Mistletoes of Western Australia 148631614X, 9781486316144

Table of contents :
Cover
Title Page
Copyright
Contents
Acknowledgements
Preface
1: Introduction
2: Parasitic plants: an overview
3: Reproduction
4: Do mistletoes have roots?
5: Host variety
6: Do mistletoes mimic their hosts?
7: Fire and other threats
8: Biogeography of Western Australia’s mistletoes
9: The relevance of mistletoes
10: Keys to mistletoe families, genera and species in Western Australia
11: Species accounts
Loranthaceae
Santalaceae
Checklist and broad distribution records of Western Australian mistletoes
Glossary
References and further reading
Index

Citation preview

MISTLETOES OF WESTERN AUSTRALIA TONY START AND KEVIN THIELE

MISTLETOES OF WESTERN AUSTRALIA TONY START AND KEVIN THIELE

© Antony Start and Kevin Thiele 2023 All rights reserved. Except under the conditions described in the ­Australian Copyright Act 1968 and subsequent amendments, no part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, duplicating or otherwise, without the prior permission of the copyright owner. Contact CSIRO Publishing for all permission requests. Antony (Tony) Start and Kevin Thiele assert their right to be known as the authors of this work. A catalogue record for this book is available from the National Library of ­Australia. ISBN: 9781486316144 (pbk) ISBN: 9781486316151 (epdf) ISBN: 9781486316168 (epub) How to cite: Start T, Thiele K (2023) Mistletoes of Western ­Australia. CSIRO Publishing, Melbourne. Published by: CSIRO Publishing Private Bag 10 Clayton South VIC 3169 ­Australia Telephone: +61 3 9545 8400 Email: [email protected] Website: www.publish.csiro.au Sign up to our email alerts: publish.csiro.au/ earlyalert Front cover: (top) Amyema miquellii (photo by Rob Davis); (bottom, left to right) Amyema microphylla, Amyema fitzgeraldii, Amyema benthamii (photos by W.P. Muir) Back cover: (left to right) Amyema xiphophylla (photo by W.P. Muir), Dendrophthoe glabrescens (photo by Rob Davis), Amyema sanguinea (photo by W.P. Muir) Edited by Adrienne de Kretser, Righting Writing Cover design by Cath Pirret Typeset by Envisage Information Technology Printed in China by 1010 Printing International Ltd

Nov22_01

CSIRO Publishing publishes and distributes scientific, technical and health science books, magazines and journals from ­Australia to a worldwide audience and conducts these activities autonomously from the research activities of the Commonwealth Scientific and Industrial Research Organisation (CSIRO). The views expressed in this publication are those of the author(s) and do not necessarily represent those of, and should not be attributed to, the publisher or CSIRO. The copyright owner shall not be liable for technical or other errors or omissions contained herein. The reader/user accepts all risks and responsibility for losses, damages, costs and other consequences resulting directly or indirectly from using this information. CSIRO acknowledges the Traditional Owners of the lands that we live and work on across ­Australia and pays its respect to Elders past and present. CSIRO recognises that Aboriginal and Torres Strait Islander peoples have made and will continue to make extraordinary contributions to all aspects of A ­ ustralian life including culture, economy and science. CSIRO is committed to reconciliation and demonstrating respect for Indigenous knowledge and science. The use of Western science in this publication should not be interpreted as diminishing the knowledge of plants, animals and environment from Indigenous ecological knowledge systems. The paper this book is printed on is in accordance with the standards of the Forest Stewardship Council® and other controlled material. The FSC® promotes environmentally responsible, socially beneficial and economically viable management of the world’s forests.

Contents

           10 Keys to mistletoe families, genera and species in Western Australia 11 Species accounts  

iv v 1 3 7 13 19 23 27 31 37 43 49 50 128

   

139 141 143 145

iii

Acknowledgements

Tony: I have been pursuing Western Australian mistletoes over the past 40+ ­ years. During that time numerous people have put up with frequent stops on long journeys, tolerated cars and camps full of plant pieces, cooked dinner while I pressed specimens, and made room for them among the stuff on roof-racks. Particular thanks are due to Andrew Burbidge, Phil Fuller, Tricia Handasyde and Norm McKenzie for their good-natured patience and help. Many people, knowing my fascination with these plants, have taken time on their own field trips to collect specimens for me. They include Paul Armstrong, Phil Fuller, Tricia Handasyde, Penny Hussey, Sue

iv

Mather, Norm McKenzie, Irene Morcombe, Bill Muir, Joff and Joan Start, John and Helen Start, and Andy Williams. Garry Brockman, Andrew Craig, Allen Lowrie, Bill Muir and Charlie Nicholson loaned their photographs. John Huisman provided access to the Western A ­ ustralian Herbarium collections, where Skye Coffey, Shelley James and Karina Knight, in particular, have not only put up with me but facilitated my work. Kevin: I am indebted to all these people and many others who, over the years, have given me ideas through amiable discussion. Thank you.

Preface

Australian mistletoes are a diverse group of indigenous plants. Many have brightly coloured flowers that attract nectarivorous birds. Their fruits can also be an important food source for some birds, foremost in Australia the mistletoebird. Their leaves are important as a food source for many insects, including the larvae of some spectacularly colourful butterflies. Despite these attributes and the contributions mistletoes make to the communities in which they live, their reputations are often tainted by their parasitic habit. It is not uncommon or unreasonable for people to ask, ‘are they friend or foe’? Chapters 1–9 of this book look at aspects of the natural history of Western ­ Australian mistletoes, as well as their conservation and some threats to their survival. Hopefully, the topics addressed here will help readers to answer the question of

‘friend or foe’ for themselves. Chapter 10 provides keys to the families, genera and species of all Western Australian mistletoes. Chapter 11 comprises an account of each species known from Western ­ Australia. Rather than tedious full descriptions, it focuses on the diagnostic features of each species, as well as the differences between it and similar and related species. Each species account includes a summary of the array of hosts from which the species has been recorded, and its distribution. We include a Glossary of the technical terms we have used. We hope that Mistletoes of Western ­ Australia will be a user-friendly source of information on the rich and varied mistletoe flora of the western third of Australia. We also hope that it will stimulate the study of the ecological roles of mistletoes in Australian plant communities.

v

This page intentionally left blank

1 Introduction and tapeworms, parasites must be bad. On the other hand, there is a growing appreciation that many species are beautiful, and a growing awareness that they are natural components of our indigenous flora. Many native animals are dependent on them. The larvae of some of our most beautiful butterflies feed on nothing but mistletoe leaves (Figs 1.1 and 1.2). Mistletoebirds and some honeyeaters depend on mistletoe berries (the importance of this relationship between plants and birds is even reflected in the mistletoebird’s common name). Many mistletoes, including Perth’s commonest mistletoes Amyema preissii and A. miquelii, flower at the height of summer

Photo: AN Start; reproduced with permission from ­Australia Post.

Strictly speaking, ‘mistletoe’ is the name the English apply to the parasitic plant Viscum album, which grows not only in England but across much of Europe, parts of North Africa and through Asia all the way to Japan. Throughout its vast range, that mistletoe has been the subject of many cultural myths and legends. It is still an important festive symbol; for example, along with holly and ivy, it is a common component of English (and ­Australian!) Christmas trappings, as it has been since time immemorial. Anyone lingering under the mistletoe at that time of year might expect (or hope) to be kissed! Sadly, this is not the place to delve into the rich tapestry of mistletoe myths and legends from around the globe. Interested readers can find many stories by Googling the phrase ‘mistletoe legends’ or similar words. Nowadays, the term ‘mistletoe’ is applied throughout the English-speaking world to plants that have the same broad growth habit as V. album. That is, mistletoes are parasitic shrubs that grow on and obtain mineral nutrients and water from a host’s branch. Mistletoes are not completely parasitic: there is chlorophyll in their leaves and sometimes in their stems, and they photosynthesise their own products such as sugars and their derivatives (discussed further in Chapter 2). Interest in mistletoes is growing. On the one hand, some people consider them to be pests: the thinking is that, like ticks

Fig. 1.1. ­Australian postage stamps showcasing the Spotted Jezebel (Delias aginaippe). Its larvae, like those of many Jezebels, feed on mistletoes. Photo: AN Start; reproduced with permission from ­Australia Post. 1

Photo: AN Start; reproduced with permission from ­Australia Post.

MISTLETOES OF WESTERN AUSTRALIA

Fig. 1.2. ­Australian postage stamps showcasing the Amaryllis Azure butterfly (Ogyris amaryllis). Michael Braby (2004) recorded its larvae feeding on 12 species of mistletoe that occur in Western ­Australia. Other Azures also use mistletoes as larval food plants.

when many nectar-feeding birds have few other nectar sources. The natural history of mistletoes is considered in the introductory chapters of this book. The book is also an aid to identification. Brian Barlow’s revision of the families Loranthaceae and Viscaceae in Flora of ­Australia (Vol. 22, 1984) covered all A ­ ustralian species known at that time, but it is technical, not readily accessible, and not very suitable for use in the field. Moreover, since 1984, our knowledge of the taxonomy, distribution and ecology of Western ­Australian species has improved greatly. David Watson’s more recent Mistletoes of Southern ­ Australia (2019) includes species from southern WA, but not the rich mistletoe flora of the state’s north.

2

Identification keys and up-to-date accounts of all mistletoe species known from WA are the core of this book. Keys are provided to the families, genera and species of all recognised WA mistletoe taxa and are based, wherever possible, on vegetative features rather than on flowers and fruits, which are not always available. The descriptive species accounts are accompanied by photographs and distribution maps. They include diagnostic features to help distinguish similar and related species, as well as information on hosts, distribution, conservation status and threats. Where possible, technical jargon has been avoided; at the end of the book there is a glossary of terms that have been used. We make no apology for avoiding common names. Recently, many common names have been coined, but many are not appropriate for WA species. For example, ‘Wire-leaved Mistletoe’ could apply to several species of Amyema and a Lysiana, so the name is confusing and tells us nothing about the relationship of one species to another in the way that scientific names can. Moreover, if we used common names in this book, we would have had to invent some for tropical species that lack them. This book is designed for use by people with a reasonable botanical knowledge, who are generally comfortable with scientific names and appreciate the information on relationships that names provide. After all, everyone is comfortable with Banksia and Eucalyptus, so why should Amyema and Lysiana not become familiar to people interested in mistletoes?

There are many parasitic plants in Western ­Australia. Not all are mistletoes, and many are not immediately recognisable as parasites because they tap into their host’s roots and, above ground, look like normal flowering plants. Before placing mistletoes within the context of the broad array of parasitic plants, we need to understand two important features of parasitic plants: all parasitic plants are either (1) root parasites or stem parasites and (2) holoparasitic or hemiparasitic. Examples of all four combinations of these features can be found in WA. As the name suggests, root parasites obtain resources from their host’s roots whereas stem parasites obtain resources from their host’s stems or, in the case of dodders (see below), all above-ground organs including leaves. Holoparasites are wholly dependent on the host plant for all their nutrient needs. They have no functioning chloroplasts, and hence cannot synthesise their own carbohydrates. Hemiparasites have functioning chloroplasts and photosynthesise their own sugars (but obtain water and mineral nutrients from the host). Ignoring dodders for the moment (but see below), three WA families include holoparasites. One is the Orobanchaceae with two species: Orobanche minor, an introduced, purple-coloured root parasite, familiar to gardeners as broomrape (Fig.  2.1) and the indigenous species, O.  cernua, which was once widespread

Photo: AN Start.

2 Parasitic plants: an overview

Fig. 2.1.  Orabanche minor is a holoparasite well known to gardeners as broomrape.

though scattered in the south-west and is currently known in WA only from Dirk Hartog Island. Members of this family are all root parasites. Another is the Orchidaceae, a few of which (e.g. the potato orchid Gastrodia) have lost the ability to photosynthesise. Recent research has shown that some holoparasites (e.g. the potato orchid) obtain their nutrients from other vascular plants through a fungal bridge and are, therefore, a special type of parasite called a mycoheterotroph. The remarkable family Apodanthaceae is the third WA family that contains holoparasites. Here, there are three indigenous species of Pilostyles. Pilostyles 3

Fig. 2.3.  The fruits of this quandong will turn bright red on maturity.

Photo: Kevin Thiele.

­ amiltoniorum (Fig. 2.2) is quite common h in the Perth hills, where it infects some Daviesia spp. (Fabaceae). Apart from the flowers and fruits, Pilostyles live entirely within the host’s stems (and perhaps in the roots – if fire kills the host’s stems and thus the parasites within them, P. hamiltoniorum can appear in regenerating stems, suggesting it persisted in its host’s roots). Even when flowering or fruiting, P.  hamiltoniorum is easily overlooked because, at a glance, the tiny flowers or fruits can be mistaken for colonies of scale-insects. Western A ­ustralia has many more hemiparasites than holoparasites. They include familiar plants such as quandong (Santalum acuminatum; Fig. 2.3) and sandalwood (S.  spicatum), both of which are

Photo: AN Start.

MISTLETOES OF WESTERN AUSTRALIA

Photo: AN Start.

Fig. 2.4.  The white flowers of a root parasite, Buchnera sp. (Orabanchaceae).

Fig. 2.2.  The tiny purple and white flowers of Pilostyles hamiltoniorum, a holoparasitic stem parasite growing on Davesia decurrens in the Perth Hills. 4

root parasites. Besides the holoparasitic Orobanche, the family Orobanchaceae contains many hemiparasitic root parasites such as Striga and Buchnera (Fig. 2.4). These genera were previously placed in the family Scrophulariaceae until recent genetic studies revealed their close relationship to Orobanche. In Africa, some strikingly beautiful Striga species are weeds with serious economic consequences as they can reduce crop yields of struggling rural

Fig. 2.5.  Phacallaria sp. (Santalaceae), a strange Asian mistletoe growing a loranthaceous mistletoe Helixanthera ligustrina in northern Thailand. Flowers and fruit develop along its frequently unbranched leafless green stems.

Photo: AN Start.

Photo: AN Start.

2 – Parasitic plants: an overview

Fig. 2.7.  Dodder (Cassytha sp.). On maturing, the fruits will become white and translucent.

communities by as much as 30  per  cent. However, WA Striga species are integral components of our indigenous flora and are not pests. Some plants change their parasitic mode during their life cycle. Cassytha (Figs 2.6 and 2.7) and Cuscuta are familiar, wiry, often yellowish, almost leafless vines that grow over trees and shrubs and sometimes even grasses. Interestingly, there is a remarkable convergence of form in these genera, which are not closely related (­Cuscuta belongs in the morning glory family Convolvulaceae while Cassytha

belongs in the laurel family Lauraceae). In both genera, seeds germinate in the soil and the seedlings are independent plants, not parasites. To develop to adulthood, however, both must find and attach to a host, end their connection with the soil and become obligate stem parasites.

The mistletoes Mistletoes are the only hemiparasitic plants that infect their host’s stems (Watson 2019). In WA, mistletoes occur in two families, the Santalaceae1 and Loranthaceae. The Santalaceae includes many more root parasites (such as the aforementioned sandalwood and quandong) than stem parasites (mistletoes). However, all WA species in the Loranthaceae, except one, are mistletoes. The exception is the state’s iconic Christmas tree, Nuytsia floribunda, Western A ­ ustralian mistletoes in the family Santalaceae were once regarded as members of a distinct family, the Viscaceae, which was composed entirely of stem parasites (including that English species with which we started the Introduction). However, recent genetic evidence has shown that the distinction between traditional members of the Santalaceae (e.g. quandongs) and those of the Viscaceae is not warranted at the family level and the two have been merged under the earlier name, Santalaceae. Some mistletoes have always been placed in the Santalaceae (e.g. the south-east Asian genus Phacellaria (Fig. 2.5)).

Photo: AN Start.

1 

Fig. 2.6.  Dodder (Cassytha sp.) almost smothering its Melaleuca hosts.

5

MISTLETOES OF WESTERN AUSTRALIA

which is not a mistletoe because it is a root parasite. It may seem illogical that all WA members of a family are mistletoes except one, but the term ‘mistletoe’ embraces a growth form, not a taxonomic grouping. Other examples of such terms are ‘mallee’ and ‘mangrove’. Mallees are eucalypts that, after fire, generate multiple stems from lignotubers. However, not all eucalypts

6

are mallees. Mangroves include a fern, a palm and representatives of several familiar families of land plants like the genus Acanthus. As you can see, mangroves are not all closely related and do not comprise a taxonomic group. Some mistletoe species might be considered mangroves too because they can infect mangrove trees and tolerate occasional inundation at high tide.

Many plants can reproduce either vegetatively or sexually. Examples of both methods can be seen in Western Australian mistletoes, although the ­ former is rare. Dependence on a host’s stem for resources (see Chapter 2) limits a mistletoe’s opportunity to reproduce vegetatively but does not eliminate it altogether. As described in Chapter 4, some mistletoes can produce cloned offspring, usually on the same host as the parent plant. However, where host canopies overlap, epicortical roots (Chapter 4) can occasionally spread a mistletoe from one host to another. If this happens, secondary haustoria can establish in the second host and give rise to new cloned plants. This is apparently uncommon, but we have seen it at Crossing Pool in ­Millstream-Chichester National Park in the Pilbara, where Amyema sanguinea occasionally spreads between intermingling canopies of its host paperbark trees (Melaleuca leucadendra). Mistletoes most commonly reproduce sexually. Sexual reproduction requires several steps, but we’ll touch on the two fundamental ones – pollen transfer and seed dispersal. Little is known about pollination in WA’s santalaceous mistletoes Korthalsella and Viscum, both of which have minute yellow or green flowers (Fig. 3.1). They are probably pollinated by insects (small flies are known to pollinate Viscum album in Europe and Korthalsella in New Zealand).

Photo: Garry Brockman.

3 Reproduction

Fig. 3.1.  The minute flowers of Korthalsella arthroclada (Santalaceae) may be insectpollinated.

Loranthaceous mistletoes are known as the ‘showy’ mistletoes because their flowers are relatively large and often brightly coloured (Fig.  3.2). They produce copious nectar and are not strongly scented. These characters are associated with pollination by nectar-feeding birds. In WA, honeyeaters and silvereyes are often observed probing ‘showy’ mistletoe flowers (Napier et  al. 2014). Flower wasps have also been observed pushing their way deep into the flowers of some mistletoe species and may be capable of pollinating them. 7

Photo: Bill Muir.

In ­Australia, the most common flower colours for bird-pollinated species are bright red and green, or a combination of both. These colours are characteristic of the bird-pollinated WA kangaroo-paws, and the same colours characterise most showy mistletoes (see illustrations for individual species accounts). Some have all-red flowers, but many have combinations of red and green. Many Lysiana spp. have green to yellow tips above a red or yellow corolla base (Fig.  3.3). Occasionally, aberrant individuals of normally red-­ flowered species lack the red pigment and have yellow flowers. This is not uncommon in A. preissii (Fig. 3.4). However, some

Fig. 3.3.  Some forms of Lysiana casuarinae have red and yellow flowers. 8

Fig. 3.4.  Red-flowered mistletoes occasionally lack the red pigment, in this case Amyema preissii (compare with Fig. 3.2).

Photo: AN Start.

Fig. 3.2.  The red flowers of Amyema preissii are typical of many loranthaceous mistletoes.

Photo: Bill Muir.

Photo: Bill Muir.

MISTLETOES OF WESTERN AUSTRALIA

Fig. 3.5.  Some mistletoes have yellow flowers that may become orange or red as they mature, in this case the young flowers of Amyema dolichopoda.

species always have yellowish flowers (Fig.  3.5; see also figures in the species accounts). Flowers of A.  fitzgeraldii are green with red filaments (Fig.  3.6) while those of A. hilliana (Fig. 3.7) and A. maidenii are entirely green (although the latter may have pink anthers). Brightly coloured fruits are also associated with dispersal by birds, and many

Photo: Bill Muir.

Fig. 3.6.  Amyema fitzgeraldii is a green-flowered species with red filaments and a red style.

Photo: Bill Muir.

Fig. 3.8.  The colour of the fruits of Amyema fitzgeraldii show through its dense indumentum.

Photo: Bill Muir.

Photo: Bill Muir.

3 – Reproduction

Fig. 3.9.  Many forms of Lysiana casuarinae have bright red fruits.

fruits of WA loranthaceous mistletoes become pink to red when they mature (Figs 3.8 and 3.9). However, some mature to a yellow colour (e.g. A.  miraculosa, Fig. 3.10) and in one form of Lysiana casuarinae growing in the Pilbara they remain green when ripe, albeit with some bronzing where they are exposed to the sun. These fruits are still attractive to birds. The story of mistletoebirds, with their specialised digestive system and habit of wiping strings of defecated sticky seeds onto branches (Fig.  3.11), is well known. Very rapid gut-transit times (as short as 4–12  min: Murphy et  al. 1993) suggests that these mistletoe specialists are unlikely to transport seed for long distances; rather,

Photo: Bill Muir.

Fig. 3.7.  Amyema hilliana has entirely green flowers.

Fig. 3.10.  Amyema miraculosa is a yellow-fruited species.

they will tend to deposit seeds in the same host in which they found the food supply, or nearby. Other more generalist fruiteating birds are probably responsible for distributing mistletoe seeds over longer 9

Photo: JD and MJ Start.

MISTLETOES OF WESTERN AUSTRALIA

Photo: AN Start.

Fig. 3.12.  A string of Viscum articulatum seeds voided by a mistletoebird, some of which are already germinating.

distances to establish new populations. Spiny-cheeked honeyeaters are often associated with mistletoes in WA, as are silvereyes, white-fronted honeyeaters and grey honeyeaters. The fruits of most Viscum spp. are small and usually white, cream or green (Fig. 4.2). The single seed has a viscous coating and is probably dispersed by birds. In the K ­ imberley, Joff and Joan Start observed mistletoebirds feeding in V.  articulatum and photographed nearby strings of defecated seeds (Fig. 3.12). Korthalsella fruits, however, are very different. They are tiny (little more than 1 mm long) and greenish yellow in colour, like the stems on which they are borne (Fig.  3.13). Each contains a single minute seed (Fig. 3.14) with a viscid coat. They are probably ejected ‘explosively’ by hydrostatic pressure, like those of New Zealand’s Korthalsella species and Arceuthobium, 10

Photo: KR Thiele.

Fig. 3.11.  A clump of germinating mistletoe seeds, in this case Diplatia grandibractea.

Fig. 3.13.  The tiny fruits of Korthalsella arthroclada probably expel the single seed explosively.

Photo: AN Start.

3 – Reproduction

Fig. 3.14.  The tiny green seed of Korthalsella leucothrix beside the fruit from which it was extracted.

a related genus in the same family that’s common in America but extends to Asia. If this is correct, then large gaps between potential host canopies may be barriers to Korthalsella seed dispersal. Adjacent tree canopies infected by either of WA’s Korthalsella species are usually no more than 2 m apart or are completely or nearly contiguous. At one location, all the vegetation below an Acacia canopy that was heavily infected by K. leucothrix was peppered with white seeds, which appear to change from greenish to white as the outer coat dries. Nevertheless, from time to time, Korthalsella must be able to disperse over longer distances and birds are probably the inadvertent vectors (see Chapter 8). More work is needed to understand both pollination and seed dispersal in Korthalsella. Outside WA, birds other than mistletoebirds specialise on eating (and hence dispersing) mistletoe fruits. The mistletoebird is the only A ­ ustralian member of a diverse group of mostly brightly coloured mistletoe specialists known as flowerpeckers. These are important pollinators and dispersers of Asian mistletoes. They are a distinct group within the family Nectariniidae that also contains the sunbirds,

which are important pollinators of many mistletoes, especially in Africa and Asia (Cheke et al. 2001). John Rawsthorne (2017) has reported that, in the Northern Territory, although mistletoebirds appear to be the main consumers of the relatively large seeds of Decaisnina species (a genus with some common species in the Kimberley) they commonly discard the seeds onto a nearby branch without ingesting them. Nevertheless, the discarded seeds germinate. This can result in very heavy infestations in some host trees while neighbouring potential hosts remain mistletoe-free. One wonders why the bird goes to the trouble of extracting the seed if it does not eat it! In Thailand, the mature orange buds of a common mistletoe, Dendrophthoe pentandra, have swollen tips, which resemble the shape and orange colour of the same plant’s ripe fruit. Scarlet-backed flowerpeckers squeeze the tips of mature buds with their beak. This causes the petals to spring apart, allowing the bird to delve into the newly opened flower for nectar and in so doing receive a dab of pollen on the head. At the same time, the bird’s head brushes over the stigma, depositing the pollen it obtained from a previously visited flower. The same pinching action on a mature fruit causes its ‘skin’ to split around the ‘equator’. The cap is flicked off and the bird extracts and devours the seed. By displaying similar visual cues, the mistletoe advertises both nectar and fruit to the flowerpecker that both pollinates it and disperses its seeds (Start 2011, 2013a). The same situation has been reported from India, so perhaps we should look for similar adaptations in A ­ ustralia. 11

This page intentionally left blank

4 Do mistletoes have roots?

Photo: JD and MJ Start.

Most plants have two distinct sections, a root and a stem. The root supplies a plant’s water and mineral nutrients from the soil while the stem supports leaves, flowers and fruits. Fuelled by light, chloroplasts in the leaves (and sometimes stems) combine carbon dioxide with water to produce sugars and their derivatives, providing the energy the plant needs to grow. Most mistletoes also have green leaves, but a few replace or augment functional leaves with green stems (Figs 4.1 and 4.2) or even huge, persistent floral bracts (e.g.  D ­ iplatia spp.) (Fig.  4.3). Either way, mistletoes synthesise their own carbohy-

Photo: AN Start.

Fig. 4.2.  Viscum articulatum (Santalaceae), a hemiparasitic stem parasite that has no ‘ordinary’ leaves but does have chlorophyll in the stems.

Fig. 4.1.  Korthalsella leucothrix (Santalaceae), a hemiparasitic stem parasite that has no ‘ordinary’ leaves but does have chlorophyll in the stems.

drates. However, as they grow on the stem of their host, mistletoes don’t have conventional roots with which to acquire water and mineral nutrients from the soil. So, what takes the place of roots? Like other seeds, a germinating mistletoe seed produces a radicle, the progenitor of a root system (Fig. 4.4). However, to obtain minerals and water, the mistletoe radicle must penetrate a suitable host’s stem and tap into its water and mineral ‘supply line’, the xylem. It does this by developing a 13

Photo: AN Start.

MISTLETOES OF WESTERN AUSTRALIA

Photo: AN Start.

Fig. 4.5.  The base of Viscum whitei (Santalaceae). There is little indication of swelling in the host plant’s stem, despite the haustorium in it.

Photo: JD and MJ Start.

Fig. 4.3.  The large green floral bracts of Diplatia grandibractea will persist and continue photosynthesising long after the flowers and fruits have gone.

Fig. 4.4.  A germinating seed of Decaisnina angustata (Loranthaceae). The ‘head’ is the radicle or embryonic root.

highly modified structure called a haustorium. Most WA mistletoes depend on their initial (primary) haustorium throughout their lives (but see below). As we saw in Chapter 2, mistletoes occur in two WA plant families, the Santalaceae and the Loranthaceae. In santala14

ceous mistletoes, the primary haustorium develops inside the host’s branch. The host continues growing beyond the point of infection and there is little external evidence of the intercepting structure apart from a modest swelling of the host’s stem (Fig. 4.5). However, in loranthaceous mistletoes the haustorium is generally large and complex and the mistletoe component is usually externally visible. Loranthaceous haustoria come in several designs. Their anatomy has been studied and the types have been classified, but little attention has been given to the functions behind variation in form. Nevertheless, studies in WA show that some types of haustoria have important functional attributes besides tapping the host’s xylem. Not surprisingly, given the state’s fire-prone environment, one function of some types is to improve the mistletoe’s ability to survive fire (discussed further in Chapter 7). The next sections of this chapter describe the basic types of haustoria found in WA’s loranthaceous mistletoes.

Ball haustoria The commonest design is called a ball haustorium. In ball haustoria, tissues of

4 – DO MISTLETOES HAVE ROOTS?

Fig. 4.8.  The ball haustorium of Amyema linophylla (Loranthaceae). The host branch is dead beyond the point of interception (arrowed). The difference in diameter of the host branch prior to and beyond the haustorium indicates how much the host branch has grown while supporting the mistletoe.

Photo: AN Start.

Fig. 4.6.  The ball haustorium of Amyema dolichopoda (Loranthaceae). In this case, the host tissue has grown around that of the mistletoe.

Photo: AN Start.

both the mistletoe and its host expand to form a roughly rounded structure, the ball (Fig. 4.6). Inside this, the tissues of the host and parasite form a complex placenta-like interface. Although the internal structure of the haustorium cannot be seen while the mistletoe is alive, when it dies (and mistletoes are usually shorter-lived than their hosts), the mistletoe’s much-softer wood tends to rot more rapidly than that of the host, thus exposing the latter’s half of the internal structure. This ‘skeleton’ is often called a wood-rose (although it looks more like a mushroom with numerous gills than a rose!). In heavily infected hosts, it

Photo: Bill Muir.

Fig. 4.7.  A Balinese wood carving which incorporates a wood-rose formed when the carved branch was infected by a mistletoe.

is common to see many wood-roses – the remnants of dead mistletoe haustoria – scattered through the host canopy. In Bali and some other places, you can buy carvings of animals into which the artist has cleverly incorporated wood-roses (Fig. 4.7). The degree to which different species of mistletoe contribute to the mass of the haustorium varies. In some, the proportion of host to mistletoe is approximately equal, in some the mistletoe portion wraps around the host, while in others it is the other way around. Within each mistletoe species, the basic pattern remains fairly constant. As they grow, ball haustoria commonly kill the host’s branch beyond the point of infection. It can be instructive to compare the live section in front of the mistletoe with the dead section beyond it. The dead portion shows the diameter of the twig soon after it was infected, and the live portion shows how much the branch has grown since then, with nothing but the mistletoe attached (Fig. 4.8). This indicates that mistletoe seedlings usually establish on host

15

MISTLETOES OF WESTERN AUSTRALIA

Ramifying haustoria

Photo: AN Start.

Some mistletoes develop strands of haustorial tissue that spread (ramify) within the host’s stem. New plants (clones) can develop when the haustorial strands send ‘sinkers’ into the host’s xylem and leafy shoots burst through the host’s bark some distance from the parent plant (Fig. 4.9). In ­Australia, ramifying haustoria appear to be most common in mistletoes that infect long-lived arid-zone hosts, such as Amyema fitzgeraldii on mulga (Acacia aneura and its relatives), Amyema gibberula var. gibberula on Grevillea wickhamii, Amyema maidenii (which infects many hosts, especially Acacia spp.) and Amyema quandang growing on western myall (Acacia papyrocarpa). Ramifying haustoria may be far more common than the few records suggest, because they are difficult to detect. When the haustorium produces a clone, the tiny new shoots are difficult to distinguish from mistletoe seedlings, hence this method of reproduction is easily overlooked. Perhaps

Fig. 4.9.  A cloned shoot of Diplatia grandibractea (Loranthaceae) emerging from ramifying haustorial tissue in the host branch. 16

mistletoes sometimes outgrow their haustorium’s supportive capacity, so if the host is long-lived the mistletoe can have ramifying haustoria, with the advantageous ability to develop new clones back along a still-thriving part of the host’s branch.

Wedge haustoria

Wedge haustoria are only seen in Diplatia grandibractea. They may be derived from ramifying tissues that expand to form an ever-widening wedge, which progressively splits the host’s branch, usually on the upper side. New mistletoe stems arise from the freshly exposed haustorial tissue (Fig. 4.10). This rhizome-like growth-form gives D. grandibractea a potentially longer life span than species with ball haustoria, because the latter are fixed in position and eventually outgrow their host’s capacity to support them whereas the wedge grows continuously towards the base of the host’s branch, generating fresh interfaces with the host’s xylem and sending out new branches. This haustorial form in D.  ­grandibractea also helps the species survive bushfires (see Chapter 7).

Photo: AN Start.

twigs, where the bark is thin enough to be penetrated by the seedling’s radicle.

Fig. 4.10.  The wedge haustorium of Diplatia grandibractea working back along the host branch. The youngest, smallest mistletoe branches are near the advancing toe of the wedge and the dead host branch beyond the mistletoe on the right-hand side.

4 – DO MISTLETOES HAVE ROOTS?

Fig. 4.12.  An epicortical root of Amyema sanguinea var. sanguinea that has established a new cloned shoot on the branch of a cajuput (Melaleuca leucadendra) host. The new plant has already developed a leafy shoot

Photo: AN Start.

In contrast to ramifying haustorial strands within a host’s branch, some mistletoes send out ‘runners’ from their primary haustorium along the outside of a host stem. Because the anatomical structure of these ‘runners’ is root-like, not stem-like, we call them epicortical roots (from the Greek epi-, above or outside, and cortex, skin or bark). Where epicortical roots touch the host’s branches, they periodically send ‘sinkers’ down into it to form secondary haustoria (Fig. 4.11). Once established, a secondary haustorium can develop a leafy shoot (Fig. 4.12). If the interconnecting root or ‘runner’ dies, as it usually does in time, the new shoot becomes an independent clone of the parent plant. This may explain why A. sanguinea, which sends out epicortical roots, can be seen growing on the undersides of huge river gum (Eucalyptus camaldulensis) branches (Fig.  4.13) where no mistletoebird could possibly have deposited a seed! An alternative explanation is that the new plants have originated from ramifying haustorial tissue in the host’s branch. Good examples can be seen at Millstream and from the east side of the bridge where the Great Northern Highway

Photo: AN Start.

Epicortical roots

Photo: AN Start.

Fig. 4.13.  Amyema sanguinea var. sanguinea (Loranthaceae) growing from the underside of a large branch of its host Eucalyptus camaldulensis at Millstream.

Fig. 4.11.  Section of an infected host showing the epicortical root of Amyema sanguinea developing a sinker. Specimen in the WA Herbarium # 05096049.

crosses the Fortescue River, just south of Newman. The ancestors of ­Australian mistletoes are thought to have come from south-east Asia (see Chapter 8), where epicortical roots occur in many genera. In A ­ ustralia, 17

MISTLETOES OF WESTERN AUSTRALIA

epicortical roots are uncommon, which is puzzling. In WA, they are only found in three genera (Decaisnina, Dendrophthoe and just one species in the large genus Amyema, A.  sanguinea). The first two genera are restricted in WA to the Kimberley and are far more diverse in Asia than ­Australia. ­Australian species in these genera are probably derived from relatively recent south-east Asian immigrants. South of the Kimberley, A.  ­sanguinea is the only ­mistletoe that

18

grows epicortical roots (the species occurs as far south as the Gascoyne River). The absence of epicortical roots in ­Australian mistletoes is thought to be a derived condition. It is possible that the development of ramifying haustorial strands sheltered inside a host’s infected branch, instead of the exposed epicortical roots that are common in south-east Asian mistletoe, is an adaptation to A ­ ustralia’s arid and fire-prone environment.

5 Host variety A mistletoe’s habitat is its host! This essential point is often overlooked, even by botanists. For example, the labels of many herbarium specimens omit the name of the host but record in detail the soil and vegetation associated with the host’s habitat (factors that are largely irrelevant to the mistletoe). Less often, the name of the host is recorded on the label but, after taxonomic treatments have changed some names and split others, the identification may be ambiguous. Moreover, the ability of some collectors to correctly identify the host in the field can add doubt to the dependability of the information on the label. Specimens that have an identifiable portion of the host mounted alongside the mistletoe as a voucher are particularly valuable. Better still are those for which there is a separately mounted host specimen with cross-referenced collecting numbers on the labels. Why is it so important to know the species of a mistletoe specimen’s host? Many of the ecological observations in this book would have been impossible were it not for an extensive record of mistletoe–host associations sourced from herbarium specimens and built up over many years of extensive travel and collecting. Fire is undoubtedly the greatest threat to most mistletoes (see Chapter 7). Knowledge of how vulnerable a mistletoe host is to fire can indicate how fire-vulnerable the mistletoe itself is. Thus, knowledge of a mistletoe’s range of hosts can be critical

to understanding its ecology and conservation status. Because some widespread species have different hosts in different regions, understanding a mistletoe’s host preferences requires the accumulation of many records from throughout its range. A good example of a mistletoe with different host preferences in different regions is Amyema benthamii. In southern ­Western ­Australia (south of ~26°S) all of the 48 available records are of kurrajong hosts, usually Brachychiton gregorii (but occasionally other Brachychiton species that have been introduced as ornamental trees). The species is rare in the Pilbara but is known from three host species in three different families (including one species of Brachychiton). In the Kimberley, however, there are 58 host records from 32 species in 17 families (Start 2013b). In Chapter 7 we explore how this variation in host preference can affect a mistletoe’s fire ecology, particularly its persistence in fire-­ vulnerable habitats. A prominent feature of host selection by WA mistletoes is the division between those that always infect, and those that never infect, myrtaceous hosts. The great majority of myrtaceous hosts are eucalypts (­Eucalyptus or Corymbia) although ­Melaleuca is an important host for some species, particularly in southern WA, and mistletoes sometimes infect bottlebrushes in the genus Callistemon. South of the Kimberley, mistletoes that infect myrtaceous hosts are rarely if ever found 19

MISTLETOES OF WESTERN AUSTRALIA

on non-myrtaceous hosts, and vice versa. This breaks down in the Kimberley, but only among the genera Dendrophthoe and Decaisnina, which are confined to ­ northern A ­ustralia and presumably arrived relatively recently from Papua New Guinea or south-east Asia (see Chapter 8). Even among these genera, fidelity to myrtaceous or non-myrtaceous hosts is quite strong. Elsewhere in A ­ ustralia, this distinction appears to break down. For example, in eastern A ­ ustralia A.  miquelii has been recorded from 10 different host families (Downey 1998), while in WA all reliable records are from Myrtaceae and mostly from species of Eucalyptus. Some species that prefer myrtaceous hosts also show strong preferences for specific genera. For example, A.  melaleucae and A. microphylla are only known to infect species of Melaleuca, and A. ­bifurcata and the related Kimberley species A.  eburna are most frequently found on bloodwoods (Corymbia spp.). Amyema sanguinea infects many species of Eucalyptus as well as Corymbia. Amyema miquelii has odd host preferences in WA: in the southwest of the state it is as common on marri (C. calophylla) as it is on various E ­ ucalyptus spp., but in the north of the state it rarely infects bloodwoods (Corymbia spp.) even though these are very common and diverse in that region. It also seems strange that it rarely infects the very abundant jarrah (E. marginata). There are no records of any mistletoe infecting karri (E.  diversicolor), although many other smooth-barked eucalypts are good mistletoe hosts. Species that occur on non-myrtaceous hosts vary from being strongly host-­ specific to extraordinarily catholic in their choice of host. Strongly host-specific spe20

cies often prefer hosts that grow in fire-safe situations. A good example is A. thalassia, which is known only from mangroves; in WA these are usually Avicennia marina. Moderately host-specific species usually favour some species that grow in fire-­ sheltered refugia, like riparian or rocky areas, although they will also infect species in more fire-prone habitats. This versatility improves the chance that some individuals will survive in refugia through droughts or wildfire and subsequently recolonise less favourable environments when conditions allow (see Chapter 7). Some species that grow where the risk of fire is high, such as on shrubs or trees in spinifex (Triodia sp.) grasslands of the Pilbara and western deserts, can infect an extraordinarily wide range of hosts. The survival advantage of not being too fussy is examined in more detail in Chapter 7. It is likely that adaptability increases the likelihood that a mistletoe seed carried into a burnt area by a bird will be deposited on a plant that it can successfully infect, thus facilitating recolonisation after fire. An extreme example of a host-generalist species in WA is Lysiana casuarinae, which was once recorded in Karijini National Park from 11 host species in five families along 200 m of road verge in a highly flammable hummock grassland! Those hosts included a perennial mulla-mulla (­Ptilotus rotundifolius). Sadly, that remarkable population was eliminated by an extensive wildfire (Start 2011). Among the four mistletoe species that occur in Perth, two (A.  linophylla and L. casuarinae) only infect swamp sheoaks (Casuarina obesa) growing on the banks of the Swan and Canning Rivers where they are seldom, if ever, exposed to fire. Amyema

5 – Host variety

miquelii often infects tall eucalypts where height can save them from scorch, while A.  preissii infects many species of Acacia, including some weedy exotic species that were introduced for their ornamental values and subsequently escaped. Urbanisation has provided these last two species with many refugia in parks and gardens where they are at little risk of being burnt (although they are sometimes attacked by a gardener’s secateurs). A remarkable feature of mistletoe–host relationships is the occurrence of epiparasitism: the infection of one parasite by another. It is quite common to find loranthaceous mistletoes, particularly A.  ­miraculosa, growing on species of Santalum, which are themselves root parasites. Perhaps more remarkable are mistletoes that grow on other mistletoes. This is almost obligatory in two of the three WA species of Viscum: V. articulatum and V. whitei are both almost

invariably recorded (in WA) as parasites of loranthaceous mistletoes. Occasional records of other hosts may be due to error when the collector failed to realise that the Viscum’s host was in fact another mistletoe and not the tree on which its host was growing. The third WA Viscum, V.  ovalifolium, has never been recorded as an epiparasite in WA. Autoparasitism, in which a mistletoe infects another individual of its species, seems to be rare in mistletoes although there should be plenty of opportunity for this to happen. When a mistletoebird finds a copious supply of mistletoe fruits it is likely to continue foraging there and, as mistletoe seeds pass through the bird in a matter of minutes, we would expect it to defecate some seeds in the mistletoes in which it was feeding. Perhaps autoparasitism is more common than we realise but is hard to detect.

21

This page intentionally left blank

6 Do mistletoes mimic their hosts?

Photo: AN Start.

into the canopy of its host, Brachychiton gregorii, while Fig.  6.2 shows how different their leaves are. Nevertheless, it helps the mistletoe to blend visually with its host if the shape, colour and orientation of its leaves and branches resemble those of its host (Fig.  6.3). Most Western ­Australian mistletoes growing in eucalypts (Euca-

Fig. 6.1.  A large plant of Amyema benthamii blending into the canopy of its host Brachychiton gregorii.

Photo: AN Start.

Mimicry is perhaps better known in animals than in plants. Nevertheless, it does occur in many plants, particularly in relation to pollination mechanisms. For example, many orchids fool male wasps into thinking their flowers are female wasps. In attempting copulation, the male will either deliver packets of pollen or collect new packets to deliver to the next flower that deceives it. The orchids achieve this trick by mimicking the visual and/or olfactory cues used by female wasps to attract males. We cannot help marvelling at the resemblance of some mistletoe leaves to those of their hosts. Colonial botanist James Drummond in 1840 was the first of many people to comment on this. In 1977, Brian Barlow and Delbert Wiens explored this topic in more depth (Barlow and Wiens 1977). They pointed out that a fundamental prerequisite to a mistletoe’s development of host mimicry is a degree of host specificity. If a mistletoe has many dissimilar hosts, mimicking one of them is unlikely to convey much selective advantage. The phenomenon is subtle. As Barlow and Wiens succinctly put it, ‘Mistletoes present a rare example of a mimicking situation in which a whole plant blends cryptically with the background of another plant’. In other words, for mimicry to be useful, the mistletoe must resemble the overall form of its host rather than specific components, like leaves. Crypsis does not require precise replication of leaf shapes. Figure 6.1 shows how well Amyema benthamii can blend

Fig. 6.2.  Western ­Australian herbarium specimen # 07730608 showing the very different shape of the mistletoe (Amyema benthamii) leaves (lower left) and those of its host Brachytchiton gregorii (top right). 23

Photo: AN Start.

MISTLETOES OF WESTERN AUSTRALIA

Photo: AN Start.

Fig. 6.3.  Amyema xiphophylla blends with its host Acacia ziphophylla.

Fig. 6.4.  The long pendulous leaves of Amyema bifurcata are not unlike those of its eucalypt hosts.

lyptus or Corymbia), like their hosts, have long pendulous branches and long, narrow, pendulous leaves (Fig. 6.4). A good WA example is A.  linophylla, the terete grey leaves and often pendulous branches of which closely resemble the twigs of its Casuarina hosts remarkably well. However, Lysiana casuarinae can be found growing alongside A. linophylla in 24

the same host tree. While the Lysiana also has terete leaves and pendulous branches, features that help it blend with its host, its leaves are bright green and contrast with its host as well as with those of the Amyema that grows alongside it. Perhaps this difference helps the Lysiana compete with the often more numerous Amyema for the services of birds. Another often cited (eastern A ­ ustralian) case is the remarkable similarity of Dendrophthoe homoplastica leaves to those of one common host, Eucalyptus shirleyi. However, Paul Downey (1998) has recorded D. homoplastica from 14 other hosts from five families, and it does not in any way resemble most of these. For every example of a mistletoe and host with similar leaves there are many others where mistletoe and host leaves look markedly different. But what is the advantage of a mistletoe mimicking its host? Barlow and Wiens (1977) concluded that mimicry provides crypsis, a protective measure that ‘evolved primarily as a response to predation pressure by herbivorous arboreal marsupials’ (notably, they thought, leaf-eating possums). Do mistletoes really need to hide from possums? In Canberra, Peter Atsatt (1983) offered a captive brush-tailed possum the leaves of mistletoes and the hosts they mimicked, as well as leaves from elm trees. The possum invariably ate all it was offered except the mistletoe leaves, which it rejected ‘except for small exploratory bites’. An experiment on a single possum is not much, so Atsatt looked for evidence of possums eating mistletoe leaves in the wild: he found none. On the other hand, he reported another study which found that

6 – DO MISTLETOES MIMIC THEIR HOSTS?

involves similarity between the flowers or fruits of mistletoes and hosts. Indeed, it is in mistletoes’ interest to show off their flowers and berries so as to attract pollinators and seed dispersers. Figure 6.7 shows how the red flowers of A. sanguinea stand out to attract pollinating birds, while its pendulous branches and leaves blend with those of its host (in this case a cajuput, Melaleuca leucadendra). Diurnal birds that see colour are the mistletoe’s pollen and seed vectors. The juxtaposition of cryptic forms and distinctive colours may make flowers and fruits visible to birds, while hiding the mistletoe from flower and fruit predators, like nocturnal possums, which rely on form and silhouettes rather than colour (compare Figs 6.5 and 6.6). To quote Barlow and Wiens (1977) again, ‘The kind of mimicry which may be established in mistletoes is that in which vegetative structure is such that the whole plant blends cryptically with its background’. In discussing mimicry in mistletoes, perhaps it would be useful to use a more precise term, like ‘form mimicry’. There is no doubt that some similarity between the orientation and shape of

Photo: AN Start.

ringtail possums frequently build their nests in clumps of host-mimicking mistletoe, suggesting that they had no difficulty finding them. Another study (Reid 1997) has cited many cases in which possums do eat mistletoe leaves. Interestingly, Atsatt’s captive brushtailed possum would reach out between the bars of its cage for mistletoe fruits, which it devoured, and Reid reported possums readily devouring both immature and ripe mistletoe fruits as well as flowers. This suggests that one advantage for mistletoes of ‘hiding’ from possums may be reduced predation of flowers and seeds. David Watson (2019) pointed out that host mimicry also occurs in New Zealand mistletoes, where it would have evolved in the absence of arboreal herbivorous mammals (there were none in New Z ­ ealand before A ­ustralian possums were introduced). Nevertheless, in New Zealand, Australian possums eat New Zealand’s ­ native mistletoes so much that they may have played a part in the extinction of one species (Norton 1991). However, Ogle (1997) reviewed possum herbivory of New Zealand’s mistletoes and made a more moderate suggestion. It is now recognised that the lack of pollinating birds is a significant factor. Both Watson (2019) and Atsatt (1983) pointed out that insects, particularly butterfly larvae, are probably more significant predators of mistletoe leaves than possums are. Butterflies usually use chemical, not visual, cues to identify suitable plants on which to lay eggs. It’s unlikely that mistletoes are able to hide from butterflies. While mimicry in mistletoes frequently, but not always, involves the shape of the leaves of parasite and host it never

Fig. 6.5.  In daylight, the colour of Amyema miquelii plants growing in a marri (Corymbia calophylla) stand out over a considerable distance to animals with colour vision, such as birds. 25

Photo: AN Start.

MISTLETOES OF WESTERN AUSTRALIA

a mistletoe’s leaves and those of its host can help the whole plant to blend cryptically with the host. However, mimicking leaf shapes alone does not always achieve crypsis – we again stress the importance of distinguishing mimicry from crypsis. It is possible that the mimicry seen in some species is simply a consequence of the parasitic habit. Atsatt suggested that host hormones that influence leaf shape may be absorbed from the host by mistletoes, resulting in the two species having similarly shaped leaves. But there are numerous cases of mistletoe and host leaves

26

Photo: AN Start.

Fig. 6.6.  The scene depicted in Fig. 6.5 as it might appear to a nocturnal animal like a possum.

Fig. 6.7.  The leaves of Amyema sanguinea var. sanguinea are hard to distinguish from those of its Melaleuca host but its bright red flowers are highly visible to birds.

looking vastly different. There are many unanswered questions involving potential mimicry and crypsis of mistletoes and their hosts, and much more research is needed.

7 Fire and other threats

Photo: AN Start.

Most Western ­Australian plant communities are affected by fire, and the plants that occur in them have evolved various survival strategies. Some, like grasstrees and palms, protect their apical buds deep within their crown. Many eucalypts and shrubs can resprout from their rootstock or from dormant buds protected under thick insulating bark. Banksias and hakeas store long-lived seed in thick woody capsules that open after fire, allowing the seed to fall on a nutrient-rich ash-bed, while many acacias build up long-lived seed-banks in the soil. Mistletoes are perhaps unique among WA flowering plants living in fire-prone habitats in that they can do none of these things. Mistletoe plants (and any fruits they may be carrying) are killed if scorched by a bushfire (Fig.  7.1). And as only fresh fruits are eaten by dispersing birds and the seeds must germinate almost as soon as they’re deposited, most mistletoes have no

Fig. 7.1.  Fire has killed both the host and the mistletoes it supported. Wood-roses show where Amyema miquelii was growing on a marri tree (Corymbia calophylla).

means of survival after severe scorching by a bushfire. Recolonisation of burnt areas requires the dispersal of fresh seeds from unburnt areas. As we saw in Chapter 5, many mistletoes are host-specific, at least to some extent. If an acceptable host can live through a bushfire and regenerate its canopy, as eucalypts do, recolonisation by mistletoes can begin almost straight away. However, if fire kills the hosts (as in many acacias) or even just the above-ground stems (like mallees), recolonisation can’t begin until a new generation of host stems has matured, usually several years after the fire. Birds that distribute mistletoe seeds (Chapters 3 and 9) tend to stay close to their food resources. However, fires can be very extensive, particularly in the Pilbara, Kimberley and deserts, where they often burn for weeks and cover many thousands of hectares (Figs 7.2 and 7.3). This means that the distance between a potential host and a source of seeds may be very large. For this reason, mistletoe recolonisation of burnt areas tends to be very slow. This is a serious challenge when fires become either more frequent or more extensive, which has happened over much of WA. Mistletoes have been incrementally almost eliminated by wildfires over huge areas, particularly in the spinifex-dominated Pilbara and deserts and the tropical savannas of the Kimberley. The continuing decline of mistletoes has cascading consequences for biodiversity 27

Photo: S van Leeuwen.

MISTLETOES OF WESTERN AUSTRALIA

Photo: S van Leeuwen.

Fig. 7.2.  Hot spinifex- (Triodia-) fuelled fires in the Pilbara and other hummock grass-dominated landscapes completely scorch the canopies of most shrubs and trees, thereby killing most mistletoes.

Fig. 7.3.  Few canopies, or mistletoes, survive hot fires that sweep across vast areas like this in the Fortescue Valley, with the Hamersley Range in the distance.

(see Chapter 9). The good news is that no known species (except, perhaps some rare species in the Northern Kimberley bioregion) are at risk of becoming extinct, because all the others have at least some means of persisting somewhere. The most common survival strategy is to include potential hosts that grow in relatively fire-safe refugia such as riparian areas, rocky gorges, scree slopes or rainforest patches. Mulga woodlands (which seldom burn unless there is a lot of spinifex in the understorey) safely support good populations of several mistletoe species, either 28

because the mulga itself is a suitable host or because preferred hosts also find shelter in those woodlands. A second (and safer) way for a mistletoe species to ensure fire is not a risk is to be highly host-specific to hosts that only grow in fire-safe places. A prime example is Amyema thalassia, which only infects mangroves – and mangrove stands very rarely carry fire. A third, less common, strategy adopted by a few species is to have a very wide host range, so that almost any shrubby plant or tree is likely to be an acceptable host. The situation described in Chapter 5, where Lysiana casuarinae was recorded growing on 11 host species in five families in one small area, is a good example. Sadly, this strategy is far from secure – a wildfire has eliminated that population from all its hosts. The strategy may have worked better historically, when Indigenous peoples created mosaics of small burns of different intensities, but it is no longer effective against today’s often severe fires that scorch entire canopies over vast areas. Some species, such as A.  benthamii, employ all three strategies in different parts of their ranges. In the fire-prone Kimberley this species has a remarkably wide range of hosts, some of which grow in relatively fire-safe places. In southern WA it has only one host, the relatively fire-safe kurrajong (Brachychiton gregorii). In the Pilbara it has three recorded hosts, one of which, Terminalia circumalata, grows in fire-safe, gravelly riverbeds. There are a few cases of mistletoes that can sometimes survive a fire. As we saw in Chapter 4, A. sanguinea can extend ramifying haustorial tissue within the stems of the eucalypt host’s branches. After canopy

scorch, new mistletoe shoots may emerge alongside new shoots of the regenerating host. Diplatia grandibractea also sometimes regenerates in this way (Fig. 4.9). This is not a fully secure survival strategy, however. A population of D. grandibractea near M ­ illstream in the Pilbara was eliminated by a fire that was so hot that it killed all the host’s infected branches. An intriguing type of haustorium with fire-survival capability can sometimes be seen on Dendrophthoe acacioides, one of the most common species in fire-prone savannas in the Kimberley. This species grows on a very diverse range of hosts, including boabs (Adansonia gregorii), the kapok bush (Cochlospermum fraseri) and corkybark wattles (Vachellia spp.). These and some other host species can survive bushfires, regenerating scorched canopies from dormant buds under the thick insulating bark. In this case, the mistletoe’s haustorium is a ball type (Chapter 4) but it is common for the host component to grow upward and around the mistletoe component, so the parasite’s haustorium sits in a ‘cup’ of insulating host tissue (Fig. 7.4). A bushfire may defoliate the host and kill all the mistletoe’s exposed branches but, provided the host’s infected branch survives the fire, the mistletoe haustorium, sheltered in its cup of fire-resistant host tissue, can sprout new branches. Fire is probably the most serious and ubiquitous threat to mistletoes, but there are others. Land clearing has been the most pervasive threat in agricultural areas, where the threat of fire can be somewhat alleviated by stricter fire management practices and road verges can provide useful refugia. The benefit of fire exclusion is

Photo: AN Start.

7 – Fire and other threats

Fig. 7.4.  A section of Cochlospermum fraseri infected by Dendrophthoe acacioides. The haustorium is nestled in a cup of fire-resistant host tissue. The mistletoe has intercepted the host branch which has died and broken away beyond the mistletoe.

clearly seen in urban environments, notably in some wheatbelt towns. Perhaps surprisingly, some feral animals can be a threat. In the Goldfields and southern deserts, the only native host for A. benthamii is the kurrajong (B. gregorii). Unfortunately, camels browse kurrajongs and their mistletoes so heavily that there are concerns for the survival of the trees themselves, and their mistletoes (Edwards et al. 2008). Rabbits provide another example, this time of an indirect long-term threat. Around the Nullarbor’s treeless plain the dominant tree, western myall (Acacia papyrocarpa), is host to Amyema quandang. Western myall seedlings are heavily grazed and killed by rabbits. Although the trees are very long-lived, they eventually die. The only significant myall regeneration since rabbits became abundant over a century ago has occurred when diseases such as myxomatosis and rabbit haemorrhagic disease (RCD) temporarily decimated rabbit numbers. Rabbits don’t kill mature myall trees, but unless the rab29

MISTLETOES OF WESTERN AUSTRALIA

bit population is permanently reduced to allow regeneration, A. quandang is in danger of eventually running out of hosts in that area. Another, perhaps more worrying, situation where rabbits are threatening a mistletoe concerns Korthalsella leucothrix, which is known in WA from very few sites. At one site the mistletoes grow on various Acacia

30

species on a large granite rock. The vegetation on the rock is fire-protected but the hosts are relatively short-lived and many are senescing. As on the Nullarbor, rabbits abound on and around the rock and are killing the Acacia seedlings there. They are putting a rare and localised mistletoe at risk of running out of hosts at one of its few known localities.

8 Biogeography of Western ­Australia’s mistletoes Loranthaceae In Western A ­ ustralia, mistletoes occur in two families. Most are in the family of showy mistletoes (the Loranthaceae) which occurs in South America, Africa, India, China and other parts of southeast Asia to Papua New Guinea, as well as in A ­ ustralia, New Zealand and many islands, most of which are Gondwanan fragments. The family probably arose in that ancient southern continent before it broke apart. A few species have moved north, off the Gondwanan plates into southern North America and southern Europe. Genetic evidence suggests that the more ‘primitive’ mistletoes in this family still occur on the landmasses that were once part of Gondwana (i.e. Africa, South America, New Zealand and ­Australia) with isolated occurrences elsewhere in south-east Asia and the Pacific. The only ‘old’ or ‘primitive’ member of the family in WA is the Christmas tree, Nuytsia floribunda, which is not a mistletoe (see Chapter 2). Apart from some ‘old’ genera of the eastern seaboard, whose ancestors were probably on our plate when Gondwana broke apart, A ­ ustralia has eight genera of showy mistletoes that probably came here from south-east Asia via New Guinea and Cape York. Five of these eight ‘new’ genera still occur in those parts but three of the eight are endemic to ­Australia. Benthamina (of coastal Queensland and New South Wales) is closely related to the large Asian

and African genera Scurrula and Taxillus. Diplatia has probably evolved in A ­ ustralia from the diverse and widespread genus Amyema, which has undergone extensive radiation here. The third, Lysiana, is the only endemic ‘new’ genus without obvious strong links to New Guinea and south-east Asia. Nevertheless, its genetic structure of 12 chromosomes and its floral anatomy suggest it is probably derived from ancestral south-east Asian stock. This pattern suggests that the ancestors of most ­Australian loranthaceous mistletoes left Gondwana on the Indian plate. After that plate collided with Eurasia, they were able to disperse through Asia, including south-east Asian islands. Sea level changes since then have resulted in many changes to the connectivity of landmasses in the region, including ­Australia’s links to New Guinea through Cape York. Thus, there have probably been several occasions when the descendants of those Indian plate mistletoes have been able to enter A ­ ustralia. Once here, there have been extensive opportunities for diversification as species spread and adapted to new and challenging environments. This interpretation is supported by biogeographical patterns within ­ Australia. Table 8.1 shows the number of genera in each ­Australian state, the Northern Territory and, for comparison, Papua New Guinea. It is hardly surprising that all eight ‘new’ ­Australian genera of showy mistletoes occur in Queensland because, in the 31

MISTLETOES OF WESTERN AUSTRALIA

Table 8.1.  Number of ‘new’ Loranthaceous genera in ­Australian states and territories Papua New Guinea

QLD

NSW

NT

VIC

SA

TAS

WA

Amyema

✓

✓

✓

✓

✓

✓

×

✓

Lysiana

×

✓

✓

✓

✓

✓

×

✓

Diplatia

1

×

✓

✓

✓

×

✓

×

✓

Dendrophthoe

✓

✓

✓

✓

✓

×

×

✓

Decaisnina

✓

✓

×

✓

×

×

×

✓

Amylotheca

✓

✓

✓

×

×

×

×

×

Benthamina

×2

✓

✓

×

×

×

×

×

Dactylophora

✓

✓

×

×

×

×

×

×

Total

5

8

6

5

3

3

0

5

The ­Australian Capital Territory is included with New South Wales. The ‘primitive’ genera that were probably on the A ­ ustralian plate when it separated from Gondwana are omitted as none occur in WA. Thus, the included genera are those whose ancestors are thought to have migrated into northern ­Australia. ✓ = present, × = absent. 1 ­Australian endemic genus, derived from Amyema. 2 ­Australian endemic genus, related to the Asian genera Taxillus and Scurrula.

Table 8.2.  Latitudinal cline in the number of loranthaceous mistletoe genera in each WA botanical province Province Mistletoe genus

Torresian (tropical)

Eremaean (arid)

South-west (mesic)

Amyema

✓

✓

✓

Lysiana

✓

✓

✓

Diplatia

✓

✓

×

Dendrophthoe

✓

×

×

Decaisnina

✓

×

×

No. of genera

5

3

2

No. of species

24

22

9

Symbols; ✓ = present, × = absent.

past, there have been land bridges between Cape York and New Guinea. From Cape York they spread south but, the number of genera diminishes: New South Wales has six, Victoria three and modern Tasmania has none, although mistletoe pollen has been found in cores of ancient Tasmanian lake sediments. Tasmanian mistletoes were probably eliminated by the last ice age and the island has not been recolonised as rising sea levels flooded Bass Strait and the connection with the mainland was lost. The mistletoes also spread west across northern A ­ ustralia and from there some 32

moved south, adapting to A ­ustralia’s more arid interior. The pattern of decreasing numbers of genera and species along the eastern seaboard is mirrored in the Northern Territory and WA. The ­Northern Territory has five genera and South ­Australia has three, all of which also occur in the Northern Territory. Table 8.2 shows the number of genera in WA’s three botanical provinces, ranging from the northern (Torresian Province) through the arid mid-latitudes (Eremaean Province) to the temperate-mesic (South-west Province).

8 – Biogeography of Western ­Australia’s mistletoes

The decreasing north–south trend in numbers of genera in WA is mirrored in species numbers. There are 24 species in the Kimberley and 23 of those occur in the wettest bioregion, the Northern Kimberley. There are only nine species in the Southwest Province and none in its wettest part, the Warren bioregion (which includes the karri forests) or the western part of WA’s south coast. However, there are three in the more arid Hampton bioregion at the eastern end of WA’s south coast.

The mistletoe flora of the arid ­Eremaean Province is relatively rich (22 species versus 24 in the Torresian and nine in the South-west Provinces). Although lacking the recent immigrant genera Dendrophthoe and Decaisnina and other Torresian endemic species, the Eremaean flora is augmented by species that evolved elsewhere in A ­ ustralia’s arid areas and moved west. Table 8.3 shows the number of species in each bioregion, which emphasises the relative richness of the mistletoe flora of the

Table 8.3.  Number of species in each WA bioregion Province

Bioregion

Code

Area (ha)1

No. of species in WA

Torresian

Central Kimberley

CEK

7 675 587

15

Dampierland

DAL

8 360 871

15

Northern Kimberley

NOK

8 420 100

23

Ord Victoria Plains

OVP

(12 540 703)

14

Victoria Bonaparte

VIB

(7 301 242)

15

Carnarvon

CAR

8 430 172

11

Central Ranges

CER

(10 164 044)

8

Coolgardie

COO

12 912 209

8

Gascoyne

GAS

18 075 257

14

Gibson Desert

GID

15 628 918

9

Great Sandy Desert

GSD

(39 486 135)

4

Great Victoria Desert

GVD

(42 246 564)

7

Hampton

HAM

(1 088 198)

3

Little Sandy Desert

LSD

11 089 857

10

Murchison

MUR

28 120 554

14

Nullarbor

NUL

(19 722 774)

6

Pilbara

PIL

17 823 126

14

Tanami

TAN

(25 997 277)

5

Yalgoo

YAL

5 087 577

7

Avon Wheatbelt

AVW

9 517 104

8

Esperance Plains

ESP

2 921 327

3

Geraldton Sandplains

GES

3 142 149

7

Jarrah Forest

JAF

4 509 074

6

Mallee

MAL

7 397 559

3

Swan Coastal Plain

SWA

1 525 798

5

Warren

WAR

844 771

0

Eremaean

South-west

Bioregional data obtained from IBRA 7 (http://www.environment.gov.au/land/nrs/science/ibra). 1 Figures in brackets are total areas of those bioregions that extend beyond WA’s borders.

33

MISTLETOES OF WESTERN AUSTRALIA

Torresian Province; in WA, this equates to the Kimberley.

Santalaceae Mistletoes in the other WA family, the Santalaceae, are diverse and widespread in northern parts of the ‘old’ and ‘new’ worlds. From there, some have spread southwards into Central America, Africa, Asia and A ­ ustralia. They have reached New Zealand and even some Pacific islands. In WA, there are five species in two genera: Viscum, with three northern (Kimberley) species and Korthalsella, with two southern species. You may recall we opened the introduction with the ‘English’ mistletoe, Viscum album. Two of our three species of Viscum (V. articulatum and V. ovalifolium) extend to India, Burma and China, but our third species (V. whitei) is endemic to northern ­Australia and has presumably evolved on our continent. The other WA genus, Korthalsella, occurs from Africa through Asia to A ­ ustralia but it does not reach Europe. Astonishingly, some species have colonised islands in the Indian Ocean (e.g. ­Madagascar) and the Pacific (from Hawaii to New Zealand). This enormous distribution is unique among mistletoe genera. It is worth contemplating how they managed it. All Korthalsella species are diminutive plants; those in WA seldom reach 12  cm in length. They are widespread in eastern A ­ ustralia. Their flowers are minute and their ripe fruits, which are little more than 1 mm long, contain a single tiny seed coated with a sticky substance (Fig 3.14). Like the closely related North American and Asian dwarf mistletoes in the genus 34

Arceuthobium, many Korthalsella species have been shown to use hydrostatic pressure to expel their seeds explosively. Observations on both WA species suggests they do so here. Up to 50 plants of K. leucothrix can occur in a single host canopy, but a gap of 2–3 m between host canopies is enough to prevent a crossing. Nevertheless, longer distance dispersal must occur sometimes. There is indirect evidence that birds are responsible for long-distance transfer of Korthalsella species. The purported carriers fall into two groups: small passerines that forage in tree canopies and seabirds that nest on oceanic islands, some of which have naturally occurring Korthalsella species. However, in both cases, the dispersal would have been inadvertent; that is, probably brought about by the carrier preening off seeds that had stuck to its legs or feathers and wiping them onto a branch, rather than passing them through their digestive system in the way mistletoebirds disperse showy mistletoe seeds. (See Chapter 3 for a discussion of local Korthalsella seed dispersal mechanisms.) The two WA species of Korthalsella belong to the group of species presumed to be transferred by small passerine birds. One, K. leucothrix, is known in WA from three locations between the north-­eastern edge of the wheatbelt and the South ­Australian border. It appears to be more common in arid parts of South ­Australia but does not occur further east or further north. The other species, K. arthroclada, is only known from Melaleuca trees growing in and around one group of salt lakes near Eneabba, north of Perth. It is probably a

8 – Biogeography of Western ­Australia’s mistletoes

WA endemic that evolved locally from ancestors that occurred elsewhere. However, there have been suggestions that it is related to K. dacrydii from south-east Asia or K.  salicorniodes from New Z ­ ealand.

Transfer of those species to south-western WA does not correspond with any known bird migration routes and seems unlikely. The solution will need work on the genetics of these and other related species.

35

This page intentionally left blank

9 The relevance of mistletoes Watson and Herring (2012) demonstrated experimentally that the removal of mistletoes resulted in a decline of ­­Australian woodland bird species, including, surprisingly, species that did not feed on mistletoes. Their explanation was that, in eucalypt woodlands where there are mistletoes, the mistletoe leaves are shed sooner and decompose faster than the hard sclerophyllous leaves of the eucalypts. A flow-on effect of faster recycling of nutrients was healthier eucalypts, leading to better quantity and quality of food resources for insects and hence the birds that preyed on them. The larvae of many butterflies and moths feed on mistletoes (Figs 1.1, 1.2 and  9.1).

Photo: Jean and Fred Hort.

All A ­ ustralian mistletoes are native and are natural components of our rich floral heritage. In reviewing the contributions that mistletoes make to global biodiversity, David Watson (2001) credited them with the status of ‘keystone’ species. In architecture, a keystone is the special stone placed at the apex of an archway that locks all the other stones in place and prevents the arch from collapsing. In ecological terms, keystone species serve an analogous function. If they are removed from an ecosystem, there is danger of a cascading collapse of interdependent members of the community in which they lived.

Fig. 9.1.  A larva of the mistletoe-dependent moth Comocrus behrii on Amyema preissii. 37

MISTLETOES OF WESTERN AUSTRALIA

When a natural community loses its mistletoes, it loses an important food resource on which many animals depend, at least at certain times of year. The two commonest mistletoes in south-western ­Australia, Amyema miquelii and A. ­preissii, flower and fruit from summer through autumn into early winter (Figs 9.2 and 9.3), a time when food resources for honeyeaters, silvereyes and mistletoebirds are scarce (Napier et al. 2014). Besides supporting birds with food in the form of insects, fruit and nectar, the branches of many mistletoes form dense clumps and are ideal places for birds to nest. Cooney et  al. (2006) recorded 217 ­Australian bird species nesting in mistletoe clumps and it has recently been shown that, in the Strzelecki Desert, red kangaroos seek out shady spots under mistletoe

Photo: AN Start.

Some feed on nothing else; some are so iconic that they have featured on ­Australian postage stamps illustrating the nation’s wildlife heritage. Watson (2001) tallied insects from six orders that are known to feed on ­ Australian mistletoes. These in turn are food for other animals, including other insects and birds, both of which may be important to the ecology of components of the communities in which they live. So important are mistletoes to native insects that the Brisbane-based Butterfly and Other Invertebrates Club published a book on the mistletoes of subtropical Queensland, New South Wales and Victoria (Moss and K ­ endall 2016). An entomological club publishing a book on mistletoes shows the value of mistletoes as larval food plants for some of the insects in which the club is interested.

Fig. 9.2.  The masses of flowers of Amyema miquelii provide an important food resource for many nectar-feeding birds at times when other food is scarce. 38

Photo: AN Start.

9 – The relevance of mistletoes

Fig. 9.3.  The masses of fruits of Amyema preissii provide an important food resource for many frugivorous birds, at times when other food is scarce.

clumps to rest through the heat of the day (Chu et al. 2021). In Western ­Australia there is even a weevil, found only on mistletoes, that avoids the attention of potential predators by resembling a bird dropping. In Chapter 7 we indicated that mistletoes have disappeared from vast tracts of the spinifex deserts. What else has gone with them? We may never know. Given all the benefits that mistletoes contribute to healthy ecosystems, it seems perverse that they can be regarded as undesirable pests that need to be controlled. Nevertheless, they are often perceived that way. Several authors have explored this conundrum, including a brief discussion under the title ‘Friend

or foe?’ in the magazine Western Wildlife (Start 1999). As Table 9.1 shows, mistletoes are only perceived as a pest in the south-west of WA. In all cases (except pricklybark in kwongan, which is discussed below) the problem occurs in highly disturbed sites where natural processes including fire regimes and hydrology have been altered and pollutants including phosphate-rich fertilisers have drifted in. Such issues are typically only found in road verges, paddocks and small patches of remnant vegetation. These sites also tend to be highly visible, often giving casual observers the impression that the phenomenon may be more extensive than it really is (Fig. 9.4). 39

MISTLETOES OF WESTERN AUSTRALIA

Table 9.1.  Western ­Australian mistletoe species regarded as pests Affected hosts

Locations

Situation

A. miquelii

Wandoo (Eucalyptus wandoo)

Along Albany and Brookton Highways and the western wheatbelt. Also, southeastern suburbs of Perth.

Road verges, paddocks and remnant vegetation.

A. miquelii

Marri (Corymbia calophylla)

Some of Perth’s south-eastern suburbs (Gosnells/Martin, Byford, Mundijong etc.).

Paddocks and remnant vegetation.

A. miquelii

Pricklybark (Eucalyptus todtiana)

Geraldton Sandplains south from Eneabba.

Patchy in kwongan heaths including many conservation reserves.

A. preissii

Jam wattle (Acacia acuminata)

Northern and western wheatbelt, notably Toodyay-York and MullewaNorthampton areas. Also, along parts of Albany Highway north of Mt Barker.

Patchy. Usually on road verges.

A. preissii

Eastern states Acacia species

Perth, notably the hills and adjacent suburbs.

Gardens and ornamental escapes.

A. fitzgeraldii

Various acacias and notably jam wattle (Acacia acuminata)

Wheatbelt, north from Mullewa and Northampton.

Patchy. Usually road verges and remnant vegetation.

Photo: AN Start.

Mistletoe

Fig. 9.4.  A eucalypt growing in a paddock beside a road is heavily infected by Amyema miquelii. Maybe the tree is in poor health from phosphate-rich fertilisers and an altered hydrology, which may have made it susceptible to the heavy parasite load.

There is also a psychological element: like tapeworms, ticks and fleas, mistletoes are parasites and hence must be inherently 40

bad and therefore eradicated. This attitude can be found among teachers, conservation groups and farmers alike. For

9 – The relevance of mistletoes

example, a Scouts project aimed to eliminate mistletoes from marri trees in a conservation reserve near Perth. The Scouts got good coverage in the local newspaper but, thankfully, failed in their objective. If all parasites are bad, then WA Christmas trees and quandongs must be bad too. Perceptions, even wrong ones, are very real to those who hold them, so it is worth exploring the situations represented in Table 9.1 a little more. It seems that mistletoes can sometimes deal a coup de grace to their hosts. Some eastern A ­ustralian acacias that were imported for their ornamental values (e.  g.  Acacia baileyana and Acacia podalyrifolia) are so susceptible to infestations by mistletoes near Perth that they may be killed. Of course, in this case this may not be a bad thing, as both the wattles have become weeds. Where garden specimens are nurtured, mistletoes can easily be pruned. It is gratifying that a few gardeners who are keen to add diversity and colour to their gardens have ‘planted’ mistletoe seeds on their acacias! Three native WA hosts are perceived to experience more severe problems from mistletoe infestations than most: the jam wattle (Acacia acuminata), wandoo (­Eucalyptus wandoo) and marri (Corymbia calophylla).

A. preissii, jam wattle and road verges Like many other acacias, the jam wattle is a relatively short-lived tree. Old stands become senescent and eventually collapse. Regeneration only occurs by seed, and fire is the principal factor triggering regeneration. Pest-level infections of the mistletoe A.  preissii are common on stands of jam

wattle on some road verges, and there is no doubt that the mistletoes have hastened the final demise of some trees. But such high infection levels are not seen in jam wattle growing in more extensive areas of natural bushland. Proponents of mistletoe control often cite dead or dying plants with heavy mistletoe infections but overlook the effect of natural senescence and other factors, notably altered fire regimes, weed invasion and phosphate pollution, which contribute to the problem and increase the difficulty of finding solutions. Blaming the mistletoes distracts from the more fundamental pressures the jam wattles face. Regenerating fires (which would have both stimulated wattle germination and reduced mistletoe populations) have largely vanished from road verges in agricultural landscapes, and weedy grasses are now so prolific that they may outcompete most jam wattle seedlings. On many road verges the jam wattle is already dying from causes that will continue with or without mistletoes.

A. miquelii, marri and wandoo Marri and wandoo eucalypts differ from jam wattle in being long-lived trees that usually survive hot fires by regenerating their canopy from epicormic buds. However, as with jam wattle, the absence of fires on road verges and in paddocks that are hot enough to occasionally scorch the trees’ canopies and control mistletoes is a likely reason for excessive infections. Mistletoe occurrence at ‘pest’ levels is usually patchy. In a wandoo-lined road reserve, there may be sections of a few hundred metres every few kilometres in which infestations seem excessive. On both road reserves and in paddocks, some trees may 41

MISTLETOES OF WESTERN AUSTRALIA

be so heavily infected that their health appears to be affected. In extreme cases the trees appear to have been killed by mistletoes. But look carefully – away from paddocks and road verges, there will often be ‘clean’ trees among the infected ones. One explanation is that the mistletoes are exploiting trees which, like ageing jam wattle, may be unhealthy for reasons other than mistletoe infection, just as people who have a cold are more susceptible to other infections. Perhaps, with time, nature will find a balance. Meanwhile, birds will no doubt enjoy an abundance of mistletoe flowers, fruits and insects (and perhaps be accused of spreading the dreaded parasite along the road).

A. miquelii and pricklybark

Amyema miquelii is occasionally regarded as a serious pest of pricklybark (Eucalyptus todtiana) in kwongan communities north of Perth. Pricklybark generally occurs as a sparse overstorey tree in kwongan heathlands. Many dead trees have numerous wood-rose scars, and heavily infected live trees may have sick-looking canopies. Fire is often rare in these communities. However, in places where fires kill the

42

mistletoes, the trees regenerate healthy crowns. It appears that if fire is permanently excluded, mistletoes could eliminate or reduce pricklybark populations in some communities. However, occasional fires likely result in healthy pricklybark populations. The impact of fire on mistletoe loads and canopy health in these communities should be investigated experimentally. This might also shed light on the contentious issue of appropriate fire management in kwongan. In this book, we have cited the role of fire as both a problem and a solution, because it is probably the most important single factor controlling the density and abundance of mistletoes (see ­Chapter 7). So, are mistletoes friend or foe? Mistletoes are intrinsically important members of natural, healthy native plant communities. When such communities are left undisturbed they are enhanced, not debilitated, by mistletoes. However, in places where human activities have created the conditions for overdominance of mistletoes, they may need to be controlled, preferably by restoring the conditions under which the whole community, including its mistletoes, can thrive.

10 Keys to mistletoe families, genera and species in Western ­Australia This chapter features keys to aid in the identification of mistletoes in Western Australia. Words in italics (other than plant names) can be found in the Glossary. 1.  Key to the families of Western A ­ ustralian mistletoes 1a

Plants appearing leafless (leaves reduced to minute scales).

Santalaceae (part)

1b

Plants with normal leaves.

2

2a

Flowers very small (2 cm long, not crowded (internodes on mature stems >1 cm long).

A. preissii

3b

Leaves 1 cm long).

A. microphylla

4a

Leaves slender, usually 1 mm diameter. Inflorescence a 2-rayed umbel, each ray supporting a diad of 2 robust flowers. Hosts Proteaceae.

5

5a

Mature leaves glabrous, dark green. Hosts Hakea.

A. gibberula var. tatei

5b

Mature leaves with a pale indumentum. Hosts Hakea or Grevillea.

A. gibberula var. gibberula

6a

Mature leaves with an obvious, if short, indumentum.

7 but see 22a

6b

Leaves glabrous.

11

7a

Inflorescence a head of 1 central and 3(4) lateral, sessile flowers (very rarely expanded into a panicle). Flowers green with red filaments. Ripe fruits globular, pink. Leaves usually 6:1). Inflorescence with 2 rays with flowers in diads. Hosts Myrtaceae (usually Corymbia). Kimberley only.

A. eburna

8b

Leaf length:width ratio usually 5mm long, sometimes a little shorter), each ray supporting a single triad of flowers, the lateral flowers with very short but definite pedicels.

A. hilliana

44

10 – Keys to mistletoe families, genera and species in Western A ­ ustralia

10b

Leaves usually 3–4 cm long. Inflorescence a head with 6 sessile flowers arranged in 2 rows of 3 flowers.

A. maidenii

11a

Hosts (in WA) Myrtaceae, usually Eucalyptus, Corymbia, rarely Melaleuca, or Callistemon.

12

11b

Hosts from many families but not (in WA) Myrtaceae.

16

12a

Inflorescence an umbel with 2 rays, each with a pair of flowers (i.e. diads).

13

12b

Inflorescence (when normally developed) an umbel with 1 or 3 (never 2) flowers per ray. (Beware damage that has removed 1 or 2 flowers – look for scars.)

14

13a

Large mistletoe with bronze-green foliage and robust red flowers, sometimes with a scurfy indumentum on young shoots and/or flowers and fruits. Fruits barrel-shaped. Hosts mostly Corymbia but also Eucalyptus. Kimberley, Pilbara and Little Sandy Desert.

A. bifurcata

13b

Leaves usually not distinctly bronze-green. Fruits pear-shaped. Northern Kimberley including some islands (a poorly known species).

A. pyriformis

14a

Usually large, tear-shaped plants with yellowish to bronze-green, pendulous stems and foliage (but colour can vary with season and host). Epicortical roots absent. Inflorescence commonly with 4 (3–7) rays, each with a triad of pedicellate red flowers. Style without a prominent black stigma. Throughout WA.

A. miquelii

14b

Leaves never bronze-green. Epicortical roots usually present. Inflorescence an umbel with 1 flower per ray. Style with a prominent black stigma.

15

15a

Stems and leaves typically strongly pendulous, the latter somewhat falcate, linear to narrowly lanceolate with a length:width ratio >4:1. Hosts Myrtaceous trees (Eucalyptus, Corymbia or Melaleuca).

A. sanguinea var. sanguinea

15b

Stems and leaves typically not strongly pendulous, the latter broadly elliptic, thick and leathery with a length:width ratio 2.5:1.

A. dolichopoda

24b

Leaves elliptic to ovate, somewhat leathery, length:width ratio (4)–5:1

D. brittenii

3b

Leaves broadly lanceolate, length:width ratio variable but usually 30 mm long, almost completely enclosing the buds and flowers.

D. grandibractea

1b

Round, much-branched plants. Bracts