Overview of Feral and Managed Honeybees in Australia

Distribution, Abundance, Extent of Interactions with Native Biota, Evidence of Impacts and Future Research
David C Paton, Department of Zoology, The University of Adelaide
for Australian Nature Conservation Agency
Environment Australia, May 1996
ISBN 0 6422 1381 X

2. The Distribution and Abundance of Honeybees in Australia

Managed hives of honeybees

Honeybees Apis mellifera were first introduced to Australia in 1810 by Samuel Marsden who imported an unknown number of colonies from England. These, however, failed to establish and eventually died out as did the hive introduced to Tasmania in 1821 (Ziegler 1993). A second introduction in 1822 to mainland Australia was successful and further introductions to other parts of the continent over the next 50-60 years introduced other races and apiaries were established in each state (Eagland 1958; Manning 1989, 1992; Wills 1989; Oldroyd et al.1993; Ziegler 1993). Despite these introductions honeybees were probably not widespread until about 1930 when beekeepers were able to travel more easily; from then the honeybee industry grew steadily (Manning 1992). For example, the average yearly production of honey in Australia from 1921-25 was 3200 tonnes but by 1949- 54 there were some 442,000 commercially managed hives producing about 12,500 tonnes of honey annually (Eagland 1958). By 1971-72 524,000 hives in Australia produced 22,000 tonnes of honey per annum (Read 1972) and in the 1980s approximately 546,000 hives produced 25,400 tonnes of honey annually (Wills 1989). Thus since the 1920s honey production has increased eight-fold despite further and substantial clearance of native vegetation (eg Harris 1976; Cook 1987; Saunders et al.1990).

Wills (1989) compiled statistics on the numbers of beekeepers and managed hives in each of the states in the 1980s (table 1). Of the states, New South Wales (NSW) supported the largest number of beekeepers and the most beehives (over 210,000). Queensland (Qld), Victoria (Vic) and South Australia (SA) each had around 84, 000 hives, Western Australia (WA) 44,000, and Tasmania (Tas) 12,000. Statistics for the Northern Territory (NT) and Australian Capital Territory (ACT) were not provided but these territories have at most a few thousand managed hives.

Table 1
Approximate numbers of hives of honeybees maintained by beekeepers in each Australian state during the 1980s (based on Wills 1989). The table shows the average numbers of hives kept per year in each state over a period of up to five years between 1979 and 1986; numbers of managed hives fluctuated between years by up to 30,000 hives or 25% of the mean in some states. Only beekeepers with at least 40 hives are included. At least another 3000 amateur beekeepers kept fewer than 40 hives each (Read 1972); these are included in no. of hives. The discrepancy between the state and Australian totals is because different five-year periods were used
 
State
No. of beekeepers
No. of hives
Queensland
370
83, 110
New South Wales
818
211, 614
Victoria
408
84, 487
Tasmania
59
12, 480
South Australia
344
85, 406
Western Australia
151
43, 640
Australia
2213
546, 200

Precise information on the distribution of managed hives in each of the states is not available. This reflects the difficulty of collating accurate information on the movements of beekeepers; the movements of apiaries vary from year to year with the availability of floral resources. Manning (1992, 1993b), however, provides information on the distribution of 3475 apiary sites in WA (2436 registered apiary sites on crown land; 1039 sites on private land based on responses from 69% of WA's beekeepers) and some information on patterns of seasonal and annual use. The apiary sites were clustered into a number of key regions mostly in an area south of Geraldton and west of Albany. Both commercial and non-commercial beekeepers used these sites and not all sites were used in each year. Sites used by commercial beekeepers were usually stocked with about 100 hives for an average of 12 weeks, while those used by noncommercial beekeepers were stocked on average with about 60 hives for 26 weeks. A few sites were stocked with beehives throughout the year and others for as little as 3-4 weeks at a time.

Elsewhere in Australia, commercial beekeepers typically stock apiary sites at similar densities (ca 100 hives) for similar durations (2-4 months; table 2) and have some sites that are used each year and others that are used less depending on flowering levels. Detailed knowledge of the distribution of managed colonies, however, is not available. Most states hold files that list the locations of registered apiary sites within reserves but do not maintain records on the frequency and extent to which those sites are used by beekeepers.

Table 2
Patterns of use of floral resources by commercial beekeepers in different parts of Australia. The symbols show the months of the year when particular floral resources are used by beekeepers. Only a selection of resources are shown to illustrate patterns of use. Information- provided by beekeepers, State Apiary Officers and others (eg Wills 1989; Manning 1992). No information provided on patterns of use of floral resources by commercial beekeepers for NSW and Qld. Most regions broadly defined and local movements of apiaries between floral resources may occur. Thus these patterns may overestimate the length of time an apiary site is occupied.
 
State and region
J F M A M J J A S O N D
Main floral resources
TAS                          
N & NW
O
O
 
 
 
 
 
 
 
 
O
O
blackberry, clover
E coast
 
 
 
 
 
 
 
 
O
O
O
 
Eucalytpus globulus
S
O
O
 
 
 
 
 
 
 
 
O
O
prickly box, clover etc
W
O
O
O
O
 
 
 
 
 
 
 
 
Eucryphia spp
SA
 
 
 
 
 
 
 
 
 
 
 
 
 
SE region
 
 
 
 
 
 
 
 
 
 
 
 
 
Salt Creek/Ngarkat
 
 
 
 
O
O
O
O
 
 
 
 
Banksia ornata
Keith-Coonalpyn
 
 
 
 
 
 
 
O
O
O
O
 
various eucalypts
Bordertown-Padthaway
O
 
 
 
 
 
 
 
 
 
O
O
Eucalyptus leucoxylon
Peebinga-Lameroo
O
O
O
O
 
 
 
 
O
O
O
O
various mallee eucalypts
Keith-Bordertown
O
O
O
 
 
 
 
 
 
 
 
 
lucerne
Bangham area
 
 
O
O
 
 
 
 
 
 
 
 
E. baxteri
Millicent area
O
O
O
 
 
 
 
 
 
 
O
O
conola, carrots, clover
Robe area
 
 
O
O
 
 
 
 
 
 
 
 
Melaleuca lanceolata
Riverland
 
 
 
 
 
 
 
 
 
 
 
 
 
Loxton-Waikerie
O
 
 
 
 
 
 
 
 
 
 
O
various eucalypts/mallees
Loxton-Waikerie
 
 
 
 
 
 
 
O
O
O
 
 
almonds, oranges, apricots
Mt Lofty Ranges/Adelaide Plains
 
 
 
 
 
 
 
 
 
 
 
 
 
McLaren Vale
 
 
 
 
 
 
 
O
O
 
 
 
almonds
Mt Lofty Ranges
O
 
 
 
 
 
 
 
 
 
 
O
Eucalyptus camaldulensis
Fleurieu Peninsula
 
O
O
O
 
 
 
 
 
 
 
 
E. baxteri, E.,obliqua
Mt Lofty Ranges
 
 
O
O
O
 
 
 
 
 
 
 
E. microcarpa
Mt Lofty Ranges
 
 
 
O
O
O
 
 
 
 
 
 
E. cosmophylla
St Vincent Gulf
 
O
O
 
 
 
 
 
 
 
 
 
Avicennia officianalis
Mt Lofty Ranges
 
 
 
O
O
O
O
O
O
O
O
O
E. leucoxylon
Mid-North/Flinders R
 
 
 
 
 
 
 
 
 
 
 
 
 
S Flinders Ranges
 
 
 
 
 
 
 
 
O
O
O
O
E. leucoxylon, Echium
S Flinders Ranges
O
O
 
 
 
 
 
 
 
 
 
 
Eucalyptus cladocalyx
S Flinders Ranges
O
O
O
 
 
 
 
 
 
 
 
 
E. microcarpa
N Spencer Gulf
O
O
O
 
 
 
 
 
 
 
 
 
Avicennia officinalis
Mid-North
O
O
O
 
 
 
 
 
 
 
 
O
Strawberry clover
Mid-North
 
 
 
O
O
O
O
O
O
O
O
O
E. leucoxylon
Mid-North
 
 
 
 
 
 
 
 
O
O
O
O
Echium
Mid-North
 
 
O
O
 
 
 
 
 
 
 
 
Eucalyptus microcarpa
Yorke Peninsula
 
O
O
O
O
 
 
 
 
 
 
 
Melaleuca lanceolata
Yorke Peninsula
 
 
 
 
 
O
O
O
O
 
 
 
E. diversifolia, Acacia
Yorke Peninsula
O
O
 
 
 
 
 
 
 
 
O
O
various mallee eucalypts
Eyre Peninsula
 
 
 
 
 
 
 
 
 
 
 
 
 
Elliston area
 
 
 
O
O
O
O
O
O
 
 
 
E. diversifolia
lower Eyre Peninsula
O
O
O
 
 
 
 
 
 
 
 
 
E. cladocalyx
scattered sites
O
O
O
O
O
O
O
O
O
O
O
O
mallee eucalypts
Kangaroo Island
 
 
 
 
 
 
 
 
 
 
 
 
 
forests
O
O
O
 
 
 
 
 
 
 
 
 
E. cladocalyx
mallee-heaths/farmland
 
 
 
O
O
O
 
 
 
 
 
 
E. porosa, E. cosmophylla
mallee-heath/scrub
 
 
 
 
 
 
 
 
O
O
O
O
E. diversifolia/ fasciculosa
WA
 
 
 
 
 
 
 
 
 
 
 
 
 
Perth-Geraldton (N sandplain)
 
 
 
 
O
O
 
 
 
 
 
 
various Banksia
Perth-Geraldton (N sandplain)
 
 
 
 
 
O
O
O
O
O
 
 
various shrubs/trees
Payne's Find area
 
 
 
 
 
 
 
 
O
O
O
O
E. loxophleba
Eastern Goldfields
O
 
 
 
 
 
 
 
 
O
O
O
various eucalypts
Perth-Bunbury (forests)
O
O
O
 
 
 
 
 
 
 
O
O
E. marginata, E. calophylla
Bunbury-Walpole area
 
 
O
O
 
 
 
 
 
 
 
 
E. diversicolor
Ravensthorpe area
O
O
O
O
O
O
 
 
 
 
O
O
E. occidentalis/ coast.hths
N State Forests
O
 
 
 
 
 
 
 
 
 
 
O
E, marginata, B. grandis
S Forests
 
O
O
O
O
O
 
 
 
 
 
 
E. diversicolor, E. calophylla
Yanchep/Dongara (coastal pln)
 
 
 
 
O
O
O
O
O
O
O
 
Banksia, Leucopogon, Dryandra, Hakea
NT
 
 
 
 
 
 
 
 
 
 
 
 
 
Top End
 
 
 
 
 
 
 
O
O
O
O
O
Melaleuca sp
Katherine area
O
O
O
 
 
 
 
 
 
 
O
O
melons, other crops

The presence of 100 hives at an apiary suggests that there will be intense use of floral resources immediately around that apiary. However, although the density of colonies at the site of an apiary is high, bees from these hives forage over an extensive area around the apiary. Overseas figures 2-3 km of their hives but may travel as far as 14 km when resources are poor (Gary et al. 1977; Eickwort and Ginsberg 1980; Visscher and Seeley 1982; Seeley 1985; Winston 1987; Roubik 1991). Sugden and Pyke (1991) estimated that honeybees foraged out to distances of about 2 km from hives in Nadgee Nature Reserve, NSW during spring and summer, while Paton et at. (MS 1; unpubl.) showed that most bees from commercial apiaries foraged mainly within 1 km of their hives while exploiting floral resources of suggest that most honeybees forage within Banksia ornata in Ngarkat Conservation Park (CP) during winter. The bees travelled further distances from their hives in warmer weather and when resources were less abundant. If we assume that most honeybees forage within 2 km of their hives then the hives present at an apiary are effectively distributed over an area of about 12 km² (1200 ha). With a hundred hives this is equivalent to an effective density of 0.1 colonies per hectare (col/ha). In most states the distances between commercial apiary sites on public land (reserves) is around 1.5-2 km. If all sites are occupied at the same time then effective densities for managed hives may approach 0.2-0.3 col/ha but usually not all sites are occupied simultaneously (G Cotton pers. comm.; D Paton pers. obs.). In Queensland registered sites on public land are 0.8 km apart while in WA they are usually at least 3.0 km apart (information from S Bryce, CALM, WA).

Feral colonies of honeybees

Historical information on the distribution and abundance of feral honeybees in Australia is largely lacking. Feral colonies would have established shortly after the introduction of managed hives and slowly spread away from these sites. In Tasmania, feral swarms were first noted in 1838 just 7 years after successful establishment, and before long were reported widely over the island (Ziegler 1993). Similarly, in New Zealand feral colonies of A. mellifera were reported occupying many hollow trees within nine years of the initial introduction (Donovan 1980). How quickly they spread across the Australian mainland is difficult to judge. The distances moved by swarms from a parent colony in other countries varies from as little as 50 m to over 4 km, the distances depending on the race of honeybee and presumably on the availability of resources (Winston 1987). Most swarms, however, settle within 1 km of the parent colony. Given sufficient time feral colonies may have preceded the spread of commercial apiaries into some areas. For example, Manning (1992) states that swarms of honeybees were first seen in the Ravensthorpe area of WA in about 1925 but commercial beekeepers did not start operating in that area until 1930. Feral colonies are now widespread though patchily distributed over the Australian mainland (table 3). Areas believed to have high densities include: mallee regions and riparian habitats along watercourses in north-eastern Victoria particularly redgum/black box habitats; coastal woodland areas of southwestern WA; coastal regions, creeklines and sugar gum woodlands of Kangaroo Island; woodland areas in the south-eastern parts of South Australia (particularly west of Keith); parts of Yorke Peninsula and the west coast of South Australia; and possibly the wetter coastal woodland areas of Queensland. Areas assumed to have low densities include alpine areas and inland areas away from water courses. Factors that may limit the distribution of feral colonies in these regions include: availability of water; availability of food; availability of suitable hollows; various bee diseases; and cold or wet weather. Quantitative data on the numbers of feral colonies in different parts of Australia are largely lacking. Manning (1992) collated qualitative information on the numbers of feral colonies in regions of WA. Each region was defined by a 1:50,000 Topographic Series map (equivalent to about 600 km²). Feral colonies were present in virtually all regions throughout the south-western third of WA. However, the relative densities estimated by beekeepers on a crude scale of low, medium and high often varied within a region and these densities were not defined. Paton et at.(MS2) have counted and monitored feral colonies in an area of about 200 ha around the SA National Parks and Wildlife Service (NPWS) Headquarters at Rocky River within Flinders Chase National Park (NP) on Kangaroo Island. The area searched consisted of extensive areas of tall sugar gum, Eucalyptus cladocalyx forest and woodland interspersed with stands of E. divers folia, E. ovata and E. fasciculosa and areas of mallee heath and modified grasslands with scattered trees. In this area densities fluctuated from 0.2 to 0.4 col/ha over a three year period but were usually maintained near 0.3 col/ha. Within localised areas the highest densities exceeded 10 col/ha. In other parts of South Australia, densities ranged from 0.001- 0.004 col/ha (in mallee heath) to 0.1-0.4 col/ ha in remnant eucalypt woodland in the Mt Lofty Ranges (table 4; Paton et al.MS2, unpubl.). The only other data available are for the river redgum, E. camaldutensis and black box, E. largiflorens habitats of Wyperfeld National Park where densities have been measured at 0.77 col/ha (Oldroyd et al. 1994).These densities, however, overestimate the densities within the reserve, since the survey was restricted to riparian woodlands known to 16 house feral colonies. Riparian woodlands represent only a small proportion of the reserve, and the other more extensive habitats within the reserve are likely to be relatively free of feral colonies. Limited data also suggest that the densities of feral colonies of honeybees in other countries are typically no more than a few colonies (0.5- 8) per km² though in some localised areas densities equivalent to over 100 col/km² have been recorded (Visscher and Seeley 1982; Roubik 1982a; 1988, 1991; Schneider and Blyther 1988; Danka et at. 1992, 1994).

Table 3
Qualitative information on the distribution of feral colonies of honeybees in Australia.
 
State General distribution Areas with high(est) densities Areas with low(est) densities Sources
Tas throughout   higher altitudes K Ziegler
H Ayton
Vic throughout NW mallee regions and riparian habitats ironbark woodlands; NE regions; drier woodlands in SW wetter forests of E Vic G Arnold
NSW throughout   alpine areas G Pyke
SA throughout SE remnant woodlands; Murray Mallee; Kangaroo Island; Yorke Peninsula; West Coast arid interior away from watercourses D Woodward
D Paton
G Cotton
SA Apiarists
WA most of state SW regions & coastal plains virtually absent from dry desert country E of Kalgoorlie and N of Geraldton K Spurge
R Manning
L Allan
NT widespread in low numbers along water courses; rural habitats near Darwin & Katherine; Douglas-Derby area patchy in Top End arid areas away from water courses A Anderson
B Walsh
M Fleming
Old widespread more likely in eucalypt forests than rainforests coastal, high rainfall areas; open eucalypt forests and woodlands Cape York Peninsula P Arming
T Bartareau
Table 4
Abundance of feral colonies in locations in SA and Vic. Range of values represent annual fluctuations in numbers of colonies recorded in tIe study area. Note that in most cases only small areas have been systematically searched and densities calculated from those studies should be treated cautiously. All data collected by Paton et al. (unpubl.) except for Wyperfeld NP data from Oldroyd et al.(1994).
 
Location
Habitat
Area (ha)
No. of feral colonies
Col/ha
Cromer CP (SA) open woodland
15
6
0.4
Scott CP (SA) open woodland
9
1
0.1
Ngarkat CP (SA) mallee-heath
3600
4
0.001
Mt Rescue CP (SA) mallee-heath
2000
7
0.004
Flinders Chase NP (SA) mallee-forest
200
45-80
0.2-0.4
Wyperfeld NP (Vic) riparian woodland
35
27
0.77

Differences between feral and managed colonies of honeybees

Feral and managed colonies of honeybees are often assumed to differ in size and resource acquisition as well as in patterns of dispersion. For example, feral colonies are often assumed to be smaller than managed colonies because of the limited capacity of many of the hollows and cavities they use. Consequently on a per hive basis, managed colonies may harvest more floral resources than a feral colony. In addition, beekeepers continually remove surplus honey and pollen from managed hives stimulating increased foraging activity and increased collection of food resources. Beekeepers also regularly shift their hives to new resources when floral resources decline in an area. Consequently the strength of managed hives may be maintained more consistently throughout the year than feral colonies that experience periods of limited floral resources.

Seeley (1985), Thorp (1987) and Winston (1987) provide general information on the sizes of feral colonies of honeybees in North America, where feral colonies ranged in size from 10, 000-30,000 bees, occupied nest cavities that ranged in volume from 15-80 L and produced on average 20 kg of honey per year. The cavities used usually had small (<75 cm2) entrances at the bottom. Feral colonies usually swarmed at least once per year and survival of these swarms was low during the first year (25%) but subsequently high (80%) for those that survived the first year. For comparison, commercially-managed colonies in North America usually consisted of 20,000-60,000 bees, had hives with volumes of 125-250 L and produced 50-100 kg of honey per annum. Survival was high and only about 25% of the colonies in commercial apiaries swarmed each year. Sizes of commercially-managed hives in Australia are similar, with the most productive hives containing at least 40,000 bees (Winn 1972; Turner et at.1972).

North American, South American, European and African statistics for feral colonies are often used to estimate sizes of feral colonies in Australia (Wapshere 1988; Stace 1988; Manning 1989; 1993a; Oldroyd et al.1994). However, there have been no measurements of the sizes of cavities used by feral honeybees in Australia and no measures of the sizes of colonies in terms of total numbers of bees. Some estimates suggest feral colonies may contain as few as 3000-5000 honeybees (J Davies in Wills 1989) yet there are also reports of individual colonies producing comb at least 1.5 m in length (Pyke 1990, Woodward 1993). A colony with a comb of that size would probably contain over 30,000 bees. In part, the lack of information on the sizes of feral colonies in Australia reflects the difficulty of measuring cavities particularly when these are occupied by colonies of bees. Paton et al.(MS2, unpubl.) used a different technique to estimate the size of feral colonies. In parts of Flinders Chase NP they estimated the strength of feral colonies at different times of the year by recording the rates at which honeybees returned to feral colonies at regular intervals (45 min) throughout the day. In this area individual colonies differed in strength by up to 10 fold. Strong colonies had over 60,000 bee returns per day in goodweather during late summer when E. cladocalyx was flowering, while weaker colonies managed less than 20, 000 bee returns per day. Honeybee activity was lowest in winter when strong hives had 20,000 bee returns' per day and weaker colonies fewer than 2000 bee returns per day. The proportion of bees returning with pollen also varied seasonally and between colonies. Typically 20-50% of the bees that returned had collected pollen, but many of these had also collected considerable volumes of nectar (Paton et al.MS2, unpubl.).

The quantities of nectar and pollen carried home by these bees were also measured. For pollen loads this involved removing the hindlegs of pollen-gathering bees, and placing these inn small plastic vials in 1 mL of lactophenol. The pollen present was then identified and counted in a haemacytometer. Nectar loads were determined by gently squeezing the bees and collecting the fluid in capillary tubes. The volume of fluid was then measured and the sugar concentration determined with a 0-90% Otago sugar refractometer. Volumes and concentrations were converted to mg of sugar. The actual net quantities of sugar collected were determined by subtracting the quantities that samples of - honeybees carried with them when they left the hive from the loads that bees carried home.

During periods of moderate to high flower abundance and suitable weather, feral colonies harvested on average between 100 and 300 g of sugar/day (equivalent to 120-360 g honey/ day) in Flinders Chase NP. During winter when flowers were less abundant and weather conditions and ambient temperatures restricted foraging time, colonies harvested as little as 10 g of sugar/day on days when foraging was possible. Some small colonies only harvested about 2 g of sugar on those days. Poorer foraging conditions and more dilute nectar contributed to the lower returns in winter since individual bees returned with smaller volumes (10-11N1) and more dilute nectar 18 (14-20% sugar wt/wt), equivalent to only 1.5- 2.0 mg sugar/load. At other times in the year each bee returned with an average load of 5-9 mg sugar, 13-19 pl at a concentration of 25- 36%. These loads are typical of the average loads collected by honeybees in other countries (eg Rinderer et a1.1984), though Roubik et al.(1986) reported that African honeybees (A. mellifera scutella) collected average loads of 30 0 in lowland forest in Panama. Over a year the gross intake for these feral colonies was equivalent to about 40 kg of honey; this is not the surplus available but the total produced and much of it will be consumed by the bees themselves. For example, a honeybee requires about 0.7 mg sugar/h or 15 mg/day (Winston 1987), so a colony of 10,000 bees would need 150 g/day or 55 kg/year. These calculations suggest that the average size of feral colonies in Flinders Chase on Kangaroo Island is less than 10,000 bees (averaged over a year).

Comparable figures for managed hives in Australia are not available. However, records are kept on the quantities of honey produced by beekeepers. Since the 1970s the quantities of honey harvested from managed beehives in Australia has averaged around 50-60 kg per year per productive hive (Read 1972; Wills 1989). About 75-80% of the managed hives in Australia are productive (Read 1972). However, when managed properly 180-200 kg can, be harvested per productive hive in a year (Winn 1972; G Cotton pers. comm.). These quantities are typical of the average quantities produced in other countries that range from 50-100 kg per productive hive (eg Thorp 1987). In addition to the quantities of honey harvested from hives an additional 60-80 kg of honey are estimated to be consumed by the colony in a year (Winston 1987). Managed hives, therefore, produce something of the order of 100-200 kg of honey/year and must harvest, on average, between ,300 and 600 g of sugar/day throughout the year. During favourable conditions managed colonies of honeybees will harvest 1-2 kg/day (G Cotton pers. comm.). This rate of intake can be met with about 150, 000 foraging trips, about 3 times the number of foraging trips made by a strong feral colony on Kangaroo Island. Winston (1987) suggests 163,000 foraging trips can be made daily by workers from strong colonies of managed hives. Thus feral colonies do not appear to reach the same size as managed colonies. Information on the quantities of pollen harvested by feral and managed honeybees in Australia is also limited. According to Winston (1987) estimates of a managed colony's annual requirements for pollen vary from 15 to 55 kg/ year. Given this, managed colonies of honeybees need to harvest 40-150 g of pollen per day throughout the year, and given that some days are not suitable for foraging substantially more than those amounts needs to be harvested when conditions are suitable. The maximum daily intakes that have been recorded for managed hives in Australia has been around 1-1.5 kg of pollen/day during a honey flow on E. camaldulensis (Winn 1972; Doull 1973). This is equivalent to about 250, 000 million Eucalyptus pollen grains (1 million pollen grains weigh about 5 mg; Doull 1973), or equivalent to the bees harvesting all of the pollen produced by about 500,000 Eucalyptus flowers. The amounts of pollen produced by individual flowers of various eucalypts varies from 160,000 to 820,000 pollen grains depending on the species (Paton unpubl.).

Five feral colonies in Flinders Chase NP were estimated to collect on average about 20 g of pollen/day but varying through the year from 12 g to 39 g (Paton et a1.MS2, unpubl.). The strongest feral colony harvested an estimated 105 g of pollen/day when E. cladocalyx was flowering (Paton unpubl.). These intakes suggest that feral colonies in Flinders Chase NP, on average, consume about 7 kg of pollen annually and substantially less than that needed by a managed colony. These estimates of pollen harvest suggest that feral colonies are substantially smaller than managed colonies.

Summary

Feral and managed colonies of honeybees have been present in Australia for about 160 years, but their distribution and abundance has increased dramatically over the last 60 years. Information on the current distribution and abundance of feral and managed colonies of honeybees in Australia is limited. There are over 540,000 managed colonies of honeybees in Australia and an unknown number of feral colonies. Managed and feral honeybees are present in all states and territories but are largely absent from alpine areas and from arid inland areas away from water.

Managed colonies of honeybees are patchily distributed in time and space. For example, managed colonies are usually shifted into areas of woodland, forest or heath for 2-4 month periods to exploit flowering peaks of certain plants (often species of Eucalyptus) before being shifted to other areas, including agricultural areas to pollinate horticultural crops. Although around 100 hives are usually _stationed at an apiary at one time, bees from this apiary will forage over an extensive area around this site with individual bees foraging several kilometres from their hives. This reduces the effective density of hives at an apiary to about 0.1 col/ ha, a density not dissimilar to the densities recorded for feral colonies. Feral colonies are also patchily distributed but quantitative data are limited to a few relatively small areas. In those areas densities have been as low as 0.001 col/ha and as high as 0.77 col/ha. The high densities reported in some areas may reflect the localised availability of suitable nesting habitat (hollows) for colonies and not reflect the functional or effective density of colonies in an area as a whole. Information on the sizes of feral colonies in Australia and on the population dynamics of feral colonies is limited to just one ongoing (and as yet unpublished) study on feral honeybees in Flinders Chase National Park on Kangaroo Island. Feral colonies were present throughout the year in this area and maintained a fairly constant density of 0.3 col/ha. These feral colonies, however, were generally smaller than managed colonies, based on the numbers of foraging honeybees returning to colonies and on the quantities of nectar and pollen that they harvested.

When present in an area a managed colony of honeybees was likely to harvest 3-5 times the quantity of nectar and pollen being harvested by a feral colony.

Considerably more information on the distribution and abundance of both feral and managed colonies of honeybees is still needed but measuring densities alone will not be sufficient to assess the likely effect of honeybees on natural systems. Ideally the distribution and abundance of floral resources that are used by honeybees and the extent to which those resources are limiting to both native fauna and honeybees must also be established. Such assessments should also extend to assessing the availability of nesting sites for feral colonies and whether honeybees compete with any native fauna for these resources as well. The first step in this process involves identifying the native plants and animals that now interact with honeybees.