ROLES FOR MODERN PLANT SYSTEMATICS IN DISCOVERY AND CONSERVATION OF FINE-SCALE BIODIVERSITY

11 Pages·2000·9.9 MB·English

by  Bruce G Baldwin

Madrono, Vol. 47, No. 4, pp. 219-229, 2000
ROLES FOR MODERN PLANT SYSTEMATICS IN DISCOVERY AND
CONSERVATION OF FINE-SCALE BIODIVERSITY
Bruce G. Baldwin
Jepson Herbarium and Department of Integrative Biology,
1001 Valley Life Sciences Building #2465, University of California,
Berkeley, CA 94720-2465
Abstract
Systematic methods involving the use ofDNA dataandgenealogical analysishavebeen widely applied
to higher-level phylogenetic questions in plants but much less commonly to discovering plant lineages
corresponding to minimal-rank taxa (i.e., species, subspecies, and varieties) or to refining plant classifi-
cation at the finest levels. Recent research in the Jepson Herbarium integrating extensive field sampling,
biosystematic data, and molecularphylogenetics providesexamples fromtheCaliforniafloraforassessing
the value of modern systematic approaches as a means ofdiscovering fine-scale plant diversity. Results
have sometimes led to taxonomic changes at the levels most important for biodiversity assessment and
have allowed resolution ofsystematic questions important to establishing conservation strategies. Angio-
sperm groups newly resolved with molecular data include both morphologically distinctive and morpho-
logically cryptic lineages that have been previously treated within more broadly circumscribed species,
subspecies, or varieties. Taxonomic recognition of such newly resolved lineages is often necessary if
taxonomy is to reflect monophyletic groups and fine-scale units of biodiversity. To promote discovery,
recognition, andconservation ofplant lineages, systematistsareadvised to samplewidely withinminimal-
rank taxa (including rare taxa) in the field and in herbaria, to consider previous taxonomies, to voucher
all collections, to examine multiple lines of systematic evidence, and to publish taxonomic changes,
including nomenclatural changes. Scientists involved in biodiversity management and conservation are
advised to regard circumscriptions of all taxa as hypotheses of natural groups, to recognize that those
hypotheses are subject to change, and to welcome taxonomic and nomenclatural changes that reflect an
improved understanding ofnatural groups. Conservation biologists are urged to bear in mind that species
or infraspecific taxa are not necessarily the minimal units of biodiversity. To conserve evolutionary
lineages (and potential for future evolution), plant managers must seek to conserve representative popu-
lations oftaxa from throughout their geographical and ecological distributions, must resist indiscriminate
use of non-local germplasm in restoration efforts, and must consider cryptic biodiversity in regional
conservation planning.
Systematics is fundamental to understanding of mal-rank taxa (i.e., species, subspecies, and varie-
biodiversity. The most widely recognized organis- ties), which are ofmost immediate concern to con-
mal units ofbiodiversity, i.e., species, may not rep- servation biologists, ecologists, and floristicians
resent natural evolutionary groups or may not re- (e.g., Rieseberg et al. 1988; see Soltis—et al. 1992).
flect the finest-scale natural groups that can be re- Studies of phylogeographic diversity fine-scale,
solved and described by systematists. Modern sys- geographically structured evolutionary lineages
tematic approaches that allow a genealogical corresponding to "Evolut—ionarily Significant
perspective on biodiversity hold great promise as a Units" (sensu Moritz 1994) also have been ex-
means of achieving a refined taxonomy that better tremely limited for plants (e.g., Fujii 1997; Soltis
reflects evolutionary lineages of organisms et al. 1997; Olsen and Schaal 1999; Tremblay and
throughout the tree of life (e.g., Angiosperm Phy- Schoen 1999; Shaw 2000; also see Schaal et al.
logeny Group 1998). To date, modern systematic 1998; Schaal and Olsen 2000), especially by com-
approaches to resolving evolutionary relationships parison with the rich literature on animal phylo-
have been applied by plant systematists mostly to geography (reviewed by Avise 2000). As noted by
questions concerning groups of recognized species Moritz (1995, 1999) and Coates (2000), conserving
or higher-level taxa (see Soltis et al. 1998). Recent independently evolving sets ofpopulations not only
studies of higher-level plant phylogeny have yield- preserves biodiversity but also may be the best
ed insights into broad-scale evolutionary and bio- strategy for conserving the potential for future evo-
geographic patterns that are directly relevant to bio- lution. On a regional scale, refined understanding
diversity assessment and prioritization of conser- of phylogeographic patterns across organismal
vation efforts (e.g.. Vane-Wright et al. 1991; Mish- groups may allow for improved resolution of bio-
ler 1995; Faith 1996; reviewed by Soltis and diversity hot-spots and identification of critical ar-
Gitzendanner 1998). eas for conservation attention (see Moritz and Faith
Modern systematic methods have been applied 1998).
less commonly to testing the naturalness of mini- In this paper, I present examples of previously
220 MADRONO [Vol. 47
undetected diversity resolved from studies of the America has been associated with the rise of di-
California flora conducted in the Jepson Herbarium. verse lineages of annuals and perennials that are
These studies illustrate both the potential and the largely or entirely restricted to the region (Axelrod
proven value of applying modern systematic meth- 1992; Graham 1999). The high degree of endem-
ods to discovery offine-scale plant diversity and to ism, ca. 50% of species, in the California Floristic
refining classifications of minimal-rank plant taxa. Province, i.e., the Mediterranean-climatic region of
Finally, I make general recommendations for how western North America (Raven and Axelrod 1978),
systematists and other biodiversity scientists and largely reflects high diversity in neoendemic groups
planners may promote discovery and conservation wherein often only minimal morphological diver-
of plant diversity. gence has occurred between evolutionary lineages,
Systematists in California and elsewhere in west-
Exploration in the Field and Laboratory ern North America have long appreciated the com-
plexity of the regional flora and the need for in-
Well-focused field exploration (see Ertter 2000a) depth evolutionary investigations to reveal natural
and detailed systematic analysis are complementary units of biodiversity. The San Francisco Bay Area
components ofan effective strategy for discovering botanists Harvey Monroe Hall, Ernest Babcock, G.
plant diversity. As reviewed by Ertter (2000b), bo- Ledyard Stebbins, Jens Clausen, David Keck, Wil-
tanical field exploration in western North America, liam Hiesey, and others pioneered the incorporation
often by non-academic professionals and amateurs, of genetic principles and experimental approaches
has been a continuing source of floristic novelties. into systematics in a highly successful effort to re-
Modern systematic approaches can contribute solve evolutionary lineages and understand com-
greatly to the process of discovery by offering a
plex patterns of variation in the California flora
rigorous means of resolving the systematic status (e.g., Babcock and Hall 1924; Stebbins 1950; Clau-
of recently discovered plant populations. For ex-
ample, DNA sequence variation may clarify wheth- dseenvel1o95p1m;enatlssoinseseystSemmoactoivcistinsow19a9l7l).owSeuvbesneqmuoernet
er phenotypically unusual populations or sets of
progress in discriminating natural plant groups and
populations belong within previously described,
refining the taxonomy of western North American
minimal-rank taxa or represent undescribed evolu-
plants.
tionary groups (e.g., Baldwin 1999a). Data from
DNA also may allow confident taxonomic place- Advances in phylogenetic theory and methodol-
ogy, together with the development of accessible
ment of newly discovered plants that are evidently
high-speed computers, now permit simultaneous
distinct from any described minimal-rank taxon but
of uncertain position (e.g., Boyd and Ross 1997). analysis of large numbers of variable characters to
Conversely, fine-scale systematic studies depend on produce rigorous hypotheses of evolutionary rela-
extensive field sampling across the geographical tionships within plant groups, as well as estimates
and ecological distribution of taxa for assessing of support for resolved lineages (see Swofford et
naturalness of groups and detecting any unrecog- al. 1996). Character changes (resulting from mu-
nized diversity within a group. Phylogeographic tations) allow diagnosis of monophyletic groups
studies have demonstrated the potential for discov- (=evolutionary lineages orclades), the mostnatural
ery of geographically distinct, and often morpho- groups recognized by system—atists (Hennig 1966;
logically cryptic, evolutionary lineages within both see Mishler 1995, 2000a, b) plants belonging to
widespread and narrowly distributed species (see monophyletic groups are more closely related to
Soltis et al. 1997; Moritz 1999; Avise 2000; Riddle one another than to plants in other groups. Access
etthaatl. 2i0n0v0o)l.veSysextteemnastiivcestgudeioegsraapnhdicflorsiastmipclsiunrgveyosf ctoharaancteevresr-firnocmreDasNinAg sneuqmubeenrcesofhamsacernohmaonlceecdultahre
widespread taxa as well as locally restricted taxa prospects for systematists to attain fine-grained, ro-
are therefore advisable to maximize the potential bust resolution of evolutionary lineages (see Hillis
for discovering unrecognized plant diversity. et al. 1996; Soltis et al. 1998).
The prospect for floristic discoveries to result
from more detailed systematic analyses of western Examples of Recent Plant Discoveries from
North American plant groups appears great. Young Systematic Studies
lineages, which account for much of the endemic
plant diversity in western North America, e.g., Cal- To illustrate the utility of modern systematic
ifornia (Raven and Axelrod 1978), can be expected methods for discovery of plant groups, I present
to exhibit mosaic or cryptic phenotypic variation below some examples from research conducted in
from minimal divergence, differential sorting ofan- my lab at the Jepson Herbarium, principally on Cal-
cestral polymorphism through descendant lineages ifornian angiosperm lineages. Although categori-
(see Maddison 1995), or hybridization (see Arnold zation of the examples is somewhat artificial, three
1997). Climatic and geologic upheaval and exten- general types ofproblems are addressed: confusing
sive species turnover seen in the plant fossil record variation within taxa, resolution of cryptic biodi-
during the mid to late Cenozoic in western North versity, and questionably distinct rare taxa.
— —
2000] BALDWIN: DISCOVERY AND CONSERVATION OF PLANT LINEAGES 221
L Variation within taxa reexamined. The first D. siunbcrseps.ceinnscrescens
category of examples comprises groups wherein
morphological variation within a taxon was of un-
D. increscens
certain systematic significance until phylogenetic subsp. villosa
studies were undertaken.
Deinandra bacigalupii: A narrow endemic mis-
placed in a widespread species. Deinandra is a D. bacigalupii
species-rich genus of tarweeds (Madiinae, Com-
positae) reinstated formembers ofHemizonia sensii D.corymbosa
Keck (1959) that are most closely related to Hol- subsp.corymbosa
ocarpha (Baldwin 1999b). Until 1999, an ca.
8-rayed Deinandra from alkaline meadows in the D.csuobrsypm.bomsacarocephaia
Livermore Valley, California, was treated within D.
[Hemizonia] increscens subsp. increscens, a mostly
coastal taxon known otherwise from Santa Barbara D.kelloggii
County to Monterey County, California (Tanowitz
1982), more than 75 km south ofLivermore Valley.
D.pallida
Morphologically, the Livermore Valley tarweed is
highly similar to D. increscens except in anthercol-
or and pappus characteristics. Robert F. Hoovercol- D.pentactis
lected and left unidentified to species the Livermore
Valley tarweed as early as 1966. Rimo Bacigalupi
annotated the UC accession of the first known col- D.lobbii
lection (by Hoover) as not matching any published
species of Hemizonia.
D.halliana
Dean Kelch and Robert Preston independently 5changes
c1o9l9l0e'csteadndthberouLgihvtersmpoerceimeVnasllteoymetarwiwtehedconincertnhse aFinga.lys1i.s oTfh1e8Sm-o2s6tSpnaurcsliemaornriiobuossotmraeel DfrNoAm psheyqluoegnecneestiocf
that the plant was not identifiable with The Jepson the external and internal transcribed spa—cers forthe north-
Manual: Higher Plants of California (Hickman ern lineage ofDeinandra (Compositae Madiinae; Bald-
1993). The characteristics of yellow to brownish, win unpublished data). The outgroup (D. minthornii) used
rather than "black" (i.e., dark purple), anthers in for rooting the tree is not shown, nor are tree statistics
the Livermore Valley plants was in conflict with and support values (to be published elsewhere). Note the
extensive divergence ofthe Livermore Valley tarweed (D.
placement in Deinandra [Hemizonia] increscens
bacigalupii) from other representatives ofDeinandra and
and ray laminae of the plants were much too short the remote phylogenetic position of D. bacigalupii from
for D. [Hemizonia] pallida. Further examination of D. increscens, the species in which the Livermore Valley
the plants in comparison with other deinandras re- tarweed was earlier treated.
vealed that the pappus was much shorter and more
irregular than in other populations then assigned to
D. increscens (Baldwin 1999a). Chromosome
plants from the "northern lineage" of Deinandra
counts ofthe Livermore Valley tarweed of2n = 12
(Baldwin, unpublished data).
II, the modal chromosome number in Deinandra Evidence from DNA substantially augmented
(as in D. increscens), were inconclusive about re- morphological evidence for distinctiveness of the
lationships of the plants (Baldwin 1999a). Livermore Valley tarweed fromD. increscens. Rec-
bosReosmualltsDoNfAph(yrloDgNeAn)etisceqauneanlcyesisdatoaf, niunclceoanrcerrit- cogrneistcieonns iofmpDr.ovbeascioguarluupnidierasstadnidsitningctoffdriovmersDi.tyiinn-
with the morphological evidence, led me to con- Deinandra and rare plants in general in the Spring-
clude that the Livermore Valley tarweed is not a town wetlands area near Livermore, California.
member ofDeinandra increscens or any other pre- Deinandra bacigalupii has been regarded as an ex-
viously recognized species of Deinandra (Fig. 1; ample of a plant species that was discovered "in
Baldwin 1999a). Representatives of the two most front of the bulldozer" (Ertter 2000b), i.e., that
divergent groups ofD. increscens, i.e., D. i. subsp. came close to being driven to extinction prior to
increscens and D. i. subsp. villosa, were resolved being recognized as distinct.
as a well-supported monophyletic group to the ex- Layia gaillardioides: Interpopulational variation
clusion of representatives of the other six recog- of phylogenetic significance. Layia gaillardioi-
nized species of the "northern lineage" of Dein- des, the woodland layia, has been regarded as an
andra and the Livermore Valley tarweed. The Liv- example of a species displaying substantial mor-
ermore Valley tarweed does not appear to be of phological variation among populations (Clausen
recent hybrid origin based on 10 unambiguous 1951). Ray laminae may be uniformly deep yellow
rDNA mutations not shared with any other sampled or have white, greenish, or pale yellow apices de-
MADRONO
222 [Vol. 47
Tomales Knoxville
MuirBeach Isabel Creek
\\M\
Saratoga Summit Lewis Creek
Fig. 2. Basal leafvariation in the woodland layia, Layia gaillardioides (modified from Clausen 1951). Leaves sepa-
rated by the vertical line correspond to outer (left) and inner (right) Coast Range populations in central and northern
California. The populational differences shown here correspond to variation across three divergent molecular lineages
that do not appear to constitute a natural group (Baldwin, unpublished data).
pending on the population examined. Clausen Lessingia: Problems in species and—varietal cir-
(1951) noted that inner and outer Coast Range pop- cumscriptions in the "yellow group." Systematic
ulations differ in stem thickness and degree oflob- investigations by Markos (2000; also see Markos
ing of the basal leaves, characteristics that he con- and Baldwin 2001) revealed an outstanding exam-
cluded were heritable and ecologically significant ple ofmisinterpreted morphological variation in an-
(Fig. 2). other lineage of annuals in the California Compos-
Phylogenetic analysis of Layia has revealed ev- itae, namely, in the "yellow group" of Lessingia
idence that L. gaillardioides as circumscribed at (Astereae). Markos (2000) found that annuals in
presentrepresents an unnatural group (Baldwin, un- Lessingia constitute two ma—jor lineages that differ
published data). Populations shown earlierby Clau- in disc corolla coloration a pink-to-white-flow-
sen to be morphologically distinct constitute three ered lineage and a yellow-flowered lineage. Within
distinct lineages that apparently are not most close- the "yellow group," Markos (2000) used morpho-
ly related to one another. Two lineages of L. gail- logical and molecular data to resolve three major
lardiodes are more closely related to L. carnosa, L. natural groups that span the boundaries of widely
hieracioides, and L. septentrionalis than to a third recognized species and varieties.
lineage ofL. gaillardioides. Each of the groups re- Markos (2000) found that different representa-
solved within L. gaillardioides is well-supportedby tives of each of three taxa {Lessingia gladulifera,
unique rDNA mutations, and relationships among L. glandulifera var. glandulifera, and L. lemmonii)
the lineages and related species are likewise robust belong to different major lineages within the "yel-
based on rDNA data. Evolutionary lineages within low group." Morphologically, differences in shape
the woodland layia conform to a typical phylogeo- ofthe style-branch appendages and presence or ab-
graphic pattern except that the natural groups with- sence of a maroon band in the corolla throat diag-
in L. gaillardioides do not constitute a clade and nose the three primary lineages of yellow-flowered
instead represent a paraphyletic or (conceivably) lessingias. Markos (2000) concluded from her phy-
polyphyletic group (see Riddle et al. 2000 for sim- logenetic data that the accepted taxonomy of Les-
ilar examples). Recognition that L. gaillardioides singia underrepresents the actual biodiversity ofthe
has been circumscribed too broadly and comprises group and warrants substantial revision (S. Markos,
multiple natural groups, each probably warranting in prep.).
taxonomic distinction, may be an important con-
servation concern given the evidently scattered dis- IL Cryptic biodiversity. Modern systematic
tribution and small size of woodland layia popula- methods have promising potential for allowing dis-
tions. Sampling ofadditional populations and DNA covery of natural plant groups that are morpholog-
regions is now underway to resolve the precise de- ically indistinguishable (or nearly so) from one an-
limitation of each evolutionary lineage within L. other but may be geographically or ecologically
gaillardioides s. lat. prior to describing segregate distinct. As noted above, lineage diversity across
taxa. the geographic distribution of a morphological or
2000] BALDWIN: DISCOVERY AND CONSERVATION OF PLANT LINEAGES 223
biological species, i.e., phylogeographic diversity, cent molecular phylogenetic study by Chan (2000),
has been widely reported in vertebrates but has not who sampled widely across populations of each
been extensively studied in plants (see Soltis et al. taxon recognized by Ornduff (1966, 1993). Chan
1997; Avise 2000; Schaal and Olsen 2000). In Cal- found strong evidence from cpDNA and nuclear
ifornian groups of annuals, members of my lab rDNA sequences for morphologically cryptic lin-
have found various examples of cryptic diversity eages in L. californica, the most widespread species
associated with geography. Two examples are par- recognized by Ornduff (1993).
ticularly important for illustrating groups that are Chan's (2000) cpDNA and nuclear rDNA data
not only morphologically cryptic but, based on led him to propose the hypothesis that one set of
multiple lines of evidence, must be recognized as populations of L. californica sensu Ornduff (1993)
distinct taxa because the well-supported but mor- is most closely related to the outer coastal, endemic
phologically indistinct lineages are evidently not Californian taxa L. macrantha subsp. macrantha
most closely related—to one another. and L. m. subsp. bakeri. Chan concluded that the
Downingia yina. Schultheis (2000) examined three taxa constitute a well-supported group exclu-
relationships throughout Downingia (Lobeliaceae) sive ofthe Pacific Northwest endemic L. macrantha
with special attention to a lineage corresponding to subsp. prisca and other populations of L. califor-
three morphological species: D. bacigalupii, D. ele- nica sensu Ornduff(1993). Chan (2000) also found
gans and D. yina. Earlier work by Weiler (1962) that the two lineages corresponding to L. califor-
and Foster (1972) established that D. yina is cyto- nica sensu Ornduff (1993) have somewhat distinct
logically highly unusual, with a broad dysploid se- (but overlapping) geographic distributions and mi-
ries of chromosome numbers, i.e., 2n = 6, 8, 10, nor pappus differences, although some individuals
and 12 IL Chromosome "races" of D. yina are ofboth groups are epappose and cannot be reliably
mostly geographically distinct but could not be dis- distinguished morphologically. Lasthenia California
tinguished reliably on the basis of morphology us- sensu Ornduff (1993) appears to represent another
ing multivariate analyses and other approaches example, parallel to Downingia yina, of lineages
(Schultheis 2000). that do not constitute a natural group but have re-
Schultheis (2000) extensively sampled D. baci- mained morphologically similar while related lin-
galupii, D. elegans, and D. yina throughout their eages have undergone substantial morphological
geographic ranges in an attempt to discern the evo- change.
lutionary significance of chromosomal variation in Taxo—nomic recognition of cryptic plant
the group. She found strong phylogenetic signals groups. Morphologically indistinct evolutionary
from sequences ofboth chloroplast DNA (cpDNA) groups such as the two examples discussed above
and nuclear rDNA for three major lineages with (within Downingia yina and Lasthenia californica)
cytological and geographic integrity that each in- present a special challenge to plant systematists and
clude populations of D. yina. One lineage corre- the botanical community. Cryptically distinct lin-
sponds to populations west of the Cascade Range, eages that together constitute a monophyletic group
all with chromosome numbers of 2n = 6, 8, or 10 may or may not be viewed as warranting taxonomic
II, i.e., D. elegans (In = 10 II) and populations of recognition. In the interests ofaccurate biodiversity
D. yina with 2n = 6, 8, or 10 II. The second lineage assessment (which typically relies on taxa as units),
corresponds to a pocket in the southern Cascade I suggest that cryptic groups that are: (1) well-sup-
Range of Oregon wherein populations of D. yina ported by different lines of molecular or other ev-
with 2/7 = 10 II are found. The third lineage cor- idence, and (2) geographically or ecologically dis-
responds to populations east of the Cascades, with tinct should be recognized as taxonomically dis-
2n = 12 II, i.e., D. bacigalupii and D. yina. tinct. Well-supported, cryptically-distinct groups
Schultheis (2000) concluded that the three well- that are not even most closely related to one anoth-
supported evolutionary lineages in D. yina warrant er (e.g., groups in D. yina and in L. californica)
taxonomic recognition despite being only crypti- leave systematists committed to natural classifica-
cally distinct morphologically. Congruence be- tion with no choices other than to recognize the
tween two lines of molecular data leave minimal cryptic groups as taxonomically distinct or to treat
doubt that D. yina represents an example of diver- all members of the minimal monophyletic group
gent evolutionary lineages that remained morpho- encompassing the cryptic lineages and the related
logically static while closely related lineages (cor- group(s) as a minimal-rank taxon. I believe that the
responding to D. bacigalupii and D. elegans) un- second option is undesirable because it under-rep-
derwent considerable morphological change. Dif- resents biodiversity.
ferences among the three major groups of D. yina Practicality ofclassification is a concern forplant
in geographic distribution and nuclear genomic ar- systematists and other botanists, especially for
rangements conceivably extend to physiological those faced with accurately identifying plant taxa
differences of fundamental importance to survivor- with minimal time and resources. A system ofclas-
ship in distinct ecologica—l settings. sification that does not allow some plant taxa to be
Lasthenia californica The goldfield genus Las- identified on the basis of macroscopic morpholog-
thenia (Compositae) has been the subject of a re- ical characteristics alone generally has been resisted
MADRONO
224 [Vol. 47
by vascular plant systematists (but not bryologists sources. Skinner et al. (1995) identified over 150
or other botanists). Insofar as geography or habitat examples of rare, minimal-rank taxa ofCalifornian
characteristics often aid identification ofcryptic lin- vascular plants that needed systematic attention. j
eages, practical problems arising from giving for- The two rare taxa discussed below are examples of ^
:
mal taxonomic status to morphologically indistin- groups that were studied systematically in part to '
guishable groups may be limited. Eventually, ad- determine whether they warrant continued recog-
DNA
vances in analysis and computer technology nition and, forBlepharizonia, to determine whether
may trivialize the process ofscreening for diagnos- gene flow between species represents a conserva-
tic genetic markers, even in the field, thereby al- tion concern.
lowing botanists to be less reliant on morphological Blepharizonia plumos—a: Rare species or minor
characteristics for plant identification. I recognize, morphological variant? Baldwin et al. (2001) ex-
however, that for plant groups wherein morphology amined biosystematic and phylogenetic data for
has evolved even faster than DNA sequences in Blepharizonia to assess whether the common big
commonly examined, rapidly-evolving gene tarweed, B. plumosa subsp. viscida, warrants tax-
regions, e.g., in some examples of insular adaptive onomic distinction from the rare big tarweed, B.
radiation (see Baldwin et al. 1998), morphology plumosa subsp. plumosa. Keck (1959) regarded the
may provide a finer-scale perspective on evolution- two taxa as allopatric but recent field work by Rob-
ary lineages than most easily obtained DNA data. ert Preston established that the two taxa are mosa-
Strict adherence to a criterion of macroscopic ically sympatric in the northern Mt. Hamilton
diagnosability for all vascular plant taxa places un- Range, California, where Baldwin et al. (2001)
acceptable and unnatural limits on the information sampled the big tarweeds for crossing and molec-
content of our taxonomy and potentially jeopard- ular studies. Phylogenetic analysis of rDNA se-
izes important segments of biodiversity because of quence data yielded evidence for ancient diver-
a human bias toward recognizing only what can be gence between the two taxa of Blepharizonia rela-
seen with minimally-assisted human eyesight. From tive to timing of divergence between taxa in the
a biological standpoint, cryptic differences between sister-genus, Hemizonia (Fig. 3). Low interfertility
plant groups, e.g., in characters associated with of artificial hybrids corroborated phylogenetic evi-
ecophysiology, can be more important to the integ- dence for greater divergence between the big tar-
rity and fitness ofplant groups than differences that weed taxa than implied by Keek's (1959) taxono-
humans can perceive visually. From a conservation my.
perspective, taxonomic recognition of cryptically Baldwin et al. (2001) concluded that the two taxa
distinct natural groups may be important to ensure of Blepharizonia should continue to be recognized
their legal protection (e.g., only formally-named and warrant treatment as separate species, B. laxa
plant lineages are eligible for protection under the i=B. plumosa subsp. viscida) and B. plumosa (=B.
U.S. Endangered Species Act). Taxonomic status plumosa subsp. plumosa). Baldwin et al. (2001)
DNA
for cryptic groups also may help to ensure their also concluded from and artificial hybridiza-
protection from misguided restoration efforts that tion data that natural hybridization between B. laxa
result in combining germplasm from different evo- and B. plumosa probably does not pose a threat to
lutionary lineages treated within the same minimal- the continued existence of the rare big tarweed, B.
rank taxon, with consequent loss of lineage integ- plumosa. Preliminary evidence forphylogeographic
rity and possible outbreeding depression (see Mo- diversity uncovered within the rare B. plumosa
ritz 1999). From a more general perspective, ad- (Baldwin et al. 2001) should serve as a caution
herence to the belief that plant systematics is a against any conceivable restoration efforts that in-
science that seeks to discern real entities ofnature, volve moving germplasm of B. plumosa between
i.e., evolutionary groups, dictates that plant taxon- populations (especially north or south of the Liv-
omy should reflect rigorous hypotheses ofrelation- ermore Valley), at least until continuing studies of
pslhiisptircatahsesremthbalnagceosn.veBnaiseendt bountaavratiifliacbilaleoerviodveenrcsei,m-I linSeiadgaelcdeiaverkseictkiyi:inABlmeipnhoarrizvoa—rniiaantaroefcSo.mpdliepltoesdc.y-
suspect that widespread recognition of cryptic taxa pha or a distinct, rare lineage? Phylogenetic stud-
would result in only a modest refinement, not a ies ofSidalcea (Malvaceae) by Andreasen (in prep.;
major overhaul, of plant taxonomy. see also Andreasen and Baldwin, 2001) helped to
clarify evolutionary lineages in the genus, a group
III. Conservation prioritization. Mishler (1995) previously regarded as highly variable, taxonomi-
and others have discussed the potential value of cally difficult, and in need of systematic attention
phylogenetic data on the age and position of lin- (Hill 1993). Among the issues ofconservation con-
eages for prioritizing conservation efforts (re- cern examined by Andreasen was the evolutionary
viewed by Soltis and Gitzendanner 1998). On an status ofS. keckii, a narrowly endemic species from
even more fundamental level, modern systematic Tulare County, California, long thought to be ex-
data can help to resolve whether rare taxa ofques- tinct until rediscovered in 1992 (see Skinner and
tionable naturalness truly represent evolutionary Pavlik 1994). Assigning conservation priority to S.
lineages worthy of conservation attention and re- keckii has been complicated by uncertainty about
2000] BALDWIN: DISCOVERY AND CONSERVATION OF PLANT LINEAGES 225
Hemsiuzbosnpi.a cluoznugleifsotlaia classification at the lowest taxonomic levels, the
following recommendations are presented for plan-
ning systematic studies:
Hemizoniacongesta
subsp.lutescens
• Sample widely within accepted taxa. To test tax-
onomic hypotheses and to detect cryptic lineage
I-Viscida diversity, sampling within taxa across the range
(MPoanrtkefireelydCGor.a)de, ofphenotypic variation and across the geograph-
ical and ecological distribution has been produc-
-Viscida tive (see above). Examining only one exemplar
(LLNL,
SanJoaquinCo.) per taxon cannot reveal hidden diversity or prob-
lems in circumscription at the taxonomic level of
Viscida sampling.
(nwofBlackButte,
. SanJoaquinCo.) • Study herbarium collections. Apart from yielding
valuable data on morphological, ecological, and
V(TiessciiadaRoad, geographic variation within minimal-rank taxa,
AlamedaCo.) studies of herbarium specimens may reveal un-
described diversity more readily than new field
Plumosa
(MarshCreekRoad, DexNplAoration. Feasibility of extracting sufficient
ContraCostaCo.) for genetic analyses from small fragments
of dried plant material may allow both morpho-
p P(LlLuNmLo,sa logical and molecular characterization of new
species discovered in herbaria (e.g., Vargas et al.
1998).
- P(TleusImaosRoaad, • Take seriously the old ta.xonomic literature. A
AlamedaCo.) sampling focus on taxa recognized only in the
Fig. 3. Chronogram of one of two maximally parsimo- most recent taxonomic revision of a plant group
nious trees from phylogenetic analysis of nuclear riboso- may ensure a repeat of errors made in that sys-
mal DNA sequences of Blepharizonia and Hemizonia tematic treatment, especially if sampling within
(modified from Baldwin et al. 2001). Branch lengths were taxa is minimal. In addition to taking afresh look
optimized by maximum-likelihood to conform to an hy- at variation within a group, systematists may find
bpoethreesjiesctoefdeuvsoilnugtiaontarreye-rwaitedecolnisktealnichoyo,d-wrhaiticohtceostu.ldNontoet that taxa no longer recognized in modern treat-
the extensive divergence between the two, minimally in- ments represent evolutionary lineages warranting
terfertile, mosaically sympatic taxa of Blepharizonio rel- recognition (e.g., Chan 2000).
ative to divergence between the two representatives of • Voucher all specimens examined. Vouchering
Hemizonia. Biosystematic andphylogeneticdataledBald- specimens for systematic studies of groups cor-
win et al. (2001) to conclude that the two taxa of Ble- responding to minimal-rank taxa is perhaps even
pthraeraitmzeonntiaaseadcihstcionrctressppeocnidest.oAnbabtrureavliagtrioounpss: tPhlatumwoasrraan=t nmoomriecelsesveenltsiatlotehnasnurfeorthasttutdhieesideanttihtiigehseorftsaaxmo--
BV.ispcliudamos=a Bs.ensluaxastr{ic=tBo.[p=Bl.umpolsuamossuabsspu.bsvpi.scpildua]m.osSae];e pled plants are not misinterpreted by others. Doc-
Baldwin et al. (2001) fortree statisticsand supportvalues. umentation ofdetailed collection data is also crit-
ical for studies involving fine-scale sampling
within minimal-rank taxa (see Huber 1998).
• Examine multiple lines ofsystematic evidence. A
distinctiveness of the species from the morpholog-
ically similar, widespread species S. diploscypha. single line of systematic evidence (e.g., one
Andreasen (in prep.; see also Andreasen and gene) can be potentially misleading about rela-
tionships within a group (see Wendel and Doyle
Baldwin, 2001) sampled both species in a genus-
wide phylogenetic analysis of rDNA spacer se- 1998). Lineage sorting and hybridization are
qaunednSc.eskeicnkiSiidwaelrceea.moSshtecfloousneldytrhaetlaSt.edditpoloosnceypahna- ymoourneglpilkaenlty gtroouapfsfectthanevionluotlidonlairnyeagpeast.teErnxsami-n
ining multiple, unlinked gene regions or molec-
other, as expected, but that representatives of each
species constituted highly divergent lineages. Based ular and other lines of data (e.g., morphology or
on her findings, Andreasen concluded that S. keckii cytology) should increase the likelihood of
is worthy of continued taxonomic recognition and achieving an accurate understanding of natural
conservation attention. plant groups.
• Sample the rare taxa. The potential value to con-
Recommendations for Systematic Studies servation biology of gaining additional system-
atic data on rare plants makes the efforts required
To promote further progress by systematists in to sample rare taxa worthwhile. Most modern
the discovery of plant diversity corresponding to molecular systematic methods involve use ofthe
minimal-rank taxa and in the refinement of plant polymerase chain reaction (PCR), which requires
MADRONO
226 [Vol. 47
only minimal DNA (see Hillis et al. 1996). Mo- base management and communication, changes
lecular data from rare plants can be obtained that reflect an improved understanding ofnatural
from minute amounts offresh or dried (e.g., her- groups are valuable and worth adopting. From a
barium) tissue without impacting populations or conservation perspective, names are expendable;
significantly damaging voucher specimens. natural plant groups are irreplaceable.
• Study biological characteristics oftheplants. In- • Bear in mind that recognized species or infra-
cluding an experimental biosystematic compo- specific taxa are not necessarily minimal units of
nent (e.g., from artificial hybridizations or com- biodiversity. As noted above, unrecognized, ev-
mon gardens) and field component (e.g., polli- olutionarily distinct lineages may exist within a
nation ecology, demography) in modern system- species, subspecies, or variety (also see Soltis
atic studies can yield valuable biological data for and Gitzendanner 1998). Research efforts to dis-
resolving fine-scale diversity within a group and cern any undetected diversity within minimal-
may lead to insights into evolutionary processes rank plant taxa using modern systematic ap-
affecting diversification (see Baldwin 1995). proaches (e.g., phylogeographic studies) have
Studies of population-genetic structure within been minimal (see above). Available data sug-
lineages can provide critical biological data for gest that cryptic lineages often show some de-
resolving microevolutionary dynamics of popu- gree of geographic distinction (see Avise 2000).
lations and for refining conservation strategies Efforts to protect taxa throughout theirgeograph-
(e.g., Bushakra et al. 1999). ical and ecological ranges are therefore warrant-
• Communicate with other field botanists. Close ed not only to ensure survival oflocally adapted
communication and cooperation with profession- populations and overall allelic diversity within a
al and amateur field botanists is especially valu- group (Endler 1977; Chesser 1983) but also to
able for promoting discovery and conservation preserve potentially distinct evolutionary lin-
of plant diversity. The reciprocal flow of knowl- eages.
edge that can develop between systematists and • Resistproposals to use non-local germplasm in-
other field-immersed plant biologists enriches discriminately in restoration efforts. Use ofnon-
botany in general and can lead to a more inten- local germplasm in restoration efforts may result
sive, well-focused effort toward detecting and in extensive hybridization between evolutionarily
conserving diversity than would be otherwise distinct but cryptic lineages and consequent loss
possible (see Ertter 2000a, b). of biodiversity (see Storfer 1999). This concern
• Publishfindings andfollow through on taxonom- is especially important given the increasing prev-
ic changes. Other biologists, especially those in- alence of mitigation efforts seeking to augment
volved in biodiversity assessment and conserva- rare plant populations in protected areas with
tion (e.g.. Skinner and Pavlik 1994), rely on for- propagules or mature plants translocated from
mal taxonomic treatments and other publications other populations slated for destruction. The
by the systematic community. Translating perti- well-intentioned practice of augmentation may
nent results of systematic studies into taxonomic be justifiable to prevent or overcome inbreeding
changes is a potentially tedious but necessary depression, e.g., if the populations involved are
step to ensure that newly discovered evolution- remnants of a more continuous metapopulation
ary lineages and new understanding of the cir- fragmented by human-related activities or if ge-
cumscriptions and positions of monophyletic neological and population genetic data indicate
plant groups in general are recognized by others. that declining populations are of a common re-
gional lineage and share similar genetic structure
Recommendations to the Conservation (see Moritz 1999). Indiscriminate translocation
Community of plants from one population to another has po-
tential to do much harm to biodiversity and to
Based on the evidence from phylogenetic studies
our prospects for understanding the evolution or
that circumscriptions of some minimal-rank taxa
population-genetic structure ofnatural plant pop-
misrepresent or under-represent biodiversity, I sug- ulations.
gest that the following recommendations be adopt- In the absence ofgenealogical and population-
ed by the conservation community in the interests genetic data, proposals for augmentation of nat-
of preventing loss of natural plant groups:
ural populations with non-local seed should be
• Regard taxa as hypotheses of natural groups viewed with the same skepticism as the univer-
subject to change. Some refinements to our un- sally objectionable idea of intermixing germ-
derstanding of the composition and position of plasm from unquestionably distinct but interfer-
natural plant groups are inevitable and desirable tile, naturally allopatric species. Even ifthe pop-
to ensure that conservation efforts are well di- ulations to be intermixed do not represent highly
rected. divergent evolutionary lineages, potential still
• Accept and encourage taxonomic changes based exists foroutbreeding depression from loss oflo-
on solid evidence of natural groups. Although cal adaptation or breakdown of co-adapted gene
taxonomic changes create difficulties in data- complexes (see Templeton 1986; Slatkin 1987;
—
2000] BALDWIN: DISCOVERY AND CONSERVATION OF PLANT LINEAGES 227
Moritz 1999; Storfer 1999). Planting of wild- Arnold, M. L. 1997. Natural hybridization andevolution.
flowers along roads and highways is another Oxford University Press, New York, NY.
widespread practice with similar potential for re- AviSE, J. C. 2000. Phylogeography: The history and for-
ducing biodiversity and confounding scientific bmraitdigoen, MofA.species. Harvard University Press, Cam-
investigation of natural plant populations. Axelrod, D. I. 1992. Miocene floristic change at 15 Ma,
• Consider cryptic biodiversity in conservation Nevada to Washington, U.S.A. Palaeobotanist 41:
planning. Even if systematists decline to recog- 234-239.
nize cryptic lineages as taxa, conservationists Babcock, E. B. and H. M. Hall. 1924. Hemizonia con-
can plan forpreservation ofcryptic groups in the gesta: A genetic, ecologic, and taxonomic study of
interests of preserving unnamed, as well as the hay-field tarweeds. University ofCalifornia Pub-
named, biodiversity. Under the U.S. Endangered lications in Botany 13:15-100.
Species Act, evolutionarily significant but mor- Baldwin, B. G. 1995. A new prospect for California bot-
phologically indistinct and unnamed lineages of any: Integrating biosystematics and phylogenetics.
Madrofio 42:154-167.
vertebrates are eligible for protection; similar
1999a. Deinandra hacigalupii (Compositae
protection for cryptic plant groups may be pos- Mad.iinae), a new tarweed from eastern Alameda
sible to achieve. Geographical and ecological cri- County, California. Madrono 46:55-57.
teria have been used effectively for recognizing . 1999b. New combinations and new—genera in the
cryptic vertebrate lineages (e.g., salmonids) and North American tarweeds (Compositae Madiinae).
also may be useful for identifying various cryptic Novon 9:462-471.
plant groups. , D. J. Crawford, J. Francisco-Ortega, S.-C.
Kim, T. Sang, and T. F Stuessy. 1998. Molecular
Conclusions phylogenetic insights on the origin and evolution of
oceanic island plants. Pp. 410-441 in P. S. Solds, D.
The potential for modern systematics to play an E. Soltis, and J. J. Doyle (eds.). Molecular system-
important role in the discovery and conservation of aticsofplantsII: DNA sequencing. KluwerAcademic
fine-scale plant biodiversity is enormous and most- Publishers, Boston, MA.
ulyseunotfappmeodl.ecTuhlaermaonvdempehnytloogfenseytsitcemmateitchsodtsowahrads wax,.R.20E0.1.PrAesbtioons,ystBe.maLt.icWeasnsdap,hyalnodgenMe.tiWceatshseesrs--
been perceived by some botanists as an alarming men—t of sympatric taxa in Blepharizonia (Composi-
tae Madiinae). Systemafic Botany 26:184-194.
diversion from the urgent business of finding and Boyd, S. andT. S. Ross. 1997. Sibaropsis (Brassicaceae),
describing previously undetected and, usually, en- a new monotypic genus from southern California.
dangered plant diversity ("fiddling while Rome Madrofio 44:29-47.
burns"). I suggest that the use of modern system- Bushakra, J. M., S. A. Hodges, J. B. Cooper, and D. D.
atic approaches, far from posing a threat to ad- Kaska. 1999. The extent ofclonality and genetic di-
vancing our knowledge of fine-scale biodiversity, versity in the Santa Cruz Island ironwood, Lyono-
can be an invaluable means of achieving rapid tharrmusfloribimdus. Molecular Ecology 8:471-475.
progress in the discovery and conservation ofplant Chan, R. K.-G. 2000. Molecular systematics ofthe gold-
fieldgenusLasthenia (Compositae: Heliantheaesensu
lineages.
lato). Ph.D. dissertation. University of California,
Berkeley.
Acknowledgments
Chesser, R. K. 1983. Isolation by distance: Relationship
Special thanks to Katarina Andreasen, Raymund K.-G. to management ofgenetic resources. Pp. 66-77 in C.
Chan, Staci Markos, and Lisa M. Schultheis for gener- Schonewald-Cox, S. Chambers, B. MacBryde,andW.
ously allowing me to present examples from their pub- Thomas (eds.). Genetics and conservation: A refer-
lished dissertations or (for KA) from an in-press postdoc- ence formanaging wildanimal andplantpopulations.
toral research article for this symposium presentation. Benjamin Cummings Publishing Company, London.
TJohhannkLs.aSltsrootthoerDafnoirehlelJ.pfCurlacwfoomrmde,ntBsrenotnDt.heMmiasnhulesrc,riapntd. ClauCsoernn,elJl. 1U9n5i1v.erSstiatgyesPirnestsh,e eItvhoalcuat,ioNnYof(prleapnrtisnpteecdiebsy.
The research discussed here was funded in part by the Hafner Publishing Co., New York, NY).
National Science Foundation (DEB-9458237), by gifts COATES, D. J. 2000. Defining conservation units in a rich
from Roderic B. Park and other generous Friends of the and fragmented flora: Implications for the manage-
Jepson Herbarium, and by the Lawrence R. Heckard En- ment ofgenetic resources and evolutionary processes
dowment Fund ofthe Jepson Herbarium. in south-west Australian plants. Australian Journal of
Botany 48:329-339.
Literature Cited Endler, J. A. 1977. Geographic variation, speciation and
clines. Princeton University Press, Princeton, NJ.
Andreasen, K. and B. G. Baldwin. 2001. Unequal evo- Ertter, B. 2000a. Our undiscovered heritage: Past and
lutionary rates between annual and perennial lineages future prospects for species-level field exploration.
ofchecker mallows (Sidalcea, Malvaceae): Evidence Madrono 47:237-252.
from 18S-26S rDNA internal andexternal transcribed . 2000b. Floristic surprises in North Americanorth
spacers. Molecular Biology and Evolution 18:936- ofMexico. Annals ofthe Missouri Botanical Garden
944. 87:81-109.
Angiosperm Phylogeny Group. 1998. An ordinal classi- Faith, D. P. 1996. Conservation priorities and phyloge-
fication for the families of flowering plants. Annals netic pattern. Conservation Biology 10:1286-1289.
of the Missouri Botanical Garden 85:531-553. Foster, R. I. 1972. Explosive chromosome evolution in
MADRONO
228 [Vol. 47
Downingia yina. Ph.D. dissertation. University of (ed.).TheJepsonmanual: HigherplantsofCalifornia.
California, Davis. University ofCalifornia Press, Berkeley, CA.
FuJii, N. 1997. Phylogeographic studies inJapanesealpine Raven, P. H. and D. I. Axelrod. 1978. Origin and rela-
plants based on intraspecific chloroplast DNA varia- tionships of the California flora. University of Cali-
tions. Bulletin of the Biogeographical Society ofJa- fornia Publications in Botany 72:1-134.
pan 52:59-69. Riddle, B. R., D. J. Hafner, L. F. Alexander, and J. R.
Graham, A. 1999. Late Cretaceous and Cenozoic history Jaeger. 2000. Cryptic vicariance in the historical as-
ofNorth American vegetation: North ofMexico. Ox- sembly of a Baja California Peninsular Desert biota.
ford University Press, New York, NY. Proceedings ofthe National Academy ofSciences of
HiLLis, D. M., C. MoRiTZ, AND B. K. Mable (eds.). 1996. the United States ofAmerica 97:14438-14443.
Molecular systematics. Sinauer Associates, Sunder- Rieseberg, L. H., D. E. Soltis, and J. D. Palmer. 1988.
land, MA. A molecular reexamination of introgression between
Hennig, W. 1966. Phylogenetic systematics. University of Helianthus annuus and H. bolanderi (Compositae).
Illinois Press, Urbana, IL. Evolution 42:227-238.
Hickman, J. C. (ed.). 1993. The Jepson manual: Higher Schaal, B. A. and K. M. Olsen. 2000. Gene genealogies
plants of California. University of California Press, andpopulation variation in plants. Proceedings ofthe
Berkeley, CA. National Academy of Sciences of the United States
Hill, S. R. 1993. Sidalcea. Pp. 755-760 in J. C. Hickman of America 97:7024-7029.
(ed.).TheJepsonmanual: HigherplantsofCalifornia. , D. A. Hayworth, K. M. Olsen, J. T. Rauscher,
University ofCalifornia Press, Berkeley, CA. and W. a. Smith. 1998. Phylogeographic studies in
HuBER, J. T. 1998. The importance ofvoucher specimens, plants: Problems and prospects. Molecular Ecology
with practical guidelines for preserving specimens of 7:465-474.
the majorinvertebratephylaforidentification.Journal Schultheis, L. M. 1999. SystematicsofDowningia(Cam-
of Natural History 32:367-385. panulaceae) and population structure ofD. bicornuta.
Keck, D. D. 1959. Subtribe Madiinae. Pp. 1106-1129 in Ph.D. dissertation. University ofCalifornia, Berkeley.
P. A. Munz, A California flora. University of Cali- Shaw, A. J. 2000. Molecular phylogeography and cryptic
fornia Press, Berkeley, CA. speciation inthe mosses,Mielichhoferiaelongataand
Maddison, W. p. 1997. Gene trees in species trees. Sys- M. mielichhoferiana (Bryaceae). Molecular Ecology
tematic Biology 46:523-536. 9:595-608.
Markos, S. 2000. Evolutionary patterns in Lessingia Skinner, M. W. and B. M. Pavlik(eds.). 1994. California
(Compositae, Astereae) using morphological, molec- Native Plant Society's Inventory of rare and endan-
ular, and biosystematic data. Ph.D. dissertation. Uni- gered vascularplants ofCalifornia, 5th ed. California
versity ofCalifornia, Berkeley. Native Plant Society, Sacramento, CA.
AND B. G. Baldwin. 2001. Higher-level relation- , D. P Tibor, R. L. Bittman, B. Ertter, T. S.
ships and major lineages of Lessingia (Compositae, Ross, S. Boyd, A. C. Sanders, J. R. Shevock, and
Astereae) based on nuclear rDNA internal and exter- D. W. Taylor. 1995. Research needs for conserving
nal transcribed spacer(ITS and ETS) sequences. Sys- California's rare plants. Madrono 42:211-241.
tematic Botany 26:168-183. Slatkin, M. 1987. Gene flow andthegeographic structure
MiSHLER, B. D. 1995. Plant systematics and conservation: of natural populations. Science 236:787-792.
Science and society. Madrono 42:103-1 13. Smocovitis, V. B. 1997. G. Ledyard Stebbins, Jr. and the
. 2000a. Deep phylogenetic relationships among evolutionary synthesis (1924-1950). American Jour-
"plants" and their implications for classification. nal ofBotany 84:1625-1637.
Taxon 49:661-683. Soltis, D. E., P S. Soltis, and B. G. Milligan. 1992.
. 2000b. The needforintegrated studiesoftheCal- Intraspecific chloroplast DNA variation: Systematic
ifornia flora. Madrofio 47: this issue. and phylogenetic implications. Pp. 117-150 /// P. S.
MoRiTZ, C. 1994. Defining "Evolutionarily Significant Soltis, D. E. Soltis, and J. J. Doyle (eds.). Molecular
Units" forconservation. Trends in Ecology and Evo- systematics ofplants. Chapman and Hall, New York,
lution 9:373-375. NY.
vati.o1n.995P.hiUlsoseosphoifcamlolTercaunlsaarctpihoynlsogoefnitehse fRoorycaolnsSeor-- lecu,lRarS.sySsotletmiast,icasndofJ.plJa.ntDsoyIlI:eD(eNdsA.).se1q9u9e8n.ciMnog-.
ciety of London. Series B, Biological Sciences 349: Kluwer Academic Publishers, Boston, MA.
113-118. , M. A. GiTZENDANNER, D. D. StRENGE, AND P. S.
. 1999. Conservation units and translocations: Soltis. 1997. Chloroplast DNA intraspecific phylo-
Strategies for conserving evolutionary processes. geography of plants from the Pacific Northwest of
Hereditas 130:217-228. North America. PlantSystematicsandEvolution 206:
, AND D. P. Faith. 1998. Comparative phylogeog- 353-373.
raphy and the identification of genetically divergent Soltis, P. S. and M. A. Gitzendanner. 1998. Molecular
areas for conservation. Molecular Ecology 7:419- systematicsandtheconservationofrarespecies.Con-
429. servation Biology 13:471-483.
Olsen, K. M. and B. a. Schaal. 1999. Evidence on the Stebbins, G. L., Jr. 1950. Variation and evolution in
origin of cassava: Phylogeography ofManihot escu- plants. Columbia University Press, New York, NY.
lenta. Proceedings of the National Academy of Sci- Storfer, a. 1999. Gene flow and endangered species
ences ofthe United States ofAmerica96:5586-5591. translocations: A topic revisited. Biological Conser-
Ornduff, R. 1966. A biosystematic survey of the gold- vation 87:173-180.
field genus Lasthenia. University of California Pub- SwoFFORD, D. L., G. J. Olsen, P. J. Waddell, and D. M.
lications in Botany 40:1-92. HiLLis. 1996. Phylogenetic inference. Pp. 407-514 in
. 1993. Lasthenia. Pp. 298-300 in J. C. Hickman D. M. Hillis, C. Moritz, and B. K. Mable (eds.). Mo-