feeding periods (CTT).

On 5 October, each colony was given a 0.45-kg (1-lb) patty of pollen substitute diet, BeePro. At four times during the fall feeding period, and three times in the spring build-up feeding period, the hives were opened up, and the remaining BeePro was weighed. If a patty was mostly consumed, it was replaced with a preweighed fresh patty. Supplementary granulated sugar and sugar syrup (when the weather was warm) were added equally to each colony at the beekeeper's discretion, but their consumption was not measured.

As colonies grew to at least 12 frames and had sufficient strength as judged by the beekeeper in spring–summer 2009 and in the spring 2010, they were split into two colonies, and supplementary Carniolan queens were placed in the daughter colonies. The date of each split was recorded, and each daughter colony received a permanent plastic tag so the performance of mother and daughter colonies could be monitored. Daughter colonies were not evaluated in measurements in fall 2009. In spring 2010, daughter colonies received the same SuperBoost or control treatment as their parent colonies. For evaluations of brood and honey comb area and adult population size in the spring 2010, parent and daughter colonies were treated as one colony.

SuperBoost devices were retrieved for analysis of release rate at two times. The first was on 16 July 2009, after devices had been in hives for 42 d (N = 7). The second was on 7 May 2010, after 37 d in hives (N = 15). Weights of retrieved pouches were measured on an analytical balance, and the mean weight loss determined by subtraction from the mean weight of eight freshly loaded pouches (469.9 ± 4.0 mg in 2009 and 520 ± 4.9 mg for the slightly larger pouch used in 2010).

Statistical Analysis.

Because there were 58 SuperBoost-treated and 52 control colonies at the start of the experiment, data on numbers of honey frames taken and honey weights from the entire group of treated colonies were normalized by multiplying them by 0.897. Chi-square tests were then run comparing the actual distributions against an expected 50:50 distribution for normalized percentages of total honey frames taken and total kilograms of honey harvested. A chi-square test also was used to compare the percentages of treated and control colonies producing a honey harvest. For those colonies that produced honey, t-tests were run to compare the means for numbers of frames taken and kilograms of honey produced between treated and control colonies.

Population vigor at the end of summer was evaluated using t-tests to compare means between SuperBoost-treated and control colonies for honey and brood comb area and numbers of adult bees. After spring build-up feeding, population vigor was assessed by analysis of variance (ANOVA) followed by Tukey's honestly significant difference (HSD) test (if P < 0.05) comparing colony means among the four treatment cohorts (TTT, TCC, CTT, and CCC).

Consumption of BeePro pollen substitute diet at the end of fall and spring build-up feeding was analyzed by ANOVA followed by Tukey's HSD test (if P < 0.05) comparing colony means among the four treatment cohorts. The same tests with the data converted to mean diet consumed per bee (larvae + adults, measured before fall feeding and after spring build-up feeding) were used to determine if differences among colonies were due solely to population size or if individual bees actually differed in the amount of diet consumed.

For continuity of survival percentages after the summer of 2009, the numbers of colonies in each of the four fall and spring build-up feeding cohorts were normalized to the number of colonies surviving in treated and control groups at the end of summer: 42 for TTT and TCC cohorts and 35 for CCC and CTT cohorts. This allowed the survival percentages to be calculated for the entire year. Proportions of colonies surviving among the four treatment groups were then compared using a modified Student–Newman–Keuls test with arcsine-transformed data (Steel and Torrie 1980). In all cases, α = 0.05.

Results and Discussion

Pheromone Release Rate.

Weight loss evaluation of retrieved pouches revealed that the mean pheromone release rate from 4 June to16 July 2009 was 0.81 ± 0.07 mg/day. Similarly, from 1 April to 7 May 2010, the release rate was 0.70 ± 0.11 mg/day. Assuming that one worker larva contains 560 ng of brood pheromone (Pankiw 2004), these release rates are equivalent to 1,446 and 1,250 larval equivalents per day, respectively.

The above-mentioned release rates are somewhat higher than the daily rates of 0.37 ± 0.25 and 0.54 ± 0.23 mg determined for two 5-wk periods during spring build-up feeding of established colonies in spring 2009 (Moeri et al. 2011), and the daily release rates of 0.29 ± 0.04–0.35 ± 0.04 mg obtained by removing exuded pheromone by swabbing the low-density polyethylene pheromone release membrane in the laboratory (Pankiw et al. 2011). Nonetheless, they are well within the target rates of 0.1–1.0 mg/day set by Pankiw et al. (2011). Moreover, the higher release rates can be attributed to greater activity caused by more contact than a daily swabbing of the membrane, as well as warmer temperatures later in the season than for the measurements taken by Moeri et al. (2011). When colonies were opened up for examination, clusters of bees were frequently observed around the release device, supporting the interpretation that bees physically remove pheromone by contact with the release membrane and that disruption of the pheromone film on the membrane surface allows more pheromone to be released.

Honey Harvest.

For both number of honey frames and kilograms of honey harvested, the cumulative values tracked each other closely for approximately a month before the yield diverged sharply upward in favor of SuperBoost-treated colonies (Figs. 1 and 2). The final values revealed 75.7% more frames and 84.3% more honey harvested from SuperBoost-treated than control colonies (Table 2). Both values were significantly divergent from an expected 50:50 distribution if there were no effect of treatment. The apparent absence of stimulation caused by the third treatment on 16 July 2009 (Figs. 1 and 2) suggests that two treatments at the beginning of nectar flow will suffice to enhance honey production of newly established package bee colonies.

Fig. 1

Comparison between untreated control package bee colonies and colonies treated three times with SuperBoost for 15 wk beginning on 30 April 2009 for cumulative numbers of honey frames harvested. Final values followed by different letters are significantly different (P < 0.05; chi-square test). See Table 2 for results of statistical analysis.

Fig. 2

Comparison between untreated control package bee colonies and colonies treated three times with SuperBoost for 15 wk beginning on 30 April 2009 for cumulative kilograms of honey harvested. Final values followed by different letters are significantly different (P < 0.05; chi-square test). See Table 2 for results of statistical analysis.

Table 2

Summary of honey production data and statistics compiled from 4 June to 26 August from 58 package bee colonies treated with SuperBoost and 52 untreated control colonies

The significant improvements in honey harvest could not be explained by the single fact that 31.8% more treated colonies produced a honey harvest (not significant), nor could they be explained by the 32.6% greater amount of honey harvested per producing colony (not significant) or by the 39.7% higher number of frames harvested per producing colony (not significant; Table 2). Rather, the combination of these three factors apparently resulted in the significantly higher percentages of total frames and kilograms of honey harvested.

There is abundant evidence that treatment with brood pheromone causes increased foraging and more pollen to be brought back to the hive (Pankiw et al. 1998, 2004; Pankiw and Page 2001; Pankiw and Rubink 2002; Pankiw 2004, 2007), but the lack of long-term studies has precluded the collection of data on honey production. Therefore, the data in Fig. 1 and Table 2, as well as a later study by Foster et al. (2011a), constitute the first demonstrations that treatment with brood pheromone can enhance the production of honey. In separate studies at three different times in the year, colonies treated with SuperBoost had more adult bees or produced more splits than control colonies (Foster et al. 2011a, Moeri et al. 2011, Pankiw et al. 2011). It is thus possible that the divergence of the cumulative honey harvest curves in early July (Figs. 1 and 2) is simply a consequence of higher numbers of foragers in SuperBoost-treated colonies.

Colony Strength at End of Summer.

Examination of the strength of all surviving colonies from 28 August to 8 September 2009 revealed a significantly higher (35.4%) adult population in colonies that had been treated with SuperBoost from 30 April to 20 August (Table 1) than in untreated control colonies (t = 2.7277, df = 85, P = 0.0077; Fig. 3). Because the size of the colony within the hive determines in part the survival of colonies over the winter (Furgala and McCutcheon 1992), these data suggest that colonies treated in the spring and summer would be better prepared for the winter than control colonies.

Fig. 3

Comparative evaluation at the end of summer (26 August–8 September 2009) for honey and brood comb area and adult population size of untreated control package bee colonies and colonies treated three times with SuperBoost for 15 wk beginning on 30 April 2009. Bars with different letters are significantly different (P < 0.05; t-test).

There was no difference in brood comb area between treated and control colonies (t = 0.8259, df = 85, P = 0.4112; Fig. 3). However, the 36.9% greater honey comb area in treated than control colonies approached significance (t = 1.8392, df = 85, P = 0.0694; Fig. 2). These data suggest that after termination of honey harvest, colonies treated previously with SuperBoost retained a better capacity for producing honey than control colonies. Again, this would have enhanced their ability to survive the coming winter.

Consumption of Diet During Fall and Spring Build-Up Feeding.

There were no significant differences (F_3,37 = 1.333; P = 0.270) among the four cohorts in consumption of BeePro over the month-long fall feeding period in 2009 (Table 3). This lack of any significant effect of treatment with brood pheromone does not agree with the results of a fall feeding study done at the same time by R. R. Sagili (personal communication) with established colonies in two Oregon locations. In his experiments, Sagili found that treatment with SuperBoost caused significantly greater consumption of supplementary pollen substitute diet, and resulted in significantly more adults and larger areas of brood comb. Unfortunately, the daughter colonies from splits were not tracked and included in the fall feeding experiments as part of their parent colonies.

Table 3

Consumption of BeePro diet during two feeding periods, fall (5 October–3 November 2009) and spring build-up (25 February–13 April 2010)

In spring 2010, continuously treated colonies (TTT, splits included) consumed significantly more (F_3,41 = 3.257; P = 0.031) BeePro in the spring build-up feeding period than colonies denied SuperBoost after the summer (TCC; Table 3). This result is similar to that found by Pankiw et al. (2008) with synthetic brood pheromone delivered to colonies daily on glass plates. Although the CCC and CTT cohorts were intermediate in diet consumption, the 50.8% greater consumption by TTT than CCC colonies is almost identical to the significantly greater consumption of SuperBoost-treated over control colonies found by Moeri et al. (2011).

When the amounts were calculated on a per bee basis (larvae + adults, as measured 2 mo before fall feeding ended, or up to a month after spring build-up feeding ceased; Table 1), there was no difference among treatments (fall 2009: F_3,37 = 1.784, P = 0.158; spring 2010: F_3,41 = 0.889, P = 0.455) (Table 3). Therefore, it seems that treatment with SuperBoost may not always stimulate individual bees to eat more supplementary diet. Rather, the positive effect on colonies that occurs through the production of more brood and higher numbers of adults could be caused by differentially greater consumption by nurse bees, leading to better nutrition of the queen and her brood through an enriched protein diet in the secretions of the mandibular and hypopharyngeal glands (Peters et al. 2010).

Colony Strength at the End of Spring Build-Up Feeding.

Evaluation of colony strength from 26 April to 11 May 2010 revealed a significantly greater area of honey comb in continuously treated colonies (TTT) than in all other treatments (Table 4). This is the third instance of enhanced honey production caused by treatment with brood pheromone. TTT colonies also had the highest number of adults (splits included with parent colonies), but the differences among cohorts were not significant. There was no difference in the area of brood comb among the four treatments.

Table 4

Comparative evaluation at the end of spring build-up feeding (26 April–11 May 2010) for honey and brood comb area and adult population size of untreated control package bee colonies (CCC, N = 6), original control colonies switched to SuperBoost-treated colonies for the fall and spring build-up feeding periods (CTT, N = 11), colonies that had received SuperBoost treatment three times in spring—summer 2009 that were not treated thereafter (TCC, N = 15), and colonies that received SuperBoost at all three periods (TTT, N = 13)

Survival Over Entire Year.

By combining colony death with the number of daughter colonies resulting from splits over the experimental year, the performance of colonies in different treatment groups (Table 5) can be calculated as a percentage of the starting population of colonies (Fig. 4). Over the first summer, SuperBoost-treated and control colony percentages tracked each other fairly closely, with the treated colony percentage at the end of summer being 72.4% (42 of 58 colonies) and the control percentage somewhat lower at 67.3% (35 of 52 colonies). Survival from 19 March 2010 onward was determined as a percentage of these numbers.

Table 5

Summary of package bee colony die-off, number of splits and survival for two treatment groups in 2009, and four treatment groups in 2010

Fig. 4

Comparative year-long percentages of the number of starting colonies for untreated control package bee colonies (CCC), original control colonies switched to SuperBoost-treated colonies for the fall and spring build-up feeding periods (CTT), colonies that had received SuperBoost treatment three times in spring–summer 2009 that were not treated thereafter (TCC), and colonies that received SuperBoost at all three periods (TTT). Percentages on 21 April 2010 followed by different letters are significantly different (P < 0.05; chi-square test for multiple proportions).

Untreated control colonies (CCC) had very poor survival over winter 2009–2010, and only one split in the following spring, resulting in 31.4% of the original number of colonies remaining 1 yr after establishment of the colonies from 1.3-kg packages (Fig. 4). Similarly, colonies that were originally controls, but were treated with SuperBoost during the fall and spring build-up feeding periods (CTT) had slightly (but not significantly) better survival after 1 yr.

Colonies that were treated three times in succession with SuperBoost after establishment, but were not treated with SuperBoost during the fall and spring build-up feeding periods (TCC) also were numerically, but not significantly better than CCC colonies that did not receive SuperBoost at any time (Fig. 4). In contrast to the other three cohorts, TTT colonies that received SuperBoost during all three treatment periods suffered the least mortality of all treatment groups. The percentage of TTT colonies surviving fell to 57.2% by 19 March 2010, but splits in the spring (Table 4), apparently enabled because of the vigor of TTT colonies that survived the winter, resulted in substantial replacement of colonies that died, resulting in only 20% decline in the population of colonies over the entire year.

These results dispel the hypothesis (frequently raised in discussion with beekeepers and researchers) that exposure to brood pheromone in the fall could inhibit the production of winter bees that are best prepared for overwintering survival (Matilla and Otis 2007). Similarly, the year-long better performance by SuperBoost-treated than control colonies seems to dispel any suggestion that prolonged exposure to brood pheromone harms the queen or the worker bees. We hypothesize that the energy expended in greater work performed by foragers (Pankiw et al. 1998, 2004; Pankiw and Page 2001; Pankiw and Rubink 2002; Pankiw 2004, 2007), and increased oviposition by the queen (Sagili and Pankiw 2009), is compensated for by enhanced vigor resulting from the more nutritious diet fed to the queen and the larvae (Peters et al. 2010).

Economic Analysis.

Table 6 summarizes the actual sequential costs and benefits of treating 100 newly acquired package bee colonies six or (hypothetically) five times with SuperBoost for 1 yr. Only costs and benefits above and beyond those accruing to untreated control colonies are included in the analysis. The analysis assumes that Canadian and U.S. dollars are at par. The cost of each treatment with SuperBoost includes $5.00 for the product itself and a labor cost of $1.00 per colony, very conservatively assuming that it costs a beekeeper $1.00 every time a colony is opened and a treatment is applied. Thus the initial financial outlay was substantial, $1,800 for six treatments and $1,200 for five treatments.

Table 6

Analysis of costs and benefits above and beyond those that would occur with no SuperBoost treatment that would accrue to Campbell's Gold Honey Farm and Meadery

The production of honey from successive pollination of blueberries, raspberries, and cranberries allows the sale of highly priced varietal honey from an on-farm store at $17.75/kg ($8.05/lb). Therefore, even though the increase in honey production was relatively small (1.68 kg [3.7 lb]), the extra production yielded an increase in honey sales of $2,982, and a net profit at the end of summer of $1,182 for a six-treatment regime and $1,782 for a five-treatment regime. In subsequent experiments (data not shown) and a study in New Zealand by Foster et al. (2011a), much higher increases in yield have been found. Therefore the increase in income from additional honey harvested in this study is very conservative.

Further costs in the fall and winter included $1,800 for three additional treatments with SuperBoost and $130 for extra BeePro consumed during fall and spring build-up feeding. These costs returned the running balance to negative figures. However, they were substantially offset in SuperBoost-treated colonies by the large benefit of having to replace only 20 colonies to reach the starting total of 100 colonies, compared with 69 replacements needed for untreated control colonies. Thus the net balance was positive for both six and five SuperBoost treatments: $6,602 and $7,202 (or $66.02 and $72.02 per colony), respectively.

In conclusion, our results demonstrate that placement of newly established package bee colonies on a year-long treatment regime with SuperBoost can be beneficial to the colonies and can provide a substantial economic benefit to the beekeeper. Treatment with SuperBoost resulted in enhanced honey production, larger colonies before overwintering, and greater numbers of splits during the spring build-up period. One year after establishment, the number of viable colonies in the cohort treated six times with SuperBoost was 80.0% of the starting number of colonies, 2.5 times the proportion in the untreated control cohort. Therefore, the future income from honey and pollination services would be greater, and the replacement costs and labor needed to establish new colonies would be considerably less than if SuperBoost were not used. A similar prolonged study has not been done with established colonies. However, in other studies with established colonies, positive results caused by SuperBoost treatment on pollen foraging (Pankiw et al. 2011), honey production (Foster et al. 2011a), and colony strength after spring build-up and fall feeding (Moeri et al. 2011; R. R. Sagili, personal communication) suggest that placing established colonies on a year-long SuperBoost treatment regime could have beneficial effects similar to those shown for package bee colonies.

We thank Contech Enterprises Inc. for supporting the research.