Royal Botanic Gardens Melbourne cO o o CM i cO VO GO NATIONAL HERBARIUM OF VICTORIA Celebrating 150 years of plant research in Australia Muelleria publishes research papers on Southern Hemisphere plant, algal and fungal systematics, particularly relating to Australia and the states of Victoria and Tasmania, and to the collections of the National Herbarium of Victoria. Acceptable submissions include: taxonomic revisions; phylogenetic and biogeographical studies; short papers describing new taxa, documenting nationally significant new records, or resolving nomenclatural matters; historical analyses relevant to systematics; any other research contributing to our knowledge of plant, algal or fungal diversity. Muelleria is published annually or semiannually by the National Herbarium of Victoria, Royal Botanic Gardens Melbourne. Manuscripts should be sent in triplicate to: The Editor, Muelleria Royal Botanic Gardens, Melbourne Birdwood Avenue South Yarra Vic. 3141 Australia Muelleria @rbg. vie .gov. au Muelleria may be found online at http://www.rbg.vic.gov.au/muell/. A cumulative index to scientific names as well as the Instructions to Contributors may also be found at this address. Twenty-five reprints of each accepted paper are provided free of charge. Subscription details can be obtained from the address above. Editors Marco Duretto Teresa Lebel Editorial Advisory Committee Tom May Jim Ross Neville Walsh Volume 16 Spring/Summer 2002 © Royal Botanic Gardens Melbourne 2002 ISSN 0077-1813 CONTENTS Volume 16 2002 Contributed Papers Page The Hygrophoraceae of Tasmania — A.M. Young and A.K. Mills 3 Genetic evidence supports reclassification of Agrostis billardierei var. filifolia and A. aemula R.Br. var. setifolia as a single species, A. punicea (Poaceae) — E.A. James, M.C. Ryan, Y.J. Fripp and A.J. Brown 29 Notes on Conothamnus Lindl. with the description of a new section, sect. Gongylocephalus Craven (Myrtaceae) — L.A. Craven 39 Calotis cuneata var. pubescens (Asteraceae), change in rank and notes on its distribution and ecology — N.G. Walsh and K.L. McDougall 43 Successful DNA amplification from Acacia (Leguminosae) and other refractory Australian plants and fungi using a nested/semi-nested PCR protocol — E Udovicic and D.J. Murphy 47 A systematic study of Acacia calamifolia s.l, with special emphasis on A. euthycarpa in Victoria — S.H. Wright, J.W. Grimes, and P.Y. Ladiges 55 Agyrium Fr., Bryophagus Nitschke ex Arnold and Racodium Fr., lichen genera previously unrecorded for Australia — G. Kantvilas 65 Alexander Clifford Beauglehole OAM (26 August 1920-19 January 2002) — J.H. Ross 71 Some new combinations and a new hybrid genus in Orchidaceae: Diurideae, for eastern Australia — J.A. Jeanes 81 Nymphoides simulans (Menyanthaceae): a new species from northern Australia — H.I. Aston 83 Variation within Asterolasia asteriscophora sensu lato (Rutaceae: Boronieae) and the recognition of new taxa in eastern Australia — B.J. Mole, M.F. Duretto, P.Y. Ladiges and E.A. James 87 Muelleria 16 : 3-28 ( 2002 ) The Hygrophoraceae of Tasmania AM. Young '’ 3 and A.K Mills 2 1 Hon. Assoc., Queensland Herbarium, Brisbane Botanic Gardens Mt Coot-tha, Mt Coot- tha Rd, Toowong, Qld, 4066 2 50 Patons Rd, Penguin, Tas. 7316 3 Author for correspondence: Bee Cottage, Langton Rd, Blackbutt, Qld 4306. tyoung@bigpond.com Abstract Twenty-six taxa of Tasmanian Hygrophoraceae are listed or described. New taxa are: Hygrocybe franklinensis A.M.Young & A.K.Mills, Hygrocybe roseoflavida A.M.Young & A.K.Mills and Hygrophorus involutus var. albus A.M.Young & A.K.Mills. Full descriptions and diagrams are pro¬ vided where necessary for all new taxa. Previously described taxa found in Tasmania are listed together with details of any new information and all known herbarium collections for the Tasmanian species. Keys to the species are included. Introduction Little has been published on the Hygrophoraceae of Tasmania other than as minor inclusions in publications that relate to other mycological or botanical investigations. Massee (1899) described the new species Hygrophorus rodwayi Massee from a location near Hobart but fur¬ ther information on Tasmanian Hygrophoraceae was not published for a little over 90 years. Young and Wood (1997) commented on the probable richness of the Tasmanian flora with¬ in family Hygrophoraceae as suggested by the work of Monks (1989) while Young (2000a) indicated that Tasmanian collections had recently been made of both Hygrocybe rodwayi (Massee) A.M.Young and H. lewelliniae (Kalchbr.) Brittleb. ex A.M.Young. A recent paper dealing solely with the Tasmanian Hygrophoraceae is that of Monks and Mills (1991) who provided a more concise description of Camarophyllus rodwayi (Massee) Monks & A.K.Mills. This was followed by the description of the new species Hygrocybe erythrocrenata Monks & A.K.Mills and the demonstration of coloured spore prints for both this new species and Hygrocybe lilaceolamellata (G.Stev.) E.Horak (Mills & Monks 1993). A small but very beautiful collection of coloured photographs of Tasmanian Hygrophoraceae is contained in the booklet on Tasmanian rainforest fungi (Fuhrer & Robinson 1992). Unfortunately, little other information is provided, however many of the 16 taxa depicted are easily recognisable, excellent definitive illustrations for these species are provided, and are cited as such within this paper. There is no longer any doubt as to the large number of species of Hygrophoraceae in Tasmania; however this paper should be considered only as a preliminary survey of the Tasmanian taxa. Although three new taxa are described, the paper’s principal aim is to provide a foundation for future studies on the Tasmanian Hygrophoraceae by providing a census of previously described taxa that occur in Tasmania together with any pertinent data. Collections made by both authors during the seasons of 1998 and 1999 indicate that a considerable number of taxa remain to be formally described. The Tasmanian species of Hygrophoraceae differ considerably in known habitat when compared to the mainland flora. Mainland taxa are found in a wide variety of habitats which include grassland, heath, dry sclerophyll woodland, dry and wet sclerophyll forest, various forms of ra inf orest and beech forest ( Nothofagus spp.). This is not the case for Tasmanian taxa which are almost exclusively found in cool temperate rainforests that are usually dominated by Nothofagus cunninghamii (southern beech) and Artherosperma moschatum (sassafras) with an under-story of tree ferns ( Cyathea and Dicksonia spp.). 4 A.M. Young and A.K. Mills The floors of these forests contain not only abundant, moist litter, but also immense car¬ pets of moss, and the combination of these two substrates (together with the conditions of temperature and humidity maintained under the forest canopy) seems to be particular¬ ly suitable for the fruiting of species of Hygrophoraceae. Assuming the normal seasonal rainfall, some species commence to produce fruiting bodies in early April but a maximum of both species fruiting and basidiomes produced by each species occurs during May and early June. Some species continue to produce considerable numbers of fruiting bodies in July and a few species may still produce scattered basidiomes as late as mid-August. Two instances where taxa occurred outside the cool temperate rainforest were noted during 1999 when Hygrocybe cantharellus (Schwein.) Murrill was collected from amongst moss in heath and a small collection of an undescribed species of Hygrocybe was found on a moss bank beside a path in coastal eucalypt forest. Extensive collecting by the second author has demonstrated that when taxa of Hygrophoraceae have been col¬ lected in the Tasmanian wet sclerophyll forests (dominated by Eucalyptus spp.), then the collection sites are marginal zones where cool temperate rainforest tree species intermin¬ gle with the wet sclerophyll forest species. There are no known collections of Tasmanian Hygrophoraceae from that state’s dry sclerophyll forests and only a single collection from pasture is here recorded. Generally, the basidiomes of species as they occur in Tasmania, have characteristics which conform very closely to the known characters of the taxa as they occur on the Australian mainland, however they do occasionally differ. Spore sizes in some Tasmanian Hygrophoraceae are sometimes larger than those of either or both the holotype or the Australian mainland representatives of the respective species. This is true for the taxa Hygrocybe cheelii A.M.Young and Hygrocybe irrigata (Pers.: Fr.) Bon. Other micro¬ scopic differences observed include much longer basidia in basidiomes of material cur¬ rently accepted as Hygrocybe stevensoniae T.W.May & A.E.Wood and spore shape vari¬ ations in both Hygrocybe rodwayi and Hygrocybe astatogala (R.Heim) Heinem. Biogeographically, the Tasmanian species of Hygrophoraceae also show relationships with the Hygrophoracee of New Zealand. A number of taxa occur in both geographical locations and New Zealand is the holotype location for several species. Links with South America are less well defined, however the species Hygrocybe reesiae A.M.Young is undoubtedly related to the South American taxon Camarophyllus adonis Singer. Several taxa such as Hygrocybe cantharellus and H. miniata (Fr.: Fr.) P.Kumm. are cosmopolitan, although first described from Europe, but others such as H. astatogala have tropical distributions. Materials and Methods All material collected in Tasmania during 1998-1999 was air dried and preserved for later microscopic examination. Field notes were made from the fresh material and the accom¬ panying colour codes refer to Komerup and Wanscher (1981). Photographic slides were made either in the field or in the laboratory. The 1998-1999 material collected by the first author forms the basis for this paper supplemented with material from the second author’s herbarium. Distribution notes for each taxon are limited to Tasmania, however many of the taxa occur on the Australian mainland and in New Zealand. The Tasmanian material (other than holotypes) cited from the 1998-1999 collections has been deposited in either the Queensland State Herbarium (BRI) or the National Herbarium of Victoria (MEL). Holotype collections and all cited material from the Mills collection have been deposited in the Tasmanian Herbarium (HO). Collection numbers from each author’s personal herbarium (cited as ‘ hb . Young ’ and ‘ hb . Mills' respectively) are provided for reference purposes. Material was also studied from AD, K, MICH, PDD, UNSW, ZT (Holmgren et al. 1990). Material in collections with ZT numbers has been divided: part remains in ZT and part is in BRI. Hygrophoraceae of Tasmania 5 All microscope work was done on an Olympus CX40 binocular microscope with drawing tube attachment. The drawing tube was calibrated to provide scale drawings using an Olympus standard 1 mm slide. Material intended for microscopic analysis was rehydrated in ammoniated congo-red and gently warmed if necessary. Illustrations are provided for the new taxa and for those species which are either not illustrated in previous papers (Young & Wood 1997; Young 1999; Young 2000a; Young et al. 2000) or which require additional diagrams as a result of new information. The habit- sketch shows basidiome dimensions. The microstructures of the pileus, hymenophoral trama and stipe are not depicted because they conform to standard forms (Young & Wood 1997). For each illustrated specimen, 20 spores and 10 basidia were selected at random, drawn and measured. The derived parameter ‘Q’ is defined as the quotient of the length divided by the width of the relevant spore or basidium; the mean ‘Q’ is the quotient of the mean length and width respectively. This paper lists several species of Hygrophoraceae originally collected and described from Europe that are stated to have no type (Boertmann pers. com.). This problem has already been addressed (Young 2000a) and where types for European taxa do not exist, the species concepts of Boertmann (1995) are used. Taxonomy Family Hygrophoraceae Lotsy, Vortr. Bot. Stammesg. 1: 706 (1907). Typical genus : Hygrophorus Fr. Basidiome small to medium sized, stipitate. Pileus conical, convex, umbilicate or infundibuliform; sometimes perforate; surface dry, moist, viscid or glutinous, smooth to squamulose or fibrillose. Lamellae generally thick, waxy, and distant; free or adnexed to decurrent. Stipe central, often brittle, with similar surface moisture or structures to pileus. Universal veil generally absent. Context soft, frequently thin, waxy and translucent. Spore print white, cream, pale violaceous or magenta. Spores small to large, subglobose to ovoid, ellipsoid or cylindrical, sometimes constricted, smooth, more rarely nodulose or echinulate, hyaline or rarely with dark contents, inamyloid rarely amyloid. Basidia often long and narrow. Cheilocystidia sometimes present, pleurocystidia rare and then as pseu- do-pleurocystidia. Hymenophoral trama regular, irregular or bilateral. Pileipellis a cutis, trichoderm, ixocutis or ixotrichoderm, rarely a hymeniderm or epithelium. Clamp con¬ nections present or absent. Development gymnocarpic, occasionally hemiangiocarpic. Terrestrial, rarely lignicolous, mycorrhizal or saprobic. Key to the tribes of Hygrophoraceae 1. Lamellae with regular to irregular trama, never divergent .Tribe 1. Hygrocybeae 1. Lamellae with divergent trama .Tribe 2. Hygrophoreae TRIBE 1. HYGROCYBEAE Kiihner, Bull. mens. Soc. linn. Lyon 48: 621 (1979). Typical genus : Hygrocybe (Fr.) P.Kumm. Hymenophoral trama regular to irregular; not forming ectomycorrhizae. Key to the genera of Hygrocybeae 1. Pileipellis composed of hyphae forming a cutis, ixocutis, trichoderm or ixotrichoderm of non-inflated, hyphal elements .Genus 1. Hygrocybe 1. Pileipellis an hymeniderm but sometimes approaching an epithelium and then com¬ posed of inflated elements (one species known for Tasmania) . .Genus 2. Camarophyllopsis 6 A.M. Young and A.K. Mills GENUS 1. HYGROCYBE (Fr.) P.Kumm., FUhr. Pilzk.: 26 (1871); Hygrocybe Fr., Syst. Mycol. 1: 101 (1821); Camarophyllus Fr., Syst. Mycol. 1: 98 (1821); Camarophyllus (Fr.) P.Kumm., Fiihr. Pilzk.: 2 (1871). Typical species: Agaricus conicus Schaeff., Fungi Bavariae 4: 2 (1774). Basidiome fleshy, often watery or waxy in texture, collybioid, mycenoid or omphaloid, generally small to medium sized but occasionally large; variously coloured, often bright red, orange, yellow, green and lilac or combinations of these colours. Pileus opaque or hygrophanous, striate or not, dry to glutinous, smooth to squamulose or fibrillose. Lamellae usually sub-distant to distant, free to adnate or decurrent, thick to very thick and with waxy appearance when fresh; velar structures absent. Universal veil absent. Stipe dry to glutinous, smooth to squamulose or fibrillose; spore print white, cream coloured, pale magenta or pale lilac. Spores hyaline, smooth or rarely spinose, non-amyloid (for known Australian taxa). Basidia sometimes long (50-70 pm), Q: 2.5-10.0, 2-and 4- spored forms frequent, clamp connections usually present. Cheilocystidia present in some species either as true or pseudo-cheilocystidia. Pleurocystidia very rare and then as pseu- do-pleurocystidia. Hymenophoral trama regular, subregular to irregular, tramal elements from very long (> 1000 pm) to very short (< 30 pm); clamp connections usually present. Pileipellis a cutis, ixocutis, trichoderm or ixotrichoderm. Development gymnocarpic and stipitocarpic. Habitat and Distribution: Solitary to gregarious, terrestrial, soil, humus or moss, rarely on rotten wood; found in various ecosystems from grasslands to forest and con¬ sidered to be saprobic. Cosmopolitan from subarctic or subantarctic to tropics and alpine regions. Key to the subgenera of Hygrocybe 1. Hymenophoral trama irregular, composed of short (20-150 pm) interwoven hyphal elements; basidiome colours often subdued (white, brown, dull lilac-grey) but may be orange, apricot or bright lilac; lamellae arcuate to decurrent; clamps present, occasionally rare in the hymenophoral trama.subgen. 1. Cuphophyllus Key 1. 1. Hymenophoral trama regular to subregular (if subregular, then basidiome brightly coloured) and composed of parallel hyphal elements which are either ‘long tubular’ or chains of short elements; basidiome often very brightly coloured (red, orange, yellow, green, lilac); lamellae variously attached; clamps present, at least at the bases of the basidia.2 2(1). Hymenophoral trama very regular, composed of very long (1000-3000 pm), asep- tate, parallel, tubular elements with tapered ends; lamellae free, ascending or nar¬ rowly adnate; tissues may blacken on bruising; basidia usually short (mean length 30-40 (45) pm); except for the aseptate hymenophoral trama, clamps usually pres¬ ent throughout the basidiome, rarely absent in some taxa with 2-spored basidia (one species known for Tasmania) .subgen. 2. Hygrocybe 2. Hymenophoral trama regular to subregular, composed of parallel chains of short, sometimes inflated hyphal elements (usually 20-400 pm); lamellae more or less free to adnate or arto decurrent; tissues never blackening on bruising; basidia some¬ times long (40-60 pm); clamps either present throughout the basidiome or present only at the bases of the basidia.3 3(2). Clamps present throughout the basidiome and of medallion form or not . .subgen. 3. Pseudohygrocybe Key 2. 3. Clamps absent throughout the basidiome except at the bases of the basidia and then frequently of medallion form.subgen. 4 Humidicutis Key 3. Hygrophoraceae of Tasmania 7 Key 1: Tasmanian species of subgenus Cuphophyllus 1. Pileus off-white to cream coloured and often with biscuit brown tints at the depressed centre .4. H. rodwayi 1. Pileus yellow, yellow-orange or some shade of lilac or pinkish lilac.2 2(1). Pileus yellow or yellow-orange.1. H. aurantiopallens 2. Pileus a shade of lilac to pinkish lilac.3 3(2). Pileus lilac to greyish lilac, hygrophanous; margins not involute when juvenile; stipe base lilac.3. H. reesiae 3. Pileus bright pinkish lilac, not hygrophanous; pileus margins involute when juve¬ nile; stipe base yellow.2. H. cheelii Key 2: Species of subgenus Pseudohygrocybe 1. Pileus glutinous to viscid and pileipellis always an ixotrichoderm .2 1. Pileus dry and pileipellis a cutis or trichoderm, or viscid and pileipellis an ixocutis .7 2(1). Pileus white or yellow.3 2. Pileus green, a shade of brown or grey.4 3(2). Pileus and stipe white; lamellae adnexed to narrowly adnate.15. H. leucogloea 3. Pileus and stipe yellow; lamellae decurrent.8. H. chromolimonea 4(2). Lamellae margins fertile and without a gluten thread; hyphal cheilocystidia absent .5 4. Lamellae margins sterile and with a gluten thread; hyphal cheilocystidia present .. .6 5(4). Pileus green.20. H. stevensoniae 5. Pileus grey to grey-brown.13. H. irrigata 6(4). Pileus green, lamellae green; spores 8.5-10.5 x 5-8 pm; dried material dull green .18. H. pseudograminicolor 6. Pileus green to brown, lamellae white or white with green or brownish tints; spores 5.5-7.5 x 3.5-5 pm; dried material brick-pink .12. H. graminicolor 7(1). Spores dimorphic; macrospores 11-18 pm long .10. H. firma 7. Spores monomorphic; spores <11 pm long.8 8(7). Pileipellis a trichoderm (at least at the centre).9 8. Pileipellis a cutis or ixocutis.10 9(8). Lamellae deeply decurrent, cream coloured to yellowish.7. H. cantharellus 9. Lamellae broadly adnate with at most a decurrent tooth, yellow with pink flush .... .17. H. miniata 10(8). Pileus viscid; pileipellis an ixocutis.11 10. Pileus dry; pileipellis a cutis.13 11(10). Pileus conical; lamellae narrowly adnate to ascending-adnate . .11. H. franklinensis 11. Pileus convex; lamellae arcuate to decurrent.12 12(11). Lamellae pallid pink; stipe base yellow.19. H. roseoflavida 12. Lamellae cream-coloured; stipe base orange-red or red. .6a. H. anomala var. anomala 13(10). Stipe yellow.14. H. julietae 13. Stipe red, orange or brown.14 14(13). Pileus and stipe brown, stipe base usually mauve tinted 16. H. lilaceolamellata 14. Pileus red to red-brown or orange-brown, stipe red to reddish orange, stipe base concolorous.15 15(14). Lamellae delicately tinted mauve or lilac; spinose spores always present . .6a. H. anomala var. ianthinomarginata 15. Lamellae white, pinkish white, reddish grey or lilac red; spinose spores absent .9. H. erythrocrenata A.M. Young and A.K. Mills Key 3: Species of subgenus Humidicutis 1. Basidiomes wholly white .22. H. mavis 1. Basidiomes wholly pale lilac to violet.21. H. lewelliniae Species Information and Descriptions Subgenus 1 Cuphophyllus Donk, Beih. Nova. Hedwigia 5: 45 (1962). Typical species: Agaricus pratensis Pers.: Fr. [= Camarophyllus pratensis (Pers.: Fr.) PKumm.] Basidiome dull coloured or rarely with bright colours in apricots, pinks or lilac to mauve; lamellae mostly decurrent; hymenophoral trama irregular; cystidia mostly absent; clamp connections frequent throughout the basidiome. 1. Hygrocybe aurantiopallens (E.Horak) A.M.Young in Young & Wood, Austral. Syst. Bot. 10: 921 (1997). Camarophyllus aurantiopallens E.Horak, Beih. Nova Hedwigia 43: 122 (1973). Type: New Zealand, Lake Rotoiti, 29.iv.1968, E.Horak s.n. (holotype PDD 27088). Misappl.: Hygrophorus aurantius Murrill sensu G.Stev., Kew Bull. 16: 382 (1963). Illustrations: Fuhrer & Robinson (1992), p. 38; Young & Wood (1997), p. 922. Habitat and distribution: gregarious on soil amongst litter in cool temperate rainfor¬ est, often at the bases of tree ferns. The species is widespread in southern, central and north-western Tasmania. Material: Mt Field National Park, 30.iv.1998, A.M.Young ( hb. Young 1974; BRI); Julius River, 6.V.1998, AM.Young (hb. Young 1985: BRI); Milkshake Reserve, 6.V.1998, A.M.Young (hb. Young 1993: BRI); Franklin R., 7.v. 1998, A.M. Young (hb. Young 2001: BRI); Tahune, 27.iv. 1998, A.M. Young (hb. Young 1961: MEL 2087780 ); Tahune, ll.v.1998, AM.Young (hb. Young 2012: MEL 2087777 ); Growlingswallet nr Mt Field, 19.V.1999, A.M.Young (hb. Young 2227: MEL 2087772 ); Franklin R., 21.V.1999, AM.Young (hb. Young 2241: MEL 2087769 ); Cradle Mt, 3.vi.l992, T.W.May s.n. (MEL 259607): Arve Loop Rd nr Geeveston, iv.1998, A.K.Mills (hb. Mills 1510: HO 508592). Remarks: Basidiomes of Hygrocybe aurantiopallens can vary in their colouration from brilliant orange to orange-yellow or apricot-yellow. Young (1999) noted the proba¬ ble misidentification of Fuhrer and Robinson (1992) of a specimen considered to be Camarophyllus apricosa (E.Horak) E.Horak. No Tasmanian material has yet been found which can be assigned to C. apricosa which has a conical pileus and distinctly ellipsoid spores. The misapplication of Hygrophorus aurantius Murrill by G. Stevenson is covered by Horak (1990), p. 278. 2. Hygrocybe cheelii A.M.Young, Austrobaileya 5: 547 (1999). Type: New South Wales. Gladesville, 17.vi.1916, J.B.Cleland, s.n. (holotype AD 3418). Cantharellus lilacinus Cleland & Cheel, Trans. & Proc. Roy. Soc. S. Australia 43: 271 (1919). Type: New South Wales. Gladesville. 17.vi.1916. J.B.Cleland, s.n. (holotype AD 3418). Camarophyllus lilacinus (Cleland & Cheel) E.Horak, New Zealand J. Bot. 28: 203 (1990); non Hygrocybe lilacina (C.Laest. ex P.Karst.) M.Moser, Die Rohrlinge und Blatterpilze (Agaricales) 3 ed., 64 (1967). Illustrations: Young (1999), p. 547; Willis (1963), plate 9, fig.l as Cantharellus lilac¬ inus ; Cleland & Cheel (1919), Plate 29, fig.l. Habitat and distribtuion: gregarious on soil amongst leaf litter or moss in cool tem¬ perate rainforest; occasionally in pasture grass. Hygrocybe cheelii is widespread and fre¬ quent in southern and central Tasmania. Hygrophoraceae of Tasmania 9 Material examined; Sandspit Forest Reserve nr Wielangta, 27.ix.1997, G.Gates (hb. Mills 7595; HO 508608)-, Growlingswallet nr Mt Field, 1.x. 1998, A.K.Mills (hb. Mills 1596; HO 508609 ); Growlingswallet nr Mt Field, 1.x. 1998, A.K.Mills (hb. Mills 1597; HO 508610); Marion Bay nr Dunally, 5.vi.l999, G.Gates (hb. Mills 1649; HO 508615; MEL 2087782); Jacksons Bend, 14.X.1998, G.Gates (hb. Mills 1602; HO 508571). Notes; These are the first collections from Tasmania assigned to this taxon. The spores of these collections measure (6.5-)8-10(-10.5) x (5.5—)6—6.5(—8) pm, mean range 8.3-9.1 x 6.3-6.6 pm, Q: 1.1-1.6(-1.8), range of mean Q: 1.31-1.45 and are somewhat larger than those of the holotype which measure 6-8.5 x 4.5-6 pm, mean 7.2 x 5 pm, Q: 1.2-1.7, mean Q: 1.43 although they have more or less the same dimension ratio as indi¬ cated by the similarity of the two sets of Q measurements. The basidia have no signifi¬ cant dimensional differences. It is interesting to note however, that a collection of H. cheelii from Victoria (Young 2000c) has spores measuring 7 -10 x 4.5-5.5(-6.5) pm, mean 8.4 x 5.4 pm, Q: 1.3-1.8, mean Q: 1.56. One possible hypothesis is that spore size increases with latitude, but more collections of the taxon will be required to confirm this. It may also be that the larger spores found in the Victorian and Tasmanian collections are ‘normal’ while the smaller spores of the holotype represent a taxon variety with unusual¬ ly small spores. Collection hb. Mills 1649 is of interest as it contains a collection in which most of the basidia are 2-spored and only very occasional basidia are 1-, 3- or 4-spored. Clamps are absent or extremely rare throughout the basidiome and only a single clamp connection was sighted in cuticular tissues during the examination. Similar clampless basidiomes where the basidia are predominantly 2-spored are well known in the Hygrophoraceae (Boertmann 1995; Young 2000a). 3. Hygrocybe reesiae A.M. Young in Young & Wood, Austral. Syst. Bot. 10: 923 (1997). Type; New South Wales. Lane Cove Bushland Park, 17.vi. 1990, R.Kearney & B.Rees s.n. (holotype UNSW 90/205). Illustration; Young & Wood (1997), p. 924. Habitat and distribution; gregarious on leaf mould in cool temperate rainforest; fre¬ quently amongst moss or litter. The species is frequent in southern and central Tasmania. Material; Arve Loop Rd nr Geeveston, ll.v.1998, B.Rees (hb. Young 2016; BRI); Mt Field, 24.iv.1997, A.K.Mills (hb. Mills 1456; HO 508586); Mt Field, 24.iv.1997, A.K.Mills (hb. Mills 1460; HO 508589); Location and date unknown; A.J.Monks (ZT 4574; BRI); Mt Field, 16.iv.1991, E.Horak (ZT 4350; BRI). Remarks; Hygrocybe reesiae is at first bright lilac with a hygrophanous pileus. As dry¬ ing proceeds, the lilac colour becomes fainter and the pileus surface changes to a lilac tinted buff. There is never a distinctly yellow stem base as occurs in H. cheelii. Some Tasmanian collections have exhibited a tendency to produce more ellipsoid spores with very few subglobose spores. As a result, spore measurements remain very similar to those of the holotype, but Q’s are greater (1.4-1.7, mean 1.48) when compared to those of the holotype (1.1-1.5, mean 1.3). There are no other major differences. There is no doubt that H. reesiae is very closely related to the South American taxon Camarophyllus adonis, and both taxa almost certainly have a common ancestry. Basidiomes of C. adonis have spores which measure 6-9 x 4.5-6.5 pm and basidia which measure 40-60 x 5-7 pm (Horak 1979) and these dimensions are almost identical to those found in basidiomes of H. reesi¬ ae. Camarophyllus adonis has a robust basidiome, branching lamellae which are at most lilac tinted and a stipe which is whitish, yellowish or earthy brown; basidiomes of H. reesiae are more slender, have simple lamellae which are bright lilac or violet and violet coloured stipes which dry slowly to buff. Future genetic analysis may show that H. reesiae is the Australian variant of C. adonis but until this is done, the geographical and 10 A.M. Young and A.K. Mills macrocharacter differences of the two taxa are considered sufficient reasons to maintain the separation of the two species. 4. Hygrocybe rodwayi (Massee) A.M.Young in Young & Wood, Austral. Syst. Bot. 10: 923 (1997); Hygrophorus rodwayi Massee, Bull. Misc. Inf. Kew 1899: 178 (1899). Camarophyllus rodwayi (Massee) A.J.Monks & A.K.Mills in Banks et al. (eds), Aspects of Tasmanian Botany-A Tribute to Winifred Curtis 13 (1991). Type: Tasmania. Kingston Rd (nr Hobart), undated, L.Rodway 137 (holotype K). Illustrations: Fuhrer & Robinson (1992), p. 39; Young & Wood (1997), p. 925. Habitat and distribution: gregarious to caespitose amongst moss or on soil amongst lit¬ ter in cool temperate rainforest. Hygrocybe rodwayi is common in southern and central Tasmania. Material: Little Florentine River, 13.v. 1998, A.M.Young (hb. Young 2023: BRI); Johns Rd nr Geeveston, 19.V.1998, AM.Young (hb. Young 2042: BRI); Johns Road nr Geeveston, 19.V.1998, A.M.Young (hb. Young 2043: MEL 2087776 ); Little Florentine River, 26.v. 1999, AM.Young (hb. Young 2268: MEL 2087766): Five Road, 24.iv.1997, A.K.Mills (hb. Mills 1451: HO 508572 ); Mt Field, 24.iv.1997, A.K.Mills (hb. Mills 1454: HO 508574 ); Johns Road nr Geeveston, 24.ix.1997, A.K.Mills (hb. Mills 1471: HO 508575 ); Johns Road nr Geeveston, 24.ix.1997, A.KMills (hb. Mills 1472: HO 508576): Johns Road nr Geeveston, 24.ix.1997, A.K.Mills (hb. Mills 1473: HO 508577: MEL 2087783): Johns Road nr Geeveston, 24.ix.1997, A.K.Mills (hb. Mills 1477: HO 508578): Arve Loop Rd nr Geeveston , ll.v.1998, A.K.Mills (hb. Mills 1535: HO 508580): Sandspit Forest Reserve nr Wielangta, 3.vi.l998, A.KMills (hb. Mills 1553: HO 508582): Sandspit Forest Reserve nr Wielangta, 3.vi.l998, A.K.Mills (hb. Mills 1554: HO 508583): Rutherford Rd nr Geeveston, 18.vi.1998, A.K.Mills (hb. Mills 1570: HO 508570): Location and date unknown, A.J..Monks (ZT 4575: BRI). Remarks: The decurrent lamellae and the cream discolouration at the centre of the other¬ wise white pileus are characteristic of this taxon. Occasional large basidiomes (pilei approaching 40 mm diameter) may be encountered, but all other characters remain constant. Microscopically, the small, subglobose spores measuring (4.5-)5-7(-7.5) x 4.5-5.5(-6) pm are very distinctive. Macroscopically, H. rodwayi could be confused with H. virginea (Wulfen: Fr.) P.D.Orton & Watling, but the latter species is easily separated microscopically because it has larger, oblong to ellipsoid spores measuring 7.0-12.5 x 3.5-7.5 pm. Subgenus 2 Hygrocybe Basidiome frequently vividly coloured (red, orange, yellow); pileus often conical; lamel¬ lae free, adnexed or narrowly adnate; hymenophoral trama strictly regular, composed of very long (500-3000 pm), tubular, aseptate elements with tapered ends; cystidia some¬ times present; clamp connections generally present throughout the basidiome. One species in Tasmania. 5. Hygrocybe astatogala (R.Heim) Heinem., Bull. Jard. Bot. Etat 33: 436 (1963); Bertrandia astatogala (R.Heim) R.Heim, Rev. Mycol. 31: 155 (1966). Type: Madagascar, (holotype P, n.v.) Illustrations: Fuhrer & Robinson (1992), p. 8; Young & Wood (1997), p. 933. Pileus conical, often red or yellow but rapidly becoming black; all tissues rapidly black¬ ening on bruising and exuding a pale, watery yellow fluid if cut, pileal and stipe surfaces covered in abundant, blackish fibrils. Habitat and distribution: solitary, gregarious or sometimes in troops amongst leaf lit¬ ter, moss or directly on soil in cool temperate rainforest; often in very sheltered locations. The species is widely distributed in central, northern and western Tasmania. Hygrophoraceae of Tasmania 11 Material : Arve Loop Rd nr Geeveston, 27.iv.1998, A.M.Young (hb Young 1953; BRI); Mt Field, 30.iv.1998, A.M.Young (hb Young 1981; BRI); Tahune, 17.V.1999, A.M.Young (hb Young 2214; MEL 2087775 ); Tasman Pen., 25.v. 1999, A.M.Young (hb Young 2252; MEL 2087768 ); Fire Rd nr Geeveston, 24.iv.1997, A.K.Mills (hb Mills 1452; HO 508573 ); Johns Rd nr Geeveston, A.K.Mills (hb Mills 1478; HO 508579 ); Sandspit Forest Reserve nr Wielangta, 3.vi. 1998, A.K.Mills (hb Mills. 1552; HO 508581); Franklin R., 13.iv.1991, E.Horak fZT 4323; BRI); Collinsvale, 3.iv.l991, E.Horak fZT 4806; BRI). Remarks; Hygrocybe astatogala is widespread and often abundant in Tasmania; troops of 20-30 basidiomes are frequently encountered. Most collections are from amongst leaf litter rather than moss. Although juvenile basidiomes are brilliant red, mature basidiomes are often jet black except for the yellow lamellae. Monks (1989) described Tasmanian variants of H. astatogala which differed only in that they had ellip¬ soid spores [8.5-10.5 x 5.5-7 pm, mean 9.5 x 6.2 pm, mean Q: 1.47] rather than the usual sub-globose spores [7.5-10 x 6-8.5 pm, mean 8.7 x 6.8 pm, Q: 1.0-1.6, mean Q: 1.28]. Two Mills collections (hb. Mills 1452 and hb. Mills 1478 ) confirm Monks’ findings because both of these collections have ellipsoidal spores (many with darker contents) measuring 7-10 x 5.5-6.5 pm, mean 8.8 x 6.3 pm, Q: (l.l-)l.2-1.6, mean Q: 1.39; all other characters of these basidiomes with ellipsoidal spores conform to those of the basidiomes with subglobose spores. Monks also suggested that the difference in spore shape may be a function of the substrate and that moss/litter/peat substrates yield basid¬ iomes with sub-globose spores while substrates derived at least in part from siliceous rocks yield basdiomes with ellipsoid spores. It is not known if the Tasmanian collections detailed above conform to Monks’ hypothesis. Subgenus 3 Pseudohygrocybe Bon, Doc. Mycol. 24: 42 (1976). Typical species; Hygrocybe coccinea (Schaeff: Fr.) P.Kumm. Basidiome variously coloured often brightly (red, orange, yellow, green, lilac); pileus conical, convex or umbilicate; lamellae narrowly adnate to decurrent; cystidia sometimes present as cheilocystidia, rarely as pseudo-pleurocystidia; hymenophoral trama regular, subregular to slightly irregular, composed of short, cylindrical to inflated elements 20-300 pm long (rarely up to 700 pm); clamp connections generally present throughout the basidiome. 6. Hygrocybe anomala A.M.Young, in Young & Wood Austral. Syst. Bot. 10: 919 (1997). Type; New South Wales, Blackheath, 23.vi. 1983, A.E.Wood s.n. (holotype UNSW 83/991). 1. Pileus at first viscid; without lilac tints on pileus or lamellae.6a. var. anomala 1. Pileus always dry; lilac tints present on either or both pileus and lamellae. .6b.var. ianthinomarginata 6a. Hygrocybe anomala A.M.Young var. anomala Illustration; Young & Wood (1997), p. 920. Habitat and distribution; gregarious amongst moss in cool, temperate rainforest; vari¬ ety anomala is known from locations in southern, central and north-western Tasmania. Material; Julius River nr Smithton, 6.vi.l998, A.M.Young (hb. Young 1987; BRI); Arve Loop Rd nr Geeveston, 1 l.v.1998, A.M.Young (hb. Young 2020; BRI); Little Florentine River, 26.v. 1999, A.M.Young (hb. Young 2263; BRI); Arve Loop Rd nr Geeveston, 27.iv.1998, A.M.Young (hb. Young 1963; MEL 2087779); Arve Loop Rd nr Geeveston, 17.v. 1999, A.K.Mills (hb. Young 2222; MEL 12 A.M. Young and A.K. Mills 2087774 ); Arve Loop Rd nr Geeveston, 17.V.1999, A.K.Mills (hb. Young 2224 ; MEL 2087773 ); Little Florentine River, 26.V.1999, A.M.Young (hb. Young 2264] MEL 2087767). Remarks: Hygrocybe anomala var. anomala has both a viscid pileus and stipe and there are no lavender/lilac tints on either pileus or lamellae. It appears to be widespread and fairly co mm on in Tasmania. 6b. Hygrocybe anomala A.M.Young var. ianthinomarginata, A.M.Young, Austrobaileya 5: 551 (1999). Type: New South Wales, Lane Cove Bushland Park, 13.vi. 1998, R. & E.Kearney & A.M.Young (hb. Young 2111 ) (holotype DAR 73918). Illustration: Young (1999), p.552. Habitat and distribution: gregarious amongst moss or humus in cool temperate rain¬ forest: variety ianthinomarginata is known from southern and north-western Tasmania. Material: Julius River, 6.V.1998, A.M.Young (hb. Young 1986 ; BRI); Johns Road, 19.V.1998, A.M.Young (hb. Young 2048 ; BRI); Franklin River, 21.v. 1999, A.M.Young (hb. Young 2247 ; BRI); Franklin R., 7.V.1998, A.M.Young (hb. Young 2002 ; MEL 2087778 ); Growlingswallet nr Mt Field, 19.v. 1999, G.Gates (hb. Young 2231 ; MEL 2087771). Remarks: Tasmanian material identified as this variety always has a dry pileus and stipe. Microscopic examination shows that both pileus and stipe have a simple cutis and never display the ixocutis present in var. anomala. The type description states that the pileus has a lilac margin and may be slightly umbonate, whereas Tasmanian material is often distinctly umbonate and the lilac marginal tints are weak or lacking. However, the lamellae are always distinctly lilac tinted and microscopic examination always shows the presence of spinose spores. Despite the slight differences, the Tasmanian material is con¬ fidently assigned to var. ianthinomarginata. 7. Hygrocybe cantharellus (Schwein.) Murrill, (as Hydrocybe), Mycologia 3: 196 (1911); Agaricus cantharellus Schwein., Schriften Naturf. Ges. Leipzig 1: 88 (1822); Hygrophorus cantharellus (Schwein.) Fr., Epicr. Syst. Mycol.: 329 (1838). Type: none designated. Illustrations: Boertmann, D (1995): p. Ill; Young & Wood (1997), p. 962. Habitat and distribution: gregarious amongst moss and humus in cool temperate rain¬ forest or gregarious amongst moss in heath. The species appears to be widespread in southern and central Tasmania. Material: Tahune, 17.v. 1999, A.M.Young & A.K.Mills (hb. Young 2213 ; BRI); Arve Loop Rd nr Geeveston, 27.iv.1998, A.M.Young (hb. Young 1968 ; MEL 2088590 ); Limebay Reserve, 12.V.1999, A.K.Mills (hb. Young 2209\ MEL 2088608 ); Arve Loop Rd nr Geeveston, 17.V.1999, A.K.Mills (hb. Mills 1514 ; HO 508596). Remarks: Tasmanian material has the usual distinguishing characteristics of this taxon: a dry, red pileus with a finely velvety surface (at least at the centre), bright red, dry stipe, and pale yellow, decurrent lamellae. The basidiomes usually have stipes that are at least 2-3 times longer than the pileus diameter. The species does not appear to be partic¬ ularly common. Hygrocybe cantharellus is one of the very few species of Tasmanian Hygrophoraceae that has been found outside the cool temperate rainforests; there is a sin¬ gle collection (hb. Young 2209) from heath. 8. Hygrocybe chromolimonea (G.Stev.) T.W.May & A.E.Wood, Mycotaxon 54: 147-150 (1995); Hygrophorus chromolimoneus G.Stev., Kew Bull. 16: 383 (1962). Gliophorus Hygrophoraceae of Tasmania 13 chromolimoneus (G.Stev.) E.Horak, Beih. Nova Hedwigia 43: 167 (1973). Type : New Zealand. Lake Rotoiti, 16.v. 1956, E.B.Kidson (Stevenson 1088) (holotype K). Illustrations; Fuhrer and Robinson (1992): p. 41; Young & Wood (1997), p. 964. Habitat and distribution; gregarious to caespitose amongst moss or humus in cool temperate rainforest; sometimes in association with old tree fem bases. The species is widespread and common. Material. Arve Loop Rd nr Geeveston, 27.iv.1998, A.M.Young (hb. Young 1967; BRI); Milkshake Reserve, 6.v. 1998, A.M.Young (hb. Young 1997; BRI); Julius R., 6.v. 1998. A.M.Young (hb. Young 1990; MEL 2088592 ); Little Florentine R., 13.V.1998, A.M.Young (hb. Young 2027; MEL 2088603 ); Growlingswallet nr Mt Field, 19.V.1999, G.Gates (hb. Young 2232; MEL 2088611 ); Arve Loop Rd nr Geeveston, 17.V.1998, A.K.Mills (hb. Mills 1512; HO 508594 ); Arve Loop Rd nr Geeveston, 17.V.1998, A.K.Mills (hb. Mills 1513; HO 508595 ); Roger River Reserve, 6.V.1998, A.K.Mills (hb. Mills 1520; HO 508599 ); Sandspit Forest Reserve nr Wielangta, 3.vi.l998, A.K.Mills (hb. Mills 1550; HO 508605 ); Sandspit Forest Reserve, 23.iv.1991, E.Horak (ZT 4957; BRI); Franklin R., 13.iv.1991, E.Horak (ZT 4336; BRI). Remarks; The viscid and wholly bright yellow basidiomes with convex, umbilicate pilei are very distinctive. The greyish, glutinous margin on each lamella can be seen with the unaided eye but is very easily seen with a xlO lens. A difficulty when examining herbarium material is that the cheilocystidia often collapse and adhere strongly to each other within the gluten layer but the latter does remain very clearly visible as a translu¬ cent region on the lamella margin. Gentle heating in the microscope mountant is often required before the cheilocystidia will separate so that they can be clearly observed. Tasmanian material of Hygrocybe chromolimonea sometimes has slightly larger spores [(8-)8.5-10.5(-12) x 4.5-6.5 pm] than either the holotype (7.5-9.5 x 4.5-6 pm) or typical Australian mainland material [(6.5—)7—9(—11) x 4-6(- 6.5) pm] however because there is so much overlap of spore ranges and all other aspects of the basidiomes remain constant, this difference in spore size is not considered important. 9. Hygrocybe erythrocrenata Monks & A.K.Mills, Mycotaxon 46: 87 (1993). Type; Tasmania. Moonlight Ridge nr Lune River, viii.1989, A.J.Monks s.n. (holotype HO 131196). Illustration; Mills & Monks (1993), p. 89. Spores (6-)7-9.5 x 3.7-5.5 pm, mean 7.5 x 4.3 pm, Q: 1.4-2.1, mean Q: 1.73, ellipsoid to subcylindrical, smooth, hyaline, occasionally constricted. Basidia 41-59 x 6-8.5 pm, mean 48.4 x 7.2 pm, Q: 5.3-8.8, mean Q: 6.71, 2- or 4-spored, clamped. Hymenophoral trama regular to subregular and consisting of hyaline, thin-walled, cylindrical to ellip¬ soid, inflated elements 29-72 x 5.5-17 pm, clamp connections abundant. Pileipellis a cutis of cylindrical, hyaline, septate, thin-walled hyphae 2.5-8.5 pm diam., clamp con¬ nections abundant. Stipitipellis a cutis of cylindrical, hyaline, septate, thin-walled hyphae 1.5-5 pm diam., clamp connections abundant. Habitat and distribution; solitary or gregarious on soil, in humus or amongst moss in cool temperate rainforest. The species appears to be widespread and reasonably common. Material; Arve Loop Rd nr Geeveston, 27.iv.1998, A.M.Young (hb. Young 1959; BRI); Little Florentine R., 13.v. 1998, A.M.Young (hb. Young 2025; BRI); Little Florentine R., 13.v. 1998, A.M.Young (hb. Young 2032; MEL 2090262 ); Little Florentine R., 13.V.1998, A.K.Mills & G.Gates (hb. Young 2270; MEL 2090267 ); Moonlight Ridge nr Lune River, viii.1989, A.J.Monks s.n. (holo¬ type HO 131196). Remarks; Re-examination of the holotype has provided the additional microcharacter data listed above. Spore measurements taken from different basidiomes in the holotype demonstrate that otherwise identical basidiomes can produce different sized spores. The 14 A.M. Young and A.K. Mills spore sizes obtained from the holotype are a little larger than those of the original descrip¬ tion [(5.5-)6-8 x 3.5-5 pm], but these larger spore dimensions correlate almost perfect¬ ly with the three collections made of this taxon during 1998. The protologue of Hygrocybe erythrocrenata states that colour variations of the sub¬ decurrent to decurrent lamellae occur, and indicates that such variations may be found within a population of basidiomes occurring in a restricted area (and by implication from the same mycelium). Mills and Monks (1993) stated that lamellae colour variations range from “white or yellowish white to pinkish white (7-10A2) or pallid reddish grey (12B2)”. Collections made during 1998 and positively assigned to this taxon have adnate lamellae with at most a decurrent tooth; the lamellae are coloured either dark red (10C7) or viola¬ ceous red (12A3-12A4). These collections also produce the violet-magenta spore print described in the protologue, have the same red pileus and stipe and their microcharacters show only the minor variations that can be normally expected within a species. Hygrocybe erythrocrenata thus appears to be a taxon which exhibits considerable colour variation of the lamellae. The variants with whitish lamellae appear to produce white spore prints while the dark red to violaceous red lamellae variants produce white spore prints that show the violet-magenta tints when the spores are scraped together into a small mass. All basidiomes have the same colours and surfaces for the pileus and stipe and their microcharacters. The vivid red stipe reported in this taxon after drying (Mills & Monks 1993) fades with time. Material inspected immediately after the drying process is complete has brownish pilei with intensely carmine-red stipes, however the carmine-red colouration typically slowly disappears and stipes of dried material that is two or three years old are usually only faintly red tinted or brownish. The holotype shows this tendency to fade: eleven years after collection, only a single stipe has any trace of the carmine-red color and the remainder are yellowish brown or have a slight orange tint. 10. Hygrocybe firma (Berk. & Broome) Singer, Sydowia 11: 355 (1957). Hygrophorus firmus Berk. & Broome, J. Linn. Soc., Bot. 11: 563 (1871). Type : Sri Lanka. Peradeniya, i. 1869, H.K.Thwaites (Thwaites 880 ) (holotype K, n.v.) Illustrations : Young, Bougher & Robinson (2000), p. 43; Young (2000b), cover photo; Fuhrer & Robinson (1992), p. 44 (as Hygrocybe procera ). Habitat and distribution: gregarious to caespitose in moss beds in cool temperate rainforest. The species appears to be common and widespread in central and north-west¬ ern Tasmania. Material : Franklin R., 7.v. 1998, A.M.Young (hb. Young 2004: BRI); Arve Loop Rd nr Geeveston, 17.V.1999, A.M.Young (hb. Young 222T, BRI); Growlingswallet nr Mt Field, 19.v. 1999, A.M.Young (hb. Young 2238: BRI); Franklin R., 7.v. 1998, A.M.Young (hb. Young 2010: MEL 2090261 ); Franklin R., 21.V.1999, A.M.Young (hb. Young 2245: MEL 2090264 ); Tasman Peninsula, 25.V.1999, A.M.Young (hb. Young 2256: MEL 2090265 ); Rutherford Rd nr Geeveston, 18.vi.1998, A.K.Mills (hb. Mills 1580: HO 509005 ); Mt Field, 20.iv.1991, E.Horak (ZT 4942: BRI). Remarks : Hygrocybe firma seems to be common in Tasmanian moss forests and is sometimes mistaken for Hygrocybe miniata, which appears to be relatively rare. The two are readily separated microscopically as H. firma has dimorphic spores, with the macrospores measuring 11-18 pm in length while H. miniata has monomorphic spores measuring 7-9.5(-10) pm in length. Macroscopically, Tasmanian specimens of H. firma seem to produce extremely long stipes and the pilei tend to remain hemispheric and umbilicate rather than expanding to the more or less plane pilei with central depressions that are encountered in some of the tropical variants. Initial observations on lamellae samples suggested both spore and basidia dimor¬ phism. Since a spore print (or similar deposit) may be considered to contain a very high Hygrophoraceae of Tasmania 15 proportion of mature spores, two samples of stipe surface immediately underneath the lamellae were examined for deposited spores and all spores on each sample were meas¬ ured. This procedure aimed to maximise a random selection of fully matured spores. A combined total of 85 macrospores and microspores were measured and the results for¬ warded to Dr David Ratkowsky of the University of Tasmania for statistical analysis (Univariate Procedure - SAS System, SAS Institute 1990). Dr Ratkowsky’s analysis con¬ firmed the bimodal distribution and provided further evidence that the Tasmanian popu¬ lations belong to H. firma. Hygrocybe firma is an exceptionally variable taxon with both yellow and red pilei which may be hemispheric, convex or umbilicate (Corner 1936; Dennis 1970; Pegler 1983, 1986). Corner’s seventeen varieties (Corner 1936) have not been widely used and scepticism with respect to the constancy of the varietal characters (Heim 1967) led to his concept of H. firma as a variable species. Pegler (1983) used H. firma var. firma but later (Pegler 1986) preferred to consider the taxon as variable and used only the epithet of H. firma. His concept is followed here. Western Australian material considered to be H. firma (Young et al. 2000) has red, convex to deeply umbilicate pilei which fade to orange or yellow-orange and this agrees wholly with material from the type locality of Sri Lanka (Pegler 1986). The Tasmanian material has brilliant red pilei and stipes and reddish tint¬ ed lamellae. This also conforms with known variations of the taxon. There remains the cluster of taxa reported from New Zealand (Horak 1990) which includes Hygrocybe firma, H. miniceps (G.Stev.) E.Horak and H. procera (G.Stev.) E.Horak, all of which have yellow to red basidiomes, trichoderms on the pilei and large spores. Currently, the last two taxa can be separated from H. firma because they are not considered to have dimorphic spores or basidia and H. procera is then separated on the basis of its amygdaliform spores compared to the ellipsoidal spores of H. miniceps. The two latter taxa are obviously closely related and may be variants of a single taxon. Neither has yet been confirmed for Australia. 11. Hygrocybe franklinensis A.M.Young & A.K.Mills, sp. nov. Pileus 9-18(-25) mm, scarlatinus, conicus vel papillato-umbonatus deinde lato-coni- cus, glaber, viscidus, striatus, ad marginem concolorum vel flavidum. Lamellae adnatae vel subadnatae, aurantiaco-rosae, ad marginem subflavidae. Stipes 25-45 x 2-5 mm, scarlatinus, siccus, laevis, cylindricus, ad basim flavidum. Sporae (7-)8-9.5(-10.5) x 5-6 (-6.5) pm, ovoideae vel ellipsoideae, hyalinae, aliquot subconstrictae. Basidia 38-52 (-60) x 7-9.5(-10.5) pm, 4-spora, fibulata. Cystidia nulla. Trama hymenophoralis regu- laris vel subregularis, fibulata. Epicutis pilei ixocutem formans. Gregaria in musco sylvestri. Type: Tasmania. Franklin R., 21.V.1999, A.M.Young ( hb. Young 2243 ) (holotype HO 508993 ) Pileus 9-18(-25) mm, brilliant red (10A8), acutely conical and usually with a sub-papil¬ late umbo expanding to broad conical, smooth, viscid, striate when moist, margin some¬ times concolorous but usually yellow tinted, even or a little ragged occasionally slightly crenulate. Lamellae narrowly adnate and then usually with a slight decurrent tooth or ascending-adnate, orange-pink (6A5-8A5), margins are pale yellow (4A6) and even. Stipe 25-45 x 2-5 mm, red (10A8) near the pileus but yellow towards the base (3A4), smooth, dry, hollow, cylindrical. Spores (7-)8-9.5(-10.5) x 5-6 (-6.5) pm, mean 9.0 x 5.6 pm, Q: 1.5-1.8 (-2.0), mean Q: 1.65, ovoid to ellipsoid, hyaline, smooth, occasionally a little constricted. Basidia 38-52 (-60) x 7-9.5(-10.5) pm, mean 48.1 x 8.1 pm, Q: 4.3-6.8(-7.2), mean Q: 5.93, 4- spored (occasionally 2-spored), clamped. Cystidia absent. Hymenophoral trama regular to sub-regular, composed of thin-walled, hyaline, inflated, cylindrical to fusiform or 16 A.M. Young and A.K. Mills moniliform elements 25-95 x 7-25 jam, clamp connections abundant. Pileipellis an ixo- cutis composed of repent, cylindrical, septate, thin-walled hyphae, 2-4.5 pm diam., clamp connections abundant. Stipitipellis a cutis of repent, cylindrical, thin-walled, sep¬ tate hyphae 2-5(-7) pm diam., clamp connections abundant. (Fig. 1) Habitat and distribution: solitary or gregarious amongst moss in cool temperate rain¬ forest. The species appears to be widespread in southern and north-western Tasmania. Material : Franklin R., 7.V.1998, A.M. Young ( hb. Young 2007\ BRI); Arve Loop Rd Harz Mtn, 17. v. 1999, A.M.Young (hb. Young 2228 ; BRI); Arve Loop Rd nr Geeveston, 17. v. 1999, A.M.Young (hb. Young 2223 ; MEL 2090263 ); Franklin R., 21.V.1999, A.M.Young (hb. Young 2243 ; HO 508993, holotype). Remarks: Hygrocybe franklinensis is similar to H. xanthopoda A.M.Young which also has a scarlet, conical, viscid pileus but differs by having pure yellow, adnexed (occasion¬ ally ascending-adnate to very narrowly adnate) lamellae and a yellow to orange-yellow, inflated stipe. Both H. franklinensis and H. xanthopoda have an ixocutis on the pileus and a dry stipe which separates them from the New Zealand species H. cavipes E.Horak which has an ixotrichoderm on the pileus and a viscid stipe. Etymology: After the type locality of Franklin River, Tasmania. Figure. 1. Hygrocybe franklinensis (holotype). A habit; B basidia; C spores. Habit and T/S sketch, bar = 10 mm; microcharacters, bar = 10 pm. Hygrophoraceae of Tasmania 17 12. Hygrocybe graminicolor (E.Horak) T.W.May & A.E.Wood, Mycotaxon 54: 147-150 (1995). Gliophorus graminicolor E.Horak, Beih. Nova Hedwigia 43: 176 (1973). Type: New Zealand. Ngahere, 21.iii.1968, E. Horak s.n. (holotype PDD 27096). Hygrocybe batesii A.M.Young, Austral. Syst. Bot. 10: 956 (1997). Type: Australia. New South Wales, Monga State Forest, 16.V.1984, A.E.Wood & N.B.Gartrell s.n. (holo¬ type UNSW 84/522). Gliophorus pallidus E.Horak, Beih. Nova Hedwigia 43: 164 (1973). Type: New Zealand. Auckland, 27.vi.1968, E. Horak s.n. (holotype PDD 27090). Illustrations: Fuhrer & Robinson (1992), p. 40; Young & Wood (1997), p. 975 and p. 958 as H. batesii. Habitat and distribution: gregarious to caespitose in humus, litter and moss in cool temperate rainforests. The species appears to be very common and widespread in central and north-western Tasmania and often occurs in large troops. Material: Arve Loop Rd nr Smithton, 27.iv. 1998, A.M.Young ( hb. Young 1955 ; BRI); Tahune, 27.iv.1998, AM.Young (hb. Young 1957 ; BRI); Tahune, 27.iv.1998, A.K.Mills (hb. Young I960: BRI); Tahune, 27.iv.1998, A.M.Young (hb. Young 1962: BRI); Mt Field, 30.iv.1998, AM.Young (hb. Young 1973: BRI); Franklin R., 7.V.1998, A.M.Young (hb. Young 2005: BRI); Milkshake Reserve nr Smithton, 6.V.1998, A.M.Young (hb. Young 1995: MEL 2088594): Lake Chisholm, 6.V.1998, AM.Young (hb. Young 2000: MEL 2088595): Fra nklin R., 7.v. 1998, AM.Young (hb. Young 2008: MEL 2088596): Franklin R., 7.V.1998, A.M.Young (hb. Young 2009: MEL 2088597): Arve Loop Rd nr Geeveston, 11 .v. 1998, A.M. Young (hb. Young 2015: MEL 2088598): Tahune, 17.v. 1999, A.M. Young & A.K.Mills (hb. Young 2218: MEL 2088610): Five Road, 24.iv. 1997, A.K.Mills (hb. Mills 1458: HO 508587): Five Road, 24.iv.1997, A.KMills (hb. Mills 1459: HO 508588): Five Road, 24.iv.1997, A.K.Mills (hb. Mills 1461: HO 508590): Johns Rd nr Geeveston, 24.ix.1997, A.K.Mills (hb. Mills 1475: HO 508994): Arve Loop Rd nr Geeveston, 17.V.1998, A.K.Mills (hb. Mills 1511: HO 508593): Arve Loop Rd nr Geeveston, 17.V.1998, A.K.Mills (hb. Mills 1515: HO 508597): Arve Loop Rd nr Geeveston, 17.V.1998, A.KMills (hb. Mills 1516: HO 508598): Franklin R., 7.V.1998, A.KMills (hb. Mills 1530: HO 508600): Arve Loop Rd nr Geeveston, ll.v.1998, A.K.Mills (hb. Mills 1536: HO 508601): Sandspit Forest Reserve nr Wielangta, 3.vi.l997, A.KMills (hb. Mills 1551: HO 508606): Sandspit Forest Reserve nr Wielangta, 9.vi.l997, A.K.Mills (hb. Mills 1562: HO 509002): Jacksons Bend, 14.X.1998, G.Gates (hb. Mills 1605: HO 508611): Jacksons Bend, 14.X.1998, G.Gates (hb. Mills 1606: HO 508612): Sandspit Forest Reserve, 23.iv.1991, E.Horak (ZT 4956: BRI); Collinsvale, 3.iv.l991, E.Horak (ZT 4809: BRI); Milks hake Reserve, 9.iv.l991, E.Horak (ZT 4886: BRI). Remarks: The green to brown colour variations of this taxon have already been fully described in Young (1999). Hygrocybe graminicolor is often extremely spectacular in the size of its troops amongst moss and 50 or more basidiomes may appear. Occasional vari¬ ants appear (hb. Mills 1562) in which the cheilocystidia are greatly reduced in number. These variants correspond to the normal characteristics in every other way (including spore size 6.5-7.7 x 4-4.5 pm) but cheilocystidia are only occasionally seen and the lamellae margins become fertile. Without the cheilocystidia present, the gluten thread tends to disappear as the basidiomes age. 13. Hygrocybe irrigata (Pers.: Fr.) Bon, Doc. mycol. 6: 41 (1976). Agaricus irrigatus Pers., Syn. meth. Fung.: 361 (1801). Agaricus irrigatus Pers.: Fr., Syst. Mycol. 1: 101 (1821). Hygrophorus irrigatus (Pers.: Fr.) Fr., Epicr. Syst. Mycol.: 329 (1838). Type: none designated. Illustrations: Boertmann (1995), p. 89; Young, Kearney & Kearney (2001, in edit.) Habitat and distribution: gregarious amongst moss in cool temperate rainforest. The species is known from central and south-eastern Tasmania. Material: Little Florentine River, 13.v. 1998, AM. Young (hb. Young 2030: MEL 2088605): Sandspit Forest Reserve nr Wielangta, l.xii.1998, G.Gates & D.Ratkowsky (hb. Mills 1612: HO 509012). 18 A.M. Young and A.K. Mills Remarks: In the field, this material resembles an ‘all-grey to grey-brown’ and extremely glutinous version of Hygrocybe graminicolor, however there is never a tint of either yellow or green at any stage on any part of the basidiome and the pilei are more or less conical rather than the convex and usually umbilicate pileus found in H. graminicol¬ or. Dried material is greyish to brown and not the brick-pinlc associated with dried mate¬ rial of H. graminicolor. The gluten often has the appearance of a thickly applied layer. The field notes with hb. Young 2030 recorded that the fresh lamellae margins had a gluti¬ nous thread, however, microscopic examination showed no traces of the gluten, no cheilocystidia and fertile lamellae margins. It is assumed that the observed thread was accidental and this assumption is supported by the field notes with collection hb. Mills 1612 which state that no glutinous thread was present in fresh material. The material corresponds almost perfectly with H. irrigata (Pers.: Fr.) Bon sensu Boertmann (1995). The two Tasmanian collections show minor variations of colour and pileus shape (varying from conic to convex) but all are within the range of characters exhibited by European collections. Boertmann (pers. comm.) examined material from hb. Young 2030 and agreed that apart from slightly larger spores (7.5-9.5 x 5-7 pm, mean 8.7 x 6.0 pm, Q: 1.2-1.6, mean Q: 1.5; European material (5-)6.5-8(-9.5) x (3.5-)4.5-5.5(-6) pm, Q: 1.2-2.0, mean Q: 1.4—1.6), the Tasmanian material was well within the range of macrocharacters exhibited by European collections. 14. Hygrocybe julietae (G.Stev.) E.Horak, New Zealand J. Bot. 9: 431 (1971). Hygrophorus julietae G.Stev., Kew Bull. 16: 377 (1962). Type: New Zealand. Wellington, l.vii.1949, G.Stevenson (Stevenson 697) (holotype K). Illustrations: Horak (1990), p. 269; Stevenson (1963), plate 6 fig. 9. Pileus 15-18 mm, at first dark greyish yellow (near 4B5) becoming light greyish yellow to clear yellow (near 3A6), convex, dry, margins striate and crenulate or occasionally a little lobed. Lamellae decurrent, yellowish grey (4C5/B5) to light yellow (near 3A6), dis¬ tant, margins even and concolorous. Stipe 50-60 x 1.5-2.5 mm, dry, yellow (near 3A6) tapering downwards, solid then becoming narrowly cavernous. Spore print white. Spores 6-8.5 x 3.5-5 pm, mean 6.9 x 3.9 pm, Q: 1.4-2.2, mean Q: 1.80 , smooth, hyaline, cylindrical to ovoid or narrowly ellipsoid, mostly constricted. Basidia 33-40 x 6.0-8.5 pm, mean 37 x 7.2 pm, Q: 4.3-6.0, mean Q: 5.13, 4-spored, clamped. Cystidia absent. Hymenophoral trama regular or occasionally subregular, composed of more or less parallel chains of hyaline, inflated, thin-walled, clamped elements 22-58 x 5-20 pm. Pileipellis a dry cutis with some hyphal gelatinisation or a weakly formed ixocutis, hyphae hyaline, thin-walled, cylindrical, septate, 2.5-4.2 pm, clamp connections present (clamp connections sometimes only occasional on upper cutis elements but common on lower cutis elements); lactifers present in subcuticular layers as highly refractive, often contorted or convoluted hyphae, 1-3 pm diameter. Stipitipellis a dry cutis with some hyphal gelatinisation or a weakly formed ixocutis with elements that are hyphal, cylin¬ drical, hyaline, thin-walled, clamped, 1-3 pm diameter. (Fig. 2) Habitat and distribution: gregarious in leaf mould in cool temperate rainforest. The species is known only from the single location in southern Tasmania. Material: Arve Loop Road nr Geeveston, ll.v.1998, A.M. Young (hb. Young 2014 ; BRI). Remarks: This taxon with its yellow convex pileus and constricted spores is neither Hygrocybe blanda E.Horak with its distinctively conic or campanulate pilei and ellipsoid, non-constricted spores, nor H. cerinolutea E.Horak, which does have a convex pileus but has larger spores (8-10 x 5.5-7 pm) without constrictions. The material when collected in the field was quite dry, however microscopic examination clearly demonstrated exten¬ sive hyphal gelatinisation on both pileus and stipe together with very firm adherence of Q OQQ .00 00 Figure. 2. Hygrocybe julietae ( hb. Young 2014; BRI). A habit; B basidia; C spores. Habit and T/S sketch, bar = 10 mm; microcharacters, bar = 10 pm. spores to the gelatinised hyphae. Both the protologue and the description of Horak (1990) indicate that H. julietae may be either viscid or dry in the field (almost certainly depend¬ ent upon local weather conditions); this Tasmanian collection conforms to this character¬ istic quite well. 15. Hygrocybe leucogloea A.M.Young, Austral. Syst. Bot. 10: 983 (1997). Type : New South Wales. Mt Wilson, 29.iv.1989, A.E.Wood s.n. (holotype UNSW 89/87). Illustration : Young & Wood (1997), p. 984. Pileus 35 mm, white, more or less convex, viscid. Lamellae very narrowly adnate, white, margins even and concolorous. Stipe 50 x 8 mm, white, dry, smooth, pith filled. Spores 6.0-7.7 x 3.7-5.0 pm, mean 6.8 x 4.3 pm, Q: 1.4-1.8 (-1.9), mean Q: 1.57, ellipsoid to narrow ellipsoid, hyaline, smooth. Basidia 42-50 x 6-8 pm, mean 45.3 x 6.8 pm, Q: 5.9-7.5, mean Q: 6.69, 4-spored, clamped. Cystidia absent. Hymenophoral trama 20 A.M. Young and A.K. Mills regular and composed of cylindrical to inflated ellipsoid elements which are hyaline, thin-walled, septate and 20-78 x 3-11 pm, clamp connections abundant. Pileipellis a very well developed ixotrichoderm up to 170 pm deep, composed of very loosely interwoven and aerial hyphae embedded in gluten; the ixotrichodermal hyphae are thin-walled, cylin¬ drical, hyaline, 2-3.5 pm diam., clamp connections abundant and usually of medallion form. Stipitipellis a cutis (or possibly a very weak ixocutis as some gelatinisation seems to be present) composed of repent, cylindrical, hyaline, thin-walled hyphae 1.5-6 pm diam., clamp connections abundant. Habitat and distribution : solitary in moss in cool temperate rainforest. The species is known only from the single locality. Material. Tasman Peninsula, 25.V.1999, A.K.Mills ( hb. Young 2251; MEL 2088614 ). Remarks : The single basidiome in this collection was rather old but there is no doubt as to its identity. Hygrocybe leucogloea is the only known Australian taxon within the Hygrocybeae which has pure white basidiomes and an ixotrichoderm on the pileus; these characters are shared by the Tasmanian material. The American species, Hygrophorus purus Peck, is similar macroscopically but differs in that it belongs to sub-genus Humidicutis and does not have clamp connections anywhere in the basidiome except at the bases of the basidia. The holotype of Hygrocybe leucogloea has spores that measure (6.3-)6.5-7.9(-8.5) x 4.0-5.6 pm, mean 7.2 x 4.8 pm, Q: 1.2-1.7, mean Q: 1.5 while those of the Tasmanian material measure 6-7.7 x 3.7-5.0 pm, mean 6.8 x 4.3 pm, Q: 1.4-1.8 (-1.9), mean Q: 1.57. The differences in dimensions are not considered important at this stage. A Victorian collection (MEL 261035 ) contained spores measuring 5.3-7.7 x 4.0-5.3 pm, mean 6.8 x 4.5 pm, Q: 1.3-1.8, mean Q: 1.5. This co nfir ms the slight variations in spore dimensions that are to be found in this taxon (Young 2000a). The Tasmanian material dis¬ played very narrowly adnate lamellae instead of the adnexed lamellae present in the holo¬ type, however this is presently believed to be simply a variation likely to be encountered in future collections of this taxon. If all future Tasmanian collections of this taxon show similar lamellae attachments, a separate Tasmanian variety could be considered. 16. Hygrocybe lilaceolamellata (G.Stev.) E.Horak, New Zealand J. Bot. 9: 434 (1971). Hygrophorus lilaceolamellatus G.Stev., Kew Bull. 16: 378 (1963). Type : New Zealand. Wellington, 2.vi.l949, G.Stevenson (Stevenson 619) (holotype K). Illustrations; Horak (1990), p. 269; Young & Wood (1997), p. 985; Fuhrer & Robinson (1992), p. 42. Habitat and distribution; gregarious on soil or amongst leaf mould or moss in cool temperate rainforest. The species is widespread and common in southern, central and north-western Tasmania. Material; Little Florentine R., 13.V.1998, A.M.Young (hb. Young 2029; BRI); Growlingswallet nr Mt Field, 19.V.1999, A.M.Young (hb. Young 2226; BRI); Julius R., 6.V.1998, A.M.Young (hb. Young 1989; MEL 2088591 ); Arve Loop Rd nr Geeveston, 1 l.v.1998, A.M.Young (hb. Young 2019; MEL 2088601); Little Florentine R., 26.v. 1999, A.K.Mills & A.M.Young (hb. Young 2262; MEL 2088615); Roger River Reserve nr Smithton, 6.v. 1998, A.K.Mills (hb. Mills 1523; HO 508996); Rutherfords Rd nr Geeveston, 18.vi.1998, A.K.Mills (hb. Mills 1578; HO 509003). Remarks; This species is very distinctive with its warm brown pilei and contrasting lilac lamellae. The taxon is very variable with respect to spore size and shape. Mean Q has been found to vary from 1.65-1.97 indicating a variation from predominantly long- ellipsoid spores in the basidiomes of some collections to predominantly cylindrical spores in others. Spore constrictions may be frequent or only occasionally displayed in various collections. Sectioned lamellae from some collections have displayed a weakly Hygrophoraceae of Tasmania 21 divergent trama and further study may result in the transfer of this species to the genus Hygrophorus. 17. Hygrocybe miniata (Fr.: Fr.) P.Kumm., Fiihr. Pilzk.: 112 (1871); Agaricus miniatus Fr.: Fr., Syst. Mycol. 1: 105 (1821); Hygrophorus miniatus (Fr.: Fr.) Fr., Epicr. Syst. Mycol. : 330 (1838). Type'. Sweden. Smoland, 21.ix.1980, M. Moser 80/372 (neotype IB, n.v., designated by Arnolds 1986, p. 148). Illustrations: Horak (1990), Plate 4, fig. 2; Young & Wood (1997), p. 989. Habitat and distribution : gregarious on humus amongst moss in cool temperate rain¬ forest. The species is known from only a single collection in south-eastern Tasmania. Material: Arve Loop Rd nr Geeveston, 27.iv.1998, A.M.Young (hb Young 1964: BRI). Remarks: The single known collection suggests that Hygrocybe miniata is possibly rare in Tasmania but this should be regarded as very tentative. There may be several rea¬ sons for the single collection during the 1998-1999 Tasmanian trips including: the species may fruit abundantly at a different time of year; the most important areas where the fungus occurs have not yet been located; and the years 1998 and 1999 were not suit¬ able years for optimum occurrence of the species. Much more collecting is required over a number of years to assess the species’ true abundance and distribution. The basidiomes of this collection strongly resemble those of H. firma, but very careful analysis of the spores (including a spore print) has confirmed that they are not dimorphic and that they measure 7.5-9.5 x 4.5-6.5 pm, mean 8.4 x 5.4 pm, Q: 1.3-1.9, mean Q: 1.56. There is no doubt that this collection represents a variant of H. miniata. 18. Hygrocybepseudograminicolor A.M.Young, Austral. Syst. Bot. 10: 992 (1997). Type: Australia. New South Wales. Mt Wilson, 26.iii.1994, F.Taeker s.n. (holotype UNSW 94/22). Illustration: Young & Wood (1997), p. 994. Habitat and distribution: gregarious to caespitose amongst leaf litter or in moss of cool temperate rainforests. The species is widespread and common in southern central and north-western Tasmania. Material: Julius R., 6.V.1998, A.M.Young (hb. Young 1988: BRI); Arve Loop Rd nr Geeveston, 1 l.v.1998 ,(hb. Young 2018: MEL 2088600 ); Little Florentine R., 13.V.1998, A.M.Young (hb. Young 2035: MEL 2088606 ); Roger River Reserve nr Smithton, 6.v. 1998, A.K.Mills (hb. Mills 1518: HO 508995 ); Arve Loop Rd nr Geeveston, 11 .v. 1998, A.K.Mills (hb. Mills 1543: HO 508998 ); Sandspit Forest Reserve nr Wielangta, 3.vi.l998, A.KMills (hb. Mills 1549: HO 508999 ); Sandspit Forest Reserve nr Wielangta, 3.vi.l997, A.K.Mills (hb. Mills 1561: HO 509001 ); Sandspit Forest Reserve nr Wielangta, 14.vii.1998, A.K.Mills (hb. Mills 1587: HO 509010 ); Franklin R., 13.iv.1991, E.Horak (ZT 4316: BRI). Remarks: Although Hygrocybe pseudograminicolor does resemble H. graminicolor and both have cheilocystidia, the bright lime green lamellae of the former species serve to distinguish it in the field from the latter taxon which has white or at most green tinted lamellae. The spores of H. pseudograminicolor (8.5-10.5 x 5.0-7.7 pm) are larger than those of H. graminicolor (5.3-7.3 x 3.3-5 pm). The green pigments of H. pseudo¬ graminicolor also differ from H. graminicolor in that the former species becomes dull green when dried for herbarium storage, but the latter becomes brick-pink. The species is comparatively rare at the type locality but is quite common in Tasmania. A peculiarity of this taxon is that the lamellar trama alters with maturation of the basidiome. During the early stages of development, the trama is regular but in late matu¬ rity it becomes more or less irregular. 22 A.M. Young and A.K. Mills Figure. 3. Hygrocybe roseoflavida (holotype). A habit; B basidia; C spores. Habit and T/S sketch, bar = 10 mm; microcharacters, bar = 10 pm. 19. Hygrocybe roseoflavida A.M.Young & A.K.Mills, sp. nov. Pileus 9-20 mm, pallido-roseus, convexus, umbilicatus, viscidus, plicato-stria- tus, ad marginem aequalum vel subcrenulatum. Lamellae decurrentes, pallido-roseae, ad marginem concolores. Stipes 13-20(-40) x 1-2 pm, super pallido-roseo-brunneus, flavidus, viscidus, laevis, cylindricus, cavus, ad basim atroflavidum. Sporae (5-)6-7.5 x 3.5-4.5 pm, ovoideae vel ellipsoideae vel lacrymoideae, hyalinae, nunquam constrictae. Basidia (17-)23-28(-32) x 5.5-7 pm, (2-)4-spora, fibulata. Cystidia nulla. Trama hymenophoralis subregularis vel regularis, fibulata. Epicutis pilei ixocutem formans. Gregaria vel caespitosa in humo sylvestri. Type: Tasmania. Sandspit Reserve nr Wielangta, 3.vi.l998, A.K.Mills {hb. Mills 1559) (holotype HO 509000) Pileus 9-20 mm, pallid pink (6A2) but often with brownish tints, convex becoming depressed at the centre or umbilicate and finally more or less plane but remaining cen¬ trally depressed, plicate-striate but otherwise smooth, viscid to glutinous, margin even to slightly crenulate. Lamellae decurrent, pallid pink (13-14A2 or similar to the pileus with brownish tints), distant, without a glutinous thread to the margins, margins even and con- colorous. Stipe 13-20(-40) x 1-2 mm, pallid pi nki sh brown (6A2-9A2 or similar to pileus) near the lamellae then becoming pale yellow (2A3) and darker towards the base, viscid to glutinous, cylindrical but sometimes a little inflated at the base, hollow. Spores (5-)6-7.5 x 3.5-4.5 pm, mean 7.0 x 3.9 pm, Q: 1.4-2.0(-2.2), mean Q: 1.79, ovoid to ellipsoid or lacrymoid, hyaline, smooth, non-constricted. Basidia (17—)23—28(—32) x 5.5-7 pm, mean 24.4 x6.4 pm, Q: (2.5-)3.5-4.4(-4.7), mean Q: 3.82, (2-)4-spored, clamped. Cystidia none. Hymenophoral trama regular to sub-regular, com¬ posed of hyaline, inflated, thin-walled elements (17-)30-60(-90) x 7-30 pm, clamp con¬ nections abundant to occasional. Pileipellis an ixocutis of gelatinised, hyaline, thin walled repent hyphae 1.5-7 pm, clamp connections abundant. Stipitipellis an ixocutis of Hygrophoraceae of Tasmania 23 hyaline, repent, septate, t hin -walled hyphae 2.5-6.0 pm, clamp connections present on smaller hyphae (< 4 pm diameter) but absent from larger hyphae. (Fig. 3) Habitat and distribution : gregarious to caespitose on humus rich soil amongst moss and tree fern debris in rainforest gully, often on the bases of old tree ferns in cool temperate rain¬ forest. This species is so far only known from locations in south-eastern Tasmania. Material: Johns Rd nr Geeveston, 19.v. 1998, A.M.Young (hb. Young 2045 ; BRI); Mt Wellington, 28.iv.98, A.M.Young (hb. Young 1969 ; MEL 2090260 ); Tasman Peninsula, 25.V.1999, A.K.Mills (hb. Young 2258; MEL 2090266 ); Sandspit Forest Reserve nr Wielangta, 3.vi.l998, A.K.Mills (hb. Mills 1559; holotype HO 509000 ); Sandspit Forest Reserve nr Wielangta, 14.vii. 1998, A.K.Mills (hb. Mills 1585; HO 509009). Remarks: This often tiny, glutinous taxon is rather interesting for its unique coloura¬ tion of pallid pink or pinkish brown pileus and lamellae but with contrasting yellow stipe. The basidia are also unusual in that they are amongst the smallest in the Hygrophoraceae. No other taxon has these colour characteristics coupled with the very small basidia. It is quite difficult to obtain good representative collections of this species as only a very few basidiomes are normally found at each location. 20. Hygrocybe stevensoniae T.W.May & A.E.Wood, Mycotaxon 54: 148 (1995). Hygrophorus viridis G.Stev., Kew Bull. 16: 383 (1963); Gliophorus viridis (G.Stev.) E.Horak, Beih. Nova Hedwigia 43: 173 (1973); non Hygrocybe viridis Capelari & Maziero, Mycotaxon 33: 192 (1988). Type: New Zealand. Levin, 26.vi.1948, G. Stevenson 338 (holotype K). Misapplied: Hygrophorus psittacinus sensu Cleland and Cheel (1919) and Willis (1963); Hygrocybe psittacina sensu Shepherd and Totterdell (1988). Illustrations: Fuhrer & Robinson (1992), p. 41; Young & Wood (1997), p. 998. Habitat and distribution: solitary to gregarious in cool temperate rainforest amongst moss or humus. The species is widely distributed but does not appear to be as common as the very similar glutinous taxon, Hygrocybe graminicolor. Material: Arve Loop Rd nr Geeveston, ll.v.1998, A.K.Mills (hb. Mills 1540; HO 508997 ); Sandspit Forest Reserve nr Wielangta, 14.vii. 1998, A.K.Mills (hb. Mills 1582; HO 509006); Sandspit Forest Reserve nr Wielangta, 14.vii.1998, A.K.Mills (hb. Mills 1583; HO 509007); Jacksons Bend, 14.x. 1998, A.K.Mills (hb. Mills 1601; HO 509011). Remarks: This taxon is presently accepted as occurring in Tasmania, but material from both Australia and New Zealand requires futher study. The four collections cited here exhibit macrocharacters consistent with the holotype as follows: all display glutinous basidiomes with a well developed ixotrichoderm on the pileus, the lamellae have fertile margins (there is no glutinous thread with embedded hyphal cheilocystidia) and the lamellae are adnate or at most have a decurrent tooth. The spores in all four collections are ellipsoidal and measure 7-9(-10) x 4-5.5(-6) pm, means 7.3-8.0 x 4.4^4.9 pm, Q: 1.4-1.8, mean Q: 1.62 -1.66. This also compares quite well with the holotype which has the same macrocharacters and in which the spores measure 6.5-9(-10) x 4-6.5 pm, mean 8.0 x 4.9 pm, Q: 1.4-1.8 (-2.1), mean Q: 1.63. The basidial lengths of Tasmanian material [ 42-64(-70) x 5-8(-9) pm, mean 47.6 -57.5 x 6.8 -7.8 pm, Q: 5.7-9.6, mean Q: 6.98 -7.66] differ consistently from the holo¬ type collection which contains basidia measuring 35-45 x 6-7.5 pm, mean 38.9 x 6.1 pm, Q: 4.7-7.0(-9), mean Q: 6.35. The larger basidia found in Tasmanian collections are also known from several collections accepted as Hygrocybe stevensoniae from the Blue Mountains, NSW which had basidia measuring 42-56 x 5-8 pm, mean 46.8-51.8 x 6.8-7.5 pm, Q: 5.4-10, mean Q: 6.21-7.12. These collections had basidiomes which exhibited the usual macrocharacters of the species and which had spores agreeing with 24 A.M. Young and A.K. Mills the dimension range exhibited by the spores of the holotype. The longer basidia of the Tasmanian material are therefore not considered to be important. Two Tasmanian collections (hb. Mills 1582, 1583 ) showed a colour variations in which pinkish tones predominated. Whether this was due to rain wash, bacterial infection or genetic variation is unknown, however all four collections cited here dried to the usual brick-pink colour exhibited by herbarium material of this taxon. Subgenus 4 Humidicutis Singer, Sydowia 2: 28 (1948). Typical species: Hygrophorus marginatus Peck. Basidiome variously coloured white, pink, dull orange, yellow, or lilac; pileus usually conical becoming umbonate or plane and frequently splitting radially; lamellae narrowly adnate, adnexed or more or less free; cystidia absent; hymenophoral trama regular, com¬ posed of short, cylindrical to inflated (often moniliform) elements 20-300 pm long; clamp connections absent throughout the basidiome except at the bases of the basidia and then frequently of medallion form. 21. Hygrocybe lewelliniae (Kalchbr.) Brittleb. ex A.M.'Young in Young & Wood, Austral. Syst. Bot. 10: 1011 (1997); Hygrophorus lewelliniae Kalchbr., Proc. Linn. Soc. New South Wales 7: 105 (1882). Type: Victoria. Western Port, 14.vi.1880, M.M.R.Lewellin (holotype Rare Book Mss All, MEL). Uncertain recombination: Brittlebank, Cat. Austral. Fungi 181 (1940). Illustrations: Willis (1957); Cole, Fuhrer & Holland (1978), plate 3. Habitat and distribution: often solitary but occasionally gregarious, usually amongst moss but occasionally in humus in cool temperate rainforest. Hygrocybe lewelliniae is common and widespread in southern, central and north-western Tasmania. Material: Arve Loop Rd nr Geeveston, 27.iv.1998, A.M.Young {hb. Young 1965: BRI); Mt Field, 30.iv.1998, A.M.Young (hb. Young 1976: BRI); Lake Chisholm, 6.V.1998, A.M.Young (hb. Young 1999: BRI; MICH 28502): Franklin R., 7.v. 1998, A.M.Young (hb. Young 2003: BRI); Little Florentine R., 13.V.1998, A.M.Young (hb. Young 2021: BRI); Fran kli n R., 21.V.1999, A.M.Young (hb. Young 2246: BRI); Arve Loop Rd nr Geeveston, 11 .v. 1998, A.M. Young (hb. Young 2017: MEL 2088599): Little Florentine R., 13.V.1998, A.M.Young (hb. Young 2022: MEL 2088602): Johns Rd nr Geeveston, 19.V.1998, A.M.Young (hb. Young 2047: MEL 2088607): Tahune, 17.v. 1999, A.M.Young (hb. Young 2212: MEL 2088609): Franklin R., 21.V.1999, A.M.Young (hb. Young 2242: MEL 2088612): Rutherfords Rd nr Geeveston, 18.vi. 1998, A.K.Mills (hb. Mills 1579: HO 509004). Remarks: Colour variations of Hygrocybe lewelliniae include very pale lilac to near¬ ly deep violet but all collections have shown the same microscopic characters as those of the holotype. Macroscopically, all collections have displayed the conical to umbonate, dry pilei with radial splitting that creates a medial split along the trama in the lamella directly beneath the split in the pileus so that each lamella half remains attached both to the pileus and along the lamella margin. The recombination by Brittlebank (1940) remains doubtful due to uncertainties with respect to effective means of publication and so the paper by Young and Wood (1997) is used to validate Brittlebank’s recombination. 22. Hygrocybe mavis (G.Stev.) E.Horak, New Zealand J. Bot. 9: 434 (1971); Hygrophorus mavis G.Stev., Kew Bull. 16: 377 (1962). Type: New Zealand. Levin, 18.vi.1949, G. Stevenson , Stevenson 654 (holotype K). Misappl.: Hygrophorus purus Peck, sensu E.Horak, New Zealand J. Bot. 28: 294 (1990). Illustration: Young & Wood (1997), p. 1016. Hygrophoraceae of Tasmania 25 Habitat and distribution: generally solitary but occasionally gregarious and occurring in moss or on soil or amongst humus in cool temperate rainforest. This species is frequent and widespread in southern, central and north-western Tasmania. Material: Mt Field, 30.iv.1998, A.M.Young ( hb. Young 1971; BRI); Lake Chisholm, 6.V.1998, A.M.Young {hb. Young 1998; BRI); Johns Rd nr Geeveston, 19.v. 1998, A.M.Young {hb. Young 2044; BRI); Little Florentine R., 26.v. 1999, A.M. Young {hb. Young 2265; BRI); Milkshake Res., 6.V.1998, A.M.Young {hb. Young 1994; MEL 2088593 ); Little Florentine R., 13.V.1998 , A.M.Young {hb. Young 2028; MEL 2088604 ); Tasman Peninsula, A.K.Mills & A.M.Young {hb. Young 2249; MEL 2088613 ); Sandspit Forest Res. nr Wielangta, 14.vii.1998, A.K.Mills {hb. Mills 1584; HO 509008); Julius R nr Smithton, 9.iv.l991, E.Horak (ZT 4854; BRI). Remarks: Hygrocybe mavis shares with H. lewelliniae the remarkable characteristic in which there is combined splitting of both the pileus and lamellae. Whether H. mavis is the white variety of H. lewelliniae remains to be resolved. The characters of Tasmanian material conform well with those of the holotype. Genus 2. Camarophyllopsis Herink, Shorn. Severoesk. Mus., Plir. Vdy 1: 61 (1958). Typical species: Camarophyllopsis schulzeri (Bres.) Herink. Basidiome thin to fleshy, small, dull coloured in grey to ochre or brown; pileus convex to umbilicate, dry and often hygrophanous; lamellae distant, broadly adnate to arcuate or decurrent; universal veil absent; stipe dry, often with small dots or pruinose punctate; spore print white. Spores hyaline, smooth, non-amyloid, subglobose to broadly ellipsoid, small (up to 7 pm long); basidia narrowly clavate, 20-70 x 4.5-8.5 pm, Q: 4.5-10.0, mostly 4-spored; cystidia absent or inconspicuous; hymenophoral trama regular to sub¬ regular and composed of short elements up to 170 pm long; pileipellis an hymeniderm; clamp connections present or absent; development monovelangiocarpic and stipiticarpic. Solitary to subgregarious, terrestrial in forests or open sites, apparently saprophytic. Mostly in temperate North America, Asia and Europe, but also known from subtropical South America and Asia. 23. Camarophyllopsis kearneyi A.M.Young, Austrobaileya 5: 562 (1999). Type: New South Wales. Lane Cove Bushland Park, 13.vi.1998, R. & E.Keamey s.n. (holotype DAR 73919). Illustration: Young (1999), p. 563. Habitat and distribution: gregarious to caespitose on soil or deep humus or amongst moss in very sheltered parts of cool temperate rainforest. The two Tasmanian collections come from nearby locations in the south east region. Material: Growlingswallet nr Mt Field, 19.V.1999, A.M.Young (hb. Young 2236; MEL 2089735); Little Florentine R„ 26.V.1999, A.M.Young (hb. Young 2269; MEL 2089736). Remarks: These two collections conform macroscopically extremely well to the type collection from Lane Cove Bushland Park. Both Tasmanian collections were made in very sheltered locations and the material in hb. Young 2269 was collected under old tree ferns in very dense shade. The Growlingswallet collection consisted of basidiomes grow¬ ing gregariously but the second collection had a number of basidiomes additionally dis¬ playing a caespitose habit. Spore sizes from the two collections were 4.5-6 x 4.5-5.5 pm, mean 5.4 x 5.2 pm, Q: 0.9-1.2, mean Q: 1.05, and 4.5-6.5 x 4-5.5 pm, mean 5.2 x 4.7 pm, Q: 1.0-1.3, mean Q: 1.10 respectively. This compares well with the holotype which has spores measuring (4 -)4.3-5.7 x 4.0-5.3(-5.7) pm, mean 4.9 x 4.6 pm, Q: 1.0-1.2(-1.3), mean Q: 1.1. The basidial ranges similarly overlap and variations are not considered important. 26 A.M. Young and A.K. Mills The prescence and structure of cheilocystidia reported from the holotype appears to vary and is now in need of further clarification. At the time of publication, only material collected from the type location was known and possible character ranges were therefore limited to those exhibited by the holotype. The two Tasmanian collections do contain lamellae ele¬ ments that appear to be identical to the cheilocystidia of the holotype, however these ele¬ ments may be partially or wholly hidden within the marginal structure of the lamellae. The Tasmanian collections indicate that this type of variation may be normal for the species and is now being encountered for the first time. In addition, the cheilocystidia often resemble immature basidia/basidioles and the ‘fine line’ between cheilocystidia and immature repro¬ ductive elements is now being considered. Further mature and immature collections of this taxon are required to determine whether the cheilocystidia reported from the holotype are either a variable character or an artifact of the basidiome age/maturity. Hygrocybe miniata (Arnolds 1990; Boertmann 1995) is another taxon in which various collections may or may not exhibit the presence of cheilocystidial elements. TRIBE 2. HYGROPHOREAE P. Henn. in Engl. & Prantl, Nat. Pflanzenfam. 1: 209 (1898). emend. Kiihner in Bull. mens. Soc. linn. Lyon 48: 617 (1979). Type genus: Hygrophorus Fr., Gen. Hymenomyc .: 8 (1836). Hymenophoral trama divergent; forming ectomycorrhizae. Genus 1. Hygrophorus Fr., Gen. Hymenomyc. 8 (1836). Typical species: Hygrophorus eburneus (Bull.: Fr.) Fr. Basidiome tricholomatoid to omphaloid, fleshy to thin, small to large; pileus variously coloured but usually of dull colours, not hygrophanous, mostly viscid to glutinous; lamel¬ lae spaced to distant, broadly adnate to decurrent, thick, waxy; universal veil often pres¬ ent and glutinous; partial veil sometimes present; stipe often glutinous or viscid, fre¬ quently with small dots punctate at the apex; spore print white. Spores hyaline, smooth, non-amyloid, basidia narrowly clavate, 30-90 x 6-15 pm, Q: 4.5-9.0; cystidia absent or inconspicuous; hymenophoral trama divergent from a central line and made of short ele¬ ments up to 200 pm long; pileipellis mostly an ixocutis or an ixotrichoderm, rarely a cutis or trichoderm; clamp connections present; development gymnocarpic to pseudoangio- carpic and stipitocarpic. Solitary to gregarious, terrestrial, always near trees or shrubs and apparently ectomycorrhizal principally with Pinaceae, Betulaceae and Fagaceae. Mostly in temperate zones of the Northern Hemisphere, but some taxa in similar climatic regions of Southern Hemisphere. One species known for Tasmania. 24. Hygrophorus involutus G.Stev., Kew Bull. 16: 373 (1962). Type: New Zealand. Butterfly, 2.vi.l958, G. Stevenson {Stevenson 1347) (holotype K). Pileus convex, white, yellow or apricot-yellow, viscid; trichoderm present; stipe con- colorous with pileus. Key to varieties of Hygrophorus involutus 1. Pileus, lamellae and stipe pure white. 24a. var. albus 1. Pileus, lamellae and stipe shades of creamish yellow to apricot-yellow. .24b. var. involutus Hygrophoraceae of Tasmania 27 24a. Hygrophorus involutus G.Stev. var. albus A.M.Young & A.K.Mills, var. nov. A H. involutus basidiomatibus albis differt. Type : Tasmania. Arve Loop Rd nr Geeveston, A.K.Mills ( hb. Mills: 1541 ) (holotype HO 508602; isotype hb.Mills.) Illustration : Fuhrer & Robinson (1992), p. 39 as Camarophyllus niveus. Basidiomes wholly pure white but otherwise identical to those of var. involutus. Habitat and distribution: gregarious amongst soil, humus or moss in cool temperate rainforest. The variety is reasonably common and often appears amongst normally coloured basidiomes. Material'. Myrtle Forest Track nr Collinsvale, 18.iii.1999, G.Gates (hb. Mills 1628; HO 508614 ). Remarks; The pure white basidiomes of Hygrophorus involutus var. albus differ from the normal light orange to yellowish basidiomes of H. involutus var. involutus only in that they are without any coloured pigments. Only two collections are here listed, however anecdotal evidence indicates that this white variety is widespread in suitable Tasmanian habitats. 24b. Hygrophorus involutus G.Stev. var. involutus Illustrations; Fuhrer & Robinson (1992), p. 45; Young & Wood (1997), p. 1020. Habitat and distribution; gregarious amongst soil, humus or moss in cool temperate rainforest. This species is widespread and common in southern, central and north-west¬ ern Tasmania. Material; Arve Loop Rd nr Geeveston, 27.iv. 1998, A.M.Young (hb. Young 1958; MEL 2087781); Mt Field, 30.iv.1998, A.M.Young (hb. Young 1972; BRI); Little Florentine River, 13.V.1998, A.M.Young (hb. Young 2040; BRI); Growlingswallet nr Mt Field, 19.V.1999, A.M.Young (hb. Young 2233; MEL 2087770 ); Arve Loop Rd nr Geeveston, ll.v.1998, A.KMills (hb. Mills 1542; HO 508603); Arve Loop Rd nr Geeveston, ll.v.1998, A.K.Mills (hb. Mills 1544; HO 508604); Sandspit Forest Reserve nr Wielangta, A.K.Mills (hb. Mills 1588; HO 508607); Myrtle Forest Track nr Collinsvale, G.Gates (hb. Mills 1627; HO 508613); Franklin R., 13.iv.1991, E.Horak (ZT 4317; BRI). Remarks; Hygrophorus involutus remains the sole representative of the genus for Australia. Collection hb. Young 2040 was used to co nfir m previous remarks in Young (2000a) regarding the weakly divergent trama. Inspection has shown that the divergent structure is best observed if the section is taken from the larger lamellae of older pilei and at a point about one third of the distance of the pileus radius commencing at the margin. The divergent hyphae can be seen in the middle part of the lamellae. Several Tasmanian collections exhibit basidiomes which have smaller mean spore sizes and Q’s (5.5-7.5 x 3.5-4.5 pm, mean 6.5 x 4.2 pm, Q: 1.3-1.8, mean Q: 1.55) than those previously reported (Young & Wood 1997) from mainland material (6.0-7.5 (-9) x (3-)3.5-4.5(-5) pm, mean 7.2 x 3.9 pm, Q: 1.4-2.3, mean Q: 1.9). However, because there is so much overlap of dimensions and since all other aspects of the basidiomes are similar, the difference is considered to be normal species variation. Acknowledgements The authors thank Dr David Ratkowsky of the University of Tasmania for assistance with statistical analysis; Dr Gintaras Kantvilas of the Tasmanian Herbarium for assistance with locating holotype material; Mr Rod Henderson of the Queensland Herbarium for assis¬ tance with locating holotype material and advice on format; Dr Tom May and Martine Pauli of the National Herbarium of Victoria for assistance with locating material; and Professor Egon Horak of the Herbarium Turicense for the generous gift to the 28 A.M. Young and A.K. Mills Queensland Herbarium of his relevant Tasmanian collections. This investigation was completed under a grant provided by the Australian Biological Resources Study for research into the species of the Hygrophoraceae of south-eastern Australia. References Arnolds, E. (1990). Tribus Hygrocybeae (Ktihner) Bas & Arnolds, in C. Bas, Th.W. Kuyper, M.E. Noordeloos and E.C. Vellinga (ds), Flora Agaricina Neerlandica 2. A.A.Balkema: Rotterdam. Boertmann, D. (1995). The genus Hygrocybe. Fungi of Northern Europe 1, 1-184. Cleland, J.B. and Cheel, E.C. (1919), Australian fungi: notes and descriptions. No. 3. Transactions and Proceedings Royal Society of South Australia 43, 262-315. Cole, F.M., Fuhrer, B.A. and Holland, A.A. (1978). A Field Guide to the Common Genera of Gilled Fungi in Australia. Inkata Press: Melbourne. Comer, E.J.H. (1936). Hygrophorus with dimorphic basidiospores. Transactions of the British Mycological Society 20, 157-184. Dennis, R.W.G. (1970). Fungus Flora of Venezuela and adjacent countries. Kew Bulletin Add. Series 3, 1-531. Fuhrer, B.A. and Robinson, R. (1992). Rainforest Fungi of Tasmania and South-East Australia. CSIRO: Melbourne. Heim, R. (1967). Hygrophores tropicaux recuielles par Roger Heim 1: Especes de Guyane fran§aise et de Nouvelle-Guinee australienne, Revue de Mycologie 32, 16-27, pi IV. Hesler, L.R. and Smith, A.H. (1963). North American Species of Hygrophorus. University of Tennessee Press: Knoxville. Holmgren, P.K, Holmgren, N.H. and Barnett, L.C. (1990). Index Herbariorum. Part 1 The Herbaria of the World (8 th edn). New York Botanical Garden: New York. Horak, E. (1979). Fungi, Basidiomycetes, Agaricales y Gasteromycetes secotioides. Flora Criptogammie de Tierra del Fuego 11, 1-524. Horak, E. (1990). Monograph of the New Zealand Hygrophoraceae (Agaricales). New Zealand Journal of Botany 28, 255-309. Kornerup, A. and Wanscher, J.H. (1981). Taschenlexikon der Farben. Muster-Schmidt Verlag: Gottingen. Massee, G. (1899). Fungi exotici, II. Bulletin of Miscellaneous Information Kew 1899, 164-185. Mills, A.K. and Monks, A.J. (1993). Unusual Spore Print Colouration within the family Hygrophoraceae. Two distinctive taxa recorded from Tasmania. My cotaxon 46, 85-91. Monks, A.J. (1989). A Preliminary Survey of the Hygrophoraceae of Tasmania. Bsc. Honours Thesis, The University of Tasmania. Monks, A.J. and Mills, A.K. (1991). ‘ Camarophyllus rodwayi (Hygrophoraceae), a rediscovered fungus’, in M.R.Banks et al. (Eds), Aspects of Tasmanian Botany - A Tribute to Winifred Curtis, 13-14. Pegler, D. N. (1983). Agaric Flora of the Lesser Antilles. Kew Bulletin, Addit. Ser. 9, 1-668. Pegler, D.N. (1986). Agaric Flora of Sri Lanka. Kew Bulletin, Addit. Ser. 12, 1-519. SAS Institute (1990). SAS Procedures Guide, Ver. 6. (3 rd edn). Cary: NC, USA. Stevenson, G. (1963). The Agaricales of New Zealand: IV. Kew Bulletin 16, 373-384. Willis, J.H. (1957). Rediscovery of a rare Victorian toadstool ( Hygrophorus lewelliniae Kalchr.). Victorian Naturalist 74, 71-72. Willis, J.H. (1963). Victorian Toadstools and Mushrooms, 3rd edn. The Field Naturalists Club of Victoria: Blackburn. Young, A.M. (1999). The Hygrocybeae (Fungi, Basidiomycota, Agaricales, Hygrophoraceae) of the Lane Cove Bushland Park, New South Wales. Austrobaileya 5, 535-564. Young, A.M. (2000a). Additions to the Hygrophoraceae (Fungi, Agaricales) of south-eastern Australia. Muelleria 13, 3-36. Young, A.M. (2000b). Common Australian Fungi: A Bushwalker’s Guide. (Revised ed.) University of New South Wales Press: Sydney. Young, A.M. (2000c). Additions to the Hygrocybeae of Victoria. I. Muelleria 14, 51-64. Young, A.M., Bougher, N.L. and Robinson, R.M. (2000). Hygrophoraceae of Western Australia II. Further Taxa. Australasian Mycologist 19, 41-48. Young, A.M. and Wood, A.E. (1997). Studies on the Hygrophoraceae (Fungi, Homobasidiomycetes, Agaricales). of Australia. Australian Systematic Botany 10, 911-1030. Muelleria 16: 29-38 (2002) Genetic evidence supports reclassification of Agrostis billardierei var. filifolia and A. aemula R.Br. var. setifolia as a single species, A. punicea (Poaceae) EA. James 1,2,5 , M.C. Ryan 3 , Y.J. Fripp 2 and A.J. Brown 4 1 National Herbarium of Victoria, Royal Botanic Gardens Melbourne, Birdwood Ave, South Yarra, Vic. 3141 2 Genetics Department, La Trobe University, Bundoora, Vic. 3083 3 Botany Department, La Trobe University, Bundoora, Vic. 3083 4 State Chemistry Laboratory, Department of Natural Resources and Environment, 621 Sneydes Rd, Werribee, Vic. 3030 5 Author for correspondence: Elizabeth.James@rbg.vic.gov.au Abstract The genetic relationship between Agrostis billardierei R.Br. var. filifolia Vickery and A. aemula R.Br. var. setifolia Vickery was investigated using random amplification of polymorphic DNA. Three random primers generated 49 products of which only one was present in all samples. Agrostis aemula var. setifolia was found to be more similar to A. billardierei var. filifolia than to A. aemula var. aemula supporting reclassification of A. billardierei var. filifolia and A. aemula var. setifolia as a single species, A. punicea A.J.Brown. & N.G.Walsh. Introduction Many populations of native grass species exist as isolated remnants because the once extensive grasslands on the western basalt plains in Victoria have declined to less than 1% of their previous extent (Scarlett et al. 1992). Interest in the preservation and utilisation of these species in Australia has highlighted some problems in the delineation of some taxa. Before management strategies for maintaining biodiversity can be implemented, there is a need to improve our understanding of the relationships between taxa and the level of diversity within recognised taxa. A number of native species of Agrostis R.Br. are found in lowland Victoria: A. adamsonii Vickery, A. aemula R.Br., A. avenacea J.F.Gmel., A. billardierei R.Br., A. robusta A.J.Brown & N.G.Walsh, A. punicea A.J.Brown & N.G.Walsh, A. rudis Roem. & Schult. and A. venusta Trin. Agrostis punicea is a recently described species based on morphological examination of former varieties of A. aemula and A. billardierei (Brown & Walsh 2000). Doubts had existed about the legitimacy of A. billardierei var. filifolia Vickery and A. aemula var. setifolia Vickery because they are superficially very similar and their separation is based on small morphological differences. When describing the two taxa, Vickery (1941) noted that they showed a strong resemblance with the major dif¬ ference being the hairy lemma of A. aemula var. setifolia. Brown and Walsh (2000) found that, besides the difference in lemma hairiness, A. billardierei var. filifolia and A. aemu¬ la var. setifolia were separated morphometrically, on average, only by the slightly short¬ er inflorescences and slightly longer lemmas, lemma-setae and awns of the latter taxon. Brown and Walsh (2000) also found that A. billardierei var. filifolia and A. aemula var. setifolia had similar ecological preferences. While populations commonly contain either one of these taxa, populations have been observed where both occur, suggesting that they may interbreed. In contrast, the typical varieties of each species are generally distinct in form and growth habit (Walsh 1994). The distribution of the taxa is somewhat different (Brown & Walsh 2000). Agrostis. aemula var. setifolia and A. billardierei var. filifolia is confined to moist, generally open, lowland environments of southern Victoria, south-east South Australia and Tasmania. Agrostis aemula var. aemula is fairly common from coastal to subalpine environments 30 E.A. James et al. across Victoria, New South Wales, South Australia and Tasmania with isolated occur- ances in Western Australia and Queensland. Agrostis billardierei var. billardierei is wide¬ ly distributed around the coastlines of southern Australia with a few inland occurrences. This study was undertaken to see whether there was genetic support for the reclassifi¬ cation by Brown and Walsh (2000) of A. billardierei var. filifolia and A. aemula var. setifo- lia as a single species, A. punicea. The similarity between these taxa is investigated using RAPDs which are dominant markers thought to sample randomly across the genome (Stammers et al. 1995). RAPDs have been found to be useful markers for taxonomic stud¬ ies at the species level or for species complexes where the number of morphological char¬ acters can be insufficient between taxa and other DNA-based methods do not detect suffi¬ cient variation (Gonzalez & Ferrer 1993; Kazan et al. 1993; Stammers et al. 1995). Agrostis avenacea was included for comparison in this study because it is considered a distinct though variable, common and widespread species, found in all regions of Victoria, in all states except the Northern Territory, and in Polynesia (Walsh 1994). Also, it frequently occurs near A. aemula and A. billardierei. It is distinguished from A. aemu¬ la by a number of features used collectively, rather than a sharp discontinuity in features, suggesting differences in the genetic basis of many genes which characterise the species. An earlier study on genetic diversity in native Agrostis species found that A. avenacea was less similar to A. billardierei var. filifolia than to A. billardierei var. robusta Vickery or A. adamsonii (James & Brown 2000). Methods PLANT MATERIAL Wild populations of the study species were sampled from sites in western Victoria (Fig. 1). Seed collections for the populations, ‘HD’: Hadden (Agrostis billardierei var. filifo¬ lia), ‘EM’: East Mortlake (A. aemula var. aemula), ‘WM’: West Mortlake and ‘SHR’: Ballyrogan (A. avenacea ) were made between December and February 1996/97. Agrostis avenacea was represented by both large weeping and small upright forms (WM and SHR respectively). Currently, these forms are not formally recognised. Seed was collected from populations ‘SBL’: South Bulart and ‘LR’: Lake Repose, Glenthompson (A. bil¬ lardierei var .filifolia) and ‘DIG’: Dartmoor (A. aemula var. setifolia) between December and February 1998/99. Seed was obtained from up to 10 individuals in each population, with collections maintained as individual seed-lots. Approximately 20 seeds per individ¬ ual were placed in tapering tubes in a pinebark-based medium, covered in a thin layer of mix with particle size < 2 mm and left to germinate in a glasshouse for up to 10 weeks. A single seedling per individual seed-lot was chosen at random and potted on to provide fresh leaf material for DNA extractions. DNA ISOLATION Genomic DNA was extracted from fresh leaf material using acetylmethylammonium bro¬ mide (CTAB) method modified from Rogers and Bendich (1985). DNA quality was assessed visually after electrophoresis on a 1.0% agarose gel containing ethidium bro¬ mide. To further assess DNA quality, 0.5 pg DNA from a subsample of plants was digest¬ ed separately with restriction enzymes EcoRl and Dral in 25 pi volumes in the appro¬ priate buffers at 37°C overnight (c. 17 h) and OD 260 /OD 280 ratio was measured. DNA AMPLIFICATION Forty synthetic decamer primers (kits A and B) from Operon Technologies, Inc. (Alameda, Calif., U.S.A.) were tested in PCR reactions according to the protocol of Williams et al. (1990). Three primers (OPB-8, OPB-12, OPB-20) giving discrete, repro¬ ducible amplification products were chosen for further analysis. Reactions were per¬ formed in a volume of 25 pi containing 5-10 ng DNA, 1 U Taq polymerase (Gibco-BRL), Genetic evidence supports Agrostis punicea 31 Figure 1 . Distribution of Agrostis study sites in Victoria’s Western District. A. aemula var. aemula: Wooriwyrite (EM); A. aemula var. setifolia: Digby (DIG); A. avenacea: Ballyrogan (SHR), Connewarren (WM); A. billardierei var. filifo- lia: Haddon (HD), Lake Repose (LR), South Bullart (SBL). 1 x Taq polymerase buffer (Gibco-BRL), 160 pM each dATP, dCTP, dGTP and dTTP, 1.5 mM MgCl 2 and 15ng primer, using a Corbett Research FTS 960 thermocycler. For the PCR, initial strand separation and amplification was initiated with one cycle of 94°C for 5 min , 35°C for 2 min, 72°C for 1 min, followed by 40 cycles of 94°C for 30 sec, 38°C for 30 sec (except for primer OPB-08 where annealing temperature was 44°C), 72°C for 30 sec. and a final extension step of 72°C for 3 min . Amplified products were resolved by electrophoresis on a 2% agarose gel containing ethidium bromide and run in TAE buffer at 80-100 v for 2 - 4 h. Images were visualised and photographed using a Sci Tech CCD video camera module and saved to computer using UVIdoc image capture software. DATA ANALYSIS RAPD products were scored as either present (1) or absent (0) for each individual, based on the assessment of a minimum of two independent PCR runs. Fragments migrating at the same size were considered to be homologous. Genstat 5 version 4.1 (PC/Windows NT) (Lawes Agricultural Trust Rothamstead) was used for data analysis. Chi-squared tests were performed to determine which band fre¬ quencies were different between populations. Two similarity matrices were calculated using simple-matching coefficient (SMC) (Gordon 1981). The first compared individuals and was used as the basis for ordination by principle co-ordinate analysis (Gower 1966). The second compared populations. A dendrogram (average-link) was constructed based on the second, reduced similarity matrix, to compare populations. Diversity indices were calculated using the Shannon-information index (Russell et al. 1993) and presented as a single index, averaged over all loci, to provide an average per locus diversity (Chakraborty & Rao 1991). The amount of variation partitioned within and among pop¬ ulations was calculated from the diversity indices (King & Schaal 1989). 32 E.A. James et al. Results Genetic variation was detected by RAPD PCR. RAPDs generated reliable and repro¬ ducible polymorphic patterns that enabled an assessment of the relationship between the former species: A. aemula (including both var. aemula and var. setifolia ), A. billardierei (only var. filifolia represented) and A. avenacea. Scorable RAPD fragments ranging from 0.38 to 2.5 kb in size were amplified by three decamer primers (OPB-08, -12 and -20). They generated a total of 49 products with an average of 16 products per primer. A fourth primer, whilst successfully amplifying DNA for most samples did not reliably amplify DNA from population ‘EM’ and so was elimi¬ nated from the analyses. The three primers revealed polymorphisms among the three species examined and only one amplification product was present in all individuals regardless of species. Polymorphic products per population varied from 0% (SBL) to 81.3% (SHR), and per species varied from 54.5% (A. billardierei) to 96.4% (A. aemula) (Table 1). The high level of polymorphism in A. aemula was due to the small number of RAPD fragments (3) that occurred in both the ‘DIG’ and ‘EM’ populations. Table 1. Number of RAPD bands (% polymorphic), range in no. band differences between RAPD phenotypes* for each population and each species. *(populations only) Species Popn Within populations Within species No. RAPD bands (% polymorphic) Range in no. band differences between RAPD phenotypes No. RAPD bands (% polymorphic) A. billardierei var. filifolia HD 14(14.3) 1-2 var. filifolia SBL 15 (0.0) 0-0 22 (54.5) var. filifolia LR 20 (45.0) 1-9 A. avenacea upright form SHR 17 (82.3) 1-11 24 (87.5) weeping form WM 19 (63.2) 1-6 A. aemula var. aemula EM 18 (66.7) 1-10 28 (96.4) var. setifolia DIG 13 (15.4) 1-2 Similarity between individuals ranged from 41.5-100%. Seventy-nine percent of RAPD PCR products varied significantly (P = 0.01) in frequency between popula¬ tions. The average similarity between species was 56.2% for A. billardierei vs A. ave¬ nacea, 66.7% for A. billardierei vs A. aemula and 70.5% for A. avenacea vs A. aem¬ ula. The number of phenotypes present in populations is listed for each population (Table 2). Similarity between populations ranged from 50.0-94.3% (Table 3). Of note, is the low similarity (62.0%) between the populations DIG and EM formerly considered to be dif¬ ferent varieties of A. aemula. Likewise, similarity between EM (A. aemula var. aemula) and A. billardierei wax. filifolia is only 51.8%. In contrast, the similarity between DIG (A. aemula var. setifolia) and A. billardierei var. filifolia is 81.7%. The relationship between populations is depicted in Fig 2. The population of A. aem¬ ula var. setifolia (DIG) clearly clusters with the populations of A. billardierei var. filifo¬ lia (HD, SBL and LR). Importantly, DIG shows a greater affinity with HD (84.0%) rather than EM (62.0%) despite HD and DIG being separated by almost twice the geographic distance separating DIG and EM (Fig. 1). Genetic evidence supports Agrostis punicea 33 Table 2. Sample numbers, population size and genetic diversity indices for each population. Variety Population (popn size) No. samples No RAPD phenotypes Ho A. billardierei var. filifolia HD (100-250) 9 4 0.0097 var. filifolia SBL (500-1000) 7 1 0.0000 var. filifolia LR (>2500) 7 6 0.0487 A. avenacea upright form SHR (100-250) 10 9 0.0808 weeping form WM (50-100) 10 9 0.0539 A. aemula var. aemula EM (100-250) 10 10 0.0648 var. setifolia DIG (>2500) 10 4 0.0068 Table 3. Reduced similarity matrix based on RAPD data comparing all populations. * = no. plants sampled. Species A. billardierei A. avenacea A. aemula Popn HD 9 * SBL 7 LR 7 SHR 10 WM 10 EM 10 DIG 10 HD _ SBL 85.6 - LR 84.8 94.3 - SHR 59.5 56.2 56.3 - WM 58.0 54.8 52.4 85.7 - EM 52.6 52.7 50.0 78.0 78.2 - DIG 84.0 81.2 79.8 62.5 63.1 62 A.b.filtfolhi jpjpj _ A.b.fi!ifoUa^QY^ A.b.fMfolia-^j^ A.a. setifoliajyjQ "SHR - A.avenaceac A.avenacea^^ _ AM.aemuIa^y^ _ ~T~ 95 90 "1 85 80 75 70 65 \ 60 % Similarity Figure 2. Cluster of populations based on reduced similarity matrix (simple matching coefficient). 34 E.A. James et al. b PCO 3 vs 2 0.3 - 0.2 - 0.1 - % 0 ' < - 0.1 - - 0.2 - -0.3 ' -0.4 — -0.4 + + + - 0.2 —i-1— 0 0.2 Axis 2 0.4 Figure 3. Principle co-ordinate analysis of RAPD phenotypes for all Agrostis individ¬ uals. Axes 1, 2 and 3 account for 73.8 % of variation, a. PCO vector 2 vs 1; b. PCO vector 3 vs 2. OHD, DSBL, ALR, xSHR, OWM, #EM, +DIG. (Number of symbols does not equal number of individuals due to some indi¬ viduals sharing phenotypes). Genetic evidence supports Agrostis punicea 35 PRINCIPAL COORDINATE ANALYSIS OF INDIVIDUALS The first two axes of the principal co-ordinate analysis accounted for 52.4% and 11.3% of total variation, respectively, with a further 9.9% accounted for by the third axis. Individuals within the populations of A. billardierei var. filifolia (HD, SBL and LR) group together and the populations form a discrete cluster (Fig 3a). Clustering for A. avenacea is similar, although some separation of the SHR population has occurred. However, for A. aemula, whilst the individuals within each variety cluster together (populations EM and DIG), the two varieties form two distinct and quite separate clusters in both princi¬ pal co-ordinate plots (Figs 3a, b). DIVERSITY INDICES Estimates of individual population diversity (H Q ) were derived from RAPD phenotypes (Table 2). The highest value of H Q = 0.0808 was for population SHR of A. avenacea. No variation was detected in SBL and low levels were found in HD and DIG. While levels of diversity were variable depending on populations, they were not necessarily correlat¬ ed with population size. Average diversity for each species was calculated separately as Ablll // sp =0.1526 for A. billardierei, Aaven // sp =0.2281 for A. avenacea and Aaem // sp =0.2722 for A. aemula (Table 4). The lower value for A. billardierei reflects the low population diversity values for pop¬ ulations HD and SBL. The comparatively high value for A. aemula results from the two populations sharing very few RAPD characters, whereas for A. avenacea, it results from high diversity within each population. Average genus diversity (calculated for all populations combined) was ALL // GEN =0.2559. Average species diversity (calculated for all populations combined) was ALL // Sp =0.2176. 7/po P provided measures of the average population diversity for each species, ranging from 0.0427 to 0.1540, and for the genus, 0.0378. PARTITIONING OF VARIATION Partitioning of variation within and among populations and species can provide an insight into their genetic structure. When variation is partitioned within and among populations (H po /H gen ), using all seven populations in the analysis, only 14.8% of the observed variation is found within populations (Table 4). The proportion of diversity varies with species (Table 4). For A. billardierei , 28.0% of variation was present within populations whereas only 21.8% was present within populations of A. aemula. When indices are cal¬ culated for A. billardierei var.filifolia and A. aemula var. setifolia as a single species, the within-population variation decreases to 17.4%. For A. avenacea, 67.5% of variation was found wit hin populations. Table 4. Genetic diversity indices and partitioning of variation for species analysed sep¬ arately and also combined. Species Diversity index Partitioning of variation Within popns Among popns Agrostis spp. combined ALL 7/pop=°. °3 7 8 ALL ^gen= 0 - 2559 14.8% 85.2% A. billardierei Ahill H POP =o.(mi AbilI H s p =0.1526 28.0% (17.4%)* 72.0% (82.6%)* A. avenacea Aavew // pO p=0.1540 Aam 7/ sp =0.2281 67.5% 32.5% A. aemula Aam // pop =0.0593 Aaew W SP =0.2722 21.8% 78.2% *Partitioning of variation when population DIG is included with A. billardierei. 36 E.A. James et al. Discussion Grass species have often been difficult to define because of minor differences in mor¬ phology, phenotypic plasticity in response to the environment, and frequently, a degree of intergradation between species. An advantage of using RAPDs is that they can be used to determine the affiliation of plants showing morphological characters with values com¬ mon to more than one taxon. This approach has been used successfully for a number of grass species: Agrostis adamsonii (James & Brown 2000), Hordeum (Gonzalez & Ferrer 1993) , Lolium/Festuca complex (Stammers et al. 1995; Pasakinskiene et al. 2000). The use of a molecular method in this study complements the set of characters used in the morphological analysis of Brown and Walsh (2000). COMPARISON OF AGROSTIS AEMULA VAR. SETIFOLIA AND A. BILLARDIEREI VAR. FILIFOLIA Agrostis aemula var. setifolia has been shown to be morphologically distinct from A. aemu- la var. aemula (Brown & Walsh 2000) and A. billiardierei var. filifolia has been shown to be both morphologically and genetically distinct from other varieties of A. billiardierei (Brown & Walsh 2000; James & Brown 2000; Ryan, unpublished data). This study shows a high degree of genetic similarity between A. aemula var. setifolia and A. billardierei var. filifolia. Hierarchical clustering of populations (Fig. 2) shows three main clusters. The three pop¬ ulations of A. billardierei var. filifolia plus the population of A. aemula var. setifolia form a cluster that separates from A. avenacea and A. aemula var. aemula at a similarity of 62.0%. In the principal co-ordinate analysis, A. aemula var. setifolia (DIG) individuals are clearly separated from populations of A. billardierei var. filifolia, although the most sim¬ ilar population is always HD. The clustering of the A. billardierei var. filifolia and A. aem¬ ula var. setifolia populations with 79.8% similarity is strong evidence for their reclassifi¬ cation as a single species or at least as individual species; both separate from A. bil¬ lardierei and A. aemula. Cluster analysis using morphological characteristics found the same relative clustering pattern for A. billardierei var. filifolia, A. aemula var. setifolia and A. aemula var. aemula but at greater similarities (87% between the first two taxa and 79% between the first two and the third) (Brown & Walsh 2000). The two populations of A. avenacea (WM and SHR) cluster at 85.7% similarity, and can be regarded as a single species, despite showing morphological variation and being separated by 70 km. Agrostis avenacea shows a similarity of 78.2% to A. aemula and indicates a closer similarity than would be normally accepted as delineating species. As there is no sharp morphological separation between A. avenacea and A. aemula (Walsh 1994) , their taxonomic status requires further investigation, using a wider range of pop¬ ulations than is reported on here. PARTITIONING OF VARIATION The genetic structure of species is determined by the evolutionary and ecological history of individual species. The breeding systems of grasses influence the partitioning of variation, with selling species having more genetic divergence (nearly half) among populations than mixed-mating and outcrossing species (Godt & Hamrick 1998). For outcrossing grass species, most diversity is found within populations. For example, in Hordeum spontaneum, which is known to have high levels of selling, 43% of variation was found within popula¬ tions (Dawson et al. 1993) whereas 72.9-80.5% within-population-variation was found within Buchloe dactyloides, a diploid out-crossing species (Huff et al. 1993). Only 14.8% of observed variation in the combined Agrostis species genetic analysis is found within populations (Table 4). This is similar to the value (14.3%) found in anoth¬ er study of Agrostis species (James & Brown 2000) and is consistent with the low level of gene flow which is expected if the populations studied, comprise different species. The proportion of variation partitioned within-populations for A. billardierei var. filifo¬ lia (28.0%) and A. aemula (21.8%) is much lower than the value of 73% found for other grass species (Godt & Hamrick 1998), the 78% reported for outcrossing species (Hamrick Genetic evidence supports Agrostis punicea 37 et al. 1991) or the near 50% values reported for selling species. The results for these Agrostis species could be due to self-fertility and the result of limited gene exchange among populations due to fragmentation of habitats, or some asexual reproduction. The low levels of variation in populations HD and DIG (Table 2) support the low levels commonly found in species with restricted ranges (Hamrick et al. 1991) but the complete absence of varia¬ tion in SBL also indicates possible differences in reproductive strategies. In A. avenacea, the variation within populations (67.5%) is si mil ar to that found pre¬ viously for A. avenacea (57.9%) using RAPDs (James & Brown 2000). Although the level of selfing in A. avenacea is not known, the variation within populations is lower but still comparable to an out-crossing species. Agrostis avenacea is a widespread, more common species than the others studied here and although some selfing may occur, its gene flow may not be as restricted. POPULATION SIZE, GENETIC DIVERSITY AND BREEDING SYSTEM IMPLICATIONS The mating system of plants plays an important role in determining the genetic structure and diversity of populations. Compared to other plants, grasses, in general, have higher levels of genetic diversity but there is more genetic differentiation among populations. For example, about 27% of total genetic variation is partitioned among grass populations compared with 22% for other plant species (Godt & Hamrick 1998). Differences in the amount of variation in the populations of A. billardierei var. filifolia suggest that there may be variations in the breeding system within this taxon. It also sug¬ gests a flexible breeding system. It has been found that selfing species of grasses are genet¬ ically depauperate, relative to mixed mating and outcrossing species, both within popula¬ tions and within species and in general, most are annuals whose population sizes often fluc¬ tuate (Godt & Hamrick 1998). It is possible that some of the Agrostis populations are ephemeral and sourced from larger, permanent populations where reproduction is mainly sexual, but apomixis occurs periodically. Population LR has a high level of diversity (// O =0.0487) and could consist of plants which almost always reproduce sexually. On the other hand, populations HD (i/ o =0.0097) and SBL (H Q = 0.0) may have breeding systems with more emphasis on apomixis or selfing. SBL plants were grown from seed collected from different parents and the lack of variation in the sample suggests that seed may be apomictic in origin and that the population was founded by seed from a single parent plant. Similarly for DIG, there may be years where seed production is largely apomictic, leading to a cohort of genetically identical seedlings. Alternatively, the low diversity in DIG may be a result of repeated genetic bottlenecks if plant numbers fluctuate widely over time. If apomictic populations of Agrostis species do occur, or if apomixis is a component of the breeding system, the genetic structure of populations may change rapidly. These results highlight the need for detailed breeding system studies on individual species or groups to understand their genetic structure. Conclusions Genetic variation detected with RAPD-PCR supports the conclusions resulting from a morphological study by Brown and Walsh (2000) placing A. aemula var. setifolia and A. billardierei var .filifolia in a new species, A. punicea. Other taxonomic issues still exist in the lowland Agrostis species in south-eastern Australia and a critical appraisal of addi¬ tional closely related taxa, including assessment of reproductive structures, ploidy levels and breeding systems is still required to resolve them. Acknowledgments Our thanks are extended to John Reynolds, Peter Franz and Daniel Isenegger (Department of Natural Resources and Environment, Victoria) for their biometrical advice and practical assistance in the data analysis. We are also grateful to nursery staff at the Royal Botanic Gardens Melbourne for maintaining the living plant collections. 38 E.A. James et al. References Brown, A J. and Walsh, N.G. (2000). A revision of Agrostis billardierei R. Br. (Poaceae). Muelleria 14, 65-90. Chakraborty, R. and Rao, C.R. (1991). ‘Measurement of genetic variation for evolutionary studies’, in R. Chakraborty and C.R. Rao (eds), Handbook of Statistics, Vol 8. Elsevier Science Publishers: B.V. Dawson, I.K., Chalmers, K.J., Waugh, R. and Powell, W. (1993). Detection and analysis of genet¬ ic variation in Hordeum spontaneum populations from Israel using RAPD markers. Molecular Ecology 2,151-159. Godt, M.J.W. and Hamrick, J.L. (1998). ‘Allozyme diversity in the grasses’, in G.P. Cheplick (ed.), Population biology of grasses, pp. 183-208. Cambridge University Press: UK. Gonzalez, J.M, and Ferrer, E. (1993). Random amplified polymorphic DNA analysis in Hordeum species. Genome 36, 1029-1031. Gordon, A.D. (1981). Classification. Chapman and Hall: London. Gower, J.C. (1966). Some distance properties of latent root and vector methods used in multivari¬ ate analysis. Biometrika 53, 325-338. Hamrick, J.L., Godt, M.J.W., Murawski, D.A. and Loveless, D.M. (1991). ‘Correlations between species traits and allozyme diversity: implications for conservation biology’, in D.A. Falk and K.E. Holsinger (eds), Genetics and Conservation of Rare Plants, pp. 75-86. Oxford University Press: New York. Huff, D.R., Peakall, R. and Smouse, PE. (1993). RAPD variation within and among natural popu¬ lations of outcrossing buffalograss [Buchloe dactyloides (Nutt.) Engelm.]. Theoretical and Applied Genetics 86, 927-934. James, E.A. and Brown, A.J. (2000). Morphological and genetic variation in the endangered Victorian endemic grass Agrostis adamsonii Vickery (Poaceae). Australian Journal of Botany 48, 383-395. Kazan, K., Manners, J.M. and Cameron, D.F. (1993). Genetic relationships and variation in the Stylosanthes guianensis species complex assessed by random amplified polymorphic DNA. Genome 36, 43^-9. King, L.M. and Schaal, B.A. (1989). Ribosomal-DNA variation and distribution in Rudbeckia mis- souriensis. Evolution 43, 1117-1119. Pasakinskiene, I., Griffiths, C.M., Bettany, A.J.E., Paplauskiene, V. and Humphreys, M.W. (2000). Anchored simple-sequence repeats as primers to generate species-specific DNA markers in Lolium and Eestuca grasses. Theoretical and Applied Genetics 100, 384-390. Rogers, S.O. and Bendich, A.J. (1985). Extraction of DNA from milligram amounts of fresh, herbarium and mummified plant tissue. Plant Molecular Biology 5, 69-76. Russell, J.R., Hosein, F., Waugh, R. Powell, W. (1993). Genetic differentiation of cocoa (Theobroma cacao L.) populations revealed by RAPD analysis. Molecular Ecology 2, 89-97. Scarlett, N.H., Wallbrink, S.J. and McDougall, K. (1992). Native Grasslands. Victoria Press: South Melbourne. Stammers, M., Harris, J., Evans, G.M., Hayward, M.D. and Foster, J.W. (1995). Use of random PCR (RAPD) technology to analyse phylogenetic relationships in the Lolium/Festuca complex. Heredity 74, 19-27. Vickery, J.W. (1941). A revision of the Australian Agrostis Linn. Contributions from the New South Wales National Herbarium 1, 101-119. Walsh, N.G. (1994). ‘Poaceae’, in N.G. Walsh and T.J. Entwisle (eds), Flora of Victoria 2, 356-627. Inkata Press: Sydney. Williams, J.G.K., Kubelik, A.R., Livak, K.J., Rafalski, J.A. and Tingey, S.V. (1990). DNA poly¬ morphisms amplified by arbitrary primers are useful as genetic markers. Nucleic Acids Research 18, 6531-6535. Note added in proof: Since completion and submission of the study reported in this paper, the genus Lachnagrostis has been recognised to occur in Australia (Jacobs, Telopea 9, 439-448, 2001; Jacobs, Telopea 9, 837-838, 2002). The current names for taxa referred to in the text, with syn¬ onyms in brackets, are L. aemula (R.Br.) Trin. (syn. Agrostis aemula ), L. billardierei (R.Br.) Trin. (syn. A. billardierei), L. fdiformis (Forst.) Trin. (syn. A. avenacea), L. scabra (Beauv.) Nees. ex. Steudel (syn. A. rudis ), L. adamsonii (Vickery) S.W.L.Jacobs (syn. A. adamsonii ), L. robusta (Vickery) S.W.L.Jacobs (syn. A. billardierei var. robusta), L. punicea ssp. punicea (A.J.Brown & N.G.Walsh) S.W.L.Jacobs (syn. A. punicea var. punicea, A. aemula var. setifolia) and L. punicea ssp .filifolia (Vickery) S.W.L.Jacobs (syn. A. punicea var. filifolia, A. billardierei var. filifolia). Muelleria 16: 39-42 (2002) Notes on Conothamnus Lindl. with the description of a new section, sect. Gongylocephalus Craven (Myrtaceae) L.A. Craven Australian National Herbarium, Centre for Plant Biodiversity Research, CSIRO Plant Industry, GPO Box 1600, Canberra, ACT 2601. Lyn.Craven@csiro.au Abstract New morphological observations of Conothamnus Lindl. are reported and the new section, Gongylocephalus Craven, is described to accommodate Trichobasis Turcz. nom. illeg. which has been included in Conothamnus without taxonomic recognition until now. Introduction Conothamnus Lindl. was established by Lindley in 1839, based upon the single species C. trinervis Lindl. In 1852 Turczaninow described the genus Trichobasis Turcz. with T. aurea Turcz. its sole species. The name T. aurea is typified by the KW set of the collec¬ tion Drummond 5th coll. 147. Turczaninow’s generic name is illegitimate, being a later homonym of Trichobasis Leveille, published in 1849. Bentham (1867) in effect trans¬ ferred Trichobasis aurea to the previously monotypic Conothamnus when he described C. divaricatus Benth. the name of which is typified by the K set of the type collection of T. aurea. In 1904 Diels added a third species to Conothamnus, C. neglectus Diels. All three species are restricted to the southwest of Western Australia. Conothamnus, including Trichobasis, has been treated as one of the several genera related to Melaleuca L., from which it has been distinguished by previous authors (e.g. Bentham 1867; Johnson & Briggs 1983; Rye 1987) by possession of a single ovule in each locule compared to several ovules per locule in Melaleuca. My observations of the three species of Conothamnus, C. aureus, C. neglectus and C. trinervis, were incongru- ent with those generally recorded in the literature. Primarily, the differences noted between the three species of Conothamnus and Melaleuca are: C. trinervis, C. aureus, C. neglectus : Ovules 2 per locule, one on each side of an axile- basal and sometimes flange-like placenta, laterally or laterally-basally attached. Seeds 1 or 2 per locule, semi-obovoid, or narrowly or flattened obovoid, concave on the placen¬ tal side, the testa coriaceous or thinly coriaceous, white or whitish-brown. Abortive/ster¬ ile ovules not developing into chaff. Melaleuca : Ovules several to numerous, irregularly arranged on an axile-median to axile- basal peltate placenta, basally attached. Seeds few per locule, angular oblong-obovoid to subobovoid, the testa membranous, brown. Abortive/sterile ovules developing into chaff. Within Conothamnus two groups may be distinguished as follows: C. trinervis: Flowers in triads. Placenta well developed and distinctly flange-like. Seed semiobovoid, strongly concave on the placental side, the testa coriaceous, white. C. aureus, C. neglectus: Flowers in dyads. Placenta not well developed. Seed narrowly or flattened obovoid, the testa thinly coriaceous, white or whitish-brown. The above differences are indicative of two evolutionary lines within Conothamnus and taxonomic recognition at sectional level is warranted. Although the epithet Trichobasis is available for use at infra-generic rank, the section that includes the nomen- clatural type of Trichobasis, T. aureus, is described as new with a new epithet, Gongylocephalus. Features of the placentation and seed of Conothamnus and the type species of Melaleuca, M. leucadendra (L.) L., are illustrated in Fig. 1. 40 L.A. Craven Figure 1. The ovary and seed of Conothamus trinervis (E, F, G, H) and Melaleuca leu- cadendra (A, B, C, D). A,E: longitudinal view through ovary locule; B, F: cross-sectional view through ovary locule (the flange-like placenta in F dis¬ torted as an artefact of drying). C, G: seed. D, H: cotyledon form. A-D from Lazarides & Adams 322 (CANB); E from Hoyle 211 (CANB); F-H from Hnatiuk 760250 (PERTH). Conothamnus 41 Conothamnus Lindl., Edwards’s Bot. Reg. Appendix vols 1-23: ix (1839). Melaleuca sect. Conothamnus (Lindley) Baill., Hist. pi. 6: 359 (1876). Species typica: C. trinervis Lindl. Conothamnus sect. Conothamnus Shrubs. Leaves decussate or rarely temate; venation parallel. Inflorescences capitate or spicate, pseudoterminal with the apex usually growing on after anthesis, several- to many-flowered. Flowers in triads, each triad subtended by a bract and with the con¬ stituent flowers all ebracteolate or variably bracteolate; not stipitate; perigynous. Hypanthium adnate to the ovary for the proximal 1/4 to 1/2 of the ovary. Sepals 5. Petals 5. Stamens indefinite; filaments fused into 5 antepetalous bundles; anthers dorsifixed, versatile, 2-celled, dehiscing by longitudinal slits. Ovary 3-locular; ovules 2 per locule, one each side of an axile-basal flange-like placenta, laterally attached. Stigma puncti- form. Fruit, dry, not or scarcely woody, the locules dehiscing by valves. Seeds 1 or 2 per locule, semi-obovoid, concave on the placental side, the testa coriaceous, white. Embryo with the cotyledons about 1/2 the embryo length and flattened planoconvex. C. trinervis Lindl., Edwards’s Bot. Reg. Appendix vols 1-23: ix (1839). Typus: Western Australia, Drummond s.n. (holotypus CGE, n.v, photo in CANB). Melaleuca cuspidata Turcz., Bull. Soc. Imp. Naturalistes Moscou 35: 327 (1862). Typus : Western Australia, Drummond 7th coll. 77 (holotypus KW; isotypi MEL, NSW). Conothamnus trinervis occurs in the Eneabba-Badgingarra district and in the Kalamunda area. Conothamnus sect. Gongylocephalus Craven, sect. nov. A sectione typica floribus duplicatis; petalis praesentibus vel carentibus; filamentis staminalibus liberis et staminis 5-aggregatis vel dispersis vel staminis coalitis in quinque fasciculis; stylis usque 5 mm longis; et placenta non distincte pteroidea differt. Species typica: Conothamnus aureus (Turcz.) Domin Trichobasis Turcz., Bull. Cl. Phys.-Math. Acad. Imp. Sci. Saint-Petersbourg 10: 337 (1852), nom. illegc, non Leveille (1849). Typus: T. aureus Turcz. Shrubs. Leaves decussate to subdecussate; venation parallel-pinnate. Inflorescences cap¬ itate or spicate, pseudoterminal and sometimes also in distal leaf axils, with the apex usu¬ ally growing on after anthesis, several- to many-flowered. Flowers in dyads, each dyad subtended by a bract and with the constituent flowers ebracteolate; stipitate or not; perig¬ ynous. Hypanthium adnate to the ovary for the proximal 1/4 to 2/3 of the length of the ovary. Sepals 5 (rarely 6). Petals 5 (rarely 6) or absent. Stamens indefinite; filaments free and the stamens grouped in antepetalous clusters (sometimes dispersed around the hypan¬ thium apex) or fused in 5 antepetalous bundles; anthers dorsifixed, versatile, 2-celled, dehiscing by longitudinal slits. Ovary 3-locular; ovules 2 per locule, one on each side of a small axile-basal non-peltate placenta, laterally-basally attached. Stigma punctiform. Fruit dry, not or scarcely woody, the locules dehiscing by valves. Seeds usually 1 per locule, narrowly or flattened obovoid, the testa thinly coriaceous, white or whitish-brown. Embryo with the cotyledons about 1/2 the embryo length and flattened planoconvex. The new sectional epithet is derived from the Greek gongylos (ball, sphere) and kephale (head) in reference to the shape of the inflorescence. C. aureus (Turcz.) Domin, Mem. Soc. Roy. Sci. Boheme. Prague 1921-2 2: 91 (1923). Trichobasis aurea Turcz., Bull. Cl. Phys.-Math. Acad. Imp. Sci. Saint-Petersbourg 10: 42 L.A. Craven 337 (1852). Typus : Western Australia, Drummond 5th coll. 147 (holotypus KW, n.v., photo in PERTH n.v.; isotypi G, K n.v., MEL, NY n.v.). Conothamnus divaricatus Benth., FI. Austral. 3: 164 (1867). Typus: Western Australia, Drummond 5th coll. 147 (holotypus K, n.v.; isotypi G, KW n.v. photo in PERTH n.v., MEL, NY n.v.). This species occurs from the Stirling Range east to Israelite Bay. C. neglectus Diels in Diels & Pritzel, Bot. Jahrb. Syst. 35: 430 (1904). Syntypi: Western Australia: Mt Melville near King George Sound, Mueller s.n. (B 1 ", ?MEL n.v.); near Cranbrook, Diels 4438 (B 1 ; PERTH fragm); about 15 km N of Albany, Diels 6034 (B 1 ). C. neglectus occurs from Walpole east to Albany and as far north as Borden. The name C. neglectus is not being lectotypified here as a thorough search for other syntypes or isosyntypes has not been made. More ample materials of Diels 4438 and/or 6034 than the fragment of Diels 4438 in PERTH may exist. Material from Mount Melville, ascribed to Mueller by Diels in the protologue, has not been seen from MEL but, in view of the wide distribution of specimens from MEL to European and north American herbaria, such material may be extant. Key to the species of Conothamnus l. Flowers in triads; style 12-15 mm long (petals present).C. trinervis 1. Flowers in dyads; style 2-5 mm long 2. Petals absent; style 3.4-5 mm long.C. aureus 2. Petals present; style 2-2.5 mm long.C. neglectus Acknowledgments The directors and curators of the following herbaria are thanked for the opportunity to study collections in their care: AD, BRI, CANB, KW, MEL, NSW, PERTH. Julie Matarczyk is thanked for her assistance with bibliographic work and data collection. The illustration was prepared by Rob Warren. Preparation of this paper in part was supported by the Australian Biological Resources Study. References Bentham, G. (1867, ‘1866’). ‘Myrtaceae’. Flora Australiensis 3, 1-289. Lovell Reeve & Co.: London. Johnson, L.A.S. and Briggs, B.G. (1983). ‘Myrtaceae’, in B.D. Morley and H.R. Toelken (eds), Flowering Plants in Australia, pp. 175-185. Rigby: Willoughby: Sydney. Rye, B.L. (1987). ‘Myrtaceae’, in N.G. Marchant, J.R. Wheeler, B.L. Rye, E.M. Bennett, N.S. Lander and T.D. Macfarlane (eds), Flora of the Perth Region 1, 377-429. Western Australian Herbarium: Perth. Muelleria 16: 43-45 (2002) Calotis cuneata var. pubescens (Asteraceae), change in rank and notes on its distribution and ecology N.G. Walsh ] ’ 3 and K.L. McDougall 2 1 National Herbarium of Victoria, Royal Botanic Gardens Melbourne, South Yarra, Vic. 3141 2 Threatened Species Unit, New South Wales National Parks and Wildlife Service, PO Box 2115, Queanbeyan, NSW 2620 3 Author for correspondence: neville.walsh@rbg.vic.gov.au Abstract Calotis cuneata (F.Muell. ex Benth.) G.L.Davis var. pubescens (F.Muell. ex Benth.) G.L.Davis is elevated to species rank as Calotis pubescens (F.Muell. ex Benth.) N.G.Walsh & K.L.McDougall and its ecology and conservation status are discussed. Background Until recently Calotis cuneata (F.Muell. ex Benth.) G.L.Davis var. pubescens (F.Muell. ex Benth.) G.L.Davis was known from only four collections: F. Mueller’s 1854 type from ‘grassy mountains on the Mitta Mitta’, Victoria (MEL), two 1956 collections by M. Mueller from Nungar Plain and Snowy Plain, Kosciuszko National Park, New South Wales (CANB), and a 1967 collection by W. Bryant from Nungar Plain (NSW). The taxon has not been recollected in Victoria and may now be extinct in that state although Ross (2000) regarded it as ‘poorly known’. Many likely areas of occurrence in F. Mueller’s rather vague collection area are now largely converted to pasture, although the possibility exists that it still occurs on grassy plains of the Alpine National Park, Bogong Unit. Surveys by one of us (KMcD) in February 2001 on Nungar Plain, recorded C. cuneata var. pubescens at four sites, and further surveys by both of us in December 2001 located the taxon at 19 individual sites on the same plain. Surveys of similar subalpine grassy plains in the area (e.g. Long Plain, Happy Jacks Plain, Boggy Plain, Gulf Plain) have not resulted in further discoveries of the taxon. A survey in February 2001 of Snowy Plain in Kosciuszko National Park, where C. cuneata var. pubescens had been collected by M. Mueller in 1956, failed to locate the taxon (R. Rehwinkel, NPWS, pers. comm.). The recent gatherings of the taxon have allowed more detailed comparisons with the typical variety of C. cuneata. Following these comparisons we are convinced that, despite superficial similarities in the mature cypselas, C. cuneata var. pubescens is rather dis¬ tantly related to C. cuneata sens. str. and is at least as closely related to C. scabiosifolia Sond. & F.Muell. (within which Bentham (1867) originally included both varieties of C. cuneata ). In erecting Calotis cuneata Davis (1952) distinguished it from C. scabiosifolia (in which she retained two of Bentham’s (1867) original six varieties) by the presence of a second series of fine plumose awns within the ring of peripheral awns on the apex of the cypsela. The two varieties of C. cuneata were distinguished by foliar characters, indu¬ mentum, and by the presence of a patch of appressed hairs on the central part of the body of the cypsela of var. pubescens. Although significant within C. cuneata, this last feature also occurs on C. scabiosifolia var. integrifolia F.Muell. ex Benth. The main cypsela awns of C. cuneata var. pubescens are unique amongst all four taxa within C. scabiosifolia and C. cuneata in being non-scabrous. More detailed comparison of the cypselas of the two varieties of C. cuneata shows some significant discriminating features not noted by Davis or in the account of the genus by Everett (1992). The central awns of C. cuneata var. cuneata are united into a solid col¬ umn in their basal half. This column encircles the base of the corolla. The central awns 44 N.G. Walsh and K.L. McDougall of C. cuneata var. pubescens are much finer and free to their bases. The margins of the cypselas of var. cuneata are narrow and acute while those of var. pubescens are broadly thickened and rounded rather like those of C. scabiosifolia. The general indumentum of C. cuneata var. pubescens differs from the typical vari¬ ety and from both varieties of C. scabiosifolia. The last three taxa have strigose, sep¬ tate hairs of varying density and coarseness, but all are evenly tapered from base to apex. Calotis cuneata var. pubescens has hairs with a coarse, erect septate base that is rather abruptly attenuated into a distinctly longer and finer apical part. This apical fil¬ ament is often lost from the hairs of older and/or exposed parts of the leaves and stems leaving the persistent basal stub which results in a coarse hispid indumentum on these parts. The distribution of C. cuneata as it is currently circumscribed gives a pattern that can¬ not be reconciled with a notion of relatively recent evolution of two entities from a com¬ mon ancestor (as might be inferred from their varietal status). The typical variety is wide¬ ly distributed from inland northern New South Wales to central Queensland while var. pubescens is highly localised in the subalps of north-eastern Victoria and southern New South Wales. For the reasons outlined above we here elevate C. cuneata var. pubescens to specific rank. Taxonomy Calotis pubescens (F.Muell. ex Benth.) N.G.Walsh & K.L.McDougall, stat. nov. Calotis scabiosifolia Sond. & F.Muell. var. pubescens F.Muell. ex Benth., FI. Austral. 3: 503 (1867). Calotis cuneata (F.Muell. ex Benth.) G.L.Davis var pubescens (F.Muell. ex Benth.) G.L.Davis, Proc. Linn. Soc. New South Wales 77: 178 (1952). Lectotype: ‘Grassy mountains on the Mitta Mitta’, F. Mueller s.n., 1854 (MEL) fide G.L. Davis loc. cit. Ecology At Nungar Plain, C. pubescens occurs in a herbfield community (in which it may be dominant) on gentle slopes between Eucalyptus pauciflora woodland and the valley floor which is vegetated by a mosaic of Poa-dominated tussock grasslands, open heaths dominated by Hovea montana and Cyperaceae-rich wetland communities. Soils are of the alpine humus type developed on a parent material of Silurian siltstone and shale of the Tantangara Formation. The altitude range is small, between c. 1340 and 1380 m a.s.l. Colonies of C. pubescens may comprise a single genet developed by rhizomatous growth and can be up to 10m in diameter. Typically associated species include Bulbine glauca, Coprosma nivalis, Leptorhynchos elongatus, Oreomyrrhis argentea, Plantago euryphylla, Poa petrophila, Poa hookeri and Wahlenbergia densifolia. Calotis glandulosa F.Muell. is also relatively abundant on the plain. A comprehensive checklist of the flora of Nungar Plain will be published elsewhere (McDougall & Walsh in prep.). Conservation Status Based on Briggs and Leigh (1996), an appropriate conservation code for Calotis pubes¬ cens is 3ECi. The species is threatened by feral pigs, which have excavated large areas of vegetation on Nungar Plain, especially the herbland community containing C. pubescens. Acknowledgements We are grateful to the two anonymous referees for useful comments on drafts of this paper. Calotis pubescens 45 References Bentham, G. (1867). Flora Australiensis 3. Lovell Reeve & Co: London. Briggs, J.D. and Leigh, J.H. (1996). Rare or threatened Australian plants. CSIRO Publishing: Collingwood. Davis, G.L. (1952). Revision of the genus Calotis R.Br. Proceedings of the Linnean Society of New South Wales 77, 146-188. Everett, J. (1992). ‘ Calotis ’ in G.J. Harden (ed.), Flora of New South Wales 3, 169-174. University of New South Wales Press: Sydney. McDougall, K.L. and Walsh, N.G. (in prep.). The flora of Nungar Plain, Kosciuszko National Park, New South Wales. Cunninghamia. Ross, J.H. (2000). A census of the vascular plants of Victoria (6 th edn). Royal Botanic Gardens Melbourne: Melbourne. Muelleria 16: 47-53 (2002) Successful DNA amplification from Acacia (Leguminosae) and other refractory Australian plants and fungi using a nested/semi-nested PCR protocol Frank Udovicic 1,3 and Daniel J. Murphy 1,2 1 National Herbarium of Victoria, Royal Botanic Gardens Melbourne, Private Bag 2000, South Yarra, Vic. 3141 2 School of Botany, The University of Melbourne, Vic. 3010 3 Author for correspondence: Frank.Udovicic@rbg.vic.gov.au Abstract Despite the promise of the polymerase chain reaction (PCR) as a method for rapid generation of informative phylogenetic characters, it has proven difficult or impossible to amplify the common¬ ly used internal transcribed spacer (ITS) 1+2 regions using standard approaches for several groups of Australian plants and fungi. This paper describes a two step nested or semi-nested PCR protocol that was utilised to provide the first consistent amplification of the ITS region from Acacia Mill. (Mimosoideae: Leguminosae). Subsequently these methods have proven to provide reliable ampli¬ fication of ITS from previously intractable Australian members of the Rutaceae and Cortinariaceae. In addition to use with fresh material, this method has also proven effective with herbarium mate¬ rial up to 26 years old. It is hoped these methods may facilitate use of the ITS region from a broad¬ er range of plants and fungi, including herbarium material. Introduction In recent years the use of PCR for obtaining DNA sequences for systematic s has expand¬ ed greatly. Of the DNA regions sequenced, the internal transcribed spacer (ITS) region of 18S-26S nuclear ribosomal DNA has proven to be particularly useful for plant (Baldwin et al. 1995) and fungal (Edel 1998) systematics. The relatively small size of ITS and its high copy number facilitates its amplification and sequencing (Baldwin et al. 1995), and rapid concerted evolution of the nuclear ribosomal DNA multigene family acts to pro¬ mote intragenomic uniformity of repeats (Arnheim et al. 1980; Appels & Dvorak 1982; Arnheim 1983; Hillis et al. 1991), decreasing the likelihood of paralogy confounding phylogenetic reconstructions based on ITS sequences. Ease of alignment, and sufficient sequence variation between closely related species, has also contributed to the use of ITS for systematics. The popularity of ITS has resulted in more than 45,000 records being submitted to the GenBank DNA sequence database, which makes comparison of sequences with other taxa, worldwide, extremely easy. The many advantages of the ITS region has seen it used for examining relationships between several Australian plant groups, e.g. Leguminosae: Papilionoideae (Crisp et al. 2000), Hovea R.Br. (Leguminosae, Thompson et al. 2001), Beaufortia R.Br. suballiance (Myrtaceae, Ladiges et al. 1999; Brown et al. 2001) and Eucalyptus L’Her. (Myrtaceae, Steane et al. 1999; Udovicic & Ladiges, 2000). Likewise, when it was necessary to sequence a region of nuclear DNA for acacias, ITS was deemed to be the ideal candidate. Initially, amplification of DNA from Acacia Mill, was attempted using primers that had proven successful for another legume genus, Lupinus L. (ITS 18, ITS26, S3, S4, S5 and S6; Kass & Wink 1997). PCR reactions with these primers, in a variety of combinations, and using standard protocols either amplified nothing or multiple fragments. When sin¬ gle bands (from samples that had multiple fragments amplified) were isolated from agarose gel and sequenced, the sequences invariably matched fungal DNA sequences in GenBank. Consequently it was necessary to use primers more specific for higher plants in an attempt to avoid amplifying these fungal sequences by PCR. Angiosperm-specific primers, 17SE and 26SE (Sun et al. 1994), were tried. These 48 F. Udovicic and D .J. Murphy primers also resulted in multiple PCR fragments and preferential amplification of unwanted fungal DNA or chloroplast ribosomal DNA. If higher annealing temperatures (above 62°C) were used, no DNA was amplified. When the 17SE and 26SE primer sequences were compared to the partial 18S and 26S sequences available for Acacia in GenBank - from A. fimbriata A.Cunn. ex G.Don (Martin & Dowd 1993) - it was found that there were four bp differences between the 18S sequence and the 17SE primer at the 5’ end. J. Miller (CSIRO Canberra) cloned the ITS region for several Acacia taxa and designed a new forward primer in the 18S (AcF). This primer was found to amplify the ITS region in combination with 26SE. High stringency PCR conditions (annealing tem¬ peratures above 60°C and hotstart PCR) were required to produce a single amplicon. Difficulties were experienced in obtaining a high yield of PCR product, especially when higher annealing temperatures were used. To overcome this problem a two step, nested or semi-nested PCR strategy was utilised, which incorporated a PCR additive (Q-solution, QIAGEN) to change the melting properties of the double stranded template DNA. The nested PCR approach was originally developed for detection of hepatitis C viral sequences in blood samples (Garson et al. 1990), and because of its sensitivity for ampli¬ fication of low-copy number templates it is frequently used for characterisation and iden¬ tification of micro-organisms. Here, the ultra-sensitivity of nested PCR is used to assist in amplification of the ITS region of Australian acacias, Rutaceae and fungi. Materials and Methods Material of Acacia, Rutaceae ( Phebalium Vent, and allied genera) and fungi (Cortinariaceae: Cortinarius (Pers.) Gray and Dermocybe (Fr.:Fr.) Wtinsche) was sam¬ pled from the field, living collections or herbarium collections. Fresh material was stored at 4°C until isolation of DNA was performed within a week of collection. Alternatively, material collected in the field was dried rapidly in silica gel and stored until convenient to isolate (Chase & Hillis 1991). Where possible herbarium material was chosen from collections under five years of age, although some collections sampled were up to 26 years old. Genomic DNA was isolated using a CTAB protocol (Udovicic et al. 1995), followed by purification with Qiagen Tip 20 columns or by using Prep-a-Gene (Bio Rad). Herbarium samples were preferentially isolated using commercial kits, QIAGEN DNeasy min i kits or NucleoSpin kits (Macherey-Nagel, Dtiren, Germany), because of the small quantity of starting material needed and the consistent quality of resulting DNA. The polymerase chain reaction (Mullis & Faloona 1987; Saiki et al. 1988) was used to amplify the ITS 1 + 2 region using a two step process (Fig. 1). Total volume of all DNA amplifications was 50 pi. First round reactions contained 0.2 mM dNTPs, 3 mM MgCl 2 , 10 pmol each primer (Table 1), 1.25 units HotStar Taq DNA polymerase (Qiagen), 30-100 ng of template DNA and 10 pi Q-solution (Qiagen). Thermal cycling was performed on an Eppendorf Mastercycler gradient thermal cycler with one hold at 95°C for 15 min preceding 30 cycles of 94°C for 30 s, 64°C for 30 s, 72°C for 20 s, and followed by one hold at 72°C for 5 min. In second round reactions a 0.5-1 pi aliquot of PCR product from the first round reac¬ tion was used as the template. Alternatively, template DNA was obtained from agarose gels of first round reactions by “stabbing” the target band with a 10 pi pipette tip, fol¬ lowed by agitation in 20 pi of ultrapure water, 10 pi of which was used in the second round PCR. Apart from primers for nested or semi-nested PCR (Table 1), reaction com¬ ponents and conditions were unchanged from the first round. Products of amplifications were visualised by agarose gel electrophoresis and stained with ethidium bromide. Successful DNA amplification 49 First round of PCR using primers IF and 1R IF 2F Product of first round of PCR Second round of PCR Nested PCR with primers 2F & 2R OR Semi-nested PCRs for ITS1 using primers IF & iR and for ITS 2 using primers iF & 1R H V 2F IF iF Figure 1. Diagram of nested and semi-nested PCRs, including location of primers. IF and 1R denote primers used for the first round of PCR, 2F and 2R are primers used for nested PCR, and iF and iR are internal primers used for semi-nest¬ ed PCR. See Table 1 for primers used for different taxa. Note that DNA regions and relative positions of primers are not to scale. Table 1. PCR primers used for different taxa. Taxa 1 st Round Primers 2 nd Round Primers Internal Primers IF IR 2F 2R iF iR Acacia AcF 1 26SE 2 S3 3 S4 3 S6 3 S5 3 Rutaceae ITS18 3 ITS26 3 S3 3 S4 3 S6 3 S5 3 Fungi ITS1 4 ITS4-B 5 ITS1 4 ITS4 4 ITS3 4 ITS2 4 Notes: Origins of primers as denoted by superscript numerals: 1= J. Miller, CSIRO, Canberra; 2= Sun et al. (1994); 3= Kass and Wink (1997); 4= White et al. (1990); 5= Gardes and Bruns (1993). Results and Discussion For Acacia , the first round of amplification resulted in a DNA fragment of approximate¬ ly 750-760 bp. This product was generally only present in low quantities and was diffi¬ cult or impossible to visualise. The second round of nested PCR usually amplified a sin¬ gle fragment of DNA approximately 650 bp long (Fig. 2). If the second round of ampli¬ fication resulted in a low yield of product (Fig. 2, lane 2), multiple fragments (Fig. 2, lane 50 F. Udovicic and D .J. Murphy 1 23456789 2000 bp 850 bp 650 bp Figure 2. Second round of nested PCR amplification of Acacia DNA using primers S3 and S4. Aliquots of 5 pi taken from 50 pi PCR reactions, electrophoresed in a 1.5% agarose gel, stained with ethidium bromide and visualised with UV light. 1: lkb Plus molecular weight ladder; 2: Acacia verticillata (L’Her.) Willd., note low yield of product; 3: A. data Benth.; 4: A. latisepala Pedley; 5: A. translucens Cunn. ex Hook.; 6: A. tumida F.Muell. ex Benth.; 7: A. ampliceps Maslin; 8: A. colei Maslin & Thompson; 9: A. platycarpa F.Muell., note the multiple amplification products. 9), or was unsuccessful, semi-nested PCR was then used. Semi-nested PCR amplified fragments of approximately 400 bp for ITS 1 and 450 bp for ITS 2 (Fig. 3). Subsequent use of this protocol for Rutaceae and fungal DNA yielded comparable results to Acacia. Primer selection was found to be of particular importance for amplification of ITS. Use of non-specific primers resulted in little or no amplification of the target template or preferential amplification of non-target templates. However, problems with amplification were not always due to non-specificity of primers; for example, in this study a new for¬ ward primer was synthesised to target the 3’ end of the 18S rDNA gene of Acacia. Although being a perfect match to a published sequence for A. fimbriata, the Acacia-spe¬ cific primer failed to amplify the ITS region. Such problems may be due to secondary structure in the target area of the primer. Alternatively, the species of Acacia trialed may have had mismatches at the priming site, although, this is unlikely given the conservative nature of the 18S rDNA sequence. In the case of Rutaceae, primers used for legumes (Kass & Wink 1997) worked satisfactorily for nested/semi-nested PCR. For fungi, primers ITS1 and ITS4-B were used for the first round of amplification. These were determined to give the best amplification of DNA from Dermocybe and Cortinarius after trialing the commonly used universal primers, ITS1 and ITS4 (White et al. 1990) and fungal (ITS1-F) and basidiomycete (ITS4-B) specific primers (Gardes & Bruns 1993) in all combinations. The advantage of the nested PCR approach has been the ability to use a high anneal¬ ing temperature, allowing high stringency PCR for avoidance of non-target amplification. Subsequent sequencing revealed a high G + C content (>70%) in the ITS region of Acacia, compared to a G + C content of 50-60 % in the fungal contaminants. This could explain the preferential amplification of contaminants at standard annealing tempera¬ tures, because at these temperatures unmelted G + C regions can prevent primer binding (Borman et al. 2000). Use of high annealing temperatures reduced the incidence of non- Successful DNA amplification 51 Figure 3. 2000 bp 500 400 bp bp 1 2 3 4 5 6 7 A comparison of PCR products resulting from semi-nested and nested amplifi¬ cation of Acacia DNA. Aliquots of 5 pi taken from 50 pi PCR reactions, elec- trophoresed in a 1.5% agarose gel, stained with ethidium bromide and visu¬ alised with UY light. Lanel: lkb Plus molecular weight ladder; Lanes 2 and 3 show semi-nested PCR of ITS 1 for A. mearnsii De Wild, and A. paradoxa DC., respectively; Lanes 4 and 5 show semi-nested PCR of ITS 2 for A. mearn¬ sii and A. paradoxa , respectively; Lanes 6 and 7 show nested PCR of ITS 1+2 for A. mearnsii and A. paradoxa, respectively. specific amplification, and while resulting in a low yield in the first round of amplifica¬ tion because of the unstable nature of the primer and template hybrid (Varadaraj & Skinner 1994), the second round of amplification served to greatly increase the yield of desired product. Despite the high annealing temperature used for PCR, for Acacia a small proportion (approximately 10%) of second round amplifications resulted in non-target templates still being amplified (Fig. 2, lane 9). In these cases it was possible to amplify the Acacia ITS region using semi-nested PCR in the second round of amplification. For Acacia, as described in materials in methods, aliquots of the PCR product from the first round of amplification served as the template for the second round of amplifica¬ tion, and this method was also used for fungal ( Dermocybe and Cortinarius) DNA. Occasionally in Rutaceae this approach resulted in a background of smeared DNA, in addition to the expected product, when visualised on an agarose gel. This problem may be caused by nonspecific amplification in the original PCR (Hengen 1995), but this was unlikely in this case due to the high stringency conditions used. Another common reason for the occurrence of DNA smears is the use of far too much DNA template for the sec¬ ond round of amplification, but as found by Hengen (1995) dilution of the primary PCR product up to 10 000-fold may not resolve the smear into a discrete band after the second round of amplification. Alternatively, a greatly reduced amount of template for the sec¬ ond round PCR can be obtained by band-stab PCR, in which a needle or toothpick is used to “stab” the appropriate DNA band on a gel, and subsequently the needle or toothpick is dipped into the second round PCR reaction mix (Bjourson & Cooper 1992; Kadokami & Lewis 1994). This method achieves the aim of a greatly reduced amount of template for the second round PCR and isolates the desired fragment from the smeared DNA back¬ ground. It is also quicker than cutting the desired band out of the gel and isolating it. Instead of toothpicks or needles, pipette tips were used to stab the band, and then, rather than mixing this directly into the PCR reaction, it was agitated in a separate tube con- 52 F. Udovicic and D .J. Murphy taining 20 pi of water. This allowed storage of this template DNA in a freezer for later use, and enabled multiple second round reactions to be carried out from the one sample. A useful feature of nested PCR is the ability to amplify the ITS region from low con¬ centrations of genomic DNA. As found in this study, this is particularly relevant when using DNA isolated from small amounts of material, such as herbarium specimens, which in addition, typically yield small quantities of degraded genomic DNA (Jansen et al. 1999). Previously, nested PCR has been used with some success for amplification of the rbcL region from DNA isolated from herbarium material (Savolainen et al. 1995). The use of herbarium material allows ready access to many taxa, including rare or even extinct species (Jansen et al. 1999). These specimens are already vouchered and have often been identified by specialists, thus increasing the reliability and repeatability of molecular sys¬ tematic research. The use of herbarium material is time and cost effective, reducing the need to re-collect specimens (Wood et al. 1999). The nested PCR methodology described here may be an effective tool for application in systematic studies using plant and fungal material preserved in herbaria. Acknowledgements We would like to thank Bryan Mole (School of Botany, The University of Melbourne) for applying this technique to Rutaceae, and Rodney Jones (School of Botany, The University of Melbourne) for work on fungi. Joe Miller (CSIRO Plant Industry, Canberra) is thanked for primer design and helpful discussion. We acknowledge ARC and ABRS funding for projects that have led to these findings. References Appels, R. and Dvorak, J. (1982). Relative rates of divergence of spacer and gene sequences with¬ in the rDNA region of species in the Triticeae: Implications for the maintenance of homogeneity of a repeated gene family. Theoretical and Applied Genetics 63, 361-365. Amheim, N. (1983). ‘Concerted evolution in multigene families’, in M. Nei and R. Koehn (eds), Evolution of Genes and Proteins, pp. 38-61. Sinauer Associates, Sunderland: Massachusetts. Amheim, N., Krystal, M., Schmickel, R., Wilson, G., Ryder, O. and Zimmer, E. (1980). Molecular evidence for genetic exchanges among ribosomal genes on nonhomologous chromosomes in man and apes. Proceedings of the National Academy of Sciences USA 77, 7323-7327. Baldwin, B.G., Sanderson, M.J., Porter, J.M., Wojciechowski, M.F., Campbell, C.S. and Donoghue, M.J. (1995). The ITS region of nuclear ribosomal DNA: A valuable source of evidence on angiosperm phylogeny. Annals of the Missouri Botanical Garden 82, 247-277. Bjourson, A.J. and Cooper, J.E. (1992). Band-stab PCR: a simple technique for the purification of individual PCR products. Nucleic Acids Research 20, 4675. Borman, J., Schuster, D., Li, W-B., Jessee, J. and Rashtchian, A. (2000). PCR from problematic templates. Focus 22, 10-11. Brown, G.K., Udovicic, F. and Ladiges, P.Y. (2001). Molecular phylogeny and biogeography of Melaleuca , Callistemon and related genera (Myrtaceae). Australian Systematic Botany 14, 565-585. Chase, M.W. and Hillis, H.H. (1991). Silica gel: an ideal material for field preservation of leaf sam¬ ples for DNA studies. Taxon 40, 215-220. Crisp, M.D., Gilmore, S. and Van Wyk, B-E. (2000). ‘Molecular phylogeny of the genistoid tribes of papilionoid legumes’, in PS. Herendeen and A. Bruneau (eds) Advances in Legume Systematics part 9, pp. 249-276. Royal Botanic Gardens: Kew. Edel, V. (1998). ‘Polymerase Chain Reaction in Mycology: an Overview’, in P.D. Bridge, D.K. Arora, C.A. Reddy and R.P. Elander (eds), Applications of PCR in Mycology , pp. 1-20. CAB International: Wallingford: Oxford. Gardes, M. and Bruns, T.D. (1993). ITS primers with enhanced specificity for basidiomycetes - application to the identification of mycorrhizae and rusts. Molecular Ecology 2, 113-118. Garson, J.A., Tedder, R.S., Briggs, M., Tuke, P., Glazebrook, J.A., Trute, A., Parker, D., Barbara, J.A.J., Contreras, M. and Aloysius, S. (1990). Detection of hepatitis C viral sequences in blood Successful DNA amplification 53 donations by “nested” polymerase chain reaction and prediction of infectivity. Lancet 335, 1419-1422. Hengen, P.N. (1995). Reamplification of PCR fragments. Trends in Biochemical Sciences 20, 124-125. Hillis, D.M., Moritz, C., Porter, C.A. and Baker, R.J. (1991). Evidence for biased gene conversion in concerted evolution of ribosomal DNA. Science 251, 308-310. Jansen, R.K., Loockerman, D.J. and Kim, H-G. (1999). ‘DNA sampling from herbarium material: a current perspective’, in D.A. Metsger and S.C. Byers (eds), Managing the Modern Herbarium - an inter-disciplinary approach , pp. 277-286. Society for the Preservation of Natural History Collections: Washington DC. Kadokami, Y. and Lewis, R.V. (1994). Repeated PCR of a gel band can be used to obtain a single PCR band. BioTechniques 17, 438. Kass, E. and Wink, M. (1997). Molecular phylogeny and phylogeography of Lupinus (Leguminosae) inferred from nucleotide sequences of the rbcL gene and ITS 1+2 regions of rDNA. Plant Systematics and Evolution 208, 139-167. Ladiges, P.Y., McFadden, G.I., Middleton, N., Orlovich, D.A., Treloar, N. and Udovicic, F. (1999). Phylogeny of Melaleuca, Callistemon, and related genera of the Beaufortia suballiance (Myrtaceae) based on 5S and ITS-1 spacer regions of nrDNA. Cladistics 15, 151-172. Martin, P.G. and Dowd, J.M. (1993). Partial sequences of ribosomal RNA of Papilionaceae and related fami li es. Phytochemistry 33, 361-363. Moritz, C. and Hillis, D.M. (1996). ‘Molecular systematics: context and controversies’, in D.M. Hillis, C. Moritz and B.K. Mable (eds), Molecular Systematics (2 nd edn), pp.1-13. Sinauer Associates: Sunderland: Massachusetts. Mullis, K.B. and Faloona, F.A. (1987). Specific synthesis of DNA in vitro via a polymerase-catal¬ ysed chain reaction. Methods in Enzymology 155, 335-350. Saiki, R.K., Gelfand, D.H., Stoffel, S., Scharf, S.J., Higuchi, R., Horn, G.T., Mullis, K.B. and Erlich, H.A. (1988). Primer-directed enzymatic amplification of DNA with a thermostable DNA polymerase. Science 239, 487-491. Savolainen, V., Cuenoud, R, Spichiger, R., Martinez, M.D.P, Crevecoeur, M. and Manen, J-F. (1995). The use of herbarium specimens in DNA phylogenetics: evaluation and improvement. Plant Systematics and Evolution 197, 87-98. Steane, D.A., McKinnon, G.E., Vaillancourt, R.E. and Potts, B.M. (1999). ITS sequence data resolve higher level relationships among the eucalypts. Molecular Phylogenetics and Evolution 12, 215-223. Sun, Y., Skinner, D.Z., Liang, G.H. and Hulbert, S.H. (1994). Phylogenetic analysis of Sorghum and related taxa using internal transcribed spacers of nuclear ribosomal DNA. Theoretical and Applied Genetics 89, 26-32. Thompson, I.R., Ladiges, PY. and Ross, J.H. (2001). Phylogenetic studies of the tribe Brongniartieae (Fabaceae) using nuclear DNA (ITS-1) and morphological data. Systematic Botany 26, 557-570. Udovicic, F. and Ladiges, PY. (2000). Informativeness of nuclear and chloroplast DNA regions and the phylogeny of the eucalypts and related genera (Myrtaceae). Kew Bulletin 55, 633-645. Udovicic, F., McFadden, G.I. and Ladiges, PY. (1995). Phylogeny of Eucalyptus and Angophora based on 5S rDNA spacer sequence data. Molecular Phylogenetics and Evolution 4, 247-256. Varadaraj, K. and Skinner, D.M. (1994). Denaturants or cosolvents improve the specificity of PCR amplification of a G + C-rich DNA using genetically engineered DNA polymerases. Gene 140, 1-5. White, T.J., Bruns, T., Lee, S. and Taylor, J. (1990). ‘Amplification and direct sequencing of fun¬ gal ribosomal RNA genes for phylogenetics’, in M.A. Innis, D.H. Gelfand, J.J. Sninsky and T.J. White, (eds), PCR Protocols: A Guide to Methods and Applications, pp. 315-322. Academic Press, Inc.: New York. Wood, E.W., Eriksson, T. and Donoghue, M.J. (1999). ‘Guidelines for the use of herbarium mate¬ rials in molecular research’, in D.A. Metsger and S.C. Byers (eds), Managing the Modern Herbarium - an inter-disciplinary approach, pp. 265-276. Society for the Preservation of Natural History Collections: Washington DC. Muelleria 16: 55-64 (2002) A systematic study of Acacia calamifolia s.l. , with special emphasis on A. euthycarpa in Victoria Stephen H. Wright *’ 3,5 , James W. Grimes 2 ’ 4 , and Pauline Y. Ladiges 1 1 School of Botany, The University of Melbourne, Vic. 3010 2 National Herbarium of Victoria, Royal Botanic Gardens Melbourne, Birdwood Ave, South Yarra, Vic. 3141 3 Present address: Department of Physiology and Pharmacology, James Cook University, Townsville, Qld 4811 4 Present address: 10 Ormerod Court, Gisbourne, Vic. 3437 5 Author for correspondence: Stephen.Wrightl@jcu.edu.au Abstract Taxonomic revision of the Acacia calamifolia Sweet ex Lindl. complex shows that A. euthycarpa (J.M.Black) J.M.Black is distinct, and that the two species have allopatric distributions. Typical A. calamifolia has moniliform legumes, narrowly linear phyllodes, united sepals, and an elongated funicle half-encircling the seed. It occurs in New South Wales, and in the Flinders Ranges and North Mount Lofty Ranges of South Australia. Acacia euthycarpa has linear legumes, narrowly lin¬ ear to oblanceolate phyllodes, free to united sepals, and a longer funicle that entirely or nearly entirely encircles the seed. It occurs in Victoria, on the Eyre Peninsula, Kangaroo Island, South Mount Lofty Ranges, and the Murray Lands of South Australia. Multigroup discriminant function analysis of A. euthycarpa using phyllode characters resolved two subspecies. Acacia euthycarpa subsp. euthycarpa has narrowly linear phyllodes. Acacia euthycarpa subsp. oblanceolata Stephen H. Wright, is newly described to accommodate populations from only two localities in Victoria and three in South Australia which have narrowly oblanceolate to oblanceolate phyllodes. A key to dis¬ tinguish the new subspecies and the morphologically similar taxa is presented. Introduction Acacia calamifolia Sweet ex Lindl., as currently understood, has a disjunct distribution throughout much of south-eastern Australian in semi-arid regions with a mean annual rainfall of 150-550 mm. It occurs on a range of soils in open scrub or woodland/open woodland communities and on rocky outcrops. Acacia calamifolia s.l is generally a rounded to Y-shaped, multistemmed shrub 2-4 m high, but occasionally grows to a small, single-stemmed tree to 10 m high in parts of west-central Victoria. It generally has nar¬ row phyllodes 1-5 mm wide, with four main veins (one main vein per face) and a dis¬ tinctive, curved tip. It has legumes that are linear to moniliform, and an elongated funi¬ cle that half to entirely encircles the seed. Acacia calamifolia var. euthycarpa J.M.Black (Black 1923) was distinguished from A. calamifolia var. calamifolia by its linear legume. Later, when elevating the taxon in rank to A. euthycarpa (J.M.Black) J.M.Black, Black (1945) again made the comparison in shape of the legume, but added that the phyllode of A. euthycarpa had three nerves. In spite of these differences, the name has long been considered synonymous with A. calam¬ ifolia (e.g. Whibley 1986; Entwisle et al. 1996; Maslin 2001). The complex that includes A. calamifolia and A. euthycarpa has been considered extremely variable (Whibley 1986; Entwisle et al. 1996). A variant of A. calamifolia with broad, oblanceolate phyllodes occurs from South Australia to west-central Victoria (Willis 1972; Entwisle et al. 1996). This variant differs from typical A. calamifolia not only in size and shape of phyllodes, but also by habit, often becoming a substantial tree to 10 m, and with a funicle almost entirely encircling the seed (Willis 1972). These populations have often been erroneously known as A. microcarpa F.Muell. var. linearis J.M.Black (e.g. Whibley 1986). 56 S.H. Wright et al. A taxonomic revision of A. calamifolia s.l. is given in the following. A morphometric analysis of phyllodes clarifies the taxonomy of broad and narrow phyllode forms in Victoria and South Australia, and supports the recognition of two distinct species, as well as a new subspecies of A. euthycarpa. Methods Ten populations were sampled in the field for a multivariate analysis of phyllode charac¬ ters, which are the most variable features of these populations. Legumes and seeds were not available for a quantitative comparison of populations, but were described from both field observations and from herbarium material. Originally flower characters were also measured, but not completed, as they exhibit almost no variation in size. Differences in fruit size were not measured for lack of adequate sample size. Field sampling was restricted to Victoria because nearly all of the variation in phyllode characters was repre¬ sented in populations from that State. Four populations were sampled from the Wimmera and six from west-central Victoria (Fig. 1). Twenty individuals were sampled across the range of each population, subjectively from small populations (less than + 50 individu¬ als) and along a walk-transect for larger populations. One typical mature phyllode was selected from each pressed specimen. Only one phyllode per individual was measured since measuring more phyllodes per individual was found to give no better estimate of plant population means (for those characters included in the analysis) and no less variances. The following phyllode meristic charac¬ ters were scored: total length, length of pulvinus, width at widest point, and thickness at widest point. Vein number was noted for each population. Multigroup discriminant function analysis (MDA) was chosen to analyse the phyllode characters because it is useful for comparing variability within and between populations. This analysis uses weighted predictor variables (i.e., weighted phyllode characters), in the form of linear functions, to provide the best discrimination between predetermined groups, in this case plant populations (Tabachnick & Fidell 1989). Each individual phyllode within a group obtains a discriminant value for each function based on its values for predictor vari¬ ables. Individuals may be classified into populations using probabilities based on the dis- Figure 1. Distribution of A. calamifolia s.l in Victoria, with populations sampled for morphometric analysis numbered 1-10. Acacia calamifolia 57 criminant scores. The analysis also inf ers the relative contributions of each predictor vari¬ able to the classification. Variables were transformed, when appropriate, to meet the assumptions of multivariate normality and equality of variance-covariance matrices. Following analysis of the wild populations, phyllodes measured from herbarium spec¬ imens at MEL were classified by the MDA to determine their relationship with the pop¬ ulations of phyllodes sampled in the field. All statistical methods were performed on a PC computer using the SPSS Windows package (S.P.S.S. 1990). Results Two discrete groups were separated by the first discriminant function in the MDA (Fig. 2), which accounted for 96% of the between-population variation in phyllode characters. The transformed variable log (width) was highly correlated (0.80) with this first function, inferring that width was the dominant character for separating populations (Table 1). Of the two groups separated, one had narrow phyllodes, averaging 1.1 mm wide (population 1-8), and the other had broad phyllodes, averaging 3.3 mm wide (populations 9-10 from Wychitella and Yowang Hill, respectively, in west-central Victoria). The means for the untransformed phyllode characters in the narrow and broad groups are shown in Table 2. Table 1. Correlation between phyllode variables and the first two discriminant functions of the MDA. Function 1 Function 2 Log (width) 0.80 0.13 Log (thickness) -0.23 0.65 Length -0.06 -0.60 Log (pulvinus) -0.05 -0.20 Table 2. Means of phyllode characters for the two groups resolved by MDA. Standard Deviations are in parentheses. Narrow phyllode group Broad phyllode group Width (mm) 1.1 (0.26) 3.3 (0.53) Thickness (mm) 0.48 (0.07) 0.37 (0.05) Length (mm) 46.7 (9.1) 40.4 (6.9) Pulvinus length (mm) 1.53 (0.31 1.65 (0.21) Within the narrow phyllode group (Fig. 2) populations 1, 2 and 3 from the Little Desert National Park in the Wimmera, and 4 and 5 from near Inglewood in west-central Victoria, have very narrow phyllodes, averaging 0.9 mm. In comparison, populations 4 from Mt Arapiles in the Wimmera, and 7 and 9 from near Wychitella and Wedderbum respectively, in west-central Victoria, are characterised by less narrow phyllodes, averaging 1.3 mm broad. The difference in means of phyllode-width between the populations is not consid¬ ered useful in delimiting taxa, especially since no other character distinguishes them. There was no obvious pattern of separation along the second function (Fig. 2), which accounted for only 3% of the between-population variation. The variables length and log (thickness) were the most correlated (-0.60 and 0.65 respectively) with this function (Table 1). By convention, only correlations between variables and functions that were in excess of 0.30 (9% of variance) were interpreted (Tabachnick & Fidell 1989). Discriminant Function 2 58 S.H. Wright et al. Figure 2. Plot of discriminant function values for phyllodes from individuals from the ten populations sampled in Victoria, and values for the population means. Individuals from populations 1-8, which comprise the narrow-phyllode group, are represented by filled circles; individuals from populations 9-10, which comprise the broad-phyllode group, are represented by open triangles; popula¬ tion means are presented by their population number in shaded circles. Figure 3. Plot of discriminant function values for phyllodes from individuals from the ten populations sampled in Victoria, values for the population means, and val¬ ues for phyllodes measured from herbarium specimens from across the geo¬ graphic range of Acacia calamifolia s.l. Individuals from populations 1—8, which comprise the narrow-phyllode group, are represented by filled circles; individuals from populations 9-10, which comprise the broad-phyllode group, are represented by open triangles; population means are presented by their population numbers in shaded circles; individual herbarium specimens are represented by open squares. Acacia calamifolia 59 Discriminant function values for phyllode variables from herbarium specimens fell within, or near, the range from those sampled in the field in Victoria (Fig. 3). Discussion THE SPECIFIC CIRCUMSCRIPTION OF ACACIA CALAMIFOLIA AND A. EUTHYCARPA As legume type is generally considered constant with species of Acacia (New 1984), the differences in shape between the pods of A. calamifolia s.s. and A. calamifolia var. euthy- carpa, in conjunction with differences in seed characters, provides evidence that A. calamifolia and A. euthycarpa are distinct species. Furthermore, A. euthycarpa generally has two to four capitula per raceme and a rachis length of 1-8 mm. Based on examination of herbarium material, greater variation was detected in A. calamifolia s.s., with two to eight capitula per raceme and a rachis length of 5-25 mm. Sepals are united in A. calamifolia s.s, and on specimens of A. euthycarpa from Kangaroo Island and South Mount Lofty Ranges. Sepals are free on specimens of A. euthy¬ carpa from other areas, suggesting geographic variation in sepal-fusion in this species. Although nervature of the phyllode was not mentioned in the protologue of A. calamifo¬ lia var. euthycarpa (Black 1923), A. euthycarpa was described as having 3-nerved phyllodes (‘sed phyllodiis arete trinervibus,’ Black 1945: 310), and on a note on one of the types (AD 97333008 ) J.M. Black wrote ‘phyllodes... about 3-nved (3-striate) on each face, obscurely 3-nvd.’ This character would distinguish it from A. calamifolia. In reality this is incorrect; the phyllodes of A euthycarpa are four nerved, with one nerve on each face, as in A. calamifo¬ lia. With age each face of the phyllode becomes faintly striate due to dessication. Care must be taken to distinguish true nerves from striations caused by long-term drying. Figure 4. Distribution map of taxa resolved within the Acacia calamifolia species com¬ plex, showing geographical regions. Distribution of A. calamifolia s. s. is shown by hatched areas; distribution of A. euthycarpa subsp. euthycarpa is shown by shaded areas; localities of A. euthycarpa subsp. oblanceolata is shown by filled circles. Geographical regions are numbered; Eyre Peninsula (1); Flinders Ranges (2); North Mount Lofty Ranges (3); South Mount Lofty Ranges (4); Murray Lands (5); Kangaroo Island (6); Wimmera (7); Mallee (8); west-central Victoria (9); Swan Hill district (10); Griffith-Cobar region (11); Broken Hill district (12). 60 S.H. Wright et al. Acacia calamifolia s.s. and A. euthycarpa have allopatric distributions. Acacia calamifo- lia s.s. occurs in New South Wales, the Flinders Ranges and the North Mount Lofty Ranges of South Australia (Fig. 4). It grows on shallow powdery calcareous loams, grey-brown cal¬ careous earths, and red duplex soils. Acacia euthycarpa occurs in Victoria, on the Eyre Peninsula, Kangaroo Island, South Mount Lofty Ranges, and the Murray Lands of South Australia (Lig. 4). It grows on bleached to brownish sands, grey-brown calcareous earths, and mottled-yellow to red duplex soils. Both species grow in scrub or as an understorey species in woodland or open woodland, often on rocky sites. They both grow in a range of rainfall conditions from 150-500 mm mean annual rainfall (arid to moderately arid conditions). PHYLLODE LORMS IN VICTORIA - TWO SUBSPECIES OF A. EUTHYCARPA The MDA distinguished two forms wit hin A. euthycarpa : the broad-phyllode and narrow- phyllode forms. Acacia euthycarpa subsp. euthycarpa is referrable to the narrow-phyl- lode form, while the name A. euthycarpa subsp. oblanceolata is proposed for the broad- phyllode form. The mean phyllode width of A. euthycarpa subsp. oblanceolata (3.3 mm) is about three times that of A. euthycarpa subsp. euthycarpa (1.1 mm). There are also lesser differences in the means of the other characters, with A. euthycarpa subsp. oblanceolata tending to have shorter, thinner phyllodes with longer pulvini and a shorter distance to the widest point. Some of the important differences are illustrated in Figure 5. Acacia euthycarpa subsp. oblanceolata is geographically widespread over a range of approximately 850 km, but is known only from two localities in Victoria. It should be pointed out that A. euthycarpa is much more common in Victoria than A. calamifolia (A. calamifolia is known to us only from few, very old collections), and that the description of the latter provided by Willis (1972) applies to A. euthycarpa. The phyllodes of A. x gray ana are very similar to those of A. euthycarpa subsp. oblance¬ olata, and the two taxa are difficult to tell apart in flower. However, A. x grayana has sub- moniliform fruit as well as pubescent new shoots and peduncles (Entwisle et al. 1996). Acacia euthycarpa subsp. oblanceolata grows on grey-brown calcareous earths, mot¬ tled-yellow to red duplex soils, and sometimes sand. It occurs in areas of mean annual rainfall ranging between 300 and 500 mm, suggesting that it has a lower tolerance to arid conditions than A. euthycarpa subsp. euthycarpa (range 150-500 mm). Localities of A. euthycarpa subsp. oblanceolata are on the periphery of the distribution of A. euthycarpa subsp. euthycarpa, which may suggest a different ecological preference, or it may be a pattern associated with agricultural clearing. Key to taxa 1. Legumes (submoniliform to) moniliform; seed 6-9 mm long, the funicle encircling the seed by about '/2 the circumference; phyllodes 0.7-4.5 mm wide, if 2 mm wide or wider, then new shoots and peduncles subglabrous to covered in hairs 2. Phyllodes 0.7-1.2 mm wide . 1. Acacia calamifolia 2. Phyllodes 2-4.5 mm wide. 2. A. x grayana 1. Legumes linear; seed 4-6 mm long, the funicle entirely encircling the seed; phyllodes 0.7-6.0 mm wide, if 2.0 mm or wider, the new shoots glabrous to glabrate 3. Phyllodes 0.7-2.0 mm wide, narrowly linear. .3A. A. euthycarpa subsp. euthycarpa 3. Phyllodes 2.5-6 mm wide, narrowly oblanceolate. .3B. A. euthycarpa subsp. oblanceolata 1. Acacia calamifolia Sweet ex Lindl., Bot. Reg. 10, t. 389 (1824). Type citation : ‘brought by Mr John Richardson, to Mr. Colvill, from the south-west interior of New Holland.” Type not seen, descripition (‘ Legumina arcuata articulata...’) and illustration decisive. Acacia pulverulenta A. Cunn. ex Benth., London J. Bot. 1: 342 (1842), nom. illeg., non Acacia calamifolia 61 A. pulverulenta Schltdl. (1838). Syntypes : interior of New Holland [between the Loddon R. and Pyramid Hill, Vic.], 8 July 1836, T.S. Mitchell “230”; Mt Flinders [one of the peaks adja¬ cent to L. Brewster which is c. 10 km S of Lachlan R. and c. 130 km N of Griffith, N.S.W.], June 1817, A. Cunningham 403 (not found). Neither of these syntypes of A. pulverulenta were located at K, and the Cunningham collections could not be found at MEL. The only syntype seen by us is the one collected on MitchelTs journey, with the number ‘230.’ Shrub 2-4 m high, plants glabrous throughout. Phyllodes narrowly linear, 4-10 cm long, 0.7-1.2 mm wide, 0.4-1.0 mm thick, terete to flat, green to grey-green, somet im es scurfy, shortly acuminate with delicate, curved apex; main longitudinal veins four in all, not promi¬ nent and the mid-veins often somewhat impressed; pulvinus 1-2 mm long; the small obscure gland inserted 1-10 nun above pulvinus. Inflorescences of simple axillary capitula, or of 2-8(-14)-headed racemes, rachis 5—25(—40) mm long, peduncles 4-10 mm long; the capit¬ ula with 25^40 golden yellow flowers, these 5-merous; calyx-lobes all united, corolla-lobes all free. Pods moniliform, woody to coriaceous, wrinkled, straight, curved or twisted, to 16 cm long, 3-6 mm wide, grey to brown, ± glaucous; seeds longitudinal in pod, oblong to ellip¬ tic in outline, 6-9 mm long, 2.5^1.5 mm wide, dull to slightly shiny, dark brown to black, the filiform funicle mostly 1-folded, encircling half of seed, aril clavate. Habitat and Distribution: Found mostly in scrub and sometimes in woodland or open woodland, occasionally on rocky sites; mostly on shallow powdery calcareous loams, grey-brown calcareous earths, or red duplex soils. Flinders Ranges (SA), North Mount Lofty Ranges (SA), Griffith-Cobar region (NSW) and near Broken Hill (NSW). Phenology : Flowers mostly September to November, but often found sparsely flow¬ ering throughout the year. Selected specimens examined : SOUTH AUSTRALIA: Flinders Ranges, near Wilpena Pound, ca. 45 km NNE of Hawker, R. Hill 6, 14 Jul 1955 (MEL); Northern Flinders Range, Northern Chace’s Range, rocky banks of Anginoonor Creek, ca. 30 km NE of Hawker, D.N.Kraehenbuehl 310 , 12 June 1961 (MEL); Oraparinna Natl Park, near head¬ quarters, J.Z.Weber 2648, 19 Sep 1971 (AD, MEL); Morgan to Eudunda road, 2.8 km WSW of Sutherlands, NW side of road, F.E.Davies 1391, 22 Nov 1989 (AD, MEL). 2. Acacia x grayana J.H.Willis, Victorian Naturalist 73: 155 (1957). Type citation : ‘Wraigworm Parish, south of Kiata and about 14 miles east of Dimboola. A.J.Gray s.n., 9.IX.1951.’ (holotype MEL 1500364, also on the sheet, a sterile specimen of A. x grayana, same locality as type, leg. 3/1951; and a fruiting specimen collected from a tree raised from seed off the type at Wail Nursery, leg. 3/11/1953; also a specimen of A. hilliana and a spec¬ imen of A. euthycarpa subsp. oblanceolata, placed there ‘for comparison’ (J.H.Willis).) Acacia microcarpa F.Muell. var. linearis J.M.Black, Flora of South Australia 4: 687 (1920). Type citation : ‘Near Monarto South’ (holotype AD). Shrub or small tree to 3 m tall, glabrous save for the glabrate to densely appressed hairy peduncles and new shoots. Phyllodes elleptic-oblanceolate to broadly oblanceolate, 2.5-5.75 mm long, 2.0-4.5 mm wide, 0.25-0.8 mm thick, flat, dull green to brown-green, acuminate with a curved apex; longitudinal nerves four, one on each broad face of the phyllodes, the two lateral ones appearing as marginal nerves, often becoming more or less irregularly striate with dessication; pulvinus 0.75-1.25 mm , the obscure to subprominent gland inserted 4-9 mm above pulvinus. Inflorescences of solitary capitula or short racemes of 2-5-capitula; rachis 2-7 mm long, peduncles 4-12 mm long; the capitula with 20-35 flowers, these 5-merous; calyx-lobes free to the base, corolla-lobes free. Pods (sub)moniliform, to 7(-7.5) cm long, 6 mm wide; seeds longitudinal in pod, + 6 mm long, the funicle encircling about half the seed. Habitat and Distribution : In sandy soils, often where moist, and often with Callitris. 62 S.H. Wright et al. Known from north-western Victoria near Wyperfeld National Park and the Little Desert National Park. Phenology : Blooming as early as June, but mostly through November. Selected specimens examined : VICTORIA: Wimmera. Winiam, S of Nhill. I.O.Maroske s.n., 11 Oct 1992 (MEL); west of Yarto, just within the eastern boundary of Wyperfeld Nat’l Park, I.O.Maroske s.n., 25 Jun 1960 (MEL); V 2 mile [0.8 km south of Kiata Store Kiata, E.M.Canning 2981, 11 Oct 1969 (CANB, MEL). 3. Acacia euthycarpa (J.M.Black) J.M.Black, Trans. & Proc. Roy. Soc. South Australia 69: 310 (1945) (Typus infra sub subsp. euthycarpa indicatur). Shrub 2-4 m high, or occasionally a small tree to 10 m high; plants glabrous throughout. Phyllodes narrowly linear, narrowly oblanceolate to oblanceolate, 3-8 cm long, 0.7-6.0 mm wide, 0.3-0.8 mm thick, terete to flat, green to grey-green, sometimes scurfy, shortly acumi¬ nate with delicate, curved apex; main longitudinal veins four in all (one per face), not promi¬ nent and the mid-veins often somewhat impressed; pulvinus 1-3 mm long; the small obscure gland inserted 0-7 mm above pulvinus. Inflorescences of simple axillary capitula, or of 2-4(-6)-headed racemes; rachis l-8(-14) mm long, peduncles 4-10 mm long; the capitula with 25-40 golden yellow flowers, these 5-merous; calyx-lobes free or united, corolla-lobes all free. Pods linear, coriaceous to crustaceous, smooth or nearly smooth, straight, curved or twisted, to 16 cm long, 3-6 mm wide, brown; seeds longitudinal in pod, oblong to elliptic in outline, 4-6 mm long, 2.5—4 mm wide, dull to slightly shiny, dark brown to black, the fili¬ form funicle mostly 2-3-folded, entirely encircling the seed, aril clavate. 3A. Acacia euthycarpa (J.M.Black) J.M.Black subspecies euthycarpa. Acacia calami- folia var. euthycarpa J.M.Black, Trans. & Proc. Roy. Soc. South Australia 47: 269 (1923), 5 . 5 . Type citation : ‘Southern districts: Yorke [sic] Peninsula, Kangaroo Island, Eyre Peninsula.’ (lectotype, here designated, AD 97333008, the fruiting specimen on the right- hand side of the sheet labelled ‘Barossa Ranges, Nov. 1912;’ syntype, also on AD 97333008: labelled ‘Amo Bay [Eyre Peninsula],’ sterile spec.; but excluding a specimen labelled ‘Nurioota’ which is part of the Mount Lofty Ranges, not the York Peninsula; AD 97333011, labelled ‘Port Lincoln [Eyre Peninsula]’; AD 973330006 ). Phyllodes narrowly linear, 3-8 cm long, 0.7-2.0 mm wide, 0.3-0.8 mm thick, terete to flat; the gland inserted 0-7 mm above pulvinus. Habitat and Distribution: Found mostly in scrub, woodland or open woodland, occasionally on rocky sites; mostly on bleached to brownish sands, grey-brown calcare¬ ous earths, and mottled-yellow to red duplex soils. Common in west-central Victoria, western Victoria, Kangaroo Island (SA), Eyre Peninsula (SA), South Mount Lofty Ranges (SA) and the Murray Lands (SA). Phenology: Flowers mostly August to November. Selected specimens examined: VICTORIA: 8.2 km S of Wychitella on Wychitella- Wedderbmn Road, S. Wright 13 and J. Grimes, 7Feb 1998(AD, BRI, CANB, MEL, MELU, PERTH); State Park west of Inglewood, comer of Barry Rock Road and the road from the Logan-Inglewood Road to Melville Caves, S. Wright 14 and J. Grimes, 7 Feb 1998 (AD, CANB, MEL, MELU, PERTH); 8.6 km west of Inglewood on the Logan-Ingelwood road, S. Wright 15 and J. Grimes, 7 Feb 1998 (MEL); 5.5 km SE of Wedderbum, Calder Highway between Wedderbum and Inglewood, J. Connock 348, 12 Sep 1992 (AD, MEL); Summit of Mt Arapiles, Mount Arapiles State Park, P.G. Abeel 525 and C. Herscovitch, 17 Dec 1986; 21.9 km N of Kaniva on the Broughton Road, J. Grimes 3434 and B. Meurer-Grimes, 13 Aug 1996 (AD, BH, CANB, MEL, NSW, NY); Wimmera. Lawlot Range, on Western Highway, P.C. Jobson 3704, 27 Aug 1995 (AD, BRI, CANB, MEL, NSW). Acacia calamifolia 63 Figure 5. Acacia euthycarpa subsp. oblanceolata, and comparative illustrations of A. euthycarpa subsp. euthycarpa, and A. calamifolia. A-B. A. euthycarpa subsp. oblanceolata. A habit x 0.7; B phyllode x 1 (A from S. Wright 18; B from M.G. Corrick 5383); C-E. A. euthycarpa subsp. euthycarpa. C phyllode x 1; D fruit x 1 and E seed x 3 (from Maroske s.n., Dec. 1966). Acacia calamifo¬ lia F-H. F phyllode x 1; G fruit x 1; H seed x 3 (from R. Filson 3488). 64 S.H. Wright et al. 3B. Acacia euthycarpa subspecies oblanceolata Stephen H. Wright, subsp. nov. a A calamifolio legumine lineari diversa; a A. euthycarpa subsp. euthycarpa phyllo- dio latiore (2.5-6 nee 0.7-2.0 mm) diversa (Fig. 5a-e) Typus: VICTORIA, Yowang Hill, northeast of St Arnaud, 36°29’S 143° 22’E, S. Wright 18 (holotype MEL 2058890; isotypes AD, BRI, CANB, MELU, NSW, PERTH). Phyllodes narrowly oblanceolate to oblanceolate, 3-6 cm long, 2.5-6.0 mm wide, 0.3-0.5 mm thick, flat, the gland inserted 0-5 mm above pulvinus. Habitat and Distribution : Found in scrub or open woodland, often on rocky sites; on grey-brown calcareous earths and mottled-yellow to red duplex soils, sometimes on sand. Rare, occurring at only six known localities: Wychitella Flora and Fauna Reserve and Yowang Hill (west-central Victoria), Gawler Ranges and near Kima Eyre Peninsula (SA) and near Murray Bridge (the Murray Lands SA). The specimen near Murray Bridge was collected in 1848 and this subspecies may no longer occur there. Phenology. Flowers mostly August to October. Remarks : Appears similar to A. x grayana, a hybrid between A. euthycarpa subsp. euthycarpa and A. brachybotrya, occurring near Kiata in the Little Desert, but may be dis¬ tinguished from A. x grayana by its glabrous new shoots and peduncles and rectilinear pod. Conservation status: A ROTAP code of 3R is proposed for Victoria. The subspecies is fairly widespread but occurs in relatively isolated populations. Though on both crown land and in reserves, in many places it is threatened by sheep grazing. Selected specimens examined : SOUTH AUSTRALIA, Eyre Peninsula: S side of Eyre Highway, 30 km W of Port Augusta, P.C. Jobson 2726, 28 Oct 1993 (AD, MEL); c. 65 km W of Kimba, c. 27 km east of Kyancutta, T.R.N. Lothian 5409, 10 Oct 1986 (BRI, MEL, PERTH); 2.9 km south of Wychitella, on Wychitella-Wedderburn road, S. Wright 12 and J. Grimes, 7 Feb 1998 (CANB, MEL 2058891, MELU, PERTH). Acknowledgments Research on the phylogeny of Acacia by PYL and JG has been supported by ABRS. Their support is gratefully ackowledged. We thank Mali Moir for the accompanying line drawing. References Black, J.M. (1923). Additions to the flora of South Australia, No. 21. Transactions and Proceedings of the Royal Society of South Australia 47, 367-370. Black, J.M. (1945). Additions to the flora of South Australia. Transactions and Proceedings of the Royal Society of South Australia 69, 309-310. Brooks, D.R. and McLennan, D.A. (1991), Phylogeny, Ecology and Behaviour. A Research Program in Comparative Biology. The University of Chicago Press: Chicago. Court, A.B. (1972). Acacia Mill.’, in Willis, J.H., A Handbook to Plants in Victoria. Melbourne University Press: Melbourne. Entwisle, T.J., Maslin, .R. and Court, A.B. (1996). Acacia' 1 , in N.G. Walsh and T.J. Entwisle (eds), Flora of Victoria 3, 586-656. Inkata Press: Melbourne. Leach, G.J. and Whiffin, T. (1978). Analysis of a hybrid swarm between Acacia brachybotrya and A. calamifolia (Leguminosae). Botanical Journal of the Linnaean Society 76, 53-69. Maslin, B.R. (2001). Acacia calamifolia’’ , in A.E. Orchard and A.J.G. Wilson (eds.), Flora of Australia 11A, 268-269. Australian Government Publishing Service: Canberra. New, T.R. (1984). A Biology of the Acacias. Oxford University Press: Melbourne. S.P.S.S. Statistical Data Analysis Package (1990). Base System User’s Guide. S.P.S.S. Inc.: Chicago, Illinois. Tabachnick, B.G. and Fidell, L.S. (1989). Using Multivariate Statistics. Harper ad Row: New York. Whibley, D.J.E. (1986). ‘Subfamily Mimosoideae’, in J.P Jessop and H.R. Toelken (Eds), Flora of South Australia 2, 511-561. South Australia Government Printing Division: Adelaide. Muelleria 16: 65-70 (2002) Agyrium Fr., Bryophagus Nitschke ex Arnold and Racodium Fr., lichen genera previously unrecorded for Australia Gintaras Kantvilas Tasmanian Herbarium, GPO Box 252-04, Hobart, Tas. 7001. gkantvilas@tmag.tas.gov.au Abstract Agyrium rufum (Pers.) Fr. (Agyriaceae), Bryophagus minutissima (Vezda) D. Hawksw. (Gyalectaceae) and Racodium rupestre Pers. (incertae sedis ) are recorded from Tasmania, repre¬ senting the first reports of these lichen genera for Australia. Morphological and anatomical data, as well as information on the distribution and ecology of each species is presented. Introduction As a result of ongoing research on the Australian lichen flora, new records for the whole continent and especially for individual States continue to be reported; for example, see the ongoing series of papers in the journal Australasian Lichenology. However, most such records concern species, and new generic records are relatively infrequent. Studies on crustose lichens from Tasmania, and comparisons with authentic reference material, mainly from the Northern Hemisphere, have led to the identification of three genera pre¬ viously unrecorded for Australia. Two, Agyrium Fr. and Racodium Fr., are cool temper¬ ate taxa and further demonstrate the strong floristic links that exist between Tasmania and the temperate, oceanic regions of the Northern Hemisphere. The third, Bryophagus Nitschke ex Arnold, has a more scattered distribution. Methods The work is based on the author’s collections from Tasmania, held in the Tasmanian Herbarium (HO) and on comparative material in the Natural History Museum, London (BM), the Royal Botanic Garden, Edinburgh (E) and the Karl-Franzens-Universitat, Graz (GZU). Observations and measurements of apothecial tissues and ascospores were made on hand-cut sections and squashes mounted in dilute KOH solution. Agyrium Fr. The genus Agyrium is characterised by a poorly developed, immersed thallus, apothecial ascomata with a reduced annular exciple, richly branched paraphyses, eight-spored, Trapelia-lype asci and simple, non-halonate, thin-walled, ellipsoid ascospores (Dennis 1981; Purvis & James 1992a; Lumbsch 1997). Although non-lichenised in the strictest sense, the mycelial hyphae are frequently associated with and penetrate green coccoid algae, especially in the vicinity of the apothecia, a condition described by Lumbsch (1997) as ‘facultative parasitism’. Thus Agyrium superficially looks like a lichen, and its closest relatives are all lichens, including, in the Australian lichen flora, Lithographa Nyl., Placopsis (Nyl.) Lindsay, Placynthiella Elenkin, Rimularia Nyl., Trapelia M.Choisy, Trapeliopsis Hertel & G.Schneider and Xylographa (Fr.) Fr. These and some additional genera not known from Australia comprise the family Agyriaceae and are dis¬ cussed extensively in the comprehensive paper of Lumbsch (1997). Although several taxa have been included in Agyrium in the past, many of these have been transferred formally to unambiguously non-lichenised, unrelated genera. Only one species, A rufum (Pers.) Fr. is currently recognised in the Northern Hemisphere (see Dennis 1981 and Purvis & James 1992a for descriptions) and occurs widely in Europe, 66 G. Kantivilas mainly on dead wood. This species is here recorded from Tasmania, representing the first record of the genus for Australasia. Two additional species of Agyrium are known from southern South America: A. antarcticum Rehm, reported from rotting Nothofagus, and A. chilense Speg., from rotting Lobelia. Agyrium rufum (Pers.) Fr., Systema Mycologicum 2: 232 (1822); Stictis rufa Pers., Observationes Mycologicae 2: 74 (1799). Type: n.v. Thallus immersed to absent, usually defined by pale, bleached patches of the substratum. Algal cells occasional, globose to broadly ellipsoid, 4-7 x 4-6 pm, mostly occurring around or beneath the apothecia. Apothecia scattered, pale orange to orange-red to red- brown, waxy or matt, convex, irregularly roundish, sometimes rather wrinkled or con¬ torted, immarginate, 0.2-0.5 mm wide, 0.12-0.2 mm thick. Exciple (in section) very reduced, 20-40 pm thick, pale to deep orange-brown, K+ orange-red to red, sometimes fleetingly or soon becoming + persistent pale yellow. Subhymenial tissues poorly differ¬ entiated, sometimes also with irregular patches pigmented as above. Hymenium 70-80 pm thick, intensely 1+ blue, with an orange-brown pigmented epithecial layer c. 10 pm thick, sometimes uneven and extending deeper into the hymenium; reaction in K as above. Asci broadly clavate, 60-75 x 12-22 pm, of the typical Trapelia-type: outer wall amyloid; apex broadly rounded; tholus thick and well-developed, with a thin amyloid cap and weakly amyloid flanks, ± non-amyloid in the remainder; ascoplasm truncate to con¬ cave at the apex, without an ocular chamber (Fig. 1A; also see Rambold & Triebel 1990). Paraphyses richly branched, c. 1 pm thick, with somewhat swollen apices 1.5-2 pm thick. Ascospores 12-8 x (5-)6-8(-9) pm, broadly ellipsoid or egg-shaped, hyaline at first, becoming grey to deep brown when old, typically with one or several vacuoles. Remarks: The above description is based on the Tasmanian specimens cited below. Their ascospores are somewhat larger than commonly seen in European material, although there is no major difference between the two populations. Brown spores are quite common in the Tasmanian material but were rarely seen in specimens from Europe. Similarly the pigmentation of apothecial tissues (and consequent K reaction) is more intense in Tasmanian specimens. These differences are likely to relate to the age of the apothecia and the degree of exposure of the microhabitat. All Tasmanian specimens fall within the concept of A. rufum as accepted by Northern Hemisphere workers, but also display some critical morphological and ecological differ¬ ences between them. One specimen is from alpine heathland where Cetraria australien- sis W.A.Weber ex Karn. and species of Cladia Nyl., Cladonia Hill ex P.Browne and Cladina Nyl. predominate. It grew on the dead, decorticated, bleached twigs of an unknown shrub, possibly a member of the Epacridaceae, associated with several other unusual crustose lichens including species of Xylographa (Fr.) Fr. and Japewia Tpnsb. It has rather small, orange-red, waxy apothecia and is identical to European material, espe¬ cially that which occurs on dead, bleached Calluna (Ericaceae) stems. Specimens from decorticated, bleached lignum of eucalypt logs in wet forests are virtually identical to European specimens from conifer lignum. The discovery of Agyrium in this habitat, espe¬ cially as a pioneer in post-logging regeneration in forestry coupes, suggests the species may be quite widespread in Tasmania and has been previously overlooked. The specimen from cool temperate rainforest, growing on the bark of a very old, dry trunk of a mature Nothofagus cunninghamii tree, is unusual: the apothecia are a very deep reddish brown, and some of the younger apothecia have a scattered orange pruina; both young and old fruiting bodies are deeply pigmented within, and the K+ red reaction is very intense and persistent, especially in the excipulum; brown spores predominate in this specimen. These characters are rather extreme in the context of all the material studied, but until Agyrium, Bryophagus and Racodium 67 more material is found, the specimen is included within A. rujum. It is rather similar to British material of Lecidea grumosa Leight. (=A. rufum ) which is also from bark and is rather intensely pigmented within. Authentic material of the South American species could not be located for compari¬ son, but on the basis of descriptions both differ clearly from A. rufum on spore size alone. In A. antarcticum the spores are 10-12 x 8-10 pm and clearly broader than those of A. rufum (Rehm 1899), whereas in A. chilense they are 10-11 x 2-2.5 pm (Saccardo & Trotter 1913) and narrower. Australian specimens examined: Tasmania: Arthur River, 250 m altitude, on Nothofagus cunninghamii in rainforest, 13.ii. 1982, G Kantvilas s.n. (HO); summit of Wild Dog Tier, 41°47’S 146°35’E, 1390 m altitude, on dead wood in alpine heathland, 11 .iii.2001, G. Kantvilas 375/01 (BM, HO); track to Mother Cummings Peak, 41°41’S 146°32’E, 1150 m altitude, on decorticated eucalypt wood in subalpine scrub, 3.iii.2002, G. Kantvilas 147/02 (HO); west of Tahune Bridge, ‘Small Coupe’, 43°06’S 146°42’E, 100 m altitude, on decorticated bleached eucalypt log in regenerating logging coupe, 19.ii.2002, G. Kantvilas 171/02 (GZU, HO). Selected comparative material examined: United Kingdom: West Ross (V.C. 105), Kinlochewe, Beinn Eighe NNR, Coille na Glas-leitir, c. 100 m altitude, on lignum of fall¬ en, decorticate Pinus, 12.iv.2001, B.J. and AM. Coppins s.n. (HO); East Ross (V.C. 106), Amat Forest, on Betula lignum, 28.V.1975, B.J. Coppins 2224 & F. Rose (E); Perthshire, Tulloch, near Killiecrankie, on decorticate Calluna twigs in burn-out, 24.vi. 1972, R. Watling (E); Easterness (V.C. 96), Abernathy Forest, on decorticated branch of Pinus, 1250 feet altitude, 24.V.1976, B.J. Coppins 3162 and L. Tibell (E); South Hants (V.C. 11), New Forest, Brockenhurst Woods, Bakers Copse, on decorticated Fraxinus, 314.1982, P.W. James (BM). Ireland: Galway, Ballynahinch, on pine bark, ii. 1877, Larbalestier (BM). Austria: Steiermark, Riicken der Aflenzer Staritzen NE iiber Aflenz, 1700-1800 m altitude, Figure 1. A: Asci and ascospores of Agyrium rufum with amyloid parts stippled; B: portion of thallus filaments of Racodium rupestre (above) and Cystocoleus ebeneus (below), showing the different arrangement of the dark fungal hyphae on filaments of the alga Trentepohlia. Scales: A= 10 pm; B= 20 pm. 68 G. Kantivilas 2.ix.l984, I. Brodo and J.Poelt (GZU); Salzburg, Pinzgau, Nationalpark Hohe Tauern, Wildgerlostal, zwischen Gasthaus Finkau (1420 m) und Trisslalm (1583 m), 23.vii.1992, C. Scheuer (GZU). Bryophagus Nitschke ex Arnold Bryophagus is a genus of three species currently included in the family Gyalectaceae: B. gloeocapsa Nitschke ex Arnold from Europe, B. similis (Vezda) Kalb, from tropical America, and B. minutissima (Vezda) D. Hawksw. from New Guinea. This last species is here recorded from Tasmania, representing the first report of the genus for Australia. Accounts of the genus are provided by Vezda (1966, 1973) and Purvis and James (1992b). In the past, nomenclatural problems caused all these species to be included in the genus Gloeolecta Lettau, but this name is now regarded as a synonym of Bryophagus (Hawksworth et al. 1980). Bryophagus is superficially similar to several other genera of tiny, inconspicuous crustose lichens, especially to Gyalecta Ach. and Absconditella Vezda (neither presently known in Tasmania), and Gyalidea Lettau (represented in Tasmania by G. hyalinescens (Nyl.) Vezda). Taxa in these genera typically have a crustose, sometimes + gelatinous thallus, small to minute, gyalectoid, pale-coloured to translucent apothecia, elongate, clavate asci typically with only a slightly thickened apex, simple paraphyses and hyaline, frequently 3-septate ascospores. Separation of these genera is not straightforward. In Gyalecta, the photobiont is Trentepohlia whereas the other genera have a coccoid photo- biont; in the case of Bryophagus, the photobiont is the distinctive Gloeocystis in which the small, ± globose or ellipsoid cells are grouped within a thick gelatinous sheath. Although having asci with an amyloid wall after pretreatment in KOH, the asci of Absconditella have a rather prominent, thickened tholus whereas those of Bryophagus are only very slightly thickened. Gyalidea and Gyalecta have asci rather similar to Bryophagus but these are entirely non-amyloid. In addition, the ascospores of Gyalidea have a thin gelatinous perispore whereas those of Bryophagus do not. Bryophagus minutissima (Vezda) D.Hawksw., in D.E. Shaw, Microorganisms in Papua New Guinea : 248 (1984); Gloeolecta minutissima Vezda, Folia Geobot. Phytotax. 8: 312 (1973). Type: Papua New Guinea: eastern Highlands, Bismarck Ranges, Mt Wilhelm, Imbuku Ridge above Lake Auende, 3450 m, 26.vi.1968, W.A. Weber & D. McVean (holo- type COLO L48422b, n.v.). Thallus not visible, evident only as necrotic patches over the bryophyte substratum. Photobiont cells globose, 2-3 p, with a thick gelatinous sheath to c. 5 pm thick. Apothecia scattered over or slightly embedded in the substratum, lecideine, 0.6-1.3 mm wide, hya¬ line, gyalectoid, with a markedly and persistently concave disc, partly obscured by an incurved, cup-shaped proper margin when young, becoming gaping and excavate when old. Excipulum hyaline, c. 20 pm thick. Hymenium 30-40 pm thick, hyaline, 1+ faint yel¬ low-brown before pre-treatment with KOH, persistently 1+ pale blue after pre-treatment. Asci cylindrical-clavate, 26-35 x 4-4.5 pm, 8-spored; apex slightly thickened, I-; outer wall 1+ blue, somewhat thicker at the apex of the ascus. Paraphyses simple, mostly ± straight, 0.5-0.8 pm thick, with apices very slightly thickened. Ascospores bacilliform, 3- septate, 8-13 x 1 pm. Pycnidia unknown. Remarks: The Tasmanian specimen is very sparingly fertile and with most apothecia over-mature and with eroded hymenia. Hence, the anatomical data given above are based mainly on the type description by Vezda (1973), supplemented where possible with observations from the Tasmanian material. In Tasmania, this species occurs over a mat of Agyrium, Bryophagus and Racodium 69 hepatics on disturbed, sandy soil in an abandoned copper-mining area, a habitat also favoured by the Northern Hemisphere species, B. gloeocapsa. Bryophagus minutissima is extremely inconspicuous and easily overlooked, especially as the thallus is almost absent and the apothecia vary from small to minute, are ± translucent and very difficult to detect even with a lens. Indeed the best clue to its presence is provided by the death in small circular patches of its bryophyte hosts. Bryophagus minutissima is well separated from the other species of the genus by its markedly smaller apothecia and ascospores. In B. gloeocapsa, the apothecia are 0.2-0.5 mm wide, rather immersed in a glossy, gelatinous thallus, and the spores are 20-30 x 1.5- 2 pm. In B. similis, the apothecia are 0.3-0.5 mm wide and the spores are 12-15 x 2.5- 3 pm. Australian specimen examined: Tasmania: Queenstown, opposite old Mt Lyell Mine Office, 42°05’S 145°33’E, on hepatics over gravelly bank colonised by bryophytes, 200 m altitude, 6.ii.l984, G. Kantvilas & P.W. James 191/84 (BM, herb. Vezda, HO). Racodium Fr. The genus Racodium is an obligately sterile genus of filamentous lichens. In the past, it has also included non-lichenised representatives but, as discussed in detail by Hawksworth (1970), it is correctly based on the lichenised hyphomycete R. rupestre Pers. and should be considered monotypic. This species is known from Europe and North America and has recently also been recorded from New Zealand (Wirth 1997). Racodium rupestre Pers., Tent. Disp. Meth. Fung.: 76 (1797). Type: n.v. Thallus filamentous, forming irregularly spreading, black, felt-like patches on the shel¬ tered faces of rocks in cool, damp environments; filaments loosely entangled, 10-20 pm wide. Photobiont Trentepohlia, forming long chains of cells. Fungal hyphae c. 1 pm thick, arranged in a parallel manner along the length of the filaments and forming a rectangu¬ lar, net-like pattern. Apothecia and pycnidia unknown. Remarks: The Tasmanian specimens accord completely with authentic material from Europe (for description see Dalby 1992). Racodium rupestre is superficially very similar to Cystocoleus ebeneus (Huds.) Hariot, a lichen which occurs in identical habitats but appears to be far more common and widespread. The two taxa can only be separated by high-power microscopy. Cystocoleus ebeneus differs by having fungal hyphae which are of noticeably uneven thickness, arranged irregularly over the surface of the algal fila¬ ments (see Fig IB and Purvis et al. 1992). Racodium rupestre is uncommon in Tasmania although it may well have been over¬ looked. In contrast, it is significant that the similar C. ebeneus is very common and has been very frequently collected. The sheltered, shaded, often overhanging faces of large boulders in generally moister vegetation types are the typical habitat for both species. Both taxa are typically found at higher elevations in open woodland and heathland, although C. ebeneus is also known to occur in the lowlands, even on coastal rocks. Australian specimens examined: Tasmania: track to Mother Cummings Peak, 41°41’S 146°32’E, on sheltered faces of large dolerite boulders at scrubby rainforest edge, 1000 m altitude, 3.iii.2002, G. Kantvilas 149/02 (HO); Mt Sprent, 42°48’S 145°58’E, on Precambrian rocks in alpine heathland, 1050 m altitude, 17.ii. 1987, G. Kantvilas s.n. (HO); Mt Arrowsmith, 42°12’S 146°05’E, on rocks, 960 m altitude, 14.xi.1964, G.C. Bratt 1778 (HO). 70 G. Kantivilas Comparative material examined: Sweden: Harjedalen Province, southern slope of Mt Gruvvalen, 62°43’N 12°25’E, on exposed rock in the upper part of the subalpine region, c. 900 m altitude, l.ix.1970, R. Santesson 22567 (. Lichenes Selecti Exsiccati Upsaliensis 45) (GZU). Acknowledgements The major part of this work was undertaken at the Botany Department, Natural History Museum, London, and I thank the Keeper and staff for their generous support in the course of my visit. I also thank Dr Brian Coppins and Dr Christian Scheuer for the loan of specimens, and Dr Antonin Vezda for confirming the identity of Bryophagus. References Dalby, D.H. (1992). ‘ Racodium Fr. (1829)’, in O.W. Purvis, B.J. Coppins, D.L. Hawksworth, P.W. Janies and D.M. Moore (eds), The Lichen Flora of Great Britain and Ireland , pp. 523-524. Natural History Museum Publications: London. Dennis, R.W.G. (1981). British Ascomycetes (revised edn). J. Cramer: Vaduz. Hawksworth, D.L. (1970). A nomenclatural note on Racodium Pers. Transactions of the British Mycological Society 54, 323-326. Hawksworth, D.L., James, P.W. and Coppins, B.J. (1980). Checklist of the British lichen-forming, lichenicolous and allied fungi. Lichenologist 12, 1-115. Lumbsch, H.T. (1997). Systematic studies in the sub-order Agyriineae (Lecanorales). Journal of the Hattori Botanical Laboratory 83, 1-73. Purvis, O.W. and James P.W. (1992a). ‘ Agyrium Fr. (1822)’, in O.W. Purvis, B.J. Coppins, D.L. Hawksworth, P.W. James and D.M. Moore (eds), The Lichen Flora of Great Britain and Ireland, p. 67. Natural History Museum Publications: London. Purvis, O.W. and James, P.W. (1992b). ‘ Bryophagus Nitschke ex Arnold (1862)’, in O.W. Purvis, B.J. Coppins, D.L. Hawksworth, P.W. James and D.M. Moore (eds), The Lichen Flora of Great Britain and Ireland, p. 124. Natural History Museum Publications: London. Purvis, O.W., Coppins, B.J., Hawksworth, D.L., James, P.W. and Moore, D.M.(1992). The Lichen Flora of Great Britain and Ireland. Natural History Museum Publications: London. Rambold, G. and Triebel, D. (1990). Gelatingia and Phaeopyxis, three helotialean genera with lichenicolous species. Notes from the Royal Botanic Garden Edinburgh 46, 375-389. Rehm, H. (1899). Ascomycetes Fuegiani a P. Dusen collecti. Bihang till Kongliga Svenska Vetenshaps-Alcadamiens Handlingar 25, Afd. 3(6), 3-21. Saccardo, PA. and Trotter, A. (1913). Sylloge Fungorum 22, Supplementum Universale 9(1), 586- 587. Vezda, A. (1966). Flechtensystematische Studien III. Die Gattungen Ramonia Stiz. und Gloeolecta Lett. Folia Geobotanica et Phytotaxonomica 1, 154-175. Vezda, A. (1973). Flechtensystematische Studien VIII. Drei neue Arten der Gyalectaceae sensu amplo aus Neu-guinea. Folia Geobotanica et Phytotaxonomica 8, 311-316. Wirth, V. (1997). Additional lichen records from New Zealand 21. Candelariella coralliza, Lepraria eburnea, Racodium rupestre, Rinodina olivaceobrunnea, Rinodina pyrina and Trapeliopsis flexuosa. Australasian Lichenological Newsletter 40, 11-13. Muelleria 16: 71-79 (2002) Alexander Clifford Beauglehole OAM (26 August 1920-19 January 2002) Cliff Beauglehole, the youngest of three sons of Richard and Margaret Beauglehole, was born in Portland in south-west Victoria (A.C. Beauglehole, undated). For the first five years of his life he lived on his parents farm south of the Portland-Bridgewater road in the Parish of Trewalla. In 1926 the family moved to his parents other farm at Gorae West. Cliff’s schooling was of short duration as he only attended the tiny Gorae State School, eight kilometres away, for six years. To reach school he rode each day on a pony along rough bush tracks. Possessing a natural curiosity and encouraged by his parents who were both interested in the natural world, it was during these formative years that his interest in natural history was initiated and stimulated. Thus began an interest in the natural world that was a life-long preoccupation. The Beaugleholes had as family friends the Holmes family at Gorae. Members of the Holmes family were very keen nature lovers who shared freely their knowledge of the bush. By the age of ten Cliff knew and could name scientifically about sixty species of orchids from the district. By the time Cliff left school he had a working knowledge of much of the local flora and fauna, but particularly of orchids and birds. On leaving school Cliff assisted his parents on their mixed farm at Gorae West to which the family had moved. Orchids were of special interest to Murray Holmes and this interest transmitted itself to Cliff. Together they covered hundreds of kilometres on horse, pushbike and on foot searching for native orchids. He received encouragement also from Mrs. Floss Mellblom, another local orchid enthusiast, who executed coloured illustrations of the local orchid species to enable others to identify them more readily. One of Cliff’s prized possessions was a set of illustrations that she presented to him. Encouraged by Murray and Floss, in 1934 at the age of fourteen Cliff began corresponding with and sending specimens for identification to W.H. Nicholls in Melbourne, one of the leading Australian orchid spe¬ cialists of the time. Nicholls (1934) replied to what was possibly Cliff’s first letter to him by stating ‘Yes. I know a little about Orchids - no one knows everything.’ The following year, on account of illness, Nicholls (1935) wrote to Cliff ‘.Would you think me unkind to ask you to write to The Government Botanist (Mr. Rae) and forward all speci¬ mens to him for determination. The new 15000 Pound Herbarium is now opened and they have all the facilities there to give you full particulars of all the flora known. You will find them very obliging.’ Thus began an association with the National Herbarium of Victoria that was to last for over sixty-five years. 72 J.H. Ross Soon after leaving school Cliff started a systematic survey of the flora of the Portland region in his spare time, in the process doubling the number of species recorded for the area. On account of this activity he was christened ‘King of the Gorae Forest’ (A.C. Beauglehole, undated). By the early 1940s Cliff was corresponding with Rev. H.M.R. Rupp and supplying him with specimens of orchids. The years 1941 and 1942 were espe¬ cially good seasons for orchids during which he discovered three previously undescribed species. His first paper ‘Orchids of the Portland District’ was published in the Victorian Naturalist 60: 23-25 (1943). In this paper were listed the scientific name, co mm on name, date of first collection and the localities at which each species of orchid was collected. The compilation of species lists was a format that Cliff favoured throughout his career. Rev. Rupp suggested to Cliff that he should study all plants. Cliff readily accepted this challenge and was encouraged by the botanists including Jim Willis at the National Herbarium of Victoria to do so. He extended his botanical explorations in south-west Victoria in search of mosses, liverworts, fungi, lichens, and freshwater and marine algae. Together with Jim Willis he conducted field work in the Portland area, Grampians and elsewhere. In September 1941 Cliff began to contribute to B E Carthew’s column ‘Nature Notes’ in the ‘Portland Observer.’ Between 1941 and 1949 Cliff contributed almost monthly to this column, and a substantial proportion of about one hundred articles contain informa¬ tion on Cliff’s observations. In October 1949 Cliff married Hilda Mary Oakley. Life on the farm, which he pur¬ chased from his parents, was very busy with potato growing, an apple orchard and dairy cows, especially following the birth of two daughters, Valerie and Yvonne. The change from dairy cows to beef cattle made life a little easier for Cliff to pursue his natural his¬ tory interests. Hilda always took a keen interest in, and often helped with, Cliff’s various projects. Throughout these years on the farm Cliff continued his botanical explorations leaving much of the work on the farm during his absences to Hilda. In 1968 Cliff and the family moved to Portland to allow Cliff to devote more time to botany. The 1950s appear to have been particularly busy. Long interested in native bees, Cliff sent his first specimen to the then National Museum of Victoria for identification. Entomologist Tarlton Rayment (1950a) replied: ‘. Even your first catch is a New Record for the State.’ In response to a request from Tarlton Rayment and detailed instruc¬ tions on how to proceed, Cliff began studying the life histories and collecting local native bees in 1950. This work led to a very close collaboration between the two. Numerous detailed requests from Rayment followed, some of which required several days of work, or a constant vigil over a long period for a male or a female of a given species. On one occasion Cliff spent two days digging out one Nomia australica nest at Bats Ridge, the winding shaft and lateral cells extending almost 1.2 metres to a special moist clay. This work on bees culminated in 1953 with the publication by the Portland Field Naturalists’ Club of the book ‘Bees of the Portland Region’ by Tarlton Rayment. Altogether Rayment described thirty new species of native bees from Cliff’s collections. Of these, three were named after Cliff, four commemorate Portland and two Gorae West. Subsequent to the publication of ‘Bees of the Portland Region ’, Cliff collected another twenty species for the Region. Rayment also made a study of the parasitic acarid mites found on the native bees sent by Cliff. While working on the native bees Cliff extended his studies and started collecting wasps for Rayment. Rayment (1950b) wrote to Cliff: ‘Keep your eyes skinned, Cliff, for greenish wasps [ Sericophorus ] about same size as the bees ( Parasphecodes wellingtoni) in stumps, they make shafts in sandy ground, and I am anxious to get some from your dis¬ trict.’ It was only in 1953 that Cliff found the first Sericophorus shafts along Cape Nelson Road. He worked on the shafts for fourteen days and subsequently found others at Mt Richmond and at Gorae West. Rayment (1951) wrote: ‘Well, Cliff, it looks at first glance that you have been remarkably successful. There were six of the wasps, and one that I A.C. Beauglehole OAM 73 have worked is new - and I think that there will be more.’ Cliff collected at least four new species of wasps, one of which was named after him. It is unfortunate that Rayment died before the work on native bees and wasps was concluded as it terminated the prospect of a book on the wasps of the Portland District by the two men. In addition to the bees and wasps, Cliff also collected the type of a new species of blow-fly (Pollenia tragica Rayment). Cliff began a serious study of ants following a meeting early in 1954 with Father John McAreavey, a Master at Xavier College, Melbourne, who was preparing a catalogue of Australian ants (McAreavey, 1954). In July 1954 McAreavey wrote: ‘I have just com¬ pleted naming of your ants, and as you guessed enjoyed it. They are in such good condi¬ tion and so clearly set out that it is a pleasure and an interest to go through them.’ In October 1955 McAreavey wrote: ‘You got a nice lot [of ants] -I make it 118 different species representing nearly all the subFamilies.’ In addition to new records for Victoria, according to McAreavey, apparently twelve of the ants collected by Cliff represented undescribed species. Unfortunately Cliff had no further contact with McAreavey once the latter left Xavier College. Once again a very fruitful collaboration had terminated pre¬ maturely for Cliff. Cliff’s interest in birds was encouraged by people such as Les Chandler and Noel Learmonth. The latter, who published ‘Birds of the Portland District’, indicated that the book ‘contains many species that would not have been recorded but for his [Cliff’s] field observations, indeed the author would not care to have attempted the book without hav¬ ing the ornithological work of Cliff Beauglehole to draw on.’ (Learmonth, 1967). Together with Hilda and other helpers in 1951 he started to collect and record sea birds washed up on the wild storm-lashed 50 kilometre coast of Discovery Bay, The carcasses of birds not collected were deposited behind the first dune to ensure that they were not counted on subsequent beach patrols. Between 1951 and 1963 almost 5000 carcasses were collected by Cliff and his helpers. In one day during August 1959 Cliff and Hilda collected 950 bird carcasses (Corrick, 1971). The flesh was allowed to decompose natu¬ rally on wire racks erected on top of the hay shed on the Beauglehole farm. It is said that somet im es it was possible to smell the Beauglehole farm before actually seeing it! Skins and clean skeletons of numerous birds were donated to the National Museum of Victoria and other Australian museums and to Harvard University, the United States of America. This work extended knowledge of the sea birds visiting our shores, and included several new records of species visiting Australia (Corrick, 1971). In 1957 a group of members of the Victorian Field Naturalists Club of Victoria visit¬ ed Portland, amongst them the conchologist Charles J Gabriel. Gabriel was hoping to find someone locally to collect tiny land snails for him. Needless to say Cliff took up the chal¬ lenge. In January 1958 Gabriel wrote: ‘Now old chap not in my wildest dreams did I anticipate an answer so quickly to my “S.O.S” for W-Vict Land Shells. It was excellent and please accept my many thanks for your efforts, ....’. Some years later Cliff continued to send molluscs to Fred and Jan Aslin in Mount Gambier. During the 1950s Cliff, together with Fred Davies of Portland, started a systematic survey of the bone deposits in many of the lim estone caves in south-west Victoria. This work arose out of concern at the damage being caused to the caves through destructive activities such as road works and the dumping of rubbish. As in many of his endeavours, Cliff obtained the support of local naturalists and in 1963 they started to excavate McEachern’s Death Trap Cave, a chimney cave about 1.2 metres in diameter and 15 metres deep. Several tons of sediments were hauled to the surface and sieved. In this hard physical labour Cliff was assisted by his wife and daughters and collectively they spent hundreds of hours on the project. The identifiable remains of over 2000 animals were recovered and some of the material dated back 5,000 years to the Pleistocene. The high¬ light of the deposit was the discovery of the remains of a Tasmanian Tiger (Carthew, 1964). In 1964-5 Norman Wakefield became involved in the excavation and assessment 74 J.H. Ross of the material (Wakefield, 1967). Study of this material added greatly to the knowledge of the earlier fauna of the region. During the 1960s and 1970s Cliff extended his travel visiting Lord Howe Island twice, the Darwin region of the Northern Territory twice, Central Australia seven times and the Kimberley Region of Western Australia four times. In addition to plants, Cliff also col¬ lected s kink s, geckoes, frogs, lizards, molluscs in Central and Western Australia. However, it was in Victoria that Cliff concentrated his efforts and there is scarcely a road or track in Victoria on which he did not travel at least once. Cliff received financial support in 1968 from the Royal Botanic Gardens Research Trust Melbourne, a Trust associated with the Maud Gibson Trust, to survey the Grampians. Although the flora of the Grampians was thought to be reasonably well doc¬ umented, there was a belief that many additional records awaited discovery. This was confirmed by Cliff’s studies. The Grampians study was followed by funding from gov¬ ernment agencies (mainly the National Parks Service and the Land Conservation Council of Victoria) to subsidise the survey of East Gippsland and other areas of Victoria, and from the Utah Foundation towards a survey of the alpine areas. In his survey work Cliff also received some financial and other support from the Western Victorian Field Naturalists’ Clubs Association, from individual clubs, other organisations and individu¬ als. In time the whole of Victoria was surveyed. The results of Cliff’s work was published by the Land Conservation Council of Victoria in the form of conservation recommenda¬ tions. Cliff subsequently published checklists for the entire state in a series of thirteen reports under the title The distribution and conservation of vascular plants in Victoria. Cliff had a very discerning eye especially when it came to detecting differences or unusual forms and missed very little in the field. In a letter supporting Cliff’s nomination for the Australian Natural History Medallion, Williams (1967) wrote ‘Many of these mosses are almost microscopic in size and indeed I find that I cannot identify many of these without the use of a microscope but Cliff identifies these lowly plants in the field - and he will be 95% right!!’ It has not been unusual for many years, and sometimes decades, to elapse before a taxon that Cliff had identified as different was recognised for¬ mally and described by a botanist. He seldom resorted to using keys to identification but relied instead on visual recognition and his prodigious memory and ability to recall diag¬ nostic details. In the field when preparing to conduct his surveys, Cliff would plan his col¬ lecting sites carefully and ensure that each different community and habitat was visited. On reaching a site, he would walk around and collect and record taxa until he could find no additional taxa before moving on to his next site. During his years of field work he amassed a huge collection of specimens and his collecting numbers exceeded 95,000. Cliff very generously donated the bulk of his private herbarium to the National Herbarium of Victoria many years ago, but he also sent specimens for identification to botanists in a number of other herbaria and his specimens are to be found in each main Australian herbarium and in several overseas. Cliff is the largest single contributor of specimens to the National Herbarium of Victoria and it is unlikely that his record will be bettered. In the case of Victoria, Cliff intended that his specimens would serve as vouch¬ ers for his species lists and he attempted to collect each taxon present within the min or 10’xlO’ grid ‘squares’ then employed for recording distribution within the state. Many of the specimens were of necessity sterile, but nevertheless they are an important record and often are the only record of the taxon in question from that part of Victoria or indeed from Victoria. Often it is a Beauglehole specimen that alerts one to the existence of an inter¬ esting or unusual taxon at a particular site and enables one to re-visit the site. Cliff had tremendous drive, energy and commitment, working from dawn to dusk in the field col¬ lecting and recording and half of the night documenting and processing his collections. Having witnessed the destruction of native habitats in his youth, Cliff was always very environmentally aware and with the passing of the years this awareness intensified. Deeply concerned by the disappearance of natural communities due to agricultural and A.C. Beauglehole OAM 75 urban developments, he was actively involved with a number of other naturalists in many campaigns for environmental conservation. Cliff was one of the foundation members of the Western Victorian Conservation Committee, a sub-co mmi ttee of the Portland Field Naturalists Club, that was established to spearhead for the fight for the preservation of the Little Desert. The Committee was active in bringing to the attention of successive gov¬ ernments the need to change practices of land management. He was also instrumental in the preservation of the Mount Richmond and the Lower Glenelg National Parks. Although not opposed to the Portland Aluminium smelter, he was opposed to it being sited on the rich wet coastal heathland of Point Danger, arguing that the smelter could readily be moved and built on disturbed pasture land one kilometre inland. Cliff had a long-standing interest in a number of sites in or near Portland such as Fawthrop Lagoon where his contribution is commemorated by a walking track that bears the name ‘Cliff’s Walkway.’ Cliff also led a campaign to prevent McDonald’s Family Restaurants from occupying a prime corner site in Portland that resulted in the City Council re-zoning the land and setting it aside permanently as a park. He was concerned not only by the incur¬ sion of alien exotics into every plant community in Victoria but also by the spread of native species such as Acacia longifolia (Andrews) Willd. subsp. sophorae (Labill.) Court and Pittosporum undulatum Vent, that were spreading rapidly beyond their former natu¬ ral ranges and having an adverse environmental effect in every respect as detrimental as that of the worst introduced species. Cliff also lamented the loss of much natural bush to pine and latterly eucalypt plantations. Always very generous with his time, Cliff was always willing to assist visitors to his area conducting field work, or to give precise directions to one of his collecting localities or provide supplementary information. Whenever a new or challenging piece of field research was required in SW Victoria, Cliff rose to the occasion and responded. Such was his industry, application and enthusiasm, that Cliff invariably surpassed with his contri¬ bution the expectations of the person(s) who made the request, more often than not over¬ whelming them with specimens or records of observations. With an insatiable thirst for knowledge of the natural world, Cliff was more or less the complete naturalist, a rarity in this day of increasing specialisation. What makes Cliff’s contributions prior to 1968 when the family moved to Portland more remarkable is that in addition to devoting so much energy to his natural history interests, he also had to earn a living operating his farm although in this latter enterprise Hilda played a crucial role. Cliff always relied initially on specialists to identify his collections. In the case of the specialist entomologists with whom he collaborated, each predeceased Cliff by many years and consequently the collaborative projects lapsed prematurely which would have been a great disappointment for Cliff. Unfortunately the government herbaria that Cliff turned to for assistance were basically unable to cope with the quantity of specimens and meet his deadlines which were often incompatible with institutional priorities. I recall vividly receiving several consignments of specimens from Cliff. Unlike most other col¬ lectors who sent in an occasional box of specimens, Cliff would send in cubic metres of specimens. Cliff was a foundation member of the Portland Field Naturalists Club and of the Western Victorian Field Naturalists Clubs Association. Always an active member of the Portland Field Naturalist Club, he encouraged many country naturalists to develop an interest in their environment and to monitor and document the vegetation. Cliff became a very important figure and an inspiration to many of these naturalists who sent specimens to him for identification or sought his advice. Cliff received Life Membership of the Portland Field Naturalist Club in 1962 in recognition of his contributions. In 1982 he was elected an Honorary member of the Field Naturalists Club of Victoria in recognition of 40 years of continuous membership. Always very active, he was a member of many other Clubs, Associations and Societies. In 1971 Cliff was awarded the Australian Natural History Medallion for services to 76 J.H. Ross natural history. In 1984 he was awarded the ‘Medal of the Order of Australia’ (OAM) for his services to botany, conservation and ornithology. Much of what Cliff accomplished would not have been possible without the support of Hilda. It was very fitting that The Portland Field Naturalists Club recognized her con¬ tribution and presented her with a Life Membership certificate on 10 October 1962 with the following words inscribed; ‘presented to Mrs. Hilda Beauglehole who in so large a measure has assisted her husband in the splendid work he has achieved.’ Cliff’s legacy will live on through his collections which constitute a huge resource from which future generations will continue to derive benefits, his publications, and from his vision in campaigning actively for several natural areas to be set aside as national parks. Cliff is survived by his wife, two daughters, seven grandchildren and two great-grand¬ children. Taxa named after Cliff: Algae Helminthocladia beaugleholei Womersley Nitella tasmanica Muell. ex A.Braun subsp. gelatinifera R.D.Wood var. beaugleholei R.D.Wood Moss Phascum beaugleholei I.G.Stone Vascular plants Bassia beaugleholei Ising (now a synonym of Sclerolaena diacantha (Nees) Benth.) Caladenia beaugleholei D.L.Jones (now a synonym of Arachnorchis flavovirens (G.W.Carr) D.L.Jones & M.A.Clem.) Epilobium brunnescens (Cockayne) P.H.Raven & Engelhorn subsp. beaugleholei K.R.West & P.H.Raven Lobelia beaugleholei Albr. Prasophyllum beaugleholei Nicholls (now a synonym of Corunastylis nuda (Hook.f.) D.L.Jones & M.A.Clem.) Solarium beaugleholei D.E.Symon Stylidium beaugleholei J.H.Willis Utricularia beaugleholei R.J.Gassin Villarsia umbricola Aston var. beaugleholei Aston Heterodea beaugleholei Filson Exoneura cliffordiella Rayment Hylaeus cliffordiellus Rayment Megachila cliffordii Rayment Serocophorus cliffordii Rayment Major publications Beauglehole, A.C. (1943). Orchids of the Portland District. Victorian Naturalist 60, 23-25. Beauglehole, A.C. (1944). Ferns of the Portland District. Victorian Naturalist 60, 193-195. Beauglehole, A.C. (1947). Checklist of the indigenous flora of the Lower Glenelg. Victorian Naturalist 64, 78-86. Beauglehole, A.C. and Learmonth, N.F. (1955). Proposed reserve at Mt Richmond, S.W. Victoria. Victorian Naturalist 12, 136-7. Beauglehole, A.C. and Learmonth, N.F. (1956). Fern flora of the Portland District. Victorian Naturalist 73, 40-42. Beauglehole, A.C. and Learmonth, N.F. (1957). The Byaduc Caves. Victorian Naturalist 73, 204-210. Lichen Bees Wasp A.C. Beauglehole OAM 77 Beauglehole, A.C. and Finck, E.W. (1967). Indigenous vascular flora of Port Campbell National Park, Victoria. Report to the National Parks Authority. Beauglehole, A.C. (1967). The indigenous vascular flora of the Mount Richmond National Park, Victoria. Report to the National Parks Authority. Beauglehole, A.C. (1968). Native flowering plants and ferns of the proposed Glenelg National Park. In ‘The case for a Lower Glenelg National Park’, Western Victorian Conservation Committee, Portland: i-xiv. Beauglehole, A.C. and Finck, E.W. (1969). A critical survey of the vascular flora of the National Parks of East Gippsland (with special reference to four adjacent areas for comparative purposes, i.e. proposed extensions). Report to the National Parks Authority, 37pp. Beauglehole, A.C. (1971a). The Howe Range area of East Gippsland (in relation to the vascular floras of seven combined East Gippsland National Parks, with special refer¬ ence to Mallacoota National Park, Victoria). Report to the National Parks Authority. Beauglehole, A.C. and Rogers, K.C. (1971). Vascular Plant Checklist of proposed Snowy River National Park, Victoria. Report to the National Parks Authority. Beauglehole, A.C. (197 lb). Vascular plant Checklist - Comparison of Proposed Cobberas and Snowy River National Parks with Alfred, Glenaladale, Captain James Cook, Lind, The Lakes and Wingan Inlet National Parks, Victoria. Report to the National Parks Authority. Beauglehole, A.C. (1972). Botanical Survey of East Gippsland. Victorian Naturalist 89, 96-98. Beauglehole, A.C. (1974). The vascular flora of Bulga and Tarra Valley National Parks. Report to the National Parks Authority. Beauglehole, A.C., Carr, G.W. and Parsons, R.F. (1975). A checklist of the vascular flora of the Holey Plains, Gippsland, Victoria. Proceedings of Royal Society of Victoria 87, 251-270. Beauglehole, A.C. (1976). Report on the vascular flora of the proposed extension to Captain James Cook National Park, Victoria. National Parks Authority. Beauglehole, A.C., Carr, G.W. and Parsons, R.F. (1977). A floristic checklist of the Otways Region, Victoria. Proceedings of Royal Society of Victoria 89, 99-122. Beauglehole, A.C. (1978a). Alterations and additions to the vascular flora of Victoria - Part 1. Victorian Naturalist 95, 61-14. Beauglehole, A.C. (1978b). Alterations and additions to the vascular flora of Victoria - Part 2. Victorian Naturalist 95, 198-203. Beauglehole, A.C. (1979). The Distribution and Conservation of native vascular plants in the Victorian Mallee. Western Victorian Field Naturalists Clubs Association, Portland. 99pp. Beauglehole, A.C. (1980a). The Distribution and Conservation of vascular plants in the Corangamite - Otway area, Victoria. Western Victorian Field Naturalists Clubs Association, Portland. 108pp. Beauglehole, A.C. (1980b). Victorian Vascular Plant Checklists. Western Victorian Field Naturalists Clubs Association, Portland. 206pp. Beauglehole, A.C. (1980c). Erroneous or doubtful Victorian vascular plant grid records. Victorian Naturalist 97, 213-216. Beauglehole, A.C. (1980d). Deletion of Victorian vascular plants and their grid distribu¬ tion. Victorian Naturalist 97, 247-249. Beauglehole, A.C. (1981a). Alterations and additions to the vascular flora of Victoria. Victorian Naturalist 98, 53-58. Beauglehole, A.C. (1981b). The Distribution and Conservation of vascular plants in the Alpine area, Victoria. Western Victorian Field Naturalists Clubs Association, Portland. 110 pp. 78 J.H. Ross Beauglehole, A.C. (1981c). The Distribution and Conservation of vascular plants in the East Gippsland area, Victoria. Western Victorian Field Naturalists Clubs Association, Portland. 124 pp. Beauglehole, A.C. (1982). The Distribution and Conservation of vascular plants in the North Central area, Victoria. Western Victorian Field Naturalists Clubs Association, Portland. 102 pp. Beauglehole, A.C. (1983a). The Distribution and Conservation of vascular plants in the Melbourne area, Victoria. Western Victorian Field Naturalists Clubs Association, Portland. 156 pp. Beauglehole, A.C. (1983b). The Distribution and Conservation of vascular plants in the Ballarat area, Victoria. Western Victorian Field Naturalists Clubs Association, Portland. 94 pp. Beauglehole, A.C. (1984a). The Distribution and Conservation of vascular plants in the South Gippsland area, Victoria. Western Victorian Field Naturalists Clubs Association, Portland. 90 pp. Beauglehole, A.C. (1984b). The Distribution and Conservation of vascular plants in South West Victoria. Western Victorian Field Naturalists Clubs Association, Portland. 124 pp. Beauglehole, A.C. (1984c). Three important orchids. Victorian Naturalist 101, 168-169 Beauglehole, A.C. (1985). The Distribution and Conservation of vascular plants in the Gippsland Fakes Hinterland area, Victoria. Western Victorian Field Naturalists Clubs Association, Portland. 98 pp. Beauglehole, A.C. (1986). The Distribution and Conservation of vascular plants in the Murray Valley area, Victoria. A.C. and H.M. Beauglehole, Portland. 81pp. Beauglehole, A.C. (1987). The Distribution and Conservation of vascular plants in the Wimmera area, Victoria. A.C. and H.M. Beauglehole, Portland. 87pp. Beauglehole, A.C. (1988). The Distribution and Conservation of vascular plants in the North East area, Victoria. A.C. and H.M. Beauglehole, Portland. 98pp. In addition, Cliff contributed about 100 articles that were published under the title ‘Nature Notes’ in the Portland Observer edited by B.E. Carthew during the years 1941-1949, and numerous articles to the Australian Conservation Foundation Portland Chapter Newsletter and to the Western Victorian Conservation Committee Newsletter. Acknowledgements I am most grateful to Hilda Beauglehole and Valerie and Yvonne for sharing with me memories of some of the early days with Cliff and for providing the photograph repro¬ duced above, and to Margaret Corrick for sharing recollections her association with the Beauglehole family when she lived in southwestern Victoria. References Beauglehole, A.C. (undated). Handwritten manuscript, Archives, Royal Botanic Gardens Melbourne. Carthew, B.E. (1964). ‘Nature Notes’, The Portland Observer, 22 May 1964. Corrick, M.G. (1971). Australian Natural History Medallion 1971. Victorian Naturalist 88, 344-346. Corrick, M.G. (2002). Alexander Clifford Beauglehole. Victorian Naturalist 119, 81-2. Gabriel, C.J. (1958). Copy of letter from C.J. Gabriel to A.C. Beauglehole dated 17 January 1958. Archives, Royal Botanic Gardens Melbourne. Learmonth, N.F. (1967). Letter supporting the nomination of A.C. Beauglehole for the Australian Natural History Medallion. Archives, Royal Botanic Gardens Melbourne. McAreavey, J. (1954). Letter from J. McAreavey to A.C. Beauglehole dated 17 July 1954. Archives, Royal Botanic Gardens Melbourne. A.C. Beauglehole OAM 79 McAreavey, J. (1955). Letter from J. McAreavey to A C Beauglehole dated 21 October 1955. Archives, Royal Botanic Gardens Melbourne. Nicholls, W.H. (1934). Letter from W.H. Nicholls to A.C. Beauglehole dated 29 Sept. 1934. Archives, Royal Botanic Gardens Melbourne. Nicholls, W.H. (1935). Letter from W.H. Nicholls to A.C. Beauglehole dated 9 October 1935. Archives, Royal Botanic Gardens Melbourne. Rayment, T. (1950a). Letter from T. Rayment to A.C. Beauglehole dated 3 March 1950. Archives, Royal Botanic Gardens Melbourne. Rayment, T. (1950b). Letter from T. Rayment to A.C. Beauglehole dated 3 November 1950. Archives, Royal Botanic Gardens Melbourne. Rayment, T. (1951). Letter from T. Rayment to A.C. Beauglehole dated 26 Jan 1951. Archives, Royal Botanic Gardens Melbourne. Wakefield, N.A. (1967). Preliminary Report on McEachern’s Cave, S.W. Victoria. Victorian Naturalist 84, 363-383. Williams, L.D. (1967). Letter from L.D. Williams to M.G. Corrick dated 1 June 1967. Archives, Royal Botanic Gardens Melbourne. J.H. Ross National Herbarium of Victoria, Royal Botanic Gardens Melbourne, Birdwood Ave., South Yarra, Vic. 3141. Jim.Ross@rbg.vic.gov.au Muelleria 16: 81-82 (2002) Some new combinations and a new hybrid genus in Orchidaceae: Diurideae , for eastern Australia Jeffrey A. Jeanes National Herbarium of Victoria, Royal Botanic Gardens Melbourne, Birdwood Avenue, South Yarra, Vic. 3141. Jeff.Jeanes@rbg.vic.gov.au Abstract New combinations and a new hybrid genus are created for the Orchidaceae tribe Diurideae in east¬ ern Australia. The following new combinations are made in Arachnorchis D.L.Jones & M.A.Clem. and Simpliglottis Szlachetko —Arachnorchis Xvariabilis (Nicholls) Jeanes, Simpliglottis gramma- ta (G.W.Carr) Jeanes, Simpliglottis jeanesii (D.L.Jones) Jeanes and Simpliglottis triceratops (D.L.Jones) Jeanes. The following new hybrid genus is created followed by a new combination within that genus — xChilosimpliglottis Jeanes, xChilosimpliglottis pescottiana (R.S.Rogers) Jeanes. Introduction The past couple of years have seen a flurry of activity in the reclassification of parts of the primarily Australian orchid tribe Diurideae (Hopper & Brown 2000, 2001a, 2001b; Jones et al. 2001; Szlachetko 2001a, 2001b; Jones et al. 2002). These various works have given rise to conflicting classifications at the generic and subgeneric levels as well as the publication of many invalid names. Furthermore, the authors have overlooked several taxa and hence some of the necessary combinations have not been made into these new taxonomic systems. The opportunity is here taken to create the necessary new combina¬ tions at the generic level for those taxa occurring in eastern Australia. The creation of these new combinations will benefit flora writers, compilers of flora lists and land man¬ agement authorities who often work within a legislative framework that demands validly published binomials for the taxa with which they deal. Taxonomy Arachnorchis Xvariabilis (Nicholls) Jeanes, comb. nov. Basionym: Caladenia variabilis Nicholls, Victorian Naturalist 66: 223, figs L & M (1950). An apparent natural hybrid between Arachnorchis orientalis (G.W.Carr) D.L.Jones & M.A.Clem. and Arachnorchis tessellata (Fitzg.) D.L.Jones & M.A.Clem. from south¬ eastern Victoria. Natural hybrids of similar appearance can be derived from hybridiza¬ tion between other taxa (Jeanes & Backhouse 2001). xChilosimpliglottis Jeanes, hybrid gen. nov. This hybrid genus is the result of natural hybridization between the genera Chiloglottis R.Br. and Simpliglottis Szlachetko. One named taxon is currently recognised. xChilosimpliglottis pescottiana (R.S.Rogers) Jeanes, comb. nov. Basionym: Chiloglottis pescottiana R.S.Rogers, Proc. Roy. Soc. Victoria new ser. 30: 139, t.25 (1918). This natural hybrid between Chiloglottis trapeziformis Fitzg. and Simpliglottis valida (D.L.Jones) Szlachetko occurs in New South Wales and Victoria, and has been observed at a number of sites where the ranges of the parent species overlap. 82 J.A. Jeanes Simpliglottis grammata (G.W.Carr) Jeanes, comb. nov. Basionym: Chiloglottis grammata G.W.Carr, Indigenous Flora & Fauna Association Miscellaneous Paper 1: 20 (1991). Simpliglottis jeanesii (D.L.Jones) Jeanes, comb. nov. Basionym: Chiloglottis jeanesii D.L.Jones, Orchadian 12(5): 233 (1997). Simpliglottis triceratops (D.L.Jones) Jeanes, comb. nov. Basionym: Chiloglottis triceratops D.L.Jones, Contributions to Tasmanian Orchidology- 3: A Taxonomic Review of Chiloglottis R.Br. in Tasmania, Australian Orchid Research 3: 66 (1998). Note : Simpliglottis Szlachetko differs from Chiloglottis in a number of important morphological characters. In Simpliglottis the leaves are generally broader and lack undulate margins, the scape is generally shorter (although it does elongate after anthesis) and stouter, the flower is usually larger, the petals are spreading or incurved (deflexed against the ovary in Chiloglottis ), the labellum is extremely mobile (more or less fixed in Chiloglottis), elliptic, ovate or cordate in shape (rhomboid or trapezoid in Chiloglottis) and the lamina calli are generally fewer, less crowded and of fairly uniform appearance. The results of recent molecular studies conducted on the group by Jones et al. (2002) demonstrate monophyly for Simpliglottis within Chiloglottis sens. lat. although they have chosen to recognise Simpliglottis at the subgeneric level only. Acknowledgments My thanks go to David Jones (CANB), Jim Ross (MEL) and Neville Walsh (MEL) for kindly checking an earlier draft of this paper. References Hopper, S.D. and Brown, A.P. (2000). New Genera, Subgenera, Combinations, and Species in the Caladenia Alliance (Orchidaceae: Diurideae). Lindleyana 15, 120-126. Hopper, S.D. and Brown, A.P. (2001a). Contributions to Western Australian Orchidology: 1. History of early collections, taxonomic concepts and key to genera. Nuytsia 14, 1-26. Hopper, S.D. and Brown, A.P. (2001b). Contributions to Western Australian Orchidology: 2. New Taxa and Circumscriptions in Caladenia (Spider, Fairy and Dragon Orchids of Western Australia). Nuytsia 14, 27-307. Jeanes, J.A. and Backhouse, G.N. (2001). Wild Orchids of Victoria, Australia. Zoonetic: Seaford. Jones, D.L., Clements, M.A., Sharma, I.K. and Mackenzie, A.M. (2001). A New Classification of Caladenia R.Br. (Orchidaceae). Orchadian 13, 389^-17. Jones, D.L., Clements, M.A., Sharma, I.K., Mackenzie, A.M. and Molloy, B.P.J. (2002). Nomenclatural Notes Arising from Studies into the Tribe Diurideae (Orchidaceae). Orchadian 13(10), 437-468. Szlachetko, D.L. (2001a). Genera et Species Orchidalium. 1. Polish Botanical Journal 46(1), 11-26. Szlachetko, D.L. (2001b). Nomenclatural adjustments in the Thelymitroideae (Orchidaceae). Polish Botanical Journal 46(2), 137-144. Muelleria 16: 83-86 (2002) Nymphoides simulans (Menyanthaceae): a new species from northern Australia Helen I. Aston National Herbarium of Victoria, Royal Botanic Gardens Melbourne, Birdwood Avenue, South Yarra, Vic. 3141. Helen.Aston@rbg.vic.gov.au Abstract Nymphoides simulans Aston sp. nov., a white-flowered species from northern Queensland and north-eastern Northern Territory, is described and its diagnostic features are discussed. The species is almost identical with N. spongiosa Aston in leaves and seeds, but differs from that species in sev¬ eral floral features. Introduction This paper is the sixth in a series connected to a revision of Nymphoides Seguier in Australia. Ten new species have been published in prior papers (Aston 1982, 1984, 1986, 1987, 1997) but material of the species now being described has been set aside for some time awaiting fuller investigation. Recent examination of seeds by Scanning Electron Microscope (Aston, in preparation) has assisted in clarifying the taxon’s status. Nymphoides simulans Aston belongs in the informal ‘indica group’ defined in Aston (1982, p. 35). Taxonomy Nymphoides simulans Aston, sp. nov. Nymphoides spongiosa Aston simulans arete habitu, foliis, fructis et seminibus, sed floribus 4-partitis (interdum 3-partitis vel 5-partitis) homostylis, antheris stigmata aequantibus praecipue differt. Type : Queensland, Cape York Peninsula, 3.2 km E of “Musgrave” along the “Musgrave” to “Marina Plains” road, 13 May 1982, H.l. Aston 2255 (holotype MEL 612170; isotypes BRI, MEL 612171, MEL spirit). Apparently annual. Petiole-like stems few to many (>20), arising from the plant base, slender, flexuose, 3.5-42 cm long, <1 mm diam. True petiole apparently absent. Leaf blades floating, elliptic to elliptic-oblong to broad-ovate in outline, deeply cordate (the lobes c. (25%-)40%-50%(-55%) of the total blade length and separated by a sinus of (20°-)30°- 75°(-100°) angle), obtuse, entire, (1.0-)2.0-3.5(-4.2) cm long, (0.7-)L5-2.5(- 3.5) cm wide, deep green or maroon-purple and shining above, white-translucent and spongy beneath; spongy tissue smooth-surfaced, not rugose. Juvenile leaves basal, sub¬ merged, very thin-textured, light green and almost translucent, not spongy; blades vari¬ able in shape, narrow-elliptic, oblanceolate to obovate, or broad-ovate to deltoid, c. 7-22 mm long, tapered into dorso-ventrally flattened petioles c. 5-35 mm long. Inflorescence as for the ‘indica group’; pedicels subtended by ± broad-ovate, membranous, translucent bracts c. 2-3.5 mm long. Pedicels 5-ll(-20), emerging erect through the sinus when in flower, very slender, 4-15(-20) mm long, < 0.5 mm diam. Flowers (3)4(5)-partite, homostylous. Calyx lobes lanceolate, acute, membranous, mostly purplish-translucent, often slightly outcurved at the apex when in fruit, 1.6-2.7 mm long. Corolla (4-)5-7 mm span, white with a yellow throat. Corolla lobes narrowly oblanceolate (almost linear) to broadly oblanceolate; mid-section glabrous except for c. 8-20 fine papillae spaced in a transverse band just above its base, the band sometimes discontinuous, the papillae then only at the sides or at the sides and centre of the band position; central longitudinal keel 84 H.I. Aston or wing on mid-section absent; side-wings extending from the apex of the lobe to half or two-thirds of the lobe length, varying from near-entire with 1-few crenulations or short laciniae at the apex, to strongly crenulate or with several deep-cut laciniae spaced along the whole margin. Corolla tube papillae short, each several cells long, clustered c. 8-12 together at the summit of a short common stalk, or the cluster sometimes sessile. Stamens with straight filaments 0.50-0.85 mm long; anthers versatile, ± broad-obloid with the Figure 1. Nymphoides simulans. A-G Outline of floating blades of mature leaves, x 1: A ( Craven 3203), B-D {Aston 2251), E-F {Aston 2247), G {Aston 2255). H- M Outline of juvenile leaves, x 1: H-K {Aston 2255), L-M {Craven 3203). Nymphoides simulans 85 Figure 2. Nymphoides simulans. A-E Outline of corolla lobes to show the crenulate to laciniate wing margins, x 15: A-B {Aston 2247), C-E {Aston 2255). F Outline of gynoecium showing ovary shape and gentle contraction into the style, x 20 {Aston 2255). length slightly > to slightly < width, (0.25-)0.30-0.45 mm long, (0.225-)0.25-0.40 mm wide. Gynoecium c. (1.20-)1.3-1.5(-1.7) mm long; ovary obovoid to broadly obovoid, the shoulders at the summit contracted gently rather than abruptly into the style; placentas 2, short, about one-quarter to one-third of the capsule length, positioned centrally down the ovary wall; ovules c. 5-19; style (0.15-)0.20-0.25(-0.30) mm long; stigmas 2, each a broad, papillate, slightly lobed wing c. (0.20-)0.25(-0.35) mm long. Capsule broadly obovoid or sometimes broadly obloid, from a little less than to equal to or rarely greater than the calyx, 2.0-2.8 mm long, 1.4-2.05 mm wide. Seeds (4-)7-19 per capsule; body of seed near-globose but slightly to moderately laterally compressed, 0.575-0.85 mm long, 0.55-0.775 mm wide, 0.35-0.60 mm thick, cream-straw to light brown-grey or brown- black when mature, usually densely covered with very short, usually broadly conical but obtuse, occasionally convex, tubercles which arise one from each cell and give the seed surface a densely granular appearance; seed faces sometimes smooth, the tubercles pres¬ ent only on or close to the seed edges and gradually diminishing in length from the edge towards the face; basal caruncle present, circular, pale, typically thick and conspicuous, sometimes thin. (Figs 1& 2). Phenology: Flowers and fruits recorded April to June. Etymology: The specific epithet refers to the similarity of N. simulans to N. spongiosa except in floral characters. Distribution and Conservation Status: Known from the Northern Territory and Queensland, north of 18°S latitude and east of 132°E longitude. Apparently most prolific 86 H.I. Aston in Queensland, on Cape York Peninsula, particularly in the “Musgrave” to Hann River region but extending south to about Mt Molloy. There are only four widespread records from the remainder of the known range, namely two from the East and South Alligator River regions of the Northern Territory, and two from the southern edge of the Gulf of Carpentaria (Bing Bong, N.T. and near Westmoreland, Qld). Nymphoides simulans is probably well under-collected across its wide geographical range and is apparently under no particular conservation threat at this time. Habitat: Occurs in still, sometimes gently flowing, temporary fresh waters to 50 cm deep in seasonally flooded swamps, lagoons, or roadside ditches, persisting for a while on saturated marginal zones as waters recede. Substrate usually sand or sandy gravel, sometimes mud. Recorded “in a sedge swamp through a Melaleuca flat” ( Aston 2259) and in “ Melaleuca viridiflora closed woodland on a depression” (A. V. Slee et al 2880). In two populations {Aston 2247 and Aston 2255) N. simulans occurred intermingled with N. triangularis Aston and in another population ( Aston 2251) it was intermingled with N. parvifolia (Griseb.) Kuntze. Notes’. Nymphoides simulans resembles N. spongiosa (Aston 1982, pp. 45-48, fig. 4) very closely in habit, leaf, fruit and seed, but differs chiefly in having 4-partite flowers (occasionally 3- or 5-partite), and in being homostylous with anthers and stigmas held at the same level at anthesis. There are also other floral differences. Flowers which I have seen either in the field, or in spirit, or softened from dried collections, are smaller (corolla span c. 5-7 mm) than those from N. spongiosa (10-20 mm, occasionally as short as 7 mm), and have the wings of the corolla lobes usually shorter in relation to the lobe length, narrower, and crenulate to laciniate rather than entire. Anthers of N. simulans are smaller (0.25-0.45 mm long) and squatter, somet im es even wider than long, whereas those of N. spongiosa are about 0.5-0.7 mm long and 1.5 t im es as long as broad. The ovary of N. simulans also dif¬ fers, that of N. spongiosa being more globular and abruptly contracted into the style. Representative Specimens Examined (17 collections examined): Queensland: Cape York Peninsula, 5.3 km NW of the Hann River crossing of the Laura to Coen road, 11 May 1982, H.I.Aston 2247 (MEL); 11.6 km SE of the Morehead River crossing of the Laura to Coen road, 12 May 1982, H.I.Aston 2251 (BRI, MEL); 5.4 km S of Musgrave along the Laura to Coen road, 14 May 1982, H.I.Aston 2264 (BRI, MEL, NSW); About 140 km S of Cooktown on the Cairns road, 9 April 1975, L.A.Craven 3203 (BRI, CANB, MEL); “Westmoreland” homestead lagoon, 11 May 1974, S.Jacobs 1547 (MEL, NSW). Northern Territory: Bing Bong Station, 8 June 1971, C.R.Dunlop 2250 (DNA); Grown [at CSIRO Berrimah] from soil collected on East Alligator floodplain, 12° 14’ S, 133°01’ E, 29 April 1994, S.Seddifield s.n. (DNA). Acknowledgements I thank Neville Walsh for preparation of the Latin diagnosis from my English draft, and Mali Moir for preparation of the figures. References Aston, H.I. (1982). New Australian species of Nymphoides Seguier (Menyanthaceae). Muelleria 5, 35-51. Aston, H.I. (1984). Nymphoides triangularis and N. elliptica (Menyanthaceae): two new Australian species. Muelleria 5, 265-270. Aston, H.I. (1986). Nymphoides disperma (Menyanthaceae): a new Australian species. Muelleria 6, 197-200. Aston, H.I. (1987). Nymphoides beaglensis (Menyanthaceae). A new Australian species. Muelleria 6, 359-362. Aston, H.I. (1997). Nymphoides spinulosperma (Menyanthaceae): a new species from south-east¬ ern Australia. Muelleria 10, 21-25. Muelleria 16: 87-112 (2002) Variation within Asterolasia asteriscophora sensu lato (Rutaceae: Boronieae) and the recognition of new taxa in eastern Australia Bryan J. Mole l,3 ; Marco F. Duretto 2 Pauline Y. Ladiges 1 and Elizabeth A. James 2 1 School of Botany, The University of Melbourne, Vic. 3010 2 National Herbarium of Victoria, Royal Botanic Gardens Melbourne, Birdwood Avenue, South Yarra, Vic. 3141 3 Author for correspondence: b.mole@pgrad.unimelb.edu.au Abstract Asterolasia asteriscophora (F.Muell.) Druce sensu lato exhibits considerable variation in a range of floral and vegetative characters. Phenetic analysis of 14 morphological characters for 17 popu¬ lations of A. asteriscophora sensu lato has elucidated three new taxa: A. asteriscophora subsp. alb- iflora B.J.MoIe, A. rupestris B.J.Mole and A. rupestris subsp. recurva B.J.Mole, and supports the recognition of A. buckinghamii (Blakely) Blakely and A. buxifolia Benth. as distinct species. Isozyme electrophoresis and subsequent interpretation proved problematic, however allele fre¬ quency data for a limited number of enzyme systems were broadly congruent with morphometric analyses. A key and detailed taxonomic descriptions are provided. Introduction Asterolasia asteriscophora (F.Muell.) Druce is a variable taxon of eastern Australia cur¬ rently suspected to contain several infraspecific taxa (Wilson 1970, 1980, ined. ; Duretto 1999). The species was first described by Ferdinand von Mueller (1855) as Phebalium asteriscophorum F.Muell., from material he collected in 1852 at Mt Disappointment, north of Melbourne, Victoria. Bentham (1863) placed it in Asterolasia F.Muell. as A. muelleri Benth. Druce (1917) noted that Bentham had not used the correct epithet when transferring the species to Asterolasia and so made the correct combination. As currently circumscribed (see Porteners 1991; Duretto 1999), A. asteriscophora is a slender, erect, many-branched shrub, 1-2 m in height. Leaves, stems and the abaxial surface of the petals are covered with an indumentum of stalked, multiangular, stellate tri- chomes. The leaves are highly variable in shape and size, spathulate to obovate, oblong- cuneate or elliptic, 3-34 mm long, and 2-12 mm wide. A form with obcordate leaves is known from Mt Kaputar National Park in New South Wales. The inflorescences are ter¬ minal or axillary, solitary or in few flowered umbels, with the terminal flower reaching anthesis first, followed by the lateral flowers. Petals are usually yellow though plants from the Emerald area (Vic.) have white petals. The range of the species’ extends from the Macedon Ranges in Victoria to the Torrington district in northern New South Wales (Fig. 1). Although the species is widespread, populations are generally small and disjunct. The form of A. asteriscophora with white petals noted by Wilson (ined.) and Duretto (1999) from Emerald, Victoria, has been described as Eriostemon spathulifolius Gand. (Gandoger 1913), but has not been formally recognised in recent flora treatments (Willis 1973; Duretto 1999). This form is located within a residential zone and is now only known from three localities in the Emerald-Avonsleigh district, c. 45km ESE of Melbourne, Victoria. Formal taxonomic recognition of this form would have immediate conservation implications as the populations are threatened directly and indirectly by res¬ idential development. Wilson (1998) noted that in the Flora of New South Wales (Porteners 1991) A. buxi¬ folia Benth. was not accounted for. This is also the case for A. buckinghamii (Blakely) Blakely, although this was not noted by Wilson (1998). Wilson (1970), however, listed A. buckinghamii as an accepted taxon. In order to resolve taxonomic problems in the A. asteriscophora species group, specimens were subjected to a phenetic analysis. It was B .J. Mole et al. Figure 1 . Known distribution of Asterolasia asteriscophora sensu lato (black circles), A. buxifolia (shaded circle), and A buckinghamii (open circles). All popula¬ tions were sampled in this study except for Mt Canobolas, Upper Genoa River, and A. buxifolia, which are represented by herbarium specimens. The black square represents the single population of A. correifolia sampled for isozyme analysis. Under the new classification proposed here, populations from Mt Kaputar and Mt Canobalas are A. rupestris subsp. rupestris', those from Parlour mountain are A. rupestris subsp. recurva; those from Emerald are A. asteriscophora subsp albiflora\ other populations from Victoria and NSW are A. asteriscophora subsp asteriscophora. evident prior to analyses that A. buxifolia was distinct based on its glabrous ovaries, (character 13), however, as this taxon had apparently been treated as conspecific with A. asteriscophora, it was included in analysis 1. Methods and Materials POPULATION SAMPLING Seventeen populations representing the known geographic range of A. asteriscophora sensu lato (Fig. 1) were sampled between November 1998 and January 1999. For the morpho¬ metric analysis, five individuals were sampled from each population. The exception was the population at Pine Mountain which consisted of only six individuals, from which only a small sample from one individual was taken. Herbarium specimens from BRI, CANB, MEL, MELU, NE, NSW, and K were used to supplement field collections for areas not sampled during field work. Herbarium acronyms follow Holmgren et al. (1990). A subset of 12 populations was sampled for isozyme analysis. For each population, 3-12 leaves were collected from 15 individuals. Fifteen individuals of a closely related but morphologically distinct species, A. correifolia (A.Juss.) Benth., were also sampled for comparison. Three voucher specimens were collected for each population. Variation in Asterolasia asteriscophora s.l. 89 MORPHOLOGICAL CHARACTERS One hundred and sixteen specimens (Appendix 1) were scored for 14 morphological characters (Table 1) reflecting the variation in leaf shape, leaf size, density and type of indumentum, petiole length, petal colour, peduncle and pedicel length, and number of flowers per inflorescence. All continuous characters were scored as an average of meas¬ urements from five organs (where five organs were available). Most characters are self explanatory, but a few require further clarification. The prophylls of A. asteriscophora sensu lato are leaf-like and may be confused with leaves. They can be distinguished on the basis of their position. Prophylls subtend the pedicels, and are similar in shape and size to one another but consistently different from leaves subtending an inflorescence or axillary shoot. Leaves subtending an inflorescence or axillary shoot were interpreted as true leaves. Leaf size is highly variable within populations and appears to be related to the age of the plant. Relatively young, vigorous plants tend to have larger leaves than more mature plants. Only mature specimens were used in this analysis. Ratio characters, leaf length deaf width (character 2), leaf length: distance to widest point from leaf base (character 3), and petal length:petal width (character 10), were used to quantify the shape of leaves and petals respectively. Since the shape of leaves and petals was consistent on any given individual, length and width measurements were con¬ sidered to be measuring the same character (leaf size). Therefore width measurements for both leaves and petals were removed prior to analysis. Table 1. Characters scored for morphometric analyses. Characters were used in both analyses unless indicated Continuous characters 1. Leaf length (along midvein) (mm) 2. Leaf length : leaf width 3. Leaf length : distance to widest point from leaf base (mm) 4. Length of petiole (mm) 5. Trichome density (per mm square) on adaxial leaf surface 6. Average number of flowers per unit inflorescence 7. Length of peduncle at anthesis (mm) 8. Length of pedicel at anthesis (mm) 9. Length of petal (mm) 10. Petal length : petal width Binary characters 11. Leaf margins recurved/not recurved 0/1 12. Indumentum on abaxial petal surface multiangular/globular stellate 0/1 (analysis 1 only) 13. Ovary glabrous/stellate indumentum 0/1 (analysis 1 only) Multistate characters 14. Petal colour white(0)/pale lemon(l)/yellow (2) Trichome density of the adaxial surfaces of leaves (character 5) was calculated using transparent grid paper (1 mm 2 ). The grid was placed over a leaf surface midway along the leaf, avoiding the midrib, and the number of trichomes in five separate grids were count¬ ed, and the average recorded for each leaf. The number recorded for each specimen is the average of five leaves. 90 B .J. Mole et al. Peduncle length (character 7) and pedicel length (character 8) showed considerable variation within a single individual but meaningful comparisons could be made by meas¬ uring only those subtending flowers at anthesis. Generally, pedicel length was greater at anthesis than in bud and greater still in fruit. Some specimens had solitary flowers while others had umbels of three or more flowers. When specimens had umbels of three or more flowers, the pedicel of the central flower was measured for consistency. SCANNING ELECTRON MICROSCOPY Adaxial and abaxial leaf surfaces of specimens were examined to ascertain the structure of trichomes present, and to illustrate variation in trichome density. Leaves stored in 70 % ethanol were air dried and mounted on stubs with the aid of carbon adhesive discs and conductive silver paint. Samples were coated with gold using an Edwards Sputter Coater S150 B and examined and photographed using a Phillips XL 30 Field Emission Scanning Electron Microscope. MULTIVARIATE PATTERN ANALYSIS Multivariate pattern analyses were completed using the computer package PATN (Belbin 1987). A distance matrix was constructed using the Manhattan metric distance measure. The Manhattan metric depends on the range of all characters used for its scaling factor (Sneath and Sokal 1973), therefore, all data were range standardised prior to each analy¬ sis to ensure all characters were of equal weight (Sneath & Sokal 1973; Milligan & Cooper 1988). Two sets of analyses were performed, the first included all specimens, while the second excluded specimens of A. buxifolia. Both sets of analyses included an Unweighted Pair Group Method of Averaging (UPGMA) hierarchical classification to identify discrete clusters of individuals and a Non Metric Multidimensional Scaling (NMDS) ordination to identify continuous variation and clusters of individuals. Co-phenetic correlation coefficients for dendrograms were calculated using a Spearman Rank correlation coefficient in the computer program SPSS (Norusis 1990). Cramer values, which give a measure of the discriminating powers of characters for each group identified in a classification (Belbin 1987), were also calculated. Cramer values have a scale of 0-1, with a higher value indicating a higher discriminating power for the relevant character. Twenty different starting configurations were used in each NMDS analysis to ensure that the ordination which best fits the data was obtained. This is identified by the run with the lowest stress value, which is an indication of the degree of correlation of the distances between individuals in the ordination and distances in the original distance matrix. There was little difference in stress values for all twenty runs, indicating global minima were obtained (Kruskal & Wish 1973). Three-dimensional and two-dimensional analyses were performed. Because three dimensional ordinations provided only marginally greater information on groups than two-dimensional ordinations, only two-dimensional ordina¬ tions are presented here. A Minimum Spanning Tree (MST) was also calculated. A MST connects all individ¬ uals with single linkages so that the shortest possible tree is constructed with no closed loops. It gives a more accurate portrayal of inter-entity distance than an ordination, which loses some of this information, particularly when the number of dimensions used is low (Belbin 1987). ISOZYMES One hundred and ninety five individuals from 13 populations (15 individuals per population) were assayed for isozyme polymorphisms. Small pieces of leaf were ground in the grinding buffer of Warburton et al. (2000) in Eppendorf tubes using a hand held electric drill. Protein electrophoresis was carried out using the Titan III cellulose acetate system (Helena Laboratories) as described by Coates (1988). Thirteen different enzyme systems often found to be polymorphic in plants were tested using three continuous running Variation in Asterolasia asteriscophora s.l. 91 buffers: (1) TEM, (2) CM (Warburton et al. 2000) and TG (Hebert & Beaton 1989). Only four enzymes run in TG running buffer gave reliable banding patterns for the majority of samples: aspartate aminotransferase (AAT, EC 2.6.1.1), fumarate (FETM, EC 4.2.1.2), glucose-6-phosphate isomerase (GPI, EC 5.3.1.9) and phosphoglucomutase (PGM, EC 5.4.2.2). After electrophoresis for 20-30 min, depending on the enzyme, gels were stained using an agar overlay of 6 ml of 1% agar at 45°C added to the appropriate stains, described by Hebert and Beaton (1989) but scaled down to a volume of 1.2 ml. Each zone of activity on gels was assumed to represent a single locus. Alleles were numbered according to their electrophoretic mobility and recorded in a table for further analysis. ANALYSIS OF ISOZYME DATA The BIOSYS-1 program of Swofford and Selander (1981) was used to analyse allele fre¬ quency data. Phenetic analysis was carried out using Nei’s genetic distance (Nei 1978) and the UPGMA hierarchical clustering technique of Sneath and Sokal (1973). Phylogenetic analysis was carried out using the modified Rogers’ Distance (Wright 1978) and the Distance Wagner method. Both methods gave similar results, and only the Distance Wagner method is presented here. TAXON DESCRIPTIONS Descriptive terminology follows Weston (1990) for inflorescence structure, though we use pedicel instead of anthopodium, and Hewson (1988) and Theobald et al. (1979) for trichomes. Conservation codes follow Briggs and Leigh (1996). CULTIVATION Plants from Emerald and Torrington, with extremes in trichome density (measured on the adaxial leaf surface), were propagated asexually. Cuttings were prepared from firm young growth and treated with Indole Butyric Acid 3000 ppm. Cuttings formed adventitious roots in 4-6 weeks. These were grown under similar conditions in a glasshouse to deter¬ mine if trichome density reflects genetic differences or is a response to environmental conditions. Results and Discussion MORPHOMETRIC ANALYSIS Analysis 1 (specimens 1-116, characters 1-14): Analysis 1 included 116 specimens of A. asteriscophora, A. buckinghamii and A. buxifolia. Both the UPGMA classification (Fig. 2) and the ordination (Fig. 3) indicate two major groups (groups A and B). The high co-phenetic correlation coefficient for the classification (0.722) and the low stress value for the ordination (0.182) indicate a minimal and acceptable loss of information from the original distance matrix. Group A incorporates two type specimens (specimen 116 - lec- totype) and (specimen 115 - residual syntype) of A. buxifolia from the Blue Mountains, New South Wales. They are distinct from all other specimens in having a glabrous ovary (character 13), a glabrous adaxial leaf surface (character 5), and globular stellate tri¬ chomes on the abaxial surface of petals (character 12). Group B (specimens 1-114) consists of specimens from New South Wales and Victoria characterised by a stellate indumentum on the ovary (character 13) and adaxial surface of the leaf (character 5), and stalked, multiangular stellate trichomes on the abax¬ ial surface of petals (character 12). The classification indicates that A. asteriscophora sensu lato contains a number of subgroups. While the ordination indicates A. bucking¬ hamii is a distinct group, other subgroups identified in the classification are not as clear¬ ly separated due to the compression of ordination space by the inclusion of A. buxifolia. Therefore the data were re-analysed omitting A. buxifolia. 92 B .J. Mole et al. Figure 2. Dissimilarity 0.46 — 0.37 _ 0.29 _ 0.20 _ 0.11 _ 0.0 _ 1-5,100, 103-1 OB 6 , 23 , 32 - 41 , 72 - 76 , 88-59, 03 , 95 , 101 12-13, 7-11, 57-66, 67-71, 115-116 15-22, 14,70, 109-114 102 24-31, SI-83. 42 - 56 , 05 - 37 ' 00.84. 90-92, 94. 96-99 Group B Group A UPGMA classification, analysis 1, specimens 1-116, truncated as indicated by the dashed line. Group A: Asterolasia buxifolia. Group B: A. bucking- hamii, and A. asteriscophora. Co-phenetic correlation coefficient = 0.722. o°& ; oo • i • '• Group B ■<. # # % • U 11 i Group A •* • 11 • Figure 3. NMDS ordination, vector I versus vector II, analysis 1, specimens 1-116. Symbols are: Asterolasia buxifolia. Group A (shaded circles); A. bucking- hamii. Group B (open circles); A. asteriscophora sensu lato Group B (solid circles). Stress = 0.182. Variation in Asterolasia asteriscophora s.l. 93 Figure 4. UPGMA classification, analysis 2, specimens 1-114. Co-phenetic correla¬ tion coefficient = 0.707. Numbered specimens are referred to in text. i Figure 5. NMDS ordination, vector I verses II, analysis 2, specimens 1-114. Symbols are: Asterolasia buckinghamii. Group B1 (open circles); Group B2.1 (solid circles); Group B2.2 Melbourne region (solid squares); Group B2.2 East Gippsland and southern NSW (open squares); Group B2.2 Torrington (shaded squares); Group B3.1 (open triangles); and Group B.3.2 (solid triangles). Numbered specimens referred to in text. Stress = 0.182. 94 B .J. Mole et al. Figure 6. Minimum spanning tree, Analysis 2, specimens 1-114. Symbols are: Asterolasia buckinghamii, Group B1 (open circles); Group B2.1 Emerald (solid circles); Group B2.2 Melbourne region (solid squares); East Gippsland and southern N.S.W. tablelands (open squares); Group B.2.2 Torrington (shaded squares); Group B3.1 (open triangles) and Group B3.2 (solid trian¬ gles). Numbered specimens are referred to in text. Analysis 2 (specimens 1-114, characters 1-11, 14): Analysis 2 included the 114 spec¬ imens in group B of analysis 1. Characters 12 and 13 were invariate after removal of A. buxifolia and were deleted. Data were range standardised before reanalysis. Five groups are evident from the UPGMA classification (Fig. 4), ordination (Fig. 5) and MST: three major groups (groups Bl, B2 and B3) and two pairs of subgroups (B2.1 & B2.2 and B3.1 & B3.2). The high co-phenetic correlation coefficient for the UPGMA classification indi¬ cates that the branch lengths in the dendrogram are highly correlated with the distances in the original distance matrix. Group Bl (specimens 57-66, 109-114) corresponds to A. buckinghamii. Groups B2 and B3 represent two different forms within A. asteriscophora sensu lato. Group B2 includes specimens from the Emerald district of Victoria (B2.1) and a widespread group that contains the lectotype of A. asteriscophora (B2.2). Group B3 includes specimens from the Armidale region of New South Wales (B3.1) and the Mt Kaputar and Mt Canobolas districts of New South Wales (B3.2). In the ordination, spec¬ imens from Torrington are positioned on the periphery of group B2.2 (typical A. aster¬ iscophora), and adjacent to group B3 (Mt Kaputar and the Armidale region) indicating they have similarities to both groups. However, the Torrington specimens cluster with Victorian and southern New South Wales specimens in group B2 in the UPGMA classi¬ fication (Fig. 4), which is supported by the Minimum Spanning Tree (Fig. 6). The positions of specimens 100, 101 and 107 are problematic. Specimen 101 from Deepwater in New South Wales is positioned between B3.2 and B2.2 in the MST (Fig. 6) Variation in Asterolasia asteriscophora s.l. 95 but is included in B2.2 because all other specimens from this region clustered clearly within B2.2. Specimen 107 from Mt Canobolas in New South Wales is considered misplaced in the MST where it is linked to specimens of group B2.2, as it clusters with¬ in group B3.2 in the classification and ordination. Otherwise, the MST supports the groups identified in both the classification and the ordination. Specimen 100 from Cabramurra in New South Wales is clustered in group B3.2 in all analyses. However, it does not fit well in this group because, other than the high trichome density and short petiole, it is very similar to specimens in group B2.2, and all other specimens from this region are clustered within B2.2. Means and Cramer values for the most discriminating continuous characters between groups are given in Table 2. Table 2. Cramer values, means and ranges for the four most discriminating continuous characters between groups. Character Cramer Value Bl.l Group Number B2.1 B2.2 B3.1 B3.2 Petiole length (mm) 0.7263 Mean 3.0 2.5 3.7 0.0 0.5 (character 4) Maximum 6.1 3.7 7.0 0.0 1.7 Minimum 1.8 1.7 1.3 0.0 0.0 Peduncle length (mm) 0.8192 Mean 1.0 4.6 7.1 5.8 5.9 (character 7) Maximum 1.7 4.9 14.3 6.3 7.7 Minimum 0.0 3.8 4.6 5.5 4.8 Pedicel length (mm) 0.7404 Mean 1.4 8.5 7.8 8.4 11.0 (character 8) Maximum 1.9 11.5 3.8 9.3 14.3 Minimum 0.9 5.7 15.4 7.3 8.9 Flower number 0.8469 Mean 1.1 4.2 4.6 4.6 4.1 per inflorescence Maximum 2.2 4.8 6.6 5.7 6.0 (Character 6) Minimum 1.0 3.3 3.0 4.0 3.0 Group B1 (specimens 57-66, 109-114) includes two syntypes (specimens 109 and 110) of A. buckinghamii and is distinct from other groups in the analysis by the combi¬ nation of the following features: flowers solitary or almost so (character 6); a reduced or absent peduncle (character 7; mean 1 mm); a reduced or absent pedicel (character 8; mean 1.4 mm); and petiolate leaves (character 4; mean petiole length 3 mm). Group B2 (specimens 6-56, 72-101) may be distinguished from group B1 in com¬ bining the following features: inflorescence an umbel of three or more (character 6), a peduncle length of 3.5-15 mm (character 7), and a pedicel length of 3.5-15.5 mm (char¬ acter 8); from group B3 by the longer petiole length (character 4; range 1.7-7 mm), a lower average trichome density on the adaxial leaf surface (character 5), and leaf shape. Group B2.1 (specimens 7—11, 85-87) consists of those specimens that have white or, very rarely, pale lemon petals (character 14), short petioles (character 4; mean length 2.5 mm), and an obovate to spathulate lamina with a rounded apex. Group B2.2 (specimens 6, 12-56, 72-84, 88-99, 101) includes the lectotype (speci¬ men 83), and two isolectotypes (specimens 81 and 82) of A. asteriscophora. This group is defined by specimens with bright yellow flowers in umbels of three or more (character 6; mean 4.6), long peduncles (character 7; mean 7.1 mm), long pedicels (character 8; mean 7.8 mm ) and petiolate leaves (character 4; mean petiole length 3.7 mm ) that are elliptic to spathulate, with a rounded apex. Group B3 (specimens 1-5, 67-71, 100, 102-108) is readily distinguished from group B1 in combining the following features: inflorescence an umbel of three or more (char¬ acter 6), a peduncle length of 4.5-8 mm (character 7), and a pedicel length of 7-15 mm 96 B .J. Mole et al. (character 8); from group B2 by the short or absent petiole (character 4) and a higher average trichome density on the adaxial leaf surface (character 5). Group B3.1 (specimens 67-71, 102) contains specimens distinguished by the combi¬ nation of the following characters: leaves sessile or almost so (character 4), obcordate to obdeltate, margins strongly recurved (character 11), high trichome density (character 5; mean 21 trichomes per mm 2 ), and a shorter leaf length/distance to widest point (charac¬ ter 3) than all other groups. Group B3.2 (specimens 1-5, 100, 103-108) is defined by the same characters as group B3.1, but differs in the subsessile leaves (character 4, mean petiole length 0.5 mm), and the non-recurved leaf margins (character 11). These five groups (Bl, B2.1, B2.2, B3.1 and B3.2) correspond to geographic regions. Specimens in group B1 are all from the Penrose, Wingello and Berrima districts of the cen¬ tral tablelands of New South Wales. Specimens in group B2.1 are all from a small area in the Emerald-Avonsleigh district of Victoria. Group B2.2 has the largest distribution, rang¬ ing from the Macedon district north of Melbourne to the southern tablelands of New South Wales. There is a disjunct occurrence of group B2.2 at Tonington in northern New South Wales. Interestingly, this population is more similar to specimens from the southern New South Wales tablelands and Victoria than it is to other northern New South Wales popula¬ tions (group B3). Specimens in group B3.2 are restricted to Mt Kaputar NP in northern New South Wales, while specimens in group B3.1 occur near Palour Mountain north west of Armidale in New South Wales. The recognised groups do not overlap geographically. Figure 7. Adaxial leaf surfaces, illustrating variation in density of stalked, multiangu- lar, stellate trichomes. (a - d) Asterolasia asteriscophora sensu lato (e - f) A. buckinghamii : (a) specimen 11 from Emerald; (b) specimen 77 from Wallaby Creek; (c) specimen 2 from Mt Kaputar; (d) specimen 73 from Torrington; (e) specimen 62 from Mittagong; and (f) specimen 57 from Medway. Variation in Asterolasia asteriscophora s.l. 97 The ordination identifies variation in trichome density, with a trend in Groups B2 and B3 of increasing density on the adaxial leaf surface correlated with decreasing latitude, although latitude per se is unlikely to be the causal factor. Specimens from the greater Melbourne region have the lowest average trichome density, while specimens from Torrington and Mt Kaputar have the highest (Fig. 7). Indumentum density correlating with latitude change was also observed in Boronia grandisepala F.Muell sensu lato. (Duretto & Ladiges 1997), where increased indumentum density correlated with increased aridity. Cultivated A. asteriscophora plants originating from Torrington and Emerald growing under controlled glasshouse conditions were observed to have similar trichome densities to specimens sampled from those populations in the field. ISOZYME ANALYSIS Four enzyme systems corresponding to seven loci were resolved. These were: AAT (one locus); FUM (two loci); GPI (two loci); and PGM (two loci). An additional locus for AAT was scored for phenotype only. The low number of enzyme systems resolved is insuffi¬ cient to describe genetic diversity between populations. However, with the exception of A. buckinghamii, the results broadly correlate with morphological results. Problems were encountered with the resolution of some enzyme banding patterns making interpretation of allele mobility difficult. A characteristic of the Rutaceae is the presence of aromatic volatile oils in the pellucid oil glands of the leaves. It appears that these oils and/or other unknown compounds have co-migrated with some alleles during electrophoresis making interpretation of band patterns difficult. Similar problems to those encountered in this study have been documented in other studies of Rutaceae (eg. Durham 1998). There were difficulties in particular with interpreting the AAT system. AAT is a dimeric system where a homozygote is represented by one band and a het¬ erozygote by three. The number of bands for individuals ranged from two to four. However, it was decided to interpret the different banding patterns as phenotypes, and record the frequencies of each. This method has previously been used by Faville et al. (1995) in a study on allozyme variation in the grass Agrostis capillarisL. In addition, the need to freeze some samples reduced enzyme activity in those samples and resulted in fewer enzyme systems being satisfactorily resolved. Consequently, only results for pop¬ ulations from the Melbourne region are presented here, although allele frequency data for all populations are recorded in Table 3. A Distance Wagner Tree (Fig. 8) for populations of A. asteriscophora in the greater Melbourne region based on four enzyme systems and seven loci (Table 3) indicates that these populations are genetically very similar. The distances in the Wagner Tree correlate Distance from root .00 .02 .04 .06 .09 .11 .13 I-1-1-1-1-1-1 Emerald -Toolangi Wallaby Creek Gladysdale Mt To wrong Total length of tree = .369 Figure 8. Wagner Tree for isozyme data rooted at the midpoint for five populations of Asterolasia asteriscophora in the Melbourne region. Co-phenetic correlation coefficient 0.974. 98 B .J. Mole et al. with the geographic isolation of these populations from one another. The Mt Towrong population near Macedon is the most distinct and also the most geographically isolated population. The population with white petals from Emerald is not distinct based on these four enzyme systems, but is distinct based on analysis of phenotype for AAT. Table 3. Allele frequencies for 13 populations of Asterolasia. Sample size per locus = 15 individuals. Populations are: 1-7 A. asteriscophora sensu lato Victorian popula¬ tions, 1 Emerald; 2 Gladysdale; 3 Toolangi; 4 Wallaby Creek; 5 Mt Towrong; 6 Mt Bowen; 7 Ben Cruachan Creek. 8-12 A. asteriscophora sensu lato New South Wales populations, 8 Mt Kaputar; 9 Torrington; 10 Mittagong; 11 Medway; 12 Tumut River. 13 A. correifolia Boonoo Boonoo New South Wales. Locus Allele Population number 1 2 3 4 5 6 7 8 9 10 11 12 13 Pgm-1 1 0 0 0 0 0.466 0 0.833 0 0.5 0.04 0 0.133 0 2 1 1 1 0.844 0.533 0.833 0.177 0.1 0.5 0.966 1 0.877 1 3 0 0 0 0.166 0 0.177 0 0.9 0 0 0 0 0 Pgm-2 1 0 0 0 0 0.1 0 1 1 0 0 0 0 0.5 2 1 1 1 0.893 0.9 0.714 0 0 1 1 0.8 0.846 0.5 3 0 0 0 0.107 0 0.286 0 0 0 0 0.2 0.154 0 Gpi-1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 Gpi-2 1 0 0 0 0 0 0 0 0.466 0 0 0 0 0 2 0 0 0.033 0.2 0.2 0.233 0 0 0 0 0.7 0.1 0.5 3 0.633 0.8 0.466 0.5 0.733 0.777 0.633 0.544 0 0.633 0 0.633 0.5 4 1 0.166 0.266 0.1 0.033 0 0 0 0.433 0 0.044 0.1 0 5 0.233 0 0.166 0.166 0.033 0 0.377 0 0.577 0.377 0.266 0.177 0 6 0 0.033 0.066 0.033 0 0 0 0 0 0 0 0 0 Fum-1 1 0 0.033 0 0 0 2 0.133 0.033 0 0 0 3 0.867 0.933 1 1 1 Fum-2 1 0 0 0.066 0.033 0.167 2 0.9 1 0.934 0.967 0.833 3 0.1 0 0 0 0 Aat-1 1 1 1 1 1 1 Taxonomic conclusions Morphological analyses have contributed to elucidating two major (groups B2 & B3) and four minor groups (groups B2.1, B2.2, B3.1 & B3.2) within Asterolasia asteriscophora sensu stricto, and confirm that specific status for A. buxifolia (group A) and A. bucking- hamii (group Bl) is appropriate. Although there is insufficient information to describe genetic diversity between populations based on isozyme analyses due to the low number of enzyme systems resolved, fixed allele differences support the morphological findings. In the classification proposed here, Group B2 and B3 are recognised at the species level. Group B2 corresponds to A. asteriscophora, and the subgroups B2.1 and B2.2 are recognised here as subspecies. Group B2.2 contains the type for A. asteriscophora, there¬ fore it is designated A. asteriscophora subsp. asteriscophora. Group B2.1 is recognised here as A. asteriscophora subsp. albiflora B.J.Mole. It is given sub-specific rank because, other than petal colour, it is morphologically s imil ar to typical A. asteriscophora. Petal colour has been used as a character to delimit species of Asterolasia in Western Australia Variation in Asterolasia asteriscophora s.l. 99 (Wilson 1980). Specimens 85 and 87 (Fig. 4) sampled near the white-petalled population over 10 years ago, have flowers which are very pale lemon in colour, indicating speci¬ mens from this locality may grade into typical A. asteriscophora. During subsequent field work at these localities, however, only white-petalled individuals were observed. Some white-petalled specimens collected at this site during field work for this research ( Mole 73-77 ) had discoloured to a pale lemon colour 12 months after pressing. In some cases petals may turn pale lemon on drying, however information recorded on specimen 87 would indicate flower colour for that specimen was “pale lemon” at the time of collec¬ tion. Even so, it is considered by the authors and the collector of that specimen (Walsh 2001 pers. comm.) to be consistent in other characters with white flowering specimens from the same locality. Asterolasia asteriscophora subsp. albiflora is geographically iso¬ lated from the typical subspecies and commences flowering several weeks earlier than populations of the latter in the Melbourne area. This characteristic has been observed over several years (Gross 1998 pers. comm.). Isozyme phenotype differences for Aspartate aminotransferase locus 2 (. Aat-2 ) distinguish A. asteriscophora subsp. albiflora from the typical subspecies. Specimens from the Torrington region are assigned to A. asteriscophora subsp. aster¬ iscophora as the ordination for morphological data (Fig. 5) places them with, but peripher¬ al to, specimens in this taxon, despite their geographic proximity to group B3 described below. Both the MST and the classification also support the inclusion of Torrington speci¬ mens in subsp. asteriscophora. Isozyme frequency data (Table 3) also indicate that speci¬ mens from Torrington are more similar to southern populations of A. asteriscophora (although still distinct from them) than they are to the Mt Kaputar population (group B3.2). Group B3 constitutes a new species, Asterolasia rupestris B.J.Mole, and the two sub¬ groups within group B3 constitute subspecies. Asterolasia rupestris includes specimens from Mt Kaputar NP (group B3.2, specimens 1-5, 103-105), Mt Canobolas (group B3.2, specimens 106-108) and the Parlour Mountains (group B3.1, specimens 68-71, 102). Asterolasia rupestris differs from typical A. asteriscophora in having obcordate leaves and a reduced or absent petiole. Asterolasia rupestris subsp. rupestris (group B3.2), rep¬ resented by specimens from Mt Kaputar and Mt Canobolas, has plane leaf margins and A. rupestris subsp. recurva B.J.Mole. (group B3.1), represented by specimens from The Parlour Mountains near Armidale, has recurved leaf margins. The differences in leaf mar¬ gins were consistently distinct between specimens assigned to each subspecies. The spec¬ imens from Mt Canobolas in New South Wales are somewhat intermediate between A. rupestris and A. asteriscophora in that the leaves are narrowly obcordate. They are included in A. rupestris because of the short or absent petiole (character 4), high trichome density (character 5) and a leaf length:distance to widest point ratio (character 3) which is greater than that of A. asteriscophora. Key to the A. asteriscophora sensu lato group. 1. Flowers usually solitary and subsessile (rarely in umbels of two or three and then only occasionally on each individual plant); peduncles absent or to 2 mm long, pedicel length at anthesis 0-2 mm long 2. Adaxial surface of the leaves glabrous; abaxial surface of petals with an indu¬ mentum of sessile, globular stellate trichomes; ovary glabrous. 1. A. buxifolia 2: Adaxial surface of the leaves with an indumentum of stalked multiangular stellate trichomes; abaxial surface of petals with an indumentum of stalked multiangular stellate trichomes; ovary with an indumentum of multiangular, stellate trichomes .2. A. buckinghamii 1: Flowers in umbels of three or more (usually five); peduncles 3-15 mm long; pedicel length at anthesis 3-15 mm long 3. Feaves sublanceolate, elliptic-obovate, or sometimes spathulate, adaxial sur¬ face sparsely or densely covered with stellate trichomes, petiole (2-)4-7mm 100 B .J. Mole et al. long; petiole terete, not appressed to the stem; the base of lamina often adaxi- ally V-shaped in section, giving the appearance of an extended petiole. .3. A. asteriscophora 4. Petals bright yellow.3a. A. asteriscophora subsp. asteriscophora 4: Petals white, rarely pale lemon.3b. A. asteriscophora subsp. albiflora 3: Leaves obcordate to obdeltate, adaxial surface densely covered with stellate trichomes, sessile or petiole < 2 mm long, when present somewhat thickened and flat, often appressed to the stem; the lamina base not adaxially V-shaped in section.4. A. rupestris 5. Leaf margins not recurved.4a. A. rupestris subsp. rupestris 5: Leaf margins strongly recurved.4b. A. rupestris subsp. recurva Taxonomy 1. Asterolasia buxifolia Benth., FI. Austral. 1: 351 (1863); Eriostemon cunninghammii F.Muell., Fragm. 9: 107 (1875). Type: Bells Road, Blue Mountains, N.S.W., 1834, R.Cunningham (lectotype K, fide Wilson, Nuytsia 12: 83-88, 1998); Botanic Gardens Sydney, 1839, A.Cunningham s.n. (residual syntype MEL 4546). Shrub to 2 m high. Stems with a dense indumentum of stellate trichomes. Leaves obovate, 10-18 mm long, 3-10 mm wide, coriaceous, apex rounded or slightly emarginate; adax¬ ial surface glabrous; abaxial surface with an indumentum of stalked, multiangular, stel¬ late trichomes; petiole 2-7 mm long. Flowers axillary, solitary; peduncles absent; pedicels 1-1.5 mm long, subtended by two white petaloid bracts. Sepals inconspicuous, c. 1 mm long. Petals five, elliptic, 6-7 mm long, yellow; abaxial surface with an indu¬ mentum of sessile, globular, stellate trichomes; adaxial surface glabrous. Stamens 10, fil¬ aments glabrous; anthers 1—1.5 mm long, each with a terminal gland. Carpels five, glabrous; style glabrous; stigma hemispherical. Cocci glabrous, beaked. Seed not seen. Additional specimens examined: NEW SOUTH WALES: 1838, Anonymous (MEL 708653 ); “Port Jackson”, 1838, Theod. Scenes [sic.] ex herb. Sond. (NSW 468151 ); Blue Mountains, Vicary s.n. (MEL 708652, MEL 708654 ); New Holland, Anderson 52 (MEL 4827); Blue Mountains, Hartley area, October 2000, R.O.Makinson 1791 (CANB n.v., MEL, NSW n.v.). Distribution and conservation status: Asterolasia buxifolia was presumed extinct because it had not been located in the field since the 1830s and recent attempts to relo¬ cate it in the Blue Mountains had not been successful (Wilson 1998; Keith Ingram 1999 pers. comm.). However, a collection of the species that is consistent with the type speci¬ men was made in October 2000 in the Hartley area of the Blue Mountains (B. Makinson pers comm.; Auld 2001; Benson & McDougall 2001). A conservation code of 2E is con¬ sidered appropriate because the taxon is currently known from only one population of between 50-100 individuals (B. Makinson pers comm.). Habitat: The species is apparently restricted to rocky watercourses, with a granitic substrate. Phenology: Flowering specimens have been collected in October. Etymology: The specific epithet buxifolia presumably means foliage like the genus Buxus, however there is no particular resemblance between the leaves of A. buxifolia and those in the genus Buxus. 2. Asterolasia buckinghamii (Blakely) Blakely, Austral. Naturalist 11: 12 (1941); Phebalium buckinghamii Blakely, Austral. Naturalist 10: 246 (1940). Type citation: “Gold Gully, Penrose, W.F.Blakely, Jeane and W.J.Buckingham, and E.Murphy, 15/10/1938; 2 miles N.E. ofWingello railway station, same collectors, 30/9/1939.” Type: Gold Gully Penrose [N.S.W], 15.x. 1938, W.F.Blakely, J. and W.J.Buckingham and Variation in Asterolasia asteriscophora s.l. 101 E.Murphy (lectotype, here designated, NSW 2274; isolectotypes MEL 232746 , CANB 81750 n.v.); 2 miles [c. 3.2 km] NE of Wingello railway station, 30.x. 1939, W.F.Blakely, W.J.Buckingham and E.Murphy (residual syntype NSW 468135). Slender upright shrub to 1.5 m high. Stems covered with an indumentum of stellate tri- chomes. Leaves orbicular to broadly obovate, 4-15 mm long, 3-7 mm wide, lateral halves adaxially concave, apex rounded to slightly emarginate; adaxial surface with a sparse to dense indumentum of stalked, multiangular, stellate trichomes (range 1-23 tri- chomes per mm 2 ); abaxial surface with a cobwebbed indumentum of stalked, multiangu¬ lar, stellate trichomes; petiole 2-4 mm long. Flowers axillary or terminal, usually soli¬ tary, rarely in a two or three flowered umbel; peduncle 0-2 mm long; pedicel 0-2 mm long. Sepals inconspicuous, to 0.5 mm long. Petals five, elliptic, 4-8 mm long, yellow; abaxial surface with an indumentum of multiangular stellate trichomes; adaxial surface glabrous. Stamens 10, filaments glabrous; anthers 1-1.5 mm long, each with a terminal gland. Carpels 5; ovary stellate-tomentose; style glabrous; stigma hemispherical. Cocci beaked. Seed not seen. Additional specimens examined: NEW SOUTH WALES: Medway Colliery, 10.x. 1991, M. Kennedy 153 (NSW 249646)-, c. 1.5 km north of Mittagong P.O., 27.vii.1995, S.Donaldson 560, G.Corsini and J.Toby (CANB 9513536)-, Flying Fox Road SW of Medway, 22.xii. 1995, G.Errington 30 (NSW 264105); Medway area, 18.X.1998, B.J.Mole 155-160 and CA.Mole ( BJM155 - MEL, MELU, NSW; BJM156-160 - MEL); Nattai River, Mittagong area, 19.x. 1998, B.J.Mole 161-165 and C.AMole (BJM161 - BRI, CANB, MEL, MELU, NSW; BJM162-165 - MEL). Distribution and conservation status: Asterolasia buckinghamii has been recorded from five localities (from Mittagong south to Penrose and Wingello) in the Central Tablelands of New South Wales. Two populations were located during this study at Mittagong and Medway. The population at Medway consisted of less than 30 individuals, while the popu¬ lation at Mittagong consisted of more than 200 individuals but was restricted to a small area, approx 100m 2 . Populations from the type locality near Penrose and Wingello were not located during this study despite searches in these areas during October and November of 1998. This region has been extensively cleared for agriculture and silviculture, and further fieldwork is recommended to establish the status of the species at this locality. A tentative conservation code of 2E is considered appropriate because the species has a geographic range of less than 100 km 2 , is currently known from only two localities, and because no plants are known to occur within a conservation reserve. Habitat: Populations from the type locality and the Nattai River north of Mittagong are recorded as growing in gullies, on river flats in sandy alluvial soils derived from sandstone, the Mittagong population also extending up a gentle north west facing slope for c. 100 m. The vegetation community at the Mittagong locality is riparian on the river flats grading to open-forest further upslope. The population at Medway, however, occurs on a south facing slope at the top of a cliff in shallow soils over sandstone in open-forest. The species there¬ fore is not restricted to damp localities, as previously thought (Blakely 1940). Phenology: Flowering plants have been collected from early October to late November. Notes: This species differs from A. asteriscophora by the usually solitary flowers and the smaller (often absent) peduncles and pedicels, and from A. buxifolia by the stellate trichomes on the ovary and abaxial surface of leaves, and the type of indumentum on the abaxial petal surface. Etymology: The specific epithet honours one of the collectors of the type specimen, Mr William J. Buckingham of Lindfield, New South Wales. 3. Asterolasia asteriscophora (F.Muell.) Druce, Bot. Soc. Exch. Club Brit. Isles 4: 606 (1917); Phebalium asteriscophorum F.Muell., Trans. & Proc. Victorian Instit. Advancem. 102 B .J. Mole et al. Sci. 31 (1855); Asterolasia muelleri Benth., FI. Austral. 1: 350 (1863), nom. illeg; Asterolasia correifolia var. muelleri Maiden & Betche, Proc. Linn. Soc. New South Wales 26: 80 (1901). Type: Mt Disappointment, October 1852, F.Mueller s.n. (lectotype MEL 708656, fide Wilson, Nuytsia 12: 84, 1998; isolectotypes MEL 708636, MEL 708637, MEL 708638). Slender, upright shrub to 2 m high. Younger stems densely stellate-tomentose, indumentum becoming sparse with age. Leaves obovate-elliptic or spathulate, 4-35 mm long, 3-15 mm wide, papery, apex rounded, rarely obtuse or slightly emarginate, flat or slightly concave adaxially; adaxial surface with a sparse or dense indumentum of multiangular, stellate tri- chomes (range = 1-28 trichomes per mm 2 ); abaxial surface cobwebbed with stalked, mul¬ tiangular, stellate trichomes; petiole 2-7 mm long. Inflorescence a terminal or axillary umbel of 3-8 flowers', peduncles 3-11 mm long; pedicels 3-15 mm at anthesis, longer in fruit. Sepals inconspicuous, 0.5-1 mm long. Petals 5, elliptic 4-9 mm long, bright yellow, pale lemon or white; abaxial surface with an indumentum of stellate trichomes; adaxial sur¬ face glabrous. Stamens 10; filaments glabrous; anthers 1-1.5 mm long each with a termi¬ nal gland. Carpels 5; ovary stellate; style glabrous; stigma hemispherical. Cocci beaked. Seed oblong, 2-2.5 mm long, 1-1.2 mm wide, testa smooth, black and shiny; endocarp thin and brittle, a dull mustard colour, deciduous from seed (Fig 9a-h). Distribution: This species is widely distributed, although often in small disjunct pop¬ ulations, along the Great Dividing Range from the Macedon and Emerald districts in Victoria, to the Tumut district in the Southern Tablelands of New South Wales. The species also occurs disjunctly in the Torrington district in the Northern Tablelands of New South Wales. Habitat: The species is found in a range of vegetation types, including low open-for¬ est, open-forest and riparian communities and is known to occur on a various substrates including krasnozem soils, alluvial soils, granitic sands and skeletal rocky ridgetops. Notes: Asterolasia asteriscophora differs from A. buckinghamii by its three or more flowered umbels, and longer peduncles and pedicels, and from A. buxifolia by the stellate trichomes on its ovaries, stellate adaxial leaf surface, and the multiangular stellate tri¬ chomes on the abaxial surfac eof the petals. Etymology: The specific epithet is derived from the Greek asterios (starry) and pho- rum (carrier) referring to the stellate trichomes found on the stems, leaves, petals and ovaries of this species. 3a. Asterolasia asteriscophora (F.Muell.) Druce subsp. asteriscophora Slender, upright shrub to 2 m high. Leaves obovate-elliptic or spathulate, 6-35 mm long, 3-15 mm wide, papery, apex rounded, rarely obtuse or slightly emarginate, flat or slight¬ ly conduplicate; adaxial surface with a sparse or dense indumentum of multiangular, stel¬ late trichomes (range 1-22 trichomes per mm 2 ); petiole 2-7 mm long. Inflorescence an umbel of 3-8 flowers; peduncles 5-11 mm long; pedicels 3-15 mm at anthesis, longer in fruit. Petals elliptic 4-9 mm long, bright yellow (Fig. 9a-f). Representative specimens examined: NEW SOUTH WALES: Tangster via Deepwater, x.1913, J.D.Lynch (NSW 374566 ); Geehi region, 4.x. 1957, M.E.Phillips s.n. and J.Raeder-Roitysch (NSW 262309 ); Near Cabramurra, 7.xi.l961, M.E.Phillips 99 (CANB 005413 ); Gilmore Creek, Bago State Forest, 4.xi.l962, K.Giles s.n. (NSW 468149 ); Dingo Creek, Silent Grove Road north of Torrington, ix.1971, H.J.Wissmann s.n. (NE 029226 ); Geehi Gorge road, l.xii.1971, D.J.Wimbush s.n. (NSW 261158 ); Kosciusko NP Peak River, 100 m upstream of junction with Wildhorse Creek, 31.x. 1993, N.Taws 235 and A.Scott (CANB 9314794 ); Kosciusko NP, Peak River, 6.xi.l993, N.Taws 239 and A.Scott (CANB 9314798, MEL 278248); Mt Kosciusko NP, Geehi area, 14.x. 1998, B.J.Mole 133-142 and C.A.Mole ( BJM133 - BRI, MEL, MELU, NE, NSW; BJM134-142 - MEL); 5.7 km along Elliot Way from O’Hares Rest area towards Cabramurra, 16.x. 1998, B.J.Mole149-153 and C.A.Mole ( BJM149 - BRI, MEL, MELU, NE, NSW; BJM150-153 - MEL); Variation in Asterolasia asteriscophora s.l. 103 Blather Creek, NNE of Torrington, 21.X.1998, B.J.Mole 177-181 and C.A.Mole ( BJM177 - BRI, MEL, MELU, NE, NSW; BJM178-181 - MEL). VICTORIA: Plenty Ranges, x.1902, G.Weindorfer s.n. (MEL 5199); Upper Genoa River, 28.ix.1947, N.A.Wakefields.n. (MEL 543047 ); Upper Genoa River, far East Gippsland, 29.ix. 1947, N.A.Wakefield s.n. (MEL 709144 ); Upper Genoa River, 25.ix.1948, NA.Wakefield 3201 (MEL 543046 ); Wallaby Creek, x.1952, D.H.Ashton s.n. (MELU 14864)- Cascades, Kinglake West, ll.x.1962, A.M.Gill s.n. (MELU 19798); Pine Mt, 17.xi.1964, J.H.Willis s.n. (MEL 709146 ); Far north-east. Black Mt top of Mount Burrowa, 17.xi.1971, J.H. Willis s.n. (MEL 502061 ); Mt Bowen, East Gippsland, 28.xi.1984, D.Parkes EG217 (MEL 678435 ); Burrowa-Pine Mt NP, 14.xi.1988, F.E.Davies 677, M.J.Winsbury and S.Donaldson (CANB 8804043 , MEL 228142); East Gippsland, Yambulla Creek, 8.ix.l988, D.E.Albrecht 3699 (MEL 268505 ); Yambulla Creeek, 1 km upstream of its confluence with Genoa River, 9.ix.l988, N.G.Walsh 2126, D.E.Albrecht and J.Westaway (MEL 268564 ); Pine Mtn, 20.ix.1998, B.J.Mole 71 (MEL); Hazeldene Road south of Gladysdale, 8.x. 1998, B.J.Mole 78-82 (BJM78 - CANB, MEL, MELU, NE, NSW; BJM79-82 - MEL); Toolangi, Chum Creek Road, 8.x. 1998, B.J.Mole 85-89 (BJM85 - CANB, MEL, MELU, NE, NSW; BJM86-89 - MEL); west side of Mt Towrong, Mt Macedon Regional Park, 11.x. 1998, B.J.Mole 90-94 and C.A.Mole ( BJM90 - CANB, MEL, MELU, NSW; BJM91-94 - MEL); Ben Cruachan Creek NW of Ben Cruachan summit, 12.x. 1998 B.J.Mole 98, 100-103 and J.B.Mole ( BJM98 - CANB, MEL, MELU, NE, NSW; BJM100-103 - MEL); East Gippsland, Pheasant Creek Track, 12.X.1998, B.J.Mole 111-119 and J.B.Mole (BJM111 - CANB, MEL, MELU, NSW; BJM112-119 - MEL); East Gippsland, Warbisco Track [Mt Bowen], 13.x. 1998, B.J.Mole 123-132 and J.B.Mole ( BJM123 - CANB, MEL, MELU, NSW; BJM124-132 - MEL); Mt Kosciusko NP, Geehi area, 14.x. 1998, B.J.Mole 133-142 and J.B.Mole ( BJM133 - CANB, MEL, MELU, NSW; BJM134-139 - MEL); Mt Buffalo NP, Eurobin Creek Falls, 16.x. 1998, B.J.Mole 144 and C.A.Mole (MEL, NSW). Distribution and conservation status: Asterolasia asteriscophora subsp. asteriscophora is widely distributed, although usually in small disjunct populations, particularly along the Great Dividing Range from the Macedon and Gladysdale districts in Victoria, to the Tumut district in the New South Wales southern tablelands. A disjunct occurrence of the subspecies is known from the Torrington district in northern New South Wales. Populations from Woods Point and the upper Genoa River (Victoria), as indicated by herbarium records, could not be located despite searching in those areas during October and November of 1998. The subspecies is widespread with several populations in national parks and others in proclaimed conservation reserves, and appears to be secure. Phenology: Flowering specimens have been collected from October to November. Populations of this subspecies in the Melbourne region were observed to commence flowering 3-4 weeks later than A. asteriscophora subsp. albiflora 3b. Asterolasia asteriscophora subsp. albiflora B.J.Mole, subsp. nov. A subspecie typica foliis et floribus generaliter parvioribus, foliis obovatis-spathu- latis, pedunculis saepe brevioribus et petalis albis differt. It differs from the typical subspecies in the generally smaller leaves and flowers, obo- vate - spathulate leaves, frequently smaller peduncles, and white petals. Type: Victoria, Emerald - Avonsleigh Road, Southern end of Country Club golf course, 8.x. 1998, B.J.Mole 73 (holotype MEL 2120867; isotypes CANB, MEL 2068578, MELU, NE, NSW). Eriostemon spathulifolius Gand., Bull. Soc. Bot. France 60: 458 (1913). Type citation: “Australia, Victoria ad Esmerald [probably Emerald], MacLennan, n.v. (possibly at LYON fide Paul G. Wilson pers. comm.). Equated with A. asteriscophora ‘form of Asterolasia found at Emerald...’ by Wilson (1970, p. 120). Slender, upright shrub to 1.5 m high. Leaves obovate to spathulate, 4-16 mm long, 3-12 mm wide, flat, apex rounded, adaxial surface with a sparse (1-3 trichomes per mm 2 ) indu¬ mentum of multiangular, stellate trichomes; lamina papery; petiole 1-4 mm long. Inflorescence an umbel of 3-5 flowers; peduncles 3-5 mm long; pedicels 6-15 mm at 104 B .J. Mole et al. Figure 9. a-e Asterolasia asteriscophora subsp. asteriscophora (Mole 100): a Flowering sprig x 1; b Seed x 5; c Endocarp x 5; d Fruits x 4; e Leaf (abax- ial view) x 1. f Leaf of A. asteriscophora subsp asteriscophora {Mole 177). g-h A. asteriscophora subsp. albiflora {Mole 75): g Flowering sprig x 1; h Leaf (abaxial view) x 1.5. i-k A. rupestris subsp. rupestris {Fox s.n. CANB 406009): i Flowering sprig x 1; j Leaf section x 2; k Leaf (abaxial view) x 1.5. 1-q A. rupestris subsp. recurva {Mole 172): 1 Flowering sprig x 1; m Flower x 2; n Leaf section x 2; o Leaf (abaxial view) x 1.5; p Petal (adaxial view) x 2.5; q Petal (abaxial view) x 2.5. Variation in Asterolasia asteriscophora s.l. 105 anthesis, longer in fruit. Petals elliptic, 4-6 mm long, white, rarely pale lemon (Fig. 9g-h). Representative specimens examined: VICTORIA: Emerald, 27.x. 1906, P.R.H.St.John s.n. (NSW 262302)-, Dandenong Ranges, Belgrave, January 1933, J.H.Willis s.n. (MEL 709148); between Beaconsfield and Emerald, 28.ix. 1933, T.S.Hart s.n. (MEL 709145); Dandenong Ranges, between Emerald and Avonsleigh, 7.x. 1934, J.H.Willis s.n. (MEL 709149); Dandenong Ranges, Emerald-Clematis, 5.x.1945, A.C.Beauglehole 7016 (MEL 2101021); Upper Femtree Gully [local¬ ity uncertain, Hemphill pers. comm. 1998], 23.ix.1965, B.Hemphill s.n. (MEL 2101022); Eastern Highlands, Emerald-Monbulk Road, 24.ix.1993, N.G.Walsh 3507 (MEL 2020508); Emerald- Avonsleigh Road, southern end of Country Club golf course, 8.x. 1998, B.J.Mole 74-77 (MEL). Synonymy: Eriostemon spathulifolius is probably a synonym of Asterolasia aster¬ iscophora subsp. albiflora but as the type material was not seen this can not be verified (cf. Wilson 1970; McGillivray 1973). The type for Eriostemon spathulifolius is probably at LYON (P.G. Wilson pers. comm. 1998) and was requested in 1998 but has not been located. Distribution and conservation status: This subspecies is known only from three local¬ ities in the Emerald - Avonsleigh district of Victoria, all of which are threatened by urban development. Two of these are located in residential areas, while the third is located with¬ in a quarry site. It was previously recorded from Menzies Creek, ‘Paradise’ and Clematis. The single record from the Grampians (ix.1937, C. French Jnr. s.n., MEL 709143) is most likely incorrectly labelled. No other record of A. asteriscophora is known from the Grampians. A conservation code of 2Ei is appropriate; the subspecies is known from a geographic range of less than 100 km, is in serious risk of disappearing from the wild within 10-20 years if present residential development and associated pressures continue, and is not known to occur within a conservation reserve. Phenology : This subspecies flowers from early October to late November and has been observed to commence flowering 3-4 weeks earlier than the typical subspecies. Etymology: The subspecific epithet is derived from the Latin, albus (white) and floreo (to flower), alluding to the white colour of the petals. 4. Asterolasia rupestris B.J.Mole, sp. nov. Asterolasiae asteriscophora e (F.Muell.) Druce affinis sed foliis generaliter brev- ioribus et subsessilibus vel saepe sessilibus, lamina obcordata vel obdeltata et supra dense stellato-tomentosam differt. Similar to Asterolasia asteriscophora but differs in the generally shorter leaves, which are subsessile, or frequently sessile, the obcordate - obdeltate lamina, and the adaxial lamina surface, which is densely stellate. Type: New South Wales, walking track to the Governor, Mt Kaputar NP, 27.xi.1987, J.M. Fox s.n. (holotype CANB 406009). Upright shrub to 1.5 m tall. Stems with a stellate indumentum. Leaves shortly petio- late, or frequently sessile; lamina obcordate or obdeltate, 9-20 mm long, 6-15 mm wide, papery, apex emarginate, sometimes truncate, base attenuate, slightly conduplicate, mar¬ gins recurved or flat; adaxial surface with a dense indumentum of hyaline multiangular stellate trichomes (15-31 trichomes per mm 2 ); abaxial surface cobwebbed with stalked, multiangular stellate trichomes; petiole when present somewhat thickened and flat, often appressed to the stem. Inflorescence a terminal or axillary umbel of 3-5 flowers; pedun¬ cles 4-9 mm long; pedicels 6-15 mm long. Sepals inconspicuous, 0.5-1 mm long. Petals 5, elliptic, 5-9 mm long, yellow, abaxial surface with an indumentum of rust coloured stellate trichomes; adaxial surface glabrous. Stamens 10; filaments glabrous; anthers 1-2 mm long, each with a terminal gland. Carpels 5, stellate-tomentose; style glabrous; stig¬ ma hemispherical. Cocci beaked. Seed not seen (Fig. 9i-q). Distribution: This species is only known from Mt Kaputar and Mt Lindsay in Mt Kaputar NP east of Narrabri, and from near Parlour Mountain northwest of Armidale; all 106 B .J. Mole et al. of these populations are apparently restricted to trachyte outcrops. Collections have been made from Mt Canobolas, southwest of Orange, but is possibly extinct at this location (see notes under A. rupestris subsp. rupestris). Phenology : The flowering period for the species is from late September to early November. Notes : The species was previously included in A. asteriscophora (e.g. Porteners 1991; Wilson 1998) from which it can be readily distinguished by the obcordate to obdeltate leaf shape and sessile to subsessile leaves. Specimens from the Mt Kaputar area lodged at CANB have been determined as Asterolasia sp. nov. by I.R. Telford and A. Lynne. Etymology : The specific epithet, which is derived from the Latin rupestris, means rock dwelling, and refers to the rocky outcrops to which this species appears to be co nfin ed. 4a .Asterolasia rupestris B.J.Mole subsp. rupestris Leaves plane. Inflorescence an umbel of 3-5 flowers', peduncles 4-6 mm long; pedicels 7-10 mm long. Petals 5-8 mm long (Fig.9i-k). Representative specimens examined'. NEW SOUTH WALES: Mt Lindsay, Nandewar Mountains, 5.xi.l909, R.H.Cambage s.n. (NSW 262327); Springside via Orange, lO.x.1948, W.E.Giles s.n. (NSW 262271); Springside, near Orange, 14.x. 1956, W.E.Giles s.n. (NSW 468148); Devils Hole [Mt Canobolas], Towac, 7 mi les SW of Orange, 8.xi.l960, E.F.Constable s.n. (MEL 2100530, NE 29227, NSW 52954 ); Eckfords (sic) Lookout, Mt Kaputar NP, 4.ix.l976, G.L.Harden s.n. (NE 38917)', The Governor, Mt Kaputar NP, 21.xi.1976, R.Coveny 8937 and S.K. Roy (NSW 468152, CANB 373743); Eckards Lookout, Mt Kaputar NP, 25.xi.1987, J.M. Fox 87/111 (CANB 406306); Mt Kaputar NP, “The Governor”, 15.ix.1998, B.J. Mole 41-51 and WA. Gebert ( BJM41 - CANB, MEL, MELU, NE, NSW; BJM42-51 - MEL). Distribution and conservation status: This subspecies has been collected from only a small number of localities at Mt Lindsay and Mt Kaputar in Mt Kaputar National Park. A population recorded from Mt Canobolas during the 1950s was not located during this study despite searching in that area during October 1998. The locality recorded on herbarium specimens is now farmland and/or pine plantations, and so, at this locality the taxon is possibly extinct. Further searches of remnant vegetation in the Mt Canobolas area are required to establish the status of the taxon at this locality. A conservation code of 3E is considered appropriate. Habitat : The subspecies grows in heathlands, shrublands and low open forest on skeletal gravelly soils on trachyte outcrops, tending to favour sheltered positions but also found in exposed sites. 4b. Asterolasia rupestris subsp. recurva B.J.Mole, subsp. nov. A subspecie typica lamina recurva differt. It differs from the typical subspecies in the recurved lamina margins. Type : New South Wales, northwest of Armidale, 4.2 km north east of Parlour Mtn, 20.x. i998, B.J.Mole 171 and CA.Mole (holotype MEL 2120868; isotypes CANB, MEL 2068579, MELU, NE, NSW). Leaf margins recurved. Inflorescence an umbel of 3-6 flowers; peduncles 4-8 mm long; pedicels 8-15 mm long. Petals 6-7 mm long (Fig. 91-q). Additional specimens examined: NEW SOUTH WALES: The Parlour, about 6 miles from Booralong, 28.x. 1951, G.L.Davies s.n. (NSW 374569); 4.2 km NE of Parlour Mt, 20.x. 1998, B.J.Mole 172-175 and CA.Mole (MET.). Distribution and conservation status : This subspecies is known only to the authors from one locality near Parlour Mountain north west of Armidale, New South Wales. The population is on private land, which the landholders wish to manage for conservation pur¬ poses (Brian Hardaker & Jeremy Bruhl, 1998 pers. comm.). Approximately 30-40 plants Variation in Asterolasia asteriscophora s.l. 107 were observed at the locality, restricted to an area c. 50 m 2 along a small creek. Further searches in the Parlour Mountain district are required to establish the extent of this taxon. A conservation code of 2E is appropriate as the taxon has a geographic range of less than 100 km 2 and is known only from two collections, both outside a conservation reserve. Habitat : The subspecies grows in low open forest on skeletal gravelly soils, along gullies. Etymology : The subspecific epithet refers to the recurved margins of the leaves, char¬ acteristic of this subspecies. Acknowledgments We thank the directors of BRI, CANB, K, NE, and NSW for loan of material; the New South Wales Parks and Wildlife Service, New South Wales State Forest Department, and Department of Natural Resources and Environment (Victoria) for permission to collect on public land; Dr Jeremy Bruhl and staff at NE for assistance with locality information; Brian and Shirley Hardaker for permission to collect plants from their property; School of Botany, The University of Melbourne for funding field expenses; Dr Yvonne Fripp, and the Latrobe University Genetics Department for the use of facilities for isozyme analysis; Jocelyn Carpenter for assistance with SEM; Paul Wilson for comments on the manuscript and for allowing citation of his unpublished manuscript for Flora of Australia; Ian Thompson and Stephen McLoughlin for comments on the manuscript; Neville Walsh for editing the Latin diagnoses; Keith Ingram for attempting to locate Asterolasia buxifolia ; Wayne Gebert, Cathy Mole and John Mole for assistance with field work; Mali Moir for the excellent illustration; Ian Maher (Parks Victoria) for organising access to the Wallaby Creek catchment area; and all the helpful staff at MEL. References Auld, T.D. (2001). The ecology of the Rutaceae in the Sydney Region of Southeastern Australia: Poorly known ecology of a neglected family. Cunninghamia 7, 213-239. Belbin, L. (1987). ‘PATN.’ Pattern Analysis Package. Users, Command and Reference Manuals. CSIRO Division of Wildlife and Ecology: Canberra. Benson, D.H. and McDougall, L. (2001). Ecology of Sydney plant species Part 8: Dicotyledons fa mi lies Rutaceae to Zygophyllaceae. Cunninghamia 7, 241-462. Bentham, G. (1863). Flora Australiensis. Vol. 1 . Reeve and Co: London. Blakely, W.F. (1940). Additions to the Flora of New South Wales. The Australian Naturalist 10, 246-247. Briggs, J.D. and Leigh, J.H. (1996). Rare or threatened Australian plants. CSIRO Publishing: Collingwood. Coates, D.J. (1988). Genetic diversity and population genetic structure in the rare Glittering Grass wattle, Acacia anomola Court. Australian Journal of Botany 36, 273-86. Druce, G.C. (1917). Nomenclatural Notes: Chiefly African and Australian. Botanical Society Exchange Club of the British Isles 4, 606. Duretto, M.F. (1999). Rutaceae. Flora of Victoria 4, 153-197. Duretto, M.F and Ladiges, P.Y. (1997). Morphological variation within the Boronia grandisepala Group (Rutaceae) and the description of nine taxa endemic to the Northern Territory, Australia. Australian Systematic Botany 10, 249-302. Durham, K.L. (1998). Morphological variation within Leionema bilobum (Rutaceae). Honours thesis, School of Botany, The University of Melbourne. Faville, M.J. Adam, K.D. and Wedderbum, M.E. (1995). Allozyme variation within and between three populations of browntop (Agrostis capillaris). New Zealand Journal of Agricultural Research 38, 65-70. Gandoger, M. (1913). Untitled. Bulletin De La Societe Botanique De France 23, 458. Hebert, P.D. and Beaton, M.J. (1989). Methodologies for allozyme analysis using cellulose acetate electrophoresis. Helena Laboratories: Beaumont, Texas. Hewson, H. (1988). Plant indumentum: A handbook of terminology. Australian Flora and Fauna series No. 9. Bureau of Flora and Fauna: Canberra. 108 B .J. Mole et al. Holmgren, P.K., Holmgren, N.H. and Barnett, L. (1990). Index Herbariorum. Part 1. The herbaria of the world (8th Edn). New York Botanic Gardens: New York. Kruskal, J.B. and Wish, M. (1973). Multidimensional Scaling. Bell Laboratories, Sage Publications: USA. McGillivray, DJ. (1973). Michel Gandoger’s names of Australian plants. Contributions from the New South Wales Herbarium 4, 319-365. Milligan, J.W. and Cooper, M.C. (1988). A study of standardisation of variables in cluster analy¬ ses. Journal of Classification 5, 181-204. Mueller, F. (1855). Description of fifty new Australian plants, chiefly from the colony of Victoria. Transactions of the Philosophical Society of Victoria 1, 9-10. Nei, M. (1978). Estimation of average heterozygosity and genetic distance from a small number of individuals. Genetics 89, 583-590. Norusis, M.J. (1990). SPSS Base System User’s Guide. SPSS: Chicago. Porteners, M.F. (1991). Asterolasia. Flora of New South Wales 2, 263-264. Sneath, PH.A. and Sokal, R.R. (1973). Numerical Taxonomy. The principles and practice of numerical classification. W.H. Freeman and Company: San Francisco. Swofford, D.L. and Selander, R.K. (1981). A computer program for the analysis of allelic variation in genetics. Journal of Heredity 72, 281-283. Theobald, W.L., Krahulic, J.L. and Rollins, R.C. (1979). ‘Trichome description and classification’, in C.R. Metcalf and L. Chalk (eds), Anatomy of the Dicotyledons, pp. 40-53. Clarendon Press: Oxford. Warburton, C.L., James, E.A., Fripp, Y.J., Trueman, S.J. and Wallace, H.M. (2000). Clonality and sexual reproductive failure in remnant populations of Santalum lanceolatum (Santalaceae). Biological Conservation 96, 45-54. Weston, PH. (1990). Notes on Boronia (Rutaceae) in New South Wales, including descriptions of three new species. Telopea 4, 121-128. Willis, J.H. (1973). A Handbook to Plants in Victoria, Vol. 2, Dicotyledons. Melbourne University Press: Melbourne Wilson, P.G. (1970). Taxonomic notes on the family Rutaceae, principally of Western Australia. Nuytsia 1, 197-207. Wilson, P.G. (1980). A new species of Urocarpus (Rutaceae) from Western Australia. Nuytsia 3, 211-213. Wilson, P.G. (1998). Nomenclatural notes and new taxa in the genera Asterolasia, Drummondita and Microcybe (Rutaceae: Boronieae). Nuytsia 12, 83-88. Wilson, P.G. ( ined .). In Asterolasia. (manuscript for Flora of Australia). Wright, S. (1978). ‘Evolution and the genetics of populations; a treatise.’ Vol. 4. Variability within and among natural populations. University of Chicago Press: Chicago. Appendix 1 . Specimens and data used in the phenetic analyses of Asterolasia asteriscophora sensu lato. Principal collector given only. For continuous characters (see table 1), mean values are shown. Missing data is coded as 999. Variation in Asterolasia asteriscophora s.l. 14 - - - - - 666 o o O O 13 - - - - - 666 - 12 o o O o o On ON On o o o O 11 - - - - - - - - - - On cn c-~ oo On On , VO NO On © p cn ON ON O) CN On CN cn ci On On +-* CN CN ’■4 r- o o VO On On o O O O 1 , ON ^H cn NO p ON ON ’ — 1 VO tp p M to NO NO NO p ON ON VO VO of P -a £ © oo ON ON ON ON ON ON o VO O O CN o z S_I ON ON ON ON ON ON ON On On ON On ON p oo O oo to to a On On On On On On o o o o i> ON On ON ON ON ON oo O) oo p a On On On On On On cn 3J-’ 3j-‘ of — U O O o vo O On o o O o NO vo CN o p p ON p oo VO p cn cn cn cn cn On 3J-’ p of IT) 8 o CN o NO o oo o oo O oo o NO o oo o oo o °0 On cn CN On On On CN O — 1 CN o — 1 O O O O O O o 0 O o Tt OO vo On On r~- cn CN o cn p O o O O o cn CN cn CN cn cn oo oo cn On CN O oo OO o CO cn p CN ON p VO 3f_ p p N -1 hh 1—1 1—1 1—1 1—1 1-1 o- 1 On c— P oo r- o o cn NO CN vo vo o vo P_ NO VO CN P hh ci hh —< ci ci ci CN o o o o O o o o o o cn p On oo O P. >—' p p p ci cn cn cn cn vo NO cn of CN CN £ £ £ £ £ c/3 oo oo oo oo Z Z Z Z Z o «l c5 o5 c3 c3 > to i> o to > to i> S to -c s 3 O- 3 3 O- 3 3 9 3 9 d s T3 3 ■3 3 03 3 03 3 a to £ to to to to to © p -|_i N_» 3 B g g P Z s s s ffi w w w w p p O) p p p P P P hJ hJ h to w w w w w w w w w w a ja © cc S s S S S S S S S s to to EC of vo On cn oj- NO ^f 'Sf P Ns K Cv fs fs a to to >£ "to .to "to "to "to "to s ^ ^ ^ ,a ■s ’S ’S *s •s *s •s ■s •s *s o Z CQ CQ CQ oq CQ cq oq «5 CQ «5 a to Si a to © '8 X! a CN cn *# oo NO c~ oo On o © i/5 z 109 o - - - - - - - - - - - - - - - - - - - - OH o O O o O o o o o O O O O O O o O - - - - - - - - - - - - - - OH - VO On oo oo oo On cn o cn cn oo cn CN c- j. On NO CN p cn CN Os OO CN CO vo p On cn oq OO oo O CN ON ON *-» ci ci ci ci o2 ci ci H H ’ ci O o O O o O 0 0 o o o cn O o o VO o p T-1 p cn cn p o O On oo p ■—i oo c- p CN ■—i of Of VO of Of of vo vo NO NO vo oo NO NO p t~i NO O O o p o p 0 o O O OO O On ON VO c- O O oo p of Ns»^ of o cn cn p ON ON CN p oq cn NO ON oo NO - On oo P vo ON On VO of P ri O O o p 0 O p 0 o o o On On 0 c- On o P p C] of o p p o o~ 1—1 NO On ON o NO ON CN of vo oo NO NO vo vo ci ND NO r~i On ON vo P On ON O cn cn O VO O o c- O VO c- O c- 0 vo cn o p cn cn CN p p p p OO p T-1 vo '—1 O p cn vo of VO of cn cn of vo of of of VO of vo of cn vo vo O o o o o O o o O o 0 O o O o O o p °° p p p p p Cl ON NO o oo p p p NO oo ' -H of CN of CN CN 1 CN ON O H ci vo of ci On o O o 0 o 0 o o o o O o o o o 0 O oo p O) O p o cn ON oo °° oo vo oo cn o ND CN cn cn of VO vo vo VO of of of of cn ci of vo ND OO o cn cn o ON cn O NO C- o vo NO of CN O p ON oo Os p p O p p O p p cn p cn p Os CN ci ci ci ci ci ci o-l o-l o-l H ci oo CN cn CN CN CN Of cn cn of c- vo ND of CN __! oo P oo CN oo Of p OO p oo oo ’—1 of cn OO O ND CN CN ON ci ci cn ci cn ci ci ci ci oi oi ci ci o o o o o o o o o o o o o o O o o o On oo IC¬ ON oo cn of oo f-H p CN P Os o NO cn p ON ci ci fi oo fi O P of On cn cn ON CN CN CN CN CN CN CN CN CN CN CN to to to to to to > to to to to to > > > > > to > > > > > to to to to to 3 3 3 3 3 3 i> dn dn 23 2h 23 > > > j> > O T3 O 03 O 03 O T3 o 03 33 3 3 3 3 3 to to to to to 3 03 -a "O "O 03 00 00 00 00 00 o o to O to g 3 to >% 03 Pb, 03 >s 03 03 '>> 03 3 23 3 23 3 5 3 23 3 3 3 s 3 s 3 s 3 s 3 s u 3 -2 3 3 3 3 'o 'o ’o ’o Ch 3 w 3 O 3 3 3 e2 e2 e2 e2 e2 § s % % % to « hJ j J j p p P p p p p p p p p p p w m w m w w w w w w w w w w w w w 2 s s s s s S S S 2 2 s s s s t\ oo PN CN vo c. °o on o CN cn ,5 f OO fo Cs 00 oo oo Oo °o Oo oo 0O On Os O, On ON on -23 -23 -23 -23 -23 -23 -23 -23 -23 -23 -23 to to .to .to A "to "to "to "to to "to "to "to to "to ^ ^ ^ ^ ^ •s ■s ■s *S ■s ■s ■s ■s ’S “S S s *s "S *S ■S P CQ cq oq cq oq cq oq cq oq oq oq cq oq oq oq oq oq CN cn of vo NO c- oo On o CN cn of vo NO c- CN CN CN CN CN CN CN CN 110 14 - - - - - - - - - - - - - 13 - - - - - - - - - - - - - 12 o o o o o o o O o o o o o O 11 - - - - - - - - - - - - - oo m m oo r- r- o 0 _. t"~ N" o © m ri cn oo p CN oo o CN p IN IN IN cn cn cn (N CN CN CN p p P o o © o 0 o o O 0 o o o O o 1 _. os a-. ON On in O cn CN i—i O cn N- ON N p in p p P NO in in P in NO N 2 in >n £ S n o k | o o o CN o o o o o o o o o o u oo N CN in as p l> oo N z S_i O On oo O c~~ p N 2 N 2 P N 2 NO 0> u 3 o O o o o o o o O O O 8 O O in in in SO a- cn o N" in p oo oo OO es as in oo t> NO NO NO p NO in NO p NO NO 43 U o NO o o r~ r- o cn o o O o o O © in OO o NO p P P oo p CN CN CN °° p P in in p cn cn in cn N-’ cn cn P p P p p p r~; N-_ r~; cn p p cn c4 cn CN l -1 1—1 1—1 N -1 o m , a - cn , N" m cn . , cn O o > O > o j> E> M > E> M i> M o O O o a a a a 0J cu cu > > i> > o O O 9 u u u u u a a a a a £» Si a C3 a a 0) CU CU CU cu 13 rj J8 £ 2 u 2 u 2 u 2 u fi c3 c§ g « a §§ » C3 cS a « c§ 1 m 1 PQ 1 PQ 1 PQ 1 PQ © P © Z CC OJ PQ Ch <0 PQ J P P p J P p p P p P p p P *c « m W W w w W w w W w W m w w © pfi & S S 2 S S S S S S S S S S S 5h cu sc «3 © ''i o "o "o "S "S ~o "o "2 "o "o "o "o "o "o 5 S3 S3 "a .a P p p •S ■s S ■s *s •s s "S O Z CQ cq CQ CQ cq CQ CQ CQ cq CQ CQ CQ CQ QQ & o> U £ a> pfi oo o o CN CO N" NO r- oo o O cd © cn cn cn CO cn CO cn CO cn cn cn cn N" N- a n © Z B .J. Mole et al. T ~~' T—1 r " H y ~~ i T— 1 T—H T ~~ i , “ H y ~~ t 1 - - - - - - - O o O o o o © © © © © © © © © © - - - - - - - - - - - - NO o m oo in cn IN cn m in O OO 0 © p OO ON p IN IN © i—i as p N-_ CN ON © © P P p P P P P P P P P P P P o O o o 0 o © © © © © 0 0 © © © p p oo p o m p ON cn CN © © CN CN cn >n N 2 in N 2 in NO N 2 P P NO OO oo ON OO P so o O o 0 o 0 © p © © © >—i © © © 0 © ' 1 © p oo p O p o p o P oo © CN © J _ cn N 2 N 2 p in N 2 NO - P NO ON ON 00 ON - 1N cn o o O o 0 IN 0 as © © cn 0 0 0 © p cn N" as. m © © © as. CN © © © IN N 2 P NO Os NO o On NO p as p ON OO OO O o o cn 0 o m IN © © oo o IN 0 IN m p N-_ p NO o p IN P © p °° oo ■—i © i—i ■—i N 2 in in NO in ^l 2 N 2 P N 2 N 2 p p P p P 8 8 8 O OO 8 8 © oo 8 8 © © © © OO © CN © CN 8 o o o © p © p © p p p NO P NO p ' 1 1 1 CN o o o o 0 o © © © 0 0 © © © 0 0 m p N" 1—1 o CN p p p © © p © SO © © P N 2 NO cn p N 2 p ^ct" P p p p P p P P o o cn oo o 0 ^1- © cn OO O > o > o > o > o S-H cu > a > 5 > is > OO Z; 00 Z 00 z oo z oo z oo Z 3 3 1 § lH s ,S 5 S s S >> a n *n >n •n N N N N N N N N N N 1 — i N N ~s "o "3 o "5 "S "o 'o ~s "o "o "3 "o o S3 S3 p S N' N' *s N N‘ p S •N p p *s ■N p P cq CQ CQ CQ CQ 0Q CQ cq CQ CQ CQ cq QQ QQ CQ CQ CN cn •*t © IN oo © o (N cn 'pi¬ m r- N- a- a- a- a- a- m in m rn in in m m Variation in Asterolasia asteriscophora s.l. Ill i-H CN D X s s z cs 43 u si I i iJ Z i - II £ W 03 II S „ s £ 8 | #Z 000000000000000000000000000000 00000 M M M M M © © p p cn cn r4 cn O O cn o O cn t"~ no cn cn 0000000000000000 CN p NJ CN cn i_J CO OO 1— 0 'inOOOC''-inC 0 CNC 0 CN 00 O' h\din«iniri«inh^^«o6r'^^'0'0'oint'n OOOOOOO ON ON ON NO ON NO O ^h‘ -h’ O Z ^ On On t"- 00 r- NO c-~ c-~ 00 m m o- 0000 o p 00 p in no no in no 00 On OOO no >n cn © —; o 1 O) cn >n cn o-_ c 4 cl d c 4 CN CN 0 0 m 0 0 O O in >>>^^^ ^^££$zzzzz Z Z Z Z bX)toX)toX)bX)OX) 1>n >n >n ■>. fl d £h toA tofl bQ bi) bO ,22225 £ £ 00 cn Z Z £ £ cn cn Z Z fs £ gf oo-o g g g fl H S 2 £ S S O Ih ?h -Ih Ih Ih 0 3 0 0 0 CD (D D D .' 5=5 . 1=5 -05 -O! -05 Q . p .■ p | p , p | £ £ £ cn cn cn z z z 000 000 bO bb 00 o e c ■a : ■„ £ £ e 2 £ £ cn cn Z Z s q o q DODD > > > > M M M M D D D D P D D D n *H *H Nh O D D O > > c "3 D D 3 3 c c D > D > Z* , 3 ! O D o .5 .5 .5 ,3 ,0 ,0 x x .2 .2 rO -O H H o3 c3 c3 c3 -j—» -+-» £ £ £ £ Q Q 2 S af~ | a s s t R a « 3 ^ au S 2 Ph Z> PQ W > > I s wwwwwwwwwwwwpqpqwwwwa, _ sssssssssssssssssssS^dddddald P °o jN H l~H >—( "o ”3 "0 "O "o "o "S "O "0 "o "5 ”5 5! 51 5 5 5 5 5 5 5 5 5 5 5 ’S ‘S ■s •s *—s ■s P p p p p p p p p P CQ CQ CQ CQ cq CQ CQ CQ cq' CQ CQ CQ CQ CQ CQ CQ ^ * *£> £ *§ 53 ^ <0 •n & S ^ a s a id ^ ^ ^ ^ ^ ^ 5 ; O bn ^ b- O Da s ^ 112 B .J. Mole et al. ro i-H CN U a ) s s Z u O) cs 43 U •3 I li P z i - ! J ^ cn £ S3 «8 | b II 31 S >- s £ £ I oooooooo x x On On On On OOOOOOOOOOOOOOOOO On On On On On On X CN -h CN nf O p cn *st no cn in cn n av o on o On X O O O (N NO OO o o c4 o o o o 5 > a S P O (U > c e in as in 0 0 o~ o X cn o o X CN CN cn in in in in ^J- o o o o o CN ^J-. oo CN CN NO in CN cn O ’“S P o O o O o X cn in cn X 1—1 X NO cn cn cn 't CN r~ t~- X j cn as «n X X nt; *—j X 1—1 1—1 1—1 CN i -1 oo oo o O , oo X oo as cn On CN cn cn n4 CN CN o o o O O OO p oo p o oo oo in O nf- NO On X CN CN CN p •P CN o O o O > > > i> »H N- Nh Vh 0J 0J CD CD _> _> > _> 5 2 2 2 03 c3 c3 c3 O « O — O o O o ao CD CD CD CO co o o O O z Z t-H V- »H Nh 0J cu CD CD IS IS X X X X CD CD X X X X CD CD 2p 2P 23 o o n C\ X X oo X X N- N- X ' J n X '~H X X oo cn X cn CN Vo X 's- X 'CJ- X C\J «N •n tN X CN hJ m P m p w P W C/^ CQ z S S s S z < u X CN X o- Os CN in in X X X oo CN nf cn cn ON O cn CN X X X n|- CN CN CN c4 X CN c4 c4 CN P CN CN ^2 CN O 0 O 0 X 0 cn o O o O cn o O i—i O X o X o in r~- CN X oo X nf oo NO X yn X NO vd in ND OO p N- N" X p cn cn cn S'- On in m On On ^P • ^ « ON N ON ON ^ ^ NO On o\ On oti N- x x N- x r~~ o o o o o o X X o O X X oo -—i X X 1—1 X o o o X X N" t> t> X O o o 0 0 0 p o o o N- X o CN o X o p o CN cn X CN cn o o o o X p X o l -1 CN CN r- r~- X -—i cn N" ? P On 9 2 N; cn oi o o iri no on no h on oo on o o o n- r- o P- © O o o o o © © o o o o o o oocNcNooxxc---n|-xxc----oo o o p p cn cn cn cn o cn r- x 3 o- o » ° £ 43 cn oo Z Z 23 r3 o o XI X o o s s % s U U U “II £ £ cn cn Z Z O > x J J X o bJO c3 S* •S .S ^ £ s s N- 0O On On IN tN N- N- a o pp pp u u X X X '—i No NO nj. >n ^ in > > o N K O ^ ^ O £ £ PP cn cn U Z Z °0 CN CN j pp ^ ^ W ^ cn m S Z Z On cn k cn cn ^s; d Jjo .“41 .In Co Q s: tcj Hj fe; s x o2 P p o c*i ^ cj pp cp cp .»h N? V N? 55 ^ 32 hK Oh ftj X Co D3 sf sf S - —J X X cn CN X X CN X K5 S3 -s; Nj sf .m J? On oo IN "SJ- S’ 5 *5 sc to «n sc &5 V H« $ Nd 3 -3 bo .S p X sc 3 12 s ^5 Oh 5 Q § g s k