VOL. 129, PART 1 31 MAY, 2005 Transactions of the Royal Society of South Australia Incorporated INCORPORATING THE Records of the South Australian Museum Contents Watts, C. H. S. & Leys, R. Review of the epigean species of Australian Limbodessus Guignot (Insecta: Coleoptera: Dytiscidae) Martin, H. A. & Specht, R. L. Sclerophyll (heathy) Loins e's in the Mount Lofty Ranges, South Australia White, J. M. & White, T. C. R. Macro-invertebrates captured in artificial substrates in the restored Watervalley Wetlands in South Australia — - ~— — Horr, G., Pring, A. & Zbik, M. The Kimba meteorite: An (H4) chondrite from South Australia Anstis, M. & Tyler, M. J. ciel bi casey of Litoria microbelos (C ea (Anura: Hylidae) — Clark, P., Holz, P. & Spratt, D. M. Hepatozoon tachyglossi sp. nov. (Haemogregarinidae), a protozoan parasite from the blood of a short-beaked echidna, Zachyglossus aculeatus -— — - -— —- Kemper, C. M. Records of humpback whales Megaptera novaeangliae in South Australia — - = — - Smales, L. R. A redescription of Odilia emanuelae (Nematoda: Trichostrongylina: Heligmonellidae) from Australian rodents with a key and comments on the genus Odilia Tanner, J. E. Three decades of habitat change in Gulf St Vincent, South Australia Souter, N. J. Flood regime change in the Hattah Lakes Victoria resulting from regulation of the River Murray = Brief Communication: O’Callaghan, M., Reddin, J. & Lehmann, D. Helminth and protozoan parasites of feral cats from Kangaroo Island = PUBLISHED AND SOLD AT THE SOCIETY’S ROOMS SOUTH AUSTRALIAN MUSEUM, NORTH TERRACE, ADELAIDE, S.A. 5000 81 Transactions of the Royal Society of S. Aust. (2005), 129(1), 1-13. REVIEW OF THE EPIGEAN SPECIES OF AUSTRALIAN LIMBODESSUS GUIGNOT (INSECTA: COLEOPTERA: DYTISCIDAE). by C. H. S. WatTs* & R. Leyst Summary watts, C. H. S. & Leys, R. (2005). Review of the epigean species of Australian Limbodessus Guignot (Insecta: Coleoptera: Dytiscidae). Trans. R. Soc. S. Aust. 129(1), 1-13, 31 May, 2005. DNA sequencing of the CO1 and 16s-tRNA-NDI fragments of the mitochondrial genome was used to support the morphological results. Nine species are recognised; Limbodessus amabilis (Clark), L. capeensis sp. nov., L. compactus (Clark), L. gemellus (Clark), L. inornatus (Sharp), L. occidentalis (Watts & Humphreys), L. praelargus (Lea), L. shuckardii (Clark), and L. rivulus (Larson). Limbodessus dispar (Sharp) is synonymised with L.' shuckardii (Clark). A key to the species is provided. Key Worbs: Coleoptera, Dytiscidae, Limbodessus, Taxonomy, DNA. Introduction Among the commonest diving beetles in southern Australia are members of the genus Limbodessus Guignot of the tribe Bidessini, which are often abundant in still to slightly moving, shallow water. The genus level classification of these beetles has recently been revised by Balke and Ribera (2004), who, based on morphological and DNA sequence data, synonymised the genera Boongurrus Larson, Tjirtudessus Watts and Humphreys, the Australian members of Liodessus Guignot and Limbodessus Guignot, placing them all in Limbodessus. Cooper et al. (2002) had previously studied DNA sequence data of species in these genera and tentatively come to similar conclusions but made no taxonomic decisions regarding them. The species level taxonomy of the species formally placed in Tjirtudessus, Boongurrus and Limbodessus has recently been dealt with by Watts and Humphreys (2003), Larson (1994) and Balke and Sato (1995) respectively. No similar revision has been done for the six Australian species previously placed by Watts (1978) in Liodessus. Increased collecting since 1978 has highlighted the great similarity between these species and_ the consequential difficulties in their identification. The need for a soundly based taxonomy of these species has become more urgent with the realisation that they are congeneric with the numerous subterranean species, previously placed in Tjirtudessus, being discovered in Western and Central Australia (Watts & Humphreys 1999, 2003, “ Department of Entomology, South Australian Museum, North Terrace, Adelaide, South Australia 5000. t Evolutionary Biology Unit (and Centre of Evolutionary Biology and Biodiversity) South Australian Museum, North Terrace, Adelaide, South Australia 5000 2004). To understand the evolution of these stygobitic species a better understanding of the taxonomy of the epigean members of the genus is needed. This paper mainly deals with the species level taxonomy of those Australian species of Bidessini previously classified as Liodessus (Watts 1978, 2002). Utilising the extensive collection in the South Australian Museum, it is based on adult morphology, in particular the male genitalia, supported by DNA sequence analysis of the CO1 and 16S-tRNA-NDI fragments of the mitochondrial genome of key specimens. For details of the DNA procedures used see Cooper ef al. (2002). For completion, brief notes are given on the three other epigean Limbodessus species that were previously placed in Limbodessus and Boongurrus (Watts 2002; Watts and Humphreys 2004). Unless otherwise noted all specimens were collected by C. H. S. Watts. Abbreviations BMNH The Natural History Museum, London. SAMA South Australian Museum, Adelaide. TABLE |. Observed DNA sequence divergence among Limbodessus species previously placed in Liodessus. Comparison % Sequence divergence Within the Limbodessus spp. 2.9-10.8 given below. Between sister species: L. inornatus (3) - L. gemellus (3) 2.9-3.7 L. amabilis (1) - L. praelargus (1) 353. Within species: L. gemellus (3) 0.1-0.3 L. shuckardii (1)/L. dispar (3) 0.0-0.7 (n) = number of sequenced specimens per species. hm Results The combined morphological and DNA results recognise six species, one of which, L. capeensis, is new. Limbodessus dispar (Sharp) from the Southwest is morphologically indistinguishable from the eastern L. shuckardii (Clark). DNA sequence data confirm a close relationship with specimens from Western Australia and Victoria differing at only 0.0% — 0.7% sequence divergence (Table 1). We consider them to be the same species. Specimens morphologically identified as L. gemellus (Clark) (see keys and under species descriptions) from Tasmania (19 km W. Maydeena), Adelaide (Onkaparinga gorge) and the Flinders Ranges (Moro Gorge) are closely similar biochemically (Table 1). These differ by 2.9 — 3.7% sequence divergence from specimens identified as L. inornatus (Sharp) from the Southwest. These are morphologically close but differ in the distal shape of the penis. We consider that the sequence and morphological differences are sufficient to consider them separate species. Limbodessus amabilis (Clark) and L. praelargus (Lea) are morphologically indistinguishable other than by their differently shaped penises. The sequence data confirm their specific separation (Table 1). Systematics Diagnosis The specimens discussed here share the following morphological features. Epigean Australian Bidessini. Elongate-oval, without sutural striae, with well-developed elytral and pronotal plicae, with long metacoxal lines, paramere two segmented, with well-developed finger-like apical lobe on distal segment of paramere (Figs 11-18). (Subterranean species of Limbodessus are more varied in shape and may lack the elytral and pronotal plicae and metacoxal lines (Watts and Humphreys 2003).) Key to epigean species of Limbodessus Including Allodessus bistrigatus (Clark) which is closely related (Balke and Ribera 2004) and is often confused with species of Limbodessus. 1. Extreme front of elytral epipleuron with raised transverse carina delineating a basal pit............... Pen ee ners pier pare Limbodessus compactus (Clark) — Elytral epipleuron without basal pit... 2 2. Head with fine line between back edges of eyes (GCeOrviCal Strid)... ccscissssccnssccavsessscsccessenesssesvaseets 4 — Head without cervical stria or, if present, only weakly and partially (Pilbara, Yilgarn and Atherton Tableland )........ccccccccsssceccseesesseeeesnees 3 5 J. C.H. 8. WATTS & R. LEYS Eyes of normal size, hind edges reaching well beyond anteriolateral corners of pronotum. Western Australia (Fig. 26)........... L. occidentalis poten heWeatedsahog sadertanenbte badcetante (Watts & Humphreys) Eyes reduced to about 60% normal size, hind edges not, or only just, reaching anteriolateral corners of pronotum. Atherton Tableland (Fig. QEY-cteschacoeddeartpenvenleclonsersieedosegt L. rivulus (Larson) Paramere with apical segment simple, finger-like (Fig. 11). Dorsal surface reddish/yellow to greyish, elytra with diffuse darker markings; ventral surface in female with black metathorax and metacoxae and yellow abdomen, males black except prosternum and head; segments of female antenna with a basal groove; elytra densely but relatively weakly punctate, coxal plates moderately punctate ......0 cece eee eeeeeeeeneereeetees or thptes Desh sthoiehs cleesae Allodessus bistrigatus (Clark) Paramere with apical segment strongly lobed, often hook-like (Figs 13-20). Segments of female antennae without basal groove; ventral surface usually more uniformly coloured; punctures Watia ble: assesssizeescrtceyssdicrzeceracns\nteescanssnapsot eae 5 Metacoxal plates with numerous large punctures; (Northeast Queensland, Fig. 22)... Saaedgedabiens tebe outed ateaecneade L. capeensis sp. nov. Metacoxal plates virtually impunctate ............... 6 Pronotal plicae straight; never with dorsal surface mat; antenna relatively thin (Fig. 1); pro and mesotarsi in males moderately expanded (Fig. 5): paramere with apical lobe without setae (Fig. 13). SOUHSAS Aree seid. oh resi carstieemntaeresaqeegpanents any 7 Pronotal plicae slanted or curved inwards; may have mat dorsal surface; antenna, particularly in females, stout (Figs 3, 4); pro and mesotarsi strongly expanded, particularly in the males (Figs 7, 8); paramere with apical lobe with long setae (Fig. 20). Southeast and Southwest... 8 Penis with apical quarter relatively broad, noticeably narrowing only close to tip (Fig. 13).. stale deste boL Qs bob uted MeTah sgh badh sled iaeiieny L. amabilis (Clark) Penis with apical quarter narrowing to blunt point (Fig, 1B) nits snceiatteeneansbiebens L. praelargus (Lea) Penis suddenly narrowing to narrow apical quarter (Fig. 20); length 2.5 — 3.0 mm; pronotum with disc usually diffusely darker; elytron with colour pattern tending linear; pronotal plicae tend to curve inwards in apical half; antenna, particularly in female, stout (Fig. 4)... eee spaamiehau nea eanibig ate necks aathad enna be L. shuckardii (Clark) Penis evenly narrowing to tip in apical quarter (Figs 15, 16); length 2.2 — 2.8 mm; pronotum never with disc diffusely darker; elytron with colour pattern blotchy; pronotal plicae sinuate; antennae not as stout (Fig. 3)... eee eee 9 Penis with apical portion long and thin, parallel sided (Fig. 15); protibiae in female moderately g 2 g 7 | g & eb ¢ REVIEW OF THE EPIGEAN SPECIES OF AUSTRALIAN LIMBODESSUS GUIGNOT -—— — 2 : 4 g a 6 8 Figs 1-4. Male and female antennae of: 1, Limbodessus amabilis; 2, L. capeensis; 3, L. inornatus; 4, L. shuckardii. Figs 5-8. Male and female protarsi of: 5, Limbodessus amabilis, 6, L. capeensis, 7, L. inornatus; 8, L. shuckardii. 4 C. H.S, WATTS & R. LEYS broad (Fig. 10); Southeast (Fig. 24)... HO Gestadeth ers yeetlofttconmt tbh eels L. gemellus (Clark) Penis with apical portion smoothly narrowing to tip (Fig. 16); protibiae in female very broad (Fig. 9); Southwest (Fig. 25)........ L. inornatus (Sharp) Descriptions. (In alphabetical order.) Limbodessus amabilis (Clark) Hydroporus amabilis Clark, 1862, p. 420. Liodessus amabilis (Clark). Guignot, 1939, p. 54. Limbodessus amabilis (Clark). Balke and Ribera, 2004. Types Lectotvpe (designated by Watts 1978). Upper specimen of two mounted on separate cards on same pin, “amabilis Clark” (yellow label), no data, BMNH. Paralectotypes 1, pinned under lectotype, “amabilis Clark” (yellow label); 3 females, | male mounted on same card, “67.56/ amabilis Clark”; 5, “S. Australia” “Bakewell 59,24”. All BMNH. Description (number of dissected males examined, 27). Figs 1, 5, 13, 21. Habitus. Length 2.6 — 3.3 mm; moderately convex; not constricted at junction of pronotum/elytra; narrowly oval. Head. Dark reddish, often lighter towards front. Narrower than elytra. Smooth, shiny, punctures small but deep, rather sparse, stronger and denser towards the rear; cervical stria well marked. Antenna with segment | cylindrical, segment 2 as long as segment 1, barrel-shaped, segment 3 as long as segment 2, narrower, narrowing towards base, segment 4 shorter than segment 3, segments 5-10 subequal, segment | | about twice length of segment 10, narrower (Fig. 1). Maxillary palpus elongate, segment 4 as long as segments 1-3 combined. Pronotum. Reddish-yellow, diffusely darker towards rear in some. A little narrower than elytra; anteriolateral angles projecting strongly forward; base not constricted, posteriolateral angles square, surface shiny, punctures relatively sparse, uneven in size, small on disc, larger behind, particularly inwards from plicae; basal plicae strong, excavated somewhat on inside, straight, reaching about half way to front margin of pronotum. Elytra. Dark reddish, most specimens with indistinct light/dark pattern, occasionally distinct. Elongate, widest at middle; shiny, moderately densely and somewhat unevenly covered with moderate punctures; plicae well impressed, straight, about as long as pronotal plicae. Elytron with well developed inner ridge near apex (ligula). Epipleuron lacking basal carina, relatively broad in anterior quarter, then progressively narrowing to near apex. Ventral surface. Meso and metathorax, metacoxae and abdomen reddish-brown to black, appendages, pronotum and head lighter. Prosternal process narrow between coxae, reaching mesothorax, apical half relatively broad with parallel raised ridges on each side, not arched in lateral view. Metathorax with wings short, broadly rounded in midline behind, with scattered very small punctures. Metacoxal plates large, shiny; punctures very small, scattered; metacoxal lines raised, distinct, moderately widely spaced, reaching to metasternum, weakly diverging; closely adpressed to ventrite 1. Ventrites | and 2 fused, with a few very large deep punctures, sutural line distinct, ventrites 3 to 5 mobile, with a few small punctures, somewhat denser and larger on ventrite 5. Legs. Protibia triangular, outer edge straight or weakly bow-shaped, widest at apex where it is about four times its basal width; protarsus weakly expanded, segment | about 2x as long as broad, segment 2 a little wider than segment 1, about half as long, segment 3 about as long as segment 1 and same width, deeply bifid, segment 4 very small, hidden within lobes of segment 3, segment 5 narrow, cylindrical, a bit longer than segment 3, segments | to 3 with dense covering of adhesive setae (Fig. 5); claws short and simple. Mesotrochanter elongate, subrectangular, with a row of relatively long thin setae on inner edge; mesofemur with 2 to 3 moderately strong setae near base on hind margin, stronger than those on mesotrochanter, mesotarsus slightly longer than protarsus. Metatrochanter tip rounded; metafemur elongate, lacking spines; metatibia narrow, moderately curved, widening towards apex; metatarsus relatively stout, segment | longest, segment 5 longer than segment 4, segments | and 2 in combination about as long as others; claws weak. Male. Little external difference between the sexes except that antennae in the female are a little stouter (Fig. 1). Median lobe of aedeagus moderately broad, apex rounded (Fig. 13). Paramere quite broad, apical segment with apical lobe overlapping apex of rest of segment which is broadly rounded (Fig. 13). Remarks A common species in South Australia, Victoria and Tasmania, often occurring together with L. gemellus, L. shuckardii or L. praelargus. Indistinguishable from L. praelargus other than by the broad tip to the penis, it is separable from the other two species by its relatively unexpanded pro and mesotarsi and straight rather than curved or sinuate pronotal plicae. Never with a mat female form such as occurs in L. shuckardii and L. gemellus. REVIEW OF THE EPIGEAN SPECIES OF AUSTRALIAN LIMBODESSUS GUIGNOT 5 The lectotype is a female and hence, on present knowledge, cannot be separated from L. praelargus. Among the paralectotypes is an immature male. The teneral penis appears more like that of L. praelargus than L. amabilis (M. Balke pers. com.) suggesting that some of the paralectotypes are in fact L. praelargus. We have however retained the usage of Watts 1978 and associated the name L. amabilis with this species. Specimens examined (localities of males with genitalia extracted). New South Wales. Berry 1/68; 14 km W Delegate, 4/11/97. Vietoria. Buangor, 9/11/97; 1 km S Drik Drik 11/10/97; Mt Emu Creek, 1/59; 3 km S Fisherman’s Rest, 6/11/97. South Australia. Adelaide, 11/60; Hindmarsh Island, 8/62; Chain of Ponds, 12/62; 3 km W Yunti, 5/10/95; Mt Compass, 8/61; 10 km E Mt Compass, 10/9/97; 1 km S$ Nangwarry, 5/10/00; Williamstown, 10/61. Tasmania. 25 km E Bridport, 23/1/00; 5 km E Bridport, 23/1/00; Cradle Valley Cradle Mountain- Lake St Clair NP, 19/1/00; Ellendale, 1/12/00, 2 km SW Ellendale, 2/12/00; 2 km W Fingal, 23/1/00; Swansea Jan; Haartz Mt, 15 km W_ Geeveston, 3/12/00; Harcus River, 14 km SW Montagu, 22/1/00; Hobart, Jan; 12 km N Hobart, 2/12/00; 8 km N Kingston, 3/12/00. Limbodessus capeensis sp. nov Types Holotype Male. “Mcllwraith Rng Weather Stn N. Qld 23/7/82. C.Watts.” SAMA, SAMA Data Base # 25- 002891. Paratypes 4, as for holotype, SAMA, SAMA Data base # 25- 009337; 4, “Captain Billy Ck N.Q. 10/83 C. Watts.”, SAMA, SAMA Data base # 25- 009338. Description (number examined 9). Figs 2, 6, 14, 22. Habitus. Length 1.9 — 2.1 mm; moderately convex; very weakly constricted at junction of pronotum /elytra; narrowly oval. Head. Reddish-yellow. Narrower than elytra. Smooth, shiny, punctures small but deep, rather sparse; cervical stria well marked. Antenna with segment | cylindrical, segment 2 as long as segment 1, barrel shaped, segment 3 as long as segment 2, narrower, narrowing towards base, segments 4-10 becoming progressively slightly broader, segment 1 1 about twice length of segment 10, narrower (Fig. 2). Maxillary palpus elongate, segment 4 as long as segments 1-3 combined. Pronotum. Reddish-yellow. A little narrower than elytra; anteriolateral angles projecting strongly forward; base very weakly constricted, poster- iolateral angles square, surface shiny, punctures relatively sparse, uneven in size, small on disc, larger behind, particularly inwards from plicae; basal plicae strong, excavated somewhat on inside, sinuate, nearly reaching front margin of pronotum. Elytra. Rather uniformly dark reddish with tips lighter, some specimens with indistinct light/dark pattern. Elongate, widest behind middle; shiny, quite densely and evenly covered with strong punctures; plicae well impressed, straight, sloping inwards a little, about as long as pronotal plicae. Elytron with well developed inner ridge near apex (ligula). Epipleuron lacking basal carina, relatively broad in anterior quarter, then progressively narrowing to near apex. Ventral surface, Meso and metathorax, metacoxae and abdomen reddish-brown, appendages, pronotum and head lighter. Prosternal process narrow between coxae, reaching mesothorax, apical half relatively broad with parallel raised ridges on each side, not arched in lateral view. Metathorax with wings short, broadly truncated in midline behind, with large deep punctures. Metacoxal plates large, shiny; punctures moderately dense, large, deep; metacoxal lines raised, distinct, moderately widely spaced, reaching to metasternum, weakly diverging; closely adpressed to ventrite 1. Ventrites | and 2 fused, with moderate number of very large deep punctures, sutural line distinct, ventrites 3 to 5 mobile, sparsely covered with small punctures, somewhat denser and larger on ventrite 5. Legs. Protibia triangular, outer edge weakly bow- shaped, widest at apex where it is about four times its basal width; protarsus very weakly expanded, segment | about 2x as long as broad, segment 2 a little wider than segment 1 and a little shorter, segment 3 as long as segment | and a little broader, deeply bifid, segment 4 very small, hidden within lobes of segment 3, segment 5 narrow, cylindrical, about same lengtli as segment 3, segments | to 3 with dense covering of adhesive setae (Fig. 6); claws short and simple. Mesotrochanter elongate, sub- rectangular, with a few relatively long thin setae on inner edge; mesofemur with 2 to 3 moderately strong setae near base on hind edge, much stronger than those on mesotrochanter, mesotarsus slightly longer than protarsus. Metatrochanter tip rounded; metafemur elongate, lacking spines; metatibia narrow, moderately curved, widening towards apex; metatarsus relatively stout, segment | longest, segment 5 longer than segment 4, segments | and 2 in combination about as long as others; claws weak. Male Little external difference between the sexes. Median lobe of aedeagus moderately broad in 6 C.H.S. WATTS & R. LEYS middle, narrowing in apical quarter to blunt point (Fig. 14). Paramere elongate, apical segment with stout apical lobe well separated from rest of segment (Fig. 14). Remarks Very similar to some Australian Leiodytes in size, colour and strong punctation. Differs from Leiodytes in the lack of a slightly raised margin to the front of the head and in the lobed parameres. The small size, unexpanded pro and mesotarsi and very strong ventral punctures readily separate this species from all other Australian Liodessus. The apical lobe of the paramere is less well developed than in other Australian species. Known only from Cape York. Etvmology From its locality — Cape York. Limbodessus compactus (Clark) Hydroporus compactus Clark, 1852, p. 421. Bidessus compactus (Clark). Sharp, 1882, p. 362. Limbodessus compactus (Clark). Guignot, 1939, p. 53. = Bidessus neoguineensis Regimbart. Balke & Sato, 1995, p. 188. = Uvarus tokarensis Sato. Balke & Sato, 1995, p. 188. Limbodessus compactus is the type species of the genus. It has recently been comprehensively dealt with by Balke and Sato (1995) and because of this is not redescribed here. Diagnosis 1.6 — 2.5 mm long, boat-shaped, stout antennae, shiny, uniformly reddish, relatively convex, with cervical stria, with basal carina on elytral epipleuron, male genitalia as in Fig.12. Remarks The species is widely distributed in Japan, SE Asia, New Guinea and Australia (Balke & Sato 1995) (Fig. 23). It is readily separated from other Limbodessus species by its uniform reddish colour, boat-like shape and the presence of transverse carina at the base of the elytral epipleura. Within Australia it is widespread (see below), favouring the littoral zone of still water. In pools in sandy riverbeds in inland Australia the species has been collected interstitially for a short distance beyond the edge of freestanding water. Limbodessus occidentalis and L. rivulus occur in similar places but penetrate further interstitially than L. compactus. Specimens examined New South Wales. 1, Armidale, 1/61; 1, ditto, 21/3/63; 1, Barrington, 17/8/97; 25, 2 km N Batemans Bay, 2/11/97; 2, Berry, 1/63; 5, Collector, 2/61; 2, 2 km N Collector, 26/11/98; 10, 8 km N Failford, 18/8/97; 3, Maclean, 1/61; 3, Nyngan, 16/3/63; Smith’s Lake, 5/70. Northern Territory. 1, 1 km SE Batchelor, 12/4/66, N McFarland; 2, Cahills crossing, Kakadu National Park, 9/6/73, Upton & Feehan; 25, Cannon Hill Kakadu National Park, 10/10/98; 1, 10 mi E Daly River, 26/6/72, BK Head; 3, Darwin, 13/5/63; 1, Gubara, Kakadu National Park, 12/10/98; 1, Howard Springs, 23/3/98; 2, Jabiru, 12/10/98; 3, Mt Borradaile, 8/10/98: 1, 4 mi W Timber Creek, 14/4/66, N McFarland. Queensland. 1, Alligator River, 20 km S Townsville, 25/3/96; 1, 8 km N Bluewater, 3/11/95; 1, Brisbane, 1/61; 17, Bushland Beech, 20 km N Townsville, 26/2/98, AJ. Watts; 3, ditto, 15/3/98: 1, ditto 16/1/98; 4, ditto, 6/2/98; 4, Caloundra, 27/3/63; 5, ditto, 7/3/63; 1, Cairns, 16/4/63; 1, Greenvale, 27/3/96; 2, 70 km SW Greenvale, 28/3/98, AJ Watts; 1, ditto, 11/3/96; 2, Mackay, 4/4/63; 2, Malanda, 13/4/63; 1, 5 km NW Mt Molloy, 5/2/97; 4, Nardello’s Lagoon, 29/3/96; 2, Stanthorpe, 1/61; 2, 25 km § Townsville, 25/3/96; 4, 37 km S Townsville, 6/11/95; 1, Townsville, 31/10/95. South Australia. 1, Naracoorte, 9/3/71; 1, 25 km NE Mt Gambier, 26/3/82, FA. Forrest. Tasmania. 4, 6 km N Pioneer, 13/1/00. Victoria. 2, Dartmoor, 12/61; 7, Fern Tree Gully, 12/61; 6, Healesville, 12/68; 1, 12 km SW Orbost, 30/11/98; 1, 12 km SW Orbost, 5/11/973; 1, Stratford 7/11/97. Western Australia. 2, Cane River HS (old), 22/5/01; 1, Cane River, 22/5/01; 1, 12 km S Newman, 27/5/01; 4,1 km N Red Hill stn., 22/5/01; 3, 30 km N Red Hill stn., 22/5/01; 8, Wittenoom Gorge, 26/5/01. Limbodessus gemellus (Clark) Hydroporus gemellus Clark, 1862, p. 421. Bidessus gemellus (Clark). Sharp, 1882, p. 362. Liodessus gemellus Clark). Watts, 1978, p. 49. = Bidessus mundus Sharp, 1882, p. 362. Watts, 1978, p. 49. Limbodessus gemellus (Clark). Balke and Ribera 2004. Types Bidessus gemellus Clark. Lectotype “gemellus Clark Australia” (yellow label), BMNH. Paralectotypes. 2, “gemellus Clark” (yellow label), BMNH. Bidessus mundus Sharp. Lectotype. ‘Type 88/Australia / n.sp. australia / Bidessus mundus/ (male symbol)”, BMNH. Paralectotypes. 1, “Type 88/australia”; 1, Australia”; both BMNH. “88 / REVIEW OF THE EPIGEAN SPECIES OF AUSTRALIAN LIMBODESSUS GUIGNOT Zt VV VE. Figs 9-10. Male and female protibia of: 9, Limbodessus inornatus; 10, L. shuckardii. Description (number examined, 125.) Figs 15, 24. As for L. amabilis except as follows. Habitus. Length 2.5 — 3.0 mm; weakly constricted at junction of pronotum and elytra. Head. Reddish/yellow, darker towards rear. Pronotum. Reddish-yellow, diffusely darker along front boarder; basal plicae strong, excavated somewhat on inside, curving inwards slightly, straightening towards front. Elvira. Elongate, widest at middle; moderately densely and evenly covered with moderate punctures; plica well impressed, straight, slanting inwards slightly, about as long as pronotal plicae. Legs. Protibia bow-shaped, widest near apex where it is about four times its basal width, protarsus moderately expanded (cf. Fig. 7). Male Dorsal surface shiny; antennae, tarsi and protibiae relatively stout (cf. Figs 3, 7, 9). Median lobe of aedeagus broad at base, narrowing to long, thin, apical portion (Fig. 15). Paramere broad, apical segment weakly triangular, apical lobe large, overlapping rest of apical segment at base, with fine setae (Fig. 15). Female Dorsal surface duller, often with weak to moderate reticulation. Antennae a little stouter. Protibia wider and more triangular (cf. Fig. 9); pro and mesotarsi slightly less expanded (cf. Fig. 7). Remarks Similar in size to L. amabilis and L. praelargus but more uniformly coloured, with reticulate females, sinuate pronotal plicae, broad pro and mesotarsi and, in the female, broad pro and mesotibiae and stout antennae. Separable from the slightly smaller L. shuckardii by the lack of a darker disc on the pronotum, the pronotal plicae more sinuate, and the absence of vague linear markings on the elytra. Limbodessus inornatus from the Southwest is very similar morphologically but differs in having a longer, parallel-sided distal portion to the penis (Fig. 16) and female L. inornatus have somewhat broader pro and mesotibiae (Fig. 9). With a more inland distribution than L. amabilis and L. praelargus but not to the same extent as L. shuckardii. Specimens examined New South Wales. 3, 2 km N Batemans Bay, 18/4/97; 1, Berry, 1/68; 1, 3 km N Bulli, 27/11/98; 2, 8 km N Failford, 18/8/97; 1, Gosford 1/61; 1,2 km S Nowra, 27/1/00; 3, 2 km S Nowra, 27/11/98: 1, Quaama, 18/1/97; 2, Ulladulla, 2/1/973; 1, Windsor, Lea; 2, Waterfall, 1/6/82. Queensland. 1, Queensland, Blackburn coll. South Australia. 1, Adelaide, Griffith coll: 1, Chain of Ponds, 10/11/96; 1, Dalhousie Spr. at light, 6/10/87, J.A.Forrest; 1, Flinders Range, May 59; 1, 10 km N Forreston, 17/9/96; 1, Kuipo 35° 14’ § 145° 138’ E, 5/10/95; 4, Moro Gorge, Flinders Ranges R. Leys; 13, Mt Gambier, 12/16; 3, Myponga, 9/12/96; 3, 13 km W Meadows 36° 11’ S 138° 96’ E, 28/9/96; 2, 10 km E Mt Compass, 10/9/97; 1, Mt Lofty, Lea; 2, Myponga, 5/1/88; 3, 1 km S Nangwarry, 9/10/97; 3, Port Lincoln, Blackburn; 3, ditto, 1/82; 1, Wood Point River Murray, 4/11/80, P. Waller. Tasmania. 1, 5 km E Bridport, 23/1/00; 1, Cradle Mt N P, Jan; 1, 17 km SW Derwent Bridge, 29/11/00; 3, 18 km N Derwent Bridge, 24/1/00; 1, Ellendale, 1/12/00; 1, 2 km SW Ellendale, 1/12/00: 1, 5 km SE Gormanston, 29/11/00; 1, Hatfield River, 15 km SW, 28/11/00; 3, Hobart, 8/61; 2, Hobart, Griffith; 3, 12 km N Hobart, 2/12/00; 1, King Isl, collector unknown; 1, 8 km W Kingston, 3/12/00; 1, 9 km Maydena, 1/12/00; 11, Narcissus Bay Lake St Clair, 30/11/00; 1, 6 km N Pioneer 23/1/00; 3, 8 C.H.S. WATTS & R. LEYS 2 km W Port Latta, 27/11/00; 1, St Helens, Jan; 1, Swansea, Jan; 1, 16 km N Waratah, 28/11/00; 1, 9 km N Queenstown on B28 Road, 28/12/00. Victoria. 1, Buangor, 9/11/97; 4, Dartmoor, 11/10/97; 2, 17 km SW Derwent Bridge, 29/11/00; 2, Healesville, 12/68; 3, 8 km W Kingston, 3/12/002; 1, Omeo, 6/11/97; 3, 30 km W Portland, 10/10/97; 1, Sardine Ck 30 km, N Orbost, 16/1/97; 4, Stratford, 7/11/97. Limbodessus inornatus (Sharp) Bidessus inornatus Sharp, 1882, p. 360. Liodessus inornatus (Sharp). Watts, 1978, p. 48. = Bidessus biformis Sharp, 1882, p. 362. Watts, 1978, p. 48. Limbodessus inornatus (Sharp). Balke and Ribera, 2004. Types Bidessus inornatus Sharp. Holotype Male. “Type 81/West Australia/ Bidessus inornatus n. sp’’, BMNH. Bidessus biformis Sharp Lectotype. “W. Australia/ type 89 d Bidessus biformis n.sp. K.G.Sound”, BMNH. Paralectotypes. 3, same data as lectotype, BMNH. 3, “W. Australia, Swan River” BMNH. 6, “W. Australia”, BMNH. Description (number examined, 351) Figs 3, 7, 9,16, 25: As for L. gemellus except as follows. Male Median lobe of aedeagus relatively broad, narrowing in apical quarter to blunt tip (Fig. 16). Paramere broad, apical segment rounded at apex, apical lobe long, overlapping rest of segment, with fine setae and some peg-like structures on front edge (Fig. 16). Female Pro and mesotibiae expanded (Fig. 9). Remarks Restricted to the Southwest where it is sympatric with L. shuckardii. Can be separated from this species by its larger size, uniform colour on the pronotum, and the pronotal plicae sinuate rather than curved. Female L. inornatus have slightly broader pro and mesotibiae (Fig. 9). The front of the apical lobe of the paramere has several peg-like structures not found in other Australian Limbodessus (Fig. 16). Indistinguishable from the eastern L. gemellus apart from a stouter distal portion to the penis and the peg- like structures on the parameres. Specimens examined Western Australia. 3, Armadale, 3/62, D. Edwards; 3, Bickley Swamp, Rottnest Island, 10/58, D. Edwards; 1, Blackwood River, Nannup, 20/10/96; 1, Bonanup, 17/10/96; 13, Bridgetown, 11/9/31, Darlington; 3, 50 km W along Broke inlet Rd Nr Walpole, 18/9/00; 4, 3 km N Bullsbrook, 16/10/96; 1, Bushy Swamp 15 km WNW _ Woodanilling, 21/9/00; 16, Byenup Lagoon NR, 21/9/00; 6, Corio Spring Rottnest Island, 10/58, D. Edwards; 4, 5 km W Cowaramup, 22/10/00; 1, 2 km SW Dandalup, 23/9/00; 2, 20 km SE Donnybrook, 18/10/96; 3, 30 km S Dwellingup, 17/10/96; 1, 8 km S Dwellingup, 17/10/96; 1, Ellen Brook NR, 14/9/00; 8, 19 km S Fremantle, 24/10/96; 1, Geraldton 10/31, Darlington; Hay River, collector unknown; 13, Ironstone Gully Falls 13 km SW Donnybrook, 22/10/96; 15, 10 km E Kalamunda, 16/10/96; 7, Kodjinup NR, 21/9/00 : 7, Lake Pleasant View, 17/9/00; 1, Lake Parkeyerring, 15/9/00; 1, Lake Poorginup, 20/9/00; 17, Maidavale, 27/4/90; 11, Margaret River, 10/31, Darlington; 7, 4 km N Mumballup, 23/9/00; 1, Nalyerin Lake, 22/9/00; 13, 4 km S New Norcia, 15/10/96; 15 km NW Pemberton, 17/5/87; 1, 20 km W Pemberton, 20/10/96; 7, Pemberton, 10/31, Darlington; 2, 15 km NE Pemberton, 8/10/96; 69, 6 km S_ Pinjarra, 23/9/00; 48, 30 km N Perth, 14/10/96; 1, Rottnest Island, 10/31, Darlington; 3, 12 km W Serpentine, 24/10/96; Stirling Range, 10/72; 2, Swan River, Lea; 9, Swan River, 10/57, D. Edwards; 11, 20 km W Strachan, 21/9/00; 1, 1 km S Wagin, 21/9/00; 1, Wilgarup River, 6/58, D. Edwards; 1, 16 km N Woodanilling, 15/9/00; 4, Yallingup, 22/10/96. Limbodessus occidentalis (Watts and Humphreys) Boongurrus occidentalis Watts and Humphreys, 2004. Limbodessus occidentalis (Watts & Humphreys) This species has only recently been described (Watts and Humphreys 2004) and hence will not be redescribed here. Diagnosis 1.9 — 2.3 mm long, relatively flat, elongate-oval, light reddish, elytra darker, without cervical stria, without basal carina on elytral epipleuron, eyes of normal size, male genitalia as in Fig. 17. Remarks Limbodessus occidentalis differs from all other epigean Limbodessus except L. rivulus, by the absence of a cervical stria between the hind edges of the eye. Like L. rivulus it is small (1.9 — 2.3 mm long), flattened with a rugose surface but unlike L. rivulus the eyes are not reduced in size. It occurs at the headwaters of sandy/gravely streams, or at the REVIEW OF THE EPIGEAN SPECIES OF AUSTRALIAN LIMBODESSUS GUIGNOT 9 GOP G Figs 11-18. Dorsal view of penis, lateral view of penis and paramere of: 11, Allodessus bistrigatus; 12, Limbodessus compactus; 13, L. amabilis; 14, L. capeensis; 15, L. gemellus; 16, L. inornatus; 17, L. occidentalis; 18, L. praelargus. \ 10 C. H. S. WATTS & R, LEYS upstream edges of pools in drying riverbeds, in the Pilbara and Yilgarn regions of Western Australia (Fig. 26), often interstitially several meters away from the water’s edge. It has also occasionally been taken together with a true stygobitic fauna in shallow calcrete aquifers (Watts and Humphreys 2004). Specimens examined (Holotype & Paratypes) Western Australia. 9, 10 km NW Eerala Stn, 23/5/01; 1, Killara Station, 6/6/02,W. F. Humphreys & R. Leys); 3, Moorarie Station, 8/6/02, W. F. Humphreys & R. Leys; 1, Wagga Wagga Station, 4/6/02, W. F. Humphreys & R. Leys; 12, Wittenoom Gorge, Town Pool, 26/5/01, C. H. S. Watts & G. A. Watts. Limbodessus praelargus (Lea) Bidessus praelargus Lea, 1898, p. 523. Liodessus praelargus (Lea). Watts, 1978, p. 51. Limbodessus praelargus (Lea). Balke and Ribera, 2004. Types Holotype Male, “praelargus Lea TYPE 6d Forest Reefs”; dissected and remounted this study, SAMA. SAMA data base # 25-001525. Paratypes 5 females, same locality; dissected and remounted this study, SAMA.SAMA data base # 25-009153. Description (number of dissected males examined, 13) Figs 18, 27. As for L amabilis except: Male Medial lobe of aedeagus moderately broad, smoothly narrowing to a blunt point (Fig. 18). Paramere broad, apical lobe stout, overlapping apical portion of apical segment which is relatively narrow and has the apex pointed rather than rounded (Fig. 18). Remarks Indistinguishable from L. amabilis except for the pointed penis and narrower paramere. The range of the two species is broadly similar. Watts (1978) correctly (but fortuitously) associated the name praelargus with the species with the pointed penis. See also under L. amabilis. Specimens examined (localities of males with genitalia extracted.) Australian Capital Territory. 30 mi S Canberra, 1/61. New South Wales. 14 km W Delagate, 4/11/97. Victoria. | 1 km E Bruthen, 6/11/97; 4 km S Glenista, 24/9/98; 10 km NE Mirranatwa, 12/10/97; 5 km NW Portland, 10/10/97. South Australia. Mt Gambier, 12/61; 1 km S Nangwarry, 9/10/97. Tasmania. S end of Lake St Clair, 30/11/00; Little Pine Lake, 8 km W Miena, 23/10/00; 2 km W Port Latta, 27/11/00. Limbodessus rivulus (Larson) Boongurrus rivulus Larson, 1994. Limbodessus rivulus (Larson). Balke and Ribera, 2004. This species has recently been treated in detail (Larson 1994) and hence will not be redescribed here. Diagnoses 1.8 — 2.2 mm long, rather rectangular, relatively flat, dull reddish with darker areas, strongly reticulate, lacking (or virtually lacking) cervical stria, lacking basal carina on elytral epipleuron, eyes much smaller than normal for epigean Limbodessus, male genitalia as in Fig. 19. Remarks Together with L. occidentalis the only epigean Limbodessus lacking a cervical stria between the rear of the eyes although Larson (1994) reported its partial presence in a small number of specimens. Separated from L. occidentalis by its obviously reduced eyes and characters of the male genitalia (Watts and Humphreys 2004). The species occurs at ( Eee 20 Figs 19-20. Dorsal view of penis, lateral view of penis and paramere of: 19, Limbodessus rivulus; 20 L. shuckardii. REVIEW OF THE EPIGEAN SPECIES OF AUSTRALIAN LIMBODESSUS GUIGNOT he A OP Se TF ew ‘. a '. aN al } wk ae Sy { a ‘, ao: Le ie of yee 27 ‘eg! a4 ts. ky ate “yeu y é r: i \y a ~ oh SYD f *, ‘ vA be ~. irr; : f i \ ; ~s ‘ ae Ce —— yds a rd \ _ ‘ ‘Rtas Na bP . bo ae 23 28 hin \y 4 : ho. my ” { i eo. 24 ; et \ nce “Ny * : —— ‘ ae 5 Te oa Ma. ew Pa) ‘od Figs 21-29. Distribution maps of specimens of Limbodessus in the South Australian Museum Entomological collection. 21, 28, L. rivulus; 29, L. shuckardii. L. amabilis; 22, L. capeensis: 23, L. compactus; 24, L. gemellus; 25, L. inornatus: 26, L. occidentalis; 27, L. praelargus; 12 C.H.S. WATTS & R. LEYS the headwaters of small, gravely, spring-fed streams in areas of open eucalypt woodland on the Atherton Tablelands. Here it occurs in gravel at the edge of the water or interstitially for at least a metre from the waters edge (pers. observation). Very recently (August, 2004), I have collected a single specimen of this, or a closely related species, from similar habitat on Mt Tamborine in southeast Queensland. Specimens examined Queensland. |, | km E Watsonville, 31/3/96; 11, 1.5 km E Watsonville, 3/8/03. Limbodessus shuckardii (Clark) Hydroporus shuckardii Clark, 1862, p. 420. Bidessus shuckhardii (Clark). Sharp, 1882, p. 361. Liodessus shuckhardi (Clark). Watts, 1978, p. 47. Limbodessus shuckardii (Clark). Balke and Ribera, 2004. Bidessus dispar Sharp, 1882, p. 363. Liodessus dispar (Sharp). Watts, 1978, p. 48. = Bidessus elegans Lea, 1898, p. 523. Watts 1978, p. 48. Limbodessus dispar (Sharp). Balke and Ribera, 2004. =Limbodessus shuckardii (Clark). Syn. nov. Types Hydroporus shuckardii Clark Lectotype. *67.56 /shuckardii Clark australia’ (yellow label), BMNH. Paralectotypes. 1, ‘shuckardii (yellow label); 2, ‘S. Australia. Bakewell, 59.24; 10, ‘australia’; all in BMNH. Bidessus dispar Sharp. Lectotype. ‘Type 90 2 a / Swan River / W. Australia/ W. Australia / Sharp Coll 1905-313 / Bidessus dispar Sharp 2 a Type’ (in Balfour- Browne’s writing). Paralectotypes. 2, same locality; 1, ‘W. Australia’; both in BMNH. Bidessus elegans Lea. Holotype ‘elegans Lea TYPE Beverley’; left-hand specimen on card marked with TY, SAMA. SAMA data base # 25-001582. Paratype Same data as Holotype; right-hand specimen on card, SAMA. SAMA data base # 25-009340. Description (number examined, 244.) Figs 4, 8,10, 20, 29. As for L. amabilis except: Habitus. Length 1.9 — 2.6 mm; very weakly constricted at junction of pronotum and elytra. Head, Light reddish-yellow with darker patches towards rear. Antenna (male) stout, segment | cylindrical, segment 2 as long as segment 1, barrel- shaped, segment 3 as long as segment 2, narrower, narrowing towards base, segment 4 shorter than segment 3, segments 5-10 subequal, slightly wine- glass shaped, segment 11 about twice length of segment 10, narrower (Fig. 4). Pronotum. Light reddish-yellow with diffuse reddish-brown area on disc. Basal plicae strong, excavated somewhat on inside, curving inwards quite strongly, reaching about two-thirds of way to front margin of pronotum. Elytra. Light reddish/yellow to reddish yellow with reddish brown pattern, 2-3 thin longitudinal lines partially discernible in most specimens. Elongate, widest at middle; shiny, moderately densely and evenly covered with moderate punctures; plicae well impressed, slanting inwards, about as long as pronotal plicae. Ventral surface. Light reddish-yellow, meso and metathorax and metacoxae darker. Metathorax with wings short, truncated or broadly rounded in midline behind, virtually impunctate. Legs. Protibia bow-shaped, widest before apex where it is about 3x its basal width; protarsus moderately expanded, segment | a little longer than wide, segment 2 a little wider than segment 1, about half as long, segment 3 about as long as segment | and same width, deeply bifid, segment 4 very small, hidden within lobes of segment 3, segment 5 narrow, cylindrical, a bit longer than segment 3, segments | to 3 with dense covering of adhesive setae. Male Shiny, antenna and tarsi as above (Figs 4, 8). Median lobe of aedeagus moderately broad, narrowing a little quite abruptly in apical quarter in both dorsal and lateral views (Fig. 20), Paramere moderately broad, apical segment rather small, apical lobe relatively long, overlapping rest of apical segment, with long setae (Fig. 20). Female Both dorsal and ventral surfaces weakly to moderately reticulate. Antenna stouter, segments 5- 10 almost bead-like (Fig. 4). Pro and mesotarsi broad, but less so than in male (Fig. 8). Remarks A relatively small species recognised by the broad pro and mesotarsi, pronotal plicae curving inwards and the centre of the pronotum diffusely slightly darker than the rest. The elytra have indistinct thin REVIEW OF THE EPIGEAN SPECIES OF AUSTRALIAN LIMBODESSUS GUIGNOT 13 linear markings reminiscent of those of Hydroglyphus grammopterus (Zimmerman) which can be faintly seen on most specimens. Other Australian Limbodessus can have quite marked linear elytral markings but these are broader and interrupted in the central region. A more inland distribution than other species, present as far north as the Mount Isa — Greenvale region of north Queensland, seemingly absent from Tasmania. Specimens examined New South Wales. 1, Collector, 20/1/97; 1, Forbes, 15/3/63; 2, Gilgandra, 19/11/92; 2, Ditto, 9/2/62; 2, Grenfell, collector unknown; 9, Nyngan, 16/3/63. Queensland. 11, Camooweal, 30/4/93; 2, Charters Towers, 23/4/63; 1, Cloncurry, 29/4/63; 1, Coorabulka, 7/71; 1, 70 km SW Greenvale, at light, 28/3/95 to 7/4/95, A. J. Watts; 1, ditto, 21-31/10/95; 1, ditto, 29/1/97 to 4/2/97; 1, ditto, 3-10/10/96; 1, Lake Buchanan, 25/9/83, B Timms. South Australia. 2, Adelaide; 1, Alligator Gorge, 6/58; 4, Chain of Ponds, 12/62; 1, 20 km N Coober Pedy, 2/68; 2,10 km N Forreston, 3/9/99; 1, Leigh Creek, Blackburn’s coll: 2, | km S Nangwarry, 9/10/97. Victoria. 1, Albury, 16/7/89, P. Waller; 1, 12 km W Brimpaen, 23/9/98; 5, Buangor, 9/11/97; 1, Dartmoor, 24/9/98; 2, Dodswell Bridge, 10/10/98, D. Churches; 12, Fyans Creek, 15 KS Stawell, 13/1/97; 2, 4 km S Glenista, 24/9/96; 6, Grampians, 2/63; Halls Gap, 13/1/97; 11, Healesville, 12/68; 1, Lake Hattah, Light trap, 28/10/67, G. W. Anderson; 8, 12 km N Mirranatwa, 12/10/97; 2, 10 km NE Mirrantwa, 12/10/97; 1, 5 km NW Portland, 10/10/97; 3, Nathatia, 9/6; 2, Turret Falls, 5 km NW Halls Gap, 13/1/97; 1, Wellington River 4 km N Licola, 30/11/98. Western Australia. 2, Armadale, 7/62, D. Edwards: |, Beverley, Lea; 8, Bridgetown, 9/11/31, Darlington; 1, 5 km N Bushy Swamp nr Wagin, 21/9/00; 2, Byenup Lagoon NR, 21/9/00; 4, 8 km S Dwellingup, 17/10/96; 69, 6 km S§ Pinjarra, 23/10/96; 2, 8 km N Pinjarra, 23/10/96; 1, Ironstone Gully Falls, 13 km SW Donnybrook, 22/10/96; 4, 10 km S Fremantle, 24/10/96; 1, 4 km N Mumballup, 23/10/96; 1, Nalyerin Lake, 22/9/00; 1, 5 km E Lake Nalyerin, 22/9/00; 1, Riffle Range Swamp, Rottnest Island, 10/59, D. Edwards; 16, 12 km W Serpentine, 24/10/96; 1, 10 km S Yallingup, 22/10/96. Acknowledgments We would like to thank Howard Hamer for drawing the illustrations, Debbie Churches and Archie McArthur for helping with the manuscript and Micheal Balke for information on the types in the BMNH and comments on the manuscript. References BALKE, M. & Ripera, I, 2004. Jumping across Wallace’s line: Allodessus and Limbodessus revisited (Coleoptera: Dytiscidae, Bidessini) based on molecular phylogenetic and morphological data, Australian Journal of Entomology 43,1 14-128. Cooper, S. J. B., Hinze, S., Leys, R., Watts, C. H. S. & Humpureys, W. F, 2002. Islands under the desert: molecular systematics and evolutionary origins of stygobitic water beetles (Coleoptera: Dytiscidae) from central Western Australia. /nvertebrate Systematics 16, 589-598, Larson, D. J., 1994. Boongurrus rivulus, a new genus and species of water beetle (Coleoptera: Dytiscidae: Bidessini) from northern Queensland, Australia. Journal of the Australian Entomological Society 33, 217-221. Nitsson A. N., 2001. World Catalogue of insects Vol 3 Dytiscidae. Apollo Books, Stenstrup, Denmark. Watts C. H. S,, 1978. A revision of Australian Dytiscidae (Coleoptera). Australian Journal of Zoology. Supplementary Series, 57, 1-166 , 2002. Checklists & guides to the identification, to genus, of adult & larval Australian water beetles of the families Dytiscidae, Noteridae, Hygrobiidae, Haliplidae, Gyrinidae, Hydraenidae and the — superfamily Hydrophiloidea (Insecta: Coleoptera. /dentification & Ecology guide no. 43, 109pp. Murray Darling Freshwater Research Centre, Albury. __ & Humpnreys W. F., 2001. A new genus and six new species of Dytiscidae (Coleoptera) from underground waters in the Yilgarn palaeodrainage system of Western Australia. Records of the South Australian Museum 34, 99-114. & , 2003. Twenty-five new Dytiscidae (Coleoptera) of the genera Tjirtudessus Watts & Humphreys, Nirripirti Watts & Humphreys and Bidessodes Regimbart, from underground waters in Australia. Records of the Foust Australian Museum 36, 135-187. , 2004. Thirteen new Dytiscidae (Galopiens) of the genera Boongurrus Larson, Tjirtudessus Watts & Humphreys and Nirripirti Watts & Humphreys, from underground waters in Australia. Trans. R, Soc. S. Aust. 128, 99-129, Transactions of the Royal Society of S. Aust. (2005), 129(1), 14-24. SCLEROPHYLL (HEATHY) UNDERSTOREYS IN THE MOUNT LOFTY RANGES, SOUTH AUSTRALIA by H. A. Martin* & R. L. SPECHT Summary Martin, H. A. & SPECHT, R. L. (2005) Sclerophyll (heathy) understoreys in the Mount Lofty Ranges, South Australia. Trans. R. Soc. S. Aust. 129(1), 14-24, 31 May, 2005. Floristic patterns in the sclerophyll (heathy) understoreys, which are characteristic of the nutrient-poor soils of the Mount Lofty Ranges, are surveyed: — on Pre-Cambrian schists at Inglewood (rainfall c. 800 mm per annum); on Pre-Cambrian quartzites at Morialta (rainfall c. 800 mm per annum); on Pre-Cambrian quartzites in Waterfall Gully (rainfall c. 1150 mm per annum); on truncated laterites of Early Tertiary in National Park Belair (rainfall c. 800 mm per annum); and on deep Mid-Tertiary sands at Blewitt Springs (rainfall c. 800 mm per annum). Seven Floristic Groups have been defined objectively using Goodall’s ‘positive interspecific correlation’ technique, adapted for computer-analysis. The floristic patterns appear to be determined by the annual cycle of available soil water at each site and the slightly different nutrient levels of sandy versus clayey surface soils. Key Woros: Floristic Groups, heath; sclerophyll, understorey, objective classification, nutrient-poor soil. Introduction Two distinct formations, termed Sklerophyllen- Wald and Savannen-Wald by Diels (1906), are characteristic of the vegetation of the Mount Lofty Ranges (Adamson & Osborn 1924; Specht & Perry 1948: Specht er a/. 1961). The former formation, translated as ‘dry sclerophyll forest’ (now heathy open-forest, Specht 1970), is invariably associated with extremely nutrient-poor soils. The latter formation, translated as ‘savanna forest/woodland’ (now grassy open-forest/woodland, Specht 1970) is on more fertile, but not nutrient-rich, soils (Specht & Rundel 1990; Specht & Specht 1999). In the Mount Lofty Ranges, the transition from one formation to the other is determined by the geological substrate that changes abruptly over a short distance (Sprigg 1946). Remnants of the very infertile, lateritic soils that were formed on the Late Cretaceous to Early Tertiary peneplain persisted until the uplift of the Mount Lofty Ranges and still overlie the geological formations on the Eden-Moana Fault Block (Stephens 1946, 1971; Specht & Perry 1948). The original understorey of the savanna (grassy) woodlands on the more fertile soils was dominated by Danthonia (Adamson & Osborn 1924; Davies & Sim 1931; Wood 1937), but is now invaded by many introduced species (Specht & Perry 1948; Specht er al. 1961; Specht 1972, 1975; Specht & Clifford 1991; Specht 2000, 2001, 2002). The nutrient-poor “School of Biological, Environmental and Earth Sciences, University of New South Wales, Sydney, New South Wales 2052. ‘ Emeritus Professor of Botany, The University of Queensland. Current address: 107 Central Avenue, St Lucia, Queensland 4067. soils that support sclerophyll (heathy) open- forests/woodlands are unsuitable for the establish- ment of invasive plants, unless the soil phosphate levels are increased after bush-fires or from pollution (Specht 1963; Heddle & Specht 1975: Specht & Specht 1989, 1999; Specht 2001, 2002). The floristics of many stands of heathy open- forests in the Mount Lofty Ranges were collated in the ecological survey conducted in 1946-47 (Specht & Perry 1948) and again in ‘The Vegetation of South Australia’ (Specht 1972). In the late 1950s, Helene Martin examined in detail the distribution of the component species in the gradient from heathy woodland to heathy open-forest on Pre-Cambrian schists (34° 48’ S, 138° 48’ E) about 2.5 km north- east of Inglewood, north of the Torrens Gorge (Martin 1961; Martin & Specht 1962). In this paper, the changes in floristics of sclerophyll (heathy) understorey vegetation are examined in detail on the nutrient-poor soils that have developed along the rainfall isohyet of c. 800 mm per annum: (1) Under Eucalyptus obliqua — E. goniocalyx syn. E. elaeophora (+ E. fasciculosa) open-forest on Pre-Cambrian schists north-east of Inglewood (Specht e¢ al. 1961; Martin 1961); (2) Under E. obliqua — E. baxteri (+ E. fasciculosa) open-forest on Pre-Cambrian quartzites at Morialta (Specht & Perry 1948); (3) Under E. microcarpa syn. E. odorata — E. leucoxylon woodland on truncated laterites of the Early Tertiary in National Park Belair (Specht & Perry 1948); (4) Under E. fasciculosa woodland on deep Mid- Tertiary sands at Blewitt Springs (Specht & Perry 1948), SCLEROPHYLL VEGETATION IN THE MOUNT LOFTY RANGES 15 a i <= a> +8 As sores na me === \ = eae be Rivet ST. VINCENT GULF DISTRIBUTION of FORMATIONS =| Dry sclerophy!! = Banksia ornata = =| dominant wa ~ES) Banksia marginata - t ns dominant C] Savanna km oO i 3 83 a a | Fig. |. Location of study sites. Morialta (M), Waterfall Gully (WG), National Park Belair (NPB) and Blewitt Springs (BS), in the sclerophyll (heathy) vegetation of the Mount Lofty Ranges (after Specht & Perry 1948). The study site at Inglewood is 12 km NNE of Morialta (Specht ef al. 1961; Martin 1961). Banksia Ornata and Banksia marginata are dominants in the heathy understorey, not the overstorey. 16 H. A. MARTIN & R. L. SPECHT Group 1 (726) + Hibbertia stricta + Acacia myrtifolia Group 4 Group 5 (208) (442) + Astroloma conostephioides + Leptospermum myrsinoides Group 8 Group 9 Group 6 Group 7 (82) (126) (254) (188) + Platylobium + Hibbertia obtusangulum acicularis + Epacris impressa Group 10 Group 11 (104) (150) Group 14 Group 15 Group 12 Group 13 (62) (64) (141) (47) Fig. 2. Association analysis of sclerophyll (heathy) understoreys in the Mount Lofty Ranges. Blewitt Springs — Group 2; Inglewood — Groups 10, 11, 12, 13 and 15; Morialta — Groups 8 and 13; National Park Belair — Groups 12 and 15; Waterfall Gully — Group 14. SCLEROPHYLL VEGETATION IN THE MOUNT LOFTY RANGES 17 TABLE |. Major floristic composition of sclerophyll (heathy) understoreys in Floristic Groups (Fig. 2) in the Mount Lofty Ranges. Species that enabled a dichotomous split in the ‘association analysis’ (Fig. 2) are shown in bold. Group 2 — Blewitt Springs (42 spp.) Tall shrubs (M): Acacia pycnantha Low shrubs (N): Acacia myrtifolia, A. spinescens, Allocasuarina muelleriana, Banksia ornata, Calytrix tetragona, Dillwynia sericea, Hakea ulicina, Leptospermum myrsinoides, Olearia ramulosa, Xanthorrhoea semiplana Sub-shrubs (Ch): Astroloma conostephioides, A. humifusum, Hibbertia stricta, H. virgata Graminoids (H): Lepidosperma carphoides Group 4(8) — Morialta (61 spp.) Low shrubs (N): Acacia myrtifolia, Epacris impressa, Hakea rostrata, Ixodia achillaeoides, Leptospermum myrsinoides, Pultenaea daphnoides, Xanthorrhoea semiplana Sub-shrubs (Ch): Acrotriche serrulata, Astroloma conostephioides, Hibbertia sericea, Leucopogon virgatus, Tetratheca pilosa Graminoids (H): Lepidosperma semiteres Group 6(10) — Inglewood (39 spp.) Low shrubs (N): Dillwynia hispida, Hakea rostrata, Leptospermum myrsinoides, Platylobium obtusangulum, Pultenaea largiflorens, Xanthorrhoea semiplana Sub-shrubs (Ch): Acrotriche serrulata, Astroloma humifusum, Hibbertia acicularis, H. sericea, Leucopogon virgatus, Pimelea linifolia Graminoids (H): Lepidosperma semiteres Group 6(11) — Inglewood (44 spp.) Low shrubs (N): Allocasuarina muelleriana, Hakea rostrata, Leptospermum myrsinoides, Xanthorrhoea semiplana Sub-shrubs (Ch): Acrotriche serrulata, Astroloma humifusum, Hibbertia acicularis, H. sericea Graminoids (H): Lepidosperma semiteres Group 7(12) —Inglewood and National Park Belair (43 spp.) Tall shrubs (M): Acacia pycnantha Low shrubs (N): Pultenaea daphnoides, Xanthorrhoea semiplana Sub-shrubs (Ch): Acrotriche serrulata, Astroloma humifusum, Hibbertia acicularis, H. sericea Graminoids (H): Lepidosperma semiteres Group 7(13) — Inglewood and Morialta (48 spp.) Tall shrubs (M): Acacia pycnantha Low shrubs (N): Olearia ramulosa Sub-shrubs (Ch): Astroloma humifusum, Hibbertia sericea Group 9(14) — Waterfall Gully (47 spp.) Low shrubs (N): Acacia myrtifolia, Acrotriche fasciculiflora, Epacris impressa, Hakea rostrata, Hakea ulicina, Ixodia achillaeoides, Leptospermum myrsinoides, Platylobium obtusangulum, Pultenaea daphnoides, Xanthorrhoea semiplana Sub-shrubs (Ch): Acrotriche serrulata, Leucopogon virgatus, Tetratheca pilosa Graminoids (H): Lepidosperma semiteres, Lomandra fibrata Group 9(15) — Inglewood and National Park Belair (54 spp.) Low shrubs (N): Acacia myrtifolia, Leptospermum myrsinoides, Platylobium obtusangulum, Pultenaea daphnoides, Xanthorrhoea semiplana Sub-shrubs (Ch): Astroloma humifusum, Hibberta acicularis, H. sericea Graminoids (H): Lepidosperma semiteres The floristics of the sclerophyll (heathy) Methods understorey vegetation on Pre-Cambrian quartzites at Morialta (annual rainfall c. 800 mm) and under the The location of the study sites at Morialta (34° 50’ E. obliqua — E. baxteri (+ E. cosmophylla) open- 8, 138° 40’ E), Waterfall Gully (35° 00’ S, 138° 40’ forest in Waterfall Gully (rainfall c.1150 mm per _ E), Belair National Park (35° 02' S, 138° 38’ E) and annum) are compared. Blewitt Springs (35° 10’ S, 138° 34’ E) are shown on 18 H. A. MARTIN & R. L, SPECHT Fig. 1 (after Specht & Perry 1948). The study site at Inglewood (34° 48’ S, 138° 47’ E) is 12 km NNE of Morialta. At each site, the species composition in quadrats (20m x Im) was recorded along. transects, established on both north- and south-facing slopes and spaced 20m along the top of the ridge. Nomenclature of the species follows Jessop & Toelken (1986). In all, 852 quadrats were recorded: 414 on Pre- Cambrian schists at Inglewood; 117 on Pre- Cambrian quartzites at Morialta; 141 on Pre- Cambrian quartzites at Waterfall Gully; 89 on truncated laterites of the Early Tertiary in Belair National Park; and 91 on deep Mid-Tertiary sands at Blewett Springs. The data was collated in 1960. Species densities (per 20m? quadrat) were plotted for major species in the understorey at each site to construct isonome maps (Pidgeon & Ashby 1942; Rayson 1957; Brewer née Pidgeon 1995). The isonome maps for the large data-set of 414 quadrats in the sclerophyll (heathy) vegetation in the climatic gradient north-east of Inglewood were published by Martin (1961); the smaller data-sets at the other sites, where rainfall is relatively uniform, are less variable in floristic patterns. Floristic Groups in the large data-bank of 852 quadrats (20m x Im), collected from the five sclerophyll (heathy) sampling sites, were defined objectively by ‘association analysis’, using Goodall’s ‘positive interspecific correlation’ technique (Goodall 1953), adapted for computer analysis by W. T. Williams of C.S.I.R.O. Division of Computing Research (Clifford & Stephenson 1975). Firstly, the chi-squared values for every pair of species in the data-bank are calculated. Next, the sum of all the chi- squared values for each species is totalled; the species with the highest chi-squared sum is considered as the ‘indicator species’ for the positive group of the dichotomous split of the data-bank. These positive and negative floristic groups are considered to be distributed in different micro- habitats in the vegetation under study (Clifford & Specht 1979; Specht & Specht 1999), Results The large data-bank of 852 quadrats (20m x Im) recorded in the sclerophyll (heathy) understoreys of the Mount Lofty Ranges (Appendix 1), from Inglewood, north of the Torrens Gorge, to Blewitt Springs in the south, was analysed objectively by the classification program ‘association analysis’ (Fig. 2 and Table 1). The objective analysis recognized eight Floristic Groups: Floristic Group 14, with Acacia myrtifolia, Leptospermum myrsinoides, Pultenaea daphnoides, and Epacris impressa common in the understorey, is a relatively uniform community developed in high rainfall on the quartzitic outcrop between Mount Lofty Summit and Waterfall Gully. Floristic Group 8, with Acacia myrtifolia, Leptospermum myrsinoides, Astroloma cono- stephioides, and Hibbertia sericea prominent in the understorey, occupies most of the quartzitic outcrop above the Waterfalls at Morialta. Floristic Groups 10, 11, 12, 13 and 15 form a graded series of communities from the wettest to the driest part of the Inglewood District on podsolic soils developed on Pre-Cambrian schists (Martin 1961). The following species are either the most frequent (or the most indicative) in the understorey: Floristic Group 11 — Hakea — rostrata, Leptospermum myrsinoides, Hibbertia acicularis, H. sericea, Lepidosperma semiteres. Floristic Group 12 — Platylobium obtusangulum, Lepidosperma semiteres (+ Pultenaea daphnoides, Lomandra fibrata) Floristic Group 10 — Hibbertia acicularis, H. sericea, Pimelea spathulata (+ Pultenaea largiflorens). Floristic Group 15 — Hibbertia acicularis, Lepidosperma semiteres. Floristic Group 13 — Hibbertia sericea (+ Acacia pycnantha). The Floristic Groups 13 and 15 in the driest part of the sclerophyll continuum at Inglewood tend to approach the herbaceous understorey of the grassy savanna. The sclerophyll understorey in the driest part of Morialta is similar to Floristic Groups 13 found at Inglewood. The sclerophyllous understoreys on the truncated lateritic podsol in Belair National Park belong to the same Floristic Groups 12 and 15 found in the Inglewood study. Floristic Group 2, with Banksia ornata, Xanthorrhoea semiplana, and Hibbertia stricta widespread in the understorey, is found on the deep sandy podsols near Blewitt Springs. Discussion Association analysis of the 852 quadrats (20m x Im in dimensions) recorded in the sclerophyll (heathy) vegetation over 40 km distance north-south in the Mount Lofty Ranges indicated that the Floristic Groups found on the Pre-Cambrian quartzite of Waterfall Gully (Group 14) and at Morialta (Group 8) were distinctly different from the Floristic Groups found on the Mid-Tertiary sands at Blewitt Springs (Group 2) and in the rainfall gradient on Pre-Cambrian schists at Inglewood (Groups 10 to 13 and 15). The ecotonal Floristic Group 13 between sclerophyll and savanna vegetation at Inglewood, SCLEROPHYLL VEGETATION IN THE MOUNT LOFTY RANGES 19 north of the Torrens Gorge, was also found at the driest margin of Moritalta to the south. Both Floristic Groups 12 and 15 on Pre-Cambrian schists at Inglewood also occurred on the truncated lateritic podsol near Pines Oval in Belair National Park. The shrubs, Acacia myrtifolia, Epacris impressa, Hakea_ rostrata, Leptospermum — myrsinoides, Pultenaea daphnoides and Xanthorrhoea semiplana were common on the Pre-Cambrian quartzites in Waterfall Gully (Group 14) and at Morialta (Group 8). These shrubby species were associated with sub- shrubs such as Acrotriche serrulata, Hibbertia sericea, Leucopogon virgatus and Tetratheca pilosa, while Astroloma conostephioides became prominent in the drier Morialta understorey (Group 8). The tussock sedge Lepidosperma semiteres was common in the ground stratum in both Groups. Acacia myrtifolia, Allocasuarina muelleriana, Banksia ornata, Leptospermum myrsinoides and Xanthorrhoea semiplana, with Hibbertia stricta as a diagnostic sub-shrub species, dominated the sclerophyll understorey on the deep sands at Blewitt Springs (Group 2). This understorey Floristic Group is very similar to the treeless heathland that dominates the deep sands in the Ninety-Mile Plain (Specht & Rayson 1957; Rayson 1957; Specht eg al. 1958). In the Floristic Groups on the clayey soils of the Pre-Cambrian schists at Inglewood, the wettest Groups (Groups 10, 11 and 15) are dominated by shrubby species such as Hakea rostrata, Leptospermum myrsinoides and Xanthorrhoea semiplana, together with sub-shrubs Acrotriche serrulata, Astroloma humifiusum, —Hibbertia acicularis and H. sericea, with the tussock sedge Lepidosperma semiteres in the ground stratum. Floristic Group 15, with distinctive species such as Acacia myrtifolia, Platylobium obtusangulum and Pultenaea daphnoides that occur in the wetter section of the rainfall gradient at Inglewood, is also defined in the sclerophyll understorey on the clayey truncated lateritic soils in Belair National Park. The golden wattle, Acacia pycnantha occurs as scattered small trees in the ecotone between the sclerophyll (heathy) and the savanna (grassy) vegetation at Inglewood, Morialta and Belair National Park (Groups 12 and 13), where shrubby species are rare, but sub-shrubs such as Astroloma humifusum and Hibbertia sericea persist. Ecological interpretation of the objective classification, in relation to the climate and soils of the Mount Lofty Ranges, was summarised for ‘The Vegetation of South Australia’ (Specht 1972) and expanded in this paper. Conclusions The distribution of species in the sclerophyll (heathy) understorey of the Mount Lofty Ranges is determined by a combination of (1) the annual cycle of available soil water (Martin & Specht 1962; Rayson 1957; Specht 1957a, 1957b) and (2) the interaction of infertile sandy surface soils versus clayey surface soils (Specht 1988; Specht & Rundel 1990; Specht & Specht 1999). The ability of the sclerophyll (heathy) vegetation to survive, since the Late Cretaceous, on such infertile soils, and to be replaced by savanna (grassy) understorey on more fertile soils, was the major stimulus for the long-term nutrition experiments on Dark Island heath (Specht 1963; Heddle & Specht 1975; Specht & Specht 1989, 1999), Acknowledgements We are indebted to Ms Paulene Riessen, then of the Botany Department, University of Adelaide, for assistance with field work. Dr W. T. Williams of C.S.1.R.0. Division of Computing Research analysed the quadrat data in 1970, References ADAMSON, R.S. & OsBorn, T. G. B. (1924) The ecology of the Eucalyptus forests of the Mount Lofty Ranges (Adelaide District), South Australia. Trans. R. Soc. S. Aust. 48, 87-144. BREWER (née PiDGEON), I, M, (1995) Isonome mapping: graphic analysis of patterns of species distribution. Proc. Linn. Soc. N. S. Wales 115, 259-279. Cuirrorb, H, T. & Specut, R. L. (1979) “The Vegetation of North Stradbroke Island” (with notes on the fauna of mangrove and marine meadow ecosystems by Marion M. Specht). (University of Queensland Press, St Lucia.) & STEPHENSON, W. T. (1975) “An Introduction to Numerical Classification and Taxonomy”. (Academic Press, New York.) Davies, J. G. & Sim, A. H. (1931) The influence of frequency of cutting on the productivity, botanical and chemical composition, and the nutritive value of “natural” pastures in southern Australia. Counc, Sci. Industr. Res. Pamphlet No. 18. Discs, L. (1906) “Die Pflanzenwelt von West-Australien stidlich des Wendenkreises . (Wilhelm Engelmann, Leipzig.) GOODALL, D. W. (1953) Objective methods for the classification of vegetation. 1. The use of positive interspecific correlation, Aust. J. Bot. 1, 39-63. HeppLe, E. M. & Specut, R. L. (1975) Dark Island heath (Ninety-Mile Plain, South Australia). 8. The effect of fertilizers on composition and growth, 1950-1972. Aust. J. Bot. 23, 151-164. 20 Jessop, J. P. & TOELKEN, H. R. (Eds) (1986) ‘Flora of South Australia”. Parts I - IV. Fourth Edition. (S. Aust. Govt. Printing Division, Adelaide.) Martin, H. A. (1961) Sclerophyll communities in the Inglewood District, Mount Lofty Ranges, South Australia. Trans. R. Soc. S. Aust, 85, 91-120. & SpecHT, R. L. (1962) Are mesic communities less drought resistant? A study of the moisture relationships in dry sclerophyll forest at Inglewood, South Australia. Aust. J Bot. 10, 106-118. PipGEon, 1. M. & Asupy, E. (1942) A new quantitative method of analysis of plant communities. Aust. J. Sci. 5, 19-21. Rayson, P. (1957) Dark Island heath (Ninety-Mile Plain, South Australia). 2, The effects of microtopography on climate, soils, and vegetation. Aust. J Bot. 5, 86-102. Specut, R. L. (1957a) Dark Island heath (Ninety-Mile Plain, South Australia). 4. Soil moisture patterns produced by rainfall interception and stem flow. Aust. J. Bot. 5, 137-150. (1957b) Dark Island heath (Ninety-Mile Plain, South Australia). 5. The water relationships in heath vegetation and pastures on the Makin Sand. Aust. J. Bot. §, 151-172. (1963) Dark Island heath (Ninety-Mile Plain, South Australia). 7. The effect of fertilizers on composition and growth. Ausv. J. Bot. 11, 67-94. (1970) Vegetation. Pp. 44-67. Jn: G. W. Leeper (Ed.) “The Australian Environment”. Fourth Edition. (C.S.I.R.O. & Melbourne University Press, Melbourne.) (1972) “The Vegetation of South Australia”. Second Edition, (Govt. Printer, Adelaide.) (1975) A heritage inverted, our endangered. Search 6, 472-477. (1988) Foliar analyses. Pp. 63-80. Jn: R. L. Specht (Ed.) “Mediterranean-Type Ecosystems. A Data Source Book”. (Kluwer Academic Publ., Dordrecht.) (2000) Savanna woodland vegetation in the South East District of South Australia: the influence of evaporative aerodynamics on the foliage structure of the understorey invaded by introduced annuals. Austral Ecology 25; 588-599. (2001) Phosphorus toxicity and pollution: a threat to our Gondwanan heritage. Ecological Management & Restoration 2, 228-230. flora Mile Plain, South Australia). 1. H. A. MARTIN & R. L, SPECHT (2002) Phosphate pollution and soil nitrate: threats to biodiversity in Australia. pp. 53-70. In: A. J. Franks, J. Playford & A. Shapcott (Eds) “Landscape Health of Queensland”. (Royal Society of Queensland Publication, Brisbane.) & CLirForp, H. T. (1991) Plant invasion and soil seed banks. Control by water and nutrients. pp. 191- 204. In: R. H. Groves & F. Di “Biogeography of Mediterranean (Cambridge University Press, Cambridge.) & Perry, R. A. (1948) The plant ecology of part Castri (Eds) Invasions”. of the Mount Lofty Ranges (1). Trans. R. Soc. S. Aust. 72, 91-132. & Rayson, P. (1957) Dark Island heath (Ninety- Definition of the ecosystem. Aust. J. Bot. 5, 52-85. & RUNDEL, P. W. (1990) Sclerophylly and foliar nutrient status of mediterranean-climate plant communities in southern Australia. Aust. J Bot. 38, 459- 474. & SPECHT, A. (1989) Species richness of sclerophyll (heathy) plant communities in Australia. Aust. J. Bot. 37, 337-350. & (1999) “Australian Plant Communities. Dynamics of Structure, Growth and Biodiversity”. (Oxford University Press, Melbourne.) (Paperback Edition 2002). , Rayson, P. & JACKMAN, M. E. (1958) Dark Island heath (Ninety-Mile Plain, South Australia). 6. Pyric succession: changes in composition, coverage, dry weight, and mineral nutrient status. Aust. J. Bot, 6, 59-88. , BROWNELL, P. F. & Hewitt, P. N. (1961) The plant ecology of the Mount Lofty Ranges, South Australia. 2. The distribution of Eucalyptus elaeophora. Trans. R. Soc. S. Aust. 85, 155-176. Spricc, R. C. (1946) Reconnaissance geological survey of portion of the western foothills of the Mount Lofty Ranges. Zrans. R. Soc. S. Aust. 70, 313-347. STEPHENS, C. G. (1946) Pedogenesis following the dissection of lateritic regions in southern Australia. Bull. Counc. Sci. Industr. Res. (Aust.) No. 206. (1971) Laterite and silcrete in Australia. Geoderma 5, 5-52. Woop, J. G. (1937) “The Vegetation of South Australia”. (Govt. Printer, Adelaide.) APPENDIX TABLE |. Sclerophyll (heathy) understorey in the Mt Lofty Ranges, South Australia. Floristic Groups: Group 2— Blewitt Springs; Group 4(8) — Morialta; Group 4(9) — Waterfall Gully, Inglewood and National Inglewood; Group 7(12) - Inglewood and National Park Belair; Group 7(13) - Inglewood and Morialta; Group 9(14) — Waterfall Gully; Group 9(15) - Inglewood and National Park Belair: Life form: MM Mesophanerophyte; M Microphanerophyte; N Nanophanerophyte; Ch Chamaephyte; H Hemicryptophyte Park Belair; Group 6(10) — Inglewood; Group 6(11) (semi-evergreen) Statistics: Number of quadrats (20m x Im) that were recorded containing the species, with percentage of the total number of quadrats in parentheses below Species Life Group 4 Group 6 Group 7 Group 2 Group 9 form 4(8) 4(9) 6(10) 611) 712) 713) 2 9(14) 915) Asteraceae Ixodia achillaeoides N 6l 33 20 11 23 5 - 25 8 (74) (26) (19) (9) (16) (11) (40) (13) Olearia ramulosa N 15S 13 Il 6 5 21 65 7 6 (18) (10) (1) (4) (4) (45) (86) (11) (9) Casuarinaceae Allocasuarina verticillata M 8 6 - 3 2 10 - 2 4 (10) (5) (2) (1) (21) (3) (6) Species A. muelleriana Cyperaceae Life form N Lepidosperma carphoides H L. laterale L. semiteres L. viscidum Dilleniaceae Hibbertia acicularis H. sericea H. stricta H. virgata Epacridaceae Acrotriche fasciculiflora A. serrulata Astroloma conostephioides N A. humifusum Epacris impressa Leucopogon rufus L. virgatus Lissanthe strigosa Fabaceae Daviesia leptophylla D. ulicifolia Dillwynia sericea D. hispida Eutaxia microphylla H H Platylobium obtusangulum N Pultenaea daphnoides P. involucrata P. largiflorens SCLEROPHYLL VEGETATION IN THE MOUNT LOFTY RANGES 21 Group 4 Group 6 Group 7 Group 2 Group 9 4(8) 4(9) 610) 611) 712) 7013) 2 914) 9(15) 21 27 35 75 19 1 43 12 15 (26) Ql) (34) (50) (14) (2) (57) (19) (23) 10 7 - 1 - 3 51 6 1 (12) (6) (1) (6) (67) (10) (2) 2 - - 1 - - 18 - - (2) (1) (24) 44 94 101 133 103 9 - 48 46 (54) (75) (97) (89) (73) (19) (78) (72) - / 1 - - 2 - 4 3 (6) (1) (4) (6) (5) 13 65 102 121 141 - - 17 48 (16) (52) (98) (81) = (100) (27) (75) 72 81 97 135 73 35 - 13 64 (88) (64) (93) (90) (52) (75) (21) (100) - 1 z - - - 76 - 1 (1) (100) (2) 3 2 - - - - 45 - 2 (4) (2) (59) (3) 6 45 1 3 24 4 - 31 14 (7) (36) (1) (2) (17) (9) (50) (22) 46 72 78 100 58 6 16 41 3 (56) (57) (75) (69) (AID (13) Q1) (66) (4) 81 - 1 10 6 13 58 - - (99) (1) (7) (4) (28) (76) 32 43 50 87 61 21 50 17 26 (39) (34) (48) (58) (43) (45) (66) (27) (41) 46 62 1 9 4 - e 62 2 (56) (49) () (6) (3) (100) 1 I 2: - 1 - 21 : 1 (1) (1) (2) (1) (28) (2) 54 43 45 45 18 5 - 26 17 (66) (34) (43) (30) (13) (11) (42) (27) - 6 16 30 23 3 - 1 5 (5) (15) (20) (16) (6) (2) (8) 5 23 3 2 21 - - 13 10 (6) (18) GB) dy) (5) (21) (16) 2 4 3 1 1 - - 3 1 (G3) (3) (3) (1) (1) (5) (2) 2 1 - 1 - ] 46 - 1 (2) (1) (1) () (61) (2) 28 24 54 38 32 1 20 10 14 G4) (19) (52) (25) (23) (2) (26) (16) (22) 2 " - = * 1 1 : F (2) (2) (tr.) 11 63 104 - 45 2 6 28 35 (13) (50) (100) (32) (4) (1) (45) (55) 51 92 32 21 72 5 - 50 42 (62) (73) (31) (14) (51) (1) (81) (66) 8 15 1 1 1 1 - 15 - (10) (12) () (1) (1) (2) (24) 19, 14 55 48 38 2 - 3 11 Q1) qd) (53) (32) 27) (4) (5) (17) 22 H. A. MARTIN & R. L. SPECHT Species Life Group 4 Group 6 Group 7 Group 2 Group 9 form 4(8) 4(9) 610) 611) 712) 713) 2, 9114) 915) Haloragaceae Gonocarpus tetragynus Ch 17 ) 1 1 3 9 3 3 2 (21) (4) (1) (1) (2) (19) (tr.) (5) (3) G. humilis Ch 21 - - 7 - a 26 a r (26) (34) Mimosaceae Acacia continua N 5 - - - 7 = bs a - (6) A, melanoxylon MM - 8 - 1 - 3 - 5 3 (6) (1) (6) (8) (5) A. myrtifolia N 82 126 - - - 1 62 62 63 (100) (100) (2) (82) (100) (98) A. pycnantha M 15 28 21 43 56 27 60 12 16 (18) (22) (20) (29) (40) (58) (79) (19) (25) A. retinodes M 1 2 - 1 1 - - 2 - (1) (2) (1) (1) (3) A. rotundifolia N - 6 16 10 5 1 6 1 5 (5) (15) (7) (4) (2) (1) (2) (8) A. rupicola N - - - - - 2 6 - - (4) (1) A, spinescens N 2 - - 1 - 1 50 ~ - (2) (1) (2) (66) Myrtaceae Calytrix tetragona N 30 11 1 12 1 5 48 8 3 (37) (9) (1) (8) (1) (11) (63) (13) (5) Leptospermum myrsinoides N 70 89 104 150 - - 54 57 32 (85) (71) (100) = (100) (71) (92) (50) L. prickly (syn. L. juniperinum) N 2. 4 5 3 6 2 5 3 1 (2) (3) (5) (2) (4) (4) (1) (5) (2) Pittosporaceae Bursaria spinosa M 3 4 - 2 6 2: - - 4 (4) G3) (1) (4) (4) (6) Cheiranthera alternifolia N 9 - - - 2 I - - - (11) () (2) Poaceae Danthonia sp. & Stipa sp. H 9 15 - 5 16 14 31 6 9 (1) (12) G3) (1) (30) (41) (10) (14) Proteaceae Adenanthos terminalis N 2 2 - 1 - - 26 - 2 (2) (2) (1) (34) (3) Banksia marginata M 22 23 1 5 > 6 5 18 B) (27) (18) (1) (3) (4) (13) (1) (29) (8) B. ornata N 2 2 - 1 - - 10 - 2 (2) (2) (1) (13) (3) Conospermum patens N - - - 1 - - 3 7 hs (1) (tr.) Grevillea lavandulacea N 22 14 3 9 6 2, 18 9 5 (27) (1) (2) (6) (4) (4) (24) (15) (8) Hakea rostrata N 65 58 78 111 29 3 26 42 16 (79) (46) (15) (74) 2) (6) G4) (68) (25) H. rugosa N 1 1 2 5 3 - 26 - 1 (1) () (2) G) (2) G4) (2) H. ulicina N 28 32 3 8 - 3 41 27 5 (34) (25) G) (5) (6) (54) (44) (8) SCLEROPHYLL VEGETATION IN THE MOUNT LOFTY RANGES 23 Species Life Group 4 Group 6 Group 7 Group 2 Group 9 form 4(8) 4(9) 6(10) 6(11) = 7112) 7(13) 2 914) 9(15) Isopogon ceratophyllus — N 25 13 - 5 - 1 21 13 - (30) (10) (3) (2) (28) (21) Persoonia juniperina N 3 7 - 1 - - - 6 l (4) (6) (1) (10) (2) Rhamnaceae Cryptandra tomentosa N 4 - - - - - 11 - - (5) (14) Spyridium parvifolium N 1 7 15 20 8 1 - 3 4 (1) (6) (14) (13) (6) (2) (5) (6) S. spathulatum N 3 1 - - - 3 - - 1 (4) (1) (6) (2) S. vexilliferum N 12 6 - - - 1 - 5 1 (15) (5) (2) (8) (2) Rubiaceae Correa reflexa N 2 - - 1 - - 21 - - (2) (1) (28) Santalaceae Exocarpos cupressiformis M 13 26 2 4 9 5 4 17 a (16) (21) (2) (3) (6) (1) (1) (28) (14) Sapindaceae Dodonaea viscosa N 6 3 - 1 1 8 - 1 2 (7) (2) () d) (17) (2) (3) Scrophulariaceae Prostanthera behriana N - - - - - - 8 - - (11) Stackhousiaceae Stackhousia monogyna = Ch 15 2 - 1 - - - 2 - (18) (2) (1) (3) Thymelaeaceae Pimelea flava subsp. dichotoma N - l - - 1 - 14 - 1 (1) (1) (18) (2) P. glauca Ch - 1 - - - - 22 - 1 (1) (29) (2) P linifolia N 20 30 44 52 35 6 - 7 23 (24) (24) (42) (35) (25) (13) (11) (36) P. octophylla N 2 - - - - - 8 - - (2) (11) Tremandraceae Tetratheca pilosa N 51 6l 46 40 55 2 - 29 32 (62) (48) (44) (27) (39) (4) (47) (50) Xanthorrhoeaceae Lomandra fibrata H 7 44 27 31 34 1 - 25 19 (9) (35) (26) (21) (24) (2) (40) (30) Xanthorrhoea quadrangulata) M 11 2 - 1 - 8 - - 2 (13) (2) (1) (17) (3) X. semiplana N 62 46 57 79 35 5 68 24 22 (76) (37) (55) (53) (25) (21) (89) (39) (34) Transect sites Inglewood 1 33 102 130 97 18 - - 33 x=414 (2) (26) (98) (87) (69) (38) (52) Morialta 60 15 - 6 6 15 - 10 5 x=117 (73) (12) (4) (4) (32) (16) (8) 24 H. A. MARTIN & R. L. SPECHT Species Life form Waterfall Gully x= 141 Belair N.P. x= 89 Blewitt Springs x=91 Total transects LU= 852 Group 4 Group 6 Group 7 Group 2 Group 9 4(8) 4(9) 6110) 6(11) 7012) 713) 2 9114) — 9(15) 11 55 2 7 7 4 - 46 9 (13) (44) Q) (4) (5) (9) (74) (14) 6 20 - 6 31 6 - 6 14 (7) (16) (4) (22) (12) (10) (22) 4 3 - 1 - 4 76 - 3 (5) (2) (1) (9) (100) (4) 82 126 104 150 141 47 76 62 64 (100) (100) (100) (100) (100) (100) (100) (100) = (100) Transactions of the Royal Society of S. Aust. (2005), 129(1), 25-38. MACRO-INVERTEBRATES CAPTURED IN ARTIFICIAL SUBSTRATES IN THE RESTORED WATERVALLEY WETLANDS IN SOUTH AUSTRALIA. by J. M. WuiTe* & T. C. R. WHITE* Summary Wuire, J. M. & Waite, T. C. R. (2005). Macro-invertebrates captured in artificial substrates in the restored Watervalley Wetlands in South Australia. 7rans. R. Soc. S. Aust. 129(1), 25-38, 31 May, 2005. Ninety seven species of macro-invertebrates were collected from six wetlands in the upper south east of South Australia between 1992 and 1995, During the study the salinity of the wetlands fluctuated between fresh or slightly saline and moderately saline; in two instances the salinity temporarily exceeded that of sea water. These changes in salinity had no apparent negative affect on the abundance of any group of macro-invertebrates with the possible exception of Trichoptera in one wetland. This paper provides basic data about these wetlands before the start of a major drainage scheme currently under construction. The information gathered here will improve the active management of these wetlands when the scheme is completed. Key Worps: Macro-invertebrates, wetlands, fluctuating salinity, tolerance to salinity. Introduction The Watervalley Wetlands are a series of alkaline, shallow and fresh or slightly saline to moderately saline, temporary, semi-permanent or permanent wetlands in the upper south east of South Australia (White 1999a). They were hydrologically restored between 1984 and 1991. They are mostly connected either by natural broad, shallow watercourses or by drains. In historic times these wetlands were connected to the ocean via Salt Creek and the Coorong (Fig. 1) but this connection was disrupted by drains further to the south. The Watervalley Wetlands are of national conservation significance (Jaensch & Auricht 1989; Nicholson 1993) and qualify for listing as Wetlands of International Importance under the Ramsar Convention although they have not yet been so listed. The wetlands, at the time of the study, were fed mainly by fresh to very slightly saline water which flows through the system of drains and watercourses from the south-east towards the north-west in all but extremely dry years. The median annual rainfall at Naracoorte, the nearest weather station upstream of the wetlands, is 577 mm. Rainfall decreases sharply towards the north and is 456 mm at Keith, the nearest station downstream (Commonwealth Bureau of Meteorology). As the gradient of the watercourses is of the order of only 1:5000, except at a few isolated points, the rate of flow is extremely slow. The wetlands vary in size from 100 hectares (Jip Jip) to 2700 hectares (Mandina Marshes) and are managed “School of Natural and Built Environments, Mawson Lakes Campus, University of South Australia, Mawson Lakes, SA 5095. * School of Agriculture and Wine, Waite Agricultural Research Institute, University of Adelaide, Glen Osmond, SA 5064. primarily for conservation. White & Brake (1995) described them and outlined their history. The chemistry of their waters, particularly with respect to salinity, was discussed by White & Brake (1995) and White (1999a). Six of the wetlands (Jip Jip, Mandina Marshes, Mandina Lakes, Cortina Lake, Bonneys Camp South and Bonneys Camp North) are discussed in this paper. All of the Watervalley Wetlands are actively managed with the aim of maintaining or enhancing their value as habitat for native wildlife, particularly native and migratory waterbirds. A program of monitoring water chemistry and various biological attributes of the wetlands commenced in April 1992. The purpose was to observe succession in these newly rehabilitated or restored wetlands and to monitor their ongoing “health” in a region that is subject to rising groundwater and increasing dryland salinity (Clark ef a/. 1991). This dual problem is currently being addressed by a drainage scheme to lower the ground water: the Upper South East Dryland Salinity and Flood Management Programme. As a result of this scheme the wetlands are currently (2004) receiving more saline water from the new groundwater drains than they received previously and, overall, salinity has increased and the periodical influx of freshwater has diminished. When the scheme is completed it should allow for the restoration of the natural late winter and spring inflow of fresher water. The importance of invertebrates in the diet of waterbirds is well documented (e.g. Hill et al. 1986; Cooper & Anderson 1996; Cox et al. 1998), This paper discusses the results of monitoring macro- invertebrates between 1992 and 1995. The results provide baseline data for any future monitoring of the effects of the drainage scheme and its impact, J.M. WHITE & T. C. R. WHITE c i ah Us ‘ .» Pretty a8 _Lagoon — Seaors he Johnnys: SOUTH AUSTR = “2 t ee me \ Jip Jip i Naracoorte Skm East —> | 440° 00 Fig. 1. The upper south east of South Australia showing the Watervalley Wetlands. either positive or negative, on the native animals and plants of the wetlands. Methods Invertebrates were collected at intervals of four to six weeks between August 1992 and the late spring or early summer of 1994. Further collections were made in 1995. One invertebrate sampling site was established at each of the six wetlands between April and July 1992. Five artificial substrates were used at each site. They were suspended from star pickets placed five metres apart along a transect at an angle of 90° to the shore commencing where the water was 60 cm deep at the time the transect was established. This gave a range of potential depths of up to one metre along a transect. At Bonneys Camp South this configuration was not feasible as the lagoon shelves very steeply to a depth of three to four metres. The transect at this site ran parallel to the shore at a depth of 60 cms when first established. The artificial substrates were similar to those used by Bennison ef al. (1989) and made of plastic mesh Gutterguard™ baskets containing one and a half onion bags made of red, loosely woven nylon mesh with a total surface area of 4256 cm’. The baskets were each 17 cm high with a 16 cm diameter circular MACRO-INVERTEBRATES IN RESTORED WETLANDS 27 base. Two small plastic vials filled with sand were placed in the bottom of each basket to act as ballast. The substrates were suspended from the star pickets by nylon rope so that they rested on the bottom of the wetland. They were left undisturbed for four to six weeks. The macro-invertebrates were then collected by placing a dip net under each substrate with as little disturbance as possible and bringing it to the surface. The contents of the substrate were then washed into the dip net and concentrated in a vial attached to the base of the net. Small fish and yabbies (Cherax destructor) were released. Samples were preserved in 70% ethanol on site. At the same time as the invertebrates were collected the depth of water was recorded and later converted to the Water Level Index of Tamasier & Grillas (1994). Conductivity, turbidity and major chemicals in the water were analysed (reported in White & Brake 1995). Conductivity is an indirect measure of salinity and, in the text where conductivity measured in mS/cm is converted to salinity in g/L, the conversion is g/L = 0.68 x conductivity in mS/cm (Hart et al. 1991). In this paper the term “saline” is used to describe water with conductivity greater than 4.4 mS/cm (3 g/L). Water with conductivity of 4.4 mS/cm to 14 mS/cm is described as being slightly saline and with conductivity 14 mS/cm to 50 mS/cm as moderately saline. From August 1993, the dominant species and total cover of submerged plants, as a percentage, were recorded in a 50cm quadrat surrounding each point. Percentage cover of vegetation and depth of water was then reduced to a mean for each transect. In the laboratory each invertebrate sample was tipped into a glass Petri dish, 20 cm in diameter, resting on a sheet of paper. Lines had been ruled on the paper to partition the area of the bottom of the dish into eight equal “pie” segments. The sample was stirred and allowed to settle as evenly as possible across the whole dish. Counting was done under a x 4 dissecting glass as follows: * All animals in the dish: large animals such as adult coleopterans, late instar odonates, spiders, and the larval cases of trichopterans (only the cases of trichopterans were counted as some, but not all, larvae left their cases when placed in alcohol). ¢ All animals in the right hand half of the dish: earlier instars of the above when they were numerous, and small snails. * All animals in the top right hand quarter of the dish: small and numerous early instars. * All animals in the top right hand eighth of the dish: very small and very numerous amphipods, larval chironomids and ostracods. For purposes of comparison and calculation all counts were standardised at one-cighth-equivalents of the total sample. Counted animals were placed in a separate vial and the residue of the sample returned to the vial in which it had originally been placed in the field. Identified specimens from each site are lodged with the Museum of South Australia. Statistical calculations were made using Pop Tools™ version 2.5.8 (Hood, 2003) and Microsoft Excel™, Results Bonneys Camp North and Bonneys Camp South held water throughout the sampling period. The sites at Cortina Lakes and Mandina Lakes dried briefly in the autumn of 1994, although other areas of Cortina Lakes continued to be inundated. Except for the spring of 1994, all but several deeper waterholes downstream of the sampling area at Mandina Marshes, were dry from January 1994 to July 1995. These periods of drying were due to low rainfall. Jip Jip was drained for management purposes in the early winter of 1993 and it did not begin to fill again until September 1995. The water level at all sites fluctuated seasonally. The number of times that each site could be sampled was dictated by water levels. Bonneys Camp North was sampled 21 times; Bonneys Camp South 22; Cortina Lakes 23; Jip Jip seven; Mandina Lakes 16 and Mandina Marshes 21. Full details of the water chemistry at each site during the study were published in White & Brake (1995): a summary of the range and median value of conductivity, nitrates and orthophosphate is given in Table 1. Water levels were published in White (1999a). Conductivity at all sites varied during the TABLE |. Range (and median value) of certain chemical attributes from June 1992 to December 1995. Bonneys Camp North Bonneys Camp South Cortina Lakes Lake Mandina Mandina Marshes Jip Jip Conductivity mS/em —4,30-16.30 (7.61) 4.2-9.59 (6.45) Nitrate mg/L 0.2-2.0 (0.8) 0, 1-2.6 (0.7) Orthophosphate mg/L 3.32-69.30 (8.91) <0,1-2.6 (0.7) 0.01 - 0.88 (0.04) <0.10 - 0.40 (0.05) 0.01 - 0.83 (0.04) 0.01 - 0.76 (0.15) 1.30-5.58 (2,26) —2.37-116.00 (14.63) 1.60-16.80(5.91) <0.1-5.3 (0.4) <0.1 - 3.7 (0.6) <0.1 - 0.6 (0,2) <0.01 - 0.94 (0.60) <0.01 - 0.65 (0.05) 28 study (Table 1; Appendix) with Cortina Lakes and Mandina Lakes each having a spike of extremely high conductivity in March 1994 as water levels receded. The water at all sites, except Jip Jip, was slightly saline for most of the time (conductivity < 14 mS/cm or salinity < 10 g/L). Jip Jip was fresh to very slightly saline. The spikes of high salinity in both Cortina and Mandina Lakes, which occurred in March 1994, exceeded the salinity of sea water (sea water = 35g/L or approximately 50 mS/cm). Bonneys Camp North, Cortina Lakes and Mandina Lakes were well vegetated but Mandina Marshes and Bonneys Camp South were sparsely vegetated in the earlier part of the study. Jip Jip was sparsely vegetated throughout. The hydrology of the two Bonneys Camp wetlands was restored in 1991 and, in the first year of the study, little aquatic vegetation had developed on the bare sandy bottom of Bonneys Camp South but the sedgeland which covers the bed of Bonneys Camp North was apparently unaffected by inundation. This sedgeland was relatively dense (mean cover of all species combined = 50%) and dominated by Baumea juncea and Hypolaena fastigata. In contrast Bonneys Camp South was devoid of vegetation until December 1993 when a submerged community dominated by Lepilaena spp and charophytes (mean combined cover = 25%) developed. From the beginning of the study until November 1994 the sampling area at Cortina Lakes TABLE 2. List of species of invertebrates recorded at each site. J.M. WHITE & T. C. R. WHITE was dominated by emergent Baumea arthrophylla and submerged charophytes with a mean combined cover of 65%. Between that time and the next sampling in November 1995 the water level receded below the sampling site for over six months and the sedges and charophytes were replaced by Ruppia megacarpa when the water level rose again. The sampling site at Mandina Lakes was dominated by a mix of Ruppia megacarpa and Lepilaena spp with a mean cover of 65%. The sampling site at Mandina Marshes had very sparse vegetation until it reflooded with fresh water after the drought and the site was colonised by a mix of Potomogeton pectinatus and P. crispus. Ninety seven species of macro-invertebrates were collected during the study (Table 2) and of these, 25 species were found at only one site. Fifteen species were found at two sites, 15 at three sites, seven at four sites, 11 at five sites and 20 were common to all sites. Four species could not be assigned to a specific site because they had been identified from preliminary general samples and were not found again. Cortina Lakes (59 species), Mandina Lakes (56 species) and Mandina Marshes (54 species) were the richest in species, Bonneys Camp North (51 species) and South (49 species) were slightly less so, and Jip Jip (37) had fewest. The overall mean number of each of the major taxonomic groups of invertebrates at each site is Taxa BCN BCS CL JJ ML MM Cnidaria (hydra) Clavidae Cordylophora sp Hydra indet Total species of 2 Cnidaria at each site Mollusca Gastropoda( snails) Planorbidae Glyptophysa tenuilirata Hydrobiidae indet Helicarionidae Echonitor nr cyrtochilus Total species of l 1 Gastropoda at each site Hirudinea (leeches) Glossiphonidae Glossiphonia sp (site not specified) Hirudinidae Bassioanobdella sp (site not specified) Total number of species = 2 Crustacea Ostracoda: seed shrimps Cyprinidae (indet) eNO ONS MACRO-INVERTEBRATES IN RESTORED WETLANDS TABLE 2. List of species of invertebrates recorded at each site cont. Taxa BCN BCS CL JJ Mytilocypris sp v v | Notodromatidae (indet) v ov Total species of Ostracoda 2 2 Amphipoda (scuds) Ceinidae Austrochiltonia australis v v Austrochiltonia subtenuis Crangonyctidae (indet) site not specified Eusiridae (indet) v Total n. species 1 1 2 3 of Amphipoda Decapoda (freshwater crayfish) Parastacidae Cherax destructor v v v v Total species of Decapoda 1 1 1 1 Arachnida: Acarina (mites) Arrenuridae Arrenurus sp | v vo v v Arrenurus sp 2 Limnocharidae Rhyncholimnochares sp v v v v Eylaidae Eylais sp v Hydrachnidae Hydrachna sp v Pionidae Piona sp v Total species Acarina 2, 2 5 Araneae( spiders) Lycosidae Trochosa tristicula phegea v Trichosa sp v Tetragnathidae Tetragnatha sp v Linyphiidae Erigona prominens v v Total species Araneae l I 3 Insecta Ephemeroptera (mayflies) Baetidae Cloeon sp Total species Ephemeroptera Odonata: Zygoptera (damselflies) Coenagrionidae Coenagrionidae juv. v Austroagrion coeruleum Jv Jv Austroagrion watsoni Ischnura heterosticta v Ischnura sp Xanthagrion erythroneurum v Lestidae Austrolestes annulosus v Austrolestes analis v Austrolestes juv v SN \ X< N\ X So ONESCNEN — 30 J. M. WHITE & T. C. R. WHITE TABLE 2. List of species of invertebrates recorded at each site cont. Taxa BCN BCS CL JJ ML MM Megapodagrionidae Argiolestes icteromelas (site not specified) Total species Zygoptera Odonata: Anistoptera (dragonflies) Aeschnidae Aeschna brevistyla Hemianax papuensis Hemicorduliidae Hemicordulia tau Libellulidae Austrothemis nigrescens Diplacodes bipunctata Diplacodes haematodes Total species Anisoptera Total species Odonata Hemiptera (bugs) Corixidae Agraptocorixa eurynome Agraptocorixa sp Diaprepocoris barycephala Micronecta sp Sigara australis Sigara sp Naucoridae Naucoris congrex Notonectidae Anisops sp Anisops thienemanni Total species Hemiptera Coleoptera ( beetles) Dytiscidae Platynectes decempunctatus Allodessus bistrigatus Antiporus femoralis Antiporus gilberti Hyphydrus elegans Megaporus gardeneri Megaporus hamatus Megaporus larva Necterosoma penicillatum Sternopriscus tasmanicus Rhantus suturalis Lancetes lanceolatus Hydrophilidae Limnoxenus zealandicus Enochrus eyrensis Paracymus pygmaeus Stenophilus marginicollis Berosus majusculus Berosus discolor Berosus veronica Berosus larva Halplidae Haliplus sp Total species Coleoptera 5 Ss¥vNNN VNNNN NNN \N\N SONGN Sen SN NN \N NNN SNe SONTREN ON NN NNN ANS MACRO-INVERTEBRATES IN RESTORED WETLANDS 31 TABLE 2. List of species of invertebrates recorded at each site cont. Taxa BCN BCS CL JJ ML MM Diptera (flies) Ceratopogonidae v Nilobezzia sp v v Chironomidae Ablabesmyia sp Chironomus sp v v SoS NS N\ Cladotanytarsus sp Cladopelma sp Cricotopus sp Dicrotendipes sp Larsia sp Sani SON SENSORS Parachironomus sp Paramerina sp SO NENEN NANA SS SNES Polypedilum nubifer Polypedilum sp Procladius sp Tanytarsus spp Chironominae indet Ephydridae indet Stratiomyidae indet Tabanidae indet Total species Diptera Trichoptera (caddis flies) Ecnomidae Ecnomus cygnitus v v v Ecnomus sp v v Hygrocloatidae indet Leptoceridae Notalina salina Notalina spira Oecetis spp Symphitoneuria opposita Triplectides australis v v v v Total species Trichoptera 5 6 5 3 Lepidoptera (moths) Pyralidae Nymphulinae indet v Nymphalinae sp 40! Total species Lepidoptera 1 Total number of species 51 49 59 37 54 BCN = Bonneys Camp North; BSC = Bonneys Camp South; CL = Cortina Lakes; JJ = Jip Jip; ML = Mandina Lakes; MM = Mandina Marshes. |= sp 40 in SA Water Reference Collection. SSN a) QVNNNNNNRNNNRS By, ANY \o \QN\N SONA = SNS N oo \N SSN Bm WON ONSEN Se NoSNG SNe BSS SREN SN =a NN ONE \ AN IRENE NyQAN TABLE 3. Overall mean number (and standard deviation) of major invertebrate groups at all sites. Taxa BCN BCS CL JJ ML MM Amphipoda 50.19 (81.2) 54.9 (36.1) 113.4 (97.6) 410.4 (355.4) — 188.3 (303.9) 583.8 (555.5) Ostracoda 36.57 (32.5) 29.9 (41.4) 12.0 (20.1) 4.75 (5.4) 45.9 (31.9) 2.5 (3.1) Trichoptera 7.24 (7.4) 7.9 (6.4) L121) 28.0 (24.0) 9.3 (5.0) 6.8 (6.0) Zygoptera 3.04 (3.8) 4.8 (3.4) 2.8 (2.1) 7.7 (5.95) 3.4 (4.5) 7.6 (6.7) Anisoptera 0.2 (0.27) 0.2 (0.3) 0.4 (0.3) 0.3 (0.3) 0.3 (0.1) 1.3 (3.0) Diptera 36.57 (21.45) 27.4 (23.2) 10.1 (10.0) 12.4 (4.2) 24.9 (48.4) 22.6 (17.8) Coleoptera 3.22 (1.8) 1.7 (1.8) 0.2 (0.2) 1.7 (1.4) 4.6 (4.8) 0.3 (0.3) Hemiptera 0.06 (0.08) 0.1 (0.1) - 1.0 (1.6) - - Acarina 9.44 (10.95) 2.4 (7.8) 0.4 (0.4) 0.55 (0.7) 0.9 (0.7) 6.7 (11.2) Gastropoda - - 0.6 (1.6) 10.9 (9.8) 0.2 (0.3) 10.9 (3.7) 32 J, M. WHITE & T. C. R. WHITE given in Table 3. Histograms showing the mean number for each sampling date, together with conductivity at the time, can be found in the appendix. Cortina Lakes was the only site where enough samples were collected both before and after the spike of extremely high conductivity (69.3 mS/cm) to get a statistically valid indication of the effects on macro-invertebrates of such relatively extreme conditions. Student’s ¢ test for the difference between these two sets of samples showed a significant increase in abundance of Ostracoda, Odonata and Acarina and a significant decrease in Trichoptera. There was no significant difference in the abundance of Amphipoda, Diptera, or Coleoptera (Table 4). Discussion All of the wetlands except Jip Jip are part of the Bakers Range Watercourse and are connected in times of high flow (Fig. 1). When Jip Jip (in the Marcollat Watercourse, several kilometres to the east of the Bakers Range Watercourse) overflows, water from it passes through a drain into the Bakers Range Watercourse and thence to the other wetlands studied (White & Brake 1995). This provides for a degree of migration by flightless macro-invertebrates throughout the system. The wetlands of the Upper South East are dependent on fresh water flowing north west from the upper reaches of the catchment which extends TABLE 4. Results of Student’ ¢ test comparing mean numbers of major invertebrate groups after spike in salinity in March 1994 at Cortina Lakes. A93-M94 Ap94-N94 Amphipoda Mean=104.6 Mean=130.6 Var=7104.8 Var=3071.2 P=0.207 Ostracoda Mean=14.9 Mean=73.7 Var=373.4 Var= 1981.6 P =0.010 Trichoptera Mean=0.8 Mean=0.2 Var=0.6 Var=0,02 P=0,.006 Odonata Mean=6.0 ean=13,2 Var=44.6 Var=24.7 P=0.008 Diptera Mean=16.0 Mean=8.1 Var=209.4 Var=67.2 P=0.063 Coleoptera Mean=0.5 Mean=1.1 Var=0.57 Var=1.31 P=0,147 Acarina Mean=0.7 Mean=1 1.2 Var=0.61 Var=27.0 P=0.002 into western Victoria. Local run-off provides some inflow in winter although the amount is probably much less important. The relationship between the wetlands and groundwater is unknown. The very slight gradient to the north west means that the flow of water is slow and evaporation rates are high, accounting for the higher salinity in the more northerly wetlands. Mandina Lakes and Cortina Lakes are fed only by water entering through controlled inlets from Mandina Marshes. Neither of these wetlands has an outlet, therefore as water levels recede, their salinities are relatively higher than the other wetlands. The natural flow throughout the system has been disturbed by drainage but the current scheme provides for both surface water and drainage water to be managed to benefit the wetlands (NRCSA 1993). This should allow the hydrological regime of the wetlands before drainage to be re- established, thus potentially conserving the habitat for species currently found in the wetlands. Mandina Marshes and Cortina Lakes are the least disturbed of all the sites. Even after the overland flow to the more northerly wetlands was reduced by 50% in 1964 because of drainage (EWS 1991), the deeper pools of both these wetlands retained water permanently (T. K. Brinkworth pers.com.). Both the riparian and the aquatic vegetation of these wetlands appear to support this notion (J. M. W. unpublished data) and both areas have surrounding terrestrial vegetation that has not been cleared. This relative stability is consistent with these two bodies of water having a greater number of species than most of the others, although it does not seem to explain the high number of species on Mandina Lakes. The whole of this wetland was grazed pasture from the 1960s until its restoration in the late 1980s (T. K. Brinkworth pers. com.). Jip Jip was only sampled seven times because it was drained in the autumn of 1993 and it did not receive any water until the spring of 1995. This necessarily reduced sampling effort could account for the paucity of species there; although the fact that its hydrology was disturbed both prior to and during the study could also be a factor. There was no apparent negative relationship between salinity (measured as conductivity) and the abundance of any groups of macro-invertebrates found at any of the study sites except, possibly, for Trichoptera at Cortina Lakes. It is frequently stated that increases in salinity lead to a decline in the number of species (e. g. Hart ef a/. 1991; Nielsen et al. 2003) and saline wetlands are therefore sometimes regarded as being “less important” than freshwater systems. However, this decrease in species occurs at levels of salinity generally below | g/L, below any yet measured in the Watervalley Wetlands cited here except rarely and briefly in Jip Jip. Furthermore, while there may be fewer species MACRO-INVERTEBRATES IN RESTORED WETLANDS 33 in more saline water, they can be much more abundant (Kingsford & Porter 1994). This could make saline wetlands much more important as a source of food for waterbirds. In this study, however, the greatest number of species was found in the two most saline lakes: Cortina (59) and Mandina (56). Mandina Marshes, with 54 species also had a relatively rich fauna. The freshest wetland, Jip Jip, had fewest species but because it was sampled less frequently than the others it would be unwise to make too much of this observation. The methods of collection used in this study trap only those species which will colonise substrates and micro-crustaceans were not counted, therefore our species lists do not represent the true number of species present in any of the wetlands. At Cortina Lakes there was a. statistically significant increase in the abundance of Ostracoda, Odonata, and Acarina after the spike in salinity (measured as conductivity) in autumn 1994 whereas the abundance of Trichoptera decreased after that event. There was a twofold increase in the abundance of Coleoptera but the variance was such that it was not statistically significant. However, caution is advised in attributing these changes to the sharp rise in salinity. The increase in abundance in both Odonata and Coleoptera began before the event and there was a very long lag time for Ostracoda and Acarina. It is possible, however, that the decline in the abundance of Trichoptera was due to higher salinity. This has been reported elsewhere (Hart ef al. 1991), but Kefford et al. (2003) found that Notalina spira and species of the genus Trip/ectides other than those found in Cortina were tolerant of conductivity greater than 25 mS/cm. So, even for this group, it is not certain that the observed changes can be attributed to changes in salinity. This study was devised to find out whether there were discernable trends in the abundance of macro- invertebrates following restoration, not to measure the effects of increases in salinity on individual species or orders. Further studies are needed to ascertain the tolerance of the species present to the sometimes quite large seasonal increases in salinity because the hydrology, in regards to quantity, quality and timing of flows, of these wetlands will be actively managed through the Upper South East Dryland Salinity and Flood Management Scheme. The tolerance of some of the species present is known (see Hart ef al. 1991; Kefford et al. 2003) and it seems, not unexpectedly, that these macro- invertebrates are able to cope with such widely fluctuating salinities. However, it is also known that the tolerance to increases in salinity varies at different stages of the life cycle (Hart et a/. 1991), Therefore, it does not follow that all species present could tolerate constant salinities around the median levels observed during the study. Nielsen ef al. (2003) found that hatching of zooplankton and germination of plants was affected by increases in the salinity of fresh water. Similar effects could be present in these systems whose salinity can fluctuate widely within a single year and it may be that the late winter-spring inflow of less saline water is essential for reproduction of some species. This, too, requires further study. Our results show that the Watervalley Wetlands experienced large seasonal fluctuations in salinity and supported a diverse macro-invertebrate fauna before the current drainage scheme commenced. It would be prudent, therefore, to manage water regimes in the Watervalley Wetlands so that these seasonal fluctuations in salinity are maintained unless it can be demonstrated that these slightly to moderately saline systems can maintain their mix of species of both plants and animals without seasonal periods of lower salinity for successful recruitment. Acknowledgements Travel and collection of field data for this paper were supported by grants from the University of South Australia and the Wildlife Conservation Fund of South Australia. We would like to thank the following people for their assistance in this project: Lynn Brake (University of South Australia) who helped devise and set up the monitoring program and assisted with initial field work; many students of Conservation and Park Management (University of South Australia) who assisted with field work; Tracy Corbin, Chris Madden and Paul McEvoy (Australian Water Quality Centre), Chris Watts and John Bradbury (SA Museum), and John Hawking (Murray-Darling Freshwater Research Centre, Albury) who identified specimens and Dr Roger Clay (University of South Australia) for advice on statistics. David Britten (University of South Australia) prepared the base map used for this paper. We are also grateful to Pat and Tom Brinkworth of Ninga Ninga Station, Kingston SE for their hospitality during field trips and for allowing us unfettered access through their property to the wetlands. 34 J. M. WHITE & T. C. R. WHITE References BENNISON, G. L., HILLMAN T. J., & SuTER P. J., (1989). Macro-invertebrates of the River Murray. Murray- Darling Basin Commission Water Quality Report no. 3. CLarRK, R., Kotwicki, V. & TOMLINSON, G. (1991). Hydrological study of Bakers Range and Marcollat Watercourses. Engineering and Water Supply Department South Australia. Cooper, C. B. & ANDERSON, S. H. (1996). Significance of invertebrate abundance to dabbling duck brood use of created wetlands. Wetlands. 16 (4):557-563. Cox, R. R., Roy, C. R., Eutiss, N. H. Jr. & JOHNSON, D. H. (1998). Mallard duckling growth and survival in relation to aquatic invertebrates. J. Wildl. Manage. 62 (1): 124 -133. Hart, B. T., BAILEY, P., Eowarps, R., HorTLE, K., JAMES, K., McMAnHon, A., MEREDITH, C. & SWADLING, K. (1991). A review of the salt sensitivity of the Australian freshwater biota. Hydrobiologia 210: 105-144. Hitt, D., WriGHT, R. & Street, M. (1986). Survival of Mallard ducklings Anas platyrhynchus and competition with fish for invertebrates on a flooded gravel quarry in England. /bis. 129: 169-167. Hoop, G. M. (2003). PopTools version 2.5.8. Available on the internet. URL http://www.cse.csiro.au/poptools. JAENSCH, R. & AuRICHT, C. (1989). Waterbirds in the South East of South Australia. Adelaide. South Australian Ornithological Association. Kerrorb, B. J., PApas, P. J. & NuGEGoDA, D. (2003). Relative salinity tolerance of macro-invertebrates from the Barwon River, Victoria, Australia. Mar. Freshwat. Res. 54: 755-765. KINGSFORD, R. T. & Porter, J. L. (1994). Waterbirds on an adjacent freshwater lake and salt lake in arid Australia. Biological Conservation. 69: 219-228. NICHOLSON, C. (1993). Conservation parks and wetlands in the upper south east — surface water recommendations and an assessment of impact associated with rising groundwater. Department of Environment and Land Management South Australia. NIELSEN D. I., Brock, M. A., CROSSLE, K., HARRIS K., HEALEY, M. & JARosINskI, I. (2003). The effects of salinity on aquatic plant germination and zooplankton hatching from two wetland sediments. Freshwat. Biol. 48: 2214-2223. NRCSA (Natural Resources Council of South Australia). (1993). Upper South East Dryland Salinity and Flood Management Plan. Department of Environment and Natural Resources SA. TaMaASIER, A. & GRILLAS, P. (1994). A review of habitat changes in the Camargue: an assessment of the effects of loss of biological diversity on the wintering waterfowl community. Biol. Conserv. 70: 39-47. Wuirte, J. M. (1999a). Seasonal variation in salinity in the Watervalley Wetlands in the south east of South Australia. Trans R. Soc. S. Aust. 123 (2), 77-80. (1999b). Watervalley Wetlands and Heritage Agreement Areas Management Plan. Printed by the University of South Australia for Wetlands and Wildlife, lst Floor, 187 Wakefield Street, Adelaide, South Australia 5000. & BRAKE, L. A. (1995). Description, history, and water chemistry of the recently restored Watervalley Wetlands in the southeast of South Australia. Wetlands 15 (3), 247-257. MACRO-INVERTEBRATES IN RESTORED WETLANDS APPENDIX. Mean number of macro-invertebrates and conductivity (mS/cm) at each wetland. Amphipoda Bonneys Camp North | Bonneys Camp South 400 —_ 20 | 1000 — 20 350 beeeuecee (mE A mphipoda | @ 300 |, —e—Conductivity 415 | yp 800 [=e Conductivity] ° 1 15 = 3 250 al 2% 600 € § 2 200 105 | & 8 105 | os a 2 400 s i age | S & 150 “ =o ey go £ < 100 © 45 © 300 We 5 | he | | | | | | Aaalla a. ato o an, Me, afl ans, ie 5 POZO GL MSSSEVOULNE>SSEP= EE Z | BOZO TEES EPOOO TEPESEPZ zPZ @ oO B 2 | oO © + Date N Date a Cortina Lakes Jip Jip | 400 — zu ____+ 80 350 + Saee 70 1200. {mmm A iphipoda] 7° | = 300+ —e—Conductwity] gq | 1000 [—e—Conductvty| 15 z3 | a os Z 8 250 50e| 2 800 foe & = 200 +408 | es S| 2B 150 130 © ae S| ees 100 20 < 1 5 50 I] I 10 200 | al ° powoenessy TEEPE Aer Ee 0 MT | ee er ee Cae PEs Peale an ¢ a = BOZO NESS POO SP ESSEZ FF ue Date oi 8 Bate 8 { Mandina Marshes | Mandina Lakes 2000 20 1000 80 ] [aero ] | | {ummm Amp hipoda ] | g 1800 + —#Conduernty 15 |g 800 |e conductivity +60 | | 28 |, §) 2 8 600 = io 5 | 8 & 1000 [106] g 38 + 405 se k ” |. £| | 2 2 400 £ < 500, 5 2 20 li II. 200 « om MiMer, wee, Sei eo o WIT eee, Lo Fa POI a 4 & 1g woe OOo Se BE BOZUST ERE EM OURNE oO i] Date “ | Ss al Date Ostracoda Bonneys Camp South | Bonneys Camp North 200 ——— 20 | | 140 + eT oe + 20 mmm Ostracoda ea mmm Ostracoda | | 150 —#— Conductivity 3 15 ie y [Re conductivity | <8 a0: 22 60 =o | S 404 20 | | () : AL. - o | | PUZOL TEES POOUL TEP ESEEZ zeez BOZOCNESES PNOULMEPESLPHZ ZY Z oO 8 g ive} oO oe oonx< Cc o | Date OY f s °° Tite & Cortina Lakes Jip Jip | 160 80 20 = ws 23) | 140 | SMM Ostracoda 70 | | {mmm Ostracoda | 5 g 120 {—e—conauetiity, @ || og 15 + |—e—conduetiity!| 45 | 3 iS 3 €] § 8 tag! | § 8 104 +109 2a 130E}) se | 9 ili | 20 1) 7 ls 10 () ol, ae, See ell lo Zz | BOZO EES 2000 MP ESEEZ =zPze | | a 8 & = | s ‘Date a Mandina Lakes Mandina Marshes 120 7 = __+ 80 a. | mmm Ostracoda | 79 | noe Mmmm Ostracoda | nm A 100 [Conductivity | 6g j 8 [=e—conducivty) 45 z 80 ae ae 225 a | aren <8 6 §| 8 Sf, 7 106 | = 8 4 a = a ” E a I ELL Ie o 1a Masteeto | 0 Me Pa Jer ~_ POZOS NEES POOOS io WS PESSEZ 8 © S POZOSNEESPUOOL TE DESC EZ = DP © 8 s fa S Bate a 35 36 J.M. WHITE & T. C. R. WHITE Trichoptera va Pa — oy a Bonneys Camp North Bonneys Camp South _ 20 40 20 28 es Tiichopters T ‘mmm Trichoptera ] 20 {w= conductivity | 15 ge 30 {Tem Conduetivity ‘s +18 o z A §| ls | | 22 Ei |22 | | BE = i ~ Cortina Lakes | Jip Jip 6 80 a BD [mmm Trichoptera [ | Trchoptera || s [—e=Conauctvty |. g @ 60 [—e—Conductivity|) 45 Ze 4] a Zo £ | $2 g 2 0+ |B | =2 ol | ss | E | ec “| | F 204 5 { { 0 onal of aan flo BOZO SESE (OMG SETS Sscerz = Ps dy oe 8 g % | 8 & “A | Bonneys Camp South i | fpraiseslanr’ rea mes —#—Conductivity | ie +45 za) 5 & 3 4} } 10 2 Sree | OND | 5 o Lam f MB im, to EVZOK NESS POO MEPESSE= = = Zz oO o BS © 8 Cate a OO — m5 Jip Jip ey (mmm Coleoptera [70 est —e—Conductivity) 45, i 25 all E a 322 € 1o} 1 5 ) | oan Bere reeer 0 2OZ0G NEES PV OULEP ESS E= zez e 8 g& S | a Date a 4 + 20 | ptera a3 Zo eos] $3? =o ” ou 0 ve cord o 38 Gastropoda Acarina Mean N Gastropoda Mean N Mean N Acarina Mean N Acarina Mean N Acarina J. M. WHITE & T. C. R. WHITE Cortina Lakes Jip Jip 6 earned ipa sacsnoie] 2° ; | —e—Conduetivity 60 s 7 | ~e—conductivity 15 z ¢ | 50 c a it § 3 og|| gs 10 @ | > 30 £ } | =8 40 {| o£ | 20 5 1 \ | : i) to bb htt a te oO 0 9 adeeaeeen eee tee is | Meade = 2 SOZOCNZES PHOOC MES ESE DZ BO BSE gS esee BE = e o@ axe o 9 <98e oO oO ‘ © o Date 5 = frontier a es actly <2 —— —_ — ~ Mandina Lakes. Hint Mandina Marshes | ey eee 8 25g a qm Gastropoda 60 \} e —e—Conductivity| teoe} | 2 Qa 0 € + 408 Bo 15 10 3 t SE + 30£ = 8 10 ” E + 20 ce | +10 ttt tt tt 0 0 ici Racal te | _TIZEE POOU TEPES ED > POO-GE?ESSEZ 3 Pz Boze Ese STs SSEPZ = Bz | | Date o " Date i. | Bonneys Camp North Bonneys Camp South 20 0 16 (mM Acarina | GE Acarina ] 44 —e—Conductivity| 445 | es —e—Conductivity 42 g| | i 10 7B] Ss 8 9 ~ | & « 6 £& o 5 o 4 | J Pe tees | al tae at+to | eee, ne min +0 | POZOS TEES POOUSTEPESSPZ FP Z BOZO MEESE COU ME? ESEE= zP2 ° 3 & © © 8 & S Cate a N Cate i ‘Cortina Lakes Jip Jip 80 | 4 ] 20 ‘i B 5 |—e-contctvty 18 405| | z 2 7 105 so E]) § . 20 ; 31 10 i ] Lo obey FOZOS NESS PVOOL NSP ESS EA = i BOZOETEESEOOO NE>SS SSP= 5 2 z 8 val Date & ' S Bate a Mandina Lakes “Mandina Marshes 4 Ym A carina Ae ummm Acarina oy 3, |e _ Conductivity & 30 —o— Conductivity] 4.5 ae 2 > 20 103 ei6 Pood € 1 & 10 5 2ES, I o 0 = aon Pat | hear! | cea rer aperon rarer 1c Sr poz Cig MEE SBOO0S MEP ESEEZ ze: POZOSNESS POCO MZ PSSSEZ = PZ 8 8 8 8 od kal & Date Date Transactions of the Royal Society of S. Aust. (2005), 129(1), 39-42. THE KIMBA METEORITE: AN (H4) CHONDRITE FROM SOUTH AUSTRALIA by G. Horr!, A. PRING!? & M. ZBIK!3" Summary Horr, G., PRING, A. & ZBIK, M. (2005). The Kimba meteorite: An (H4) chondrite from South Australia. Trans. R. Soc. S. Aust. 129(1), 39-42, 31 May, 2005. The Kimba meteorite is a single stone of 1.492 kg found between 5 and 10 km south of Kimba, South Australia (33° 13'S 136° 25’ E). It consists of olivine (Fay). n = 15), low-Ca pyroxene (Fsi67Woi2 n = 10), high-Ca pyroxene and troilite. All of the iron-nickel metal and most of the troilite appear to have oxidized to a mixture of iron and iron nickel oxides and hydroxides. Based on its texture and mineral chemistry, the Kimba meteorite is classified as an H4 chondrite of shock stage S1 and weathering grade W4. Key Worpbs: Kimba, meteorite, chondrite. Introduction A single mass of the Kimba meteorite was found by Mr Lyall Cliff of Kimba more than 20 years ago, on one of his two properties (sections 43 and 21) south of Kimba, South Australia (33° 13’ S 136° 25’ E) (Fig 1). The meteorite lay unrecognized on the windowsill in a shed on Mr Cliff's property until November 1995, when Byron Smith, a member of a local farming family, noticed the stone and suggested that it was a meteorite. Mr L. M. Smith brought the stone to the South Australian Museum, where the identification was confirmed. Unfortunately, Mr Cliff was unable to remember exactly when he picked up the stone, or on which of his two properties it was originally found: they are both south of the township of Kimba and about 5 km apart. Under legislation enacted by the Government of South Australia, all meteorites found in the state are the property of the Crown. The Board of the South Australian Museum decided to reward both Mr Cliff and Byron Smith for their part in finding this meteorite. In accordance with the guidelines on meteorite nomenclature, the meteorite has been named Kimba after the geographical locality closest to its site of discovery. The meteorite’s name and petrological classification have been approval by the Meteorite Nomenclature Committee. Physical Description The Kimba meteorite resembles in outline one half of a rectangle broken upon a diagonal (Fig. 2). The ' Department of Mineralogy, South Australian Museum, North Terrace, Adelaide, South Australia 5000. ? Department of Geology and Geophysics, University of Adelaide, Adelaide, South Australia 5005, ‘Jan Wark Research Institute, University of South Australia, Mawson Lakes, Adelaide, South Australia 5095. * Author to whom all correspondence should be addressed triangular section measures about 11 x 11 x 15 cm, and is about 6 cm deep. The stone has a deep fissure caused by weathering, with subsidiary cracks branching from it, but the whole is very coherent and solid. The surface is dark reddish brown and rough because of the adherence of grains of quartz. In the surface oxidation crust, a number of black spots or blisters up to 2 cm across have developed, probably caused by weathering and oxidation of iron-nickel metal and troilite (Fig. 2). Petrographic Features One corner of the stone was sawn from the mass and a polished thin section was prepared from the off-cut. This section was used for the petrographic examination and for electron microprobe analysis. The interior of the meteorite is dark-grey, and chondrules and some troilite are apparent under low magnification. The chondrules and chondrule fragments are partly recrystallised and have well- defined boundaries, so are recognisable even without using crossed polars. They are typically less than 0.5 mm in diameter but some chondrules measure up to 2 mm. Olivine and pyroxene grains up to 0.5 mm across, along with chondrules and their fragments, are embedded in dark, opaque, ferruginous matrix. Troilite occurs as finely disseminated grains throughout the matrix. Several varieties of chondrules are apparent. Barred olivine chondrules (BO type of Wasson (1993)) are most distinctive in the studied thin section. They appear as skeletal olivine with thin lamellae of turbid, glassy mesostasis with a Na- dominant feldspar composition. Chondrules of a second type, radial pyroxene (RP), are also present, and several of these display a cryptocrystalline structure with wavy extinction. A number of granular olivine - pyroxene (GOP) chondrules and porphyritic pyroxene (PP) chondrules are present, containing 40 G. HORR, A. PRING & M. ZBIK 136° 137° 138° 139° 32 SOUTH AUSTRALIA PORT AUGUSTA 33° PORT PIRIE 3a" Fig. 1. Map showing the location of Kimba. bObern ‘ \ | { j i | { | | Fig. 2. Photograph of the Kimba meteorite, showing the deep weathering cracks and the large blisters on the surface. THE KIMBA METEORITE: AN (H4) CHONDRITE FROM SOUTH AUSTRALIA 41 large, euhedral, polysynthetically twinned pyroxene grains. A single porphyritic olivine (PO) chondrule is present in the thin section and it displays coarse, euhedral olivine grains with turbid intergranular mesostasis. Cryptocrystalline pyroxene chondrules (C) are numerous, the largest being about 2 mm in diameter. In addition, there are lithic fragments up to 4 mm in diameter of sub-rounded aggregates of olivine crystals and microcrystalline groundmass. The olivine grains in Kimba display normal extinction (unstrained) and are unaltered but show irregular cracks that are stained by iron oxide. While some troilite is evident by reflected light microscopy, all of the iron-nickel metal appears to have been oxidized. Mineralogy Compositions of the silicate minerals were determined using an electron microprobe (CAMECA SX51 with Moran analysis package) at the University of Adelaide Centre for Electron Microscopy and Microstructure Analysis (now Adelaide Microscopy). Analyses were made using an accelerating voltage of 15 keV, a sample current of 20 nA, and beam width of 0.1 Um. Olivine in the Kimba meteorite is equilibrated, with a mean fayalite content of Faio.2+03 (n=15){Mg/(Mg+Fe)}. The orthopyroxene shows only small variations in chemical composition, with a mean ferrosilite content of Fsi67::3 (n=10) {Mg/(Mg+Fe)} and a wollastonite content of 1.2 mol% (n=10){Ca/(Cat+Mg+Fe)}. The composition of the high-Ca pyroxene varies from grain to grain. The glassy mesostasis has _ feldspar-like composition (approximately Abs2Ani3Ors) and is probably not original as the meteorite is heavily weathered. Classification The Kimba meteorite has been classified as an H4 chondrite. The olivine (Fa,y..) and low-Ca pyroxene (Fs\,7) compositions are within the range of the H chondrites (Keil & Fredriksson 1964). The well- defined chondrule boundaries, equilibrated olivine and low-Ca pyroxene compositions, abundance of high-Ca pyroxene which varies in composition, and feldspar-like mesostasis predominantly consisting of microcrystalline material, suggest that the Kimba meteorite belongs to the type 4 classification of Van Schmus & Wood (1967). The wollastonite content in the low-Ca pyroxene lies in the higher range of H4 chondrites (Scott et al., 1986). Olivine and pyroxene crystals in the Kimba chondrite display irregular fractures and sharp optical extinction, which all indicate that the meteorite does not seem to be shocked after metamorphism. According to the classification scheme of Stoffler e¢ a/. (1991), the shock facies is estimated to be S1, i.e. unshocked. Near complete oxidation of metal and troilite, but no alteration of silicates in the meteorite indicates weathering state W4 on the classification scheme of Wlotzka (1993). Related Meteorites Kimba is the eighth H4 chondrite to be reported from South Australia, but the only one from the Eyre Peninsula. Four of the others, Witchelina, Kittakittaooloo, Coonama and Myrtle Springs, were found in the northeastern region of South Australia and have recently been reviewed (Zbik and Pring 2004). The other three H4 chondrites (Cook 004, Cook 007 and Cook 009) are all from the Cook area of the Nullarbor Plain, some 600 km to the west of Kimba (Grady 2000, Bevan and Pring 1993). It is unlikely that any of the other South Australian H4 chondrites would be part of the same fall as Kimba, as the nearest site, Myrtle Springs, is some 330 km to the North. The Eyre Peninsula of South Australia is one of the most productive areas of the state for meteorite finds. This is a product of both its dry climate and the extensive clearing of the land for cropping (see Wallace and Pring (1991) for a brief summary). Four other H-group chondrites have been found in the area: Buckleboo (H6), Kaldoonera Hill (H6), Kappakoola (H6) and Kielpa (H5). The distinction between type 4 and 5 in the Van Schmus and Wood classification is based on the abundance of clinopyroxene, with type 5 chondrites having only very minor amounts, While classification into the petrological types can be subjective, we think that it is very unlikely that Kimba and Kielpa are related, even thought the average olivine compositions are somewhat similar (Fais» for Kielpa and Fai9.2 for Kimba). Acknowledgments The authors are thankful to Ben McHenry for drafting Fig. | and Trevor Peters for the photograph of the meteorite (Fig. 2). We also wish thank Mr Angus Netting and Mr John Terlet, of Adelaide Microscopy in the University of Adelaide, for assistance with the electron microprobe analyses. Dr W. D. Birch and Dr A. W. R. Bevan provided constructive comments on an earlier version of this manuscript. 42 G. HORR, A. PRING & M. ZBIK References BevAN, A. W. R. & PRING, A. 1993. Guidelines for the naming of new meteorite finds from the Nullarbor Region, South Australia. Meteoritics, 28, 600-602. Grapy, M. M. 2000. Catalogue of meteorites. 5th edition. Cambridge University Press London. 690pp. KeIL, K. & FREDRIKSSON, K. 1964. The iron, magnesium and calcium distribution in coexisting olivines and rhombic pyroxenes of chondrites. J. Geophys. Res. 69: 3487-3515. Scort, E. R. D., TAYLor, G. J. & KEIL, K. 1986. Accretion, metamorphism, and brecciation of ordinary chondrites: Evidence from petrologic studies of meteorites from Roosevelt County, New Mexico. Proc Lunar Planet. Sci. Conf. 17th, E115-E123. STOFFLER, D., KEIL, K. & Scott, E. R. D. 1991. Shock metamorphism of ordinary chondrites. Geochim. Cosmochim. Acta. 55: 3845-3867. VAN ScHMus, W. R. & Woop, J. A. 1967. A chemical- petrologic classification for the chondritic meteorites. Geochim. Cosmochim. Acta 31: 747-765. WALLACE, M. E. & PRING, A. 1991. The Mangalo meteorite, a new (L6) olivine-hypersthene chondrite from South Australia. Trans. R. Soc. S. Aust. 115, 89-91. Wasson, J. T. 1993. Constraints on chondrule origins. Meteoritics 28: 14-28. WLOTZKA, F. 1993. A weathering scale for the ordinary chondrites. Meteoritics: 28, 460. ZBIK, M. & PRING, A. 2004. The Myrtle Springs meteorite: a chondrite (H4) from South Australia. Trans. R. Soc. S. Aus. 128, 33-36. Transactions of the Royal Society of S. Aust. (2005), 129(1), 43-48. BREEDING BIOLOGY OF LITORIA MICROBELOS (COGGER) (ANURA: HYLIDAE) by M. Anstis* & M. J. TYLER* Summary Anstis, M. & Tyrer, M. J. (2005) Breeding biology of Litoria microbelos (Cogger) (Anura:Hylidae). Trans. R. Soc. S. Aust. 129(1), 43-48, 31 May, 2005. The embryonic and larval development of the Javelin Frog, Litoria microbelos, breeding sites and form of the egg mass are described. At 0.78 mm diameter, the ovum is the smallest of all currently known Australian frog species. Two clutches included 259 and 277 eggs respectively. The tadpole can be distinguished by its very small size, body shape and features of the oral disc. Key Worps: Litoria microbelos, hylid, clutch size, embryonic development, larval development, habitat. Introduction The Javelin Frog, Litoria microbelos (Cogger) is the smallest known species of Australian hylid frog and its distribution extends from coastal north- western Australia to north-eastern Queensland (Barker et al., 1995), where breeding occurs in ephemeral water bodies during the wet season. A brief account of the breeding behaviour and early development is given by Tyler ef a/. (1983), but no material beyond stage 27 (Gosner, 1960) was available for that study and no illustrations were provided. The present paper provides a more complete description of the embryonic and larval development of this species. Materials and Methods Embryos: Two egg clutches laid by two females collected near Coolalinga, Northern Territory were studied. Two females and two calling males were collected individually at 2030 hrs on 5. ii. 2003 from the edge of a shallow, temporary flooded depression in grass near a larger water body. Each male was placed with a gravid female in a separate inflated plastic bag containing rainwater and some grasses. Amplexus took place and egg clutches from each pair were raised to hatching stages in shallow water to a depth of 3 cm, at 25-27°C. Embryos between stages 11 and early 25 were studied using the staging descriptions of Gosner (1960), and Anstis (2002) between stages 19-25. Larvae: Tadpoles studied included those raised from the above clutches and samples collected at Marrara near Darwin and Coonjimbah Billabong “ 26 Wideview Rd., Berowra Hts, N.S.W. 2082, Australia ' Department of Environmental Biology, University of Adelaide, S.A. 5005, Australia. near Jabiru, N.T. Larvae collected in the field were maintained to metamorphosis to confirm identity. Tadpoles were raised outdoors in available sunlight/shade conditions within containers of pond water to a depth of 18 cm, on a substrate of sand and rooted vegetation. They were fed on goldfish flakes, frozen lettuce and small protein food sticks once they reached stage 35, Specimens were measured with an ocular micrometer attached to a Wild M5 stereoscopic microscope and vernier calipers. Live specimens were examined under the microscope and then preserved in 4% phosphate-buffered formalin at various developmental stages (Gosner, 1960). Measurements of embryonic and larval stages are presented in Tables | and 2. Descriptive terminology follows Anstis (2002). Illustrations were made by Anstis, with the aid of a drawing tube attached to the microscope. Morphometric measurements in mm (see Anstis ef al., 1998) were taken on a small sample of stage 36- 39 tadpoles anaesthetised in 1% chlorbutol solution. Abbreviations are as follows: Lateral view: TL = total length; BL = body length; BD = body depth; BTM = depth of tail musculature at base of tail; TD = maximum tail depth; DF = depth of dorsal fin (at TD); VF = depth of ventral fin (at TD); TM = depth of tail musculature (at TD); SS = snout to uppermost corner of opening of spiracle; SE = tip of snout to anterior rim of eye; SN = tip of snout to anterior rim of naris; ED = eye diameter. Dorsal view: BW = body width; EBW = body width at level of eyes; BTMW = width of tail musculature at base of tail; IO = inter-orbital span measured between inner edge of each eye; EN = anterior edge of eye to posterior edge of naris; IN = internarial span, measured between inner rim of each naris. Ventral view: ODW = maximum width of oral disc. UL} [RSIOP UO SYOTF SUTJ OUIOS “9]OSNUT [1B] [BSIOp IOAO soyoied moy[ak pue spueq yaep ‘diy [rej pue ojosnur [1e} [e.1o}e] “UUINSIOP I9AO Surjddiys yoryq “jounsip Mou ads [e19ye] yep ‘UMOIq MOT[EYS SyIBOYS Mel ‘ded jerpowi INOYUM jd ‘Any dopaasp 0} MOL 4SP] ed ‘Pastunesoy Aqyty 0} 1eou pidnd punose Sul pos ‘deS mowieu AOA uado aqny (Aji) jYSI] oposnu [1e) pue Apog SMOI 100} [[V poonpoy oInsstj plo1oyD qua ‘apoeds UIU ¢ SIU 89 «STZ TeI]d JOJUDA ‘Q[ISNU [IB] JO 9dBJAINS [esp ops yor ssoroe soyoqed yor[q proiq ye oeyyided [elo JepNSos ‘uswopge ojo ysnf OUIOS *O[QISIA 0} 049 YSNoIY} NOUS WOT Ayoreq MOL 7V peoy Jo apis uo odiys yep ‘PasTUNeIOy Es ‘gfosnur [Ie] UO pue UOISAL Apvurey winjnosedo = [BIQOLIOA pure urleIg JO OpIs SMOI Y}00} 2 pojuowsid JO a8po je WIS B opoesids 2 Joyo UO sttopids mojoq “d pue syyeoys ‘poonpal winjnoiedo pasoyo soy.ny suado Apreau aqny judA s soroydopmt pros asreds Mel Jo sospy Apysi[s Aq palsaod = AANSSTJ PlOIOYD ~—« POUT Jap [JOM S[IOo yeuNsayUT UI OZ SIU LP FT 2 suawsods ounssl} SUIWWIO} OSvARIIO 2 awios ul sdunyo Moy uid [eoepouioys asp [eigr| (O] “SIq) proroys s0J ydaoxo euNsoyul puoses SApYsI]s in B SUILUIOJ ‘[ISNUI [12] UO punose yuosoid — JAMO] JO JouI09 juowidojaasp —- yep SLI “Ieayo Suuesjd pue peoy punose (poyoyey ysnf moj ysey) 4 Sursvosour sasoydourayy soSpll [eiqe’] yors Mojog UINUIXeYY «SOU BaUIOD ~=—_— papuedxa soyjany [jem Apog ulul QZ ‘SIYSZ ZZ s 0} SuruuIsaq suly ‘onbedo OSVARI]O [VULJSOIUL ISILJ Sso] URI dAoge [eM Apog ‘ApJorojsod Apdnige ssodey “‘Q]OSNUT [12] JOAO pue Y[OA JO quid yeoepowioys S]UOWIR| LJ AyyIu BIUIOS ‘ApIoLIojue Jodaap ses yok DILJINS [BSIOP JOAO “UILIG JO 19UI09 Yeo JOMO| €-Z pure “a[qeuIoosIp ‘popuedxo Apysiys (Suryoyey ysnf a10u) punowe sysoyy oroydourjayy MO[Iq OI] YW syuoureyry soddn ¢-7 uorser ondO, ulerq aAoge [jem Apog UI QZ SIU GT IT apy & Jodaop opi] (Suryoyey ysnf ows) aaoge SV aaoge SV aaoge SV uayisugy stp aaoqe SV ®B SUI} SQAOge SY uw Sp sIyL 07 umolg AYsnp surf quowsid dIQISIA sid [eeu ‘UMOIG Yep puRls Suroiey Joreao YM pospe ‘popunol A[peoiq dy SuMOIg AUUBOID OBS YJOA suodsap yd sdy pajuiod ATIUBITS ‘O[QISTA ‘loys [rey syuouTUOId ‘uMOIG 9[ISNUI [Ie] pue Apog eoepowo}s ‘\UIUTUOIg s]]I3 Jo sdiy. = quaprAo Ajoreg aqinb uorsai peay UI Sp SIU S-0 61 6] aseIs 18 quowidojsacq sue3lo Sits) quowido[sacq adeys Suryoyey ISA JUSWISIg are) OAISOupy yeusoyxg ondo pel /Apog WO SINOH SRS yt “SOJAQOIDIUT BIIOWT JO soduquia fo sainjvaf jojuaudojaaap Aay ‘| ATaVE, BREEDING BIOLOGY OF LITORIA MICROBELOS (COGGER) (ANURA: HYLIDAE) 45 TABLE 2. Measurements of embryos, mean with ranges in parentheses in mm. Stage n Ovum Capsule 11 11 0.78 (0.77-0.82) 2.84 (2.05-3.28) 12 7 0,81 (0.77-0.86) 2.65 (2.05-3.11) 17 8 1.57 (1.35-1.72) Total Length Body Length 19 14 2.89 (2.7-3.11) 1.8 (1.72-1.88) 21 15 3.71 (3.6-3.85) 1,86 (1.8-1.96) 22 9 4.0 (3.85-4.18) 1.88 (1.8-1.92) 24 8 4,5 (4.34-4.67) 1.91 (1.84-1.96) Fig. |. a—Stage 17, b— Stage 19, c — Stage 21. Results Breeding Sites Adult males call from grass stems, low branches of shrubs, and on the ground beside temporary pools, flooded ditches and larger semi-permanent ponds, swamps and billabongs. At the Coolalinga site, many males were calling in a swampy area by 1600 on 5 February, 2003 after a heavy afternoon storm. That night, most frogs were active near recently formed small shallow puddles in grassland beside the main swamp. Four calling males were on a single small grass tussock and there were numerous such tussocks (and low shrubs) supporting frogs around the edges of the swamp and flooded areas nearby. Oviposition and embryonic development Amplexus is axillary. Oviposition occurred after 2130 during the night of 5 February but was not observed. The egg clutches contained 277 and 259 eggs. Most eggs were laid singly on the bottom of the bag, but some were adhering in small clusters of 2- 10 eggs, and others were attached to vegetation or fine twigs. Early development is quite rapid and embryos first examined at 0930 on 6 February, were at stages 11 and 12. The ovum is very small (mean diameter 0.78 mm, stage 11) and at stage 11, the animal pole is dark brown and the yolk is white. The capsule is a single fluid layer, initially spherical but becoming more fluid and expanded as the embryo develops. Measurements of embryos are provided in Table 1. At 1210 on 6 February, embryos were at stage 13 and appeared dark brown macroscopically but lighter brown under the microscope (x6 magnification). By 1530 most embryos were at early stage 17: head broad in dorsal view, arched in lateral view; tail bud points upwards; gill arches quite prominent; adhesive organs well developed; stomodaeal pit visible; hatching gland lightly pigmented. By 2000 all embryos had reached stages 17-18. Details of embryonic development from hatching are presented in Table 1-3 and Figure 1. Larvae The period of greatest growth is between stages 25 and 30, when TL and BL measurements treble. From 40. BL increments are smaller per stage (0.5 mm or less). A composite description of fully grown live tadpoles at stages 36-39 is given, with some comparison to stage 25, Body: small, cylindrical, slightly wider than deep; snout broadly rounded in dorsal view, rounded in lateral view; eyes lateral, prominent and with slight dorsal tilt, iris mostly dense gold with thin copper ring around pupil; nares closer to snout than to eyes, widely spaced, directed dorsally, quite large and cavernous relative to size of body; spiracle well below body axis, opens dorsoposteriorly, posterior to midpoint of body; vent tube dextral, short, opens midway up ventral fin (type a, Anstis 2002). Tail: fins shallow to moderate, dorsal fin begins at base of body, arches acutely to greatest depth anterior to midpoint and tapers gradually to narrowly rounded tip; ventral fin mostly fairly shallow with a slight arch near midpoint, then tapers evenly to tip. Musculature moderate and tapers to a fine point. Pigmentation in Life Dorsal view: broad, dark brown band extends from snout, between nares, over chondrocranium and along vertebral region; head region less pigmented on either side of band then darker again at each side anterior to eye, fine melanophore flecks all over anterior half of body; pigment less dense over 46 M. ANSTIS & M. J. TYLER TABLE 3. Measurements of larvae and metamorphs; mean with ranges in parentheses in mm. Stage n Total Length Body Length 25early 17 5.05 (4.71-5.33) 2.16 (2.05-2.29) 25 late 17 8.05 (6.27-9.17) 3,54 (2.89-4.02) 30 2 15.45 (15.13, 15.77) 6.35 (6.27, 6.44) 31 5 16.38 (15.13-16.59) 6.47 (6.27-6.60) 33 2 17.30 (16.74,17.87) 7.0 (6.60,7.40) 34 1 16.58 6.92 35 7 18.34 (17.54-9.32) 7.11 (6.92-7.24) 36 6 19,27 (18.5-20.0) 7.37 (7.24-7.56) 37 4 19,90 (18.83-21.0) 7.42 (7.32-7.56) 38 2 20.75 (20.5, 21.0) 7.84 (7.64, 8.05) 39 5 22.65 (20.8-23.7) 7.98 (7 40-8.21) 40 3 21.40 (20.2-22.8) 7.72 (7.40-8.05) 41 1 22.2 7.88 44 3 10.67 (9.48-12.39) 7.18 (6.76-7.56) 45 2 7.80 (7.56, 8.05) 7.12 (7.08, 7.16) 46 11 7.46 (6.44-8.21) TABLE 4. Measurements of morphometric characters (mm) for one larva at Stage 36 and 39. Character Stage 36 Stage 39 TL 20.4 23.7 BL 8.05 8.21 BD 3.7 4.02 BW 4.02 4.5 EBW 3.54 3.7 BIM 1.61 1.61 BIMW 1.61 1.61 TD 4.34 4.5 DF 1.77 2.09 T™ 1.45 1.28 VF 1.13 1.13 10 1.93 1.93 IN 0.8 0.8 EN 0.96 0.88 N 0.32 0.32 SS 5.15 4.99 SN 0.8 0.64 SE 1.93 1.77 ED 1.28 1.28 ODW 2.09 2.09 abdomen, where it becomes somewhat speckled on either side of darker band. Dorsal surface of tail musculature patchy black and gold, at times appearing banded. Lateral view: sides of body less pigmented over upper half of abdomen (intestines partly visible) with some gold clusters, merging with opaque silver/white over lower half. Dark lateral stripe from snout to eye, broadens posterior to eye over anterior upper half of abdomen, then becomes less dense and mostly obscured by a fine layer of golden iridophores. Gold continues over branchial region with clumps of melanophores and gold clusters Fig. 2. a — Stage 37 (dorsal view), b — Stage 37 (lateral view). Fig. 3. Oral Disc. beneath eye. Dorsal surface of limbs with patches of melanophores and gold; a line of melanophores and gold borders each toe. Sides of tail musculature have a mostly continuous dark stripe down the middle bordered above and below by a clearer or non- pigmented stripe highlighted with fine gold clusters. Fins densely speckled with small black clumps of melanophores and fine gold clusters of iridophores, otherwise clear. Preserved specimens at stage 25 have only stippled melanophores over most of the darker dorsal areas, while larvae in later stages from about 35 have dense dark areas, readily visible once the gold iridophores have gone. Ventral view: dense, opaque copper sheen over abdomen and clear over anterior half apart from a few fine gold clusters. The copper sheen on the ventral surface is thin and patchy in live individuals at stage 25 and the dark stripe down the lateral surface of the tail musculature is less defined. Oral Disc (Fig. 3, specimen at stage 39): directed ventrally, quite wide (e.g. ratio ODW/EBW = 0.59, stage 36); wide anterior medial gap in marginal papillae, one row of papillae on either side of gap increasing to 2-3 laterally and one row around BREEDING BIOLOGY OF L/TORIA MICROBELOS (COGGER) (ANURA: HYLIDAE) 47 posterior margin with a few additional submarginal papillae in some specimens. Two upper and three lower tooth rows; A! is the longest row, A? with narrow but distinct medial gap. Rows P!24 entire and P? may be very slightly shorter than P! and P?. Jaw sheaths medium (Anstis, 2002), upper sheath with long lateral processes, lower sheath quite acutely V-shaped. Metamorphosis It was not possible to maintain the larvae hatched from eggs until metamorphosis. Larvae collected from the field at stages 38-40 on 6 February, 2003 began to metamorphose on 10 February and strongly resemble the adult in shape. Dorsum yellow-gold with bronze sheen developing, iris golden and a small silver line across base of each finger and toe disc; venter mostly opaque white. Lateral stripes of adult not yet defined. Six anaesthetised metamorphs from Marrara at stage 46 have a mean length of 7.87 mm (7.24 — 8.21) and five preserved specimens from Coonjimbah Billabong have a mean of 6.98 mm (6.44 — 7.72). Larval Behaviour L. microbelos tadpoles have been observed most commonly on the substrate in captivity, but will at times range through midwater regions of the water column, In the field, they are often found on the substrate in shallow water at the sides of ponds or flooded ditches, hidden amongst vegetation. They have been observed grazing on the substrate, over rocks and on vegetation. Discussion Breeding Sites Adult Litoria microbelos are abundant during the wet season, frequently calling after rain during late afternoon and night. Males often select very small, shallow water-filled depressions to call from, many of which dry up during the heat of the next day or two. If breeding takes place in these sites, many embryos would not survive unless adequate daily follow-up rain occurred. However, large numbers of calling males are also found around the edges of larger billabongs, swamps and flooded ditches, and populations are obviously thriving in many areas observed from near Darwin to Jabiru. Tyler e¢ al. (1983), state that males only call from the base of grass tussocks if males of bicolor are absent. In the presence of L. bicolor the L. microbelos call from higher positions. Oviposition and embryonic development Although oviposition was not witnessed, it appears, from the position of the eggs when found, that they are laid in small groups and either become attached to vegetation or are scattered by the adults over the substrate. The ova measured for this study had already reached stages 11-12 and, as a very small increase in diameter occurs from about stage 10, it is likely that the diameter of ova in stages 1-9 is slightly less than that given here. However, even at stage 11, this species has the smallest mean diameter of all known Australian frog species, the nearest being Crinia tinnula (0.85 mm, stage 10, Anstis, 2002). The combination of the extremely small size of hatchlings at stage 19 (mean TL = 2.89), short tail and very broadly rounded tail tip, help to distinguish this species from all other sympatric hylid embryos at this stage (Anstis unpubl., Anstis, 2002). Embryonic development to hatching at stage 19 was reached at a minimum of 34 hours 15 min after the eggs were laid. This is quite rapid in comparison to that known for other sympatric species such as Litoria rubella (stage 20, after 72 hrs at 30°C, Tyler et al., 1983), Uperoleia inundata (stage 20, after 48 hrs at 27-28°C, Anstis, unpubl.) and Crinia bilingua (stage 20, after 48 hrs at 25- 27°C, Anstis, unpubl.). However, embryos of the sympatric species, Limnodynastes ornatus, hatched at stage 20 after only 18 hours at 30°C (Tyler ef al., 1983). A mean of 4.1 mm (4.0-4.2) for embryos at stage 24 is given by Tyler e¢ al. (1983), similar to the mean of 4.5 for specimens in the present study (Table 2). Larvae Tadpoles raised from eggs remained in stage 25 for at least 20 days. More observations on specimens raised in warmer, shallower water (and other varied environmental conditions) need to be made to test whether this is a typical rate of development for this species or whether development slowed as a result of other factors during culture. Tyler ef al, (1983) provide a mean TL of 11.4 mm for larvae at stage 27, which is 3.35 mm longer than the mean TL of 8.05 for late stage 25 specimens in the present study. This species can be distinguished from all other fully grown conspecific larvae by a combination of its very small maximum size, cylindrical body shape, ventral oral disc, tooth row formula of 2(1)/3, narrowly rounded tail tip (rather than finely pointed) and pigmentation patterns. In relation to the body types of Anstis (2002), it most closely affiliates with a small version of type 2, (Litoria latopalmata complex) in shape and behaviour, but has a slightly more ventrally directed oral disc and larger nares. Small Type 2 hylid larvae at early stage 25 commonly sympatric with L. microbelos, such as L. nasuta, L. inermis and L. tornieri can be distinguished from the former by their lack of lateral stripes and more finely pointed tail tip, and 48 M. ANSTIS & M. J. TYLER L. tornieri is easily separated from all species at stage 25, by their red tail fins (Anstis, unpublished). Type 1 sympatric hylid tadpoles also at stage 25, such as L. rothi, L. bicolor and L. caerulea, while somewhat similar in body shape to larvae of L. microbelos, can be distinguished by their finely pointed tail tips, surface-dwelling behaviour and anteroventral oral disc, with all but L. bicolor having an LTRF (Alltig, 1970) of 2(1)/3(1). Other conspecific bottom- dwelling larval species closer to L. microbelos in maximum size range (prior to about stage 27) are Crinia bilingua (Tyler et al., 1983), Uperoleia mimula (Richards and Alford, 1993) and U. lithomoda (Davies et al., 1986), but all of these species have a more oval to rounded body shape, different tail shape, a different oral disc possessing a medial posterior gap in the marginal papillae and lack the distinctive lateral stripes of L. microbelos larvae. Metamorphosis Larvae collected in the field at stages 2 30 advanced steadily to metamorphosis, especially those at stages 38 — 40, which began meta- morphosing four days after capture. Larval life span could not be determined for any one group. Acknowledgements Research was undertaken by Marion Anstis under the NT Frogwatch Licence no. 16881. The WWF Frogs! Program and Stan Orchard, the co-ordinator are gratefully acknowledged for their financial assistance and support during this study. Jeanne Young and the Technical Staff of the Zoology Department at Charles Darwin University are gratefully acknowledged for assistance with field work, localities and materials. References Attia, R. A. (1970) A key to the tadpoles of the continental United States and Canada. Herpetologica 26: 180-207. Anstis, M. (2002) Tadpoles of South-eastern Australia: a guide with keys. Reed New Holland. , ALForp, R. A. & GILLEsPIE, G. R. (1998) Breeding biology of Litoria booroolongensis (Moore, 1961), and Litoria lesueuri (Dumeril & Bibron, 1841) (Anura: Hylidae) and comments on population declines of L. booroolongensis. Trans. R. Soc. S. Aust. 122(1), 33- 43. BARKER, J., GRIGG, G. C. & TyLer, M. J. (1995) A Field Guide to Australian Frogs. Surrey Beatty, Chipping Norton, N.S.W. Davies, M., MCDONALD, K. R. & CorBEN, C. (1986) The genus Uperoleia in Queensland, Australia. Proc. R. Soc. Vict. 98(4), 147-188. Gosner, K. L. (1960) A simplified table for staging anuran embryos and larvae with notes on identification. Herpetologica 16, 183-190. RICHARDS, S. J. & ALFORD, R. A. (1993) The tadpoles of two Queensland Frogs (Anura: Hylidae, Myobatrachidae). Mem. Old Mus. 33(1), 337-340. TyLer, M. J., Crook, G. A. & Davies, M. (1983) Reproductive biology of the frogs of the Magela Creek system, Northern Territory. Rec. S. Aust. Mus. 18(18): 415-440. Transactions of the Royal Society of S. Aust. (2005), 129(1), 49-52. HEPATOZOON TACHYGLOSSI SP. NOV. (HAEMOGREGARINIDAE), A PROTOZOAN PARASITE FROM THE BLOOD OF A SHORT-BEAKED ECHIDNA, TACHYGLOSSUS ACULEATUS by P. CLarkK!, P. Hotz? & D. M. Spratt? Summary CLARK, T., HOLz, P. & Spratt, D. M. (2005) Hepatozoon tachyglossi sp. nov. (Haemogregarinidae), a protozoan parasite from the blood of a short-beaked echidna, Tachyglossus aculeatus. Trans. R. Soc. S. Aust. 129(1), 49-52, 31 May, 2005. Hepatozoon tachyglossi sp. nov. is described from monocytes in the peripheral blood of a debilitated short- beaked echidna from the Healesville region of Victoria. Of the Hepatozoon Miller, 1908 species known to occur in Australian native mammals, all of those in marsupials occur in erythrocytes and only H. muris from introduced and native rodents oceurs in monocytes. H. tachyglossi is distinguished from H. muris by its larger size and the lack of a capsule. Key Worps: Hepatozoon tachyglossi, new species, echidna, Tachyglossus aculeatus. Introduction The blood of the short-beaked echidna, Tachyglossus aculeatus (Shaw & Nodder, 1792), has yielded relatively few haemoparasites. Those identified have been restricted to the Piroplasmidae and include Babesia tachyglossi (Backhouse & Bolliger, 1959) and Theileria tachyglossi (Priestly, 1915) (Backhouse & Bolliger 1957, 1959; Mackerras 1959). We report a novel species of Hepatozoon from monocytes in the peripheral blood of a short-beaked echidna from the Healesville region of Victoria and compare the species with other members of the genus previously described from Australian native mammals. Clinical history and methods A male, juvenile, short-beaked echidna was presented to the veterinary service of Healesville Sanctuary. The animal had several injuries including a damaged tongue and a fractured humerus. During captive management and treatment, the animal developed severe dyspnoea and a mucopurulent nasal discharge. It was euthanased after a poor response to treatment with antimicrobial drugs and supportive therapy. A sample of blood was collected ante-mortem from the bill sinus. The morphology of haematological cells in a blood film, stained with Wright’s and Giemsa stains, was assessed by light microscopy. Tissue School of Clinical Sciences, Division of Veterinary and Biomedical Sciences, Murdoch University, South St, Murdoch, Western Australia, 6150. ? Healesville Sanctuary, P.O. Box 248, Healesville, VIC, 3777. ‘CSIRO Sustainable Ecosystems, GPO Box 284, Canberra, ACT, 2601 samples of bone marrow, liver and spleen, were collected post-mortem and fixed in 10 percent buffered formalin. The tissues were processed using standard histological methods; sections were stained with haematoxylin and eosin stains and examined using light microscopy. The organisms and their host cells were digitally photographed. A stage micrometer was photographed at the same magnification and used to insert scale bars to all micrographs. The micrometer was also used for all measurements which are presented in microns as the mean + standard deviation followed by the range in parentheses. Results Intracellular organisms were observed within leukocytes in the peripheral blood (Fig. 1). These were observed in 24/50 monocytes but were not observed in any granulocytes. Hepatozoon tachyglossi sp. nov. (FIG. 1) Host Tachyglossus aculeatus. Location Healesville, Victoria. Type Slide in South Australian Museum No. 28751. Typically the organisms were oval to elongate in shape but were quite pleomorphic with some pyriform and round forms observed. The organisms were 9.9 + 1.4 (7.8-12.4) in length and 4.7 + 0.7 (3.8- 5.9) in width (n = 18). Most organisms had an eccentric, subterminal nucleus and some exhibited a 50 P. CLARK, P. HOLZ & D. M. SPRATT Fig. 1. Examples of Hepatozoon tachyglossi sp. nov. within monocytes from the peripheral blood of a short-beaked echidna. Bar = 10 um. small amount of punctate, dark brown — black pigment in the cytoplasm. No distinct capsule was observed and there was only a subtle difference in colour between the cytoplasm of the organism and the cytoplasm of the cell. Only one organism was evident per cell; in some cases this caused displacement of the host cell nucleus. No consistent changes in cell morphology were evident. No extracellular organisms were noted. Examination of histological sections of bone marrow, liver and spleen did not reveal schizonts. Discussion The organism described is morphologically similar to a species of Hepatozoon Miller, 1908, the only coccidian genus that inhabits the blood of mammals. Desser (1990) highlighted the taxonomic confusion arising from the problem of differentiating species of Hepatozoon from that of Haemogregarina Danilewsky, 1885 on the basis of gamonts in the blood of the vertebrate host. Differentiation was based primarily on the size of the oocysts and the presence or absence of sporocysts in the invertebrate definitive host. Desser (1990) tested and confirmed the hypothesis of Landau ef al. (1972) that tiny cystic stages (cystozoites) in the liver and lungs represent a common feature in all species of Hepatozoon and, as a consequence, differentiation of the genera Hepatozoon and Haemogregarina based solely on the stages in the vertebrate host is possible. These findings further support the evidence of Landau ef al. (1972) that transmission of species of Hepatozoon may be by predation, as well as by ingestion of the infected arthropod vectors. The genus has not been reported previously in HEPATOZOON TACHYGLOSSI, SP. NOV. IN AN ECHIDNA 51 Fig. 2. Examples of Hepatozoon muris from the blood of Rattus lutreolus. Bar = 10 um. monotremes. However, H. peramelis (Welsh & Dalyell, 1909), H. dasyuroides Mackerras, 1959, H. dasyuri (Welsh, Dalyell & Burfitt, 1909), H. petauri (Welsh & Barling, 1909) and H. pseudocheiri Mackerras, 1959, have been described from Australian marsupials (see Mackerras, 1959). In addition, O'Donoghue and Adlard (2000) reported several unidentified species of Hepatozoon from Australian marsupials. All reports have recorded the parasite within erythrocytes or ‘free’ in the blood (Mackerras 1959; Speare et al. 1984; Bettiol e¢ al. 1996). Also, H. muris (Balfour, 1906) has been recognised in both the introduced Rattus norvegicus (Berkenhout, 1769) and R. rattus (Linneaus, 1758), and the native rodents, R. fuscipes (Waterhouse, 1839) and R. sordidus (Gould, 1858) (Mackerras 1959; O’Donoghue & Adlard 2000). Hepatozoon muris, in contrast to the species reported in marsupials, infects leukocytes (Soulsby 1982). The organisms in the current study were oval to elongate in shape, as are many species of Hepatozoon, and were similar in size to H. peramelis (9.0 — 10.0 um by 3.0 — 3.5 um; Welsh & Dalyell 1909) and slightly larger than H. petauri (7.5 — 8.0 um by 3.5 — 4.0 lm; Mackerras 1959), H. pseudocheiri (8.0 — 13.0 um by 1.5-3.0 um; Mackerras 1959), and Hepatozoon sp (8.7 + 0.2 um by 2.1 + 0.4 um, Bettiol et al. 1996). In contrast, H. dasyuroides (12 — 13 tum by | — 2 um; Mackerras 1959) and H. dasyuri (12 um by 4 [tm; Welsh Dalyell & Burfitt 1909, 1910) are longer, narrower parasites. Hepatozoon tachyglossi sp. noy., like H. muris, occurs in the monocytes of its host. However, H. muris is a smaller (7.0 — 8.0 [tm by 3.0 — 3.5 um; Mackerras, 1959) more ovoid, and typically less morphologically variable parasite, than H. tachyglossi (Fig. 2). Additionally, H. muris has an eosinophilic capsule that was not evident in A. tachyglossi. Typically, the definitive hosts and vectors of species of Hepatozoon are blood feeding arthropods. The mite Laelaps echidninus Berlese, 1887 fills this role for H. muris. In the current case, ticks identified as Aponomma concolor Neumann, 1899, were evident on the echidna but were not examined for sporocysts and could not be proven to be the definitive host. Although the life history of this species of Hepatozoon remains unknown, we consider that specific status is warranted on the basis of morphological features and the occurrence of the parasite in monocytes of the monotreme, Tachyglossus aculeatus. The effect on the host was not determined due to the other pathological processes and the animal’s debilitated state may have allowed the organism to proliferate. Further work to identify the prevalence of this organism in short-beaked echidnas, assess any pathogenic effects on the animal and phylogenetic studies to determine its relationship to other species of Hepatozoon in Australia need to be undertaken. Acknowledgements We thank Dr Peter O’Donoghue for sharing his knowledge of protozoan organisms with us and two anonymous referees whose comments improved the presentation of the manuscript. References Backnouse, T. C., & BOLLIGER, A. (1957) A piroplasm of the echidna (7achyglossus aculeatus). Aust. J. Sci. 19, 24-25. & (1959) Babesia tachyglossi n. sp. from the echidna Zachyglossus aculeatus. J. Protozool. 6, 320-322. BeETTIOL, S. S., GoLDsmID, J. M., Le, D. D. & Driessen, M. (1996) The first record of a member of the genus Hepatozoon in the eastern barred bandicoot (Perameles gunnit) in Tasmania. J. Parasitol. 82, 829-830. Desser, S. S. (1990) Tissue “cysts” of Hepatozoon griseisciuri in the grey squirrel, Sciurus carolinensis: the significance of these cysts in species of Hepatozoon. J. Parasitol. 76, 257-259. Lanbau, I., MICHEL, J. C. & CHasaup, A. G, (1972) Cycle biologique d’Hepatozoon domerguei: discussion sur les caractéres fondamentaux d’un cycle de Coccidie. Z. Parasitenk, 38, 250-270. MacKerras, M. J. (1959) The Haematozoa of Australian mammals. Aust. J. Zool. 7, 105-135. 52 P. CLARK, P. HOLZ & D. M. SPRATT O’ DONOGHUE, P. J. & ADLARD, R. D. (2000) Catalogue of protozoan parasites recorded in Australia. Mem. Qld. Mus. 45, 1-163. Soutssy, E. J. L. (1982) “Helminths, arthropods and protozoa of domesticated animals”. 7th ed. (Bailliére Tindall, London.) SPEARE, R., HAFFENDEN, A. T., DANIELS, P. W., THOMAS, A. D., & SEAWRIGHT, C. D. (1984) Diseases of the Herbert River ringtail, Pseudocheirus herbertensis, and other North Queensland rainforest possums. pp. 283-302 In Smith, A. P. & Hume, I. D. (Eds.) “Possums and Gliders”. (Surrey Beatty & Sons, Chipping Norton.) WELSH, D. A., & DALYELL, E. J. (1909) Haemogregarina peramelis: a free haemogregarine of an Australian bandicoot. J Path. Bact. 14, 547-549. & Burritt, M. B. (1909) Haemogregarina dasyuri. A preliminary note on an undescribed haemogregarine of the Australian native cat. Aust. Med. Congr. for 1908. 2, 333-337 (1910) Haemogregarina dasyuri: a haemo- gregarine of the Australian native cat. J. Path. Bact. 14, 542-546. Transactions of the Royal Society of S. Aust. (2005), 129(1), 53-58. RECORDS OF HUMPBACK WHALES MEGAPTERA NOVAEANGLIAE IN SOUTH AUSTRALIA by C. M. KEMPER Summary Kemper, C. M. (2005), Records of humpback whales Megaptera novaeangliae in South Australia. Trans. R. Soc. S. Aust. 129(1), 53-58, 31 May, 2005. Opportunistic sightings and museum specimens of humpback whales (Megaptera novaeangliae) in South Australia (n = 116) were collated to the year 2003. Records were made in all months of the year, with 57% during June and July, Timing of this peak and the presence of neonates presumably related to the northward migration from high latitudes to breeding grounds in the tropics. Sightings and beach-washed carcasses were widely distributed, from Head of Bight to the Victorian border, with apparent concentrations in eastern Gulf St Vincent, Kangaroo Island and Victor Harbor regions, and to a lesser extent in the south-east of the State and southern Eyre Peninsula. Most sightings involved single whales or groups of two (range 1 — 4). All five carcasses were neonates or juveniles with total lengths of 3.88 m to approx. 10 m. Estimated lengths (5 — 15 m) were available for 17 live whales. The geological age of one museum specimen is uncertain but may predate European settlement. This and whaling records from Fowler Bay during 1840 indicate that humpback whales have been present, and may not have been uncommon, in South Australia since at least the early 19th century. Research is needed to determine the relationship of South Australian humpback whales to populations migrating off the western and eastern coasts of Australia. Key Worps: Humpback whale, South Australia, sightings, strandings, distribution. Introduction Humpback whales (Megaptera novaeangliae) occur in all major ocean basins and from the polar regions to the tropics (Clapham 2002). Populations in the southern hemisphere are broadly delimited by longitudinal degrees: for the Australasian region these are Group IV between 70° and 130° E, and Group V between 130° and 170° E (Mackintosh 1965). These populations migrate along the west coast of Australia, and the east coast of Australia and throughout New Zealand, respectively (Dawbin 1966). They have different genetic make-up (Baker et al. 1994), songs (Cato 1991; Dawbin & Eyre 1991) and migration routes (Dawbin 1966). At 1999, the size of the Group IV population was estimated to be 8207-13 640, increasing at a rate of about 10% per annum (Bannister & Hedley 2001) and that of Group V, 36004440 increasing at about 11% per annum (Paterson e¢ al, 2001). Humpback whales in the Australasian region were taken by pelagic and shore-based whalers during the 19th and 20th centuries (see overviews by Bannister & Hedley 2001 and Paterson 2001). Between the early 1900s and 1963, most humpbacks off Australia were taken from shore-based whaling stations along the east and west coasts of the continent. There were also significant catches by pelagic whalers during the mid to late 1930s (Chittleborough 1965) and illegal whaling by the Soviet fleet took many animals South Australian Museum, North Terrace, Adelaide, South Australia 5000. south of the continent until 1973 (Tormosov 1995; Mikhalev 2000). There was no targeted, shore-based whaling of humpback whales in South Australia during the 20th century but some animals were taken there opportunistically during 19th century whaling for southern right whales (Eubalaena australis). There are few published records of humpback whales in South Australia and as a result, it is generally thought that these represent vagrants. However, in 1840 the whale ship Amazon took eight humpback whales at Fowler Bay (Bannister 1986). Chittleborough (1965) published two sightings — one whale at the head of the Great Australian Bight in August 1952 and a female with newborn calf in Gulf St Vincent in winter 1961. Museum specimens and beach-washed carcasses from South Australia were reported and /or illustrated in Aitken (1971), Kemper & Ling (1991) and Judd et al. (1992). This study documents the geographic and temporal distribution of sightings and carcasses of humpback whales in South Australia using opportunistic records from various sources. Some information is also provided on relative age/size of animals, as well as number of animals seen in each group. Avenues for further research are explored. Material and Methods Records of opportunistic sightings (n = 109) of humpback whales were obtained from the whale sightings database at the South Australian Museum. The Museum database has been compiled from various sources, including the South Australian 54 Whale Centre (Victor Harbor), marine mammal researchers, public, fishers, and the published records listed above. Sighting records dated from 1948 to 2003. Records of carcasses and South Australian Museum specimens (n = 7) dated from ‘before 1913’ to 2000. Records were coded for reliability of identification: 1 (n = 11) = photographic evidence; 2 (n = 50) = certain/probable (experienced observer or distinctive features noted; i.e. white underside of flukes, trailing edge of flukes scalloped, very long flippers, dark lumps on head, small irregular dorsal fin); 3 (n = 29) = some doubt (inexperienced observer, not enough features recorded to be confident of identification); 4 (n = 19) = no description of features provided, observer experience unknown but identification may be correct. Latitudes and longitudes, including the accuracy of localities, were calculated by Museum staff and trained volunteers. Distance from shore, number of whales sighted and other details were sometimes supplied by observers. Carcasses provided accurate body length data as well as the degree of physical maturity of the prepared skeleton. Whale length was estimated quite accurately for three live animals: one by the author from photographs of the flukes and two by fishers C. M. KEMPER who related the length of the whale to that of the boat. Relative size and/or rough estimates of whale length were noted by some observers. Results Of the seven records of humpback whale mortalities, two were bones of unknown age (of which one was possibly subfossil) and therefore no details were available of the carcasses, circumstances or when the animals died (Table 1). One juvenile (S12) probably floated in dead, and one neonate (M21309) was very fresh when first discovered on the beach (Fig. 1) and therefore was probably alive nearby before it stranded or washed up. M15187, a juvenile, was reported as being alive in the vicinity for several days before washing up dead. Two animals (M12778 and M16971) were in an advanced state of decomposition when discovered but because of the remote locations of the carcasses, they could have been on the beach some time before being discovered. Neither had characteristics of animals that had been floating dead for some time before washing up. Humpback whales have been seen alive in many parts of coastal South Australia, as well as up to 250 km offshore in the Great Australian Bight and up to 100 km offshore in the south-east of the State (Fig. 2). TABLE |. Specimen records of Megaptera novaeangliae in South Australia. Date is when first seen or reported on beach or in case of unknown age of material, when registered into collection. No. Date Locality Length Sex Relative Decomposition Material at SAM (m) Age M149 __ before Largs Bay, Adelaide — — a unknown right and left bullae 1913 34° 49’ S, 138° 29’ E (provenance unknown) M5120 _ before ‘west coast of SA’ a — —o unknown radius, ulna, humerus, 1944 scapula (provenance unknown) S12 8 Aug. 14 km E Vivonne, ~10 U juvenile — advanced, none 1984 Kangaroo Island much skin 36° 00' 00" S, 137° 19’ 30” E sloughed off M12778 24 Oct. 12kmSW 5.53 —, neonate — quite almost full skeleton 1985 Streaky Bay, decomposed Great Australian Bight 32° 52’ 10” S, 134° 07' 00" E MI15187_ 1 Mar. 3 km N Point Yorke, 7.25 male — juvenile intact but almost full skeleton, 1989 Investigator Strait starting to baleen, barnacles 35° 12’ S, 137° 12' E decompose M16971 late Dec. 12.5 km SSE Ceduna, 7.10 as juvenile — very skull collected, 199] Great Australian Bight decomposed barnacles 32° 17'S, 133° 39’ E M21309 27 July 7.4 km NW Elliston, 3.88 female neonate very fresh full skeleton, baleen, 2000 Great Australian Bight organs, genetics tissues, 33° 36’ 00” S, 134° 50 20" E toxicology tissues HUMPBACK WHALES IN SOUTH AUSTRALIA 55 Many sightings were made from land—some being seen as close as 50 m from shore. Concentrations in sightings are apparent for eastern Gulf St Vincent, Kangaroo Island and Victor Harbor regions, and to a lesser extent off the lower Eyre Peninsula and in the south-east of the State. Humpback whales have been seen well within the gulfs, including a reliable record north-east of Whyalla and several north of Adelaide. When whale movements were ‘tracked’ by successive observations south of Adelaide, the pattern was northbound. Local residents reported Fig. 1. Female Megaptera novaeangliae neonate (M21309) that humpback whales were often sighted off washed up at Anxious Bay, 7.4 km NW Elliston, SA on 27. _ Elliston, generally moving north (Marie Clark, pers. July 2000. Photo: C. Kemper/South Australian Museum. comm.). | | | | | | Z ; South Australia | | | | ! | | gg Head of Bight | a — | ms Fowder Bay | saa: ata | . 7 an, | aa Wrserg iN | he i 2 | E P I : vA “ee yre Peninsula aS %, es. Cag Elliston £7 as | ~ r Pon ¢ 4 Mf gt ios era Southern Ocean a oF | ~ | emia | « ? t a | . ena Th | j < ' N a vhs Les | 3 - 0 70 140 280 ae Nautical Miles ae y A Fig. 2. Geographic distribution of Megaptera novaeangliae in South Australia. Closed triangles = sightings reliability | and 2, closed circles = sightings reliability 3 and 4, open squares = South Australian Museum specimens, * = towns/cities mentioned in text. nA an Ww ow oun 25 Mild 3 and4 Old tand 2 No. records nN o J oF MAM J J Month A S ON D Fig. 3. Month of observation for sightings and carcasses of Megaptera novaeangliae in South Australia. Id = reliability of identification (see Material and Methods). *= one carcass record, Fig. 4. Megaptera novaeangliae trapped in tuna feedlot off Boston Island, Spencer Gulf (34° 44" S, 135° 55’ E) on 14 June 1993. Photo: Daryl Lawrence/NPWSA. Only five of the sightings and no intact carcasses were recorded before 1980, Between 1980 and 1989, there were 17 sightings and three carcasses. The remaining records (79%, n = 87) dated from 1990 to 2003. Sightings of live animals were made during all months of the year and carcasses were recorded in March, July, August, October and December (Fig. 3, Table 1). Sixty (57%) of the sightings were made during June and July. There did not appear to be a relationship between region and month sighted— both eastern and western parts of the State had humpback sightings in all seasons. Accurate data on body length and relative age came from carcasses collected for the South Australian Museum (Table 1). Of the five measured specimens, two were neonatal in length (5.53, 3.88 m) and three were juveniles (7.10, 7.25, ~10 m). Estimates of body length of 17 live animals (5-15 m) were available, mostly from boat-based observers. (Since boating enthusiasts are usually familiar with the length of their vessel and can compare a whale with this, these estimates are C. M. KEMPER considered reasonably accurate.) The length of a live whale entrapped in a tuna feedlot near Port Lincoln was estimated (10 m) by the author by measuring fluke width from photographs submitted (Fig. 4). A number of observers reported ‘adult and calf’, ‘subadult’ or ‘large’ and ‘medium’-sized animals travelling together but with no supporting estimation of body length or how the relative size was determined. Humpback whales have been observed interacting with southern right whales at Head of Bight during winter, one such interaction lasting for >75 min (S, Burnell and R. Pirzl, pers. comm.). Humpback whales were seen during five out of seven years of southern right whale studies at Head of Bight. The number of humpback whales observed during sighting events in South Australia often accompanied reports (n = 98). The results are summarised as follows: | animal = 44 times, 2 = 44, =8,4=2. Discussion The number of records of humpback whales for South Australia is ample evidence that this species is more than a vagrant to the south-central coast of the continent. In addition, sightings are made from year to year in some places, e.g. Head of Bight. The documenting of 19th century and recent records is evidence that humpback whales have been present off South Australia for more than 160 years. The American whale ship, Amazon, took eight humpback whales and 33 southern right whales at Fowler Bay during an 80-day period in the winter of 1840 (Bannister 1986), An excavation of the site in 1994 identified many bones of southern right whales but no humpback whales, although not all material was identifiable (Kemper & Samson 1999), Townsend (1935) compiled logbook records of American whale ships from the 19th century and noted only two humpback records for the waters off southern Australia, both near Albany, Western Australia in July. With Australian populations of humpback whales increasing since at least the early 1980s (Bannister & Hedley 2001; Paterson er al. 2001), it is expected that sightings off South Australia will become more frequent. The coast of South Australia spans the longitudes 129° to 141° E, which falls within the limits of the Group V_ population. However, the affinity of humpback whales in that State to Group IV and/or V populations is not known. Between the 1930s and 1960s research on great whale movements was based largely on Discovery marking of whales in the Antarctic and Oceania, and off Australia and New Zealand (Dawbin 1966). This technique of studying the movements of whales involved firing numbered HUMPBACK WHALES IN SOUTH AUSTRALIA 57 markers into live whales then these markers were often recovered during whaling operations. Since there were no 20th century whaling stations along the coast of South Australia (Findlay 2001), no marks were recovered there. If members of both humpback whale populations occur off South Australia they may remain discrete (possibly Group IV in the Great Australian Bight and Group V off the south-east of the State) or there may be no geographic separation. Dawbin (1966) reported limited mixing of these populations in their Antarctic feeding grounds. Humpback whales occasionally calve south of 28° S on the east coast of Australia (Janetski & Paterson 2001) but there are no known current calving grounds off southern Australia (Bannister ef al, 1996). Three ‘cows and calves’ were taken by the Amazon during June to September 1840. Since no body lengths were recorded, the ‘calves’ may have been either neonates (possibly born prematurely on the northward migration) or young of the previous year. Humpback whales breed in tropical waters of at least 25°C (Dawbin 1966) and the waters off South Australia are temperate (i.e. 10—20°C). It is theretore unlikely that this is an undescribed calving ground. In the Australasian region humpback whale migration is segregated by age, gender and reproductive status (Chittleborough 1965; Dawbin 1966; Bannister ef a/. 1996). On the northward migration (mostly June to August but some as late as October (Paterson ef al. 2001)) females with yearling calves travel first, followed by adult males and non-pregnant females, then lastly females in late pregnancy. On the southward migration, which occurs from August/September to November/ December (Paterson ef a/. 2001) females in early pregnancy travel first, followed by immature animals, then resting females with mature males and lastly cows with young calves (Dawbin 1966). There is evidence that not all females migrate north to the breeding grounds each year (Brown ef al. 1995). Thus it could be hypothesised that some of the South Australian sightings are these females. Chittleborough (1965) noted that some Group IV humpback whales meet the south coast of Western Australia on their northward migration, then travel west/south-west to continue north along the west coast of the continent. This observation, and the fact that the majority of South Australian sightings were in June and July, suggests that the latter are part of the northward migration. The presence of dead neonates in July and October is presumably linked to the last of the northbound migration, i.e. late- pregnancy females. Humpback whales are weaned at lengths of 7.5 to 9 m when less than 12 months old (Bannister et al. 1996). Of the three dead juveniles recorded in South Australia, one was 7.2 m (March) and the other 7.1 m (December). The March juvenile may have been migrating north with its mother. The timing and size of the December juvenile suggests a southbound migration as either a dependent calf or weanling. The August carcass of an animal estimated at 10 m body length was probably weaned and may have been on a northward migration. A small proportion of humpback whales feed opportunistically off Australia on their migrations north and south. Chittleborough (1965) found only five out of 197 humpbacks sampled off the Western Australian coast had food (small quantities) in their stomachs. The species has been observed feeding on the euphaustid crustacean, Nyctiphanes australis, off eastern Tasmania and on an unknown prey off southern New South Wales (Gill ef al. 1998). Pygmy blue whales have been seen feeding on N. australis during summer and autumn in the Bonney upwelling off western Victoria and south-eastern South Australia but as yet no humpback whales have been seen there (Gill 2002). A humpback whale was observed on 16 July 1993 tn Discovery Bay, Victoria, very close to the South Australian border during surveys for southern right whales (Kemper, unpublished data) but this is not when upwelling events occur. Future research on humpback whales in South Australia should include |) determining affinities to existing Australian populations by genetic studies and photo-identification (some tissues and photographs presently available at the South Australian Museum), 2) estimating relative abundance and mapping offshore distribution using platform of opportunity surveys and 3) determining sex and relative age structure through close study and biopsy of live animals. Acknowledgements Research involving opportunistic sightings of whales relies on researchers, members of the public and government officers to submit records. I thank all those who have contributed, especially Stephen Burnell and Rebecca Pirzl, Trevor Whibley, officers of National Parks and Wildlife South Australia and the South Australian Whale Centre. Sightings accumulated by John Ling and Pin Needham formed the basis of the whale records and BHP Ltd. financed their databasing at the South Australian Museum. Many people are thanked for their work on that database — Jenny Mole, Helen Owens, Tanya Taylor, Claire Taylor, Michaela Ciaglia and Tami Stone. | am grateful to the following people for assistance with collecting humpback whale carcasses: Lynette Queale, Catherine Bell, Ron 58 C. M. KEMPER Waterhouse, Ross Allen, Don Mount, Tom Gerschwitz, David Farlam, Trevor Whibley. Museum staff members (Jim McNamara, Terry Sim, Lynette Queale, Peter Cockerham, Bob Hamilton- Bruce) have played an important role in preparing and managing the cetacean collection. John Bannister provided helpful comments on a draft manuscript. References AITKEN, P. F. (1971) Whales from the coast of South Australia. Trans. R. Soc. S. Aust. 95(2), 95-103. BAKER, C. S., SLADE, R. W., BANNISTER, J. L., ABERNETHY, R. B., WEINRICH, M. T., LIEN, J., URBAN, J., CORKERON, P., CALMABOKIDIS, J., VASQUEZ, O., & PALUMBI, S. R. (1994) Hierarchical structure of mitochondrial DNA gene flow among humpback whales Megaptera novaeangliae, world-wide. Mol. Ecol. 3, 313-327. BANNISTER, J. L. (1986) Notes on nineteenth century catches of southern right whales (Eubalaena australis) off the southern coast of Western Australia. Rep. Int. Whal. Commn (Special Issue) 10, 255-259. & HEDLEY, S. L. (2001) Southern hemisphere Group IV humpback whales: their status from recent aerial surveys. Mem. Od Mus. 47(2), 587-598. , Kemper, C. M. & WARNEKE, R. M. (1996) “The Action Plan for Australian Cetaceans”. (Australian Nature Conservation Agency, Canberra). Brown, M. R., CoRKERON, P. J., HALE, P. T., SCHULTZ, K. W. & Bryben, M. M. (1995) Evidence for a sex-segregated migration in the humpback whale (Megaptera novaeangliae). Proc. R. Soc. Lond. Series B 259: 229- 234 Cato, D. H. (1991) Songs of humpback whales: the Australian perspective. Mem. Od Mus. 30(2), 277-290. CHITTLEBOROUGH, R. G. (1965) Dynamics of two populations of the humpback whale, Megaptera novaeangliae (Borowski). Aust. J. mar. and Freshwat. Res. 16, 33-128. CLAPHAM, P. J. (2002) Humpback whale pp. 589-592 /n Perrin, W. F, Wursig, B. & Thewissen, J. G. M. (Eds). “Encyclopedia of Marine Mammals” (Academic Press, San Diego). DawBIN, W. H. (1966) The seasonal migratory cycle of humpback whales pp. 145-170 Jn Norris, K. (Ed) “Whales, Dolphins, and Porpoises” (University of California Press, Berkeley). & Eyre, E. J. (1991) Humpback whale songs along the coast of Western Australia and some comparison with east coast songs. Mem. Od Mus. 30(2), 249-254. FINDLAY, K. P. (2001) A review of humpback whale catches by modern whaling operations in the southern hemisphere. Mem. Od Mus. 47(2), 411-420. GILL, P. C. (2002) A blue whale (Balaenoptera musculus) feeding ground in a southern Australian coastal upwelling zone. J. Cetacean Res. Manage. 4(2), 179-184. , EVANS, K. J. & WapsTRA, J. (1998) Feeding by humpback whales in Tasmanian waters. Rec. Queen Vict. Mus. 107, 1-5. JANETSKI, H. A. & PATERSON, R. A. (2001) Aspects of humpback whale, Megapera novaeangliae, calf mortality in Queensland. Mem. Od Mus. 47(2), 431-435. Jupp, M., KEmMPER, C., LING, J. & OLMAN, J. (1992) “A Guide to Whales and Whale Watching in South Australia” (South Australian Museum, Adelaide). Kemper , C. M. & Lina, J. K. (1991) Whale strandings in South Australia (1881-1989). Trans. R. Soc. S. Aust. 115(1), 37-52. & SAMSON, C. R. (1999) Southern right whale remains from 19th century whaling at Fowler Bay, South Australia. Rec. S. Aust. Mus. 32(2), 155-172. MackintosH, N. A. (1965) “The Stocks of Whales” (Fishing News (Books) Ltd, London). MIKHALEV, Y. A. (2000) Biological characteristics of humpbacks taken in Antarctic Area V by the whaling fleets Slava and Sovietskaya Ukraina. \n Rep. Int. Whal. Commn pp. SC/52/1A12_ (International Whaling Commission, Cambridge). PATERSON, R. A. (2001) Exploitation of humpback whales, Megaptera novaeangliae, in the South West Pacific and adjacent Antarctic waters during the 19th and 20th Centuries. Mem. Od Mus. 47(2), 421-429. , PATERSON, P. & CATo, D. H. (2001) Status of humpback whales, Megaptera novaeangliae, in east Australia at the end of the 20th Century. Mem. Od Mus. 47(2), 579-586. Tormosovy, D. D. (1995) Humpback whale catches by area and sex in the Antarctic taken by the Yurii Dolgorukiy. Rep. Int. Whal. Commn 45, 141. TOWNSEND, C. H. (1935) The distribution of certain whales as shown by logbook records of American whaleships. Zoologica, N. Y. 19, 1-50. Transactions of the Roval Society of S. Aust. (2005), 129(1), 59-64. A REDESRIPTION OF ODILIA EMANUELAE (NEMATODA: TRICHOSTRONGYLINA: HELIGMONELLIDAE) FROM AUSTRALIAN RODENTS WITH A KEY AND COMMENTS ON THE GENUS ODILIA by L. R. SMALES* Summary Smaces, L. R. (2005) A redescription of Odilia emanuelae (Nematoda: Trichostrongylina: Heligmonellidae) from Australian rodents with a key and comments on the genus Odilia. Trans. R. Soc. S. Aust. 129(1), 59-64, 31 May, 2005. Odilia emanuelae (Mawson, 1961) is redescribed from a new host, Raffus leucopus from Queensland Australia. A key to the genus Odilia, based on the number and size of the ridges of the synlophe is given. The relationships of species of the genus Odilia with their murid hosts, coevolution or host switching, are discussed and speculation as to the significance of their known biogeographic distribution is put forward. Key Worps: Nematoda, Heligmonellidae, murid, Ra/tus, Australia, phylogeny, biogeography. Introduction During a study of the helminth parasites of the Cape York rat, Rattus leucopus (Gray), specimens of three species of trichostrongyloid nematodes, Heligmonellidae, were encountered in the duodenum of several hosts. They included Nippostrongylus brasiliensis (Travassos, 1914), a cosmopolitan species and NV. magnus (Mawson, 1961), endemic to Australian rodents. Previously noted from Rattus fuscipes (Waterhouse), R. sordidus (Gould) and Melomys cervinipes (Gould) (see Smales, 1997), WN. magnus was redescribed by Beveridge & Durette- Desset (1992) from R. fuscipes and experimentally infected Rattus norvegicus (Berkenhout). The third species found was Odilia emanuelae (Mawson, 1961). Odilia emanuelae was described originally from R. sordidus as R. conatus and R. fuscipes as R. assimilis by Mawson (1961) but she did not describe some features including the cuticular ridges of the synlophe. Subsequently a brief description of the synlophe was given by Durette-Desset (1969) on the basis of one specimen. The additional material from a new host allows a more detailed description to be prepared and a key to known species of the genus given. The genus QOdilia has been reported from 11 endemic species of murids from Australia: in the subfamilies Hydromyinae, Mastacomys fuscus Thomas, Melomys burtoni (Ramsay), Melomys cervinipes, Mesembryomys gouldii (Gray), Pseudomys higginsi _(Trouessart), — Uromiys caudimaculatus (Krefft), Zvzomys argurus (Thomas) and Zyzomys woodwardi (Thomas); and Murinae, *School of Biological and Environmental Sciences, Central Queensland University, Rockhampton, Queensland 4700. Australia. Email: Lwarner@cqu.edu.au Rattus fuscipes, R. lutreolus (Gray) and R. sordidus (see Smales, 1997). Species of Odilia have also been found in the hydromyine Mallomys rothschildi Thomas from Irian Jaya, now Papua, Indonesia and the murines Rattus xanthurus Gray and Maxomys musschenbreokii Jentink from Sulawesi, Indonesia (Hasegawa & Sayaffrudin, 1994; 1995 ; Hasegawa et al., 1999), Rattus leucopus, the Cape York rat, and R. sordidus, the canefield rat, are both found in northern Queensland and the island of New Guinea, the only two endemic murines that are found on both sides of Torres Strait (Flannery, 1995). This distribution provides evidence of recent past land bridges, probably in the Pleistocene, between the two land masses (Moore & Leung, 1995). Observations on the host range of the species and the significance of the geographical distribution of the genus Odilia, given the geographical distributions of the hosts are presented. Materials And Methods The specimens from R. leucopus were fixed in 10% formalin, stored in 70% ethanol and examined in lactophenol. En face preparations and transverse sections were cut by hand using a cataract scalpel and mounted in polyvinyl — lactophenol. Measurements of 10 males and 10 females from Rattus leucopus were taken using an ocular micrometer and given as the range followed by the mean in parentheses, in micrometres unless otherwise stated. Drawings were made with the aid of an Olympus BH Nomarski interference contrast microscope and drawing tube. Specimens are held in the CSIRO Wildlife collection, Canberra (CSIRO) and the South Australian Museum, Adelaide (SAM AHC). Terminology and classification used follows 60 L. R. SMALES Figs. 1-14. Odilia emanuelae (Mawson, 1961). Female head, en face view. 2. Female, anterior body section through oesophageal region, arrow indicating axis of orientation of synlophe. 3. Male, mid body section. 4. Female, anterior end, lateral view showing origins of ridges of synlophe. 5. Female, mid body section. 6. Male, section within posterior third of body 7. Female, anterior end, lateral view. 8. Female, section within posterior quarter of body. 9. Genital cone, right ventral view. 10. Bursa, dorsal view, flattened. 11. Spicule tips, lateral view. 12. Genital cone, right lateral view. 13. Female, posterior end left lateral view. 14. Male, posterior end, right lateral view. Abbreviations d, dorsal; |, left; r, right; v, ventral. Scale bars: 1, 9, 11, 12, 10 um; 2, 3, 5, 6, 8, 20um; 4, 10, 25um; 7, 13, 14, 50m. A REDESRIPTION OF ODILIA EMANUELAE 61 Durette-Desset (1971, 1973, 1983, 1985), Beveridge & Durette-Desset (1992) and Durette-Desset ef al. (1994). Rodent classification follows Strahan (1995). Odilia emanuelae (Mawson,1961) (Figs 1-14) Heligmonoides emanuelae Mawson, 1961, pp 809- 810, figs 30-34 table 4, from Rattus conatus (syn R. sordidus ) and R, assimilis (syn R. fuscipes); Durette- Desset (1969) p 738, fig 4C. Austrostrongylus emanuelae: (Mawson,1961) Durette-Desset 1971 p 65. Odilia emanuelae: : (Mawson,1961) Durette- Desset 1973 p 517; Smales 1992 p 75. Material examined From Ratius sordidus: holotype male, allotype female, Innisfail, (17° 32’ S, 146° O1’ E) Queens- land, SAM AHC 41332: from Rattus leucopus: 55 males, 81 females, East Mc Illwraith Range, Cape York Peninsula (13° 45’ S, 143° 20’ E), Queensland coll. P. Catling, I. Mason and P. Haycock, 9. xiii. 1990, 10. iti. 1990, CSIRO N3293, N3296, N3324, N3325, N3326, N3329: from Melomys cervinipes (Gould) 10 males, 10 females, D’Aguillar Range (27° 50’ S, 152° 45’ E), Queensland, 19. ii. 1963, SAM AHC 5805, coll. Aland and Stewart, 26.viii. 1993, SAM AHC 32190, 32191, 32192. Comparison of the measurements of specimens from Rattus sordidus, type host, and R. leucopus are given in Table 1. Redescription Small coiled nematodes; prominent cephalic vesicle present; buccal capsule vestigial. Mouth opening triangular with rudimentary lips, surrounded by four double papillae, each comprising a cephalic plus externo-labial papilla and two lateral amphids. Internal labial papillae not visible. Oesophagus claviform; nerve ring surrounds oesophagus at about mid level; excretory pore and digitiform deirids at same level, posterior to nerve ring. Synlophe: Longitudinal cuticular ridges continuous, extend from posterior margin of cephalic vesicle to just anterior to bursa or vulva; 17 in anterior, 18 in mid body; axis of orientation from right ventral to left dorsal at approximately 75° to frontal axis; 7-8 in dorsal side, 9-10 in ventral side; ridges | and 1’ largest, forming typical type A carene, ridges 2, 3 smaller than ridges 1, ridges 4-6 increasing in size, ridges 7-10 decreasing in size. Posterior region of body with 15 (male), 17 (female) ridges reduced in size; dorsal side with 7-8; ventral side with 7-10 ridges. Male Length 1.3-1.64 (1.50) mm, maximum width 54-67 (60). Cephalic vesicle 42-56 (50.6) long. Oesophagus 300-420 (345) long; excretory pore 231, 340 from anterior end. Bursa asymmetrical, right lobe larger (rays of right lobe more robust); deep dorsal cleft. Dorsal ray symmetrical divided at about half its length, each branch dividing again at distal tip; terminal divisions, rays 9, 10 symmetrical; rays 8 arising at same level, right ray 8 more robust than left. TABLE 1. Measurements of Odilia emanuelae, in {um unless otherwise stated, from two host species; 10 males and 10 females from each. Data for Rattus fiscipes are from Mawson (1961). Rattus leucopus Rattus fuscipes Locality East Mclllwraith Range Innisfail Male Length, mm 1.3-1.64 2.2-2.6 Width 54-67 - Cephalic vesicle 42-56 50-60 Oesophagus length 300-420 240-270 Ant. end to excretory pore 230, 340 190-220 nerve ring - - deirid - 190-220 Spicules 245-270 250-330 Gubernaculum 17-22 - Female Length, mm 1.8-2.6 2.1-3.1 Width 63-74 Cephalic vesicle 50-56 50-60 Oesophagus length 240-410 270-290 Ant. end to excretory pore - 200-222 nerve ring - - deirid - 200-222 Tail 30-43 40-50 Eggs 50-63 x 27-34 60-80 x 40-50 Vulva to tail tip 87-119 105-130 62 L. R. SMALES Rays 4, 5, 6 with common stem, reaching margin of bursa; rays 4 and 5 robust curving anteriorly, rays 6 slender, curving posteriorly. Rays 2 and 3 with common stem, robust, diverge distally, curve posteriorly, reaching margin of bursa. Genital cone short, ventral lobe with unpaired papilla 0, lightly sclerotized; dorsal lip bifid, each lobe with single papilla 7. Spicules equal, filiform, tips pointed, 245- 270 (253) long. Gubernaculum 17-22 (19.5) long. Female Length 1.8-2.6 (2.1) mm; maximum width 63-74 (67). Cephalic vesicle 49.5-56 (51.5) long; oesophagus 240-410 (320) long. Vulva opens 87-119 (94.5) from tail tip; posterior end may or may not be flexed at right angles just behind vulva. Monodelphic, ovejector with sphincter 30, 35, shorter than vestibule 50, 60, infundibulum, about same length as sphincter. Tail 29.5-43 (35) long. Eggs in utero 49.5-63 (56.5) by 26.5-34 (30). Key to species of the genus Odilia 1. Synlophe with discontinuous ventral ridges........ Phat B hort dba O. mackerassae (Mawson, 1961) Synlophe with continuous ridges..........:0cee 2 2. Synlophe with 18 or more ridges in mid body ..3 Synlophe with fewer than 18 ridges in mid body PETTUS TRY Nt cin vir sneer 20 Canes crerceerrns 10 3. Synlophe with 18 ridges in mid body................ 4 Synlophe with more than 18 ridges in mid body Heegesesaaesuei as atin bob pasa peated ese repeqedtnontuned shies 5 4, Synlophe with fewer than 18 ridges posteriorly, 12-15 in males. Gubernaculum 17-22 long; spicule tips taper to sharp point. Female tail conical, rounded tip ......... eee O. emanuelae (Mawson, 1961) Synlophe with more than 18 ridges posteriorly, 24 in males. Gubernaculum 30-40 long; spicule tips joined distally surrounded by transparent membrane. Female tail tapers sharply from vulva to pointed tp... eee ee eee O. tasmaniensis Gibbons & Spratt, 1995 5. Synlophe with 19-20 ridges (male) in mid body, ridges becoming tiny posteriorly. Gubernaculum 19-22 long; spicule: body length 1: 6.5. Female tail 30; eggs 59-78 x 29-42........ O. mamasaensis Hasegawa, Miyata & Syafruddin, 1999 Synlophe with more than 20 ridges (male) in mid BODY ccacsegeiy lst beg cecbaa crc ne tenn btn teesestaten 6 6. Synlophe with 21 ridges (male) in mid body. Gubernaculum 28 long; spicule: body length 1: 8 pe caper eae O. mawsonae (Durette-Desset, 1969) Synlophe with more than 21 ridges in mid body 7. Synlophe with up to 35 ridges in mid body ...... 8 Synlophe with more than 35 ridges in mid body 8. Synlophe with 24-25 (male), 24-28 (female) ridges in mid body, 23 (male) 13 (female) ridges becoming minute posteriorly. Gubernaculum 32- 37; spicule: body length 1:4. Female tail conical, pointed tip; eggs 72-80 x 35-43 .....O. maxomyos Hasegawa, Miyata & Syafruddin, 1999 Synlophe with 22-29 ridges (male), 24-35 ridges (female) in mid body, 29 (male), 26 (female) ridges posteriorly. Gubernaculum absent; spicule: body length: 1:13. Female tail with prepuce; eggs 60-70 X AO bgcetescetetsrnenennrcetaneinne O. praeputialis Gibbons & Spratt, 1995 9. Synlophe with many (male), 36 (female), even sized ridges in mid body. Gubernaculum 20 long; spicule: body length ratio 1:8. Female tail twisted into 1-2 coils in front of vulva... eee Jumepnaneatedarreaeeoel O. polyrhabdote (Mawson,1961) Synlophe with 40 (male), 48 (female) ridges in mid body. Spicule tips with hair like projection 60 from distal end supporting fan-like alae; spicule: body length 1:9, Female tail conical 60- 70 long with prepuce .......... cece O. uromyos (Mawson, 1961) 10. Synlophe with 17 ridges in mid body, 20 (male), 19 (female) posteriorly. Gubernaculum 50 long; spicule: body length 1:14. Eggs 70-80 x 40-50 3 O. bainae Beveridge & Durette-Desset, 1992 Synlophe with fewer than 17 ridges in mid body SAO OT NE ALT PLEO EITT fi aey ey reer OTEET ERED 11 11. Synlophe with 16 ridges (male) in mid body, 16 ridges smaller posteriorly. Spicule: body length 1:16. Female tail 50 long; eggs 69-77 x 35-45 ... cates O. mallomyos Hasegawa & Syafruddin, 1994 Synlophe not as above ........cc cc eeeeeeeeereeeees 12 12. Synlophe with 15 (male), 16 (female) ridges in PAA DOH aged betaatza rsd bacecactesatdossscanteneteerenss 13 Synlophe with less than 15 (male), 16 (female) ridges in Mid DOY ....... eee eeeeeeer eee rreressesenees 14 13. Synlophe with 16-20 minute ridges posteriorly. Gubernaculum 32-43 long; spicule: body length 1:7. Female cuticle inflated proximally to tail..... tas O. moatensis Hasegawa & Syafruddin, 1999 Synlophe with 30 (male), 50 (female) small even ridges posteriorly. Gubernaculum 51-62 long; spicule: body length 1:8. Female cuticle not inflated proximally to tail.......... O. sulawesiensis Hasegawa & Syafruddin, 1999 14. Synlophe with 14 (male), [5 (female) ridges in mid body, 16 (female), smaller even ridges posteriorly. Gubernaculum 30 long; spicule tips pointed; spicule: body length 1: 9-1: 11. Female tail conical, rounded tip 40-50 long wu... be ates actin hs ony eg O. melomyos (Mawson,1961) Synlophe with 14 ridges (male) in mid body. Gubernaculum 25 long; spicule tips expanding, bifid; spicule: body length 1:9. Female tail conical, pointed tip, flexed sharply back on itself Babine dete coadahs deg ME O. brachybursa (Mawson,1961) A REDESRIPTION OF ODILIA EMANUELAE 63 Although Hasegawa & Sayafruddin (1995) noted Odilia sp. 1 and Odilia sp. 2 from Rattus cf. morotaiensis — from Indonesia, insufficient morphological data were provided to allow inclusion of these two species in the key. Discussion The present study of O. emanuelae revealed slight variations in the morphology of the species as compared with the description of Mawson (1961), particularly in the range of measurements, with specimens from R. leucopus smaller than those from R. sordidus. The spicule tips were described as widened and alate by Mawson (1961). Gibbons & Spratt (1995), however, commented that they broadened then tapered to a sharp tip. Examination of the type male, as well as the specimens from 2. leucopus in this study confirmed this latter form. The ridges of the synlophe were counted by Mawson (1961) as up to 20 in the mid body but given and figured as 18 by Durette-Desset (1969) for a female worm from R. sordidus, as is the case for specimens from R. leucopus. These minor morphometric differences may be due to host induced variation and are not sufficient to establish a separate species. Consequently the material from R. leucopus is assigned to O. emanuelae. The host range is accordingly expanded to include a third endemic Rattus species. The only other host records for O. emanuelae are from M. cervinipes, specimens from a single host deposited in the SAM and specimens collected during an unpublished survey of the helminths of M. cervinipes and R. fuscipes from the D’ Aguilar Ranges, south east Queensland (Aland!) in which two of 12 M. cervinipes were reported as being infected with O. emanuelae. Re-examination of this material revealed specimens of O. emanuelae in a third host, making a total of 4 infected M. cervinipes, all from the D’Aguilar Ranges. Odilia emanuelae has not been reported in other surveys of the helminths from melomys, such as that of Mawson (1961) although she examined hosts from Innisfail, the type locality (Mawson, 1961; Smales, 1997). This suggests that Raffus species are the normal hosts and that infections found in M. cervinipes in ' Aland, K. (1993) BSc Hons Thesis, Dept of Parasitology, University of Queensland. this study are an example of an occasional infection occurring where normal and alternative hosts are sympatric. The geographic distribution of O. emanuelae has been extended further north from the type locality into Cape York and south to south east Queensland. Previous interpretations of the origins of the Trichostrongylina in Australian rodents have presumed that the genus Odilia arose in Australia, co-evolving with the rodent sub family Hydromyinae (see Durette-Desset, 1985); that is with the earliest of the rodent invaders commonly known as the old endemics. The rodents are thought to have arrived in Australia some 5-10 million years ago (Watts & Aslin, 1981; Flannery, 1995). Odilia was then captured by more recent rodent arrivals, the Murinae new endemic Rattus species that crossed to Australia from New Guinea less than one million years ago (Smales, 1992; Beveridge & Durette-Desset, 1992; Gibbons & Spratt, 1995), More recently, however Odilia species have been described from several Indonesian islands and occurring in both old endemic and new endemic hosts (Hasegawa & Syafruddin 1994; 1995; Hasegawa ef al., 1999). At the same time, new fossil evidence from Australia suggests more complex evolutionary processes than had first been thought. There is now evidence for at least three phases of immigration, both direct from Southeast Asia, and through New Guinea, involving both old and new endemics (Godthelp, 2001). This new evidence suggests that ancestral forms of Odilia may have co-evolved with rodent hosts in Southeast Asia. The present host and geographic distribution therefore reflects a complex series of evolutionary events involving host switching and co- evolution as rodent faunas and their helminth communities migrate, and undergo evolutionary radiations, The more data gathered about rodent hosts and their parasites from Southeast Asia and Australasia the more complex the patterns of their relationships become. Acknowledgements I am grateful to Dr D. Spratt for making the material available. 64 L. R. SMALES References BeveERIDGE, I. & DurRETTE-DEsseT, M.-C. (1992) A new species of trichostrongyloid nematode, Odilia bainae, from a native rodent, Rattus fuscipes (Waterhouse). Trans. R. Soc. S. Aust. 116, 123-128. DuretTtE-Dessert, M.-C. (1969) Les systemes d’arétes cuticulaires chez les Nematodes H¢éligmosomes parasites de muridés Australiens. Ann. Parasitol. Hum. Comp. 44, 733-747. (1971) Essai de classification des Nématodes Héligmosomes. Corrélations avec la paléobiogéographie des hétes. Mém. Mus. natn. Hist. nat. Série A, Zoologie, 69, 1-126. (1973) Note réctificative sur le genre Austrostrongylus (Nématode). Ann. Parasitol. Hum. Comp. 48, 517. (1983) Keys to the genera of the superfamily Trichostrongyloidea, No. 10 pp.1-86 /n Anderson R. C. & Chabaud A. G. (Eds), “CIH Keys To The Nematode Parasites Of Vertebrates ” (Commonwealth Agricultural Bureaux, Farnham Royal). (1985) Trichostrongyloid nematodes and their vertebrate hosts: reconstruction of the phylogeny of a parasitic group. Adv. Parasitol. 24, 239-306. , BeveripGE, I. & Spratt, D. M. (1994) The origins and evolutionary expansion of the Strongylida (Nematoda). Int. J. Parasitol. 24, 1139-1165. FLANNERY, T. (1995) “Mammals of New Guinea” (Reed Books, Chatswood). Gipsons, L. M. & Spratt, D. M. (1995) Two new species of Odilia (Nematoda: Heligmonellidae) from Australian rodents, with comments on O. bainae Beveridge & Durette-Desset 1992. Syst. Parasitol. 31, 67-79. GopTHELp, H. (2001) The Australian rodent fauna, flotilla’s flotsam or just fleet footed pp. 319-321 /n Metcalfe, 1. Smith, J.M.B. Morwood, M. & Davidson, I. (Eds) “Faunal and Floral Migrations and Evolution in S.E. Asia — Australasia” (A.A. Balkema, Lisse). Hasecawa, H. (1996) Notes on the morphology of three nematode species of the subfamily Nippostrongylinae (Heligmosomoidea: Heligmonellidae) collected from an endemic rat of Halmahela Island, Indonesia. Biol. Mag. Okinawa, 34, 13-21. & SAYAFRUDDIN, (1994) Odilia mallomyos sp.n. (Nematoda: Heligmonellidae) from Mallomys rothschildi weylandi (Rodentia: Muridae) of Irian Jaya, Indonesia. . Helm. Soc. Wash. 61, 208-214. (1995) Nippostrongylus marhaeniae sp. n. and other nematodes collected from Rattus cf morotaiensis in North Halmahera, Molucca Islands, Indonesia. /bid. 62, 111-116. , Miyata, A. & SYAFRUDDIN, (1999) Six new nematodes of the Heligmonellidae (Trichostrongylina) collected from endemic murines of Sulawesi, Indonesia. J. Parasitol. 85, 513-524. Mawson, P. M. (1961) Trichostrongyles from rodents in Queensland, with comments on the genus Longistriata (Nematoda: Heligmosomatidae). Aust. J. Zool. 9, 791- 826. Moorg, L. A. & LEUNG, L. (1995) Cape York Rat Rattus leucopus (Gray, 1867) pp. 635-655 Jn Strahan, R. (Ed.) “The Mammals of Australia” (Reed Books, Chatswood). SMat_es, L. R. (1992) A survey of the helminths of Rattus sordidus with a description of Ancistronema coronatum n.g., n.sp. (Nematoda: Chabertiidae). Syst. Parasitol. 22, 73-80. (1997) A review of the helminth parasites of Australian rodents. Aust. J. Zool. 45, 505-521. STRAHAN, R, (1995) “The Mammals of Australia” (Reed Books, Chatswood). watts, C. H. S. & ASLIN, H. J. (1984) The Rodents of Australia. (Angus & Robertson, London). Transactions of the Royal Society of S. Aust. (2005), 129(1), 65-73. THREE DECADES OF HABITAT CHANGE IN GULF ST VINCENT, SOUTH AUSTRALIA. by J. E. TANNER" Summary TANNER, J. E. (2005). Three decades of habitat change in Gulf St Vincent, South Australia. Trans. R. Soc. S. Aust. 129(1), 65-73, 31 May, 2005. Benthic habitats in Gulf St Vincent, South Australia, have changed considerably in the period between the 1960s and 2000/2001. Diver surveys in the 1960s indicated the presence of an extensive area of deep-water Heterozostera seagrass in Investigator Strait, and Ma/leus-Pinna bivalve assemblage in the south-eastern area of the gulf. Neither of these habitat types were detected in remote video and diver surveys conducted in 2000/2001, with these areas instead being dominated by relatively barren sand flats. In the central section of the gulf, the cover of bryozoans, and density of scallops, also declined over the three decades between the two surveys. Further north there have been fewer changes, with the original Pinna assemblages and seagrass meadows still present. It is thought that these changes are primarily due to anthropogenic influences, with the two major candidates being increased turbidity (due to terrestrial inputs from sewage, stormwater runoff, agricultural runoff and dust storms), and direct damage from prawn trawling. As a consequence of these changes, the habitat complexity in Gulf St Vincent has decreased substantially, with likely consequences for other fauna such as fish and mobile invertebrates. Introduction While the effects of anthropogenic disturbance on many terrestrial systems are obvious and well documented, much less is known about what changes are occurring in marine systems, especially in waters deeper than a few metres. It is often thought that marine systems are stable, and little affected by change on a broad-scale, despite some well documented examples of change due to specific disturbances at smaller scales (e.g. Trawling: Jennings & Kaiser 1998; Hall 1999; Pollution: Neverauskas 1987; Lapointe et al, 1994; Sea Level Change: Seddon ef a/. 2000). This attitude is probably related to our inability to easily see what is happening on the ocean floor, and to the lack of long- term studies and baseline data on what systems looked like before human disturbance (or even in its early stages). While there is good evidence of change in individual habitats in some marine systems, such as seagrass loss (e.g. Neverauskas 1987; Walker & McComb 1992; Short & Wyllie-Echeverria 1996; Edyvane 1999), and changes in kelp abundance in the eastern North Pacific related to changes in otter abundance (e.g. Estes & Duggins 1995), there are few well documented studies that examine change in entire marine ecosystems over a period of decades. Most of those studies that have been done rely on the existence of earlier studies that utilised remote sampling techniques (such as benthic grabs) to examine infauna, or trawl gear to examine catches of " SARDI Aquatic Sciences, PO Box 120, Henley Beach, SA 5022. Australia. Phone: +61 8 8207 5489. Fax: +61 8 8207 5481, Email: tanner,jason@saugoy,sa.gov.au. fish and other macrofauna (e.g. Haedrich & Barnes 1997; Wilson et al. 1998; Frid et al. 1999, 2000), and have not specifically examined changes in benthic habitats. Between 1964 and 1969, Shepherd & Sprigg (1976) conducted an extensive series of diver surveys of benthic habitats in Gulf St Vincent, South Australia. They observed the intact habitat in situ, and recorded all the major components of the flora and fauna, While these surveys were primarily qualitative, with few abundance estimates, a comprehensive map documenting the various community types in the gulf was _ published (reprinted here as Fig. 1). This map was used to compare the major habitat features and benthic assemblages present in the 1960s, to those present in 2000/2001, to detect any substantial changes in the intervening period, To do this an extensive series of remote video surveys was undertaken in 2000/2001, complemented by a number of spot dives to ensure that the information being obtained from the video was reliable. There have been no extensive benthic surveys carried out in Gulf St Vincent between these two studies, and thus we do not know if any intermediate states occurred, and it is only possible to speculate on the causes of any changes observed. Gulf St Vincent is a large (~ 13000 km? including its approaches), relatively shallow (maximum depth ~ 40 m), marine embayment located on the South Australian coast. It is an inverse estuary, with salinity ranging from ~37%. at the mouth to 41%. at the head, due to high evaporation rates and low precipitation (Bye 1976). The location of Kangaroo Island across the mouth of the gulf means that exchange of water with the open ocean is restricted, leading to long 66 J. E. TANNER 137°30'E 138°0'E 138°30'E Bare sand and shoal Heterozostera-Lunulites assemblage Algal debris (Calcarenite) mostly desert Ascidian-scallop assemblage Pinna-holothurian assemblage 34°30'S Bryozoan assemblage 34°30'S Malleus-Pinna assemblage Seagrass meadows. Mostly Posidonia Boulder conglomerates Reef (kunkerised shell beds) of low relief TARE Oi i i c Aeolianate reef 2000 survey point 35°0'S 35°0'S === ee ee ee et £ oS xxx x x xx KK xX XXXXXK cK MK XK OK RK OL KK XK J Kak x Wah ahah oh x% « & x OS MM RM MM RM MM MRK M KKK KRM KM KK KO! 35°30'S 35°30'S 137°30'E 138°0'E 138°30'E Fig. 1. Benthic community composition in Gulf St Vincent in the 1960s as determined by diver spot surveys (modified from Shepherd and Sprigg 1976). The survey sites for 2000/2001 are marked for comparison. THREE DECADES OF HABITAT CHANGE 67 residence times of water within the gulf (de Silva Samarasinghe & Lennon 1987). This geography also means that the area is a low energy environment, especially in the northern section, which as a consequence, is slowly being filled in by sediment deposition. Most of the substrate is either sand or fine silt, with only a few areas of hard bottom (Shepherd & Sprigg 1976, see Fig. 1), although there are substantial areas of calcrete underlying much of the sand, The city of Adelaide (population ~ | million), is located on the eastern shore of the gulf, and is a source of domestic and industrial pollution. The only other settlements in the area are several small towns with populations < 1000. There is considerable agricultural activity along the shores of the gulf, and in its catchment, which is a further source of pollution, However, due to the arid nature of the region (precipitation < 500 mm yr") there is little natural runoff. The gulf also supports substantial recreational and commercial fisheries, including a small (10 boat) prawn fishing fleet that targets the western king prawn (Melicertus latisulcatus). The prawn fishery commenced in 1968/69, and effort reached a peak of 15200 hours of trawling in 1982/83, and a peak catch of 602 t in 1976/77, before dramatic declines in both catch (to 200-400 t yr!) and effort (to ~4000 hours yr’) for the period 1995- 2002 (Svane & Johnson 2003). During the early phase of the fishery, the mid-northern portion of the gulf was heavily targeted (but south of 34° 30’ S), but since the mid 1980s to early 1990s effort has predominantly focussed on more southern areas where the prawns tend to be larger (Morgan 1995). Materials and Methods To quantify the distribution, abundance and composition of benthic habitats throughout Gulf St Vincent, an extensive series of remote video surveys was conducted between June 2000 and June 2001. Survey sites were located every 2 nm along east-west transects across the gulf which were spaced every 5 nm (18 transects between 34° 15’ S and 35° 40’ S). Transects extended from the eastern to the western edge of the gulf (although areas less than approximately 5 m deep were inaccessible to the vessel used and therefore excluded). The western extremity of the survey area was 137° 40’ E, and the easternmost point was 138° 30’ E. In total, data were obtained for 294 sites, with an additional three sites excluded from analysis because of poor image quality (see Fig. 2 for site locations). At each site, a digital video camera was lowered to within | m of the bottom, and left to record for 10 min while the boat drifted. To determine the linear distance moved during this time, a GPS was used to record the location (+10m) when a clear image of the bottom was first obtained, and again when it was lifted off the bottom. The mean distance covered per 10 min survey was 141+] m (se). Animal taxa visible in the video footage were enumerated, with total abundance standardised to the mean distance covered in a 10 min survey (141 m). Counts were only made for segments of the footage that were clear, and standardised counts were also adjusted for the proportion of the video that could not be used reliably. The exact area covered could not be calculated, as there was some variation in the height of the camera from the substrate, however, on average a swathe of ~ 2 m wide was surveyed. For those taxa for which individuals could not be distinguished (seagrasses and algae), as well as bare substrate, the video was stopped during playback approximately every 1 min, and percent cover recorded with the aid of a grid overlying the image. The mean percent cover for all recorded frames was then calculated for each site. To ensure that the video provided a reliable record of the benthos, spot dives were made at 53 sites to examine the benthos in more detail, and a series of photographs were taken of 0.25 m? quadrats for a later comparison with the video footage. These photographs were only qualitatively assessed, and did not show any major discrepancies with the video footage, so are not considered further. Dives were conducted on most of the defined habitat types, although depth constraints prevented stations > 24 m in depth from being surveyed in this way. To objectively determine the community type at each site, cluster analysis was used on the standardised data. As the objective was to produce a map that could be compared with that produced by Shepherd & Sprigg (1976), taxa were grouped at the same level as they used (e.g. sponges, bryozoans, scallops etc), and minor taxa were removed from the analysis. The clustering technique used was Wards Flexible B, with B = -0.25 (Seber 1984). This technique was chosen as it did not produce any chaining, unlike more commonly used methods such as group-average and centroid. Initially, 15 groups were chosen for further investigation, and these were manually merged on the basis of their dominant taxa to achieve similar groupings to those used by Shepherd & Sprigg (1976). Not all of the habitat types defined by Shepherd & Sprigg (1976) were present in the 2000/2001 survey, while several new intermediate habitat types were defined based on small clusters that did not readily fit in with any of the original types. It should be emphasised that the 2000/2001 survey sites did not correspond to the survey sites used in the 1960s, as information on the location of the later was not available. Thus, any comparisons over time are based on_ the interpolations made by Shepherd & Sprigg (1976) in producing their map of habitat distributions. 68 J. E. TANNER 137°30'E 138°0'E 138°30'E Legend Barren Sand Ascidian Ascidian/Bryozoan 7 34°30'S Bryozoan fe +XxX ofm 7 a =~ “) inna a+xcox@f snnas {\ oa. 2, eer 34°30'S Scallops a eee . Se, } | Seagrass/Pinna Z Wy ae a P+ + mot x eee Seagrass (| | Ao fan docxrxs ences ane / 3 Aomexr xxop+exkmeoend AW\| / maxx me ee eae ee \ 7, — Pie Le SES tSbe! Pe ereterren Sree rre rey /\, Ais eae 35°0'S 35°0'S xx xX XX XX KKK KKXKXXK XX 35°30'S ee ne het ( ee, eS —L 35°30'S a \ ‘ _ an mxxx xX XXX xXx Ve s'm XX XE EXX 2S is. 137°30'E 138°0'E 138°30'E Fig. 2. Benthic community composition in Gulf St Vincent in 2000/2001 as determined from remote video surveys. outlines of the habitats present in the 1960s are included for comparison. The THREE DECADES OF HABITAT CHANGE Results There have been some substantial changes in the epibenthos of Gulf St Vincent since the surveys of the mid to late 1960s by Shepherd & Sprigg (1976) (compare Fig. | with Fig. 2, Table 1). Particularly noticeable is the absence of the seagrass Heterozostera tasmanica, which covered extensive areas of deep sand plains in the southern gulf and Investigator Strait in the 1960s. While there was only a sparse cover of Heterozostera in the 1960s (S. Shepherd pers. com.), this area is now completely devoid of seagrass, with none being seen either in the remote video footage or on the spot dives. Also missing is the Malleus-Pinna (bivalve) assemblage that Shepherd & Sprigg (1976) documented in the south-eastern section of the gulf. While this area still contains some scattered Pinna (<0.1 m~), there was no evidence of any Malleus (hammer oysters that grow up to 150 mm in length). It is possible in this case that individual animals would not have been detected in the remote video footage because of their cryptic nature, whilst the depth precluded diving to check for them, Nevertheless, the clumps that existed previously would have been detectable with the video sampling, and can thus be regarded as absent. Both 69 of these assemblages are characterised by relatively long-lived organisms (5-10 yrs), and thus short-term seasonal or annual variation is unlikely to explain their absence in 2000/2001. There is no evidence that new types of assemblage dominated by large macrofauna/flora have established in the place of the Heterozostera and Malleus-Pinna assemblages that have disappeared. Instead, these areas now appear to be predominantly bare sand, with scattered invertebrates, including ascidians, bryozoans, sponges and some Pinna (Table 2). There are, however, extensive areas (mostly in water greater than 30 m deep) that are very depauperate in large macrofauna, and appear to be barren sand plains (Fig. 2), at least with respect to epibenthic organisms. The other substantial changes are an apparent 80% decrease in the area dominated by bryozoans in the central part of the gulf, and a reduction in the abundance of scallops in the central eastern section. In most of the areas where scallops were formerly a noticeable part of the benthic assemblage (with densities of 0.5 — 4 m*, Shepherd & Sprigg 1976) they now only occur in very low abundance (Table 2). In 2000/2001, only three sites on the western side of the gulf, and one in the south-east, had substantial TABLE 1. Habitat change, measured as the percentage of each habitat type that changed to other habitat types, in Gulf St Vincent over the period 1964-69 to 2000-01, Left column represents habitats in the 1960s, top row habitats in 2000. It should be noted that the habitat present in the 1960s was based on interpolation from dive sites that did not correspond to the 2000/2001 survey sites, potentially inflating the differences between the two surveys. Barren Ascidian Ascidian Bryozoan Pinna Scallop Seagrass Seagrass Total Sand Bryozoan Pinna transitions Heterozostera 100% 55 Algal debris 50% 50% 2 Ascidian-scallop 29% 47% 3% 3% 13% 38 Pinna-holothurian 31% 6% 6% 56% 5% 16 Bryozoan 56% 2% 21% 21% 43 Malleus-Pinna 95% 5% 42 Seagrass 15% I% 5% 1% 10% 3% 3% 62% 78 Reef (kunkerised shell) 33% 33% 33% Aeolianate reef 100% 2 TABLE 2. Densities and percent cover of major taxa in each habitat category in 2000/2001. Habitat categories are based on a cluster analysis, as described in the text. Density values assume each transect is 2 m wide, and as transect width could not be measured, are only approximate. Numbers in parentheses are standard errors. Blank cells indicate that the taxon did not occur in that habitat. Ascidian Pinna Bryozoan Sponge Scallop Sand Algae Seagrass (%) (%) (%) Ascidian 2.2 (0.5) 0.07 (0.03) 0.01 (0.008) 0.8 (0.7) 0.05 (0.02) 89 (3) 6 (2) 4 (2) Ascidian — Bryozoan 3.6 (0.04) 0.05 (0.03) 3.6 (0.04) 0.8 (0.7) 0.07 (0.05) 73 (8) 2 (1) 13 (5) Bryozoan 0.05 (0.03) 0.3 (0.3) 4.1 (0.3) 0.2 (0.04) 0.007 (0.004) 93 (2) 1 (1) Pinna 0.06 (0.02) 0.9 (0.3) 0.09 (0.02) 0.5 (0.2) 0.003 (0.002) 84 (4) 8 (3) 5 (2) Barren Sand 0.01 (0.002) 0.02 (0.002) 0.008 (0.001) 0.2 (0.05) 0.003 (0.001) 98 (0.3) 0.4 (0.1) 0.2 (0.1) Scallops 0.05 (0.02) 0.06 (0.03) 0.9 (0.9) 0.07 (0.02) 3.7(0.08) 71 (22) 7(4)~——s«<717) Seagrass 0.02 (0.005) 0.03 (0.01) 0.04 (0.008) 0.01 (0.009) 24 (4) 4(1) 72 (4) Seagrass - Pinna 0.4 (0.3) 0.8 (0.3) 0.05 (0.004) 0.2 (0.1) 21 (2) 2 (1) 73 (6) 70 J. E. TANNER numbers of scallops (Fig. 2), whereas in the 1960s there were also large areas on the eastern side where they dominated (Fig. 1). This reduction cannot be attributed to commercial scallop fishing as no such fishery has operated in the area. Bryozoans now only occur in low abundance, mostly in the central part of the gulf. Only 12 of the 294 sites surveyed had a bryozoan cover of greater than 5%, with none greater than 25%. This compares with the 1960s, when bryozoans dominated a substantial portion of the central gulf (Fig. 1, Table 1). As was the case in the 1960s, seagrasses still dominate many of the shallower areas <10 — 12 m deep both in the gulf and along the northern shore of Kangaroo Island (Fig. 2), with 65% of the sites originally classified as seagrass still being either seagrass or seagrass/Pinna in 2000/2001 (Table 1). The dominant seagrass genus remains Posidonia, with only a few areas of Amphibolis, and Halophila occurring predominantly on the western side of the gulf (see Tanner 2002). There is also some Halophila in deeper waters, which appears to be ephemeral as it was only detected at sites surveyed during the summer. Pinna in the northern section of the gulf remained relatively unchanged, although only 56% (9 of 16) of sites originally classified as Pinna/holothurian remained as Pinna, while a further 24 sites changed from some other habitat to Pinna (Table 1). Discussion There have been some obvious changes in the benthic assemblages present in Gulf St Vincent in the period between this study and that of Shepherd and Sprigg (1976), especially in the southern part of the gulf and in Investigator Strait. The main changes are the loss of extensive deep-water Heferozostera tasmanica meadows and Malleus-Pinna assemblages in the southern region, and a reduction in the cover of bryozoans and density of scallops in the central and eastern parts of the gulf. Interestingly, it is the deeper regions further from land that are generally the most changed, although the very shallow (<5 m deep) inshore areas were not included in this survey. There are several significant anthropogenic influences that may have contributed to these changes, although natural processes may also have played a role. The city of Adelaide is responsible for discharging a large amount of pollution into the gulf, in addition to that coming from agricultural runoff, and this is likely to have imposed a substantial stress on many organisms (Miller 1982; Neverauskas 1987; Edyvane 1999), There were also several severe dust storms in the 1980s that removed large amounts of topsoil from Yorke Peninsula (on the western shore) and deposited it into gulf waters (G. K. Jones, pers. comm.). The long flushing time of the gulf (Bye 1976; de Silva Samarsinghe & Lennon 1987) will have exacerbated any effects due to increased terrestrial inputs, as they are only slowly removed from the system. Prawn trawling has also been extensive (Morgan 1995), and although now carried out in a much more sustainable manner, damage from the previously intensive fishery may take many decades to be reversed. The loss of Heterozostera is most likely a result of increased water turbidity, and a subsequent decline in the amount of light reaching the bottom, although unfortunately no historical turbidity data could be found to determine the extent of this change. Given that this species occurred predominantly in deep (30- 40 m) water, it was probably at its lower depth limit, and it would only have required a small decrease in light penetration for Heterozostera to be unable to maintain itself. The maximum recorded depth for this species is 39 m (Duarte 1991). An increase in turbidity could have come about through several different mechanisms, Firstly, increased coastal discharge, both from the city of Adelaide and from agricultural areas, may have resulted in an increase in the amount of fine sediment in the water column, It would thus be of interest to examine the sediments of these and other areas of the gulf to see if an increase in the amount of terrigenous material can be detected over the last several decades. Secondly, there has been a substantial loss of seagrass along the metropolitan coast (~ 5000 ha), possibly connected to sewage discharge (Neverauskas 1987; Shepherd er al. 1989), resulting in a substantial increase in the rate of sediment resuspension in shallow waters. Although this increase in sediment resuspension has not been reliably quantified, there is now a consistent band of dirty brown water inshore of the seagrass line, which numerous anecdotal reports suggest is a relatively recent phenomenon (occurring since the loss of seagrasses over the last 40-50 years). If these resuspended sediments include a_ substantial proportion of very fine material that can stay in suspension for long periods of time, it is possible that they may have been distributed throughout the gulf, resulting in a system-wide increase in turbidity. Finally, heavy trawling activity is well known to result in sediment resuspension (Churchill 1989; Pilskaln et a/. 1998; Palanques et a/. 2001), and this may have increased the amount of suspended material. Trawling may also have had a direct impact on Heterozostera, causing more damage than could be sustained in areas where it could only just survive. The area formerly covered by Heferozostera has experienced substantial trawling pressure throughout the lifetime of the prawn fishery (Morgan 1995), and despite the substantial decline in effort over the last decade, the system may have experienced a state change that cannot easily be reversed. THREE DECADES OF HABITAT CHANGE 71 The former Ma/leus-Pinna assemblage that existed in the south-eastern portion of the gulf coincides very closely with the current main trawl grounds for the prawn fishery. It is thus likely that this assemblage experienced substantial direct damage from the trawl gear, as well as possible negative effects from increases in turbidity, whether due to trawling or coastal activities. While most of this area could not be examined by divers because of the depth, no Malleus were seen either in the video footage, or in the shallower areas that were accessible to divers. Further north, the bryozoan assemblages would also have been susceptible to trawl activity, as they are fragile and not adapted to cope with extensive physical disturbance (Bradstock & Gordon 1983). At its peak, the prawn fishery trawled in excess of 1600 km? yr'! (assuming a trawl speed of 3 kn, that the nets sweep a 20 m wide path, and that no areas are trawled more than once ina year), and thus would have had a direct effect on a large proportion of the gulf, although currently a much smaller area is trawled (200-450 km? yr! in the 1990s, J. Tanner unpublished data), While there are many well documented cases of trawling having a negative impact on benthic organisms (e.g. Auster ef al. 1996; Engel & Kvitek 1998: Kaiser et al. 1998; Collie et al. 2000), there is considerable controversy over how great the real impact is. This controversy arises as many studies have failed to show that trawling affects the benthic community (e.g. Gibbs et al, 1980; Van Dolah ef al. 1991; Hall et al. 1993; Hannson ef al. 2000; Lindegarth et a/. 2000). In Gulf St Vincent, the current trawling practices seem to have little impact on infauna over the short term, at least in the ascidian and Pinna habitats that have been studied (Drabsch et a/. 2001), although approximately 36% of epifauna is removed or dies subsequently (Tanner 2003). This suggests that previous periods of intense trawling may well have had a substantial negative effect on benthic communities in the gulf. The loss of these macro-faunal and floral assemblages has potentially important ecosystem level consequences. As the species that have been lost are those that formed most of the structure in these otherwise relatively homogenous sand plains, there are likely to be important implications for species that require complex habitat-structure to survive. For example, in New Zealand, bryozoan beds form an important habitat for juvenile snapper and other commercially fished species and have thus been protected from destructive fishing practices (Bradstock & Gordon 1983). Given that other habitat-forming species have not replaced those that have been lost, the physical complexity of these habitats has been greatly reduced, and concomitant losses or reductions in other species are likely. It is well established, for example, that shallow-water seagrasses provide important habitats for many fish and invertebrate species (e.g. Bell & Westoby 1986; Edgar 1990; Connolly 1994; Perkins-Visser et al. 1996), and while much less is known about deep- water seagrasses, they are likely to serve a similar function. These habitat-forming species may have also provided an important food source for other species, and so their loss may have disrupted food- webs. For example, snapper (Pagrus auratus) are important predators of Malleus, and the loss of the latter species may have played a role in the decline of snapper stocks in Gulf St Vincent. In conclusion, there have been substantial changes to the benthic habitats present in Gulf St Vincent between the 1960s and 2000/2001, especially in the central and southern regions. Potential causes include increases in turbidity due to terrestrial runoff (including sewage discharge, stormwater and agricultural runoff) and dust-storms, and the direct effects of trawling. The northern (most inland) portion of the gulf seems to have experienced the least degradation, possibly due to the very limited terrestrial runoff associated with South Australia’s arid climate, and the lack of a history of trawling. The documented changes predominantly involve the loss of important structure-forming species, and thus could potentially have substantial implications for many associated species that rely on physically complex habitats for their survival. Acknowledgments I would like to thank T. Fowler, K. Jones, S. Shepherd and two anonymous reviewers for comments on an earlier version of this manuscript. Thanks also to the many people who helped with the surveys. Funding for this research was provided by FRDC grant # 1998/208. 72 J. E. TANNER References AUSTER, P. J., MALATESTA, R. J., LANGTON, R. W., WATLING, L., VALENTINE, P. C., DONALDSON, C. L., LANGTON, E. W., SHEPARD, A. N., & Babs, I. G. (1996) The impacts of mobile fishing gear on seafloor habitats in the Gulf of Maine (northwest Atlantic): implications for conservation of fish populations. Rev. Fish. Sci. 4, 185- 202. BELL, J. D. & Westosy, M. (1986) Importance of local changes in leaf height and density to fish and decapods associated with seagrasses. J) Exp. Mar: Biol. Ecol. 104, 249-274. Brapstock, M. & Gorpbon, D. P. (1983) Coral-like bryozoan growths in Tasman Bay, and their protection to conserve commercial fish stocks. New Zeal. J. Mar. Freshwater Res, 17, 159-163. Bye, J. A. T. (1976) Physical oceanography of Gulf St Vincent and Investigator Strait pp. 143-160 Jn Twidale, C.R., Tyler, M. J., & Webb, B. P. (Eds) “Natural History of the Adelaide Region” (Royal Society of South Australia, Adelaide). CHURCHILL, J. H. (1989) The effect of commercial trawling on sediment resuspension and transport over the Middle Atlantic Bight continental shelf. Continental Shelf Res. 9, 841-864. Couuik, J. S., HALL, S. JL, KAISER, M. IL, & Poiner, L R. (2000) A quantitative analysis of fishing impacts on shelf-sea benthos. J An. Ecol. 69, 785-798. CONNOLLY, R. M. (1994) The role of seagrass as preferred habitat for juvenile Si/laginodes punctata (Cuv. & Val.) (Sillaginidae, Pisces): habitat selection or feeding? J. Exp. Mar. Biol. Ecol. 180, 39-47. De SILVA SAMARASINGHE, J. R. & LENNON, G. W. (1987) Hypersalinity, flushing and transient salt-wedges in a tidal gulf - an inverse estuary. Est. Coast. Shelf Sci. 24, 483-498. Drasscu, S. L., TANNER, J. E., & CONNELL, S. D. (2001) Limited infaunal response to trawling in previously untrawled areas. ICES J. Mar. Sci. 58, 1261-1271. Duarte, C. M. (1991) Seagrass depth limits. Ag. Bot. 40, 363-378. Epaar, G. J. (1990) Predator-prey interactions in seagrass beds. II. Distribution and diet of the blue manna crab Portunus pelagicus Linnaeus at Cliff Head, Western Australia. J. Exp. Mar. Biol. Ecol. 139, 23-32. Eby VANE, K. S. (1999) Coastal and marine wetlands in Gulf St. Vincent, South Australia: understanding their loss and degradation, Wetlands Ecol. Manage. 7, 83-104. ENGEL, J. & Kvitek, R. (1998) Effects of otter trawling on a benthic community in Monterey Bay National Marine Sanctuary. Conserv. Biol. 12, 1204-1214. Estes, J. A. & Duaains, D. O. (1995) Sea otters and kelp forests in Alaska: generality and variation in a community ecological paradigm. Ecol. Monogr. 65, 75- 100. Frip, C. L. J., CLARK, R. A., & HALL, J. A. (1999) Long- term changes in the benthos on a heavily fished ground off the NE coast of England. Mar. Ecol. Prog. Ser. 188, 13-20. , HArwoop, K. G., HALL, S. J., & HALL, J. A. (2000) Long-term changes in the benthic communities on North Sea fishing grounds. /CES J Mar. Sci. 57, 1303-1309. Giss, P. J., COLLINS, A. J., & COLLetTt, L. C. (1980) Effects of otter prawn trawling on the macrobenthos of a sandy substratum in a New South Wales estuary. Aust. J. Mar. FW. Res. 31, 509-516. HAgprRICH, R. L. & BARNES, S. M. (1997) Changes over time of the size structure in an exploited shelf fish community. Fish. Res, 31, 229-239. HALL, S. J., ROBERTSON, M. R., BASFoRD, D. J., & HEANEY, S. D. (1993) The possible effects of fishing disturbance in the northern North Sea: an analysis of spatial patterns in community structure around a wreck. Netherlands J. Sea Res, 31, 201-208. (1999) “The effects of fishing on marine eco- systems and communities” (Blackwell Science, London). Hansson, M., LINDEGRATH, M., VALENTINSSON, D., & ULMESTRAND, M. (2000) Effects of shrimp-trawling on abundance of benthic macrofauna in Gullmarsfjorden, Sweden. Mar. Ecol. Prog. Ser. 198, 191-201. JENNINGS, S, & Kaiser, M. J. (1998) The effects of fishing on marine ecosystems. Adv. Mar. Biol. 34, 203- 352. Kaiser, M. J., Epbwarps, D. B., ARMSTRONG, P. J., RADFORD, K., Loucu, N. E., FLATT, R. P., & Jones, H. D. (1998) Changes in megafaunal benthic communities in different habitats after trawling disturbance. /CES J. Mar. Sci. 55, 353-361. LAPoInTE, B. E., TOMASKO, D. A., & Marzig, W. R. (1994) Eutrophication and trophic state classification of seagrass communities in the Florida Keys. Bull. Mar. Sci. 54, 696-717. LINDEGARTH, M., VALENTINSSON, D., HANSSON, M., & ULMESTRAND, M. (2000) Effects of trawling disturbances on temporal and spatial structure of benthic soft- sediment assemblages in Gullmarsfjorden, Sweden. ICES J. Mar. Sci. 57, 1369-1376. MILter, S. (1982) “South Australian land-based marine pollution” (Department of Environmnet and Planning, Adelaide). Moraan, G. R. (1995) “Assessment, management and research support for the Gulf St Vincent prawn fishery” (Primary Industries SA, Adelaide). NEVERAUSKAS, V. P. (1987) Monitoring seagrass beds around a sewage sludge outfall in South Australia. Mar. Pol, Bull. 18, 158-164. PALANQUES, A., GUILLEN, J., & PuiG, P. (2001) Impact of bottom trawling on water turbidity and muddy sediment of an unfished continental shelf. Limnol. Oceanogr. 46, 1100-1110. PERKINS-VISSER, E., WoLcoTT, T. G., & WoLcortT, D. L. (1996) Nursery role of seagrass beds: enhanced growth of juvenile blue crabs (Callinectes sapidus Rathbun). J. Exp. Mar. Biol. Ecol. 198, 155-173. PILSKALN, C, H., CHURCHILL, J. H., & Mayer, L. M. (1998) Resuspension of sediment by bottom trawling in the Gulf of Maine and potential geochemical consequences. Conserv. Biol. 12, 1223-1229. SEBER, G. A. F. (1984) “Multivariate observations” (Wiley, New York). SEDDON, S., CONNOLLY, R. M., & EpyvANE, K. S. (2000) Large-scale seagrass dieback in northern Spencer Gulf, South Australia. 4g. Bot. 66, 297-310. SHEPHERD, S. A. & SpricG, R. C. (1976) Substrate, sediments and subtidal ecology of Gulf St. Vincent and Investgator Strait pp. 161-174 /n Twidale, C. R., Tyler, M. J., & Webb, B. P. (Eds) “Natural History of the Adelaide Region”. (Royal Society of South Australia, Adelaide). . McComas, A. J., ButtHurs, D. A, NEVERAUSKAS, V. P., STEFFENSEN, D. A., & WEST, R. (1989) Decline of seagrasses pp. 346-393 /n Larkum, A. W., McComb, A. J., & Shepherd, S. A. (Eds) “Biology of seagrasses” (Elsevier, Amsterdam). Suort, F. T. & WYLLIE-ECHEVERRIA, S. (1996) Natural and human-induced disturbance of seagrasses. Environ. Conserv, 23, 17-27. THREE DECADES OF HABITAT CHANGE 73 SvANE, I. & JOHNSON, J. (2003) “Gulf St Vincent prawn (Melicertus latisulcatus) fishery” (SARDI Aquatic Sciences, Adelaide). TANNER, J. E. (2002) “Habitat modification and its influence on prawn and crab fisheries” (SARDI Aquatic Sciences, Adelaide). (2003) The influence of prawn trawling on benthic assemblages in Gulf St Vincent, South Australia. Can. J. Fish. Aq. Sci. 60, 517-526. VAN Dotan, R. F., WENDT, P. H., & LevISEN, M. V. (1991) A study of the effects of shrimp trawling on benthic communities in two South Carolina sounds. Fish. Res. 12, 139-156. WALKER, D. 1. & McComs, A. J. (1992) Seagrass degradation in Australian coastal waters. Mar. Pol. Bull. 25, 191-195. WILSON, R. S., HEISLERS, S., & PoorRE, G. C. (1998) Changes in benthic communities of Port Phillip Bay, Australia, between 1969 and 1995. Mar. Freshwater Res. 49, 847-861. Sources of Unpublished Materials Dr KEITH JoNEs, SARDI Aquatic Sciences, PO Box 120, Henley Beach, SA. 5022. Australia. Dr ScoRESBY SHEPHERD, SARDI Aquatic Sciences, PO Box 120, Henley Beach, SA. 5022. Australia. Transactions of the Royal Society of S. Aust. (2005), 129(1), 74-80. FLOOD REGIME CHANGE IN THE HATTAH LAKES VICTORIA RESULTING FROM REGULATION OF THE RIVER MURRAY by N. J. SourER" Summary Souter, N. J. (2005) Flood regime change in the Hattah Lakes Victoria resulting from regulation of the River Murray. Trans. R. Soc. S. Aust. 129(1), 74-80, 31 May, 2005. Regulation of the River Murray was found to alter the frequency and variability of flooding through the Hattah Lakes on the river’s floodplain. The height of Lake Hattah, recorded monthly since May 1908, was correlated with historical data recording the extent of flooding through the system. This allowed the flooding frequencies for six lakes in the system to be calculated using the remaining height data for Lake Hattah. An investigation of pre (1908-1935) and post (1936-1996) regulation frequencies revealed a reduction in the frequency of small and medium sized floods and a reduction in overall variability of flooding frequency throughout the system. Inflow channel remodelling undertaken in 1972/73, in an attempt to increase the flooding frequency of the lakes, increased the frequency of small floods similar to pre-regulation levels but did little to improve the variability of flooding through the system. Key Woros: River regulation, floodplain lakes, flooding frequency, environmental flows, Introduction Flooding frequency and the length of the dry period between flooding are important factors in structuring the aquatic communities of wetlands (Crome 1986; Boulton & Lloyd 1992; Wellborn et al. 1996; Sheldon & Puckridge 1998). Differences in flooding frequency have been related to differences in macroinvertebrate diversity between floodplain lakes (Van den Brink et al. 1994) and are known to effect the emergence of aquatic invertebrates from dry floodplain sediments (Boulton & Lloyd 1992). Fish are known to utilise floodplains as habitat, as adults (Welcome 1985; Reimer 1991) and as nurseries for native Australian fish (Gehrke 1992). Changes in vegetation structure may also occur as a result of changes in flooding frequency (Bren 1992; Roberts & Marston 2000; Roberts et a/. 2000). River regulation through the construction and operation of dams, weirs and locks to provide predictable flows for human use, changes the natural flow regime of * Floodplain Ecology Group, Department of Natural Resources and Environment, Kaiela Research Station, Shepparton Victoria. Current address: Department of Water, Land and Biodiversity Conservation, Surface Water Assessment Branch, GPO Box 2834, Adelaide SA 5001. Email: nsouter@adam.com.au Siepentritt, M. & WILKINSON, L. 2004 The Murray-Darling Basin’s Wetlands: Managing the Natural and Economic Riches (Department of the Environment and Heritage, Canberra). ? BAKER, T. (1976) Report on investigations into the Hattah Lakes National Park water supply problem. National Parks Service, Victoria (unpub. ). Suaw, J. R. (1985) The effects of river regulation and management of water supplies on the Hattah-Kulkyne National Park and Murray-Kulkyne park, Masters Thesis, University of Melbourne (unpub.). CummMING, P. L. F., & LLtoyp, L. N. (1993) Flood characteristics of the Hattah Lakes system. Integrated Watering Strategy Report No. 5, Floodplain Ecology Group, Department of Conservation and Natural Resources, Shepparton (unpub.). - both a river and its floodplain (Jacobs 1990; Boulton & Lloyd 1992). Environmental flows may be used to restore wetlands which have suffered reduced flooding as a result of regulation (Petts 1996; Tharme 2003). The RAMSAR listed Hattah Lakes are a system of 17 floodplain lakes on the mid to lower River Murray targeted to receive environmental flows for ecological gain (Siebentritt'!). Regulation has caused a decrease in the flooding frequency of one of the first lakes in the system, Lake Hattah (Barker?; Robinson 1966; Shaw?; Cumming & Lloyd?) whilst the effect of regulation on the other 16 lakes is unknown. If the maximum environmental benefit is to be gained by providing environmental flows to the Hattah Lakes, an estimate of the impact of regulation on the flooding frequency of the other lakes within the system is required. This is important because the dynamics of floodplain flooding are complex, as many floodplain systems have intricate patterns of connectivity and flow (Kingsford 2000). Differences in floodplain topography can also have dramatic effects, as small fluctuations may be the difference between large areas of floodplain, individual lakes and wetlands becoming inundated or remaining dry (Taylor et al. 1996). A previous attempt at increasing the rate of inflow to and flooding frequency of, the lakes was made in 1972/73 when Chalka Creek, which feeds the lakes from the River Murray, was channelised (Baker?). The impact of this modification has not previously been assessed. This paper describes a novel method, using historical records, to estimate flooding frequency through the Hattah Lakes system. Changes in flooding frequency of the Hattah Lakes caused by river regulation and the modifications to Chalka FLOOD REGIME CHANGE IN THE HATTAH LAKES VICTORIA 75 142.255° 142.32° 142.5057 Chalka Litttle Lake Hattahe Victoria Melbourne: Hattah Lake -34,785%s, , Kramen ty 734.785" wo N Oo a ° 6 q y g Fig. |. The Hattah Lakes system. The shaded lakes are those referred to in the text. 6 . 5 | 3 L yy ny 2 1 0 U U i I 1908 1916 1923 1931 1939 1946 1954 1962 1969 1977 1985 1992 Year Fig. 2. Monthly records of the height of Lake Hattah from May 1908 to October1996. 76 N. J. SOUTER Creek are estimated and assessed. From this, recommendations regarding future environmental flows are suggested. Methods Site Description Hattah Lakes are a series of 17 interconnected lakes on the Victorian floodplain of the River Murray. The system relies on overbank flow from the Murray and is fed via Chalka Creek, an anabranch of the river. Chalka Creek fills the northern lakes in succession, before rejoining the Murray to the north (Fig. 1). The lakes on the south-easterly terminal branch fill successively at similar times to those on the northern branch. Lake Kramen is filled via overbank flow directly from the River. The height of Lake Hattah, one of the first lakes in the SE terminal branch, is directly related to the height of the River Murray at Euston (Robinson 1966). Most lakes dry within 12 months after the cessation of flow (Cumming and Lloyd*). Lakes Hattah and Mournpoul are the most permanent of the system, holding water for several years. To increase the rate of inflow to the Lakes, Chalka Creek was deepened and widened in 1972/73 (Baker?). This operation was undertaken to allow water to flow to the Hattah Lakes when discharge in the River Murray at Euston exceeded approximately 36 700 ML/day instead of the previous approximate level of 48 000 ML/day (Baker’). Estimate of flooding frequency The flooding frequencies of six of the seventeen lakes in the system: Lakes Hattah, Mournpoul, Konardin, Nip Nip, Bitterang and Kramen were calculated. These lakes were chosen for their geographical spread and because _ historical information detailing their flooding was available (Fig. 1). The height, depth in metres, of Lake Hattah was recorded monthly on a gauge from May 1908 until April 1984 (Fig. 2). From May 1984 until October 1996 data were collected inconsistently. However during this period the maximum height of each flood was recorded (Hattah-Kulkyne National Park, unpublished data). Historical information for twelve floods which described or depicted the extent of flooding through the system was summarised (Table 1). From 1908 until 1996 when flooding occurred it took place in the spring/summer of each year. This enabled the maximum Lake Hattah gauge height of each yearly flood to be determined. Flooding frequencies of the six lakes were estimated by correlating the height of Lake Hattah with the extent of flooding through the system. For Lake Hattah to be considered to have flooded, rather than just having received a minor inflow of water, an arbitrary increase in lake height of 0.35 m must have occurred, regardless of the initial lake depth. The flood threshold for the other five lakes was estimated as the lowest recorded Lake Hattah gauge height at which each lake was flooded. Given this value the number of times each lake flooded over a given period was estimated by referring to the height of Lake Hattah each year and determining the extent of flooding through the system. The flooding frequency of each lake was calculated by dividing the number of times a lake was estimated to have flooded by the total number of years analysed. The frequency of dry years is equal to one minus the flooding frequency. The completion of the Hume Dam in 1936 is regarded in this study as the beginning of river regulation. This provided 28 years (1908-1935) of pre-regulation data and 61 years (1936-1996) of post-regulation data. Two by two contingency tables of the frequency of inflow years vs. no inflow years for each lake were constructed to assess the impact of regulation and Chalka Creek channelisation using a G-test (Sokal & Rolf 1995). Three comparisons between flooding frequency for each lake were made: 1) Pre-regulation (1908-1935: 28 years) vs. the period after regulation but prior to the channelisation of Chalka Creek (1936-1972: 37 years), 2) Post-regulation before (1936-1972) vs. after the channelisation of Chalka Creek (1973-1996: 24 years), and 3) Pre-regulation (1908-1935) vs. post regulation after the channelisation of Chalka Creek (1973- 1996). Results Flooding characteristics The determination of flood threshold values for each of the six lakes are: LAKE HATTAH The lack of records describing floods which exclusively filled Lake Hattah made it difficult to determine what level of inflow to Lake Hattah constitutes a flood or a minor inflow of water. On only four occasions since 1908 did water enter Lake Hattah but not fill to a level likely to move further through the system. The lake rose 0.10 m, (from 2.28 to 2.38 m) in 1910, 0.30 m (from 1.98 to 2.28 m) in 1922, 1.78 m (from 0.51 to 2.29 m) in 1928, whilst the 1987 flood entered the lake but its height did not increase (Table 1). In calculating flooding frequency only 1928 was analysed as a flood, whilst the minor inflows in 1910, 1922 and 1987 were classified as years of no inflow. FLOOD REGIME CHANGE IN THE HATTAH LAKES VICTORIA 77 TABLE 1. Summarised historical records of flooding through the Hattah Lakes system, showing the terminal lake flooded along each arm of the system and the maximum height of Lake Hattah for each flood. Source Year Terminal Lake = Terminal Lake = Maximum height of flooded (northern arm) (eastern arm) Lake Hattah Department of Natural Resources 1956 Bitterang Kramen 5.49 m and Environment (DNRE)> Robinson (1966) 1964 Bitterang Nip Nip 4.50 m Robinson (1966)° 1964 Bitterang 40m Robinson (1966)° 1964 Nip Nip 3.66 m Douglas’ 1970 Bitterang Not reported 4.27 m DNRE® 1986 Mournpoul Hattah 2.74 m DNRE® 1987 Lockie Hattah inflow but no height increase DNRE® 1989 Bitterang Nip Nip 3.75 m TM Satellite photograph (15 Dec 1989) 1989 Bitterang Nip Nip 3.75 m DNRE?® 1990 Bitterang N/R 3.76 m MDBC (1994)!° 1990 Bitterang Nip Nip 3.76m TM Satellite photograph (9 Feb 1993) 1992 Bitterang Nip Nip 4.59m (minimal input to Kramen) DNRE? 1993 Bitterang Kramen 5.0m DNRE'"! 1995 Konardin Nip Nip 3.60 m (reached but did not fill) Puckridge ef al.'* September Mournpoul - <2.57m (actively flooding 1996 when recorded) Puckridge ef al.'? October Bitterang Nip Nip 3.95 m 1996 5 Aerial photograph (unpub.) ° Robinson reported that Lake Bitterang began to fill once the height of Lake Hattah reached 4.0 m, whilst for Lake Nip Nip to fill Lake Hattah must reach 3.66 m. 7 Douglas, M. G. (1972) The water regime of the Chalka Creek and lakes of the Kulkyne Forest and Hattah Lakes National Park. Sunraysia Naturalists’ Research Trust - Ninth Report (unpub.). & Compiled during the preparation of Cumming and Lloyd* ° Brendan Atkins (pers. comm.) '0 Terrascan Aerial photograph '! Phil Murdoch (pers. comm.) 5 Puckridge, J. T., Ward, K. A. and Walker, K. F. (1997) Hydrological determinants of fish and macroinvertebrate ecology in the Hattah Lakes System: Implications for Time-Share Flooding. Part 1: 1996/97. University of Adelaide, Department of Natural Resources and Environment, Land and Water Resources Research and Development Corporation (unpub.). LAKE MOURNPOUL Lake Mournpoul was found to require a slightly larger flood than Lake Hattah for it to fill. Water was observed to reach Lake Hattah before Lake Mournpoul in 1970, 1986 and 1989, whilst in 1987 water did not enter Lake Mournpoul but flowed into Lake Hattah. The 1986 flood reached a gauge height of 2.74 m and filled only Lakes Hattah and Mournpoul. Puckridge ef. al/.'* noted that in September 1996 Lake Mournpoul was actively filling when Lake Hattah had a gauge height of 2.57 m and that Lake Mournpoul had probably flooded earlier as a result of an earlier unobserved flood peak. Using this information the flood threshold of Lake Mournpoul has been estimated at 2.57 m on the Lake Hattah gauge. LAKE KONARDIN In 1995 Lake Konardin flooded when Lake Hattah was at a height of 3.60 m and was the last lake flooded on the northern arm of the system. Prior to flooding the Lake was dry (Phil Murdoch, Hattah- Kulkyne National Park Ranger pers. com.). The flood threshold of Lake Kramen is estimated at 3.60 m on the Lake Hattah gauge. LAKE Nip Nip Robinson (1966) determined that Lake Hattah must reach about 12’ (3.66 m) before Lake Nip Nip floods. This observation was supported in 1995 when Lake Hattah rose to 3.60 m and water reached, but did not fill Lake Nip Nip. The floods of 1989, 1990, 1992 and 1996 all showed lake Nip Nip to fill 78 N. J. SOUTER TABLE 2. Lake Hattah gauge height and frequencies of no inflow and flooding for six lakes in the Hattah system. Frequencies are presented for a range of time spans: pre-regulation and post regulation both before and after Chalka Creek channelisation. No Inflow Hattah Mournpoul Konardin Nip Nip Bitterang Kramen Lake Hattah gauge height (m) - - 2.75 3.60 3.66 3.75 5.0 Pre-regulation (1908-1935) 0.32 0.68 0.64 0.50 0.50 0.43 0.07 frequency Post-regulation/pre-Chalka 0.59 0.41 0.41 0.32 0.32 0.32 0.03 modification (1936-1972) frequency Post-regulation/post-Chalka 0.33 0.67 0.67 0.38 0.33 0.33 0.13 modification (1973-1996) frequency TABLE 3. G-test results for comparisons of the flood frequency of the Hattah lakes for periods pre- (1908-1935), and post- regulation prior to (1936-1972) and after (1973-1996) Chalka Creek channelisation (* significant difference at the p<0.05 level). Lake 1908-1935 vs. 1936-1972 1936-1972 vs. 1973-1996 1908-1935 vs. 1973-1996 = p= o p= is p= Hattah 4.733 0.030* 3.931 0.047* 0.008 0.928 Mournpoul 3.550 0.059 3.931 0.047* 0.031 0.859 Konardin 1.999 0.157 0.160 0.689 0.798 0.371 Nip Nip 1.999 0.157 0.005 0.942 1.439 0.230 Bitterang 0.722 0.395 0.005 0.942 0.482 0.487 Kramen 0.606 0.436 1.984 0.159 0.387 0.534 when the gauge height of Lake Hattah was over 3.66 m (Table 1). The threshold of Lake Nip Nip is estimated at 3.66 m on the Lake Hattah gauge. LAKE BITTERANG Robinson (1966) reported that Lake Bitterang began to fill when Lake Hattah reached 4.0 m. This is considerably higher than the values recorded during 1989, 1990 and 1996. This difference cannot be explained by Lake Hattah holding water previously, as prior to the 1964 flood it had been dry for approximately ten months. The flood threshold of Lake Bitterang 1s likely to be no greater than the 3.75 m recorded for the 1989 flood and verified via two sources (Table 1). However, the threshold must be greater than 3.60 m, the height of the 1995 flood, which did not reach Lake Bitterang, suggesting that the true threshold value lies somewhere between 3.60 and 3.75 m. The threshold value has been estimated as 3.75 m on the Lake Hattah gauge. The other four floods to fill Lake Bitterang (1956, 1970, 1992 and 1993) were all well above this value. LAKE KRAMEN The threshold value for Lake Kramen lies somewhere between 4.59 and 5.0 m as the 1992 flood of 4.59 m reached Lake Kramen but did not fill it, whilst the 1993 flood of 5.0 m filled the Lake. The aerial photograph of the 1956 flood revealed significant connection directly between the River Murray and Lake Kramen, suggesting that the Lake, when it fills, receives most of its water directly from the River Murray and not via the Chalka Creek system. The threshold of Lake Kramen is estimated at 5.0 m. Whilst there was some difference between the two values used to estimate this threshold, the use of the higher value is justified in that on only two occasions (1939, 4.65 m and 1973, 4.88 m) did floods of a height between these values occur. Changes in flooding frequency Prior to regulation of the River Murray, the Hattah Lakes system was subject to a wide range of flooding frequencies (Table 2). Lakes Hattah and Mourpoul were the most frequently flooded, Lakes Konardin, Bitterang and Nip Nip formed a group of intermittently flooded lakes and Lake Kramen was only infrequently filled. Post-regulation (1936-1972) flooding frequencies for all lakes were lower than those estimated prior to regulation (Table 2). The flooding frequency of Lake Hattah was significantly lower, whilst the frequencies of other lakes were not significantly different (Table 3). Post-regulation the flooding frequencies of lakes Hattah and Mournpoul FLOOD REGIME CHANGE IN THE HATTAH LAKES VICTORIA 79 were identical, as were the frequencies of lakes Konardin, Bitterang and Nip Nip. There was little difference in the flooding frequencies of Lake Kramen either before or after regulation Lakes Hattah and Mournpoul were flooded significantly more frequently after the channelisation of Chalka Creek than in the 37 years prior (Table 3). Channelisation significantly increased the flooding frequency of Lake Hattah which returned to its pre- regulation level, as there was no. significant difference in the flooding frequency prior to regulation and post Chalka Creek modification (Table 3). Channelisation did not increase flooding frequency within the rest of the system as there was little change in the flooding frequencies of Lakes Konardin, Nip Nip and Bitterang, nor were differences in flooding frequency between lakes significant (Table 3). Discussion Regulation of the River Murray has reduced the frequency and variability of flooding within the Hattah Lakes. This was primarily due to the reduction in the frequency of small to medium sized floods. This in turn reduced variability as lakes which could be filled by small to medium sized floods were now only generally filled via larger floods which filled all of the lakes. This investigation arrived at the same conclusion as the River Murray monthly simulation model of Maheshwari ef al. (1995) which showed that regulation had caused a reduction in the frequency of small to medium sized floods in the mid-Murray. Given the positive relationship between the height of Lake Hattah and the River Murray at Euston (Robinson 1966), this result was not unexpected. Kingsford (2000) suggested that instream models are unsuitable for assessing flood regime on a river’s floodplain. The value of investigating the effect of river regulation on the floodplain, rather than just the main channel, was demonstrated by the findings of this study that regulation reduced the variability of flooding frequencies between lakes, a result of the reduction of small to medium sized floods. The channelisation of Chalka Creek in 1972/73 did, as predicted (Baker?; Shaw?), increase the flooding frequencies of lakes Hattah and Mournpoul (even restoring the pre-regulation flooding frequency of Lake Hattah). Channelisation had no impact on the systems less frequently flooded lakes and did not fully restore the variability of flooding through the system. Whilst a loss of variability in the flood regime was observed, differences in flooding frequency between periods for any of the lakes were generally not statistically significant. This does not necessarily mean that non-statistically significant changes may not be biologically important as large differences in flooding were required for significance. The relationship between flooding and the biological response in the Hattah Lakes is beyond this paper and will be addressed elsewhere. The inability to detect statistical differences in the flooding frequencies between lakes could also be due to the short time intervals examined. The longest period of 37 years (between 1936-1972) saw a considerable increase in the degree of regulation of the River (Maheshwari et al. 1995) added to which the flow regime of the River Murray is unstable within a 50 year time frame (Walker ef al. 1995). Thus it is likely that larger data sets, although unavailable, would be required to achieve significant results. Caution in these results should also be exercised, as sources of error exist in determining the flood threshold height for each lake. It is likely that multiple flood threshold values exist for each of the lakes. For example, threshold values may change according to the speed at which flooding progresses through the system. As this information was unavailable or inadequate it could not be assessed. Differences in the degree of drawdown at the start of each flood may also alter the threshold height. However most of the lakes within the system are likely to have dried between floods as most are shallow and dry within twelve months after the cessation of flow (Cumming and Lloyd*). Lakes Mournpoul (drying time, 7 years) and Hattah (3 years) are two of the deeper lakes in the system and, taking longer to dry (Cumming and Lloyd?) often have different drawdown conditions at the start of each flood. These conditions were assessed in determining the flooding frequency of Lake Hattah although they were unknown for Lake Mournpoul. Further detailed on ground investigations of flooding through the system are recommended to improve the accuracy of the flooding frequency estimates provided in this paper. The loss of flooding variability may have significant ecological impacts, given the importance of flooding frequency in structuring aquatic communities. With the error associated in calculating flooding frequency these values should be used as a guide only. However, it can be reasonably suggested that efforts to rehabilitate the Hattah Lakes, through the provision of environmental flows, would be best directed towards increasing the variability of flooding through the system, most particularly within intermittently flooded lakes such as Konardin, Nip Nip and Bitterang as these lakes have suffered reduced flooding frequency and variability as a result of regulation but have not been improved by the modifications to Chalka Creek. 80 N. J. SOUTER Acknowledgements This work arose out of the Land and Water Resources Research & Development Corporation, project VCB1, Time-share flooding of aquatic ecosystems. Keith Ward, Jim Puckridge, Alex McNee and Lance Lloyd are thanked for their support and advice on earlier drafts. Trevor Jacobs suggested that I prepare this work for publication. References BouLton, A. J. & LLoyp, L. N. (1992) Flooding frequency and invertebrate emergence from dry floodplain sediments of the River Murray, Australia. Regul. Riv. 7, 137-151. Bren, L. J. (1992) Tree invasion of an intermittent wetland in relation to changes in the flooding frequency of the River Murray, Australia. Aust. J. Ecol. 17(4), 395-408. Crome, F. H. J. (1986) Australian waterfowl do not necessarily breed on a rising water level. Aust. Wildl. Res. 13, 461-480. GEHRKE, P. C. (1992) Diel abundance, migration and feeding of fish larvae in a floodplain billabong. J Fish Biol. 40, 695-707. Jacoss, T. (1990) River regulation. pp. 39-58 In McKay, N. & Eastburn, D. (Eds.) “The Murray” (Murray-Darling Basin Commission, Canberrra). KINGsFoRD, R. T. (2000) Ecological impacts of dams, water diversions and river management on floodplain wetlands in Australia. Austral Ecol. 25(2), 109-127. MAHESHWARI, B. L., WALKER, K. F. & McMAuon, T. A. (1995) Effects of regulation on the flow regime of the River Murray, Australia. Regul. Riv. 10, 15-38. Petts, G. E. (1996) Water allocation to protect river ecosystems. Regul. Riv. 12, 353-365. REIMER, G. (1991) The ecological importance of floodplains for fish at the river March (Austria). Arch. Hydrobiol. 121, 355-363. Rosperts, J. & MARSTON, F. (2000) “Water regime of wetland and floodplain plants in the Murray-Darling Basin: A source book of ecological knowledge” (CSIRO Land and Water, Canberra). , YOUNG, B. & Marston, F. (2000) “Estimating the water requirements for plants of floodplain wetlands: a guide” (Land and Water Resources Research and Development Corporation, Canberra). ROBINSON, G. (1966) The regime of Hattah Lakes. Proc. Roy. Soc. Victoria 79(2), 561-568. SHELDON, F. & PuckripGe, J. T. (1998) Macro- invertebrate assemblages of Goyder Lagoon, Diamantina River, South Australia. Trans. R. Soc. S. Aust. 122(1): 17-31. SOKAL, R. R. & Routr, F. J. (1995) “Biometry: the principles and practice of statistics in biological research” (W. H. Freeman and Company, New York). TAYLOR, P. J., WALKER, G. R., HODGSON, G., HATTON, T. J. & Corre, R. L. (1996) Testing of a GIS Model of Eucalyptus largiflorens health on a semiarid, saline floodplain. Envir. Manag. 20, 553-564. THARME, R. E. (2003) A global perspective on environ- mental flow assessment: emerging trends in the development and application of environmental flow methodologies for rivers. River Res. Applic. 19, 397-441. VAN DEN BRINK, F. W. B., BELJAARDS, M. J., Boots, N.C. A. & VAN DER VELDE, G. (1994) Macrozoobenthos abundance and community composition in three lower Rhine floodplain lakes with varying inundation regimes. Regul. Riv. 9, 279-293. WALKER, K. F., SHELDON, F. & PucKRIDGE, J. T. (1995) A perspective on dryland river ecosystems. Regul. Riv. 11, 85-104. WELCcoME, R. L. (1985) River fisheries. £A.O. Fish Tech. Pap. 262, p. 330. WELLBORN G. A., SKELLY D. K. & WERNER E. E. (1996). Mechanisms creating community structure across a freshwater habitat gradient. Annu. Rev. Ecol. Syst. 27, 337-363. Transactions of the Royal Society of S. Aust. (2005), 129(1), 81-83. BRIEF COMMUNICATION HELMINTH AND PROTOZOAN PARASITES OF FERAL CATS FROM KANGAROO ISLAND There are limited data on the helminth and protozoan parasites of the feral cat, Felis catus, on Kangaroo Island although the carnivore is well established there. Five helminth species from feral cats were reported in a list compiled from parasites that had been submitted to a veterinary laboratory in South Australia'. Surveys of the helminth species in feral cats conducted elsewhere have identified a varying prevalence dependent on the available food sources and climate*+5°7, Few of these surveys, however, identified the protozoan parasites. A significantly greater prevalence of the protozoan parasite, Toxoplasma gondii, has been identified in sheep on Kangaroo Island and attributed to an apparent presence of large numbers of feral cats*. In this study, we present data on the occurrence of the helminth and protozoan parasites detected in a sample of feral cats from Kangaroo Island. The gastro-intestinal tract, heart, lungs, whole blood and faeces from 46 cats were submitted for examination. Organs were opened, washed and the parasites removed and counted using a dissecting microscope. The stomach wall was examined for nodules. Mucus from the stomach wall was examined specifically for Ollulanus tricuspis using a squash preparation on a glass slide and a compound microscope. The stomach and small intestinal wall was scraped, fixed in formalin and examined for helminths. A faecal sample was examined for helminth eggs, nematode larvae and protozoan cysts using a centrifugal flotation method in saturated MgSO, solution and additionally in saturated KI solution to determine the presence of Cryptosporidium oocysts’. The Baermann method! was used to recover Aelurostrongylus abstrusus larvae from faeces and lung parenchyma. Faecal samples were stored in 2% K2Cr,O; so that coccidia could sporulate for identification. Nematodes were preserved in alcohol, cleared in lactophenol and identified using a compound microscope. Armed scoleces of preserved Tuenia specimens were removed, mounted and cleared in DeFauré’s medium for the examination of rostellar hooks. Identification of Zaenia specimens was made by counting and measuring the large and small rostellar hooks and TABLE 1. Helminth and protozoan parasites found in 46 feral cats from Kangaroo Island. Parasite Site Prevalence (%) Abundance (mean) Acanthocephala Oncicola pomatostomi Intestine 7 2-7 (4) Nematoda Aelurostrongylus abstrusus Lung 11 - Ancylostoma tubaeforme Intestine 15 1-16 (4) Cyathospirura dasyuridis Stomach 15 1-78 (98) Cylicospirura felineus Stomach 57 1-39 (7) Ollulanus tricuspis Stomach 2 6 Toxocara cati Intestine 76 1-53 (9) Cestoda Dipylidium caninum Intestine 4 25-31 (28) Spirometra erinacei Intestine 39 1-26 (5) Taenia taeniaeformis Intestine 63 1-31 (5) Trematoda Brachylaima cribbi Intestine 2 1 Protozoa Faeces Cryptosporidium f4 Giardia 2 Isospora felis 15 Isospora rivolta 4 Sarcocystis 7 Toxoplasma (serology) IHAT 87 82 TABLE 2. Comparison between the detection of helminths in the intestine and the detection of eggs in faecal samples. Helminth Intestine Faeces Number infected Positive for eggs Ancylostoma tubaeforme 7 3 Brachylaima cribbi ] 1 Cyathospirura dasyuridis & Cylicospirura felineus 26 4 Oncicola pomatostomi 3 0 Spirometra erinacei 18 14 Toxocara cati 35 28 Taenia taeniaeformis 29 13 comparing the data with those of Verster'!. An indirect the prevalence of these two parasites may be haemagglutination test (Toxo HAI commercial test kit {Fumouze]) and a direct agglutination test (Antigene Toxo- AD commercial test kit [bioMéricux]) were used for detecting Toxoplasma antibody in serum samples from 47 cats. A dilution of 1:80 was regarded as positive for the HAI test and 1:4 for the DA test following the manufacturers instructions. All of the cats examined were mature; six were regarded as young adults. Twenty-one cats were female and 25 male. Ten of the female cats were pregnant and two were in lactation. The majority of the parasites found (Table 1) have been reported previously in surveys of feral cats. This study confirms that Tuenia taeniaeformis, Spirometra erinacei and Toxocara cati are common parasites of feral cats. Ancylostoma tubaeforme is regarded as the common hookworm of cats and has been reported in feral cats from the Northern Territory’ and from around Sydney’. Cylicospirura felineus and Cyathospirura dasyuridis were found in tumour-like nodules on the stomach wall. Cylicospirura felineus was the predominant species and occurred together with Cyathospirura dasyuridis in seven cats. The trematode, Brachylaima cribbi, occurs in a variety of mammals and birds, has helicid snails as intermediate hosts and is infectious to man!?. Oncicola pomatostomi was detected in 65% of 188 feral cats from the Northern Territory suggesting that birds, as paratenic hosts, constituted a significant part of their diet’. In this study, the parasite was less common, however there is no information on the abundance of this parasite in birds from Kangaroo Island preventing further inference. The prevalence of Isospora felis was higher than that reported in feral cats elsewhere. Infections by /sospora spp. are considered age dependent, occurring more commonly in younger animals*5, Whilst Giardia sp. was found in a faecal sample from only one cat in this study, higher prevalence reported elsewhere! has implicated the feral cat in the transmission of the parasite to wildlife and man. It has been suggested that Cryptosporidium sp. recovered from domestic cats is not of zoonotic significance and the oocysts are smaller than those recovered from man'*. Furthermore, cats appear to carry different species such as C. baileyi and C. muris'* essentially from birds and rodents respectively. We were unable to determine the identity of the species detected in this study. McGlade ef a/.'> reported a significantly greater prevalence of Cryptosporidium sp, and Giardia sp. in cats using PCR in comparison to microscopy and consequently underestimated here. The dimensions of Sarcocystis sp. sporocysts here are consistent with those of S. gigantea (x length 13.2 tm x ¥ width 9.4 um, n = 22), a species infecting sheep. Faecal examination may also underestimate the prevalence of toxoplasmosis. The shedding of Toxoplasma gondii oocysts in faeces occurs for a short time and only once following infection, usually when the young cat begins hunting rodents and birds'*. Serological tests indicate previous exposure to the parasite. The high prevalence of Toxoplasma antibodies in feral cats is consistent with the high prevalence of antibodies in sheep on Kangaroo Island. There was a poor correlation between the detection of adult nematodes and the detection of nematode eggs in faecal samples (Table 2). Zoxocara cati eggs were not detected in faecal samples on seven occasions, six being due to infections with immature nematodes. Hookworm eggs were detected in the faeces of three of the seven cats infected. Of the four negative samples, three contained a single worm and the fourth contained three, non-gravid female worms. The detection of cestode and acanthocephalan eggs in faeces of cats infected with adult worms was similarly inconsistent. Infections of up to 31 cestodes remained undetected by faecal examination. These results indicate that coprological surveys for helminth parasites may underestimate the prevalence of infections. Aelurostrongylus abstrusus are small worms, <10 mm long, occurring in lung parenchyma. Adults were not recovered from lungs when examined macroscopically, however, using the Baermann technique, Aelurostrongylus larvae were recovered from lung tissue and faecal samples of positive cats. The information presented in this study identified similar helminth and protozoan parasites in feral cats from Kangaroo Island to those detected elsewhere in Australia and confirms that feral cats may act as reservoirs for these parasites. The study suggests that the feral cat is responsible for the high prevalence of Yoxoplasma in sheep on Kangaroo Island and poses a potential disease risk to wildlife. It also establishes that feral cats in Australia are infected with Cryptosoporidium. 'O’Callaghan, M. G., Moore, E. & Ford, G. E. (1984) Helminth and arthropod parasites from dogs and cats in South Australia. Aust. vet. Practit. 14, 159-161. 2Coman, B. J. (1972) A survey of the gastro-intestinal parasites of the feral cat in Victoria. Aust. vet. J. 48, 133-136. ‘Ryan, G. E. (1976) Gastro-intestinal parasites of feral cats in New South Wales. Aust. vet. J. 52, 224-227. ‘Gregory, G.G. & Munday, B. L. (1976) Internal parasites of feral cats from the Tasmanian midlands and King Island. Aust. vet. J. 52, 317-320. °Coman, B. J., Jones, E. H. & Driesen, M. A. (1981) Helminth parasites and arthropods of feral cats. Aust. vet. J. 57, 324-327. °Coman, B. J., Jones, E. H. & Westbury, H. A. (1981) Protozoan and viral infections of feral cats. Aust. vet. J. 57, 319-323. 'O’Callaghan, M. G. & Beveridge, I. (1996) Gastro-intestinal parasites of feral cats in the Northern Territory. Trans. R. Soc. S. Aust. 129, 175-176. ‘O’Donoghue, P. J., Riley, M. J. & Clarke, J. F. (1987) Serological survey for 7oxoplasma infections in sheep. Aust. vet. J. 64, 40-45. °O’ Donoghue, P. J. (1995) Cryptosporidium and cryptosporidiosis in man and animals. Int. J. Parasitol. 25, 139-195. "Georgi, J. R. (1974) Parasitology for Veterinarians (W. B. Saunders, Philadelphia. USA.) pp 134-135. 83 "Verster, A. (1969) A taxonomic revision of the genus Taenia Linnaeus, 1758 s. str, Onderstepoort J. vet. Res. 36, 3-58. "Butcher, A. R. & Grove, D. I. (2001) Description of the life-cycle stages of Brachylaima cribbi n. sp. (Digenea: Brachylaimidae) derived from eggs recovered from human faeces in Australia. Sys. Parasitol. 49, 211-221. Milstein, T. C. & Goldsmith, J. M. (1997) Parasites of feral cats from southern Tasmania and their potential significance. Aust. ver. J. 75, 218-219. “Sargent, K. D., Morgan, U. M., Elliot, A. & Thompson, R. C. A. (1998) Morphological and genetic characterisation of Cryptosporidium oocysts from domestic cats. Vet. Parasitol. 77, 221-227. 'McGlade, T. R., Robertson, I. D. Elliot, A. D., Read, C. & Thompson, R. C. A. (2003) Gastro-intestinal parasites of domestic cats in Perth, Western Australia. Vet. Parasitol. 117, 251-262. ‘Dubey, J. P. & Beattie, C. P. (1988) Toxoplasmosis of animals and man. (CRC Press Boca Raton, Florida. USA) pp 18-19. M. O’CALLAGHAN, South Australian Research and Development Institute, GPO Box 397, Adelaide, SA 5001, E-mail: ocallaghan.micko@saugov.sa.gov.au, J. REDDIN, Primary Industries and Resources, PO Box 469 Murray Bridge, SA 5253 and D. LEHMANN, Kangaroo Island Veterinary Clinic, Kingscote, SA 5223. ao Ss eo 6)VOL. 129, PART 2 30 NOVEMBER, 2005 RSA . A special issue of the Transactions of the Royal Society of South Australia Incorporated containing papers on the 2004 Expedition to the Althorpe Islands, South Australia Guest Editors: Sue Murray-Jones (BSc, PhD) Scoresby Shepherd (B.A., LL.B., M.Env.St., Ph.D.) INCORPORATING THE Records of the South Australian Museum Contents. Murray-Jones, S. & Shepherd, S. A. An expedition to the Althorpe Islands, South Australia: Introductory narrative and conservation recommendations Zang, W. L. Geology of Althorpe Island Radford, A. Human settlement on Althorpe Island and condition of the lighthouse complex - Lawley, E. F. & Shepherd, S. A. Land use and vegetation of Althorpe Island, South Australia, and a floristic comparison with South Neptune Islands Lawley, E. F., Lawley, J. J. & Page, B. Effects of African boxthorn removal on native vegetation and burrowing of short-tailed shearwaters on Althorpe Island, South Australia _- Baldock, R. N, & Womersley, H. B. S. Marine benthic algae of the Althorpe Islands, South Australia Baker, J. L., Edgar, G. J. & Barrett, N.S. Subtidal macroflora of Althorpe and Haystack Islands, South Australia Benkendorff, K. Intertidal molluscan and echinoderm diversity at Althorpe Islands and Innes National Park, South Australia Staples, D. A. Pycnogonida of the Althorpe Islands, South Australia -— — Walker-Smith, G. K. A new species of Neopeltopsis (Copepoda, Harpacticoida, Peltidiidae) from Althorpe Island, South Australia -— — - Shepherd, S. A., Edgar, G. J. & Barrett, N. S. Reef fishes of the Althorpe Islands and adjacent coasts of central South Australia Shepherd, S. A., Teale, J. & Muirhead, D. Cleaning symbiosis among inshore fishes at Althorpe Island, South Australia and elsewhere -— - -— — Shepherd, S.A. & Brook, J. B. Foraging ecology of the western blue eases Achoerodus gouldii, at the Althorpe Islands, South Australia — - Einoder, L. D. & Goldsworthy, S. D. Foraging flights of short-tailed shearwaters (Puffinus tenuirostris) from Althorpe Island: assessing their use of neritic waters PUBLISHED AND SOLD AT THE SOCIETY’S ROOMS SOUTH AUSTRALIAN MUSEUM, NORTH TERRACE, ADELAIDE, S.A. 5000 85 90 94 100 111 116 128 145 158 170 183 193 202 209 Transactions of the Royal Society of S. Aust. (2005), 129(2), 85-89. AN EXPEDITION TO THE ALTHORPE ISLANDS, SOUTH AUSTRALIA: INTRODUCTORY NARRATIVE AND CONSERVATION RECOMMENDATIONS by S. MurrAy-JONES* & S. A. SHEPHERD* Summary Murray-Jones, S, & SHEPHERD, S. A. (2005) An expedition to the Althorpe Islands, South Australia: Introductory narrative and conservation recommendations. Trans. R. Soc. S. Aust. 129(2), 85-89, 30 November, 2005. A general overview of the 2004 Expedition to the Althorpe Islands group is provided, including a description of the islands, an outline of the research program, a brief introduction to the papers presented in this volume, and some recommendations for management. Key Worps: Southern Australia, Althorpe Is, lighthouse, geology, terrestrial ecology, heritage, marine conservation. Description of islands and habitats The Althorpe Is consist of one main island and five small islets (Fig. 1), of which Althorpe I. is the southernmost and largest (96 ha, and 93m high). The five small islets (the Western Isles) are off the western side of Althorpe I. (Baker 2005). To the NE, and closer to the coast, are Seal |., and Haystack I. (43 m high). Together, the islands and islets are informally known as the Althorpe Is Group. The islands are the remains of once prominent granite hills surmounted by Bridgewater calcarenite. The islands were probably isolated from the mainland some 9 — 10 000 yrs ago (Zang 2005). All islands in the Althorpe group have sub-tidal granite basement reefs, with some calcarenite blocks present in the nearshore area in some places, where they have eroded from the island “capping” above. On Althorpe I., jointing in the massive granite humps has resulted in gaping, many-branched crevasses and chasms along the coastline, that alternate with long fingers of rock, in turn dissected into segmented chains of islets. The small Western Isles are the eroded and segmented remnants of a granite rise along a series of joints or dykes (Robinson ef al. 1996). Mooring Bay, bounded by cliffs on the NE side of Althorpe I., provides the only anchorage * Coast Protection Branch, Natural & Cultural Heritage, Department for Environment & Heritage (DEH), GPO Box 1047, Adelaide, South Australia 5001. * South Australian Research and Development Institute (SARDI), Aquatic Sciences, PO Box 120, Henley Beach, South Australia $022. ' The names given in Fig. | were gazetted in the SA Government Gazette on 13th Oct. 2005 (p3685). The main island of the Althorpe Is group (commonly known as Althorpe I.) is in fact unnamed. A proposal to name it Laubadere I., after the name given to it by Freycinet in 1824, is under consideration by the Geographical Names Advisory Committee of the Government of SA. (Robinson ef al. 1996). The names adopted in Fig. | are mainly those in use by lighthouse staff and the Friends of the Althorpe Is Conservation Park, with a few adopted by the 2004 expedition. All names have been accepted by the Geographical Names Advisory Committee and await ratification!. Althorpe Islands The base of the main island is made up of granite rock rising up to ~15 m above sea level. Sitting on top of this granite base are layers of sandstone, which are the remains of ancient sand dune systems that once covered the region. All this is capped by layers of calcarenite. The geology of Althorpe I. is described in more detail in Wang (2005, this volume). Althorpe I. rises steeply to a plateau 93 m above sea level. On the plateau there is an abandoned airstrip, a lighthouse, and three lighthouse keepers’ cottages. The northern islet of the Western Isles actually consists of four islets separated by narrow channels. The base of all four islets is granitic, with the two middle islets capped with calcarenite. These form two peaks rising to 23 m and 22 m respectively above sea level. The two outer islets are bare granite only. The southern islet, also informally called ‘Tern I” consists of bare granite projecting 12 m above sea level. Haystack Island This island has a narrow wall of sheer cliffs, made of sandstone that has been undermined and indented. The cliffs sit on a supporting ridge of granite that lies well below sea level. The island has been eroded into a series of tall lobes connected by thin necks of rock, two of the lobes rising to 44 m and 41 m respectively above sea level. The sandstone is capped with limestone, which forms a series of stepped terraces on top of the domes (Robinson e¢ al. 1996). 86 S. MURRAY-JONES & S, A. SHEPHERD Stenhouse Bay e! Innes NP : Haystack Island Seal Island __Althorpe Island Fig. 1. Map of the Althorpe Islands Group. Seal Island This island rises 35 m above sea level on a hump of granitic gneiss, intruded by amphibolite dykes. The sea has eroded deep grooves and indentations that channel the swell into a surge that reaches the island platform of sandstone capped with limestone. These surges undermine the limestone cap, the fringe of which is strewn with jagged fragments of rock from collapsed overhangs. The broadest joints and dykes have been worn to deep indentations, or completely penetrated, to form the segmented islets off the western tip of Seal I. (Robinson et al. 1996). Marine habitat The habitat diversity in the area includes: near- shore benthic granitic basement reefs with a diversity of forms such as platforms, ledges, boulders, caves, chasms and crevasses, overhangs, vertical rock walls, isolated reef outcrops and near-shore fringing reef, patches of broken calcarenite blocks and rubble; areas of small sandy beaches in Mooring Bay; mixed sand / granite reef / calcarenite block / rubble reef intertidal habitats; subtidal seagrass beds (e.g. NE Althorpe I. and Haystack I.); and benthic sand habitat (Robinson et al. 1996; Edyvane 1999; Edyvane & Baker 1998). 2004 Expedition The Royal Society of South Australia, with a long history of sponsoring expeditions to the State’s offshore islands (Robinson ef al. 2003), promoted this expedition, with major funding support by South Australian Research and Development Institute and Department for Environment and Heritage. Previous visits to the Althorpe Is include some bird observations by Perryman (1937), and a brief survey by a National Parks & Wildlife SA party on 24 — 25 Nov. 1982 of the plants, mammals, birds and reptiles (Robinson et al. 1996), Since 1982, the Friends of the Althorpe Is Conservation Park (FoATCP) have INTRODUCTORY NARRATIVE & CONSERVATION RECOMMENDATIONS ON ALTHORPE ISLANDS, SA 87 made regular visits to the island, and accumulated much additional data (e.g. Lawley & Shepherd 2005). The scientific program that is the focus of this special edition took place from 29 Jan. to 12 Feb. 2004, and comprised a terrestrial and a marine component. The terrestrial party occupied the lighthouse cottages on Althorpe I., with members of FoAICP, while the marine party stayed either aboard the fisheries research vessel Ngerin, present from | — 8 Feb. 2004, or on shore. Small boats were used to ferry scientists between islands. The following persons took part in the expedition’s scientific programs: Terrestrial program Alison Radford (land-use history and archaeology), assisted by Vicki Cheshire; Rob Fitzpatrick (soils) assisted by Alison Fitzpatrick; Brad Page and Shelley Harrison (seabird feeding), assisted by Vicki Cheshire and Erika Lawley; vegetation surveys (Erika and John Lawley). Marine program Robert Baldock, (algal systematics), Anthony Cheshire (algal ecophysiology); Bayden Russell and Jarrod Stehbens (ecology of marine macroalgae); Scoresby Shepherd (feeding ecology of groper; cleaning behaviour of fishes); Kirsten Bilgmann and Sue Gibbs (dolphin genetics and habitat use); Alastair Hirst (macroepifauna of algae); Rob Lewis (rock lobster population assessment, assisted by James Brook); David Staples (systematics of pycnogonids); Chris Halstead, Graham Edgar, Neville Barrett, Ali Bloomfield and James Brook (biodiversity assessment); Thierry Laperousaz (animal collections for SA Museum); Kirsten Benkendorff and Alex Gaut (intertidal biodiversity). Sue Murray-Jones assumed responsibility for the logistics and assisted in the algal ecophysiological, fish and dolphin habitat work. Friends of Althorpe Islands Conservation Park (FoAICP) FoAICP was formed in 1996, after attempts by DEH to find a lessee for the Island failed. The current membership is about 150. The Friends undertake quarterly visits to the island throughout the year to carry out maintenance on buildings, the solar/wind power energy system, equipment and pathways, and other tasks such as weed and feral animal control. Funding for these is usually obtained through project grants. Participants and their roles included: John Lawley (island caretaker, logistics and coxswain duties); Les Harper (coast guard radio); John Webster and David White (cottage cleaning and preparation, water, electricity and mouse control); Erika Lawley (kitchen coordination, guide to island locations, shearwater research assistant); Stefania Madonna and Christine Lawley (domestic duties, terrestrial work, weed control). Operations support Dave Kerr (Master of Ngerin), Chris Small (Ngerin Mate/Engineer), and Michael Clark (Dive Supervisor), all from SARDI; and from DEH, Rick James (acting Mate); Danny Doyle, Caroline Paterson, and Tim Collins (general support), as well as support from FoAICP. Results The terrestrial studies in this special issue have consolidated historical and biological information not generally accessible. These include: the history of human occupation of Althorpe I. and the condition of the lighthouse complex (Radford 2005); and a brief history of the land use of the island, with its likely effects on the vegetation (Lawley & Shepherd 2005). The response of the island’s vegetation and the nesting of the short-tailed shearwater, Puffinus tenuirostris following removal of an African boxthorn infestation is considered by Lawley ef al. (2005), while Einoder & Goldsworthy (2005) describe the at-sea movement and marine habitat use of the shearwaters. In addition, a description of the geology (Zang 2005) of Althorpe I. is included, although the study was done during independent visits to the Island. The marine studies in this issue have substantially extended knowledge of benthic and intertidal communities of the region. The diversity of intertidal molluscs and echinoderms is described by Benkendorff (2005). Staples (2005) describes the pycnogonid fauna of the area, including two new species, Pseudopallene watsonae and P. inflatus. Walker-Smith (2005) describes a new species of Neopeltopsis and documents the occurrence of other species of Peltidiidae from Althorpe I. Baldock & Womersley (2005) summarize past and recent algal collections by providing a species list of benthic algae. Baker et al. (2005) describe the algal diversity and community structure at Althorpe and Haystack Is, using data collected both in 1993 and during the 2004 expedition. A description of the reef fishes of Althorpe and Haystack Is, as well as of some mainland control sites, is provided by Shepherd er al. (2005a). Shepherd & Brook (2005) describe the feeding behaviour of Western Blue Groper, while Shepherd ef al. (2005b) describe cleaning symbioses amongst inshore fishes, both for the Althorpe Is and elsewhere, the first such account for South Australia. 88 S. MURRAY-JONES & S. A. SHEPHERD Recommendations After the boxthorn removal program described by Lawley et al, (2005), native vegetation was found to recolonise bare patches, and there is some evidence that shearwater burrows may have expanded into bare areas as well. Any boxthorns re-introduced to the islands by birds flying from mainland infestations will continue to need to be controlled, although the gradual re-establishment of native plant species in disturbed areas may ultimately limit opportunities for new boxthorn plants to establish. The destruction or degeneration of habitat, or the introduction of predators, can cause the demise of many native species, both plants and animals. The monitoring of introduced species, and control where necessary, will help conserve both the native vegetation and faunas of these islands. A systematic and sustained program to eradicate the introduced population of house mice from the island could be planned as part of future conservation management. Generally the light-station complex is in fair condition. The light-station and other evidence of human occupation should be conserved to protect their cultural significance and adapted as necessary for ongoing use. Any new development should be to the south of the existing cottages. Regular maintenance checks of the building should be undertaken to mitigate the effects of weather, the saline environment and impacts of plants and animals. Limited access to the island by air and water should be maintained. If possible occupancy should be encouraged with comprehensive waste management strategies developed. The marine studies at Althorpe and Haystack Is support the conclusions of earlier island studies (e.g. the Encounter 2002 expedition; Robinson ef al. 2003) that islands support a greater diversity of habitats than coastal locations. This is due partly to a greater range of depths close to shore, and the higher productivity of islands, which can capture the production of surrounding waters and upwelling regions (the offshore island effect). The greatest biodiversity would be protected by maximizing the number of habitats conserved, as shown by Benkendorff (2005). As wave energy and light are key forcing functions for benthic communities, a Marine Protected Area should contain representatives of habitats on exposed to sheltered areas from the intertidal to deep water, embracing all islands in the group. Acknowledgments In addition to the major sponsors of the expedition, SARDI (Aquatic Sciences) and DEH, we thank Rob Lewis and Anthony Cheshire for their strong support. FoAICP supported the terrestrial program with a grant of $3,520 from the Australian Government Envirofund, sponsored by Friends of Parks Inc., and the Department for Environment and Heritage provided a small boat and various support staff. We thank the Ngerin crew, and all support staff for their contribution, and in particular Mick Clark (Dive Supervisor), for ensuring the safe conduct of a diverse diving program. References Baker, J. L. (2005) “Towards a System of Ecologically Representative Marine Protected Areas in South Australian Marine Bioregions-Technical Report’. Report to Dept. Environment and Heritage, S. A. Accessed 8 Aug. 2005. URL http://www.environment.sa.gov.au/ coasts/pdfs/mpa_report/2005report.pdf Baker, J. L., EpGAR, G. J., & BARRETT, N. S. (2005). Subtidal macroflora of Althorpe and Haystack Islands, South Australia. Trans. R. Soc, S. Aust. 129, 128-144. BALbock, R. N, & WomeErSLEY, H. B. S. (2005). Marine benthic algae of the Althorpe Islands, South Australia, Trans. R. Soc. S. Aust. 129, 116-127. BENKENDORFF, K. (2005). Intertidal molluscan and echinoderm diversity at Althorpe Island and Innes National Park, South Australia. Trans. R. Soc. S. Aust. 129, 145-157. EpyvaNneg, K, S, (1999) Conserving Marine Biodiversity in South Australia. Part 2 — Identification of areas of high conservation value in South Australia. Adelaide, PIRSA/SARDI. & BAKER, J. L. (1998) Marine benthic survey of Investigator Strait - Gambier Isles, South Australia. Report to Environment Australia (Marine Protected Areas Program) (Stage 4). SARDI 34 pp. Etnoper, L.D. & GoLpswortHy, 8S. D. (2005) Foraging flights of short-tailed shearwaters (Puffinus tenuirostris) from Althorpe Island: assessing their use of neritic waters. Jrans. R. Soc. S. Aust. 129, 209-216. LIGHTHOUSES OF AUSTRALIA INC. (2005) Lighthouses of South Australia, Accessed 7th June. http://www.light house.net.au/lights/SA/Althorpe%20Island/Althorpe%2 OIsland.htm#History LAWLEY, E. F. & SHEPHERD, S. A. (2005) Land use and vegetation of Althorpe Island, South Australia, and a floristic comparison with South Neptune Islands. 7rans. R. Soc. S. Aust. 129, 100-110. , LAWLEY, J. J. & Pace, B. (2005) Effects of African boxthorn removal on native vegetation and burrowing of short-tailed shearwaters on Althorpe Island, South Australia. Trans. R. Soc. S. Aust. 129, 111-115. PERRYMAN, C. E, (1937) Notes from Althorpe Islands. S.A. Ornith. 14(1), 14-19. Raprorb, A. Human settlement on Althorpe Island and condition of the lighthouse complex. Zrans. R. Soc, S. Aust. 129, 94-99. Rosinson, A. C. CANTY, P., Mooney, P & RUDDUCK, P. (1996) “South Australia’s Offshore Islands”. Australian Heritage Commission. (Australian Government Publishing Service). INTRODUCTORY NARRATIVE & CONSERVATION RECOMMENDATIONS ON ALTHORPE ISLANDS, SA 89 , Murray-JONES, S., SHEPHERD, S. A. & WACE, N. M. (2003) The Encounter 2002 expedition to the Isles of St Francis, South Australia: Formation of the islands, introductory narrative & marine conservation recommendations. Trans. R. Soc. S. Aust. 127, 69-73. SHEPHERD, S. A. & Brook, J. B. (2005) Foraging ecology of the western blue groper, Achoerodus gouldii at the Althorpe Islands, South Australia. Trans. R. Soc. S. Aust. 129, 202-208. , EDGAR, G. J., & BARRETT, N. S. (2005a) Reef fishes of the Althorpe Islands and adjacent coasts of central South Australia. Trans. R. Soc. S. Aust. 129, 183-192. , TEALE, J. & MUIRHEAD, D. (2005b) Cleaning symbiosis among inshore fishes at Althorpe Island, South Australia and elsewhere. Trans. R. Soc. S. Aust. 129, 193-201. StapLes. D. A. (2005) Pycnogonida of the Althorpe Islands, South Australia. Trans. R. Soc. S. Aust. 129, 158-169. WALKER-SMITH, G. K. A new species of Neopeltopsis (Copepoda, Harpacticoida, Peltidiidae) from Althorpe Island, South Australia. Trans. R. Soc. S. Aust. 129, 170-182. ZANG, W-L. (2005) Geology of Althorpe Island. Trans. R. Soc. S. Aust. 129, 90-93. Transactions of the Royal Society of S. Aust. (2005), 129(2), 90-93. GEOLOGY OF ALTHORPE ISLAND by W. L. ZANG* Summary ZANG, W. L. (2005) Geology of Althorpe Island. Trans. R. Soc. S. Aust. 129(2), 90-93, 30 November, 2005. Althorpe Island is located near the southern margin of the Gawler Craton and contains a Palaeoproterozoic granite basement, which is capped by Quaternary calcarenite of the Bridgewater Formation. The basement (Donington Suite) was multiply deformed and intruded by the Tournefort Dyke Swarm. The Bridgewater Formation contains mainly calcarenite and palaeosol, and in the middle part, some shelly fossils occur. The island was probably isolated from the mainland some 9-10 000 yrs ago. Introduction The geology of Althorpe Island is relatively well known and the rocks on the island can be correlated with those on Yorke Peninsula. Tectonically, the island is located within the southern Gawler Craton, which contains rocks ranging from ~2550 Ma (million years before present) to ~1500 Ma. Early geological mapping by Crawford (1965) suggested the island contains metamorphic basement and Quaternary sedimentary cover; the basement rocks were tentatively assigned to the Archaean. Subsequent surveys by Major (1973) and Rankin et al. (1991) interpreted the basement rocks to be of Palaeoproterozoic age. Generally the island comprises a coastal platform of granite basement (Donington Suite, 1850 — 1855 Ma), capped by Quaternary Bridgewater Formation. An updated geological map of the island and mainland has been published recently (Zang 2003; Fig. 1). Basement Rocks Basement rocks comprise mainly adamellite, and are best displayed in the southern coastal areas (Fig. 2a). The adamellite is pink to pink-brown and contains quartz (30 — 35%), microcline (20 — 45%) plagioclase (15 — 30%), biotite (3 — 5%), magnetite (1 — 4%) and accessory hornblende, sphene, apatite and zircon. An imprecise Rb-Sr date derived from adamellite from Althorpe I. and nearby islands suggests an age older than 1794 Ma (Webb et al. 1986). A fine-grained biotite-hornblende micro- * Geological Survey, Department of Primary Industries and Resources, South Australia. Email: zang.wen-long@saugov.sa.gov.au Purvis, A. C. (1999) Mineralogical report Nos 7770 and 7834. Pontifex and Associates Pty Ltd. Petrographic reports (unpublished). Schaefer, B. F. (1998) Insights into Proterozoic tectonics from the southern Eyre Peninsula, South Australia. Ph.D. thesis, University of Adelaide. w adamellite also occurs on the island as isolated pods or dykes in the host adamellite and the presence of orthopyroxene (~2%) suggests high igneous or metamorphic temperatures (Purvis'). The oldest rock on the island is migmatitic paragneiss in the northern part of the island. The paragneiss is layered with orthoclase-plagioclase- rich bands of garnet-biotite-spinel-sillimanite- corundum and _ plagioclase-orthopyroxene-quartz assemblages. This was possibly derived from a silica-deficient siltstone (Purvis'). The mineral assemblages suggest upper amphibolite to granulite facies metamorphism (700 — 800 °C). Similar rocks at Corny Point, Yorke Peninsula have been dated by U-Pb zircon geochronology: cores of zircon in paragneiss range in age from ~2400 Ma to ~1920 Ma, suggesting a maximum age for the sediment protoliths, and newly-grown outer zones indicate a metamorphism age of 1845 Ma (Zang & Fanning 2001). Dark-coloured dykes of Jussieu Metadolerite, intruding the Donington Suite granite, occur in the north of the island. The metadolerite is mylonitic and contains relics of back-veining textures, in which the mafic rock is broken into angular fragments by the injection of felsic magma from the host rock. This suggests that the emplacement and crystallisation of the mafic dykes had proceeded when host rocks were still partly molten, and therefore syntectonically (~1850 Ma). There are also several ultra-mafic (S10, = 39.2%, Rankin ef al. 1991) enclaves which comprise hornblende, clinopyroxene, plagioclase and opaques. The ultra-mafic rocks are fine to medium- grained and granoblastic; several outcrops in the field record the presence of an earlier fabric (clinopyroxene — hornblende elongation). Undeformed, younger conjugate mafic dykes (Tournefort Dyke Swarm) trend NW-SE and NE-SW, showing a conjugate relationship. These dykes are generally altered and consist mainly of plagioclase, hornblende, clinopyroxene, biotite, sphene, opaques etc. They may be as old as ~1810 Ma (Schaefer?). GEOLOGY OF ALTHORPE ISLAND 28,981 91 Ae Af <= re b St Kilda Formation ® Ultramafics | Bridgewater __| Formation 6 Microademellite 7 7) Tournefort Jussieu 7 /7\ Dyke Swarm* Metadolerite ¢ S RAS Donington © Paragneiss 9 bi 0 Suite 3 METRES Fig. 1. Geological map of Althorpe I. Sedimentary Cover Quaternary Bridgewater Formation is ~90 m thick on Althorpe I. (Fig. 2b), and comprises a succession of Pleistocene beach-dune calcarenite complexes (common along coastal settings in South Australia, Sprigg & Boutakoff 1953). Three units can be recognised on Althorpe I. The lower unit contains Late-Middle Pleistocene medium to coarse-grained, large-scale bedded calcarenite, interbedded with red- brown sandy clay (palaeosols). Wilson? suggested that this unit could be as old as 0.7 Ma. The middle unit is equivalent to the Glanville Formation, from which two pulmonate gastropods were reported (Ludbrook 1973). The formation was dated at 0.1-0.13 Ma by the Uranium, TL and Sr methods (Belperio 1995; Zang 2003). The upper unit contains fine to medium- grained aeolianite and bioclastic calcarenite and was dated by the '*C method at Gleesons Landing, SW * Wilson, C. C. (1991) Geology of the Quaternary Bridgewater Formation of southwest and central South Australia. Ph.D. thesis, Flinders University of South Australia). Yorke Peninsula. Charcoal from a solution hole was dated at 22400 + 800 years and an underlying shell horizon at >27 000 years B.P. (Wilson*). The dates may limit the uppermost Bridgewater Formation to ~22 000 years. The Bridgewater Formation on the island is capped by calcrete. The calcarenites or calcareous aeolianites of the lower Bridgewater Formation are considered to have been deposited as a series of spatially separated, shoreline-parallel dunes or barriers. Morphologically, individual dunes display weakly lenticular to tabular geometries and are capped by a red-brown palaeosol layer. Generally, the calcarenites consist of homogeneous, fine to medium-grained, well-sorted bioclastic grainstone with tiny fragments of molluscan, bryozoan, foraminiferal, echinoid, algal, peloidal and lithoclast allochems (Wilson’). Near the jetty, modern beach sand (St Kilda Formation), mostly silica sand with some carbonate shells, has been deposited. This beach provides the only safe boat-landing spot on the island. The top of the island is capped by a thin layer of W. L. ZANG Fig. 2. A) basement granite of the Donington Suite, Mooring Bay, Althorpe 1. (1950-1855 Ma); B) Quaternary sediments of the Bridgewater Formation, Mooring Bay, Althorpe I. (~0.5 Ma). GEOLOGY OF ALTHORPE ISLAND 93 calcrete. Calcrete and soil weathering profiles are extensively developed within the calcareous aeolianites and play an important role in the stabilisation and preservation of these sediments. Calcretisation was obviously developed during several pedogenic stages of biochemical and physiochemical processes to a completely indurated or fossilised profile. Calcrete was used for building material in the early years and is still mined for road building material on Yorke Peninsula. Palaeogeography Althorpe I. and the surrounding isles are considered to have been linked with Yorke Peninsula during deposition of the Bridgewater Formation ca. half a million years ago. Repetitive transgressions of the sea across the continental shelf caused successive calcarenite barriers to form (Belperio 1995). Climate and sea level change were driven by repetitive global Quaternary glaciations and deglaciations. Complete separation of the islands from the mainland by gulf waters may have occurred as recently as 9-10 000 yrs ago when the land bridge was eroded and submerged during the latest sea level rise. Some possible marsupial bones in shallow caves on the island may support this suggestion. Discussion The geological history of the Althorpe Is can be traced back to ~2000 Ma years ago when the region of Yorke and Eyre Peninsulas was a palaeo-ocean bounded by the Archaean terrains of the northern Gawler Craton and southern Antarctic continent. Remnants of metamorphosed siliciclastic and carbonate deposits on the island and Yorke Peninsula may suggest a shallow marine environment in the areas (Zang & Fanning 2001). Those shallow marine deposits were intruded by granites of the Donington Suite during ~1850 - 1855 Ma and metamorphosed by deep-crust sourced heat, up to ~800 °C. The granites, in return, were intruded by mafic Tournefort Dyke Swarm during ~1800 Ma. No geological history has been recorded on the island until the middle Quaternary. Middle Pleistocene (~0.5 Ma) deposits of the lower Bridgewater Formation in the region might be related to a short glacier meltdown and deposition of regressive calcarenites, of which the uppermost is always weathered or veneered by palaeosol. The glacial-transgression deposits of the Upper Pleistocene Glanville Formation reached the Althorpe Is, and as far as the southern margin of Yorke Peninsula. A Late Pleistocene uplift event is recognised by the formation of Peesey Swamp to the east of Warooka township, Yorke Peninsula, where the Middle Pleistocene Bridgewater Formation forms the ridge to the west of the Warooka Fault zone, The Glanville Formation and overlying aeolianites were deposited in the Peesey Swamp depression. Related to this uplift, the uppermost Bridgewater Formation in the areas contains mainly palaeosol and weathered calcareous sand. The Holocene St Kilda Formation is probably deposited following more recent — glacial transgression, starting some 6000-7000 years ago in South Australia (Belperio 1995). This transgression reached southern Yorke Peninsula, since Posidonia- rich deposits occur in the Marion Lake area (von der Borch et al. 1977). The relative sea level in the Althorpe Is and southern Yorke Peninsula areas seems to have risen slightly during the last 6000 years. References Bevcrerio, A. P. (compiler) (1995) Quaternary. /n: Drexel, J. F & Preiss, W. V. (Eds), The geology of South Australia, Vol. 2: The Phanerozoic. S. A. Geological Survey Bulletin 54, 219-281. Crawrorb, A. R. (1965) The geology of Yorke Peninsula. Geological Survey. S. A. Department of Primary Industries and Resources. Bulletin 39, 1-96. Lupsrook, N. H. (1973) Two pulmonate gastropods from Bridgewater Formation, Althorpe Island. S.A. Department of Mines. Report Book 73/00206. Major, R. B. (1973) Preliminary report — geology of islands of the western continental shelf of South Australia. S. 4. Department of Mines. Report Book 73/226. RANKIN, L. R., FLINT, R. B. & BELPERIO, A. P. (1991) Precambrian geology of islands in the Investigator Strait area, South Australia. S. A. Department of Mines and Energy. Report Book 91/57, Sprica, R. C. & BouTakorr, N, (1953) Summary report on the petroleum possibilities of the Gambier Sunklands. Mining Review, Adelaide 95, 41-62. VON DER Borcu, C. C., BOLTON, B. & WARREN, J. K. (1977) Environmental setting and microstructure of subfossil lithified stromatolites associated with evaporites. Marion Lake, South Australia. Sedimentology 24, 693-708. Wess, A. W., THOMSON, B. P., BLissetTT, A. H., DALY, S. J., FLINT, R. B. & Parker, A. J. (1986) Geochronology of the Gawler Craton, South Australia. Aust. J Earth Sciences 33,119-143. ZANG, W. L. (2003) Maitland Special Map Sheet. S. A. Geological Survey. Geological Atlas 1:250 000 Series, sheet S153-12 and portion SI53-16. & FANNING. C. M. (2001) Age of the Kimban Orogeny revealed — U-Pb dates on the Corny Point Paragneiss, Yorke Peninsula, South Australia. MESA Journal 23, 28-33. Transactions of the Royal Society ofS. Aust. (2005), 129(2), 94-99. HUMAN SETTLEMENT ON ALTHORPE ISLAND AND CONDITION OF THE LIGHTHOUSE COMPLEX by A. RADFoRD! Summary Raprorp, A. (2005). Human settlement on Althorpe Island and condition of the lighthouse complex. Trans. R. Soc. S. Aust, 129(2), 94-99, 30 November, 2005. Althorpe I. became important for coastal navigation with construction of the lighthouse in 1879. The island was occupied from then until 1991 when the light was automated. Evidence of human occupation during that period included: various buildings associated with the lighthouse, runway, jetty and flying fox, grave sites and guano mining. The present significance of the island in terms of its cultural heritage, and other values is discussed. Key Woros: Althorpe I., history, lighthouse, jetty, flying fox, exploration, shipwrecks, South Australia. Introduction The Althorpe Is are located 8 km from Cape Spencer at the southern tip of Yorke Peninsula. During the period of European exploration the island was first visited by Matthew Flinders (Cooper 1953). Louis de Freycinet, cartographer to Nicholas Baudin’s French expedition, named the island group “Iles Vauban” (Brown 2004) and the main island Ie Laubadére (Péron 1824). About 10 years later, when Flinders drew up his maps, the island and the nearby Western Isles were named Althorpe? Islands. They were named after Viscount of Althorp, son of Earl George John Spencer (Farlex Inc. 2005)>. Earl George Spencer, 2nd Earl Spencer, was then First Lord of the Admiralty and Spencer Gulf and Cape Spencer were named after him. The main island is a calcarenite plateau with cliffs ~90 m high over a granitic base (Zang 2005), and provides limited opportunities for access. The beach at Mooring Bay on the northern side was used to construct a jetty and obtain access. Althorpe I. became an important part of South Australia’s coastal navigation with the development of a lighthouse and associated structures in 1879. This paper provides a brief historical review of the European occupation of Althorpe I., lists the shipwrecks that have occurred, and describes the present state of the island’s various structures, which comprise its heritage values. Heritage Branch, Department for Environment and Heritage, GPO Box 1047, Adelaide, SA 5001. Email: Radford.Alison@saugov.sa.gov.au The ’e was probably a spelling mistake. The courtesy title of the eldest son of the Earl Spencer is Viscount Althorp. Conservation Plan for Althorpe Island Lightstation (1991) Australian Construction Services. Es Methods Historical sources were examined and combined with a site visit on 2 — 5 Feb. 2004, as part of the 2004 expedition (Murray-Jones & Shepherd 2005), in order to: (1) update the 1991 Conservation Plan‘; (2) provide a dilapidation report on the state of the lighthouse and its precincts, keepers’ cottages, associated sheds, and the jetty precinct; and (3) prepare policies for the future of the built form and shipwrecks in the area. Other evidence of human occupation including guano mining, grave sites, and associated lighthouse keeper activities were also noted. Sites were visited, GPS coordinates taken (AMG Coordinates using Datum WGS84), measured drawings undertaken of the lighthouse precinct, and a condition audit conducted. The location of places or sites referred to are shown in Murray-Jones & Shepherd (2005). Results History Early settlement to South Australia was by sea to Port Adelaide with an uncertain route through islands, reefs and strong tidal currents. Light towers, such as pharos, were recognised as valuable aids to help ships navigate through these dangerous waters. The British Trinity House was set up Royal Charter in 1514 to serve as a charitable organisation for mariners, provide aids to navigation and to care for lighthouses (Trinity House 2005). Trinity House in Port Adelaide was founded for the same purposes on the British model. The first lighthouse in South Australia was built at Cape Willoughby, on Kangaroo I., and comm- issioned on 10 Jan. 1852. Others followed in the vicinity. Following an Intercolonial Conference in HUMAN SETTLEMENT 95 1873, a further lighthouse construction program began. One such proposed lighthouse location was Cape Spencer, but after Commander Goodbrough visited Spencer Gulf in 1874 he recommended establishing a light on Althorpe I. instead. In 1876 estimates in the Public Services Loan Bill provided £6,480 to construct a lighthouse and lightkeepers’ cottages on the island. After some delays, drawings were finally prepared by Robert Hickson, Engineer of Harbors and Jetties, South Australia in 1877, but there were further delays through lack of tenders. As a result, construction was finally undertaken by Harbors and Jetties staff in 1878, but not without accident®. Due to the steep cliffs, a jetty and an inclined ladder were constructed first. The attending cutter, Young St George, was wrecked and later, the foreman, John Anley, was killed by a falling rock. Costs escalated, due inter alia to the need to transport all the drinking water from Port Adelaide, and from delays in obtaining the Chance Brothers revolving light. The light was finally exhibited on 14 Feb. 1879 at a total cost of £11,700. Over the next century various repairs and changes were carried out including additional water storage (an ongoing problem for lightstation staff). A telephone line between the Island and Cape Spencer was laid in March 1886, following fears of a Russian invasion. The inclined ladder and tramway was replaced with a flying fox in 1904. The Commonwealth took over management of Australian lighthouses from Trinity House in 1912, and the late 1930s to early 1940s saw a major upgrading program for the site. This included new walls, a new engine house and signal box, upgrading of the cottages to provide new bathrooms, rainwater > Ibid, © [bid. 1 Thid. * Heritage Branch files — Statements of Heritage value: Lighthouse (10312): Completed in 1879, the Althorpe Island Lighthouse is significant for its association with the establishment of the network of lighthouses around South Australia’s shores to protect the shipping routes, which were crucial to the State’s economic development at the time. The lighthouse is important as an example of a well-executed stone lighthouse tower, in particular the detailing of the spiral staircase (HAS 2/2000). Keepers’ cottages (10314); The three lighthouse cottages were constructed to service the Althorpe Island Lighthouse and remain as an intact example of lighthouse keepers’ cottages of the late 1870s. They are an integral part of this remote lightstation, which was established to guide both local and interstate shipping through Investigator Strait during the period when sea transport was vital to the State’s economy (HAS 2/2000). Jetty (10318): This jetty and associated structures were built to service the Althorpe Island lightstation, established in 1879 to protect the shipping routes which were crucial to the State’s economic development at the time. The jetty represents the dependence on maritime transport as the only means of servicing the State’s island-based lightstations and reinforces the difficulties posed by the remoteness of these sites (HAS 2/2000). ” Register of National Estate Identified No. 6887. SA Government Gazette 16 March 1967. '! SA Government Gazette 14 August 1997, storage, and replacement of the rear pitch of the main roof and lean-to with a single pitch. The main light was converted to electricity in 1963 and a 3 m leading light erected in 1963. During the programme of automation of Australian lighthouses from 1975, the Althorpe I. lighthouse was finally automated in 1991, and the island vacated by the Australian Maritime Safety Authority (AMSA)’. The lighthouse and other built structures became part of the Althorpe Islands Conservation Park and were occupied occasionally. The lighthouse (10312, Fig. 1), keepers’ cottages (10314, Figs. 2, 3) and jetty (10318, Figs. 4, 5) were listed in the State Heritage Register in 1980, and Statements of Heritage prepared’. The lighthouse and cottages were also listed on the Register of the National Estate®. The lighthouse complex is located within the Unincorporated area of the State. The Western Isles above high water were gazetted as a Fauna Reserve in 1967!° and became the Althorpe Islands Conservation Park in 1972. Haystack and Seal Islands were added in 1977 with the area below high water added in 1991. Althorpe I. (Sections 13 and 61) was added to the Conservation Park in 1997 (Robinson et al. 1996)!'. The Conservation Park is important for its cultural and natural features, and as a habitat for sea birds. The Buildings The lighthouse precinct (Fig. 1) consists of a lighthouse, three keepers’ cottages, various sheds, tanks, fences and paths and the remnants of a runway. The lighthouse (Fig. 3) (Lat. 35° 22.2’ S; Long. 136° 51.7’ E) is managed by AMSA. It is a circular 13 m tall structure constructed of random rubble limestone and painted white. The exterior was altered in 1973 with the sandstone quoins being replaced with concrete. The light, exhibited at 107 m above sea level, is a 120 volt 1000 W tungsten halogen lamp with a range of ~45 km. The interior consists of a spiral stair of cut cantilevered sandstone risers with treads finished in Mintaro slate (now painted alternately red and green). The balusters are steel with an unpainted handrail. The floors are Mintaro slate supported by cast iron joists. The cottages all show a typically South Australian design unlike many lighthouse complexes elsewhere where there is a hierarchy of living conditions based on rank, Cottage | has a window from the master bedroom towards the lighthouse (absent in Cottages 2 and 3). The three cottages have received basic maintenance only, since the lighthouse was automated, and all show evidence of the harsh marine climate. 96 A. RADFORD Legend . Cottage (Principal keeper’s residence) . Cottage (2nd keeper’s residence) . Lighthouse . Generator Shed . Workshop Office Rope Locker CIDARWNE . Cottage (Ist assistant keeper’s residence) Fig. 1. Lighthouse precinct plan (adapted from Conservation Plan). Jetty precincts The jetty area consists of the flying fox (Fig. 4), jetty (Fig. 5), shed and nearby grave-site for the bodies recovered from the SS Pareora (Fig. 6). The jetty is a wooden structure of varying age, 72 m long and 2.7 m wide, with a hand crane at the seaward end standing on a wider 3.6 m platform. The jetty is now in poor condition with many structural members corroded, eaten away, missing, or, in the case of piles, replaced with newer, round, sister piles bolted beside them". A flying fox attached to the jetty terminates '2 National Archives have reference to repairs at various times but they are closed files. '3 Lighthouse log 2 May 1892 notes “employed putting up a fence around the grave”. National Archives, Adelaide Series D19 barcode 10684806. Unfortunately the logs for the time of death are missing. at the top of the cliff. While the top anchorage is in good condition, the flying fox is deemed unsafe due to the poor condition of the lower anchorage to the jetty. Graves There is much evidence of the 127-year human occupation all over the island. This includes graffiti in Cathedral Cave on The Hump, and the remains of a stone wall, which may have been a barrier to prevent animals crossing from the main island on to the Hump. Four graves can be seen above Mooring Bay. They may be memorials or may have been modified and updated by lighthouse staff and other island occupants over time. These are: (a) Julies Garbis, located near top of slope to E of HUMAN SETTLEMENT 97 Bedroom Enclosed veranda | es = 0 3000mm eel Bedroom iain | Fig. 2. Typical cottage plan (from Conservation Plan), Fig. 3. Cottage with lighthouse at rear. jetty. The grave is bounded with old bed posts'3, with wording on the grave ‘Julies Garbis died May 8th 1890 aged 42 yrs’. (b) Near sea level, to the east of the jetty, is located a white timber cross with a lighthouse carved on the top in a pile of rocks. It has no inscription but may be in memory of Arthur (‘Dick’) Johnson, a crew ‘4 Memorandum from Head Keeper to District Office Port Adelaide 11 November 1919. National Archives, Adelaide Series D14 barcode 430591. Fig. 4. Flying fox from jetty. Fig. 5. Jetty from above. member who lost his life on the wrecked cutter Rapid. His body was never found (Coroneos & McKinnon 1997), (c) Immediately west of the jetty is a white picket fenced area'* with a cross (Fig. 6) inscribed: ‘SS Pereora (sic) Wrecked 18.9.1919 JF Booth JC Braithwaite R Deebly RIP’. (d) Further west at the base of a cliff is a replica marker — an old door inscribed: ‘To the memory of G Peterson aged 48 yr died October 8th 1838’. Guano mining and sundry remains Penguins inhabit many limestone caves around the island, including some close to the cliff top. These 98 A. RADFORD Fig. 6. Pareora grave. caves often have low head room ( SARDI Aquatic Sciences, PO Box 120, Henley Beach, South Australia 5022. ‘Unpublished report by Graham Medlin for Friends of Althorpe Islands Conservation Park. 2001. 4SA Lighthouse logs. NAA: D19,1908-1910, in National Archives of Australia. ° SA Lighthouse logs. NAA: D19, 1879-1880, 1883-1888. A description of the vegetation and an accompanying floristic map were first published by Robinson er a/. (1996), following a visit in 1982. This paper summarises the history of land use of Althorpe I., describes its vegetation and the major biotic influences on the vegetation, and gives a plant species list (Table 1). For comparison, we also provide a plant species list from two neighbouring islands of similar size to Althorpe I., the South Neptune Islands, here termed the SN Lighthouse I., and SN North I. The former of these has been similarly grazed, but the latter has had no significant history of any human impact. Human impact after 1879 Nothing is known of the native flora or fauna present on Althorpe I. in 1879, when the lighthouse station was built and thereafter occupied by three families. Sub-fossil remains found in 2001 on, or just under, the surface in caves on the island indicate the relatively recent extinction of the Greater sticknest rat (Leporillus conditor), the Bush rat (Rattus fuscipes), and the Southern brown bandicoot (Isoodon obesulus?). Goats were probably introduced soon after human occupation, and by 1908 they were being hunted. By 1962 there were 120 goats, but a major cull in 1968 left only six, which bred up to 25 by 1981. By about 1990, however, all remaining goats had been exterminated. Other domestic grazers, such as a horse and groups of 30-40 sheep at a time, were introduced at various times‘, and may have affected the vegetation directly by grazing, and indirectly by introducing seed variously present in their gut or wool on arrival on the island. Firewood, which was regularly brought to the island, is another likely source of introduced seed. VEGETATION OF THE ALTHORPE ISLANDS 101 Cleared or Non-Native Grasslands Atriplex paludosa ssp. cordata Low Open Shrubland Myoporum insulare Low Open Shrubland Maireana oppositifolia | Atriplex paludosa ssp. cordata Low Shrubland Nitraria billardierei Low Shrubland Halosarcia spp. Low Closed Shrubland Bare Rock or Sand St Survey Sites Ca, a 668500 6085000 Cee 6084500 6084000 6083500 Geocentric Datum of Australia, 1994 @ |e and Hetilage 669000 Fig. 1. Floristic map of Althorpe Island showing all vegetation communities and the location of all survey quadrats, ALT 00101-ALT 010001. The vegetation was also affected through collection of building and paving materials®. Sand was removed from an area of 600 m2 to the west of the lighthouse, called the Quarry, which in 1980 was still almost without vegetation. “Seed grass”, couch grass (Cynodon dactylon), and marram grass ® SA Lighthouse logs. NAA: D19, 1881-1883. 7 SA Lighthouse logs. NAA: D19, 1908-1910; 1910-1912. * The specimens collected during 2000 were duplicated during the 2001 survey. (Ammophila arenaria), were introduced in 1910 and 1911’, but the marram grass did not become established (Table 1). In 1961, an airstrip, 450 m long, was cleared on the plateau, and sown with mixed grass seed for surface stabilization. Prior to this study, plant collections on Althorpe I. were made in 1907 by Dr R. S. Rogers (Maiden 1908), in 1910 and 1916 by Captain S. A. White (White 1916), in 1982 by Robinson ef a/. (1996), and in 2000 by S. Hawkins*. 102 E. F LAWLEY & S. A. SHEPHERD TABLE |. Plant species recorded from Althorpe Island and the two South Neptune Islands. Taxonomy follows the Census of South Australian Vascular Plants. Ed 5.1 (May 2004). Introduced species are marked with an asterisk. Species lists presented are the result of historical records and terrestrial surveys in 1982 and 2001. New records for Althorpe I. collected from 2001-2004 are marked as +. The 1982 survey of South Neptune Islands by Robinson et al. (1996) is omitted as it duplicated the list of plant species occurring on SN North I. found by Stirling et al. (1970). The Stirling 1970 list is included because some species not found in 2001, were likely present, but not all collection areas on SN North I. were revisited, due to time constraints. SNI Lighthouse I. was thoroughly searched over 10 days resulting in a comprehensive species list. g i) 2 s ¢ 2 =e =e & @ ss ss s & a 4 va Scientific Name Common Name mn & - a & 8 & Ba B@ B &B B = 3 Pp ege oF? & @ an ia.” aa 3 3 I A A A year of collection 1907 1916 1982 2001 2001 1970 2001 CASUARINACEAE Allocasuarina verticillata Drooping sheoak * (.keeper intr c.1982) URTICACEAE Parietaria debilis Smooth nettle * * Urtica urens* Stinging nettle * POLYGONACEAE Muehlenbeckia gunnii Coastal climbing lignum * * * AIZOACEAE Carpobrotus edulis* Hottentot fig * * Carpobrotus rossii Karkalla % * * * * Disphyma crassifolium Round-leaf pigface * * * ssp. clavellatum Mesembryanthemum Common iceplant * * * crystallinum* Tetragonia implexicoma Bower spinach * * * * PORTULACACEAE Calandrinia calyptrata Small-leaved parakeelya * Calandrinia granulifera Pigmy purslane * * CARYOPHYLLACEAE Polycarpon tetraphyllum* Four-leaf allseed * Sagina apetala* Common pearlwort * * * Scleranthus pungens Prickly knawel * * Silene nocturna* Mediterranean catchfly * Spergula arvensis* Corn spurrey * Spergularia marina Salt sand-spurrey * * Spergularia media* Coast sand-spurrey * * * * Spergularia rubra* Red sand-spurrey * Stellaria media* Chickweed * * CHENOPODIACEAE Atriplex cinerea Grey saltbush * * * * * Atriplex muelleri(?) Mueller's saltbush * Atriplex paludosa ssp. cordata Marsh saltbush * x * Atriplex semibaccata Berry saltbush * Chenopodium murale* Nettle-leaved goosefoot * * * * od * * Enchylaena tomentosa var. tomentosa Ruby saltbush TABLE 1. Cont. VEGETATION OF THE ALTHORPE ISLANDS 2 8 s Ss a ih Ee Ea & az} a aa a Zz z Scientific Name Common Name 8 4 4 4 FA FA g sD) i) Oo vo i= s g & & B 6: @ 2 @ | £2 2h $& G = =< @ 2 — 2 & | S 3 A A B year of collection 1907 1916 1970 2001 Halosarcia Grey samphire * halocnemoides ssp. halocnemoides Halosarcia pergranulata Samphire Halosarcia sp. Samphire Maireana oppositifolia Salt bluebush Rhagodia candolleana Seaberry saltbush * ssp. candolleana Sarcocornia blackiana Samphire * Sarcocornia quinqueflora Beaded glasswort Sclerolaena uniflora Bassia Suaeda australis Austral seablite * Threlkeldia diffusa Coast bone fruit * CRUCIFERAE Brassica tournefortii* Wild turnip Cakile maritima ssp. Two-horned sea rocket maritima* Hymenolobus Oval purse procumbens* Lepidium foliosum Peppercress * = * Sisymbrium orientale* Wild mustard CRASSULACEAE Crassula closiana Stalked crassula Crassula decumbens Spreading crassula var. decumbens Crassula sieberana ssp. Australian stonecrop tetramera LEGUMINOSAE Medicago polymorpha var. Burr-medic polymorpha* Melilotus indica* King Island melilot Swainsona lessertiifolia Coast swainson-pea * Trifolium repens* White clover OXALIDACEAE Oxalis pes-caprae* Soursob GERANIACEAE Geranium solanderi Native geranium Pelargonium australe Austral Stork's bill * ZYGOPHYLLACEAE Nitraria billardierei Nitre bush Zygophyllum billardierei Coast twin leaf EUPHORBIACEAE Euphorbia paralias* Sea spurge RUTACEAE Common correa “ Correa reflexa var. coriacea 104 TABLE 1. Cont. E. F LAWLEY & S. A. SHEPHERD Scientific Name Common Name Althorpe Island Althorpe Island South Neptunes North South Neptunes North | year of collection 1970 2001 MALVACEAE i Malva behriana Australian hollyhock * * Malva dendromorpha* Tree mallow % Malva parviflora* Small-flower marshmallow * FRANKENIACEAE Frankenia pauciflora Southern sea heath * *; , var. fruticulosa UMBELLIFERAE Apium prostratum Sea celery +7 * * var. filiforme Petroselinum crispum* Parsley * PRIMULACEAE Anagallis arvensis* Blue pimpernel * Samolus repens Creeping brookweed " LABIATAE Prostanthera serpyllifolia Small-leaved mintbush ssp. serpyllifolia SOLANACEAE Lycium ferocissimum* African boxthorn * MYOPORACEAE Myoporum insulare Common boobialla a PLANTAGINACEAE Plantago hispida Hairy plantain * GOODENIACEAE Goodenia varia Sticky goodenia * COMPOSITAE Angianthus preissianus Common cup-flower Arctotheca calendula* Cape weed i Brachyscome exilis Slender daisy - * Conyza bonariensis* Flaxleaf fleabane + Cotula coronopifolia* Waterbuttons * Cotula vulgaris Slender cotula * Craspedia variabilis Billy-buttons + Euchiton sphaericus Cudweed y Ixodia achillaeoides ssp. Coast ixodia * achillaeoides Lactuca_ serriola* Prickly lettuce * Leiocarpa supina Coast plover-daisy * Leucophyta brownii Coast cushion bush * * Microseris lanceolata Yam daisy * Olearia axillaris Coast daisy-bush Olearia ramulosa * * Podolepis rugata var. littoralis Pseudognaphalium luteoalbum Twiggy-daisy bush Pleated podolepis Cudweed VEGETATION OF THE ALTHORPE ISLANDS 105 TABLE |. Cont. 3 ‘ 3 sS ss g 2 @ 2 @ 5 5 Bas) & i & 4 Zz Zz Scientific Name Common Name ne a 2 a 2 2 A i) cP) oO o fc < =I e a & & 2 r= B |: 2@ BS & = <= @ = @ 4 G&G 3 3 s A A BAB year of collection 1907 1916 1982 2001 2001 1970 2001 Reichardia tingitana* False sowthistle * Senecio lautus Variable groundsel * Fg ‘3 Sonchus oleraceus* Common sowthistle * * * Urospermum picroides* False hawkbit * JUNCAGINACEAE Triglochin mucronatum Prickly arrowgrass + Triglochin tricophorum Arrowgrass % LILIACEAE Bulbine semibarbata Leek lily * * * Dianella revoluta Black-anther flax lily * - * var. revoluta JUNCACEAE Juncus bufonius Toad Rush + GRAMINAE Austrodanthonia White top * caespitosa Austrostipa flavescens Coast spear-grass * Avena barbata* Bearded oat bu = Bromus catharticus* Prairie grass * Bromus diandrus* Great brome * ‘i Cynodon dactylon* Couch-grass * Digitaria violascens* Emu-grass * if Elymus multiflorus Wheatgrass s Hainardia cylindrica* Common barbgrass * * Hordeum leporinum* Barley grass * * - * Lachnagrostis filiformis Blown grass * Lagurus ovatus * Hare's tail grass * Lolium perennex Ryegrass = rigidum* Lolium perenne* Perennial ryegrass fe a Lolium rigidum* Annual ryegrass r « Parapholis incurva* Curly ryegrass 7 * Poa fax Scaly poa i Poa poiformis Coast tussock grass * if var. poiformis * oe Sporobolus virginicus Salt-couch 106 E. F LAWLEY & S. A. SHEPHERD Methods Surveys, photopoints and vegetation maps In 2001 the survey of Althorpe I. described in this paper was carried out using the methods of the Biological Surveys of South Australia (Heard & Channon 1997). The survey was allotted the identifying number BS-126, in the Biological Survey Data Base, where the collected data are now stored as part of the Environmental Data Base of South Australia. Survey BS-126 comprises Althorpe I. (surveyed June and October 2001) and the South Neptune Is (surveyed October 2001). Numbered voucher specimens of all species collected were identified and deposited at the State Herbarium of South Australia. Sites were chosen to represent the different vegetation communities on the island, and quadrats located after consulting aerial photographs and the previous floristic map (Robinson ef al. 1996), and visual assessment. In each of 10 standard 30x30 m quadrats on Althorpe I., all plant species were recorded, the dominant overstorey and understorey species were noted, as well as the life form and average plant height of each species, using the adapted Muir’s table. Species’ cover abundances were estimated using the modified Braun- Blanquet scale (Heard & Channon 1997) and life stages noted. The overall vegetation association in each quadrat was described by summarizing the canopy cover for the dominant strata using the codes from Muir’s table, and the vegetation structure was described from the SA Vegetation Structural Formations table (Heard & Channon 1997). Names of vouchered samples of plant species found outside the quadrats, were added to the species list (Table 1). Physical data including landform element, site slope and aspect, outcrop cover, outcrop lithology, surface strew-size, -cover and -lithology, fire-scars, bare earth and plant litter coverage, and erosion were also collected. Vertebrate presence and surface soil texture class were examined and recorded, as was disturbance within a 30 m radius of the quadrat. The ten quadrats surveyed on Althorpe I. are shown on the floristic map (Fig. 1). Eight of the 10 quadrats are permanently marked with steel droppers carrying the standard Biological Survey aluminium disks (Nos 9431-9438), as photopoints for future reference. A 1999 aerial photo of Althorpe I. was ortho- rectified using GPS coordinates of notable island features, and used as a base to draw the vegetation map. Boundaries between vegetation communities were ground-truthed using handheld GPS. Results Vegetation Mapping The vegetation map (Fig. 1) presents five distinct native vegetation formations and a non-native grass community, which are described below. Nitre bush (Nitraria billardierei) Low SHRUBLAND (sites ALT00301, ALTO0401, ALTO0501) — Fig. 2A. This occurs on the central and eastern area of the main plateau of Althorpe I. that is covered in deep sand with occasional low dunes. Seaberry saltbush (Rhagodia candolleana ssp. candolleana), and coast twin leaf (Zygophyllum billardierei), occur in almost equal cover abundance as the nitre bush (N. billardierei), which is the tallest naturally occurring native species on the Island. Groundcovers are the native perennials, coast bone fruit (Threlkeldia diffusa), and ruby saltbush (Enchylaena tomentosa ssp. tomentosa) as well as some introduced annual forbs and grasses. African boxthorn (Lycium ferocissimum) was well established here until 1998, when it was progressively removed. The area is burrowed by short-tailed shearwaters at an average density of 12- 15 burrows per 100 m?. Low OPEN ALT 00201, Atriplex paludosa SHRUBLAND _ (sites ALT01001) — Fig. 2B. This chenopod shrubland occurs in shallow soils over calcarenite. Rhagodia candolleana ssp. candolleana, an associated plant, increases in percentage of cover where the soil deepens, whilst conversely the mat plant Disphyma crassifolium ssp. clavellatum decreases. Species on the island unique to this shrubland are the native forbs, hairy plantain (Plantago hispida), billy-buttons (Craspedia variabilis), and yam daisy, (Microseris lanceolata). Some native perennial grasses and introduced annual grasses and medics also occur in this Atriplex shrubland. Short-tailed shearwater burrows occur in this habitat at average densities of 6-8 per 100 m?. ssp. cordata ALT00101, Maireana oppositifolia and Atriplex paludosa ssp. cordata LOW SHRUBLAND (Sites ALTOO801 and ALT00901) — Fig. 2C. This low shrubland is on the island’s very exposed steep slopes, but extends onto the plateau in places where the surface calcarenite is exposed or covered by a thin veneer of soil. Robinson ef al. (1996) classified this shrubland as Leucophyta brownii Cushion bush (sub nom. Calocephalus brownii), but this species is quite uncommon and absent from many areas on the slopes. Its cover abundance was classified in quadrat ALTOO801 as N (1-10 individual plants) and in ALTO0901 as zero. Maireana oppositifolia and Atriplex paludosa ssp. cordata had a cover of 5 — 25% in quadrat ALT00801, and 25 — 50% in ALTO0901. In sheltered pockets in the calcarenite surface VEGETATION OF THE ALTHORPE ISLANDS 107 many other ground cover and shrub species occur, such as round-leaf pigface (Disphyma crassifolium ssp. clavellatum), coast bone fruit (Threlkeldia diffusa), seaberry saltbush (Rhagodia candolleana ssp. candolleana), coast twin leaf (Zygophyllum billardierei), bassia (Sclerolaena uniflora), southern sea heath (Frankenia pauciflora var. fruticulosa) and coast plover-daisy (Leiocarpa supina), in addition to grasses and numerous herbaceous plants. Less frequent on the slopes are the vivid purple flowering coast swainson-pea (Swainsona lessertiifolia), and the large golden flowering pleated podolepis (Podolepis rugata var. littoralis). The coastal Lrodia achillaeoides ssp. achillaeoides has been found only on the northern slope west of the saddle. The quarry survey site ALT 00701 (Fig. 1) is now re-vegetated with this low Maireana/Atriplex shrubland. N. billardierei grows densely and in substantial patches on the slopes where rubble fall has increased soil depth, but is at too small a scale to map. The Maireana/Atriplex. community grades into the Atriplex paludosa ssp. cordata vegetation type where the soil deepens. Halosarcia pergranulata Low CLOSED SHRUBLAND This vegetation formation, absent from Robinson et al. (1996), is only found near sea-level on the northern point, called ‘The Monuments’, where calcarenite outcrops over a granitic base partly shelter the site from the prevailing SW _ winds. Species present are: samphire (Halosarcia pergranulata), with some groundcover of sea celery (Apium prostratum var. filiforme), Frankenia pauciflora var. fruticulosa, Enchylaena tomentosa var. tomentosa, and the common _ iceplant (Mesembryanthemum — crystallinum). —Karkalla (Carpobrotus rossii) grows thickly out onto granite edging this site. Myoporum insulare LOW OPEN SHRUBLAND (site ALT00601) — Fig. 2D. This formation, absent from Robinson ef al. (1996), is located on sand at the isthmus to ‘The Monuments’. Sand over granite, grading to almost bare granite, supports 24 species of which 8 are introduced. Predominant are common boobialla (Myoporum insulare), Nitraria billardierei, coastal climbing lignum (Muehlenbeckia gunnii), Atriplex paludosa ssp. cordata and Leiocarpa supina, with groundcovers Carpobrotus rossii and Disphyma crassifolium ssp. clavellatum. Sparsely represented are: sea spurge (Euphorbia paralias), two-horned sea rocket (Cakile maritima), Mesembrvanthemum crystallinum, and four other herbaceous non-natives. ® Census of South Australian Vascular Plants. Ed 5.1. Staff and Assoc. of the State Herb. of South Australia, May 2004. A few grey saltbush (Atriplex cinerea), the only specimens on the Island, and some Australian hollyhock (Malva behriana), were also found here. The once present African boxthorn (Lycium ferocissimum), has been eradicated. Non-native Grasslands and Cleared Areas This cover type encompasses the landing strip and the area taken up by buildings and yards (Fig. 1). The vegetated area, of mainly annual non-native grasses, is found in a patch on the eastern cliff top and also on the landing strip, where in addition to the annuals, Cynodon dactylon, a perennial non-native couch grass, occurs. Maintenance of the landing strip by mowing ceased in 2001, 10 years after the light- station was automated, and shearwater burrows are gradually reappearing in the area. Island Perimeter Perpendicular sandstone cliff faces and steep calcarenite slopes, often covered in talus scree form the island’s perimeter. A typical shoreline vegetation, too narrow to be accurately mapped, covers the area between the foot of the slopes and the granite intertidal. Common species are: Cakile maritima, Disphyma crassifolium ssp. clavellatum, Apium prostratum var. filiforme, Carpobrotus rossii, and where the soil is deeper, Nitraria billardierei and the occasional Myoporum insulare or Malva behriana shrub. Species With Few Records Peppercress (Lepidium foliosum), collected by White (1916) on Althorpe and Seal Is, has not been recorded since then. Stirling ef a/. (1970) recorded the same species on the South Neptune Is, where it was also absent in 2001. Keighery ef a/. (2002) noted that Lepidium foliosum grew in association with guano-rich seabird nesting sites. Small-leaved mintbush (Prostanthera — serpyllifolia ssp. serpyllifolia), recorded by Robinson ef al. (1996), has not been found again, despite intensive searches. The native grass, Elymus multiflorus, previously recorded only in the Flinders Ranges (FR) and Eastern (EA) areas of the State (Census 2004°), was also recorded on Althorpe I. in 2001. In all, 86 plant species are recorded from Althorpe I., of which 46 are native and 40 introduced. Discussion Vegetation change over time, introductions, removals and regeneration The changes in Althorpe I.’s vegetation must have been severe following human settlement. Goats arguably had the greatest impact on the vegetation, even that on the steep slopes, as witnessed by the 108 E. F LAWLEY & S. A. SHEPHERD Fig. 2. Photos of four of the vegetation communities on Althorpe Island. A: Nitraria billardierei_ Low SHRUBLAND. B: Atriplex paludosa ssp. cordata Low OPEN SHRUBLAND. C: Maireana oppositifolia and Atriplex paludosa ssp. cordata Low SHRUBLAND. D: Myoporum insulare LOw OPEN SHRUBLAND. many goat tracks still visible many years after goat eradication. South Neptune I. has a_ similar occupation history, and Dr. R. S. Rogers wrote in 1907, only six years after the establishment of its Lighthouse: “There is a large number of big goats on the island, and these animals have already changed the vegetation” (Maiden 1908). Besides the impact of goats and sheep, more subtle changes may have been caused by the disappearance (Medlin’) of the herbivorous greater-sticknest rat (Leporillus conditor), and the arrival of garden snails (Helix asper), house mice (Mus musculus), and introduced plants on the island. Although the species present prior to settlement in 1879 can never be known, it may be inferred from a floristic comparison with other islands and the mainland coast, and from records of species’ responses to grazing or competition from introduced species. Adair & Groves (1998) found that most Australian studies quantifying weed impact demonstrated a decline in species richness, canopy cover or frequency of native species. Introduction of grazers may reduce native vegetation and lead to extinctions of native plant species, and the removal of the same grazers can then lead to the spread of invasive exotics (Cox et al.1967 in Timmins & Geritzlehner 2003), and subsequent loss of those native species vulnerable to competition. Abbott et al. (2000) found a 37% reduction in plant species over 40 years on Carnac I., Western Australia since 1951. They attributed the remarkable vegetational changes inter alia to the removal of rabbits and the arrival of the introduced species, Mesembryanthemum crystallinum, Malva parviflora, and more recently, Lycium ferocissimum. Althorpe I. is likely to have suffered a similar loss of native plant species. Comparison with the mainland coast shows that Althorpe I. has all but one, Melaleuca lanceolata, of the eight perennial plant species, which occur in all geomorphic regions along the coast (Opperman 1999). Species likely to have been on Althorpe I. at the time of human settlement include some of the 12 species that occurred in over 25% of the SA Coastal survey quadrats (Opperman 1999). These are: Dianella brevicaulis, Olearia axillaris, Leucopogon parviflorus, Isolepis nodosa, Pimelea serpyllifolia ssp. serpyllifolia and Clematis microphylla. The island’s physical environment and soils are also suitable for growth of Exocarpos aphyllus, Beyeria lechenaultii, Scaevola crassifolia, Helichrysum VEGETATION OF THE ALTHORPE ISLANDS 109 leucopsideum, Alyxia buxifolia and, at the base of the island, Swaeda australis. The disappearance of Lepidium foliosum might be due to the 1980s decline of the fairy penguin (Eudyptula minor), and the consequent loss of bird guano favoured by this species (Keighery ef al. 2002). The South Neptune Islands (SNI) species list (Table 1) provides further insight. Seven SNI North I. native species were not found, and may have disappeared from SNI Lighthouse I.'°, the most obvious being Correa reflexa var. reflexa and coastal lignum, Muehlenbeckia gunnii. The latter species has persisted on Althorpe I. in very low numbers, notably in crevices among rocks. On SNI North I. coast tussock grass, Poa poiformis, forms an extensive native grassland community, but it is rare on SNI Lighthouse I., possibly due to heavy grazing by resident Cape Barren Geese preventing its re- establishment after goat removal. These birds rarely visit Althorpe I. and do not breed there. P poiformis, which requires a certain level of salt spray (Opperman 1999), is absent from Althorpe I. despite the presence of suitable habitats on its lower slopes. It was possibly grazed to extinction. A comparison with the previously occupied 809 ha St Francis I., Nuyts Archipelago, shows some habitats similar to those on Althorpe I. (Robinson e¢ al. 2003). The Nitraria billardierei community on the cliff top sands on St Francis I. contains Olearia axillaris and Myoporum insulare, both missing from the equivalent community on Althorpe I. Of the 131 plant species recorded from St Francis I. 26% are exotic (Robinson ef al. 2003) compared with the 47% of exotic species among the 86 species on Althorpe I. (Table 1). This difference may be due to the greater proximity of Althorpe I. to the mainland, its smaller size or greater modification. The spatial extent of some species has apparently also become much reduced. Bower spinach, 7etragonia implexicoma, is usually a common and widespread coastal and island species, which may rapidly go extinct under grazing (Underwood & Bunce 2004), It has survived on Althorpe I. only on the moister, steep, southerly-facing slopes intensely burrowed by shearwaters, and unattractive to grazing goats. Two native grasses are poorly represented on Althorpe I. Coast spear-grass, Austrostipa flavescens, occurring in various areas on shallow sand near the plateau perimeter, is favoured by the shearwaters as nesting habitat, possibly because the tussocks bind the soil, preventing erosion and burrow collapse. The palatable and nutritious white-top Austrodanthonia caespitosa, (see Jessop & Toelken 1986) is now ‘© A number of native species are recorded for SNI Lighthouse I. but not recorded for SNI North I., they are possibly present but survey time on North Island was 8 hours, on Lighthouse Island 10 days. restricted to a few sites on the cliff top edges, and was likely grazed to near extinction. Exotic annual grasses are now widespread on Althorpe I. and may bind the soil and reduce soil erosion in areas used by shearwaters. Exotic grasses were also sown on the newly graded landing strip at South Neptune Lighthouse I. in 1961, but failed to establish there and persist only among granite boulders at the northern end of the strip. Since goat eradication by 1990 on Althorpe I. an increase in cover of native species has been observed. Whereas Robinson ef al. (1996) noted that introduced grassland at the western end of the plateau was common in 1982, by 2001 Atriplex paludosa ssp. cordata, had become the predominant canopy species (Fig. 1). Nitraria billardierei had also spread, obliterating former tractor tracks, likely because the large seed in its palatable fruit is readily spread by gulls and other birds (unpublished observations). Stirling et al. (1970) also observed N. billardierei seed in regurgitate of the Pacific gull, Larus pacificus, on North South Neptune I. between 1967-1970, although N. billardierei was then absent. However, by 2001 the species had established there in small numbers. The old landing strip on Althorpe I. now has the groundcovers, Threlkeldia diffusa and Enchylaena tomentosa var. tomentosa among the perennial couch grass, (Cynodon dactylon), and annual grasses. An African boxthorn control program has been in operation since 1998 and the introduced herbaceous biennial, tree mallow, Malva dendromorpha, once common on the plateau, has been manually removed since 1996, and is now in low numbers. Conclusions The vegetation of Althorpe I. has been severely disturbed by goats and human occupation between 1879 and 1991, with the resultant likely loss of native species and the introduction of many exotics. Since 1991, greater environmental awareness has stimulated efforts to eliminate introduced animals, and eradicate exotic perennials. Goat eradication and boxthorn control on Althorpe I. are beginning to show benefits, and providing that boxthorn re- establishment is prevented, the documented trend of native vegetation recovery is expected to continue. The 2001 and later vegetation surveys, with fixed photopoints, have established a valuable data-base and baseline, against which to compare future changes in the vegetation. Acknowledgements We thank the Native Vegetation Council for the grant enabling the 2001 vegetation survey, Innes 110 E. F LAWLEY & S. A. SHEPHERD National Park staff for cooperation and support, Lee Heard and Sue Kenny of (then) Planning SA, for training Friends of Althorpe I. Conservation Park, supporting the 2001 survey and entering electronic data. Rosemary Taplin and Helen Vonow at the State Herbarium of South Australia assisted with identifications, searched early records, and resolved many difficulties. Sue Kenny helpfully modified Figure 1, Peter Canty provided information and advice on the 1982 survey, John Lawley provided historical information, Sherilee Hawkins carried out preliminary survey work, and Brad Page provided much helpful advice in assembling the data. We also thank Rob Lewis and Anthony Cheshire of SARDI for promoting the Island research, and Sue Murray- Jones for her administrative skills. Wendy Stubbs, Helen Vonow, Bob Baldock and referees provided helpful comments on the manuscript. References ABBOTT, I., MARCHANT, N. & CRANFIELD, R. (2000) Long- term change in the floristic composition and vegetation structure of Carnac Island, Western Australia. Journal of Biogeography. 27, 333-346. AbaIR, R. J. & Groves, R. H. (1998) “Impact of Environmental Weeds on Biodiversity, A Review and Development of a Methodology”. (Biodiversity Group Environment Australia, Canberra). Bevperio, A. P., HAILS, J. R. & Gostin, V. A. (1983) A review of Holocene sea levels in South Australia. /n Hopley, D. (Ed.) “Australian Sea Levels in the Last 15000 Years: a Review”. Department of Geography, James University of North Queensland. Monograph series, Occasional Paper 3, 37-47. Brown, J. H. & Lomo tino, M. V. (2000) Concluding remarks: historical perspective and the future of island biography theory. Glob. Ecol. & Biogeogr., 9, 87-92. Cox, J. E., TAyLor, R. H., & MAson, R. (1967) Motunau Island, Canterbury, New Zealand. An ecological survey. New Zealand DSIR Bulletin 178. HEARD, L., & CHANNON, B. (Eds) (1997) “Guide to a Native Vegetation Survey using the Biological Survey of South Australia Methodology. First Edition”. (Geographic Analysis and Research Unit, Information and Data Analysis Branch, Department of Housing and Urban Development, Adelaide). Jessop, J. P. & TOELKEN, H. R. (1986) “Flora of South Australia” 4th ed. South Australian Government Printer Adelaide. KEIGHERY, G. J., ALFORD, J. J. & LONGMAN, V. (2002) A vegetation survey of the islands of the Turquoise Coast from Dongara to Lancelin, south-western Australia. Conservation Science W. Aust. 4, 13-62. McArTuurR, R. H., & WILSON, E. O. (1967) “The Theory of Island Biogeography”. (Princeton University Press, Princeton, N. J.) MAIDEN, J. H. (1908) A contribution to the botany of South Australia. Trans. R. Soc. S. Aust. 32, 252-286. OPPERMAN, A. (1999) “A Biological Survey of the South Australian Coastal Dune and Clifftop Vegetation” (Coast and Marine Section, Environment Protection Agency, Adelaide). Rosinson, A., CANTY, P., Mooney, P. & RuDDUCK, P. (1996) “South Australia’s Offshore Islands”. Australian Heritage Commission. (Australian Government Publishing Service, Canberra). RoBINSON, A. C., CANTY, P. D., WACE, N. M. & BARKER R. M. (2003) The Encounter 2002 expedition to the Isles of St Francis, South Australia: Flora and vegetation. Trans. R. Soc. S. Aust. 127, 107-128. STIRLING, I., STIRLING, S. M., & SHAUGHNESSY, G. (1970) The bird fauna of South Neptune Islands, South Australia. Emu 70, 189-192. TIMMINS, S. & GERITZLEHNER, J. (2003) Do weeds respond to Pest Animal Control? DOC Science Internal Series 121. (Department of Conservation, Wellington). Unpberwoop, M. & BuNCcE, A. (2004) The breeding biology of the White-faced Storm Petrel (Pelagodroma marina) on Mud Islands, Port Phillip Bay, Victoria. Emu 104, 213-220. White, S. A. (1916) “The Cruise of the Avocet”. (W. K. Thomas & Co., Adelaide). Transactions of the Royal Society of S. Aust. (2005), 129(2), 111-115. EFFECTS OF AFRICAN BOXTHORN REMOVAL ON NATIVE VEGETATION AND BURROWING OF SHORT-TAILED SHEARWATERS ON ALTHORPE ISLAND, SOUTH AUSTRALIA by E. FE. Lawtey!, J. J. LAWLEY? & B. PAGs Summary Law ey, E. F, LAwLey, J. J. & Pace, B. (2005). Effects of African boxthorn removal on native vegetation and burrowing of short-tailed shearwaters on Althorpe Island, South Australia, Trans. R. Soc. S. Aust. 129(2), 111-115, 30 November, 2005. Althorpe Island’s African boxthorn infestation was incrementally removed from 1998 and the effect of this removal on the vegetation and on the nesting of the short-tailed shearwater, Puffinus tenuirostris, was examined. During the study period, native perennial mat plant vegetation succeeded an initial flush of annual non-native plants at most sites examined. A weak trend of increasing burrow density of shearwaters within two metres from boxthorn stumps after removal of the canopy was also observed. Key Worps: Lycium ferocissimum; African boxthorn; Puffinus tenuirostris; short-tailed shearwater; succession; Althorpe Island Introduction The increased soil nutrient loads (through guano) and disturbance (excavating and trampling) caused by seabirds have been shown to decrease both the density of plants and the species richness in seabird colonies (Crooks 2002; Bancroft et al. 2005). Vegetation within colonies of burrowing seabirds is typically dominated by only a few species (reviewed in Walsh ef al. 1997). Within Australian colonies of shearwaters (Puffinus spp.) the dominant vegetation is typically the succulents: Tetragonia spp., Rhagodia spp. and/or Carpobrotus spp. (Walsh et al. 1997). However, where soil nutrient loads are naturally low, the combination of increased guano deposition and disturbance are thought to create conditions that favour annual and/or exotic plant species (Lamont 1984). In that study, a shearwater colony in Western Australia, was dominated by short-lived, succulent exotics, whereas areas surrounding the colony were dominated by long-lived heath species. The removal of invasive alien species from natural ecosystems can have unexpected outcomes. If the alien species are at different trophic levels, then mesopredator release, cascade and other indirect effects can occur in the system after their removal (Zavaleta et al. 2001). Hence it is critically important to undertake post-eradication monitoring in order to assess the effect of eradication. Althorpe I. (Lat. 35° 23’ S; Long. 136° 51’ E), located 8.5 km off Cape Spencer on Yorke Peninsula, ' PO Box 89, Tarlee, South Australia 5411. Email: lawley@chariot.net.au > Tbid } SARDI Aquatic Sciences, PO Box 120, Henley Beach, South Australia 5022 South Australia, is 92 ha in area. The island has high conservation value as it is the nesting site of about 20,000 — short-tailed = shearwaters, — Puffinus tenuirostris, which dig nesting burrows in the sandy soil (Robinson ef al. 1996). Cats, Felis catus, and goats, Capra hircus, were introduced during human occupation (Lawley & Shepherd 2005) and may have affected shearwater breeding success by predation and trampling respectively. The goats were eradicated by 1990, and the cats by 2003. The African boxthorn, Lycium ferocissimum, introduced to Althorpe I. around the 1930s, proliferated after the removal of goats, and became abundant in virtually all habitats (Lawley & Shepherd 2005). Observations suggested that shearwaters were unable to burrow under the boxthorns, which develop a fine, dense root mass under the bole. From 1998 a boxthorn eradication program was undertaken. An estimated 10000 mature boxthorns were removed, and in the succeeding six years up to 2000 seedlings were removed annually. Following the boxthorn eradication program, this study was undertaken to determine: (1) the effect of boxthorn removal on_ the surrounding vegetation, and in particular, the process of recolonisation in the bare patches produced by boxthorn removal; and (2) whether shearwaters expanded their burrowing activities into the above bare patches. Materials and Methods Recolonisation of bare patches Twenty stumps of large boxthorns, left in situ on removal of the canopies, were selected for 112 E. EF LAWLEY, J. J. LAWLEY & B. PAGE monitoring in May 2001. Each stump was marked with paint and a tag bearing the year of clearing and a sequential number, and its GPS location recorded. A hoop of 1.5 m diameter, divided into segments to facilitate estimation of cover, was placed around the stump, and the enclosed quadrat photographed. The number of plants of each species appearing in the quadrats was recorded in three categories: recently germinated seedlings; immature plants >5 cm high; and mature plants. For spreading perennials, the percent cover was estimated visually. Percentage cover of annual seedlings was not plotted because germination date and growth rate of these are closely linked to autumn rain events, the onset of which may vary by several weeks. A difference in cover of annuals at the time of the survey may not be a true reflection of the temporal trend. The quadrats were resampled in the spring of 2001, and in the autumn and spring of 2002 and 2004. Shearwater burrow density Studies were concentrated solely in a Nitraria billardierei shrubland where burrows were deepest, to about | m depth, and at relatively high densities. Four transects (named A-D), each 60 m long (except transect B which was 45 m), were set on level ground, with similar densities of boxthorn among uniform native vegetation, and with similar soil depths. Transect D was placed in an area cleared of boxthorn in 2000, A in an area cleared in 2001 and Transects B and C in areas cleared in early 2002. GPS coordinates for the southern extremities of transects were noted, and marked with permanent droppers with tags recording the transect number and year of clearing. Distances between shearwater burrows and boxthorn stumps were measured in April 2002, 2003 and 2004 according to the following protocol. Observers walked from south to north on either side of a tape measure stretched between the extremities of the transect, holding a 2.5 m measuring stick at right angles to the tape. The exact location of each boxthorn stump or bush with a trunk diameter >50 mm, within 2.5 m of the transect line, was recorded. Shearwater burrows were included in the study if they showed signs of having recently been used by shearwaters (guano deposits, recent excavations and/or chick present). Each shearwater burrow’s distance to a stump, within a radius of 2 m of each stump, was recorded by measuring from the centre of the burrow entrance to the centre of the nearest stump, to an accuracy of 10 cm, in the categories 0 — 0.5 m, >0.5 — | m, or >1— 2 m distance from stumps. In Transects A-D there were 6, 8, 14 and 10 stumps respectively. Based on the methods outlined above, the mean burrow density was calculated for each transect, in each of the distance categories from stumps (0 — 0.5 m = 0.79 m’, >0.5 — 1 m = 2.36 m2, or >1 — 2 m= 9.43 m?). The number of burrows around boxthorn stumps was recorded from the period before boxthorn removal until four years after removal (Fig. 1). To monitor average burrow density over time for each transect area, a burrow count was conducted in a 10 x 10 m quadrat at the ends of each transect. Statistical analyses, including the Spearman rank correlation, r, were done with the SPSS statistical software (version 11, SPSS Inc., Chicago). All statistical tests are two-tailed, unless stated, with the o level of statistical significance set at 0.05. Results Spontaneous recolonisation In the first year after boxthorn removal, many species appeared in the quadrats, including annual non-native forb and grass seedlings, such as common sowthistle, Sonchus olearaceus (up to 260 seedlings per quadrat), ryegrass, Lolium sp. (up to 120 seedlings per quadrat), common __ iceplant, Mesembryanthemum crystallinum (up to 200 seedlings per quadrat). Perennial native seedlings persisted through the summer, and, together with vegetative expansion of plants outside the quadrats, an overall increase of cover by perennial native plants was observed over the four years observation period (Fig. 2 and Fig. 3). Plots showing the changes in mean percent cover of the four most common native species over four years are given in Fig. 3. Seaberry saltbush, Rhagodia candolleana ssp. candolleana (Fig. 3A), appeared in six quadrats, showing an increase in percent cover in five quadrats, but disappeared from the sixth where it was overgrown by the common iceplant, Mesembryanthemum crystallinum, and wild turnip Brassica tournefortii. A mat plant, coast bone fruit, Threlkeldia diffusa (Fig. 3B), colonized rapidly, appearing in nine quadrats after one year, and in 17 after four years. Another mat plant, ruby saltbush, Enchylaena tomentosa var. tomentosa (Fig. 3C), appeared in one quadrat after one year, but was present in 11 quadrats after four years. Coast twinleaf, Zygophyllum billardierei (Fig. 3D), initially appeared in seven sites but subsequently declined in some quadrats. By spring 2004 some of the stumps, four years after cutting, had disappeared completely under the native matplants with a cumulative cover of almost 100% (Fig. 2). Shearwater burrow density In the N. billardierei shrubland overall burrow density, as measured in the quadrats at the ends of the transects, ranged from 0.12 — 0.19 burrows m*°. There were too few data to examine inter-annual changes in burrow density. Along the transects, the BOXTHORN REMOVAL EFFECTS ON NATIVE VEGETATION AND SHEARWATERS 113 0.30 0.25 0.15 Number of burrows per square metre 0.10 0.05 —~o—0.0 to 0.5m tA 05to1.0m —B—1.0to2.0m —B— 0.0 to 2.0m a er eee | 1 year prior 1 year after 2 years after 3 years after 4 years after [ Number of years since boxthorn removal Fig. |. The mean number of short-tailed shearwater burrows around individually numbered boxthorn stumps, measured before and after boxthorn removal. For clarity, standard error bars are shown only for summed density data, 0 — 2 m from stumps. mean burrow density was positively correlated with the number of years since boxthorn removal; however, none of the regressions were significant (0-0.5 m: r = 0.286, df = 3, P = 0.36; 0.5 — 1.0 m: r= 0.417, df = 3, P = 0.29; 1.0 — 2.0 m: r = 0.686, df = 3, P = 0.16; Fig. 1). The overall mean burrow density between 0 — 2 m was positively correlated with the number of years since boxthorn removal and the regression approached significance (0 — 2.0 m: r= 0.833, P = 0.08, Fig. 1). Discussion The eradication of goats from Althorpe I. most likely had pervasive ecosystem consequences before this study began. Boxthorn plants, previously kept at low levels by goats, proliferated after their eradication, and boxthorn became abundant in virtually all habitats. The rapid spread of boxthorn was aided by at least two vectors common on Althorpe L., the little raven, Corvus mellori, and the starling, Sturnus vulgaris, which eat boxthorn berries and spread seed in regurgitates and droppings (EFL, unpublished data). In this study, the immediate effect of boxthorn removal was the exposure of bare patches. Apparently at Althorpe I., the shading by the boxthorn canopy, combined with its extensive root system, was able to out-compete and eliminate understorey growth (see Belsky 1994, Holmgren et al. 1997). After clearing of boxthorns, annual weeds invaded the clearings, and in succeeding years were gradually replaced by native perennials in most of the 20 quadrats (Fig. 3). A similar succession of exotic winter annuals, followed by native mat plants in bare areas, was found by Brown ef al. (1993) in Tasmania. However, in that study the bare areas were a consequence of shearwater trampling, not boxthorn removal. 114 E. RF LAWLEY, J. J. LAWLEY & B. PAGE 10 Rhagodia candolleana . o > 6 oO = fay 2 fo) a 0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 45 15 Enchylaena tomentosa o@ 10 > fe} is) = o 2 o oa 5 0 05 1.0 1.5 2.0 2.5 3.0 3.5 4.0 45 Number of years since boxthorn removal Threlkeldia diffusa 1p Zygophyllum billardierei 10 0) r — r 1 1 0 05 1.0 1.5 20 2.5 3.0 3.5 4.0 45 Number of years since boxthorn removal Fig. 3. Mean percent cover of Rhagodia candolleana (A), Threlkeldia diffusa (B), Enchylaena tomentosa (C), and Zygophyllum billardierei (D) in marked quadrats, following the removal of boxthorn (Lycium ferocissimum). Vertical bars are standard errors. Whilst this study has shown only a weak temporal trend in increased burrowing near boxthorn stumps, this may be because too few years have elapsed for the fibrous root system under the stumps to break down sufficiently for shearwaters to burrow through them. Further, more time may be needed for young shearwaters to excavate new burrows in the greater area made available. The shearwater burrow densities in the quadrats at the end of each transect encompassed the range observed in the study sites, particularly those observed up till two years after removal. The data, as yet inconclusive, suggest that burrow density is rising. BOXTHORN REMOVAL EFFECTS ON NATIVE VEGETATION AND SHEARWATERS 115 Conclusions The removal of African boxthorn has had a positive effect on the native flora, by allowing native shrubs, particularly mat plants, to recolonise the bare patches. Whether shearwaters have expanded their burrowing activities into the areas vacated by boxthorn is less clear. Continued monitoring, perhaps triennially, both of shearwater burrow density and of the plant quadrats, is recommended to determine long-term trends. Short-tailed shearwater colonies are absent from the southern Australian mainland and found only on a few suitable offshore islands. The destruction or degeneration of habitat can cause the demise of these sea-bird colonies, and Jones ef al. (2003) documented such declines in New Zealand, largely due to introduced predators. Hence, the eradication of feral cats and the unremitting control of aggressive alien species, such as boxthorn, will help conserve both the native vegetation and seabird colonies on these islands. Acknowledgements We thank the Friends of Althorpe Islands Conservation Park for facilitating the field trips and participation in fieldwork, and the staff at Innes National Park for support. We also thank Rob Lewis and Anthony Cheshire of SARDI for promoting the Island research, Scoresby Shepherd for encouragement, helpful comments and for editing the manuscript, and referees for improvements to the paper. References Bancrort, W. J., ROBERTS, J. D. & GARKAKLIS, M. J. (2005) Burrowing seabirds drive decreased diversity and structural complexity, and increased productivity in insular-vegetation communities. Aust. J. Botany 53, 231- 241. Betsky, A. J. (1994) Influences of trees on savanna productivity: tests of shade, nutrients, and tree-grass competition. Ecology 75, 922-932. Brown, M. J., MARUYAMA, N. & WILLIAMS, K. J. (1993) Ecological studies of vegetation in short-tailed shearwater colonies in Tasmania. Pap. Proc. R. Soc. Tasm. 127, 11-16. Crooks, J. A. (2002) Characterizing ecosystem-level consequences of biological invasions: the role of ecosystem engineers. Oikos. 97, 153-166. Lamont, B. B. (1984) Specialised modes of nutrition. pp. 126-145 In Pate, J. S. & Beard, J. S. (Eds) “Kwongan, Plant Life of the Sandplain”. (University of Western Australia Press, Perth). HOLMGREN, M., SCHEFFER, M. & Huston, M. A. (1997) The interplay of facilitation and competition in plant communities. Ecology 78, 1966-1975. Jones, C., BETTANY, S., MOLLER, H., FLETCHER, D. & Cruz, J. D, (2003) Burrow occupancy and productivity at coastal Sooty shearwater (Puffinus griseus) breeding colonies, South Island, New Zealand: can mark- recapture be used to estimate burrowscope accuracy? Wildlife Research 30(4), 377-388. Lawey, E. F & SuHepHerp, S. A. (2005) Land use and vegetation of Althorpe Island, South Australia, and a floristic comparison with South Neptune Islands. Trans R. Soc, S. Aust. 129, 100-110. Rosinson, A. C., CANTY, P., Mooney, P. & Rupbuck, P. (1996) “South Australia’s Offshore Islands”. Australian Heritage Commission (Australian Government Publishing Service, Canberra). ZAVALETA, E. S., Hops, R. J. & Mooney, H. A. (2001) Viewing invasive species removal in a whole-ecosystem context. Trends in Ecology and Evolution 16, 454-459. Watsh, J. B., Kirkpatrick, J. B. & Skira, I. J. (1997) Vegetation patterns, environmental correlates and vegetation change in a Puffinus tenuirostris breeding colony at Cape Queen Elizabeth, Tasmania. Aust. J. Botany 45, 71-79. Transactions of the Royal Society of S. Aust. (2005), 129(2), 116-127. MARINE BENTHIC ALGAE OF THE ALTHORPE ISLANDS, SOUTH AUSTRALIA by R. N. BALDock! & H. B. S. WOMERSLEY? Summary Batpock, R. N, & WomersLey, H. B. S. (2005) Marine benthic algae of the Althorpe Islands, South Australia. Trans, R. Soc. S. Aust. 129(2), 116-127, 30 November, 2005. Deep water to intertidal marine algae have been collected from the Althorpe Islands on four occasions. Present collections include 15 species of Chlorophyta, 37 Phaeophyta and 92 Rhodophyta, probably a fraction of the total species present. A canopy of Acrocarpia paniculata and Ecklonia radiata occurs in most subtidal regions. Algal communities at a cave site and those exposed to high wave energy and sheltered conditions are described. Kry Worpbs: Marine benthic algae; Althorpe Islands; offshore islands; algal communities; intertidal algae; subtidal algae; algal canopy Introduction The Althorpe Islands lie in Investigator Strait, South Australia and consist of a single, large, steep- sided calcarenite island and several lower outlying isles. The group was separated from southern Yorke Peninsula, S. Australia, by sea level rise during the last interglacial period. They are underlain by jointed granites and amphibolite dykes (see Robinson ef al. 1996, p.288) supporting vertical, orange, calcarenite cliffs with caves on exposed coastlines. Unlike parts of the adjacent mainland there are no flat calcarenite shore platforms, and only a single relatively narrow sandy beach, the Mooring Bay. A granite rise with joints and dykes on the western side has been eroded to five isles, the Western Isles. Subject to the full effects of cyclonic cells moving east, the Althorpes are exposed to periodic high wave energy on all sides. Algae deposited in the State Herbarium have also been collected from the adjacent southern Yorke Peninsula mainland over a number of years by Shepherd & Womersley, and Edyvane & Baker collected at nearby Haystack, Chinaman Hat and Seal Islands for the construction of the Interim Marine and Coastal Regionalisation of Australia (IMCRA). (See Edyvane & Baker 1996.) An Investigator Strait collection (J. Watson, January 1971, unpublished) from two deep-water transects between Foul Bay/Hillock Point on the mainland and Cape Cassini/Cape Dutton on Kangaroo Island has also been made. The present study although confined to the algae of the main Althorpe Island and its five outliers, the ' University of South Australia, Mawson Lakes campus, South Australia, 5095. Email: Robert.Baldock@unisa.edu.au, and State Herbarium. 2 State Herbarium, Botanic Garden, Hackney Rd, Hackney, South Australia 5000, and Department of Environmental Biology, University of Adelaide. Western Isles, brings together data from both the 2004 expedition and several previous studies to describe algal distribution of these islands. Collections for these studies have been made by the following: 1. By divers aboard the vessel SAORI, 4 January, 1964 while undertaking a marine exploration of S. Australian Gulf waters; 2. By J. Baker and K. Edyvane, 26-27 October, 1993, during the collection of data for IMCRA; 3. By A. Gaut, 2001 from the intertidal, but also containing drift specimens; 4. By E. Lawley, 28 September, 2001 and 28 November, 2004 from a saline soak 50 m inland and 10 m above sea level; 5. By R. Baldock, J. Brook, R. Lewis and N. Barrett, January 31 — February 11, 2004, mainly using SCUBA during the expedition conducted by the South Australian Research and Development Institute (SARDI) aboard the research vessel Ngerin. There is also a single specimen of Cystophora platylobium collected by M. Retallick, from 18 m deep off the N. end, on 3 March 1987. Collation of collections in the current study will permit a database for further ecological work and comparisons with other local regions such as the Gambier Island group, suggested for Marine Protected Areas. Names and classifications below follow “The Marine Benthic Flora of Southern Australia” (Womersley 1984, 1987, 1994. 1996, 1998, 2003), with changes from later phycological studies. Methods In 2004, seven collecting sites were chosen on the basis of accessibility and contrasting habitat conditions. They have been described in terms of physical conditions and floristics; 2004 MARINE BENTHIC ALGAE OF THE ALTHORPE ISLANDS, SOUTH AUSTRALIA 117 identifications of species are generally by the first author, and the remainder by H. B. S.Womersley, who checked all determinations. The species list consists exclusively of specimens deposited in the State Herbarium. Biomass differences at several depths between “The Boulders”, with moderate wave energy, and “The Hump”, a high energy habitat, were also determined in order to expand the site descriptions. Sub-tidal algae were collected from four 1 m? quadrats at several depths with the help of A. Hirst, then identified and weighed. Encrusting lithophytic algae and sea grasses were not collected. Results Site Descriptions for the 2004 Collections I. “The Boulders” (BO), E corner of Mooring Bay This is a granitic headland exposed to moderate wave energy and tidal current. The substratum slopes at about 10° to a sandy bottom at about 8 m depth. There are scattered submerged flat boulders and several emergent rounded ones that give the locality its name and where waves break continuously. A band of red turf algae, bleached during summer, consisting predominantly of Chondrophycus tumidus, Polyopes constrictus, and Gelidium australe occurred in the sublittoral fringe zone as defined by Womersley (1981), with sporadic denuded plants of the fucoid Cystophora intermedia. The sublittoral canopy layer with up to 90% cover was dominated by patches of the fucoid Acrocarpia paniculata and the spinous form of the laminarian, Ecklonia radiata. Prominent but sporadic patches of green Caulerpa flexilis var. muelleri and C. brownii also occurred. Wet weights of algae collected during 2004 at 5 and 8 m depths (Fig. 1) indicate increased dominance of Ecklonia and decreased dominance of Acrocarpia and Cystophora moniliformis with depth. An understorey of encrusting coralline red algae (not collected), other red algae and Caulerpa spp was present. 2. “The Hump” (HU), adjacent to a SE headland of high wave energy This site has near-vertical, jointed granitic rock walls to a depth of 5 m then a stepped slope of about 20° to a coarse sandy bottom at 25 m depth. The region is subject to high wave energy, with considerable surge even on relatively calm days. A distinct band of red algal turf, bleached in upper parts during summer, with some Ulva occurred in an extended sublittoral fringe, the result of high wave energy. Asparagopsis armata and mats of dark red Gelidium australe were present with Pterocladia lucida prominent in shallow water. At about 5 m depth the articulated corallines Cheilosporum sagittatum and Haliptilon roseum occurred in patches and as an understorey to a dense canopy of Acrocarpia paniculata and Ecklonia. Larger clumps of the red algae Melanthalia concinna and Ballia callitricha with the brown algae Halopteris pseudospicata, Homoeostrichus sinclairii and Lobospira bicuspidata also occurred. 2500 Se ] = 2000 [__] 5m depth | 2 Hi sm depth ~ 1500 St | a g 1000 | oD | = 500 | | | o- — ee = Se 89 £2 § § gf 8. 2 6 of eo & © = = © x = o> £o on i ae} SO := aD @ Oo 5 Qa a. 2 oy = nS == (oa oma ox See + AS 9° (7p) fob) = 3 O + Qn © Qs os Oo ® o nO & ome) 0} Qe qa D> Ee > O Cc oe axe pom O O€£ fe) Om o Qs. x & > © O 6 o = Lu iu Be oO Fig. 1. Dominant algal species collected at “The Boulders” in 2004: differences with depth. 118 R. N. BALDOCK & H. B. S. WOMERSLEY 7000 5m depth _~ 6000 15 N : | m = 5000 _ = 4000 25m a @o = 3000 o = 2000 1000 0 _ & 8 E Zg oo 2 0 So | as xd Oo 0 o oO x oc ee | oO © aS O= LU Oo & oc qt a 20 o oO O Fig. 2. Canopy species collected at “The Hump” in 2004: changes with depth. 250 ec E 200 5m — = 150 mm 15m [o)) D | @ = 50 0 | | — | —_ —— £ 3 o 2 ae mets & no © = = © a 2 a) cn = 2 Onc a0 o> = ec Oo = Go ec te’ 8 ES £ © 5 2 7A} On 8-0 OH > fe) = Qs 2". on noe) © oO ro) ° ce) 5 12) o Ss) co) 2 oO fe) © Ss £ o's oY a £8 xr ® oH oe Q a w oO € £ Oo a. je) ac is Fig. 3. Understorey algae collected at “The Hump” in 2004: The brown canopy species and articulated corallines persisted to 25 m deep, where the deeper- water brown algae Scytothalia dorycarpa, Glossophora nigricans, Carpoglossum confluens and Cystophora platylobium occurred sporadically. Wet- weights of canopy species (Fig. 2) indicate a reverse changes with depth. relationship with depth to that found at “The Boulders”, and may be related to the effect of the higher wave energy and steeper slope at the site. The red algae Phacelocarpus peperocarpos and Plocamium preissianum became increasingly common with depth. (See Fig..3.) MARINE BENTHIC ALGAE OF THE ALTHORPE ISLANDS, SOUTH AUSTRALIA 119 Fig. 4. “The Monument”. Photo: A. Hirst. Fig. 5. Inshore of “The Monument” with bleached algal turf in the lower intertidal zone. 3. “The Monument’(MO), a low rocky promontory on the NW side of the Mooring Bay This is a relatively flat promontory of jointed granitic rock extending about 300 m from the base of a calcarenite cliff (see Fig.4). It is almost bisected halfway along its length, and there is a shallow channel to 3 m deep on its eastern outer end. There is moderate wave action near the cliff face, with some sand scour near the shore. The shallow channel is subject to high wave surge and the seaward extremity of the promontory experiences high wave energy. The different conditions along the length of the promontory are reflected in the change of algal dominants. Inshore, calcified Liagora harveyana occurred in shallow calmer water in gutters formed in the jointed rock, with Cladosiphon vermicularis in the lower eulittoral on shaded walls or in rock pools. Cartilaginous red algae Chondrophycus tumidus and Gigartina densa formed a distinct band, bleached in summer, in the sublittoral fringe zone with some articulated coralline algae (Fig. 5). Cystophora moniliformis was prominent in shallow seaward areas subject to wave surge, with Caulerpa brownii in the lower eulittoral — upper sublittoral and some Caulerpa papillosa. 120 R. N. BALDOCK & H. B. S. WOMERSLEY Ecklonia radiata occurred in the sublittoral, but had shorter stipes and was less dominant than at other sites. Large patches of co-dominant Lobospira bicuspidata and Cystophora monilifera occurred mixed with some Acrocarpia paniculata. Patches about 2 m in diameter of Caulerpa flexilis var. muelleri, sometimes mixed with Caulerpa brownii and Caulerpa geminata, occurred on horizontal rocks 1 —4 m deep midway along the promontory. Caulerpa obscura appeared in crevices and shaded slopes almost to the low water mark, and Dictyopteris muelleri was found at the entrance of caves, 2 — 3 m deep. The farthest extent of the promontory, exposed to higher wave energy, had a prominent sublittoral fringe with Ulva australis, Gelidium australe, and denuded Cystophora intermedia. 4. Mooring Bay (MB) This is a concave, sandy bay about 300 m across bounded in the east by “The Boulders” and in the north by “The Monument”. A flat, granitic substratum partly sand covered with a mono-specific Liagora harveyana community occurred at 1 — 2 m depth on the eastern side. Shepherd & Brook (2005) reported it to be sheltered, with smooth granite boulders of 1 —2 m relief, depth 1 — 3 m with Cystophora and Sargassum spp and Liagora, and patches of the articulated corallines Corallina officinalis and Jania verrucosa. 5. Sea cave (SC), SE coast A fracture in the granitic basement rock has been eroded to a deep channel in the cliff face some 50 m long and 15 m deep underwater, narrowing to a cave 11 m deep underwater ending abruptly at a rock fall 7 m deep underwater with a hole in the slumped calcarenite ceiling above water (see Fig. 6). Near- vertical sides and cave environment supported deep- water and shade-loving bryozoans, hydroids and algal species. These latter included the red algae Rhodophyllis membranacea, Carpopeltis phyllophora, Halymenia plana and Sonderopelta coriacea, and the brown alga Homoeostrichus sinclairit. Western Isles (WI) A string of five isles about 10 m high on the exposed western side of the main island have been formed by erosion of an amphibolite dyke (Robinson Fig. 6. Sea cave entrance, with the jointed basement granitic rock overlain by slumped calcarenite. MARINE BENTHIC ALGAE OF THE ALTHORPE ISLANDS, SOUTH AUSTRALIA et al. 1996). The calcarenite topping has been eroded extensively leaving mainly conical granitic mounds above water. Shepherd (pers. comm.) who also dived at the site reports the sea floor to be generally 15 — 25 m deep, with moderate to strong surge, smooth granite substratum, | — 2 m relief and in some places undercut with caves. He noted that Ecklonia, Acrocarpia and Cystophora spp formed the canopy TABLE |. Site Codes used for algal collections. 121 layer. A form of Cystophora retorta with narrow branches was collected. NW coast of the main island. This region is about 100 m offshore from “The Monument”. Collections at 10, 16, and 23 m were made by divers on 6 February, 2004. Code Site Depth (m) Collector(s) recorded Date on herbarium sheets Collections prior to 2004 NS __ N side of Althorpe I., SAORI expedition 9-13 R. Baldock 04.1.1964 EX coast exposed to high wave energy 5, 10, 15 J. Baker and K. Edyvane 26.x.1993 SH sheltered coast 5, 10, 15 J. Baker and K. Edyvane 27.x.1993 IT intertidal/drift collection A. Gaut 27-8.ix.2001 SS saline soak, S of “the Hump”, 50m inland E. Lawler 28ix.2001; 28.xi.2004 2004 expedition BO “The Boulders” 0-8 R. Baldock 31.1.2004 HU = “The Hump” 4-25 R. Baldock 31.1.2004 MO “The Monument” 0-4 R. Baldock 01.11.2004 SC Sea cave 7-11 R. Baldock 03.11.2004 NW _ Northwest coast 10, 16, 23 J. Brook and R. Lewis 05.11.2004 Wi Western Isles 10, 27 J. Brook, R. Lewis and 06.11.2004 N. Barrett MB_ Mooring Bay 1-2 S. Shepherd 11.11.2004 TABLE 2. List of Species with Depths in metres Cyanophyta Rivulariales-Rivulariaceae Calothrix ? Rivularia firma Womersley Chlorophyta Ulvales-Ulvaceae Blidingia marginata (J. Agardh) P. Dangeard tEnteromorpha clathrata (Roth) Greville ? ¥ Enteromorpha intestinalis (Linnaeus) Link? T Ulva australis Areschoug Cladophorales-Cladophoraceae Apjohnia laetevirens Harvey Chaetomorpha aerea (Dillwyn) Kitzing Cladophora valonioides Sonder Codiales-Codiaceae Codium pomoides J. Agardh Caulerpales-Caulerpaceae Caulerpa brownii (C. Agardh) Endlicher Caulerpa flexilis Lamouroux Caulerpa geminata Harvey Caulerpa obscura Sonder Caulerpa papillosa J. Agardh Caulerpa scalpelliformis (R. Brown) C. Agardh Caulerpa vesiculifera Harvey IT, on Chaetomorpha aerea MO, mid-eulittoral SS IT, in shallow pool IT, in shallow pool BO, MO, 0-3m BO, 8m IT, in shallow pool HU, 17-25m MO, Im BO, MO, NS, SH, 0-13m BO, MO, NS, SH, 0-13m MO on Ecklonia base, 0-3m IT, as drift; MO, NS, SH, 0-13m BO, MO, SH, 0-10m NS, NW, MO, 0-23m NS, 9-13m T the recent merging of Enteromorpha with Ulva by Hayden, H. S. et al. (2003) has not been recognized 122 R. N. BALDOCK & H. B. S. WOMERSLEY TABLE 2. Cont. Phaeophyta Chordariales-Ralfsiaceae Ralfsia verrucosa (Areschoug) J. Agardh Chordariaceae Cladosiphon vermicularis (J. Agardh) Kylin Sphacelariales-Sphacelariaceae Sphacelaria novae-caledoniae Sauvageau Stypocaulaceae Halopteris paniculata (Suhr) Prud’homme van Reine ? Halopteris pseudospicata Sauvageau Phloiocaulon spectabile Reinke Cladostephaceae Cladostephus spongiosus (Hudson) C. Agardh Dictyotales-Dictyotaceae-Dictyoteae Dictyopteris muelleri (Sonder) Reinbold Dictyota diemensis Kiitzing Dilophus fastigiatus (Sonder) J. Agardh Glossophora nigricans (J. Agardh) Womersley Pachydictyon paniculatum (J. Agardh) J. Agardh Zonarieae Distromium flabellatum Womersley Distromium multifidum Womersley Exallosorus olsenii (Womersley) Phillips Homoeostrichus sinclairii (Hooker & Harvey) J. Agardh Lobospira bicuspidata Areschoug Zonaria spiralis (J. Agardh) Papenfuss Zonaria turneriana J. Agardh Sporochnales-Sporochnaceae Sporochnus radiciformis (R. Brown) C. Agardh Scytosiphonales-Scytosiphonaceae Colpomenia sinuosa (Mertens) Derbes et Solier Scytosiphon lomentaria (Lyngbye) Link Laminariales-Alariaceae Ecklonia radiata (C. Agardh) J. Agardh Fucales-Hormosiraceae Hormosira banksii (Turner) Decaisne Seirococcaceae Scytothalia dorycarpa (Turner) Greville Cystoseiraceae Acrocarpia paniculata (Turner) Areschoug Carpoglossum confluens (R. Brown) Kiitzing Caulocystis cephalornithos (Labillardiere) Areschoug Cystophora intermedia J. Agardh Cystophora monilifera J. Agardh Cystophora moniliformis (Esper) Womersley & Nizamuddin Cystophora platylobium (Mertens) J. Agardh Cystophora retorta (Mertens) J. Agardh ? Cystophora siliquosa J. Agardh Sargassaceae Sargassum paradoxum (R. Brown ex Turner) Hooker & Harvey Sargassum vestitum (R. Brown) C. Agardh Sargassum: Arthrophycus? (base only) MO, on Cellana shell, lower eulittoral MO, 0-3m BO, on Metagoniolithon radiatum, 8m HU, 0-4m BO, MO, NW, 0-16m HU, 17-25m MO, 0-3m MO, 1.5m MO, NS, 0-13m NS, NW, WI, 9-16m HU, SC, WI, 7-27m BO, IT, MB, NS, SH, 0-13m IT as drift, SH, 10m SH, 0-10m SC, 7-1 1m BO, EX, HU, NS, NW, SH, WL 5-27m HU, MO, NS, 0-13m BO, SH, 8-25m HU, WI, 17-27m NS, 9-13m BO, MO, 1-8m IT MO, and widespread but not collected MO, 0-3m HU, 17-25m BO, MO, NS, 1-13m HU, NS, WI, 9-27m IT, as drift BO, IT as drift, MO, 0-1m BO, MO, 1-3m BO, MO, 1-3m HU, NS, 17-25m WI, 10m (form with narrow branches) MO, 3m IT, as drift BO, 0-5m BO, MO, 1-8m MARINE BENTHIC ALGAE OF THE ALTHORPE ISLANDS, SOUTH AUSTRALIA TABLE 2. Cont. 123 Rhodophyta Balliales’-Balliaceae Ballia callitricha (C. Agardh) Kiitzing Nemaliales-Liagoraceae EX, HU, IT as drift, NS, NW, WI, 5-27m Helminthocladia australis Harvey Liagora harveyana Zeh Gelidiales-Gelidiaceae Gelidium australe J. Agardh Pterocladia lucida (Turner) J. Agardh Pterocladiella capillacea (Gmelin) Santelices & Hommersand Gigartinales-Peyssonneliaceae Peyssonnelia capensis Montagne Peyssonnelia foliosa Womersley Peyssonnelia novae-hollandiae Kiitzing Sonderopelta coriacea Womersley & Sinkora Polyidaceae Rhodopeltis australis Harvey Halymeniaceae Carpopeltis phyllophora (Hooker & Harvey) Schmitz Halymenia plana Zanardini Kallymeniaceae Austrophyllis alcicornis (J. Agardh) Womersley & Norris? Callophyllis lambertii (Turner) J. Agardh Cirrulicarpus nanus (J.Agardh) Womersley Gigartinaceae Gigartina brachiata Harvey? Gigartina densa Edyvane & Womersley Polyopes constrictus (Turner) J. Agardh Dicranemataceae Peltasta australis J. Agardh Areschougiaceae Callophycus laxus (Sonder) Silva Callophycus oppositifolius (C. Agardh) Silva Plocamiaceae Plocamium angustum (J. Agardh) Hooker & Harvey Plocamium cartilagineum (Linnaeus) Dixon Plocamium costatum (C. Agardh) Hooker & Harvey Plocamium dilatatum J. Agardh Plocamium leptophyllum Kiitzing ? Plocamium preissianum Sonder Phacelocarpaceae Phacelocarpus apodus J. Agardh Phacelocarpus peperocarpos (Poiret) Wynne, Ardre & Silva Nizymeniaceae Nizymenia conferta (Harvey) Chiovitti, Saunders & Kraft MB, MO, Im MB, MO, 0-3m BO, EX, MO, NS, NW, SH, WI, 0-16m EX, HU, NW, SC, WI, 5-16m IT in pool, SH, 0-5m EX, WI, 15-27m EX, 15m EX, 10-15m EX, HU, SC, WI, 7-27m WI, 27m EX, HU, NS, SC, WI, 5-27m SC, 7-11MSC 7-11m NW, 23m NS, NW, WI, 9-16m SH, ?m MO, 0-3m MO, 0-3m BO, EX, HU, MO, 0-10m WI, 27m WI, 10m WI, 10m BO, EX, MO, NS, NW, SH, 0-23m EX, IT, as drift; SH, 0-15m EX, SH, 5-15m WI, 27m SH, 5m EX, HU, NW, SH, 5-23m HU, MO, NS, NW, 0-27m EX, HU, NW, SC, SH, WI, 5-27m NS, 9-13m ‘ tentatively placed in this systematic position on the basis of genetic sequencing by Choi et al. (2000) 124 R. N. BALDOCK & H. B. 8S. WOMERSLEY TABLE 2. Cont. Cystocloniaceae Rhodophyllis membranacea (Harvey) Hooker & Harvey ex Harvey Rhodophyllis multipartita Harvey Hypneaceae Hypnea ramentacea (C. Agardh) J. Agardh Hypnea valentiae (Turner) Montagne Mychodeaceae Mychodea marginifera (Areschoug) Kraft Gracilariales-Gracilariaceae Curdiea angustata (Sonder) Millar Curdiea obesa (Harvey) Kylin Melanthalia abscissa (Turner) Hooker & Harvey Rhodymeniales-Champiaceae Champia zostericola (Harvey) Reedman & Womersley? Rhodymeniaceae *Rhodymenia foliifera Harvey Rhodymenia verrucosa Womersley Corallinales-Sporolithaceae Sporolithon durum (Foslie) Townsend & Woelkerling. Corallinaceae Arthrocardia flabellata (Kitzing) Manza ssp. australica Womersley & Johansen Cheilosporum sagittatum (Lamouroux) Areschoug Corallina officinalis Linnaeus Haliptilon roseum (Lamarck) Garbary & Johansen Jania affinis Harvey Jania micrarthrodia Lamouroux Jania verrucosa Lamouroux Metagoniolithon radiatum (Lamarck) Ducker Metagoniolithon stelliferum (Lamarck) Weber-van Bosse Metamastophora flabellata (Sonder) Setchell Synarthrophyton patena (Hooker & Harvey) Townsend Bonnemaisoniales-Bonnemaisoniaceae Asparagopsis armata Harvey Delisea hypneoides Harvey Delisea pulchra (Greville) Montagne Leptophyllis conferta (R. Brown ex Turner) J. Agardh Ptilonia australasica Harvey Ceramiales-Ceramiaceae-Wrangelicae Wrangelia nobilis Hooker & Harvey Wrangelia plumosa Harvey Dasyphileae Muellerena wattsii (Harvey) Schmitz SC, 7-11m HU, NW, WI, 10-27m BO, NS, NW, 0-16m MO, 0-3m MO, 0-3m SC, 7-1 1m WI, 27m EX, MO, NS, HU, 0-15m BO on Laurencia in lower eulittoral, EX, HU, NW, SH, 5-25m EX, MO, 15-16m BO, 4-5m MO, 0-3m BO, EX, HU, NS, NW, SC, 4-16m BO, MB, 0-1m BO, on Ecklonia bases and Acrocarpia, EX, HU, MO, NW, SH, 0-25m WI, on Ballia, 27m MO, 0-3m MB, SH, 1-10m BO, EX, MO, NS, 0-15m SH, 10m HU, SC, WI, 7-27m WI, on Ballia and Phacelocarpus, 27m BO, MO, SH, 0-S5m HU, NS on Acrocarpia paniculata, 5-27m NW, WI, 23-27m WI on Ballia, 27m WI, 27m NW, 16m IT, eulittoral pool HU on Phacelocarpus peperocarpos, 15m * The earliest name available for this complex, previously known as R. australis Sonder or R. sonderi Silva. (H.B.S.Womersley) MARINE BENTHIC ALGAE OF THE ALTHORPE ISLANDS, SOUTH AUSTRALIA TABLE 2. Cont. 125 Antithamnieae Acrothamnion preissii (Sonder) Wollaston Antithamnion biarmatum Athanasiadis Antithamnion hanovioides (Sonder) De Toni Pterothamnieae Inkyuleea mariana Choi, Kraft & Saunders Heterothamnieae Elisiella arbuscula (J. Agardh) Womersley ? Griffithsieae Griffithsia elegans Baldock Ptiloteae Euptilota articulata (J. Agardh) Schmitz Spyrideae Spyridia dasyoides Sonder Ceramieae Centroceros clavulatum (C. Agardh) Montagne Ceramium filiculum Womersley Ceramium pusillum Harvey Dasyaceae Dasya baldockii Parsons & Womersley Heterosiphonia gunniana (Harvey) Reinbold Heterosiphonia microcladioides (J. Agardh) Falkenberg Thuretia quercifolia Decaisne Delesseriaceae-Nitophylloideae Acrosorium ciliolatum (Harvey) Kylin Crassilingua marginifera (J. Agardh) Papenfuss Haraldiophyllum erosum (Harvey) Millar & Huisman Hymenena multipartita (Hooker & Harvey) Kylin Rhodomelaceae-Polysiphoniae Polysiphonia decipiens Montagne Polysiphonia sertularioides (Grateloup) J. Agardh Herposiphonieae Herposiphoniella plurisegmenta Womersley Lophothalieae Haplodasya urceolata (Harvey) Parsons Amansieae Osmundaria prolifera Lamouroux Laurencieae Chondrophycus tumidus (Saito & Womersley) Garbary & Harper Laurencia clavata Sonder Laurencia elata (C. Agardh) Hooker & Harvey Laurencia filiformis (C. Agardh) Montagne Laurencia majuscula (Harvey) Lucas Laurencia spp HU on Zonaria, W1 on Inkyuleea, 17-27m SC on Metamastophora, 7-\\m BO, MO on Polyopes constrictus, 0-3m SC, WI, 7-27m BO on Acrocarpia, 0-1m SC on Glossophora, 7-\\m NW, SC, WI, 7-23m WI, 10-27m IT rockpool BO on Acrocarpia, 0-1m BO on Acrocarpia, MO on Lobospira, 0-3m NS, 9-13m NS, 9-13m HU epiphytic, 17-25m NW, 23m HU on Homoeostrichus sinclairii, SC on Glossophora bases , 5-1 1m NW on Callophyllis, 16-23m WI on /nkyuleea, 27m NW on Haliptilon, 16m BO on Acrocarpia, MO, 0-3m MO, 0-3m NS on Prerocladiella capillacea, 9-13m BO on Cystophora intermedia, 0-1m NS, 9-13m BO, MO, 0-3m BO, 5m EX, 5m IT as drift, SH, 5m SH, 10m IT, MO, 0-3m 126 R.N. BALDOCK & H. B. S. WOMERSLEY Species for which the Island is the type locality: Dasya_ baldockii, and ~~ _Herposiphoniella plurisegmenta. Extension of Distribution Range The distributions for Peltasta australis and Antithamnion biarmatum, previously known mainly from Tasmania and Victoria, have been extended westwards. Discussion The Althorpes are characterised by crystalline granitic substrates, temperate waters and high wave energy. Canopy species of subtidal communities are largely Acrocarpia paniculata and Ecklonia radiata, with Cystophora monilifera and C. moniliformis also prominent. Large patches of Caulerpa spp. and Phacelocarpus spp. predominate at depth and as the understorey. Shepherd & Sprigg (1976) and Edyvane & Baker (1998), believe the high macroalgal diversity and productivity of islands at the entrance to South Australian gulfs including the Althorpes, are probably a result of a number of geographic factors including high wave energy, summer nutrient upwelling and geographic position between oceanic and gulf waters with appreciable tidal flows from the gulf. Edyvane & Baker pointed to the unique position of the Gambier and Althorpe islands at the boundary of autumn and summer water masses at gulf entrances (Petrusevics 1993). This higher algal diversity is, however, typical of the whole west coast of Eyre Peninsula and south coast of Yorke Peninsula. Phillips (2001) added habitat hetero-geneity, marine transgressions and regressions and lack of mass extinctions to the list of factors contributing to the high macroalgal biodiversity and endemism of southern Australian regions. Womersley & Edmonds (1958) had earlier classified the South Australian coast on the basis of intertidal and upper subtidal organisms (mainly algae), wave action and substrate. Two categories of coasts are relevant. One with extreme wave action and steeply sloping crystalline rocks occurs on the southwest tip of Yorke Peninsula, and west and south Kangaroo Island. The other lies between areas of the first and has more sheltered conditions, moderate wave action, and flat calcarenite shore platforms interspersed with sandy beaches, It is prominent on southern Yorke Peninsula and the north coast of Kangaroo Island. The Althorpes appear to lie geographically at the junction of these two types, but ecologically with the first. The Althorpe Islands are positioned at the boundary of the St Vincent Gulf and Eyre Bioregions (as classified by IMCRA Technical Group 1998), and contain features of both bioregions according to Baker (2005). Edyvane & Baker (1998) noted the higher species richness of the region’s outlying islands, particularly Haystack Island. However, this high diversity is typical of the high wave energy coasts of Yorke and Eyre Peninsulas and Kangaroo Island in general (see Womersley & Edmonds 1958). Although only 144 species are documented in this present paper, further detailed collecting in the intertidal and sublittoral throughout the year would undoubtedly increase the known flora from this algal-rich southern Yorke Peninsula region. Acknowledgements SARDI financed the expedition, and provided the research vessel Ngerin and diving facilities. A. Hirst helped crop algae for biomass determinations and provided photographs. The Friends of Althorpe Is kindly hosted research teams, and they and the Innes National Park staff helped greatly with the collection and transfer to shore of equipment and supplies. The State Herbarium provided research facilities including curatorial support. In particular, Ms Carolyn Ricci was invaluable in collating and documenting collections. References Baker, J. (2005). “Towards a System of Ecologically Representative Marine Protected Areas in South Australian Marine Bioregions — Technical Report”. Report for Coast and Marine Conservation Branch, Department for Environment and Heritage, South Australia. 1250p. Cnot, H-G., Krart, G. T. & SAUNDERS, G. W. (2000). Nuclear small-unit rDNA sequences from Ballia spp. (Rhodophyta): proposal of the Balliales ord. nov., Balliales fam. noy., Ballia nana sp. nov. and Inkyuleea gen. nov. (Ceramiales). Phycologia 39:272-287 EpyvANE, K. S. & Baker, J. L. (1996). “Marine Biogeography of Spencer Gulf, South Australia”. Final report to the Nature Conservation Agency. (S. Australian Research and Development Institute (SARDI) Aquatic Sciences, West Beach, S. Australia.) 29pp + & (1998) “Marine benthic survey of Investigator Strait-Gambier Isles, South Australia”. Report to Environment Australia (Marine Protected Areas Program) Project D801, stage 4. 23pp + (SARDI, Aquatic Sciences, West Beach, South Australia.) MARINE BENTHIC ALGAE OF THE ALTHORPE ISLANDS, SOUTH AUSTRALIA 127 Haypen, H. S., BLomster, J., Macas, C. A., Siva, P. C., STANHOPE, M. J. & WAALAND, J. R. (2003). Linnaeus was right all along: Ulva and Enteromorpha are not distinct genera. Eur. J. Phycol. 38: 277-294. IMCRA TecHNICAL Group (1998). Interim Marine and Coastal Regionalisation for Australia: an ecosystem- based classification for marine and _ coastal environments, Version 3.3. Environment Australia, Canberra. Petrusevics, P. M. (1993). SST fronts in inverse estuaries, South Australia: Marine Biogeography. Consultancy Report to PISA, S. Australia. (SARDI, West Beach, S. Australia.) Puitiips, J. A. (2001). Marine macroalgal biodiversity hotspots: why is there high species richness and endemism in southern Australian marine benthic flora? Biodiv. and Cons. 10:1555-1577. RosINson, T., CANTY, P., Mooney, T. & Ruppuck, P. (1996). “South Australia’s offshore islands”. Dept. of Environment and Natural Resources South Australia. SHEPHERD, S. A. & BROOK, J. B. (2005) Foraging ecology of the western blue groper, Achoerodus gouldii, at Althorpe Islands, South Australia. Trans. R. Soc. S. Aust. 129, 202-208. SHEPHERD, S. A. & SpricG, R. C. (1976). Substrates, sediments and subtidal ecology of Gulf St Vincent and Investigator Strait. Jn Twidale, C & Webb, B. P. (Eds.) “Natural History of the Adelaide Region.” pp. 161-174. (R. Soc. S. Australia, Adelaide.) WENIU, C., SCHAHINGER, R. B. & LENNON, G. W. (1990). Layered models of coastal upwelling: a case study of the South Australian region. Jn Davies, A. M. (Ed.) “Modelling Marine Systems” pp. 73-91. (CRC Press, Boca Raton, Florida.) Womers_Ley, H. B. S. (1981) The marine ecology of temperate coasts. Jn Clayton M.N. & King, R. J. (Eds), “Marine Botany: an Australasian Perspective”, Ch. 11, pp. 294-306. (Longman Cheshire, Melbourne.) (1984) “The Marine Benthic Flora of Southern Australia. Part I” (Govt. Printer, Adelaide). (1987) “The Marine Benthic Flora of Southern Australia. Part II” (Govt. Printer, Adelaide). (1994) “The Marine Benthic Flora of Southern Australia. Rhodophyta - Part IIA” (ABRS, Canberra). (1996) “The Marine Benthic Flora of Southern Australia. Rhodophyta - Part IIIB” (ABRS, Canberra). (1998) “The Marine Benthic Flora of Southern Australia. Rhodophyta - Part IIIC” (State Herbarium of South Australia). (2003) “The Marine Benthic Flora of Southern Australia. Rhodophyta - Part IID” (ABRS, CANBERRA). & Epmonps, S. J. (1958). A general account of the intertidal ecology of South Australian coasts. Aust. J. Mar. Freshw. Res. 9 (2):217-260, plates 1-12. Transactions of the Royal Society of S. Aust. (2005), 129(2), 128-144. SUBTIDAL MACROFLORA OF ALTHORPE AND HAYSTACK ISLANDS, SOUTH AUSTRALIA by J. L. BAKER!, G. J. EpGAR? & N. S. BARRETT? Summary Baker, J. L., EpGar, G, J. & BARRETT, N. S. (2005), Subtidal macroflora of Althorpe and Haystack Islands, South Australia. Trans. R. Soc. S. Aust. 129(2), 128-144, 30 November, 2005. The benthic macroflora at sites around the Althorpe and Haystack Is was surveyed in October 1993 and January 2004. Nearshore reefs were dominated by the kelp Ecklonia radiata, the fucoid Acrocarpia paniculata and mixed Cystophora species, whereas reefs at Haystack I. were dominated by a multi-species canopy of mixed Sargassum and Cystophora species, with Ecklonia and Seirococcus axillaris also present. On sandy bottoms off Althorpe I. and Haystack I., seagrass communities comprised mixed beds of up to five seagrass species, within four genera (Amphibolis, Halophila, Posidonia, and Heterozostera). Biomass and species composition of the canopy macroalgae were similar at those sites, which were surveyed in both 1993 and 2004, particularly at sites where few species dominated the canopy. For the upper mid-sublittoral samples at two sites around Althorpe L., a similar average number of understorey species was recorded per m? in both 1993 and 2004. At comparable sites, the understorey species composition differed between the two time periods; however, Caulerpa flexilis was a dominant component during both 1993 and 2004. The islands are species-rich, with 267 species within 151 genera so far recorded to 27 m depth. During the 1993 survey, at least 180 species of macroalgae were recorded in 29 quadrat samples, many of these from Haystack I. The macroal gal species richness at Althorpe and Haystack Is is comparable with that of the Investigator Group and Nuyts Archipelago. These figures highlight the importance of these islands in terms of macroalgal species richness in SA. Key Worpbs: Macroalgae, subtidal survey, Althorpe Is, southern Australia, temperate macroalgal distributions, macroalgal species richness. Introduction Macroalgae are a dominant feature of sublittoral reefs in the temperate waters of southern Australia, which is a particularly species-rich region (Womersley 1984, 1987, 1990, 1994, 1996, 1998, 2003; Huisman ef al. 1998). In South Australia (SA), which is especially rich in macroalgae, surveys at several islands and bays during the past four decades have contributed significantly to the knowledge of macroalgal structure, distribution and zonation, and the physical factors influencing those patterns (e.g. Shepherd & Womersley 1970, 1971, 1976). This paper adds to the catalogue of work from island groups in SA coastal waters. In 1993, reef macroflora and seagrass communities at Althorpe and Haystack Is were sampled during a state-wide benthic survey program (Edyvane & Baker*). During the 2004 expedition some of the South Australian Research & Development Institute, 2 Hamra Ave, West Beach, South Australia 5024 and Dept Applied and Molecular Ecology, University of Adelaide, Waite campus, Urrbrae, South Australia 5064. > Present address: 8 Fairfield Ave, Somerton Park, South Australia 5044, Email: jjbaker@senet.com.au Tasmanian Aquaculture and Fisheries Institute, University of Tasmania, Private Bag 49, Hobart, Tasmania 7011. Edyvane, K. S. & Baker, J. L. (1998). “Marine Benthic Survey of Investigator Strait — Gambier Isles, South Australia”. Report to Environment Australia (Marine Protected Areas Program): Project D801 (Stage 4). (SARDI Aquatic Sciences, South Australia). Es same sites at Althorpe and Haystack Is were revisited and sampled for biomass (Baldock & Womersley 2005). Further sampling of percent cover was undertaken by the second and third authors during surveys to assess the area for habitat value as part of potential Marine Protected Areas. This paper presents the results of those surveys from the two time periods, particularly the data on biomass and species richness, and considers the biogeographic relations of the island group Regional oceanography and site description Frontal systems form seasonally in southern Spencer Gulf, and the Althorpe Is are at the boundary of these temperature fronts, where warmer gulf waters and cooler deeper waters off western Kangaroo I. meet, causing strong benthic temperature and salinity differentials (Bruce & Short 1992; Petrusevics 1993). Sea surface temperatures around the Althorpe Is are ~15 — 17° C in winter and 19 — 21° C in summer, and summer upwellings off western Kangaroo I. and associated nutrients may penetrate to the entrance of Investigator Strait (Petrusevics 1993; Kampf et al. 2004). The prevailing swell diffracts around Althorpe 1., such that differences in water movement between various sides of the island are slight (Shepherd ef a/. 2005). The Althorpe Is lie at the boundary of the Eyre (EYR) Bioregion and the Gulf St Vincent (GSV) Bioregion, two of eight “meso-scale” biogeographic SUBTIDAL MACROFLORA 129 L = Yorke Peninsula <>! Ati. a 1 d me Zz Nell / se) o a e > _ Investigator Strait ee | | ' : = O-PS Ne Wi pal ¢ - (a Kangaroo Island Fig. 1. Map of the Althorpe and Haystack Is showing location in western Investigator Strait, at the boundary of the Eyre and Gulf St Vincent Bioregions. Site numbers (see Table 1) are shown in white. regions recognised for SA (IMCRA Technical Group 1998) (Fig. 1). The Althorpe Is subtidally are of smoothly sloping granitic basement reefs, interspersed with gaping crevasses and chasms, except for the occasional fallen calcarenite block in the shallows, whereas Haystack I. has a substratum of eroded calcarenite overlying a granitic spine. The geology and topography of the Althorpe and Haystack Is are described in some detail by Rankin ef al. (1991), Murray-Jones & Shepherd (2005) and Zang (2005). Methods Sampling and analysis 1993: Sampling was undertaken from 24 — 31 Oct. 1993 with | m? quadrats. Mean % cover of macroalgae and seagrasses was estimated visually, For two sites each at Althorpe I. and Haystack I. (Fig. 1), the contents of four replicate quadrats (taken at places which were representative of the habitat) at 5, 10, and on occasion 15 m depths were harvested, and subsequently preserved, weighed and identified (Table 1). 2004: Algal biomass data were collected at five sites by Team A (Table 1, and see Baldock & Womersley 2005), while Team B, as part of a wider survey assessing biodiversity, used 20 replicate 50- point quadrats (1 m2) placed along a series of 50 m transects at 5 m and 10 m depths to estimate percent cover at six sites (Table 1). Multivariate classification procedures such as non- metric multi-dimensional scaling (MDS) (Kruskal & Wish 1978; Field ef al. 1982: Clarke 1993) and cluster analysis (Romesburg 1984, Clarke 1993) were used to elucidate spatial patterns in the 1993 and 2004 macroalgal biomass data. All weights given are wet weights in g or kg m2. The data, which range over four orders of magnitude, were transformed (logi)) to increase the number of species that contribute to the computation of dissimilarity between samples, and to prevent the few species of highest biomass from dominating analyses. Canopy and understorey data were ordinated separately by an MDS procedure in SYSTAT y. 11, using the Bray- Curtis dissimilarity measure for the species by sites matrices, and normalised Euclidean distance as the constant in the Minkowski metric, for computing distances between points in the MDS configuration. Kruskal’s F-Stress Formula 1 (Kruskal & Wish 1978) was used as the goodness-of-fit statistic for the regression of 2-dimensional distances versus dissimilarity values. The resulting positions in the MDS plot were validated using a cluster analysis (Romesburg 1984) procedure in SYSTAT vy. 1 1, with the Bray-Curtis dissimilarity measure as a distance metric, and the UPGMA clustering algorithm. Other methods used to display patterns of sampled species composition and/or abundance include summary tables and charts, and a species accumulation curve. Results Macrofloral Composition Ecklonia radiata is generally the dominant canopy species, with Acrocarpia paniculata as a co- dominant, on hard substrates in the mid-sublittoral (5 ~— 15 m depth) around Althorpe I. (Table 2, Fig. 2A, B), except near the sand line near seagrass beds (e.g. 10 m depth, Boulders - see Table 2). At Site 2, 130 JL. BAKER, G. J. EDGAR & N. S. BARRETT TABLE 1. Sites sampled, and collecting methods used during surveys of Althorpe and Haystack Is in 1993 and 2004. Site numbers are shown in Fig. 1, B= biomass in g wet weight m2. PC = percentage cover m?. Team, Site Number Sampling Method Place location, and depth and variable measured 1993 Site 1. The Boulders, 5, 10 m 1.0 m? quadrats B, PC Althorpe I. Site 2. Off Chain Islet, 5, 10, 15 m A-2004 Site 1, The Boulders, 0 — 8 m 0.25 m?— 1.0 m? Althorpe I. Site 2. Off Chain Islet, 4-27 m quadrats Salmon Inlet / Sea Cave, 7— 11m B Site 3. NW Bay, 10, 16, 23 m Site 5. Western Isles, 10, 27 m B-2004 Site 1. The Boulders, 5, 10 m 50 points per quadrat, Althorpe I. Site 2. Off Chain Islet, 5, 10 m 20 quadrats per site. PC Site 3. NW Bay, 10 m Site 4, Swallowtail Bay, 5, 10 m Site 5, Western Isles 5, 10 m 1993 Site 6,5, 10m 1.0 m? quadrats. B, PC Haystack I. Site 7, 10m B-2004 Site 6,5 m; Site 7, 5m 50 points per quadrat, Haystack I. 20 quadrats per site. PC Ecklonia comprised 65 — 75% cover and ~2 —8 kg m? from 5 mand 15 m depth in 1993, and similarly high cover and biomass in 2004. Elsewhere, at Sites 1, 3, and 4, biomass and % cover of Ecklonia were also high in 1993 and 2004 (Table 2, Fig. 2A,B). In the upper mid-sublittoral (5 — 10 m), the cover and biomass of Acrocarpia were high at Site 2, with values of 1.5 — 3.6 kg m? recorded in some quadrats. At the same depths sampled off Haystack I., this species was only a minor component of the canopy flora. A few species of Cystophora, mainly C. moniliformis and C. monilifera, were also present in the mid-sublittoral around Althorpe [., although the latter species was not recorded in 1993 at Site 2 (Table 2). At Althorpe 1, during both the 1993 and 2004 surveys (Table 1), the fucoid Seirococcus axillaris, was largely absent and, when present, occurred only as a minor component of the canopy (i.e. average <100 g m? at Site 2 at 15 m). In contrast, at Site 7, Haystack [., Seirococcus was a major canopy species at 10 m depth (Table 2), with 1.0 — 4.2 kg m? recorded in 1993. The 2004 survey also recorded S. axillaris at Site 6, Haystack I., but in low densities. The fucoid Scytothalia dorycarpa (closely related to Seirococcus) was uncommon or absent at most sampled sites around Althorpe and Haystack Is in 1993 and 2004 (Table 2, Fig. 2A,B); however it was recorded at 5 m depth at Site 5 (Fig. 2B), and at 23 m at Site 3, with a biomass of 960 g m>°. Five species of seagrass were recorded on sand at 10 m depth at Site 1, Althorpe L, and Site 6, Haystack I. (Table 2). At Site 1, Posidonia sinuosa was dominant but P angustifolia was not recorded, whereas at Site 6, the opposite was found. At both sites Halophila australis, Heterozostera nigricaulis (formerly H. tasmanica), and Amphibolis antarctica were also recorded. At Site 5, upper mid-sublittoral canopy composition was similar to that recorded around Althorpe L., with Ecklonia as the dominant cover, and lesser quantities of Acrocarpia and species of Cystophora (Fig. 2A,B). Differences in the canopy at Site 5 included the presence of Scytothalia dorycarpa and differences in the composition of the Cystophora species present (e.g. C. retorta, recorded at 5 m). In deeper water (23 m) at Site 3, Acrocarpia (1690 g m2), Scytothalia dorycarpa (960 g m”), Seirococcus axillaris (865 g m*) and Cystophora platylobium (670 g m) occurred among the dominant Ecklonia ( ~10 kg m); however, samples at 27 m depth at the same site revealed Ecklonia (biomass 0.7 — 2 kg m7”) as the only large brown alga, among a predominantly rhodophyte community (biomass 800 — 1700 g m*). Common species were: Delisea pulchra, Ballia callitricha, Inkyulea mariana, Rhodophyllis multipartita, Phacelocarpus peperocarpos, Sonderopelta coriacea, and Spyridium dasyoides, and understorey brown algae such as Homoeostrichus sinclairti and Zonaria spiralis. Other red macroalgae recorded in the deeper waters at Site 3 included Carpopeltis phyllophora, Callophycus — oppositifolius, Curdiea obesa, Leptophyllis conferta, Peltasta australis, Plocamium dilatatum, Ptilonia australasica, Rhodopeltis australis, the arborescent coralline Metamastophora flabellata, the coralline Jania pulchella (= J. affinis) SUBTIDAL MACROFLORA 131 0 Ecklonia Acrocarpia Cystophora monilifera C. moniliformis C. retorta C. retroflexa Sargassum sonderi Sargassum fallax Sargassum sp. Ecklonia Acrocarpia Scytothalia dorycarpa Seirococcus axillaris Cystophora monilifera L Scytothalia dorycarpa NW Haystack NE Haystack Swallowtail Bay The Hump / off Chain Islet The Boulders Swallowtail Bay NW Bay The Hump / off Chain Islet Western Isles The Boulders Fig. 2. Mean percentage cover of canopy macroalgae recorded at (A) 5m depth and (B) 10 m depth at selected sites around Althorpe I. and Haystack I. in 2004. Site numbers and locations are described in Table 1. and the small calcareous epiphyte Synarthrophyton patens. In contrast to Althorpe Is, Acrocarpia was uncommon at Haystack I. and Ecklonia was co- dominant at 5 — 10 m depth with other taxa including Sargassum and Cystophora species and Seirococcus axillaris (Table 2; Fig. 2A). Haystack I. was rich in Sargassum species, with 10 species recorded from the 1993 and 2004 surveys combined. The major component canopy species were variously S. linearifolium, S. verruculosum, S. sonderi, S. fallax, C. moniliformis, and C. monilifera, with some differences in relative abundance between the two sampling periods (Table 2; Figs 2A, 4). At 10 m cl € maaquny sypaydo]rD 91 SYDAISND D]JA}DIUBUOLT c 17 v iS LOL Ol S6 DYMLAIN IVI DIVE GI stusongyy sypdydossny g $6 6Le val 87 piouap sisdospindsy 6 pIVINsY VIsNOYISaAp 8 I DISABUOI DISNOYISAAP ce I suasaajan] piuyoldy sarsadg Aat0jssapuy) (I) popsosay sarseds e €1 €l € Z € € Cc € v Ss ¢ 9 9 Adour Jo Joquiny [20 x 0961 I 96 SLUD]JIXD SNIIOIOMAS & ov 87 pduvoA.op vypyjoyAag a cI SI (¢) UNIISAaA WNssPBADS nH clv WNSOJNINAAIA WNSSDBADS Z wv 9E1 wnaasynuids wnssps.vg gs v0 wnxoppind WNSSD3.ADg a 61 6001 (4 wintpofiinauly WNSsvBADS & 81 es wNYyd1owosajaYy WNSSDBADG ao 16 10€ 99 xD]JDf WNSSDB.ADS i ‘3 ‘6 SI suaidizap wnssps.Ang S.. “Eye vcol (406) 8rr9 SLES pO6E p9Sl ecsp TIL 9Iv £681 a DIDIPDA DIUO [YI & 8 €Z pypujoinfqns vsoydojsay m4 cl | psonbyis vsoydojsy a 661 € wnigojainjd vaoydojsy i Te £97 orl €Il Ivvl SII plafymuou psoydojs¢y 901 COL 88I 6ST vss Tes L¢l LI smuofyiuow vsoydojsay 9Il C8 CIT suanfuor unssojsod4v) SV 0s9 CCOT €Ig OLS eri 6L C60 871 (0) pypjnaiund midavz0doy sarsadg Adour (1) £661 £661 £661 P00 £661 r00T ~—- £661 r00C £661 P00 £661 r00T ~— £661 ADAING JO ILIA Ol ¢ x01 g ¢ Ol Ol a al ¢ ¢ 8 x01 Lous 9 US 9 ONS TOUS TOUS zoug Tas «6 Z ONS TOUS = ONS POMS CT OHS TOUS SOUS uoy) ssa] fo av8jvo.19pul fo saizads Aaso} aip ‘wninp uoyy[o1ods sp yons ‘saul]]D409 asojsn’) ‘sajduips JDApo' 2 pun 666] Ul ST yonIsdoyy pun adsoysjy ay) 10 ‘yjdap wu ¢I-¢ wos p: NX Lon) “‘PaPlOIAL ILIM SASSDABDIS YIIYM JD SAIS = x “(ul) Y. SUaPUN SSDULOIG MO] Sapnjoul ,, dap 0} tafad ‘sSuippay uuinjoo ul CI ‘OT ‘¢ S#aquinn “| xipuaddp ut pajst] ‘zu 8 ¢ satsady Aa1ojsiapuy) ‘ON JPIOL,, “1 A1QPL Ut passt] 24D Sans fo suolp207T ‘2jqn} sspuolg ays Ul papnjoul jou nb aynoydas woul paspiaan (;.1u 8) sspuolg 1yS1am ysaif aio DID *[ A/GRI Ul Pq! apsoras (Al) saydydida parpiossy pup (111) sasspa8pas ‘an3]posopUl (11) Aasojssapun pun (1) Adoun) °Z FAVA UISAP AAD SUOIDIOT “FOOT 133 SUBTIDAL MACROFLORA eS wWNIUUDYII]DI DIUOYdISOsa]aH] 6£ psafiy]aquin DIpojIOAAJaE] vl if pIsa1ojf DIUIUIA]OH] 8EC I€ 8r pyoaidsopnasd si4ajdojoHy ST usstaid pusajdojopy if i €l ¢ 8 tl v € uinasod uopydiDoET €C Sasa] DISYIYfIUH € € v Ol sunasiu pl6oydossojH IZ DIDIYIDAG DULADSID) II € vl Ol 8 € es 3 L9 apagsnD UNIpI]eD I¢ Lajjanu WintUojIOLYJAAT 0c € c snyoisyspf snydopiq 8P ppyaoinf DIOAIIIG ai pUOojOYyoIp DJOAIIG ov psoppia DASDG if 9 | saplo1inzabUu DASDG 61 snaopjuaups sndip20padspay I I 6S 8r syoujoyffo puyjn.10y 9€ pyundo wnayjipjaoD L9 saplowod wnipoy at layjanul un1pod (a6 wnsojpiuiniu wnipoy cis unjpalDs wnipoD 9 snupu sndavaynaay ¢ «AS, Dlupuoyy q if Cc € (4 TE I 8S WNIDIJISDS wnaodsopiay) 9S c 9CT I pjoaisajsoz pidupyy L Z vsoyidpd pduajnvy 79 pinasqo ndi1ajnvy (ayjanu ‘IeA Suipnypout) 9€1 81 p6T =: CERT €0l 8 syixayf pdiajnoy Cc I I I 8s € 9LT 5 psoydoyjAyd syjadodivy ov I € puilafisuns siydydoyo) saedg Aor0jsaapuy (1) £661 £661 £661 FOOT £661 FOOT £661 FOOT £661 FOOT £661 FOOT £661 AdAINS JO 19, Ol iS x01 iS ¢ Ol Ol SI SI ¢ iS 8 x01 Lous 9 SUS 9 OS as TOUS CHS TMS THS CHS LHS TMS Lees Tes SOUS ‘MOD °T AIAVE, J. L. BAKER, G. J. EDGAR & N. S. BARRETT 134 Nam a t+ N 07 Cl 61 9 -— ONAN cs sa Le SC LY SOI At wo + L9 es Ol val a eT Ic SL IT ST LI eT LOC $6 181 lon el oI (a0) vs 6r 0€ LOI 81 al 9€ 87 cf 68 sypaids p1ipuoz saplouyonids passoquoasaqgay psoydaudsos pdiprouuag paan1409 vyjadosapuos DSOMAMA DIUAUAPOYY SYDAISND DiuauApoyYy supjoa syjdydopoyy psaslanjo DIUuOpqoyy pplony vIpvjz041ag paanpp1dd2 VIpvjI01A suaidiap piuoydisajog unupissiaid WNIWUDIO[ lsuajsiau WNUDI0, g UNAUISYINADI UNIUDIOLY WNISNSUD WNIUDIO[ sodupso1adad sndups0jaavyd snpodp sndupsojaanyg avIPUD]JOY-avAOU DYaUUOssdad psoyof pyauuossdag sisuadpo pyauuossdag wnjojnaiupd uodjoipAyang piafijoid DLaDpunuscE psojnuips papoyodyy slupjnaian papoyoay wunsafiyjays UoYIYOUOSHIaW UNIDIPDA UOYIOIUOSVIAWy pUULIUOD DIEDYJU_jaW DSSIOSqD DIIDYJU_IaW pyopidsnaig bidsoqoT SIUAOfIf DIDUANDT DIDJa DIDUAINDT paonjuaups paudAyy IAIDIOUIS SNYI1AJSOAOWOFT satsadg Aat0jsaapuyy (1) £661 £661 Ol ¢ Lous 9 aus P00 ¢ CUS £661 r00E ¢ Ol Cc OUS TOUS £661 Ol Z aus rO0T SI TOUS £661 SI CUS £661 ¢ T oS r00E 8 T oS AQAING JO Ted, sous UuoD “7 WAV, 135 SUBTIDAL MACROFLORA L9 91 9S IT LV c8T p9E sol £661 £661 £661 vl Or Lous Si 9 US x01 9 as el £661 SI TOUS 14 r00e Or TOUS 6l &661 Ol TOUS suaasa1siu WUNAJUBIOAUP synvaiunyd siapydalqyjvy vyojjaaipad piuafxa,0g wnjlf uoydisopy]D sojyAydidy sseaseasg (AD) 8s9 psonuis pIuOpIsSOg O€I SYDAISND DIUOPISOg pyofisnsnp piuopisog 197 SIJMDILABIU DABJSOZOAIJAL] 89 sypasnp ppydoppy] OI poyospjup syoqiydup sarsadg sseasvas (I) II TZ ST IZ 91 €€ Popsosdy so1edg Aosojssopuy) JO Joquiny [BIO], DUDLIGUAN] DIADUOZ satsodg Ao103s10puyQ (1) vOOT £661 r00E £661 POOT £661 ABAING JO TRI, ST SI ¢ iS 8 x01 CHS THUS TUS THUS Lugs Tus SOUS OD °T WAVE, 136 J. L. BAKER, G. J. EDGAR & N. S. BARRETT 2800 30 % _ 2400 25 “E 2000 $ S 20 ° 1600 a 15 & «wo 1200 © 7 a) S 800 10 6 2 ° soll! GoUetool. 3? o _ Gl a rae 2 ; : ek ; xe & &. Cs wes . Ss s se s RS Se s & RRS RS Rs Se SOS SF RRS ees ¥ Sd Seo BH OOO FY 2? Fe ES ss @? © RS ee SS Po Ne Ss NS ve 2 is sre Se 3 © < oe ea RS SSS ee oe e S eS SS ry & a & Roe & e o G & i) xa ee oe Fig. 3. Dominant canopy macroflora sampled in 1993 (grey bars) and 2004 (black bars), at Sm depth, Site 6, Haystack I. Data for 1993 are biomass (g wwt / m7’), averaged for replicate quadrat samples, with SD bars. Data for 2004 are percentage cover. Similarity in canopy species composition between 1993 and 2004 is indicated by overlapping bars (1.c. black on grey). F-Stress = 0.067 T T HUMP1593 HUMP 1093 L BOULS504 _| HUMP504 HUMPS93 @ © cies HUMP 1004 E0uLs4@ HUMP 1504 | HUMP504@ = @ Al HUMP 1004 @ HUMP1504 UM ES2s, HUMP 1093 @ BOUL804 @ BOUL 1093 BOUL504 | | BOUL1093 @ BOUL593 BOUL593 if T T 1 0.0 0.5 1.0 1.5 Distances F-Stress = 0.089 HUMP1593 HUMP1504 | @HumPs93— | HUMP1093 HUMP 1593 @ HUMP1004 UMP504 auirrieds HUMP504 2 HUMP 1504 @ ® B0UL1093 _ HUMP593 BOUL1093 HUMP 1004 | @ @ BOULS593 BOUL593 L rai | BOUL804 @ s0utsos soe 4 [ T T T T 1 0.0 0.1 02 #03 O04 O65 Distances Fig. 4. Non-metric multi-dimensional scaling and cluster analysis of the log-transformed biomass data for (A) canopy species, and (B) understorey species, recorded at Site 1 (The Boulders) and Site 2 (The Hump) at Althorpe I. in 1993 and 2004. Numbers after the site name indicate depth (m) and year (93 or 04). SUBTIDAL MACROFLORA 137 depth, canopy species recorded at Haystack I. that were not common at other sampled sites included Carpoglossum confluens and Seirococcus axillaris. At the Althorpe I. sites, common understorey species in the mid-sublittoral included: species of Caulerpa (particularly C. flexilis); various turfing browns; and numerous erect and encrusting red algae (Table 2). During the 1993 survey, Caulerpa flexilis was especially abundant at 5 m at Site | (Table 2). Reefs in the upper mid-sublittoral Sites 6 and 7, Haystack I., were particularly rich in understorey species, whose species composition differed considerably from that recorded at Althorpe I. (Table 2, Appendix 1). Comparison of 1993 and 2004 Collections The 2004 survey has shown some consistency over time in the composition of the main canopy species over the 11-year period, particularly the dominance of Ecklonia and Acrocarpia on mid-sublittoral reefs around Althorpe L, and the co-dominance of Ecklonia amongst a mixed canopy of Cystophora and Sargassum species on reefs around Haystack I. (Table 2, Fig. 2); however, one noticeable difference between the two sampling periods was the absence of Acrocarpia in samples taken at Site | (5 m) in 1993, and its dominance there in 2004. Haystack I. (Site 6) also showed striking differences over the 11-year period in species composition and abundance of some species at 5 m, notably Sargassum (Table 2, Fig. 3). Sargassum sonderi was the major species recorded in 2004, whereas S. linearifolium, S. paradoxum, S. spinuligerum and S. verruculosum were recorded most frequently in 1993. A multivariate analysis of the canopy and understorey data for the two survey periods indicated little difference between samples at 5 and 10 m depths at Site 2 (off The Hump). In both 1993 and 2004 these pairs of data had a low calculated Bray- Curtis distance, and thus appeared close together in the ordination space and the cluster tree (Fig. 4A). Similarly, the 10 m samples at Site 2 in 1993 and 2004, and the 5 and 8 m samples at Site 1 (The Boulders) in 2004 were relatively close (Fig. 4A), whereas the 5 m sample from 1993 was relatively TABLE 3, Numbers of understorey species per m? recorded in 1993 and 2004, at 5 m depth at Sites 1 and 2. Understorey Site 2 Site | No. m? (1993, 2004) 13, 14 21, 16 Sum of all species recorded 22 31 in 1993 and 2004 No. species >4 g m? recorded I 4 in both 1993 and 2004 Total no. and % of species recorded 6 6 in both 1993 and 2004 (27%) (19%) isolated, primarily because this was the only Althorpe I. site sample at which Acrocarpia paniculata was not recorded. For the understorey data, the most similar pairs of samples were from 5 and 8 m depths at Site | (sampled in 2004), and from 5 and 10 m at Site 2 (also 2004). Generally, the multivariate analysis showed that samples from comparable depths taken at the two different time periods were not similar, with few 1993 and 2004 samples grouping closely together in the ordination space. The average number of understorey species recorded per m? at 5 m depth at Sites 1 and 2 was remarkably similar over time (Table 3), although the species composition differed substantially (Table 2). Thus, <30% of understorey species were common to both sites at 5 m depth in both years (Fig, 2, Table 3). Notable understorey species recorded during both 1993 and 2004 included Ballia callitricha at Site 2, and Gelidium australe, Plocamium angustum and Metagoniolithon radiatum at Site 1. For all samples taken between 5 and 15 m deep at Site 2, 40% of the understorey species were recorded during both time periods. Spatial Patterns in Species Composition and Richness After averaging replicate samples for each depth, the number of canopy species per m? was strongly correlated with the number of understorey species per m? at Althorpe and Haystack Is sites (r = 0.79). Where a dense cover of Ecklonia (and sometimes Acrocarpia) were dominant (e.g. at 5 — 15 m depth at Site 2), few other canopy species were present, and fewer understorey species, compared with sites (e.g. Sites 6, 7) where Ecklonia and/or Acrocarpia were less dominant (Table 2). The sampled species composition in the mid- sublittoral differed both within and between islands. At the finest scale (replicate quadrats) there were few species in common between samples. For example, of the 27 species recorded at 5 m at Site | in 1993, about half were found only in one of four replicate quadrats, and only four species (the common Cystophora moniliformis, C. monilifera, Caulerpa flexilis, and Plocamium angustum) were found in all quadrats. Similarly at the 10 m site, about half of the species were recorded in only one quadrat, and only four of the 45 species were found in all four replicate quadrats. At the 5 m site at Site 2, only three (Ecklonia radiata, Ballia callitricha and Peyssonnelia novae-hollandiae) of the 23 species were recorded in three replicate quadrats. At Althorpe I., the understorey species composition differed sub- stantially between Sites 1 and 2 (Table 2; Fig. 4B: Appendix). For example, during the 1993 survey, only 12% of macroalgal species recorded on reefs at 138 J. L. BAKER, G. J. EDGAR & N.S. BARRETT 10 m deep were common to both sites. Of the 50 understorey species that were collectively recorded from 5 — 10 m at the two sites in 1993, only 8 (16%) were found at both sites. In the 2004 sampling, 27% of the macroalgal species recorded at 5 m were common to the two sites. Differences in species composition were also observed between sites of similar exposure at two different islands. For example, only 18% of the species from 5 m depth samples at Sites 1 and 6 were common to both. Haystack I. was particularly rich in macroalgal genera and species, both in the canopy and the understorey, compared with Althorpe I. Using the averaged biomass data (Table 2) for replicate quadrats, 2 — 6 canopy species per m? were recorded for the Althorpe I. sites in 1993, compared with 13 at the Haystack I. sites. Similarly, for understorey species, <30 species per m* were recorded at most Althorpe I. sites in 1993, compared with 49 — 67 at Haystack I. (Table 2, Appendix). Within genera, examples of the species richness at Haystack I. include the chlorophyte Codium (4 species), and the rhodophytes Mychodea (5 species) and Hymenocladia (3 species). The family Rhodomelaceae was well represented at Haystack I., with at least 15 species recorded during the 1993 survey (Table 2, Appendix 1). The Althorpe Is and Haystack I. together were rich in macroalgae, with at least 232 species within 136 genera recorded from the depth range 3 — 27 m. In addition to those cited above for Haystack I., other examples of the species richness at the Althorpe Is include: the canopy genera Sargassum (10 species) and Cystophora (7 species); the understorey genera, Caulerpa (10 species), Dictyota (5 species), Phacelocarpus (4 species), Plocamium (7 species), Laurencia (5 species), Griffithsia (3 species) and Heterosiphonia (3 species) (Table 2, Appendix). While neither survey was designed to determine species richness, it is notable that during the 1993 survey alone, at least 180 species of macroalgae were recorded in 29 quadrat samples, from Sites 1 and 2, Althorpe I., and Sites 6 and 7, Haystack I. (Table 2; Fig. 5; Appendix). The “steps” in the curve of cumulative species in Fig. 5 result from sampling at species-rich Haystack I., where most of the 11 x 1m? quadrat samples yielded species not previously recorded in other samples. The curve in Fig. 5, which does not reach an asymptote, shows that further sampling from Haystack I. would likely yield even more species of macroalgae. Discussion Macroalgal Biogeography of Althorpe and Haystack Is Due to its central geographic location in the Flindersian Province, the Althorpe Is contain many species from the warmer western part of the Flindersian Province, and the cool water Maugean Sub-Province, as well as species with a broad Flindersian distribution, extending to SE NSW (Womersley 1990). The islands also lie between the Eyre and Gulf St Vincent Bioregions, and have various geomorphological, oceanographic and biological features that are characteristic of each of these bioregions. These include: the high species 200 4 5 | eeeee ® 160 + 9 | geeee a 120 > oe ‘ | eee ra 80 | 7) o ] > = 40 = Ss 8 5 0 ——— ee 4 0 5 10 16 20 25 30 35 L no. samples Fig. 5. Species accumulation curve for 29 x 1 m? quadrat samples from Althorpe and Haystack Is Samples were taken at 5, 10 and 15 m, at Sites 1,2 Althorpe L., and Sites 6,7 Haystack I. Samples are plotted in randomised order, and include all canopy and understorey species (except for encrusting corallines). SUBTIDAL MACROFLORA 139 diversity and abundance of red macroalgae, and the seasonal influence of nutrient-rich, cooler water upwellings, characteristic features of the Eyre Bioregion (IMCRA Technical Group 1998, Baker 2005). Ecklonia, Acrocarpia, and widely distributed species of Sargassum and Cystophora together form the major structural components of the reef system. Examples of species with a western Flindersian distribution (Cowan 2000) include the green Cladophora valonioides, and the rhodophytes Cliftonaea pectinata, Dasya_ baldockii, and Heterosiphonia_ callithamnium. Maugean Sub- Province species include the canopy brown Sargassum vestitum, the turfing brown Exallosorus olsenii, and the reds Gigartina muelleriana, Peltasta australis, Ptilonia australasica, Muellerena wattsii, Melanthalia abscissa, Shepleya wattsii, Mychodea hamata, and Gloiocladia fruticulosa. The canopy species composition recorded at Althorpe and Haystack Is shares some common features with canopies of island reefs both to the west (e.g. Investigator Group off western Eyre Peninsula) and east (e.g. West Island, Encounter Bay). Similarities include the dominance of Ecklonia, Acrocarpia paniculata, and the presence of mixed Cystophora species (Shepherd & Womersley 1970, 1971; Baker & Edyvane 2003). At Haystack I., the mixed canopy of various Sargassum and Cystophora species, co-dominant with Ecklonia, is a feature of other less wave- exposed reefs throughout the central SA coast, such as the Sir Joseph Banks Group, Spencer Gulf (JLB unpublished data). The widely distributed Flindersian species Scytothalia dorycarpa, and the more easterly (Maugean) species Seirococcus axillaris, were not common at the Althorpe Is or Haystack I., although the former species is dominant elsewhere in central SA, and the latter at islands further east (e.g. West I.). The records of six species in this survey extend their known geographic range to the west. These species, mostly Maugean, are: Exallosorus olsenii (sub nom. Homoeostrichus olsenii in Womersley 1987), Sargassum vestitum, Cystophora retroflexa, Gigartina muelleriana, Peltasta australis, and Ptilonia australasica. Also of note from these surveys are records of species with limited known spatial distributions in SA. Examples include: (i) the uncommon coralline Amphiroa gracilis, a sub-tropical species known from WA, SA and Qld (Cowan 2000). In SA it is found mainly in the Gulfs region and around Yorke and Eyre Peninsulas (Womersley 1996; Plant Biodiversity Centre and DEH 2004); and (ii) Arthrocardia wardii, an articulated coralline whose distribution extends to NSW and Tas. In SA, the species has been found only in Investigator Strait and at Kangaroo I. (Womersley 1996; Plant Biodiversity Centre and DEH 2004). Canopy Species Composition in Space and Time A number of studies have shown that on reefs where Ecklonia is dominant, the corresponding understorey is low in biomass and/or species due inter alia to effects of shade, scour, and occupation of available space (Kennelly 1987, 1988; Schiel 1990; Kendrick et al. 1999; Fowler-Walker & Connell 2002; Turner & Cheshire 2003; Kendrick er al, 2004). The Althorpe I. reefs were mostly dominated by canopies of Ecklonia and Acrocarpia, with few other canopy species, and a consequent low diversity and abundance of understorey species. The higher species richness at Haystack I. than at Althorpe Is may be due to a combination of factors. The highly rugose, actively eroding, calcareous substrate at Haystack I., compared with the smooth granite substrate at Althorpe Is, creates more micro- habitats and greater opportunity for algal propagules to settle, particularly within the Rhodophyta. Also, the mixed Sargassum-Cystophora canopy at Haystack I. promotes a more diverse understorey (e.g. Kendrick et al. 2004), possibly through the provision of more space and light, and less scour, compared with canopies of Ecklonia. A feature of increased species richness is the presence of many rare or uncommon species (Hubbell 2001), as we found at Haystack I. Despite the fact that the exact sites were not revisited in 2004, the apparent consistency over time of canopy composition at these sites at Althorpe I. (a feature previously noted by Baker & Edyvane (2003) at Nuyts Archipelago) suggests a measure of stability in these assemblages. Although alternative states in algal assemblages may occur (Turner & Cheshire 2003), the prevailing exposure at any given site may ensure the long-term persistence of particular canopy genera (e.g. Fig. 4, Turner & Cheshire 2003). At Haystack I. the main differences observed over the 11 years were in the species and cover of Sargassum at Site 6. These could be variously due to shifts from one alternative state to another (see above), and to seasonal differences in time of sampling between surveys. As shown by Edgar (1983), some Sargassum species seasonally shed reproductive branches and laterals. Understorey Species Composition in Space and Time At sites investigated in this study, generally <30% of the understorey species were recorded in both 1993 and 2004, although a similar number of understorey species per m? were recorded during both time periods. The apparent persistence of Caulerpa flexilis at 5 m at Site 1 at both time periods 140 J. L. BAKER, G. J. EDGAR & N. S. BARRETT was presumably facilitated by its stolonic matte, which can pre-emptively occupy space and exclude competitors (Shepherd 1981). Changes in species composition over time are likely for many reasons. These include: fine-scale patchiness of species; rarity of many species, especially when species richness is high (Baker & Edyvane 2003); opportunistic and ephemeral strategies of some species (Shepherd 1981); and effects of shading, storms, grazers, sediment and scour (Kennelly 1987, 1988; Connell 2005). Species Richness Intertidal and subtidal collecting at Althorpe and Haystack Is to 30 m depth have so far yielded a total of 276 species within 151 genera. This compares with 240 species in the Investigator Group (Shepherd & Womersley 1971; Baker & Edyvane 2003), ~230 species at St Francis Isles (Shepherd & Womersley 1976; Womersley & Baldock 2003), 364 species at Waterloo Bay (Shepherd & Womersley 1981), and 132 species at West I. (Shepherd & Womersley 1970). Hence the Althorpe and Haystack Is have a similarly rich flora to other islands of the eastern Great Australian Bight, which is notable for its algal diversity. Thus, of the estimated 961 macroalgal species recorded in SA waters (Womersley 1984, 1987, 1994, 1996, 1998, 2003; Cowan 2000), at least 28% of them, excluding the crustose corallines, have been found to date at these islands. Acknowledgments We thank the many organisations and individuals who assisted in these surveys and their supporting organizations: the former Australian Nature Conservation Agency, and the Australian Heritage Commission for funding the 1993 survey; SARDI for provision of MRV N¢gerin, and diving facilities in 1993 and 2004; R. N. Baldock for collecting and processing algal biomass data in the 2004 expedition, and for algal identifications; A. Bloomfield and J. Brook for assistance in the collection of percent cover and species data; the Marine Protected Area program of Department for Environment and Heritage for funding the visit and work of the second and third authors; the Friends of Althorpe Islands for hosting research teams in 2004; Innes National Park staff for logistic assistance; the State Herbarium for research facilities and curatorial support; especially Professor H. B. S. Womersley for algal identifications; K. Edyvane, the survey team leader in 1993; divers assisting with algal collections, namely G. Andrews, R. Baldock, J. Brook, B. Davies, A. Hirst, M. Kinloch, R. K. Lewis, I. McGrath, P. Preece, S. A. Shepherd, and G. Westphalen; the crew of the MRV N¢eerin for their tireless assistance; G. Lorenzin, A. Doonan, V. Day, S. Deakin and C. Ricci for laboratory assistance; and lastly the editors and anonymous reviewers for helpful criticism and improvements to the paper. References Baker, J. L. (2005) “Towards a System of Ecologically Representative Marine Protected Areas in South Australian Marine Bioregions-Technical Report”. Report to Dept. Environment and Heritage, S. A. URL http://www.environment.sa.gov.au/coasts/pdfs/mpa_ report/2005report.pdf & EbDYVANE, K. S. (2003) Subtidal macrofloral survey of St Francis and Fenelon Islands, South Australia. Trans. R. Soc. S. Aust. 127, 177-187. BALpbock, R. N. & WoMERSLEY, H. B. S. (2005) Marine benthic algae of the Althorpe Islands, South Australia. Trans. R. Soc. S. Aust. 129(2), 116-127. Bruce, B. D. & Suort, D. A. (1992) Observations on the distribution of larval fish in relation to a frontal zone at the mouth of Spencer Gulf, South Australia. Jn Hancock, D. A. (Ed.) “Larval biology. Australian Society for Fish Biology Workshop, Hobart, 20 August 1991”. Proceedings No, 15, pp. 124-137 (Bureau of Rural Resources, Canberra). CLARKE, K. (1993) Non-parametric multivariate analyses of changes in community structure. Aust. J. Ecol. 18, 177- 183. CONNELL, S. D. (2005) Assembly and maintenance of subtidal habitat heterogeneity: synergistic effects of light penetration and sedimentation. Mar. Ecol. Prog. Ser. 289, 53-61. Cowan, R. A. (2000). “Australian Marine Algal Name Index: A database of the taxonomy, nomenclature and distribution of Australian marine macroalgae”’. Accessed 8 Aug. 2005. URL http://dseweb.murdoch.edu.au/wise/ Epaar, G. J. (1983) The ecology of south-east Tasmanian phytal animal communities. 2. Seasonal change in plant and animal populations. J Exp. Mar. Biol. Ecol. 70, 159-179. Fieip, J., CLARK, K. & Warwick, R. (1982) A practical strategy for analysing multi-species distribution patterns. Mar. Ecol. Prog. Ser. 8, 37-52. FowLer-WALKER, M. J. & CONNELL, S. D. (2002) Opposing states of subtidal habitat across temperate Australia: consistency and predictability in kelp canopy-benthic associations. Mar. Ecol. Prog. Ser. 240, 49-56. HusBeELL, S. P. (2001) “The Unified Neutral Theory of Biodiversity and Biogeography”. (Princeton University Press, Princeton, USA). Huisman, J. M., Cowan, R. A. & ENTWISLE, T. J. (1998) Biodiversity of Australian marine macroalgae-a progress report. Bot. Marina 41, 89-93. INTERIM MARINE & COASTAL REGIONALISATION OF AUSTRALIA (IMCRA) TECHNICAL Group (1998) “Interim Marine and Coastal Regionalisation of Australia. An Ecosystem Classification of the Marine and Coastal Environments of Australia”. Version 3.3. (Environment Australia, Canberra), SUBTIDAL MACROFLORA 141 KAmpr, J., DOUBELL, M., GRIFFIN, D., MATTHEWS, R. L, & Warp, T. M. (2004) Evidence of a large seasonal coastal upwelling system along the southern shelf of Australia. Geophys. Res. Letters 31, 1-4. Kenprick, G. A., LAvery, P. A., & PHILLIPS, J. C. (1999) Influence of Ecklonia radiata kelp canopy on structure of macro-algal assemblages in Marmion Lagoon, Western Australia. Hydrobiol, 398/399: 275-283. , HARVEY, E. S., WERNBERG, T., HARMAN, N. & GOLDBERG, N. (2004). The role of disturbance in maintaining diversity of benthic macroalgal assemblages in southwestern Australia. Jap. J. Phycol. 52 (Supplem.) 5-9. KENNELLY, S. J. (1987) Physical disturbances in an Australian kelp community. 1. Temporal effects. Mar. Ecol. Prog. Ser. 40, 145-153. (1988) Effects of kelp canopies on understorey species due to shade and scour. Mar. Ecol. Prog. Ser. 50, 215-224, Kruskal, J. & Wisu, M. (1978) “Multidimensional Scaling". Sage University Papers. (Sage Publications, California. USA). MurraAy-JONES, S. & SHEPHERD, S. A. (2005) An expedition to the Althorpe Islands, South Australia: introductory narrative and conservation recommendations. Trans. R. Soc. S. Aust. 129(2), 85-89. PetRusevics, P. (1993) SST fronts in inverse estuaries, South Australia - indicators of reduced gulf-shelf exchange. Aust. J. Mar, Freshw. Res. 44, 305-323. PLANT BIlopiveRsity CENTRE & DEPARTMENT FOR ENVIRONMENT AND HERITAGE (2004) “Electronic Flora of South Australia”, The Census of SA Plants, Algae and Fungi. Accessed 8 Aug. 2005. http://www.flora.sa. gov.au/census.html RANKIN, L. R, FLInt, R. B. & BELPERIO, A. P. (1991). “Precambrian Geology of the Islands of the Investigator Strait Area, South Australia”. Department of Mines and Energy Report DME 425/89. (Department of Mines and Energy, South Australia). Rosinson, A. C., CANTY, P. D., Mooney, P. & RubDDICK, P. (1996) “South Australia’s Offshore Islands”. (Australian Government Publishing Service, Canberra). RoMEsBuRG, H. (1984) “Cluster Analysis for Researchers” (Wadsworth Publishing Co., California, USA). SculeEL, D. R. (1990) Macroalgal assemblages in New Zealand: structure, interactions and demography. Hydrobiologia 192, 59-76. SHEPHERD, S. A. (1981) Ecological strategies in a deep water red algal community, Bot. Marina 24: 457-463. & WomeERSLEY, H. B. S. (1970) The sublittoral ecology of West Island, South Australia. J. Environmental features and algal ecology. Trans. R. Soc. S. Aust. 94, 105-138. & (1971) Pearson Island expedition, 1969. 7. The subtidal ecology of benthic algae. Trans. R. Soc. S. Aust. 95, 155-167. (1976) The subtidal algal and seagrass ecology of St Francis Island, South Australia. Trans. R. Soc. S. Aust. 100, 177-191, (1981) The algal and seagrass ecology of Waterloo Bay, South Australia. Aquat. Bot. 11, 305-371. , Epaar, G. J. & BARRETT, N. S. (2005) Littoral reef fishes of Althorpe Islands and adjacent coasts of central South Australia. Trans. R. Soc. S. Aust. 129(2), 183-192. TURNER, D. J. & CHESHIRE, A. C. (2003) Encounter 2002 expedition to the Isles of St Francis, South Australia: Structure and productivity of benthic macroalgal communities. Zrans. R, Soc. S. Aust 127, 153-166. WoMERSLEY, H. B. S. (1984) “The Marine Benthic Flora of Southern Australia. Part I”. Handbook of the Flora and Fauna of South Australia. (Government Printer, Adelaide.) (1987) “The Marine Benthic Flora of Southern Australia, Part I”. Handbook of the Flora and Fauna of South Australia. (Government Printer, Adelaide.) (1990) Biogeography of Australasian marine macroalgae. /n: Clayton, M. & King, R. (eds.) “Biology of Marine Plants.” (Longman Cheshire, Melbourne). (1994) “The Marine Benthic Flora of Southern Australia. Rhodophyta - Part Illa”. Flora of Australia Supplementary Series: no. | (Australian Biological Resources Study, Canberra), (1996). “The Marine Benthic Flora of Southern Australia. Rhodophyta - Part IIIb”. Flora of Australia Supplementary Series: no. 5 (Australian Biological Resources Study, Canberra). (1998) “The Marine Benthic Flora of Southern Australia, Rhodophyta - Part IIIc” (State Herbarium of South Australia, Adelaide). (2003) “The Marine Benthic Flora of Southern Australia. Rhodophyta - Part IIId” (Australian Biological Resources Study, and State Herbarium of South Australia, Canberra). & BALbock, R. N, (2003) The Encounter 2002 expedition to the Isles of St Francis, South Australia: marine benthic algae. Trans. R, Soc. S. Aust. 127, 141-151. ZANG, W-L. (2005) Geology of Althorpe Island. Trans. R. Soc, S. Aust. 129(2), 90-93. Appendix List of sites at which minor species (<5 g m7?) were recorded at Althorpe and Haystack Is. A full list of species is available from the first author. Names of collectors are given by Baldock & Womersley (2005). Code Site Depth (m) Date NS N side of Althorpe I. 9-13 4.1.1964 NW Site 3. Althorpe L. 10, 16, 23 4-11.11,2004 SE Site 2, Althorpe I. 5, 10, 15 26.x.1993, 31.i-11.ii1.2004 NE Site 1, Althorpe L. 5, 10 27.x.1993, 31.i1-4.11.2004 H1 Site 6, Haystack I. 5, 10 29.x.1993 H2 Site 7, Haystack I, 10 29.x.1993 MO The Monuments, Althorpe |. 0 - 4 1.11.2004 sc Sea Cave, Althorpe I. 7-11 3.i1.2004 142 J. L. BAKER, G. J. EDGAR & N.S. BARRETT SW Site 4, Althorpe I. 5, 10 4-11.ii.2004 W Site 5, Western Isles. 5, 10, 27 4-11.ii.2004 Hl Site 6, Haystack I. 5, 10 29.x.1993; 4-11.ii.2004 H2 Site 7, Haystack I. 5, 10 29.x.1993; 4-11 .ii.2004 Chlorophyta Caulerpales - Caulerpaceae Caulerpa brownii (C. Agardh) Endlicher NE, MO, NS, SE, SW, H1, H2, 0-13m Caulerpa scalpelliformis (R. Brown) C. Agardh W, NS, NW, NE, MO, H1, H2, 0-23m Phaeophyta Chordariales - Chordariaceae Polycerea nigrescens (Harvey ex Kiitzing) Kylin H1, 10m Dictyotales - Dictyotaceae - Dictyoteae Dicytota alternifida J. Agardh NE, 10m Pachydictyon polycladum (Kiitzing) Womersley H1, 5m Dictyotales - Zonarieae Dictyopteris muelleri (Sonder) Reinbold — MO, SE, 1.5-10m Distromium flabellatum Womersley NE, H2, 10m Distromium multifidum Womersley NE, 5m Lobophora variegata (Lamouroux) Womersley ex Oliveira H1, H2, 5-10m Zonaria angustata (Kiitzing) Papenfuss H1, H2, 5-10m Zonaria crenata J. Agardh H2, 10m Fucales - Cystoseiraceae Myriodesma integrifolium Harvey H1, 5m Rhodophyta Gelidiales-Gelidiaceae Gelidium crinale (Turner) Gaillon H1, 5m Gigartinales - Gigartinaceae Gigartina muelleriana Setch. & N.L. Gardner H2, 10m Gigartina pinnata J. Agardh SE, H1, 5m Polyopes constrictus (Turner) J. Agardh SE, NE, MO, 0-10m Gigartinales - Dumontiaceae Dudresnaya australis J. Agardh ex Setch. H2, 10m Gigartinales - Kallymeniaceae Polycoelia laciniata J. Agardh SE, 10m Glaphrymenia pustulosa J. Agardh H2, 10m Gigartinales - Areschougiaceae Callophycus oppositifolius (C. Agardh) Silva NE, W, 10m Erythroclonium angustatum Sonder H1, 10m Rhabdonia coccinea (Harvey) Hooker & Harvey H2, 10m Rhabdonia verticillata Harvey H1, 5m Solieria robusta (Greville) Kylin H1, H2, 5-10m Gigartinales - Plocamiaceae Plocamium costatum (C. Agardh) Hooker & Harvey SE, NE, 5-15m Gigartinales - Phacelocarpaceae Phacelocarpus sessilis Harvey ex J. Agardh NE, 10m Gigartinales - Nizymeniaceae Nizymenia conferta (Harvey) Chiovitti, Saunders & Kraft NS, H1, H2, 5-13m Gigartinales - Cystocloniaceae Rhodophyllis membranacea (Harvey) Hooker & Harvey ex Harvey H1, SC, 5-11m Rhodophyllis multipartita Harvey | SE, NW, W, 10-27m Gigartinales - Mychodeaceae Mychodea hamata Harvey H1, 10m Mychodea marginifera (Aresch.) Kraft H1, 10m Mychodea pusilla (Harvey) J. Agardh H1, 10m SUBTIDAL MACROFLORA 143 Gracilariales - Gracilariaceae Gracilaria cliftonii Withell, A. Millar & Kraft H1, 10m Rhodymeniales - Rhodymeniaceae Botryocladia sonderi Silva H1, 5m Erythrymenia minuta Kylin H2, SE, 10-15m Gloiocladia australis (J.Agardh) Norris H1, 10m Gloiocladia fruticulosa (Harvey) Norris H2, 10ra Gloiosaccion brownii Harvey H1, 10m Hymenocladia chondricola (Sonder) Lewis H1, 10m Hymenocladia divaricata Harvey H1, 10m Hymenocladia usnea (R. Brown ex Turner) J. Agardh H1, 10m Rhodymeniales - Champiaceae Champia affinis (Hooker & Harvey) J. Agardh H1, 10m Champia viridis C. Agardh NE, H1, H3, 5-10m Rhodymeniales - Lomentariaceae Lomentaria sp. H2, 10m Corallinales - Corallinaceae Amphiroa anceps (Lamouroux) Decaisne SE, NE, NW, H1, 5-15m Amphiroa gracilis Harvey NW, SE, SW, 5-10m Arthrocardia wardii (Harvey) Areschoug H1, 10m Jania pulchella (Harvey) Johansen & Womersley NE, H1, 5-10m; W, on Ballia, 27m Jania verrucosa Lamouroux NE, 1-10m Corallinales - Mastophoroideae Metamastophora flabellata (Sonder) Setchell SE, SC, W, 7-27m Ceramiales - Ceramiaceae Perischelia glomulifera (J. Agardh) J. Agardh = H2, 10m Ceramiales - Ceramiaceae - Wrangelieae Anotrichium “sp.1” NE, H2, 5-10m Euptilocladia spongiosa Wollaston H2, 10m Hirsutithallia formosa (Harvey) Wollaston & Womersley H1, H2, 10m Involucrana crassa (Hooker & Harvey) Gordon-Mills H2, 10m Shepleya wattsii (Harvey) Gordon-Mills © H2, 10m Wrangelia nobilis Hooker & Harvey NW, 16m Wrangelia velutina (Sonder) Harvey H1, 10m Ceramiales - Ceramiaceae - Crouanieae Ptilocladia australis (Harvey) Wollaston H1, H2, 5-10m Ptilocladia pulchra Sonder NE, 10m Ceramiales - Ceramiaceae - Dasyphileae Dasyphila preissii Sonder H2, 10m Ceramiales - Ceramiaceae - Antithamnieae Antithamnion hanovioides (Sonder) De Toni SE, NE, MO on Polvopes constricus, 0-10m Ceramiales - Ceramiaceae - Spongoclonieae Spongoclonium “sp. 3” NE, 10m Spongoclonium sp. H2, 10m Ceramiales - Ceramiaceae - Griffithsieae Griffithsia monilis Harvey H1, 5m Ceramiales - Ceramiaceae - Spyrideae Spyridia dasyoides Sonder H1, W, 5-27m Ceramiales - Ceramiaceae - Ceramieae Ceramium pusillum Harvey H2, 10m Ceramium rubrum C. Agardh H1, 10m Dasyaceae Heterosiphonia gunniana (Harvey) Reinbold NS, H1, 5-13m Delesseriaceae-Nitophylloideae Acrosorium ciliolatum (Harvey) Kylin H1, 5-11m Apoglossum spathulatum (Sonder) Womersley & Shepley H1, 10m Crassilingua marginifera (J. Agardh) Papenfuss SE, NW on Callophyllis, 10-23m 144 J. L. BAKER, G. J. EDGAR & N. S. BARRETT Hymenena endiviaefolia (Hooker & Harvey) Womersley H1, H2, 5-10m Hypoglossum revolutum (Harvey) J. Agardh H1, 10m Laurencieae Laurencia majuscula (Harvey) Lucas NE, 10m Rhodomelaceae Cliftonaea pectinata Harvey H2, 10m Coeloclonium tasmanicum (Harvey) Womersley H2, 10m Dictvomenia harveyana Sonder_ H1, H2, 5-10m Epiglossum smithiae (Hooker & Harvey) Kutz. H1, H2, 5-10m Herposiphonia versicolor (Hooker & Harvey) Reinbold H1, 10m Laurencia brongniartii J. Agardh ~H1, 5m Lenormandia pardalis J.Agardh H1, 5m Protokuetzingia australasica Montagne H1, 5m Transactions of the Royal Society of S. Aust. (2005), 129(2), 145-157. INTERTIDAL MOLLUSCAN AND ECHINODERM DIVERSITY AT ALTHORPE ISLAND AND INNES NATIONAL PARK, SOUTH AUSTRALIA by K. BENKENDORFF* Summary BENKENDORFF, K, (2005). Intertidal molluscan and echinoderm diversity at Althorpe Island and Innes National Park, South Australia. Trans. R. Soc. S. Aust. 129(2), 145-157, 30 November, 2005. Species inventory data provide the first step towards understanding local marine communities. Here I apply rapid biodiversity assessment to ten intertidal sites, with two levels of habitat complexity: rock platforms with and without boulder fields. Five sites were located on Althorpe I., with a further five reefs in and around the Innes National Park on Yorke Peninsula. One hour, timed-search surveys were used to determine the species richness and rarity of macromolluscs and echinoderms. In total 82 molluscan species were found, but only eight echinoderms. A large proportion of species was found to be numerically rare (<5 individuals per site) or spatially rare (only observed at one or two sites). Reefs with boulder fields supported the highest species richness, and overall more species were found on Althorpe I, (66) than the mainland (59). Multivariate analyses revealed that different molluscan communities occur on rock platforms with and without significant boulder habitat. Different molluscan communities also appear to occur on Althorpe |. when compared to the mainland sites, Thus, a comprehensive marine reserve system should include representation from these two distinct intertidal rocky reef habitats, at both offshore and coastal locations on Yorke Peninsula. Key Worbs: Intertidal reefs; macromollusc; echinoderm; community composition; marine reserve selection; species inventory. Introduction Species inventory data provide the foundation for conservation and management of marine habitats. Unfortunately, however, inadequate inventory data are the “rule rather than the exception” for most Australian marine habitats (Smith 2005). The problem of access to many marine habitats means that they are poorly known compared to their terrestrial counterparts. As a consequence the application of systematic reserve selection procedures has been very limited in the marine environment (Pressey & McNeill 1996; Gladstone 2002) and, instead, site selection for protective measures has often been based on educated guesswork (Smith 2005). The limited resources available for marine species inventories have led to the need for development of rapid biodiversity assessment methods (see Benkendorff & Davis 2002; Gladstone 2002; Benkendorff 2003; Smith 2005). Typically, these rapid assessment programs aim to establish whether certain locations are representative and/or of particular ecological importance to the region, within a limited time frame. Species richness (a diversity) is often used as an indicator of ecological importance because it captures one important aspect of biological diversity and can be reliably assessed with "School of Biological Sciences, Flinders GPO Box 2100, Adelaide, South Email: kirsten.benkendorff@flinders.edu.au University, Australia S001. relative ease (Benkendorff 2003). Species richness inventories can be used to detect biodiverse hotspots and facilitate the prioritization of sites for conservation (Benkendorff & Davis 2002; Gladstone 2002), as well as produce estimates of the regional (y) diversity (e.g. Warwick & Light 2002), and the turnover of invertebrate assemblages between locations (B diversity e.g. Oliver & Beattie 1996). Species richness measures are nevertheless limited to describing only one aspect of the potential variation in ecological communities. Thus, many ecologists prefer to collect abundance data to calculate diversity indices (Bisby 1995), quantify spatial patterns of intertidal diversity (e.g. Davidson et al. 2004) or to undertake multivariate analyses of the community composition (e.g. Underwood & Chapman 1998). However, given the large natural variation that is known to occur in invertebrate communities on intertidal reefs, abundance data will only be useful if sufficiently replicated in space and time (e.g. Underwood & Chapman 1998). Benkendorff (2003) found that the collection of quantitative abundance data using replicated quadrats can greatly reduce the area of reef that can be sampled in limited time frames, thus negatively affecting the measure of overall species richness. An alternative approach used by Smith (2005) is to provide a semi-quantitative ranking using defined abundance categories for each species observed during search surveys. This scoring system can be used in multivariate comparisons of the species composition across habitats or locations, whilst not 146 K. BENKENDORFF adding greatly to the time taken to complete each survey. Rarity in most marine species is poorly understood (Chapman 1999) and thus any assessment of numerical abundance and spatial distribution will add to our current state of knowledge. Rapid biodiversity assessment will typically involve the use of one or more indicator taxa. Previous studies on intertidal reefs in NSW, Australia have demonstrated that macro-molluscs are effective surrogates for intertidal diversity, with high correlations with over-all species richness (Gladstone 2002; Smith 2005). Molluscs also have the advantage of being widespread and easily sampled, with relatively well understood taxonomy and ecology compared to other invertebrate taxa (Benkendorff & Davis 2002, Smith 2005). The species richness (a diversity) of intertidal reef macromolluscs can be reliably assessed using short (1 hr) unreplicated timed search surveys (Benkendorff 2003), and these surveys have been shown to produce accurate site rankings when compared to longer-term species richness data (Benkendorff & Davis 2002). Mollusc diversity can vary substantially between intertidal reefs, with habitat complexity likely to be a major driving factor. In particular, species-rich hotspots on the NSW coast are characterized by the presence of suitable boulders for molluscan egg mass deposition (Benkendorff & Davis 2004). In southern Australian waters, 95% of marine molluscs are estimated to be endemic (Allen 1999). Overall, the southern temperate flora and fauna is reported to have much higher levels of endemism than those in the tropical north (Zann 2000), thus highlighting the need for effective conservation in this bioregion. However, most of the marine estate under protection or management by the Commonwealth Government is on the Great Barrier Reef or off-shore e.g. in the Great Australian Bight (Edyvane 1996; DEH 2004). Currently, only 0.26% of the South Australian marine jurisdiction is under protection in aquatic reserves (DEH 2003). In Gulf St Vincent and Spencer Gulf, significant intertidal reefs are protected at Aldinga, Troubridge Pt and Whyalla. Notably, these are all limestone reefs with no representation of granite or basalt reefs on the S.A. mainland, Some near-shore islands with granite reefs, such as West I. and Goose I. are included in aquatic reserves. These islands may provide intertidal habitat that is relatively less impacted by humans than the mainland. Thus, an additional reserve around the Althorpe Is could prove beneficial for the conservation of local marine biodiversity. A ' Gaut, A, (2000). A preliminary amateur intertidal study of Althorpe Island Conservation Park. Friends of the Althorpe Islands Conservation Park, Unpublished report. comparative investigation of the community structure is needed to determine whether both offshore and near-shore reefs, encompassing the full range of substrata found in the region, should be represented in future marine conservation areas. Only one unpublished study has previously investigated the biodiversity of intertidal reefs on Althorpe I'. Molluscs were found to be the most numerous group, with a total of 17 species identified and shell evidence for several more. This list is likely to be an underestimation of biodiversity due to high swell conditions and the broad scope of the survey. Consequently, further assessment of the intertidal biodiversity is warranted. The main objective of this study was to use rapid biodiversity assessment surveys to obtain species inventories of the macromollusc and echinoderm fauna occurring at representative rocky intertidal sites on Althorpe I. and around Innes National Park on Yorke Peninsula mainland. The numerical and spatial rarity of each species in this region was also investigated. Multivariate analyses on the molluscs were then used to compare the community composition between the mainland and island locations. The influence of habitat was also assessed by comparing simple rock platforms to reefs with significant boulder fields across both locations. This study should help inform planning for a comprehensive and representative system of marine reserves in South Australia. Methods This study was undertaken at five intertidal sites along the rocky coast of Althorpe I., with a further five sites on the mainland of Yorke Peninsula, South Australia (Fig. 1). The mainland sites are collectively referred to as Innes National Park, although only four sites actually occur in the Park, with the remaining one ~10 km north at Marion Bay. The surveys were all conducted within the same week (6th — 12th February, 2004) over low tide (0.12 — 0.16 m, Table 1). At both the island and mainland locations, sites were selected according to the accessibility of habitat in the lower littoral zone and thus all sites are relatively gently sloping with some protection from the prevailing SW swell. All sites encompass areas of rock platform with rock pools and crevices. At each location, three sites were complex granite rock platforms with mixed granite, basalt and limestone boulder fields encompassing >10% of the total search area (Table 1). The remaining two sites at each location were predominantly rock platforms with <1% boulder habitat. These reefs were composed of granite, with the exception of one mainland site, Stenhouse Bay, which is a limestone reef (Table 1). Comparable limestone reefs do not occur on INTERTIDAL COMMUNITIES OF ALTHORPE I. AND MAINLAND, S.A. 147 YORKE PENINSULA { Gym Beach eenuneee INNES “Sennen, NATIONAL * PARK Marion Bay 5 Stenhouse Bay 4 Cape Spencer 5 ALTHORPE I. 4 INVESTIGATOR STRAIGHT NOT TO SCALE Fig. 1. Intertidal study sites on Althorpe Island and Innes National Park, Yorke Peninsula, South Australia. 148 Kk. BENKENDORFF TABLE |. Description of the intertidal sites surveyed for molluscs and echinoderms on: a) Althorpe I.; and b) Innes National Park, Yorke Peninsula, South Australia. a) Althorpe I. Site GPS Reading Tide Habitat 1) Western Channel S 35° 22.081" 0.15 Granite channel filled with basalt and some limestone boulders E 136° 51.333’ 2) Western Platform S 35° 22.046’ 0.15 Granite rock platform E 136° 51.352’ 3) The Fangs S 35° 22.195" 0.14 Granite rock platform with deep pools E 136° 51.999’ 4) East of Mooring Bay S 35° 22.165’ 0.14 Granite rock platform extending into a shallow basalt, granite E 136° 51.793’ and limestone mixed boulder field 5) West of Pareora S 35° 21.989’ 0,12 Granite platform with channels of granite, basalt and Monument E 136° 51.603’ limestone boulders b) Mainland Site GPS Reading Tide Description 1) Gym Beach S 35° 09.230" 0,12 Granite rock platform and boulder field E 136° 54.445’ 2) Marion Bay South —S 35° 14.978" 0.12 Granite platform with some deep pools E 136° 58.937’ 3) Cape Spencer SWS 35° 17.954" 0.15 Basalt and granite boulders extending into granite platform E 136° 52.811" S 35° 17.932" 0.15 E 136° 53.016’ S 35° 16.025’ E 136° 57.250' 4) Cape Spencer NE 5) Stenhouse Bay Mixed granite, limestone and basalt boulder-filled inlet with granite outcrop and platform Limestone platform Althorpe I., although they are common on the coast around Innes National Park; consequently Stenhouse Bay was included to provide a more comprehensive survey of the regional biodiversity. Using measuring tapes, a 20 x 20 m quadrat was defined as the search area, commencing from the lowest possible level of the shore at each site. A GPS reading was taken from the middle of the site (Table 1). One hour timed search surveys were then conducted at each site according to Benkendorff (2003), with the modification that the search was divided into 6 x 10 min time intervals to enable the construction of species accumulation curves. The surveys involved searching for and recording all molluscan and echinoderm species (>5 mm) found within the entire range of intertidal habitats present within the 20 x 20 m plot. Molluscs were identified according to Lamprell & Healy (1998), Lamprell & Whitehead (1992), Wilson (1993), Edgar (1997), Jansen (2000), Coleman (2003) and Rudman (2004). Voucher specimens were collected from species that could not be easily identified and some of these were later identified with the assistance of Peter Hunt (SA Malacological Society). Higher taxonomic groupings of the Mollusca have been assigned according to the classification outlined in Beesley et al. (1998). Echinoderms were identified according to Edgar (1997). Two measures of species rarity have been assessed in this study: 1) spatial rarity; and 2) numerical rarity. Spatial rarity is simply defined by the frequency of sites at which each species was recorded. Numerical rarity was assigned according to maximum local abundance of each species observed at each site in an hour. Species were assigned to one of the following categories: a) rare, <5; b) uncommon, 5 — 20; c) common, 21 — 100, and; d) abundant, >100 individuals. Frequency distributions for numerical rarity are calculated for the region by taking the maximum abundance for each species at any one site. Multivariate analyses were undertaken to investigate the molluscan communities using the Primer Software package Version 5 (Clarke & Gorely 2001). Using semi-quantitative abundance scores, the similarities between each pair of sites were analysed using the Bray-Curtis measure according to Smith (2004). Non-metric multi-dimensional scaling (nMDS) was performed on the similarity matrix and portrayed in two-dimensional plots using the grouping factors of: a) location (Althorpe I. vs. mainland) and b) habitat (rock platforms vs. platform INTERTIDAL COMMUNITIES OF ALTHORPE I. AND MAINLAND, S.A. 149 TABLE 2. Species list for a) Mollusca and b) Echinodermata, at ten intertidal rocky reef sites on Athorpe I. and in or near Innes National Park, Yorke Peninsula. * = egg masses were also observed. No. sites refers to the number of sites each species was recorded from within each location (max = 5). The relative rarity of each is defined according to the maximum number of individuals observed at any one site within the location: Rare (R) <5; Uncommon (U), 5-20; Common (C), 21 — 100; Abundant (A), >100 individuals. a) Family Species Althorpe Island Innes National Park No. sites Rarity No. sites Rarity Eogastropoda Patellidae Scutellastra chapmani Patella peronii Nacellidae Cellana tramoserica Cellana solida Lottidae Notoacmea petterdi Notoacmea mayi Patelloida cf. flammea Patelloida alticostata Patelloida latistrigata AMANNOMNBNN AMaAWaAnnNnN © Orthogastropoda Neritopsidae Nerita atramentosa Haliotidae Haliotis rubra Fisurellidae Scutus antipodes Clypidina rugosa Notomella candida * QACAAASF ANAANAPPPrACA B * Trochidae Austrocochlea rudis Austrocochlea adelaidae Austrochoclea odontis Notogibbula preissiana Clanculus brunneus Granata imbricata Talobis rotunda Canthrindella picturata Herpetoma pumilo ARRAYS! CAAA QPS SSF COACH SFPAYPLrSrIrA: ms Thalotia c.f. chlorostoma Talopena c.f. gloriola Turbinidae Turbo undulatus Turbo torquatus Micrastaea aurea Qn: ZFaQ'A Austroliotia botanica a Subninella gruneri Astralium aureum Cererthidae Bittium granarium Littorinidae Bembicium nanum Littorina acutispira Littorina unifasciata Nodilittorina praetermissa Rissoina crassa Rissoina fasciata Calyptraeidae Crepidula aculeate Epitoniidae Opalia australis Ranellidae Sassia sp. Cancellariidae Cancellaria c.f. granosa Buccinidae Cominella lineolata Cominella eburnea Unidentified sp. 1 Columbellidae Mitrella pulla FB OP OCOKF NON COWWHENWNNWN | HN * aN * ANF SSC: AAAAAQVPSPePCA! QO 1 wi KEEP ENON HKFNHKHNUUNUNWWHK OH HOW SCH OH KE UNSCANHK HK HY An! o) SOF ON OCHONUUN AACLY! 150 TABLE 2. Cont. K. BENKENDORFF a) Family Species Althorpe Island Innes National Park No. sites Rarity No. sites Rarity Muricidae Lepsiella vinosa 2 i, 2 € Dicathais orbita 5 Cc 5 Cc Lepsiella flindersii 1 R 1 R Marginellidae Austroginella sp. 1 R 0 - Conidae Conus anemone 2 R 3 R Mitridae Mitra badia 2 R 0 - Cerithiopsidae Euselia pileata 1 R 0 - Unidentifed Unidentified sp. 2 1 R 0 - Siphonartidae Siphonaria funiculata 5 C I U Siphonaria zelandica 2 U 2* € Siphonaria diemenensis aie A 4 A Siphonaria c.f. baconi 4 Cc 4 A Anapsidae Aplysia parvula ii R 1 R Notospidae Berthellina citrina 1 R 0 - Unidentified sp. 3 0 - 1 R Nudibranchia Austreolis ornata 1 R 0 Rostangia bassia 1* R 0 - Unidentified eggs 0 - 1* R Flabellina poenicia 0 - 1 R Polyplacophora Ischnochitonidae Ischnochiton elongatus 5 C 3 C Ischnochiton variegatus 5 C 2 U Ischnochiton australis 0 - 2 R Mopalidae Ischnochiton sp. 1 R 0 a Plaxiphora albida 3 U 4 Cc Bivalvia Galeommatidae Lasea australis 5 € 3 C Carditidae Cardita sp. 3 R 1 R Ostreidae Ostrea angasi 0 - 1 U Saccostrea glomerata 1 R 0 - Mytilidae Hormomya erosa 0 - 1 R Xenostrobus pulex 1 C + A Venerupis c.f. anomala 2, R 2; R Pteriidae Pinctada sp. 1 R 0 - Archidae Barbatia riculata 0 - 1 R Teredinidae Thraciopsis subrecta 1 R 0 - Unidentified Unidentified sp. 4 0 - 1 R Cephalopoda Spirulidae Spirula spirula 0 - 1 R b) Family Species Althorpe Island Innes National Park No. sites Rarity No. sites Rarity Asteroidea Asterinidae Patiriella calcar 2 R 2 U Asteriidae Coscinasterias muricata 1 R 0 - Goniasteridae Tosia australis 1 R 0 - Oreasteridae Nectria sp. 1 R 0 - Ophiuroidea Ophionereididae Ophionereis schayeri 3 R 2; R INTERTIDAL COMMUNITIES OF ALTHORPE I. AND MAINLAND, S.A. 151 TABLE 2. Cont. b) Family Species Althorpe Island Innes National Park No. sites Rarity No. sites Rarity Ophiactidae Ophiactis resiliens 1 R 0 = Echinoidea Temnopleuridae Holopneustes inflatus 1 R 0 7 Echinometridae Heliocidaris erythrogramma l R 0 7 No. Species 0 10 20 30 Time (mins) = Western Channel @- Western Platform ’- The Fangs - East of Mooring Bay @- West of Monument —O— Gym Beach ——*— Marion Bay —*— SW Cape Spencer -——e— NW Cape Spencer —*— Stenhouse Ba’ 40 50 60 Fig. 2. Species accumulation curves for molluscs recorded during one-hour search surveys at five intertidal sites on Althorpe Island (broken lines) and five intertidal reefs on the Yorke Peninsula mainland (Innes National Park, solid lines). with boulder fields). Significant differences in community structure were then explored using a two- way ANOSIM. SIMPER was undertaken to identify the species contributing to the differences in community structure. These analyses were repeated using a presence/absence transformation on the species matrix. Results A total of 82 species of molluscs and eight echinoderms was recorded during the 10 surveys on Yorke Peninsula and Althorpe I. (Table 2). Sixty-six of the molluscan species were observed on Althorpe I., including 23 species that were not found on the mainland (Table 2a). Fifty-nine species of molluscs were recorded in Innes National Park, with 15 not observed on Althorpe I. The egg masses of five gastropods were observed on Althorpe I., whereas the egg masses from four species were observed on the mainland (Table 2). Eight species of Echinodermata were recorded and all of these were found on Althorpe I., including six species that were not observed on the mainland (Table 2b). The maximum number of species recorded at any one site was 46 molluscs (Fig. 2) and three echinoderms in the Western Channel at Althorpe I. Nearly half this number of species was recorded at Stenhouse Bay in Innes National Park, with 25 molluscs (Fig. 2) and no echinoderms. Species 152 K. BENKENDORFF 30 25 @ Molluscs O Echinoderms 2) a 5) ® 2. oO a) 2 ® xe) Ee 3 Zz 1 2 3 4 5 6 7 8 9 10 Number of sites 45 40 35 no 30 a) .) oy b) f % Oo © 20 ce! E | wae 15 10 is) 0 T T 1 rare uncommon common abundant Fig. 3. The frequency distribution of rarity in molluscs and echinoderms recorded during ten intertidal surveys on Althorpe Island and at Innes National Park; a) spatial rarity as determined by the number of sites occupied by each species; b) numerical rarity, as determined by the maximum abundance at any site: Rare (R) < 5; Uncommon (U), 5-20; Common (C), 21-100; Abundant (A), >100 individuals. INTERTIDAL COMMUNITIES OF ALTHORPE I. AND MAINLAND, S.A. 153 Stress: 0.0 a Althorpes y_ Innes National Park A. Mixed boulders v_ Granite platform Limestone platform Fig. 4. Two-dimensional nMDS plot for the molluscan community data based on semi-quantitative abundance rankings from 10 intertidal sites. The plots show dissimilarities between sites grouped according to: a) location (Althorpe I. vs. mainland); and b) habitat (rock platforms without boulders vs. mixed boulders with some rock platform). richness accumulation curves show that new species were still being recorded in the last 10 minutes of the survey at most sites (Fig. 2). In particular, the species accumulation curves were still rising steeply for the two most species-rich sites, the western channel on Althorpe I and Gym Beach in Innes National Park. Overall, there was no apparent difference in cumulative species richness for sites on Althorpe I. when compared to the mainland (Fig. 2). However, the four sites which are relatively simple rock platforms have the lowest accumulated species richness compared to those sites with significant boulder habitat (Fig. 2, Table 1). The majority of species recorded in these surveys was found to be spatially and/or numerically rare (Fig. 3). Thirty four percent of molluscs and 75% of echinoderm species were only detected at a single site (Fig. 3). Furthermore, a maximum of only one or two individuals was recorded for 49% of mollusc and 87.5% of echinoderm species (Fig. 3b). No echinoderm species were recorded at more than half of the sites or were numerically common at any one site (Fig. 3a). By comparison, 10 species of molluscs were recorded at all sites and these species, along with an additional four were observed to be numerically abundant (Fig. 3a), with 100s of individuals at most or all sites. The nMDS ordination of the full data set using semi-quantitative abundance categories indicates that discrete molluscan assemblages occur on Althorpe I. compared to those on the mainland (Innes National Park, Fig. 4a). Furthermore, segregation of the sites can be observed according to the molluscan communities occurring on the different types of intertidal habitats (Fig. 4b, stress 0.07). The sites with mixed boulder fields are 154 kK. BENKENDORFF clustered reasonably tightly compared to the rock platforms. The limestone platform at Stenhouse Bay is somewhat separated from the granite reefs. Two- way ANOSIM reveals significant differences in the community composition between rock platforms and boulder reefs (Global R = 0.542, p = 0.03), but differences between the island and mainland locations are only significant at the 10% level (Global R = 0.514, p = 0.067). ANOSIM using presence/absence transformed data supports the above trends with significant differences according to habitat (R = 0.604, p = 0.03), but not location (R = 0.4, p = 0.133). SIMPER analyses using the abundance matrix indicate that the average dissimilarity between Althorpe I. and the mainland locations is 37.67% of the species composition. Cumulative contributions from 51 species are required to account for over 90% of the total dissimilarity between locations; 30 of these species have dissimilarity/standard deviation ratios above or close to 1, and are therefore considered reliable indicators. A subset of these species that contribute to >3% of the average dissimilarity are presented in Table 3a. SIMPER analyses on the presence/absence transformed data also revealed Siphonaria funiculata and Xenostrobus pulex to be reliable indicators, with the former more prevalent on Althorpe I. and the latter more prevalent on the mainland. Investigation of the dissimilarity between rock platform and mixed boulder communities using SIMPER on the abundance matrix revealed an average dissimilarity of 37.8%. Again, 51 species contribute to >90% of the total dissimilarity and 35 have dissimilarity/standard deviation ratios above or close to one. The subset of species contributing to over 3% of the dissimilarity is listed in Table 3b. In particular, Austrocochlea spp. appear to be useful indicators for the different habitats, with A. concamerata and A. odontis more abundant in boulder communities, whereas A. adelaidae is more prevalent on rock platforms. Presence/absence transformation supports use of A. concamerata as an indicator for molluscan communities in boulder habitats. Discussion A relatively high species richness of molluscs, but low richness of echinoderms can be found on rocky intertidal reefs around Althorpe I. and Innes National Park. This study has contributed an additional eight echinoderm species and 50 molluscan species to the preliminary study of the Althorpe intertidal communities conducted by Gaut'. Species lists are an important first step towards conservation and management (Oliver & Beattie 1993; Soule & Kohm 1989; Stork 1994), because they facilitate the identification of rare species and enable quantitative measures of biodiversity. However, comprehensive species lists are very difficult to obtain due to temporal variation in marine communities. So this study should only be regarded as a representative snapshot of the local fauna. It can be difficult to compare the results of marine biodiversity surveys between studies and across regions due to lack of standardization in the survey methods (Benkendorff 2003). Nevertheless, the mean species richness recorded per site for molluscs in this study (31.7+6.4) is comparable to previous timed search surveys on the NSW Coast (e.g. Gladstone 2002, mean = 31+8.9 from three-hour searches; Benkendorff 2003, mean = 38.9+12.2 from one-hour searches). The total number of molluscs (y diversity) recorded on Yorke Peninsula during this study (81 from 10 hours searching, Table 2a) also compares favorably to that recorded from intertidal reefs along the south east coast of Australia. For example, Gladstone (2002) reports a total of 80 molluscan species from a total of 90 search-hours at 15 locations along the NSW Coast, and Smith (2005) reports 170 molluscs from 25.8 total search hours on eight headlands around the Solitary Is National Park, NSW. In the same study by Smith (2005), 30 echinoderm species were recorded, indicating the relatively low diversity for this Phylum on intertidal reefs around Yorke Peninsula, S.A. (eight species, Table 2b). Across both island and mainland locations, sites with boulder habitat were found to have higher species richness than the simpler rock platforms (Fig. 2, Table 1). This finding is consistent with the Habitat Diversity Hypothesis (Connor & McCoy 1979), which predicts that a greater diversity of species will occur where a greater diversity of habitats exists. A boulder-filled channel on the NW side of Althorpe I. (Fig. 1) was found to be particularly rich in intertidal molluscs (Fig. 2). According to Benkendorff & Davis (2002), this site could be regarded as a regional “hotspot” of molluscan diversity by applying the definition of two standard deviations above the regional mean. No other sites were found to fit this “hotspot” definition, although Gym Beach in Innes National Park is close (Fig. 2). The continuously rising species accumulation curves at both these sites suggest that species richness may be underrepresented by the one-hour search surveys used. Benkendorff & Davis (2002) have demonstrated that it is significantly more difficult to obtain a comprehensive species list at species-rich intertidal hotspots. Furthermore, relatively few egg masses were recorded at any site during this study, most likely due to the timing of the surveys (late summer). However, both Gym Beach INTERTIDAL COMMUNITIES OF ALTHORPE I. AND MAINLAND, S.A. 155 TABLE 3. SEMPER results for the difference in molluscan communities between: a) Alihorpe I. and mainland (Innes NP) locations, and b) rock platforms and mixed boulder reefs. The Bray Curtis dissimilarity measure is based on a categorized rarity matrix with 999 permutations and several repeated computations. The average abundance is taken from semi- quantitative abundance rankings averaged across each group of sites for each species, such that 0 = not present; 1 = rare (<5 individuals); 2. = uncommon (5-20 individuals), 3 = Common (21-100 individuals); 4 = Abundant (>100 individuals). Only those species with a dissimilarity/standard deviation ratio close to and above one are listed as consistent indicators and only those contributing to greater than 3% of the total dissimilarity are presented here. a) Species Average Abundance Ratio % Cumulative Althorpe I. Innes NP Diss/SD contribution — Contribution % Siphonaria funiculata 3 > 0.4 2.86 4.66 4.66 Xenostrobus pulex 0.6 < 2.8 1.47 4.28 8.94 Turbo undulatus 1.2 < 2.8 1.4 3.12 12.05 Bembicium nanum 2 < 2.1 1.36 3.08 15.14 Plaxiphora albida 0.8 < oy 1.66 3.04 18.18 Austrocochlea odontis | < 1.6 0.98 3,02 21.20 b) Species Average Abundance Ratio % Cumulative Mixed boulders Rock platforms Diss/SD = -% contribution Contribution % Austrocochlea concamerata 3 > 0.25 3.59 4.9] 4.91 Austrocochlea adelaidae 0 < 2 0.97 3.51 8.42 Austrocochlea odontis 2 > 0.25 L.11 3.38 11.79 Bembicium nanum 2.5 > 1.25 1.43 3.35 15.14 Xenostrobus pulex 1.67 < 1.75 1.07 3.29 18.43 Notogibbula lehmanni 2.83 > I 1.52 3.22 21.65 and the Althorpe channel provide extensive boulder habitat that would be suitable for egg mass deposition by a wide range of gastropods (pers. obs., see also Benkendorff & Davis 2004). Consequently, additional surveys at these sites during late spring when more molluscs are breeding (pers. obs.) may further highlight the importance of these sites as habitat for intertidal molluscs. Species richness, community composition and rarity should all be taken into consideration for biodiversity conservation and management. Species turnover (B diversity) was high between sites in both locations, with 28 molluscs and six echinoderms only recorded at a single site (Fig. 3a). This suggests that any one hotspot will not fully represent all the spatially rare species observed in this study. However, a new marine conservation reserve around the Althorpe Is would encompass all the associated intertidal reefs surveyed here. All but 15 molluscan species that were recorded solely on the mainland in this study would be represented in such a reserve (Table 2). Overall, a higher species richness and more rare species were recorded on Althorpe I. when compared to the mainland (Table 2). Whilst this supports prioritization of Althorpe I. for protection in anew aquatic reserve, this location should still not be considered representative of the mainland. Some distinct clustering could be seen for the Island versus the mainland sites in the nMDS ordination plot (Fig. 4a) and ANOSIM indicated a marginally significant difference in the community composition between these two locations. A large number of species were found to contribute to the average dissimilarity between locations, with many species such as Xenostrobus pulex being more common on the mainland. However some, such as Siphonaria Juniculata, are more prevalent on Althorpe I. (Table 3a). The reasons for these differences in molluscan community structure between Althorpe I. and Innes National Park are currently unclear but are likely to be due to a combination of biotic and abiotic factors. In particular, distance from source populations and current patterns would influence the chance of populations establishing at the various sites. In addition to representation of the coastal and mainland locations, there is also a clear need to protect the different types of intertidal habitats in marine reserves. Distinct molluscan communities were found on rock platforms, compared to those sites with mixed boulder fields, as shown by nMDS ordination (Fig. 4b). Indeed, ANOSIM revealed that the difference between habitats was more significant 156 Kk. BENKENDORFF than that between the mainland and island locations. The sites are separated for habitat type along a different axis to that for location (Fig. 4), suggesting that different molluscan species are driving these two patterns, as supported by SIMPER analysis (Table 3). Again, a large number of species were found to contribute to the dissimilarity between habitat types, and whilst many of these species are specialized to live under boulders (data not shown), this does not apply to those species with the greatest percent contribution (Table 3b). Several Austrocochlea spp. were identified as possible indicators for the different communities, with A, adelaidae only recorded on rock platforms without boulders (Table 3b). Not all of the variation in molluscan communities can be explained by the two factors used in this study. Notably, there was a fairly high degree of separation between all sites in the nMDS ordination (Fig. 4). In particular, there is an apparent separation between the granite platforms compared to the limestone reef at Stenhouse Bay (Fig. 4b). However, the lack of replicate sites prevents any further conclusions being drawn about the effects of substratum type on intertidal molluscan community composition. At the mixed boulder sites, the total area covered by boulders and the proportion of different rock types may account for some more of the variation. Thus, the effects of substratum and other abiotic factors, such as wave exposure, are worthy of further investigation. The patterns in species richness and community composition observed in this study appear to be largely driven by rare species (Fig. 3). Rare species are usually regarded as those that persist with naturally low abundances and/or restricted geographic ranges (Gaston 1994; Chapman 1999). The species abundance distribution (numerical rarity) for intertidal molluscs and echinoderms in this study is skewed to the right (Fig. 3b), indicating that the majority of species are found in very low numbers and most of the individuals belong to a few abundant species. This pattern is consistent with previous studies on a wide range of communities (e.g. Fisher et al. 1943; MacArthur 1957). The frequency distribution for the spatial occurrence of intertidal molluscs in this study is also skewed right, but tending toward bimodality (Fig. 3a). This indicates that there is one group of molluscs whose presence is predictable, on most, if not all intertidal reefs in this S.A. bioregion, whilst most other species are highly unpredictable with patchy or localised distributions. Bimodal distributions for the number of locations that intertidal species are recorded have been previously observed at the regional scale (e.g. Gladstone 2002). Gaston (1994) suggests that these distributions could, in part, reflect sampling artifacts because the left hand mode is likely to be inflated due to the fact that species with low abundances are harder to find and therefore have a lower probability of being recorded at any site. In support of this, the majority of numerically rare species were only found at one or two sites in this study (Table 2). Species that are both numerically and spatially rare are often termed accidentals, vagrants, immigrants or incidentals (Gaston 1994) because they do not appear to maintain viable populations in the study area. Indeed, the interconnectedness of intertidal and subtidal habitats could create problems for assessing the rarity of intertidal species due to vertical migration. Nevertheless, from an_ ecological perspective, immigrants should be regarded as a part of the community because they could contribute to species interactions (Gaston 1994). The potential conservation importance of these rare species should not be overlooked until more is known about their source populations. In conclusion, this study demonstrates that the intertidal reefs around Althorpe I. and Innes National Park support a high species richness of molluscan fauna, as well as a high proportion of apparently rare molluscan and echinoderm species. Due to differences in the community composition, a comprehensive marine reserve system should include both island and mainland sites. Furthermore, representative sites need to encompass the full range of different intertidal habitats that occur in the region, at least including boulder fields and rock platforms. The importance of geology should be further investigated since only limestone reefs are currently protected on Yorke Peninsula mainland. Granite reefs occur on both Althorpe I. and in Innes National Park. These are often accompanied by mixed basalt, granite and limestone boulder fields providing high habitat complexity and greater species richness compared to the rock platforms. One such site on Althorpe I. was identified as a hotspot of species richness that could be prioritised for conservation. But ultimately, the large number of spatially rare species found in the region and the large variability in community composition between sites indicates the need to protect a suite of sites rather than any one location. Acknowledgements 1 am extremely grateful to Alex Gaut for her assistance in the field. This research would not have been possible without the support of R. K. Lewis, and the help of Sue Murray-Jones and Scoresby Shepherd for organizing this scientific expedition to Althorpe Is, including access to the island on the SARDI research vessel Ngerin. Thanks also to the Friends of the Althorpes for providing INTERTIDAL COMMUNITIES OF ALTHORPE I. AND MAINLAND, S.A. 157 accommodation on the Island and for their interest conserving the local environment. I am also grateful to the Rangers in Innes National Park for allowing access and collection permits for the intertidal reefs within the Park. The assistance of Peter Hunt in the identification of some smaller molluscan species is much appreciated. Constructive comments from two anonymous reviewers have improved the manuscript. References Auten, T. J. (1999) Building pathways for marine invertebrate conservation. pp 407-412 Jn Ponder, W. & Lunney, D. (Eds) “The other 99%, The conservation and biodiversity of invertebrates” Trans. R. Zool. Soc. NSW (Surrey Beatty & Sons, Chipping Norton). BEESLEY, P. L. Ross, G. J. B. and WELLS, A. (Eds) (1998) “Mollusca: the Southern Synthesis. Fauna of Australia Vol. 5” (CSIRO Publishing, Melbourne). BeNKENDOREF, K. (2003) Rapid biodiversity assessment on rocky intertidal reefs: An examination of molluscan species richness. Rec. S. Aust. Mus. Monographs Series 7, 67-78. & Davis, A. R. (2002) Identifying hotspots of molluscan species richness on rocky intertidal reefs. Biodiversity & Conservation 11, 1959-1973. (2004) Gastropod egg mass deposition on a temperate, wave exposed coastline in New South Wales Australia: Implications for intertidal conservation. Aquatic Conservation: Marine and Freshwater Ecosystems 14, 263-280. Bissy, F. A. (1995) Characterisation of biodiversity pp 21- 106 In Heywood, V. H. (Ed.) “Global Biodiversity Assessment” (United Nations Environment Programme, Cambridge University Press, Great Britain). CLARKE, K. R. & GorLey, R. N. (2001) “PRIMER v5: User Manual/Tutorial” (PRIMER-E Ltd, Plymouth, U.K.) Cuapman, M. G. (1999) Are there adequate data to assess how well theories of rarity apply to marine invertebrates? Biodiversity and Conservation 8, 1295-1318. Coteman, N. (2003) “1001 Nudibranchs. Catalogue of Indo-Pacific sea slugs” (Neville Coleman’s Underwater Geographic Pty Ltd. Springwood, QLD, Australia). Connor, E. F, & McCoy E. D. (1979) The statistics and biology of the species-area relationship. Am. Nat. 113, 791-833, Davipson, I. C., Crook, A. C. & BARNES, D. K. A. (2004) Quantifying spatial patterns of intertidal biodiversity: is movement important? Mar. Ecol, 25, 15-34. DEH (2003) Marine Conservation - Existing Marine Protected Areas in South Australia. South Australian Department for Environment and _— Heritage. http://www.denr.sa.gov.au/coasts/marine_existing_paps. html (accessed Oct., 2004). DEH (2004) Commonwealth Marine Protected Area Estate. Australian Government Department of Environment and Heritage. http://www.deh.gov.au/coasts/mpa/common wealth/manage/estate.html (accessed Oct., 2004) Epaar, G. J. (1997) “Australian marine life: the plants and animals of temperate waters.” (Reed Books, Victoria). EpyvaNng, K, (1996) The role of marine protected areas in temperate ecosystem management. pp. 53-67 Jn Thackway, R. (Ed.) “Developing Australia’s Representative Marine Protected Areas: Criteria and Guidelines for Identification and Selection” (Department of the Environment, Sport and Territories, Canberra). FISHER, R. A., CorBeT, A. S. & WILLIAMS, C. B. (1943) The relationship between the number of species and the number of individuals in a random sample of an animal population. J Animal Ecol. 12, 42-58. Gaston, K. J. (1994) “Rarity” (Chapman and Hall, Melbourne). GLADSTONE, W. (2002) The potential value of indicator groups in the selection of marine reserves. Biol. Cons. 104, 211-220. JANSEN, P. (2000) “Seashells of South-East Australia” (Capricornica Publications, Lindfield, Australia). LampreLL, K, & HEALY, J. (1998) “Bivalves of Australia Volume 2” (Backhuys Publishers, Leiden). & WHITEHEAD, T. (1992) “Bivalves of Australia Volume 1” (Crawford House Press, Bathurst). MacArtuur, R. H. (1957) On the relative abundance of bird species. Proc. Natl. Acad. Sci. USA 43, 373-387. Outver, I. & Beattie, A. J. (1996) Designing a cost- effective invertebrate survey: A test of methods for rapid assessment of biodiversity. Ecol. Applic. 6, 594- 607, Pressey, R. & MCNEILL, S. (1996). Some current ideas and applications in the selection of terrestrial protected areas: are there any lessons for the marine environment? pp. 125-133 In: Thackway, R. (Ed.) “Developing Australia’s Representative Marine Protected Areas: Criteria and Guidelines for Identification and Selection” (Department of the Environment, Sport and Territories, Canberra). RupmMaAN, W. (2004) Online Slug Forum. Australian Museum http://www.austmus. gov.au/science/division/ invert/mal/forum/index.htm (accessed Feb., 2004) Smit, S. D. A. (2005) Rapid assessment of invertebrate biodiversity on rocky shores: where there’s a whelk there’s a way. Biodiversity and Conservation In press. Souté, M. E. & Kon, K. A. (1989) “Research Priorities for Conservation Biology” (Island Press, Washington, D.C.). Stork, N. E. (1994) Inventories and biodiversity: More than a question of numbers. pp. 81-100 Jn Forey, P. L., Humphries, C. J. and Vane-Wright, R. I. (Eds) “Systematics and Conservation” (Oxford University Press, New York). Unperwoop, A. J. & CHAPMAN M. G. (1998) Spatial analyses of intertidal assemblages on sheltered rocky shores. Aust. J Ecol. 23, 138-157. Warwick, R. M. & Licurt, J. (2002) Death assemblages of molluscs on St Martin’s Flats, Isles of Scilly: A surrogate for regional biodiversity? Biodiv. Cons. 11, 99-112. Witson, B. (1993) “Australian Marine Shells” (Odyssey Publishing. Kallaroo, W. A.). ZANN, L. P. (2000) “The Marine Environment. Technical Annex |. State of the Marine Environment Report for Australia” (Department of the Environment, Sport and Territories, Canberra). Transactions of the Royal Society of S. Aust. (2004), 129(2), 158-169. PYCNOGONIDA FROM THE ALTHORPE ISLANDS, SOUTH AUSTRALIA by D. A. STAPLES* Summary StapLes, D. A. (2005). Pycnogonida of the Althorpe Islands, South Australia. Trans. R. Soc, S. Aust. 129(2), 158-169, 30 November, 2005. Six species of shallow-water Pycnogonida are recorded from the Althorpe Islands. Pseudopallene watsonae sp. nov. and Pseudopallene inflata sp. nov. are described. The adult male of Spasmopallene reflexa (Stock, 1968), described on the basis of a juvenile, is described and the species rediagnosed. The genera Spasmopallene Stock, 1968 and Pallenella Schimkewitsch, 1909 are synonymised with Pseudopallene Wilson, 1878. The holotype of Ps. ambigua Stock, 1956 has been re-examined and compared with the new species. Callipallene emaciata micracantha Stock, 1954 is raised to species status. Stlopallene cheilorhynchus Clark, 1963 and Achelia transfugoides Stock, 1973 are recorded. Key Worps: Pycnogonids; Althorpe Islands; Callipallenidae; Ammotheidae. Introduction The pycnogonid fauna of South Australia is well represented in museum and private collections principally as a result of research and monitoring programs conducted over many years. Most specimens have been collected in Spencer Gulf and Gulf St Vincent, with few records from offshore islands, and none from the Althorpe Islands. Six species referable to four genera and to two families are represented. This collection provides an opportunity to reconsider the generic status of Spasmopallene and Pallenella and describe new species. Specimens were collected using SCUBA at depths ranging from 2 to 27 m. Background information to the Althorpe Is and expedition details are given by Murray-Jones & Shepherd (2005). Specimens examined are lodged in the South Australian Museum (SAM) and Museum Victoria (NMV). CALLIPALLENIDAE Hilton, 1942 Pseudopallene Wilson, 1878 Pseudopallene Wilson, 1878: 202; Stock, 1954. Pallenella Schimkewitsch, 1909b: 1-13 (new synonomy). Spasmopallene Stock, 1968: 39-40 (new synonomy). Type species Pseudopallene circularis (Goodsir, 1842). Diagnosis Scapes l-segmented, chelae fingers smooth; proboscis conical distally, tip mamilliform; ovigers “Museum Victoria, GPO Box 666. Melbourne, Victoria 3001. Email: dstaples@museum.vic.gov.au 10-segmented, sexually dimorphic, terminal claw well developed; palps and auxiliary claws absent. All Australian species recorded to date share the same habitus as the holotype; body compact, segmented; anterior region of cephalon bulbous; neck short; proboscis directed ventrally, not visible in dorsal view. Scapes carried ventrally and angled outwards, chelae directed inwards such that the fingers lay directly below the mouth; cephalon and chelifores in combination, assuming a somewhat triangular aspect when viewed anteriorly. Juveniles and subadults differ from adults in the shape of the proboscis and chelifores, which are angled towards posterior; chelae geniculate, directed forward and upward; movable fingers distorted, digitiform; propodal heel more pronounced; coxa 2 short; oviger spine numbers low. Remarks The genus is common in southern Australian waters. Species determination is confused by a range of intermediate and overlapping characters, which no doubt contributed to Clark’s (1963: 33) observation that no two males (or females) of this group have been found to be exactly alike. To a large extent this variability can be attributed to ontogenetic changes previously noted by Stock (1973a: 116) and Staples (1997: 1053). In the present material, limited variation is also evident in adult specimens, particularly in the heel and oviger spine numbers. Stock (1968: 39) erected the genus Spasmopallene to accommodate specimens from the Great Australian Bight, which he described as S. reflexa and S. clarki Stock, 1968, at the same time embracing Pseudopallene dubia Clark, 1963. It is evident by association of larvae, juveniles and subadults with adult Pseudopallene and the sharing of a common substrate, that the genus PYCNOGONIDA FROM THE ALTHORPE ISLANDS, SOUTH AUSTRALIA 159 2.00MM Fig. 1. Pseudopallene ambigua Stock, 1956, holotype: A, cephalon ventro-anterior; B, ocular tubercle; C, oviger claw; D, trunk, lateral; E, third leg; F, trunk dorsal view; G, trunk, anterior view; H, propodus and tarsus, leg 3; I, chela. Spasmopallene is based on juvenile Pseudopallene. Sexual dimorphism of the ovigers is clearly evident in subadults, perhaps contributing to an assumption that the type specimens of Spasmopallene were sexually mature. Neither author mentioned the presence of genital pores in type material. The genus Pallenella was erected by Schimkewitsch (1909) to accommodate Pallene laevis Hoek, 1881, principally based on that species having 2-segmented scapes. In all other respects, diagnosis of the genus agrees with Pseudopallene. Hoek’s description of the scapes is, however, rather ambiguous and appears to have been misinterpreted by Schimkewitsch. His figures, which show the scapes to be clearly segmented (PI.X1 Figs 8 and 9), are inconsistent with his description of the scapes as being “constricted at the base, and indistinctly divided into two joints”. Examination of specimens in this collection, and from elsewhere in southern Australia, shows that a basal constriction of the scape is a variable character present in more than one species of Pseudopallene. In some instances the restriction is well defined and with a suggestion of a suture, but never segmented. It can be present in 160 D. A. STAPLES subadult and adult specimens and, in some cases, in specimens from the same collection. Pallenella has remained monotypic and retention of the genus is not supported by these observations. The genera Spasmopallene and Pallenella are synonymised with Pseudopallene. Key to adults of species of southern Australian Pseudopallene 1. a) Long segments of legs each with 2 distinct annular constrictions ............... P. pachychiera b) Leg segments without distinct constrictions .2 2. a) Propodal heel with spines arranged in 4 pairs P reflexa b) Propodal heel spines not paired...............0 3 3. a) Anterior margin of cephalon evenly rounded4 b Anterior margin of cephalon with mid-dorsal MOU sscazsascesepeceesnenyedees itp svednnenreesreanacoaseionaeee 5 4. a) Terminal oviger claw, smooth, crenulate or serrated on 1 margin only .........ceceeeeeeseeeeeee 5 b) Terminal claw serrated on both margins....... 6 5. a) Terminal oviger claw smooth or crenulate...... J cbbeaantn nested dobetaleg th abeiiaga renege eeetezssaaae P. ambigua b) Terminal oviger claw serrated............ P. laevis 6. a)Chela fingers about '/+ length of palm, immovable finger strongly curved, tip blunt... ohh ES asta Sek So de at RENE aE P. watsonae b) Chela fingers longer than '/2 length of palm, immoveable finger weakly curved or straight, pointed .......e cc eeeeeeeeeeeeeees P. inflata Pseudopallene ambigua Stock, 1956 Fig | A-I Pseudopallene ambigua Stock, 1956: 40-42, Fig. 5 a-i; Clark, 1963: 31-33, Fig. 16 A-F; -Stock, 1973a: 115-117, Fig. 7 c, g, e. Material examined Holotype, male. Bass Strait, Museum Hamburg, K17 680. Remarks The holotype of Pseudopallene ambigua has been re-examined for comparison with the new species. The following observations can be made but generally there is little to add to the original description. Stock remarked on the unusual shape of the ocular tubercle, which he illustrated with a stepped anterior margin (Stock, 1956: Fig. 5). No such unusual feature is now evident, suggesting perhaps, temporary handling damage to the tubercle. The scapes have a slight basal constriction and coxa 2 is more inflated distally than figured. Re- examination of the oviger terminal claw using high resolution digital imaging shows that the striations observed by Stock are in fact crenulations (Fig. 1, C). Pseudopallene ambigua shares much in common with P /aevis. Both species are of similar size (leg spans > 43 mm and about twice the size of the new species described herein) and share Bass Strait as their type locality. In both type specimens, only the inner margin of the oviger terminal claw is serrated or bears striations, whereas all specimens in this collection bear distinct serrations on both margins. I am grateful to Dr. H. Dastych of the Hamburg Museum who kindly re-examined the R ambigua holotype to confirm my earlier observations. Based on Hoek’s description, P /aevis can be distinguished by the more compact trunk and conical proboscis. A further point of difference is the curvature of the leg segments. Given that constriction of the scape segment appears to be a variable character, these differences are perhaps not significant, but serve to distinguish the species pending a detailed review of all available material. Additional figures of P ambigua are provided. Pseudopallene watsonae sp. nov. (Figs 2, 3 J-K) Pseudopallene ambigua.-Stock, 1973a: Fig. 7 a, b, d, f. (non P ambigua Stock, 1956), Material examined Holotype SAM E3414, male (ovigerous), Althorpe Is. 35° 22' S 136° 51’ E, Western Isles, S.A., on bryozoan Orthoscuticella cf ventricosa, 27 m, R. Lewis, 6 Jan 2004. Allotype NMV J53155, female, Port Phillip Bay Heads, Vic., bryozoan Scuticella sp., J. E. Watson, Mar 1980. Paratypes NMV J53163, 1 male (ovigerous), Port Phillip Bay Heads, Vic.,15 — 25 m, J.E. Watson, Mar 1980; NMV J53164, 1 female, Port Phillip Bay Heads, Vic., 16 m, D.A. Staples, Nov 1978. Description Holotype. Leg span 21.3 mm. Trunk smooth, segmented, separated from lateral processes by distinct cuticular line, lateral ecdysial lines well- defined, arthrodial membrane broad. Neck short, cephalon with low mid-dorsal mound anterior to ocular tubercle divided longitudinally by a cuticular band extending the length of the neck. The band is darkly pigmented, fading to base of tubercle. Lateral processes about as long as own maximum diameter, each with several tiny dorsodistal spines, each process separated by < '/; own basal diameter, separation decreasing between posterior processes, Abdomen slightly inclined, short, inflated, narrows PYCNOGONIDA FROM THE ALTHORPE ISLANDS, SOUTH AUSTRALIA 161 Yu) O.1 mm QS Fig. 2. Pseudopallene watsonae sp. nov., SAM E3414, male holotype: A, trunk, lateral view; B, trunk, dorsal view; C, trunk, anterior view; D, third leg; E, tarsus and propodus, third leg; F, oviger; G, oviger claw; H, chela, left , interior; I chela, left, exterior. 162 D. A. STAPLES abruptly to a cleft tip, reaching to about half length of 4th lateral process. Ocular tubercle a little wider than tall, bearing 2 prominent dorsal papillae, 4 large pigmented eyes of equal size, lateral sense organs not evident. Proboscis directed ventrally, mid-region gently curved inward, tapering to oral surface, conspicuous fringe of dense setae immediately preceding tip. Scape one-segmented, little shorter than length of proboscis, slight basal constriction. Chelae robust, fingers short, tips blunt, touching when closed; immoveable finger straight, less that '/s length of palm; moveable finger strongly curved and rounded, a little off-set distally from immovable finger, prominent lobe on each finger corresponds with indentation on inner surface of the opposing finger. Palm broad, outer surface strongly inflated, inner and outer surfaces covered in short setae with a group of longer setae on inner surface at base of immoveable finger. Oviger segment 5 longest, about 25% longer than segment 4, curved, distal apophysis prominent, strigilis absent, terminal claw, distally acute, partially folded over, margins serrated, distal serrations fine, spine formula 14:10:10:10, proximal spine smallest, distal-most spine placed dorso-laterally, spines relatively narrow and long, many partially folded distally, similar to the main claw, segments 7 — 10 with scattered surface spinules, bifurcate setae visible under high magnification. Measurements Oviger male (mm). Seg. 1, damaged; Seg. 2, 0.24; seg. 3, 0.27; seg. 4, 0.95; seg. 5, 1.27; seg. 6, 0.25; seg. 7, 0.42; seg. 8, 0.26; seg. 9, 0.25; seg. 10, 0.25, claw 0.17. Legs slender, scattered short spines on longer segments, most spines with basal process. Coxa 2 is 2.3 times as long as coxa 3, tibia 2 longest segment, femur and tibia | subequal, length of tibia 2 about 5.0 times maximum width; tarsus bearing one large and many smaller spines; propodus weakly curved, heel low, with 2 (3) large spines of slightly variable shape followed by two smaller spines in single row, of the larger spines, the distal-most is largest, proximal spine slightly curved, sole with median row of about twelve spines flanked by numerous smaller spines. Main claw of varying length, from !/2 — */s length of propodus, Genital pores small and appear to be on legs 3 and 4 only. Conical area on dorsal surface coxa 2 of all legs may indicate the presence of a gland. Measurements Holotype (mm). — Length of trunk (frontal margin of cephalic segment to tip of 4th lateral process), 3.40: width (across 2nd lateral process), 1.50; proboscis length (lateral), 0.77; Third leg; coxa I, 0.42: coxa 2, 0.97; coxa 3, 0.42; femur, 2.00; tibia 1, 2.0; tibia 2, 2.4; tarsus, 0.22; propodus, 0.75; claw, 0.67. Allotype Slightly smaller but otherwise in close agreement with male. One oviger lost, other, segment 10 damaged, partly missing. Segment 5 marginally longer than segment 4, segment 4 slightly swollen proximally. Lateral processes narrowly separated. Scape with strong proximal constriction. Femora swollen, each with about 10 eggs. Genital pores large, all legs. Variation The oviger spine formulae are variable, Port Phillip Bay male, (J53163) 13:10:10:10; female (J53264) 14:9:8:10. The oviger distal claw shape was constant. Scapes of female (J53164) strongly constricted at base. Remarks L have not examined Stock’s (1973a:115) material from Pearson I.; however, based on his description (Figs 7 a, b d, f) it is clear that he has attributed the juvenile specimens of PR watsonae to P ambigua. This new material, consisting of mature specimens, confirms the independent status of R watsonae. The most conspicuous difference between P watsonae and P ambigua is in the shape of the chelifore fingers. Pseudopallene ambigua is further distinguished by the overall size, being about twice that of P watsonae and the more slender legs of which tibia 2 is about 7.2 times as long as the maximum width. The mid-dorsal region of cephalon is evenly rounded with no suggestion of a mid-dorsal process. The proboscis is slightly constricted at about one-third and two-thirds its length before tapering distally. The curvature of oviger segment 5 is stronger and the terminal claw lacks well-defined teeth. This ovigerous holotype carried several larvae amongst the eggs. The larvae are hyaline, the 4 legs each terminated in a claw as long as the propodus and tarsus together, heel with single spine, propodal sole with 2 spines (Fig. 3, J). In contrast to the robust chela fingers of the adult, which are well suited to crushing bryozoan zooids, the larval chelae are delicate. The moveable finger is distorted and digitiform, terminating in a needle-like process which appears to connect with internal (glandular?) tissue (Fig. 3 K). The shape of the digitiform process suggests a specialized purpose, and it is possible that it may be used to manipulate the manubrium covering the frontal pore of each zooid facilitating insertion of the proboscis. Many of the bryozoan zooids were empty with the operculum of the frontal pore displaced to a vertical position. PYCNOGONIDA FROM THE ALTHORPE ISLANDS, SOUTH AUSTRALIA 163 Fig. 3. Pseudopallene inflata sp.nov., SAM E 3417, female allotype: A, trunk, anterior view; B, trunk, dorsal view; C, trunk, lateral; D, oviger; E, oviger claw; EF, oviger spines; G, chela; H leg 4; I. tarsus and propodus leg 4; Pseudopallene watsonae SAM E3414,larval form: J, tarsus and propodus; K, chela. 164 D. A. STAPLES Etymology: This species is named for Dr. Jeanette Watson in recognition of her outstanding contribution to Australian pycnogonid collections. Pseudopallene inflata sp. nov. Fig 3 Holotype SAM E3417, female, Althorpe I., 35° 22’ S; 136° 51' E; 300 m N-NW of The Monuments, on bryozoan Orthoscuticella cf. ventricosa, 23 m, R. Lewis, 6 Feb 2004. Paratypes SAM E3416, 1 female (subadult), Althorpe Is, Western Isles, S.A. on bryozoan Orthoscuticella cf ventricosa, 13 m, D. A. Staples and T. Laperousaz, 6/02/04; SAM E3415, 1 female, eastern Mooring Bay, Althorpe I. on bryozoan Scuticella, 3-5 m, S.A. Shepherd, 12/2/05. Female (holotype): Leg span 15.5mm. Trunk smooth, segmented, lateral ecdysial lines well-defined. Neck short, mid-dorsal mound on cephalon conical, divided by longitudinal band; dorsal and ventral surface of all trunk segments inflated, Lateral processes little longer than maximum width, separated by about half own basal diameter, each with several tiny dorso-distal spines. Abdomen short, rounded, inclined at about 45 degrees. Ocular tubercle taller than width at base, tapered toward tip, apical papillae prominent, frontal margin convex; 4 eyes, oval, pigmented; lateral sense organs not evident. Proboscis directed ventrally slightly constricted at about one- third and two-thirds length. Scape strongly constricted proximally, chela fingers pointed, cutting edges smooth, slightly irregular, immoveable finger little more than '/2 length of palm, few setae. Oviger segments 4 and 5 about equal length, segment 4 swollen proximally, spine formula segments 7 — 10, 13:9:9:8, terminal claw elongate, greater than half length segment 10, scoop-shaped, margins of distal two-thirds bluntly serrated, serrations larger proximally. Measurements Oviger holotype (mm). Seg. 1, 0.10; seg. 2, 0.25; seg. 3, 0.28; seg. 4, 0.55; seg. 5, 0.58; seg. 6, 0.17; seg. 7, 0.31; seg. 8, 0.25; seg. 9, 0.23; seg. 10, 0.21, claw 0.12. Legs; coxa 2 about 2 times length coxa 3, inflated distally, tibia 2 longest segment, femur slightly curved, inflated, surface with scattered spines, little longer than tibia 1, both tibiae spinous dorsally, individual spines mounted on small processes, propodal heel low, 3 - 4 heel spines, distal-most spine longest, longer than '/2 width of segment, sole armed with about 10 smaller spines. The remaining third leg is damaged. Genital pores large, all legs. Measurements Holotype (mm) — Length trunk (frontal margin of cephalic segment to tip of 4th lateral process, lateral), 1.91; width (across 2nd lateral process), 1.15 proboscis length (lateral), 0.98. 4th leg; coxa I, 0.3; coxa 2, 0.77; coxa 3, 0.37; femur, 2.09; tibia 1, 2.0; tibia 2, 2.18 tarsus, 0.12; propodus, 0.67; claw, 0.45. Variation The oviger spine formulae variable, distal claw shapes constant. The spine formulae adult female SAM E3415 was 12:9:910, SAM 3416 was 11:8:7:7 Remarks The presence of an acute swelling on the mid dorsal surface of the cephalon anterior to the ocular tubercle, together with inflated trunk segments most readily distinguish this species from its congeners. The longer heel spines, also, appear to be a useful diagnostic character. Etymology The specific name alludes to the rounded or inflated trunk segments. Pseudopallene reflexa (Stock, 1968) comb. nov. Fig 4, Fig 5 F-G Spasmopallene reflexa Stock, 1968: 40-42, Fig. 15 a-h. Type locality Galathea Stn. 571. Great Australian Bight. (38° 47' S, 142° 41’ E). Material examined SAM E3418, | male (ovigerous), 4 subadults, 3 juveniles; Althorpe Is, Western Isles, S.A. amongst bryozoan Orthoscuticella cf ventricosa, 13 m, D. A. Staples and T. Laperousaz, 6 Feb 2004. NMV J53160, 1 subadult female, Popes Eye, Port Phillip Bay, Vic., on bryozoan Orthoscuticella ventricosa, 3m, T. O’Hara, 10 Mar 2005. Description Male Leg span 20.0 mm. Trunk smooth, completely segmented, lateral ecdysial lines not evident. Lateral processes little longer than maximum width, first and second pair separated by about two-thirds own basal diameter, decreasing to about '/s diameter between segments 3 and 4. Mid-dorsal region of cephalon, anterior to ocular tubercle, rounded in PYCNOGONIDA FROM THE ALTHORPE ISLANDS, SOUTH AUSTRALIA 165 Fig. 4. Pseudopallene reflexa, SAM E3418, male: A, trunk, anterior view; B, trunk, lateral view; C, trunk, dorsal view: D, third leg; E, tarsus and propodus, third leg; F, oviger; G, oviger claw; H, chela, left interior; I, chela, left exterior. 166 D. A. STAPLES anterior view, smooth, no evidence of mound or longitudinal division. Neck short. Abdomen follows downward curvature of trunk, inflated, barrel- shaped, cleft at tip, little shorter than distal margin of 4" Jateral process. Ocular tubercle broader than tall, four well pigmented eyes, 2 low apical papillae; lateral sense organs not evident. Proboscis very slightly inflated at about two-thirds length, almost parallel sided, narrows sharply distally, dense fringe of oral setae, proboscis and basal arthrodial membrane extended. Scape with slight basal constriction, chelae fingers smooth, pointed, immoveable finger little longer than moveable finger and little less than '/2 length of palm; palm glabrous. Oviger segment 5 longest, strongly curved, distal apophysis prominent, acute; segments 4 and 5 with scattered, fine setae; bifurcate setae visible under high magnification on segments 4 — 10, spine formula 12:9:9:8, terminal claw of same form in adults and subadults, elongate, ladle-like, margin lined distally with about 20 well-defined teeth, increasing in length distally. Oviger measurements Male holotype (mm). Seg. 1, 0.25; seg. 2, 0.30; seg. 3, 0.40; seg. 4, 0.90; seg. 5, 1.53; seg. 6, 0.35; seg. 7, 0.42; seg. 8, 0.30; seg. 9, 0.27; seg. 10, 0.25, claw 0.18. Legs; coxa 2, 2.7 times as long as coxa 3, coxa 2 with low, conical swelling on dorsal surface at about 3/4 length of segment, surface of femur and tibiae with scattered spines, longer spines about '/s segment width but mostly shorter, dorsal spines particularly abundant, most surmounted on a basal process, longer segments uneven, with low swellings on surface; tibia 2 longest, 4.7 times as long as greatest width; femur little longer than tibia 1, low swelling ventrally at '/2 to */3 length, dorsal surface linear, dorsal surface of tibiae irregular, ventral surface linear with few spines; tarsus short, numerous scattered setae on dorsal surface, single larger distal spine; propodal heel pronounced, spine arrangement distinctive, typically with two basal median spines, distal-most spine slightly larger, followed by 4 pair of slightly smaller spines angled in a ‘V’ arrangement; sole with about 10 spines. Main claw little less than 2/3 length of propodus. Genital pores small, ventral surface coxa 2 legs 3 and 4. Auxiliary claws absent. Measurements (male mm). Length trunk (frontal margin of cephalic segment to tip of 4th lateral process), 3.55; width (across 2nd lateral process), 1.42; proboscis length (lateral), 1.37. Third leg; coxa 1, 0.37; coxa 2, 1.32; coxa 3, 0.50; femur, 2.00; tibia 1, 1.70; tibia 2, 2.37; tarsus, 0.22; propodus, 0.65; claw, 0.47. Remarks These specimens have taken up the orange-brown colour of the gut contents and match the Scuticellid bryozoan on which they were found. The juvenile specimen is in close agreement with the form described as Spasmopallene reflexa Stock, 1968 first described from the region. In particular, it agrees in the shape of the proboscis and distal oviger claw. Unfortunately, however, | have been unsuccessful in confirming the characteristic arrangement of heel spines on the holotype. Stock recorded 4 — 5 heel spines but it is possible that 4 of the spines observed by Stock obscured matching pairs and that the 5th spine represents a median basal spine. Based on Stock’s description, and the material before me, I have little reason to doubt that the same species is represented. The 2nd coxae of subadults and juveniles are short, 1.6 — 1.8 times longer that coxa 3, and the propodus is much more strongly curved with a more prominent heel. The gut follows the contours and irregular surface of the legs, narrowing and dilating intermittently giving it a globular appearance. The proboscis of one subadult female tapers from about !/2 its length, distal portion plug- like, expanding slightly before again narrowing to setose tip (Fig. 5 G). Pseudopallene reflexa is readily distinguished from other described species by the irregular surface of the longer leg segments, the distinctive paired arrangement of the propodal heel spines, the proboscis shape, the stronger curvature of the 5th oviger segment and the form of the terminal oviger claw. Distribution Bicheno, Tas.; Bass Strait; Port Phillip Bay, Vic. at depths 3 — 72 m. Stylopallene Clark, 1963 Stylopallene cheilorhynchus Clark, 1963 (Fig 5 A, B) Stvlopallene cheilorhynchus Clark, 1963: 36-38, Fig. 19 A-I.? Stock, 1973a: 117.?Stock, 1973b: 92. Type locality Port Arthur, Tasmania. Material examined SAM E3420, 2 subadults, 1 larval form, Althorpe Is, 300 m N-NW of The Monuments, S.A. 23 m, R. Lewis, 6 Feb 2004; SAM E3421, 5 immature specimens, 2 larval forms, Western Isles, 27 m, R. Lewis, 6 Feb 2004; SAM E3422, 7 males, 4 females, 6 subadults, | larval form, Western Isles, on bryozoan Amathia wilsoni, 13 m, D. Staples and T. Laperousaz, 6 Feb 2004; SAM E3423, 1 male, eastern Mooring Bay, Althorpe L, 3 — 5 m, S.A. PYCNOGONIDA FROM THE ALTHORPE ISLANDS, SOUTH AUSTRALIA 167 Fig. 5. Stylopallene cheilorhynchus SAM E3422: A., trunk, male, lateral: B, chela, juvenile. Callipallene micracantha SAM £3424, male, C, oviger spines; D, leg 3; SAM E3425, female, E, leg 3; P. reflexa, SAM E3418, F, propodus, juvenile leg 3; G, proboscis, sub adult. Shepherd, | 2/02/05; SAM E3429, | male ovigerous, Nora Creina, S.A., snorkel collection, A.I.M.S collecting Team, 20 Feb 1989; SAM E3438, 1 subadult, Cape Northumberland, S.A. in algae, 15 m, S.A. Shepherd, Feb 1977; SAM E3430, | male ovigerous, St Francis I., off SW point, S.A., rocks and sand 35 m, W. Zeidler, 29/01/1982; SAM E3431, 1 female, Pearson I., S.A., 30 — 33 m, (no additional data); SAM E3432, 9 larval forms and juveniles, off Cape Jaffa, approx 3 km WNW of Margaret Brock Reef Lighthouse, S.A., ridges crevices, overhangs, algae, few sand pockets, W. Zeidler and K.L. Gowlett-Holmes, 17 Feb 1989; SAM E3433, 2 females, Dutton Cove, NE Kangaroo [., S.A. in night time plankton tow, J. Ottaway, 19 Jan 1979; SAM E3428, | male ovigerous, Cape Thomas, between Godfrey I., S.A., 3 — 7 m, red algae, soft rock, reef, sand wreck, W. Zeidler and K. Gowlett-Holmes, 16 Feb 1989; SAM E3434, 1 female, Kingscote, Kangaroo I., S.A., 5 m, on hydroids, N. Coleman, 13 Mar 1978; SAM E3435, | Juv., Robe, S.A., | km off- shore, 15 m, D. Staples, 16 Apr 1976; SAM E3436, | male ovigerous, West I., Encounter Bay, in sortings, 12 m, J. Ottaway, 26 Nov 1978; SAM E3437. 1 male ovigerous, Hopkins |., SE Eyre Peninsula, S.A., 17 m, 30 Nov 1995. Remarks A compact species, easily confused with closely related but more elongate, S. longicauda Stock, 1973a, and shares similar body markings. The distal 'o of the longer segments of the legs is black, contrasting against the proximal off-white colouring giving a banded appearance to the specimens. There 168 D. A. STAPLES is little to add to Clark’s (1963) description. Clark described the femur and tibia 2 as subequal; in this material tibia 2 is consistently about 1.2 times longer than the femur. Female genital pores are prominent and swollen. Male genital pores are tiny and obscure in comparison. Additional figures showing the adult in lateral view and juvenile chela are provided. This species was present in vast numbers on colonies of the arborescent bryozoan Amathia wilsoni. The large proportion of egg-bearing adults and the presence of so many immature forms may indicate a summer breeding season. Distribution St Francis Isles, S.A. to Coffs Harbour, N.S.W. and Tasmania. Low-water to 90 m. Very common in SE Aust. Callipallene Flynn, 1929 Callipallene micracantha Stock, 1954 Fig 5 C-E Callipallene emaciata micracantha Stock, 1954: 44-46, figs 19, 20 a-b, — Clark, 1963: 21-23, Fig. 11 A-I. — Child, 1975: 10. Type locality Off Cape Everard, Victoria (37° 05’ S, 150° 05’ E). Material examined SAM _ E3424, 1 ovigerous male, Althorpe Is., Western Isles, S.A., 27 m, R. Lewis, 6 Feb 2004; SAM E3425, 2 males, | female, Backstairs Passage, J. E. Watson, (undated); SAM E3439, 1 female, Cape Northumberland, S.A. in algae, 15 m, S.A. Shepherd, 1 March 1975; SAM E3440, | female, Gulf St Vincent, S.A. on Posidonia sinuosa, S.A. Shepherd, May 1985; SAM E3441, 1 female, Wallaroo, Spencer Gulf, S.A., J. E. Watson, Jan 1983. Remarks These specimens substantially agree with Stock’s (1954) description of the male holotype. Stock erected the subspecies micracantha to accommodate a single male specimen from eastern Bass Strait, which differed from other species and subspecies of the genus in the spination of the lateral processes and crop. His specimen was also notable in the presence of a double row of denticles on the immoveable finger, a character shared with C. cuspidata Stock, 1954 and Neopallene antipoda Stock, 1954. The double row of denticles is not present in this male specimen; however examination of additional material from the Victorian coastline and Bass Strait indicates this to be a variable character in adults of both sexes. The ventral spine-tipped swellings of the femur and to a lesser extent tibia 1, are present in all southern Australian specimens but more pronounced in the female, particularly egg-bearing specimens (Fig. 5, E). The crenulate propodal spines and serrated auxiliary claws characteristic of C. emaciata (Child 1979:41) are not present in the material examined. In the shape of the femur and tibia | the Australian material agrees with C. emaciata subspec. (Stock, 1954:46) from New Zealand but significant differences in the length of the neck and shape of the proboscis distinguish the species. Stock (1954) was reluctant to give his specimen species status on the basis of his solitary male, but given the wide range of material now available and the consistency in significant morphological characters, specific rank is well justified. The oviger spines are dimorphic. The distal-most spine on oviger segments 7 — 9 is off-set and larger than preceding spines. The teeth are irregular, well defined and larger on the distal margin. Dimorphism is not unique in the genus but in general shape, the distal spine morphology is close to that found in C. panamensis Child, 1979, C. brevirostris novaezealandiae (Thomson, 1884) and C. phantoma (Dohrn, 1881). Spines preceding the distal spine are uniform and finely denticulate. Spines at the base of the chelifore implants are variable in number, 3-4 most common, often with about 3 additional spines on the lateral margin. Female genital pores are larger than those of the male and present on all legs. Distribution Pearson I., S.A. to Batemans Bay, NSW and Tasmania 10 — 135 m AMMOTHEIDAE Dohrn, 1881 Achelia Hodge, 1864 Achelia transfugoides Stock, 1973 Achelia transfugoides Stock, 1973a: 104-106, Fig. 2. — Stock, 1973b: 92. Type locality Toad Head, West I., South Australia. Material examined SAM E3426, 1 female, Althorpe Is, Western Isles, S.A., in sortings, 13 m, D. Staples and T. Laperousaz, 6 Feb 2004: SAM E3427, | male, eastern Mooring Bay, Althorpe I., S.A. on bryozoan Amathia sp, 3 —5 m, S.A. Shepherd, 12/02/05; SAM E3442, 1 subadult, 7 females, 5 males ovigerous, Fanny Point, Boston I., Spencer Gulf, S.A. on Sargassum, reef, 2- 8 m, W. Zeidler, K.L. Gowlett-Holmes, 17 Feb 1988, Remarks | have tentatively assigned this material to A. transfugoides, the principal difference being in the PYCNOGONIDA FROM THE ALTHORPE ISLANDS, SOUTH AUSTRALIA 169 shape of the proboscis. Stock (1973a:104) described the proboscis with a swollen basal part which is consistent with his Fig. 2 c showing maximum width at the base. This figure is however, inconsistent with his figures b, e, in which he shows the basal part narrower than the inflated mid- region and with which the present material agrees. The tubiform part of the proboscis is a little shorter than illustrated by Stock (Fig. e), and in some other material I have examined, but otherwise closely agrees with the type material. Distribution Perth, WA to Wilsons Promontory, Vic. Acknowledgements I am grateful to the scientific party and crew of the RV Ngerin for their assistance and support. In particular, | thank Sue Murray-Jones for facilitating my participation, and fellow divers and collectors, Thierry Laperousaz, Scoresby Shepherd, Rob Lewis, and James Brook. Special thanks to H. Dastych of the Zoological Museum Hamburg for lending the holotype material, to Robin Wilson and Angelika Brandt for transporting the specimens, and to Phil Bock for identifying bryozoans. Finally I acknowledge the helpful criticism of G.C.B.Poore, and of the two reviewers, Franz Krapp and Roger Bamber. References Cuitp, C. A. (1975) Pycnogonida of Western Australia. Smithson. Contr. Zool. 19, 1-29. (1979) Shallow-water Pycnogonida of the Isthmus of Panama and the coasts of Middle America. Smithson. Contr. Zool. 293, 1-86. Crark, W. C. (1963) Australian Pycnogonida. Rec. Austr: Mus. 26 (1), 1-81. Donen, A. (1881) Die Pantopoden des Golfes von Neapel und der angrenzenden Meeresabschnitte. Fauna Flora Golfe. Neapel 3, 1-252. Hoek, P. C. C. (1881) Report on the Pycnogonida dredged by HMS Challenger 1873-76. Rep. Scien Results Explor Voyage Challenger 3 (10), 1-167. Murray-JONnes, S. & SHEPHERD, S. A. (2005) Introductory narrative and conservation recommendations. Trans. R. Soc. S. Aust. 129, 85-89. SCHIMKEWITSCH, W. (1909) Nochmals tiber die Periodizitit in dem System der Pantopoden. Zool. Anz. 34 (1), 1-13. StapLes, D. A. (1997) Pycnogonids. pp.1040-1072 In Shepherd, S. A. & Davies, M. (Eds) “Marine Invertebrates of Southern Australia Part 111’. (SARDI and Fauna and Flora of SA Handbooks Committee, Adelaide). Stock, J. H. (1954) Pycnogonida from Indo-West Pacific, Australian and New Zealand waters. Vidensk Meddr dansk naturh. Foren. 116, 1-168. (1956) Pantopoden aus dem Zoologischen Museum Hamburg. Mitt. Hamb. zool. Mus. Inst. 54, 33- 48. (1968) Pycnogonida collected by the Galathea and Anton Bruun in the Indian and Pacific Oceans. Vidensk. Meddr dansk naturh. Foren. (131), 7-65. (1973a) Pycnogonida from south-eastern Australia. Beaufortia 20, (266), 99-127. (1973b) Achelia shepherdi n sp and other Pycnogonida from Australia. Beaufortia. 21, (279), 91- THomson, G. M. (1884) On the New Zealand Pycnogonida, with descriptions of new species. Trans. Proc N. Z. Inst. 16, 242-248. WILson, E. B. (1878) Descriptions of two new genera of Pycnogonida. Am. J. Sci. 15, 200-203. Transactions of the Royal Society of S. Aust. (2005), 129(2), 170-182. A NEW SPECIES OF NEOPELTOPSIS (COPEPODA, HARPACTICOIDA, PELTIDITIDAE) FROM ALTHORPE ISLAND, SOUTH AUSTRALIA by G. K. WALKER-SMITH! Summary WALKER-SMITH, G. K. (2005) A new species of Neopeltopsis (Copepoda, Harpacticoida, Peltidiidae) from Althorpe Island, South Australia. Trans. R. Soc. S. Aust. 129(2), 170-182, 30 November, 2005. A new species of Neopeltopsis Hicks, 1976 is described from algae collected from Althorpe Island, South Australia. The new species is separated from its congeners by a number of morphological characters related to the A2, mandible, P1, P2, PS and the male Al. A revised generic diagnosis for Neopel/topsis is provided and intra- and intergeneric relationships are discussed. The occurrence of other species of Peltidiidae, collected from algal washings from Althorpe I., is documented. Neopeltopsis appears to be restricted to the Southern Hemisphere, with previous records limited to New Zealand (N. pectinipes Hicks, 1976) and Argentina (N. hicksi Pallares, 1979). Key Worns: Algae, Harpacticoida, Neopeltopsis, new species, Peltidiidae, South Australia. Introduction Harpacticoid copepods are microcrustaceans, usually less than 1 mm in length, that occur in marine, estuarine, and freshwater habitats. There are approximately 50 families and 460 genera of Harpacticoida and estimates of the total number of described species range between 3000 (Huys et al. 1996) and 4000 to 4500 (Giere 1993). The Australian harpacticoid fauna is abundant and diverse, but remains largely undescribed. One hundred and thirty-three species of Harpacticoida have been recorded in Australia, with major contributions being made by Nicholls (1941, 1942, 1945a-c, 1957), Hamond (1971, 1973a-e, 1974, 1987), Harris (1994, 2002), Harris & Robertson (1994), Bartsch (1993, 1994, 1995, 1999) and Karanovic (2004). Thirty-one species have been recorded from South Australia, including 11 Australian endemics described by Nicholls (1941, 1942, 1945a) and Hamond (1971, 1973c, 1973e). In February 2004, algal samples were gathered from Althorpe L, South Australia (SA) and harpacticoids living among the thalli were extracted. Among these harpacticoids, a new species of Peltidiidae Sars, 1904 belonging to the genus Neopeltopsis Hicks, 1976 was _ discovered. Worldwide, the family Peltidiidae consists of eight genera and 58 species. Species in this family are typically algal-dwellers, with dorsoventrally flattened bodies and modified appendages that ' Marine Invertebrate Section, South Australian Museum, North Terrace, Adelaide, South Australia 5000, Australia, Present address: School of Zoology, University of Tasmania, Private Bag 5, Hobart Tasmania 7001 and The Tasmanian Museum and Art Gallery, GPO Box 1164, Hobart Tasmania 7001. Email: geneforw@postoffice.utas.edu.au enable them to exist on the surface of the algal thalli (Hicks 1986). Prior to the present study, six species of Peltidiidae were known to occur in SA: Alteutha depressa (Baird, 1837); Alteutha spinicauda Nicholls, 1941; Parapeltidium cristatum Nicholls, 1941; Parapeltidium dubium Nicholls, 1941; Peltidium proximum Nicholls, 1941 and Peltidium simplex Nicholls, 1941. Peltidium speciosum Thompson & A. Scott, 1903, was recorded in South Australia by Nicholls (1941); however, the validity of this species is doubtful; Wells & Rao (1987) believed it could not be distinguished from several other species of Peltidium Philippi, 1839. The only other species of Peltidiidae recorded from Australia is Alteuthellopsis corallina Humes, 1981, which was found associated with scleractinian corals in Queensland (Humes 1991). This paper provides the description of a new species of Neopeltopsis, collected from Althorpe I., as well as a revised diagnosis of the genus. Neopeltopsis currently comprises only two species: N. pectinipes Hicks, 1976, recorded from New Zealand and N. hicksi Pallares, 1979 from waters off Argentina. The new species, described here, is separated from its congeners by morphological characters related to the A2, mandible, Pl, P2, P5 and the male Al. This paper also documents the occurrence of other species of Peltidiidae found in algae collected from Althorpe I., SA. Material and Methods Algae were collected by hand. Algal samples were washed in a bucket of freshwater and this water was then poured through a 63 4m mesh sieve. Retained material was fixed in 95% ethanol and _ later transferred to 70% ethanol. Samples were examined NEW SPECIES OF NEOPELTOPSIS 171 under a Wild MB8__ stereomicroscope and harpacticoids were extracted using fine forceps. Harpacticoids were dissected in a drop of glycerol on a microslide using electrolytically-sharpened tungsten needles. Appendages were transferred to new microslides, mounted in Gurr’s Aquamount and coverslips were sealed with clear nail varnish. Microslides were examined using either a Leitz Dialux 22 compound microscope with phase contrast or a Leica DMR compound microscope with interference contrast. Illustrations were made with the aid of a camera lucida. Selected specimens were examined using a Philips XL20 scanning electron microscope (SEM). These specimens were dehydrated in a graded ethanol series, critical point dried using CO, and gold coated prior to examination under the SEM (K V=10, spot size 3). Terminology used follows Huys & Boxshall (1991). Abbreviations used are: Al, antennules or first antennae; ae, aesthetasc; A2, antennae or second antennae; Md, mandible; Mxl, maxillule; Mx, maxilla; Mxp, maxilliped; P1 — P4, swimming legs 1 — 4. Individual segments of Pl — P4 rami are written (for example) as Pl exopod-3, for the third (or terminal) segment of the Pl exopod. P5 and P6 refers to the fifth and sixth leg respectively. Total length measurements are from the tip of the rostrum to the posterior margin of the caudal rami (excluding caudal setae). Armature formulae for swimming legs are constructed following Lang’s (1934) method (also see Huys & Boxshall 1991: 29), The term “armature” refers collectively to articulating elements such as setae and spines. All material examined is held in the collections of the South Australian Museum (SAM) and the Tasmanian Museum and Art Gallery (TMAG). Neopeltopsis Hicks, 1976 Neopeltopsis Hicks 1976: 363-370.—Hicks 1986: 356, 360-361. Diagnosis Body distinctive, broad, dorsoventrally flattened with simple pattern of chitinous thickening. P1- bearing somite incorporated into large cephalosome; P2 — PS somites free, epimeral plates well- developed; remaining abdominal somites fused, much shorter than prosome. Anal somite and caudal rami free. Urosome-caudal rami complex analogous to that of Porcellidium (Porcellidiidae Sars, 1904) (Hicks 1986). Rostrum broad, prominent, not defined at base; 2 Al 8-segmented, d Al 8 — 9 segmented. A2 exopod reduced, 1- or 2-segmented, with 2 or 4 setae in total; basis without, and endopod- 1 with, abexopodal seta; endopod-2 with 1 lateral and | distal large, pectinate (comb-like) seta. Mandibular gnathobase elongate, palp with 1- segmented exopod and endopod. Maxillule arthrite (i.e. praecoxal endite) with 6 — 8 spinose spines on distal margin and 2 setae on anterior surface; coxal endite with 2 — 3 setae; basis elongate; exopod distinct, 1-segmented, with 2 — 3 setae; endopod incorporated into basis and represented by 2 — 3 setae. Maxilla with 3 syncoxal endites, proximal most widely separated, with 4, 2 and 3 setae respectively; endopod fused to basis (forming allobasis), endopod represented by 2 setae, allobasis with | spinose and 2 naked setae distally. Maxilliped subchelate; pedestal well-developed; syncoxa armed with | — 2 setae, basis expanded, ovoid with pad-like distal seta; endopod drawn out into elongate, narrow claw, sometimes with small setae on lateral surface of claw. Pl coxa and basis elongate, orientated at right angles; exopod 3-segmented, exopod-3 reduced, indistinctly separated from exopod-2, bearing 4 broadly flattened, pectinate setae, without accessory armature, geniculate seta absent; endopod 2-segmented, reduced, much shorter than exopod. P2 — P4 rami 3-segmented. Armature formulae for both sexes as follows: Exopod Endopod p2 0.1.22 (1 —2) 0—1.0— 1.1 — 2) 20 P3 0.1.322 1.1.220 P4 0.1.322 1.1.220 P5 both sexes 2-segmented or sometimes indistinctly 2-segmented. Genital double somite and distal somites fused, expanded posterolaterally and almost surrounding the caudal rami. 2 genital apparatus comprising paired genital apertures located ventrally on urosome, apertures covered by P6; copulatory pore located on ventral midline (slightly) posterior to genital apertures, covered by operculum. Eggs in single egg-sac. Caudal rami short and subrectangular with 7 setae, principal terminal setae distinct (i.e. not fused to one another at base). 5 P6 left and right identical, large, lobe shaped; not known for N. pectinipes or N. hicksi. Male with 1(?)—2 spermatophores (not known for N. hicksi). Sexual dimorphism in body size, Al, PI and PS. Species Neopeltopsis pectinipes Hicks, 1976; N. hicksi Pallares, 1979; Neopeltopsis sp. nov. described herein. Distribution Wellington, New Zealand (Hicks 1976); Argentina (Atlantic Ocean and Tierra del Fuego) (Pallares 1979); Althorpe I., SA. 172 G. K. WALKER-SMITH Fig. 1. Neopeltopsis althorpensis sp. nov.: A, 2 habitus, dorsal (holotype, SAM C6219). B, d habitus, dorsal (paratype, SAM C6220). Terminal caudal setae cut short. CO = copulatory operculum, CP = copulatory pore, SR = seminal receptacle, P6 = sixth leg, S = spermatophore. Habitat Phytal. Recorded from: Pterocladia lucida, P. pinnata and Caulerpa brownii in New Zealand; Macrocystis (Phaeophyta) and species in the family Delesseriaceae (Rhodophyta) off the coast of Argentina; Pterocladia sp. and Lobospira bicuspidata in SA. Remarks Herein the generic diagnosis of Hicks (1986) has been expanded to better define the genus. Neopeltopsis althorpensis sp. nov. (Figs 1 — 8) Material examined Holotype. Althorpe I., SA (35° 22.02’ S, 136° 51.08’ E), from washings of Pterocladia sp. (Rhodophyta), depth ~2 m, coll. A. J. Hirst, 01 Feb. 2004, SAM C6219 (ovigerous & , dissected, mounted on 9 slides). Paratypes Collected with holotype. SAM C6220 (1 6 in ethanol, allotype); SAM C6221 (1 6, dissected, mounted on 3 slides. P3 and P4 lost); TMAG G5474 (1 6, partially dissected, in ethanol; P2 mounted on 1 slide); TMAG G5475 (1 2, mounted on | slide); SAM (C6222 (2 ovigerous 2°, 4 dd, | juv., in ethanol); TMAG G5476 (2 ovigerous ° 2, | non- ovigerous 2, 3 od, 5 juv., in ethanol). Other material Collected with holotype. SAM C6223 (1 2 mounted on SEM stub), SAM C6224 (2 36, mounted on SEM stub). NEW SPECIES OF NEOPELTOPSIS 173 SS a a a SS a= KS “<< Coo Fig. 2. Neopeltopsis althorpensis sp. nov.: A, 2 A1, dorsal (holotype, SAM C6219). B, ¢ Al, dorsal (paratype, SAM C6220); C, 3 Al, ventral (paratype, SAM C6220). 174 G. K. WALKER-SMITH Ss EF D {0.05 mm 0.05 mm Fig. 3. Neopeltopsis althorpensis sp. nov., 2° holotype, SAM C6219: A, A2; B, Md; C, Mxl; D, Mx; E, Mxp, anterior; F, Mxp, posterior (basis and endopod only). NEW SPECIES OF NEOPELTOPSIS 175 Diagnosis Al & with curved, thorn-like projections on segment 7. A2 exopod |-segmented, with 4 setae. Mandible exopod length approximately equal to width. Pl exopod-2 0.25 size of exopod-l. P2 endopod-2 without inner seta. P2 exopod-3 with only 1 outer spine. P5 exopod partially fused to baseoendopod in both sexes. Description of female Mean total body length 0.86 mm + 0.09 mm (n = 3). Body dorso-ventrally flattened, simple pattern of chitinous thickening (Fig. 1A), integument with numerous pores and sensillae (as found in male, see Fig. 7B), Caudal rami short and subrectangular, with 7 setae. Rostrum broad and prominent, fused at base (Fig. 1A). Caudal rami (Fig. 6C) length ~ width; inner distal margin finely serrate; with 7 setae, seta I shortest, seta IT dorsal to seta I, seta V well-developed and pinnate, seta VII dorsal and triarticulate at base. Al 8-segmented (Fig. 2A); all seta bare. Segment | with | seta and few fine setules; segment 2 with 11 setae and some short setules; segment 4 with aesthetasc fused basally to 1 seta; segment 8 with a shorter aesthetasc fused basally to 2 setae (i.e. a acrothek), Armature formula for Al: 1-[1], 2-[11], 3- [8], 4-[3+ (I+ae)], 5-[2], 6-[2], 7-[2], 8-[3+acrothek] A2 (Fig. 3A) basis and endopod-1 separate; basis with few fine setules and unarmed; endopod-1 with | abexopodal seta; endopod-2 with 2 surface frills distally; lateral armature consisting of | pectinate spine, | spine and 2 setae basally fused; distally with I pectinate spine and 4 geniculate setae (innermost basally fused to naked seta); exopod 1-segmented, with 4 bare setae. Labrum and paragnaths not illustrated. Mandible (Fig. 3B) coxa narrow and elongate, expanding distally to small gnathobase; basis with | seta; exopod |-segmented, small (about as long as wide), with 3 terminal setae; endopod with 1 short and 3 long setae distally and 1 proximal seta. All setae naked. Maxillule (Fig. 3C) arthrite of praecoxa with 8 pinnate and 2 smooth setae; no setae observed on medial surface; coxal endite with 2 bare setae; basis elongate, with 5 smooth setae, | geniculate seta and I serrate spine; exopod |-segmented, with 2 smooth setae; endopod completely incorporated into basal segment, but represented by 2 setae. Maxilla (Fig. 3D) syncoxa with 3 endites; proximal endite with 4 setae; middle and distal endites with 2 spinulose setae and 3 bare setae respectively; allobasis bearing | pinnate spine and 2 naked setae; endopod completely incorporated and represented by 2 setae. Maxilliped (Figs 3E — F, 8A) subchelate; pedestal well-developed; syncoxa smallest segment bearing 2 plumose setae; basis ovoid with a short, stout, pad- like seta with tiny spinules (arrowed in Fig. 8A); with few spinules proximally along the palmar margin; endopod drawn out into recurved claw, as long as basis, 2 small setae on lateral surface of the claw. Pl (Figs 4A, 8C). Coxa without armature. Basis, inner seta inserted above endopod; with few spinules around insertion of outer basal seta and | tube pore on outer anterior surface and patch of tiny spinules as figured. Exopod 3-segmented; exopod-3 small and indistinctly separated from exopod-2; exopod-1 with smooth outer seta and patch of tiny denticles as figured; exopod-2 with bare outer seta dorsolaterally and inner seta strongly serrate and posteriorly displaced, with a patch of tiny denticles on the dorsal surface (arrowed in Fig. 8C); exopod-3 with 4 strongly pectinate flattened setae. Endopod reduced, 2-segmented; endopod-1 ~ 4 x length of endopod-2, without setae; endopod-2 tiny, with 3 bare setae. P2 — P4 (Figs 4B, SA — B) rami 3-segmented. Armature formulae as follows: Exopod Endopod P2 0.1.221 1.0.220 P3 0.1.322 1.1.220 P4 0.1.322 1.1.220 P2 (Fig. 4B). Praecoxa and coxa not illustrated, coxa with anterior pore. Basis elongate with bare outer seta. Exopod-1! with fine setae on inner margin and spinulose spine on laterodistal corner; exopod-2 with plumose inner seta and pinnate outer spine; exopod-3 shorter than 2 preceding segments, with 2 inner, plumose setae, 2 distal setae with spinules and setules and | pinnate outer spine. Endopod-1 with | inner seta and fine setules along outer margin; endopod-2 without setae; endopod-3 with 2 inner setae and 2 distal setae. P3 (Fig. SA). Praecoxa and coxa not illustrated. Basis transversely elongate, with bare outer seta. Exopod-1 with patch of fine spinules around distal outer corner, spinulose spine, no inner seta; exopod- 2 with plumose inner and pinnate outer seta, with | anterior and 2 posterior spinule rows and a pore in the distal margin of the segment (arrowed in Fig. 5A); exopod-3 with 3 inner plumose setae (2 proximal most with additional pinnules), 2 distal setae with pinnules and setules, 2 pinnate outer spines, 2 spinule rows posteriorly. Endopod-1 with | plumose inner seta; endopod-2 with inner seta (broken off in holotype, present in paratype, position stippled in Fig. 5A); endopod-3 with 2 inner setae and 2 distal setae. P4 (Fig. 5B). Praecoxa and coxa not illustrated. G. K. WALKER-SMITH 176 pee LILES se FP LAP Fig. 4. Neopeltopsis althorpensis sp. nov., 2 holotype, SAM C6219: A, PI anterior; B, P2 anterior. 177 S OF NEOPELTOPSIS NEW SPECIE ma Fig. 5. Neopeltopsis althorpensis sp. nov., 2 holotype, SAM C6219: A, P3 anterior; B, P4 posterior. Arrows indicate pore. 178 G. K. WALKER-SMITH yy < Chee SS St Fig. 6. Neopeltopsis althorpensis sp. nov.: A, 2 P5 anterior (holotype, SAM C6219); B, ¢ PS anterior (paratype, SAM C6221); C, 2 right caudal ramus, dorsal (holotype, SAM C6219), setae numbered; D, ¢ right caudal ramus, dorsal (paratype, SAM C6220). NEW SPECIES OF NEOPELTOPSIS 179 Basis elongate with | naked outer seta. Exopod-1 with pinnate outer spine; no inner seta; exopod-2 with 2 rows of small spinules, outer spine pinnate and elongate, nearly twice as long as entire exopod- 2, inner seta plumose, 2 rows of spinules on posterior surface and a pore in the distal margin of the segment (arrowed in Fig. 5B); exopod-3 with 3 inner setae (proximal with additional row of pinnules), 2 distal setae, 2 pinnate outer spines, 2 inner, posterior spinule rows. Endopod-! with | plumose inner seta; endopod-2 without seta; endopod-3 with 2 inner setae and 2 distal pinnate setae. P5 (Fig. 6A) exopod partially fused to baseoendopod. Baseoendopod outer seta smooth and arising from setophore; endopodal lobe with 4 bare setae and | tube pore. Exopod with 5 pinnate setae, 1 naked seta and | ventral pore. Po (Fig. 1A) small, kidney shaped, with 1 seta; covering genital apertures. Single copulatory pore (Fig. 1A) located on ventral midline (slightly) posterior to genital apertures, covered by operculum (Fig. 1A). Description of male Mean total body length 0.68 mm + 0.05 mm (n= 5). Body (Figs 1B, 7A). Sexual dimorphism in body size, Al, Pl and P5. Al (Figs 2B — C, 7C) haplocer, 8-segmented, segment 7 with 2 curved, thorn-like projections, aesthetascs on segments 3 and 5. Armature formula for Al: 1-[1], 2-[11], 3-[5 + (1+ae)], 4-[2], 5-[5 + (1+ae)], 6-[0], 7-[0], 8-[9]. Oral appendages as in female. Swimming legs as for female except: dP1 proportionally longer than that of the female; extends to distal edge of caudal rami whereas 2 PI only reaches to distal margin of egg sac. P5 (Fig. 6B) exopod partially fused to baseoendopod; outer basal setophore bearing | naked seta; endopodal lobe with | naked seta and | tube pore; exopod with 3 smooth and 3 spinulose setae; ventral exopodal surface with pore. P6 (Fig. 1B) left and right identical, large, semi-circular, without setae. Urosome with 2 spermatophores. Etymology The specific name althorpensis is derived from the type locality, Althorpe L, SA. Variability The left P2 endopod-2 and endopod-3 of the holotype were partially fused along the inner margin but the right P2 endopod was clearly 3-segmented, which is the normal condition. The left PS of one of the male paratypes was smaller than the right one; the right PS represented the normal size as it was the same size as both the left and right P5 in other male paratypes. Remarks This species was found (rarely) on Lobospira bicuspidata (Phaeophyta). The family level diagnosis of Huys et al. (1996) and Boxshall & Halsey (2004) must be emended as N. althorpensis and N. pectinipes males have two spermatophores, not one. Discussion Neopeltopsis is distinguished from other Peltidiidae genera by two autapomorphies: the possession of four pectinate setae on the Pl exopod- 3 and the medial fusion of the abdominal somites. No other characters are unique to just one genus of Peltidiidae (Hicks 1986), and for this reason Neopeltopsis can be considered the best defined genus of the Peltidiidae. The possession of five armature elements on the Pl exopod-3 is considered the plesiomorphic condition and while there are other genera of Peltidiidae with four armature elements on the terminal segment of P1, they are not pectinate (i.e. Peltidium and Parapeltidium A. Scott, 1909). As the new species from Althorpe I. possesses four pectinate setae on the P] exopod-3 and exhibits fusion of the abdominal somites, it is placed in the genus Neopeltopsis. In addition to these two apomorphies, Neopeltopsis shares many other character states with N. hicksi and N. pectinipes; however, it is separated from its congeners by: 1) Al of male with two curved, thorn-like projections on segment 7 (N. pectinipes with one thorn-like projection on segments 6 and 7; N. hicksi has only one projection on segment 7). Based on the illustrations of Hicks (1976) and Pallares (1979) it is assumed that the segmentation of the Al of the male was misinterpreted. It is most probable that segment 4, a small sclerite, was mistakenly considered to be part of segment 3. It is also possible the terminal segment observed by Hicks (1976) is in fact fused to the preceding segment. If this is not the case, and Hicks (1976) did overlook the small somite which is segment 4, then the dAI of N. pectinipes has 10 segments; 2) A2 exopod l-segmented and with 4 setae (N. pectinipes \-segmented with 2 setae; N. hicksi 2-segmented, endopod-! with | seta and endopod-2 with 3 setae); 3) Mandible exopod length ~ width (N. pectinipes length ~2 x width, i.e. cylindrical; NV. hicksi same as for N. althorpensis); 4) Pl exopod-2 0.25 size of exopod-1 (N. pectinipes 0.75 size of exopod-1; N. hicksi approximately equal to exopod-1); 5) P2 endopod-2 without inner seta (N. pectinipes and WN. hicksi with | inner seta): 180 G. K. WALKER-SMITH Fig. 7. Neopeltopsis althorpensis sp. nov., 5 paratype, SAM C6224: A, habitus, dorsolateral; B, integument; C, Al, arrow indicates thorn-like projection. 6) P2 endopod-3 with four setae (NV. hicksi has three and N. pectinipes has four); 7) P2 exopod-3 with only 1 outer spine (N. pectinipes and N. hicksi with 2 outer spines); 8) P5 exopod partially fused to baseoendopod in both sexes (not fused in N. pectinipes or N. hicksi). The absence of an inner seta on P2 endopod-2 has not been reported for any other member of the Peltidiidae. Careful examination of male and female paratypes of N. althorpensis revealed all specimens lacked an inner seta on the P2 endodpod-2 and there was no scar to indicate the seta had broken away. Fusion of the P5 baseoendopod and exopod has arisen independently in three other peltid genera: Parapeltidium, Alteuthella A. Scott, 1909 and Alteuthellopsis Lang, 1944 (Hicks 1986). Humes (1981) recorded the incomplete fusion of the PS exopod and baseoendopod in A/teuthellopsis corallina but Hicks (1986) believed this to be an illusion created by the orientation of the slide mount, which disappeared when the limb was rotated. This does not appear to be the case for N. althorpensis. All paratypes examined exhibited partial fusion of the P5 baseoendopod and exopod. NEW SPECIES OF NEOPELTOPSIS 18] Fig. 8. Neopeltopsis althorpensis sp. nov.. A, 2 Mxp (paratype, SAM C6223), arrow indicates pad-like seta; B, 3 PI, arrow indicates ventral pectinate spine (paratype, SAM C6224); C, 2? Pl, left arrow indicates ventral pectinate spine, right arrow indicates patch of tiny denticles (paratype, SAM C6223). Based on the following apomorphic characters; the fusion of the A2 exopod segments, the reduced size of the Pl exopod, the reduction in setation of P2 endopod-2 and exopod-3, and the partial fusion of the PS exopod and baseoendopod, it is suggested that N. althorpensis is the most derived species of Neopeltopsis (currently). Neopeltopsis althorpensis appears to be most closely related to N. pectinipes, sharing one apomorphic character state: A2 exopod I-segmented. With a 2-segmented A2 exopod, N. hicksi can be considered the least derived within the genus; however, this species does possess an advanced character state: 3 setae on the P2 endopod- 3 instead of 4, as found in N. althorpensis and N. pectinipes. Hicks (1986) suggested Neopeltopsis was most closely related to Eupelte Claus, 1860 and Alteuthellopsis since they all had a 2-segmented P1 endopod (see Hicks 1986; Fig. 4). Hicks (1986) also believed Neopeltopsis and Alteuthellopsis were related by the possession of two outer spines on the exopod-3 of P2—P4. Strictly speaking, this character state can no longer define the terminal clade of Neopeltopsis and Altheuthellopsis because the P2 exopod-3 of N. a/thorpensis only has one outer spine. However, since Neopeltopsis and Altheuthellopsis both possess <3 spines on the P2 exopod-3 and have two outer spines on the exopod of P3 and P4, this clade still stands. Relationships between the other peltid genera, as suggested by Hicks (1986), also remain unchanged. Neopeltopsis althorpensis is the first species of Neopeltopsis recorded from Australia and the first species of Peltidiidae described since Alteutha polarsternae Dahms, 1992 was described from the Weddell Sea (Antarctica). In addition to N. althorpensis, five other species of Peltidiidae were collected from Althorpe I., SA: Peltidium simplex, Alteutha depressa, and three new species of Alteutha, which will be described in a future paper. Acknowledgements This work was supported by a grant from the Australian Biological Resources Study. Sincere thanks go to Alastair Hirst for collecting the specimens and identifying the algae. I would also like to thank the organisers of the Althorpes 2004 Expedition for providing the opportunity to conduct this research. Thanks to Lyn Waterhouse from Adelaide Microscopy for her assistance with the SEM, and to the anonymous referees whose comments helped improve the manuscript. 182 G. K. WALKER-SMITH References Bartscu, I. (1993) Syngastes craterifer, ein neuer Tegastidae (Harpacticoida, Copepoda) von Siidwesteraustralien. Mitt. Hamb. Zool. Mus. Inst. 90, 187-195. Bartscu, I. (1994) Three new species of Syngastes from south-western Australia. Spixiana 17, 161-173. (1995) A new tegastid (Tegastidae: Harpacticoida: Copepoda) from southwestern Australia: Syngastes dentipes sp. nov. Rec. W Aust. Mus. 17, 221-226. (1999) Three new species of Syngastes (Tegastidae, Harpacticoida, Copepoda) from Western Australia. pp. 295-313 In Walker, D. I. & Wells, F. E. (Eds) “The Seagrass Flora and Fauna of Rottnest Island, Western Australia” (Western Australia Museum, Perth). BOxXsHALL, G. A. & HALsEy, S. H. (2004) “An Introduction to Copepod Diversity” (The Ray Society, London). GiereE, O. (1993) “Meiobenthology: the microscopic fauna in aquatic sediments” (Springer-Verlag, Berlin, Germany). Hamonp, R. (1971) The Australian species of Mesochra (Crustacea: Harpacticoida), with a comprehensive key to the genus. Aust. J. Zool. Supp. 7, 1-32. (1973a) A review of Cletodes (Crustacea: Harpacticoida), with the description of a new species from Queensland. Mem. Old. Mus. 16, 471-483. (1973b) Four new copepods (Crustacea: Harpacticoida: Canuellidae), simultaneously occurring with Diogenes senex (Crustacea: Paguridae) near Sydney. Proc. Linn, Soc. NSW. 97, 165-201. (1973c) Robertgurneya smithi nov. sp. (Crustacea; Harpacticoida), with notes on other species of the genus. J. Nat. Hist. 7, 65-76. (1973d) The Australian species of Robertsonia (Crustacea, Harpacticoida), with a revised key to the genus. Rec. Aust. Mus. 28, 421-435. (1973e) The harpacticoid copepods (Crustacea) of the saline lakes in southeast Australia, with special reference to the Laophontidae. Rec. Aust. Mus. 28, 393- 420. (1974) Australonannopus aestuarinus gen. nov., sp. nov. (Crustacea: Harpacticoida), an aberrant cletodid from Australian brackish waters. Pap. Proc. R. Soc. Tas. 107, 239-245. (1987) Non-marine harpacticoid copepods of Australia. I. Canthocamptidae of the genus Canthocamptus Westwood s. lat. and Fibulacamptus, gen. noy., and including the description of a related new species of Canthocamptus from New Caledonia. /nvert. Taxon. 1, 1023-1247. Harris, V. A. (1994) New species belonging to the family Porcellidiidae (Harpacticoida: Copepoda) from Kioloa, New South Wales, Australia. Rec. Aust. Mus. 46, 303-340. (2002) A new genus belonging to the family Porcellidiidae (Crustacea: Copepoda: Harpacticoida) with three new species from Australia. Rec, Aust. Mus. 54, 1-24. & Robertson, H. M. (1994) New species belonging to the family Porcellidiidae (Harpacticoida: Copepoda) from the southern coast of New South Wales, Australia. Rec. Aust. Mus. 46, 257-301. Hicks, G. R. F. (1976) Neopeltopsis pectinipes, anew genus and species of seaweed dwelling copepod (Harpacticoida: Peltidiidae) from Wellington, New Zealand. N.Z. J. Mar. Freshwat. Res. 10, 363-370. (1986) Phylogenetic relationships within the harpacticoid copepod family Peltidiidae Sars, including the description of a new genus. Zool. J. Linn. Soc. 86, 349-362. Humes, A. G. (1981) Harpacticoid copepods associated with Cnidaria in the Indo-west Pacific. 4 Crust. Biol. 1, 227-240. (1991) Copepoda associated with scleractinian corals on the Great Barrier Reef, northeastern Australia, with a key to the genera of the Lichomolgidae, J Nat. Hist. 25, 1171-1231. Huys, R., & BOxsHALL, G. A. (1991) “Copepod Evolution” (The Ray Society, London). , Gee, J. M., Moore, C. G., & Hamond, R. (1996) “Marine and brackish water harpacticoid copepods, part 1” (Field Studies Council, Shrewsbury, England). Karanovic, T. (2004) “Subterranean copepods from arid Western Australia. Crustaceana Monographs, 3”. (Brill, Leiden). Lana, K. (1934) Marine Harpacticiden von der Campell- Insel und einigen anderen siidlichen Inseln. Acta Uni. Lund, New Series Avd. 2, 30, 1-56. NICHOLLS, A. G. (1941) Littoral Copepoda from South Australia. I — Harpacticoida. Rec, S. Aust. Mus. 6, 381- 427. (1942) Marine Copepoda from Western Australia. I. Littoral harpacticoids from Rottnest Island. J. R. Soc. W. Aust. 27, 135-141. (1945a) Marine Copepoda from Western Australia. I[I. Littoral harpacticoids from Port Denison. J. R. Soc. W. Aust, 29, 1-16. (1945b) Marine Copepoda from Western Australia. IV. Psammophilous harpacticoids. J. R. Soc. W Aust. 29, 17-24. (1945c) Marine Copepoda from Western Australia V. A new species of Paramesochra, with an account of a new harpacticoid family, the Remaneidae, and its affinities. J R. Soc. W. Aust. 29, 91-105. PALLARES, R. E. (1979) Copepodos Harpacticoides marinos de Tierra del Fuego (Argentina) Isla de los Estados III. Contr. Cient. CIBIMA 142, 1-28. WELLS, J. B. J. & Rao, G. C. (1987) Littoral Harpacticoida (Crustacea: Copepoda) from Andaman and Nicobar Islands. Mem. Zool. Surv. India. 16(4), 1-385. Transactions of the Royal Society of S. Aust. (2005), 129(2), 183-192. REEF FISHES OF THE ALTHORPE ISLANDS AND ADJACENT COASTS OF CENTRAL SOUTH AUSTRALIA by S. A. SHEPHERD!, G. J. EDGAR? & N. S. BARRETT Summary SHEPHERD, S. A., EDGAR, G, J. & BARRETT, N. S. (2005). Reef fishes of the Althorpe Islands and adjacent coasts of central South Australia. Trans. R. Soc. S. Aust. 129(2), 183-192, 30 November, 2005. Reef fish populations were surveyed by visual census from 2 — 20 m depth at a range of sites at the Althorpe Is, Haystack I., and nearby mainland reference sites. Mean densities ranged mainly from 0.13 — 0.18 m2, and comprised mainly benthic carnivores, herbivores and omnivores. Densities are comparable to those elsewhere in SE Australia, but much lower than those in the eastern Great Australian Bight. Forty fish species were recorded at Althorpe Is, 28 at Haystack I. and less at the other sites. Most species were too low in abundance to examine depth-related patterns, but the three most common species, blue-throated wrasse, herring cale and magpie perch, decreased in abundance with depth. The main faunal differences encountered were between mainland and island sites, and with depth. Biogeographically, the reef fish fauna was largely of widely ranging southern Australian temperate species, but included six species with SW Australian affinities, and three species with SE Australian affinities. Key Worbs: Reef fishes; southern Australia; depth distributions; temperate fishes; fish abundance patterns. Introduction Prior to this survey, the fish assemblages of shallow, temperate reefs of the central South Australian coast were not formally described, and could only be extrapolated from descriptions for the islands of the eastern Great Australian Bight (Kuiter 1983; Branden et al. 1986; Shepherd & Brook 2003), and from SW and SE Australia (Edgar 1984; Barrett & Edgar 1993; Turner & Norman 1998; Curley ef al. 2002; Harman et al. 2003). The Althorpe Islands, comprising the main Althorpe I. and the Western Isles (see Fig. 1 in Murray-Jones & Shepherd this issue), are situated at the entrance of Investigator Strait, at the juxtaposition of oceanic waters of the South Australian Sea (Bye 1976) with waters of Spencer and St Vincent Gulfs. Although only 8 km from mainland Yorke Peninsula, the islands enjoy oceanic temperatures and salinities (Bye 1976; Thomas & Shepherd 1982), and receive nutrients from the seasonal summer oceanic upwelling off western Kangaroo I., which penetrates to the entrance to Investigator Strait (Petrusevics 1993), and increases chlorophyll a levels (Dimmlich et al. 2004). These conditions may explain the moderately rich benthic fauna and flora of the islands (Baker er al. 2005), ' Senior Research Fellow, South Australian Research and Development Institute, PO Box 120 Henley Beach, S.A. 5022. Email: shepherd.scoresby@saugov.sa.gov.au School of Zoology, University of Tasmania, Private Bag 49, Hobart, Tasmania. 7001. Marine Research/Laboratories, Tasmanian Aquaculture and Fisheries Institute, University of Tasmania, Private Bag 49, Hobart, Tasmania. 7001. which give the islands significance for biodiversity conservation. Biogeographically, the fish fauna of the Althorpe Is also has interest because the islands lie in the transitional region between the south- western and south-eastern Australian fish faunas (Wilson & Allen 1987). The Althorpe Is Expedition took place from 31 January to 10 February 2002 (Murray-Jones & Shepherd this issue), and gave an opportunity to survey inter alia the littoral reef fishes. Our purpose was to describe the diversity and abundance of reef fishes at different depths around Althorpe and Haystack Is, and at nearby reference mainland sites, consider the biogeographic affinities of the fish assemblages, and assess the value of the sites as potential marine reserves. Surveys such as this are a vital component of this assessment process (Edgar e/ al. 1997). Methods Site descriptions All shores of the islands, except within the sheltered north-easterly facing Mooring Bay, are subject to moderate to strong SW swells, and all are exposed to high wind waves, with fetches variously ranging from 12 to 100 km. The mainland sites were exposed to the SW swell, and to wind waves from SE to SW for Sites 8, 14 and 15, from NE to NW for Site 7, and from NW to SW for Sites 9 and 10 (see Table 1; Fig. 1, Murray-Jones & Shepherd 2005). Rocky substratum around the Althorpe Is is granitic, comprising mainly smoothly sloping rock with occasional joints and caves. At Site 6 rock 184 S. A. SHEPHERD, G. J. EDGAR & N.S. BARRETT TABLE |. Location of sites and depths sampled, with number of replicate samples (per 500 m?), at the Althorpe Is and at adjacent mainland sites. A dashed line indicates that rock met sand above the depth shown. Site Lat. Long. Number of replicates at each depth (m) 2 5 10 15 20 Althorpe Islands 1. The Boulders 35° 22.174' S; 136° 52.252’ E 4 5 4 3 - 2. Western Isles (NE) 35° 22.071’ S; 136° 50.952’ E l 3 4 2 l 3. Chain Islet 35° 22.448’ S; 136° 51.718’ E 0 4 4 0 0 4. Northwest Bay 35° 21.871’ S; 136° 51.406’ E 0 0 4 0 0 5. Swallowtail Bay 35° 22.480’ S; 136° 51.480’ E 4 4 4 2 2 6. Mooring Bay (W) 35° 22.011' S; 136° 51.623’ E 3 - - - - 7. Haystack Island (NE) 35° 19.406’ S; 136° 54.500’ E 0 4 0 2 - 8. Haystack I. (NW) 35° 19.266’ S; 136° 54.431’ E 0 4 0 - Yorke Peninsula 9. Carney Point 34° 53.694’ S; 137° 00.264’ E 0 4 4 - - 10. Cape de Berg 34° 56.456’ S; 136° 58.580' E 0 4 4 - - 11. Chinamans Hat I. (S) 35° 17.419’ S; 136° 54.967' E 4 3 2 - - 12. Chinamans Hat I. (NW) 35° 17.914’ S; 136° 54.901’ E 2 - - - - meets sand at a depth of | — 3 m, and elsewhere around the islands at 20 — 30 m depth. At Haystack I. and Chinamans Hat I., the substratum is travertine limestone, and rock meets sand at 3 — 15 m according to exposure. At Carney Point and Cape de Berg, the substratum is granitic and rock meets sand at 12 — 15 m depth. Transect methods In all, 94 100 m transects were surveyed at 12 dive sites from depths of 2—20 m at mainly 5 m intervals (Table 1), although time constraints severely limited sampling the deeper sites. We used a standard visual census method (Barrett & Buxton 2002) to estimate the size and number of all fish species within 5 m of a 100 m or two 50 m lines set sequentially at the given depth i.e. covering 500 m2 per transect (see Shepherd & Brook 2003 for further details). Experienced divers swam along the line about | m above the algal canopy, and estimated and recorded on a slate the length of every species. Habitats at all sites were similar and variously dominated by the laminarian Ecklonia radiata or species of Cystophora, Sargassum or Acrocarpia paniculata (Baker et al, 2005). Underwater visibility was 6 — 10 m. Analysis Following Shepherd & Brook (2003), reef fishes were Classified in five ecological groups according to behaviour and life habit as follows: Group 1 — pelagic and mid-water species; Group 2 — species that live in seagrasses but wander into algal habitats; Group 3 — demersal, site-attached species that swim above and close to the algal canopy; Group 4 — demersal, site-attached species usually cryptic within the algal canopy; and Group 5 — cryptic or cave- dwelling species, active mainly at night. To examine the relationship between fish assemblages at the differing sites and depths surveyed, a Bray-Curtis similarity matrix was calculated from log (x+1) abundance data, and the resulting site relationships plotted as two- dimensional ordinations using multidimensional scaling techniques (MDS) (Clarke 1993), Data were log (x+1) transformed to avoid dominance of the most common species, and to give greater weight to rarer species. To avoid bias due to the differing number of transects between sites, data from only two transects per site were randomly selected for analysis. Abundances of common species were examined for correlations with depth. Information on life habit and nomenclature of fishes was variously extracted from the publications of Gomon ef al. (1994), Edgar (1997), Hutchins & Swainston (1999) and Shepherd & Brook (2003). Results Assemblage patterns The MDS plot for fish abundance data by site and depth (Fig. 1) shows two distinct trends. First, the Althorpe Is sites cluster together, with mainland sites (Chinamans Hat I., Carney Point and Cape de Berg) forming outliers. Secondly, within the Althorpe Is cluster, fish assemblages tend to cluster by depth for the 2, 5 and 10 m sets. We accordingly present summaries of mean fish densities for each depth interval for all Althorpe Is sites combined (Table 2) and for Haystack and Chinamans Hat I. (Table 3). Data for Carney Point and Cape de Berg are presented for 5 and 10 m depths combined (Table 3). REEF FISHES OF THE ALTHORPE ISLANDS 185 fish all depths MDS Stress: 0.23 Fig. 1. MDS ordination comparing species’ composition and abundance at 5 depths (in metres) and the 12 sites numbered in Table 1. TABLE 2. List of fish species recorded at the Althorpe Is at six sites from 2 — 20 m depth, ordered according to ecological group. Mean densities for sites are given in numbers 500 m?, and mean size in cm. % occurrence is the percentage of transects in which the species was recorded. FT is feeding type. BC = benthic carnivore; C = carnivore; H = herbivore; O = omnivore; P = planktivore. Standard errors in brackets. : a Density % amieratepeeies ie 2m 5m 10 m 1S m 20m occ. Group 1 Dinolestes lewini C 0.3 1.9 4.5 0.7 0 33 Long-finned pike (0.2) (0.9) (1.6) (0.5) Scorpis aequipinnis P 5.3 10.2 4.0 11.3 0.3 72 Sea sweep (2.8) (4.6) (1.0) (5.9) (0.2) Caesioperca rasor P 0 0.6 6.3 0.2 0.5 12 Barber perch (0.5) (5.7) (0.1) (0.3) Enoplosus armatus P 0.4 1.5 3.3 0.7 1.0 60 Old wife (0.1) (0.3) (1.1) (0.1) (0.7) Pseudocaranx dentex BC 0.1 0.1 0 0.2 0 2 Silver trevally (0.1) (0.1) (0.1) Arripis georgiana P 0.4 0 0 0 0 2 186 TABLE 2. Cont. S. A. SHEPHERD, G. J. EDGAR & N. S. BARRETT : Densit % Name:oh species rE 2m 5m 10 a 15m 20m oce. Tommy ruff (0.3) Mean density 6.4 14.3 18.1 13.0 1.8 Group 2 Myliobatis australis BC 0.06 0 0 0 0 2 Eagle ray (0.04) Upeneichthys vlamingii BC 0.9 0.5 1.3 14.3 1.8 40 Red mullet (0.3) (0.2) (0.8) (1.7) (0.9) Mean density 1.0 0.5 4.3 14.3 1.8 Group 3 Achoerodus gouldii BC 0.9 0.8 0.5 1.0 1.0 47 Western blue groper (0.4) (0.3) (0.1) (0.6) (0) Pentaceropsis recurvirostris BC 0 0 0.05 0.2 0 4 Long-snouted boarfish (0.04) (0.1) Dactylophora nigricans H 1.1 0.5 0.3 1.0 0.3 35 Dusky morwong (0.6) (0.3) (0.1) (0.4) (0.2) Kyphosus sydneyanus H 0 0.2 0.1 0 0 5 Silver drummer (0.2) (0.08) Cheilodactylus nigripes BC 6.2 3.2 2.5 1.5 2.8 84 Magpie perch (1.0) (0.5) (0.6) (0.2) (0.2) Girella zebra H 17.9 4.3 3.4 1.7 1.0 65 Zebra fish (12.1) (2.5) (1.3) (0.8) (0.7) Notolabrus tetricus BC 40.5 30.7 23.0 21.7 15.8 93 Blue-throated wrasse (7.9) (2.8) (3.1) (1.3) (3.4) Dotalabrus aurantiacus BC 0.4 0.2 0.3 0 0 21 Castelnau’s wrasse (0.2) (0.2) (0.1) Pseudolabrus parilus BC 0 0 0 1.8 0 4 Brown-spotted wrasse (1.5) Ophthalmolepis lineolatus BC 0 0 0 0.7 0 2 Maori wrasse (0.5) Austrolabrus maculatus BC 0 0 0.2 0 0 5 Black-spotted wrasse (0.1) Meuschenia flavolineata oO 1.4 0.6 0.6 1.0 0 30 Yellowstripe leatherjacket (1.2) (0.2) (0.2) (0.8) Meuschenia hippocrepis BC 0.9 0.9 0.8 1.5 0 49 Horseshoe leatherjacket (0.5) (0.2) (0.3) (1.2) Meuschenia freycineti O 0 0.03 0 0 0 2 6-spined leatherjacket (0.02) Meuschenia galii oO 0 0.2 0.3 0 0 16 Blue-lined leatherjacket (0.07) (0.1) Meuschenia venusta O 0 0 0.05 0 0 2 Stars and stripes leatherjacket (0.04) Acanthaluteres vittiger BC 0 0.08 0.10 0 0 5 Toothbrush leatherjacket (0.07) (0.05) Eubalichthys gunnii BC 0.3 0 0 0.7 0 4 Gunn’s leatherjacket (0.3) (0.5) Eubalichthys mosaicus C 0 0.05 0 0 0 2 Mosaic leatherjacket (0.04) Mean density 69.8 41.6 3237 32.7 20.8 Group 4 Aplodactylus arctidens H 0 0.06 0.05 0 0 + Southern sea carp (0.05) (0.04) Aplodactylus westralis H 0.1 0.05 0 1.3 0 9 Western sea carp (0.1) (0.04) (1.1) REEF FISHES OF THE ALTHORPE ISLANDS 187 TABLE 2. Cont. : Densit % Maung or SPevies i =m 3m Omms 15m 20m oce. Odax acroptilus BC 0 0.1 0.1 0 0 7 Rainbow cale (0.06) (0.08) Odax cyanomelas H 3.2 512 1.5 1.2 1.0 75 Herring cale (1.4) (1.3) (0.5) (0.8) (0) Pictilabrus laticlavius BC 0.5 1.7 2.0 3.8 1.5 67 Senator wrasse (0.3) (0.3) (0.6) (1.7) (0.4) Eupetrichthys angustipes BC 0 0 0.1 0 0 + Snakeskin wrasse (0.06) Siphonognathus beddomei BC 0 0 2.1 0 0 11 Pencil weed whiting (1.2) Parma victoriae H 1.3 1.2 0.4 0.5 0.8 49 Victorian scalyfin (0.4) (0.4) (0.05) (0.4) (0.2) Tilodon sexfasciatum BC 0.4 0.2 0.2 0.2 0 26 Moonlighter (0.3) (0.1) (0.1) (0.1) Diodon nichthemerus BC 0 0.05 0 0 2 Globefish (0.04) Mean density D5 8.5 6.5 7.0 3.3 Group 5 Parascyllium variolatum BC 0 0.06 0 0 0 2 Varied catshark (0.04) Centroberyx lineatus P 0 0 0.05 0 0 2 Swallowtail (0.04) Pempheris multiradiata Cc 0 4.0 6.7 0.2 1.8 26 Common bullseye (0.9) (3.4) (0.1) (1.3) Mean density 0 4.1 6.8 0.2 1.8 Total density 82.7 69.0 65.4 67.2 29.5 Total number of species 21 28 30 23 13 TABLE 3. Mean densities of fish species (in numbers 500 m? at Haystack 1., Chinamans Hat I, Carney Point (CP) and Cape de Berg (CdB) at various depths, ordered according to ecological group. % occurrence (the percentage of transects in which the species was recorded) is given for Haystack 1, Chinamans Hat 1. and for Carney Point and Cape de Berg combined. FT is feeding type. BC = benthic carnivore; C = carnivore; H = herbivore; O = omnivore; P = planktivore. Standard errors (in brackets) are given for replicate sites or, in the case of Carney Point and Cape de Berg, for two depths. Haystack I. Chinamans Hat I. CP CdB Name of species FT 5m 15m % 2m 5m 10m % 5S, 5, % Occ Occ 10m 10m Occ Group 1 Dinolestes lewini GC 1.1 0 20 0 0 0 0 0 0.5 6 Long-finned pike (0.8) (0.4) Scorpis aequipinnis P 12.8 10.5 80 2.9 4.4 0.5 67 10.3 12.8 69 Sea sweep (0.2) (6.4) (1.7) (0.8) (5.9) (8.0) Scorpis georgianus P 0 0 0 0 0 0.5 7 0 0 0 Banded sweep Enoplosus armatus P 2.0 0.5 60 0 0 0 0 1S 1.3 25 Old wife (0.4) (0.4) (1.2) (0.9) Arripis georgiana P 5.0 0 10 0 0.2 0 7 0 0 0 Tommy ruff (3.6) (0.1) Mugil cephalus P 0 0 0 0 0.5 0 13 0 0 0 Mullet (0.4) Mean density 20.9 11.0 2.9: 5.1 1.0 11.8 14.6 188 TABLE 3. Cont. S. A. SHEPHERD, G. J. EDGAR & N. S. BARRETT Haystack I. Chinamans Hat I. GP CdB Name of species FT 5m Sm % 2m 5m 10m % By 35 % Occ Occ 10m 10m Occ Group 2 Upeneichthys vlamingii BC 0.4 0 10 0 0.2 0 7 1.3 2.0 63 Red mullet (0.3) (0.1) (0.5) (0.4) Parequula melbournensis BC 0 0 0 0 0.4 0 13 0 0 0 Silverbelly (0.3) Siphamia cephalotes P 0 0 0 0 0 0 0 0 4.5 19 Wood's siphonfish (2.5) Mean density 0.4 0 0 0.6 0 1.3 6.5 Group 3 Achoerodus gouldii BC 0.6 0 40 4.4 1.9 0.5 73 0.3 0 6 Western blue groper (0.3) (0.4) (1.6) (0.2) Dactylophora nigricans H 1.3 1.0 60 1.3 1A 0 53 0.8 0.3 19 Dusky morwong (0.1) = (0) (0.9) (0.8) (0.6) (0.2) Kyphosus sydneyanus H 2.1 0 40 4.4 3.3 0 40 12.5 75 56 Silver drummer (1.0) (0.8) (1.1) (8.2) (5.3) Cheilodactylus nigripes BC 2.8 7.0 90 0.5 133 255: 47 3.5 5.8 81 Magpie perch (0) (2.1) (0) (0.9) (0) (1.6) Girella zebra H 8.5 5.0 90 5.5 3.6 0 53 0.8 0.8 19 Zebra fish (0.5) (0.7) (0.4) (2.3) (0.5) (0.2) Notolabrus tetricus BC 26.6 44.5 100 11.5 13.8 11.5 100 17.8 11.3 100 Blue-throated wrasse (4.0) (3.2) (0.4) (1.3) (2.7) (4.1) Dotalabrus aurantiacus BC 23 3.0 80 0 0 0.5 af 0.8 0.3 19 Castelnau’s wrasse (0) (0) (0.6) (0.2) Austrolabrus maculatus BC 0.4 0 10 0 0 0 0 1.0 0.3 19 Black-spotted wrasse (0.3) (0.7) (0.2) Meuschenia flavolineata O 3.0 5.0 80 0 1.4 1.5 27 4.0 0.5 50 Yellowstripe leatherjacket (1.0) (2.1) (0.9) (3.2) (0.4) Meuschenia hippocrepis BC 5.0 1.0 70 1.1 Lo: 2.0 60 9.5 4.8 75 Horseshoe leatherjacket (1.8) (1.0) (0.3) (6.3) (0.7) (2.7) Acanthaluteres browniii BC 1.1 0 10 0 0 0 0 0 0 0 Spiny-tailed leatherjacket (0.8) Meuschenia galii BC 0.4 0 20 0 0.2 0 7 0.3 0.3 13 Blue-lined leatherjacket (0.1) (0.1) (0.2) (0.2) Acanthaluteres vittiger O 0.1 0 10 0.1 0 0 7 0 0.5 13 Toothbrush leatherjacket (0.1) (0.1) (0.4) Eubalichthys bucephalus BC 0 2.0 20 0 0 0 0 0 0 0 Black reef leatherjacket (1.4) Eubalichthys cyanoura = BC 0 0 0 0 0.2 0 7 0 0 0 Blue-tailed leatherjacket (0.1) Chelmonops curiosus BC 0.8 0 40 0 0.1 1.5 13 0.5 0.5 13 Western talma (0.2) (0.1) (1.1) (0.3) (0.4) Mean density 55.0 68.5 28.8 34.8 19.5 51.8 32.9 Group 4 Odax acroptilus BC 0.5 0 20 0 0 0 0 3.8 0 25 Rainbow cale (0) (2.7) Odax cyanomelas H 10.0 1.5 100 116 7.0 1.5 100 3.5 1.5 50 Herring cale (4.1) (0.4) (0.6) (3.5) d.8) 0 (1) Pictilabrus laticlavius BC 1.9 0 40 0.3 0.1 0.5 20 0.8 3.0 44 Senator wrasse (1.0) (0.2) = (0.1) (0.2) (1.1) Siphonognathus beddomei BC 0.8 0 50 0 0 0 0 14.5 1.3 50 Pencil weed whiting (0.2) (8.5) (0.9) Parma victoriae H 44 0 60 3.3 2.0 0.5 87 0.8 0.8 19 REEF FISHES OF THE ALTHORPE ISLANDS 189 TABLE 3. Cont. Haystack I. Chinamans Hat I. CP. CdB Name of species FT 5m 5m % 2m 5m 10m % 5; 5, % Oce Occ 10m 10m Occ Victorian scalyfin (1.0) (0.2) (1.1) (0.6) (0.5) Tilodon sexfasciatum BC L.1 1.0 70 1.6 1.0 1.5 53 2.8 0.5 44 Moonlighter (0.1) (0.8) (0.3) (0.6) (0.5) (0.4) Heteroclinus johnstoni = C 0.1 0 10 0 0 0 0 0 0 0 Johnston’s weedfish (0.1) Mean density 18.9 2.5 16.8 10.1 4.0 26.2 7.1 Group 5 Paraplesiops meleagris — C 0.4 0.5 50 0 0 0 0 0 0 0 Western blue devil (0.1) (0.4) Pempheris klunzingeri — C 0 0 0 0 0 0 0 0.3 1.5 13 Rough bullseye (0.2) (1.1) Pempheris multiradiata C 1.5 1.5 60 0.8 0.7 1.0 27 0.8 2.8 25 Common bullseye (0.5) (1.1) (0.5) (0.5) (0.2) (1.2) Mean density 1.9 2.0 0.8 0.7 1.0 1.1 4.3 Total density 97.0 84.0 49.3 513 25.5 92.2 65.4 Total number of species 28 14 14 21 13 23 24 Species richness, density and trophic groupings The abundance, Amp, of magpie perch, In all, 40 species were recorded at the Althorpe Is, 29 species at Haystack I., 24 species each at Chinamans Hat I. and Cape de Berg, and 23 species at Carney Point. A comparison of Tables 2 and 3 shows that 16 species were present at offshore sites (Althorpe Is or Haystack I.), but not at mainland ones, and three species at mainland, but not offshore, sites. However, these species were not abundant anywhere, and their absence has little significance. Mean fish densities (numbers 500 m-?) were lowest at Chinamans Hat I. (42.0 SE 6.8) and in the range 65 — 92 for other sites, except at 20 m depth at Althorpe I. with 29.5. The most abundant groups were the benthic feeders (Groups 3 and 4), comprising 51% at the Althorpe Is, and variously 61 — 89% elsewhere. The second most abundant group (Group 1) was the pelagic or mid-water species, which comprised 12 —23% of fish numbers at all sites except at Chinamans Hat I. where they were in low numbers. Group 2 were in low numbers except where transects passed near soft bottoms, and Group 5 were present where transects crossed cryptic habitats. Depth distributions: Three common species at Althorpe Is showed trends of declining abundance with depth. For the blue-throated wrasse, Notolabrus tetricus, abundance of both juveniles (<10 cm), Aj, and post-juveniles (>10 cm), Ajj, in numbers 500 m2, declined with depth, D, in metres, the former exponentially and the latter linearly. The respective regression equations are: Aj=5.61—2.11InD (R? =0.78; P< 0.001) Api = 34.94 1.01 D (R2 = 0.33; P< 0.05) Cheilodactylus nigripes, also declined exponentially with depth, giving the regression equation: Amp = 7.20—2.16 In D (R? = 0.33; P < 0.05) Schools of zebra fish, Girella zebra, were sometimes encountered at shallow depths and less commonly with increasing depth. Due to their schooling behaviour, variability in density was high and the decline in abundance with depth was not significant. We also examined the abundance of the herring cale, Odax cyanomelas vs mean cover of its principal diet, Ecklonia radiata, and also vs depth. Ecklonia cover varied between 20 and 85% (except at the shallow western end of Mooring Bay (Site 6), where it was virtually absent), but was not correlated with herring cale abundance (r = 0.13; ns). If the data from Site 6, where herring cale was also absent, are omitted, the abundance, A,,, of herring cale declined exponentially with depth, and the following equation was fitted: Ate = 9.36 - 3.05 In D (R? = 0.52; P < 0.05) Discussion Biogeography The visual search method used in this study is likely to under-estimate or overlook cryptic and nocturnal species associated with reefs, and only by chance includes species typical of deeper water, soft bottoms or seagrasses. Nevertheless, the biogeo- 190 S. A. SHEPHERD, G. J. EDGAR & N. S. BARRETT graphic affinities of the reef fauna are evident. Wilson & Allen (1987) considered that the southern Australian temperate region contained four overlapping faunal components: a south-western component, a south-eastern component, an endemic southern component, and a broader component extending up the east and west Australian coasts. The Althorpe Is lie in an area of transition between the first two components, but the reef fish fauna is dominated overwhelmingly by widely ranging southern Australian species. Three species characteristic of the SW region extending to Althorpe Is were: the western sea carp, Aplodactylus westralis, whose eastern limit is SW Yorke Peninsula; the spiny-tailed leatherjacket, Acanthaluteres brownii, with eastern limits in Investigator Strait; and the western talma, Chelmonops curiosus, with its eastern limit in eastern S.A. Another three species, the brown-spotted wrasse, Notolabrus parilus, the western blue devil, Paraplesiops meleagris, and the western blue groper, Achoerodus gouldii, extend further east to western Victoria. Only three south- eastern species with their western limit in Investigator Strait or Spencer Gulf were present: Gunn’s leatherjacket, Eubalichthys gunnii, the southern sea carp, Aplodactylus arctidens, and Johnston’s weedfish, Heteroclinus johnstoni. Finally, two southern species with disjunct distributions, absent from the colder SE Australian waters were present: the maori wrasse, Ophthalmolepis lineolatus, (with an eastern range now extending to SE Tasmania — unpublished data), and the black- spotted wrasse, Austrolabrus maculatus. None of the above species were common at Althorpe Is. Species abundances Total mean densities of fish in this study (mostly 0.13 — 0.18 m~) are low in comparison with offshore islands of the eastern Great Australian Bight, but similar to those recorded elsewhere in SE Australia (reviewed in Shepherd & Brook 2003). The low densities are largely due to the low abundance of open water feeders (Group 1), which comprise only 6 — 28% of individuals, compared with benthic feeders (Groups 3, 4), which comprise 59 — 92%. Hence, the ‘offshore island effect’, by which islands capture the productivity of the surrounding ocean, and which is reflected mainly in the increased abundance of mid-water species, is apparently absent from the Althorpe Is and nearby mainland sites. Differences in terms of species composition and abundance between sites and depths were minor, as shown by the MDS plot (Fig. 1). We attribute this relative homogeneity between sites to the general topographic and habitat similarity between sites at Althorpe Is. Algal habitats were similar around the islands, as all sites, except those within shallow bays, were exposed to the prevailing SW swell (Baker e7 al. 2005). The few differences that did exist can be attributed to habitat differences (mixed rock, sand and seagrass) and to the modified algal communities in more sheltered sites, as in Mooring Bay, a relative outlier in Fig. 1 (see Shepherd & Brook 2005). Elsewhere, topographic heterogeneity, as at Haystack L., with a high abundance of caves and crevices in an eroded limestone substratum, favoured high fish abundance and greater species diversity. Although mainland sites such as Carney Point, Cape De Berg and Chinamans Hat I. are also relative outliers (Fig. 1), their fish assemblages overall differed little from those of the offshore sites (Tables 2 & 3). Notable differences were: more yellow-stripe and horseshoe leatherjackets, silver drummer, and small blue groper (sheltered sites only), and fewer zebra fish at mainland sites. Overall, there were no obvious differences in habitats between Althorpe Is and mainland sites, other than substratum at several sites, and the obvious lack of deeper rocky habitat (15 mand 20 m) on the Yorke Peninsula coastline due to the shallower soft bottoms found there. The latter difference is important, however, as, while shallow water assemblages are similar, the presence of assemblages and reef habitat > 40 m deep at Althorpe Is distinguishes them from the adjacent coastline and adds significantly to their conservation value. With the exception of three species, referred to below, the abundances of individual species were generally too low to show meaningful patterns on the depth gradient. The three most abundant species all declined in density with depth. This is likely due to several factors. In the case of the wrasse, N. fetricus, juveniles apparently recruit into shallow water, as shown in this and the earlier study of Shepherd & Brook (2003), and move into deeper water with increasing size. The blue groper, A. gouldii, has the same recruitment pattern (Shepherd 2005), but its numbers were generally low at Althorpe Is, although much higher densities of small groper were recorded at sheltered mainland sites. In the case of the magpie perch, C. nigripes, the abundance patterns may be due to several factors. First, the majority of the 2 m depth samples were in partly sheltered bays where topographic complexity (abundance of granite boulders, and crevices) was often high, compared to deeper water. Species like C. nigripes, the scalyfin, Parma victoriae, and wrasses, which are site-attached and rest in shelter-holes, tend to be more abundant in more complex habitats (Cappo 1995; Lincoln Smith & Jones 1995; Curley et al. 2002). Thirdly, the algal mats in which C. nigripes forages (Lowry & Cappo 1999; Wellenreuther & Connell 2002), may decline in extent with increasing depth. Lastly, in the case of the herring cale, the decline in abundance with depth, as REEF FISHES OF THE ALTHORPE ISLANDS 191 at St Francis Is (Shepherd & Brook 2003), may be associated with the general distribution of its preferred food, the kelp Ecklonia radiata, although we found no correlation with Ecklonia cover measured at the scale of our sampling. Percentage occurrence data give a different aspect of abundance. Species with high values (see above) are mainly either site-attached species (Groups 3, 4) with small ranges, requiring cryptic habitat in which to shelter, or mid-water species (Group 1), which concentrate near vertical faces. In the case of the bullseye (Group 5) the % occurrence value likely reflects the frequency of large caves. Low values indicate relative rarity. Habitats Curley et al. (2002) suggested that habitat-related features were the principal determinants of the composition and abundance of temperate fishes, and that sites for marine parks should be selected on the basis of conserving the maximum range of habitat types. Topography, such as degree of relief and substratum type (limestone or granite), depth, degree of shelter, and algal community type are all well- known habitat factors determining the species present and their abundance (Harman ef al. 2003; Lincoln Smith & Jones 1995; Shepherd & Brook 2003). There may also be benefits in conserving island habitats, which have less terrigenous and anthropogenic influences, but have — local oceanographic and climatic forces, contributing to a unique range of habitats (review of Brown & Lomolino 2000). Habitats themselves are a good surrogate for species, because protected areas that incorporate the maximum number of habitat types will also provide refuge for species, often various life stages, and ecological linkages (Ballantine 1997; Nowlis & Friedlander 2004). The Althorpe Is encompass a wide depth range (including depths >15 m not found on the adjacent coastline), and a range of exposures, giving a good representation of granitic habitats, while Haystack I., and Chinamans Hat I. provide a contrasting limestone topography with high topographic complexity, and a range of exposures and depths. Together, these sites provide a gradient in exposure, from rough conditions at the Althorpe Is to moderate shelter at Chinamans Hat, and other mainland sites, and encompass a wide range of habitats, which emir ently meet the above criteria for marine park status. Acknowledgements The expedition was funded by the South Australian Research and Development Institute and _ the Department for Environment and Heritage. We thank Rob Lewis, Anthony Cheshire and Bryan McDonald for their strong support, and Sue Murray-Jones for her organizational skill. We thank the captain and crew of RV Ngerin, Michael Clark for manning diver support craft, Tim Collins for providing the zodiac, and John and Erika Lawley for their hospitality on- shore. Chris Halstead, Ali Bloomfield, James Brook and Renate Velzeboer assisted in the diving. We also thank John Carragher, David Turner and Mike Steer for helpful comments on the paper. References Baker, J. L. EoGar, G. J. & BARRETT, N. S. (2005) Subtidal macroflora of Althorpe Islands, South Australia. Trans. R. Soc. S. Aust, 129, 128-144. BALLANTINE, W. J. (1997) No-take marine reserve networks support fisheries. /n Hancock, D.A., Smith, D. C., Grant, A. & Beumer, J. P. (Eds) “Developing and Sustaining World Fishery Resources: the State of Science and Management”. Proceedings of the Second World Fisheries Congress. Pp.702-706 (CSIRO Publishing, Melbourne). Barrett, N. S. & Buxton, C. (2002) Examining underwater visual census techniques for the assessment of population structure and biodiversity in temperate coastal marine protected areas. Tasmanian Aquaculture & Fisheries Institute. Technical Report Series No. 11, 114 p. Barrett, N.S. & EpGar, G. J. (1993) A survey of potential marine reserve locations in Bass Strait. Tasmanian Dept Primary Industry and Fisheries. Report to Australian National Parks and Wildlife Service, 42 p. Branpen, K. L., EoGar, G. J. & SHEPHERD, S. A. (1986) Spatial reef fish populations of the Investigator Group, South Australia: a comparison of two census methods. Trans. R. Soc, S. Aust. 110, 69-76. Brown, J. H. & Lomo.ino, M. V. (2000) Concluding remarks: historical perspective and the future of island biogeographic theory. Glob. Ecol. Biogeogr. 9, 87-92. Bye, J. A. T. (1976) Physical oceanography of Gulf St Vincent and Investigator Strait. /n Twidale, C. W., Tyler, M. J. & Webb, B. P. (Eds) “Natural History of the Adelaide region”. Pp. 143-160 (Royal Society of South Australia, Adelaide). Cappo, M. (1995) The population biology of the temperate reef fish Cheilodactylus nigripes in an artificial reef environment, Trans. R. Soc. S. Aust. 119, 113-122. CLARKE, K, R, (1993) Non-parametric multivariate analyses of changes in community structure. Aus/. J. Ecol. 18, 117-143. Cur.ey, B. G., KincsForp, M, J. & GILLANDERS, B. M. (2002) Spatial and habitat-related patterns of temperate reef fish assemblages: implications for the design of marine protected areas. Mar. Freshwater Res. 53, 1197- 1210. DimMLicu, W. F., BREED, W. G., GEDDES, M. & WARD, T. M. (2004) Relative importance of gulf and shelf waters for spawning and recruitment of Australian anchovy, Engraulis australis, in South Australia. Fish. Oceanogr. 13, 310-323. 192 S. A. SHEPHERD, G. J. EDGAR & N. S. BARRETT Epaar, G. J. (1984) General features of the ecology and biogeography of Tasmanian subtidal rocky shore communities. Pap. Proc. R. Soc. Tasm. 118, 173-186. (1997) “Australian Marine Life: the Plants and Animals of Temperate Waters”. (Reed Books, Melbourne). , Mover-ey, J., BARRETT, N. S., PETERS, D. & REED, C. (1997) The conservation related benefits of a systematic marine biological sampling program; the Tasmanian reef bioregionalisation as a case study. Biol. Cons. 79, 227-240. Gomon, M. F, Giover, C. J. M. & Kuirer, R. H. (1994) “The Fishes of Australia’s South Coast”. (The Fauna and Flora of South Australia Handbook Committee, Adelaide). Harman, N., Harvey, E. S. & KENDRICK, G. A. (2003) Differences in fish assemblages from different reef habitats at Hamelin Bay, south-western Australia. Mar. Freshwater Res. 54, 177-184. HutTcHIns, B. & SWAINSTON, R. (1999) “Sea Fishes of Southern Australia”. 2nd Ed. (Gary Allen, Sydney). Kuirer, R. H. (1983) An annotated list of fishes of the Investigator Group, South Australia. Fish. Res. Pap. Dep. Fish (S. Aust.) No. 7. 12 p. LINcoLn SmitH, M. P. & Jones, G. P. (1995) Fishes of shallow coastal habitats. Jn Underwood, A. J. & Chapman, M. G. (Eds) “Coastal Marine Ecology of Temperate Australia”. Pp. 240-253 (University of NSW Press: Sydney). Lowry, M. & Cappo, M. (1999) Morwongs. /n Andrew, N. (Ed.) “Under Southern Seas: the Ecology of Australia’s Rocky Shores”. Pp. 172-179 (University of NSW Press: Sydney). Murray-JONES, S. & SHEPHERD, S. A. (2005) Introductory narrative and conservation recommendations. Zrans. R. Soc. S. Aust. 129, 85-89. Now is, J. S. & FRIEDLANDER, A. (2004) Design and designation of marine reserves. /n Sobel, J. A. and Dalgren, C. P. (Eds) “Marine Reserves: A Guide to Science, Design, and Use”. Pp. 128-163. (Island Press, Washington). Prrrusevics, P. M. (1993) SST fronts in inverse estuaries, South Australia — indicators of reduced Gulf-shelf exchange. Aust. J. Mar. Freshwater Res. 44, 305-323. SHEPHERD, S. A. (2005) Ontogenetic changes in diet, feeding behaviour and activity of the western blue groper, Achoerodus gouldii. In Wells, F. E., Walker, D. I. and Kendrick, G. A. (Eds) “The Marine Fauna and Flora of Esperance, Western Australia”. (Western Australian Museum, Perth) (in press). & Brook, J. B. (2003) Encounter 2002 expedition to the Isles of St Francis, South Australia: reef fishes. Trans. R. Soc. S. Aust. 127, 269-279. Tuomas, P. M. & SHEPHERD, S. A. (1982) The marine environment. /n Shepherd, S. A. and Thomas, I. M. (Eds) “Marine Invertebrates of Southern Australia”. Pp. 11-25 (Government Printer, Adelaide). Turner, M. L. & NorMaAN, M. D. (1998) Fishes of Wilsons Promontory and corner Inlet, Victoria: composition and biogeographic affinities. Mem. Mus. Vict. 57, 143-165. WELLENREUTHER, M. & CONNELL, S. D. (2002) Response of predators to prey abundance: separating the effects of prey density and patch size. J. Exp. Mar. Biol. Ecol. 273, 61-71. WILson, B. R. & ALLEN, G. R. (1987) Major components and distribution of marine fauna. /n Dyne, G. R. and Watson, D. W. (Eds) “Fauna of Australia. General Articles”. Volume 1A pp. 43-68 (Australian Government Publishing Service, Canberra). Transactions of the Royal Society of S. Aust. (2005), 129(2), 193-201. CLEANING SYMBIOSIS AMONG INSHORE FISHES AT ALTHORPE ISLAND, SOUTH AUSTRALIA AND ELSEWHERE by S. A. SHEPHERD!, J. TEALE? & D, MUIRHEAD? Summary SHEPHERD, S, A., TEALE, J, & MUIRHEAD, D, (2005). Cleaning symbiosis among inshore fishes at Althorpe Island, South Australia and elsewhere. Trans. R. Soc, S. Aust. 129(2), 193-201, 30 Novernber, 2005. Host-cleaner symbiosis among fish, not previously reported for South Australia, was investigated at Althorpe I., West I., Cape Jervis and other sites over short to long periods. Five cleaner fish species, one crustacean cleaner species, and 23 client fish species are recorded. The cleaning symbiosis is described in some detail for two cleaners, the moonlighter, Tilodon sexfasciatus, and the pencil weed whiting, Siphonognathus beddomei, and in less detail for three other cleaners, the western cleaner clingfish, Cochleoceps bicolor, the black-spotted wrasse, Austrolabrus maculatus, and the old wife, Enoplosus armatus. Cleaning behaviour appears to be facultative, and its timing dependent on cleaner behaviour. At West I. it was restricted to a few occasions per day from late morning to late afternoon during summer months at a fixed cleaner station, whereas at Althorpe I. and Cape Jervis cleaner stations were often not fixed, and cleaning took place opportunistically throughout the day. A third cleaner, the western cleaner clingfish, Cochleoceps bicolor, appears to clean mainly deeper water species. Communicative and contextual signals between client and cleaner are described. They include: guild signs by the cleaner; posing by the client, in which the body is inclined at various angles; and fixed cleaner stations defined variously by topography, or presence of sponges. At Althorpe I. zebra fishes that posed at an angle head up received more attention from cleaners than those that posed horizontally, and those that posed in small host groups received more attention than those in larger groups. Kry Worps: Fish cleaning; cleaning symbiosis; posing behaviour; fish behaviour; ectoparasites; southern Australia; cleaner; Tilodon sexfasciatus, Siphonognathus beddomei, Cochleoceps bicolor, Austrolabrus maculatus, Enoplosus armatus. Introduction Cleaning symbiosis among fishes is a common mutualism among tropical fishes in which cleaner fish benefit by feeding on ectoparasites on client species, which in turn benefit by their removal (Losey et al. 1999). Information on the similar temperate symbiosis is scarce, and in southern Australia is available only from the accounts of Hutchins (1979, 1991a) in southwestern Australia, and brief references elsewhere (e.g. Hutchins 1991b; Edgar 1997; Hutchins & Swainston 1999; Morrison & Storrie 1999; Kuiter 2000). However, studies of cleaning symbiosis in other temperate waters of the world are more numerous e.g. the northern and southern Atlantic (Henriques & Almada 1997; Zander et al. 1999; Sazima ef al. 2000), the Mediterranean (reviews of Moosleitner 1980 and Zander et al. 1999; Zander & Nieder 1997; Zander & Sétje 2002), the Black Sea (Darkov & Mochek 1980), the Gulf of California (McCourt & Thomson 1984), California (Hobson 1971), and New Zealand (Ayling & Grace 1971). ' Senior Research Fellow, South Australian Research and Development Institute, PO Box 120 Henley Beach, South Australia 5022. Email: shepherd.scoresby@saugov.sa.gov.au ? Plant Biodiversity Centre, Hackney Rd, Hackney, South Australia 5069. * 9 Giles St., Glenelg, South Australia 5045. The prevalence of such cleaning behaviour among labrids in particular has recently generated much interest in their use as a benign method of reducing parasite loads on fish farms (Young 1996). Ectoparasites are a problem in southern Australian fish farms, and observations on the symbiosis, and the cleaner and host species involved, will be valuable in assessing the merit of such biological control. In this paper we describe aspects of the ecology of cleaning behaviour from accumulated observations (a) at West I., South Australia (35° 37’ S, 138° 35' E) over 15 years, (b) during the 2004 expedition and a later visit to Althorpe I. (34° 22' S; 136° 52’ EB), (c) at Cape Jervis (35° 35’ S; 138° 6’ E), and (d) under Rapid Bay jetty (35° 32.8’ S; 138° 12.4’ E) with many casual observations elsewhere. Methods During long-term underwater studies of abalone at West I. (Shepherd 1998), visits were made there by SAS at 1-2 month intervals for 1-2 days from 1983- 2000, and less often after that. Fish cleaning behaviour was first observed in Dec. 1985 in Abalone Cove at 4 m depth, near an abalone monitoring site, at the junction of a vertical granite wall and a seagrass bed of Amphibolis antarctica. Thereafter, the cleaning station was monitored for evidence of the behaviour 194 S. A. SHEPHERD, J. TEALE & D. MUIRHEAD up to eight times per visit at different times of day until June 2001 when the study terminated. Concurrently with observations of the behaviour of the blue groper, Achoerodus gouldii (Shepherd & Brook 2005), and the moonlighter, Tilodon sexfasciatus, fish cleaning by the latter species was monitored in Mooring Bay, Althorpe I. near shore at a depth of 3 m among granitic blocks and sheet granite. The site was monitored several times per hour between 0900 and 1700 h from 2-11 February and 9-13 April, 2004 and from 7-11 February 2005. In 2005 fish cleaning was recorded mainly during concurrent time budget studies on the moonlighter. At Cape Jervis cleaning was observed at a sheltered site 250 m north of Cape Jervis lighthouse at about 3 m depth among reef of high relief with numerous caves, in which five moonlighters resided. At the Rapid Bay jetty site, throughout its >300 m length, the array of vertical piles and the mass of old piles, concrete cladding and debris strewn on the bottom, together formed a complex reef habitat where three cleaner species, the western cleaner clingfish, Cochleoceps bicolor, the black-spotted wrasse, Austrolabrus maculatus, and __ the moonlighter, 7? sexfasciatus, were common. Cleaning occurred at depths of 3 — 10 m at stations adjacent to piles, among debris or near sponge. For each cleaning event observed, data recorded were: (a) the specific identity, number and size of the host or client fish and cleaner fish present; (b) the duration of the behaviour, including anticipatory hovering or slow swimming in the vicinity; (c) pose of the client species; and (d) the number of bites observed for each fish. In 2004 and 2005 at Althorpe I. we noted more carefully the angle of pose of the client, and categorised poses as vertical (head up), inclined (with head up at ~ 40 — 60° from horizontal), horizontal, or nose down (at an angle of ~ 20 — 40° from horizontal). In the horizontal category it was not always clear whether a fish was a posing client, or a hovering onlooker, and in cases of doubt it was included in the latter category. From observations on the frequency of sightings of cleaning events at West [., an estimate of the expected number of cleaning events there on any day was derived as follows (see Dixon et al. 2005). Let x be the expected mean number of daily cleaning events during time period f, (in minutes). If s is the total number of sightings of fish cleaning on 7 inspections, and if the effective observation time by the observer watching the cleaning station is fj min. each during the period fp, then it can be shown that: s/n.ty X/tp Le. x =S8.tp/ nly The standard error of s, a binomial estimated quantity, is given by: V/s.(n-sy/n Hence, from this relation, and knowing the variance of t,, the standard error of x can be calculated. Note that for each observed cleaning event lasting e minutes, if the average diver observation time is 3 min. (see below) then ¢, = e+ 6, because a positive sighting will result if the diver observations commence up to 3 min. before and after time period e. Results Five cleaner fish species, one cleaner shrimp and 23 client species, are recorded from South Australia (Table 1). The posing position of the client fish varied within and between species, and also according to the cleaner species. Typically the client fish lay motionless, dorsal side uppermost, and with fins erect or extended. Some client species posed inclined with head upward at various angles (e.g. Fig. 5), while other species posed mainly horizontally (Figs 1-4). In the presence of a western cleaner clingfish or a pencil weed whiting, the client fish often opened its mouth and gills (Fig. 2). A summary of fish cleaning observations at various sites, together with data on the species involved, the duration of cleaning events, size of client groups and number of bites per fish, is given in Table 1, and considered in more detail below for the main sites investigated. West Island At West I. fish cleaning occurred at the same station at the junction of Amphibolis antarctica seagrass beds and a rock wall during the 15-year period, but was not observed elsewhere despite frequent searches. Normally a small school of zebra fish arrived at the cleaning station, between seagrass and a vertical wall, and hovered in the vicinity. Client fish variously posed horizontally, inclined or vertically, while others swam slowly or hovered nearby. Cleaners then emerged from the seagrass and swam about a host, searching for and apparently picking ectoparasites from its flanks. On four occasions other client species were present, in addition to the zebra fish (Table 1). A blue-throated wrasse occasionally paused nearby or was pursued by a cleaner, but cleaning was only observed once. On the five occasions when the inception of cleaning was observed, the client(s) initiated cleaning four times, by posing first and waiting for cleaners to emerge, and on one occasion the cleaner approached a hovering fish. All cleaning events were recorded between December and March between 1100 and 1500 h between 1-5 h (mean 2.4 h; CLEANING SYMBIOSES 195 Figs 1-6. 1. Blue devil, Paraplesiops meleagris, being cleaned on the left flank by a western cleaner clingfish, Cochleoceps bicolor. Seacliff Reef, 12 m depth. 2. Western blue groper, Achoerodus gouldii, posing for cleaning by a group of small, almost transparent, western cleaner clingfish, C. bicolor, on the client’s flank. Western River, Kangaroo I., 16 m depth. 3. Spotted boarfish, Paristiopterus gallipavo, posing for cleaning by western cleaner clingfish, C. bicolor, which is approaching the client near the mouth. Wirrina Reef, 22 m depth. 4. Harlequin fish, Othos dentex, being cleaned on the flank by a western cleaner clingfish, C. bicolor. Snug Cove, Kangaroo I., 14 m depth. 5. Magpie perch, Cheilodactylus nigripes, being cleaned on the flank by a western cleaner clingfish, C. bicolor. Note the angled pose of the client fish. Seacliff Reef, 12 m depth. 6. Rainbow cale, Odax acroptilus, posing for a pencil weed whiting, Siphonognathus beddomei, approaching from the rear. Note the bright eye-spot of the cleaner. Moana Reef, 10 m depth. All photos D. M. s.e. 0.4 h) after low water. In all there were 508 monitoring occasions distributed across all months of the year (mean number of occasions 42.3 (s.e. 4.8) per month), and fish cleaning was observed only on 16 occasions, between the months described, with sea temperatures >16° C. We tested for independence between the number of cleaning events and the cumulative number of monthly monitoring occasions for the 15-year period, and rejected the null hypothesis of non-independence (x? = 26.6; P<0.001) i.e. there was a highly significant 196 S. A. SHEPHERD, J. TEALE & D. MUIRHEAD TABLE 1. Cleaner and host species with their respective length range (L), mean client abundance (A) per cleaning event, location of cleaning, and duration, D. N = number of cleaning episodes observed. Standard errors in brackets. Nd = no data. All sites are in South Australia, except where indicated otherwise. Cleaner species Client species Pencil weed whiting Siphonognathus beddomei resident in seagrass Amphibolis antarctica. L= 5-8 cm A=1.6 (s.e. 0.1) Pencil weed whiting Siphonognathus beddomei resident in seagrass L=5-8 cm; A=1-2. Moonlighter Tilodon sexfasciatus resident in caves L= 10-20 cm A=1.2 (s.e. 0.1) Moonlighter T. sexfasciatus resident in caves L= 8-17 cm; A=1. Western cleaner clingfish Cochleoceps bicolor L= 3 cm; A=3. Western cleaner clingfish C. bicolor L= 3 cm; A=1. Resident among sponge, ascidians and rarely bryozoans Zebra fish Girella zebra L=15-30 cm; A=4.1 (s.e. 0.4) Magpie perch Cheilodactylus nigripes L=25 cm; A=1 Bluethroated wrasse Notolabrus tetricus L=20-25 cm; A=1 (1) Spiny-tailed leatherjacket Acanthaluteres brownii L=20 cm; A=1 Bluethroated wrasse N. tetricus L=15 cm; A=1 (2) Rainbow cale Odax acroptilus L=30 cm; A=1 (Fig. 6) (3) Senator wrasse, Pictilabrus laticlavius A= Maori wrasse Ophthalmolepis lineolata A=1 Rainbow cale Odax acroptilus A=1 Red mullet Upeneichthys vlamingii A=1 Zebra fish G. zebra L=20-35 cm; A=2.5 (s.e. 0.5) Magpie perch C. nigripes L=30-35 cm; A=1 Bluethroated wrasse N. tetricus L=35 cm; A=1 Silver drummer Kyphosus sydneyanus L=25 cm Blue groper Achoerodus gouldii L=80 cm Old wife Enoplosus armatus L=20 cm Moonlighter 7) sexfasciatum L=15 cm Zebra fish G. zebra L=20-25 cm; A=3.6 (s.e.0.7) Magpie perch C. nigripes L=22 cm; A=1 Blue groper A. gouldii L=27-35 cm; A=1 Dusky morwong Dactylophora nigricans L=45 cm; A=1 Blue groper A. gouldii L=30 cm; A=! (e.g. Fig. 2) Harlequin fish Othos dentex Magpie perch C. nigripes L=25-30 cm (Fig. 5) Zebra fish G. zebra L=30 cm; A=1 Harlequin fish O. dentex L=70 cm; A=I (Fig. 4) Blue groper A. gouldii L=30-80 cm; A=1 (Fig. 2) N Location and notes 16 Abalone Cove, West I. near vertical wall and seagrass D=3 Isec. (s.e. 3) Bites/fish 3.2 (s.e. 0.5) (1)Rapid Head, (2) Moana reef, (3) Second Valley. ((3) A.Hirst pers. comm.) Poses generally horizontal, but maori wrasse with inclined pose. Mooring Bay, Althorpe I. D=50 sec. (s.e. 9). Bites/fish — see Table 2. Poses of the zebra fish and magpie perch varied (see text), but poses of remaining species horizontal. Cape Jervis, 300 m north of lighthouse, 4 m deep, near rock wall and cave. D=53 sec. (s.e. 15) Bites/fish 1.5 (s.e. 0.2) Recherche Archipelago, W.A. Hutchins (199 1a,b) Morrison & Storrie (1999) Rapid Bay jetty, Western River, K.I., Snug Cove, K.I. Second Valley, Moana reef, Aldinga drop- off, Hallett Cove reef, Seacliff reef, Glenelg dredge site. Magpie perch and zebra fish occasionally posed inclined CLEANING SYMBIOSES 197 TABLE |. Cleaner and host species with their respective length range (L), mean client abundance (A) per cleaning event, location of cleaning, and duration, D. N = number of cleaning episodes observed. Standard errors in brackets. Nd = no data. All sites are in South Australia, except where indicated otherwise (cont.). Cleaner species Client species N Location and notes Dusky morwong D. nigricans L=50-80 cm; A=1 Boarfish Peniaceropsis recurvirostris l L=30 cm; A=1 NO upward. All other species posed horizontally. Overall D=30-60 sec. Spotted boarfish ] Paristiopterus gallipavo L=35 cm; A=I (Fig. 3) Red mullet U. vlamingii L=25 cm; A=1 ] Spiny tailed leatherjacket 1 Acanthaluteres brownii L=30 cm; A=1 Scaly fin Parma victoriae 1 L=15 cm; A=1 Blue devil Paraplesiops meleagris 3 L=25-30cm; A=1 (Fig. 1) Bullseye Pempheris multiradiata | L=15 cm; A=1 Bue groper A. gouldii 3 L=25- 35 cm; A=1 Red mullet U. vlamingii L=25cem; A=5 I Dusky morwong D. nigricans 1 L=33 cm; A=1 Black-spotted wrasse Austrolabrus maculatus Rapid Bay jetty. All horizontal poses, except zebra fish (this study and D. Cowan (pers. comm.)) Zebra fish G. zebra L=22 cm; A=4 1 Old wife E. armatus L=20 cm; A=3 Cleaner shrimp Periclimenes aesopius Pp. Blue groper A. gouldii L= 1.2 m; A=1 3 Leafy seadragon Phycodurus eques 1 Fam. Tetraodontidae 1 Anchorage Cove, Pearson I. (Shepherd 2005) Horizontal pose Victor Harbor Port Victoria R.H. Kuiter (pers.comm.) association between fish cleaning and the summer season. Next we tested for independence between the number of cleaning events and the cumulative number of monthly monitoring occasions during the above four summer months, and accepted the null hypothesis of non-independence (y? = 4.5; ns) i.e. the probability of cleaning occurring was approximately the same during each month from December to March. Next we examined the data for time of day at which cleaning took place. There were 230 monitoring occasions at the cleaning station between December and March, with a cumulative mean of 23.0 (s.e. 4.1) monitoring occasions for each hour from 0800 to 1800 h, with cleaning events observed only between 1100 and 1500 h. The null hypothesis of non-independence of cleaning events and the number of observations per hour throughout the day was rejected (x? = 29.4; P<0.001), and we concluded that fish cleaning was restricted to the period 1100 - 1500 h. Lastly we calculated from the above formula that the expected number of cleaning events per day from December to March inclusive was 3.2 (s.e. 0.2), on the basis that: (1) , the number of monitoring inspections, = 132, averaging 3 min. duration each (monitoring inspections were ~2 min. duration, except one in four of them, which was ~6 min. during quadrat monitoring nearby, thus averaging 3 min.); (2) each cleaning event, including pre-cleaning hovering, lasted 3.1 (s.e. 0.3) min., derived from five measured events, i.e. f7 = 9.1 (see above); and (3) both inspections and cleaning events occurred randomly during the period 1100-1500 h. Althorpe Island The behaviour of the cleaner moonlighter, Tilodon sexfasciatus, differed dramatically between 2004 and 2005 in Mooring Bay. In February 2004 three moonlighters sheltered in a cave at the cleaning station, and emerged from time to time to clean groups of client species, which gathered at the cave entrance at various times of day. Eight of the nine cleaning events observed took place during the period 1500-1800 h, 4-6 h after low tide (mean time 198 S. A. SHEPHERD, J. TEALE & D. MUIRHEAD of cleaning 1558 h) and one at 1045 h at about low tide. As previously, we tested for independence between monitoring observations and cleaning events, and rejected the null hypothesis (x? = 12.6; P <0.005) of non-independence. We concluded that cleaning behaviour strongly favoured mid- to late afternoon. In April 2004 no cleaning took place, despite frequent observations over a week, and some posing by zebra fish at the cleaning station. In 2005 several cleaner moonlighters were also present in the same area, but sheltered in other crevices nearby. These cleaners rarely cleaned at the cleaning station used in the previous year, but foraged individually or in pairs over a wide area, and were approached or followed by client fishes in small to large groups, seeking their attention. Cleaning took place wherever and whenever a moonlighter responded to a client, and was recorded throughout the day when observations took place (i.e. 0900 - 1700 h), and up to 5 h before and after low tide. Over the two years, 84% of the fishes cleaned, or posing for cleaning, were zebra fish, while six other species were observed posing on rare occasions (Table 1). The zebra fish variously adopted all poses, although the head down pose was rare, occurring on only 1.2% of occasions. The magpie perch mainly, and the silver drummer and blue groper always posed horizontally, while the moonlighter posed nose down, and the blue-throated wrasse mostly horizontally, but sometimes inclined head up. Sea sweep, Scorpis aequipinnis, occasionally posed, but were never observed being cleaned, while the horseshoe leatherjacket, Meuschenia hippocrepis, on occasions joined a group of client species, but never posed, and sometimes seemingly disrupted the event. Cleaners mainly searched the flanks of a client, and occasionally the pectoral fins, tail or head and mouth. On only two occasions (i.e. 2.8%) did we observe cheating, when the cleaner fish aggressively bit a fish, causing the client to jolt noticeably and flee. Next we examined whether attention from a cleaner depended on the size of the client group. The mean number of feeding bites per fish was significantly higher for group sizes of one and two combined, compared with larger groups (t = 3.13; P < 0.002) (Table 2). The difference in mean number of bites per fish between client groups of | - 2 fish and 3 or more fish was significant (t = 3.13; P < 0.002). Moreover the proportion of fish receiving cleaner attention declined significantly (x? = 30.6; P<0.001) with increasing size of client group, i.e. from 87.5% for single fish down to 25 — 45.6% for groups of 4 fish (Table 2), We then examined whether non- horizontal posing by a client resulted in more cleaning bites from a cleaner. Fish posing vertically received a mean of 2.5 (s.e. 0.3) bites per fish, those posing inclined upward received 2.0 (s.e. 0.2), those head down received 1.8 (s.e. 0.8) while the remainder posing or hovering horizontally received 0.5 (s.e. 0.3) bites per fish. The difference in mean bites per fish between those posing vertically and those inclined upward was not significant (t = 1.2; ns), but the difference between those posing horizontally and those non-horizontally was highly significant (t = 5.9; P<0.001). Those adopting a head-down pose were too few to test statistically. From Table 2 it is evident that in general 50 — 80% of fish posed vertically or at an angle for host group sizes | — 4, but in larger groups the incidence of posing non- horizontally was less. Cape Jervis Periodic visits were made to the cleaning station over 15 months (Table 1), but cleaning was observed only from December to February between 1300 and 1730 h and 2-5 h after low tide. Several moonlighters apparently resided in a cave, near the entrance to which fishes sometimes hovered and posed for cleaning. However, the cleaning station was very loosely defined, as at Althorpe I. in 2005, and cleaners often made feeding excursions away from TABLE 2. Althorpe I. 2004 and 2005. Mean number of cleaning bites per fish for different sized groups of host fishes (all species combined), mean percentage of fish in host group cleaned, and proportion of different posing types in each group size category. Group Size = number of host fishes present during a cleaning event, E = number of cleaning events, and N = total number of fish. Standard errors in brackets. For the four categories of pose, A = hovering vertically head up, Al = inclined head up, > = hovering horizontally, and Ni = inclined head downward. Group BE N Bites/ % Pose Size fish cleaned ” 4) > Ny 1 32 32 1.9 (0.3) 87.5 0.16 0.37 0.41 0.06 2 17 34 1.9 (0.3) 79.4 044 O41 0.12 0.03 3 4 12 1,3 (0.5) 58.3 0.17 0.25 0.58 0 4 5 20 0.7 (0.4) 25 0.35 0.35 0.30 0 25 6 57 1.3 (0.2) 45.6 0.23 0.19 0.54 = 0.04 CLEANING SYMBIOSES 199 the cave, during which they cleaned zebra fish and other species that sought attention. Rapid Bay jetty and elsewhere During the many visits to Rapid Bay jetty, cleaning by the black-spotted wrasse and western cleaner clingfish has been observed mainly from late morning to mid-afternoon. The western cleaner clingfish has been observed cleaning numerous species at many deeper reefs (10 — 20 m) in Gulf St Vincent and Investigator Strait where erect, stalked, caliculate (goblet-shaped) or bilamellate sponges (e.g. Callyspongia bilaminata), and large solitary ascidians (e.g. Ascidia sydneiensis and Herdmania momus) occur. The clingfish appeared to reside within the folds, lamellae or cup, as the case was, of the sponge, or near the ascidian, and establish a cleaning station close by. Cleaning by the clingfish has been observed in all months of the year, except July and August, but this could be due to fewer diver observations in winter months. Client species and sites are listed in Table 1. As yet we have no data on bite rates, due, in the case of the clingfish, to the difficulty of obtaining data in poor light conditions, and, in the case of the black-spotted wrasse, to the tendency of the wrasse to retreat in the presence of a diver. Discussion Most studies of cleaning in temperate regions note that in these waters cleaning behaviour is incidental, transient or occasional. Except for Hobson’s (1971) study which noted a critical temperature of 12 — 13 °C, below which cleaning was rare, this is the first study in temperate waters with evidence of seasonality in the behaviour, presumably due to the lower incidence of ectoparasitism in colder water (Ayling & Grace 1971; Cété & Molloy 2003; but cf Henriques & Almada 1997). At West I. the mean sea temperature range 1s 13.5 °C (winter) to 21.5 °C (summer) (Shepherd & Womersley 1970), but cleaning was observed only from December with temperatures >16 °C. However, if 16 °C were a temperature threshold sensu Hobson (1971), cleaning should occur until late May, when in fact it was not observed after March. Other factors, which inhibit cleaning, as noted by Hobson (1971), include turbidity and surge. The former is unlikely to have biased our results at West I., where water transparency was generally higher in winter than summer (Shepherd & Womersley 1970), but the effect of surge or swell is not known because monitoring observations were strongly biased toward days with low swell and better diving conditions. From evidence that gentle tactile stimuli are rewarding to fish (Losey 1974; Thresher 1977), Losey (1979) proposed that cleaning behaviour evolved independently in many parts of the world due to the tactile reward to the client. Cleaner precursors exploited the positive response by clients to such stimulation to gain ready access to a food supply. Communicative signals evolved to increase the probability of the response. These included: adoption of specific poses by the client, and striking colouration (a guild sign — see Eibl-Eiblesfeldt 1955) by the cleaner. Contextual signals also evolved, including cleaning stations, ectoparasite loads on the client, and prospective food for the cleaner. In this study some signals were present and others absent, as might be expected from a behaviour that is transient and occasional. Zebra fish, and occasionally magpie perch and blue-throated wrasse, but not other species, adopted a characteristic (non-horizontal) pose, while other species adopted a horizontal pose albeit with other indicia such as erect fins or opened gills. Guild signs, in the form of conspicuous transverse stripes, are conspicuous in the moonlighter, the western cleaner clingfish, and the old wife, but not the pencil weed whiting, which is a uniform green colour. However, another possible guild sign is the false eyespot, present in small moonlighters, old wives, black-spotted wrasse and female pencil weed whiting. The last species also has very conspicuous eyes (Fig. 6), which may serve the same function. In this study cleaner stations, when present, had well characterised topographies, close to rock outcrops or vertical walls, or specific sponge or ascidian habitats where cleaners variously resided. However, their existence, or persistence apparently depended on the idiosyncratic behaviour of the cleaner species. Some moonlighters had cleaner stations, whereas others did not. The pencil weed whiting and western cleaner clingfish seemed to have permanent stations. At Althorpe I. in 2004, cleaner moonlighters adopted a ‘stay-close-to-home’ behaviour, and had fixed cleaner stations, whereas in 2005 they regularly foraged up to 30 m from their shelter holes and cleaned opportunistically. Their reduced foraging range may have been induced by seal predators, which were commonly seen at the site in 2004 but not in 2005 (cf Connell 2002). Conspicuous topographic features characteristic of cleaning stations were noted by Darkovy & Mochev (1980) in the Black Sea, and by Moosleitner (1980) and Zander & Sétje (2002) in the Mediterranean, whereas Ayling & Grace (1971) and Hobson (1971) found that cleaning stations at fixed sites were rare, or, if present, loosely defined. Diurnal peaks of cleaning activity have been previously recorded by Johnson & Ruben (1988) and Sazima et al. (2000), who suggested that they corresponded with post-feeding periods of host 200 S. A. SHEPHERD, J. TEALE & D, MUIRHEAD species, when they had time to seek a cleaning station, At West I. schools of zebra fish often arrived at specific points in their ranges at about the same time of day. This may have been influenced by the tidal cycle as cleaning was always recorded on an incoming tide. We hypothesize that peaks of cleaning activity are influenced by the activity periods of the client. It is curious that at both West I. and Althorpe I. these temporal cleaning peaks occurred only at fixed cleaning stations, suggestive of a combined temporal/topographic cue. Cleaning a client’s pharynx and gills by entering the mouth and opercular openings has been observed only in the case of the western cleaner clingfish with several client species (e.g. blue groper, blue devil, dusky morwong, magpie perch, and harlequin fish). This is not surprising as this behaviour is only possible with a small cleaner and a relatively large fish (cf. Ayling & Grace 1971). The 23 client species identified in this study (Table 1) are all common in reef habitat, and are likely site- attached but with differing range sizes. If species with large ranges had access to more than one cleaner fish or station (termed choosy clients), then market theory predicts that they should have priority of access to a cleaner fish, compared with resident fish without such choice (Bshary & Noé 2003). At Althorpe I., Cape Jervis and West I., zebra fish seemed to include both residents (some were recognizable by markings) and roving groups, but we could not distinguish them with certainty. Hence we could not tell whether choosy clients claimed priority over residents. However, queuing for cleaner service occurred when larger client groups were present, and we never observed queue-jumping. The most common cleaner in shallow waters, the moonlighter, had a mean density of 4.3 (s.e. 0.9) per 2000 m? at 48 fish survey sites in southern Gulf St Vincent and Investigator Strait (SAS unpublished data), so potential cleaners are often spaced apart at a scale of tens to hundreds of metres on coastal reefs. Hence we hypothesize that these cleaners may be in relatively short supply at many shallow coastal reefs sites. Darkov & Mochek (1980) also believed that there was a general shortage of cleaners in their waters. The black-spotted wrasse (see Shepherd & Brook 2003) and the western cleaner clingfish are common deeper water species, but their abundance on offshore reefs is still largely unknown Casimir (1969), cited in Zander & Sétje (2002), suggested that the degree of inclination in the pose of a client species indicated its appetitive behaviour, which is greater with steeper slope, and Coté ef al. (1998) found that fish with an inclined posed enjoyed twice the chance of cleaner attention compared with non-posing fish. Others (McCourt & Thompson 1984; Zander ef al. 1999) have reported variable posing behaviour within a species. In this study the 2005 Althorpe I. data suggested a more complex picture for zebra fish. While an inclined pose overall attracted greater cleaner attention, the number of fishes in a client group during a cleaning event was also an important modifying factor. For small client group sizes of one or two, the probability of cleaner attention was higher, and hence the benefits of inclined posing for attention were less, and the incidence of such poses was relatively low. For client groups of five or more, only a minority adopted an inclined posed, of which 96% received cleaner attention, consistent with the hypothesis that the rewards for inclined posing were high. We hypothesize that inclined posing imposes an increased risk of predation, especially at sites frequented by seals, and hence inclined posing mainly occurs when there is competition for cleaner attention and the rewards of ectoparasite removal are substantial. Our data at other sites are consistent with this view, although of themselves inconclusive. The low frequency (~3 — 4 events per day) and the seasonality of cleaning, as recorded at West I., and possibly existing at Althorpe I. and Cape Jervis, indicate that in these temperate waters the cleaner species involved are facultative cleaners, obtaining only a fraction of their food from cleaning activities. Cleaner species in other temperate seas are also facultative (Hobson 1971; Ayling & Grace 1971; Zander & Sétje 2002), and, according to the last- named authors, there may be a geographic cline from tropical to boreal waters from professional to occasional cleaners. However, the western cleaner clingfish may be a full-time cleaner, as at one site gut contents were entirely parasitic crustaceans (J. B. Hutchins pers. comm.). Hobson (1971) suggested that cleaning stations developed mainly at sites where fish tended to aggregate, usually in places of high habitat diversity ie. with some vertical relief, an observation that accords with our experience. If cleaner species or cleaner stations prove to be locally scarce in southern Australia, and cleaning is a significant benefit to fish health as suggested by some authors (Losey ef al. 1999), then their presence may provide a strong argument for establishing reserves in their vicinity. Many of these and related questions, such as the ectoparasitic loads on temperate fish, fish cleaning by deeper water species, the effectiveness of cleaning (Grutter 1996), and the adaptive value of ectoparasite removal in temperate waters have yet to be answered. Acknowledgements We thank R. K. Lewis and A. C. Cheshire for promoting the expedition to Althorpe I., the crew of RV Ngerin for logistic support for SAS, John and CLEANING SYMBIOSES 201 Erika Lawley for their hospitality on Althorpe I., and Michael Clark, James Brook, and Renate Velzeboer for assistance in the field. Alastair Hirst, Sandra Leigh, Steve Reynolds, Rudie Kuiter and Barry Hutchins kindly provided data and/or advice, and Rick McGarvey gave statistical advice. Comments by Sue Murray-Jones, Paul Jennings and Sharon Drabsch improved the manuscript. References AYLING, A. M. & Grace, R. V. (1971) Cleaning symbiosis among New Zealand fishes. N. Z. J. Mar: Freshw. Res. 5, 205-218. BsHAry, R. & Noe, R. (2003) Biological markets: the ubiquitous influence of partner choice on the dynamics of cleaner fish — client reef fish interactions. /n Hammerstein, P. (Ed.) “Genetic and Cultural Evolution of Cooperation” pp. 168-184. (MIT Press, Cambridge, MA). CAsimiR, M. J. (1969) Zum Verhalten des Putzerfisches Svmphodus melanocercus (Risso). Z. Tierspsych. 26, 225-229, ConneLL, S. D, (2002) Effects of a predator and prey on a foraging reef fish: implications for understanding density-dependent growth. J Fish Biol. 60, 1551-1661. Core, 1. M. & Mo_toy, P. P. (2003) Temporal variation in cleaner fish and client behaviour: does it reflect ectoparasite availability? Ethology 109, 487-499. , ARNAL, C, & REYNOLDS, J. D. (1998) Variation in posing behaviour among fish species visiting cleaning stations. J. Fish Biol. 53 (Supplement A), 256-266. Darkoy, A. A. & Mocuek, A. D. (1980) Cleaning symbiosis in Black Sea fishes. J /chthvology 20, 152-155. Dixon, P. M., Ettison, A. M. & GoreLLi, N. J. (2005) Improving the precision of estimates of the frequency of rare events. Ecology 86, 1114-1123. Epaar, G. J. (1997) “Australian Marine Life: the Plants and Animals of Temperate Waters”. (Reed Books, Melbourne.) EIBL-EIBESFELDT, [. (1955) Uber Symbiosen, Parasitismus und andere besondere zwischenartliche Beziehungen tropischer Meeres-fische. Z. Tierspsych. 16, 19-25. Grutter, A. (1996) Parasite removal rates by the cleaner wrasse Labroides dimidiatus. Mar. Ecol. Prog. Ser. 130, 61-70. HENRIQUES, M. & ALMADA, V. C. (1977) Relative importance of cleaning behaviour in Centrolabrus exoletus and other wrasse at Arrabida, Portugal. J. mar. biol. Ass. U.K. 77, 891-898. Hosson, E. S, (1971) Cleaning symbiosis among California inshore fishes. Fish. Bull. 69, 491-523. Hutcuins, J. B. (1979) “The Fishes of Rottnest Island” (Creative Research, Perth), (199la) Southern Australia’s clingfishes, Aust. Nat. History 23, 627-633. (1991b) Description of three new species of gobiesocid fishes from southern Australia, with a key to the species of Cochleoceps. Rec. West Aust. Mus. 15, 655-672. & SWAINSTON, R. (1999) Sea Fishes of Southern Australia. 2nd Ed. (Swainston Publishing, Perth). JOHNSON, W. S. & RUBEN, P. (1988) Cleaning behavior of Bodianus rufus, Thalassoma_ bifasciatum, Gobiosoma evelynae and Periclimenes pederseni along a depth gradient at Salt River Submarine Canyon, St Croix. Envir, Biol. Fish. 23, 225-232. Kuirer, R. H. (2000) “Coastal fishes of South-eastern Australia”. (Gary Allen, Sydney). Losey, G. S. (1974) Cleaning symbiosis provides a positive reinforcer to fish. Science 184, 179-180. enigmatic (1979) Fish cleaning symbiosis: proximate causes of host behaviour. Anim. Behav. 27, 669-685. , GRUTTER, A. S., ROSENQUIST, G., MAHON, J. L. & ZAMZzow, J. P. (1999) Cleaning symbiosis: a review. pp. 379-395 In Almada, V. C., Oliveira, R. F. & Goncalves, E. J. (Eds) “Behaviour and conservation of littoral fishes”. (Instituto Superior de Psicologia Aplicada, Lisbon). McCourt, R. M. & THomson, D, A. (1984) Cleaning behavior of the juvenile Panamic Sergeant Major, Abudefduf troschelii (Gill), with a resumé of cleaning associations in the Gulf of California and adjacent waters. Calif: Fish and Game 70, 234-239, MOoosLeltner, V. H. (1980) Putzerfische und-garnelen im Mittelmeer (Cleanerfish and Cleanershrimps in the Mediterranean). Zool. Anz., Jena 205, 219-240. Morrison, S. & STORRIE, A. (1999) “Wonders of Western Waters: the Marine Life of South-Western Australia”. (Dept Conservation & Land Management, Perth). SAzIMA, 1., SAZIMA, C., FRANCINI-FILHO, R. B. & Moura, R. L. (2000) Daily cleaning activity and diversity of clients of the barber goby, E/acatinus figaro, on rocky reefs in southeastern Brazil. Env. Biol. Fish. 59, 69-77. SHEPHERD, S. A. (1998) Studies on southern Australian abalone (genus Haliotis) XIX. Long-term juvenile mortality dynamics, J. Shel/fish Res. 17, 813-825. (2005) Ontogenetic changes in diet, feeding behaviour and activity of the western blue groper, Achoerodus gouldii. Rec. W. Aust. Mus. (in press). & Brook, J. B. (2003) Encounter 2002 expedition to the Isles of St Francis, South Australia: reef fishes. Trans. R. Soc. S. Aust. 117, 269-279. & (2005) Foraging ecology of the western blue groper, Achoerodus gouldii, at the Althorpe Islands, South Australia. Trans. R. Soc. S. Aust. 129, 202-208. & Womers_ey, H. B. S. (1970) The sublittoral ecology of West Island, South Australia. 1. Environmental features and algal ecology. Trans. R. Soc. S. Aust. 94, 105-138. THompson, S. K. (1992) “Sampling”. (John Wiley & Sons Inc.: New York). THRESHER, R. (1977) Pseudo-cleaning behavior of Florida reef fishes. Copeia 1977, 768-769. Youna, C. M. (1996) Wrasse as cleaner fish. pp. 204-210. In Sayer, M. D. J., Treasurer, J. W. & Costello, M. W. (Eds) “Wrasse: Biology and Use in Aquaculture” (Fishing News Books Ltd, Oxford). ZANDER, C. D. & Nieper, J. (1997) Interspecific associations in Mediterranean fishes: feeding communities, cleaning symbiosis and cleaner mimics. Vie Milieu 47, 203-212. & SotiE, 1. (2002) Seasonal and geographical differences in cleaner fish activity in the Mediterranean Sea, Helgol. Mar: Res. 55, 232-241. . Meyer, U. & SCHMIDT, A. (1999). Cleaner fish symbiosis in European and Macronesian waters. /n: Almada, V. C., Oliviera, R. F & Gongalves, E. J. (Eds) “Behaviour and Conservation of Littoral Fishes’’, pp.397- 422. (Instituto Superior de Psicologia Aplicada: Lisboa). Transactions of the Royal Society of S. Aust. (2005), 129(2), 202-208. FORAGING ECOLOGY OF THE WESTERN BLUE GROPER, ACHOERODUS GOULDI, AT THE ALTHORPE ISLANDS, SOUTH AUSTRALIA by S. A. SHEPHERD’ & J. B. BROOK* Summary SHEPHERD, S. A. & BROOK, J. B. (2005) Foraging ecology of the western blue groper, Achoerodus gouldii, at the Althorpe Islands, South Australia. Trans. R. Soc. S. Aust. 129(2), 202-208, 30 November, 2005. The foraging behaviour of small to large western blue groper, Achoerodus gouldii, was examined at Althorpe I. Small fish foraged mainly by taking feeding bites in algal canopies, or small mussels from bare rock. With increasing size, fish switched to more efficient suction-bite feeding in which epifaunal aggregates were sucked into the mouth, and foraged increasingly on epifaunal aggregates within algal mats. Small and medium-sized fish tended to select foraging habitats with the highest densities of epifauna, whereas large fish, with an expanded home range, captured large prey as well as algal epifauna. Fish at all sizes were most active, in terms of feeding bite rate, in the morning and late afternoon, and least active in the early afternoon on an incoming tide. Small fish, 18-25 cm length, spent about half their time in shelter and foraged, swam or rested for the remainder. With increasing size, bite rates declined, the period of emergence increased, and the proportion of time when emergent spent swimming increased, while that spent resting decreased, and that spent foraging changed little. Key Worbs: Western blue groper; Achoerodus gouldii; foraging selectivity; foraging behaviour; feeding mode; time budget. Introduction The western blue groper (WBG), Achoerodus gouldii (Richardson, 1843), is the largest carnivorous fish on littoral rocky reefs in southern Australia, reaching a length of 1.7 m (Gomon ef al. 1994). The species ranges from the Abrolhos Is, Western Australia to Port Phillip Bay, Victoria (Gomon ef al. 1994), but is rare on rocky reefs east of Investigator Strait, South Australia (Shepherd & Brook'). Small WBG are found in sheltered, shallow reef areas, and with increasing size migrate to deeper water (Shepherd & Brook'). The species has been targeted by spear- and line-fishers, and its abundance declined seriously on mainland reefs of South Australia (Johnson 1982) until it was protected in the Gulf waters of S.A., and its capture outside the Gulfs made subject to size and boat limits. WBG is proposed by the Australian Society for Fish Biology (Crook 2001) for listing as Vulnerable (lower risk — conservation-dependent) in the IUCN red list category. Except for the study of Shepherd (2005) and earlier surveys (e.g. Shepherd & Brook'), the biology of WBG is unknown, although it appears similar in many respects to that of its allopatric congener, Achoerodus viridis, studied by Gillanders & Kingsford (1993, 1998), and Gillanders (1995a,b, *Senior Research Fellow, South Australian Research and Development Institute, PO Box 120 Henley Beach, South Australia 5022. Email: shepherd.scoresby@saugov.sa.gov.au ' PO Box 111, Normanville, South Australia 5204. ' Shepherd, S. A. & Brook, J. B. (2003) A survey of the western blue groper on southern Yorke Peninsula, Reefwatch Report. Conservation Council of South Australia. 12 pp. 1997, 1999). In that earlier study (Shepherd 2005), the diet and foraging behaviour of WBG was examined near Esperance and in the Investigator Group, in the western and eastern Great Australian Bight respectively. A common feeding mode was shown to be suction-bite feeding in algal mats in which the fish took epifaunal invertebrate prey non- selectively (cf. Gillanders 1995a). In this paper we further examine the pattern of foraging and feeding modes employed in different algal habitats by medium-sized to large WBG within the framework of optimal foraging theory. The study was carried out in Mooring Bay, Althorpe I. (34° 22’ S; 136° 52’ E), during the February 2004 expedition there (Murray-Jones & Shepherd 2005), and during a later visit from 9 — 13.4.04. Specifically we addressed the questions: (1) does foraging behaviour change with fish size; and (2) is choice of foraging habitats influenced by prey abundance and availability? Methods Site Description The eastern part of Mooring Bay, Althorpe I. has a smoothly sloping, granite substratum, intersected by several shallow gullies, overlain with blocks and boulders, and meeting sand at a depth of 3 m. The study site extended for 135 m alongshore with an average width of 35 m of rocky substratum, where we followed the foraging behaviour of seven resident medium-sized WBG, 18 — 45 cm TL, and two large WBG, 75-80 cm TL, which foraged within and beyond the study site. FORAGING ECOLOGY OF THE WESTERN BLUE GROPER 203 We recognised eight distinct habitats, defined by dominant algae or substratum features in the area, and estimated the proportional cover of each habitat with the line-intercept procedure. We set 100 m transect lines parallel to shore over the length of the study site, at 1.5 and 2.5 m depths and recorded the distance under the line covered by each habitat type. The transects covered the spatial extent of the range of small to medium-sized WBG (see below). The habitats recognised were: (1) a fucoid community, comprising mainly, in descending order of abundance, Cystophora monilifera, Ecklonia radiata, Dictyota sp., C. subfarcinata, C. moniliformis, and a sparse turf understorey of Caulerpa brownii, Haliptilon sp. and Pachydictyon paniculatum; (2) Caulerpa flexilis patches; (3) a coralline turf, comprising mainly Corallina officinalis and Jania parva; (4) a monospecific Liagora harveyana community; (5) a monospecific P paniculatum community; (6) cryptic habitat such as vertical faces, overhangs or caves with a low (5-30%) cover of Lobospira bicuspidata, Metagoniolithon stellifera, Haliptilon sp., and Gelidium sp.; (7) bare rock; and (8) sand. In this paper we use the term ‘turf’ to describe structurally homogeneous, algal tufts, 3 — 5 cm high, in habitats (2), (3) and (6). Sampling Eight samples (each 400 cm?) in habitat (1), and four in habitats (2) — (7), were taken by placing a 0.5 mm mesh net over the canopy/substratum, freeing the attached algae/epifauna with a chisel, and placing the contents in a plastic bag, which was then sealed. Samples were preserved, and later rinsed and weighed in the laboratory, algae identified, and animals > 0.5 mm enumerated in the following categories: amphipods (mostly gammarids); isopods (including tanaids, ostracods and fragments); shrimps (mainly Palaemonidae and Synalphidae); crabs; gastropods; _— bivalves; _—_— polychaetes; sipunculans; and ascidians. Activity budget To record the activity-time budget, four categories of activities were recognised within three size groups of WBG; 18 — 25 cm total length (TL), 30 — 45 cm TL, and adults 75 — 80 cm TL. The activities recognised (see Pavlov & Kasumyan 1998, Fulton & Bellwood 2002) were: (a) swimming above the algal canopy; (b) foraging, which included slow swimming close to the algae while inspecting the habitat, biting the substratum or epibiota, and winnowing rejected material; (c) resting near the bottom; and (d) interactions with other fish. After allowing an initial 5 — 10 min. in a day for acceptance time, a focal fish was then followed for 10 min., and the proportion of time spent in each activity recorded during three sampling periods, 0900 — 1200 h, 1200 — 1500 h, and 1500 — 1800 h. We counted the number of feeding bites in each habitat, and recorded whether bites were directed among algal fronds within the algal canopy, or on the substratum, (which included the base of the algal turf). Interactions with other fish were also recorded during the same periods. The data on feeding bite frequency were analysed by 2-way ANOVA for three periods of day and three size-classes of fish (after removing three 10 min. sample observations at random from small- and medium-sized fish classes in order to equalise replication). To test statistically the foraging habitat preferences of WBG we used Ivlev’s (1961) electivity coefficient, E, which measures the degree of selection of a particular habitat. The equation is: E= (7% —pj)/(r; + p;) where 7; is the proportion of feeding bites in habitat i relative to the total number of bites, and p; is the relative availability of that habitat, estimated by its proportional cover within the fishes’ home range. The index ranges from —1, for total avoidance, to +1, for maximum selection, and values around zero indicate random foraging. We subdivided the fucoid habitat into two components, the canopy and understorey, due to different foraging behaviour of fish in each. Correlations between rankings of E and epifaunal abundances were tested by the Spearman rank correlation coefficient, p. The number of emergent individuals, each identifiable by size, was counted on 17 occasions over 15 days at different times of day between 0900 and 1800 h. Each count was done by swimming twice throughout the sub-adult home range, in order to cover visually the whole area, and noting all emergent WBG. The method gives a reasonably precise estimate of the mean proportion of the daytime in which fish were active, and was used by Shepherd & Clarkson (2001) and Shepherd (2005). Home ranges were estimated by the polygon method i.e. by recording the maximum spatial extent alongshore and offshore of foraging by the small- and medium size-classes of WBG during the period of the study, totalling > 30 h of observations. Results Foraging, range and habitat preferences Small to medium-sized WBG are diurnal, benthic feeders, intermittently resting in shelter and emerging to feed. The smallest group (18 — 25 cm) had a home range estimated to be 80 x 25 m ie. 2000 m? in the most sheltered part of the study site, while the middle-sized group (30 — 45 cm) ranged 204 S. A. SHEPHERD & J. B. BROOK 0 20 40 Fish size (cm) 60 80 100 Fig. 1. Plot of the mean (+ SE) proportion, P, of feeding bites per 10-min sample directed at the substratum vs length (cm) of western blue groper, A. gouldii, for each fish sampled. The remaining proportion of bites was directed at the algal canopy. further into deeper and less sheltered waters over 120 x 35 mi.e. 4200 m?*. The adult WBG foraged over at least 15 000 m? in the Bay, and beyond its limits, but the limited visibility of 8 — 10 m prevented us estimating their full home range, which extended to >15 m depth. All WBG foraged solitarily in the various habitats by swimming slowly ~1 — 1.5 m above the sub- stratum, and periodically diving to the algal canopy, turf or substratum to take one or more feeding bites. Bites (or pecks) by small fish were directed at individual prey in foliose algae or in algal canopies, and with increasing size, proportionally more feeding bites were made on the substratum. Further, with increasing size fish more frequently employed suction-bites, in which they closed the expanded jaws over the algal turf close to the substratum and sucked prey into the mouth by rapid expansion of the buccal cavity, later ejecting the debris (see Shepherd 2005). As it was not always possible to detect whether fish took bites at single prey or used suction-bites, we used data on the precise location of bites (among algae or on the substratum), to estimate the shift in feeding mode. A plot of the mean proportion, P, of feeding bites on the substratum vs fish length, L, (Fig. 1) shows the shift in feeding location from bites among algal fronds to bites on the substratum. The curve of best fit was the logarithmic regression: =-1.92+ 0.68 InL (R*= 0.96; P < 0.001) Large fish (75 — 80 cm) foraged mainly by taking suction-bites among algal turfs or other algal habitats, and less often, by taking a gastropod or crab by aggressive bite or ram-and-bite in cryptic habitat (see Shepherd 2005). The proportional cover of the eight habitats in the range of small to large fish, the number of feeding bites in each habitat, and the proportion directed at the substratum/turf, are given in Table 1. Although large fish (75 — 80 cm TL) foraged in deeper water beyond the range of medium-sized fish (30 — 45 cm TL), we used the proportional cover values for the habitats in the home range of the latter group to calculate habitat availability for large fish, when foraging at the study site. None of the three size groups foraged in the various habitats in proportion to their availability within the fish’s range (x? = 18.4 for the smallest size group, x? = 152.2 for the middle- sized group, and x? =121.6 for large fish; in each case P<0.001). For adults, we add the caveat that only the shallow part of their range is considered here. Values for Ivlev’s electivity coefficient, E, for the three size groups of WBG (Table 2) indicate selected and avoided habitats. The two smallest groups, but not the large group, selected the coralline turf and Caulerpa flexilis habitats for foraging. Selection for the fucoid canopy decreased monotonically with increasing fish size, while selection for the substratum increased with size. The Pachydictyon and Liagora habitats were avoided, except by the smallest group, whereas strength of avoidance of FORAGING ECOLOGY OF THE WESTERN BLUE GROPER 205 TABLE 1. Proportional cover of eight habitats in the respective ranges of the 18 — 25 cm and 30 — 45 cm size groups of western blue groper, A. gouldii, and the total number of feeding bites recorded in each habitat. The proportion of feeding bites (P,) recorded in each habitat, and the mean proportion (P.) of those feeding bites directed at the substratum for the three size-classes of WBG are given for each habitat. Proportional cover of habitats for WBG (35 — 40 cm) applies also to large WBG (75 — 80 cm) while foraging at the study site. WBG (18 — 25 cm ) WBG (30 — 45 cm) WBG (75 — 80 cm) Habitat Cover Pi P, Cover Pi, P, Ph Pe Coralline turf 0.33 0.37 0.19 0.28 0.41 0.65 0.28 1.0 Fucoids-canopy 0.28 0.26 0 0.42 0.27 0 0.03 0 - substratum 0.02 l 0.11 I 0.44 1 Caulerpa flexilis 0.09 0.10 0.20 0.11 0.20 0.59 0.12 1.0 Pachydictyon 0.07 0.07 0.03 0.03 0.005 0 0.02 1.0 Liagora 0.11 0.14 0.06 0.08 0.002 0 0.02 1.0 Cryptic habitat 0.01 0 0 0.02 0.005 1.0 0.09 1.0 Bare rock 0.09 0.05 1 0.05 0 0 0 0 Sand 0.02 0 0 0.02 0 0 0 0 Total number of bites 478 606 129 TABLE 2. Values of Ivlev’s electivity coefficient, E, for three cryptic habitat declined with increasing size, until at size classes of western blue groper, A. gouldii, in the largest size WBG strongly selected that habitat. different foraging habitats in Mooring Bay, Althorpe I. Next we examined the epifaunal abundances in Negative values indicate avoidance, and positive values algal samples from various habitats, where WBG preference. foraged. The highest epifaunal densities were in the Fish Size fucoid, coralline and Pachydictyon habitats with Small Medium Large ~1800 — 3100 individuals per sample (Table 3). This Habitat 18-25 cm__ 30-45 cm__75-80 cm was due, in the case of the fucoid habitat, to the many Coralline turf 0.06 0.19 0 epiphytic mussels in the canopy, in the case of the Fucoid — canopy 0.17 0.07 -0.79 coralline turf, to the mass of spionid polychaetes on - substratum -0.90 -0.13 0.44 the substratum, and, in the case of the Pachydictvon Caulerpa flexilis 0.05 0.29 0.04 habitat, to the many tiny rissoid gastropods on the Pachydictyon 0 “0.71 -0.20 algal blades. Cryptic habitat and bare rock had the Liagora 0.12 -0.95 -0.60 lowest epifaunal densities. Cryptic habitat “I -0.60 0.64 To determine whether a relationship existed of rock = : E between epifaunal abundance within habitats and habitat selectivity by fish, we ranked habitats (a) by TABLE 3. Mean abundance of epifaunal groups (numbers 400 cm”) in seven different habitats in Mooring Bay, Althorpe I. Algal biomass is given in grams fresh weight 400 cm. Standard errors in brackets. Coralline Fucoid habitat Caulerpa Pachy- Liagora Cryptic Bare Species Turf Canopy Substrate —_flexilis dictyon Habitat Rock Gastropods 134 (22) - 405 (18) 10 (2) 14093 (346) 206 (53) 2 (1) - Bivalves 233 (68) 2900? (171) 65 (4) 150 (71) 59 (22) 1 (1) 4 (1) 54 (7) -Amphipods 210 (11) 125 (12) 57 (9) 25 (5) 242 (57) 39 (14) 68 (4) - Isopods - 5 (2) 24 (4) IL (2) 39 (20) 2: (1) 2 : Shrimps - : 8 (3) - 40 (6) ‘ - L Crabs - - 4 (1) - - - - - Polychaetes 1423' (685) 51 (8) 59 (9) 76 (43) - - 16 (3) - Sipunculans 18 (8) - - 2 (2) - - - - Ascidians - - 34 (5) - - - - - Holothurians - - 3 (1) 2 (1) - - - - Others I - - - - - - Total nos 2019 (782) 3081 (129) 659 (23) 276 (129) 1789 (274) 248 (67) 90 (6) 54 (7) Biomass 203 (8) 150 (11) 43 (6) 180 (10) 108 (8) 76 (4) 27 (6) <1 ‘mainly spionids. * mainly epiphytic mussels. 3 mainly rissoid gastropods. 206 S.A. SHEPHERD & J. B. BROOK Bite rate 0900-1200 1200-1500 Time of day 1500-1800 Fig. 2. Plot of feeding bite rate (numbers 10 min."') for three size classes of western blue groper, A. gouldii, during three time periods. increasing value of E for each size group, and (b) by increasing mean epifaunal density per habitat, using data in Table 3, for each size group of fish. The rankings were significantly correlated for small fish (p = 0.70, P<0.025) and for medium-sized fish (p = 0.72, P<0.025), but not for large fish (p = 0.27, ns). Hence, fish up to 45 cm TL tended to forage more in habitats with greater epifaunal densities, a trend not present in large fish. Time budget Mean feeding bite rates for each WBG size group were relatively high during the morning, low during the early afternoon, and high again late in the day (Fig. 2). Bite rates were highest for the small and middle-size groups, and lowest for the largest fish. The ANOVA analysis showed that bite rate differences were significant both for the fish size- classes (F = 10.2; P<0.01) and time of day (F = 5.25; P<0.01), with no interaction. We then repeated the analysis for the two smallest size-classes of fish, this time including all the data, and found no significant difference between bites rates of the small and medium size-classes (F = 4.07; P>0.05), although time of day remained significant (F = 5.14; P<0.01). In the adult group, the female bite rates were 75% greater than the male’s, but the difference was not significant (t = 1.1; ns). During the relevant period of observations, the average time of low tide was 0940 h, and of high tide 1545 h. Thus the maximum feeding rates occurred around the change of tide shortly after low and high water respectively. A time budget for the three size classes of WBG (Table 4) shows ontogenetic changes in time spent emergent, and on different activities when emergent. However, the time budget for adults is partial only, as TABLE 4. Activity-time budgets for three size groups of WBG, A. gouldii, showing the percent of daytime spent emergent, the allocation of time when emergent, to foraging, swimming, resting and social encounters in Mooring Bay. Standard errors in brackets. WBG WBG WBG (18-25 cm) (30-45 cm) (75-80 cm) No. of 10-min 24 24 15 samples % of total time 46.3 (6.1) 65.9 (2.6) 100 emergent % of time emergent spent: Foraging 49.7(9.2) 50.3 (4.0) 46.2 (7.1) Swimming 32.1(2.1) 41.2(1.5) 51.7 (7.3) Resting 18.1 (8.1) 8.3 (4.0) 0 Encounters: Intraspecific 0.02 0.08 1.4 Interspecific - Cleaning 0 0 0.7 - Aggression 0.04 0.18 - they also foraged outside the study site, for which no data could be obtained. The time in shelter, including time resting during foraging excursions, declined almost linearly with increasing size, and concomitantly, swimming and foraging time increased monotonically with increasing size. Intra- and inter- specific encounters (see Shepherd 2005; Shepherd ef al. 2005 for details) were low, indicative of a virtual absence of competition for food (Jones 1984). Discussion This study provides valuable information on the foraging ecology of a poorly known species near the FORAGING ECOLOGY OF THE WESTERN BLUE GROPER 207 eastern end of its range. An earlier study (Shepherd 2005) has shown that the diet of WBG switched from algal epifauna and small molluscs, in small- to medium-sized fish, to larger fauna, such as crabs, abalone and sea-urchins, in large fish. The feeding mode also changed from bites or pecks at single prey to the more efficient suction-bites on prey aggregates, a feeding mode which became the dominant method of feeding. The foraging data in this paper reinforce those conclusions, as we were able, by quantifying the precise location of bites (algal canopy or substratum), to estimate the rate of switching from single bites to the more efficient bite- suction feeding mode with increasing fish size. Feeding bites in the algal canopy or among algal fronds are likely directed at individual crustacean epifauna or attached molluscs, such as small mussels or rissoid gastropods, which live diffusely among foliose algal fronds. By comparison, bites on the substratum are directed at the resident epifaunal aggregates (Shepherd 2005), which live among the accumulated detritus, rhizomes and low algal turf (Fenwick 1976). Habitats with turfs or rhizomes forming a low mat e.g. coralline turfs, C. flexilis, and the substrate under fucoids, have their epifauna concentrated within 1 ~ 2 cm of the substrate, effectively making that habitat more profitable for suction-feeding (Shepherd 2005). But does epifaunal abundance in algal habitats wholly explain the selective foraging by WBG? While association does not logically entail preference (Underwood er al. 2004), the strong correlation between epifaunal abundance and E for small and medium-sized WBG does suggest, in the absence of other explanations, that differential prey densities in habitats may strongly influence fishes’ choice of foraging habitat. Wellenreuther & Connell (2002) also found for another suction-feeding species, Cheilodactylus nigripes (see Cappo 1995), a preference for feeding in algal mats with high prey densities. However, other factors, such as algal architecture, may also influence selection. Shepherd (2005) noted that very small WBG (5 — 10 cm TL) preferred to forage in algae with an open architecture and lower epifaunal abundances, compared with algae with dense fronds, and suggested that foraging efficiency favoured a habitat that was more easily searched visually for individual prey. This may explain some of the noise in the data, for example why the Caulerpa flexilis and Liagora habitats, with lower biomass and lower frond density compared with Pachydictvon, were favoured by some size groups of fish. A third factor, algal patch size, was also found by Wellenreuther & Connell (2002) to positively affect bite rates; however, that factor seems unlikely in this study in a between-habitat comparison. Patches of Caulerpa flexilis, coralline turf, and the fucoid understorey ranged in area from 0.5 — 5 m?, much smaller that the patches of Pachydictvon and Liagora, which were mostly >10 m2? (unpublished data). Yet selectivity values, E, were higher for the former than the latter group among the two largest WBG size groups which employed mainly suction feeding, suggesting that the influence of patch size per se, if anything, was negative. The changing pattern of habitat selection with increasing size of fish (Table 2) is best explained by the shift in feeding mode from single bites to suction-feeding, which is most efficient when directed at the substratum. The further shift by large fish toward capture of larger prey, such as sea urchins and abalone, with less reliance on suction feeding (see Shepherd 2005) explains the high selectivity value for cryptic habitat (Table 2). Omitting that habitat and also the fucoid canopy, in which suction feeding is ineffective, from a comparison of rankings of E and epifaunal abundances, there is a significant correlation (p = 0.714; P = 0.05) between the two sets of rankings; this suggests that, in habitats where suction feeding is employed by large fish, epifaunal abundances influence habitat selection. Overall, adult WBG appear to have a dual foraging strategy for exploiting large, but sparse and patchy, prey, and tiny, but variably abundant, prey. By foraging over a large range, it can exploit cryptic habitats for large prey (crabs, urchins and molluscs), and, at the same time, take many ‘popcorn snacks’ on algal epifauna in the ubiquitous algal habitats. In terms of time allocation to foraging, resting and swimming, we predict a dynamic trade-off, changing with size, between maximising time in shelter, with its adaptive value for predator avoidance and reducing energy expenditure, and maximising food intake (Hobson ef a/. 1981). Small fish are the most vulnerable to predation and large ones need the most food. This study exemplified these trade-offs, but showed some differences between WBG activity at Althorpe I., and at other sites studied by Shepherd (2005). At Althorpe L., bite rates were lowest for the 3 h period after midday on an incoming tide, whereas they were highest at Pearson and Ward Is for the same period also on an incoming tide (Shepherd 2005). It is possible that diurnal activity periods are contingent on the local ecology and unrelated to tidal movement. Overall, home range size, foraging preferences and time allocation of WBG can be broadly explained by fish size, feeding mode, epifaunal abundances and considerations of foraging efficiency. Thus, foraging range expands with increasing fish size and requirement for food, habitat selection depends on profitability, the feeding mode changes with fish size (and a corresponding larger mouth gape) to increase feeding efficiency, while swimming and foraging 208 S. A. SHEPHERD & J. B. BROOK time increase accordingly. However, foraging preferences are also dependent on a fish’s learning and memory systems, such as spatial memory of habitats, and memory for cues related to patch profitabiltity within habitats (Warburton 2003), both little known in benthic fishes. Optimal foraging theory is likely to provide at best a heuristic and simplified framework for studying fish foraging behaviour in a complex environment. Acknowledgements We thank Rob Lewis and Anthony Cheshire of the South Australian Research and Development Institute for their strong support for the expedition, and John and Erika Lawley and other friends of Althorpe I. for their support onshore. Dive officer, Michael Clark, provided boat and logistic support, and the crew of RV Ngerin manned diving dinghies. Tim Collins of Innes National Park provided a boat during the second visit. Renate Velzeboer and Sue Murray-Jones assisted in the field, Bob Baldock provided algal identifications, and Janine Baker gave much statistical assistance. Lastly we thank Simon Bryars, John Carragher, Sue Murray-Jones and Tony Fowler for helpful comments on the ms. References Cappo, M. (1995) The population biology of the temperate reef fish Cheilodactylus nigripes in an artificial reef environment, 7rans. R. Soc. S. Aust. 119, 113-122. Crook, D. (2001) Threatened Fishes Committee Report. Aust. Soc. Fish Biol. Newsletter 31 (1), 33-42. Fenwick, G. D. (1976) The effect of wave exposure on the amphipod fauna of the alga Caulerpa brownii. J. Exp. Mar. Biol. Ecol. 25, 1-18. Futton, C. J. & BeLLWwoop, D. R. (2002) Patterns of foraging in labrid fishes. Mar. Ecol. Prog. Ser. 226, 135- 142. GILLANDERS, B. M. (1995a) Reproductive biology of the protogynous hermaphrodite Achoerodus viridis (Labridae) from south-eastern Australia. Mar. Freshw. Res. 46, 999-1008. (1995b) Feeding ecology of the temperate marine fish Achoerodus viridis (Labridae): size, seasonal and site-specific differences. Mar. Freshw. Res. 46, 1009-1020. (1997) Patterns of abundance and size structure in the blue groper, Achoerodus viridis (Pisces, Labridae): evidence of links between estuaries and coastal reefs. Envy. Biol. Fish. 49, 153-173. (1999). Blue groper pp 188-193 /n Andrew, N. (Ed.) “Under Southern Seas: the Ecology of Australia’s Rocky Reefs”. (Univ. of N.S.W. Press, Sydney). & KincsForb, M. J. (1993) Abundance patterns of Achoerodus viridis (Pisces, Labridae) on estuarine and exposed rocky reefs: possible linkages pp 93-98 In Battershill, C. N., Schiel, D. R., Jones, G. P., Creese, R. G. & MacDiarmid, A. B. (Eds) “Proceedings of the 2nd International Temperate Reef Symposium 1992, Auckland, N.Z”. (NIWA Marine, Wellington). & (1998) Influence of habitat on abundance and size structure of a large temperate fish, Achoerodus viridis (Pisces: Labridae). Mar. Biol. 132, 503-514, Gomon, M. F,, Gover, C. J. M. & Kuirer, R. H. (1994) “The Fishes of Australia’s South Coast” (State Print: Adelaide). Hosson, E. S., MCFARLAND, W. N. & CuHess, J. R. (1981) Crepuscular and nocturnal activities of California near- shore fishes, with consideration of their scotopic, visual pigments, and the photic environment. Fish. Bull. U.S. 79, 1-30. IvLEY, V. S. (1961) “Experimental Ecology of the Feeding of Fishes”. (Yale University Press, New Haven, Connecticut). Jounson, J. E. (1982) Blue groper: situation report. S4FIC 6(10), 26-28. Jones, G. P. (1984) The influence of habitat and behavioural interactions on the local distribution of the wrasse, Pseudolabrus celidotus. Envir. Biol. Fishes 10, 43-58. Murray-Jones, S. & SHEPHERD, S. A. (2005) Expedition to Althorpe Islands, South Australia: introductory narrative and conservation recommendations. Trans. R. Soc. S. Aust. 129, 85-89. Paviov, D. S. & KAsuMYAN, A. O. (1998) The structure of the feeding behaviour of fishes. J Ichthyol. 38, 116-128. SHEPHERD, S. A. (2005) Ontogenetic changes in diet, feeding behaviour and activity of the western blue groper, Achoerodus gouldii. Rec. W. Aust. Mus. (in press). & CxLarkson, P. S. (2001) Diet, feeding behaviour, activity and predation of the blue-throated wrasse, Notolabrus tetricus. Mar. Freshw. Res. 52, 311- 322. , TEALE, J. & MUIRHEAD, D. (2005) Cleaning symbioses among inshore fishes at Althorpe I., South Australia, and elsewhere. Trans. R. Soc. S. Aust. 129, 193-201. Unperwoop, A. J., CHAPMAN, D. G. & Crowe, T. P. (2004) Identifying and understanding ecological preferences for habitat or prey. J. Exp. Mar. Biol, Ecol. 300, 161-187. Warsurton, K. (2003) Learning of foraging skills by fish. Fish and Fisheries 4, 203-215. WELLENREUTHER, M. & CONNELL, S. D, (2002) Response of predators to prey abundance: separating the effects of prey density and patch size, J Exp. Mar. Biol. Ecol. 273, 61-71. Transactions of the Royal Society of S. Aust. (2005), 129(2), 209-216. FORAGING FLIGHTS OF SHORT-TAILED SHEARWATERS (Puffinus tenuirostris) FROM ALTHORPE ISLAND: ASSESSING THEIR USE OF NERITIC WATERS by L. D. Einoper!? & S. D. GoLDsworTHy? Summary Etnoper, L. D. & GoLpswortuy, S. D. (2005) Foraging flights of short-tailed shearwaters (Puffinus tenuirostris) from Althorpe Island: assessing their use of neritic waters. Trans. R. Soc. S. Aust, 129(2), 209-216, 30 November, 2005. The at-sea movement and habitat use of the short-tailed shearwater, Puffinus tenuirostris, were examined usi ng satellite transmitters fixed dorsally to five parents provisioning chicks on Althorpe I., South Australia. Only foraging trips of short duration were targeted by this study, and a range of flight parameters including trip duration, foraging location, total distance covered, and maximum speed were recorded and analysed in order to determine the behaviour of individuals undertaking short duration foraging trips. All short trips were performed during the daytime, and lasted between 16 — 18 hours. During these trips birds foraged exclusively over neritic waters 35 — 70 km to the southwest of the colony. Variable wind conditions did not seem to influence either the direction of flight, or location of foraging activity. Most flight tracks were characterised by fast and direct outbound and return flight, with concentration of time spent (putative foraging activity) in a particular area, suggesting that birds were ‘travelling’ to a known foraging area. However, one track appeared less direct, as the bird frequently changed direction implying that ‘searching’ flight was being performed. Key Worbs: Short-tailed shearwater; Puffinus tenuirostris; satellite tracking: foraging; Althorpe Island; modes of flight. Introduction The short-tailed shearwater or mutton-bird, Puffinus tenuirostris, is a highly pelagic seabird, as it spends much of its life wandering the ocean in search of food. An individual may travel tens of thousands of kilometres each year during an annual migration from its breeding grounds along the southern coastline of Australia, to the northern Pacific Ocean (Marchant & Higgins 1990). During the austral spring, shearwaters return to their breeding grounds, and through summer and autumn they forage over much of the Southern Ocean and eastern Indian Ocean, from Australian neritic waters to the sea-ice edge of Antarctica (Johnstone & Kerry 1976; Nicholls e¢ al. 1998; Klomp & Schultz 2000). Much of the research on the short-tailed shearwater has been colony-based, and a number of long-term studies, running for up to 55 years (Bradley ef al. 1991), have provided an understanding of their population dynamics and breeding ecology. Knowledge of their at-sea behaviour, however, is very limited, due primarily to the many challenges of recording seabird activities away from the colony. As a result, little is known for this species about the location of their primary foraging grounds, or the relative importance of neritic and oceanic waters during the breeding season. Much of our knowledge ' School of Earth and Environmental Sciences, Adelaide University, Adelaide ? South Australian Research and Development Institute (Aquatic Sciences), PO Box 120, Henley Beach, South Australia 5022. of the at-sea movement and behaviour of short-tailed shearwaters is based on observations from research or fishing vessels (Wood 1993; Hunt et al. 1996; Ito 2002), or from the known distribution of prey items that are returned to the colony by parents (Weimerskirch & Cherel 1998; Connan ef al. 2005). Their potential foraging range has also been based on estimates of flight speed and their physiological capabilities (Weimerskirch & Cherel 1998; Baduini et al. 2001). However, it is now possible to monitor the real-time location of individuals of known sex, status and from a known colony as they travel at sea through the use of satellite telemetry (Klomp & Schultz 2000; Schultz & Klomp 2000a). In the only other satellite tracking study on this species during the breeding season, breeding adults were recorded undertaking flights to the productive waters of the Antarctic whilst provisioning their chicks on their island colonies in southern New South Wales (Klomp & Schultz 2000). These long duration trips (10 — 30 days) are common during the provisioning phase as adults alternate long trips with short (I — 3 day) foraging trips in a dual foraging strategy (Weimerskirch et a/. 1998; Schultz & Klomp 2000b). This alternating pattern of foraging is common among. pelagic seabird species (Weimerskirch et al. 1994; Weimerskirch 1998; Cartard et al. 2000; Baduini & Hyrenbach 2003; Ropert-Coudert et al. 2004), and is thought to represent a compromise between providing energy flow to chicks at the nest and the maintenance of parental body condition (Weimerskirch 1998; Schaffer et al. 2003), Energy transfer to chicks is 210 L. D, EINODER & S. D. GOLDSWORTHY most efficient following short trips, but at the expense of adult condition. To restore their condition, adults undertake longer ‘oceanic’ foraging trips, during which chicks may fast for extended periods (Wiemerskirch & Cherel 1998; Schultz & Klomp 2000b). The increased frequency of feeds associated with successive short trips gives chicks energy for growth and body maintenance during these fasting periods (Bradley et a/. 2000; Schultz & Klomp 2000b). Thus, despite the short-tailed shearwater’s exploitation of distant resources, waters adjacent to colonies also play an important role in breeding success. Colony-based studies can provide information on chick provisioning rates and the duration of parental foraging trips, but provide little insight into the distance travelled, path taken, or location of foraging areas, As a result we still have a_ limited understanding of the actual at-sea behaviour of provisioning adults, including their general movements, habitat use, and how their foraging behaviour relates and adjusts to varying physical and biological oceanographic conditions that may dictate the location and availability of prey. This study presents the findings of preliminary research into the at-sea behaviour of breeding short-tailed shearwaters from Althorpe I., South Australia (SA). Satellite transmitters were deployed on chick-provisioning adults, which were conducting one-day foraging trips in order to address the following questions: (1) What are the characteristics of short provisioning flights? (2) When conducting short flights do adults forage over neritic waters, or travel to nearby oceanic waters off the continental shelf? (3) Do adults return to specific areas to forage, or move randomly in search of food? Methods Site, burrow monitoring, and adult condition South Australia supports a population of about 2 million breeding short-tailed shearwaters distributed over 33 breeding colonies, all located on offshore islands (Copley 1993; Robinson ef al. 1996). Althorpe I. is 96 ha in size, with an estimated 22,420 breeding birds (Robinson et al. 1996). The study was carried out on Althorpe I. from 20 February — 13 March 2005. At this stage in the breeding season, chicks are 2 — 4 weeks old (Marchant & Higgins 1990). A total of 32 burrows containing good-sized chicks were monitored throughout the study period. Adults that were undertaking short foraging trips were targeted for tracking, and two methods were used to determine the likelihood of an adult undertaking a short trip: (1) assessing adult body condition upon capture; and (2) monitoring the attendance patterns of adults to chicks in the burrow. Adult body condition was assessed by weighing the bird to the nearest 5 g with a 200 g spring balance. This method was used as further evidence of the type of foraging trip that had just been completed, as it is thought that after multiple short trips parental body condition is low, but is increased following a long trip (Wiemerskirch & Cherel 1998). Adults weighing less than 600 g were deemed in low condition (Baduini ef al. 2001), more likely to initiate a long foraging trip (Wiemerskirch & Cherel 1998), and were not selected for tracking. The attendance patterns of adults were monitored by visiting burrows daily at ~1900 h when chicks were weighed to the nearest g with a 100 g spring balance. The visitation of a parent could be identified by an increase in chick weight between successive burrow inspections. Weight increases of over 10 g were attributed to the delivery of a meal. After monitoring the weight changes of chicks for a period of 10 days it was possible to identify those chicks, which had not yet received a feed. Parents of these chicks were deemed to have undertaken a long trip, given the alternating foraging pattern of this species, as trips are either 1 — 2 days, or 7 — 30 days duration (Weimerskirch & Cherel 1998; Klomp & Schultz 2000). At 10 of these burrows, trapdoors were installed at the entrance, with the purpose of retaining the adult in the burrow (Weimerskirch 1998). Sticks were then placed at the entrance in order to keep the swinging door open, and so that a visit by an adult could be detected by the displacement of the stick, and closure of the door, When closed, the swinging door still allowed the entrance of other birds, but prevented birds from leaving the burrow. After the installation of the trapdoors, burrow inspections were done every hour during the night. If a visit was detected, the adult was caught when chick begging had ceased, i.e. when feeding had stopped. Tracker attachment and analysis of foraging tracks We used four 30 g KiwiSat 202 Satellite transmitters (single AA Cell, Sirtrack Ltd, North Havelock, New Zealand) to monitor the movements of five short-tailed shearwaters, each weighing 600 — 700 g with a wingspan of 95 — 100 cm. Similar sized satellite transmitters have been used successfully on this species (Nicholls et al. 1998; Klomp & Schultz 2000; Schultz and Klomp 2000a). Transmitters were glued to the back feathers (Nicholls e¢ a/. 1998; Cartard et al. 2000; Klomp & Schultz 2000; Schultz & Klomp 2000a), using Loctite 401 (Intek Adhesives Ltd, Northumberland, England), and upon the birds’ return and capture at the colony, trackers were removed. The movement of shearwaters at sea was monitored by Service Argos Inc. (Toulouse, France), which uses two NOAA FORAGING OF SHORT-TAILED SHEARWATERS 211 R v Eyre Peninsul * Yorke Peninsula Althorpe Is Kangaroo Is 0 25 50 kilometers Fig. |. The combined at-sea movements of seven foraging trips showing the exclusive foraging activity to the south west of Althorpe I., over neritic waters. The isobaths of the continental shelf at 200 — 1000 m are shown. (National Oceanic and Atmospheric Administration, USA) satellites to receive a signal sent by transmitters. The data stored on the satellites are relayed to CNES (Centre National des Etudes Spatiales) in Toulouse, France. The units had a transmission interval of 60 s. Location data from each foraging trip was analysed to determine the: (1) location; (2) distance flown; (3) maximum distance from colony; and (4) average travel speeds (ground speed) between fixed locations. Tracking data were analysed with the program ‘TimeTrack’ (version |.0- 9, M. D. Sumner, University of Tasmania, Hobart) which filters the data by applying the filter described by McConnell (1992) based on a maximum travel speed of 60 km/h. We interpolated a point every 15 minutes (time) between satellite fixes, after assuming straight-line travel between locations, and a constant travel speed. Distance flown was calculated as the total cumulative distance between all positional fixes along the foraging track. For each location point the straight-line distance from the colony was calculated and expressed as a proportion of the maximum distance attained during the foraging trip. This revealed the outbound and inbound component of each trip. The relationship between the distance from the colony and average horizontal speed (km h') was analysed using regression (Spearman rank) to compare the rates of travel during the departure and return flights from a foraging area. Travelling speed between each point was calculated, in order to assess changes in flight speed during a trip and to infer two different modes of activity. ‘Travelling’ was inferred from high speed flight and ‘foraging’ from slow speed flight. This is based on the rationale that birds actively exploiting a prey patch would spend more time in a certain area than when commuting between feeding patches (Gremillet et al. 2004). L. D. EINODER & S. D. GOLDSWORTHY 212 J aon 8 J ae A Neal B Qe 4 eee oo ' ees Is t ? / va A gp He fo 2 f | at ee _— % LA § 8 1 a i a 10 20 Track 1 f Tometers Track 2 fa aomelsrs i y = 7 J Pare 8 ¥ FA rate & & ue \ Ln \ m Cc Kt D ee Q Pp @ / ~» Althorpe I Althorpe Is ys i ithorpe Is at Ry + a / ¢ ‘ 4 2 ¢ . y a4 ¢ ieee 7 wy ari oe — js 8 a 20 { a 19 20 Track 3 f ~~ Kilometers Track 4 et ~ filometers g d J a ms As a 3 ee \ iN f E wer F Soe 2 ae Alth if bate “sg Althorpe Is q i oF my tas \ ie av — as / \ ‘ gq B \ & Ben we - * 4 Se = a ? a 724 ~~ Be out A L § S8by ——_—— BE ed Pe me wg BF pee Y ot a ars f risen ; a 10 20 Track 5 € ake Track 6 ra kilometers Fig. 2. A-F The at-sea distributions of six flight paths of short-tailed shearwaters during the study period. Points represent the output of the time-track program giving a location every 15 minutes (ie both the satellite fixes and the predicted locations). The location of the Cape Borda weather station is shown by a star. TABLE 1. Flight parameters from all six foraging tracks (* the same individual tracked undertaking successive trips). Max. distance Total Average Max. Trip Trip time from colony distance bearing speed no. Date period (hrs) Duration (km) (km) (degrees) (km hr!) ‘? 1 25.2.05 0540-2252 17hrs12mins 70 198 232 46 is 2 26.2.05 0516-2240 17hrs 24mins 54 167 229 88.7 3 26.2.05 0410-2209 17hrs 59mins 45 143 232 53 + 27.2.05 0356-2033 16hrs 37mins 32 130 221 44.6 5 28.2.05 0545-2154 16hrs 09mins 70 312 215 28.5 6 1.3.05 0441-2100 16hrs 19mins 37 133 240 46 FORAGING OF SHORT-TAILED SHEARWATERS 213 Wind conditions Information on wind direction and wind strength during the study period was obtained from the Australian Bureau of Meteorology website for the weather station at Cape Borda, Kangaroo I., 48 km SSW of Althorpe I. Results Foraging movements Four one-day foraging trips were monitored from four individuals, as well as two consecutive one-day foraging trips from a fifth individual during the study (Table 1). Adults were away from the colony between 16 — 18 h duration. Adults began their foraging trip either during the main departure phase from the colony between first light (~0500 h) and sunrise (0605 — 0609 h during the study period), or slightly earlier at ~0400 h, as fewer adults chose to depart before first light. All birds travelled over neritic waters to the SW of Althorpe I. (Table 1; Figs 1, 2A- F). The maximum distance from the colony ranged from 32 — 70 km, and the total distance travelled during foraging trips ranged between 130 — 312 km (Table 1). The proportion of maximum distance was negatively correlated with the average horizontal speed on both the outbound (Spearman’s rho: r = 0.563, P = 0.01) and return (Spearman’s rho: r = 0.172, P = 0.05) foraging trips, indicating that shearwaters travelled more rapidly when they were close to the colony, compared with when they were close to their maximum distance (Fig. 3A). This result is due to the similar foraging behaviour performed during five of the six tracks, involving ‘travelling’ movement to an area followed by a period of ‘foraging’, before returning to the colony (Figs 2A-D, and F, and Fig. 3B). In contrast, Track 5 showed a different pattern of movement, undergoing many changes in direction (Fig. 2E), more variable flight speeds throughout the trip duration (Fig. 3B), reaching maximum distance towards the end of the trip (Fig. 3A), and covering a greater total distance (Table 1). These features are indicative of ‘searching’ flight (Arnould e¢ al. 1996; Weimerskirch ef al. 1997), and suggest that the individual was actively searching for a suitable patch of prey. The location of this individual’s movements also differed from that of the four other individuals involved in the study. The short period of ‘foraging’ flight was located close to the coastline of Kangaroo I., whereas the other birds foraged further from the Kangaroo I. coast, a few km to the north and west. Relationship with wind conditions The wind strength was moderate to strong during the study period, varying from 11 — 33 km hr! (Table A Distance from colony [%] Se 0-10 10-20 20-30 30-40 40-50 50-60 60-70 70-80 80-90 90-100 Stage of foraging trip (+) Flight speed ({kmfh]} 0-10 10-20 20-30 30-40 40-50 50-60 60-70 70-80 80-90 90-100 Stage of foraging trip (+) Fig. 3, Relationship between (A) the average distance from Althorpe I., as a proportion of maximum distance (%), and (B) the average horizontal speed (km hr) travelled between 15 minute points, at different stages of the foraging trip (% duration). Vertical bars are standard errors. Solid circles show the mean for Trips 1, 2, 3, 4, and 6 and open circles represent Trip 5. TABLE 2. Wind direction and speed (km Ir!) at Cape Borda, Kangaroo 1. Date Time Direction Speed 25.02.05 900 SE 11 1500 SW 31 26.02.05 900 ESE 15 1500 N 19 27.02.05 900 NE 24 1500 NNE 17 28.02.05 900 NE 19 1500 WSW 13 1.03.05 900 S 30 1500 S 33 2). Wind direction was variable during the five consecutive days of the study. The fact that all foraging trips were directed to the SW of Althorpe I. indicates that neither the strength nor direction of the wind had a measured influence on the flight behaviour of tracked birds. 214 L. D. EINODER & S. D. GOLDSWORTHY Effect of transmitters on shearwaters The duration of the foraging trips of tracked individuals, as well as their successful provisioning of chicks upon return to the colony, suggest that the attachment of trackers had minimal adverse effects on foraging behaviour. In addition, tracked birds departed and returned to the colony with the main movement of the rest of the colony. Discussion Habitat utilisation These results represent the most detailed tracking data for the short duration foraging trips of this species, and provide an insight into their foraging ecology over the neritic waters surrounding their colony. This study revealed that adults concentrated their foraging efforts over a body of water 30 — 70 km SW of the colony in the week of the study. A number of studies on other pelagic seabirds have revealed that tracking data can be used to draw inferences as to the location and predictability of patches of prey in the marine environment (Anderson et al. 1997; Weimerskirch et al. 1997; Ropert- Coudert et al. 2004). The concentration of foraging effort by five individuals to the SW of the colony suggests that adults were not foraging randomly from the colony. Individuals may be re-visiting a known prey patch exploited the day before, either on their previous short trip, or on their return from a long trip. The bird that was tracked undergoing two short trips to the same area exemplified this notion of an individual returning to an area of successful foraging over successive day trips. The flight capabilities of the short-tailed shearwater (Klomp & Schultz 2000; Baduini er al. 2001) reveal that provisioning adults are quite capable of conducting a 200 km round trip to the edge of the continental shelf, or beyond, in the 16-18 hours they are absent from their colony. Nevertheless, this study suggests that during foraging trips of short duration they are closely linked to shelf areas of less that 100 m depth. Only one other tracking study has recorded the short trip movements of this species during the chick- provisioning phase (Schultz & Klomp 2000b). That study, conducted of a colony in NSW, recorded two successive day trips of a single breeding adult, and revealed that foraging was restricted to the strip of neritic water south of the colony. The foraging parameters including flight speed and distance covered were comparable to this study. Fisheries interactions The results of this study show that breeding adults from Althorpe L. concentrated their foraging effort in an area close to that utilised by the SA pilchard (Sardinops sagax) fishery (Rogers et al. 2004). The role of pilchards in the diet of the shearwater population in SA is unclear because no dietary studies have been conducted in this region. However, shearwaters are known to feed on small pelagic fish species in other parts of their breeding range to the east (Montague et al. 1986; Skira 1986; Weimerskirch & Cherel 1998). Foraging location and dietary data for short-tailed shearwaters are needed in order to determine the degree of spatial overlap and competition between shearwaters and commercial fishers. Distribution of food resources Many seabird species have proven to be a useful tool in monitoring the marine environment, because fluctuations in seabird population, breeding, or nutritional parameters are commonly due to perturbations lower in the food chain (Cairns 1992; Croxall et al. 1988). Foraging parameters are particularly sensitive to changes in prey availability (Cairns 1992; Montevecchi 1993), and thus can.be of value as indicators of changes in the quality, abundance or availability of prey. The link between shearwater foraging behaviour over neritic waters of Australia, and the availability of marine food resources is not well understood. Despite a small sample size, the area of concentrated foraging activity to the SW of Althorpe I. suggests that this body of water supported adequately profitable prey patches during the weeklong study period. The replication of these tracking methods over a number of successive weeks or months during the chick- provisioning phase would enable further assessment of the dynamics of prey availability over neritic waters surrounding Althorpe I. Oceanographic features There is growing evidence that marine birds do not randomly sample the environment for prey, but focus their search where prey can be located predictably. Spatially and temporally predictable feeding sites, such as those associated with upwelling, are of importance to breeding Procellariiformes that depend upon the availability of prey near their colonies in order to raise young (Ainley ef al. 1998). A series of regional coastal upwelling centres occur along the SA coastline, driven by SE winds in summer (Kampf ef al. 2004); however, the spatial and temporal variation in upwelling intensity in the eastern Great Australian Bight has only recently received attention, and the reliance of many local apex predator populations on the productivity associated with these areas is little known (Gill 2002). The shearwater population on Althorpe I. is well located to exploit the marine productivity associated with these events, as a site of strong FORAGING OF SHORT-TAILED SHEARWATERS 215 upwelling is located along the western coastline of Kangaroo I., between 60 — 90 km away (McClatchie et al. 2005). The concentration of foraging effort to the SW of the colony reported in this study may have been in response to the increased marine productivity associated with upwelling activity in that area; however, the strength of this relationship is reliant upon an assessment of the wind conditions, sea surface temperatures, and data on the primary and secondary productivity in the area during the week of the study. Future research into this topic requires the co-ordination of oceanographic studies in the area, with tracking and dietary studies of this species, on the same spatial and temporal scales. Conclusions This study provides the first information on the at- sea behaviour of shearwaters when undertaking short foraging trips over neritic waters in the eastern Great Australian Bight. Due to the co-ordinated departure and return of tracked birds with the rest of the colony, and their successful provisioning of chicks, it appears that 30 g satellite tracking devices are a feasible method of monitoring the normal at-sea behaviour of provisioning adults. Most tracked foraging flights clearly revealed two modes of travel: direct ‘travelling’ flight to a particular area, and slower flight associated with ‘foraging’ activity. This enabled the identification of a concentrated foraging area over neritic waters to the SW of the colony. The consistencies in flight direction from the colony suggest that birds do not forage randomly, and that foraging flights are undertaken towards a specific, possibly previously known, area. As the short-tailed shearwater is one of the most abundant pelagic seabirds over the neritic waters of SA in summer, they must consume a large amount of the area’s marine resources. Of primary concern is their interaction with the harvest of pilchards (Sardinops sagax) by commercial fishers. Such competition is of growing interest to environmental and fisheries’ managers in many parts of the world (Tasker et al. 2000; Goldsworthy ef al. 2001; Furness 2003). As fishing effort is concentrated 50-100 km west of Althorpe I. (Rogers et al. 2004) and may expand to other areas in future years, the potential exists for competition between shearwaters and commercial fishers. Satellite tracking and dietary work is pivotal to achieving an understanding of such competition. Acknowledgements We thank those who assisted in the collection of data whilst in the field, including Brad Page, James Thiessen, Sarah Pennington, Deb Fraser, and Jon Pipitone. To John Lawley for a guided tour of Althorpe I., and for his interest and enthusiasm in the research project. Thanks also to Tim Collins and the staff of Innes National Park for the loan of equipment, and assistance in transport when in the field. The program ‘Timetrack’ was designed by Mike Sumner, of the University of Tasmania, and kindly provided for the analysis and filtering of the tracking data. We also thank referees and editors for improvements to the manuscript. References AINLEY, D. G., JAcoss, S. S., Ripic, C. A., & GAFFNET, I. (1998) Seabird distribution and oceanic features of the Amundsen and southern Bellingshausen seas. Antar. Sci. 10(2), 111-123. ANDERSON, D. J., SHWANDT, A. J. & DOUGLAS, H. D. (1997) Foraging ranges of Waved Albatross in the eastern tropical Pacific Ocean. Jn Robertson, G. & Gales, R. (Eds) “Albatross Biology and Conservation” pp.180-185 (Surrey Beatty and Sons, Chipping Norton). ARNOULD, J. P. Y., BRiGGs, D. R., CROXALL, J. P., Prince, P. A. & Woop, A. G. (1996) The foraging behaviour and energetics of wandering albatrosses brooding chicks. Antar. Sci. 8 (3), 229-236. BAbuINI, C. L. AND HYRENBACH, K.D. (2003) Biogeography of procellariform foraging strategies: does ocean productivity influence provisioning? Mar. Ornith. 31, 101-112. BabuINI, C. L., Lovvorn, J. R. & Hunt Jr., G. L. (2001) Determining the body condition of short-tailed shearwaters: implications for migratory flight ranges and starvation events. Mar. Ecol. Prog. Ser, 222, 265-277. BRADLEY, J. S., SKIRA, I. J. & WooLter, R. D. (1991) A long-term study of the short-tailed shearwater Puffinus tenuirostris on Fisher Island, Australia. /bis 133 (Suppl.1), 55-61. , Cox, J. M., NicHOLson, L. W., WooLLer, R. D., Hamer, K. C., & HILL, J. K. (2000) Parental influence upon the provisioning schedules of nesting short-tailed shearwaters Puffinus tenuirostris. J. Av. Biol. 31, 522- Cairns, D. K. (1992) Bridging the gap between ornithology and fisheries science: use of seabird data in stock assessment models. The Cond. 94, 811-824. CARTARD, A., WEIMERSKIRCH, H. & CHEREL, Y. (2000) Exploitation of distant Antarctic waters and close shelf- break waters by white-chinned petrels rearing chicks. Mar. Ecol. Prog. Ser. 194, 249-261. CHEREL, Y., & WEIMERSKIRCH, H. (1995) Seabirds as indicators of marine resources: black-browed albatrosses feeding on ommastrephid squids in Kerguelen waters. Mar. Ecol. Prog. Ser. 129, 295-300. 216 L. D. EINODER & S. D. GOLDSWORTHY ConnaN, M., MAyzAub, P., BoUTOUTE, M., WEIMERSKIRCH, W., & CHEREL, Y. (2005) Lipid composition of stomach oil in a procellariform seabird Puffinus tenuirostris: implications for food web studies. Mar. Ecol. Prog. Ser. 290, 277-290. Cop.ey, P. B. (1993) The status of seabirds in South Australia. Jn Ross, A. J. B., Weaver, K., and Greig, J. C. (Eds) “Proceedings of the National Seabird Workshop” pp. 139-179. (Biodiversity Group, Environment Australia, Canberra). CROXALL, J. P., MCCANN, T. S., PRINCE, P. A. & ROTHERY, P. (1988) Reproductive performance of seabirds and seals at South Georgia and Signy Island, South Orkney Islands, 1976-1987: implications for Southern Ocean monitoring studies. Jn Sahrhage, D. (Ed.) “Antarctic Ocean and Resource Variability” pp. 261-285. (Springer, Berlin), Furness, R. W. (2003). Impacts of fisheries on seabird communities. Scientia Marina 67 (Suppl.2): 33-45. Gill, B.C. (2002) A blue whale (Balaenopterus musculus) feeding ground in a southern Australian coastal upwelling zone. J. Cet. Res. Mgmt 4(2), 179-184. GoLpsworthy, S. D., He, X., Tuck, G. N., Lewis, M. & WILLIAMS, R. (2001) Trophic interactions between the Patagonian toothfish, its fishery, and seals and seabirds around Macquarie Island. Mar. Ecol. Prog. Ser. 218, 283- 302. GREMILLET, D., DELUOmo, G., RYAN, P. G., PETERS, G., Ropert-Coupert, Y., & WEEKS, S. J. (2004) Offshore diplomacy, or how seabirds mitigate intra-specific competition: a case study based on GPS tracking of Cape gannets from neighbouring colonies. Mar. Ecol. Prog. Ser, 268, 265-279 Hunt, G. L., CoyLe, K. O., HOFFMAN, S., DECKER, M. & Fuint, E, N. (1996) Foraging ecology of short-tailed shearwaters near the Pribilof islands, Bering Sea. Mar. Ecol. Prog. Ser, 141, 1-11. Iro, S. (2002) Foraging areas of the short-tailed shearwater during their northward migration along the Pacific coast of northern Japan. Ornith. Sci. 1, 159-162. JOHNSTONE, G. W. A., & Kerry. K. R. (1976) Ornithological observations in the Australian section of the Southern Ocean. Proc. XVI .Int. Ornith. Congr. 5, 725-736. KAmpr, J., DoUBELL, M., GRIFFIN, D., MATTHEWS, R. L., & Warob, T. M. (2004) Evidence of a large seasonal coastal upwelling system along the southern shelf of Australia. Geoph. Res. Let. 31(L09310):doi:10.1029/2003GLO0 19221. Ktomp, N. L, & Scnuttz, M. A. (2000) Short-tailed shearwaters breeding in Australia forage in Antarctic waters, Mar. Ecol. Prog. Ser. 194, 307-310. Marcuant, S. A, & Hicains. P. J. (1990) “Handbook of Australian, New Zealand and Antarctic birds” (Oxford University Press, Melbourne). McCratcuie, S., MIDDLETON, J.F. & Warp, T.M. (2005) Water mass analysis and alongshore variation in upwelling intensity in the eastern Great Australian Bight. J.Geophys Res. (in press) McConneELL, B. J., CHAMBERS, C., NICHOLAS, K. S. & FepAK, M. A. (1992) Satellite tracking of grey seals (Halichoerus grypus). J. Zool, (London) 226, 271-282 MONTAGUE, T. L., CULLEN, J. M. & FITZHERBERT, K. (1986) The diet of the short-tailed shearwater Puffinus tenuirostris during its breeding season. Emu 86, 207-213. MonTEVECCHI, W. A. (1993) Birds as indicators of change in marine prey stocks. /n Furness, R. W. & Greenwood, J. D. D. (Eds) “Birds as Monitors of Environmental Change” pp. 217-265. (Chapman and Hall, London.) NICHOLLS, D. G., STAMPTON, P., KLomp, N. I. & SCHULTZ, M. (1998) Post-breeding flight to Antarctic waters by a short-tailed shearwater Puffinus tenuirostris. Emu 98, 79-82. Ropinson, A. C., CANTY, P, Mooney, T. & Ruppock, P. (1996) “South Australia’s offshore islands.” (Australian Heritage Commission, Canberra.) Rocers, P. J., STEPHENSON, P., MCLEay, L. J., DIMMLICH, W. FE, & Warp, T. M. (2004) Sardine (Pilchard) Sardinops sagax Fishery Assessment Report. SARDI Report Series No 84. 75 p. RopPERT-COUDERT, Y., WILSON, R, P., DAUNT, F. & Kato, A. (2004). Patterns of energy acquisition by a central place forager: benefits of alternating short and long foraging trips. Behav. Ecol. 15(5), 824-830. SCHAFFER, S. A., Costa, D. P. & WEIMERSKIRCH, H, (2003) Foraging effort in relation to the constraints of reproduction in free-ranging albatrosses. Func. Ecol. 17, 66-74. ScuuLtz, M. A., & KLomp, N. I. (2000a) Does the foraging strategy of adult short-tailed shearwaters cause obesity in their chicks? J. Av. Biol. 31, 287-294. & Kiomp. N. I. (2000b) Chick-provisioning behaviour of two shearwaters breeding in south-eastern Australia." Austral Ecol. 25, 319-320. Sxira, I. J. (1986) Food of the short-tailed shearwater, Puffinus tenuirostris, in Tasmania. Aust. Wildl, Res. 13, 481-488. TAsKer, M. L., CAMPHUYSEN, C. J., Cooper, J., GARTHE, S., Monteveccul, W. A. & BLABER, J. M. (2000) The impacts of fishing on marine birds. /CES J. Mar. Sci. 57, 531-547. WEIMERSKIRCH, H., DONCASTER, C. P. & CUENOT-CHAILLET, F. (1994) Pelagic seabirds and the marine environment: foraging patterns of the wandering albatross in relation to prey availability and distribution. Proc. R. Soc. Lond. Ser. B 255, 91-97. , CHASTEL, O., ACKERMANN, L., CHAURAND, T., CUENOT-CHAILLET, F., HINDERMEYER, X. & JUDAS, J. (1994) Alternate long and short foraging trips in pelagic seabird parents. Anim. Behav. 47, 472-476. , Witson, R. PB & Lys, P (1997) Activity patterns of foraging in the wandering albatross: a marine predator with two modes of prey searching. Mar Ecol. Prog. Ser. 151, 245-254. & CHEREL, Y. (1998) Feeding ecology of the short-tailed shearwaters: breeding in Tasmania and foraging in the Antarctic? Mar. Ecol. Prog. Ser. 167, 261- 274. (1998) How can a pelagic seabird provision its chick when relying on a distant food source? Cyclic attendance at the colony, foraging decision and body condition in sooty shearwaters. J. Anim, Ecol. 67, 99- 109, Woop, K. A. (1993) Feeding behaviour, offal preferences and tarsus shape of Puffinus shearwaters off central New South Wales. Notornis 40, 123-127. ROYAL SOCIETY OF SOUTH AUSTRALIA INCORPORATED Patron: HER EXCELLENCY MARJORIE JACKSON-NELSON, AC, MBE GOVERNOR OF SOUTH AUSTRALIA OFFICERS FOR 2004-2005 President: R. W. FITZPATRICK, Bsc Agric, MSc Agric, PhD Past President: O. W. WIEBKIN, BSc, PhD Vice-Presidents: A. PRING, Bsc(Hons), PhD Secretary: Minutes Secretary: J. H. LOVE, BA, BD N. F MORCOM, BSc(Hons), PhD Editor: Treasurer: O. MAYO, BSc(Hons), PhD, DSc, FAA, FTSE J.T: JENNINGS, BSc(Hons), PhD Guest Editors: Programme Secretary: S. MURRAY-JONES N. J. SOUTER, BSc(Hons), PhD S. SHEPHERD Membership Secretary: Editorial Board: A. J. McARTHUR, OAM, BE P. CLARKE J. JENNINGS Librarian: K. LEE O. MAYO N. MORCOM 1. RILEY Other Members of Council: A.S. GACKLE, BSc J. E. PATTISON, MA, BSc, MSc, Grad Cert Ed C. H. S. WATTS, BSc(Hons), PhD P. A. CLARKE, BSc, BA, PhD Printed by Graphic Print Group, 10-14 Kingston Avenue, Richmond, S.A. 5033