To investigate Eucalyptus (gum tree) pollen allergy in children in relation to geography, particularly vegetation, and its relationship to asthma.
Males (n = 180) and females (n = 200) aged 9 to 14 participated. Some were healthy (asymptomatic), some had asthma, and some had other symptoms associated with atopy. School students were from three urban coastal schools and one school from a nearby semi-rural elevated area (range) near Brisbane, Australia. Coastal and range locations featured different distributions of Myrtaceae family vegetation (including Eucalyptus, Melaleuca, Leptospermum species). Skin prick test (SPT) responses to 15 commercial allergens were compared. As well, responses from coast versus range groups, and ‘asthma’ (n = 97) versus ‘healthy’ status (n = 107) groups, were compared.
SPT responses (≥3mm wheal diameter) indicate that children with asthma are 31.1 times more likely to be allergic to Eucalyptus pollen extract (OR: 31.1; 95%CI 4.1- 235.7) compared to healthy children. Dust mite (p = .018), Eucalyptus (p = .046) and cockroach (p = .047) allergen SPT responses (wheals ≥3mm) were significantly greater in participants located on the coast versus range as determined by Fisher’s Exact Test (α .05). For each location, percentage of positive responses (wheals ≥3mm) was greatest for ‘dust mite’ (30.9%-46%), ‘cockroach’ (18.1% -35%) and ‘Bermuda grass’ (10.6%-19.4%).
The results support the hypothesis that proximity to Myrtaceae vegetation is related to positive SPT response and that Eucalyptus is an important allergen for children with asthma. Substantial response to olive allergen, in the absence of olive trees, suggests that the response may be driven by substances in other plants, perhaps Melaleuca quinquenervia, which abounds in coastal areas.
Citation: Gibbs JEM (2015) Eucalyptus Pollen Allergy and Asthma in Children: A Cross-Sectional Study in South-East Queensland, Australia. PLoS ONE 10(5): e0126506. https://doi.org/10.1371/journal.pone.0126506
Academic Editor: Kazuhiro Ito, National Heart and Lung institute, UNITED KINGDOM
Received: September 16, 2014; Accepted: April 2, 2015; Published: May 4, 2015
Copyright: © 2015 Jane E. M. Gibbs. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited
Data Availability: All relevant data are within the paper.
Funding: The South East Queensland Regional Organization of Councils (SEQROC)funded the project.The National Research Centre for Environmental Toxicology (now ENTOX) provided a scholarship. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The author declares that no competing interests exist.
Asthma prevalence in Australia is high compared to other countries and Queensland’s estimate of the ‘current asthma’ rate of 11.8% exceeds the national average of 9.9%, about two million people nationwide.
For more than 40 years in Brisbane, Queensland’s capital, seasonal asthma peaks in autumn with a smaller one in spring, have been identified .Those peaks are not well understood nor is the high prevalence of asthma in Australia and New Zealand generally. Inside homes, avoidance of house-dust mite and parental smoking, in conjunction with breast-feeding and increased ingestion of omega-3 fatty acids, has shown promising reduction in wheeze in children . The association between asthma and cockroach allergy has also been established  but has only limited explanatory value for ‘current asthma’ . Outside the home fungal spores, pollens  and anthropogenic pollutants  are associated with respiratory symptoms but adequate explanatory variables are elusive.
Research into the role of Australian native vegetation, much of which belongs to the Myrtaceae family, is lacking. The Eucalyptus or ‘gum’ tree (Fig 1) is a prominent member.
This investigation seeks to enhance the currently poor knowledge base regarding allergy and plants from Australasia, and challenge the assumptions that deter scrutiny of common Australasian plants as possible respiratory symptom triggers.
Myrtaceae plants have been long been regarded as safe garden alternatives to wind pollinated (anemophilous) plants for those with asthma and allergies. This tradition continues despite a paucity of evidence to support that position. Instead of wind, insect (entomophilous) and animal pollinated (zoophilous) plants rely upon the emission of volatile chemicals, largely terpenoids, to deliver specific scent messages to preferred pollinators, often from the brush-like flowers (Fig 2). Some of these substances have been shown to be sensitizing[10–12].
Lewis and Vinay, identify the Myrtaceae family as producing problematic ‘inhalant effects’ in Hawaii. Stablein et al  investigated Melaleuca quinquenervia, introduced to Florida from Australia, and concluded that reports of allergic effects from M.Quinquenervia (cream bottlebrush) were probably due to its cross-reactivity with Bahia grass pollen.
Floral frames for the asthma picture
Northern NSW and South-Eastern Queensland are home to vast numbers of Melaleuca quinquenervia  (Fig 3) which starts flowering (Fig 4) in late summer, peaks in autumn, and finishes in early winter. Flowering coincides with the asthma peak in S.E. Queensland. Casuarina (she-oak or Australian Pine) species are common and can flower from autumn to spring. The majority of prevalent Eucalyptus and Acacia (wattle) species flower in winter to spring in this region although some flower at other times. Leptospermum (tea-tree) species flower mostly in spring. These plants feature on the populated flat coastal strip that separates the Great Dividing Range from the Pacific Ocean.
Travelling inland, as the landform elevates, the vegetation is much more varied. Crops, orchards, rich grazing pastures, mixed and commercial forests and woodlands feature. Eucalyptus, Callitris, Casuarina, Pinus, and Acacia species are often seen.
Brisbane is a sprawling city of over 2 million people. It has grown simultaneously with the popularity of Australian native plants as garden specimens. What began as increased interest in the 1960s became a ‘boom’ by the 1980sand replaced a formerly European garden tradition. Eucalyptus, Melaleuca, Callistemon and Leptospermum species can be seen extensively planted along streets, in gardens, around schools and in most public green spaces.
This research investigates allergen sensitization patterns that may differ according to geography, specifically, the vegetation. Child participants from three coastal schools (average elevation 14m), from greater Brisbane, and a Blackall Range school (elevation 416m) from the Sunshine Coast region (Fig 5) were enlisted. Melaleuca quinquenervia only grows extensively on the coastal wetlands and does not grow naturally on the elevated range area selected. Proximity to large tracts of these Myrtaceae trees determined selection of the coastal schools.
The coastal schools are suburban consisting of mostly detached homes set on blocks of average 600sqm. The Range school in Maleny is part of a rural tourist town of approximately 4000 population where most of the children also live in detached homes on average blocks in town, and some on small acreage.
It was hypothesised that any response to Eucalyptus flower pollen allergen, representative of Myrtaceae family plants, would be similar in the three coastal schools and differ from the Range school because of proximity to Myrtaceae family vegetation, either natural or purposefully planted.
Acacia and Casuarina (Australian Pine) are found similarly in both areas and were not expected to differ between coastal and range schools.
Response to privet was also of interest. It infests the Maleny area but not obviously the urban coastal areas. Olive (Olea), which cross-reacts with privet, was used here due to unavailability of privet allergen. Olives require dry summers, not wet ones that feature in Queensland and are not apparent in the area in question. Linalool, a terpenoid, from olive is also found in Melaleuca  and privet (Ligustrum) (Fig 6) and is the familiar honey scent that attracts bees. Will children from coastal schools respond to olive allergen perhaps because coastal Myrtaceae trees contain similar substances to olive or privet?
Ethics approval was sought and obtained from the Griffith University Human Research Ethics Committee. Informed written consent was obtained from parents or guardians and children before children were tested.
Recruitment of participants
Following ethics approval, children were recruited from four state schools of the same mid-range socio-economic band (designated by the Queensland Education Department): Kippa Ring (KR), Maleny (M), Boondall (B) and Slacks Creek (SC).
Volunteers from grades 4 to7 with and without respiratory and allergic symptoms were requested via the school newsletter. Parents or guardians were promised copies of skin test results. Children were given an ISAAC QUESTIONNAIRE (International Study of Asthma and Allergies in Childhood), a personal details form and an “informed consent” form for completion by carers. Children were tested, each school in turn, in one session each, one week apart in September/October 1998. The same staff administered allergens and were utilised in all sessions. Table 1 shows the number and sex of the children recruited at each school.
All volunteers were accepted and assessed by the attendant respiratory physician and spirometry recorded. Asthma status was allocated on the basis of short interview and medical examination, their spirometry result and their ISAAC questionnaire. All children with a doctor’s diagnosis of asthma were accepted as such. After SPT, children with asthma in the previous 12 months were classified as the ‘asthma’ (n = 97) group and children who had no allergic or respiratory symptoms at all were classified as ‘healthy’ (n = 107). The ‘asthma’ group included children with asthma of varying severity based upon the number of asthma episodes in the last 12 months: ‘mild’, 1 to 3; ‘moderate’, 4 to12; and ‘severe’, more than 12. Others were designated ‘dry cough’ (dry cough only), ‘atopic’ (hay fever and eczema symptoms only) and ‘inactive asthma’ (asthma not in last 12 months). No further analysis of these last three groups features here. Health status groups are shown in Table 2.
All children were skin prick tested with commercial allergens in class groups according to names in alphabetical order, with a positive histamine and negative saline control. Allergens were dropped onto a grid drawn on the volar forearm, skin pricked with a lancetter in the middle of the allergen droplet. The resultant wheals were compared to controls and the diameter measured after 10 minutes.
Allergens were obtained from Bayer Corporation, Sydney. Melaleuca and Ligustrum (privet) were requested but were not available. Allergens used were (as shown on the label of each):Standardized mite Dermatophagoides farina; Cockroach Mix 6585 (American, German); Standardized cat pelt; Dog hair—dander; Hormodendrum cladosporioides; Alternaria tenuis; Johnson grass, Sorghum halepense; Bahia grass, Paspalum notatum; Ryegrass, perennial, Lolium perenne; Bermuda grass, Cynodon dactylon; Olive, Olea europea; Acacia, Golden, Acacia longifolia; Ragweed, mix GS (giant short); Pine, Aus/SW; She-oak, Casuarina glauca; and Eucalyptus/blue gum, Eucalyptus globulus.
For all SPT measures, a positive result was wheal diameter 3mm or greater that formed after 10 minutes from skin prick. Irregular shaped wheals were calculated by taking the width and adding the length and dividing by two.
Schools groups comparisons
Each school group of children with and without symptoms was compared against each other school. For the four schools, the percentage of positive responses (wheals ≥3mm) for ‘dust mite’ (30.9%M-46%B), ‘cockroach’ (18.1% M-35%SC), ‘Bermuda grass’ (10.6%M-19.4%B) and ‘Bahia grass’ (10.6%M-18.5%B) were highest, and responses for ‘dog’ (1.6%KR-4%B), the lowest. All schools SPT response rates were most uniform for ‘Cladosporium’ (5%SC-6.4%M). These can be seen charted in Fig 7.
Coastal versus range school comparisons
For the range and coast comparisons, three coastal schools were compared, as a group, with Maleny participants. Fisher’s Exact Test applied to range and coastal SPT positive (wheals≥3mm) responses revealed that dust mite, Eucalyptus and cockroach allergen responses, respectively, significantly distinguished coastal from range scores as shown in Table 3 (α set at .05).
No significant differences in severity between coast and range groups were found (t = 1.41, df 95, p = .16). Asthma severity comparisons between coast and range groups were determined by assigning a weighted variable for ‘mild’(1), ‘moderate’(2) and ‘severe’(3) asthma to each participant in those groups. The measure reflects number of episodes of asthma noted on the participant’s ISAAC questionnaire.
‘Healthy’ and ‘asthma’ comparisons
Percentage positive response comparisons between groups comprising only ‘healthy’ and ‘asthma’ groups from all schools combined are shown in Table 4 and charted in Fig 8. Odds of a response to allergens with wheal size ≥3mm by a child with asthma, compared to a healthy child, are also shown in Table 4.
Positive response was greatest for dust mite, cockroach mix, and Bermuda grass for both groups respectively. The children of the ‘asthma’ group demonstrated a higher positive response to all allergens compared to the ‘healthy’ group.
For the ‘asthma’ group, of the plant allergens, Bermuda grass (26%) exceeded Eucalyptus (24%) and Bahia grass (23%) followed by Australian Pine (Casuarina) (22%), Johnson grass (22%) and ryegrass (20%). Cat (19%) and Alternaria (17%) featured among animal and fungal responses.
The allergen response that resulted in the highest odds ratio (Table 4) was Eucalyptus (OR 31) followed by Cladosporium (OR 18). Thus the odds of a SPT response to Eucalyptus, greater than 3mm wheal, was 31 times more for a child with asthma than for a child with no symptoms of allergy or asthma. A post-hoc analysis of Eucalyptus allergen responses (two-way Anova) revealed no significant differences between males or females (F = .052; p = .857). There was no significant interaction between ‘group’ (asthma or healthy) and ‘sex’ (male or female) (F = .029; p = .864). Eucalyptus scores for asthma or healthy groups differed significantly (F = 1586; p = .016). Background data for Table 4 is shown in Table 5. Despite a significant sex difference in group membership, the impact of sex upon Eucalyptus scores is not significant.
The findings support the hypothesis that nearby vegetation may be related to an increased SPT response to those plants commonly found in coastal SE Queensland.
Response to Eucalyptus allergen significantly differed when comparing Maleny to the three coastal schools together (Table 3). This may be due to the large tracts of Myrtaceae family, trees on the coast.
The substantial SPT response is despite the use of commercial allergen Eucalyptus globulus, or Tasmanian Blue Gum, found naturally in Tasmania, 2000 km away, not in S.E.Queensland. SPT response to Eucalyptus does not appear to be species specific perhaps because the volatiles emitted from foliage of local Eucalyptus tereticornis are similar. A possible explanation for the phenomenon comes from Gonzalez et al  in their explanation of how people sensitized to olive pollen can react to unrelated plant pollens. They conclude that allergic reactions to many different sources can be attributable to a few cross-reactive components.
Cross-reactive responses may also explain the substantial response to olive allergen here. Olive growth in Brisbane is restricted to the occasional garden specimen which typically does not fruit. Olive allergen was included in the test battery as a surrogate for privet, a weed on the range area. Neither privet nor olive is present in substantial numbers on the coast. Cross reactivity of olive has been established with privet, pine, birch, mugwort and cypress . Plants from completely different families, genera and species can be cross-reactive. Extensive plantation pine forests that lie between northern Brisbane and Maleny could be relevant in explaining the substantial response in all schools. As well, or alternately, the response may be linked to common volatile terpenoids like linalool which occurs in olive, privet, Melaleuca and Eucalyptus species. Terpene enzymes, such as linalool synthase , 3-carene synthase  and could play a role in cross-reactive allergen responses. Molecular weights of these enzymes are within range of IgE binding components identified in Melaleuca.
Proximity to highways and increased traffic could result in higher response rates for many allergens for the coastal schools compared to Maleny. This is consistent with views by Behrendt et al.  however responses here to Acacia, Alternaria and Cladosporium do not fit this explanation. Some responses are lower than those of the Maleny group.
The olive result is a reminder that the responses to particular plant allergens must be interpreted carefully. The response to olive on the coast, despite is apparent absence, is similar to that for Hirschwehr et al  in their description of allergic responses to ragweed allergen in the absence of likely sensitization to ragweed in Germany. They note that ragweed is rare in Germany. This study highlights the problems due to local responses and cross-reactivity inherent in the interpretation of SPT results.
Results here invite investigation into chemical similarities to other plants associated with asthma and allergies. Given the role of cross-reactions in responses to SPT with plant allergens, is the substantial response to Bahia grass in this study because of its demonstrated cross-reactivity with Melaleuca?
The key to the cross-reactivity door may lie in the organic compounds that make up the volatile organic emissions of plants world-wide. The volatile emissions from Eucalyptus are also found in other species [32–36] associated with allergy and asthma. Eucalyptus, native to Australia, has been planted in U.S.A. since the 1860s and extensive plantations can be found in India, Brazil, China, Africa, Spain, Portugal and Pakistan . It may be time to explore the contribution of floral volatiles from Eucalyptus and other members of the Myrtaceae family, especially Melaleuca, to asthma and allergic response.
Results supported the hypothesis that pollen allergen response from Eucalyptus flowers would differ between coastal and range schools. As well, cross-reactivity of a local plant with olive species, not growing in the area is likely. Analysis of chemical compositions would suggest Melaleuca quinquenervia and/or Eucalyptus species, because of common terpenoid components. Appreciation for the complexity of chemical relationships among different species and genera can improve understanding of national and international patterns of allergic response to plants.
Thanks to Associate Professor Charles Mitchell for supervision, guidance and examining the participants; Dr Shannon Rutherford for assistance; and the children, teachers, parents and carers who participated.
Conceived and designed the experiments: JG. Performed the experiments: JG. Analyzed the data: JG. Contributed reagents/materials/analysis tools: JG. Wrote the paper: JG.
- 1. Australian Centre for Asthma Monitoring. Asthma in Australia 2011. Canberra: 2011.
- 2. Derrick EH. Asthma and the Brisbane climate. Aust NZ J Med. 1972;3:235–346. pmid:4508245
- 3. Masoli M, Fabian D, Holt S, Beasley R. The Global Burden of Asthma. Global Initiative for Asthma, (GINA), 2004.
- 4. Peat JK. Can asthma be prevented? Evidence from epidemiological studies of children in Australia and New Zealand in the last decade. Clin Exp Allergy. 1998;28(3):261–5. pmid:9543073.
- 5. Arruda LK, Vailes LD, Platts-Mills TA, Hayden ML, Chapman MD. Induction of IgE antibody responses by glutathione S-transferase from the German cockroach (Blattella germanica). J Biol Chem. 1997;272(33):20907–12. pmid:9252418.
- 6. Sporik R, Squillace SP, Ingram JM, Rakes G, Honsinger RW, Platts-Mills TA. Mite, cat, and cockroach exposure, allergen sensitisation, and asthma in children: a case-control study of three schools. Thorax. 1999;54(8):675–80. pmid:10413718; PubMed Central PMCID: PMC1745561.
- 7. Rutherford S, Simpson R, Williams G, Mitchell C, McCall B. Relationships between environmental factors and lung function of asthmatic subjects in south east Queensland, Australia. J Occup Environ Med. 2000;42(9):882–91. pmid:10998763.
- 8. Petroeschevsky A, Simpson RW, Thalib L, Rutherford S. Associations between outdoor air pollution and hospital admissions in Brisbane, Australia. Arch Environ Health. 2001;56(1):37–52. pmid:AN 2001:230102.
- 9. NSW AFo. Asthma Friendly Gardens. 2013.
- 10. Matura M, Goossens A, Bordalo O, Garcia-Bravo B, Magnusson K, Wrangsjo K, et al. Oxidized citrus oil (R-limonene): a frequent skin sensitizer in Europe. J Am Acad Dermatol. 2002;47(5):709–14. pmid:12399762.
- 11. Skold M, Borje A, Harambasic E, Karlberg AT. Contact allergens formed on air exposure of linalool. Identification and quantification of primary and secondary oxidation products and the effect on skin sensitization. Chem Res Toxicol. 2004;17(12):1697–705. pmid:15606147.
- 12. Skold M, Hagvall L, Karlberg AT. Autoxidation of linalyl acetate, the main component of lavender oil, creates potent contact allergens. Contact Dermatitis. 2008;58(1):9–14. pmid:18154552.
- 13. Lewis W, Vinay P, Zenger V. Airborne and Allergenic Pollen of North America. Baltimore: The Johns Hopkins University Press; 1983.
- 14. Stablein JJ, Bucholtz GA, Lockey RF. Melaleuca tree and respiratory disease. Ann Allergy Asthma Immunol. 2002;89(5):523–30. pmid:12452213.
- 15. Specht RL. Conservation atlas of plant communities in Australia. Lismore: Centre for Coastal Management in association with Southern Cross University Press; 1995.
- 16. Yates A. Yates Garden Guide Centenary Edition. 37th ed. Sydney: Willliam Collins; 1987.
- 17. Pajaron MJ, Vila L, Prieto I, Resano A, Sanz ML, Oehling AK. Cross-reactivity of Olea europaea with other Oleaceae species in allergic rhinitis and bronchial asthma. Allergy. 1997;52(8):829–35. pmid:9284982.
- 18. Ireland BF, Hibbert DB, Goldsack RJ, Doran JC, Brophy JJ. Chemical variation in the leaf essential oil of Melaleuca quinquenervia (Cav.) S.T. Blake. Biochem Syst Ecol. 2002;30(5):457–70. pmid:AN 2002:303064.
- 19. He ZD, Ueda S, Akaji M, Fujita T, Inoue K, Yang CR. Monoterpenoid and phenylethanoid glycosides from Ligustrum pedunculare. Phytochemistry. 1994;36(3):709–16. pmid:7765002.
- 20. Boland D, Brophy JJ, House APN. Eucalyptus leaf oils. Use, chemistry, distillation and marketing. Sydney: ACIAR/CSIRO 1991.
- 21. Barata EN, Pickett JA, Wadhams LJ, Woodcock CM, Mustaparta H. Identification of Host and Nonhost Semiochemicals of Eucalyptus Woodborer Phoracantha semipunctata by Gas Chromatography Electroantennography. J Chem Ecol. 2000;26(8):1877–95.
- 22. Gonzalez EM, Villalba M, Rodriguez R. Allergenic cross-reactivity of olive pollen. Allergy. 2000;55(7):658–63. pmid:10921466.
- 23. Flamini G, Cioni PL, Morelli I. Volatiles from leaves, fruits, and virgin oil from Olea europaea Cv. Olivastra Seggianese from Italy. J Agric Food Chem. 2003;51(5):1382–6. Epub 2003/02/20. pmid:12590485.
- 24. Street RA, Hewitt CN, Mennicken S. Isoprene and monoterpene emissions from a Eucalyptus plantation in Portugal. Journal of Geophysical Research. 1997;102(D13):15875–87. pmid:AN 1997:546181.
- 25. Fischbach RJ, Zimmer I, Steinbrecher R, Pfichner A, Schnitzler JP. Monoterpene synthase activities in leaves of Picea abies (L.) Karst. and Quercus ilex L. Phytochemistry. 2000;54(3):257–65. pmid:10870179.
- 26. Cseke L, Dudareva N, Pichersky E. Structure and evolution of linalool synthase. Mol Biol Evol. 1998;15(11):1491–8. pmid:2003061907.
- 27. Hoelscher DJ, Williams DC, Wildung MR, Croteau R. A cDNA clone for 3-carene synthase from Salvia stenophylla. Phytochemistry. 2003;62(7):1081–6. pmid:AN 2003:124843.
- 28. Sweeney M, Hosseiny S, Hunter S, Klotz SD, Gennaro RN, White RS. Immunodetection and comparison of Melaleuca, bottlebrush, and Bahia pollens. Int Arch Allergy Immunol. 1994;105(3):289–96. pmid:7920031.
- 29. Behrendt H, Kasche A, Ebner von Eschenbach C, Risse U, Huss-Marp J, Ring J. Secretion of proinflammatory eicosanoid-like substances precedes allergen release from pollen grains in the initiation of allergic sensitization. Int Arch Allergy Immunol. 2001;124(1–3):121–5. doi: 53688. pmid:11306946.
- 30. Hirschwehr R, Heppner C, Spitzauer S, Sperr WR, Valent P, Berger U, et al. Identification of common allergenic structures in mugwort and ragweed pollen. J Allergy Clin Immunol. 1998;101(2 Pt 1):196–206. pmid:9500752.
- 31. He C, Murray F, Lyons T. Monoterpene and isoprene emissions from 15 Eucalyptus species in Australia. Atmos Environ. 2000;34(4):645–55.
- 32. Harrison D, Hunter MC, Lewis AC, Seakins PW, Nunes TV, Pio CA. Isoprene and monoterpene emission from the coniferous species Abies Borisii-regis—implications for regional air chemistry in Greece. Atmos Environ. 2001;35(27):4687–98.
- 33. Tollsten L, Mueller PM. Volatile organic compounds emitted from beech leaves. Phytochemistry. 1996;43(4):759–62.
- 34. Pio CA, Silva PA, Cerqueira MA, Nunes TV. Diurnal and seasonal emissions of volatile organic compounds from cork oak (Quercus suber) trees. Atmos Environ. 2005;39(10):1817–27.
- 35. Dormont L, Roques A, Malosse C. Cone and foliage volatiles emitted by Pinus cembra and some related conifer species. Phytochemistry. 1998;49(5):1269–77.
- 36. Blanch J-S, Llusia J, Niinemets U, Noe SM, Penuelas J. Instantaneous and historical temperature effects on α-pinene emissions in Pinus halepensis and Quercus ilex. J Environ Biol. 2011;32(Copyright (C) 2012 American Chemical Society (ACS). All Rights Reserved.):1–6. pmid:21888223
- 37. Wrigley J, Fagg M. Eucalypts a celebration. Sydney: Allen & Unwin; 2012.