122 Plant Protection Quarterly Vol.14(4) 1999 during winter, but only in the coldest dis- tricts. The stems of all species are erect, with thin greyish bark and prominent lenticels. The entire leaves occur in oppo- Review site pairs along the twigs, and isolated unpruned plants form dense rounded canopies. The small flowers occur in dense masses at the ends of terminal and lateral twigs, and consist of a short (0.5–1.0 mm) The Biology of Australian Weeds. 36. Ligustrum tubular four lobed calyx, a white to cream corolla tube of similar length ending in lucidum Aiton and Ligustrum sinense Lour. four spreading lobes 2–4 mm long, two small white (mauve to purple in L. sinense) J.T. SwarbrickA, S.M. TimminsB and K.M. BullenC spreading stamens, and a central ovary A Weed Science Consultancy, 15 Katoomba Crescent, Toowoomba, containing several loculi surmounted by a short columnar style. The flowers of Queensland 4350, Australia. B all species of privet have distinct and to Department of Conservation, PO Box 10-420, Wellington, New Zealand. many people unpleasant odours based on C University of Queensland, Gatton, Queensland 4343, Australia. trimethylamines (Mabberley 1997). The fruits are ovoid or spherical dark blue to black berries with a very thin skin, oily Name and related Australian south-eastern Australia and New Zealand fleshy pulp, woody or papery endocarp species and L. ovalifolium is at least occasionally and 1–3 seeds. The European privet (Ligustrum vulgare L.) naturalized in parts of New Zealand, nei- was apparently known as ligustrum about ther is a significant environmental weed. Key to Ligustrum species naturalized in 2000 years ago by Virgil (Gledhill 1985). Australia and New Zealand (after The name may have meant ‘binder’ with Description Goulding 1973) reference to some use of the rather flexible Ligustrum species are deciduous, semi- a Young shoots and branchlets without twigs, and was retained for the whole ge- deciduous or more commonly evergreen hairs; leaves glossy above, pointed; nus by Linnaeus. The English common shrubs and trees. Under warm moist con- flowers shortly stalked or stalkless; co- name privet appears to be obscurely de- ditions all species retain their foliage rolla lobes reflexed b rived from the 16th century English name throughout the year, although in Aus- Young shoots and branchlets hairy; primprimp for the same plant (Brown tralia both L. sinense and L. lucidum may leaves not glossy above, often bluntly 1993). Apart from this single European partially defoliate during dry weather. tipped; flowers stalked; corolla lobes species the genus ranges from India to In New Zealand they tend to defoliate spreading, not reflexed c north-eastern Australia, with most species in south-eastern and eastern Asia (Kiew 1978, Mabberley 1997). The endemic Aus- tralian species L. australianum F.Muell. is found in rainforest margins throughout sub-coastal central and north-eastern Queensland, whilst the south-eastern Asian L. glomeratum Blume occurs natu- rally in rainforests on Christmas Island (Green 1993). About 12 species of Ligustrum are in cultivation throughout the world as orna- mental or hedge plants, at least four of which have become naturalized in Aus- tralia and New Zealand. L. vulgare L. (Eu- ropean or common privet) occurs across much of Europe and northern Africa, from whence it has been introduced into Aus- tralasia. Both L. lucidum Aiton (broad- leaved or tree privet) and L. sinense Lour. (Chinese privet) were introduced indi- rectly from eastern Asia via Europe as ornamentals. The Japanese species Ligustrum ovalifolium Hassk. probably ar- rived in this region by the same route and for the same purpose. Ligustrum lucidum and L. sinense are widely naturalized throughout sub- coastal eastern Australia and the North Is- land of New Zealand, in both of which they are more or less serious environmen- tal weeds. Both species are sparsely natu- ralized in the South Island of New Zea- land. Although L. vulgare is naturalized in Figure 1. Ligustrum lucidum – broad-leaved privet. Plant Protection Quarterly Vol.14(4) 1999 123 Ligustrum sinense (Figure 2) (Dellow 1983, Hnatiuk 1990, Carr et al. is usually a multi-stemmed 1992, Queensland Herbarium 1997). It also shrub 2–6 m tall. Canopies are naturalized on Norfolk Island (Flora of rounded and are composed of Australia 1984). thin mid to dark green non- Ligustrum sinense (Figure 4) is also natu- shiny leaves which are 3–5 cm ralized in sub-coastal eastern Australia long, with rounded to bluntly from northern Queensland to Victoria pointed tips and tapering bases. (Hnatiuk 1990, Carr et al. 1992, Queens- The leaves are finely hairy land Herbarium 1997). It is a potentially when young and are similarly serious environmental weed on both Nor- coloured above and below. The folk and Lord Howe Islands (Flora of Aus- flowers occur in dense tapering tralia 1984). masses about 10 cm long at the The Australian distribution of both spe- ends of lateral and terminal cies probably reflects both their spread in twigs, and are followed by cultivation and their climatic tolerances. shortly stalked 5 mm long Both species are expected to spread fur- blackish-purple berries. ther in the future. Ligustrum vulgare is a single to multi-stemmed shrub 3–4 m New Zealand tall. Unpruned canopies are Ligustrum lucidum is widely naturalized in rounded and consist of thick the North Island of New Zealand, espe- shiny leaves 3–6 cm long. The cially in Northland, Auckland, Waikato, leaves have bluntly pointed tips the Bay of Plenty and Coromandel. It oc- and tapering bases and are curs in waste places, forest margins, open finely hairy when young. The forests and coastal cliffs, and can invade flowers occur in dense masses closed forests wherever gaps occur. Its 3–6 cm long at the ends of the height and longevity mean that once es- twigs, and are followed by tablished it can dominate the native veg- shortly stalked glossy blackish etation (Cameron 1990). 6–8 mm long berries. Ligustrum sinense is abundantly natu- Ligustrum ovalifolium is a ralized on roadsides and cliffs in and 3–5 m tall shrub. It resembles around towns in the North Island, particu- broad-leaved privet in having larly in the north and west, where it com- Figure 2. Ligustrum sinense – Chinese privet. hairless young twigs and petes with shrubby native plants. It also glossy leaves, but the leaves are occurs around Nelson, Christchurch and b Leaves 5–13 cm long; corolla tube only 3–6 cm long, the flower tube is twice Jackson in the South Island (Webb et al. shorter or as long as the lobes; anthers as long as the petals and the blackish ber- 1988, Williams and Karl 1996)). exserted from tube; fruits bluish-black, ries are stalkless. on short smooth stalks L. lucidum Habitat Leaves 3–6 cm long; corolla tube more Variation Climatic requirements than twice as long as lobes; anthers Variegated forms of L. lucidum and L. Both L. lucidum and L. sinense grow best barely exserted from tube; fruits black, sinense are common in cultivation. Vari- where there is moderate to high soil mois- stalkless L. ovalifolium egation ranges from white through yellow ture and air humidity throughout the c Lower leaf midribs finely hairy, mar- to pinkish and may be either marginal or year, e.g. in sheltered gullies on slopes and gins smooth to wavy; corolla tube blotchy across the leaves (Griffiths 1994). tablelands which receive higher than av- shorter than lobes; anthers mauve to Variegation is most prominent in well-lit erage local rainfall or runoff. In these situ- purple, exserted; fruits blackish-pur- leaves, tends to disappear in heavily ations they generally grow vigorously and ple, on hairy stalks L. sinense shaded parts of the canopy, and has not retain their leaves throughout the year, Lower leaf midribs not finely hairy, been observed in naturalized plants. whilst in more exposed areas with lower margins never wavy; corolla tube humidities their growth is usually more slightly longer than lobes; anthers History stunted and they tend to defoliate during white, barely exserted; fruits glossy Ligustrum lucidum was collected by Euro- dry seasons. black, on short slightly hairy stalks pean horticulturists in China in 1794 and L. vulgare L. sinense in 1852 (Chittenden 1951), both Substratum Ligustrum lucidum (Figure 1) is a small to probably from cultivated plants. They Ligustrum lucidum and L. sinense grow on medium sized tree, generally with a sin- were widely cultivated in Europe before a very wide variety of soils, ranging from gle trunk 3–15 m tall. Trees growing in the being brought to Australia and New Zea- sands through loams to nearly pure clays open have dense pyramidal crowns of land during the nineteenth century and (Westoby et al. 1983). L. lucidum grows rigid dark green shiny leaves 5–13 cm planted as hedges and shade trees. Nei- well in limestone karst around Molong in long with pointed tips and rounded bases, ther species appears to have become natu- central New South Wales. In the Sydney which are hairless when young and are ralized in New Zealand until the 1950s area L. lucidum prefers soils that are higher distinctly paler below than above; mature (Esler 1987). in clay content, whilst L. sinense prefers leaves snap when folded backward across more sandy soils, though both grow best the midrib. The flowers occur in dense ta- Distribution in areas of increased nutrition due to ur- pering masses up to 20 cm long at the tips Australia ban runoff, soil disturbance and removal of the twigs. The corolla tubes are much Ligustrum lucidum (Figure 3) is widely of the native vegetation (Adamson 1977, the same length as the petals, and are fol- naturalized in sub-coastal eastern Aus- Buchanan 1989). Neither species grows lowed by bluish-black 8 mm long berries tralia from northern Queensland to Victo- well on the poorer sandstone soils of ridge on short stalks. ria, especially around towns and cities tops around Sydney (Fox and Adamson 124 Plant Protection Quarterly Vol.14(4) 1999 smothering first the ground flora then shrubs and tree seedlings, until no natives are left except emergent eucalypts or podocarps. Ligustrum lucidum tends to form monospecific stands with plants of differ- ent sizes and ages (van Aalst 1992, Breaden 1996). The larger plants out- compete the smaller native plants and privet seedlings for light and possibly other resources. The plants are long-lived, and once such a monospecific stand has been formed it is capable of maintaining itself for very long periods. Ligustrum lucidum most closely fits into Garwood’s (1989) category of tropical for- est trees as a long-lived pioneer, since it has large numbers of efficiently dispersed seeds which can germinate in either sun or shade, grows well in gaps and, whilst it tends to dominate late secondary forest Figure 3. Distribution of Ligustrum Figure 4. Distribution of Ligustrum succession, may also be a component of lucidum in Australia and New sinense in Australia and New the mature forest canopy (van Aalst 1992). Zealand. The plant is not known to Zealand. The plant is not known to By contrast L. sinense with its reduced seed be naturalized in central or Western be naturalized in central or Western dormancy, lower stature and lower toler- Australia. Australia. ance of shade more closely fits Garwood’s definition of a short-lived tropical forest 1979). High soil phosphorus levels in- regrowth from basal and lateral buds after pioneer. crease growth (Fox and Adamson 1979). stem damage, widely spaced pairs of Soil disturbance is a major factor in the leaves that minimize self-shading and Growth and development initial establishment of both species, at bending forces on the slender stem, and a Physiology least around Sydney (Adamson and dense mass of shallow fibrous roots which Starch levels and regrowth potential of cut Buchanan 1974). It both removes competi- efficiently exploit soil nutrients, moisture stems of L. lucidum were measured at two tion from existing native plants and in- and oxygen (Fox and Adamson 1979). monthly intervals over a year by van Aalst creases the availability of soil nutrients Once L. lucidum plants reach the light (1992). Starch levels remained at about 1% and moisture. On volcanic soils around they branch repeatedly to form a rounded from March to August or September 1990, Toowoomba in southern Queensland, or oval canopy, with many twigs receiv- rose to 3-6% between November 1990 and however, L. lucidum readily invades ing full or partial sunlight. From this stage March 1991, and then fell again. Re- unburnt and generally undisturbed euca- onwards profuse flowering occurs at sprouting occurred only from cuttings lypt (Eucalyptus spp.) forest through dense shoot tips. In years of good soil moisture taken in November and January. The lantana (Lantana camara L.) understorey seed production may reach three or more higher levels of starch in the branches dur- (J.T. Swarbrick unpublished). million per large tree (Fox and Adamson ing summer resulted from increased pho- 1979, van Aalst 1992). tosynthesis at this time of year. Van Aalst Light regime Ligustrum lucidum stem cuttings grown also recorded rapid reshooting of cut Both Ligustrum species can germinate and under 0, 30, 50, 70 and 90% shade pro- stumps throughout the year, explaining grow under very low light intensities duced longer stems (due to greater inter- the discrepancy between reshooting abili- (Buchanan 1989, van Aalst 1992, Breaden node lengths) under lower light ties of separated shoots and cut stumps by 1996). Dense germination of both species intensities. These plants also produced the much greater starch reserves of the occurs at 1–5% full sunlight under sweet larger and thinner leaves with lower chlo- lower trunks and upper root systems. Her pittosporum (Pittosporum undulatum rophyll levels than the more brightly lit observations help to explain the season- Vent.) canopies in Sydney, with L. lucidum cuttings. Overall root, shoot and leaf dry ality of flowering and fruiting. showing 349 seedlings and L. sinense 624 matter levels did not, however, differ sig- Both privet species have a number of seedlings m-2 (Buchanan 1989). In New nificantly between light intensities (van stress tolerant features, including toler- Zealand L. lucidum germinates and grows Aalst 1992). ance to short term drought, temporary under the dense shade of the secondary waterlogging (Dasciano et al. 1994), heavy forest tree mahoe (Melicytus ramiflorus Plant associations shade, serious stem damage and dust and Forst. & Forst.f.). Although seedlings can Both privet species are typical rain forest other forms of aerial pollution. Further- persist under such dense shade, very few plants, with the ability to germinate and more both species take advantage of en- survive more than a few years unless the survive in dense shade, rapid sapling hanced nutrient levels in urban runoff canopy is broken, e.g. by clearing mature growth, tall slender trunks with thin bark, (Adamson and Buchanan 1974). privet in an attempt to stimulate the re- and bird dispersed seeds. This group of generation of the native vegetation (van characteristics allows them to invade, Phenology Aalst 1992). compete with and suppress many types of Seedlings of both species of privet show Ligustrum lucidum seedlings grow Australasian native vegetation. Both spe- epigeal germination (Buchanan 1981). faster than native plant seedlings under cies invade eucalypt forests and wood- Cotyledons of both species abscise early dense shade in Sydney bushland. They do lands in Australia and open coastal under heavily shaded conditions (Buch- so because of a number of effective com- podocarp and broad-leaved forests in anan 1989). Seedling growth is rapid un- petitive strategies. These include the pro- New Zealand. Once established they grow der good light conditions, with the well duction of slender single stems, rapid upward like a rising tide (Figure 5), spaced pairs of leaves increasing in size Plant Protection Quarterly Vol.14(4) 1999 125 upwards along the stem. The taproot of anther sacs split down their outer sides have been recorded feeding on them. both species may be either vertical (espe- and recurve to display the pollen. The pol- These include blackbirds (Turdus merula cially in better soils) or horizontal len grains adhere lightly to the inner sides L.), silvereyes or waxeyes (Zosterops latera- (Buchanan 1989), the latter occurring es- of the anther walls and to each other and lis Latham), tui (Prosthemadera novae- pecially in stony soil and allowing easier do not float away in the wind, although seelandia Gmelin), New Zealand native pi- uprooting of small seedlings. Dense some can be shaken from the anthers. Pol- geons (Hemiphaga novaeseelandia Gmelin), masses of fine fibrous roots which rapidly lination is presumably by medium sized starlings (Sturnus vulgaris L.), pied exploit the resources of the soil surface are insects such as flies and beetles, which are currawongs (Strepera graculina Shaw), produced by seedlings of both species attracted by the distinctive scent and the white-headed pigeons (Columba leucomela (Breaden 1996). moving masses of pale flowers contrasted Temminck) and king parrots (Alisterus Buchanan (1989) illustrates L. lucidum against the dark green foliage. Under suit- scapularis Lichtenstein) (Buchanan 1978, as suckering from the collar region, but ably warm, moist conditions pollination Loyn and French 1991, van Aalst 1992, unless damaged, plants normally grow in L. lucidum at least is highly effective and Timmins and Mackenzie 1995, Bass 1996, into single-stemmed trees. L. sinense fre- most flowers in an inflorescence may de- Williams and Karl 1996). Silvereyes both quently suckers from surface roots, but it velop into mature fruits. peck away the flesh to leave naked seeds is unclear whether this happens naturally on the parent tree and swallow and def- or only after root damage. J.T. Swarbrick Seed production and dispersal ecate whole seeds (van Aalst 1992), whilst (unpublished) has noted cut roots of L. A wide variation (5–51%, average 22%) pied currawongs normally regurgitate sinense about 1 cm thick suckering from occurred in the percentage of fruits con- them in pellets containing up to 50 viable the proximal ends to form new plants. taining more than one seed on four L. luci- seeds (Adamson 1977, Bass 1996), some- Whole L. lucidum trees were harvested dum trees in Auckland (van Aalst 1992). A times up to 1 km from the feeding site in Argentina by Dascanio et al. (1994) to significant number of fruits in both privet (Buchanan 1978). determine the biomass. Trunks and stems species also lacked effective embryos In Brisbane, T. Low (personal commu- represented 60.8%, branches 34.6%, twigs (9.2% for L. lucidum and 17.8% for L. nication) has noticed flocks of pied 2.0% and leaves 2.6% of the estimated 2337 sinense), giving an average of 110 and 85 currawongs feeding on L. lucidum in gar- t biomass ha-1. intact seeds per 100 fruits. Two seeds oc- dens that carry heavy loads of fruits in dry Basal shoot regrowth was studied in curred in 30% of L. lucidum fruits and 3% winters, then dispersing seeds back into mature L. lucidum privet trees by van Aalst of L. sinense fruits (Burrows and Kohen nearby bushland. (1992), who cut trees down 10–20 cm 1983). A low number (0.4%) of fruits in above the soil surface and left the cut each species have been shown to contain Physiology of seeds, germination and soil stumps untreated for one to two years. three fully formed seeds (van Aalst 1992). seed banks Most shoots arose from the upper part of Large well lit 35–40 year old L. lucidum Both species of privet have high initial the cut trunk, and stump shoots grew trees in Auckland may produce 1–3 mil- seed viability. Adamson (1977) found it to faster and to greater lengths than the lion seeds per tree per year, and larger be about 90% for both species in Sydney, shoots of either seedlings or rooted trees could probably produce even more whilst Burrows and Kohen (1983) noted cuttings. (van Aalst 1992). Well lit trees in Sydney about 90% for L. lucidum and 75% for L. Ligustrum lucidum forms flower buds in were shown to produce an average of 400 sinense. October–November and flowers between fruits m-2, with L. sinense producing 1300 The presence of the fleshy mesocarp November and March. The fruits are ripe (Westoby et al. 1983). Fruit numbers per was shown to prevent germination of by May, and have mostly been dispersed tree and per branch vary with stem diam- fresh fruits of both species by Burrows and by August (van Aalst 1992, J.T. Swarbrick eter and the degree of shading, with Kohen (1986), although this was demon- unpublished). L. sinense flowers and fruits smaller or more heavily shaded stems set- strated only for L. lucidum by van Aalst over the same periods, but with a low ting less fruit than larger or better lit stems (1992). Burrows and Kohen ascribed the level of flowering and fruiting throughout and branches (Westoby et al. 1983, van lack of germination in fresh fruits to ex- the year (Webb et al. 1988, van Aalst 1992). Aalst 1992). ceedingly low water potentials (below The trunks of L. lucidum thicken mainly Flowering is limited on L. lucidum trees -3,000 j kg-1 or -30 bars) in the presence of by an annual flush in early summer, al- less than 7 cm thick at the base, but once both fresh and rehydrated flesh, and lowing age determination from cross- flowering has commenced it occurs on showed that the woody endocarp of L. sections of the stem (van Aalst 1992). most well lit branches over 2 cm thick. The lucidum may delay germination for up to Timmins and Mackenzie (1995) note that number of flowers produced per tree in- 15 days. broad-leaved privet trees may live for a creases steadily at least until 30 cm diam- The germination of both species of century and, since suckers tend to develop eter, at which diameter their maximum privet at different temperatures was stud- around their bases as the trunks senesce, potential fruit production had probably ied in Sydney by Burrows and Kohen individuals are potentially immortal. not yet been achieved (van Aalst 1992). (1983). Freshly cleaned seeds of both spe- Fruit production in L. lucidum may vary cies were incubated in the dark at constant Reproduction significantly between years. Westoby et al. temperatures of 5, 10, 15, 20, 25 and 30°C Both species of privet reproduce primarily (1983) found that trees that produced and germination recorded over 33 days, by seeds, which are distributed mainly by heavy crops of fruits one year often had a after which ungerminated seeds were birds. L. sinense may also reproduce lighter crop in the subsequent year. Casual transferred to the optimum germination vegetatively by root suckers. Both species observations have shown that very heavy temperature for that species for a further can be vegetatively propagated by stem fruit production may occur on well lit 14 days. The optimum germination tem- cuttings. trees in Toowoomba, Queensland in rela- perature for L. lucidum was 15°C and for L. tively moist summers, while little or no sinense 20–25°C. Seeds held at 30°C for 33 Floral biology and pollination fruit production normally occurs during days generally failed to germinate even Both species of privet flower only on well- very dry summers. when transferred to the optimum lower lit crowns and branches of established Seed dispersal in both species of privet temperature for that species. Low soil plants (Adamson 1977). The flowers in is principally by birds. Ripe fruits are pre- temperatures occur in Sydney during seed each panicle open over about a week, and sented in dense masses on the outside of fall (June to August) and the authors sug- release a distinctive heavy scent. The two the canopy, and many species of birds gest that this may stratify the seeds, 126 Plant Protection Quarterly Vol.14(4) 1999 promoting better germination when soil temperatures increase in the spring. They cite Rudolf (1974) who showed that strati- fication assists germination in L. vulgare. Rapid synchronous germination of freshly cleaned fruits has occurred in the laboratory in both species of privet under good conditions, with final germination of 88.5% for L. lucidum and 44.5% for L. sinense (Burrows and Kohen 1986). Van Aalst (1992) found that about 50% of fresh cleaned seeds of L. lucidum and fresh cleaned and uncleaned L. sinense germi- nated within 70 days under optimum con- ditions. Panetta and Robazza (1997) re- ported that under good conditions there is a flush of germination in both species dur- ing the winter and spring immediately af- ter seed-fall, with no further emergence after five months for L. sinense and seven months for L. lucidum. When the experi- ment was repeated in 1997, however, seedling emergence was neither as fast nor as synchronous as in 1996 (F.D. Figure 5. Ligustrum lucidum overwhelming successive layers of eucalypt Panetta personal communication). Panetta forest in Toowoomba, Queensland. and Robazza (1997) reaffirmed Burrows and Kohen’s 1986 observation that intact Hybrids 10–15 m tall, below which many L. lucidum fruits eventually germinate in moist soil. Hybrids between the various species of seedlings but successively fewer saplings Privet seed banks of 292–557 seeds m-2 privet have not been reported in the avail- struggle to survive (J.T. Swarbrick unpub- have been shown to occur in Sydney soils, able literature. lished). This dominance may be almost although less than 8% of these seeds were complete in fire-sheltered gullies around germinable (Burrows and Kohen 1986). Population dynamics Sydney, although both populations and Whilst anecdotal observation suggests Both species of privet are widely culti- plants are smaller on slopes and especially that the seed bank of L. lucidum in moist vated in both Australia and New Zealand, on exposed ridges subject to desiccation soil decreases rapidly after one year, good and both remain commercially available. and fire (Burrows and Kohen 1983). germination of both species was observed All naturalized populations of both spe- by Burrows and Kohen (1986) after dry cies are thought to have originated from Importance storing seed in the laboratory for two mature flowering plants in urban and sub- Detrimental years. Refrigerated L. lucidum seeds may urban gardens and parks (Adamson Both privet species are competitors in remain viable for up to two years, al- 1977). terms of Grime’s (1979) C-S-R model, in though similarly stored L. sinense seeds Birds disperse the seeds of both species that they show vigorous above and below lost viability after the first year (van Aalst of privet into a wide variety of situations ground lateral expansion, annual flower- 1992). Her studies of the soil seed bank of and vegetational complexes. Their good ing, high seed production, effective seed L. lucidum also indicated that most seeds dispersal (especially to under perching dispersal and seed dormancy (van Aalst germinate within a year of deposition, al- sites), high initial viability and tolerance 1992). though some may germinate after two. of shade, soil types, temperatures and Ligustrum lucidum has been described moisture levels means that privet seed- as a ‘problem weed’ in the northern half of Vegetative reproduction lings occur in many situations, sometimes the North Island of New Zealand. It read- Ligustrum lucidum reproduces naturally at very high densities. ily invades short forest, scrub, and forest only by seed, but may also be reproduced Seedling mortality is very high in most margins, substantially changes the native by stem cuttings. Twenty cm long basal situations, with most dying either from vegetation, and once established is diffi- stem cuttings from young saplings 3–5 cm drought in more exposed places or from cult to eliminate (Timmins and Williams thick produced shoots within two weeks lack of sufficient light under dense cano- 1987). Although exotic trees over 10 m tall and roots from three weeks when placed pies. Allelopathic inhibition does not seem are uncommon as weeds in New Zealand, in moist river sand without additional to operate in either species of privet, as the this species is an exception (Williams hormones (van Aalst 1992). Cuttings pro- seeds of both species germinate abun- 1997). In the Northland on the North Is- duced shoots and roots under all levels of dantly under parent trees. Heavy selection land it replaces mid-canopy forest trees shade. The variegated horticultural geno- for seedling survivorship and vigour such as taraire (Bielschmiedia taraire types are normally propagated by stem means that only the best adapted and most (A.Cunn.) Benth. & Hook. ex Kirk), towai cuttings. fortunately placed seedlings are likely to (Weinmannia silvicola Sol. ex A.Cunn.) and Ligustrum sinense reproduces by shoots survive beyond the seedling stage. pohutukawa (Meterosideros excelsa Sol. ex arising from surface roots as well as by Under suitable conditions both species Gaertn.) (Fromont and King 1992). It has seed, although the degree to which this of privet dominate the ground layer and also replaced much of the native canopy happens without prior wounding of the then successively taller strata of vegeta- of forests along the Coromandel Peninsula surface roots is unclear. Such wounding tion. Eventually L. lucidum at least (Figure (S.M. Timmins unpublished) and is invad- could occur naturally to some extent, and 5) may form a monospecific understorey ing forest gaps around Auckland (Cam- subsequent rotting of the interconnecting under established eucalypts. Mature L. lu- eron 1990). surface roots would result in separated cidum on moist basaltic soils in south-east Ligustrum sinense displaces the native plants. Queensland maintains a closed canopy shrub layer in both forests and Plant Protection Quarterly Vol.14(4) 1999 127 regenerating native vegetation in the defends L. vulgare against the charge of In south-eastern coastal Queensland L. North Island of New Zealand (S.M. causing hay fever and asthma on these lucidum invades various orchard crops, Timmins unpublished). This is of particu- grounds, pointing out that the Oleaceae usually starting as seedlings under the lar concern in alluvial forest remnants, are pollinated by small insects and none canopies of trees where birds have been some of which contain uncommon and have wind-borne pollen. He ascribes the feeding. Privet invades kiwi fruit orchards rare shrubs such as ribbonwood association of privet with hay fever and in the Bay of Plenty in New Zealand (Lit- (Plagianthus regius (Poit.) Hochr.) and asthma to mistaken identity brought on by tle 1982), whilst L. lucidum is a common in- heartleaf kohukohu (Pittosporum obcord- the showy and noticeably smelly flowers vader of gardens in Toowoomba, Queens- atum Raoul). Although individual plants occurring at the same time as less con- land and elsewhere along the eastern Aus- of L. sinense are short-lived in these situa- spicuous but more highly allergenic tralian seaboard. tions it may persist as a community, and grasses. Honey derived from privet flowers has has been observed persisting under the Following extensive clinical study and been reported to have a fishy smell due to light to moderate shade of New Zealand testing in Sydney, Bass (1989 and personal the triethylamine in the nectar (Mabberley native forest (Fromont and King 1992). communication) believes that the pollen of 1997), although no such annotation ap- Invasion by both privet species has ad- neither species is strongly allergenic, but pears in Clemson (1985). verse impacts on native vegetation in Aus- that cross-reactivity occurs and grass pol- tralia, especially in and around eastern len sensitive people may also become sen- Beneficial coastal and sub-coastal urban centres sitive to the pollen of L. lucidum. Since this All species of privets were introduced into (Buchanan 1989, Breaden 1996). Both spe- pollen is seldom air-borne however, it is Australia and New Zealand as ornamen- cies thrive in fire-free areas where there is suggested that significant privet pollen al- tals and as shade, shelter and hedge sufficient rainfall or urban stormwater lergy is unlikely. Bass (1989 and personal plants, and have ably fulfilled these roles and other runoff, as well as on naturally communication) suggests it may be the for over a century. Variegated forms of the fertile, anthropogenically enriched, and perfume of privet flowers which already main species are commercially available disturbed soils. Many of these areas were sensitized people find irritating rather and widely planted. The privets are also occupied in pre-European times by euca- than the pollen. To cover all eventualities very tolerant of dry, wet, shady, and dusty lypt forests, which replaced the naturally she suggests that all species of privets be conditions (Goulding 1973), and thrive climactic rain forests and scrubs as a result excluded from low allergen gardens (D.J. even under adverse industrial and urban of long periods of Aboriginal fire manage- Bass n.d.). conditions. L. lucidum was cultivated as ment. Cessation of burning has allowed Both species of privet have been associ- host for the economically very important reinvasion of these areas by rainforest spe- ated with the poisoning of stock (mainly white wax scale (Coccus pela (L.)) industry cies, including the two species of privets. horses) and people (especially children). in China in the 1870s and 1880s, as graphi- Dense L. lucidum infestation in Francis and Southcott (1967) considered cally described by Goulding (1973). Toowoomba, Queensland reduce photo- that both the leaves and fruits of all spe- synthetically active radiation levels to cies of privet were poisonous and noted, Legislation and community action 1.3% of full sunlight at mid-day (from 510 but did not support the contention, that None of the privets are Declared Plants in to 32 µE m-2 sec-1) (Breaden 1996), sup- children had died from eating the fruits. Australia. L. lucidum is a major environ- pressing grasses and other native herbs, Everist (1981) failed to find any clear Aus- mental weed in parts of sub-coastal south- shrubs and tree seedlings. Deeper leaf lit- tralian examples of stock or human poi- eastern Queensland, but applications for ter accumulation (3.8 cm compared to 1.9 soning by either the leaves or fruits of any its declaration have been unsuccessful cm under open eucalypt forest with species of privet, although he cited several since it is considered to be too widespread grasses and herbs) and increased leaf lit- overseas sources from 1939 to 1968 for and too firmly naturalized to eradicate. ter biomass (873 g compared to 318 g m-2) such toxicity. Connor (1977) considered Both species are, however, included in the also presumably help to inhibit germina- that even though both human and stock list of restricted plants in proposed tion of small-seeded native plants. Thirty toxicity was rare in New Zealand and else- Queensland legislation, making their per cent of the standing leaf litter crop un- where, it had occurred and privets should propagation or sale illegal within the der mixed L. lucidum and Argentinian na- be considered as dangerous plants around state. tive trees was due to the privet, rising to the farm. Neither McBarron (1983) nor A privet control strategy has recently 79% under more intensively invaded for- Dowling and McKenzie (1993) included been developed by Toowoomba City est (Dascanio et al. 1994). privets among the poisonous plants of Council (1997). The goals are to eradicate The short term control of privets and New South Wales or of Queensland. The both L. lucidum and L. sinense from urban other exotic rainforest trees in areas of eu- most detailed discussion of the apparently areas and to control them in peripheral calypt forest appears to be insupportable toxic principles of privet foliage and fruits bushland within five years. The strategy in the long term without a return to the is given by Watt and Breyer-Brandwijk depends on information gathering, pub- burning regimes which were responsible (1962), who suggested that the available licity, strategic planning, annual privet for rainforest suppression. Unfortunately, evidence points towards an irritant princi- control programs within the urban area in suburban eastern Australia the ground ple rather than a toxin. The bark contains (70 t removed in 1994, 400 t in 1975, 1000 t layer of perennial tufted grasses which about 7% tannin, and the fruit also con- in 1996, 2130 t in 1997, over 3000 t in 1998 carried the fires has often been lost, pre- tains a number of possibly toxic chemicals, and over 8000 t in 1999), community in- venting the frequent relatively cool fires including ligustron and the glucoside volvement, and routine privet control on that maintained the eucalypt forests be- syringopicrin. Neither Watt and Breyer- council property and in public bushland fore European management. Brandwijk (1962) nor any of the other au- throughout the year. Collected privet is Both species of privet have been impli- thors mentioned above cited any clear chipped and used as mulch. An annual cated with allergic reactions leading to cases of privet poisoning of either humans budget has been allocated for this work, hay fever and asthma, but this is difficult or stock. Fruits are avidly eaten by birds and privet may become a Declared Plant to prove. The pollen of both species is ad- without apparent ill effect (Watt and within the city when most has been re- hesive and mainly transported by insects, Breyer-Brandwijk 1962), whilst feeding moved from public and private land. and is unlikely to be present as free-float- fruit and foliage to pigs and fruit to sheep In New Zealand both species of privet ing grains in the air in sufficient quantities has resulted in no apparent ill effects currently have national surveillance sta- to cause allergic reactions. Little (1981) (Everist 1981). tus, which prohibits their propagation, 128 Plant Protection Quarterly Vol.14(4) 1999 sale, distribution or commercial display. respectively). A subsequent trial using the trunks, or applied to freshly cut stumps at They are also subject to Regional Pest same herbicides and treatment methods 1 g L-1 water or to foliage at 10 g 100 L-1 Management Strategies in some areas, es- (Dellow et al. 1985) gave similar results. water. Very small plants have been killed pecially in parts of the North Island. The A range of herbicides was tested by overall spraying with glyphosate. Northland Regional Pest Management against L. lucidum and L. sinense at Strategy already includes minor provision Whakatane in the North Island of New Physical control and fire for privet control (Northland Regional Zealand, using cut stump, soil injection Seedlings and young plants of both spe- Council 1995). and foliar application (James and cies of privet can often be hand pulled, Mortimer 1984, T.K. James personal com- and where possible this is an effective Weed management munication). Effective control (90% or method of control. Once the stems are Individual plants of both species of privet more plants dead after three months) was more than about 1 cm thick, however, the are readily controlled, especially by the achieved only with cut stump applications tap roots are too strongly established to use of appropriate herbicides. Large areas of 20% 2,4-D plus 5% picloram or 1% permit hand pulling. Larger plants may be or numbers of privets are, however, often triclopyr plus 0.25% picloram (both in dug from the ground (Whittet 1968 recom- difficult to control, since they generally spring and autumn), foliar application of mends using a mattock), but the process is infest extensive areas of degraded native 0.15% DPX T6376 in spring and in au- laborious and disruptive to the soil sur- vegetation which may be difficult to ac- tumn, and foliar application of 0.72% face and surrounding plants. Well estab- cess, have already lost the capacity to re- glyphosate plus 0.5% X-45 wetter, but lished L. lucidum can only be removed by generate the ground flora from seed banks only in spring. large mechanical equipment. L. sinense of native grasses and other species, and Madden and Swarbrick (1990) tested tends to sucker from damaged or cut be liable to rapid re-invasion by bird- five herbicides by stem injection, basal roots. dispersed seeds from outside the area. bark and cut stump applications against Although philosophically attractive for Control of nearby seed-bearing plants is L. lucidum in south-eastern Queensland weed control in native vegetation, the an essential pre-requisite to long term in winter. Good control (90% or more Bradley method by which weeds are re- management of both species of privet death) was achieved 14 months later only moved by hand without herbicides at a (Adamson 1977). The re-establishment of with 11.2% metsulfuron methyl in water rate commensurate with natural re- a regular burning program is probably es- by stem injection (96% control), 0.06% vegetation (Bradley 1971, 1988) is inap- sential to control seedling re-establish- metsulfuron-methyl in water by cut propriate to the control of established ment in infested eucalypt woodlands and stump application (96%), 4.8% triclopyr trees and shrubs. In an attempt to control forests in Australia, but would be inappro- ester in diesel oil by cut stump application established privets without using herbi- priate in New Zealand and in Australian (96%), 3% 2,4-D acid in water also by cut cides, Mowatt (1981) tried cutting back rainforests. stump application (95%), while 22.4% stumps and covering them with black glyphosate in water (also by cut stump ap- plastic, and removing the main stems with Herbicides plication) gave 86% control after 14 or without propping the proximal ends of The earliest recommendation for chemical months. The relative volumes of herbi- the lateral roots in the air. The differential control for L. lucidum in Australia was by cides used were 1.0 for stem injection, effects of these treatments on the two spe- Whittet (1968), who suggested cut stump 7.6 for basal bark and 8.7 for cut stump cies of privet are unclear from her results, treatment with 2% 2,4,5-T ester in oil. treatment. The relative labour costs were although she does comment that L. luci- Mowatt (1981) followed by testing stem 1.7, 1.0 and 8.0 respectively, and the dum failed to regenerate from cut roots injections of 25% hexazinone, 3.6 and 7.2% metsulfuron-methyl treatments were the once the main stem had been removed. glyphosate, 4.8 and 9.6% triclopyr and cheapest in terms of herbicide cost. Her other non-chemical treatments were 6.6% dicamba against both species of Current registrations for the chemical of little value apart from totally covering privet near Sydney, and found stem injec- control of both species of privet in Aus- the cut stumps with two layers of black tion with herbicides to be faster, cheaper, tralia include cut stump treatment with ei- plastic, which controlled regrowth for at easier and less disturbing to the native ther 5% triclopyr in diesel oil or 18% least nine months. None of these methods vegetation than uprooting. The two con- glyphosate in water, stem injection with have been taken up or are now in use. centrations of triclopyr were the most ef- 36% glyphosate, and foliar spraying with Ligustrum lucidum in particular is a fective of the six treatments tested. 0.00006% metsulfuron-methyl in water typical rainforest tree, with an erect slen- At Kerikeri in the North Island of New (PESKEM 1998). der trunk covered with thin bark and lack- Zealand, Little (1982) found cut stump ap- Toowoomba City Council has killed ing any fire-protective features. This plication of 5% picloram plus 20% 2,4-D very large numbers of L. lucidum trees and makes the cambium susceptible to hot fire amine (as Tordon 50-D) to be highly effec- seedlings along roadsides and in eucalypt around the base of the trunk. The main tive on unidentified privet over 2 cm thick. woodland over recent years with a difficulty in using fire to control L. lucidum The herbicide 2,4,5-T was found to be number of herbicides. These have in- is that by the time this species becomes both ineffective against L. sinense in that it cluded cut stump application of 18% dominant, the tufted perennial grasses only killed 70–75% of the plants when ap- glyphosate, which controls stumps up to such as kangaroo grass (Themeda triandra plied as a cut stump treatment and only 1.5 m tall growing through wire fences but Forssk.) which normally carry fire caused 30–40% defoliation when applied must be applied within 10 seconds of cut- through eucalypt woodlands and forests as basal bark (a result echoed by James ting the stem or trunk. Basal bark spray- have been shaded out. Experience in and Mortimer in 1984), and unsuitable for ing with either 0.5% Garlon (triclopyr) or Toowoomba has shown, however, that use in urban bushland (Dellow et al. 1983). 1.6% Access (triclopyr plus picloram) in during dry winters (June to August) Good results were obtained with cut diesel have also been very effective, as has broad-leaved privet largely defoliates on stump application of 18% glyphosate (96% overall spraying with 100 g Brush-Off exposed slopes and ridges (though less so control), 4% triclopyr in diesel (99%), 4% (metsulfuron-methyl) in 100 L water. in sheltered gullies) and the dry leaves triclopyr in diesel plus D-C-Tron sur- Escort (metsulfuron-methyl) has been together with other dry material on and factant (100%) and an unspecified rate of found to be the most effective herbicide for close to the ground are often sufficient to hexazinone (98%), with some success privet control in the Northland of New carry a hot fire. Extensive areas of sloping from all herbicides as a basal bark treat- Zealand. This may either be diluted with eucalypt forest along Toowoomba’s east- ment (95%, 67%, 55% and 57% defoliation water and injected into holes bored in the ern escarpment parks which have been Plant Protection Quarterly Vol.14(4) 1999 129 heavily invaded firstly by lantana and herbaria (AK, WAIK, MPN, WELT, CHR and L. sinense Lour. at different tem- then by L. lucidum have been successfully and WELTU), the Department of Conser- peratures. Australian Weeds 2, 130-2. burnt during winter in El Niño periods. vation weed inventory, Atkinson (1997), Burrows, F.J. and Kohen, J. (1986). Inhibi- Cutting the trunks a few weeks after such Colin Ogle, Paul Cashmore, John Sawyer, tion of germination in privet. Plant fires showed that many exhibited visible Shannel Courtney, Tom Belton, Nick Sing- Protection Quarterly 1, 107-8. browning of the phloem and sapwood. ers, Dick Veitch, Joanne O’Reilly and Pe- Cameron, E.K. (1990). The next generation Most saplings and small trees (under 5 cm ter Williams. of problem plants to the native forest diameter) were killed by the fires, as well remnants of northern New Zealand. as many much thicker trees, although in References Conference report. New Zealand Insti- some cases the interior damage extended Adamson, D. (1977). Privet and the prob- tute of Noxious Plants Officers Inc., only part-way round the trunk and the lem of control of exotic plants in urban 1990, pp. 41-7. plants regenerated from the unburnt por- bushland. Unpublished submission. Carr, G.W., Yugovic, J.V. and Robinson, tion. Following such fires it has been nec- Noxious Plants Advisory Commission K.E. (1992). ‘Environmental weed inva- essary to work through the forests in the of New South Wales, Sydney. sions in Victoria: conservation and succeeding summer (when any regrowth Adamson, D. and Buchanan, R.A. (1974). management implications’, p. 50. (De- becomes obvious) and to kill surviving Exotic plants in urban bushland in the partment of Conservation and Environ- trees by cut stump treatment. Such hot Sydney region. Proceedings of the Weed ment and Ecological Horticulture Pty. fires presumably also kill much of the Science Society of New South Wales 6, Ltd., Melbourne). privet seed bank at and near the surface of 24-7. Chittenden, F.J. (ed.) (1951). ‘Dictionary the soil, but fail to kill lantana which re- Anderson, D.J. (1996). The effect of fire on of gardening, Volume 3’, p. 1165-6. generates rapidly after fire from protected woody weeds and native plants in de- (Clarendon Press, Oxford). buds at the soil surface. graded urban eucalypt woodlands. Un- Clemson, A. (1985). ‘Honey and pollen published report. The University of flora’, pp. 183-4. (Inkata Press, Mel- Natural enemies Queensland Gatton College, Gatton. bourne). Little is known about the natural enemies Atkinson, I.A.E. (1997). Problem Weeds of Connor, H.E. (1977). ‘The poisonous of privets in south-eastern and eastern New Zealand Islands. Science for Con- plants of New Zealand’, pp. 133, 247. Asia, although it is to be expected that servation 45. New Zealand Department (Government Printer, Wellington). they are subject to a wide range of attack- of Conservation, Wellington. Dascanio, L.M., Barrera, M.D. and Frangi, ing arthropods (including Coccus pela) and Bass, D.A. (1996). Pied currawongs and J.L. (1994). Biomass structure and dry pathogens. The Australian and New Zea- invading ornamentals: what’s happen- matter dynamics of subtropical alluvial land material has mostly been imported ing in northern New South Wales. Pro- and exotic Ligustrum forests at the Rio either as seed or as healthy nursery stock ceedings of the 11th Australian Weeds de la Plata, Argentina. Vegetatio 115, via Europe, without significant arthro- Conference, Melbourne, pp. 362-5. 61-76. pods or pathogens. Bass, D.J. (1989). Weeds and human Dellow, J.J. (1983). Privet in New South In Australia the larvae of the hawk health. National Trust Bush Manage- Wales. Proceedings of a Workshop on moth Psilogramma menephron menephron ment News Sheet, September 1989. Na- the Management of Weeds in Recrea- (Cramer) eat an insignificant amount of tional Trust (NSW), Sydney tional Areas, eds G.M. and P.M. Kloot. foliage of both species (Anderson 1996). Bass, D.J. (No date). ‘The low allergen gar- SA Departments of Agriculture and of Buchanan (1989) mentions an un-named den’. (The Asthma Foundation of NSW, National Parks and Wildlife, Adelaide. and insignificant stem borer. St. Leonards). Dellow, J.J., Smith, L.W. and Milne, B.R. In New Zealand passion-vine hopper Bradley, J. (1971). ‘Bush regeneration‘. (1983). Herbicide control of privet. Pro- (Scolypopa australis (Walk.)) is a common (Mosman Parklands and Ashton Park ceedings of a Workshop on the Man- leaf-sucking insect on both species of priv- Association, Sydney). agement of Weeds in Recreational Ar- ets during summer (Goulding 1973, Bradley, J. (1988). ‘Bringing back the bush: eas, eds G.M. and P.M. Kloot. SA De- Timmins and Mackenzie 1995), but has the Bradley method of bush regenera- partments of Agriculture and of Na- negligible adverse effect on the plants. tion’. (Lansdowne Press, Sydney). tional Parks and Wildlife, Adelaide. Pirone (1978) lists a number of fungi, Breaden, R. (1996). The impact of broad- Dellow, J.J., Milne, B.R. and Smith, L.W. viruses, insects, mites and nematodes leaf privet (Ligustrum lucidum W.T.Ait.) (1985). Herbicide control of privet which attack ornamental Ligustrum spe- in an area of environmental value. Un- (Ligustrum lucidum). Australian Weeds cies in North America. published report. The University of Research Newsletter 33, 18-19. Although reportedly toxic in large Queensland Gatton College, Gatton, Dowling, R.M. and McKenzie, R.A. (1993). amounts, both sheep and cattle graze Queensland. ‘Poisonous plants – a field guide’. (De- privet foliage without ill effect if provided Brown, L. (ed.) (1993). ‘The new shorter partment of Primary Industries, Bris- with adequate alternate fodder (T.K. Oxford English dictionary on historical bane). James personal communication), and low principles, Volume 2’, p. 2360. (Claren- Esler, A.E. (1987). The naturalisation of infestations of privet seedlings may be don Press, Oxford). plants in urban Auckland, New Zea- prevented from becoming more serious by Buchanan, R.A. (1978). Pied currawongs land 3. Catalogue of naturalised plants. judicious grazing. (Strepera graculina Shaw): their diet and New Zealand Journal of Botany 25, weed dispersal in some urban bush- 539-58. Acknowledgments land, Sydney, NSW. Bulletin of the Eco- Everist, S.L. (1981). ‘Poisonous plants of The authors are grateful to Dr. Gillian logical Society of Australia 8, 6. Australia’, pp. 555-7. (Angus and Scott, botanical artist, Toowoomba for the Buchanan, R.A. (1981). ‘Common weeds Roberston, Sydney). drawings, Piet Filet of Toowoomba City of Sydney bushland’. (Inkata Press, Flora of Australia. (1984). ‘Volume 49: Council for obtaining a number of the Melbourne). Oceanic Islands 1’, p. 332. (Australian references, and Chris Edkins of the De- Buchanan, R.A. (1989). ‘Bush regenera- Government Publishing Service, Can- partment of Conservation for preparing tion: recovering Australian land- berra). the New Zealand distribution maps. The scapes’, p. 67. (TAFE, Sydney). Fox, M.D. and Adamson, D. (1979). The New Zealand distribution maps are based Burrows, F.J. and Kohen, J. (1983). Germi- ecology of invasions. In ‘A Natural on distribution records from New Zealand nation of Ligustrum lucidum W.T.Ait. Legacy’ eds H.F. Becher, D. Lunney 130 Plant Protection Quarterly Vol.14(4) 1999 and I. Dunn, p. 146-51. (Pergamon Northland Regional Council. (1995). remnants, Nelson, New Zealand. New Press, Sydney). Northland Regional Pest Management Zealand Journal of Ecology 20, 127-45. Francis, D.F. and Southcott, R.V. (1967). Strategy. Northland Regional Council, Williams, P.A. (1997). Ecology and man- Plants harmful to man in Australia. Whangarei. agement of invasive weeds. Conserva- Miscellaneous Publication No 1, Bo- Panetta, D. and Robazza, J. (1997). Ecology tion Sciences Publication 7. New Zea- tanic Garden, Adelaide, p. 29. and management of priority woody land Department of Conservation, Wel- Fromont, M. and King, S. (1992). Charac- weeds in southeast Queensland. Tech- lington. teristics and control methods of 30 nical Highlights 1996/97. Department weed species affecting conservation of Natural Resources, Brisbane. land in Northland. Unpublished re- PESKEM (1998). A custom search for port. New Zealand Department of Con- Ligustrum spp. Centre for Pesticide Ap- servation, Whangarei. plication and Safety, University of Garwood, N.C. (1989). Tropical soil seed Queensland Gatton College, Gatton. banks: a review. In ‘Ecology of soil seed Pirone, D.J. (1960). ‘Diseases and pests of banks’, eds M.A. Leck, V.T. Parker and ornamental plants, 4th edition’. (The R.L. Simpson. (Academic Press Inc., Ronald Press Co., New York). San Diego). Rudolf, P.O. (1974). Ligustrum L. In ‘Agri- Gledhill, D. (1985). ‘The names of plants’, culture handbook’, pp. 500-2. (United p. 129. (Cambridge University Press, States Department of Agriculture, Cambridge). Washington DC, USA). Goulding, J.H. (1973). Privets. Annual Queensland Herbarium (1997). Queens- Journal of the Royal New Zealand Institute land Herbarium HERBRECS Report of Horticulture 1, 45-8. 27/10/1997. Queensland Herbarium, Green, P.S. (1993). ‘Oleaceae. Flora of Aus- Brisbane. tralia, Volume 50: Oceanic Islands 2’, Timmins, S.M. and Mackenzie, I.W. pp. 375-6. (Australian Government (1995). Weeds in New Zealand Pro- Publishing Service, Canberra). tected Natural Areas Database. Techni- Griffiths, M. (1994). ‘Index of garden cal Series Number 8. Department of plants’, pp. 668-9. (Macmillan Press, Conservation, Wellington, pp. 163-70. London). Timmins, S.M. and Williams, P.A. (1987). Grime, J.P. (1979). ‘Plant strategies and Characteristics of problem weeds in vegetation processes’. (John Wiley and New Zealand protected areas. In ‘Na- Sons, Chichester). ture conservation: the role of remnants Hnatiuk, R.J. (1990). ‘Census of Australian of native vegetation’, eds D.A. Sanders, vascular plants’. Australian Flora and G.W. Arnold, A.A. Burbidge and A.J.M. Fauna Series, Number 11, p. 407. (Aus- Hopkins, pp. 241-7. (Surrey Beatty & tralian Government Publishing Service, Sons, Sydney). Canberra). Toowoomba City Council. (1997). Privet James, T.K. and Mortimer, J. (1984). Con- eradication strategy: let’s make Too- trol of privet. Proceedings of the 37th woomba a privet free zone. Too- New Zealand Weed and Pest Control woomba City Council, Toowoomba. Conference, pp. 206-9. van Aalst, M.M. (1992). Seed ecology and Kiew, R. (1978). The Oleaceae of Malesia: vegetative regrowth of tree privet The genus Ligustrum. Blumea 24, 143-9. (Ligustrum lucidum). Unpublished Little, E.C.S. (1981). A much maligned M.Sc. Thesis, University of Auckland, plant. New Zealand Farmer 25/6/1981, 135 pp. 104, 107. Watt, J.M. and Breyer-Brandwijk, M.G. Little, E.C.S. (1982). How to control privet. (1962). ‘The medicinal and poisonous New Zealand Journal of Agriculture 145, 15. plants of southern and eastern Africa’, Loyn, R.H. and French, K. (1991). Birds p. 807. (E & S Livingstone Ltd., Edin- and environmental weeds in south- burgh). eastern Australia. Plant Protection Quar- Webb, C.J., Sykes, R.W. and Garnock- terly 6, 137-49. Jones, P.W. (1988). ‘Flora of New Mabberley, D.J. (1997). ‘The plant-book. A Zealand. Volume 4. Naturalised portable dictionary of the higher pteridophytes, gymnosperms and di- plants, 2nd edition’, p. 409. (Cambridge cotyledons’, pp. 878-81. (Botany Divi- University Press, Cambridge). sion, DSIR, Christchurch, New Zea- Madden, L.E. and Swarbrick, J.T. (1990). land). Chemical control of Ligustrum lucidum. Westoby, M., Dalby, J. and Adams-Acton, Plant Protection Quarterly 5, 145-7. L. (1983). Fruit production by two spe- McBarron, E.J. (1983). ‘Poisonous plants. cies of privet, Ligustrum sinense Lour. Handbook for farmers and graziers’. and L. lucidum W.T.Ait. Australian (Inkata Press, Melbourne). Weeds 2, 127-9. Mowatt, J. (1981). Control of large-leaved Whittet, J.N. (1968). Weeds. The farmers’ privet (Ligustrum lucidum) and small- handbook series. Department of Agri- leaved privet (L. sinense) in urban culture, Sydney. bushland. Proceedings of the 6th Aus- Williams, P.A. and Karl, B.J. (1996). Fleshy tralian Weeds Conference Volume 1 fruits of indigenous and adventive pp. 165-8. plants in the diet of birds in forest