Download molecular evidence for the common origin of snap

American Journal of Botany 89(9): 1503–1509. 2002.
MOLECULAR EVIDENCE FOR THE COMMON ORIGIN
SNAP-TRAPS AMONG CARNIVOROUS PLANTS1
KENNETH M. CAMERON,2,4 KENNETH J. WURDACK,2,5
RICHARD W. JOBSON2,3
OF
AND
The Lewis B. and Dorothy Cullman Program for Molecular Systematics Studies, The New York Botanical Garden,
Bronx, New York 10458 USA; and
3
Department of Botany, University of Queensland, Brisbane, Queensland 4072 Australia
2
The snap-trap leaves of the aquatic waterwheel plant (Aldrovanda) resemble those of Venus’ flytrap (Dionaea), its distribution and
habit are reminiscent of bladderworts (Utricularia), but it shares many reproductive characters with sundews (Drosera). Moreover,
Aldrovanda has never been included in molecular phylogenetic studies, so it has been unclear whether snap-traps evolved only once
or more than once among angiosperms. Using sequences from nuclear 18S and plastid rbcL, atpB, and matK genes, we show that
Aldrovanda is sister to Dionaea, and this pair is sister to Drosera. Our results indicate that snap-traps are derived from flypaper-traps
and have a common ancestry among flowering plants, despite the fact that this mechanism is used by both a terrestrial species and
an aquatic one. Genetic and fossil evidence for the close relationship between these unique and threatened organisms indicate that
carnivory evolved from a common ancestor within this caryophyllid clade at least 65 million years ago.
Key words:
Aldrovanda; carnivorous plants; Dionaea; DNA; Droseraceae; molecular systematics; phylogeny.
Charles Darwin (1875) called Dionaea muscipula Ellis (Venus’ flytrap) ‘‘one of the most wonderful plants in the world’’
because of its rapidly closing, snap-trap leaves, which are used
to capture and digest small animals for the plants’ nutritional
benefit. Its trapping mechanism is one of the fastest movements known among plants (Iijima and Sibaoka, 1985; Juniper,
Robins, and Joel, 1989), and the monotypic genus is well
known to both schoolchildren and scientists. Few people, however, are as familiar with the aquatic waterwheel plant (Aldrovanda vesiculosa L.), which employs a similar snap-trap
mechanism for trapping prey underwater (Fig. 1).
Aldrovanda vesiculosa was first collected from India in
1696 (Kundu, Basu, and Chakraverty, 1996). The genus was
named in honor of Ulisse Aldrovandi, a prominent Italian naturalist, and the specific epithet used by Linnaeus in 1753 refers
to the fact that the leaves were originally thought to trap air
for buoyancy (de Lassus, 1861; Lloyd, 1976). Darwin (1875)
carefully studied living specimens and suspected the plant, instead, to use its leaves for small animal capture, as was proven
later to be the case. Aldrovanda is a rootless, submerged,
aquatic herb that produces whorls of 7–8 leaves per node. It
grows in dystrophic lakes and ponds throughout the Old
World. In its habit and distribution, therefore, it is similar to
many aquatic species of the carnivorous bladderworts (Utricularia spp.). Rather than producing underwater bladders for
catching prey, however, the leaves of Aldrovanda consist of a
pair of oval or round leaf lobes, 5–10 mm long, that resemble
an open bivalve shell. On the inner surface of each lobe are
approximately 20 hairs that serve as triggers for the snap-trap.
This foliar morphology is so strongly reminiscent of the leaves
Manuscript received 26 February 2002; revision accepted 2 May 2002.
The authors thank Bruce Salmon for sending us leaf tissue of Drosera regia
and Jay Horn at Duke University for material of Dionaea and Drosophyllum.
The photograph of Aldrovanda was given with permission by Mr. Barry Rice,
qwww.sarracenia.com. This research was funded by the Lewis B. and Dorothy Cullman Foundation.
4
Author for reprint requests (e-mail: [email protected]).
5
Current address: Laboratory of Molecular Systematics, Museum Support
Center A2000, Smithsonian Institution, Washington, D.C. 20560 USA.
1
of Venus’ flytrap that Darwin (1875) described Aldrovanda as
‘‘a miniature, aquatic Dionaea.’’
In its reproductive morphology, however, Aldrovanda shares
more characters with the carnivorous sundews (Drosera spp.)
than it does with Dionaea. A comparison of these taxa is presented in Table 1. Both Aldrovanda and Drosera have flowers
with five stamens and parietal placentation of their ovules,
whereas the flowers of Dionaea have 15 stamens and basal
placentation. The pistil of Aldrovanda has separate, undivided
styles like some species of Drosera, whereas the styles of
Dionaea are completely united. Both Dionaea and Drosera
display multiple operculate pores around the connection area
of their pollen tetrads (Takahashi and Sohma, 1982). Aldrovanda also sheds its pollen as tetrads, but its pollen is triporate
and differs from both Dionaea and Drosera. Consequently,
Nakai (1949) proposed a monotypic family, Aldrovandaceae,
to accommodate this enigmatic taxon rather than to classify it
in Droseraceae.
A close relationship with Drosera is further supported by
genetic evidence in the form of chromosome size, centromere
position, and CMA1 DAPI2 fluorescent banding patterns
(Hoshi and Kondo, 1998). Both Aldrovanda and Drosera have
mostly small chromosomes with no primary constriction and
2–4 CMA1 DAPI2 sites. In contrast, the chromosomes of
Dionaea are all medium in size with localized centromeres and
show no CMA1 DAPI2 sites.
The only cladistic study of phylogenetic relationships within
Droseraceae that sampled Aldrovanda included 14 morphological characters for this taxon (Williams, Albert, and Chase,
1994). The results showed a sister relationship between Aldrovanda and Drosera, with Dionaea sister to the pair, implying
that plants with adhesive, glandular flypaper-traps may have
evolved from plants with snap-traps. This scenario seems unlikely because there is evidence (Williams, 1976; Juniper, Robins, and Joel, 1989) that the trigger hairs of Dionaea are derived from glandular trichomes, not vice versa. On the other
hand, a number of molecular phylogenetic studies have demonstrated that carnivory evolved independently among several
different lineages of flowering plants (Albert, Williams, and
1503
1504
AMERICAN JOURNAL
OF
BOTANY
[Vol. 89
Fig. 1. Habit photographs of Dionaea muscipula Ellis and Aldrovanda vesiculosa L., which capture and digest small animals with modified snap-trap leaves.
(a) Dionaea is a terrestrial species. (b) Aldrovanda is a rootless, submerged aquatic.
Chase, 1992; Meimberg et al., 2000). Furthermore, convergent
evolution has selected for similar trapping mechanisms among
these unrelated lineages. For example, leaves used as pitfalltraps (i.e., ‘‘pitcher plants’’) are known to have evolved within
lineages of the orders Caryophyllales, Oxalidales, Ericales,
and Poales (sensu APG, 1998). Likewise, flypaper-traps have
evolved in at least three or four different clades.
The exact relationship between Aldrovanda, Dionaea, and
Drosera has not been addressed adequately, and molecular
data clearly have a role to play in sorting out this problem
(Givnish, 1989). With full knowledge that similar trapping
mechanisms have evolved independently among different lineages of carnivorous plants, we set out to determine the phy-
logenetic position of Aldrovanda using DNA sequence data
from both nuclear and plastid genes. The placement of this
taxon among angiosperms has implications not only for taxonomy, but also for determining whether snap-traps evolved
more than once among angiosperms and whether they were
derived from flypaper-traps or vice versa.
MATERIALS AND METHODS
Taxon sampling and gene sequencing—Partial rbcL, atpB, and matK plastid gene sequences, as well as nuclear small-subunit (18S) ribosomal DNA
sequences were obtained from 15 taxa classified in the order Caryophyllales.
DNA sequences for 18S and atpB were newly generated for Aldrovanda,
September 2002]
TABLE 1.
CAMERON
ET AL.—ORIGIN OF SNAP-TRAPS
1505
A comparison of nonmolecular characters between Aldrovanda and other members of Droseraceae s.s.
Character
Distribution
Habit
Trapping mechanism
Aldrovanda
Old World: Disjunct in Europe, Africa, Southeast
Asia, and Australia
Aquatic
Snap-trap with ;20 trigger
hairs/leaf lobe
Circinate
Sessile, nonvascular, quadrifid
adaxial glands
Dionaea
Drosera regia
Other Drosera spp.
New World: Southeast USA
Old World: South Africa
Terrestrial
Flypaper-trap
New and Old World:
esp. diverse in South
Africa and Australia
Terrestrial
Flypaper-trap
Circinate
Stalked, vascular, adaxial glands
Circinate
Stalked, vascular, adaxial glands
Stamen no.
Pollen
5
Triaperturate, spinose pollen
shed in tetrads without radial plates
Terrestrial
Snap-trap with 3 trigger hairs/
leaf lobe
Circinate
Sessile, nonvascular, adaxial
glands and stellate, abaxial
glands
15
Multi-aperturate, nonspinose
pollen shed in tetrads without radial plates
Pollen apertures
Pores opposite at tetrad connection; not operculate and
without channel openings
Pores alternate at tetrad connection; not operculate and
without channel openings
Styles
Placentation
Seed
Separate, undivided
Parietal
Large, with smooth, thickened exotesta and endotesta
United
Basal
Large, with smooth, thickened, exotesta
Chromosome no.
2n 5 48 (medium)
2n 5 32 (small)
2.94 mm2
2
0.90 mm2
None
?
?
0.51–1.69 mm2
Mostly 2 or more
2/1
2/1
1/1
2 or 1/2 or 1
Leaf vernation
Glands
Average area of chromosome complement
CMA1 DAPI2 sites
7-methyljuglone/plumbagin
Dionaea, Drosophyllum, Drosera regia, D. adelae, and D. capensis. We also
generated new rbcL sequences for Aldrovanda and Drosera adelae, as well
as new matK sequences for Aldrovanda, D. capensis, and D. adelae. All
remaining sequences used in the analyses were downloaded from GenBank
(See http://ajbsupp.botany.org/v89/ for voucher information and GenBank accession numbers). We initially determined (results not shown) that Aldrovanda
is a member of the ‘‘non-core’’ Caryophyllales (sensu Cuénoud et al., 2002)
by analyzing its rbcL, atpB, and 18S sequences within a large, published
matrix of angiosperm sequences that included representatives of most eudicot
orders (modified from Soltis et al., 2000). Based on this result and on previously published, broadscale analyses of Caryophyllales (Lledó et al., 1998;
Meimberg et al., 2000; Soltis et al., 2000; Cuénoud et al., 2002), we sampled
15 species from 13 genera and nine families (Ancistrocladaceae, Dioncophyllaceae, Droseraceae, Frankeniaceae, Nepenthaceae, Plumbaginaceae, Polygonaceae, Simmondsiaceae, and Tamaricaceae) to represent the ingroup and
appropriate outgroup taxa. The entire aligned matrix is available from the first
author or online from The New York Botanical Garden (http://www.nybg.org).
Total DNA was extracted using the FastPrep (Qbiogene, Carlsbad, California, USA) and glassmilk method from approximately 0.5 cm2 of dried tissue
as described by Struwe et al. (1998). Target loci were amplified in 50-mL
volumes using standard polymerase chain reaction (PCR) protocols that included the addition of bovine serum albumin (BSA) and betaine. Amplification and sequencing primers were the same as used by Soltis et al. (2000)
except for matK, which we amplified and sequenced using primers 390F and
1326R as cited in Cuénoud et al. (2002). This resulted in ;900 base pairs
(bp) of sequence for that gene. In all cases, resulting PCR products were
purified using QIAquickTM spin columns (Qiagen, Valencia, California, USA)
according to the manufacturer’s protocols. Cycle sequencing reactions were
completed using a combination of purified PCR template, primer, and dRhodamine Ready Reaction mix (Applied Biosystems, Foster City, California,
USA) for 20 cycles. These reactions resulted in nearly complete, overlapping,
forward and reverse strands of the target loci. Centri-Sep sephadex columns
(Princeton Separations, Adelphia, New Jersey, USA) were used according to
5
5
Multi-aperturate, spinose Multi-aperturate, spinupollen shed in tetrads
lose pollen shed in
without radial plates
tetrads mostly with
radial plates
Pores opposite at tetrad
Pores opposite at tetrad
connection; not operconnection; opercuculate and without
late and mostly with
channel openings
channel openings
Separate, undivided
Separate, mostly divided
Parietal
Parietal
Small, dust-like with
Small, dust-like with
thin, reticulate exotesthin, reticulate exotesta
ta
2n 5 34 (small)
2n 5 20, 30, 32, 40, 60,
80 (small)
the manufacturer’s instructions to remove excess dye terminators and primer
from the cycle sequencing products. These were subsequently dehydrated,
resuspended in a mixture of formamide and loading dye, and pipetted onto a
5% denaturing polyacrylamide gel. Samples were run on an Applied Biosystems ABI 377XL automated DNA sequencer, and resulting electropherograms
were edited using Sequencher 3.0 (GeneCode, Ann Arbor, Michigan, USA).
Phylogenetic analyses—The individual and combined data matrices were
analyzed using the parsimony criterion in PAUP* version 4.0b8 (Swofford,
2001). Simmondsia was specified as the single outgroup taxon based on the
topologies produced in broader phylogenetic studies (e.g., Soltis et al., 2000;
Cuénoud et al., 2002). The 59 and 39 ends of each gene sequence (typically
;30–50 bp) were excluded to minimize missing data. In addition, a hypervariable 32-bp region of matK located at ;190 bp from the 59 end was excluded because of ambiguous alignment. All equally parsimonious trees were
found by executing a branch and bound search with gaps treated as missing
data and characters unordered and weighted equally. Internal support was
determined by performing parsimony bootstrap analyses (1000 replicates, simple addition, and tree bisection-reconnection [TBR] branch swapping). Although the phylogenetic position of Drosera regia Stephens was found to be
inconsistent among the individual gene analyses (i.e., either sister to the other
species of Drosera or sister to the Aldrovanda-Dionaea clade), no bootstrap
support .50% (i.e., ‘‘hard’’ incongruence sensu Weins, 1998) was detected
among the individual trees, so combination of the four data sets proceeded.
To address the evolution of trapping mechanisms, we used both ACCTRAN
and DELTRAN optimizations in MacClade version 3.04 (Maddison and Maddison, 1992) to examine the distribution of the following character states: not
carnivorous, pitfall-trap, flypaper-trap, and snap-trap.
RESULTS
Table 2 summarizes each individual gene matrix as well as
the combined four-gene matrix. Tree statistics for each analysis
1506
TABLE 2.
AMERICAN JOURNAL
OF
BOTANY
[Vol. 89
Comparison of matrix and tree statistics.
No. characters
No. variable characters (% of total)
No. informative characters (% of variable)
No. trees
Tree length (steps)
Consistency index (CI)
Retention index (RI)
Droseraceae monophyly bootstrap support (%)
Drosera monophyly bootstrap support (%)
Aldrovanda/Dionaea monophyly bootstrap support (%)
are also presented. Droseraceae sensu stricto (s.s.) were found
to be monophyletic and with 100% bootstrap support in all
individual gene analyses. A sister relationship between Aldrovanda and Dionaea was discovered in every analysis, but it
is supported by the bootstrap only in the matK analysis. However, an alternative topology in which Aldrovanda alone is
sister to Drosera was found to be equally parsimonious in the
results of the separate 18S and atpB analyses. In these two
cases, the strict consenus trees are unresolved for the position
of these taxa (see Fig. 2).
The combined data matrix consists of 5386 characters, of
which 693 are potentially parsimony informative. We recovered a single most parsimonious tree (Fig. 3) in which every
clade is supported at .50% by the bootstrap (bts). The genus
Drosera is monophyletic, with the morphologically unique,
South African species D. regia sister to the other two species
sampled. Dionaea and Aldrovanda are sister to each other
(90% bts), and together this pair is sister to Drosera, forming
a well-supported Droseraceae (100% bts) if Drosophyllum lu-
Nuclear 18S
Plastid rbcL
Plastid atpB
Plastid matK
Combined
1739
183 (10.5)
99 (54.1)
5
317
0.700
0.639
100
69
,50
1331
327 (24.6)
159 (48.6)
3
603
0.697
0.544
100
,50
,50
1428
368 (25.8)
184 (50.0)
6
599
0.748
0.683
100
59
,50
888
439 (49.4)
188 (42.8)
1
858
0.724
0.629
100
,50
57
5386
1317 (24.5)
693 (52.6)
1
2404
0.712
0.611
100
57
90
sitanicum (L.) Link is excluded from the family. Sister to Droseraceae s.s. is a second clade (81% bts) composed of four
genera, all but one of which are carnivores and/or lianas at
some stage in their life cycle. These include the carnivorous
liana Triphyophyllum, which is sister to the noncarnivorous
liana Ancistrocladus (100% bts). The monotypic carnivore
Drosophyllum is sister to them (99% bts), and the carnivorous
pitcher-plant liana Nepenthes is sister to these three.
DISCUSSION
Despite Aldrovanda’s similarity in leaf morphology to Dionaea, the greater commonality in floral features between Aldrovanda and Drosera has consistently kept Aldrovanda classified within Droseraceae by most taxonomists (e.g., Diels,
1906), even when Dionaea is elevated to familial status as
Dionaeaceae (Small, 1933). In fact, the illegitimate combination Drosera aldrovanda F. Muell. has even been made (von
Mueller, 1876) to accommodate Aldrovanda within Drosera
Fig. 2. Alternative topologies for the relationships within Droseraceae based on individual gene analyses and on the analysis of morphological data by
Williams, Albert, and Chase (1994). Only the Droseraceae s.s. clade is shown for each as a strict consensus tree with bootstrap percentages.
September 2002]
CAMERON
ET AL.—ORIGIN OF SNAP-TRAPS
1507
Fig. 3. Phylogenetic relationships for the carnivorous plant genera in Caryophyllales inferred from parsimony analysis of combined plastid rbcL, atpB,
matK, and nuclear small-subunit (18S) rDNA sequences. Branch lengths are proportional to the number of character changes on each branch as indicated
(ACCTRAN optimization). Bootstrap percentages .50% are indicated at each node of the single most parsimonious tree. The following character states were
optimized onto the gene tree: not carnivorous (dark gray), pitfall-traps (dashes), flypaper-traps (light gray), and snap-traps (solid black).
rather than to treat it as a distinct genus. In contrast, Dionaea
has never been classified under any other generic name.
The common origin and evolution of snap-traps—Until
now, no molecular data have been available to help settle the
ongoing debate regarding the exact phylogenetic position and
classification of Aldrovanda vesiculosa. However, our results
show that Aldrovanda is sister to Dionaea, not to Drosera,
and that snap-traps evolved only once among carnivorous
plants (moving Aldrovanda away from Dionaea to a position
sister to Drosera requires an additional ten steps). We have
pointed out the numerous morphological differences between
Aldrovanda and Dionaea, but there are other likely synapomorphies in addition to the common trapping mechanism that
unite them. Boesewinkel (1989) considered the ovules and
seeds of Aldrovanda to be more similar to those of Dionaea
than to any other taxon in Droseraceae. The two genera also
share a similar trichome/gland morphology, including sessile
and stellate glands lacking vascular tissue and multicellular
trigger hairs (Juniper, Robins, and Joel, 1989). Moreover, the
1508
AMERICAN JOURNAL
traps of both Aldrovanda and Dionaea function in a similar
manner by rapid transmission (6–17 cm/s) of action potentials
between excitable cells (Iijima and Sibaoka, 1985). Some species of Drosera also generate action potentials, but their trasmission velocity is 10% slower. Finally, both Aldrovanda and
Dionaea lack 7-methyljuglone, but possess plumbagin in their
tissues (Culham and Gornall, 1994). However, there are several species of Drosera that have this same napthoquinone
profile.
Based on a number of shared morphological characters and
the structure of the molecular cladogram, therefore, we recommend the recognition of Aldrovanda, Dionaea, and Drosera, but not Drosophyllum, as members of Droseraceae s.s.
Drosophyllum is a monotypic genus from Portugal. Although
it traps insects with glandular trichomes, it differs from Drosera in a number of morphological features, and its disassociation from Droseraceae s.s. has been documented and discussed in detail by others (Williams, Albert, and Chase, 1994;
Meimberg et al., 2000). It should be treated as the sole member of Drosophyllaceae.
Furthermore, based on this cladogram, it is most parsimonious to hypothesize that the snap-traps of Aldrovanda and
Dionaea were derived from a common terrestrial ancestor that
had flypaper-traps, not the other way around. Likewise, the
pitfall-traps of the tropical pitcher plant genus Nepenthes also
must have evolved from an ancestor with adhesive, flypaper
traps. Its exact position among the genera of Caryophyllales
has been ambiguous to date; Albert, Williams, and Chase,
(1992) and Fay et al. (1997) found Nepenthes to be sister to
a Plumbago/Rheum clade, Williams, Albert, and Chase (1994)
and Lledó et al. (1998) found it sister to the entire carnivorous
clade, Meimberg et al. (2000) to the Drosophyllum/Ancistrocladus/Dioncophyllaceae clade, and Cuénoud et al. (2002) to
Droseraceae. Only one of these relationships—that of Meimberg et al. (2000), which is the same as found here—has been
supported by the bootstrap.
Lledó et al. (1998) and others have discussed the hypothesis
that the common ancestor of this mostly insectivorous, caryophyllid clade was preadapted for a carnivorous lifestyle because many of the noncarnivorous, outgroup taxa (e.g., Plumbago, Tamarix, Frankenia) possess multicellular glands that
secrete mucilage, salt, or other compounds. Modification of
these glands for animal capture and/or digestion followed by
their co-option as snap-trap trigger hairs in the ancestor of
Dionaea and Aldrovanda seems plausible and has been suggested (Juniper, Robins, and Joel, 1989; Williams, 1976). The
habitat of Dionaea muscipula frequently floods (Roberts and
Oosting, 1958). Yet even under these conditions, it has been
reported that the flytraps continue to grow and capture prey
underwater (Schnell, 1976). Thus, it is not difficult to envision
a Dionaea-like, terrestrial ancestor of Aldrovanda vesiculosa
becoming adapted to a permanently aquatic lifestyle as suggested by Arber (1920).
Implications for biogeography and conservation—Aldrovanda has a fossil pollen and seed history dating back to the
lower Tertiary (von Kircheimer, 1941; Muller, 1981; Yakubovskaya, 1991), with at least 13 different names having been
applied to these fossils (e.g., A. siberica V. Nikit., A. europaea
Negru, and A. praevesiculosa Kirchh.). Dionaea also has a
documented fossil history, with the discovery of pollen from
the middle Miocene and Pliocene of central Europe that is
comparable to modern Dionaea muscipula (Muller, 1981).
OF
BOTANY
[Vol. 89
These microfossils indicate that both genera were once more
common and widespread in distribution than they are today.
The ancient seeds of Aldrovanda also show enough morphological variation for Yakubovskaya (1991) to have hypothesized an evolutionary continuum along at least two ancestral
lines. One of these, originating from A. ovata M. Chandl., has
become extinct; the other, from A. intermediata E. Reid and
M. Chandl., ultimately led to the evolution of the modern species. Moreover, these fossils strongly support a date of origin
for this entire, highly specialized, carnivorous lineage of at
least 65 million years ago and probably much earlier. This age
may assist in explaining the current transcontinental distributions and putative Gondwanan origins of Drosera (Meimberg
et al., 2000) and Nepenthes (Meimberg et al., 2001).
Modern Aldrovanda vesiculosa is a relictual species historically distributed in parts of western Europe, central Africa,
southern India, Japan, and Queensland, Australia. It is now
presumed to be extinct in several countries (e.g., France, Italy,
and India), and is shrinking drastically in the number and size
of its extant populations (Adamec, 1995; Kaminski, Adamec,
and Breckpot, 1996; Kundu, Basu, and Chakraverty, 1996).
For 200 yr, it had been recorded (Adamec and Lev, 1999) from
more than 150 sites within Europe. Today it is estimated to
survive in fewer than 36 localities, most of which are in Poland, Ukraine, and Russia (Adamec, 1995). For this reason, it
is now considered one of the rarest aquatic plants in the Old
World (Kaminski, Adamec, and Breckpot, 1996). Likewise,
Dionaea muscipula can be considered a relictual species with
a narrow, endemic distribution of less than 300 km2 in the
southeastern United States.
Both Aldrovanda and Dionaea are severely threatened by
anthropogenic disturbance of their specialized habitats (Adamec, 1995; Kaminski, Adamec, and Breckpot, 1996; Kundu,
Basu, and Chakraverty, 1996), and Dionaea has been overcollected as a horticultural curiosity; it is currently listed on
Appendix II of CITES. Recent reintroductions of Aldrovanda
into the Czech Republic (Adamec and Lev, 1999) appear to
be successful and offer hope of conserving this remarkable
species. Introductions of Dionaea have also taken place in
Florida, New Jersey, and Deleware (USDA NRCS, 2001), but
the future survival of these two species in their native habitats
is of serious concern to conservationists. It is our hope that
the fundamental systematic data presented here have helped to
clarify the classification, phylogenetic relationships, evolutionary origin, and ecological adaptation of what Arthur Dobbs
(Dillwyn, 1843), the former Governor of North Carolina and
discoverer of Venus’ flytrap, called ‘‘the great wonder of the
vegetable kingdom.’’
LITERATURE CITED
ADAMEC, L. 1995. Ecological requirements and recent European distribution
of the aquatic carnivorous plant Aldrovanda vesiculosa L.: a review.
Folia Geobotanica and Phytotaxonomica 30: 53–61.
ADAMEC, L., AND J. LEV. 1999. The introduction of the aquatic carnivorous
plant Aldrovanda vesiculosa to new potential sites in the Czech Republic:
a five-year investigation. Folia Geobotanica and Phytotaxonomica 34:
299–305.
ALBERT, V. A., S. E. WILLIAMS, AND M. W. CHASE. 1992. Carnivorous
plants: phylogeny and structural evolution. Science 257: 1491–1495.
APG [ANGIOSPERM PHYLOGENY GROUP.]. 1998. An ordinal classification for
the families of flowering plants. Annals of the Missouri Botanical Garden
85: 531–553.
ARBER, A. 1920. Water plants; a study of aquatic angiosperms. Cambridge
University Press, Cambridge, UK.
September 2002]
CAMERON
ET AL.—ORIGIN OF SNAP-TRAPS
BOESEWINKEL, F. D. 1989. Ovule and seed development in Droseraceae. Acta
Botanica Neerlandica 38: 295–312.
CUÉNOUD, P., V. SAVOLAINEN, L. W. CHATROU, M. POWELL, R. J. GRAYER,
AND M. W. CHASE. 2002. Molecular phylogenetics of Caryophyllales
based on nuclear 18S rDNA and plastid rbcL, atpB, and matK DNA
sequences. American Journal of Botany 89: 132–144.
CULHAM, A., AND R. J. GORNALL. 1994. The taxonomic significance of naphthoquinones in the Droseraceae. Biochemical Systematics and Ecology
22: 507–515.
DARWIN, C. 1875. Insectivorous plants. John Murray, London, UK.
DE L ASSUS , A. 1861. Analyse du mémorie de Gaëtan Monti suyr
l’Aldrovandia suivie de quelques observations sur l’irritabilité des follicules de cette plante. Bulletin de la Société Botanique de France 8: 519–
523.
DIELS, L. 1906. Droseraceae. In A. Engler [ed.], Das Pflanzenreich 4: 1–136.
W. Engelmann, Weinheim, Germany.
DILLWYN, L. W. 1843. Hortus Collinsonianus; an account of the plants cultivated by the late Peter Collinson, Esq., F.R.S., arranged alphabetically.
Imprint published by W. Swansea, C. Murray and D. Rees.
FAY, M., K. M. CAMERON, G. PRANCE, M. D. LLEDÓ, AND M. W. CHASE.
1997. Familial relationships of Rhabdodendron (Rhabodendraceae):
plastid rbcL sequences indicate a caryophyllid placement. Kew Bulletin
52: 923–932.
GIVNISH, T. J. 1989. Ecology and evolution of carnivorous plants. In W. G.
Abrahamson [ed.], Plant–animal interactions, 243–290. McGraw-Hill,
New York, New York, USA.
HOSHI, Y., AND K. KONDO. 1998. A chromosome phylogeny of the Droseraceae by using CMA-DAPI fluorescent banding. Cytologia 63: 329–339.
IIJIMA, T., AND T. SIBAOKA. 1985. Membrane potentials in excitable cells of
Aldrovanda vesiculosa trap-lobes. Plant and Cell Physiology 26: 1–14.
JUNIPER, B. E., R. J. ROBINS, AND D. M. JOEL. 1989. The carnivorous plants.
Academic Press, London, UK.
KAMINSKI, R., L. ADAMEC, AND C. BRECKPOT. 1996. Report on recent sites
of Aldrovanda vesiculosa (Droseraceae) in Poland. Fragmenta Floristica
et Geobotanica 41: 291–294.
KUNDU, S. R., S. BASU, AND R. K. CHAKRAVERTY. 1996. Aldrovanda vesiculosa Linn.—its maiden appearance and disappearance from India: a
review. Journal of Economic and Taxonomic Botany 20: 719–724.
LLEDÓ, M., M. CRESPO, K. CAMERON, M. FAY, AND M. CHASE. 1998. Systematics of Plumbaginaceae based upon cladistic analysis of rbcL sequence data. Systematic Botany 23: 21–29.
LLOYD, F. E. 1976. The carnivorous plants. Dover, New York, New York,
USA.
MADDISON, W. P., AND D. R. MADDISON. 1992. MacClade: analysis of phylogeny and characer evolution. Sinauer, Sunderland, Massachusetts,
USA.
MEIMBERG, H., P. DITTRICH, G. BRINGMANN, J. SCHLAUER, AND G. HEUBL.
1509
2000. Molecular phylogeny of Caryophyllidae s.l. based on matK sequences with special emphasis on carnivorous taxa. Plant Biology 2:
218–228.
MEIMBERG, H., A. WISTUBA, P. DITTRICH, AND G. HEUBL. 2001. Molecular
phylogeny of Nepenthaceae based on cladistic analysis of plastid trnK
intron sequence data. Plant Biology 3: 164–175.
MULLER, J. 1981. Fossil pollen records of extant angiosperms. Botanical
Review 47: 1–142.
NAKAI, T. 1949. Classes, Ordines, Familiae, Subfamiliae, Tribus, Genera nov
quae attinet ad plantas Koreanas. Journal of Japanese Botany 24: 8–14.
ROBERTS, P. R., AND H. J. OOSTING. 1958. Reponses of Venus’ flytrap to
factors involved in its endemism. Ecological Monographs 28: 193–218.
SCHNELL, D. E. 1976. Carnivorous plants of the United States and Canada.
John F. Blair, Winston-Salem, North Carolina, USA.
SMALL, J. K. 1933. Flora of the southeastern United States. Published by the
author, New York, New York, USA.
SOLTIS, D. E., ET AL. 2000. Angiosperm phylogeny inferred from 18S rDNA,
rbcL, and atpB sequences. Botanical Journal of the Linnean Society 133:
381–461.
STRUWE, L., M. THIV, J. W. KADEREIT, A. S. R. PEPPER, T. J. MOTLEY, P. J.
WHITE, J. H. E. ROVA, K. POTGIETER, AND V. A. ALBERT. 1998. Saccifolium (Saccifoliaceae), an endemic of Sierra de la Neblina on the Brazilian-Venezuelan border, is related to a temperate-alpine linage of Gentianaceae. Harvard Papers in Botany 3: 199–214.
SWOFFORD, D. L. 2001. PAUP* 4.0b8: phylogenetic analysis using parsimony. Sinauer, Sunderland, Massachusetts, USA.
TAKAHASHI, H., AND K. SOHMA. 1982. Pollen morphology of the Droseraceae
and its related taxa. Science Reports, Tohoku University, 4th Series, Biology 38: 81–156.
USDA NRCS. 2001. The PLANTS database, version 3.1 (http://
plants.usda.gov). National Plant Data Center, Baton Rouge, Louisiana,
USA.
VON KIRCHEIMER, F. 1941. Über ein Vorkommen der Gattung Aldrovanda
Linné im Alttertiar Thüringens. Braunkohle 40: 308–311.
VON MUELLER, F. J. H. 1876. Fragmenta Phytographiae Australiae, vol. 10:
79–81. Published by the author, Melbourne, Australia.
WEINS, J. J. 1998. Combining data sets with different phylogenetic histories.
Systematic Biology 47: 568–581.
WILLIAMS, S. E. 1976. Comparative sensory physiology of the Droseraceae—
the evolution of a plant sensory system. Proceedings of the American
Philosophical Society 120: 187–204.
WILLIAMS, S. E., V. A. ALBERT, AND M. W. CHASE. 1994. Relationships of
Droseraceae: a cladistic analysis of rbcL sequence and morphological
data. Ameerican Journal of Botany 81: 1027–1037.
YAKUBOVSKAYA, T. V. 1991. The genus Aldrovanda (Droseraceae) in the
Pleistocene of the Belorussian SSR. Botanicheskii Zhurnal (St. Petersburg) 76: 109–118.