A revision of the genus Swainsona (Fabaceae)

155 Pages·1993·55.4 MB·English

by  Joy Thompson

Thompson, Swainsona 427
A revision of the genus Swainsona (Fabaceae)
Joy Thompson
Abstract
Thompson, Joy (National Herbarium of Neiu South Wales, Royal Botanic Gardens, Sydney, NSW,
Australia 2000) 1993. A revision of the genus Swainsona (Fabaceae). Telopea 5(3): 427-581. The
genus Swainsona is revised. A key to the 85 species precedes a formal treatment of these taxa,
of which sixteen are new (Swainsona calcicola, S. complanata, S. cornuta, S. disjuncta,
S. eremaea, S. fuscoviridis, S. halophila, S. longicarinata, S. longipilosa, S. paucifoliolata,
S. perlonga, S. rostrata, S. rotunda, S. sejuncta, S. tanamiensis, S. vestita) and two of altered
rank (S. acuticarinata (A. Lee) J. Thompson, S. elegantoides (A. Lee) J. Thompson). The species
are placed in groups according to perceived relationship and keys provided to groups and to
species within each group. The genus is restricted to Australia except for one endemic New
Zealand species.
Introduction
This work was undertaken as a precursor to a treatment of the genus for the 'Flora of
Australia'. A.T. Lee (1948) published a revision of the genus that was sound and
perceptive. She had little access to herbaria other than NSW, apart from a visit to
MEL, but received co-operation from other workers such as J.M. Black (AD) and C.A.
Gardner (PERTH) so that she was able to treat the genus on an Australia-wide basis.
Her erudition and their examination of critical specimens at K facilitated her accurate
use of nomenclature that still stands for most species. Unfortunately at the time Lee
was working there had been little botanical exploration of those parts of central and
western Australia soon to be opened up as vehicular routes and for mining explora¬
tion and development. The collections built up since that time have included many
new species and have shown some of Lee's taxa to need redefinition.
Our lack of information about important aspects of the biology of Swainsona should
not invalidate speculation about the position of this genus in the Fabaceae. There can
be no doubt that it is closely related to genera placed in the tribe Galegeae by Polhill
& Raven (1981). It is most unfortunate that Polhill did not refer to Lee's (1948) pub¬
lication and apparently assumed Swainsona to be well represented by the relatively
unmodified S. galegifolia. As a result of this assumption his keys to the subtribes, and
the genera within them, are nullified by different species of Swainsona that run down
in almost all key-leads.
Tire relationships of genera within this tribe are not well understood. Over the years
species have been moved from genus to genus, and genera split and reunited. Barneby
(1964) gave the history of the development of understanding of Astragalus (to quote
Bentham (1866), 'a very numerous genus') and Oxytropis ('a considerable genus'), and
it seems more than coincidental that the majority of character-states used by Sander¬
son (1991) in his cladistic analysis of 113 taxa of North American Astragalus would be
those chosen for a similar work on Swainsona to reflect apomorphies found in Aus¬
tralasia. The parallels between evolutionary changes within this well known group of
taxa and those that have occurred in Swainsona are striking and Sanderson (1991), in
phylogenetic studies in Astragalus, also discussed problems of relationship. It is un¬
fortunate that so many genera of the Galegeae are still little known, and obvious that
428 Telopea Vol. 5(3): 1993
they are now delimited by convention rather than by sound taxonomy. Lee suggested
that 'Swainsona is one of a group of Leguminosae whose separation is more arbitrary
than most7, and Bameby that 'the question of generic limits in the Galegeae is fraught
with difficulty'. I am accepting Sioainsona as a distinct genus, as circumscribed here,
on the basis of its geographic isolation from those genera on other continents that
appear to be closely related to it. I do not consider it to be more closely related to
African genera than to those of North America. Swainsona's uniform chromosome
number of 2n=32 is not common in other genera of the tribe where 2n=16 is common
and often uniform. Clianthus and the genera separated in Polhill & Raven (1981) as
the tribe Carmichaelieae have a uniform number of 32, and it is perhaps here that
Szvainsona's closest relatives are to be found. Much remains to be done before the
taxonomic structure of this part of the Fabaceae is well understood and, within Swain¬
sona, much to be elucidated; but now that those previously little-known areas inhab¬
ited by many species of Swainsona have become more accessible, observation will
become a less daunting task.
Swainsona is well known as a cause of stock-poisoning, the poisonous principle, swain-
sonine, affecting the nervous system. Swainsonine is also found in the genus Astra¬
galus (Fellows 1989). It is not always known, owing to past widespread misidentifica-
tion and contradictory specimen notes, which species are involved.
A few species of Swainsona have been cultivated in gardens of Europe, often by
grafting onto a Colutea stock, but they are rarely seen in horticulture now, in spite of
their beauty, perhaps owing to their failure to persist.
It can be readily observed that many, perhaps most, species are erratic in their occur¬
rence, carpeting the landscape after significant rain in areas of low and irregular
rainfall. A few are recorded as being found after fire, even after many decades of
apparent absence.
Diagnostic Characters
In generic and specific descriptions I have tried to cover much of the variation to be
found in this genus, but for many features this has not been feasible. Like many taxa
that can successfully invade inhospitable areas, Swainsona may live its life in a short
time while conditions are favourable or may persist to become much larger in its
vegetative parts, and more floriferous, when conditions allow. It is impossible and
useless to cover all variation in dimensions in these cases and in the case of many
species this is as yet unknown, each new collection altering perception of the norm.
The treatment presented here has suffered from many of the problems that beset Lee,
because Swainsona is a genus with features of significance that are difficult to assess
in the herbarium. Where possible I have followed the terminology used by Lee but
the larger range of specimens available now has shown additional useful features,
and that some character-states are more variable and therefore less useful than was
previously supposed. As the base of the plant is rarely collected, it is difficult to
ascertain whether one or several stems arise there, where the stem-bases are in rela¬
tion to soil-level, or even whether the plant is annual or perennial, in many cases.
Bameby (1964) describes the roots and stem-bases of Astragalus and regards differ¬
ences as phylogenetically significant, as does Sanderson (1991). Such observations on
Szvainsotia are too few to allow accurate descriptions for a majority of species. For the
stem descriptions I have had to rely on dried material and live plants will not show
the same ridging. The stem hairs are usually characteristic for the species but are
more variable than was assumed by Lee. Barneby's description of Astragalus vesture
Gardens’Bulletin Singapore 66(1): 3-14. 2014 3
The conservation of Peninsular Malaysian Geostachys
(Zingiberaceae)
K.H. Lau
Forest Research Institute Malaysia,
52109 Kepong, Selangor, Malaysia
[email protected]
ABSTRACT. There are currently 15 species of Geostachys in Peninsular Malaysia of which
13 are hyper endemic to their localities. Four taxa are categorised as Endangered, eight as
Vulnerable and one each as Near Threatened, Least Concern and Data Deficient. Maps showing
the Extent of Occurrence and Area of Occupancy for each species are included. Key issues
and recommendations on conservation measures, particulai'ly with respect to ecotourism, are
discussed.
Keywords. Area of occupancy, assessment, conservation, extent of occurrence, Geostachys,
Peninsular Malaysia
Introduction
Geostachys (Baker) Ridl. is a genus of lesser-known gingers distributed in Vietnam,
Thailand, Sumatra, Peninsular Malaysia and Borneo. Of the 25 taxa currently known
(The Plant List, 2013), 15 occur in higher altitude lush rain forests of Peninsular
&
Malaysia (Lau et al,, 2007; Lau Lim, 2012). This makes Peninsular Malaysia the
genus hotspot. Among the 5 taxa, 3 are hyper-endemic to their localities and the
1 1
other two are quite widespread.
In Malaysia, the main threats to lowland forest are from deforestation and
changes in land use. Despite their preference for higher altitudes, Geostachys habitats
suffer from similar disturbances. While there are several works describing the genus
in Peninsular Malaysia (Ridley, 1899, 1920, 1924; Holttum, 1950; Stone, 1980), none
address the conseivation perspective. This paper discusses threats faced by the species
&
and their conservation status. It also complements an earlier account (Lau Lim,
2012) by presenting further insights into the current conservation issues involving the
genus.
Materials and methods
Dried specimens from Universiti Kebangsaan Malaysia Herbarium (UKMB), Kepong
Herbarium (KEP), UniversityMalayaHerbarium (KLU), Singapore Herbarium (SING)
and Royal Botanic Gardens Kew Herbarium (K) were scrutinized and recorded into
4 Card. Bull. Singapore 66(1) 2014
the Botanical Research and Herbarium Management System (BRAHMS). All species
were assessed using the modified Malaysian Red List Categories (Chua, 2012). The
modification from the standard lUCN (2001) system involves the addition of a Rare
(RA) category. This category has this definition: the taxon is not exposed to any known
TUCN
direct or plausible potential threat and does not qualify under the five criteria
AND OR
occurs in not more than two sites has an Extent of OccuiTence (EOO) not
OR A
more than 100 km- Area ofOccupancy (AOO) not more than 10 knE. taxon data
information sheet was completed for each species and the infonnation was uploaded
into the Malaysia Clearing House Mechanism website (http://www.chm.frim.gov.
EOO AOO
my/). The and were calculated for each species using ArcView GIS 3.2a
software. Forest cover data was obtained from the Third National and State Forest
Inventories (1991-1993). Field efforts were made to gather fresh data and analyse
current threats.
Results
Table 1 summarises the conservation status ofeach species ofGeostachys in Peninsular
Malaysia.
Table 1. Conservation status of Geostachys following the modified Malaysian Red List
Categories.
EOO AOO
Species Category (km^)* (km") Endemicity
VUD2
G. belumensis n.a. 4 Hyper-endemic
G. decurvata EN D2 n.a. 8 Hyper-endemic
NT
G. densiflora 3,172 32 Widespread
G. elegans EC 23,108 24 Widespread
G. erectifrons VU D2 n.a. 4 Hyper-endemic
VUD2
G. leiicantha n.a. 4 Hyper-endemic
VUD2
G. megaphylla n.a. 8 Hyper-endemic
DD
G. montana n.a. 4 Hyper-endemic
EN
G. penangensis B2ab(iii) n.a. 8 Hyper-endemic
G. primulina EN D2 n.a. 4 Hyper-endemic
VUD2
G. rupestris n.a. 4 Hyper-endemic
VUD2
G. secunda n.a. 4 Hyper-endemic
VUD2
G. sericea n.a. 4 Hyper-endemic
VUD2
G. tahanensis n.a. 4 Hyper-endemic
END2
G. taipingensis n.a. 8 Hyper-endemic
*EOO can only be calculated ifat least 3 localities are present.
The species calculated to be Endangered (EN) are Geostachys decurvata
(Baker) Ridl., G. penangensis Ridl., G. primulina Ridl. and G. taipingensis Holttum,
Conservation ofPeninsularMalaysian Geostachys 5
Geographical Distribution of EN Geostachys
(Zingiberaceae) in Peninsular Malaysia
LEGEND LOCALITIES
Geostachys decurvata E Geostachys penangensis 1. Penang Hill
^ Geostachys primulina ® Geostachys taipingensis 2. G. Hijau
3. Fraser’s Hill
Fig. 1. Distribution ofEndangered (EN) Geostachys species in Peninsular Malaysia.
6 Gard. Bull. Singapore 66(1) 2014
all of which are hyper-endemic (Fig. 1). Found growing discreetly within the cool
atmosphere of hill forests in Perak, Penang and Pahang, these taxa are endangered
mainly dueto lownumbers ofindividualsperspecies. With the exception ofGeostachys
penangensis, several attempts to locate additional populations of these species in
these forest reserves proved futile. Geostachys decurvata and G. taipingensis occur
sympatrically on Gunung (G.) Hijau, Maxwell’s Flill. These two taxa have both
been very infi'equently collected and G. taipingensis has not been collected for the
herbarium since 1949. Lau (2006) reported the rediscovery of Geostachys primulina
from Fraser’s Hill for the first time in more than 80 years. Geostachys penangensis
can still be found quite abundantly on Penang Hill (Lau, 2004) with populations at
several locations on the hill such as Government Hill, Tiger Hill, Western Hill and
Moniot Road West. However, road widening activities along the main road (Jalan
Tuanku Yahaya Petra) have caused much destruction to the clumps. The increase in
human activities on the hill is of concern and long-term monitoring ofthe impacts on
the populations is required. In addition, the sites of these four taxa are also used as
eco-forest trails and are exposed to the risk ofbeing trampled.
&
The Vulnerable (VU) taxa are Geostachys belumensis C.K.Lim K.H.Lau, G.
&
erectifrons K.H.Lau, C.K.Lim K.Mat-Salleh, G. leucantha B.C.Stone, G. megaphylla
Holttum, G. rupestris Ridl., G. secunda (Baker) Ridl., G. sericea (Ridl.) Holttum and
G. tahanensis Holttum (Fig. 2). In Taman Negara National Park, three taxa are found,
namely Geostachys erectifrons, G. sericea and G. tahanensis. Their distributions are,
however, very localised. So far, Geostachys erectifrons is only known from the area
m
between Pennatang and Kubang campsites (Merapoh) at an elevation of 1000-1200
(Lau et al., 2005). Scattered clumps can be seen on ridges and slopes along both sides
of the main trail. The distribution of Geostachys tahanensis is also veiy restricted,
occurring on a different side ofthe mountain near Kuala Tahan Rest Camp, formerly
m
known as Wray’s Camp, at about 1100 above sea level. Geostachys sericea has a
wider distribution on both the Kuala Tahan and Merapoh sides, but is loosely scattered
from the lower montane forest to upper montane forest. Until more populations are
discovered beyond these well-used trails and spots, they remain vulnerable as they
are at risk of being trampled or collected. For Geostachys belumensis, some of
the populations in Gerik are now likely protected since 18,866 ha of the Amanjaya
Forest Reserve (FR) was gazetted in May 2013. The area has also been identified
as an important wildlife corridor under the Central Forest Spine (CFS) Masterplan.
This provides protection to many other flora and fauna species along the East-West
Highway. Recent discoveries include populations in Piah FR (Perak) and G. Tebu
(Terengganu) (C.K. Lim pers. obs.). However, the type location of this taxon, which
is in Gerik, is showing signs of degradation due to encroachment. In particular the
surrounding area has been cleared, probably for the harvesting of non-timber forest
products. Geostachys leucantha and G. megaphylla are found in G. Ulu Kali (Genting
Highlands) and G. Brinchang (Cameron Highlands) respectively.Although there are no
immediate direct threats to these populations, the environmental impacts arising from
non-forest land use will likely affect the viability of the populations. Infrastructure
projects, such as for resorts, power lines and slope enhancement structures, as well as
Conservation ofPeninsularMalaysian Geostachys 1
Geographical Distribution of VU Geostachys
(Zingiberaceae) in Peninsular Malaysia
100^00’ 101*00' 102*00' 103*00' 104*00'
o
O- -Oo
o O
E
1 : 3,500,000
O
CJ>
O- -ooo
o
CD
O
Oi
O- -ooo
D
o
o-
o
sr
o
CO
O- -Oo
0 o
CO
o
K)
O- -ooo
0
CN Forest cover as in National
Forest Inventory (1991-1993)
III
50 Kilometers
NFI III Courtesy of Forestry Department Peninsular Malaysia SINGAPORE
100*00’ 101*00' 102*00' 103*00' 104*00'
LEGEND LOCALITIES
Geostachys befumensis ® Geostachys rupestris 1. G. Jerai
® 2. Amanjaya FR
Geostachys erectifrons Geostachys secunda
® 3. G- Brinchang
Geostachys leucantha Geostachys sericea
® 4. G. Batu Puteh
Geostachys megaphylla Geostachys tahanensis
5. G. Ulu Kali
6. Taman Negara National Park
Fig. 2. Distribution ofVulnerable (VU) Geostachys speeies in Peninsular Malaysia.
8 Gard. Bull. Singapore 66(1) 2014
forest clearance for agriculture, especially around Brinchang town and above, have
imposed great pressures on the health of the species’ populations. For Geostachys
mpestris, which is found on G. Jerai (Kedah), dumpsites at forest fringes are slowly
degrading the quality ofthe habitat where clumps once grew. Part ofthe forested area
on G. Jerai serves as a tourist attraction. In the pristine forest ofG. Batu Puteh (Perak)
Geostachys secimda is found along trails used by the local communities and are,
therefore, at risk ofbeing trampled.
The two more widely distributed species are Geostachys densiflom Ridl. and G.
elegans Ridl. Geostachys densiflom is mauily found in the mountainous area along the
Titiwangsa Ridge in the Cameron Highlands and on Fraser’s Hill (Fig. 3). Until recently
this species was only Imown from Pahang and Perak but a 2010 expedition to G.
Chamar and G. Chamar Anak, Kelantan revealed new sites for Geostachys densiflom.
m m
These new populations occur at an elevation of 1300 and 1700 respectively. This
AGO
has expanded its to close to 32 km“. At the type locality in G. Korbu it was last
collected in 1913 and it is not known if it still occurs there; flirther investigation is
needed to verify its status.
Most populations of Geostachys elegans are restricted to the state of Pahang
although the type locality is G. Ledang in northern Johor (Fig. 4), The populations in
both Taman Negara National Park and G. Ledang State Park are relatively safe within
their Protected Areas (PAs).
Geostachys montana (Ridl.) Holttum is Data Deficient (DD) as it could not
be relocated on G. Berembun in the Cameron Highlands (Fig. 5) even after several
attempts. There are two possibilities, i.e., either the original locality details were
inaccurate or it has gone extinct in the wild. Further research is required to determine
its status in the near future.
Discussion
Protected areas remain the best places for material for taxonomic research and the most
likely place for the discovery ofnew species. In Malaysia, the pristine conditions ofits
National and State Parks and otherprotected forest areas provide the lastrefuges forrich
biodiversity that tropical forests are famed for. For instance, examples ofnew species
discovered from protected areas in Malaysia include two new Geostachys species (Lau
et al., 2005) in Taman Negara National Park, another Geostachys species in Maliau
&
Basin, Sabah (Lim Lau, 2006), a Pennilabium orchid in Endau Rompin National
Park, Johor (O’Byme et ah, 2009), two gingers from the genus Scaphochlamys in Krau
Wildlife Reserve, Pahang (Sam et al., 2010) and 13 new species oiBegonia from G.
Mulu National Park and G. Buda National Park, Sarawak (Sang et al., 2013). These,
and many others, indicate that more inventory attention should be given to areas that
are ‘botanical black holes’. It is anticipated that more new species will be found in the
future as the diversity in the protected areas continues to be studied.
With 60% ofthe world’s Geostachys species occumng in Peninsular Malaysia,
with more than 85% ofthese taxa hyper-endemic, this region is regarded as the genus
: :
Conservation ofPeninsularMalaysian Geostachys 9
Geographical Distribution of Geostachys densiflora
(Zingiberaceae) in Peninsular Malaysia
100“00' 10r00' 102“00' 103“00' 104°00'
.OOoZ lOOoZ
,00o9
.00o9
.OOot^
.OOoe
Extent of Occurrence (EOO)
Localities
3172.193 sq km
1. Gn. Chamar, Gn. Chamar Anak 6. Gn. Jasar
2. Fraser's Hill 7. Gn. Brinchang Area of Occupancy (AOO)
3. Gn. Korbu 8. Cameron Highlands 32 sq km
4. Gn. Chamar
5. Gn. Irau EOO
Forest cover within
:
97%
Fig. 3. Distribution of Geostachys densiflora Ridl. in Peninsular Malaysia.
:
10 Gard. Bull. Singapore 66(1) 2014
Geographical Distribution of Geostachys elegans
(Zingiberaceae) in Peninsular Malaysia
100°00' ioroo’ 102“00’ 103°00' 104"00'
THAILAND
1 : 3,500,000
Location based on
:
#
Herbarium specimen(s)
Forest cover as in National
Forest Inventory (1991-1993)
III
50 0 50 Kilometers
NFI III Courtesy of Forestry Department Peninsular Malaysia SINGAPORE
104°00
Localities Extent of Occurrence (EOO)
23108.1 12 sq km
Taman Negara, Gn. Tahan
1.
2. Gn. Tapis Area of Occupancy (AOO)
3. Gn. Ledang 24 sq km
4. Genting Highlands, Gn. Ulu Kali
5. Bukit Kinta F.R. EOO
Forest cover within
6. Krau W.R., Gn. Benom 53% :
Fig. 4. Distribution of Geostachys elegans Ridl. in Peninsular Malaysia.