On the origin and systematic position of the Azorean goldenrod, Solidago azorica (Asteraceae)

Abstract

Goldenrods were first collected in the azores by the German botanist Karl Hochstetter in 1838 and described as an endemic species Solidago azorica. In 1882, asa Gray placed the name into synonymy of the american seaside goldenrod, S. semper-virens. The taxonomic position and status of the plants in the azores remained unclear ever since but recent human-mediated introduction from the american coast seemed to be the most likely explanation. Here, I analyze molecular and morphological data and the historical record to test this hypothesis. While morphological differences are not clear and an overall similarity to some specimens from New foundland is striking, I find that all analyzed Solidago plants from the azores archipelago differ in their nuclear ITS and eTS sequences plus a number of microsatellite markers from american goldenrods. furthermore , the historical record suggests existence of goldenrods in the azores at the time of the arrival of the first settlers and well before Columbus' first journey. Moreover, large populations were reported from several islands in the 16th century. I conclude that the azorean plants are native to the azores and represent a distinct endemic species sharing a common ancestor with S. sempervirens. The azorean plants represent a geographically isolated, genetically distinct population that is most likely the result of a natural colonization event from the North american coast perhaps via vagrant birds. I reinstate the name S. azorica and describe the morphological differences between S. azorica and S. sempervirens.

Full text

Phytotaxa 210 (1): 047–059
www.mapress.com/phytotaxa/
Copyright © 2015 Magnolia Press Article PHYTOTAXA
ISSN 1179-3155 (print edition)
ISSN 1179-3163 (online edition)
Accepted by Mark Carine: 4 Feb. 2015; published: 29 May 2015
http://dx.doi.org/10.11646/phytotaxa.210.1.5
47
On the origin and systematic position of the Azorean goldenrod, Solidago azorica
(Asteraceae)
HANNO SCHAEFER
Plant Biodiversity Research, Technische Universität München, Emil-Ramann-Str. 2, 85354 Freising, Germany, hanno.schaefer@tum.de
Abstract
Goldenrods were first collected in the Azores by the German botanist Karl Hochstetter in 1838 and described as an endemic
species Solidago azorica. In 1882, Asa Gray placed the name into synonymy of the American seaside goldenrod, S. semper-
virens. The taxonomic position and status of the plants in the Azores remained unclear ever since but recent human-mediated
introduction from the American coast seemed to be the most likely explanation. Here, I analyze molecular and morphologi-
cal data and the historical record to test this hypothesis. While morphological differences are not clear and an overall similar-
ity to some specimens from New Foundland is striking, I find that all analyzed Solidago plants from the Azores archipelago
differ in their nuclear ITS and ETS sequences plus a number of microsatellite markers from American goldenrods. Further-
more, the historical record suggests existence of goldenrods in the Azores at the time of the arrival of the first settlers and
well before Columbus’ first journey. Moreover, large populations were reported from several islands in the 16th century. I
conclude that the Azorean plants are native to the Azores and represent a distinct endemic species sharing a common ances-
tor with S. sempervirens. The Azorean plants represent a geographically isolated, genetically distinct population that is most
likely the result of a natural colonization event from the North American coast perhaps via vagrant birds. I reinstate the name
S. azorica and describe the morphological differences between S. azorica and S. sempervirens.
Key words: cubres, ETS, Gaspar Frutuoso, ITS, trnQ-rps16, seaside goldenrod
Introduction
The genus Solidago Linnaeus (1753: 878) (Asteraceae) comprises c. 84 species of perennial herbs: 77 species in
North America (including Mexico) (Semple and Cook 2006), three to four in South America (Lopez Laphitz 2009),
one species native to Europe and Northern Africa (Tutin et al. 1976), and three native to China (Yilin and Semple
2011). In general, goldenrods are well characterized and easily recognizable by their bright yellow inflorescences,
perennial habit, and clonal reproduction. Species circumscriptions in the genus, however, are less clear-cut and based
mainly on morphological characters with a large number of intraspecific taxa and potential hybrids. All species have
a base chromosome number of x=9 and often include several ploidy levels (Semple et al. 1984, Peirson et al. 2013).
Comprehensive and well-sampled molecular studies of the genus are so far lacking but the recent study of Laureto
and Barkman (2011) shows that such approaches have great potential to help establish a stable classification of this
problematic genus.
Solidago sempervirens Linnaeus (1753: 878), Fig. 1, is a herbaceous perennial native to sand dunes and marshes
along the North American Atlantic Coast from New Foundland in the North probably south to Virginia (J. C.
Semple, pers. communication 2011). The more southerly populations from Virginia to Mexico and the Caribbean are
morphologically distinct and have been described as S. mexicana Linnaeus (1753: 879). They are currently classified as
S. sempervirens subsp. mexicana (L.) Semple (2003: 1615). Much of the herbarium material labelled S. sempervirens
or S. mexicana might also represent a different taxon, S. stricta Aiton (1789: 216) (syn. S. virgata Michaux (1803:
117)), or hybrids between S. sempervirens and S. stricta or S. rugosa Miller (1768: no. 25) (Semple and Cook 2006).
Solidago sempervirens is relatively tolerant to soil salinity and airborne salt spray and even though it does not
seem to depend on salt (not a halophyte in the strict sense), it seems to be more competitive under increased salt
conditions (Brauer and Geber 2002). Current deicing salt use practices therefore seem to favor its spread on roadsides
and railroad tracks from the coasts further inland. Populations of S. sempervirens are now found throughout the Great

SCHAEFER
48 • Phytotaxa 210 (1) © 2015 Magnolia Press
Lakes region of Canada and the Northeastern United States (Brauer and Geber 2002 and references listed therein).
Solidago sempervirens is mostly self-incompatible but selfing might result in a very small number of seeds (Innes and
Hermanutz 1988). Its pollen is relatively large and sticky, and wind transfer probably not very efficient. Honeybees
(Apis mellifera) are thus thought to be the most important pollinators today (Innes and Hermanutz 1988) but since they
are a relatively recent introduction to the American continent, native insects like bumblebees (Bombus spec.) must
have been the main pollinators over most of the lineages’ evolutionary history. The seeds are short-lived and wind-
dispersed but dispersal ability seems to be low with most seeds falling down within a radius of 10 metres from their
mother plants (Lee 1993).
FIGURE 1. Solidago sempervirens L. subsp. sempervirens, flowering stem at Duxbury beach, Massachusetts, USA, October 2011
(photographer H. Schaefer).
Two species of Solidago have been reported from the Azores, an archipelago of nine islands in the Northern
Atlantic about 3400 km east of the North American coast (see Schaefer 2003 for an overview of geography, geology,
flora and vegetation). The first, Solidago gigantea Aiton (1789: 211) subsp. serotina (Kuntze 1891: 314) McNeill
(1973: 280), is a recently introduced species first seen less than 30 years ago and in the past decade has become
invasive on some islands (Franco 1984, Schaefer 2003). I will not discuss the status of this species further since its
American origin and recent introduction to the Azores are generally accepted. The second species is widespread and
its taxonomic position and native status have been controversial. It was discovered in 1838 by one of the first botanists
visiting the Azores, the German Karl Hochstetter, and was later described as Solidago azorica Hochstetter ex Seubert
(Seubert 1844: 31 & t10;; Fig. 2, Fig. 3A–C). Some 40 years later, however, the prominent Harvard botanist Asa Gray
placed it in synonymy of S. sempervirens (Gray 1882: 192), unfortunately without a detailed discussion of its status
or comparison of morphological characters. Twenty years later, Harold St. John raised it again to variety level as S.
sempervirens var. azorica (Hochst. ex Seub.) St.John, (1915: 27) based on differences in cauline leaf morphology
(St. John 1915). Finally, in the most recent taxonomic treatment, John C. Semple raised it to subspecies level as
S. sempervirens subsp. azorica (Hochst. ex Seub.) Semple (2003:1615), mainly based on the isolated range of the
Azorean population (Semple 2003).
No native Solidago is known from any of the other middle-Atlantic islands and the genus is indeed absent from
most of the European Atlantic coast. The only Macaronesian record outside the Azores is from Madeira, where the
South American S. chilensis Meyen (1834: 311) has recently been discovered in two localities and might become
invasive (Silva et al. 2009). The absence of S. sempervirens from neighboring archipelagos and the extremely small
number of natural colonisers from the American continent in the Azores (Schaefer 2003), together with the over-
abundance of exotic invaders in those islands (Schaefer 2003, Schaefer et al. 2011c) made most modern botanists
believe that Hochstetter’s Solidago species is likely a recent human-mediated introduction to the Azores from the
American coast and arrived perhaps via whaling ships or fishing boats (Schaefer 2003, 2005; Semple and Cook 2006).
In contrast, earlier Portuguese authors like Palhinha (1966) had classified the species as “native”. The doubts about
the status of the Azorean goldenrod remained and consequently, a classification as “doubtful” was chosen in the most

AZOREAN GOLDENRODS Phytotaxa 210 (1) © 2015 Magnolia Press • 49
recent checklist for the archipelago (Silva et al. 2010) until more evidence would allow an informed decision to be
made.
The goal of this study is therefore to collect molecular sequence data and review all available morphological,
historical, and biogeographical data for the Azorean Solidago and their closest American relatives to establish if the
Azorean plants represent an introduced or native population and whether or not they are genetically distinct from the
American seaside goldenrods and establish the appropriate taxonomic status for Azorean plants.
FIGURE 2. Solidago azorica Hochst., copper engraving from Flora Azorica (Seubert 1844) based on the holotype Hochstetter 107
(TUB).

SCHAEFER
50 • Phytotaxa 210 (1) © 2015 Magnolia Press
Methods & Material
Morphology
I studied Azorean Solidago populations in their natural habitat on all nine islands during multiple visits between
1998–2013 and S. sempervirens in native coastal habitats in Massachusetts (USA) in 2010/2011 as well as plants from
an inland roadside population brought into cultivation in the gardens at the Harvard University Herbaria by Douglas
Goldman. I also studied the morphology of herbarium material of Azorean and American Solidago from AZU, AZB,
and GH.
FIGURE 3. Solidago azorica Hochst., A—large coastal population on São Jorge island, Fajã Rasa; B—flowering inflorescence, Corvo
island, June 2011; C—details of capitulae, Corvo island, with pollinating syrphid fly, June 2011 (photographers A: L. Silveira, B/C: H.
Schaefer).

AZOREAN GOLDENRODS Phytotaxa 210 (1) © 2015 Magnolia Press • 51
Sampling for Molecular Analysis and DNA Extraction
In the molecular analyses, I compared Solidago material from five different Azorean islands (Corvo, Flores, São Jorge,
Pico, and Terceira) with S. sempervirens subsp. sempervirens from Massachusetts and Michigan (USA) and Magdalen
Islands (Canada), S. sempervirens subsp. mexicana from Florida (USA), and S. uliginosa from Massachusetts (USA). In
total, I analysed 12 specimens (Tab. 1) and used silica-dried leaves to extract DNA with a NucleoSpin Plant extraction
kit (Macherey Nagel, Germany) following the manufacturer’s protocol. Then, I sequenced the nuclear ITS1 and ITS2
spacers plus intervening 5.8S gene, the 3′ end of the ETS spacer, and the chloroplast trnQ-rps16 and trnH-psbA spacers
using the primers and PCR protocols described in Laureto and Barkman (2011). Thirty-one sequences were generated
for this study and deposited in Genbank (accession numbers KP153071–KP153099). Furthermore, I downloaded
all available Solidago sequences with well-documented geographic origin from Genbank (mainly originating from
the studies by Laureto and Barkman 2011 and Urbatsch et al. 2003) and combined them with the newly generated
sequences.
TABLE 1: Material for genetic analyses and GenBank accession numbers for the different loci (“—” -amplification not successful).
Taxon Origin voucher DNA No. ETS ITS trnH-psbA trnQ-rps16
S. azorica Azores, São Jorge unvouchered HS 818 KP
153090
— KP
153071
—
S. azorica Azores, Flores H. Schaefer 2011/176 (GH) HS 922 — — KP
153072
—
S. azorica Azores, Flores H. Schaefer 2011/177 (GH) HS 923 KP
153091
KP
153081
KP
153073
—
S. azorica Azores, Corvo H. Schaefer 2011/164 (GH) HS 924 KP
153092
KP
153082
KP
153074
KP 153097
S. azorica Azores, Corvo H. Schaefer 2014/178 (TUM) SYS 376 — KP
153085
— —
S. azorica Azores, Pico H. Schaefer 2011/352 (GH) HS 925 KP
153093
KP
153083
KP
153075
KP 153098
S. azorica Azores, São Jorge H. Schaefer 2011/423 (GH) HS 993 KP
153094
KP
153084
KP
153076
KP 153099
S. gigantea Azores, Terceira H. Schaefer 2014/223 (TUM) SYS 402 — KP
153086
— —
S. gigantea Azores, Terceira H. Schaefer 2010/478 (GH) HS795 — — KP
153077
—
S. uliginosa USA, Massachu-
setts
H. Schaefer 2011/316 (GH) HS 1205 — KP
153089
KP
153080
—
S. sempervirens USA, Massachu-
setts
H. Schaefer 2011/534 (TUM) HS 792 — KP
153088
KP
153079
—
S. sempervirens USA, Massachu-
setts
H. Schaefer 2011/315 (GH) HS 1204 KP
153095
KP
153087
KP
153078
—
S. sempervirens Canada, Magdalen
Islands
M. Fernald et al. 8108 (GH) HS 1256 KP
153096
———
To test for variation within the Azores archipelago, I additionally used primers for four microsatellite regions
(SS4F, SS4G, SS19C, and SS20E) developed for S. sempervirens population studies by Wieczorek and Geber (2002).
I amplified and sequenced those regions successfully for up to six samples of Azorean Solidago from four different
islands and three samples of North American S. sempervirens and compared the length of their hypervariable regions
with those of published sequences for an inland S. sempervirens population in Watkins Glen, New York, and one
sample of Solidago gigantea subsp. serotina from Terceira, Azores. Whilst this sampling is not sufficient to reveal
population-level differentiation within the Azores, it should allow potential differences between North American and
Azores plants to be detected.
Sequence Alignment and Phylogenetic Analyzes
Sequences were edited using Sequencher 4.9 (GeneCodes Corp.), and aligned by eye in MacClade 4.08 (Maddison and

SCHAEFER
52 • Phytotaxa 210 (1) © 2015 Magnolia Press
Maddison 2005). Maximum likelihood (ML; Felsenstein 1973) tree searches and ML bootstrap searches (Felsenstein
1985) for the individual and combined data sets were performed using RAxML-HPC2 vs. 7.2.6 (Stamatakis et al.
2008) on the CIPRES cluster (Miller et al. 2009). Based on the Akaike Information Criterion (Akaike 1974), the GTR
+ Γ model (six general time-reversible substitution rates, assuming gamma rate heterogeneity) was selected, with
model parameters estimated over the duration of specified runs.
Results
Morphology
The overall morphology of S. sempervirens is extremely variable and some specimens, especially those of S. sempervirens
subsp. sempervirens from New Foundland and other parts of Northeastern Canada are almost indistinguishable from
the Azorean Solidago. For most specimens, however, the leaf characters already suggested by St.John (1915) are useful
to distinguish the Azorean plants from the American specimens: cauline leaves in Azorean Solidago are sessile, ovate
or deltoid-lanceolate, broadest just above the base (Fig. 2), and tapering gradually into the blunt, attenuate tip, whereas
cauline leaves in American S. sempervirens are usually linear to broadly lanceolate (Fig. 1), widest near the middle and
tapering equally to either end (St. John, 1915). Further differences in leaf morphology have been reported by Anderson
and Creech (1975): secretory cavities absent in leaves of Azorean plants but present (adaxial to the veins) in American
S. sempervirens; +/- bifacial mesophyll in Azorean plants and isolateral mesophyll in American S. sempervirens; sheath
extensions less abundant in Azorean plants; bundle sheath fibres of the midvein only adaxial in Azorean plants, but
adaxial and abaxial in American S. sempervirens; and finally storage parenchyma absent in Azorean plants versus
infrequent/rare in American S. sempervirens.
Solidago sempervirens subsp. mexicana is a much more delicate plant than the Azorean goldenrod and differs not
only in leaf shape but also in lower numbers of disk and ray florets and in its overall less succulent habitus.
Molecular data
The aligned ETS matrix comprises 525 nucleotides of 23 ingroup species (35 accessions) and two outgroups (Eastwoodia
elegans Brandegee (1894: 397) and Tonestus graniticus (Tiehm & Shultz (1985: 165)) Nesom & Morgan (1990: 178))
based on Roberts and Urbatsch 2003). In general, the sequenced part of the ETS region is not very variable in Solidago.
All Azorean Solidago samples share a unique substitution at alignment position 226 (Cytosine, whereas all remaining
sequenced Solidago species have a Thymine at that position). At position 286, the Azorean samples share a substitution
with all American/Canadian S. sempervirens (Thymine for Cytosine).
For the ITS1–5.8S-ITS2 region, the matrix comprises 609 aligned nucleotides of 26 ingroup species (42 accessions)
and the four outgroups Columbiadoria hallii, Eastwoodia elegans, Stenotus pulvinatus, and Tonestus graniticus (based
on Roberts and Urbatsch 2003). Azorean Solidago sequences share a substitution at alignment position 436 (Thymine
instead of Cytosine), which is not found in American/Canadian S. sempervirens but only in three more distantly related
American species (S. fistulosa, S. rugosa, and S. speciosa). All Azorean accessions share a substitution with American/
Canadian S. sempervirens at position 115 (Cytosine instead of Thymine) and at position 574 (Adenine instead of
Guanine). At position 489, all American/Canadian S. sempervirens share a unique substitution (Adenine for Guanine),
which is not found in any of the Azorean samples or in any other sequenced Solidago species.
The combined nuclear ribosomal alignment comprises 1134 aligned nucleotides for 27 ingroup species (38
accessions) and two outgroups. It has a total of 8% gaps or missing data. In the best ML tree (Fig. 4A), the Azorean
accessions (S. azorica) form a highly supported clade (bootstrap support (BS) 97%) and are sister to a moderately
supported S. sempervirens clade (BS 81%), which includes both S. sempervirens subsp. sempervirens and S.
sempervirens subsp. mexicana.
Regarding the chloroplast genome sequences, the trnQ-rps16 matrix comprises 1064 aligned nucleotides for 23
ingroup species (27 accessions) and the two outgroups Ericameria nauseosa and Euthamia graminifolia (based on
Laureto and Barkman 2011). The sequences for the Azorean Solidago, American S. sempervirens, S. riddellii, and S.
rugosa are 100% identical. The sequence matrix of the second chloroplast region, the trnH-psbA spacer, comprises
344 aligned nucleotides for 29 ingroup species and again the two outgroups Ericameria nauseosa and Euthamia

AZOREAN GOLDENRODS Phytotaxa 210 (1) © 2015 Magnolia Press • 53
graminifolia (40 accessions in total). The six S. azorica sequences are all identical and differ from two of the three S.
sempervirens sequences in a duplication at position 150. However, in the third S. sempervirens sequence (HS792), this
duplication is also lacking. The combined plastid matrix contains 1410 aligned nucleotides (33% gaps or missing data).
The best ML tree for the plastid regions is mainly unresolved (Fig. 4B).
For the combined nuclear plus plastid analysis, I removed duplicate accessions with high amounts of missing data
to keep the overall percentage of missing data low. The resulting alignment comprises 2485 aligned nucleotides (19%
gaps or missing data) for 24 ingroup species (31 accessions) and two outgroups. In the best ML tree for this matrix
(Fig. 5), S. sempervirens and S. azorica are each recovered as monophyletic groups (81% and 94% BS) and as sister
groups (92% BS).
The results for the microsatellite regions (Tab. 2) do not reveal any island-specific variation within the
Azores archipelago. For region SS4F with a core motif of (CTT)7 in S. sempervirens from New York (Wieczorek
and Geber 2002), I find (CTT)8 in all five sequenced Azorean samples, while the three S. sempervirens samples
from USA and Canada all have six repeats of (CTT). Region SS4G has a core motif of (CT)10 in S. sempervirens
from New York (Wieczorek and Geber 2002) and also in the three samples from Massachusetts and Canada, while
I find in all six analyzed Azorean samples nine repeats of (CT). For the microsatellite SS19C with a core motif of
(GAT)11(GAC)(GAT)4 in S. sempervirens from the New York population analyzed by Wieczorek and Geber (2002),
I find a motif of (GAT)4(GAC)(GAT)3 in all six analyzed Azorean samples and in a sample of S. sempervirens from
Massachusetts, while the second Massachusetts sample has (GAT)5(GAC)(GAT)3. Finally, for microsatellite region
SS20E with a core motif of (TA)4(TG)12 in the New York S. sempervirens population (Wieczorek and Geber 2002) I
find once (TA)3(TG)13 and once (TA)4(TG)7 in S. sempervirens from Massachusetts, while the two analyzed Azorean
samples (one from Flores, one from Pico) both have (TA)3(TG)10.
TABLE 2: Results of microsatellite analysis.
Taxon Population SS4F SS4G SS19C SS20E
Solidago
sempervirens ssp.
sempervirens
New York (Wieczorek
and Geber 2002)
(CTT)7(CT)10 (GAT)11(GAC)(GAT)4(TA)4(TG)12
Solidago
sempervirens ssp.
sempervirens
Massachusetts inland
(HS792)
(CTT)6(CT)10 (GAT)5(GAC)(GAT)3(TA)4(TG)7
Solidago
sempervirens ssp.
sempervirens
Massachusetts coast
(Duxbury bay-HS1204,
HS1205)
(CTT)6(CT)10 (GAT)4(GAC)(GAT)3(TA)3(TG)13
Solidago
sempervirens ssp.
sempervirens
Magdalen islands,
Canada
(HS1256)
(CTT)6(CT)10 — —
Solidago azorica Azores (Flores -2 ind.,
Corvo-HS924, Pico-
HS925, São Jorge-2 ind.)
(CTT)8(CT)9(GAT)4(GAC)(GAT)3(TA)3(TG)10
Solidago gigantea
ssp. serotina
Azores (Terceira, HS795) (CTT)5(CT)13 (GAT)6(TA)3(TG)7
Historical evidence
The earliest historical evidence for Solidago in the Azores I could find is in a 16th century description of the islands
by the Portuguese priest and chronicler Gaspar Frutuoso (Frutuoso 1998 [written 1565–1591]). He mentions “cubres”
(the Azorean name for goldenrods) as common plants on the Azorean islands Terceira, São Jorge and especially on
Flores Island, where he describes them as very common on fallow land. He also mentions a vast flat coastal area called
“Fajã dos Cubres”, São Jorge, dominated by Solidago (Fig. 3A) and furthermore states that the name “Ilha das Flores”
(Portuguese for ‘island of the flowers’) refers to the bright yellow bloom of goldenrods seen by the first colonizers
arriving in the island.

SCHAEFER
54 • Phytotaxa 210 (1) © 2015 Magnolia Press
FIGURE 4. Best maximum likelihood phylogenies, A—based on the combined nuclear ribosomal ETS and ITS regions (1134 basepairs);
B—based on the plastid trnQ-rps16 and trnH-psbA regions (1410 basepairs). Likelihood bootstrap values >60 shown at the nodes. Solidago
azorica highlighted in red, S. sempervirens in green; GB-sequence downloaded from GenBank.

AZOREAN GOLDENRODS Phytotaxa 210 (1) © 2015 Magnolia Press • 55
FIGURE 5. Best maximum likelihood phylogeny based on the combined nuclear and plastid data (2198 basepairs). Likelihood bootstrap
values >60 shown at the nodes. Solidago azorica highlighted in red, S. sempervirens in green; GB-sequence downloaded from GenBank.
Discussion
While the morphological data cannot help in this case to decide if variety, subspecies, or species-level is adequate for
the Azorean Solidago population, the described molecular differences are relatively large compared to other groups
of closely related Solidago species in the tree (Fig. 5). This genetic difference detected between North American
populations and the Azorean Solidago population is unlikely to have evolved within less than five centuries following
the earliest possibility for human-mediated introduction of Solidago seeds from the American coast: Christopher
Columbus’ stop in the Azores on his return from the first journey to the New World. Columbus arrived in the Azores in
1493 (February 18–24) and stayed at the easternmost island Santa Maria some 600 km away from Flores (Columbus
1893). Flores island had been discovered by 1452 or possibly a few years earlier (Verlinden 1986), about 40 years
before Columbus’ journey, so if Flores island was named after its large coastal Solidago populations, they must have
existed before the first contact with the New world. Even though it seems that the island was first called “São Tomás”,
“Santa Iria”, and “Corvo”, the name “Ilha das Flores” already appears in documents from 1475, which would still have
been well before the first possible introduction of Solidago seeds from the West Indies or further North (Verlinden
1986). Leaving the controversial naming issues aside, one would still have to explain, how a species that could not
have arrived before 1493, could reach the large population size and wide distribution across the entire archipelago,
reported by Frutuoso less than 100 years later. Other American species of the same genus like S. gigantea and S.
canadensis are known to be very successful colonizers outside their native range but it seems likely that a chronicler

SCHAEFER
56 • Phytotaxa 210 (1) © 2015 Magnolia Press
like Frutuoso would have reported such a rapid invasion of goldenrods that would have happened mainly during his
life time (c. 1522–1591).
Additional evidence might come from palynological studies: Connor et al. (2012) recently reported small amounts
of Solidago-type pollen from 2.500 years old lake sediments on Flores. It is important to point out that this pollen type
cannot be assigned with certainty to a particular genus and according to Connor et al. (2012) could represent the genera
Pericallis, Senecio, Solidago or related species. Given the relatively large size of Solidago pollen, it seems plausible,
however, that they may represent traces from an ancient native Solidago population. Azorean Solidago is today mostly
restricted to coastal areas below 500 m (Schaefer 2003) and can be found at higher altitudes only on the most humid
islands of Flores and Corvo. This might explain why its pollen appears in the Flores sediments but is absent from the
high altitude sediment cores from Pico Island.
If one accepts that the Azorean goldenrod has arrived in the archipelago long before the first human settlers,
the inability to detect genetic differences between the different Azorean islands is somewhat surprising, especially
compared to the variation detected between the different samples of S. sempervirens from Massachusetts. The large
and sticky pollen is unlikely to be dispersed by wind or insects across the c. 250 km sea barrier between the western
and central group of the Azores. Dispersal of S. azorica seeds between the islands also seems unlikely given the
generally low dispersal ability reported for S. sempervirens (Lee 1993). However, the current data supports a scenario
of continued gene flow between the different islands, a pattern thought to be common in the Azores (Carine and
Schaefer 2010) even though it has been questioned for most of the endemic lineages (Schaefer et al. 2011b). All that
said, a much denser sampling of Solidago populations within the archipelago and especially inclusion of material from
the eastern group (Santa Maria and São Miguel) is necessary to obtain robust data on genetic structure within and
between the islands of the Azores archipelago and to test this hypothesis further.
Conclusion
Introduction of the Azorean goldenrod through human-mediated transport seems highly unlikely because of the
considerable genetic differences detected between North American populations and the Azorean Solidago populations,
which are unlikely to evolve within just five centuries following the first possible introduction through ships arriving
from the Americas. Furthermore, the small available time window between Columbus’ stop in the Azores and the
reportedly large population of goldenrods in the archipelago less than a century later as well as the presence of
Solidago-type pollen in 2.500 years old lake sediments all lead us to the conclusion that Solidago has probably been
present in the Azores for thousands of years. The first seeds might have arrived in the Azores attached to coastal birds,
which were blown away from the Northeastern American coast by storms. In the absence of any evidence for continued
gene flow from or to the American coastal populations, it seems most appropriate to treat the plants as an endemic
species. Solidago azorica should be included in conservation programs and to ensure its survival on all islands habitat
management plans should be developed in the eastern group of the archipelago.
This study, together with the recent study of the Azorean Marsilea population (Schaefer et al. 2011a), highlights
the huge potential of molecular sequence data to address questions about origin and taxonomic status of Macaronesian
species especially in morphologically challenging groups. More studies of this kind are needed to confirm or reject
native, endemic or introduced status for all plant species in the Azores and elsewhere in Macaronesia in order to ensure
that conservation efforts and funds are directed to indigenous and endemic species only.
Emended description of Solidago azorica
Solidago azorica Hochstetter ex Seubert (Seubert 1844: 31 & t10)
Type:—PORTUGAL. Azores: Hochstetter 107 (holotype TUB!).
Plants: up to 150 cm. Stems: 1–10, erect, glabrous, with woody base. Leaves: basal rosette usually no longer present
at flower, alternate, simple, entire, +/- fleshy and amplexicaule, ovate-lanceolate, up to 25 × 5 cm, acute, glabrous,
margin entire. Flowering capitulae: up to 400, small, in dense panicles, broadly club-shaped. Peduncles: 2–3 mm,
glabrous or sparsely hairy. Involucres: 4–7 mm. Phyllaries: in 3–4 series, unequal, lanceolate, margins ciliate, apices

AZOREAN GOLDENRODS Phytotaxa 210 (1) © 2015 Magnolia Press • 57
acute. Ray florets 5–12; laminae 5–6.2 × 0.4–0.6 mm. Disc florets 7–15; corollas 3–3.2 mm, lobes 0.5–1.2 mm.
Cypselae (obconic) 1.5–3 mm, moderately strigose; pappi 3.8–4 mm. Perennial.
Distribution. The Azorean islands Corvo (widespread), Flores (widespread), Faial (widespread), Pico (scattered
along the coast), Terceira (scattered along the coast), São Jorge (locally common), Graciosa (widespread), São Miguel
(rare, two locations on the North coast), and Santa Maria (rare, two locations on the East coast).
Habitat. Locally common in coastal cliffs and on lava flows up to 500 m in the eastern and central group of the
archipelago, up to 900 m in the western group (Schaefer 2003).
Phenology. Flowering time June to August.
Conservation Status. Not endangered. The species is widespread and common in the western group and on some
of the central group islands. It is, however, restricted to very few locations on the islands of the eastern group and
potentially endangered there.
Representative specimens examined
S. azorica Hochst.:
PORTUGAL . Azores: Graciosa, C.S. Brown 132 (GH!); Flores, H. Schaefer 2011/177 (GH); Corvo, H. Schaefer
2011/219 (GH!); Pico, H. Schaefer 2011/352 (GH!); São Jorge, H. Schaefer 2011/423 (GH!).
S. sempervirens L.:
LOCALITY NOT KNOWN: “Habitat in Mexico?”, s.coll. s.n. (LINN-HL998-13 photo!) [lectotype of S. mexicana
L. designated by Taylor and Taylor, 1984]. “Habitat in Noveboracao, Canada”, s.coll. s.n, (LINN-HL998-1 photo!)
[lectotype of S. sempervirens L. designated by Taylor and Taylor, 1984].
CANADA. Nova Scotia: Victoria County, Cape Breton, E. Scamman 4455 (GH!). Yarmouth County, M.L. Fernald
& B. Long 24572 (GH!). Sable Island, H. St.John 1330, 1331, 1332, 1333, 1334 (GH!). Québec: Comté de Rimouski,
BIC, J. Rousseau 26834 (GH!). Ile à deux têtes, J. Rousseau 21406 (GH!). Magdalen Islands, M. Fernald et al.
8107 & 8108 (GH!). Kent County, New Brunswick, Kouchibouguac National Park, D. Munro 2120 (GH!). Queens
County, Prince Edward Island, M. Fernald et al. 8109 (GH!). New Foundland: Bay of Islands, M. L. Fernald et al. 443
(GH!).
U.S.A. Delaware: Kent County, ‘Kitts hummock’, E.L. Larsen 341 (GH!). New Jersey: Ocean County, Island
Beach, R.T. Clausen s.n. (GH!). Michigan: Wayne County, A.A. Reznicek 4989 (GH!). Florida: Sanibel Island, S.M.
Tracy 7248 (GH!). Texas: Port Arthur, B.C. Tharp 42–72 (GH!).
MEXICO. Tabasco: Rio Grijalva, F.D. Barlow 19/1 (GH). Veracruz: mouth of Rio de la Antigua, C.A. Purpus
6295 (GH!).
Acknowledgements
I thank M. Moura and L. Silva for hospitality during visits to AZB, and C. Dilger-Endrulat for hospitality at TUB; J.C.
Semple, S. Cappellari, S. Connor, and D. Goldman for discussion, and the Azorean Direcção Regional do Ambiente
for collection permits.
References
Aiton, W. (1789) Hortus kewensis, or, a catalogue of the plants cultivated in the Royal Botanic Garden at Kew by William Aiton Gardener
to his Majesty. Vol. I. George Nicol, London, 496 pp.
Akaike, H. (1974) A new look at the statistical model identification. IEEE Transactions on Automatic Control 19: 716–723.
http://dx.doi.org/10.1109/TAC.1974.1100705
Anderson, L.C. & Creech, J.B. (1975) Comparative leaf anatomy of Solidago and related Asteraceae. American Journal of Botany 62:
486–493.
http://dx.doi.org/10.2307/2441956
Brandegee, T.S. (1894) Two undescribed plants from the coast range. Zoë 4: 397–398.

SCHAEFER
58 • Phytotaxa 210 (1) © 2015 Magnolia Press
Brauer, J. & Geber, M.A. (2002) Population differentiation in the range expansion of a native maritime plant, Solidago sempervirens L.
International Journal of Plant Sciences 163: 141–150.
http://dx.doi.org/10.1086/324047
Carine, M.A. & Schaefer, H. (2010) The Azorean diversity enigma: why are there so few Azorean endemic flowering plants and why are
they so widespread? Journal of Biogeography 37: 77–89.
Columbus, C. (1893) The journal of Christopher Columbus during his first voyage (1492–93) and documents relating to the voyages of
John Cabot and Gaspar Corte Real. Translated by C.R. Markham. Hakluyt Society, London, 259 pp.
Connor, S., van Leeuwen, J.F.N., Rittenour, T.M., van der Knaap, W.O., Ammann, B. & Björck, S. (2012) The ecological impact of oceanic
island colonisation-a palaeoecological perspective from the Azores. Journal of Biogeography 39: 1007–1023.
http://dx.doi.org/10.1111/j.1365-2699.2011.02671.x
Felsenstein, J. (1973) Maximum likelihood and minimum-steps methods for estimating evolutionary trees from data on discrete characters.
Systematic Zoology 22: 240–249.
http://dx.doi.org/10.2307/2412304
Felsenstein, J. (1985) Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39: 783–791.
http://dx.doi.org/10.2307/2408678
Franco, J.A. (1984) Nova Flora de Portugal (Continente e Açores). Volume II Clethraceae-Compositae. Sociedade Astória Lda., Lisboa,
659 pp.
Frutuoso, G. (1998) Saudades da Terra, Vol. 6. Instituto Cultural de Ponta Delgada, Ponta Delgada, Portugal, 165 pp. Available from:
http://www.azoreangenealogy.com/livro6.htm (accessed 18 November 2014)
Gray, A. (1882) Contributions to North American Botany: Studies of Aster and Solidago in the older herbaria. Proceedings of the American
Academy of Arts and Sciences 17: 163–230.
http://dx.doi.org/10.2307/25138649
Kuntze, O. (1891) Revisio Generum Plantarum, Pars 1. Arthur Felix, Leipzig, 374 pp.
Laureto, P.J. & Barkman, T.J. (2011) Nuclear and chloroplast DNA suggest a complex single origin for the threatened allopolyploid
Solidago houghtonii (Asteraceae) involving reticulate evolution and introgression. Systematic Botany 36: 209–226.
http://dx.doi.org/10.1600/036364411X553289
Lee, P.C. (1993) The effect of seed dispersal limitations on the spatial distribution of a gap species, seaside goldenrod (Solidago
sempervirens). Canadian Journal of Botany 71: 978–984.
Linnaeus, C. (1753) Species plantarum: exhibentes plantas rite cognitas, ad genera relatas, cum differentiis specificis, nominibus
trivialibus, synonymis selectis, locis natalibus, secundum systema sexuale digestas. Tomus II. Impensis Laurentii Salvii, Holmiae,
640 pp.
Lopez Laphitz, R.M. (2009) The genus Solidago L. (Astereae, Asteraceae) in South America and related taxa in North America. Masters
thesis, University of Waterloo, Canada.
Maddison, D.R. & Maddison, W.P. (2005) MacClade 4: Analysis of phylogeny and character evolution. Version 4.08a. Note on Solidago.
Botanical Journal of the Linnean Society 67: 279–280.
Meyen, F.J.F. (1834) Reisen um die Erde. Vol. 1.Sander’schen Buchhandlung, Berlin, 494 pp.
Michaux, A. (1803) Flora Boreali-Americana Vol. 2. Levrault, Paris & Strasbourg, 340 pp.
Miller, M.A., Pfeiffer, W. & Schwartz, T. (2010) Creating the CIPRES Science Gateway for inference of large phylogenetic trees. In: IEEE
(Ed.) Proceedings of the Gateway Computing Environments Workshop (GCE), 14 Nov. 2010. New Orleans, LA, pp. 1–8.
http://dx.doi.org/10.1109/GCE.2010.5676129
Miller, P. (1768) The gardeners dictionary. 8th Edition. Printed for the author, London, 1300 pp.
Nesom, G.L. & Morgan, D.R. (1990) Reinstatement of Tonestus (Asteraceae: Astereae). Phytologia 68: 174–180.
Palhinha, R.T. (1966) Catálogo das plantas vasculares dos Açores. Sociedade de Estudos Açorianos Afonso Chaves, Lisboa, 186 pp.
Peirson, J.A., Dick, C.W. & Reznicek, A.A. (2013) Phylogeography and polyploid evolution of North American goldenrods (Solidago
subsect. Humiles, Asteraceae). Journal of Biogeography 40: 1887–1898.
Roberts, R.P. & Urbatsch, L.E. (2003) Molecular phylogeny of Ericameria (Asteraceae, Astereae) based on nuclear ribosomal 3’ ETS and
ITS sequence data. Taxon 52: 209–228.
http://dx.doi.org/10.2307/3647390
Schaefer, H. (2003) Chorology and diversity of the Azorean flora. Dissertationes Botanicae 374: 1–130.
Schaefer, H. (2005) Flora of the Azores: a fieldguide, 2nd edn. Margraf Publishers/Backhuys Publishers, Weikersheim, 346 pp.
Schaefer, H., Carine, M.A. & Rumsey, F.J. (2011a) From European priority species to invasive weed: Marsilea azorica is a misidentified
alien. Systematic Botany 36: 845–853.
Schaefer, H., Moura, M., Belo Maciel, G., Silva, L.F., Rumsey, F.J. & Carine, M.A. (2011b) Taxonomic checklists and the Linnean
shortfall in oceanic island biogeography: a case study in the Azores. Journal of Biogeography 38: 1345–1355.

AZOREAN GOLDENRODS Phytotaxa 210 (1) © 2015 Magnolia Press • 59
Schaefer, H., Hardy, O.J., Silva, L.F., Barraclough, T.G. & Savolainen, V. (2011c) Testing Darwin’s naturalization hypothesis in the
Azores. Ecology Letters 14: 389–396.
Semple, J.C. (2003) New names and combinations in Solidago (Asteraceae: Astereae). Sida 20: 1605–1616.
Semple, J.C. & Cook, R.E. (2006) Solidago. In: Flora of North America Editorial Committee (Ed.) Flora of North America vol. 20,
Asteraceae, Part 2. Astereae and Senecioneae. Oxford University Press, New York, pp. 107–166.
Semple, J.C., Ringius, G.S., Leeder, C. & Morton, G. (1984) Chromosome numbers of goldenrods, Euthamia and Solidago (Compositae:
Astereae). II. Additional counts with comments on cytogeography. Brittonia 36: 280–292.
http://dx.doi.org/10.2307/2806528
Seubert, M. (1844) Flora Azorica quam ex collectionibus schedisque Hochstetteri patris et filii elaboravit. Apud Adolphum Marcum,
Bonnae, 50 pp.
Silva, J.J., Semple, J.C., Lopez Laphitz, R. & Sequeira, M. (2009) First record of La Plata River Goldenrod Solidago chilensis Meyen
(Asteraceae), in the Island of Madeira (Portugal). Boletim do Museu Municipal do Funchal 58: 32–36.
Silva, L., Moura, M., Schaefer, H., Rumsey, F. & Dias, E.F. (2010) Lista das plantas vasculares (Tracheobionta). List of vascular plants
(Tracheobionta) In: Borges, P.A.V., Costa, A., Cunha, R., Gabriel, R., Gonçalves, V., Martins, A.F., Melo, I., Parente, M., Raposeiro,
P., Rodrigues, P., Santos, R.S., Silva, L., Vieira, P. & Vieira, V. (Eds.) A list of the terrestrial and marine biota from the Azores.
Princípia, Cascais, pp. 117–146.
Stamatakis, A., Hoover, P. & Rougemont, J. (2008) A rapid bootstrap algorithm for the RAxML web-servers. Systematic Biology 75:
758–771.
http://dx.doi.org/10.1080/10635150802429642
St. John, H. (1915) An insular variety of Solidago sempervirens. Rhodora 17: 26–27.
Taylor, C.E.S. & Taylor, R.J. (1984) New species, new combinations and notes on the goldenrods (Euthamia and Solidago-Asteraceae).
Sida 10: 176–183.
Tiehm, A.J. & Shultz, L.M. (1985) A new Haplopappus (Asteraceae: Astereae) from Nevada. Brittonia 37: 165–168.
http://dx.doi.org/10.2307/2806095
Trelease, W. (1897) Botanical observations on the Azores. Missouri Botanical Garden Annual Report 8: 77–220.
http://dx.doi.org/10.2307/2992160
Tutin, T.G., Heywood, V.H., Burges, N.A., Moore, D.M. & Valentine, D.H. (Eds.) (1976) Flora Europaea: Plantaginaceae to Compositae
(and Rubiaceae), Vol. 4. Cambridge University Press, Cambridge, 534 pp.
Watson, H.C. (1870) Botany of the Azores. In: Godman, F. du C. (Ed.) Natural history of the Azores or Western Islands, ed. F. du C.
Godman. John van Voorst, London, pp. 113–220.
Wieczorek, A.M. & Geber, M.A. (2002) Microsatellite loci for studies of population differentiation and range expansion in Solidago
sempervirens L. (Asteraceae). Molecular Ecology Notes 2: 554–556.
http://dx.doi.org/10.1046/j.1471-8286.2002.00316.x
Yilin, C. & Semple, J.C. (2011) Solidago Linnaeus. In: Wu, Z.Y., Raven, P.H. & Hong, D.Y. (Eds.) Flora of China Volume 20–21
(Asteraceae). Science Press, Beijing & Missouri Botanical Garden Press, St. Louis, pp. 632–634.