Printer Friendly

Allotetraploids in patagonia with affinities to western North American diploids: did dispersal or genome doubling occur first?



Of many trans-oceanic biogeographic patterns, the amphitropical disjunct distribution of plant taxa between western North America and western South America has intrigued botanists for over a century (e.g., Gray & Hooker, 1880; Bray, 1898, 1900; Grant, 1959; Raven, 1963; Carlquist, 1966, 1983). The floras between the temperate regions of these two continents are far from homogenous, yet the cumulative result of chance events over thousands of millennia has resulted in a discemable pattern of disjunct relationships. Raven (1963) compiled over 130 instances of species or species pairs with established distributions in temperate regions on both continents that are separated by some 7000 km. Long distance dispersal is the favored hypothesis for diaspore movement in most instances (see also Thome, 1972; Carlquist, 1983), and in the majority of cases, North America has served as the source and South America the sink. Many of these disjunct species groups share common characteristics (Raven, 1963) that function as preadaptations favoring establishment. For example, temperate disjuncts tend to be annuals from open habitats, primarily self-pollinating, and have seed, fruit, or other diaspore characteristics that aid in their attachment to migratory birds (Cruden, 1966; Carlquist, 1983). In a recent review, Wen and Ickert-Bond (2009) characterized the amphitropic disjunet pattern of the Americas as common and well-known, with a non-comprehensive list of 24 studies that have addressed this pattern to some measure within particular plant groups.


Apart from pattern, processes play an essential role in the establishment and ultimate fate of dispersed lineages. Established disjunct species evidence diaspore movement and provide a baseline for assessing the frequency of dispersal but this baseline recognizably fails to reflect dispersal with precision. Undoubtedly, many dispersal events failed to establish or have since been extirpated. Others, such as in the Gilia laciniata complex, have given rise to new species following dispersal (Morrell et al., 2000). Multiple dispersals are evident in yet other genera, such as Sanicula (Vargas et al., 1998), and multiple dispersals as well as speciation in Tiquila (Moore et al., 2006). Phylogenetic and biogeographic studies aided by DNA sequence data are illuminating the processes of speciation associated with long distance dispersal, a combination that, as noted by Mummenhoffand Franzke (2007), provides spectacular examples of evolutionary biology.

Allopolyploidy is recognizably an important mode of speciation in plants (Otto & Whitton, 2000; Soltis & Soltis, 2000, 2009). Stebbins (1942, 1947) early championed the value of allopolyploids in understanding plant biogeography. Raven (1963:157) reviewed some instances of ploidal levels that vary within particular species groups between North and South America and provided chromosome data for many other disjunct species pairs and groups. An important question when allopolyploidy is encountered in conjunction with intercontinental disjunct distributions is the timing of allopolyploidization relative to dispersal (Fig. 1). In a recent discussion, examples of both pre- and post-dispersal allopolyploidization were presented (Mummenhoff & Franzke, 2007). In Microseris, allopolyploidization is hypothesized to have occurred prior to dispersal from North America, followed by extirpation of the allotetraploid and one parent in North America and subsequent diversification of the allotetraploid in Australia and New Zealand (Vijverberg et al., 2000). Australian Nicotiana are similarly thought to have experienced allopolyploidization prior to their dispersal from South America (Goodspeed, 1954; Aoki & Ito, 2000). In contrast, the dispersal of diploids to Australia, subsequent allopolyploidization, and extirpation of the dispersed diploids from Australia is hypothesized for Lepidium (Mummenhoff et al., 2004). In Lepidium, the hypothesized timing of allopolyploidization relative to dispersal is supported by evidence that the diploid progenitors are presently established on different continents. For Microseris and Nicotiana, in contrast, the favored hypothesis is equivocal with respect to available data.


The phlox family exhibits several instances of temperate amphitropical disjunct distributions unambiguously attributable to multiple dispersal events, both with and without subsequent speciation. Polemoniaceae contains some 400 species among 26 genera (Porter & Johnson, 2000). The temperate subfamily, Polemonioideae (22 genera/340 species) is predominantly western North American in distribution, with some 70 % of this distribution occurring within California, USA. Fifteen Polemonioideae species from nine genera occur in South America (discounting putatively recently naturalized species such as Collomia grandiflora; Puntieri & Brion, 2005), primarily along the west coast, which shares many climatic and ecological similarities to western North America. Two species, Microsteris gracilis and Polemonium micranthurn, are conspecific with their North American counterparts, while the rest are unique to South America. This uniqueness implies either stasis in South America following dispersal with concomitant extirpation (or lack of discovery) in North America, or sufficient change in South America following dispersal to diagnose these lineages now as distinct species. The origins of Gilia valvidensis and Gilia laciniata have been explored by Morrell et al. (2000), and relationships of Leptosiphon (Linanthus) pusillis by Bell and Patterson (2000).

Here, we investigate the origins of Collomia biflora and Navarretia involucrata, South American species from sister genera. In both instances, a single species is thought to exist in South America that is unique relative to any of the described North American diversity. This diversity is sufficiently rich (14 and 34 species, respectively) to support a North American source and South American sink dispersal hypotheses for these two species. This hypothesis is further supported by early phylogenetic analyses (Spencer & Porter, 1997; Johnson & Johnson, 2006). Whereas Collomia biflora has long been recognized to be polyploid, Navarretia involucrata has been assumed to be diploid. We examine the pattern of diversification and timing of diversification relative to dispersal in both species.

Materials and Methods

Phylogenetic relationships in Collomia have been recently reconstructed using a variety of chloroplast and nuclear markers (Green, 2010; Green et al., in prep.), but the pattern and processes detailed here were only tangentially discussed. Based on these analyses, we confined the sampling here to the annual, linear leaved Collomia species, with C. tracyi and C. tinctoria designated as the outgroup. This sampling excludes the perennial species (Collomia section Collomiastrum) and the annual species with more than one ovule per locule (Collomia section Courtoisia), neither of which are important to understanding the amphitropical disjunct distribution of this genus. The bulk of our data for Collomia comes from Green (2010), with additional populations of Collomia biflora, Collomia grandiflora, and Collomia linearis included here. In total, our sampling includes 34 populations representing nine species (Appendix I). Based on analyses of ITS sequences and morphology (Spencer & Porter, 1997) and ongoing work in our lab, we confined our sampling of Navarretia to the monophyletic Navarretia section Navarretia, with Navarretia tagetina, the sister to the remaining species in this section, specified as the outgroup. We sampled 25 populations representing all 11 named species in this section plus three additional species we have recently discovered that will be formally named elsewhere (Johnson et al., in prep).

For all sampled populations of Collomia and Navarretia, we sequenced four chloroplast regions (5'trnK intron and adjacent 5' two-thirds of the matK gene; trnL-trnL intron and trnL-trnF intergenic spacer; trnS-trnG intergenic spacer; and thepsbM-trnD intergenic spacer) and the nuclear ITS-l, 5.8S, and ITS-2 region using primers as specified in Johnson et al. (2008). We used a somewhat reduced sampling (27 populations of Collomia for all 9 species, and 19 populations of Navarretia for 10 of the 11 named species and all three unnamed species) for low copy nuclear genes, primarily due to low quality DNA from herbarium specimens in the excluded samples; these low copy nuclear genes include portions of PISTILLATA (PI) and g3pdh in Collomia, and PI alone in Navarretia. Primers for g3pdh follow Strand et al. (1997), while primers developed in the Johnson lab by Weese (2004) for Polemoniaceae were used to access a 5' portion of the PI gene, with internal primers developed as needed. These primers include PI-7F (5'-AGAGGAAAGAJTGAGATAAAGAGG-3') and PI-450R (5'-TTCTCTYCCTCCARCATCATF-3') used for amplification and sequencing, with additional internal primers PI-900R (5'-ATCATFCTCTTTCTTGATCC-3'), PI-880F (5'-ATCCATGGACAGATCTGGTAA-3'), PiV1F (5'-CATAGGTTGGTTGAGAT CTTGG-3'), and PiV1R (5'-CTATCATTCTCTTTTTTTGATCCTG-3') used for sequencing if needed. We followed the methods of Johnson and Johnson (2006) for DNA extraction, PCR amplification, sequencing, and cloning (for the low copy nuclear genes and some ITS).

We compiled individual matrices for Collomia and Navarretia for the combined cpDNA regions, the ITS region, and each low copy nuclear gene. We manually aligned sequences within these matrices using Se-A1 2.0 (Rambaut, 2002), trimmed a few poly-N repeats with hypervariability (e.g., two introns in the PI gene contain poly CT strings of hypervariable length), and applied simple indel coding using SeqState (Muller, 2005). We made no attempt to analyze the matrices combined because visual comparison of results among the different matrices provides a stronger interpretive base for addressing dispersal hypotheses than would a single combined analysis of the chloroplast and putatively unlinked nuclear genes.

We performed parsimony analyses on each matrix using PAUP*4.0bl0 (Swofford, 2002) with equal weighting of all characters, heuristic searches employing 1000 replications of random addition, TBR branch swapping, and collapsing branches with minimum lengths of zero (amb-). We assessed branch support via 100,000 replications of fast bootstrapping, and reconstructed base substitutions using DELTRAN optimization. DELTRAN optimization favored the reconstruction of many substitution between the outgroup and ingroup rather than along the two branches attached to the root, which better matched patterns of variation observed in these genera when matrices included broader (i.e., more phylogenetically distant) taxon sampling (unpubl. data) using either ACCTRAN or DELTRAN optimization.

We assessed the presence of a molecular clock in the cpDNA and ITS sequences using the likelihood ratio test and then calculated a rough approximation for the timing of dispersal in Collomia by multiplying the divergence between C. biflora and C. grandiflora (cpDNA), C. biflora and C. linearis (ITS), and C. grandiflora from Bariloche, Argentina, and C. grandiflora from the USA (cpDNA and ITS) by a range of published substitution rates for noncoding cpDNA (0.26-0.92 % Myr-1; Alsos et al., 2005; Mummenhoff et al., 2004) and ITS (0.172-.834 % Myr-1; Kay et al., 2006) sequences. We estimated the average divergence between these taxa from branch lengths obtained from maximum likelihood analyses performed in PAUP* after first estimating the best model and parameters (using AIC) with Modeltest 3.7 (Posada & Crandall, 1998).



Parsimony analyses recovered three minimum length trees from the cpDNA matrix, two from the ITS matrix, four from the PI matrix, and one from the g3pdh matrix. Focusing solely on the inferred affinities of the South American species, Collomia biflora appears with strong bootstrap support (100 %) in a clade with C. grandiflora in the cpDNA trees. A single ITS copy was recovered for C. biflora via direct sequencing and also via sampling of cloned PCR fragments; C. biflora appears with marginal bootstrap support (68 %) in a clade with C. linear& in the ITS trees. In the PI and g3pdh trees, the two homeologs recovered from cloning PCR fragments of C. biflora appear separately in clades with C. linear& and C. grandiflora. With PI, both clades are supported with high bootstrap values (100 %), whereas with g3pdh, the C. biflora--C. grandiflora clade has moderate support (72 %) and the C. biflora--C. linearis clade has only weak support (<50 %). A population of C. grandiflora putatively naturalized in Argentina appears more closely related to the homeolog of C. biflora derived from a C grandiflora like ancestor than to C. grandiflora itself in the PI and g3pdh trees.

Dispersal time estimates range from 137,000-485,000 yr based on the average cpDNA divergence between C. biflora and the sister clade of C. grandiflora; 92,000-328,000 yr based on the average cpDNA divergence between C grandiflora from Bariloche, Argentina and the other C. grandiflora accessions within its same clade; 233,000-1,130,000 yr based on average ITS divergence between C. biflora and C linearis; and 388,000-1,881,000 yr based on the average ITS divergence between C grandiflora from Bariloche, Argentina and the other accessions of C grandiflora.


Parsiomony analyses recovered two minimum length trees from the cpDNA matrix, five from the ITS matrix, and three from the PI matrix. The South American species Navarretia involucrata is placed isolated as sister to the remaining species exclusive of N. tagetina in the cpDNA trees (this result holds true when species outside section Navarretia are used to root the tree as well; results not shown). Similar to C. biflora, a single ITS copy was recovered from both direct sequencing and sequencing of cloned PCR fragments in N. involucrata; this sequence places N. involucrata with strong bootstrap support (98 %) within a largely unresolved clade also containing N. prostrata, N. myersii, N. leucocephela, and N. willamettensis. For the PI gene, two homeologs were recovered from cloning PCR fragments not only in N. involucrata, but also in N. propinqua, N. saximontana, N. willametensis, and N. leucocephela. Focusing again on the South American species, one homeolog from N. involucrata appears within a weakly supported clade including N. prostrata, N. fossalis, N. leucocephela, and N. willamettensis; within this clade, the placement of N. involucrata as sister to N, prostrata is strongly supported (99 % bootstrap). The other N. involucrata homeolog is placed weakly (<50 % bootstrap) in a clade that includes two recently discerned but as yet unnamed species from North America.


Origins--Collomia biflora

Collomia biflora is a common annual foothills species along the Andean corridor in Argentina, Chile, and Bolivia. It is more or less upright in habit, bears flowers in head-like inflorescences, and is notable for having bright red adaxial corolla lobes. Collomia biflora is morphologically plastic and can occur in unbranched forms terminating in a single head-like inforescence, or in much branched forms, as wide as tall, with many flowering heads. This plasticity has resulted in some nomenclatural confusion, but is not unexpected given that its closest extant relatives identified here, C. linearis and C. grandiflora, are notably morphologically plastic as well (e.g., Wilken, 1978, 1982). In North America, C. linearis is the most widespread Collomia species, ranging from the west to the east coast of the United States and Canada, and absent from only the "southern states" (USDA Plants database, 2010). As with the majority of annual Polemoniaceae, it is particularly abundant in western North America and can be found in virtually any mountain range within its distribution. It has pink (seldom white) corolla lobes and occupies an ecological zone parallel to that inhabited by C. biflora in South America. Collomia grandiflora is also widely distributed in Western North America and possesses, next to C. linearis, the broadest distribution of any Collomia species. It co-occurs with C. linearis in many locations, and is unique in the genus for producing salmon to yellow chasmogamous flowers as well as minute whitish cleistogamous flowers. Underscoring the morphological plasticity in these North American species, Wherry (1944) lists five subspecific taxa for C. linearis and five for C. grandiflora, none of which are in current use.

Morphological plasticity in Collomia biflora, compounded by the rule of priority in the Botanical Code, has led to some nomenclatural confusion involving the identity of Collomia diversity in South America. Eight names have been applied to what we here consider a single species. Collomia linearis (Cav.) Nutt, C. bilfora (Ruiz &Pav.) Brand, and C. cavanillesii Hook. & Am. are the most commonly applied names in botanical literature, while C. coccinia Lehm. ex Lindl. remains popular in the horticultural trade. Wherry (1944), in opposition to Philippi (1895), argued that the epithat 'linearis' belongs to the North American species so named when Collomia was erected by Nuttall (i.e., Collomia linearis Nutt.), rather than to the species from South America originally named in the genus Phlox (i.e., Phlox linears Cav. [equivalent to] Collomia linearis (Cav.) Nutt.). This conclusion is supported by the current code (ICBN Article 11.4; McNeill et al. 2006). Wherry also cast doubt as to the true identity of C. biflora given that the protologue describes a plant with paired blue flowers and considered C. cavanillesii the correct name at the species level for South American Collomia. Having not yet viewed the type for C. biflora (as neither did Wherry), the lead author followed this line of reasoning in referring South American material to C. cavanillesii in an earlier publication (Johnson & Johnson, 2006). Other authors have also suggested South American Collomia likely represents but a single species (Reiche, 1910; Wilken et al., 1982). Soriano (1947) disagreed with Wherry regarding the validity of the combination C. linearis (Cav.) Nutt, but followed Wherry in restricting C. biflora to a species with non-red, solitary or paired flowers, noting that most of the material referred to C. biflora actually corresponds to C. linear& (Cav.) Nutt (with C. cavanillesii considered by him a synonym of this species). Grant (1959) took a different approach, and recognized both C. biflora and C. cavanillesii in South America, with suggested affinities to C. grandiflora and C. linearis, respectively (with corolla size an apparent distinguishing feature based on specimen annotations at CONC). A recent worker, E. Servat, reviewing specimens at SI recognized C. cavanillesii, C. biflora, and C. coccinea, distinguishing the three species by branching, bract and leaf similarity, and leaf lobing. This confused nomenclature deserves attention and is being addressed in conjunction with a thorough examination of types (Johnson et al. in prep.). In approaching this study, the authors' working hypothesis is that just two named Collomia species are presently in South America: Collomia biflora circumscribing tetraploid material typically with red corolla lobes regardless of corolla size and plant habit, and Collomia grandiflora circumscribing diploid material with salmon flowers reported as naturalized in Argentina (Puntieri & Brion, 2005). We were unable to include material of the "yellow collomia" from Chile referred to C. cavanillesii (Hoffmann et al., 1998) but discuss some possible relationships below.

Sugiura (1936) and Flory (1937) report Collomia coccinea as tetraploid (synonymized to C. cavanillesii by Grant, 1959 and here treated as C. biflora; a direct tetraploid count for C. biflora also exists; Wilken, 1986). Raven (1963) records a diploid count for C. cavanillesii, but it is uncertain where this count originated or if it is a typographical error given that he acknowledged Grant (1959 and pers. comm.) as his source of information for Polemoniaceae. Following Grant (1959), Raven recognized both C. cavanillesii and C. biflora in South America, but reversed their putative sister species to C. grandiflora and C. linearis, respectively. Following this taxonomy, two sister pair relationships between single North American species and single South American species requires two dispersal events to South America within Collomia. Also, given Grant's knowledge of the polyploid condition in material he referred to C. cavaniIlesii, an affinity between this material and only C. linearis would imply autopolyploidy as the mechanism of genome doubling in the South American species. Johnson and Johnson (2006) first demonstrated that Collomia biflora was an allotetraploid using evidence from nuclear idhA and idhB genes.

An allopolyploid origin for Collomia biflora opens the door for additional hypotheses of dispersal relative to diversification. Allopolyploidization could have occurred in North America prior to dispersal, with a single dispersal event to South America followed by extirpation of the allopolyploid in North America (three events: allopolyploidization, dispersal, extirpation; Fig. la). Altematively, two diploid species phylogenetically near C grandiflora and C linearis could have dispersed to South America with the allopolyploidization event occurring in South America followed by the extirpation of the two diploid species (five events: dispersal twice, allopolyploidization, extirpation twice; Fig. lb). Eliminating extirpation as necessary components of either hypothesis still favors allopolyploidization prior to dispersal from a simple parsimony perspective (Fig. la; two steps, versus three in Fig. lb).

Though less parsimonious, the second hypothesis of allopolyploidization following dispersal is supported by the data we present here. We sampled the naturalized population of Collomia grandiflora reported from Bariloche, Argentina (Figs. 2 and 3). This population occurs along the roadside in disturbed ground. The population is small, was relatively recently discovered, and is the only reported instance of this species in Argentina (Puntieri & Brion, 2005). The physical description of the plants, including their flowers, match C. grandiflora as it is occurs in the United States. Furthermore, this species has been reported as naturalized in Europe (Pysek et al., 2002) and Australia (Hussey et al., 1997). The conclusion by Puntieri and Brion (2005) that the Bariloche population represents a naturalized introduction is thus logical; nevertheless, DNA sequence data indicate this population more likely represents a remnant of a much earlier dispersal event that established in South America prior to the allopolyploid formation of C. biflora.

Specifically, PI and g3pdh sequences recover the Bariloche C. grandiflora as sister to C. biflora's maternal homeolog rather than among North American C. grandiflora (Fig. 3). Both low copy nuclear genes reveal synapomorphies between the Bariloche accession and C. biflora, as well as the lack of synapormophies in the Bariloche accession that unite the remaining accessions of C. grandiflora into a well supported clade. The ITS region in C. biflora has been homogenized through gene conversion or loss of loci to the single repeat typical of its paternal parent, as has been documented in some allotetraploids (e.g., Dierschke et al., 2009). Chloroplast DNA places the Bariloche accession, with some divergence, in a clade of C. grandiflora accessions from disparate locations in the United States that also includes C. biflora. Though cpDNA does not unite the Bariloche accession with C. biflora exclusive of C. grandiflora, the pattern of variation with both the Bariloche accession and C. biflora nesting within C. grandiflora indicates that the species that dispersed to South America giving rise to C. biflora's maternal line was C. grandiflora itself. The yellow collomia from Metropoliana Chile (Hoffmann et al., 1998) may represent additional populations of this relic or derivatives from it, a hypothesis we are actively pursuing with recently obtained material. Relictual populations of C. linearis are unknown; specimens from CONC with smaller flowers and description of interior corollas 'deep pink' are late season plants. The lead author revisited one population earlier in the season and found only red, average-sized flowers (with 'red' itself being subjective). Persistent or not, our data indicate that the allopolyploidization event giving rise to C. biflora occurred in South America, and thus at least two dispersals of diploid species from North America also occurred.

Origins - Navarretia involucrata

Navarretia involucrata inhabits vernal pools and seasonally moist depressions along the Andean corridor in Chile and Argentina. Plants are small, spreading, with small corollas and an autogamous breeding system. This habit and predisposition to seasonally moist habitats are shared with its North American relatives, particularly N. leucocephela, N. prostrata, N. myersii, and N. fossalis-species that form the 'core vernal pool clade' of Spencer and Porter (1997). Remaining members of section Navarretia, such as N. tagetina, N. subuligera, and N. interexta are considered facultative or marginal vernal pool species; they are often associated with moist depressions and other seasonally wet habitats, but are not obligately tied to vernal depressions as are the core clade members. Grant (1959) and Spencer and Porter (1997) suggested N. leucocephela ssp. minima was the likely progenitor of N. involucrata.

Prior to work in our lab, reticulation and allopolyploidization have been scarcely discussed in the context of Navarretia. Instead, diversification has been thought to be divergent and possibly, in the case of the core vernal pool species, driven by isolation following fragmentation of a mid to late Pliestocene vernal pool that once covered the central plain of California, or by dispersal from pool to pool accentuated by genetic bottlenecks and drift. Our data are consistent, however, with an allopolyploid origin of N. involucrata as well as the North American species N. propinqua (long considered a subspecies of the diploid N. interetexta), and possibly N. willametensis and at least some populations of N. leucocephela. Although our sampling is less dense in Navarretia compared to Collomia, these results indicate that allopolyploidy has occurred multiple times within the vernal pool clade, which makes discerning the timing of alloplolyploidization in N. involucrata relative to dispersal a matter of conjecture without a relictual diploid persisting in South America.

The paternal parent of N. involucrata belongs to the core vernal pool clade as evidenced by ITS sequences (Fig. 4), with particularly close affinities to N. prostrata as evidenced by PI sequences (Fig. 5). The maternal parent is ambiguous. ITS sequences again show complete homogenization to the paternal homeolog. Chloroplast DNA places N. involucrata clearly within section Navarretia, but as sister to all species except N. tagetina rather than as sister to a single species. PI places, but without substantial bootstrap support, the maternal homeolog of N. involucrata as sister to two unnamed North American species, but the divergence along branches from their hypothetical ancestor is long relative to the branch uniting these species, and trees three steps longer fail to resolve this relationship. Although we have initiated parallel studies with other low copy genes such as g3pdh and idh, our sampling is not yet sufficient to make inferences from those genes beyond confirming an allopolyploid pattern of variation in Navarretia involucrata.

We included the three as yet unnamed Navarretia discovered through our efforts, all native to North America, to see if sequence data implicated these taxa in the origins of N. involucrata. Though this does not appear to be the case, their discovery, combined with the discovery and naming of N. saximontana, N. willametensis, and a subspecies of N. leucocephela in the past decade (Spencer & Spencer, 2003; Bj6rk, 2002) along with N. myresii (with two subspecies; Day, 1993, 1995) in the decade before that, indicates that our knowledge of species diversity in this portion of Polemoniaceae may yet be imperfect. Our sampling of N. involucrata from South America is recognizably limited and completely lacking in Chilean diversity. A diploid maternal parent or even closer paternal parent than N. prostrata may yet show up with additional sampling and closer attention to morphological detail on both continents.

Timing of Dispersal

Periods during which amphitropical regions shared the greatest ecological similarity, as well as fluctuating climatic regimes, are likely the times during which the greatest opportunity for dispersal and establishment existed (Axelrod, 1950; Raven, 1963). For the western North American-western South American disjunctions explored here, dispersal likely occurred in the Pliestocene (Raven and Axelrod 1978). Though rough and assumption-laden, our estimates of divergence times in Collomia derived from a range of nucleotide substation rates for both chloroplast and ITS sequences place the time of dispersal within the Pliestocene, with average values in the mid Pleistocene (Ionian period). Elsewhere in Polemoniaceae, Morrell et al. (2000) estimated a divergence time of 1-3 Myr in the South American Gilia laciniata complex, while Leptosiphon pusillis was estimated to have diverged 1-2 Myr from its North American relatives (Bell & Patterson, 2000). Thus, the divergences estimated here are within the range of divergence times estimated for relatives with similar disjunct distributions. We did not attempt to estimate a divergence time for Navarretia given extreme differences between the available maternal and paternal nodes as well as a paucity of change in the paternal line. However, vernal pools were common on both continents during the Pliestocene and a late Pliestocene/early Holocene dispersal hypothesis is compatible with available evidence.

In the absence of persistent diploid progenitors in the same region as an allopolyploid species, distinguishing the timing of allopolyploidization relative to dispersal is largely left to conjecture if the putative parents themselves co-occur in a common region. From an events perspective, allopolyploidization prior to dispersal is more parsimonious (Fig. lb), but factors attributed to the success of allopolyploids conceptually suggest the less parsimonious hypothesis should not be discounted. We list several such intrinsic factors here without exposition and refer readers to recent reviews for greater detail (Mummenhoff & Franzke, 2007; Soltis & Burleigh, 2009): 1) buffering effect of fixed heterozygosity; 2) allievement of inbreeding depression; 3) increased biochemical diversity via higher levels of heterozygosity leading to broader ecological tolerances; 4) genetic novelties via genome rearrangements enhancing the possibility for niche separation; and 5) the heightend contribution of all of these factors when the contributing genomes are themselves divergent.

Though successful long distance dispersal between North and South America is rare relative to extant plant diversity, the replicated pattern of modern plant distributions evidence the cumulative effects of such chance events. Fluctuating climatic conditions and reoccurring glacial cycles during the Pleistocene may well have favored the persistence of genetically rich allotetraploids over their diploid progenitors still recovering from post-dispersal genetic bottlenecks, just as polyploidy may have contributed to the success of genome-doubled lineages over diploids during the K-T mass extinction (Fawcett et al., 2009). Given the propensity of allotetraploids to be more successful than their diploid progenitors, their present distribution is more apt to reflect their site of origin. Otherwise, one must account for their formation, likely range expansion prior to dispersal (i.e., all else being equal, a wide-ranging species is more likely to participate in chance dispersal events than a geographically restricted species; Raven, 1963), and then extirpation in the same range where their diploid progenitors often still persist. The discovery in our data of diploid material plausibly relictual and persistent related to at least one of the diploid progenitors of C. biflora evidences both a post-dispersal allopolyploidization hypothesis in this species, as well as the greater success (based on current established range) of the allopolyploid over its diploid progenitors. Although no diploid "smoking gun" has yet been identified for Navarretia involucrata, we favor the hypothesis that it also was formed in South America following dispersal of its diploid progenitors.

While our study is not the first to distinguish the timing of allopolyploidization relative to dispersal, it contributes to a growing body of evidence documenting the post dispersal formation of allopolyploids in diverse lineages, including well studied examples such as Tragapogon (e.g., Ownbey, 1950; Lim et al., 2008; dispersal in the past century) and Lepidium (Mummenhoff et al., 2004; Dierschke et al., 2009; dispersal also during the Pliestocene). Post dispersal allopolyploidization may well be a general pattern in groups where multiple dispersals among close diploid relatives is followed by the establishment of contact zones, hybridization, and in many cases, an eventual decline in, or extirpation of, the diploid progenitors.

DOI 10.1007/s12229-012-9101-8

Appendix I

GenBank accession numbers and voucher information. Acronyms following species names correspond to those used in Figs. 2 through 5. For Collomia, genbank numbers are given in the following order: trnk, matK, trnL, trnS, psbM, ITS, PI, g3pdh. For Navarretia, the same order was followed except for g3pdh, which was not included in this study. Semicolons are used to separate genes and commas are used to separate homeologs within genes for the polyploid species. All vouchers are housed at BRY unless indicated by an alternative herbarium within brackets.

Collomia biflora (Ruiz & Pav.) Brand--ARG1 = Sersic s.n. (DQ196463; DQ196886; DQ196955; DQ196934; HQ117098; DQ196896; HQ116874, HQ116875; HQ117235, HQ117236). ARG2 = [BAA 25056] (HQ911803; HQ911925; HQ911968; HQ911789; HQ911892; HQ911836;--,--; HQ911783, HQ911784). ARG3 = Belgrano et al. 462 (HQ911804; HQ911926; HQ911969; HQ911790; HQ911893; HQ911837; , .; , ). ARG4 = Belgrano et al. 519 (HQ911806; HQ911928; HQ911971; HQ911792; HQ911895; HQ911838; --, --; -, --). ARG5 = Belgrano et al. 532 (HQ911805; HQ911927; HQ911970; HQ911791; HQ911894; HQ911839; HQ911962, HQ911963; --,--). CHL = Thibauat et al. 156 (HQ116975; HQ116935; HQ117058; HQllT020; HQ117099; HQ116834; HQ116876, HQ116877; HQ117237, HQ117238). Collomia grandiflora Douglas ex Lindl.--AZ = Christy 449 [RSA] (HQ911808; HQ911930; HQ911973; HQ911794; HQ911897; HQ911841;--;--). B-ARG = Belgrano 550 (HQ911811; HQ911933; HQ911976; HQ911797; HQ911900; HQ911844; HQ911965; HQ911785). CA = Johnson 94-038 (HQ116983; HQ116942; HQ117065; HQ117027; HQ117107; HQ116840; HQ116886; HQ117247). ID = Johnson & Johnson 95-034 (HQ911809; HQ911931; HQ911974; HQ911795; HQ911898; HQ911842; HQ911966; HQ911786). OR = Johnson 93-086 (DQ196461; DQ196884; DQ196953; DQ196932; HQ117106; DQ196906; HQ116885; HQ117246). NV-1 = Johnson 04-151 (DQ196462; DQ196885; DQ196954; DQ196932; HQ117109; DQ196907; HQ116887; HQ117248). NV-2 = Howell s.n. (HQ911807; HQ911929; HQ911972; HQ911793; HQ911896; HQ911840; HQ911967; HQ911787). UT = Johnson & Matheson 03-073 (HQ911810; HQ911932; HQ911975; HQ911796; HQ911899; HQ911843; HQ911964; HQ911788). Collomia linearis Nutt.--CO-1 = Johnson 99-010 (HQ116990; HQ4116947; HQ117071; HQ117032; HQ117116; HQ116845; HQ116897; HQ117258). CO-2 = Atwood & Higgins 5752 (HQ911812; HQ911934; HQ911977; HQ911798; HQ911901; HQ911845; --; --). ID-1 =Johnson 92-045 [WS] (HQ116988; L34188; AF208170; EU628236; EU628359; AF208200; HQ116894; HQ117255). ID-2 = Johnson 04-168 (DQ196451; DQ196874; DQ196943; DQ196922; HQ117114; DQ196895; HQ116895; HQ117256). NE = Stephens & Brooks 24521 [KANU] (HQ911814; HQ911936; HQ911979; HQ911800; --; HQ911847; --; --). NV = Johnson & Johnson 04-104 (DQ 196449; DQ196872; DQ196941; DQ196920; HQ117113; DQ196893; HQ116893; HQ117254). OR = Johnson 97-134 (HQ911813; HQ911935; HQ911978; HQ911799; HQ911902; HQ911846;--;--). UT = Thorne et al. 4171 (DQ196450; DQ196873; DQ196942; DQ196921; HQ117115; DQ196894; HQ116896; HQ 117257). Collomia maerocalyx Leiberg ex Brand.--OR-1 = Johnson & Johnson 05-071 (HQ116991; HQ116948; HQ117072; HQ117033; HQ117117; HQ116846; HQ116898; HQ117259). OR-2 = Johnson & Johnson 05-079 (HQ116992; HQ116949; HQ117073; HQ117034; HQ117118; HQ116847; HQ116899; HQ117260). Collomia renaeta Joyal--NV = Johnson & Johnson 04-107 (DQ196455; DQ196878; DQ196947; DQ196926; HQ117119; DQ196900; HQ116900; HQ117261). OR = Johnson & Johnson 05-103 (HQ116993; HQ116950; HQ117074; HQ117035; HQ117120; HQ116848; HQ116901; HQ117262). Collomia tenella A. Gray--NV = Johnson 06-120 (HQ116998; HQ116954; HQ117078; HQ117039; HQ117126; HQ116852; HQ116908; HQ117269). UT = Johnson 01-025 (DQ196457; DQ196880; DQ196949; DQ196928; HQ117124; DQ196902; HQ116906; HQ117267). Collomia tinetoria Kellogg--ID = Porter 13769 (DQ196445; DQ196868; DQ196937; DQ196916; HQ117130; DQ196889; HQI16912; HQ117273). OR = Johnson 95-048 (HQll7000; HQ116956; HQ117080; HQ117041; HQ117128; HQ116854; HQ116910; HQ117271). Collomia tracyi H. Mason--CA-1 = Johnson 94-075 (DQ196447; DQ196870; DQ196939; DQ196918; HQ117131; DQ196891; HQ116913; HQ117274). CA-2 = Johnson 94-078 (HQll7001; HQ116957; HQ117081; HQ911802; HQ117132; HQ116855; HQ116914; HQ117275). Collomia wilkenii L.A. Johnson & R.L. Johnson--NV1 = Johnson & Johnson 04-105 (DQ196452; DQ196875; DQ196944; DQ196923; HQ117133; DQ196908, DQ196909; HQ116915, HQ116916; HQ117276, HQ117277). NV2 = Johnson & Zhang 05-166 (HQ117002; HQ116958; HQ117082; HQ117043; HQ117134; HQ116856, HQ116857; HQ116917, HQ116918; HQ117278, HQ117279).

Navarretia fossalis Moran--CA = Spencer 4416-21 [RSA] (HQ911911; HQ912008; HQ911878; HQ911987; HQ911822; HQ911855; HQ911937). Navarretia intertexta (Benth.) Hook.--CA-1 = Johnson 93-088 (HQ911918; HQ912015; HQ911885; HQ911994; HQ911829; HQ911864; HQ911939). CA-2 = Johnson 04038 (HQ911919; HQ912016; HQ911886; HQ911995; HQ911830; HQ911865; HQ911938). Navarretia involuerata Ruiz & Pav.--ARG-1 = Belgrano et al. 470 (HQ911907; EU628547; EU628507; EU628244; EU628367; EU628293; HQ911940, HQ911944). ARG-2 = Belgrano et al. 499 (HQ911908; HQ912005; HQ911875; HQ911984; HQ911819; HQ911852; HQ911941, HQ911943). ARG-3 = Belgrano et al. 551 (HQ911909; HQ912006; HQ911876; HQ911985; HQ911820; HQ911853; HQ911942, HQ911945). Navarretia leucocephala Benth. ssp. bakeri (H. Mason) A.G. Day--CA = Gowen 697(HQ911924; HQ912021; HQ911891; HQ912000; HQ911835; HQ911870; --, --). Navarretia leucocephala Benth ssp. leucocephala--CA = Johnson 04-118 (HQ911921; HQ912018; HQ911888; HQ911997; HQ911832; HQ911867; --, --). Navarretia leucocephala Benth. ssp. minima (Nutt.) A.G. Day--OR = Johnson 05-198 (HQ911922; HQ912019; HQ911889; HQ911998; HQ911833; HQg11868; HQ911946, HQ911947). Navarretia leucocephela ssp. pauciflora (H. Mason) A.G. Day--CA = Johnson 04-036 (HQ911923; HQ912020; HQ911890; HQ911999; HQ911834; HQ911869; HQ911960, HQ911961). Navarretia myersii P.S. Allen & A.G. Day ssp. myersii-CA = Popp s.n. [DAV] (HQ911920; HQ912017; HQ911887; HQ911996; HQ911831; HQ911866; --, --). Navarretia propinqua Suksd.--NV = Howell s.n. (HQ911917; HQ912014; HQ911884; HQ911993; HQ911828; HQ911862, HQ911863; HQ911948, HQ911949). UT = Johnson 04-163(HQ911916; HQ912013; HQ911883; HQ911992; HQ911827; HQ911860, HQ911861; HQ911950, HQ911951). Navarretia prostrata (A. Gray) Greene--CA = Wilken s.n. (HQ117012; HQ116967; HQ117091; HQ117052; HQ117143; HQ116866; HQ116928). Navarretia saximontana S.C. Spencer--UT = Johnson 07-036 (HQ911912; HQ912009; HQ911879; HQ911988; HQ911823; HQ911856; HQ911952, HQ911953). Navarretia sp. nov. 1--CA = Johnson et al. 09-021 (HQ911914; HQ912011; HQ911881; HQ911990; HQ911825; HQ911858; HQ911954). Navarretia sp. nov. 2--CA = Johnson et al. 09-032 (HQ911915; HQ912012; HQ911882; HQ911991; HQ911826; HQ911859; HQ911955). Navarretia sp. nov. 3--UT = Johnson & Johnson 05-197 (HQ911913; HQ912010; HQ911880; HQ911989; HQ911824; HQ911857; HQ911956). Navarretia subuligera Greene--CA-1 = Johnson & Zhang 04-135 (HQ117016; HQ116970; HQ117094; HQ117055; HQ117146; HQ116869; HQ116932). CA-2 = Johnson & Gowen 09-052 (HQ911906; HQ912004; HQ911874; HQ911983; HQ911818; HQ911851;--). CA-3 = Spencer 568-82 [RSA] (HQ911905; HQ912003; HQ911873; HQ911982; HQ911817; HQ911850; --). Navarretia tagetina Greene--CA- 1 = Johnson 04-024 (HQ911903; HQ912001; HQ911871; HQ911980; HQ911815; HQ911848; HQ911957). CA-2 = Johnson 04-046 (HQ117017; HQ116971; HQ117095; HQ117056; HQ117147; HQ116870; HQ116933). CA-3 = Johnson & Zhang 05164 (HQ911904; HQ912002; HQ911872; HQ911981; HQ911816; HQ911849;--). Navarretia willamettensis S.C. Spencer--OR = Johnson & Halse 05-208 (HQ911910; HQ912007; HQ911877; HQ911986; HQ911821; HQ911854; HQg11958, HQ911959).]

Acknowledgements This research was supported by the U.S. National Science Foundation (Grants DEB-0344837, DBI-0520978, and OISE 0530267). The authors thank M. Belgrano for help with field work and specimen information from S. America, and J. M. Porter for discussions about South American Polemonioideae.

Published online: 30 June 2012

Literature Cited

Alsos, I. G., T. Engelskjon, L. Gielly, P. Taberlet & C. Brochmann. 2005. Impact of ice ages on circumpolar molecular diversity: insights from an ecological key species. Molecular Ecology 14: 2739 2753.

Aoki, S. & M. Ito. 2000. Molecular phylogeny of Nicotiana (Solanaceae) based on the nucleotide sequence of the matK gene. Plant Biology 2: 316-324.

Axelrod, D. I. 1950. Evolution of desert vegetation in western North America. Camegie Institution of Washington Publication 590: 215-306.

Bell, C. D. & R. W. Patterson. 2000. Molecular phylogeny and biogeography ofLinanthus (Polemoniaceae). American Journal of Botany 87:1857-1870.

Bjork, C. R. 2002. A new subspecies of Navarretia leucocephala (Polemoniaceae) from vernal pools in Eastern Washington. Madrono 49: 165-168.

Bray, W. L. 1898. On the relation of the flora of lower Sonoran zone in North America to the flora of the arid zones of Chili and Argentine. Botanical Gazette 26:121-147.

-- 1900. The relationships of the Noah American flora to that of South America. Science 12:709-716.

Carlquist, S. 1966. The biota of long distance dispersal. Evolution 20: 30-48.

-- 1983. Intercontinental dispersal. Sonderb. Naturwiss. Vereins Hamburg 7: 37-47. Cruden, R. 1966. Birds as agents of long-distance dispersal for disjunct plant groups of the temperate western hemisphere. Evolution 20:517-532.

Day, A. G. 1993. New taxa and nomenclatural changes in Allophyllum, Gilia, and Navarretia (Polemoniaceae). Novon 3:331-340.

-- 1995. Sessile-flowered species in the Navarretia leucocephala group (Polemoniaceae). Madrono 42: 34-39.

Diersehke, T., T. Mandakova, M. A. Lysak & K. Mummenhoff. 2009. A bicontinental origin of polyploidy Australian/New Zealand Lepidium species (Brassicaceae)? Evidence from genomic in situ hybridization. Annals of Botany 104: 681-588.

Fawcett, J. A., S. Maere & Y. van de Peer. 2009. Plants with double genomes might have had a better chance to survive the Cretaceous-Tertiary extinction event. Proceedings of the National Academy of Sciences, USA 106: 5737-5742.

Flory, W. S. 1937. Chromosome numbers in the Polemoniaceae. Cytologia,

Fujii Jubilee Volume: 171 180.

Goodspeed, T. H. 1954. The Genus Nicotiana. Chronia Botanica, Waltham, Mass.

Grant V. (1959) Natural history of the Phlox family. Martinus Nijhoff, The Hague.

Gray, A. & J. D. Hooker. 1880. The vegetation of the Rocky Mountain region and a comparison with that of other parts of the world. United States Geological and Geographical Survey of the Territories Bulletin 6:1-77.

Green, E. S. 2010. Infrageneric relationships within Collomia (Polemoniaceae). M. S. Thesis, Brigham Young University, Provo, UT.

Hoffmann, A., M. T. K. Arroyo, F. Liberona, M. Munoz & J. Watson. 1998. Plantas altoandinas en la flora silvestre de Chile. Ediciones Fundacion Claudio Gay, Santiago.

Hussey, B. M. J., G. J. Keighery, R. D. Cousens, J. Dodd & S. G. Lloyd. 1997. Western weeds: a guide to the weeds of Western Australia. The Plant Protection Society of Western Australia, Victoria Park.

Johnson, L. A. & R. L. Johnson. 2006. Morphological delimitation and molecular evidence for allopolyploidy in Collomia wilkenii (Polemoniaceae), a new species from northern Nevada. Systematic Botany 31: 349-360.

--,--L. M. Chan, T. L. Weese, L. D. Busby & S. McMurry. 2008. Nuclear and cpDNA sequences combined provide strong inference of higher phylogenetic relationships in the phlox family (Polemoniaceae). Molecular Phylogenetics and Evolution 48:997-1012.

Kay, K. M., J. B. Whittall & S. A. Hodges. 2006. A survey of nuclear ribosomal internal transcribed spacer substitution rates across angiosperms: an approximate molecular clock with life history effects. BMC Evolutionary Biology 6: 36.

Lim, K. Y., D. E. Soltis, P. S. Soltis, J. "Fate, R. Matyasek, H. Srubarova, A. Kovarik, J. C. Pires, Z. Xiong & A. R. Leitch. 2008. Rapid chromosome evolution in recently formed polyploids in Tragopogon (Asteraceae). PLoS ONE 3: e3353, doi:10.1371/journal.pone.0003353.

McNeill, J., F. R. Barrie, H. M. Burdet, V. Demoulin, D. L. Hawksworth, K. Marhold, D. H. Nicolson, J. Prado, P. C. Silva, J. E. Skog, J. H. Wiersema, & N. J. Turland. 2006. International Code of Botanical Nomenclature (Vienna Code) adopted by the Seventeenth International Botanical Congress Vienna, Austria, July 2005. Regnum Vegetabile, 146. A. R. G. Gantner Verlag, Ruggell

Moore, M. J., A. Tye & R. K. Jansen. 2006. Patterns of long-distance dispersal in Tiquilia subg. Tiquilia (Boraginaceae): implications for the origins of amphitropical disjuncts and Galapagos Islands endemics. American Journal of Botany 93:1163-1177.

Morrell, P. L., J. M. Porter & E. A. Friar. 2000. Intercontinental dispersal: the origin of the widespread South American plant species Gilia laciniata (Polemoniaceae) from a rare California and Oregon coastal endemic. Plant Systematics & Evolution 224:13-32.

Muller, K. 2005. SeqState: primer design and sequence statistics for phylogenetic DNA datasets. Applied Bioinformatics 4:65-69.

Mummenhoff, K. & A. Franzke. 2007. Gone with the bird: late Tertiary and Quaternary intercontinental long-distance dispersal and allopolyploidization in plants. Systematics and Biodiversity 5: 255-260.

Mummenhoff, K., P. Linder, N. Friesen, J. L. Bowman, J.-Y. Lee & A. Franzke. 2004. Molecular evidence for bicontinental hybridogenous genomic constitution in Lepidium sensu stricto (Brassicaceae) species from Australia and New Zealand. American Journal of Botany 91:254-261.

Otto, S. P. & J. Whitton. 2000. Polyploid incidence and evolution. Annual Reviews of Genetics 34: 401-437.

Ownbey, M. 1950. Natural hybridisation and amphidiploidy in the genus Tragopogon. American Journal of Botany 37: 487-499.

Philippi, R. A. 1895. Plantas neuvas Chilensas, Polemoniaceae. Anales de la Universidad de Chile 90:211-221.

Porter, J. M. & L. A. Johnson. 2000. A phylogenetic classification of Polemoniaceae. Aliso 19: 55-91.

Posada, D. & K. A. Crandall. 1998. Modeltest: testing the model of DNA substitution. Bioinformatics 14: 817-818.

Puntieri, J. G. & C. A. M. Brion. 2005. Nuevas citas para la flora Argentina: Collomia grandiflora (Polemoniaceae) y Potentilla recta (Rosaceae). Hickenia 3: 227-232.

Pysek, P., J. Sadlo & B. Mandak. 2002. Catalogue of alien plants of the Czech Republic. Preslia, Praha 74: 97-186.

Rambaut, A. 2002. Se-Al sequence alignment editor version 2.0. University of Oxford, Oxford. United Kingdom.

Raven, P. H. 1963. Amphitropical relationships in the floras of North and South America. Quarterly Review of Biology 38:151-177.

-- & D. I. Axelrnd. 1978. Origin and relationships of the California flora. University of California Publications in Botany 72: 1-134.

Reiche, K. 1910. Polemoniaceae. Estudios criticos sobre la flora de Chile 5: 146-158.

Spencer, S. C. & J. M. Porter. 1997. Evolutionary diversification and adaptation to novel environments in Navarretia (Polemoniaceae). Systematic Botany 22: 649-668.

Spencer, S. C. & A. E. Spencer. 2003. Navarretia willamettensis and Navarretia saximontana (Polemoniaceae), new species from ephemeral wetlands of western North America. Madrono 50:196-199.

Soltis, D. E. & J. G. Burleigh. 2009. Suviving the K-T mass extinction: new perspectives of polyploidization in angiosperms. Proceedings of the National Academy of Sciences, USA 106: 5455-5456.

Soltis, P. S. & D. E. Soltis. 2000. The role of genetic and genomic attributes in the success of polyploids. Proceedings of the National Academy of Sciences, USA 97:7051-7057.

-- &.2009. The role of hybridization in Plant Speciation. Annual Review of Plant Biology 60:561-568.

Soriano, A. 1947. Notas sobre plantas de Patagonia. Boletin de Sociedad Argentina de Botfinica 2:99-106.

Stebbins, G. L., Jr. 1942. Polyploid complexes in relation to ecology and the history of floras. American Naturalist 766:-3645.

-- 1947. Evidence on rates of evolution from the distribution of existing and fossil plant species. Ecological Monographs 17: 149-158.

Strand, A. E., J. Leebens-Mack & B. G. Milligan. 1997. Nuclear DNA-based markers for plant evolutionary biology. Molecular Ecology 6: 113-118.

Sugiura, T. 1936. Studies on the chromosome numbers in higher plants, with special reference to cytokinesis 1. Cytologia 7: 544-595.

Swofford, D.L. 2002. PAUP* phylogenetic analysis using parsimony (* and other methods), version 4.0b10. Sinauer Associations, Sunderland. MA.

Thorne, R. F. 1972. Major disjunctions in the geographic ranges of seed plants. The Quarterly Review of Biology 47:365-411.

USDA PLANTS Database. 2010. Accessed 30 September 2010.

Vargas, P., B. G. Baldwin & L. Constance. 1998. Nuclear ribosomal DNA evidence for a western North American origin of Hawaiian and South American species of Sanicula (Apiaceae). Proceedings of the National Academy of Sciences, USA 95: 235-240.

Vijverberg, K., P. Kuperus, J. A. J. Breeuwer & K. Bachmann. 2000. Incipient adaptive radiation of New Zealand and Australian Microseris (Asteraceae): an amplified fragment length polymorphism (AFLP) study. Journal of Evolutionary Biology 13:997-1008.

Weese, T. L. 2004. Phylogenetic analyses in the phlox family (Polemoniaceae). Ph.D. Thesis, Brigham Young University, Provo, UT.

Wen, J. & S. M. Ickert-Bond. 2009. Evolution of the Madrean-Tehyan disjunctions and the North and South American amphitropical disjunctions in plants. Journal of Systematics and Evolution 47:331348.

Wherry, E. T. 1944. Review of the genera Collomia and Gymnosteris. American Midland Naturalist 31: 216-231.

Wilken, D. H. 1978. Vegetative and floral relationships among western North American populations of Collomia linearis Nuttal (Polemoniaceae). American Journal of Botany 65: 896-901.

-- 1982. The balance between chasmogamy and cleistogamy in Collomia grandiflora (Polemoniaceae). American Journal of Botany 69:1326-1333.

-- 1986. Polemoniaceae. Chromosome Number Reports XCIII. Taxon 35:899-900.

--, D. M. Smith, J. B. Harborne & C. W. Glennie. 1982. Flavonoid and anthocyanin patterns and the systematic relationships in Collomia. Biochemical Systematics and Ecology 10: 239-243.

Leigh A. Johnson (1,4) Lauren M. Chan (1,2) Raul Pozner (3) Lisa D. Glazier (1)

(1) Department of Biology and S.L. Welsh Herbarium, Brigham Young University, Provo, UT 84602, USA

(2) Department of Biology, Duke University, Box 90338, Durham, NC 27708, USA

(3) Instituto de Botanica Darwinion, Casilla de Correo 22, B 1642HYD San Isidro, Buenos Aires, Argentina

(4) Author for Correspondence; e-mail:
COPYRIGHT 2012 New York Botanical Garden
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2012 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Johnson, Leigh A.; Chan, Lauren M.; Pozner, Raul; Glazier, Lisa D.
Publication:The Botanical Review
Article Type:Report
Geographic Code:100NA
Date:Sep 1, 2012
Previous Article:Proteaceae leaf fossils: phylogeny, diversity, ecology and austral distributions.
Next Article:Foreword: a Festschrift on the occasion of Dennis Wm. Stevenson's 70th birthday.

Terms of use | Privacy policy | Copyright © 2022 Farlex, Inc. | Feedback | For webmasters |