Revised nomenclature and a new North American record for the villose cliff brake (Pellaea, Pteridophyta)

5 Pages·2003·3 MB·

by  M D Windham

Revised Nomenclature and a New North American Record for the
Villose Cliff Brake (Pellaea, Pteridophyta)
Michael I). Windham
Utah Museum of Natural History, University of Utah, Salt Uake City, Utah 84112-0050, U.S.A.
[email protected]
George Yatskievych
Missouri Botanical Garden, P.O. Box 299, St. Uouis, Missouri 63166-0299, U.S.A.
[email protected]
Abstract. Originally described hy Fee in 1857. boundaries of the Flora (Windham, 1993b), this
the villose cliff brake was not validly published for taxon is relatively widespread in Mexico and prob¬
another 136 years. In 1993, this member of the ably originated there (Windham, 1993a). Consid¬
Pteridaceae (sen.su Try on & Tryon) was named Pel¬ ering this distribution, it is not surprising that sub¬
laea ternifolia subsp. villosa Windham, acknowl¬ sequent work on the pellaeas of Mexico (Mendoza
edging its overall similarity to the ternate cliff et al., 2001; Mickel, in press) has substantially im¬
brake. Although formally including it within P. ter¬ proved our understanding of P. ternifolia subsp. vil¬
nifolia (Cavanilles) Link, Windham suggested that losa. These new data require a reconsideration of
the villose cliff brake was a probable allotetraploid both the taxonomy and the known distribution of
produced by hybridization between typical P. ter¬ this taxon.
nifolia and some other (then unidentified) taxon. Pellaea ternifolia was first described (as Pteris
Recent work by Mendoza et al. has identified a new ternifolia Cavanilles) in 1802. It has the most ex¬
species, P. ribae Mendoza & Windham, that ap¬ tensive distribution of any species in the genus,
pears to be the "missing parent." Considering the ranging from the southwestern United States to
evidence that the villose cliff brake is a reproduc-
southern Chile, with disjunct populations in Hawaii
tively competent taxon that originated as a hybrid
and Hispaniola (Tryon, 1957). Within this vast
between P. ternifolia subsp. ternifolia and P. ribae,
range, there is considerable habitat diversity and
we now feel that it should be recognized as a dis¬
ample opportunity for geographic isolation. These
tinct species. Herein, we propose a new combina¬
factors have combined to produce a puzzling array
tion, Pellaea villosa (Windham) Windham & Yat¬
of morphotypes and an equally complex synonymy
skievych, for the villose cliff brake. We also report
(Tryon, 1957: 150). Nearly all of the synonyms and
a northern range extension to Arizona, the second
proposed segregates have essentially glabrous
record for the area covered by the Flora of North
leaves like the type material of P. ternifolia from
America.
Peru (Nee s.n.; lectotype [designated by Christen¬
Key words: Arizona, North America, Pellaea,
sen. 1937]: MA not seen; isoleetotype: F). The first
Pteridaceae.
author to discuss the villose cliff brake per se was
fVe (1857), who provided an accurate description
While preparing a taxonomic treatment of Pel¬
laea for Volume 2 of the f lora of North America and the provisional name “P. lanuginosa.” Although
Project (Windham. 1993b), the senior author de¬ impressed by the apparent distinctness of this tax¬
termined that U.S. populations assigned to Pellaea on, Fee merely discussed it under P. ternifolia and
ternifolia (Cavanilles) Link (Pteridaceae: sublam. stated that he wanted to see more material before
Cheilanthoideae, sensu Tryon & Tryon, 1982) rep¬ recognizing it formally. Thus, the name “P. lanu¬
resent "three genetically distinct taxa characterized ginosa" Fee was considered a nomen provisorium
by differences in leaf morphology and chromosome by Tryon (1957), and has never been validated.
number" (Windham, 1993a: 42). The most distinc¬ Since Fee’s initial recognition of the villose cliff
tive element, a tetraploid with sparsely villous brake, its distinctive nature has become more evi¬
fronds, was segregated as P ternifolia (Cavanilles) dent. Alice Tryon noted a correlation between chro¬
Link subsp. villosa Windham, the villose cliff mosome number and leaf morphology in Pellaea
brake. Reported from a single locality within the ternifolia, stating that "the tetraploid plants gen-
Novon 13: 358-362. 2003.
Volume 13, Number 3 Windham & Yatskievych 359
2003 Villose Cliff Brake
Table 1. Comparison of selected taxa belonging to the / 1ellaea ternifolia comple x demonstrating the intermediacy
of P. villosa. The characters listed for P. ternifolia apply only to the diploid cytotype i of subspecies ternifolia (see text).
Characters I’ ternifolia P. villosa P. ribae
Rhizome scale shape Linear-subulate Linear Linear-filiform
Rhizome scale length 5—7 mm 6—12 mm 8-15 mm
Rhizome scale coloration Bicolored Mostly hicolored Concolored
Petioles and rachises Glabrous or with a ± Villous Densely villous
few scattered tri -
chomes
Distal portion of petioles Grooved or flattened Rounded to slightly Rounded adaxially
adaxially flattened adaxially
Pinnae Ternately divided ex¬ Usually entire for dis¬ Entire throughout
cept at leaf apex tal Vi—Yi
Ahaxial pinna surfaces Glabrous Sparsely villous Densely villous
Apical mucro on pinnae Well-developed Weakly developed Absent
Fargest ultimate segments Usually <18 mm Usually >18 mm Usuallv <18 mm long
long long
Mean spore size 39-45 gm 46—53 gm 53—60 gm
Chromosome number n = 29 (diploid) n = 58 (tetraploid) Unknown
erallv resemble the diploids, but the pinnae tend Although morphological intermediacy is expect¬
to be entire and the rachises more pubescent" ed in taxa of hybrid origin, it rarely extends to mean
(Tryon, 1972: 230). Based on isozyme profiles and spore size, as it does in the case of the villose cliff
additional chromosome studies, the senior author brake (Table 1). In general, allotetraploid ferns pro¬
suggested that the “tetraploids are segmental allo¬ duce spores that are somewhat larger than either
polyploids produced by hybridization between typ¬ parent due to increased DNA content. The lack of
ical P. ternifolia and other (as yet unidentified) dip¬ a chromosome count for Pellaea ribae leaves open
loid elements within the P. ternifolia complex" two possibilities. If spore sizes are strictly correlat¬
(Windham, 1993a: 42). With no specimens of the ed with ploidy level in this group, then the mate¬
“missing parent” at hand, Windham (1993a, rials of P ribae currently at hand likely represent
1993b) took a conservative approach to the taxon¬ a polyploid derivative of the “missing parent" of
omy of the group and recognized the villose cliff subspecies villosa. Alternatively, P. ribae may ex¬
brake as a subspecies of P. ternifolia. emplify a rare exception to the correlation between
Subsequent work on the cliff brakes of Mexico chromosome number and spore size. The species
(Mendoza et al.. 2001; Mickel. in press) indicates appears to be a strict gypsophile, and larger prop-
that this classification needs revision. Central to our agules might provide an adaptive advantage in such
change in thinking is the discovery of Pellaea ribae a harsh habitat. Notholaena bryopoda Maxon, an¬
Mendoza & Windham (Mendoza el al.. 2001), the other strict gypsophile that is known to be diploid,
probable “missing parent" of the villose eliff brake. also bas exceptionally large spores (Windham, un¬
As shown in Figure 1 and Table 1. subspecies vil- published). Plants of P. ribae derived from spores
losa is morphologically intermediate between dip¬ of the type collection currently are being grown at
loid P. ternifolia subsp. ternifolia and P. ribae in the Missouri Botanical Garden in hopes of testing
nearly every character examined. From minute de¬ these hypotheses.
tails of the rhizome scales to the more obvious fea¬ Regardless of the actual chromosome number of
tures of frond dissection and pubescence, the vil¬ Pellaea ribae, the data at hand indicate that the
lose cliff brake exhibits clear intermediacy. The features distinguishing subspecies villosa from typ¬
only exception is in the length of the largest ulti¬ ical P. ternifolia derive from a genome shared with
mate segments. These usually exceed Iff mm in P. ribae. This genome is strikingly divergent from
subspecies villosa and are usually less than 18 mm that of diploid P. ternifolia, confirming the sugges¬
in both diploid P. ternifolia subsp. ternifolia and P. tion (Windham, 1993a) that the villose cliff brake
ribae (Table 1). Such increases in frond segment is an allotetraploid. Although subspecies villosa
size are common in hybrid ferns (e.g., Windham. was formed originally by interspecific hybridization
1988; Paler & Barrington, 1995) and may be at¬ between P. ternifolia and P. ribae, we contend that
tributable to heterosis (Grant, 1975). it is a reproductively competent allopolyploid taxon
360 Novon
Figure 1. A—C. Diploid P. ternifolia (Cav.) Link subsp. ternifolia drawn from Mickel 3849 (NY). —A. Leaf and rhizome.
—B. Racbis and pinna attachment (adaxial surface). —C. Rhizome scale. D—F. P. villosa (Windham) Windham &
Yatskievych drawn from Rzedowski 22894 (NY). —1). Leaf and rhizome. —K. Rachis and pinna attachment (abaxial
surface). —F. Rhizome scale. G—I. P. ribae Mendoza & W indham drawn from Heil 5247 (BRY). —G. Leaf and rhizome.
—H. Rachis and pinna attachment (abaxial surface). —I. Rhizome scale. Scale bar = 5 cm (A, I). G), 1 mm (B, C,
E, F. H). Adapted from Mickel (in press).
Volume 13, Number 3 Windham & Yatskievych 361
2003 Villose Cliff Brake
lliat should be treated taxonomically as a separate tains of western Texas (Windham, 1993b). However,
species. It is amply distinct morphologically from recent botanical surveys of the Madrean “sky is¬
both P ternifolia and P ribae (Table 1). lands” of southern Arizona have identified a second
The appendix on nothotaxa in the International U.S. population:
Code of Botanical Nomenclature (Greuter et al..
U.S.A. Arizona: Cochise Co., Huachuca Mountains,
2000) states in Article H.5.1. “The appropriate
Lone Mountain Watershed, Sycamore Canyon, above Mud
rank of a nothotaxon is that of the postulated or
Spring, 31 "24'N I 10'23Vi'W elev. 5700 ft.. E-facing slope
known parent taxa." However, Article 3.2 states, “If just above arroyo bottom in pine-oak woods with Juniperus
the postulated or known parent taxa are of unequal deppeana, Quercus emoryi, Eragrostis, Muhlenbergia emer-
sleyi, Cyperus, Bouteloua, Tradescantia, Ranunculus ari-
rank, the appropriate rank of the nothotaxon is the
zonicus, Ipomoea, Calliandra, Bommeria, terrestrial; un¬
lowest of these ranks.” Given that we postulate that
common, 11 Aug. 1996, M. Fishbein 2861, with K. Hooper
P villosa arose following hybridization between a & R. Coleman (ARIZ).
subspecies of P. ternifolia and the species P ribae,
This new population is situated approximately
these statements might be interpreted to require the
300 km north to northwest of the closest known
continued recognition of P. villosa at the subspecific
localities in the Sierra Madre Occidental and 600
rather than specific level. We drawr attention to
km west-northwest of the only other U.S. collection
problems of interpretation created by this section
site. It remains to be seen whether Pellaea villosa
of the Code, especially as applied to fern nomen¬
will be found in the mountains of southern New
clature. The present practice among biosystem-
Mexico, which have fewer Madrean habitats. At the
atists, including most pteridologists, is to distin¬
moment, it belongs to a group of species (including
guish between primary hybrids and the
the ferns Cheilanthes lendigera (Cavanilles) Swartz,
reproductively competent taxa (sensu Wagner,
Notholaena aschenborniana Klotzsch, N. neglecta
1954) that evolve from them following polyploidi-
Maxon, and P. ternifolia (Cavanilles) Link subsp.
zation. Such allopolyploids are reproductively iso¬
arizonica Windham) that are known from Arizona
lated from either parent and function as full spe¬
and trans-Pecos Texas but not from intervening re¬
cies; therefore we believe they should not be
gions of New Mexico.
treated nomenclaturally as hybrids.
Although we considered validating Fee's nomen Acknowledgments. We are grateful to John
provisorium for this taxon, we have not seen the Mickel for sharing his extensive knowledge of the
type (staff at P were unable to locate the specimen). taxa under consideration and for permission to use
Thus we are uncertain as to the precise application illustrations (skillfully drawn by Haruto Fukuda)
of the name. We therefore propose the following from his upcoming book on the ferns and allied
new combination based on the epithet applied by plants ol Mexico. We thank Mark Fishbein for per¬
Windham (1993a): mission to publish his discovery of Pellaea villosa
in Arizona, and express our gratitude to the cura¬
Pellaea villosa (Windham) Windham & Yatskiev¬ tors of the cited herbaria for loans and access to
ych, comb, et stat. nov. Basionym: Pellaea ter¬ their collections. Our reviewers, R. James Hickey,
nifolia (Cavanilles) Link subsp. villosa Gerald J. Gastony, Ruth Kirkpatrick, and Alan R.
Windham, Contr. LJniv. Michigan Herb. 19: 43. Smith, provided insightful comments that improved
1993. TYPE: Mexico. Hidalgo: rocky hills, our discussion greatly. We also w ish to acknowledge
Lena Station, 8300 ft., 24 Aug. 1905, C. G. the Utah Museum of Natural History (UMNH) and
Pringle 10025 (holotype, LL; isotypes, ARIZ, the New York Botanical Garden (NY) for providing
BRY, CAS, COLO, DAO, ENCB, F, GH. INI), funding that made this work possible. Finally, we
KANG, LL, MSC. OKLA, SMU, TEX, UC, US, thank Jill Schwartz for her help with scanning and
VT, WIS). assembling the figure.
This species, occurring on various rocky sub¬ Literature Cited
strates, is widespread in the Sierra Madre Occiden¬
Christensen, C. 1937. Taxonomic fern studies III. Revi¬
tal and Sierra Madre Oriental of northern and cen¬
sion of the genera and species of ferns described by A.
tral Mexico. Mickel (in press) has identified J. Cavanilles. Dansk Rot. Ark. 9(3): 3—32, fig. 1. tab.
specimens from the states of Chihuahua, Coahuila, 1-L
Durango, Guanajuato, Hidalgo, Mexico. Michoacan, Fee, A. L. A. 1857. Iconographie des espfeces nouvelles
decrites on enumerees dans le genera filicum et revision
Nuevo Leon, Puebla, San Luis Potosi, and Sonora.
des publications ant6rieures relatives t) la famille des
In the United States, the villose cliff brake has been
fougeres, 111. Memoires sur ies Families de Fougeres 8:
reported from a single locality in the Davis Moun¬ 67—138. Veuve Berger-Levault et fils, Strasbourg.
362 Novon
Grant. V. 1975. Genetics of Flowering Plants. Columbia Tryon, A. F. 1957. A revision of the fern genus Pellaea
Univ. Press, New York. section Pellaea. Ann. Missouri Bot. Garth 44: 125—193.
Greuter, W.. J. McNeill. F. li. Barrie, H. M. Burdet, V. -. 1972. Spores, chromosomes anti relations of the
Demoulin, T. S. Filgueiras, 1). H. Nicolson, P. C. Silva. fern Pellaea atropurpurea. Rhodora 74: 220—241.
J. K. Skog. P. Trehane, N. J. Turland & 1). I,. Hawk- Tryon, R. M. & A. F. Tryon. 1982. Ferns anti Allietl
sworth. 2000. International Code of Botanical Nomen¬ Plants, with Special Reference to Tropical America.
clature (Saint Louis Code). Regnum Veg. 138. Springer-Verlag, New' York.
Wagner, W. H., Jr. 1954. Reticulate evolution in the Ap¬
Mendoza, A., M. Windham, B. Perez-Garcfa «S G. Yat-
palachian aspleniums. Evolution 8: 103—118.
skievych. 2001. Una nueva especie de Pellaea (Pteri-
Windham, M. I). 1988. The Origin and Genetic Diversi¬
daceae) del estado de San Luis Potosf, Mexico. Acta
fication of Polyploid Taxa in the Pellaea wrightiana
Bot. Mex. 57: 15-21.
Complex (Adiantaceae). Ph.D. Dissertation, University
Mickel, J. T. In press. Pellaea. In J. T. Mickel & A. R.
of Kansas.
Smith, Pteridophyte flora of Mexico. Mem. New York
-. 1993a. New taxa and nomenclatural changes in
Bot. Garth the North American fern flora. Contr. Univ. Michigan
Paler. M. H. & I). S. Barrington. 1995. The hybrid Cys- Herb. 19: 31-61.
topteris fragilis X C. tenuis (Dryopteridaceae) anti the -. 1993b. Pellaea. Pp. 175-186 in Flora of North
relationship between its tetraploid progenitors. Syst. America Editorial Committee (editors). Flora of North Amer¬
Bot. 20: 528-545. ica North of Mexico. Vol. 2. Oxford Univ. Press, New York.