AMERICAN Volume 92
FERN Number
J O U R N A q January—March 2002

QUARTERLY JOURNAL OF THE AMERICAN FERN SOCIETY

Obituary: Rolla Milton Tryon, Jr. (1916-2001)
Gerald J. Gastony, David S. Barrington, and David S. Conant 1

a Associated with the Intertracheid Pit seas) of the Woody Fern Botrychium
multifidum ela C. Morrow and Roland R. Dute 10

A New Filmy Fern from the Dominican Republic Carlos Sanchez 20

Adiantum argutum, an Unrecognized Species of the A. latifolium Group
Jefferson Prado and David B. Lellinger 23

Polypodium cseapage Plants Sporulate viacpecsndaeed in a Non-seasonal Glasshouse Envi-
ronmen a E. Simdn and Elizabeth Sheffield 30

The American Fern Society

Council for 2002

CHRISTOPHER H. HAUFLER, Dept. of Botany, University of Kansas, Lawrence, KS 66045-2016.
President
TOM RANKER, University Museum, Campus Box 265, University of Colorado, Boulder, art eu ‘65.
President
W. CARL TAYLOR, 800 W. Wells St., Milwaukee Public Museum, Milwaukee, WI eg 1478.
Secretary
JAMES D. CAPONETTI, Dept. of Botany, University of Tennessee, Knoxville, TN 37916-1110.
Treasurer
GEORGE YATSKIEVYCH, Missouri Botanical Garden, P. O. Box 200, St. oe MO 63166-0299.
rship Secretary
JAMES D. MONTGOMERY, Ecology HI, R.D. 1, Box 1795, Berwick, PA 18603-9801.
Back — Curator

R. JAMES HICKEY, Botany Dept., Miami University, Oxford, OH 45056. rnal Editor
DAVID B. LELLINGER, U.S. National Herbarium MRC-166

Smithsonian Institution, Washington, DC 20560-0166. Memoir Editor
CINDY JOHNSON-GROH, Dept. ‘of pom — Adolphus College,

800 W. College Ave., St. Peter, MN 56082 Bulletin Editor

American Fern Journal
EDITOR

R. JAMES HICKEY Botany De
ami University, Oxford, OH 45056
ph. (513) £5 Sone e-mail: hickeyrj @ muohio.edu

ASSOCIATE EDITORS
GERALD J. GASTONY .............. Dept. of Biology, Indiana University, ages aig IN 47405-6801
CHRISTOPHER H. HAUFLER ...... Dept. of Botan get University of Kansa 66045-2106
ROBBIN C. MORAN ew York Botanical Gecten: Bike NY 10458-5126
JAMES H. PECK ser of — University of Arkansas—Little Rock,
2801 S. University Ave., Litthe Rock, AR 72204

The “American Fern Journal” (ISSN 0002- inapls is an illustrated quarter! iene a the sf acting
study of ferns. It is owned by the American Fern Society, and ame at ate st St
VA 22180-4151. Periodicals postage paid at Vienna, VA, and addition

— for bare: issues, made 6 months (domestic) to 12 sani sei ‘cael after the date of issue,
and r back issues should be addressed to Dr. James D. Montgomery, Ecology Il, R.D. 1,
Berwick ies 18603. 9801.

Chan of address, dues, and applications for membership should be sent to the Membership
Secretary.

General inquiries posianeag ferns should be addressed to the Secretary.

Subscriptions $20.00 gross, $19.50 net if paid through an agency (agency fee $0.50); sent free to
members of the prorestians Fern Soaew. (annual dues, $15.00 + wecbe mailing surcharge beyond
U.S.A.; life oni ne rship, $300.00 + $140.00 mailing surcharge beyond U.S.A.

Back volumes are available for most years as printed issues or on microfiche. Please contact the
Back Issues Curator for prices and availability.

POsTMASTER: Send address changes to AMERICAN FERN JOURNAL, 326 West St. NW., Vienna, VA
22180-4151.

FIDDLEHEAD FORUM
The editor - the Bulletin of the American Fern Society welcomes contributions from members and

non-members, including miscellaneous notes, offers to ee or purchase materials, personalia,
horticultural sae and reviews of non-technical books on

SPORE EXCHANGE
Mr. Stephen McDaniel, 1716 Piermont Dr., Hacienda Hts., CA 91745-3678, is Director. pote
exchanged and lists of available spores sent on request. http://www.amerfernsoc.org/sporexy
GIFTS AND BEQUESTS
Gifts and dais ssn ps to ia Society enable it to expend it its services to members and to others interested
in ferns. Ba other gifts are always welcomed and are tax-deductible.

Tnonin lA hy - > thn < me
4) te OCCretary

American Fern Journal 92(1):1—9 (2002)

MISSoy

Ay
Obituary: Rolla Milton Tryon, Jr. g4p 1 4 2099
(1916-2001) DEN LiBRap

GERALD J. GASTONY
Department of Biology, Indiana University, Bloomington, IN 47405-3700

Davip S. BARRINGTON
Department of Botany, University of Vermont, Burlington, VT 05404-0086

Davip S. CONANT

Department of Natural Sciences, Lyndon State College, Lyndonville, VT 05851

Rolla Tryon, a member of the American Fern Society since 1932 and one of
the twentieth century’s most eminent students of pteridophytes, was born on
August 26, 1916 in Chicago, Illinois. His father, a professor of American history
and education at the University of Chicago, maintained a summer cottage in
Chesterton, Indiana in addition to his home in Chicago. Rolla’s fascination
with ferns and fern allies developed during boyhood forays from that Ches-

The photograph was taken by Dr. Walter H. Hodge in Mexico City in December, 1972 and was
made available by the Hunt Institute for Botanical Documentation.

2 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 1 (2002)

terton cottage into sand dune habitats along Lake Michigan in the northwest
of Indiana. At the age of 18, he published his first paper, relating his obser-
vations on Osmunda plants in the Indiana Dunes (see complete bibliography
below). As a boy, Rolla was greatly influenced by, and in turn influenced,
Charles Deam (author of the 1940 Flora of Indiana), advising Deam about fern
species he had found in the dunes area. When a doubting Deam appeared at
the cottage door one day asking to meet Rolla and to be shown these ferns in
situ, he was surprised to learn that Rolla was not the adult of the family but
a mere boy of 14. Thus began a productive friendship documented in Rolla’s
correspondence with Deam from May, 1935 to January, 1953. All the penny
postcards and letters he received from Deam have been carefully maintained
in one of Rolla’s files, now archived at Indiana University—fascinating reading.
Rolla’s insatiable boyhood appetite for ferns got him into a bit of trouble at
home, however, when his father learned that he had charged Bower's three
volumes on The Ferns to his account at Brentano’s bookstore in Chicago.

Rolla built a solid academic superstructure on the foundation of these boy-
hood experiences. Among his academic accomplishments were an A.A. degree
in 1935 and a B.S. degree in 1937, both from the University of Chicago, and
a Ph.M. in 1938 from the University of Wisconsin. In 1940 he earned an MS.
and in 1941 a Ph.D., both from Harvard University. During his days as a Har-
vard student, he contracted malaria in South Carolina while collecting plants
for M. L. Fernald, and during the war-torn year following completion of his
Ph.D. he served as a lab technician in the U. S. Chemical Warfare Service at
Massachusetts Institute of Technology. His father thought he should follow his
Ph.D. in botany with another, this time in chemistry, so that he could earn a
living, but instead Rolla became an Instructor in Botany first at Dartmouth
College, then at the University of Wisconsin before becoming an Assistant
Professor in Botany at the University of Minnesota in 1945. While an Assistant
Professor at the University of Wisconsin, Rolla met Alice Faber. Their marriage
in 1945 initiated not only a happy and enduring domestic partnership but also
a research synergism whose productivity has nourished pteridologists through-
out the world.

In 1947 Rolla became Associate Professor in Botany at Washington Univer-
sity, St. Louis and Assistant Curator of the herbarium of the Missouri Botanical
Garden, positions he held to 1957. During this appointment, he and Alice were
the original organizers of the Missouri Botanical Garden’s annual Systematics
Symposium, whose 48th meeting was held 12-13 October, 2001. This highly
successful annual meeting has received continuous support from the National
Science Foundation from its second year (1954) to the present (with the lapse
of a single year). From 1946 to 1957 Rolla served as curator and librarian of
the American Fern Society’s library and herbarium, responding to members’
requests for loans of materials. That herbarium and library was subsequently
entrusted to Warren H. Wagner at the University of Michigan. Following a year
as Research Associate at the University of California, Berkeley, Rolla went to
the Gray Herbarium of Harvard University as Associate Curator and Curator
of Ferns in 1958 and became Curator of the Gray Herbarium in 1967. Rolla

ROLLA MILTON TRYON, JR. 3

and Alice traveled the world extensively, attending international meetings,
conducting field work, studying specimens at major herbaria in the Americas,
Europe, and Africa, and conducting field courses on the ferns. In addition to
other services to professional societies, Rolla served for many years as Asso-
ciate Editor of Rhodora and the American Fern Journal, as Associate Editor of
Brittonia (1961-1964), as Editor-in-Chief of Rhodora (1977-1982), and as Pres-
ident of the New England Botanical Club and the American Fern Society.

A framed photograph of his revered mentor, Charles A. Weatherby (see
American Fern Journal 40[1] for a remarkable series of papers honoring this
unusually respected and beloved botanist), was always prominently displayed
on Rolla’s desk at Harvard, undoubtedly inspiring his own welcoming, patient,
and supportive response to all who entered his office seeking counsel. In 1970
Rolla initiated an annual New England Fern Conference at Harvard Forest. For
20 years this provided a stimulating intellectual setting in which students of
fern biology discussed and developed their ideas. In 1972 he became Professor
of Biology at Harvard University, holding both the Curatorship and Professor-
ship until his retirement in 1987. He remained at Harvard as Professor Emer-
itus from 1987 to 1989 when he moved to the University of South Florida in
Tampa as Adjunct Professor, bringing with him his extensive library of fern
and biogeographic literature. To mark the retirements of Alice and Rolla Tryon
from Harvard, a festschrift of 13 papers plus introduction was published in
their honor in the Annals of the Missouri Botanical Garden (vol. 77: 225-339.
1990).

At the University of South Florida, Rolla and Alice helped found the Insti-
tute for Systematic Botany and endowed the Tryon Lecture Series that brings
several internationally known botanists to the university each year. In their
research-active office on the Tampa campus, he and Alice continued their pter-
idological work, as his following bibliography indicates.

Rolla Tryon’s publication list exceeds 100 titles and includes a great breadth
of topics. Papers ranged from articles on pteridophytes for the 1943 Encyclo-
pedia Britannica to a glossary of terms relating to the fern leaf, discussions of
the history of pteridology and fern classification, a remembrance of his grad-
uate mentor and counselor Charles A. Weatherby, discussions of the formali-
ties of fern nomenclature, and many book reviews. His monographs and re-
visions focused mostly on ferns but also included angiosperms Convolvulus
and Elymus. Signal among these were his revisions of Pteridium, Doryopteris,
the Selaginella rupestris group, American Notholaena, and the Cyatheaceae.
His papers on fern biogeography began with Doryopteris in 1944, matured in
his exposition of geographic speciation in Selaginella in 1971, and continued
to his and Alice’s 1999 discussion of the phytogeography of eastern North
American ferns (honoring Ching Ren-Chang). Floristic and taxonomic notes on
ferns ranged from simple observations of growth forms and hybrids to eluci-
dations of complex taxonomic and nomenclatural issues. For his 1955 publi-
cation on the taxonomy of cycads (coauthored with students in his Washington
University class) he was awarded the 1956 Robert Montgomery award of the
Fairchild Tropical Garden for distinguished achievement in the world of palms

4 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 1 (2002)

and cycads. At the time of his death, he had a book review in press in Rhodora
and a paper in press in Bradea (coauthored with his former student Paulo
Windisch).

In addition to his numerous papers on ferns and other topics, Rolla is no-
table for his books. Among these are two editions of his Ferns and Fern Allies
of Wisconsin (1940, 1953) and Ferns and Fern Allies of Minnesota (1954,
1980). He is renowned for his knowledge of the ferns of Peru, first expressed
in his 1964 Ferns of Peru (250 pages in the Contributions from the Gray Her-
barium of Harvard University). This treatment was updated and completed in
six subsequent parts entitled Pteridophyta of Peru between 1989 and 1994,
mostly coauthored with Robert Stolze but with several portions contributed
by other pteridological specialists. His monumental 1982 book Ferns and AIl-
lied Plants with Special Reference to Tropical America, coauthored with Alice
Tryon, is an encyclopedic treatment of this subject that continues to stimulate
new research, as does his treatment of Pteridaceae, with Alice Tryon and Karl
Kramer, in volume 1 of Families and Genera of Vascular Plants edited by K.
Kramer and P. S. Green.

Rolla’s kindly and perceptive mentoring and his outstanding contributions
to our knowledge of ferns is signaled by having the following four fern taxa
named in his honor.

1) Asplenium tryonii Correll. In describing this species, Donovan Correll
(1961) said “It is a pleasure to name this species for Dr. Tryon, who has
always been most gracious in helping his fellow-workers with their never-
ending problems in the study of ferns.” Known only from Chihuahua, Mex-
ico, this species was further discussed and illustrated in Ferns and Fern
Allies of Chihuahua by Knobloch and Correll (1962).

Alsophila tryonorum Riba. The eminent Mexican pteridologist Ram6n Riba
(see American Fern Journal 90:112—118, 2000) stated that “this species is
named after Dr. Rolla M. Tryon and Dr. Alice F. Tryon for their contributions
to the taxonomy of the ferns” (Riba, 1967). The plural specific epithet rec-
ognizes the close professional relationship between this highly productive
research team. This tree fern species is now known as Trichipteris tryono-
rum (Riba) R. Tryon following its transfer by Rolla in his 1970 paper on the
classification of the Cyatheaceae.

Nephelea tryoniana Gastony. ‘“‘I am pleased to name this species for my
mentor, Dr. Rolla M. Tryon, in recognition of his outstanding contribution
to the understanding of the systematics and evolution of the family Cy-
atheaceae” (Gastony, 1973). Subsequent research by Conant (Conant and
Cooper-Driver, 1980; Conant, 1983) revealed that this tree fern species is a
reproductively stabilized diploid hybrid species that is now regarded as
Alsophila tryoniana (Gastony) Conant.

Tryonella Pichi Sermolli. This new generic name was established by Pichi
Sermolli (1974) “in honour of the eminent pteridologist R. M. Tryon, Jr.,
author of many important papers on ferns, who, inter alia, supported the
distinction of the present genus from Doryopteris, though without giving it

i)
—

ioe)
—

ns
—

ROLLA MILTON TRYON, JR. 5

anew name.” This name is currently regarded as a synonym of Doryopteris
by Tryon and Tryon (1982) and Tryon, Tryon, and Kramer (1990).

Among the doctoral graduate students he trained, Rolla counted the follow-
ing (those with asterisks received their degrees from other institutions): Alice
F. Tryon, *Karl Kramer, *Ramon Riba, Gerald Gastony, Lawrence Palkovic,
David Barrington, David Conant, Paulo Windisch, *R. James Hickey, *Robbin
Moran, Sonia Sultan, and Calvin Sperling. He also mentored Robert Stolze in
his taxonomic revision of Cnemidaria at the Field Museum. Always available
to his students, he modeled his supportive and insightful mentoring on his
experiences with his own graduate mentor, Charles Weatherby. For this he has
earned our love as well as our respect. His impact on his students, and their
students, and their students is incalculable.

In 1978 Rolla M. Tryon, Jr. was elected to honorary membership in the Amer-
ican Fern Society, a special category of membership for persons who have
made outstanding contributions to the study of ferns. In 1984 he received a
Merit Award from the Botanical Society of America “In recognition of distin-
guished achievement in and contributions to the advancement of botanical
science. Pre-eminently knowledgeable in matters of taxonomy and nomencla-
ture, this foremost pteridologist is a perceptive student of phytogeography and
of the evolutionary impact of the selective process during plant migration.”

In addition to Rolla’s botanical activities he was also highly skilled in run-
ning a family farm in Knox County, Indiana for many years. He visited the
farm, overseeing its management, a few times each year. The extensive records
he kept in managing crops and livestock illustrate his practical ability in man-
aging business as well as scientific data.

On August 20, 2001, six days before his eighty-fifth birthday, Rolla Milton
Tryon, Jr., left us to continue our work with the pteridophytes of the world,
and to delight in them, without him. We do this fortified by his writings, the
echoes of his encouraging words, and his everlasting example. He was the
beloved husband of Alice Faber Tryon, the benefactor of countless students of
pteridophytes, including many who never knew him personally, an inspiration
and counselor to many collaborators and coauthors, the advisor of doctoral
students, the teacher of innumerable undergraduates, and our dear friend and
mentor. He will be deeply missed. He already is.

LITERATURE CITED

CONANT, D. S. 1983. sis revision of the genus Alsophila (Cyatheaceae) in the Americas. J. Arnold
Arbor. 64: 333-—

Conant, D. S. and : ones DRIVER. 1980. Autogamous allohomoploidy in Alsophila and Ne-

phelea (Cyatheaceae): a new hypothesis for speciation in homoploid homosporous ferns.

mer. J. Bot. 67: 1269-1288.

CorreLL, D. S. 1961. Two Texas-Chihuahuan ferns. Wrightia 2: 200-203.

— G. J. 1973. A revision of the fern genus Nephelea. Contr. Gray Herb. 203: 81—

eure _ I. W. and D. S. CorreLL. 1962. Ferns and fern allies of Chihuahua, pace —

a Foundation, Renner, Tex
PICHI pameene R. E. G. 1974. seit a Webbia 29: 1-16.

6 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 1 (2002)

Risa, R. 1967. New taxa in the genus Alsophila. Rhodora 69: 65-68.

BIBLIOGRAPHY OF ROLLA M. TRYON, JR. (1916-2001)

TRYON, R. M., JR. 1934. Some observations on Osmundas. Amer. Fern J. 24:

TRYON, R. M., JR. 1936. Ferns of the dune region of Indiana. Amer. Mid. en 17: ite

TRYON, R. M., JR. 1936. Botrychium dissecturn and forma obliquum. Amer. sg n J.

TRYON, R. M., JR. 1938. The phenomenon of forking in ferns. Amer. Fern J. 2

TRYON, R. M., JR. 1938. Recent additions to the flora of Indiana. Proc. “la poe Sci. 47: 76—
17

TRYON, R. M., JR. 1939. Notes on the ferns of Wisconsin. Amer. Fern J. 2 9,
TRYON, R. M., JR. 1939. The varieties of Convolvulus spithamaeus and a C. sepium. Rhodora 41:
5—423.

TRYON, R. M., JR. 1940. Notes on some Indiana plants. Proc. Indiana Acad. Sci. 49: 89-90.

TRYON, R. M., JR., N. C. Fassett, D. W. DUNLOP, and M. E. 6 EMER. 1940. The ferns and fern allies
of Sip — Dept., Univ. of Wisconsin, Madiso

TRYON, R. M., JR. 0. An Osmunda hybrid. Amer. Fern J. 30: 65-66.

TRYON, R. M., JR. 1 mi a revision of the genus Pteridium. Bhai 43: 1-31, 37-67. Reprinted as
Contr. Gray Herb. 1 -7

bia R. M., JR. 1942. ae Roland, A. E., The ferns of Nova Scotia. Amer. Fern J. 32: 73—

Perey R. M., JR. 1942. A new Dryopteris hybrid. Amer. Fern J. 32: 81—

TRYON, R. M., JR. 1942. A revision of the genus Doryopteris. Contr. G ae 80.
TRYON, R. M., JR. 1942. Review of Brown, C. A. and D. S. Correll, The ferns ak 7e allies of
Louisiana. Rhodora 44: 484—485.

sia
TRYON, R. M., JR. 1943. Several articles on Pteridophyta for Encyclopedia Brittanic
TRYON, R. M., JR. ma Abstracts of the American Fern Jour nal for Biological Abstracts.

TRYON, R. M., JR. and J. W. Moore. . Notes on aquatic and prairie acca in ane
i 6.
TRYON, R. M., JR. and J. W. Moore. 1946. A preliminary checklist of the flowering plants, ferns
and fern allies of Minnesota. Univ. of Minnesota, Minneapolis.
TRYON, R. M., JR. 1946. A new Doryopteris hybrid. Amer. Fern J. 36: 4
OorE, J. W. and R. M. TRYON, JR. 1946. A new record of Isoétes inalsinpod. Amer. Fern J. 36:
89-91
vista F, K and R. M. TRYON, JR. 1948. A fertile mutant of a Woodsia hybrid. Amer. Fern J. 35:
13
TRYON, M., JR. 1948. Some Woodsias from the north shore of Lake Superior. Amer. Fern J. 38:
159-170.
Boouer, L. E. and R. M. TRYON, JR. 1948. A study of Elymus in Minnesota. Rhodora 50: 80-91.
TRYON, R. M., JR. 1950. Charles Alfred Weatherby—teacher and counselor. Amer. Fern J. 40: 9-10.
TRYON, R. M., JR. 1950. A new erect species of the Selaginella rupestris group. Amer. Fern J. 40:

69-74.
TRYON, R. M., JR. — Ferns of the Missouri Ozark region. Missouri Bot. Gard. Bull. 39: 136-138.

TRYON, R. M., JR. 1951. Review of Manton, I., Problems of cytology and evolution in the Pterido-
phyta. Ecology 32: 769-770.
TRYON, R. M., JR. 1952. A sketch of the history of fern classification. Ann. Missouri. Bot. Gard. 39:

6
TRYON, R. M., JR., N. C. Fassett, D. W. DUNLOP and M. E. DiEMER. 1953. The ferns and fern allies
of Wisconsin, Ed. 2. Univ. Wisconsin Press, Madison.
TRYON, R. M., JR. 1954. The ferns and fern allies of Minnesota. Univ. Minnesota Press, Minneap-
olis

TRYON, R. M., JR, E. BARBOUR, E. Davis, H. Kipp, R. Lonc, C. Marvin, B. MIKULA, and R. MOHLEN-
BROCK. 1955. Ancient seed plants: the cycads. Missouri. Bot. Gard. Bull. 43: 65-80.

ROLLA MILTON TRYON, JR.

TRYON, iy M., re ney ep ae ager vee its allies. Ann. Missouri. Bot. Gard. 42: 1-99.
KRAME N, JR. 1955.

eeu. new species of Doryopteris from Surinam. Ann.
Missouri Bot a < _ ner oe
Sti . — JR. 1956. A revision of the American species of Notholaena. Contr. Gray Herb. 179.

aa. . a JR. 1957. Adianturn in Peru: new species and combinations. Amer. Fern J. 47: 139—
ah

KosuskI, C. E., C. V. MorTON, M. Ownsey, and R. M. TRYON. 1958. Report of the committee for
recommendations on desirable procedures in herbarium practice and ethics. Brittonia 10:
93-95.

TRYON, R. M., JR. and A. F. TRYON. 1959. Observations on cultivated ferns: the hardy species of
tree ferns (cha and Cyatheaceae). Amer. Fern. J. 49: 129-142. Reprinted in Lasca Leaves
10: 26-33. 1960.

TRYON, R. M., Ir. 1960. The ecology of Peruvian ferns. Amer. Fern J. 50: 4
TRYON, R. 1960. A glossary of some terms relating to the en — Taxon c 108-109, Reprinted

in Russian iio in Bot. Jour. Akad. Nauk, U.S.S 736-739.
TRYON, M., JR. 1960. Review of Pteridophyta in Munz, P. a o balou rene Amer. Fern J.
50: seit

TRYON, R M. JR. 1960. A review of the genus Dennstaedtia in America. Contr. Gray Herb. 187:

23—

TRYON, * 4 JR. 1960. New species of ferns from serait mei South America. Rhodora 62: 1-10.

TRYON, R. M., JR. 1961. Taxonomic fern notes. I. Rhodora 63: 7-88

TRYON, R. M., JR. 1962. The fern genus Doryopteris in aa ee and Rio Grande do Sul,
Brazil. Sellowia 14: 51-59

TRYON, R. M., JR. 1962, Review af Wherry, E. T., The Fern Guide. Amer. Fern J. 52: 89-91.

TRYON, R. M., JR. 1962. A note on Nephrolepis pa ie cv. Duffii. Amer. Fern J. = 153-155.

TRYON, R. M., JR. 1962. Taxonomic Fern Notes. II. Pityrogramma (including Trismeria) and Ano-
gramma. Contr. Gray Herb. 189. 52-76.

TRYON, R. 1962. Taxonomic fern notes. III. Contr. Gray Herb. 191: 91—

TRYON, R. M., JR. 1962. Review of Knobloch, I. W. and D. S. Sessil ow and Fern Allies of
Chihuahua. Rhodora 64: 347-348

TRYON, R. 1962. A commentary o axpeetiudus names. Taxon 11: 116—120.

TRYON, R. 1963. Nomenclatural er a Taxon 12: 28-285.

TRYON, R. M. 1964. Evolution in the leaf of living ferns. Mem. Torrey Bot. Club 21: 73-85.

TRYON, R. 1964. erie ferns of Peru. Polypodiaceae (Dennstaedtieae to Oleandreae). Contr. Gray
Herb. 1

194: 1-—
TRYON, R. 1964. rite ie Fern Notes IV. Rhodora 66: 110—117.
PICHI-SERMOLLI, R. E. G., F. BALLARD, R. E. HOLTTuM, H. IT, F. M. JARRETT, A. C. JERMY, E. A. C.
L. E. SCHELPE, M.-L. TARDIEU-BLOT, and R. M. TRYON. 1965. Index Filicum Supplementum

Quartum pro Annis 1934-1960. Internat. Bureau for Plant Taxonomy and Nomenclature,
Utrecht, Netherlands.

TRYON, R. 1965. rig. te in the ferns. Taxon 14: 213-218.

TRYON, R. and D. M. BRITTON. 1966. A study of variation in the cytotypes of Dryopteris spinul
Rhodora 68: 59-92.

TRYON, R. 1967. Taxonomic fern notes V. New combinations in Peruvian species of Thelypteris.

odora 69: 5-
TRYON, R. 1967. paviine of Hara, H., The flora of eastern Himalaya. Rhodora 69: 456-457.
Boum, B. A. and R. M. TRYON. 1967. Phenolic compounds in ferns. I. A survey of some ferns for

c
TRYON, R. M. and A. F. TRYON. 1968. Edith Scamman kage 1967). Amer. Feri 5 58: 1-4.
TRYON, R. 1968. Nomenclatural proposals. Taxon 17: 588-590.
TRYON, R. 1969. Pteridology, Pp. 97-102 in J. Ewan (ed.) A short history of botany in the United
States. sone Publishing Co., New York.
TRYON, R. 1969. Taxonomic problems in the geography of North American ferns. BioScience 19:
iat

8 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 1 (2002)

paige R. 1970. aes of Correll, D. S. and M. C. Johnston., Manual of the vascular plants of
xas. Rhodora 72: 533-535.
Pines R. 1970. The le of the Cyatheaceae. Contr. Gray Herb. 200: 3-53.
TrYON, R. 1970. Development and evolution of fern floras of oceanic islands. Biotropica 2: 76-84.
Reprinted 1971 as Pp. 54-62 in W. L. Stern (ed.) Adaptive aspects of insular evolution.

Wash

TRYON, R. 1971. The American at ferns allied to i trata seeps ang 73: 1-19.

TRYON, R. 1971. Ferns of the Andes and Amazon. Morris Arbor. Bull. 2

TrYON, R. 1971. The process of evolutionary migration in species oe sae Brittonia 23: 89—
100.

TRYON, : 1972. Endemic areas and geographic speciation in tropical American ferns. Biotropica
4: 121-131.

TRYON, s 1972. Taxonomic fern notes, VI—New species of American Cyatheaceae. Rhodora 74:

TRYON, R. M., Jr. and A. F. TRYON. 1973. spac 7 - evolutionary relations in the

cheilanthoid ferns. Bot. J. Linn. Soc. 67: Suppl. 1:
TRYON, A. and R. TRYON. 1973. Thelypteris in se ae ee America. Amer. Fern J. 63: 65—
76

TRYON, R., B. VOELLER, A. TRYON, and R. RIBA. 1973. Fern biology in Mexico (a cl field program).
BioScience 23: 28-33.

TRYON, R., C.-J. WIDEN, A. HUHTIKANGAS, and M. LOUNASMAA. 1973. pony nana derivatives in
Dryopteris parallelogram and D. patula. Phytochemistry 12:

TRYON, A. and R. TRY 974. Ge ogaph — in temperate rence fant and some rela-
tionships in Galas Amer. Fern. J. 64: 99-104.

TRYON, R. 1975. The Benjamin D. G Gilbert fern nea Amer. Fern J. 65: 60.
TRYON, R. 7 and G. J. GAsToNny. 1975. The biogeography of endemism in the Cyatheaceae. Fern

Gaz 573-79.
TRYON, R. 4 and D. S. CONANT. 1975. The ferns of Brazilian Amazonia. re paiitias 5: 23-34.
TRYON, R. 1976. A oa of the genus Cyathea. Contr. Gray Herb. 206:
GASTONY, G. J. a . M. TRYON. 1976. Spore morphology in the Cyatheacoe Il. nae genera
Lophosoria, soe Sphaeropteris, Alsophila, and Nephelea. Amer. J. Bot. 63: 738-758.

aug R. M. 1977. Studies on the American Cyatheaceae and on the pie ees of on family.
Males. Bull. 30: 2839—

ete R M. and G. VITALE. 1977. dua for antheridogen production and its mediation of a
mating system in natural populations of fern gametophytes. Bot. J. Linn. Soc. 74: 243-249.

TRYON, R. 1979. Biogeography of the Antillean fern flora. Pp. 55-68 in D. Bramwell (ed.) Plants
and islands. Academic Press, New York.

TRYON, A. F., R. TRYON, and F. BADRE. 1980. Classification, spores, and nomenclature of the marsh
fern. Rhodora 82: 461-474

TRYON, R. 1980. Review of Sepensnck, J. A. and O. Huber, Flora of Avila. Amer. Fern J. 70: 79.

Tryon, R. M., JR. 1980. The ferns and fern allies of Minnesota, Ed. 2. Univ. Minnesota Press,
Minneapolis.

wer - ft A. TRYON. 1981. Taxonomic and nomenclatural notes on ferns. Rhodora 83: 133-—

wee - 1982. — “e a A. R., Pteridophytes in D. E. Breedlove (ed.) Flora of Chiapas
Part 2. Madrofio
TRYON, R. pane ALF. aly a Additional taxonomic and nomenclatural notes on ferns. Rho-

TRYON, R. M. and A. F. TRYON. 1982. Ferns and allied plants with special reference to tropical
America. Springer-Verlag, New

TRYON, R. 1984. An unusual new Bapheeyplesenctin from Peru. Amer. Fern J. 74: 108-110.

TRYON, R. 1985. Fern speciation and biogeography. Proc. Roy. Soc. Edinburgh, B. 86: 353-360.

TRYON, R. 1986. Dicksoniaceae. Lophosoriaceae. Metaxyaceae. Cyatheaceae. Pp. 1-59 in G. Harling
and L. Andersson (eds.) Flora of Ecuador, No. 27. Swedish Research Councils, Stockholm.

TRYON, R. 1986. Some new names and combinations in Pteridaceae. Amer. Fern J. 76: 184—186.

ROLLA MILTON TRYON, JR. 9

TRYON, R. nip — biogeography of species, with special reference to ferns. Bot. Rev. (Lancaster)
S2otks—

TRYON, R. i pee Pp. 327-338 in H. Lieth and M. J. A. Werger (eds.) Tropical rain
forest ecosystems. let Amster

TRYON, R. M. and R. G. STOLZE. 1989. Pied eae of Peru. Part I. 1. Ophioglossaceae—12. Cy-
eee. Fieldiana, Bot ‘New eries 20: 1-145.

Tryon, R. M. and R. G. STozE. 1989. Pteridophyta of Peru. Part II. 13. Pteridaceae—15. Dennstaed-

tiaceae. Fieldiana, Bot. New Series 22: 1-128.
TRYON, R. M., A. F. TRYON, and K. U. KRAMER. 1990. nbpigh Sues Pp: 230-256, in K. Kubitzki (ed.)
ee eee and genera a a plants. Vol. 1. oo and gymnosperms. Vol.

K. U. Kramer and P. S. Green. Springer-Verlag, a6
as RM and R. G. SToLzE. 1991. Pteridophyta of Peru. a IV. 17. Dryopteridaceae (with
collaboration of] T. Mickel and R. CG. Moran). Fieldiana, Bot. New Series 27: 1-176.
TrYON, R. M. and R. G. STOLZE. 1992. Pteridophyta of oF has Ill. 16. Thelypteridaceae (con-
tributed ‘ne A. R. Sensi Fieldiana, Bot. New Series 2
TRYON, R. M. and R. G. STOLZE. 1993. Pteridophyta of a < V1 ie ariaernepRi Poly-
podiaceae (with gepacieiers of B. Leén). Fieldiana, Bot. New Series —190.

TRYON, R. M. and R. G. STOLZ 1994. Pteridophyta of Peru. Part VI. 22. spe akan sir
(with collaboration of = \.H ckey and B. Ollgaard). ne apr Bot. New Series 34: 1—
TRYON, R. 1997. Systematic ee on Oleandra. Rhodora 9 -3

TRYON, R. 1997. Proposal to reject the name Acrostichurn a a (Pardaceas), Taxon 46: 339—
340

CHURCHILL, H., R. TrYON, and D. S. BARRINGTON. 1998. Development of the sorus in tree ferns:
Dicksoniaceae. Canad. J. Bot. ne 1245-1252.

TRYON, R. M. and A. F. TRYON. 1999. Observations on the phytogeography of eastern North Amer-
ican ferns. Pp 250-273 in X-C Zang and K-H Shing (eds.). Ching Memorial Volume. Institute
of Botany, Chinese Acad. Sci., Beijin

TRYON, R. 2000. Systematic notes on the Old World fern genus Oleandra. Rhodora 102: 428-438.

TRYON, R. M. 2001. Review of Hoshizaki, B. J. and R. C. Moran, Fern grower's manual, revised
and expanded edition. ego 103: 34

Winoiscu, P. G. and R. TRYON. 2001. The pane Ricardo Franco (State i “is Grosso, Brazil) as
probable migration ean and its present fern flora. Bradea 8: 267—

American Fern Journal 92(1):10—-19 (2002)

Crystals Associated with the Intertracheid Pit
Membrane of the Woody Fern Botrychium multifidum

ANGELA C. MORROW AND ROLAND R. DUTE
Department of Biological Sciences and Alabama Agricultural Experiment Station,
Auburn University, Auburn, Al 36849, USA

ABSTRACT.—CALCIUM-containing crystals have been found in the lumens of secondary tracheids
in the rhizome of the woody fern Botrychium multifidum. These crystals are styloids with rough,
pyramid- shaped ends. The crystals are usually single; however, conjoined or grouped crystals
were also found. Crystal formation apparently has no constant relation to the pit membrane, but
crystals of mature tracheids are often associated with the pit membrane or are located in the pit
areas. Crystals were also located between the helical thickenings of the lumen walls. No crystal
chamber or crystal sheath was found in association with the crystal body.

Crystals are a common feature in many plant tissues (Scurfield and Mitchell,
1973), and more than 1000 crystal producing woody plants, spanning 160 fam-
ilies, were described at the light microscopic level by Chattaway (1955, 1956).
Scanning electron microscopy has allowed for more rapid identification of
crystals in plant tissues and a clearer picture of their morphology (Scurfield
and Mitchell, 1973). Although crystals in xylem tissue have been reported in
the vessels of Intsia Thouars (Fabaceae; Hillis, 1996), Torreya yunnanesis C.Y,
Cheng & L.K. Fu (Taxaceae; Kondo et al., 1996), and Polyalthia Blume (An-
nonaceae; Scurfield and Mitchell, 1973), they are most commonly found in
the xylem parenchyma, septate fibers, or vessel tyloses (Scurfield and Mitchell,
1973).

The formation of crystals by Botrychium, the only extant fern that produces
wood (Gifford and Foster, 1989), has not been previously reported. During our
studies of the torus-bearing pit membrane in the tracheid of Botrychium mul-
tifidum (S.G. Gmelin) Rupr., we discovered occasional instances of crystals
associated with the pit membrane. This paper describes the morphology of
these crystals as observed with SEM. The discoverey of these very small crys-
tals was unexpected, and our exploration of them to this point has been strictly
descriptive. However, in our discussion we explore several possible reasons
for crystal formation in this wood.

MATERIALS AND METHODS

Rhizome samples of upright or orthotropous rhizomes of Botrychium mul-
tifidum. were collected by Dr. D. W. Stevenson (New York Botanical Garden,
New York City, U.S.A.) from Plumas County, California and fixed in FPA. The
collection site (elevation 2000 m) is rocky mountain soil at the edge of a mead-
ow and the ground is frozen for much of the year. The samples were typical
rhizomes selected as random samples and representative of the population. In

MORROW AND DUTE: CRYSTALS IN BOTRYCHIUM MULTIFIDUM 11

our lab, samples were cut transversely into 1-2 mm pieces that were placed
into 50% ethanol and then dehydrated through a graded alcohol series. Sam-
ples then were cut into small wedges, placed into hexamethyldisilazane
(HMDS) for 2 hours (Nation, 1983), and subsequently placed under a chemical
hood overnight to dry. Dry samples were attached to aluminum stubs with
double-sided sticky tape and coated with gold-palladium. For comparison pur-
poses, samples of Botrychium dissectum Sprengel and B. virginianum (L.)
Swartz from Lee County, Alabama were prepared in the same manner as
B.multifidum. Specimens were viewed with a Zeiss DSM 940 at 5,10, or 15
kV. Qualitative element identification was performed using energy dispersive
spectroscopy (Tracor Northern Micro Z II) coupled to the SEM.

RESULTS

Secondary xylem tracheids of Botrychium multifidum contain helical wall
thickenings and intertracheid circular bordered pits (Fig. 1). Thickenings, as
seen in longitudinal section, are uniform neither in height nor in distance
between gyres, and thickenings are sometimes branched (Fig. 1). The pit mem-
brane is almost always differentiated into a torus and margo (Fig. 2). Microfi-
brils of the pit membrane are loosely woven in the margo region, but tightly
woven in the torus. Tearing of the pit membrane was sometimes evident in
the margo (Fig. 2). Crystals were found in association with torus-bearing pit
membranes of tracheids (Fig. 11), as well as in tracheid lumen (Figs. 1, 3).
These crystals were not apparent at the light level. Crystals associated with
these tracheids are styloids (Frey-Wyssling, 1981; Carlquist, 1988); they are
rectangular columnar with pyramidal ends. Intact crystals have columns that
are four-sided and are smooth-surfaced. The pyramidal crystal ends consist of
four equilateral triangles, although wedge-shaped ends also were observed
(Fig. 4). Crystal ends, when visible, typically appeared to be rough (Fig. 3),
although some crystals with smooth ends were observed (Fig. 4). Crystals
ranged in size from 4.3 to 12 wm in length, and 1.14 to 2.4 pm in width (N =
12). The mean crystal length is 7.27 »m, mean width is 1.55 pm, and mean
ration of width-to-length is 1: 4.7. The crystals were not always regular in
shape and sometimes appeared to have their growth modified by the presence
of a helical wall thickening (Fig. 3). By energy dispersive spectroscopy (EDS),
these crystals were found to be composed of a calcium compound, most likely
calcium oxalate (Fig. 13).

Although single isolated crystals were most commonly found, joined double
crystals with U-shaped conjoined end were also observed (Fig. 4). Crystals in
groups of two or more were also encountered and had either parallel or per-
pendicular orientation to each other (Figs. 1,5). Crystals were found in various
positions within the tracheary lumen. They were located between either the
helical thickenings of the wall material, laying flat on the inner cell wall, or
projecting out from the pit membrane (Fig. 1).

It was clear from some specimens that the crystals were composite structures
(Figs. 4, 6-8). In some instances the subunits resembled raphides (Figs. 6,7),

12 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 1 (2002)

Fics. 14. SEM micrographs of intertracheary pit membranes and crystals. 1) Tracheary lumen
with helical wall thickenings (W), crystals (arrow), pit aperture (A), and circular bordered pit
membrane (P); scale bar = 5 um. 2) Intertracheary pit membrane with torus (T) a margo (M).

e pit border was removed when the wood was split during preparation; scale = 2 pm. 3)
Crystal entering a pit aperture (A). Note how = crystal appears to have grown tsa the helical
thickening to the right (double arrow). R = rough end of crystal; scale bar = 2 um. 4) Double
crystal joined at one end (arrow); scale bar = m.

MORROW AND DUTE: CRYSTALS IN BOTRYCHIUM MULTIFIDUM

FIGs

SEM micrographs of crystals in which crystals revea

Multiple crystals with parallel orientation and perpendicular orientation. Note subunits
crystal (arrow); scale bar = 5 pm. 6) E

| their subunit composition. 5)

s in broken
view of composite crystal formed by smaller raphide
shaped crystals (arrow); scale bar = 500 nm. 7) Composite crystal w ith styloid (arrow) and raphide
crystal (double arrow) shaped subunite: scale bar = 2 pm. 8) Composite crystal with styloid crystal
subunits (arrow); scale bar = 2 pm.

14 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 1 (2002)

whereas in others they resembled small styloid crystals that were fused to form
one large crystal (Figs. 7, 8). Both types of subunits appear to integrate into
one another (Fig. 7). One shattered example had a hollow center (Fig. 9).

Crystals were observed to traverse the pit aperture (Figs. 3, 10) and contact
the pit membrane (Figs. 11, 12). These did not appear to penetrate the pit
membrane, but we are uncertain of this point due to the poor preservation of
the pit membranes in our samples. Fig. 12 demonstrates a unique occurrence
in which a pit membrane is approached by a crystal from either side. Due to
the position of the crystal relative to the pit membrane, we were unable to
confirm the presence of a torus in each crystal-associated pit membrane; how-
ever a torus was present in the samples that exhibited a crystal behind the pit
membrane (fig. 11).

No noticeable chamber or crystal sheath was ever observed in association
with a crystal. No evidence of a surrounding membrane was discovered, al-
though the ends of the crystals often were rough (Fig. 3). Efforts to examine
crystals with TEM to determine the presence or absence of a chamber or crystal
sheath were not successful. Crystals were not observed in either Botrychium
dissectum or B. virginianum.

DISCUSSION

Three major systems of mineralization occur in plants. These include silic-
ification, calcium carbonate crystallization, and calcium oxalate crystallization
(Grimson et al., 1982). Calcium oxalate crystals, either in the monohydrate or
polyhydrate state, are the most common mineral deposits (Webb and Arnott,
1982). EDS evidence indicates that our crystals contain calcium. The bipyra-
midal shape of the crystals’ ends, and the rectangular columns, suggest that
they are crystals of calcium oxalate in the polyhydrate form (Frey-Wyssling,
1981). Usually, acid solubility tests are used to confirm crystal composition in
plants (Webb and Arnott, 1982). In addition, the oxalate nature of a calcium
crystal can be tested with cupric acetate and ferric sulphate (Deshpande and
Vishwakarma, 1992). However, due to the small size and sparse number of
crystals found in Botrychium multifidum, these tests were not performed.

The location of crystals in tracheid lumens in unusual. Crystals in wood are
most frequently found in ray or axial parenchyma cells (Chattaway, 1955,
1956), although they may also be found in septate fibers, vessel tyloses, and
even in vascular cambia (Deshpande and Vishwakarma, 1992). In Polyalthia,
vessels contained a crystalline mass (Scurfield and Mitchell, 1973). In the cur-
rent study, crystals were isolated within the tracheary lumen, and there was
no evidence suggesting attachment to cell walls. Some crystals appeared to be
touching, but were not attached to, the pit membrane. Crystals also were found
that had no apparent association with a pit membrane. Therefore, it appeared
that crystal formation was not directly related to pit membranes.

Crystals in plants often are formed in membrane-bound compartments with-
in the vacuole (Arnott and Pautard, 1970; Franceschi, 1984: Webb et al., 1995).
As proposed by Arnott and Pautard (1970), the cell membrane may control

MORROW AND DUTE: CRYSTALS IN BOTRYCHIUM MULTIFIDUM

cv

Fics. 9-12. SEM micrographs of crystals in association with pit area and pit membranes. 9
Fractured crystal with hollow center (arrow). Note the aperture behind the crystal (double arrow);
scale bar = 5 jm. 10) Crystal entering a pit aperture. Note shaping of the crystal around helical
thickening (arrow); scale bar = 2 1) Crystal behind pit membrane; scale bar = 2 pm. 12) Pit
membrane associated with crystals from contiguous tracheids; A = aperture; C = crystal; P = pit

membrane; scale bar = 2 pr

16

A. Experimental

AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 1 (2002)

PEAK LISTING
RGY AREA

cp wh

EL. AND LINE -
KA

B. Control
I

PEAK LISTING
RGY AREA

OU bh oN

EL. AND LINE a
I KA

7

Pic. 13.

EDS of tracheid. A. Spectral tracing of crystal within a tracheid. The calcium compone

nt
of the spe ectrum is conspicuous and is indicated by the peak labeled CA. B. EDS of tracheid without

represents calcium in the middle lamella. AU

= palladium; SI =

MORROW AND DUTE: CRYSTALS IN BOTRYCHIUM MULTIFIDUM 17

both shape and growth of crystals. There was no direct evidence that crystals
of B. multifidum were once enclosed in a membrane, but Scurfield and Mitch-
ell (1973) suggest that a rough area on a crystal is indicative of the adhering
remnants of membrane. In crystals of B. multifidum only the membrane’s im-
pression on a crystal would be evident because living portions of the tracheid
has undergone autolysis and no membrane remains. If the vacuole with its
membrane-covered crystal pressed against either the cell wall or a cell wall
thickening as a crystal formed, this contact could explain the shape of these
crystals.

Water flow though the xylem could also deposit crystals (no longer enclosed
by cytoplasm) randomly throughout a tracheid, including on top of a pit mem-
brane or between wall thickenings. Due to erosion, water flow might also
change crystal shape.

Another aspect of crystal development in plant cells in isolation of a crystal
by wall material or a suberized sheath after the crystal has formed within a
vacuole. This process would, in essence, externalize the crystal (Frank and
Jensen, 1970). In Agave (Agavaceae), crystals are produced in such extraplasm-
ic compartments (Wattendorff, 1976a, b). Wattendorff (1976b) found that all
styloid idioblast of Agave, where they did not touch the wall, were surrounded
by a suberized sheath. Although crystals of B. multifidum are styloids, they
appear neither to be associated with a sheath of any sort nor to be isolated by
cell wall material.

The reason a cell forms a crystal is not well understood. Crystal formation
may represent a crystallization of waste material or storage of minerals (Desh-
pande and Vishwakarma, 1992). Crystal formation also may be associated with
ionic balance, and therefore, the formation of a crystal could be a form of
osmoregulation (Franceschi and Horner, 1980). Franceschi and Horner (1979)
correlated the amount of calcium in the growth medium and the number of
crystals formed in Psychotria L. (Rubiaceae) callus. Lane (1994) has suggested
that calcium oxalate crystals may promote the polymerization of lignin which
of course, would be occurring in the developing tracheids of B. multifidum.

It is evident at times that crystal formation in plants is under genetic control
(Frey-Wyssling, 1981; Webb, 1999); however, genetic control of the formation
of all crystals has not been proven. The cell in which a crystal is produced
undergoes many changes at macro, micro, and ultrastructural levels, as well
as, changes in cell chemistry. These changes, documented in other plant taxa
during crystal formation, make it unlikely that crystal formation could be sim-
ply the result of precipitation or crystallization (Franceschi and Horner, 1980),
although crystal formation may represent crystallization of waste material or
storage of minerals (Deshpande and Vishwakarma, 1992). Deshpande and Vish-
wakarma (1992) also identified seasonal fluctuation in crystal formation after
the cessation of cambial activity. Gourley and Grime (1994) described crystals
that were more commonly found in the late wood of Acacia Mill.(Fabaceae).
The availability of water was also determined to be a factor in crystal formation
(Gourley and Grime, 1994).

It was impossible to determine for certain whether crystals in the tracheids

18 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 1 (2002)

of B. multifidum formed before or after cell death. Perhaps due to greater water
flow resistance occurring at the pit membrane, there would have been a greater
chance for calcium precipitation in the pit area rather than in the tracheary
lumen. If this were the case, crystals could at the pit membrane form after the
death of a tracheid. However, the rough ends observed on some crystals sug-
gest they may have been enclosed at one time by a membrane. Additionally,
the crystals appear to conform to the shape of the pit aperture or cell wall
thickenings and do not appear to have been randomly distributed by water
flow. Perhaps the best explanation of where these crystals develop is in mem-
brane compartment within vacuoles of living tracheids. The enlargement of a
crystal in a plane perpendicular to the cell’s axis would result in its abutting
a wall or pit membrane, thus influencing crystal shape. Crystals that elongated
parallel to a cell’s axis would not encounter these boundaries and would not
be shaped by them. Crystals that were not pressed into a cell wall or pit mem-
brane also would not develop this shaping and might settle between wall
thickenings, or after cell death, move with the xylem water stream. However,
the positions of crystals in Figures 3, 10, and 12 with respect to the pit mem-
brane suggest that crystal position is not the result of water flow.

Crystal formation has also been associated with the products of fungal me-
tabolism within the plant cells (Scurfield and Mitchell, 1973). In our study, no
fungal hyphae were found near any of the crystals. Therefore, this possibility
in B. multifidum seems unlikely.

If crystal manufacture is under genetic control, what advantage does the cell
gain from its production? This is an especially intriguing question with regards
to Botrychium as crystal production would be occurring in cells about to die.
Lane (1994) has suggested that calcium oxalate crystals play a role in lignin
polymerization and perhaps this may be true in these lignified tracheids. How-
ever, the lack of crystals in other Botrychium species indicates that this would
be true only under certain environmental conditions. As previously men-
tioned, some authors believe that crystals represent the storage of calcium that
could be either reserve calcium or waste calcium (Deshpande and Vishwak-
arma, 1992; Webb, 1999). Storage of needed calcium in a near-death cell would
be unlikely. However, these crystals may have been produced by a cell for
ionic balance or osmoregulation. Ionic balance and osmoregulation are critical
for immature cells (Franceschi and Horner, 1980). If a plant were growing on
soil with high nitrate levels, assimilation of this compound would increase
cell pH, and oxalic acid might be produced to counter this effect. The oxalate
anion could then react with calcium to form a crystal that would remove the
excess anion from cell sap (Franceschi and Horner, 1980).

Another explanation could be protection from herbivores, although crystal
production in a leaf cell would be more plausible for defense purposes. The
crystals in B. multifidum are too small and too few in number for this type of
protection. Fire protection was listed as an explanation by Gourley and Grime
(1994) for crystals in Acacia, but again this is unlikely for a rhizome.

Based on our data the best explanation for crystal formation in the xylem of
B. multifidum is that crystals are the result of excess calcium precipitation,

MORROW AND DUTE: CRYSTALS IN BOTRYCHIUM MULTIFIDUM 19

which could represent either waste, storage, or osmoregulation in the plant.
Because the crystals are located in dead cells, active resolubilization by a cell
would be unlikely; however, if crystals were dissolved by water flow in the
xylem, their minerals could be carried in the transpiration stream. Deshpande
and Vishwakarma (1992) have suggested that formation of calcium crystals

may be a reversible process in some tissues. Therefore, these crystals do not
necessarily represent a calcium loss for the plant.

LITERATURE CITED

ARNOTT, H. J., and F. G. E. PAUTARD. 1970. Calcification in plants. Pp 375-446, in maida
saledicatian: Cellular and molecular aspects. H. Schraer, ed. North-Holland, Amsterdam

CHaTTaway, M. M. 1955. Crystals in woody tissues. Part I. Trop. Woods 102:55—74,
———. 1956. Crystals in woody tissues. Part II. Trop. Woods 104:100—24.
DESHPANDE, B. P., and A. K. VISHWAKARMA. 1992. Calcium beam a in the fusiform cells
of the pli es of Gmelina arborea. [AWA Bull. N.S..
isa V. R. 1984. Developmental — : calcium ee an sand depositions in
vulgaris L. leaves. Protoplasma 12
H. T. HORNER. 1979. Use of aera eis callus in study of calcium oxalate
crystal idioblast formation. Z. Pflanzenphysiol. 92:1-75.
1980. Calcium oxalate crystals in plants. Bot. Rev. (Lancaster) 46:361—42
FRANK, E. and W. A. ae 1970. On the formation of the pattern of crystal idioblast in ade
ensiformis D.C. IV. The fine structure of the crystal cells. Planta 95:202—217.
FREY-WYSSLING, A. 1981. Crystallography of the two hydrates of crystalline calcium oxalate in
plants. Amer. J. Bot. 68:130-141
GIFFORD, E. M., and A. S. FOSTER. 1989. oe and evolution of vascular plants. 3rd ed.,
eeu and Co., Salt Lake C
Gour.ay, L. D., and G. W. GRIME. an. Calc cium oxalate . in african Acacia isa and
their says by scanning proton a (SPM IAWA Bull. N.S. 15:137-1
ee M. J., H. J. ARNOTT, and M. A. WEBB. 1982. A scanning electron eo ins of
71133-1140.

winged point — in the bean legume. Scanning Electron Microscopy III
Hits, W. E. . Formation of robinetin crystals in vessels of Intsia species. ore ng vie N.S.
x reaione
Konpo, Y., T. Fuyl, ¥. HAYASHI, and A. Kato. 1996. Organic crystals in the tracheids of Torreya

yunnanenss IAWA Bull. N.S. 17:393—403.

Lang, B. G. 1994. Oxalate, germin, and the extracellular matrix of higher plants. F.A.S.E.B.J. 8:
294-301.

NATION, J. ca 1983. A new method using sre ea ca for preparation of soft insect
tissues for scanning electron microscopy. Stain Technology

ScurFIELD, G., and A. J. MITCHELL. 1973. Crystals in woody ely Bot. J. Linn. Soc. 66:277—289.

WarTTENDoREF, J. 1976a. A third type of raphide crystal in the plant kingdom: six-sided raphides

i 4a

ioe animate of the suberized svioid crystal cells in pam leaves. Planta (Berl.)
ee pier

Wess, M. A. see Cell-mediated crystalliztion of calcium oxalate in plants. Pl.Cell 11:751—761.

, and H. J. ARNOTT. 1982. A survey of calcium oxalate crystals and other mineral inclusions

in seeds. eres Electron Microscopy III:1109-1131

ALETTO, N. C. Carpira, L. E. Lopez, and H. J. ARNOTT. 1995. The intravaculolar

oat fae associated with calcium oxalate crystals in leaves of Vitis. Plant J. 7:633-648

American Fern Journal 92(1):20—22 (2002)

A New Filmy Fern from the Dominican Republic

CARLOS SANCHEZ
Jardin Botaénico Nacional, Carretera del Rocio, km 3.5, Calabazar, Boyeros,
C. P. 19230 Ciudad de la Habana, Cuba

ABSTRACT.—A new species of Hymenophyllum subg. Hymenophyllum with entire involucral
valves is described from the Dominican Republic on the island of Hispaniola.

During the prepartion of a revision of the filmy ferns (Hymenophyllaceae)
for the Flora of the Greater Antilles project, a peculiar species was discovered
among the undetermined specimens in the Gray Herbarium.

The taxon belongs to Hymenophyllum subg. Hymenophyllum, following
Morton (1968, pp. 162-164), a subgenus represented by only two species in
the Antilles. Hymenophyllum tunbrigense (L.) Smith is known only from Ja-
maica and Hispaniola (Proctor 1985, p. 90), whereas H. fucoides (Sw.) Sw., is
more widely distributed (Proctor, 1985, p. 92; 1989, p. 58). The members of
this subgenus are characterized by having toothed segment margins, and most
have sinuous to toothed involucral valves as well. The new species is de-
scribed as

Hymenophyllum integrivalvatum C. Sanchez sp. nov. Fig. 1

Ab speciebus aliis antillanis subgeneris Hymenophylli valvis integris, stip-
itibus brevissimis, segmentis pinnarum paucis (1 vel 2), necnon laminis gla-
berrimis diversa.

TypE—Dominican Republic: Pcia. La Vega: Near the pyramid ca. 13 km from Valle Nuevo on
the road to San José de Ocoa, ca. 2500 m elev., 22 August 1957, Gastony, Jones & Norris 740 (GH;
isotype US).

Rhizomes creeping, filiform, 0.1-0.3 mm in diam., clothed with deciduous,
brownish, pluricellular trichomes, with a few conspicuous, straight roots ca.
5 mm distant. Fronds small, erect, determinate, approximate, 1.15—2.1 cm
long.; stipes 0.1-0.3 mm long., 0.2 mm in diam., very narrowly alate thorough-
out, dark brown, glabrous or with a few brownish, often 2-celled trichomes;
laminae narrowly ovate, lanceolate or oblong 1.4-1.8 cm long, X 0.8-1 cm
wide, pinnate-pinnatifid; rachises notably flexuous, narrowly and evenly alate,
the alae less than 0.1 mm wide, dark brown, glabrous; pinnae 5-8 pairs,
spreading to ascending, mostly with 2 acroscopic segments; segments narrowly
elliptic, oblong, or linear-oblong, 1.2-1.8 mm wide, glabrous, the margins dis-
tantly toothed, the teeth usually more distant than their length and ascending,
the midvein dark brown, the lamina tissue olivaceous-green when dry; sori
conspicuous in size in comparison with the length of the lamina, borne at the
lamina apex or in the distal half, subaxillary on the acroscopic side of the

SANCHEZ: A NEW FILMY FERN FROM THE DOMINICAN REPUBLIC 21

B22,

4. Cale ry
Se
Fic. 1. Holotype of Hymenophyllum integrivalvatum (GH). A. Habit of plant. B. Detail of an
ultimate segment. C. Detail of a sorus.

pinnae; involucres 1.6—2.2 mm long, 1.4 mm wide, broadly elliptic or broadly
ovate, bivalvate, the valves wider than the sterile segments, the margin entire,
the filiform receptacle included.

DISTRIBUTION.—Endemic to Hispaniola (Dominican Republic), known only
from the type collection.

Hasirat.—Epipetric in very moist burned and timbered pinelands, forming
thick mats on rocks along streams in very moist ravines, according to the in-
formation on the label.

The new species is most closely related to H. tunbrigense (L.) J. E. Smith,
which is also known from a few collections from Hispaniola. Hymenophyllum
tunbrigense differs in having larger fronds, straight rachises with wider alae,
more divided pinnae, narrower segments, and sinuous involucres. The entire
involucral valves, very small fronds, and the absence of trichomes on the re-
mainder of the lamina separate the new species from all other Antillean spe-
cies of subg. Hymenophyllum.

ACKNOWLEDGMENTS

I am indebted to the following individuals for their help and hospitality during my visit to
different North American herbaria to study Greater Antillean pteridophytes: John T. Mickel and
Thomas Zanoni (New York Botanical Garden), David B. Lellinger (United States National Herbar-
ium), and Emily Wood and David E. Boufford (Gray Herbarium, Harvard University). I am grateful
to the New York Botanical Garden and The John D. & Catherine T. MacArthur Foundation for
financial support. Thanks also to David B. Lellinger for revising the English version of the man-
uscript. I also wish to thank Manuel G. Caluff for the illustrations.

22 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 1 (2002)

LITERATURE CITED

Morton, C. V. 1968. The Genera, Subgenera, and Sections of the Hymenophyllaceae. Contr. U.
S. Natl. Herb. 38:153-214.

Proctor, G. R. 1985. Ferns of Jamaica: A guide to the pteridophytes. British Museum (Natural
History), London,

. 1989. Ferns of Puerto Rico and the Virgin Islands. Mem. New. York Bot. Gard. 53: 1-389.

American Fern Journal 92(1):23—29 (2002)

Adiantum argutum, an Unrecognized Species of the
A. latifolium Group

JEFFERSON PRADO
Secdo de Briologia e Pteridologia, Instituto de Botanica, Caixa Postal 4005, 01061-970
Sao Paulo, SP, Brasil
Davip B. LELLINGER
Department of Botany, National Museum of Natural History, Smithsonian Institution,
Washington, DC 20560-0166

ABSTRACT.—The present paper distinguishes A. argutum, an unrecognized but widespread species
from South America, from the related A. Jatifolium, and designates a lectotype for A. argutum.

Several pinnate or bipinnate Adiantum species have an indument like that
of the A. serratodentatum group, but differ in having fewer, larger, less di-
midiate pinnules and thin, very long-creeping rhizomes. Among the species
of this group are A. argutum Splitg., A. incertum Lindm., which is based on
Lindman Regnell Exped. I. A2083 (S not seen; isotypes B, GH) from Paraguay,
the widespread A. Jatifolium Lam., and A. viviesii Proctor, which is based on
Proctor 41389 (US; isotypes IJ, SJ) from Puerto Rico. Adiantum glaziovii Baker,
which is based on Glaziou 13345 (K, isotype US) from Rio de Janeiro, Brazil,
is a synonym of A. Jatifolium. The early, unplaced name A. elatum Desv.,
which is based on a Brazilian specimen from the Herb. Dombey (P-Herb. Juss.
Cat. 1421 not seen Morton photo 3153) will likely displace one of the later
named species found in Brazil. The specimen was said by Morton to have
almost glabrous segments; it needs to be examined critically.

Adiantum argutum Splitgerber, Tijdschr. voor Natuurl. Gesch. en Physiol.
[Leiden] 7: 427. 1840. Figs. 1, 2.

Lectotype (chosen here): “in sylvis montosis Surin. prope Bleauwe Berg,”
Surinam, May 1838, Splitgerber 891 (L not seen, photo US). Other syntype:
idem, id., Splitgerber 290 (L not seen, photo US).

Adiantum fovearum Raddi var. reductum Jenm. Ferns Br. W. Ind. & Guiana
87. 1899. TYPE: Not clearly stated, but presumably Guyana, Jenman (NY? not
seen).

Splitgerber (1840) described A. argutum based on material he collected in
Surinam. The author did not mention any specimens in the protologue, only
the locality. Original materials were found by Morton at Leiden (Morton photos
193 and 194, both US), and they may be considered the syntypes of Splitger-
ber’s taxon. The lectotype selected here (Splitberger 891) has an original label,
handwritten by the author with the same information he published with the

24 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 1 (2002)

Ss °
J gv -
is °
\ 10
o 2 ¢
3 o “Sa ‘4 e e
o —— a :
° e t eveinaph Ottis Pl
° en Me ind
<— aa yy. ‘, ane =
Be = é bad e
\ a \
° r bane 0
,
° a °
f
eS @

Fic. 1. Range of Adiantum argutum Splitg.

original description. The photograph of the other syntype (Splitgerber 290)
resembles A. Jatifolium Lam., although we can not place it there with certainty.

According to Splitgerber (1840), A. argutum has a long-creeping rhizome;
the laminae are lustrous adaxially and have 3 or 4 pinna pairs, acuminate
pinnules, a subrhombic terminal pinnule, reduced and flabellate basal pin-
nules, sparse, minute setiform scales abaxially, and oblong sori. In fact, these
characters distinguish this species from all others closely related to it. Other
important features to recognize A. argutum are the distant fronds and the id-
ioblasts on the abaxial surface of the pinnules.

Unfortunately, over the years, the Splitgerber species was included within
the concept of A. Jatifolium by Vareschi (1969, p. 734), Kramer (1978, p. 91),
Tryon and Stolze (1989, p. 66), and Smith (1995, p. 259). In other cases, the

PRADO AND LELLINGER: ADIANTUM ARGUTUM 25

name was synonymized under this species, by Posthumus (1928, p. 105), Lel-
linger (1989, p. 148), and Cremers and Hoff (1990, p. 19).

Adiantum argutum demonstrates its close affinity to A. Jatifolium mainly in
its slender, long-creeping rhizome, 2-pinnate fronds, and stipe and rachis cov-
ered by deltate to lanceolate scales with a pectinate base. However, A. latifol-
ium differs by its smaller, obtuse to subacute pinnules with a roundish apex
that are abaxially glaucous, glabrous, and without idioblasts.

Adiantum incertum Lindm. differs from A. argutum and A. Jatifolium in
having the scales on the abaxial surface of the pinnules hairlike and with a
few basal processes, rather than having such scales with a pectinate base or
totally lacking scales. In addition, it is a species restricted to Paraguay and
extra-Amazonian Brazil: Goids (Maurilandia, Rio dos Bois, Hatschbach 34271,
MBM, MO, NY, UC), Mato Grosso (Santa Terezinha, 21 km SW of Portal da
Amazonia, Thomas et al. 4334, NY, US), SAo Paulo (Sao Carlos, 9 km NNE of
the BR Station at Santa Eud6xia, Eiten & Eiten 3488, US), and Parana (Foz do
Iguacu, Parque Nacional das Cataratas, Hatschbach 23171, HB, MBM, MO, UC,
UPCB).

Adiantum obliquum Willd., although its fronds look much like those of A.
argutum, can be distinguished by its short-creeping rhizomes, approximate
and usually 1-pinnate fronds, and pinnules with conspicuous idioblasts on
both surfaces. It may be more related to A. Jucidum Cav. and perhaps to A.
petiolatum Desv., with which it hybridizes. These three species may form a
separate group.

Adiantum argutum has a more restricted area of distribution than its closest
relative A. Jatifolium. It occurs in northern South America (Colombia to French
Guiana) and in the Amazonian regions of Peru, Bolivia, and Brazil. It grows
in primary and secondary forests, on dark red lateritic clay soils, from 50 to
1000 m elevation.

Representative specimens of A. argutum studied:

COLOMBIA: Meta: Sierra de La Macarena, Cajfio Entrada, Philipson & Idrobo
1748 (US); Villavicencio, Pennell 1607 (GH). Boyaca: Los Llanos, Haught 2833
and 2844 (both GH). Vichada: San José de Ocumé: near Rio Vichada at Botomi,
ca. 14 km NW, Hermann 11107 (US); NE de Pto. Infrida, 3°58’N, 67°50’W,
Churchill et al. 17748 (NY).

VENEZUELA: Bolivar: La Tomasa, Williams 1295 (US); Rio Paragua, Isla El
Casabe, Killip 37301 (US); Salto Alta, Alto Orinoco, Croizat 486 (NY); Dtto.
Sifontes, Concesién Minera Oro Uno, 7 km NW of la Clarita, 6°13'N, 61°27'W,
Aymard et al. 3976 (NY); Sierra Imataca betw Rio La Reforma and Puerto Rico,
N of El Palmar, Steyermark 88012 (US). Amazonas: Around the margin of the
Rio Orinoco above Tamatama, Williams 15199 (GH); Cuenca del Rio Manapi-
are, 5°5’N, 66°03’W, Huber 435 (NY). Delta Amacuro: Rio Cuyubini, Cerro de
la Paloma, Steyermark 87649 (NY). Mérida: Near border Rio Grande de Toro,
61°44’W, 80°4'N, Breteler 3781 (US).

TRINIDAD: Fendler 2 (NY).

GUYANA: Cuyuni-Mazaruni: 8 km N of Bartica on W bank of Essequibo
River, 06°29’N, 58°38’W, Henkel & Chin 297 (US). U. Takutu-U. Essequibo:

i io”
” a INS

PRADO AND LELLINGER: ADIANTUM ARGUTUM 27

Rupununi area, Surama Village, 04°08’N, 59°04’W, Acevedo et al. P3297 (US);
Marudi River, 02°11’N, 59°11'W, Henkel et al. 2902 (NY), 3032 (US); Kuyuwini
River, 02°11'N, 59°11’W, Henkel et al. 3022 (US); NW Kanuku Mts., 3°21'N,
59°30'W, Hoffman & Foster 3510 (US); Rupununi River, Jansen-Jacobs et al.
4207 (US). Potaro-Siparuni: On 0.5 km island in Essequibo River, 1 km S of
Fairview, 4°40'N, 58°40'W, McDowell 3371 (US); Iwokrama Mts., Annai-Ka-
rupukari Rd., 04°19’N, 58°51’W, Hoffman et al. 1409 (US); River Isherton,
2°20'N, Smith 2432 (GH, NY). Barima-Waini: Head of Barima River, Ayamba
Falls, 4.5 mi W of Eclipse Falls, ca. 10 km W of Arakaka, 7°39’N, 60°09’, Pipoly
& Lall 8200 (NY); Head of Barima River, NW of Kariako River, 7°30'N, 60°35’W,
McDowell 4393 (NY); Labbakaka Creek, Tiger Creek, Sandwith 1209 (K, NY).

SURINAM: Haut Litany: Basin du Litany, 2°31'N, 54°45’W, Granville et al.
12040 (US). Nickerie: Area of Kabalebo Dam project, Lindeman & Roon 884
(US); Area of Kabalebo Dam project, 4°—5°N, 57°30’—-58°W, Lindeman et al. 165
and 343 (both NY); Sectie O, along railroad, vic. Km. 70, Maguire & Stahel
23605 (GH, NY). Brokopondo: 2.4 km S of village Gansee, Donselaar 1189 and
1276 (both GH); Zuid River, 3°10’—3°20’N, 56°29’-56°49’W, Kayser Airstrip, 45
km above the confluence with Lucie River, Irwin et al. 57697 (NY).

FRENCH GUIANA: Cayenne, Inini River, 3°28’N, 52°36'30’"W, Cremers et al.
8781 (US); Camp Eugene, Basin du Sinnamary, 4°51’S, 53°4'W, Cremers &
Granville 13727 (NY); Gobaya Soula, Basin du Maroni, 53°58’W, 3°37’S, Cre-
mers et al. 10125 (US); Saoul, 3°37’N, 53°12’W and vicinity, Route de Bélizon,
N of Eaux Claires, Heald & Yahr 56 and 65 (both NY); Comté., degrad auprés
de Crique Martineau, Oldeman 1426 (NY); Mt. Balbao, Secteur Sud, 3°35’N,
53°20'W, Granville et al. 8958 (NY).

PERU: Madre de Dios: Near the confluence of Rio Tambopata and Rio La
Torre, 39 km SW of Puerto Maldonado, 12°50’S, 69°20’W, Smith & Condor
1114 (US) and 1363 (NY); Tambopata, Vargas 18577 (GH); Tambopata, vic of
Moho towards Piedra Redonda, at the Bolivian frontier, 12°30'S, 69°40’W, Nu-
fiez et al. 9695 (GH, NY); Tambopata, SSW of Pto. Maldonado at the confluence
of the R. La Torre and the R. Tambopata (SE bank), Tambopata Nature Reserve,
12°49’S, 69°17'W, Barbour 4763 (NY), Loépez 4585 (GH).

BOLIVIA: Beni: Pcia. Ballivian: Rio Colorado, Collegio Técnico Agropecu-
ario de Rfo Colorado, 15°00’S, 67°10'W, Fay & Fay 2105, 2640, 2652, 2654,
2681 (all US); 18.4 Km E of Riberalta, then 1 km NE on old road to Cachuela
Esperanza, 11°05’S, 65°50’W, Solomon 7804 (NY); Isla Capanario, 50 m from
the San Borja—San Ignacio de Moxos road, 212 km from Campoamento Totai-
zal, Rolleri 140 (NY); Pcia. Moxos: Chimanes Forest, 15°10’S, 66°37’W, Fay &
Fay 2794 (US). Sta. Cruz: Bella Vista, Rio Blanco, Scolnik & Luti 681 (US);
Pcia. Ichilo: Old meander loop of the Rio Ichilo, 1-1.5 km SW of the Buena
Vista—-Villa Tunari Hwy., 17°18'S, 64°12’W, Nee & Moran 45225 (NY). Pando:
Nicolas Suarez, SW of Cobija on the Rio Naraueda, 11°08’S, 69°08’W, Sperling

oo

Fic. 2. Adiantum argutum Splitg. Fig. 2A. Habit. Fig. 2B. Abaxial surface of some pinnules. Fig.
2C. Abaxial surface of a pinnule showing the scales.

28 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 1 (2002)

& King 6475 (GH, NY, UEC); Ca. 20 km from Cobija towards Castro Erifia, Casas
& Sussana 8123 (NY); W bank of the R. Madeira betw Cachoeiras Madeira and
Misericordia, Prance et al. 6612 (NY). La Paz: Pcia. Iturralde, Siete Cielos, R.
Manupare, 12°27'S, 67°37’W, Solomon 16947 (NY).

BRAZIL: Amapa: Serra do Navio, bank of the Rio Amapari, Emmerich &
Andrade 745 (HB, R). Roraima: Posto Mucajai, Rio Mucajaf, vic of Mucajai
airstrip, Prance et al. 10991 (GH, R, UC). Para: Lageira, airstrip on Rio Mai-
curt, 0°55'S, 54°26’W, Strudwick et al. 3088 (NY), 3129 (MG, NY, US), 3580
(US); Curud S.A., near Alenquer, Santarém, Piggott 2547 (K, NY); Breu Branco,
ca. 40 km S of Tucurui, 4°03’S, 49°40’W, Daly et al. 1376 (GH, MO, NY, US);
Serra dos Carajas, Serra Norte, ca. 15 km W of AMZA Exploration Camp., 6°S,
50°15’W, Berg & Henderson BG472 (GH, NY, UC, UEC, US); Serra dos Carajas,
6°4'S, 50°8’W, Secco 286 (GH, K, MG, NY, SPF); BR-163, Cuiabé—Santarém
Highway, km 885.5, Prance et al. P25171 (MG, NY, UC, US); Parque Indigena
do Tumucumaque, Rio Parti de Oeste, Miss&o Tiriyo, 2°20’N, 55°45’W, Caval-
cante 2401 (K, MG, NY, US); Conceigaéo do Araguaia range of low hills ca. 20
km W of Redengao, near Sado Jodo and Troncamento Santa Teresa, 8°03’S,
50°10’W, Plowman et al. 8635 (GH, NY), 8757 (GH, NY, US); Rio Xingu, Balée
2398 (NY); Confluéncia com Rio Pardo, Vasconcelos et al. 260a (NY); Rio Ita-
caiunas, affluent of the Rio Tocantins, Serra Buritirama, 50°15’W, 5°30’S, Pires
12427 (NY); Rio Cumina, Ducke (Hb. Mus. Goeldi 8885, 15163) (both HB, MG).
Amazonas: 1—5 km road Boca do Acre to Rio Branco, Prance et al. 2533 (GH,
MG, NY, R, US); Vic. of Tototobi, Basin of the Rio Demeni, Prance et al. 10208
(NY, UC, US); Rio Curuquete, Providencia, Prance et al. 14632 (NY, UC); Sao
Paulo de Olivenga, 30 km above the mouth of the Rio Coti, Prance et al. 14444
(B, NY); Vic. of Macujai airstrip, Prance et al. 10991 (MG, NY, UC); Borba,
4°02'S, 59°06’W, W side of the Rio Cunama, Hill et al. 12868 (MO, NY). Ron-
dénia: Mineragao Campo Novo, BR-421, a 2 km a Oeste da Mineragéo Campo
Novo, 10°35'5"S, 63°37’W, Vieira et al. 517 (NY, US); Basin of Rio Madeira,
Trail north of Rio Madeira from 2 km, below confluence of Rio Abuna, Prance
et al. 8345 (K, NY, UC, US). Acre: Palacio de Castro, fazenda Mococa, ramal
no km 120 da rod. Rio Branco—Pérto Velho, Santos et al. 122 (MG, NY, US); 9
km from Rio Branco on Rio Branco—Pérto Acre road at cut-off for Colénia
Cinco Mil, Lowrie et al. 650 (MG, NY, R, US); Sena Madureira, Bacia do Rio
Purus, varagao para o Seringal Fonte Boa, 10°07’'S, 69°13’W, Silveira et al. 668
(MO, NY); Xapuri, Seringal Cachoeira, 35 km SE of Xapuri, Pinard 809 (NY),
Kainer 126 (NY). Mato Grosso: Colider, Comunidade Sao Francisco, Salino
284 (GH); Serra Ricardo Franco, 15°S, 60°W, Windisch 1503 (HRCB); Rio Pei-
xoto de Azevedo, Faz. Sao José (Cachimbo), Bokermann 6747 (UEC); Santa
Terezinha, BR-158, Vila Confresa, pr. ao aeroporto da Fazenda Confesa,
10°35'S, 5°35’W, Windisch 5987 (UC).

ACKNOWLEDGMENTS
The first author appreciates the financial support of the Brazilian Research Council CNPq (Proc.
n. 300843/93-3 and 450658/99-6) and of the Smithsonian Institution, Washington DC (Short-Term
Visitor Grant). We thank also Sra. Emiko Naruto for preparing the illustration.

PRADO AND LELLINGER: ADIANTUM ARGUTUM 29

LITERATURE CITED

Cremers, G. and M. Horr. 1990. Inventaire taxonomique des plantes de La Guyane Francaise. I-
Les ace ta Museum National d’Histoire Naturelle, Inventaires de Faune et Flore.
Fasc

MER, K. U. 1978. ors pteridophytes of Suriname. Uitgaven Natuurw. Stud. Suriname Nederl.
Antillen 93:1-19

LELLINGER, D. B. 1989 ais ferns and fern-allies of ssi a Panama and Chocé (Part 1: Psilo-
taceae through Dicksoniaceae) haope 2A

PosTHUMUS, O. 1928. The ferns of Surinam. pre ie Java. 1

SMITH, A. R. 1995. poe L. Pp. eae in P. E. Berry, B. K. Holst, ae Vatachievsdhi (eds.),
Flora of the Venezuelan Guayana, vol. 2: Pteridophytes, Srempatophytan: Acanthaceae—Ar-
aceae. Timber Press, Portland.

SPLITGERBER, F, L. 1840, Enumeratio Filicum et Lycopodiacearum quas in Surinamo legit F. L.
iteadies Tijdschr. Natuurl. Gesch. Physiol. 7:391-444.

TrYON, R. M. and R. G. STOLZE. ee ah eae of Peru, Part II. 13.Pteridaceae—15.Dennstaed-
tiaceae. Fieldiana Bot., n.s. 2

VARESCHI, V. 1969. Helechos. = oat sod in T. Lasser (ed.), Flora de Venezuela. vol. 1, Tomo 2.
Instituto Boténico, Caracas

American Fern Journal 92(1):30—38 (2002)

Polypodium vulgare Plants Sporulate Continuously in
a Non-Seasonal Glasshouse Environment

SANNA E. SIMAN AND ELIZABETH SHEFFIELD
School of Biological Sciences, 3.614 Stopford Building, The ee of Manchester,
Oxford Road, Manchester M13 9PT, U

ABSTRACT.—In their natural environments pteridophytes usually have regular sporing periods, the
onset of which is triggered by the interaction of climatic and nutritional factors. Little, however,
is known about what changes there may be in the sporing behaviour of a fern when it is transferred
from its natural habitat to an artificial environment, such as a glasshouse. We recorded sporing
behaviour in relation to vegetative growth in two genetically matched populations of Polypodium
vulgare. One population was placed in a controlled-climate glasshouse, the other was left outside.
The recruitment of new fronds was significantly higher in the indoor population than in the
outdoor population. The indoor population also maintained a high proportion of actively sporing
fronds throughout the winter. There was no net recruitment of new fronds in the outdoor ce
lation during the winter and early spring. Some elements of the glasshouse envi t, probably
the enhanced light and temperature, induced continuous sporing in this fern. Considering the
ever-increasing interest in ferns as ornamental plants, and the growing body of evidence of toxic
and allergenic effects caused by fern spores, this kind of sporing behaviour may have implications
for human health.

Ferns in their natural environments usually have regular and predictable
periods of spore production and release. In the temperate zones and the sea-
sonal tropics they tend to release their spores towards the end of the growing
season. In the wet tropics, where the growing season is much longer or even
continuous, initiation of new fronds and maturation of older fronds take place
throughout the year (Page, 1979). Most ferns have been said to show very little
fluctuation in annual spore output with variations in climate, in contrast with
good and bad seed years in angiosperms and conifers (Page, 1979). This does
not apply to all fern species. Page (1976) pointed out that for Pteridium aquil-
inum (bracken) the spore yield can vary widely between different years. Sim-
ilarly, Steeves (1959) noted that, for Osmunda cinnamomea, a hot dry summer
is usually followed by a high degree of fertility in the following spring, where-
as a cooler moister summer leads to reduced fertility. Furthermore, the onset
of the reproductive phase in fern sporophytes can be demonstrated to be reg-
ulated by the interaction of several factors, such as light exposure, temperature
and the nutritional status of the plant.

Field observations have suggested that the onset of the reproductive phase
in ferns (as in many flowering plants) is induced by particular photoperiods,
but the evidence so far published is scanty (Wardlaw and Sharma, 1963).
periments carried out by Wardlaw and Sharma (1963) indicated that there is
a more or less direct relationship between active photosynthesis and/or pho-
toperiodic perception in the expanded leaves and the induction and devel-
opment of sori in the next inner leaves of Dryopteris austriaca. Harvey and

SIMAN & SHEFFIELD: POLYPODIUM VULGARE SPORULATION 31

Caponetti (1972), however, demonstrated that increasing light intensities in-
hibited sporophyll differentiation in Osmunda cinnamomea. Maximal initia-
tion of sporangia occurred in total darkness in this species, so it may be that
green and non-green spored ferns respond to light in different ways. There
may be doubts about the importance of photoperiodic induction, but in deter-
mining the extent of the fertility in ferns, photosynthetically available radia-
tion and the duration of exposure to light are probably of major importance.
Steeves (1959) compared the incidence of fertility between O. cinnamomea
plants in heavy wood and open areas and found a greater incidence of fertility
in the latter plants. Conway (1957), Dring (1965) and Page (1976) suggested
the same property in Pteridium (bracken), in which they found a gradual de-
crease in fertility with increasing degree of shade, although vegetative growth
in the latter may be little impaired. The enhancement of sporogenesis by high
light has recently been confirmed in experiments conducted with clones of
bracken grown in high and low levels of photosynthetically available radiation
(Wynn et al., 2000).

Temperature also plays a role in the onset of the reproductive phase as
shown by Labouriau (1958). He found that initiation of sporangia was stimu-
lated by exposure of the developing outermost set of Osmunda claytoniana
fronds to a temperature of 26°C; plants kept at a lower temperature remained
sterile. Similar trends were reported in Pteridium by Sheffield (1996) and
Wynn et al. (2000).

Allsopp (1964, 1965) suggested that nutritional conditions, particularly car-
bohydrate supply, appear to be of greater importance for the induction of spo-
rangia in pteridophytes than photoperiodic or similar stimuli. Several studies
indicate that the nutritional status of the plant is indeed of great importance
for the initiation of the spore-productive stage (Wardlaw and Sharma, 1963).
Goebel (1887, 1905, 1908) and Atkinson (1896) concluded, from experiments
with Onocleoid ferns, that if carbohydrate supplies are inadequate, developing
leaves tend to remain in the vegetative state. Goebel (1928) found that imma-
ture sporophylls of certain ferns developed as vegetative fronds when the ster-
ile fronds of the plant were removed. This has been repeated in numerous fern
species (e.g. by Labouriau (1958), Wardlaw and Sharma (1963), and Steeves
and Wetmore (1953)), who thus linked induction of sporogenous tissue to car-
bohydrate supply. Sussex and Steeves (1958) cultured excised leaf primordia
of Leptopteris hymenophylloides, Todea barbara and Osmunda cinnamomea,
and found that high sucrose concentrations in the medium was essential for
the inception and early development of sori and sporangia in those species.
They also showed that an increased supply of inorganic nitrogen promotes the
onset and extent of fertility in T’ barbara. According to Wardlaw and Sharma
(1963), there is a positive relationship between the amount of photosynthesis-
ing leaf surface and the induction of fertility.

It is apparent from the above that we have some limited knowledge of what
triggers sporing in ferns in natural conditions. The sporing behaviour of ferns
transferred from their natural habitat to an artificial environment, e.g. a glass-
house, however, remains largely unexplored. Considering the ever-increasing

32 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 1 (2002)

interest in ferns as household and garden ornamentals (Gress, 1996) and the
fact that fern spores may cause adverse health effects in humans (Siman et al.,
1999), more knowledge in this field is urgently required.

The aim of the current experiment was to compare the reproductive perfor-
mance of Polypodium vulgare plants placed in either a seasonal outdoor en-
vironment or a constant high-temperature and high-light glasshouse environ-
ment.

MATERIAL AND METHODS

Polypodium vulgare plants were collected in March 1998, from stone walls
along the south-east side of road B4403, running along the south-east shore of
Llyn Tegid (Bala Lake) (N 52°53’, W 3°38’), Wales, U.K. Polypodium was cho-
sen as it represents a widespread genus, including a broad range of horticul-
tural favorites, such as P. amorphum, P. cambricum and P. interjectum, of sim-
ilar morphology and life history (Mickel, 1994).

he plants were potted in commercial potting compost within a day of col-
lection. Lengths of rhizome were split into two equal parts (i.e. bearing the
same number of fronds in each of the two pots in a pair), and each was placed
in one pot. In this way 64 pots were prepared, i.e. 32 pairs of pots containing
clones. The length of the potted rhizomes varied from 1 to 5 cm, but was much
the same within each pair. In order to minimise the impact on the results of
the growth of any apical meristems, only median pieces of rhizome were used.
All plants were allowed a six-month settling-in period (mid-March to mid-
September) outdoors, after which one pot of each pair was left outdoors, to
the north-east of a glasshouse in the Manchester University Experimental
Grounds, Manchester, U.K. These were the ‘‘outdoor population”. The other
group of the plants was put inside a glasshouse (mean day temperature: 28°C,
range 20-38; mean night temperature: 15°C, range 11—27; photosynthetically
available radiation: c. 110 pmol m~? s~'), at the aforementioned Experimental
Grounds. These were the “indoor population”. At the start, the total number
of fronds in each population was very similar. No plants were given any ad-
ditional nutrients during the course of the experiment. Those outside were
subject to ambient rainfall, those inside were watered regularly. During the
settling-in period all fronds in 14 pots, seven in each population, died. Eight
of the 14 pots belonged to matched pairs, so the aim of ensuring a genetic
similarity between the two populations was still met to a high degree.

Weekly records were taken of the number of fronds in each pot with a) no
sori, b) immature (green) sori, c) sporing (yellow-orange) sori and d) empty
(brown) sori, from the beginning of October 1998 until June 1999 for the indoor
population. The outdoor population was recorded until the end of its growing
season in mid-September 1999.

The proportions of recruited fronds during the experimental period by the
two populations were compared with a x? test. The differences in numbers
and proportions of fronds of each developmental stage in the two populations
were compared numerically.

SIMAN & SHEFFIELD: POLYPODIUM VULGARE SPORULATION 33

|= outdoors

mean number of fronds per pot

28 8
oe Cc =
ee
= =
o

18/11/98

=
8
N
=
~
Nn
i=)

&
—~
oO
=

30/12/98

8
3

10/02/99
24/02/99
10/03/99

Fic. 1. Changes in the mean number of fronds per pot in two genetically matched populations of
Polypodium vulgare. The indoor aaa was placed in a controlled-climate greenhouse (mean
day ek anarae 28°C, mean night temperature: 15°C, photosynthetically available radiation: ca 110
pmol m~? s~'); the outdoor seatiitten was left outside (U.K. natural weather conditions). The re-
cruitment of new fronds in the indoor population occurred in three waves, one from October 1998
to mid-January 1999, the second from mid-January 1999 to late March 1999, and the third from early
April 1999 to the end of the experiment. Error bars show standard error of the mean

RESULTS

The recruitment of new fronds from October 1998 to early June 1999 was
significantly higher in the indoor population than in the outdoor population
(x2-test, x = 127, df = 1, p<0.01) (Fig. 1). During this time the indoor popu-
lation increased its number of fronds more than fourfold. The recruitment of
new fronds took place in three waves (Figs. 1 and 2a), each of which increased
the number of fronds by a factor between 1.5 and 1.7. The outdoor population
increased its number of fronds by a factor 1.2 from October 1998 to June 1999
(Fig. 1). All recruitment of new fronds in the outdoor population took place
from mid-April 1999 onwards. The increase in the number of fronds in the
outdoor population continued during the summer months until mid Septem-
ber 1999 (Fig. 3a).

The majority of the new fronds recruited in the indoor population during
the course of the experiment was fertile (Fig. 2a), so the indoor population
maintained a high proportion of actively sporing fronds throughout the winter.

ach wave of recruitment in the indoor population began with a sudden in-
crease in the number of initially sterile fronds; a number which decreased as
sori began to appear. The proportion of fronds that remained sterile throughout
the wave decreased with each wave. Thus, at the end of January 1999 the
proportion of sterile fronds was 37%, in early April 1999 28% of the fronds
were sterile and in early June 1999 the proportion of sterile fronds was 17%

34 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 1 (2002)

number of fronds

100

S86&8S88a88

percentages of fronds

co
eieresetege 33a 3 g
BSses ede Sbaz2 23 5 date
non at&éF4C aw FSS Ee

Fic. 2. Sporing behaviour in a population of Polypodium vulgare kept in a controlled-climate

bars represents i) sterile fronds and fronds with ii) green sori, iii) sporing sori and ix) empty sori,
as indicated by the key in the figure. (a) Total number of fronds in the population (full bars) and
number of fronds in each of the four groups (i.e. sterile, green, sporing, empty) for each week of
the experiment. (b) Proportions of sterile, green, sporing and empty fronds, respectively, for each
week of the experiment.

(Fig. 2b). The frond mortality represented 7.8% of the total number of fronds
gained over the experimental period and was entirely due to old fronds wither-
ing and falling off.

In the outdoor population there was no net recruitment of fronds during the
winter and early spring. When the new fronds started to emerge, in late April,

SIMAN & SHEFFIELD: POLYPODIUM VULGARE SPORULATION 35

Number of fronds

3 g
co.
raed
2S

Percentages of fronds

os 8 88 8

$
3
co

Fic. 3. Sporing behaviour in a population sa painir to vulgare kept outside in the Manchester
University Experimental Grounds, U.K. (natural weather conditions) from October 1998 to Sep-
tember 1999. The subdivision of the bars nanan i) sterile fronds and fronds with ii) green sori,
iii) sporing sori and iv) empty sori, as indicated by the colour code key in the figure. (a) Total
number of fronds in the population (full bars) and number of fronds in each of the four groups
(i.e. sterile, green, sporing, empty) for each week of the experiment. (b) Proportions of sterile,
green, sporing and empty fronds, respectively, for each week of the experiment.

most of them soon turned into fertile fronds, so there was a steady increase in
the number and proportion of fertile fronds over the summer (Figs. 3a and 3b).
The proportion of sterile fronds decreased simultaneously so towards the end
of the growing season in mid-September 1999, the proportion of sterile fronds
was 17% (Fig. 3b), i.e. the same as for the indoor population at the end of its

36 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 1 (2002)

third wave in early June 1999. The increase in number of fronds in the outdoor
population was slightly impaired at two points (24/June/99 and 22/July/99) by
herbivory from snails, but the fronds thus lost represented no more than 5%
of the population.

DISCUSSION

It is clear that the conditions of a warm and illuminated glasshouse stimu-
lated the vegetative growth and spore output of the P. vulgare plants.

In the indoor population, the initial response to the glasshouse conditions
was increased vegetative growth. At a time when the outdoor population
stopped producing new fronds, the indoor population continued recruiting.
Similar continuous growth has been observed in Pteridium grown in glass-
houses (Wynn et al, 2000), but it is interesting that Thomson (2000) reports
that Pteridium plants from four places (Honshu, Japan; Kiev, Ukraine; Bridg-
ton, Maine, U.S.A. and Waterloo, Michigan, U.S.A.) require cold treatment
(4°C) for four to eight weeks to ensure successful spring emergence of croziers
when cultivated in the relative warmth of Sydney Royal Botanic Gardens
(summer temp: max. 25.5°C, min. 18.2°C; winter temp: max. 16.8°C, min.
8.7°C). It seems that fluctuating temperatures warmer than those of the natural
environment of a fern can have adverse effects on frond recruitment.

The majority of the new fronds emerging in the indoor population of the
present study became fertile, so a high proportion of fertile fronds was main-
tained throughout the winter. Steeves and Wetmore (1953) concluded, after
experiments with Osmunda cinnamomea, that the factors which determine
fertility exercise their effects during the year before the leaves expanded. As-
suming, in the present experiment, that each wave of recruitment created in
the indoor population mimicked one growing season, we could suggest that
the warm and bright indoor climate had an effect on the fertility of the second
and third waves of new fronds. The proportion of fertile fronds in each of
those two waves was higher than in its preceding wave. This may well be an
effect of the enhanced nutritional status of the population, caused by a high
production of photosynthate, which, transported as sugars to the bud primor-
dia, might induce fertility, as suggested by Harvey and Caponetti (1972).

In its natural environment P. vulgare produces ripe spores from July/August.
The ripening of the spores is a gradual process, occupying a period of several
months. Within a single sorus some sporangia shed early and others will take
longer to ripen and shed later (Wright and Wright, 1999). The persistent pro-
portion of 10-20% sporing fronds in the outdoor population from October
1998 to March 1999 is evidence of this behaviour.

Each wave of recruitment indoors, as well as the single period of recruitment
outdoors, (i.e. the growing season) increased the number of fronds by a factor
of c. 1.5. This suggests that there were, at the beginning of the experiment, an
equal number of dormant buds in the rhizomes of the populations and that
the number of dormant buds an existing number of fronds can initiate for the
next generation is restricted by something other than purely environmental

SIMAN & SHEFFIELD: POLYPODIUM VULGARE SPORULATION 37

factors. This could explain the occurrence of the proportionally similar waves
of recruitment.

The present study shows that by enhancing the light and temperature it is
possible to interrupt the strict reproductive cycle and induce continuous spor-
ing ina pteridophyte population. Evidence of similar behaviour has recently
been obtained in another study, in which dormant Pteridium rhizomes pro-
duced fertile fronds within 13 weeks of being put into warm, well lit condi-
tions (Wynn et al., 2000). Air samples taken in glasshouse and fernery envi-
ronments in the UK do include ferns spores at all times of the year (e.g. Win-
ston, 1998; Siman, 2000).

This study suggests that transfer of plants to glasshouses could benefit fern
spore collectors by inducing continuous sporing in plants. There are less
welcome implications of fern spore production in indoor environments, how-
ever, especially for species that do generate vastly more spores in glasshouse
settings than those in natural environments. A glasshouse is an enclosed en-
vironment with little chance of biological particles being blown away by
winds. This means that there is a higher risk of inhalation of fern spores in a
fern-rich glasshouse than in most places outdoors. Based on the growing body
of evidence of toxic and allergenic effects caused by fern spores (as reviewed
by Simén et al., 1999, see also Siman et al., 2000), we suggest that some pro-
tective measures (e.g. face masks) should be taken by people who regularly
work in or visit indoor fern-rich environments.

ACKNOWLEDGMENTS

authors wish to thank David Newton for looking after the plants in the Experimental
Grounds, Gareth Ballance for standing in to take records in those weeks when we were away an
the University of Manchester Research Support Fund and the F. C. Moore Studentship Fund for
financing the study.

LITERATURE CITED

ALLsopp, A. 1964. The metabolic status and morphogenesis. saab aplaiatinigs 14:1-27.
ALLSOPP, A. 1965. The significance for development of water supply, osmotic relations and nutri-
Handbuch der Pflanzenphysiologie 15:504—552.
Srewene G. F. 1896. The probable influence of disturbed nutrition on the evolution of the veg-
etative phase of the sporophyte. Amer. Naturalist 30:353-357.
Conway, E. 1957. Spore production in bracken (Pteridium aquilinum (L.) Kuhn). J. Ecol. 45:273-

284
DRING, M. Si ba The influence of shaded conditions on the fertility of bracken. Brit. Fern Gaz.
9:2

EVANS, a ie The nature of flower induction. Pp. 457-480, in L. T. Evans, ed. The induction
of flov gear Macmillan of Australia, Melbourne.

GOEBEL, K. 18 Uber kunstliche Vergrunung der Sporophylle von Onoclea struthiopteris. Ber.
Deutsch. ae Ges. 5:69.

GOEBEL, K. 1905. Onecnouraglhiy of Plants. Part II. (English ed.). Oxford University Press.

GoEBEL, K. 1908. Einleitung in die experimentelle Morphologie der Pflanzen. Druck und Verlag
von B. Me ype Leipzig

"sheng K. 1928. Organographie der Pflanzen. Teil I. png Jena.

S, A. ae oo, Hort. Week 28-Mar 1996:2

38 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 1 (2002)

Harvey, W. H., AND J. D. CAPONETTI. 1972. In vitro studies on the induction of sporogenous tissue
on leaves of cinnamon fern. I. Environmental factors. Can. J. Bot. 50:2673—2682.
LABOURIAU, L. G. 1958. Studies on the initiation of sporangia in ferns. Arq. Mus. Nac. 46:119-202.

PAGE, Cc. = 1976. The taxonomy and phytogeography of bracken—a review. Bot. J. Linn. ae 75s)

Boy Ay e 1979. Experimental aspects of fern ecology. Pp. 105-140, in A. F. Dyer, ed. The ex-

perimental papas of ferns. Academic Press, London
bance E. 1996. From pteridophyte spore to lena it in the natural environment. Pp. 541—
ar aa and R,J. Johns, eds. Pteridology in perspective. Royal Botanic Gardens,

SIMAN, . E. 2000. Fern spores and human health. PhD dissertation. University of conan eee
SIMAN, S. E., A. C. POVEY, AND E. SHEFFIELD. 1999. Human health risks from fern spores?—a revie
87.

SIMAN, S. E, A.C. POVEY, ra Warp, G.P. MARGISON AND E. SHEFFIELD. 2000. Fern spore extracts
can damage DNA. Brit. J. Cinice 83:69-73.
STEEVES, T. A. 1959. An eriieeanias al two forms of Osmunda cinnamomea. Rhodora 61:223—

230.

STEEVES, T. A., AND R. H. WETMORE. 1953. Morphogenetic studies on eae cinnamomea L.:
some aspects . “ general morphology. Phytomorphology 3:339

Sussex, I. M., AND T. A. STEEVES. 19 ie + aust on the control of ya of fern leaves in
sterile culture. ot Gaz. 119:203-—

spas J. A. Morphometric and genomic diversity in the genus Pteridium (Dennstaedtiaceae).

n. Bot. 85:77-99.
Waou E W., AND D. N. SHARMA. 1963. Experimental and analytical studies of pteridophytes.
Bot. ne 101-121,

eho D. 1998. Risk assessment of environmental exposure to fern spores. M.Sc. dissertation,
hota! of Mancheste

WRIGHT, B., AND A. WRIGHT. om The “Wright” way to collect and clean fern spores. Pteridologist
3:62 ote

Wynn, J. M., J. L. SMALL, R. J. PAKEMAN, AND E. SHEFFIELD, 2000, An assessment of genetic and
envixonmental effects on sporangial development in img (Pteridium aquilinum (L.)
Kuhn) using a novel quantitative method. Ann. Bot. 85:113-115.

3 Oe

on

re »

aang plot yh? > Elliowsen: 7

pk

. oar
- 7 : iJ
ae des 7 f
: :
- - i
7 = a, ar cam’ -
; Vai _ - ie
7 aa
4 : iat tae. yy, =
- a a : = :
_ ae ek ae wD SS ae Fe
es Se a hcinn ag te >
a a _ =, ee Sol 3 Cae 3 :
; x ae Reto al ie ara: a! :
x wo ee ee hier - f 7 7
/
; ; 7 - ir
=f —
‘

Pp = |
-
7

a 7
ele ene ih le A ite a i
2 a fe. oe on a

tanical Garden Libra

I IN

Authors are encouraged to submit manuscripts pertinent to pteridology for pub-
lication in the American Fern Journal. Manuscripts should be sent to the Editor.
Acceptance of papers for publication depends on merit as judged by two or more
referees. Authors are encouraged to contribute toward publishing costs; however,
the payment or non-payment of page charges will affect neither the acceptability
of manuscripts nor the date of publication.

Authors should adhere to the following guidelines; manuscripts not so prepared
may be returned for revision prior to review. Submit manuscripts in triplicate
(xerocopies acceptable), including review copies of illustrations and originals of
illustrations. After review, submission of final versions of manuscripts on diskette
(in PC- or Mac-compatible formats) is strongly encouraged. Use standard 8% by
11 inch paper of good quality, not “erasable” paper. Double space manuscripts
throughout, including title, authors’ names and addresses, short, informative ab-
stract, text (including heads and keys), literature cited, tables (separate from text),
and figure captions (grouped as consecutive paragraphs separate from figures).
Arrange parts of manuscript in order just given. Include author’s name and page
number in upper right corner of every sheet. Provide margins of at least 25 mm
all around on typed pages. Do not submit right-justified copy, avoid footnotes,
and do not break words at ends of lines. Make table headings and figure captions
self-explanatory. Use S.I. (metric) units for all measures (e.g., distance, elevation,
weight) unless quoted or cited from another source (e.g., specimen citations). For
nomenclatural matter (i.e., synonymy and typification), use one paragraph per
basionym (see Regnum Veg. 58:39—40. 1968). Abbreviate titles of serial publi-
cations according to Botanico-Periodicum-Huntianum (Lawrence et al., 1968,
Hunt Botanical Library, Pittsburgh) and its supplement (1991). References cited
only as part of nomenclatural matter are not included in literature cited. For shorter
notes and reviews, omit the abstract and put all references parenthetically in text.
Use ae Herbariorum (Regnum Veg. 120:1—693. 1990) for designations of her-
bari

icatetik should be proportioned to fit page width with caption on the same
page. Provide margins of at least 25 mm on all illustrations. For continuous-tone
illustrations, design originals for reproduction without reduction or by uniform
amount. In composite blocks, abut edges of adjacent photographs. Avoid com-
bining continuous-tone and line-copy in single illustrations or blocks. Coordinate
sequence and numbering of figures (and of tables) with order of citation in text.
Explain scales and symbols in figures themselves, not in captions. Include a scale
and reference to latitude and longitude in each map.

Proofs and reprint order forms are sent to authors by the printer. Authors should
send corrected proofs to the editor and reprint orders to the printer. Authors will
be assessed charges for extensive alterations made after type has been set.

For other matter of form or style, consult recent issues of American Fern Jour-
nal and The Chicago Manual of Style, 14th ed. (1993, Univ. Chicago Press,
Chicago). Occasionally, departure from these guidelines may be justified. Authors
are encouraged to consult the editor for assistance with any aspect of manuscript
preparation.

Papers longer than 32 printed pages may be sent to the Editor of Preridologia
(Memoir Editor, see cover 2).

PTERIDOLOGIA ISSUES IN PRINT

The following issues of Pteridologia, the memoir series of the American
Fern Society, are available for purchase:

1. Wagner, David H. 1979. Systematics of Polystichum in Western North
America North of Mexico. 64 pp. $10.00 postpaid.

2A. Lellinger, David B. 1989. The Ferns and Fern-allies of Costa Rica,
Panama, and the Choc6 (Part 1: Psilotaceae through Dicksoniaceae). 364 pp.
$32.00 postpaid.

Send your order with a check or money order to: American Fern Society,

Inc., c/o U.S. National Herbarium MRC-166, Smithsonian Institution, Washing-
ton, DC 20560.

AMERICAN FERN JOURNAL ON MICROFICHE

Volumes 1-61 of the American Fern Journal are available as archival quality,
silver positive microfiches. Single volumes or the entire run may be purchased.
The fiches are easily read with 10X or greater magnification (using a dissecting
microscope and transmitted illumination or a fiche reader). Silver negative micro-
fiches of vols. 1-50 are also available. The price is $4.00 per volume or $244.00
per set of 61 volumes, postpaid.

Send your inquiry or order with a check or money order to: American Fern
Society, Inc., c/o Dr. James D. Montgomery, Ecology III, Inc., R.D. 1, Box 1795,
Berwick, PA 18603.

VISIT THE AMERICAN FERN SOCIETY’S
WORLD WIDE WEB HOMEPAGE:
http://www.amerfernsoc.org/

ee
AMERICAN eee
FERN aici

J O UJ R N A 1 April-June 2002

IN MEMORIAM: WARREN HERBERT WAGNER, JR.

Obituary: Warren H. Wagner, Jr. (1920-2000) Donald R. Farrar 39
Bibliography of Warren Herbert Wagner, Jr. David B. Lellinger 50

The Mating Systems of Some stele Polypodiaceae
n-Liang Chiou, Donald R. Farrar and Tom A. Ranker 65

Belowground enseicras and Abundance of Botrychium Gametophytes and Juvenile
Sporoph

ay Cindy Johnson-Groh, Chandra Riedel, Laura Schoessler and Krissa Skogen 80
Additions to the Fern Flora of Saba, Netherlands Antilles David B. Lellinger 93
Taxonomic Notes on Hawaiian Pteridophytes Daniel D. Palmer 97
Novelties in Pteridaceae from South America Jefferson Prado and Alan R. Smith 105

Is Gametophyte Sexuality in the Laboratory a — Predictor of Sexuality in N.
m A. Ranker and Heather A. pnd 112

Additional Support for Two Subgenera of Anemia (Schizaeaceae) from Data for the
Chloroplast Intergenic Spacer Region eo and Morphology
Skog, E. A. Zimmer and J. T. Mickel 119

Intrafamilial Relationships of the Thelypteroid cy (Thelypteridaceae)
n R. Smith and Raymond B. Cranfill 131

Two New Species of Moonworts (Botrychium subg. Botrychium) from Alaska
Mary Clay Stensvold, Donald R. Farrar and Cindy Johnson-Groh 150

Isoétes X herb-wagne ri, an [ pecific Hybrid of I. bolanderi x I. echinospora

o

(Isoétaceae)
W. Carl Taylor 161

Botrychium alaskense, a New Moonwort ea the Interior of Alask
Warren Herb Wagner, Ir and Jason R. Grant 164

A New Name for an Old Fern from North Alabama
James E. Watkins, Jr. and Donald R. Farrar 171

Continued Pteridophyte Invasion of Hawaii Kenneth A. Wilson 179

The American Fern Society

Council for 2002

CHRISTOPHER H. HAUFLER, Dept. of Botany, University of Kansas, Lawrence, KS erase 2016.
resident
TOM RANKER, University Museum, Campus Box 265, University of Colorado, Boulder, CO pa 0265.
Vice President
W. CARL TAYLOR, 800 W. Wells St., Milwaukee Public Museum, Milwaukee, WI 53233-1478.
Secretary
JAMES D. CAPONETTI, Dept. of Botany, University of Tennessee, Knoxville, TN 37916-1110.
Treasurer
GEORGE YATSKIEVYCH, Missouri Botanical Garden, P. O. Box 299, St. Louis, MO 63166-0299.
Membership Secretary
JAMES D. MONTGOMERY, Ecology II, R.D. 1, Box 1795, Berwick, PA 18603-9801.
Back Issues Curator

R. JAMES HICKEY, Botany Dept.,; Miami rcebie Oxford, OH 45056. Journal Editor
DAVID B. LELLINGER, U.S. National Herbarium MRC-166,

Smithsonian Institution, Washington, DC 2 0560-0166. Memoir Editor
CINDY JOHNSON-GROH, Dept. of ae oS Adolphus College,

800 W. College Ave., St. Peter, MN 56082- Bulletin Editor

American Fern Journal
EDITOR

R. JAMES HICKEY Botany Department,
Miami University, Oxford, OH 45056
ph. (513) 529-6000, e-mail: hickeyrj @ muohio.edu

ASSOCIATE EDITORS
GERALD J. GASTONY ~..00.....6.. uae of ae Indiana Sa ee IN 47405-6801
RISTOPHER H. HAUFLER .... Dept. of Botany, University of Kansas, Lawrence, KS 66045-2106
ROBBIN C. MORAN New York Botanical Gaaa hie NY 10458-5126
JAMES H. PECK Dept. of Biology, University of Arkansas—Little Rock,
2801 S. University Ave., Little Rock, AR 72204

ian “American Fern Journal” (ISSN 0002-8444) is an illustrated quarterly vero to the Asti
study of ferns. It is owned by the American Fern Society, and published at The Am n Fern Soc
% Missouri Botanical ana P.O. Box 299, St. Louis, MO 63166-0299. Peciodicgle postage oar. é
St. Louis, MO, and addition
— aims for missing oe potste 6 months semica pry to ks fr en (foreign) after the date of i ee

rders for pg oom should be addressed to Dr. J . Montgomery, Ecology II, R.D.
cece PA 1860: i
Changes of sees dues, and applications for membership should be sent to the Membership

tary.

General inquiries concerning ferns should be addressed to the Secre

Subscriptions $20.00 gross, $19.50 net if paid through an agency (agency fee $0.50); sent free to
members of the American i ociety (annual dues, $15.00 + $5.00 mailing surcharge beyond

U.S.A.

Back volumes are available for most years as printed issues or on microfiche. Please contact the

Back ae Curator for prices and availability.
TMASTER: Send address changes to AMERICAN FERN JOURNAL, 326 West St. NW., Vienna, VA
SieeateL
FIDDLEHEAD FORUM

The editor of the Bulletin of the American Fern Society welcomes contributions from members and
non-members, including miscellaneous notes, offers to exchange or purchase materials, personalia,
horticultural notes, and reviews of non-technical books on

SPORE Ex:

n McDaniel, 1716 Piermont Dr., Hacienda Hts., CA 91745-3678, is Director. Spores
cinhanea set lists of available spores sent on request fe / html

GIFTS AND ccs
enable bed to — its eervwes to members. and to = eerened

MISSOURI BOTANICAL

&

:39-49 (2(

Acs

American Fern Journal 92(2

AUG 2. 9 2002
GARDEN LIBRARY
Obituary: Warren H. Wagner, Jr. (1920-2000)

DONALD R. FARRAR

Department of Botany, Iowa State University, Ames, IA 50011

Warren Herbert Wagner, Jr. was born on 29 August 1920 and raised
in Washington D.C., the son of Warren Herbert Wagner and Harriet Claflin
Wagner. His early interests in natural history took him. frequently to the Smith-
sonian Institution, where he became acquainted with the experts, including

40 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

the eminent pteridologists William R. Maxon and Conrad V. Morton. In college
at the University of Pennsylvania, he became the enthusiastic field compan-
ion of Edgar T. Wherry, author of the “The Fern Guide.”” Wherry was a
mineralogist who became an expert on fern habitats and the first to point
out the important associations of epipetric ferns with particular rock types.

Graduating from the University of Pennsylvania in 1942, Herb entered the
U.S. Naval Air Corps, serving first in the Atlantic, then in the Pacific Fleet,
where he was a Naval Air Navigator. In the Pacific islands he spent his off-
duty hours collecting ferns and butterflies, later publishing (with David
Grether) ‘“Pteridophytes of Guam”’ (1948) as well as articles on the pterido-
phytes and butterflies of the Admiralty Islands. During this time he also flew
into California, taking his specimens to E. B. Copeland at Berkeley, re-
nowned expert on Philippine ferns. This was the beginning of an association
that would bring him back to Berkeley for graduate study. While in the
Navy, he also began what was to become a life-long study of the ferns of the
Hawaiian Islands.

At the University of California—Berkeley, in 1945, Herb joined student
colleagues Charles Heiser, Ernest Gifford, Frank Ranzoni, Vern Grant, Art
Krukeberg, and others in courses instructed by G. Ledyard Stebbins,
Adriance Foster, and Herb’s major professor, Lincoln Constance. Copeland,
although retired, was still active and advised informally on Herb’s research.
Also among Herb’s student colleagues in systematic botany was Florence
Signaigo. Herb and Florence married in 1948 and began a lifelong productive
partnership in research and publication.

After receiving his Ph.D. in 1950, Herb spent a year as a Gray Herbarium
Fellow at Harvard, then moved to the University of Michigan in 1951, where
he remained throughout his illustrious career in teaching and research. He
chaired or co-chaired the graduate programs of more than 45 Ph.D. students.
He taught a variety of graduate and undergraduate courses, including his
highly popular “Systematic Botany” and “Biology of Woody Plants,’ both of
which he continued to co-teach after ‘‘retirement” (1991) through the fall of
1999. Herb’s public lectures and seminars were equally popular. Few biolo-
gists have been in such demand as a visiting speaker. His curriculum vitae
list of “invited lectures” totaled 169—since 1992!

From 1966 to 1971 Herb served as Director of the University of Michigan’s
Matthaei Botanical Garden. His popularity with garden clubs, amateur bota-
nists, and conservation groups made the Gardens a center of outreach
activities in this arena, as well as a center for research and display of con-
servatory plants. He chaired the Department of Botany in the Division of Bio-
logical Sciences from 1974 to 1977 and chaired many additional department
and college committees, including the University of Michigan’s Tropical Studies
Committee from 1983 through 1997. He was chairman or president of nine
professional societies, including the American Fern Society, American
Society of Plant Taxonomists, and Botanical Society of America, and council
member/trustee/advisor to dozens of organizations. He served as an editor
for the Flora of North America, co-editing the Pteridophytes in volume two.

FARRAR: WARREN H. WAGNER, JR. 41

As reviewer of countless journal manuscripts, grant proposals, and botany/
biology programs, he gave freely of his time while continuing to teach and
maintain a research program generating over 240 research publications.

“Probably the best-known botanist ever to work at the University (of
Michigan),’”’ Herb Wagner’s influence on his science was immense. He has
been referred to as “the founder of modern day systematics” in reference to
his seminal contributions to cladistic analysis of phylogenetic relationships.
His studies in the recognition of species hybrids, their value in understand-
ing species relationships and their role in speciation and evolution became
classics. His knowledge of ferns worldwide was phenomenal and allowed
unique insights into fern ecology and life history, as well as systematics.
Along with awards recognizing his contributions to systematic biology
(including Willi Hennig Fellow, National Academy of Science, and the Asa
Gray Award of the American Society of Plant Taxonomists), his name is
indelibly inscribed in cladistic literature via the universally recognized
‘‘Wagner Tree’’ representation of phylogenetic relationships.

In writing about the career of Edgar Wherry, Herb stated that Wherry “was
one of those rare individuals—a real naturalist.’’ Extended as the highest of
compliments, the same could be said of Herb Wagner. His had an exception-
ally keen ability to observe the small and intricate details of plants interact-
ing with their environment. Study the knowable. Accumulate information on
the parts. With time, dedication, and an open mind, the big picture will
emerge. These gems of Herb’s philosophy attributed value to all research, no
matter how small the project or whom the researcher. All good data were
worth getting excited about—and he did. Herb’s ability to inspire others
through his interest in their studies and their knowledge not only fostered
independent research, but also created a legion of professionals and amateurs
eager to contribute data to Herb’s projects. The total productivity of this syn-
ergism, though unquantifiable, remains hugely visible.

Herb maintained a rigorous schedule of teaching, research, invited lectures
and symposia presentations until just weeks before his death on January 8,
2000 at the age of 79. He was very much looking forward to participating in
the 2000 Botanical Society of America symposium on the ‘Biology and
Conservation of the Ophioglossaceae’’ that he helped to organize. In the
summer of 1999, Herb and Florence conducted field work in Alaska and in
southwestern Canada. From both places they returned with, of course, new
species of Botrychium.

The foregoing highlights of the career and achievements of Warren ‘‘Herb”’
Wagner fall short in communicating the extraordinary nature of his per-
sonality, his gift for teaching, and his full influence on pteridology and pterido-
logists of the last half-century. The more personal reminiscences that follow
portray a rare individual whose life we are all so fortunate to have shared.

ACKNOWLEDGEMENTS

Information on the early years of Herb’s career were graciously provided by Florence Wagner.
Factual information is taken from Herb Wagner’s 1999 curriculum vitae. Other observations

42 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

extend from my long association with the Wagners, as a graduate student in Ann Arbor and in
many subsequent field trips and discussions of plants, people, and philosophy.

Portions of this article are excerpted with permission from Taxon 49:585-592. August 2000,
“Warren H Wagner, Jr. (1920-2000),” by Donald R. Farrar. Quotations are by Julian P. Adams
and William R. Anderson of the University of Michigan in The University Record 55:2-4.
January 17, 2000, ‘Warren H. (‘Herb’) Wagner,” by William R. Anderson.

REMINISCENCES

Meeting Herb Wagner.—I first met Warren Herb Wagner, Jr., on the eve-
ning of 13 June 1980, at Flathead Lake Biological Field Station in far north-
western Montana. This meeting has stuck in my mind, perhaps because in
many respects it typified my relationship with him. I was a new student in
pteridology and had come to Flathead Lake to take his (and Florence’s) four-
week-long fern course. Earlier that year I had phoned Herb to tell him that
I had enrolled in his course and to let him know about a new, unnamed fern
hybrid I had found in the Shawnee Hills of southern Illinois, a cross be-
tween the walking fern (Asplenium rhizophyllum) and the maidenhair
spleenwort (A. trichomanes). As it turned out, Herb was a few days late for
the course because of a conflict with the final days of the International Bota-
nical Congress in Vancouver. He finally arrived at the biological station
about 10:30 p.m., when I was alone in the lab and identifying plants. The
lab door suddenly opened and Herb burst into the room. He walked about
half its length before acknowledging my presence. We introduced ourselves.
He explained that he had been collecting moonworts all day en route to the
field station, but it didn’t seem like it to me. Instead of being exhausted after
a long day of fieldwork, he was animated and lecturing me about the biology
of moonworts and ferns in general. I listened spellbound and thrilled that he
would take the time to explain it all to me. “Let’s see your hybrid fern,” he
demanded. “It’s in my cabin,” I replied, ‘and my cabin is a long way away,
and it’s pitch-black outside, and I’ve lost my flashlight. Can I show it to you
tomorrow?” Then—and I'll never forget this—he gave me the most odd sup-
plicating look, and with hands clasped together as if praying, he said,
‘““Won’t you please go get it—now!”’ How could I refuse? I stumbled back to my
cabin in darkness, found the specimens, and retraced my steps to the lab.
Herb examined the specimens. “Yes, that’s it! Asplenium rhizophyllum xX
A. trichomanes. Congratulations!’” He shook my hand; I had received his
imprimatur. Then, without further ado, he described how he wanted the lab
rearranged—by me, that is. Tables, benches, plant presses, cardboard divid-
ers, microscopes—all these were to be moved according to a plan that he had
devised while I was retrieving the specimens. After explaining where every-
thing should be relocated, he said, ‘‘Ok, gotta blow!” and he left the lab as
abruptly as he had entered. I was once again alone in the lab, excited by
what I had just learned about moonworts and spleenworts, and too motivated
to mind having been pressed into service.—Robbin C. Moran, The New York
Botanical Garden, Bronx, NY 10458-5126.

FARRAR: WARREN H. WAGNER, JR. 43

Herb at ‘“‘Bug Camp’’.—My first memory of Herb Wagner is from Spring,
1952. During my senior year at the University of Michigan, about to graduate
with a major in Biological Science, I worked as an undergraduate lab assis-
tant in botany. That job involved lugging five-gallon bottles of distilled
water, spraying the greenhouse for insects, and other tasks. One day I was
called in to meet a new faculty member, Dr. Warren H. Wagner, Jr. He asked
me to be his graduate assistant that summer at the bug camp, the University
of Michigan Biological Station at Douglas Lake. With no other plans, I
agreed

That marked my beginning as a botanist. I assisted Wagner in courses in
Phycology and Pteridology. To be a graduate student, I had to take a course,
so I signed up for an independent study with him. Since the genus Equisetum
was well represented around the bug camp, Wagner assigned me to work with
its taxonomy. Herb showed himself to be an enthusiastic, knowledgeable field
botanist. I well remember, after a hard day bent in a half-crouch searching out
Botrychium, stopping with him for ice cream on the way back to Douglas
Lake. And, I remember evenings all of us gathered around a piano to hear
him play. Most evenings were spent in the lab until 11 p.m. or so, working
with Herb, on the materials gathered during the long days in the field.

That summer, Herb advised me to go to some other university for my mas-
ter’s degree, to widen my exposure to the whole field of botany, with the
understanding that I would return to Michigan to work for my doctoral de-
gree under him. I went off to Florida State University as a graduate research
assistant on a tidal marsh study, then to the University of California,
to study the anatomy of Equisetum stomata with Dr. Adriance Foster and
get an M. A. in Botany. After a stint in the U. S. Army, I returned to Ann
Arbor, to my mentor, Dr. Warren H. Wagner, Jr., and to the taxonomy of
Equisetum.—Richard L. Hauke, 900 N. Stafford St., Apt. 1103, Arlington,
VA 22203-1844.

Unflappable Herb.—If you know something of Herb’s activities during
World War II, it is no surprise that he was unflappable. He had been an offi-
cer in the U. S. Navy, a navigator on air flights, and for a time was based on
Guam. Like the pilots, he was not called upon to fly every day; on his days
off he collected butterflies and ferns. After the end of the war, he and David
F. Grether published an account of the pteridophytes of Guam based princi-
pally on their collections and those of other U. S. military personnel. (Occ.
Pap. Bishop Mus. 19(2):25-99. 1948). He was a native of Washington, DC,
and his butterfly collection was given to the Smithsonian Institution.

The best collecting for butterflies, Herb found, was beyond the American
defense lines that were maintained by Marine guards and patrols. Some well
armed Japanese troops were still at large beyond those lines, along with
numbers of their fallen colleagues who were very attractive to butterflies.
Because of the danger, the Marine guards didn’t like Herb’s excursions, but
as an officer, he was able to go where he wanted. A butterfly net was of the
utmost importance to those excursions. It was the equivalent of a white

44 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

flag that signaled the bearer’s non-combatant status. Even so, there must
have been a considerable element of danger, although Herb said he was never
shot at by the Japanese while carrying a net. (A plant press, even loaded
with ferns, did not confer the same protection, presumably because its
function was unknown to the Japanese soldiers.

In Ann Arbor, Herb’s spring wildflower course was always popular. His
humor, knowledge, and enthusiasm were delightful. Who could forget his
earnest description of a hand lens and how it was to be used and worn
around the neck? He then pulled from within the lectern a large hand lens
attached to a red ribbon so wide that it was practically a sash! Although
Herb was generally unflappable, I do remember one exception that was, de-
pending upon your point of view, hilarious.

The year I assisted him, his lectures were presented in three classrooms
that had been converted into a long, narrow lecture hall. The slide projector,
which I manned, was placed well back in the hall, perhaps 40 feet from the
desk, lectern and screen at the front. In order to signal for a change of slides,
Herb would hold a wooden pointer vertically and drop it dramatically on
the concrete floor next to his feet. Although none of us knew it at the time,
the repeated thunk was not popular with the Dept. of Geology professors
who had offices on the floor below.

One afternoon, Herb called for darkness. The shades were drawn while
Herb turned out the room lights. I turned on the projector and put the first
slide on the screen. Naturally, I was looking at the projector when Herb drop-
ped the pointer to request the second slide. There was a tremendous det-
onation in the front of that quiet room. My head snapped up in time to see
Herb still coming down behind the desk after what must have been a terrific
leap, nearly a pole vault. He strode over to the door, clicked on the lights,
took out and lit with trembling hands one of the small cigars he smoked at
that time, inhaled deeply, and said ‘‘Take a five-minute break.” Of course,
the students could not contain themselves. By evening, most of the graduate
students knew what had happened: one of them had taped a few caps from
a cap pistol to the blunt end of the pointer. After the recess and cigar, Herb’s
lecture and slides continued as before, punctuated only by an ordina
thunk. The Geology professors were not at all amused.—David B. Lellinger,
National Museum of Natural History, Smithsonian Institution, Washington,
DC 20560-0166.

Graciousness and Support.—Like most graduate students back in the
1970's, it wasn’t long before beginning my study of ferns that I was introduced
to Herb Wagner. My first impressions were that he was brilliant, dynamic,
mesmerizing, and exactly the type of scientist that I wanted to be. Of course,
that wasn’t possible because he was certainly one of a kind. He reviewed one
of the first papers that I wrote concerning meiotic studies in Ceratopteris.
I remember that the review was less than favorable about some aspects of the
study and of the way that it was presented. He was right, of course, but for
some reason, perhaps extensive re-writing, it eventually was published.

FARRAR: WARREN H. WAGNER, JR. 45

A couple of years later, | wrote a paper that made an attempt to propose
an alternative explanation for the pattern of reticulate evolution that Herb
had described for Appalachian Dryopteris. This was the last piece of my
doctoral dissertation. I soon graduated and while I was at work as a young
assistant professor, I actually received a reprint request for this paper from
Hugh Iltis with a big “‘congratulations”’ scrawled on it! But any glory was to
be short-lived. A couple of years later, during a meeting of the Southeastern
fern group at the Duke Marine lab, I remember sitting quietly in the front
row as Herb methodically destroyed my brash interpretation of Appalachian
Dryopteris. | was humbled, but not humiliated. He was far too gracious for

at.

Although I was not one of his students in any type of academic lineage,
he was a constant and helpful influence in my career. Letters of recommen-
dation for jobs and promotions, reviews of my academic department, and oc-
casional suggestions spanned the course of over 20 years. My last personal
encounter with him was during a visit to Knoxville in the late 1990’s where
he gave an inspiring seminar to a group of faculty and students. I was teach-
ing a sophomore genetics class that semester and made a point of inviting
my students to hear his seminar. Although they weren’t botany majors and
generally had very little interest in plants, it was obvious that he had not
lost his charm, dynamism, and ability to mesmerize even these supposedly
disinterested young people in his story about moonworts! I keep a small pic-
ture of Herb behind me in my office and often turn and sneak a peek at him
for inspiration when I am trying to figure out how to do a better job of pre-
paring a lesson or writing up a laboratory exercise.—Les Hickok, Department
of Botany, The University of Tennessee, Knoxville, TN 37996.

Humanity and Professionalism.—He combined intellect with a humanity
and friendliness. Herb Wagner, as we called him when he befriended us,
traveled widely and infected nearly everyone he met with an enthusiasm for
life and for inquiry into the natural world.

His influence upon me came early in my professional career. While a stu-
dent at the University of Kentucky between 1961 and 1963, Herb had already
visited the campus, and as always, took a field trip into the Kentucky hills.
I was told that an old man walking along a road of a cove yelled, ‘“‘Hey, what
are you fellers doing down there?’ Herb’s immediate response, ‘‘We’re just
botriculating.”” And though suspicious of revenuers, the old man responded,
“Oh well, go right ahead.”

It was shortly after I arrived in Cullowhee in 1969 that Herb made a visit
to Highlands Biological Station in preparation for the fern conference in
1970. He had come across a paper that Jim Horton and I published listing
new county records that included Asplenium heteroresiliens, a hybrid on
marl known only from the coast. He visited our herbarium and in his in-
structive but gentle way said, ‘I see why you identified Asplenium tricho-
manes for heteroresiliens, but it isn’t.” Our somewhat depauperate and
poorly dried specimen didn’t exactly match either species. Herb also got me

46 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

to be more careful in my identification of other species, pointing out that
what appeared to be Populus grandidentata growing beside the science build-
ing was something else, which we later identified as Populus X canescens.

In the early 1970’s Herb was President of the American Fern Society, pop-
ularizing the profession. About this time I had a free evening and took in the
movie, “A New Leaf,” a spoof of a millionaire and his attraction to a wall-
flower type of lady who happened to be a student of ferns. To my amaze-
ment, she came to the point of verifying her identity that the tropical fern
she discovered in Alaska was confirmed by ‘‘Dr. Wagner at the University of
Michigan.” Such was his fame as it expanded beyond the ivory tower into
society in general.

Herb visited Cullowhee many times. On his last trip here in April, 1997,
not only did he give the visiting scholar lectures but also wanted to return
to some nice place before he returned home. His 4 p.m. seminar that Friday
afternoon was so well received that a couple of visitors, Bob Dellinger, local
naturalist, and Gary Kauffman, USDA Forest Service botanist, asked to tag
along with us the next day to the Joyce Kilmer area. Although Herb had in-
dulged in country ham the previous days (he was supposed to be on a low
salt diet) and was a bit tired from the activities the previous two days, he de-
lighted in finding Carex austro-caroliniana with peduncles several centi-
meters long. Such was his exuberance over those things he found
fascinating.—Dan Pittillo, Department of Biology, Western Carolina Univer-
sity, Cullowhee, NC 28723.

In His Element.—When | think of Herb, I tend to think of him in a field
trip setting. For instance, when he was invited to the annual meeting of the
Field Botanists of Ontario in Midland, Ontario a number of years ago, I re-
member the lead car leaving the sandy gravelly country road and entering an
open sandy field with a few copses of junipers. In what resembled a “cops
and robbers” film on T.V., the lead car had not come to a full stop when
Herb rolled out the front door of the car. He took a giant step towards the
copse and was almost immediately flat on the ground under the prickly Juni-
per. Suddenly there was a victory call for all and sundry within 200 yards—
“Botrychiums.” All Herb’s followers were on the spot in a few seconds and
each looked in marvel at some very small specimens between the juniper
and the sand—a gap of less than a foot. Herb was in his element in the field,
and his disciples enjoyed every minute of it!—D. M. Britton, Professor Emer-
itus, Department of Molecular Science & Genetics, University of Guelph,
Guelph, Ontario, N1G 2W1 Canada.

One of the All-time Giants.—My first contact with Herb Wagner came
about in early 1978 when I was at the University of Cambridge, England. I
had been intrigued for some time by a note in Herb’s paper on spores in rela-
tion to fern phylogeny (Ann. Missouri Bot. Gard. 61: 346. 1974) on the pres-
ence of large spores in ferns in the higher altitude rainforests of Hawaii not
being linked to polyploidy. I had noted that in the New Zealand species
of Grammitis (Grammitidaceae) there is a trend towards the production of

FARRAR: WARREN H. WAGNER, JR. 47

larger spores, and frequently fewer spores per plant, in the species occurring
at higher altitudes and latitudes. Although few chromosome counts are avail-
able this does not seem to be linked to polyploidy. I had assumed that this
wasn’t a parallel of the strategy of higher plant taxa that at higher altitudes
produce fewer and larger seeds to contain resources needed during longer
periods of dormancy, because dormancy is not an issue with Grammitid
spores, which are photosynthetic. Perhaps Herb’s ‘‘selection for precinctive-
ness’’ was working here? So I wrote to him concerning my observations; he
promptly replied that he thought it was, and that in Hawaii, for example,
any spore or any propagule that is especially likely to be carried by wind is
going to lose out. His observations on butterflies had some part in his logic,
with the most conspicuous Hawaiian butterfly being a very sluggish inhabi-
tant in mountain valleys, rather than ridges, where it was likely to be blown
away. I’m still very much undecided about selection for precintiveness in
Grammitidaceae after further work, as the high altitude New Guinea species
do not show the same increased spore size as the New Zealand ones, but
Herb’s prompt, encouraging and discursive response to a junior unknown
pteridological correspondent made a deep and lasting impression on me that
here was one of the all-time giants of pteridology.

Nine years later, when I had moved to the Royal Botanic Gardens, Kew, I
finally met both Herb and Florence at the International Botanical Congress
in Berlin in 1987, and the impression I had of Herb from his correspondence
was strongly reinforced. He enjoyed ferns and people and talk, and I really
envied all of his students for having such a stimulating mentor! A talk with
Florence at Berlin turned up a surprising coincidence—one of my M. Sc.
supervisors at the University of Auckland, New Zealand, Prof. Jack Ratten-
bury, had been best man at Florence and Herb’s wedding—it’s a very small
botanical world.—Barbara Parris, Fern Research Foundation, 21 James Kemp
Place, Kerikeri, Bay of Islands, New Zealand.

Tutor and Friend.—Herb Wagner was my tutor and friend during the last
dozen years of his life and he assisted me in gaining the skills needed to
study the Hawaiian ferns. Herb worked with the Hawaiian fern flora for
more than half a century, during which time he greatly increased the know]l-
edge of its diversity and biology. He described more than three dozen new
species, varieties, and hybrids. With his contributions, our organized under-
standing of the Hawaiian ferns was much advanced.

Herb made several long visits to Hawaii during which he went on exten-
sive field trips and spent much time in local herbaria. During these visits he
taught courses reviewing the Hawaiian ferns and took his students on field
excursions. A remarkable and enthusiastic teacher, the passionate sharing of
his deep knowledge of Hawaiian ferns inspired many in Hawaii to study this
neglected group of plants. His lectures and seminars were remarkable for his
enthusiastic, thorough, and effective presentation of material to students
with only a minimal knowledge of pteridophytes. I will remember the ballet
like pirouettes he would do at the blackboard when emphasizing a point,

48 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

and I remember the vast collection of anecdotes and jokes he used to enliven
a lecture and to help students remember the subject under discussion. He
would not let the decreasing physical vigor he experienced in later years
interfere with active field work. On his return to Ann Arbor from Hawaii he
always left behind a whirlwind of enthusiasm for the study of Hawaiian ferns.

Florence Wagner’s collaboration and organizational abilities allowed the
Wagner family to effectively pursue the twin goals of teaching and research.
Many of us in Hawaii will fondly remember a day’s end with Herb and
Florence, where a good dinner, cocktails, stimulating conversation, and en-
joyable company combined to make for a very pleasant evening.

He is missed by all who knew him in Hawaii. His stimulating visits
will be missed and the time he spent here will be remembered here with
nostalgia.—Daniel D. Palmer, 1975 Judd Hillside, Honolulu, HI 96822.

Missing Herb.—I miss Herb too much to find words to describe how misera-
ble I am at knowing he is not at the end of the phone needing a this or a that,
or welcoming my weird plant ID, or to confirm one of those peripheral vegeta-
tive idioblastic forms that my heart hoped might be something new. I remem-
ber the summers he visited Iowa Lakeside Lab, the AFS field trips,
particularly the one in northern Michigan. I did not know Herb from Michigan
as many did, as he was my academic grandfather—and father in Iowa by
remote-control, a fitting way to be for us hybrid studying types. He thought that
my following Darwinian fates of individuals and natural history approach to
fern reproduction and population structure was significant when most around
me spoke only of things homoeologous, bar codes of molecules on gels, boot-
strapping clades, and such. In comfort, Herb offered that he often no longer
knew what cladistic people meant with the language they used to describe his
ground-plan divergence method he created so long ago. I only wanted to be a
teacher, and Herb’s way with a gathering of students still is my exemplar. We
will all miss him; he will always be with us.—James H. Peck, Department of
Biology, University of Arkansas—Little Rock, Little Rock, AR 72204.

A Personal Gift.—I didn’t go to the University of Michigan to study with
Herb Wagner. My interests were in plant physiology, but in my first semester
a fellow graduate student, Bob Faden, who was then a student of Herb’s,
secured an invitation for me to accompany the annual Wagner entourage to
the hills and canyons of southern Ohio and Kentucky. Herb had just pub-
lished papers on the independent occurrence of gametophytes of Vittaria
and Trichomanes in the eastern United States. We had no trouble finding
great mats of Vittaria gametophytes and, with some squirming on our backs
under overhanging sandstone cliffs, we also found the green, threadlike
gametophytic filaments of Trichomanes. Returning with collections of these
intriguing plants, I was thrilled and convinced that with my growing
physiological expertise I would coax them in culture into producing sporo-
phytes and would resolve the mystery of their independent (without sporo-
phyte production) occurrence in the wild. Nearly 40 years later I still work
on that.

FARRAR: WARREN H. WAGNER, JR. 49

After two years of graduate school I was having doubts about my future in
academia. With considerable trepidation I informed Herb of my intention to
take some time off and join the Peace Corps. To my surprise, he supported
my decision, but then gently reminded me that I was now the ‘world’s
expert” on independent gametophytes and I ‘‘owed it to science’ to stay one
more year to publish what I had learned. By the time I did that, of course,
my thinking had changed and I was forever hooked on fern gametophyte
biology. Herb probably expected that outcome, but it wouldn’t have hap-
pened without his encouragement and confidence in me.

The excitement of field trips with Herb and Florence remain highlights of
my graduate years at Michigan, as do warm memories of holidays at the
Wagner home, cutting out snowflakes or whatever project Florence had de-
signed for their “extended family” of graduate students. This nurturing of an
Ozark farm boy a long way from home made a difference. It was a personal
gift. Yet I know it was only one of many such personal gifts, bestowed on
many others as well, by Herb and Florence. For those gifts we all say
thanks!—Don Farrar, Department of Botany, Iowa State University, Ames,
Iowa 50011.

American Fern Journal 92(2):50-64 (2002)

Bibliography of Warren Herbert Wagner, Jr.

Davin B. LELLINGER
Department of Botany, Smithsonian Institution, Washington, DC 20560-0166

Asstract.—This bibliography contains 395 entries, all the scientific publications of Warren H.

Exact publication dates were found for most journal articles. However, month or estimated dates
were used for other lire eciseem books, which usually are - datable within a =
All the entries are contained in a searchable database now maintained by Florence S. Wagne
The eral categorizes the ee according to their subject matter and includes a
more or less exact publication dates that are the basis for the onan list.

Wacner, W. H., Jk. 1941. New localities for Botrychium matricariaefolium in Maryland. Amer.
Fern J. 31:21—22.

———.. 1941. Butterfly hunting: Why and How? The Naturalist 1:6 1941. [Mimeographed publi-
cation of the Naturalists Field Club of the University of Pennsylvania]

. 1941. District of Columbia Butterfly Notes (Lepidoptera: weit tale Entomol. News

52: 196-200, 245-249. The parts were published 18 Jul and 8 Oct

. 1942. Bipinnate Christmas ferns. Amer. Fern J. 32:27—29.

9—70.

ia)

1943. New locality for a rare Hairstreak (Lepidoptera: ane Entomol. News 54:11.
1943. scat: sae by remote control. Amer. Fern J. 3 7

944, rms new to Trinidad. Amer. Fern J. 34 oe
1944, pare occurrence of the apparent hybrid Cystopteris. Amer. Fern J. 34:125—127.
1945. Fern hunt in Puerto Rico. Amer. Fern J. 3
1945. Ferns on pacific island coconut trees. Amer. Forn J. 35:74—76.
1946. Fern field notes in the Washington—Baltimore area. Castanea 11:59-60.
. 1946. Notes on the protection of rare ferns in the Wochisister Malti asces area. Bull. on
dis Washington-Baltimore Area Flora 10:5-6. [Mimeographed]

- 1946. Botrychium multifidum in Virginia. Amer. Fern J. 36:117-1
: 1946. Pteridophyta. Pp. 3-8 in Hermann, F. J. A checklist of olnnts of the Washington—
Baltimore area, ed. 2. [Mimeographed
. 1947. Tree-climbing Gleichenias. Amer. Fern J. 37:90-95

& D. F. GRETHER. 1948. Pteridophytes of Guam. Occas. Pap. Bernice Pauahi Bishop Mus.
19(2):25-99,

———.. 1948. A new fern from Rota, Mariana Islands. Pacific Sci. 2:214—-21
RETHER. 1948. The Pteridophytes of the Admiralty Islands. Unity, Calif. Publ. Bot.
23(2): 17-110, t. 5-25.

& D. F. GreTHer. 1948 [1949]. The butterflies of the Admiralty Islands. Proc. U. S. Natl.
Mus. 98:163-186.
——.. 1949. A reinterpretation of Schizostege lidgatei (Baker) Hillebrand. Bull. Torrey Bot.

Club 76:444—461.

. 1950. The Hart’s-tongue Fern. 18th Annual Calif, Spring Garden Show, April 21-28, pp.
i,

=

| . | | | |

Re |

1950. The habitat of Diellia. Amer. Fern J. 40:21—

- 1950. Ferns naturalized in Hawaii. Occas. Pap. ae Pauahi Bishop Mus. 20(8):95—
121.

- 1951. A new species of Diellia from Oahu. Amer. Fern J. 41:9-13.

LELLINGER: BIBLIOGRAPHY OF WARREN HERBERT WAGNER, JR. 51

. 1951; Sie pe a analysis of evolution in Pteridophyta. [Review of: Manton, I. 1950.
Problems of Cytology and sig in the Pteridophyta. Cambridge Univ. Press, London

sir sire York.] Ss. oe 5:177—

951. Review of: Manton, I. pi Problems of Cytology and gigerton ie in the Pterido-

ha Cambridge Univ. Press, London and New York. Amer. Fern J. 4 93.
. 1952. The flowerless plants. [Review of: Billington, C. 1952. Fems of Michigan. Cran-
— Institate oe Science, Bloomfield Hills, MI.] Sci. Monthly 75(5):3
rm Genus Diellia: Its Structure, Affinities and Ao Univ. Calif. Publ.
ree a) leah t. 1-21.

. 1952. Review of: McVaugh, R. & J. H. degen 1951. Ferns of Georgia. Univ. of Georgia
mae pe Michigan pee 58(21):366—
———. 1952. Review of: Billington, C. 1952. caine of Michigan. Cranbrook Inst. of Science,

Bloomfield Hills, MI. Cranbrook Inst. Sci. News Letter 22(1): ee

. 1952. Juvenile leaves of two Polypodies. Amer. Fern J. 42:8
. 1952. Types of foliar dichotomy in living ferns. Amer. J. i pe 578-59 92.
. 1953. The genus Diellia and the value of characters in determining fern affinities. Amer.
J. Bot. 40:34—40.
2 1953; An Asplenium prototype ne be genus Diellia. one — Bot. Club 80:76-94.
. 1953. Report of the Secretary for 2. A
. 1953. A cytological study of the pam Spleonw i re mer. Fern J. 43:109-114
1953. Review of: Hubbard, D. H. 1952. Ferns of Hawaii National Park. Hawaii Nat. Hist.

——— & D. J. HaGenan. 1954. A natural hybrid sy et apaerias lonchitis and P. acrostichoides
en the Bruce Peninsula. Rhodora 56:1-6, t.
4. A Bolivian Elaphoglossum of unique “leaf structure. Bull. Torrey Bot. Club 81:

1954. Report of the Secretary for 1953. Amer. Fern J. 44:21-24
1954. Reticulate evolution in the Appalachian Asphiiiana, Evolution 8:103-118
[Reprinted as pp. 136-151 in Orndoff, R. 1967. Papers on Plant Systematics. Little, —
Boston. ]
oes The evidence used recent classifications of the ferns. Rapports Comm. aux Sect.
: , 6, Ville Congr. Int. B
WILSON, - ci & W. H. WAGNER he pris a of: Lawalrée, A. 1950. Flore cnn de Belgi-
que: Pt spy Jardin Botanique de |’Etat, Bruxelles. Amer. Fern J. 4
Wacner, W. H. Jr. 1955. Cytotaxonomic observations on North American “on phen 57:219-
240.
. 1955. Should - American Hart’s-tongue be interpreted as a distinct species? Amer.
Fern J. 45:127-1
. 1955. Austin ie Clark. re News 9(4,5):151-157.
. 1955. Review of: Manton, I. . A. Sledge. 1954. Observations on the cytology and
acne of the sean sh For Ceylon. Phil. Trans. Roy. Soc. London, Biol. Sci.
238:127-185. Amer. Fern J. 4
. 1956. A reoviany rege mii en a C. C. Hall. Madrofio 13:195—205
D. J. Hac . 1956. A diploid variety in the Cystopteris fragilis sienpies Rhodora

58:79-87, t. ee.
. 1956. Asplenium ee x ot gla a new triploid hybrid produced under arti-
fical conditions. Amer. Fern J. 4
Voss, E. G. & W. H. WAGNER . 1956. in on Pieris virginiensis and Erora laeta—two butterflies
hitherto enikciovted from Michigan. Lepidopterists’ News 10:1
& L. P. Lorp. 1956. The morphological and cytological distinctness of Botrychium minga-
swe and B. lun in Michigan. Bull. Torrey Bot. Club —280.
D. J. emia oe [1957]. a a on some tralia: -producing populations of the
Cystopteris fragilis complex. Amer. Fern J. 46:137-146.
1957. Heteroblastic leaf moepholgy in juvenile plants of Dicranopteris linearis (Gleiche-
tincead). Phytomorphology 7

52 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

. 1957. Botanical Phase. Pp. 1-14 in J. M. Sheldon & E. W. Hewson. Atmospheric Pollu-

tion by Aeroallergens, Progress Report No. 1. Eng. Res. Inst. 2421-1-P, Ann Arbor, MI

J. DaruING, JR. 1957. Synthetic and wild Asplenium gravesii. Brittonia 9:57-63.

STEINER, E., A. S. SUssMAN & W. H. WacnerR Jr. 1957. Botany Laboratory Manual. Dryden, New
York

Wacner, W. H. Jr. & R. S. WHiTMIRE. 1957. Spontaneous production of a morphologically distinct,
fertile oo by a sterile diploid of Asplenium ebenoides. Bull. Torrey Bot. Club

seg i is GILBERT, 1957. An unusual new Cheilanthoid fern from California. Amer. J. Bot.
: 743.
Constance, L., H. Lewis, R. C. Rouuins, R. F. THORNE & W. H. WAGNER JR. 1958. Suggested outline
for at aprons botany. Pl. Sci. Bull. 4(1):1-3
esis ni H. Jr. 1958. Notes on the distribution of a kentuckiense. Amer. Fern J.
9-43.

eee Botanical Phase. Pp. 1-13 in J. M. ci & E. W. Hewson. Messiah sone Pollu-
tion rc neering a, Progress Report No. 2. Eng. Res. Inst. 2421—-2-P, Ann Arbor,
& EA 8. Perennial ragweeds fAnabocetia) in Michigan, with the lem of
a new, vito Si ie Rhodora 60:177—204
1958. Review of: Bold, H. C. 1957. Marntolosy of Plants. Harper, New York. Science
128:1079.
Dincte, A. N., J. C. Gin, W. H. Wacner Jr. & E. W. Hewson. 1959. The emission, dispersion and
deposition of ragweed pollen. Adv. Geophys. 6:367—387
Wacner, W. H. Jr. 1959. Botanical research on atmospheric pollution. Proc. 13th Ann. Meeting
Northeastern Weed Control Conf., Rutgers Univ.:262-268. [Mimeographed]
K.

wood and its relationship to a peculiar plant from West Virginia. Amer. Fern J. 48:146—
159.

- 1958 [1959]. The hybrid ragweed, Ambrosia artemisiifolia x trifida. Rhodora 60:309-

- 1959. Problems in the classification of ferns. Proc. IXth Intl. Bot. Congr. 2(Abstrs.):
418-419.

CanTLON, J. E., W. H. WAGNER JR. & W. is Riripon 1959. A range extension in Arctic Alaska for
cares lunaria, Amer. Fern J. 49:25-29.

Wacner, W. H. Jr. 1959. Hybrid swarms of mittee [Review of: Walker, T. ss 1958. Hybridization
in some — of Pteris L. Evolution 12:82—92.] Amer. Fern J. 49:4

, ed. 9. An annotated bibliography of ragweed (Ambrosia). a Ailacay Appl. Immu-

nol. . ae

- 1959. Botanical phase. Pp. 1-22 in J. M. Sheldon & E. W. Hewson. Atmospheric Pollu-

tion ky Aeroallergens, Progress Report No. 3. Univ. Mich. Res. Inst. 2421-3-P, Ann

MI.

. SCHWEMMIN & W. H. Wacner JR. 1959. Pollen release in the common ragweed
todanaie cape tagioersd Bot. Gaz. 120:235—243.
W. . 1959

can Grapeferns resembling Botrychium ternatum: A preliminary

. 1960. Evergreen Grapeferns and the pres of infraspecific categories as used in North
American Pteridophytes. Amer. Fern J. 50:32—45.

. 1960. The proliferations of Asplenium pipet Castanea 25:74-79.

. 1960. Ragweeds. Cranbrook Inst. Sci. po Letter 30(1):2-8.

- 1960. Botanical phase. Pp. 1-25 in J. M. Sheldon & E. W. Hewson. a ao
tion by sy gi Progress Report No. es Univ. Mich. Off. Res. Admin. 034
Ann

. 196 ae and pigmentation in aa subg. Sceptridium in the northeastern
United ‘Staten, Bull. Torrey Bot. Club 87:3

961. Filicineae, Frond. Pp. 396-398, pooh in P. Gray, ed. The Encyclopedia of the
nies Sciences. Reinhold, New York.

ee 9

LELLINGER: BIBLIOGRAPHY OF WARREN HERBERT WAGNER, JR. 53

. 1961. Some new data on the vernation differences of Botrychium dissetum and B. terna-
tum. Amer. Fern J. 51:31-—
. 1961. Problems in the elaneification of ferns. Pp. 841-844 in Recent Advances in Botany,

———. 1961. On the relative development of the fertile segments in Botrychium dissectum and
B. ear Amer. Fern J. 51:75-81.

1961. The Goldback Ferns of California. Review of: Alt, K. S. & V. Grant. 1960. Cytotaxo-
nomic observations on the Goldback Fern. Brittonia 12:153-170. Amer. Fern J. 51:105-106.
. 1961. Spore studies and speciation in the ferns on Hawaii. 10th Pac. Sci. Congr. Abstrs.
Symp. Pap. 135-136
. 1961. Roots and the naa differences between Botrychium oneidense and B. dissec-
tum. Rhodora 63:164-1
& K. E. Boypston. io, en new hybrid showing homology between Asplenium ebenoides
and A. pinnatifidum. Brittonia 13:286—289.

961. Nomenclature and typification of two Botrychiums of the southeastern United
States, Taxon 10:165-169.

1 Review of: Wherry, E. T. 1961. The Fern Guide. Northeastern and Midland
Eiesiliie States and Adjacent Canada. Doubleday, Garden City. Amer. Mid]. Naturalist 67:

& D. J. HaGENAH. 1962. Dryopteris in the Huron Mountain Club area of Michigan. Brit-
a 14:90—100.
962. Cytological observations on Adiantum xX tracyi C. C. Hall. Madrofio 16:158—161.
ae, grein The endemic Botrychiums of the southeastern United States. ASB Bull. 9:40.
Morzentl, V. M. & W. H. WacNER Jr. 1962. Southeastern American ‘‘Blackstem spleenworts” of
0-41

Wacner, W. H. Jr. 1962. Plant co —— and leaf production in iene multifidum “‘ssp.
icum’ and ‘forma dentatum.’’ Amer. Fern J. 52:1-18.
. 1962. The synthesis and expression of nou pennies data. Pp. 273, 276, 277 in L. Benson.
1962. Plant Taxonomy: Methods and Principles. Ronald Press, New York.
. 1962. A graphic method for orn inka: based upon group correlations of
indexes of divergence. Pp. 415 n L. Benson. Plant Taxonomy, Methods and Princi-
ples. Ronald Press, New York.
. 1962. Irregular morphological development in hybrid ferns. Phytomorphology 12:87—100.
. 1962. Ferns and their allies. [30-minute film] AIBS Secondary School Biological Sciences
Film Series, McGraw-Hill
962. oe crosses and spore characteristics in the genus Psilotum (Psilotaceae).
ages J.B
962. soiree boschianum in Arkansas. Amer. Fern J. 52:85—86.
Geanen, ri I., W. W. Payne & W. H. WacNner JR. 1962. Botanical phase. Pp. 1-16 in J. M. Sheldon
& E. W. Hewson. Atmospheric Pollution by Aeroallergens, Progress Report No. 5. Univ
sch: Off. Res. Admin. 04202, 04771, 04772, 04773-1-P, Ann Arbor, MI.
Wacner, W. H. Jr. & D. E. RAWLINGS. 1962. A sampling of pins agen Subg. Sceptridium in the
vicinity of Leonardtown, St. Mary’s Co., Md. Castanea 27:132-1
toa ips and taxonomic categories in lower mated plants. Pp. 63-71 in V.
H. Harwue . Léve, eds. Symposium on Biosystematics. Intl. Assn. for Plant Taxono-
my, Utrecht, ie ds.
. 1963. A biosystematic survey of United States ferns—Preliminary abstract. Amer. Fern J.
53:1—16.
. 1963. Gros and herbarium studies, University of Washington, June 23-28, 1963. [Mimeo-
— 12p
5 ee raat 1963. gee prothallial clones (Trichomanes?) locally abundant on Illinois
Reon Amer. J. B 0:623.
1963. Pteridology in oie Newslett. Hawaiian Bot. Soc. 2(8):117-123. [Mimeo-
graphed]

54 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

J. SHARP. 1963. A cnet reduced vascular plant in the United States. Science
142:1483-1484, cover picture
. 1963 [1964]. Pteridaphytes of the concn mye area, Giles County, Virginia, including
notes from Whitetop Mountain. Castanea 28:1

964. Chemistry and taxonomy. [Review = ie ee ed. 1963. Chemical Plant Taxon-
ony Academic Press, New York.] Science 143:1428.
. 1964. vig vce Filicinae. Taxon 13:56-64.
& K. L. Cuen. 1964. [23 new chromosome records] in Love, A. & O. T. Solbrig. IOPB
oe ae eports. I. Taxon 13:99-102
. Fern Geecihen ten in terms of divergence patterns. Abstrs. Xth Intl. Bot. Congr.

a

sie Assessment of species relationships in the Filicineae. Abstrs. Xth Intl. Bot. Congr.
147-14

Evans, A. “ & W. H. Wacner Jr. 1964. me cae i x intermedia—a natural Woodfern
cross of noteworthy morphology. Rhodor. 266.

Wacner, W. H. Jr. 1964. vig ps Pteridophyta an cai 1-15.] Pp. 1, 2, 39-48 i

. Rad-
ford, H. E. Ahles & C. R. Bell. Guide to the Vascular Flora of the Carolinas. he, a North
Carolina, Aran Hill.

Scora, R. W. & W. H. Wacner Jr. 1964. A oo chromatographic study of eastern Ameri-
can pce he Hn Fern J. 54:10

Wacner, W. oa JR. 1964. The neha “een of living ferns. Mem. Torrey Bot. Club
21(5):86—95.

. 1964 [1965]. ees creer Copeland (1873-1964) and his contributions to pteridology.

Amer. senha 177-

. 1965. Edwin Bin any Copeland 1873-1964. Taxon 14:33—41.

. 1965. ietitadies maps of Familes 1-15, Pteridophyta] Pp. 1-6 in A. E. Radford, H. E.
Abiles & C. R. Bell. Atlas of the Vascular Flora of the Carolinas. Univ. No. Car. Tech. Bull.
165.

& K. L. CHEN. 1965. eee o ie as and sporangia as a tool in the detection of Dryo-

eee hybrids. Amer. Fern J. 5

——— & F. S. Wacner. 1965. Rochester area ee Ferns (Dryopteris celsa) and their hybrids. Proc.
yal Sa . Sei. 11(2):59—

STEINER, E., A. S. SUSSMAN & W. H. ia JR. 1965. Botany Laboratory Manual, rev. ed. Holt,
Rinehart ; Winston, New York.

Wacner, W. H. Jr. 1965. Pellaea oo in North Carolina and the question of its origin. J.
Elisha Mitchell Sci. Soc. 81:9

. 1965. Bibliography. pete BioScience 15:809-810.

. 1965. Fern paraphyses: acipusaaig on recent papers. Taxon 1

» D. R. Farrar & K. L. CHEN. 1 ye A new sexual form of Pellaea sails var. glabella from
Missouri. Amer. Fern J. 55:171-178.
& J.B. Sm

IN. 1966. ae ie Sections and Societies. Botanical Sciences (G). Science

151:865—866.

MickeL, J. T., W. H. Wacner JR. & K. L. CHEN. 1966. Chromosome observations on the ferns of
Mexico. Caryologia 19:95-102.

Wacner, W. H. Jr. 1966. Two new species of ferns from the United States. Amer. Fern 1: 56:

sa

1966. Report of the Librarian and Curator for 1965. Amer. Fern J. 56:45-46,

- 1966. New data on North American Oak Ferns, oo eg 68:121-138.

- 1966. Modern Research on Evolution in the Ferns. Pp. 64-184 i . A. Jensen & L. G.
Kavaljian, eds. Plant Biology Today. Advances and os ee ed. : Sn Belmont,

. 1966. gman in the Ferns. Bull. Fairchild Trop. Gard. 21(3):4—8,17-22, cover picture.
& F

Wacner. 1966. Pteridophytes of sek Mountain Lake Area, Giles Co., Virginia: Bio-
repent beste. 3 1964-65. Castanea 31:1
966. The

e new University of Michigan psoas Gardens. Taxon 15:285—286.

LELLINGER: BIBLIOGRAPHY OF WARREN HERBERT WAGNER, JR. 55

. 1966. Illustrations of transient fern forms. Amer. Fern J. 56:101—107.
. 1967. Reports of Sections and Societies. a Sciences (G). Science 155:880.
Eva

, J. T. MickeL, D. B. LELLINGER & A. M. 967. Pteridophytes of Costa Rica: Prelimi-
nary chockliet ad species known to be or ie occurring in Costa Rica. [Mimeographed,
54 as ]

R. C. Woopsipe. 1967. Ferns for an indoor limestone boulder habitat. Amer. Hort. Mag.
oa 219-223
. E967. Rexiow of: R. L. Taylor & R. A. Ludwig, eds. 1966. The evolution of Canada’s flora.
Univ. of Toronto, Toronto. Canad. Field-Naturalist 81:206—207.
. 1968. Hybridization, taxonomy, and evolution. Pp. 113-138 in V. H. Heywood, ed. Mod-
ern Methods in Plant Taxonomy. Academic Press, London

. L. McWituiaMs. 1968. The University of Michigan Botanical Gardens special collec-
tions. Arb. & Bot. Gard. Bull. 2:43-46.
. 1968. recreate stamens and carpels of a willow from northern Michigan. Michigan
Bot. 7:113—120.

196

ge taxonomy and modern systematics. BioScience 18:96—100.

Farrar, D. : & W. H. WAGNER Jr. 1968. The gametophyte of voter holopterum Kunze.
Bot. Gaz. 129:210-219.

Wacner, W. H. Jr. 1968. Review of: R. H. Mohlenbrock. 1967. The Illustrated Flora of Illinois:
Ferns. So. Ill. Univ. Press, Carbondale. Quart. Rev. Biol. 43(4):461

———. 1968. The role of a botanical garden in a modern marae Scietiee 162:1165-

. 1969. The construction of a classification. Pp. 67-90 in Systematic Biology. or 1692
cit the National Academy of Science, A erscnts DC. (Translated by E. R. de la Sota and
published as La Construccién de una Clasificacién. Quid de la Ciencia, la Tecnologia y la

Educacion 2(18):440-448. 1983, haa 532-538. 1984.]
oe, es Richard Eric Ho sean sagen kgeray pteridologist. Amer. Fern J. 59:1-3.
——.,F.S. “Wa GNER & D. J. Ha . 1969. The Log Fern (Dryoptreris celsa) and its hybrids in
sen ig preliminary a Michigan Bot. 8:137—145.
oe . Bio —— important is it to basic evolutionary systematics? Abstrs.
xh a Bot. Congr. 230.
F. S. WacneR. 1969. A new natural hybrid in the Appalachian Asplenium complex and
its taxonomic significance. Brittonia 21:178-1
. 1969. The role and taxonomic treatment of ntti. BioScience 19:785-—789.
. 1970. . The Barnes pyres — Populus X barnesii, hybr. nov—A nomenclatural case
in eal Michigan Bot. 9:53—
——. 19 Sg-te sane wad evo ia ionary noise. Taxon 19:146—151.
. 1970. Natural oe of floating stems of Scouring-rush, Equise-
tum hyemale. ‘Miche Bot. 9:166-17
, D. R. Farrar & B. W. McALpPIN. oa epee ae _ eee Biological Station
area, ees ee J. Elisha Mitchell Sci.
& L. D. Gomez P. . Field research on ferns in co ae Ophioglossum, Dictyoxi-
phium, Geetha rae 10-12, 1971. iaiwnsaeiepties 4pp.]
1971. Evolution of naires in ia to the a Virginia Polytech. Inst.
State Univ. Res. Div. Mongr. 2:147—19
1971. The Sonthetaions Adder’s- —— Ca epee vulgatum var. pycnostichum,
ig for the first time in Michigan. Michigan Bot.
971. Ferns in biology: Some final comments. nose 21:311, 323-324.
nt Report of the President for 1970. Amer. Fern Soc. News & Views 2: —
. 1971. Lindsaea (Schizoloma) ae Swartz in Hawaii. Amer. Fern J. 61 58.
Wuirtier, D. P. & W. H. WaGNER JR. 1971. The variation in spore size and pre he in Dryo-
pteris taxa. Amer. Fern J. 61: ofeagee
1972. Botanical research at botanical gardens. Pp. 28-33 in P. F. Rice, ed. agen,
of the Symposium—A National Psion Garden System phi Canada. Techn. Bull. N
Royal Botanical Gardens, Hamilton, Ontario, Canada.

56 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

snes 2 iy by & W. H. Wacner Jr. 1972. A native Woodfern garden. Amer. Horticulturist

51(1
WAGNER, ae 7 Jr. 1972. Estimation of the evolutionary ane Bibliographic references
—— to the en gga Method. [Mimeographed,

Pp.

Cain, S. A. & W. H. WaGNER ash Dale J. Hagenah (1908-1971): An outstanding Michigan
botanist. Michigan Bot. 1 ae

Wacner, W. H. Jr. 1972. io pe the President for 1971. Amer. Fern Soc. News & Views 6:1—4.

. 1972. Dale J. Hagenah. Amer. Fern J. 62:2

, F. S. Wacner & L. D. cia P; 1972. The Cantal American fern genus Pleuroderris: its

— yr significance r. J. Bot. 59:677.
eo a nonst a little-known fern epiphyte with dimorphic stems. Amer.
ae 62:3

oar 1°

. 1972 eal Review of: R. M. Lloyd. 1971. perenne of the Onocleoid ferns. Univ.

Calif. Publ. Bot. 61:1-93. Bull. Torrey Bot. Club 99:

. 1972 [1973]. Disjunctions in homosporous a plants. Ann. Missouri Bot. Gard.

59:203—217.

& D. R. Farrar. 1973. The oo fern genus Hyalotricha and its family relationships.

lage J. Bot. 60(4, suppl.):3

973. An orchid asc for monarch butterflies (Danaidae). J. Lepidopterists’ Soc.

27: apg

. 1973. Reticulation of Holly shea ee in the western United States and ad-
janes er ia mer. Fern J. 63

973. Some future Pee of pes systematics and phylogeny. Pp. 245-256 in A. C.

Jermy, J. * Crabbe & B. A. Thomas, eds. 1973. The Phylogeny and Classification of the
Ferns. Bot. J. Linn. Soc. 7 Suppl. 1

Tuomas, R. D., W. H. Wacner Jr. & M. R. Masta 1973. Log Fern (Dryopteris celsa) and related
species in Louisiana. Castanea 38:269—27

Wacner, W. H. Jr., F. S. WAGNrR, J. A. ata & J. F, MaTTHEws. 1973 [1974]. Asplenium monta-
num X platyneuron, a new primary member of the A i te < ltiaiail complex from
useing Mountain, N.C. J. Elisha Mitchell Sci. Soc. 89(3):218-223.

——. 1974. es hackberry (Ulmaceae: Celtis tenuifolia) in the a Lakes region. Michigan
tg 13:73-9

MEsLeR, M. R. & 4 H. Wacner Jr. 1974. The venation of Ophioglossum dendroneuron and its
systematic significance. Amer. J. Bot. ey Suppl.):37

Wacner, W. H. Jr., J. T. Micke. & L. D. z P: 1974. a dubium and its bearing on
the penis of generic relationships ey Cyrtomium and Phanerophlebia. Amer. J. Bot. 61(5,
Suppl.):39

. 1974. The classification of leaf types of land plants. Amer. J. Bot. 61(5, Suppl.):67.
- 1974. Fern. Pp. 237-248 in Encyclopaedia Brittanica, ed. 15, vol. 7. H. H. Benton,
Chicas, HL.
Hit, R. H. & W. H. hiiggen ag 1974. Seasonality and spore type of the pteridophytes of Michi-
gan. Michigan Bot. 13
Wacner, ae H. Jr. 1974. aa years of botany 1947-1972: Introduction. Ann. Missouri Bot. Gard.
61:1-—
pe 25 years of botany 1947-1972: Pteridology. Ann. Missouri Bot. Gard. 61:86—111.
- 1974. Structure of spores in relation to fern phylogeny. Ann. Missouri Bot. Gard.
61:332-—353.
Dancik, B. P., B. V. BARNES & W. H. Wacner Jr. 1974. Aberrant pistillate catkins of Betula alleg-
haniensis. Michigan Bot. 13:177-179
Wacner, W. H. Jr. 1975. Review of: Hirabayashi, = sexi Cytogeographic Studies on Dryopteris
of Japan. Harashobo, Tokyo. Amer. Fern J. 6
1975. Notes on the floral biology of ci sai ane negundo). Michigan Bot. 14:73-82.
1973 | [1975]. Plant species and ecosystems: a botanist’s viewpoint. Trans. Missouri Acad.
7/8:40.

- 1975. The spoken “x” in hybrid binomials. Taxon 24:296.

LELLINGER: BIBLIOGRAPHY OF WARREN HERBERT WAGNER, JR. 57

& F. S. WacneR. 1975. A hybrid polypody from the New World tropics. Fern Gaz. 11:125—
135.

Wiptn, C.-J., D. M. Brirron, W. H. WaGNER JR. & F. S. WAGNER. 1975. Chemotaxonomic studies on
hybrids of neh sea: in eastern North America. Canad. J. Bot. 53:1554—1567.

WAGNER, ke H. Jr. 1975. A bunchberry ‘‘Last Rose of Summer.”’ gen Bot. 14:201-—202.

5. Sex he the angiosperms—another proposition. Sida 6
; rp Review of: G. L. Stebbins. 1974. Flow st Plants: ae Above the Species
Level. Belknap Press, Harvard Univ., Cambridge, MA. Amer. Sci. 702-703

oi J. D., W. H. WaGNeER Jr., D. R. FARRAR & S. W. LEONARD. 1975. Asano sea
in the nga Biological Station area, southern Appalachians. reve 40

Wacner, W. H. Jr. 1976. How to find the rare grapeferns and moonworts. thee e oaks
3(2):2.

& F. S. WacNER. 1976. The role of foliar "ial in the systematics of ferns. Bot. Soc.
Amer. Abstr., Tulane Univ., New Orleans.
L. D. Gomez & W. H. WAGNER JR. 1976. An uk triploid vie ae polypody from the
vicinity of Cartago, Costa Rica. Bot. Soc. Amer. Abstrs., Tulane Univ., New Orleans. 41-42
Wacner, W. H. Jr. & F. S. WacNneR. 1976. Asplenium ee een Sim ‘rom gues Gorge,
Greene County, Ohio—a second North American record. Ohio J. Sci. 76:9
& D. J. SCHOEN. ape yi Oak (Quercus imbricaria) and its hybride in Michigan.
Michigan Bot. 15:14
— & W. C. TAyLor. api yaa x leedsii and its westernmost station. Sida 6:224—234.
A. H. SHowa ter. 1976. Ecological notes on Celastrina ebenina (Lycaenidae). J. Lepi-
ie ong Soc. 30:310-312.
. Systematic implications of the Psilotaceae. Brittonia 29:54—-63.
NER. 1977. A diminutive grapefern (Botrychium) of the maple—basswood for-
ests in 1 the Wpper Great Lakes region. Bot. Soc. Amer. Misc. Ser. Publ. 1
& D. . 1976 [1977]. a Central American fern genus Hyaloticha aa its family
Seeneikieeks Syst. Bot. 1:348-36
Voss, J. H. BEAMAN, E. A. es F. W. Case, J. A. CHURCHILL & P. W. THOMPSON. 1977.
Bnduneoced, threatened, and rare vascular plants in Michigan. Michigan Bot. 16:99-110.
——., F. S. Wacner & L. ~ — P. 1977. An enigmatic polypody fern from Cartago, Costa
Rica. igegs 12/13:8

——. 1977. A distinctive ae form of be Marbled White Butterfly, Euchloe ae (Lepi-
dope bern in the Great Lakes area. Great Lakes Entomologist 10:107-1
. S. Wacner. 1977. Fertile- amet leaf dimorphy in ferns. Gard. Bull. ai Settlem.

sosi-267
, F. S. Wacner, C. N. MILLER Jr. & D. H. WacNeR. 1978. New observations on the Royal
Hern iii Osmunda x ruggii. Rhodora 80:92-106.
— & T. L. MELLICHAMP. 1978. Foo — habitat, and range of Celastrina ebenina (Lycaeni-
— - Lepidopterst Soc. 32:20—
eer Great om anita Pieris virginiensis nae Pieridae) in com-
prion siti outhern counterpart. Great Lakes Entomologist 11
——. 1978. Venlo idioblasts in Pteris and their systematic Sag Acta Phytotax.
ane 29:3
. 1978. ea of eo veins in modern ferns. Bot. Soc. Amer. Misc. Ser. 156:30
— & K. E. Boypston. 1978. warf coastal variety of maidenhair fern, Adiantum pedatum.
Canad. J. Bot. oe
. 1978. A probable natural ap tic of ao eurymedon and P. rutulus (Papilionidae) from
{dale J. ass eee Soc. 32
. WaGNER & L. D. GO

es 978. The singular origin of a Central American fern,
eae bee mic aes ee aes 10:254-264.
1978 [1979]. Hyalotrichopteris, a new generic name for a Central American polypodioid
fern. Taxon 27:548.
M. BEITEL. 1979. oreo variation in clones of Running Pine, Lycopodium flabelli-
forme. Michigan Bot. 18:19—

AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

- 1979. Reticulate veins in the systematics of modern ferns. Taxon 28:87—95.
, L. D. Gomez P. & F. S. Wacner. 1979. Coenosori and foliar nen in New World Pleo-
peltis and Marginariopsis. Bot. Soc. Amer. Misc. Ser. Publ. 157:48—49.

& F.

. 1979. [Accounts of i silenare or pre ae or extirpated Senet
phytes.] $0: 35-41, 108-110 in D. W. L

, LMAN, sa how Ferns (Dryopteris celsa) and their relatives in the Dismal
Swany Pp. 127-139 in P. W. Kirk Jr., ed. The Great Dismal Swamp. Univ. Press of Virginia,
Oostinc, D. P., D. J. HARVEY & W. H. WAGNER JR. 1979. Euristrymon ontario (Lycaenidae): first re-
port in Michigan. J. Lepidopterists’ Soc. 33:151-152.
Wacner, W. H. Jr. 1979. New herbarium building in Costa Rica. Taxon 28:358.
» F. S. Wacner & J. M. BerreL. 1979. An unusual occurrence of Asplenium heterochroum.
Canlele a 174-177.
SmiTH, A. R., W. H. WaGNeR Jr. & T. DUNCAN. 1980. Noteworthy Collections. sae ape lusita-
nicum L. ee i a Apr, Clausen (Ophioglossaceae). Madrofi
Wacner, W. H. & F. S. Wacner. 1980. Polyploidy in Pteridophytes. Pp. mean in an H.
aes ed. ae Popo. Biolog Relevance. Plenum Press, New York & London.
1980. Review of: R. L 8. A acne ‘eae of Equisetum subg. Equi-
peteati: Nova Hedwigia 30: eae Michigan Bot. 1
- 1980. A probable new hybrid
central pera Michigan Bot. 1
, T. F. DANIEL & M. “ HANSEN. vel o
Michigan Bot. ie 37-4
Wa ter, K. S., W. H. meet jr. & F

— cesta matricariifolium x simplex, from

ybridizing Verbascum population in Michigan.

- S. Wacner. 1980. Ecological, biosystematic, and nomencla-

tural oo on Scott’s Splocawact, x Asplenosorus ebenoides. Bot. Soc. Amer. Misc. Ser.
Publ. 158:123.

Wacner, W. a jn & M. K. Hansen. 1980. Size reduction southward in Michigan’s Mustard White
Butterfly, Pieris napi (Lepidoptera: Pieridae). Great Lakes Entomologist 13:77-88

——. 1980 . Dyer, ed. 1979. The Experimental Biology of Pan. Academic
Press, — York. Econ. Bot. 34:305-306.

. MAYFIELD. 1980. Foodplants and cocoon construction i

(Lepidoptera: Saturniidae) i in southern Michi

ussion pp. 222, 224, 227, 229. Reprinted as pp.

. F. eds. 1985. Chadinte Theory and Methodology. Van
Nostrand eSnips New York.]

Barnes, B. V. & W. H. WAGNER Jr. 1981. Michigan Trees. A ned to vad of Nein and
the Great Ladue Region. Univ. of Michigan

THOMPSON, J. W. & W.

WAGNER, w. H. a a F. Dani a "vs EITEL. 1980 [1981]. Studies on Populus heterophylla in
southern Michigan. Michigan Bot. 19:269—
Duncan, T., R. B. PHitups & W. H. AGNER, JR. ti [1981]. A comparison es —e diagrams
derived by various — and cladistic methods. Syst. Bot. 5:264—
Wacner, W. H. GNER. 1981. New s
chium (phiogonsicon) from —) Ameri
D.M SON.

.. W. H. Wacner Jr. & K. S. Wa ter. 1981. Unu

usual frond developmen in the Sensi-
tive Fern Onoclea sensibilis L.. Amer. Midl. Naturalist

105:396—400

LELLINGER: BIBLIOGRAPHY OF WARREN HERBERT WAGNER, JR. 59

Wacner, W. H. Jr., M. K. HANSEN & M. R. MayYriELD. 1981. True and false foodplants of Callosamia
paced (Lepidoptera: Saturniidae) in southern Michigan. Great Lakes Entomologist

14:159-16
fee Fems in the Hawaiian Islands. i wniengce Forum 8:43—4
——, F. S. Wacner, S. W. LEONARD & M. R. 981. A ts ai of Ophioglossum

Bsn “ 'p. St. John. Castanea 46: ne on

———. 1982. Ferns, Clubmosses, Spikemosses, Quillworts, and Horsetail. Pp. phe gt in -S.'].
yeaa pe — North American Wildlife. Readers’ Digest Assn., Pleasantville, N

—-. F. AGNER. 1982. The taxonomy of Dryopteris x poyseri Wherry. pt Bot.

21; a :
Funk, V. A. & W. H. WaGNER Jr. 1982. A bibliography of botanical cladistics: I. 1981. Brittonia

sages ere
Wacner, W. H. Jr. . S. PEIGLER. 1981 [1982]. Two notable range extensions for Callosamia
securifera lovers in Georgia and ane Carolina. J. Lepidopterists’ Soc. 35:247
F. Wacner & C. HaAuFLeR. 1982. A hybrid population hae the ‘‘all- fertile”

Paths paradoxum and the hemidimerphic B. hesperium (Ophioglossaceae). Bot. Soc.

Wa ter, K. S., W. H. WaGNER JR. & F. S. WAGNER. 1982. erie biosystematic, and nomencla-
tural notes on Scott’s Spleenwort, <x Asplenosorus ebenoi mer. F . 72:65-75.
Wacner, W. H. M. BeIreL & F. S. WAGNER. 1982. Complex niniees bits in the leaves of
puidecine Mogaphs -like leaves in Lycophytes. Science 218:793—794.
CARLSON, T. M. & W. H. Wacner Jr. 1982. The North American distribution of the genus Dryopte-
s. Contr. “pir Michigan Herb. 15:141—-162.
raetens W. H. Jr. & F. S. WAGNER. 1982. Botrychium rugulosum (Ophioglossaceae), a newly rec-
ognized species of evergreen Grapefern in the Great Lakes area of North America. Contr
Univ. Michigan Herb. 15:315-324
& C. E. NAUMAN. mii setipa tdalhampell, a spontaneous fern hybrid from southern
Florida. Amer. Fern J. 7
. 1982 [1983]. Edgar T. eo ian 1885-1982. Bull. Torrey Bot. Club 109:545-548.
& F. S. Wacner. 1983. Genus communities as a systematic tool in the study of New World
Botrychium (Ophioglossaceae). Taxon 32:51-6
. 1983. Bibliography of Edgar T. Wherry a 49:6-14.
. 1983. Plant speciation: the classical approach. [Review of: V. Grant. 1981. Plant Specia-
ae! = 2. Columbia Univ. Press, New York] Evolution 37:426-427.
3. Edgar T. Wherry and his contributions to pteridology. Amer. Fern J. 73:1-5.
———, ae SMITH & T. R. PRAY. 1983. A pou ee hybrid, Pellaea bridgesii < mucronata, and
its age ic significance. Madrofio 30
. 19 chee stics: The recognition af ae and their role in cladistics and classifica-
tion. pale 3-79 in N. I. Platnick & V. A. Funk, eds. 1983. Advances in Cladistics, vol. 2.
ee Univ. Press, New York.
M. JOHNSON. 1983. Trophopod, a commonly overlooked storage structure of potential
systematic value in ferns. Taxon 32:268-269.
983. Homology and the early diversification of vascular plants. Amer. J. Bot. 70(5, part

& F. S. WacNeR. 1983. Two Moonworts of the Rocky Mountains; Botrychium hesperium

and a new species formerly confused with it. Amer. Fern J. 73:53-62.

. 1983. Adiantum concinnum Sana rent cones Common Maidenhair Fern), ir
prceene (Helecho Lengua, Paddle Fern, Hart’s Tongue). Pp. 187, 239 in D. H. Janzen, ed.
Costa Rican i History. Univ. Chicago Press

—— & L. D. Gomez. 1983. Pteridophytes te Teens Pp. 311-318 in D. H. Janzen, ed.
Costa Rican Retinal History. Univ. Chicago Pr

MayFIELD, M. R., M. C. Cote & W. H. WAGNER Jr. Seat Ricciaceae in Michigan. Michigan Bot.
22:145—-150.

MavFIELD, M. R., M. C. Cote & W. H. WaGNeER JR. 1983. Techniques for preparing riccias for herba-
rium study. Taxon 32:616-617

60 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

Wynne, M. m W. H. Wacner Jr. & C. B. Beck. 1984. Botany at the University of Michigan. Pl. Sci.
Bull. 30:10-11
Wacner, W. Bi Jr. 1984. Review of: E. Hennipman & M. C. Roos. 1982. A monograph of the fern
genus Poaaah dei al Konin. Ned. Akad. Watecachn, Afd. Natuurk. II, 80:
1-126. Amer. Fern J
. 1984. Review of: a vic umpkin & D. L. Plucknett. 1982. Azolla as a Green Manure. Use
and Management in Crop Production. Westview Press, Boulder, Colorado. Econ. Bot. 38:260.
- 1984. Applications of the concepts of Groundplan-Divergence. Pp. 95-118 in T. Duncan
& T. F. Stuessy, eds. Cladistics: Perspectives on the Reconstruction of Evolutionary History.
Columbia Univ. site New York.
4. A comparison of taxonomic methods in Sy aime Pp. 643-654 in W. F.
Grant, ed. 1984. Plant a iy Academic sii Canada, Don Mills, Ontar
——, F. S. Wacner, C. HaurLer & J. K. EMERSON. 1984. A new si tates 0 a scaled
(Ophioglossaceac, Botrychium). Corned. J. Bot. 62: ee
. 1984. Systema ic problems in the fern genus Polystichum. Amer. J. Bot. 71(5, part 2):140.
. 1984, Review of J. H. Peck. 1982. Ferns and Fern Allies of the Driftless Area _ —
Iowa, Minnesota, eos 1S aig Milwaukee Public Mus. Contr. Biol. Geol. 0.
1982. Michigan Bot.
& E. ROuLEAu. 18, a‘ western Holly Fern, Polystichum x scopulinum, in Newfoundland.
Amer. Fern J. 74:3
——,, C. M. ALLEN & Ae i Tatas: 1984. Saks ae a Hook. & Grev. in Louisiana
and the taxonomy of O. nudicaule L. f. Cas a 49:9
——, F. S. Wacener & J. M. BEITEL. 1985. Ev rene for een hybridisation in pterido-
phytes i subterranean mycoparasitic gametophytes. Proc c. Roy. Soc. Edinburgh, ser. B,
86:273-2
& V. QuEvEDO. 1985. Polymorphism in ae a — Sw. and the taxonomic sta-
tus of A. germanii (Fée) Prantl. Amer. J. Bot. 72(6):9
- 1985. Morphological variation sets hipaa in Pie ele Amer. Fern J. 7
. 1985. Winter Grapefern. P. 13 P. C. H. Pritchard, ser. ed. 1985. Rare and a
Biota of Florida, vol. 5 Plants. eae Presses of ie Gainesville.
985. Review of: A. E. Salgado. 1982. Venation pattern in Philippine oe ferns.
De La Salle Univ. Monogr. Ser. 3:1-87. Pilipinas (J. Philipp. Stud.) 5:165-16
. 1986. Flying flowers! Butterflies and their foodplants. LSA (Publ. of the feles of Litera-
ier Science, and the Arts, Univ, of M Mich.) 9:4—9, cover.
986. Review of: G. R. Proctor. 1985. Ferns of oo A Guide to the Pteridophytes.
ice Museum pe ayes mC London. Syst. Bot. 11
J. M. Berret. mparative gai of the i aa ferns.
Amer. J. ‘Bot. 65, pea 732-7
Farrar, D. R., W. H. Wacner JRr., N. R. an & C. L. JoHNsoN-Gron. 1986. Subterranean sporo-
phytic vised in neg oe subg. Botrychium. Amer. J. Bot. 73(5, on 735-736.
—, = 4: JR. 1986. Japanese honeysuckle invasion. Michigan Bot. 2571
6. The New World fern genus Marginariopsis (Polypodiaceae): Pe example “4 leaf di-
morphy suse in generic delimitation. Bull. Torr trey Bot. Club 113:15
& Wacner. 1986. Thre ® new species of Moonworts (Botrychium nig nes
ao in western North erica. Amer. Fern J. 76:33—47.
» F. S. WAGNER & W. C. TAYLor. 106, 7 gti abortive spores in herbarium specimens of
sterile one se nergy 76:1
Wac

=
co

Paris, C. A \., F. 38 Pain’ <5 eee Cryptic species, species definition, and taxo-
nomic Ginaetics es ca homosporous ferns. Amer. J. Bot. 74(5, abstracts):7
WAGNER, a R. The generic concept in pteridophytes. Amer J. Bot. 74(5, Seog oo

menetict Bot.
1987. ae Pp. 137-143 in S. B. Parker, vol. ed. 1987. McGraw-Hill. Encyclopedia
= en and Technology, ed. 6, vol. 18. McGraw- Hill, New Yor

- 1987. Some questions about natural hybrids in ferns. Bot. Helvatica 97:195-205.

LELLINGER: BIBLIOGRAPHY OF WARREN HERBERT WAGNER, JR. 61

——. 1988. Review of: F. W. Case, Jr. 1987. Orchids of the Western Great Lakes Region. Cran-
brook Inst. Sci. Bull. 48. Jack-Pine Warbler 66:39—40.

. 1988. Status of the Hawaiian fern flora. Fiddlehead Forum 15(2):11—
. 1988. ciehere eye vs. divergence in the morphology of Hawaiian a inka Amer. J.
Bot. 75(Suppl.):144.

& F. aaa AGNER. 1988. Detecting Botrychium hybrids in the Lake Superior region. Michi-
gan Bot. 27:75-80.
When can you find them? American ferns and fern allies and their phenologies.
Fiddichead Forum 15:26-27.

988. Threatened and endangered (ferns and fern allies of North America). Fiddlehead

sti 15:32—34.
, S. TAyLor, G. Grieve, R. O. Kapp & W. K. Stewart. 1988. Simple-leaved Fraxinus (Fraxi-
nus: Oleaceae) in Michigan. Michigan Bot. 27:119-134.

. 1988. Herb Wagner, President of IAP, summarizes the programme. IAP News 4(1):1.

& M. K. HANSEN. 1989. An Angulifera dees Kentucky Lepidopterist 15(1):2.

. 1989. Kathryn E. Boydston ponies 8): Michigan’s fern hybridist and two new exam-
lis of her work. Michigan Bot. 2
Zou, Xiaoming & W. H. WAGNER Jr. Pring hank A preliminary review of Botrychium in China.

mer. Fern J. 78:122-135.
Wacner, W. H. Jr. 1989. Golden anniversary of the first Appalachian Spleenwort triangle. Fiddle-

BEITEL, J. M. & WaGNER, W. H. Jr. 1989. fora American lycopsids and their generic delimita-
i rg

A
Wacner, W. H. Jr. 1989. The role of new ae of correlating data in pteridology. Pp. 17-20
in K. H. Shing & K. U. Kramer, eds. Proceedings of the International Symposium on Sys-
tematic Pteridology. China Science & Technology Press, Beiji mer
. 1989. Phylogeny and cladistics in pteridology. Pp. 21-28 in K. H. Shing & K. U. Kramer,
eds. Proceedings of the International Symposium on Systematic Pistia: China Science
& Technology Press, nag ge
& E ENAH. 1989. A synthetic —— spa x Asplenosorus pinnatifidus X
Phys. scolopendrium var. americana.
Paris, C. A., F. S. WAGNER & W. H. WAGNER - 1989. Cryptic species, —— delimitation, and
heen practice in the homosporous ferns. Amer. Fern
Wacner, W. H. Jr. & B. G. SCHOLTENS. 1989. Michigan’s ee ae An appeal to
birders. poomea Audubon News 37(4):5
. 1989. Botrychium socorrense W. Wagner, sp. nov. Pp. 20-21 in G. A. Levin & R. sige
989. The Vascular Flora of Isla Socorro, Mexico. San Diego Soc. Nat. Hist. Mem. 1
ae M. . Moran. 1989. Lycopodium hickeyi: A new species of North pe
Chakenmes. Amer. font 79:119-121
. 1989. Hartford Fern: First tuted plant in America? Fiddlehead Forum 1
& T. Devine. 1989. Moonworts (Botrychium: +e aaa in the =e oo area, Butte
and Tehama Counties, California. Madrofio 36:1
. 1990. Ophioglossaceae. Pp. 193-197 in K. U. ak P. S. Green, eds. The Families and
Genera of Vascular Plants, vol. 1: Pteridophytes and Gymnosperms. Sco gina Berlin.
a i Wetland butterflies in Michigan. Michigan Entomol. Soc. Newslett. 35(3):1
90. Hawaii’s satchel-sorus tree ferns, Cibotium species: What is their aan
on Fiddlehead Forum 17:7-8.
& F. S. Wacner. 1990. Mocncwrnicts (Botrychium subg. Botrychium) of the upper Great
Lakes region, U.S.A. and Canada, with descriptions of two new species. Contr. Univ. Mich-
igan Herb. 17:313—325.
. More on the Hartford Fern. Fiddlehead Forum 17
& FP. S. bien 1990. Another nothospecies in the fede Asplenium complex.
Amer. Fern J. 80:44—49.
. 1989 [1990]. spo of: W. J. Cody & D. M. Britton. 1989. Ferns and Fern Allies of Cana-
e* a. Canadian Govt. Publ. Centre, Ottawa. Naturaliste Canad. 116:213-214.

62 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

. 1990. cigrieldlig issues from the Pacific standpoint. Abstrs. Papers 156th Ann. Meet-

ing, Amer. Ass
. 1990. Biclogical diversity Underlying concepts. Michigan Academician 22(4):311-317.
—— &F. S. Wacner. 1990. Notes on the fan-leaflet group of Moonworts in North America

with descriptions i two new members. Amer. Fern J. 80:73-81

Werth, C. R. & W. H. WaGnerR Jr. 1990. (13) Proposal to designate reproductively competent spe-
cies of hybrid origin by an x placed in brackets (reword Article H.3, Note 1). Taxon
39:699-—702.

MIcKEL, J. T. & W. H. WAGNER Jr. 1991. Joseph M. Beitel (1952-1991). Brittonia 43:123-125.
Wacner, W. H. Jr. 1990 [1991]. A natural hybrid of Gray Dogwood, Cornus racemosa, and
Round-leaved Dogwood, C. rugosa, from Michigan. Michigan Bot. 29:131-136.
F. S. WaGNER. 1991, Hawaii as an example of the status of Pacific Island pteridophytes.
XVIIth Pacific Sci. Congr. Abstrs. 146.
MickeL, J. T. & W. H. WacNeER Jr. 1991. Joseph M. Beitel (1952-1991). Fiddlehead Forum 18:

4, 17.

Wacner, W. H. Jr. 1. New examples of the Moonwort Hybrid, Botrychium matricariifolium x
simplex Pan Side ae Canad. Field-Naturalist 105:91—94.

MICKEL, . H. WaGNER JR. & E. M. Girrorp. 1991. Ferns and other lower vascular plants.
Pp. 163-178. The New Encyclopaedia Ruttaniea, ed. 15, vol. 19. Encyclopaedia Brittanica,
Chicago, IL.

Wacner, W. H. Jr. & F. S. WacNER. 1991. The Ophioglossaceae of Mexico. Amer. J. Bot. 78(6,
suppl.):149-150.

Morin, N. R., W. H. WaGcner Jr. & A. R. SMITH. 1991. Report on Flora of North America I. Amer.
J. Bot. spe suppl.):204

Wacner, W. H. Jr., E. M. Fuss: C. R. WertH & R. L. Bartcis. 1991. First records of —
leaved Relioomuruct, Asplenium x alternifolium, in the New World. Castanea 56:128—13

. 1991. i glandular Cinnamon Fern. Fiddlehead Forum 18:2

BRuceg, J. G., H. WAGNER Jr. & J. M. BEITEL. 1991. sh new species a hep eno gah Lycopo-
diella Lscopodincne from southwestern Michigan. Michigan Bot. 3

ag 3 3 . H. WAGNER JR. 1991. An eagiiceal leaf of Pees ee. Tete Amer.

Wacner, W. H. Jr., F. S. Wacner, A. A. REZNICEK & C. R. WERTH. 1992. senate singulare
Dryoteridacea a new fern nothogenus from Ontario. Canad. J. Bot. 70:24 hp

2. Phytogeography of Hawaiian peridphstes Amer. J. Bot. 79(6, te tae

ese Fy . WAGNER JR. & J. M 1992. Pacific Firmoss (Huperzia si ae
(Lycopodiaceas) in eastern North seins at Gros Morne National Park, Newfoundland.
Amer. Fern J. 82:63-66.

Wacner, W. H. Jr. & J. M. BEITEL. 1992. Generic classification of modern North American Lyco-
podiaceae. Ann. Missouri Bot. Gard. 79:676-686.

1992. Hiemobotrychium, a new section of Botrychium subgenus Sceptridium from the

southeastern United States. Novon 2:267-268.

- 1993. Ophioglossaceae. Adder’s-tongue Family. Pp. 98-100 in J. C. Hickman, ed. The

Jepson Manual. Higher Plants of California. Univ. of California Press, a

- 1993. Problems with biotic invasives: a biologist’s viewpoint. Pp. 1-8 in B. N. McKnight,

ed, ri Biological Pollution: The Control and Impact of Invasive Paes Species. Indiana

Academy of Science, Indianapolis, IN.

- 1993. New species of Hawaiian Pteridophytes. Contr. pent ea Herb. 19:63-82.

. 1993. Fern — and allied plants. Fiddlehead Forum 2
A. ITH. 1993. Pteridophytes of North America. ete ae in Flora of North

America Editorial Committee, eds. 1993. Flora of North America North of Mexico, vol. 1.
Oxford Univ. Press, New York.

Taytor, W. C., W. H. Wacner Jr., R. W. Hospy & F. R. WARSHAUER. a Isoétes hawaiiensis: a
previously untleecribed Quillwort from Hawaii. Amer. Fern J. 8

=e, LORD,

:67—70.
rns and fern allies of Kauai. Bull. Natl. Trop. Bot. mia 23(3):51-54.
. 1993. . new combination for a North American lycopod. Novon 3:305.

LELLINGER: BIBLIOGRAPHY OF WARREN HERBERT WAGNER, JR. 63

Montcomery, J. D. & W. H. WaGNer Jr. 1993. Dryopteris Adanson. Pp. 280—290 in Flora of North
erica Editorial Committee, eds. 1993. Flora of North America North of Mexico, vol. 2.

Oxford Univ. Press, New York.

Wacner, W. H. Jr. 1993. Schizaeaceae Kaulfuss. Curly-grass Family. Pp. 112-113 in Flora of

North America Editorial etna eds. 1993. Flora of North America North of Mexico,
vol. 2. Oxford Univ. Press, New York.

J. M. BEIrEL. 1993. Lyco taihnd Mirbel. Club-moss Family. Pp. 18-37 in Flora of
North America Editorial Committee, eds. 1993. Flora of North America North of Mexico,
vol. 2. Oxford Univ. Press, ork.

& F. S. WaGNER. 1993. Pek pas C. Agardh. Adder’s-tongue Family. Pp. 85-106 in
Flora of North America Editorial Committee, as 1993. Flora of North America North of
Mexico, vol. 2. Oxford Univ. Press, eee rk.

, R. C. Moran & C. R. WERTH. 1993. Aspleniaceae Newman. Spleenwort Family. Pp. 228-
245 in Flora of North America Editorial Committee, eds. 1993. Flora of North America
North of Mexico, vol. 2. Oxford Univ. Press, yes be

Mepeiros, A. C., R. W. Hoppy & W. H. WAGNER JR. 1 Stee on the rediscovery, status, and

ecology of the very rare Hawaiian fern Gees pene (Thelypteridaceae). Amer. Fern J.

83:86—89
Wacner, W. H. Jr. & F. S. WAGNER. 1994. Another widely disjunct, rare and local oo American
coe (Ophioglosacee Botrychium subg. Botrychium). Amer. Fern J. 8

& F. 4. Morphological trends in Ophioglossales. pes + Bot. 81(6,

Aas

. 1995. Evolution : Hawaiian ferns and fern allies in relation to their conservation status.

Pactic Sci. 49:31

5 Bs _ WAGNER cz Fin N. ee Taxonomic notes on the Pteridophytes of Hawaii. Contr.
cals. Michigan Herb. 20:241—2

SCHOLTENS, B. G. & W. H. WAGNER JR. om [1995]. Biology of the er Hemileuca (Lepidoptera:
Saturniidae) in pb ae Great Lakes ageing 2754

Wacner, W. H. Jr. 1995. ge nore Pp. 44-48 in R. C. Moran: &R. — eds. 1995. Flora
ra a vol. 1. Universidad Nal. Aut6noma de México, Méxic
. 19 e and sf cua Michigan pteridophytes. Michigan pn 28:207-208.

1996. adie vs. marcescent ferns and fern allies. Fiddlehead Forum 23:32-33.

Mepelros, A. C., W. H. WAGNER JR. & R. W. Hospy. 1996. A new Hawaiian hanging Firmoss (Lyco-
podiaceae: Spnctbagt es om the ene Hawaiian Islands. Amer. Fern J. 86:89-97.

Wacner, W. H. Jr. 1997. ona enon 35-36 in J. Boggan, V. Funk, C. Kelloff, M. Hoff,
G. Cremers & = Feuillet, eds. 1997. Checklist of the Plants of ai cena ed. 2. Biological
Diversity of the Guianas ae erecta Institution, Washington

ScHOLTENS, B. G. & W. H. Wacner Jr. 1997. An example of clinal vari odin in eastern North

merican Buckmoths (Saturniidae: Hemieuc J. Lepidopterists’ Soc. 51:47-56.

Wacner, W. H. Jr. & F. S. WAGNER. 1997. northwestern North American Moonwort metropo-
lis and its outliers. Amer. J. Bot. 84(6, Bae ppl.):168.

Noopen, L. D. & W. H. WaGner Jr. 1997 [1998]. Photosynthetic capacity and leaf reorientation in
two oa ferns, Polystichum acrostichoides and Dryopteris intermedia. Amer. Fern
J. 871

WAGNER, si H. Jr. “1998. Role of disturbed habitats in conservation of pteridophytes. Amer. J.
ie eg suppl.):99

. pwd fie hybrids for hardy garden cultivation. Hardy Fern Found. Quart.

0}80-36

F. S. Wacner. 1998. Moonwort madness: a reply. Fiddlehead Forum 25:27, 30, 31. Re-

pie in ge Fern Found. Quart. 9(2):24—26.

Zach, J. C . MANNING & W. H. WAGNER Jr. 1998. : new adder’s-tongue (Ophioglossum:
Ophislosacon for North omen Sida 18:307-31
Wacner, W. H. Jr. 1998. Review of: C Page. 1997. Sea of Britain and Ireland, ed. 2. Cam-

bridge Univ. in London and ae York. Syst. Bot. 23:249-250.

64 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

, F. S. Wacner, D. D. PALMER & R. W. Hospy. 1999. oo notes on the Pteridophytes
of poe Contr. Univ. Michigan Herb. 22:135-18
A. V. GILMAN. 2001. Dryopteris acti gin nov. (D. carthusiana x goldiana), a rare
pip Hil hybrid from Vermont. Amer. Fer 1:9-12.
WIDERLECHNER, M. P. & W. H. Wacner JR. 1998 stot Occurrence of European cenie, Rubus
caesius (Rosaceae), vein recemi in Iowa and Michigan. Michigan Bot. 37:107-11
Wacner, W. H. Jr. 2002. Leaf forms in southeastern Michigan Sassafras. seo =a 39:70-76.
& J. R. GRANT. sens ava alaskense, a new Moonwort hybrid from the interior of
Alaska. Amer. Fern J. 9 70.
ZikA, P. F., E. R. ALVERSON, ae os i: JR. & F. S. Wacner. in press. Botrychium hesperium
in the Wallowa Mountains of Oregon. Amer. Fern J.

American Fern Journal 92(2):65—79 (2002)

The Mating Systems of Some Epiphytic Polypodiaceae

WEN-LIANG CHIOU
Division of Forest Biology, Taiwan Forestry Research Institute, 53 Nan-Hai Rd.,
Taipei 100, Taiwan
Donatp R. Farrar’
Department of Botany, Iowa State University, Ames, IA 50011
Tom A. RANKER
University Museum and Department of Environmental, pecan & Organismic Biology,
University of Colorado, Boulder, CO 803

Asstract.—Genetic loads, estimated from sporophyte production by isolated gametophyte cul-

mating in C. angustifolium (in part), Microgramma heterophylla and Polypodium pellucidum.
Polyploidy characterizes the intragametophytic-selfing species, whereas the intergametophytic-
mating taxa are diploid. The duplicated loci of polyploid taxa may mitigate the expression of re-
cessive lethal alleles caused by intragametophytic selfing, whereas genetic load probably main-
tains the mating systems of the intergametophytically mating taxa. Enzyme electrophoretic
patterns of fixed heterozygosity support allopolyploid origins of C. phyllitidis and P. aureum
and confirm their intragametophytic mating systems. Antheridiogens, present in both groups,

may promote intergametophytic mating in diploids through promotion of the early development

ploids if these gametophytes delay or do not attain insensitivity to their own antheridiogen. In
the polyploids, antheridiogens may also alleviate low genetic variability through promotion of
occasional outcrossing. The perennial, clone-forming habit of epiphytic Polypodiaceae increases
the duration and the physical space occupied by derivatives of a single spore, thus expanding
the chance of interaction with a later migrant. Genetic load, duplicated genes, and antheridio-
gens, together with a perennial and clone-forming gametophyte growth habit, interact to produce
successful breeding strategies of these epiphytic species.

Three mating systems have been documented in ferns: intragametophytic
pies 2 cima dla selfing, and intergametophytic crossing (Klekowski,
1979). The more general term intergametophytic mating refers to either or
both of the Ge two systems when it is not possible to determine whether
cross-mated gametophytes are from the same (intergametophytic selfing) or
different (intergametophytic crossing) sporophytes.

Because gametophytes are potentially bisexual, it had been thought that
intragametophytic selfing was predominant in homosporous ferns (Klekowski,
1979), and high selfing rates have been reported in some diploid homo-
sporous ferns, especially those with subterranean gametophytes (e.g., Soltis
and Soltis, 1986a) and in some pioneering species (Crist and Farrar, 1983).
Intragametophytic selfing also seems to be the trend in species that are poly-

loid with respect to other species in a genus. Possibly the duplicated loci

’ Author for correspondence.

66 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

of polyploid species mitigate the problem of recessive deleterious allele
expression associated with selfing (Klekowski and Baker, 1966: Masuyama
and Watano, 1990). Intragametophytic selfing may also reflect a genetic bottle-
neck associated with the origin of polyploid species events. However, iso-
zyme evidence has shown that intergametophytic mating is the most
common breeding system in diploid homosporous ferns with above-ground
gametophytes (Haufler & Soltis, 1984; Soltis & Soltis, 1986b; 1990).

Reproductive behaviors are affected by several factors. Inbreeding depres-
sion resulting from recessive lethal genes (expressed as genetic loads in this
study) seems to be the primary factor preventing intragametophytic selfing in
many species (Haufler et al., 1990). Failure to attain bisexuality and asynchro-
nous maturation of antheridia and archegonia also promote outcrossing in po-
tentially bisexual gametophytes (Klekowski, 1968). Functional unisexuality is
augmented by antheridiogen systems that stimulate precocious antheridium
formation in some gametophytes in a population while allowing others to re-
main unisexually female (Naf, 1979). Antheridiogen can also induce previ-
ously buried spores to germinate to form gametophytes that reach the surface
and produce sperm capable of fertilizing above-ground plants (Voeller, 1971).
Variation in growth habit including indeterminate growth, branching, gemma
production, and production of perennial gametophyte clones can also influ-
ence reproductive behavior by increasing the probability of interaction be-
tween distant gametophytes and between plants established at different times
(Chiou & Farrar, 1997b; Chiou et al., 1998; Dassler & Farrar, 2001).

Comprehensive research on fern reproduction is limited, especially in
epiphytic ferns (Werth & Cousens, 1990) whose gametophytes exist in very
different environments from those of terrestrial species. The habitat of
gametophytes of epiphytic species is often within dense bryophyte mats
where interaction between gametophytes via antheridiogen or sperm transfer
may be significantly hindered compared to gametophytes of terrestrial ferns
on less complex surfaces (Dassler, 1995; Dassler & Farrar, 2001). On the other
hand, bryophyte mats may maintain free water, the medium for sperm, avail-
able for longer periods and thus promote fertilization over longer distances
than substrates lacking bryophytes.

Most species of Polypodiaceae are epiphytic. The presence of antheridio-
gen systems (Chiou & Farrar, 1997a) and clone-forming, perennial growth

MATERIALS AND METHODS
The materials used in this

study follow. G h
collected by Chiou and Farr y reenhouse materials and those

ar were used for genetic load studies. Materials

CHIOU ET AL: MATING SYSTEMS 67

for isozyme studies came from the four Florida localities indicated below by
two-letter abbreviations. Vouchers are held at Iowa State University.

Campyloneurum angustifolium (Swartz) Fée—ISU greenhouse (origin un-
known) (Farrar 02-1-8-1) and Marie Selby Botanical Garden, Sarasota,
Florida (origin unknown; Chiou 14337).

Campyloneurum phyllitidis (L.) C. Presl—ISU greenhouse (2 collections,
origin unknown) (Farrar 02-1-8-2), Castellow Hammock (CH), Fakahatchee
Strand State Preserve (FS), and Jonathan Dickson State Park (JD).

Microgramma heterophylla (L.) Wherry—ISU greenhouse (Florida) (Farrar
02-1-8-3).

Phlebodium aureum (L.) J. Smith—Fakahatchee Strand State Preserve (FS)
(Chiou 14342), Jonathan Dickinson State Park (JD) (Chiou 14300), Toso-
hatchee State Reserve (TS).

Phymatosorus scolopendria (Burm.) Pic.-Serm.—Hawaii (Farrar 92-8-20-1).
Polypodium pellucidum Kaulf.—Hawaii (Farrar 92-8-19-3).

Spores were first sown and gametophytes grown in multi-gametophyte cul-
tures on agar-solidified mineral media in plastic petri dishes (Chiou & Farrar,
1997a). Cultures were maintained under continuous, white fluorescent illu-
mination of 2000-3000 lux and at a temperature range of 20—24°C.

Genetic load (presence of sporophyte-lethal alleles) was estimated by com-
paring isolated-spore and isolated-gametophyte cultures with paired-spore
and paired-gametophyte cultures. These cultures were grown in compart-
mentalized plastic sheets (‘jelly-molds’’), each with 20 cells containing
about 6 ml of agar medium. In each sheet, a single spore was transferred
directly into each of 5 cells (isolated-spore cultures), and a one-month-old
gametophyte which was still asexual was transferred from multigametophyte
cultures into each of another 5 cells (isolated-gametophyte cultures). The
other 10 cells of each sheet received two spores (paired-spore cultures) and
two gametophytes (paired-gametophyte cultures). Five such culture sheets
were separated by clear, flat, plastic sheets and stacked in transparent plastic
boxes (vegetable crispers) for a total of 50 replicates of each culture type for
each species. Gametophytes grown from spores isolated before germination
served to test the possibility that gametophytes transferred from multi-
gametophyte cultures to isolate and pair cultures might have experienced
gametophyte interactions before transfer. The light intensity of these cultures
was maintained between 1500 lux (bottom layer) and 3500 lux (top layer).
Plants were watered with distilled water every two weeks after the gameto-
phytes were 4 months old.

Whether sporophytes were produced by syngamy was determined by
examination with a compound microscope. Genetic load was estimated
by counting the percentage of bisexual gametophytes failing to produce

68 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

sporophytes (Peck et al., 1990). The five layers of sheets in each box were
designated as blocks, and spore-isolated cultures vs. gametophyte-isolated
cultures were designated as split plots. For C. angustifolium, C. phyllitidis,
and P. aureum, since two sources of spores were used, a Latin Square was
designed. Five treatments were “A” and “‘B” (isolated plants of two different
sources), ““AA’’ and “BB” (paired plants from the same source), and “AB”
(paired plants from different sources). Isolated gametophytes were removed
when sporophytes were formed or 8 months after spore sowing and exam-
ined microscopically to determine their sexuality. Development of gameto-
phyte sexuality in multi-plant cultures was also monitored (Chiou & Farrar,
1997a).

Isozyme data were analyzed with BIOSYS-1 (Release 1.7, Swofford and
Selander, 1989).

RESULTS

aureum and Polypodium pellucidum, when bisexual plants became domi-
nant. A relatively small number of male gametophytes (<10%) were present

C. angustifolium, C. phyllitidis, and P. aureum (data not shown). Differences

CHIOU ET AL: MATING SYSTEMS 69

ABLE 1. Sex expression (%) of Polypodiaceae species in isolated spore and isolated gametophyte
cultures at 8 months after sowing spores.

Treatment Male Female’ Bisexual!

Campyloneurum angustifolium
re

19 0 81
Gametophyte 6 0 94
Campyloneurum phyllitidis
Spore 17 41(a) 42(b)
Gametophyte 8 11(b) 81(a)
Microgramma heterophylla
14 0 86
6 0 94
Phlebodium aureum
Spore 0 3 97
Gametophyte 0 3 97
Phymatosorus scolopendria
re 0 0 100
Gametophyte 0 0 100
Polypodium pellucidum
pore 7 85(a) 8(b)
Gametophyte a 35(b) 61(a)

’ Different letters in parentheses for each gender of each species indicate significant differences
in Duncan’s multiple test (95% confidence limits).

after sowing spores were present only in C. phyllitidis and in P. pellucidium
(Table 1). Whereas few or no female gametophytes remained at 8 months in
other species, in C. phyllitidis and in P. pellucidium female gametophytes
were still common, especially in cultures from isolated spores.

The sexual expression of gametophytes was not significantly different
(95% confidence limits in Duncan’s multiple test) among different layers
within cultures boxes except for C. angustifolium, where significantly more
male gametophytes developed in the lowest (and darkest) layer.

Genetic Loap.—At 8 months after sowing spores, sexual sporophyte produc-
tion was lower in the isolated cultures (of both isolated-spore and isolated-
gametophyte cultures) than in paired cultures of M. heterophylla and P.
pellucidum (Table 2). There was no significant difference between isolated
and paired cultures of P. aureum and P. scolopendria.

In C. angustifolium, gametophytes from source B failed to produce sporo-
phytes in both isolated-spore and isolated-gametophyte cultures at 8 months
after sowing spores, whereas 83% of isolated plants from source A produced
sporophytes. Thus the inferred genetic load of source B was extremely high,
whereas the genetic load of source A was quite low.

In C. phyllitidis, no difference was observed in sporophyte production be-
tween gametophytes from the two spore sources. There also was no signifi-
cant difference between sporophyte production by single gametophytes
isolated as young plants from multi-gametophyte cultures and any of the

70 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

TaBLE 2. The percentage of gametophytes producing sporophytes sexually and the estimated
genetic load expressed by bisexual gametophytes of 8 month old spore-isolated and gametophyte-
isolated cultures of Polypodiaceae species.

% Producing sporophytes Est. genetic load?
Source’ Spore® Gametophyte® Average” Spore® Gametophyte® Average®
Campyloneurum angustifolium
A 77(a) 90 83(ab) 16(b) 8(b) 12(b)
AA 92(a) 96(a) 94(a) — - ~
AB 73(ab) 86(a) 79(b) - - -
BB 44(c) 48(bc) 46(c) - - o
B O(d) O(d) 100(a) 100(a) 100(a)
Campyloneurum phyllitidis
A 9 40(a) 25(b) 74(a) 38(b) 56(a)
AA 48(a) 61(a) 55(a) 2 — -
AB 55(a) 64(a) 59(a) - ~ =
BB 49(a) 65(a) 57(a) - ~ re
) 61(a) 33(b) 83(a) 31(b) 57(a)
Phlebodium aureum
7(b) 82(ab) 79(b) 18(a) 13(a) 16(a)
AA 2(ab) 90(a 91(ab) - ~ -
AB 98(ab) 80(ab) 88(ab) - - -
BB 89(ab) 100(a) 4(a) - - ~
B 2(ab) 97(ab) 95(a) 7(a) 3(a) 5(a)
Microgramma heterophylla
A O(b) 2(b) 1(b) 100(a) 98(a) 99
AA 32(a) 24(a) 28(a) _ ~ —
Phymatosorus scolopendria
A 88(a) 90(a) 89(a) 12(a) 10(a) Ei
AA 93(a) 93(a) 93(a) - ~ -
Polypodium pellucidum
A O(c) 2(c) 1(b) 100(a) 95(a) 98
AA 63(a) 26(b) 43(a) - os =

‘ A and B indicate single gametophyte cultures; AA, AB and BB indicate paired cultures, AA
and BB indicate gametophyte pairs from the same sporophyte, AB indicates gametophyte pairs
from two different sporophytes
* Genetic load was estimated b
duce sporophytes.

* The same letter in parentheses for each item for eac
in Duncan’s multiple test (95% confidence limits).

y counting the percentage of bisexual gametophytes failing to pro-

h species indicates no significant difference

paired cultures. However, single plants isolated as spores produced signifi-
cantly fewer sporophytes than those isolated as young gametophytes or those
grown in pairs. Thus genetic load estimates from isolated-spore and isolated-
gametophyte cultures were significantly different (Table 2).

In P. pellucidum, 21% of the isolated plants produced sporophytes. How-
ever, only 1% of these were produced through syngamy, the others were
generated through apogamy. All of the apogamous sporophytes formed from

CHIOU ET AL: MATING SYSTEMS 71

Allele frequencies at polymorphic loci in three populations of re Soria
phyllitidis. Sample sizes in each population are 10. Populations were invariable at 27

Locus Allele JD! CH’ FS! Locus Allele JD’ CH’ Fs’
Lap-b Z 1.00 0.10 0.65 Pgm-3a ab 1.00 1.00 0.90
- 0.90 a5; Z - _ 0.10

Mdh-3a s - 0.10 - Pgm-3b 1 1.00 1.00 0.80
vA 1.00 0.90 1.00 Z - = 0.20

Mdh-3b ef - 0.10 - 6Pgd-2a 1 1.00 1.00 0.40
zZ 1.00 0.90 1.00 2 - - 0.60

' JD=Jonathan Dickson State Park, CH=Castellow Hammock, FS=Fakahatchee Strand State
Preserve.

IsozYMEs.—In C. phyllitidis, seventeen putative locus pairs were scored
across the eleven enzyme systems (Table 3). There was no variability within
or among the three populations for 12. These were fixed for the same allele
at four (Fbp-1, Idh, Mdh-1, and Tpi-1) and for fixed interlocus heterozygosity
at eight locus-pairs (Aco, Fbp-2, Hk, Mdh-2, 6Pgd-1, Pgi-2, Skdh, Tpi-2). Mdh-
3 was fixed for a single allele in the samples from JD and FS populations, Pgm-
2 was fixed in all populations, but for a different allele at 2b in JD, and Pgm-3
was fixed in the samples from JD and CH populations. 6Pgd-2 had fixed hetero-
zygosity in populations JD and CH. The genetic similarity among the three
populations was high (>0.934) for both Rogers’ (1972) and Nei’s (1978) genetic
coefficient (Table 4).

In population JD, only one genotype was found at each of the locus-pairs.
In population CH, Lap and Mdh-3 each had two genotypes, combining
to form three multilocus genotypes. In population FS, Lap and Pgm-3 each
contained three genotypes and the variable locus 6Pgd-2 contained two geno-
types, combining to form six multilocus genotypes (Table 5). In combination,
the three populations form 10 multilocus genotypes. Of 30 plants tested,
only one displayed recombinational heterozygosity at one locus (Lap 11/23).

In P. aureum, 22 putative loci were scored among the 13 enzyme systems.
There was no variability within or among the three populations for 21 locus
pairs. Eleven (Ald, Fbp-1, Fbp-2, Idh-1, Idh-3, Mdh-1, Mdh-3, 6Pgd-2, Pgi-1,
Skdh-1, and Skdh-2) were fixed at the same allele, and ten (Aco-1, Aco-2, Hk,
Lap, Mdh-2, 6Pgd-3, Pgi-2, Pgm-1, Pgm-2, and Tpi-2) had fixed heterozygos-

TABLE 4. Matrix of Roger’s genetic similarity (above diagonal) and Nei’s unbiased genetic identity
(below diagonal) among three populations of Campyloneurum phyllitidis in Florida.

Population’ JD CH FS
JD ae 0.938 0.934
CCH 0.946 ae 0.951
FS 0.956 0.981 a a

1jD= — Dickson State Park, CH=Castellow Hammock, FS= Fakahatchee Strand State
Preserv

v2 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

TasLe 5. Description of multilocus genotypes of two populations of Campyloneurum phyllitidis
in Florida.

Locus CH! FS!
Lap-a/b 11/22 11/33 11/33 11/22 41/22 11/33 11/33 11/23 11/22
Mdh-3a/b 29/99 11/11 29/22
Pgm-3a/b 47/11 11/22 ss | 22/22 as a 114/91
6Pgd-2a/b 41/22 22/22 11/22 11/22 22/22 22/22

Observed 10% 10% 80% 30% 10% 20% 10% 10% 20%
‘CH = Castellow Hammock, FS = Fakahatchee Strand State Preserve.

ity. Got displayed fixed heterozygosity for the same alleles in TS and JD pop-
ulations and in 90% of the samples in FS, but had fixed heterozygosity with
a different allele for Got-b in the other 10%. The genetic similarity between
population FS and the two identical populations TS and JD was high, 0.998
in Rogers’ (1972) genetic similarity and 1.0 in Nei’s (1978) genetic identity.
In populations TS and JD, there was only one genotype for all the locus-
pairs. In population FS, only two genotypes appeared, Got-a!'/b"™ (10%) and
Got-a''/b** (90%). No recombinational heterozygous individuals were de-
tected among the 30 plants tested.

DISCUSSION

In multi-gametophyte cultures of all species, male and bisexual plants ap-
peared only after a significant number of female plants had differentiated.
This is consistent with the presence of antheridiogen systems in these spe-
cies, as has been previously demonstrated (Chiou & Farrar, 1997a). Anther-
idia on bisexual gametophytes grown as isolated plants were probably
induced by their own gametophyte’s antheridiogen secretions. This could
occur either by absence of or delayed attainment of insensitivity of a gameto-
phyte to antheridiogen, or by generation of secondary lobes with reduced
physiological connection to the principal antheridiogen-producing apex. A

tion from antheridiogen. Plants that remained unisexual males at eight
months generally were relatively slow-growing and small.

In isolated-plant cultures, a significant difference in sexual expression
between gametophytes grown from isolated spores and those grown from
gametophytes isolated from multi-gametophyte cultures when they were one

pellucidum is that plants in multi-gametophyte cultures were subjected to
antheridiogen before individual gametophytes were transferred to the iso-

CHIOU ET AL: MATING SYSTEMS 73

lated-gametophyte cultures. It is also possible that transplanting actively
growing gametophytes might disrupt their normal developmental pattern.
Neither hypothesis explains the slow rate at which bisexuality is attained by
spore-isolated gametophytes of these species relative to the other species
tested.

Genetic load can be estimated by the ability of isolated bisexual plants to
produce sporophytes. Isolated gametophytes of Polypodium pellucidum,
Microgramma heterophylla, and the B source of Campyloneurum angustifo-
lium produced virtually no sporophytes through syngamy, whereas isolated
gametopytes of C. phyllitidis, Phlebodium aureum, Phymatosorus scolopen-
dira and the A source of C. angustifolium produced abundant sporophytes
through intragametophytic selfing.

Genetic load was not significantly different between isolated-spore and
isolated-gametophyte cultures, except in Campyloneurum phyllitidis where
the apparent genetic load of isolated-spore cultures was much greater. The
reason for this apparent elevation of genetic load in isolated-spore cultures
in C. phyllitidis is not clear, but from the delayed production of bisexual
plants evident in isolated spore cultures (Table 1), it is possible that anther-
idia in some bisexual plants may still have been functionally immature even
at eight months, falsely indicating a high genetic load. No similar delay in
production of antheridia by this species was evident in multispore cultures
(Chiou & Farrar, 1997a) or in isolated-gametophyte cultures. However, pair-
ing of gametophytes from different sporophytes in C. phyllitidis also failed
to fully relieve the sporophyte suppression observed in isolated gametophyte
cultures. This suggests that some genetic load in this species is perhaps
being expressed in gamete development (Klekowski, 1971), or that there was
very little genetic difference between the two sporophytes, as might well be
the case in a population reproducing primarily by intragametophytic selfing.
In fact, most sporophytes of this species were homozygous at all loci tested.

Paired-spore and paired-gametophyte cultures allowed intergametophytic
mating to relieve inbreeding depression that might prevent intragameto-
phytic selfing in either gametophyte. Thus any increase in sporophyte pro-
duction in paired cultures relative to isolated cultures of the same species is
assumed to have resulted from intergametophytic mating.

Two spore sources were used in studies of Campyloneurum angustifolium,
C. phyllitidis, and Phlebodium aureum. Genetic load estimates obtained from
the two sources were not significantly different for C. phyllitidis and P. aur-
eum. Estimated genetic load did differ significantly between the two sources
of C. angustifolium in which it was low for source A, but very high for
source B. This suggests that the B sporophyte of C. angustifolium was from a
highly outcrossing population, whereas the A sporophyte was derived from
a population with a higher level of inbreeding. Both diploid and tetraploid
chromosome counts have been reported for C. angustifolium (n=37, Evans,
1963, from Peru; n=74, Evans, 1963, from Costa Rica; Sorsa, 1966, from
Costa Rica; Knobloch, 1967, from Jamaica). Thus, judging from a correlation

74 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

143) ¥ 43. *

TABLE 6. Description of mult genoty} percentage in a combination of three Florida
populations of Campyloneurum phyllitidis.

Lap-a/b Mdh-3a/b Pgm-2a/b Pgm-3a/b 6Pgd-2a/b Percentage
11/22 22/22 11/22 11/11 1/22 33.33
11/33 22/22 11/11 TN/41 11/22 26.67
11/33 14/11 14/41 14/11 11/22 3.33
11/22 22/22 71/41 11/11 41/22 3.33
41/33 22/22 11/14 11/11 11/22 6.67
11/22 APs Lies a jg 11/22 10.00
11/23 22122. 11/11 a ala 22/22 3.33
11/33 22122 11/11 22/22 Ti/22 3.33
11/22 22/22 11/11 14/22 22/22 3.33
11/22 22/22 11/11 ag 22/22 6.67

of polyploidy with low genetic load, it is possible that the A sporophyte and
B sporophyte of C. angustifolium were tetraploid and diploid respectively.

Both diploid and tetraploid forms have also been reported in Phlebodium
aureum (n= 37, n=74, Evans, 1963), and Phymatosorus scolopendria (2n=72,
Léve et al., 1977; n=72, Tsai and Shieh, 1983). In Campyloneurum phyllitidis,
only tetraploids have been found (n=74, 2n=148, Evans 1963, Nauman
1993), whereas in Polypodium pellucidum, only diploid numbers have been
reported (Manton, 1951). Thus genetic load estimates indicate that the sample
sporophyte spore sources of C. phyllitidis, P. aureum and P. scolopendria
were probably tetraploids, whereas those of Microgramma heterophylla, and
P. pellucidum were diploid. In fact, isozyme evidence also indicates that
C. phyllitidis and P. aureum are polyploids.

Strong evidence for describing the mating system of diploid species in the
wild can be obtained from analysis of isozyme electrophoretic patterns. Elec-
trophoretic patterns for the species tested in this study, Campyloneurum
phyllitidis and Phlebodium aureum, revealed a high level of fixed hetero-
zygosity for both, indicating that these samples from Florida are polyploid,
probably allopolyploid. Because of this, accurate counts of heterozygous in-
dividuals and estimates of outcrossing from isozyme evidence are not possi-
ble, although we can state that at least one putatively outcrossed individual
(Lap-a'*/b**) was among the 10 samples of C. phyllitidis in the FS popula-
tion. No evidence of outcrossing was present in the Florida populations of
P. aureum but the extremely low level of genetic variability among sampled
plants of Pz aureum (29 of 30 plants were genetically identical) would
preclude isozymic detection of most recombinational heterozygotes if they
were formed. However, the considerable variability among sporophytes of
C. phyllitidis (10 multilocus genotypes among 30 plants tested) allows ample
opportunity for detection of unbalanced heterozygotes and three-allele hetero-
zygotes (at Lap) that would be produced if outcrossing was frequent (Table 6).

Assuming allopolyploidy, intragametophytic selfing, and no mutations, the
maximum number of genotypes per locus-pair provides an estimate of the

CHIOU ET AL: MATING SYSTEMS 75

minimum number of hybridization events involved in producing an allote-
traploid species (Ranker et al., 1994). Thus in Campyloneurum phyllitidis,
the single genotype in the population JD suggests that only one hybridization
event occurred in the ancestry of this population in Jonathan Dickson State
Park. Two genotypes at each of Lap and Mdh-3 of population CH indicate
that that population originated from at least two hybridization events. Three
genotypes at Pgm-3a/b implies that at least three hybridization events pro-
duced population FS. However, one of these genotypes (11/22) could have
been generated (11/11 x 22/22) by the low level of outcrossing demonstrated
by presence of the Lap 11/23 genotype. Because there are not great distances
separating these populations in Florida and because of the high similarity of
genotypes among the three populations, we can also consider them as a sin-
gle population. In that case the number of hybridization events responsible
for the Florida plants is at least three and may be as high as ten.

In Phlebodium aureum, the genetic similarity among sampled populations
is very high. Isozyme evidence showed no genetic variation in the samples
of TS and JD populations and only a single plant representing a distinct
genotype in the FS population. Because Campyloneurum phyllitidis and
P. aureum are widespread in tropical America and because likely parental
diploids are not present in Florida, these Florida genotypes probably repre-
sent three to ten separate spore introductions for C. phyllitidis and possibly
only one or two for P. aureum.

In an electrophoretic study of Polypodium pellucidum, Li and Haufler
(1999) found a small but significant (mean fixation index of 0.169 across
populations of epiphytic P. pellucidum var. pellucidum) excess of homozy-
gous individuals in most (but not all) populations sampled in the Hawaiian
archipelago. Since intragametophytic selfing as a dominant breeding system
would lead to much higher fixation values in relatively few generations
(Hartl & Clark, 1997), the observed level of fixation likely results from a
mixed mating system where intergametophytic selfing among gametophytes
in populations derived from the same sporophyte is occurring. Thus electro-
phoretic analysis of population structure in P. pellucidum is not inconsistent
with the implications of our results that intragametophytic selfing in this
species is rare, probably curtailed by genetic load. This constraint on repro-
duction via single isolated spores likely contributes to the genetic differen-
tiation and low values of gene flow between populations estimated by Li
and Haufler (1999).

The fact that diploidy favors intergametophytic mating whereas tetraploidy
favors intragametophytic selfing has been demonstrated (e.g., Masuyama and
Watano, 1990). The fixation of different alleles from the different parent spe-
cies of allopolyploids may mitigate the expression of recessive lethal genes
caused by intragametophytic selfing. In our study, low levels of genetic load,
as evidenced by production of sporophytes through intragametophytic self-
ing, was well correlated with polyploidy (Table 7). Isolated gametophytes of
the diploid species (Polypodium pellucidum, Microgramma heterophylla,
and the B source of Campyloneurum angustifolium) produced virtually no

76 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

TaBLE 7. Genetic load, ploidy level and probable mating systems in Polypodiaceae species.

Species Genetic load’ Ploidy” Mating system?*

C. angustifolium A* 14 P (Evans 1963) Intragam. selfing
C. angustifolium B* 100 D (Evans 1963) Intergam. mating
C. phyllitidis 35 P (Isozyme; Evans 1963) Intergam. selfing
M. heterophylla 99 D Intergam. mating
hl. aureum 10 P (Isozyme; Evans 1963) Intergam. selfing
Phy. scolopendria 11 P (Tsai & Shieh 1983) Intergam. selfing
Pol. pellucidum 98 D (Manton 1951) Intergam. mating

‘Estimated from average sporophyte production by single gametophytes isolated as spores and
as young gametophytes, except for C. phyllitidis in which the estimate is from plants isolated as
young gametophytes.

* Determined from isozyme patterns (isozyme) and/or chromosome counts (reference). D=dip-
loid, P= polyploid.

* Determined from isozymes and/or genetic load.

* C. angustifolium A and B plants are from two different sources.

sporophytes through Ssyngamy, whereas isolated gametopytes of the tetra-
ploid species (C. phyllitidis, Phlebodium aureum, Phymatosorus scolopen-
dria and the A source of C. angustifolium) produced abundant sporophytes
through intragametophytic selfing.

Gametophytes derived from isolated spores resulting from long-distance

spores can produce sporophytes through either intragametophytic selfing or
intergametophytic mating, the latter being augmented by antheridiogen-
stimulated production of male gametophytes. Species predominantly repro-
ducing by intergametophytic mating can maintain high levels of genetic
variability, including a high genetic load, but may be very limited in their
ability to migrate by long-distance spore dispersal (Peck et al., 1990). Thus a
trade-off exists between maintenance of genetic variability on one hand and
ease of migration on the other.

CHIOU ET AL: MATING SYSTEMS 77

Schneller & Hess, 1995). But, the existence of antheridiogens here could also
function to promote bisexuality in isolated plants if individual thalli have
reduced ability to attain bisexuality. Gametophytes of these species propa-
gate vegetatively by branching (Chiou & Farrar, 1997b). Maintaining an an-
theridiogen system may be adaptive in promoting antheridium formation on
new vegetatively produced thalli regardless of whether the species is in-
breeding or outbreeding.

Whether species are outbreeders or inbreeders, breeding behavior must be
adaptive to establishment and survival of individuals of those species. In
general, inbreeding may be advantageous for initiating new populations fol-
lowing long-range spore dispersal where gametophytes are likely to be de-
rived from single isolated spores. The opposite strategy, outbreeding, has the
advantage of generating and retaining genetic diversity, and high genetic
loads carried by outbreeders tend to maintain that mating system. Previous
studies have demonstrated that gametophytes of the epiphytic species
studied here grow perennially through branching and vegetative prolifera-
tions which increase their life span and effective gametophyte size (Chiou &
Farrar, 1997b) and produce antheridiogen that facilitates the production
of male gametophytes (Chiou & Farrar, 1997a). Both of these characteristics
have been proposed to increase the probability of intergametophytic mating
(Chiou & Farrar, 1997a; b; Chiou et al., 1998). Here we demonstrate that
sporophytes of these species may be produced through either outcrossing or
inbreeding, as evidenced by high or low level of genetic load, respectively.
Their mating systems may be controlled principally by genetic load and
generally correlated with ploidy level. Interwoven with these factors are
gametophyte morphology, growth habit, antheridiogen production, and envi-
ronmental parameters which together maintain successful reproduction of
these species. For outbreeding diploid species, genetic load is possibly the
driving force leading to morphological and physiological adaptations promo-
ting outcrossing.

ACKNOWLEDGMENTS

The authors thank Drs. D. B. Lellinger, C. H. Haufler, and L. Hickok for their useful comments,
Dr. Paul N. Hinz for help with statistical analysis of gametophyte culture data, the Marie Selby
Botanical Garden for allowing collection of spores of Campyloneurum angustifolium, Dr. H. Lu-
Rivero for assisting collection of Campyloneurum angustifolium, the Florida

ia?)
|
5
jor
ns)

tellow Hammock, Dr. J. B. Miller and Mr. R. E. Roberts for help with collecting Campyloneurum
phyllitidis and Phelebodium aureum in Jonathan Dickinson State Park, Mr. M. Owen for assis-
tance in collecting Campyloneurum phyllitidis and Phelebodium aureum in Fakahatchee Strand
State Preserve, Mr. Keith Fisher for help with locating Phelebodium aureum in Tosohatchee State
Reserve, and Mrs. Ming-Ren Huang for help with plant cultures and data recording. This research
was supported by the Taiwan Forestry Research Institute (Contribution No. 196).

78 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

REFERENCES

Cuiou, W.-L. and D. R. Farrar. 1997a. Antheridiogen production and response in Polypodiaceae
species. Amer. J. Bot. 84:633-640.
and D. R. Farrar. 1997b. — a morphology of selected species of the
wong? Polypodiaceae. Amer. Fern J. 87:
, D. R. Farrar, and T. A. cai a eae eee and reproductive biol-
ogy in Elaphoglossum Schott. Canad. J. Bot. 76:1967-19
CRIsT, ish Gs dees _R. Farrar. 1983. Genetic load and ue dispersal in Asplenium platy-
Canad. J. Bot. 61: ab ip
Dass, ee L 1995. Significance of gamatophyte form in tropical, epiphytic ferns. PhD diss. lowa
ate University, Ames.
—— pe D. R. Farrar. 2001. Significance . eae form in long-distance colonization
by mele bigaie ferns. Brittonia 53:352-369.
M.

Evans, A. 3. New peace a. in the Polypodiaceae and Grammitidaceae.
Carolan i. 671-6
Farrar, D. R. 1990. heat and evolution in asexually reproducing independent fern gameto-

phytes. ak Bot. 15:98-111.

Hartt, D. L. and A. G. CLark. 1997. Principles of Population Genetics. Sinauer Associates, Sunder-

and, Massachusetts.

Haurier, C. H. and G. J. Gasrony. 1978. Antheridiogen and breeding system in the fern genus
Bommeria. Canad. J. Bot. 56:1594—1601.

and D. E. Souris. 1984. Obligate mies von in a homosporous fern: field confirmation of

a laboratory prediction. Amer. J. Bot. 71:878—-881.

, M. D, WinpHaM, and T. A. Ranker. 1990. oo analysis of the Cystopteris tennes-

seensis complex. Ann. Missouri Bot. Gard. 77:314—-329.

Hooper, E. A. and C. H. HAuFer. 1997. Genetic Hii and ues system in a group of neo-
tropical Kae ferns (Pleopeltis; Polypodiaceae). Amer. J. Bot. 84:1664—-1674

K.exowsKkI, E. J. JR. 1968. Reproductive biology and evolution in the pteridophyta. PhD diss. Uni-
versity of Sateen Berkeley.

———. 1971. Ferns and genetics. BioScience 21:317-322.

. 1979. The genetics and reproductive biology of ferns. Pp 133-170 in A. F. Dyer, ed. The

Experimental ain of Ferns. Academic Press, Lon

966. Evolutionary significance of polyploidy i in the pteridophyta. Sci-

ence 153: 305-307.
Knosiocu, I. W. 1967. Chromosome numbers in Cheilanthes and Polypodium. Amer. J. Bot.
54:461 ae

Li, J., and C . 1999. Genetic variation, breeding apres and patterns of diversifica-
tion in poantonanp: Polypodinan (Polypodiaceae). Syst. Bot. 2

Love, A., D, Love, and R. E. G. oo SERMOLLI. 1977. eset renait Atlas of Pteridophyta.
mer, Vaduz, Liechtenstei

Cra
Manton, I. 1951. ee of Pabeodieae’ in America. Nature 167:3
Masuyama, S Y. Watano. 1990. Trends for inbreeding in cas pteridophytes. Pl. Sp.

Antheridiogens and antheridia development. Pp 435-470, in A. F. Dyer, ed. The
enc ldesit regen of ferns. Academic Press, London.
, C. E. 1993. Campyloneurum. Pp 328 in Flora of North America Editorial Committee,

add. Flora of eet America. Oxford Univ. Press, New York, NY.

Nei, M. 1978. Estimation of seu anes and genetic distance from a small number of
individuals. Genetics 89:583—

Peck, J. H., C. J. Peck, and D. R. viciieae 90. Influences bs life history attributes on formation of
local and distant fern populations. Amer Fern. J. 8

Ranker, T. A., LER, P. S. Soitis, and D, E aay he 1989. Genetic evidence for allopoly-
ploidy in the Neotropical fern esis eee (Adiantaceae) and the reconstruction
of an ancestral genome. Syst. Bot. 14:439-447

CHIOU ET AL: MATING SYSTEMS 79

a, : K. FLoyp, and P. G. Trapp. 1994. Multiple colonization of Asplenium adiantum-nigrum
o the ea eget Archipelago. Evolution 48:1364—1370.
Roc, i S. 1972. Measures of genetic similarity and genetic distance. Studies in Genetics,
+ Toss ate 7213:145-153
Petia J. J., C. H. Haurter, and T. A. Ranker. 1990. Antheridiogen and natural gametophyte
populations. Amer. Fern J. 80:143-15
and A. Hess, 1995. Antheridiogen syatem in the fern Asplenium ruta-muraria (Asplenia-

Soitis, D. E. and P. S. Soitis. 1986a. Electrophoretic evidence for inbreeding in the fern
Botrychium asl (Ophioglossaceae). Amer. J. Bot. 74:504—509.
and P. S. TIs. 1986b. Polyploidy and breeding systems in homosporous pteridophyta:
a eeecnhs preci Naturalist 130:219-
d, P.:S, oo 1990. Genetic i pa and among populations of ferns. Amer.

Fern J. 84:161—
Sorsa, V. 1966. ae studies in the rcs come Amer. Fern J. 56:113-119.
Sworrorp, D. L. and R. B. SELANDER. 1989 -1. A computer program for the analysis of

allelic variation in acenuanl cites or biochemical systematics, release 1.7. Illinois
Natural History Survey. Urban

Tsal, J.-L. and W.-C. SHIEH. 1983. A ptenicoienhe survey of the pie e se in Taiwan. (1)
Material collection and chromosome observation. J. Sci. Engin. 20:1

VoELLER, B. 1971. Developmental adehes of fern gametophytes: enitian hr biology. Bio-
Science 5 267-270.

Wertu, C. R. 1989. The use of — data for inferring ancestry of polyploid pteridophytes.
2 lid eo Ecol. 17:117—

——., and M. I. Cousens. 1990. yen the contribution of population studies on ferns.
Amer. Fern J. 80:183-190.

American Fern Journal 92(2):80—92 (2002)

Belowground Distribution and Abundance of
Botrychium Gametophytes and Juvenile Sporophytes

CINDY JOHNSON-GROH'’, CHANDRA RIEDEL?,
LAURA SCHOESSLER®, and KrissA SKOGEN*
Biology Department, Gustavus Adolphus College, 800 W. College Ave., St. Peter, MN 56082

phytes m_ * respectively. Botrychium hesperium also has a relatively high density of 478 gameto-
[ ith ar

The importance of propagule banks (also referred to as seed, spore, or dia-
spore banks) in community dynamics has long been recognized (Leck et al.,
1989) for flowering plants. Propagules may persist belowground for many
years, creating a secure reservoir from which aboveground plants can be

density of fern gametophytes resulting from cultivation of spore banks is
high, ranging from 57,000 spores m ? (Milberg, 1991) to 5,000,000 m~2

y Hall, Iowa State University, Ames, IA 50011.
Evolution and Behavior, University of Minnesota, 100
Ecology Bldg., 1987 Upper Buford Circle, St. Paul, MN 55108

Current address: Argonne Na | ry, ’ . ° ’
a .

JOHNSON-GROH ET AL.: BELOWGROUND DISTRIBUTION 81

of propagules (represented by germinated gametophytes) to the aboveground
density of sporophytes, all previous studies have reported the remarkable
lack of ferns present in the aboveground vegetation despite a belowground
reservoir of spores. During and ter Horst (1983), Leck and Simpson (1987),
Milberg (1991), Raffaele (1996), Rydgren and Hestmark (1997) and Strickler
and Edgerton (1976) all found evidence of high densities of fern propagules
where ferns were entirely absent from the aboveground vegetation.

Botrychium presents two problems in applying conventional propagule
bank techniques. First is the difficulty and length of time required to culture
Botrychium beyond germination (Whittier and Thomas, 1993). Botrychium
spores require darkness for germination, and in nature Botrychium gameto-
phytes require mycorrhizae for growth beyond the two- or three-celled stage.
Therefore, it is not possible to culture and quantify the propagules using
standard seed bank techniques.

The second and more significant problem relates to the different life cycle
of Botrychium, relatively little of which is visible aboveground (Fig. 1).
Plants produce spores that filter into the soil and germinate in darkness. Fol-
lowing germination, a belowground achlorophyllous, fleshy gametophyte is
produced. The gametophyte produces gametangia and sexual reproduction
occurs, resulting in a belowground juvenile sporophyte. The belowground
rhizome is upright and short with mycorrhizal roots (no root hairs) and a
single leaf-producing bud at the apex. It takes several years for this juvenile
sporophyte to produce a leaf-bearing apex and emerge aboveground (Johnson-
Groh et al., 1998). The plants generally produce one leaf annually, but it
is common for Botrychium plants to remain dormant belowground in a given
year and produce no aboveground leaf (Johnson-Groh, 1997).

In addition to these belowground stages, some species reproduce asexually
via belowground gemmae, small (0.5—-1 mm) propagules that can indepen-
dently give rise to a new plant once detached from the parent plant (Farrar
and Johnson-Groh, 1990). Gemmae form on the rhizome and abscise at ma-
turity. Upon germination, gemmae develop 4 or 5 short roots prior to the dif-
ferentiation of a shoot apex and production of leaves (Farrar and Johnson-
Groh, 1990). The first leaves formed are short and slender and do not reach
the soil surface. The presence of vegetative reproduction greatly influences
the population dynamics of these gemmiferous species. It is common in the
field to see two or more leaves of gemmiferous Botrychium emerging in close
proximity. Excavation of these clusters usually reveals a large number of
belowground sporophytes in various stages of development.

The terms propagule bank and diaspore bank as used in other studies
imply banks of structures that have been disseminated (e.g. seeds, spores).
Except for gemmae, Botrychium gametophytes and sporophytes are not struc-
tures designed specifically to propagate or disseminate. Consequently the
use of existing terms (propagule, diaspore, spore or seed bank) does not
accurately describe the belowground structures of Botrychium. The term
belowground structure bank, as used in this paper, includes all belowground
structures: gemmae, gametophytes, juvenile plants and spores.

82 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

sporophyte (2n) } spore

ground
juvenile belowground sporophyte (2n) belowground gametophyte
sporophyte (2n) with shoot and root (with gametangia)
gemmae development

gametophyte fertilization

Fic. 1. Botrychium life cycle illustrating the belowground developmental stages.

Long-term demographic studies (15 years) of Botrychium reveal that popu-
lation numbers are quite variable (Johnson-Groh, 1997). Aboveground Botry-
chium population numbers fluctuate independently within and between
populations, as well as between years and between different sites. Fire, herbi-
vory, herbicides, and timber harvests may have an immediate impact on the
aboveground sporophytes (Johnson-Groh and Farrar, 1996). However, the above-
ground populations are fairly resilient and rebound following perturbations,
although recovery may take several years.

Many species of Botrychium are considered rare. Several are listed as criti-
cal, threatened, or endangered (Minnesota Department of Natural Resources,
2002). Understanding Botrychium population dynamics, including their be-
lowground biology, is necessary to effectively manage these species. A more
complete understanding of the belowground structure bank will allow pre-
diction of the impact of various management regimes, such as fire or grazing,
on these rare species.

The goal of this research was to investigate the belowground structure
bank of several species of Botrychium. It seems likely that the belowground
structures (gametophytes, juvenile sporophytes, gemmae and spores) play a

JOHNSON-GROH ET AL.: BELOWGROUND DISTRIBUTION 83

key role in the population dynamics of Botrychium, and it is anticipated that
the number of belowground structures should be at least as many as the
number of aboveground plants.

MATERIALS AND METHODS

Sites for sampling the belowground populations of Botrychium were
selected in areas that contained a typical density of aboveground plants
(~50—400 plants in an area 200 m*). A sampling grid resembling spokes of a
wheel was established within each population, and samples were collected
with a bulb planter at one-meter intervals along the six 8-meter spokes
(Fig. 2). In addition to the 48 spoke samples, a sample was collected from the
middle. A bulb planter (5-cm diameter) assured a uniform sample volume of
approximately 200 cm® for each sample. Distance from each sample to the
nearest aboveground plant, aboveground population density, and notes on
the vegetation and general ecology were recorded.

Soil samples were collected during the summers of 1988 to 1999 from a
variety of sites; these were processed the following academic years (Table 1).
Seven species of Botrychium subg. Botrychium were investigated: Botry-
chium campestre W. H. Wagner & Farrar, B. hesperium (Maxon & R. T. Clausen)
W. H. Wagner & Lellinger, B. gallicomontanum Farrar & Johnson-Groh, B.
lanceolatum (S. G. Gmel.) Angstr., B. montanum W. H. Wagner, B. mormo

. H. Wagner, B. yaaxudakeit Stensvold & Farrar, B. virginianum (L.) Sw.
An eighth species, Botrychium virginianum (L.) Sw. from subg. Osmundopteris,
was surveyed from two different sites. Because Botrychium typically grows in
mixed species assemblages, it is difficult to locate populations of only one
species. The sites sampled for B. campestre, B. hesperium, and B. virginianum
consisted only of those species, respectively. All other sites contained more
than one species. In all cases, the named species was the dominant species
(>75% of aboveground Botrychium).

Soil samples were processed using centrifugation, which allows the lighter
plant material to be extracted for examination under the microscope (Mason
and Farrar, 1989). The soil sample was broken up and washed through a series
of soil sieves where larger roots and debris were removed. The sediment
was collected and root segments dyed in neutral red and cut to the approxi-
mate size of Botrychium gemmae (0.5 mm) were added to the sediment to
gauge the success of the procedure. Samples were sieved in successively
finer sieves and then centrifuged first in water and then in sucrose solutions.
Centrifugation separated the belowground structures from the denser soil
particles. The first centrifugation caused dead organic matter to float to the
top of the tube, leaving living organic matter and stained roots in the pellet.
A second centrifugation in sucrose caused the living material to float.

The decanted liquid containing Botrychium plants was examined under
the microscope for gemmae, gametophytes, sporophytes, and stained root seg-
ments. Gametophytes were usually small (<1 mm) and irregularly shaped, and

84 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

N
NW NE
section
SW SE
$s

Fic. 2. Sampling design for collecting 49 soil samples for analysis. Circles represent soil sam-
ples collected on one of 6 radii.

could be identified by the presence of rhizoids, antheridia, and archegonia.
Juvenile sporophytes represented a continuum of development following
the formation of the embryo (still attached to the gametophyte) through the
development of a mature plant. Except for very young sporophytes, all juve-
niles had roots that were succulent and lacked root hairs. Very young sporo-
phytes, here referred to as embryos, were small (<1 mm), spherical to
irregular in shape, lacked rhizoids, and, when detached from the gameto-
phyte, had a relatively large scar where they had been attached. Gemmae
likewise represented a continuum of development following their formation.
At abscission, gemmae were spherical and small (0.5-1 mm). As gemmae
began to elongate and form roots, it was impossible to distinguish them from
juvenile sporophytes originating from a gametophyte. Examination of rhi-
zomes of mature leaf-bearing plants determined which species regularly pro-
duce gemmae.

RESULTS AND DISCUSSION

Results of the belowground analysis are shown in Table 2. The density
of aboveground sporophytes ranged from 0.4 m ~~ for B. hesperium and B.
virginianum to 16.1 m * for B. gallicomontanum. The proportion of samples
that contained belowground structures ranged from 4% for B. campestre to
65% for B. hesperium, with an overall average of 30%. Botrychium campestre
and 8B. virginianum, had notably low frequencies. The density of below-
ground gametophytes ranged from 10 m * for B. gallicomontanum and one

JOHNSON-GROH ET AL.: BELOWGROUND DISTRIBUTION 85

TaBLE 1. Botrychium species surveyed for belowground analysis, their sampling location, and
year of collection.

Species Location Year
B. campestre IA, Plymouth County, 5-Ridge Prairie 1988
B. hesperium WA, Colville Nat. Forest, Big Muddy 1998
B. gallicomontanum MN, Norman County, Frenchman’s Bluff 1995
B. lanceolatum OR, Wallowa-Whitman Nat. Forest, Lostine Creek 1999
B. montanum WA, Colville Nat. Forest, Kerry Creek Seep 1998
B. mormo MN, Chippewa Nat. Forest, Ottertail Peninsula 1995
B. yaaxudakeit AK, Tongass Nat. Forest, Yakutat 1999
B. virginianum (So. MN) MN, Nicolett County, St. Peter 1997
B. virginianum (No. MN) MN, Superior Nat. Forest, Pike Mountain 1999

population of B. virginianum to 738 m-° for B. montanum. The density of
belowground juvenile sporophytes ranged from 0 m ° (B. gallicomontanum,
B. montanum, and B. virginianum) to 281 m ” (B. hesperium). The average
stained root recovery was 84%.

Three of the surveyed species are known to produce gemmae (B. cam-
pestre, B. hesperium, and B. gallicomontanum). For those jag that pro-
duce gemmae, the density ranged from 5,907 gemmae m * for B. campestre
to 21 m ° for B. hesperium (Table 3). Gemmae, or gemma-like structures,
were also found in samples in which the dominant species does not produce
gemmae. For example, B. lanceolatum does not produce gemmae, but gem-
mae were found in the samples. Two associated species, B. minganense and
B. pedunculosum, were also found at the B. lanceolatum site and are known
to produce sparse gemmae, so it is possible that the gemmae extracted be-
long to these associated species. (It is impossible to distinguish belowground
structures of species except sometimes through absence or presence of
gemmae.) Botrychium yaaxudakeit also does not produce gemmae, and it is
likely that the gemmae found in this survey were from B. ascendens, which
produces numerous gemmae, or even B. minganense, which produces few

ae,

The high number of gemmae in B. /anceolatum may also be due to a pecu-
liar behavior observed in this species of expelling the embryo from the ga-
metophyte. Embryos appear to be “ejected” through the degeneration of
surrounding gametophyte tissue at a stage very similar in size and morphol-
ogy to gemmae produced in other species. This behavior may allow produc-
tion of multiple embryos per gametophyte, increasing the total production of
sporophytes. Further investigations of this behavior are underway.

There is large variation (4-65%) among species with regard to their fre-
quency in the soil samples. Whether this is due to natural variation among
species in response to environmental differences (e.g., Botrychium campestre
may naturally have a lower frequency than B. hesperium) or to chance is not
easy to resolve. Two technical problems prohibit easy resolution of this.
First, centrifugation and microscopic techniques used for this investigation

TasLe 2. Result of Botrychium belowground analysis, with the maximum and minimum values for each column in boldface.

sity of
ensity of belowground Ratio of below-
aboveground Frequency of Density of juvenile Total density of ground to
sporophytes belowground gametaphytes sporophytes mere ty o belo ih cir aboveground
Species (m*) structures (%) (m~*) (m*) mae (m~*) structures (m™*) plants
B. campestre (sig 4 21 198 5907 6126 914
B. hesperium 0.4 65 478 281 21 780 1950
B. gallicomontanum 16,1 27 10 0 4170 4180 260
B. lanceolatum 3.1 61 135 10 520' 665 215
B. mo 12 41 738 0 0 738 615
B. 12.8 46 728 104 0 832 65
B. akeit 1.4 29 281 42 312" 635 454
B. virginianum (So. MN) 0.6 8 73 0 0) 73 122
B. virginianum (No. MN) 0.4 6 10 0 0 10 25
Average (all) 4.7 30 261 fa 1228! 1560 332
Average (moonworts) 6.0 40 324 91 1579 1994 332

' See discussion for explanation of “gemmae” in soil samples for non-gemmiferous species.

(2002) 2 MAAINNN 26 ANN'TIOA *TVNUNOl NYA NVOMANV

JOHNSON-GROH ET AL.: BELOWGROUND DISTRIBUTION 87

TABLE 3. Soil surveys that contained gemmae produced either by the dominant primary species or
by associated species.

Associated species

Dominant species found at the sampling
Density of known to produce site and known to
Dominant species gemmae (m*) gemmae produce gemmae
B. campestre 5907 Yes None
B. hesperium 21 Yes None
B. gallicomontanum 4170 Yes None
B. lanceolatum 520 No B. minganense,
B. pedunculosum
B. yaaxudakeit 312 No B. ascendens

are extremely labor-intensive, thereby precluding multiple repetitions of the
same species. Second, although the sampling technique used for this study
is nominally disruptive, it is prudent to minimize excessive disturbance of
rare Botrychium species. For the latter reason, B. virginianum (which is not
rare) was included in this survey (despite its position in a different sub-
genus) and was sampled twice. These samples reveal a high degree of con-
sistency, with nearly identical belowground structure banks, even though
the sites surveyed were more than 300 miles apart.

It is also likely that the frequency differences among species are increased
by the patchy distribution of Botrychium and the probability of sampling
those patches. Distributions of plants result from factors of dispersal, survi-
vorship, and habitat heterogeneity. Because the distribution of aboveground
plants is patchy the same might be expected belowground. The random
chance of sampling a dense patch could account for the variation between
species.

Spores.—Although it is difficult to determine the presence of Botrychium
spores in the soil, other spore bank studies have shown high diversity and
abundance of fern spores that persist in the soil for many years (Dyer, 1994;
Milberg, 1991). It is reasonable to assume that there is a long-lived bank of
Botrychium spores. This bank is in constant turnover, receiving a variable
annual input of spores and losing spores to predation, loss of viability, and
other environmental perturbations.

Annual spore production is the primary means of restocking the spore
bank. Spore production, however, can be expected to vary from year to year,
depending on the number of aboveground plants and their development in
any given year. Johnson-Groh and Lee (in press) found that only 55% of B.
gallicomontanum and 39% of B. mormo produced spores in 1996. The re-
mainder of the plants senesced prior to spore production. Similar results
have been found for other species (Cabin and Marshall, 2000; Houle, 1998;
Kalisz, 1991: Raffaele, 1996). Houle (1998) found temporal differences in the
seed rain for Betula alleghaniensis and seedling establishment, but a con-

88 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

stant seed bank. Houle noted that the uncoupling of the seed bank, rain, and
seedling establishment further contributes to the spatiotemporal heterogene-
ity. It seems probable that similar complex spatiotemporal relationships
influence Botrychium distribution.

Compounding the variable numbers of spores added annually to the pro-
pagule bank is the limited dispersal of spores. It is unknown how widespread
moonwort spores disperse but based on the work of Peck et al. (1990) on
Botrychium virginianum we can conclude that most spores disperse within
5 m or less. Dyer (1994) found that the largest spore banks occurred in samples
taken immediately below ferns and that at a distance of 2 m away from the
spore source, the spore bank was notably smaller. It seems likely that a few
spores may become airborne and disperse farther. Because of the ability
of Botrychium gametophytes to self-fertilize, it is reasonable to expect that
a single spore is capable of dispersing and establishing a new population
(Farrar, 1998).

Over time, a sizeable Botrychium spore bank is established in the soil.
Botrychium spores likely remain viable for long periods of time, as do
many other ferns (Lloyd and Klekowski, 1970: Miller, 1968; Sussman, 1965;
Windham et al., 1986). These spores are probably dormant until conditions
(moisture and mycorrhizae) are adequate, at which time many or all the
spores in that localized area germinate and develop.

densities of 15,000 spores m2:
have the lowest at 100 m ”. This

tion, mortality at this stage j
expect high spore densities within Botrychium populations.

JOHNSON-GROH ET AL.: BELOWGROUND DISTRIBUTION 89

Species with gemmae (B. campestre, B. gallicomontanum) have a higher
total belowground density than those without gemmae. Like spores, gemmae,
once detached from the parent plant, require mycorrhizae for further deve-
lopment. Farrar and Johnson-Groh (1990) found relatively few gemmae that
contained mycorrhizae, which could explain the low number of developing
gemmae relative to the number of gemmae produced. (Gemmae obtain myce-
lia through their connection with the parent rhizome; if unsuccessful, they
remain dormant.) The primary role of gemmae may be to maintain the popu-
lation in a microsite that has already proven successful. The frequent occur-
rence of multiple stems within a small-localized area (1-4 cm”) suggests that
gemmae are effective in local propagation. Dispersal beyond a short distance
is limited, as evidenced by the low frequency of the highly gemmiferous spe-
cies (B. campestre, B. gallicomontanum).

Species that produce profuse gemmae produce the lowest number of game-
tophytes (B. campestre, B. gallicomontanum). Gemmae, a form of asexual
reproduction, produce essentially the same genetic product that a selfing
gametophyte produces. The advantage of gemma production is the position-
ing for immediate success (mycorrhizae present). A greater reliance on repro-
duction via spores and gametophytes by most species and the higher
disperability of spores undoubtedly accounts for the higher frequencies in
soil samples of the non-gemmiferous species. The advantage of spore—game-
tophyte production allows dispersal to new sites, thereby insuring that
“assets are diversified,” which may provide a long-term advantage to the
species. To draw further from investment analogies, gemmae are short-term
investments with immediate returns, whereas spores are long-term invest-
ments with greater evolutionary payback.

Som. HETEROGENEITY.—Variations in microtopography, microclimate, parent
material, mycorrhizae, and microorganisms all influence soil heterogeneity
(Stark, 1994), creating a patchy environment. Because the spatial scale is
very small, conditions merely a few centimeters away may not be sufficient
to induce germination, thus creating a patchy distribution of plants.

Of these variable factors, mycorrhizae are probably the most important for
Botrychium. Moonworts require endophytic mycorrhizae for gametophyte
and sporophyte development and are dependent on mycorrhizae as a signifi-
cant source of carbohydrate, minerals, and water. This observation is based
on several peculiar behaviors. First, similar to orchids, moonworts do not
emerge every year. They frequently fail to emerge for one to three consecu-
tive years, with no subsequent decrease in size or other negative effects
(Johnson-Groh, 1998). Second, “albino” botrychiums have been observed
Another indication that Botrychium depends relatively little on its own
leaves for photosynthesis is the observation that these leaves frequently do
not emerge above the litter. In fact only a small proportion of the total popu-
lation of some species actually emerges from the litter (Johnson-Groh, 1998).
Herbivory and loss of leaves through fire do not affect the size and vigor
of plants in the subsequent year (Johnson-Groh, 1998). Finally, if roots and

90 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

leaves of juvenile plants are produced one per year, as in adults, 5-8 years
may be required for development from gametophyte to a mature sporophyte
with an emergent photosynthetic leaf (Johnson-Groh, 1998). Juvenile plants
must rely on mycorrhizae for carbohydrates. Thus, although there has been
no physiological studies to confirm this, it seems certain that moonworts
(Botrychium subg. Botrychium) may depend largely on mycorrhizae for carbon
from other plants, in addition to that produced by their own photosynthesis.

If photosynthesis is not critical for this subgenus and mycorrhizae are pri-
marily responsible for overall energy budget, then understanding the below-
ground biology of Botrychium is imperative. Indeed, assumptions made
about the population biology of other ferns may be irrelevant to Botrychium.
Health of the mycorrhizal connection may determine juvenile recruitment
and survivorship, and Botrychium populations may appear or disappear in
accordance with mycorrhizal health.

Mycorrhizae play an important role in nutrient acquisition. This may be
especially important for Botrychium because of the inability of its roots to
forage. Root-foraging has been observed in flowering plants (Caldwell, 1994).
It allows them to respond to small-scale nutrient patches. However,
Botrychium roots are relatively few (5-30/plant), do not have root hairs, and
do not appear to have the morphological plasticity to forage for small-scale
patches of soil nutrients. Typically roots extend almost perfectly horizontally
for their entire length (3-20 cm). Only occasionally are roots observed to
abruptly bend in another direction. Tibbet (2000) argued that mycorrhizae
are especially important for roots that do not have the morphological plasti-
city to respond to small-scale nutrient patches. Mycelia rapidly colonize
patches of soil nutrients, making them ideal foraging instruments of the au-
totroph. In Botrychium it seems highly probable that its mycorrhizal mycelia
are more important than root proliferation in nutrient acquisition.

Botrychium species that have high belowground densities generally have
high aboveground population densities. Botrychium campestre, B. gallico-
montanum, and B. mormo have the highest below- and aboveground den-
sities. Botrychium virginianum has a relatively low below- and aboveground
density. The ratio of belowground to aboveground plants ranges from 65:1 in
B. mormo to 914:1 in the gemmiferous B. campestre.

For each species in this study, a relatively small volume of soil (962 cm”)
was sampled across a large spatial grid (201 m?*). Estimations of density are
derived from these results. The patchy distribution and inability to sample
large volumes of soil make it difficult to determine precise belowground
populations. The opposite approach of intensively and completely surveying
both above- and belowground in a small area is currently being investigated
for B. virginianum and B. campestre. Preliminary results indicate that the
belowground density in a small patch is several times the aboveground
density.

Despite the difficulty of making direct comparisons below- and above-
ground, in all cases it is readily apparent that belowground structures
are much more abundant than aboveground plants. Sizeable reservoirs of

JOHNSON-GROH ET AL.: BELOWGROUND DISTRIBUTION 91

belowground structures are extremely important to the population dynamics
of Botrychium and serve to replenish the populations following environmen-
tal perturbations. Because moonworts often remain dormant in any given
season, they are essentially protected a aacoreh pe and can easily withstand
dry years, fires, herbivory, or other aboveground disturbances. Despite highly
variable aboveground populations, belowground stages provide Botrychium
populations with a high degree of buffering against local extinction. The
belowground structure bank is a reserve of plants in various stages that can
eventually produce an aboveground population, regardless of past above-
ground perturbations.

Dyer (1994) noted the importance of fern propagule banks to the conserva-
tion of fern species. Reintroducing or augmenting populations from the spore
bank broadens the options available for many species. Whereas it is difficult
to manipulate Botrychium belowground structure banks, it is important to
recognize the importance of these banks to the overall population dynamics.
Modeled Botrychium populations (Johnson-Groh et al., 1998) predict greater
stability of populations than would be concluded from monitoring only
aboveground plants. This resiliency is a result of the large belowground re-
serve of gametophytes and juvenile sporophytes capable of regenerating the
population. The long-term impact of environmental perturbations on popula-
tions is buffered by a large bank of belowground structures.

AACKNOWLEDGMENTS

We acknowledge partial funding provided by the Colville and Tongass National Forests and
Gustavus Adolphus College. We are grateful to Kathy Ahlenslager and Mary Stensvold for their
interest, support, and help with this study. We thank Don Farrar for discussions and for provid-
ing data on species examined in his laboratory.

LITERATURE CITED

CaLDWELL, M. M. 1994. ae nutrients in fertile soil microsites. Pp. 325-347 in M.

Caldwell and R. earcy, eds. Exploitation of environmental heterogeneity by ae
Academic es as mi ego.
Canin, R. J. and D. L. MarsHALL. 2000. The demographic role of soil seed banks. I. Spatial and

temporal sien of below- and aboveground populations of the desert mustard Les-
querella eee J. Ecol. 88:283-—292.

Durinc, H. J. and B. TER ip 1983. The diaspore bank of bryophytes and ferns in chalk grass-
land. Lindbergia 9:5

Dyer, A. F. 1994. Natural es spore banks—can they be used to retrieve lost ferns? Biodiversity
and Conservation 3:1

and S. Linpsay. 1992. Soil spore banks of temperate ferns. Amer. Fern J. 82:89-123.

Farrar, D. 3 1998. dh trai ones of moonwort Botrychiums. Pp. 109-113 in N. Berlin
P. Miller, J. Borovansky, U. S. Seal, and O. Byers, eds. Population and habitat abba
seni (PHVA) for hee ae fern anual mormo) Final Report. Conservation
Biology Specialist Group, Apple Valley, M

and OHNSON-GROH. 1990. rissa he i gemmae in moonwort ferns, Bo-

trychiu umm subgenus Botrychium. Amer. J. Bot. 77:1168-1175.

HamiLTon, R. G. 1988. The significance of spore <a in natural populations of Athyrium pycno-

arpon and - eriearbrra Amer. Fern J. 78:98—104.

92 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

Hous, G. 1998. Seed dispersal and seedling recruitment of Betula alleghaniensis: spatial incon-
sistency in time. geile :807-818

JOHNSON-GroH, C. L. 1 é. Population EOE ae ecology and phenology of

Botrychium mormo. x 103-108 in N. Berlin, P. Miller, J. Borovansky, U. S. Seal, and

pe Byers, eds. Population and habitat viability assessment (PHVA) for the goblin fern

Final Report. Conservation Biology Specialist Group, Apple Valley, MN.

. 1997. Field surveys for Botrychium aang rae and weg oy Botrychium mor-
mo in Minnesota. Report to the Minnesota Dept. Nat. Resources, St. Paul, MN.

—— and D. R. Farrar. 1996. The effects of fire on prairie moonworts eee subgenus
spe aga Amer. J. Bot. 83:134 (abstract).

dP. MILLER. 1998. Modeling extinction probabilities for moonwort (Botrychium)
meu: Amer. J. Bot. 85:95 (abstract).

—— and J. M. Lex. in en Phenology and demography of two species of Botrychium (Ophio-
glossaceae). Am

Kauisz, S. 1991. gece determination of seed bank age structure in the winter annual
Colinsia verna. Ecology 72:575-585.

Leck, M. A. and R. L. Smwpson. 1987, Spore bank of a Delaware River freshwater tidal wetland.
Bull. Torrey Bot. Club 114:1

, V. T. Parker, and R. L. mi 1989. Ecology of soil seed banks. Academic Press Inc.,
San Diego, CA.

Lioyp, R. M. and E. J. KLekowski. 1970. Spore germination and Latent in Pteridophyta: evolu-
niga i nae of chlorophyllous spores. Biotropica 2:1

bert N. and D. R. Farrar. 1989. Recovery of Henan gametophyts gemmae and

mature sporophytes by centrifugation. Amer. J. Bot. 79:143—
eases P. 1991. Fern spores in grassland soil. Canad. J. Bot. ee ak
LLER, J. H. 1968. Fern gametophytes as experimental material. Bot. Rev. (Lancaster) 34:
361-440.

MINNESOTA oo Or Natura Resources. 2002. Minnesota’s List of Endangered, Threatened
and ecial Concern _ Species. http://www.dnr.state.mn.us/fish_and_wildlife/
endangered. species/,

Prck, J. H., C. J. Peck, and D. R. Farrar. 1990. Influences . = pees attributes on formation of
local and distant fern populations. Amer. Fern J. 80:

RaFFAELE, E. 1996. Relationship between seed and spore ee aaa vegetation of a mountain
flood meadow (mallin) in ae Argentina. Wetlands 16:1-9.

YDGREN, K. and G. HesTmMark. 1997. The soil propagule bank in a boreal old-growth spruce
cae hapa with depth aa relitionships to aboveground vegetation. Canad. J. Bot. 75:

nee ' oe 988. Spore bank, dark germination and gender determination in Athyrium and

Pn hog Results and implication for population biology of Pteridophyta. Bot. Helv.
98:77-86.

STARK, J. re 1994. Causes of soil nutrient heterogeneity. Pp. 255-284 in M. M. Caldwell and
R. W. Pearcy, eds. Exploitation of environmental heterogeneity by plants. Academic Press,
an Diego.
StRICKLER, G. S. and P. J. EDGERTON. 1976. Emergent seedlings from coniferous litter and soil in
eastern gs a Ecology 57:801-807.

Sussman, A. S. 1965. Physiology of dormancy and germination in the propagules of Sere

plants. Pp. 297-315 in W. Ruhland ed. Handbuch der Pflanzenphysiologie XV. Springer
Verlag, Berlin.

TipBeT, M. 2000. Roots, foraging and wes ex
rhizal 5 yey Funct. sia 14
Wuirtier, D. P. and R. D. THom 1993. yin ametophytes “0 young sporophytes of Botrychium

jenmanii | in axenic culture. te J. Plant Sci. — 68-—
WinDHaM, M. D., P. G. Wotr, and T. A 986. it affecting sterol ae viability
in herbarium collections of three species es Pellaea. Amer. Fern J. 76

capone of soil nutrient patches: the role of mycor-
—399.

American Fern Journal 92(2):93—96 (2002)

Additions to the Fern Flora of Saba,
Netherlands Antilles

Davip B. LELLINGER

Department of Botany, National Museum of Natural History Smithsonian Institution,
Washington, DC 20560-0166

ABSTRACT.—Recent fieldwork on Saba, Netherlands Antilles, has resulted in the discovery of nine
fern species unrecorded for the island. All are known from other islands in the Antilles, as well
as from other tropical areas.

Saba is a volcanic island about three kilometers in diameter that rises
to about 850 m in a graceful cone called Mt. Scenery. The summit is the
highest point in the Kingdom of The Netherlands. Saba lies south of Sint
Maarten/St. Martin in the Windward Island chain at the northern end of the
Lesser Antilles. About 1,300 people reside on Saba, which is little visited by
tourists because of its small size and precipitous, rocky headlands that pre-
vent beaches from forming. A beautiful fringing reef, recently declared a na-
tional park, makes the island a choice destination for SCUBA divers, but
Saba also offers opportunities for hiking, birding, and botanising.

Although the island has been occupied for more than 350 years, a roller
coaster-like road running the length of the island was hewed out of the rock
and extensive retaining walls to contain the road were built by hand less
than a half century ago. Prior to the road, all travel on the island was on
foot, over trails that for the most part run rather circumferentially. Even so,
the trails are by no means level because of the lava ridges and valleys be-
tween them, locally called guts. A day’s hike usually includes elevation
changes of at least 500 m. In recent years, additional trails have been con-
structed on the slopes and to the top of Mt. Scenery to make more of the is-
land accessible to hikers. The road now connects the three principal villages
with a dock and jetty at the southwestern end of the island, a minute beach
(when it is exposed at all) on the western side, and a very short air field on
the northeastern end, which is suitable only for short takeoff and landing
aircraft. Ferry boats run from the dock and jetty.

Mt. Scenery is high enough to be covered by rain clouds for much of the
time, and the island is therefore more moist than Sint Maarten/St. Martin
and other nearby low islands that depend entirely upon convection for their
rainfall. The additional water has resulted in a diverse native vegetation that
has been divided into five types, which occur at different elevations above
the sea as roughly concentric circles: grassy scrub near sea level, dry wood-
land above grassy scrub on the leeward side, moist forest above these types,
rain forest above moist forest, and cloud forest at the summit of Mt. Scenery.

94 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

In the 19th and 20th centuries, much of the forest was destroyed by crop
production or grazing, especially at lower elevations and where the upper
slopes are not so steep. With improved access by boats, the residents of Saba
imported more of their food, and cultivation and grazing ceased in the less
accessible areas of the island. Native vegetation used to return slowly
through a succession of relatively temporary associations of plants, the first
of which was tree fern brakes (Cyathea arborea and C. muricata) (Hodge,
1990; Romeijn, 1989).

On the windward (eastern) side of the island, the forest vegetation was
highly degraded by hurricanes in the 1990’s. Many trees were uprooted, and
those that remained were broken and battered. What was ferny, shaded for-
est floor is now almost fully open to the sun. Because of this, the dominant
herbs are now Elephant Ear (Philodendron giganteum), various Heliconia
species, and invasive vines and weeds. In the destroyed areas, there is little
difference between the moist, rain, and cloud forests. All have open cano-
pies, and weedy species are dominant on the ground. On the leeward (west-
ern) side of the island, the forests are much more intact, and it is still
possible to distinguish between the forest types by their structure, moisture,
dominant flowering plants, and complement of ferns.

Nephrolepis multiflora is often dominant in formerly cultivated fields on
the windward slopes of Mt. Scenery, where it forms an impassable layer
ca. 1 m tall that can persist for at least a decade and probably much longer.
Only a few tree ferns appear to be capable of germinating and growing
through the mat, but they form insufficient cover to make a tree-fern brake.
They fail to shade and to kill the Nephrolepis and so cannot put in motion
the normal succession to moist forest or rain forest.

The usual processes of succession at elevations below 600 m are also im-
peded by wild goats, which cause erosion and eat the leaves and shoots of
young trees before they grow beyond reach, preventing regeneration of the
forest (Romeijn, 1989). At present, hunting for meat is insufficient to over-
come the goats’ reproduction. A small part of the northern slopes of Mt.
Scenery, encompassing all the vegetation types, and the entire summit of the
mountain are now a protected park. Fencing this area to exclude goats
would be good park management and would provide an opportunity for in-
teresting ecological studies as well.

LELLINGER: ADDITIONS TO THE FERN FLORA 95

At the suggestion of G. R. Proctor, my wife Jeannette and I spent several
days in April, 1999 hiking trails and occasionally roadsides, mostly in the
higher, forested parts of the island. (The grassy scrub might better be
searched for ferns during the wet season in the winter.) The trails around
the island are maintained by volunteers, and only the less travelled ones re-
quire the services of a guide. Considering how little botanising Saba has
received, it is no surprise that we were able to find the species in the
following list new to the island.

Ctenitis meridionalis (Poir. in Lam.) Ching.—Rare, terrestrial along trail
between Sandy Cruz and Troy, near Down Gut above Wells Bay, in moist
forest at 500-600 m elevation, Lellinger 2034 (U, US).

Endemic in the Lesser Antilles.

Diplazium cristatum (Desr.) Alston.—Rare, terrestrial in shady places
along trail above Upper Hell’s Gate W and NW to beyond Sandy Cruz, in
rain forest at 500-600 m elevation, Lellinger 2033 (U, US).

Widespread in tropical America from northern South America northward.

Nephrolepis multiflora (Roxb.) Jarrett ex Morton.—Common along road-
sides between Big Rendevous and English Quarter, in hurricane-damaged
moist forest and rain forest at 400-600 m elevation, Lellinger 2028 (U, US).

An Old World species naturalized sporadically in the Antilles, Mexico to
Peru, and Brazil.

Pteris tripartita Swartz.—Seen along the trail in moist forest along the
trail between Sandy Cruz and Troy.

An Old World species naturalized widely in the Antilles and from Nicaragua
to Venezuela and Bolivia.

Thelypteris hispidula (Dcne.) Reed var. hispidula.—Rare on street-side
rock walls at The Level, SW of Windwardside, in shade at 400-500 m eleva-
tion, Lellinger 2027 (U, US).

Widely distributed in tropical America.

Thelypteris poiteana (Bory) Proctor.—Occasional along the trail from Troy
to Bottom Hill, terrestrial in moist forest at 450-500 m elevation, Lellinger
2038 (U, US).

Widely distributed in tropical America.

Trichomanes krausii Hook. & Grev.—Rare along the trail between Sandy
Cruz and Troy, near Down Gut above Wells Bay, on shady, clay banks in
moist forest at 500-600 m elevation, Lellinger 2037 (U, US).

Widely distributed in tropical America.

Trichomanes membranaceum L.—Rare along the trail between Sandy
Cruz and Troy, near Down Gut above Wells Bay, on shady, clay banks in
moist forest at 500-600 m elevation, Lellinger 2035 (U, US).

Widely distributed in tropical America.

Trichomanes punctatum Poir in Lam. subsp. punctatum.—Occasional
along the trail between Sandy Cruz and Troy, near Down Gut above Wells Bay,
on shady banks in moist forest at 500-600 m elevation, Lellinger 2036 (U, US).

Distributed in the Antilles, Trinidad and Tobago, and Venezuela.

96 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

LITERATURE CITED

Hopce, W. 1990. Wildflower (Autumn 1990):34-39,

Kramer, K. U. 1962. aye of the Netherlands detities vol. I. Pteridophyta. Natuurw. Stud. Sur.
Ned. Antill. 25

Proctor, G. R. 1977. grin of the Lesser Antilles (Leeward and Windward jai, vol. 2. Pteri-
dophyta. Arnold Arboretum, Harvard University, Jamaica Plain, MA

RomEYN, P. 1989. The forest of Saba. BOS NiEuWSLETTER 8(19):44—51.

American Fern Journal 92(2):97—104 (2002)

Taxonomic Notes on Hawaiian Pteridophytes

DANIEL D. PALMER
1975 Judd Hillside, Honolulu, HI 96822

AssTRACT.—A review of the literature regarding Hawaiian pteridophytes, examination of herba-
rium specimens, and much field work resulted in the need to publish two new varieties and to
make thirteen new combinations.

In preparing a book reviewing the pteridophyte flora of Hawaii, the author
examined type specimens, reviewed collections at several herbaria, con-
sulted literature, observed common garden studies, and conducted much field
work. Findings from these studies necessitate the publication of thirteen new
combinations (Adenophorus pinnatifidus var. rockii, Asplenium horridum
var. glabratum, A. kaulfussii f. bipinnatum, A. kaulfussii f. gemmiparum,
A. peruvianum var. insulare, Christella x intermedia, Christella wailele,
Dryopteris crinalis var. podosorus, D. glabra var. alboviridis, D. glabra var.
hobdyana, Hypolepis hawaiiensis var. mauiensis, Microlepia strigosa var.
mauiensis, and Microsorpum spectrum var. pentadactylum) and the descrip-
tion of two new varieties (Dryopteris glabra var. flynnii, and Polypodium pel-
lucidum var. acuminatum).

Adenophorus pinnatifidus Gaudich. var. rockii (Copel.) D. D. Palmer, comb.
nov.

Polypodium rockii Copel. Philipp. J. Sci. 11:173. 1916. Adenophorus
sarmentosus (Brack.) K. A. Wilson var. rockii (Copel.) O. Deg. & I. Deg.,
Fl. Hawaiiensis, fam. 17a. 1967.

Adenophorus pinnatifidus var. rockii is distinguished from var. pinnatifi-
dus by its longer, narrower, more linear, acute-tipped segments arising at an
oblique angle from the midrib, and often by the presence of more sori on the
lobes (var. pinnatifidus with 3-4 sori per lobe, and var. rockii with 1—7).

Asplenium horridum Kaulf. var. glabratum (W. J. Rob.) D. D. Palmer, comb.
nov.

Asplenium glabratum W.J. Rob., Bull. Torrey Bot. Club 40:214. 1913.

Asplenium glabratum is here recognized as a variety. It differs from A.
horridum var. horridum in the degree of scaliness, var. horridum being very
scaly and var. glabrum nearly glabrous. However, one can find a continuum
of plants with intermediate degrees of scaliness between the two extremes.
Hillebrand first recognized this taxon as Asplenium horridum Kaulf. var. B
(Fl. Hawaiian Isl. 604. 1888).

98 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

Asplenium kaulfussii Schltdl.

This is an extremely variable species. Most plants are 1-pinnate with acute
or obtuse-tipped pinnae, but some may produce pinnae with margins that
are lacerate or deeply incised with toothed segments, and rarely the fronds
may even be 2-pinnate. Usually nonproliferous, plants in some scattered col-
onies can produce a few proliferations or sometimes proliferations can be
found on all pinnae. Variant fronds are sometimes found on the same rhi-
zome as the typical 1-pinnate form. I have chosen to recognize these variants
as forms of A. kaulfussii. The following key separates the forms.

ecabvsnae ccosvathea! ‘esi nice GEE TOE OO ET OEE FUT ECC Te eae 2
1. Fronds 1-pinnate-pinnatisect to 2-pinnate.... 2.0.0.0... 000 ccc cece cece eee... ,
2(1). Pinna margins entire, lacking proliferations........................... . kaulfussii
2. Pinna margins entire, proliferations occasional to plentiful on the lateral veins of pinnae ..

metal Retin Silke ce tartey wirhe att atten aeok Gy een ners am tnehee EIN ee TTL, bes NN Ly nied Wk ed - gemmiparum

Maes 2 horse Ca Area ah Otc epee aga Bee ep eM ao ies aicaican ag al ava i wi /Saearn skeoa om & f. bipinnatum
3. Fronds 1-pinnate-pinnatisect to fully 2-pinnate, the ultimate segments linear with 2(4) ob-
Mise lobes nn am éupundind apex 20 dk ee ccc ccc ecccacn, . dareoides

Asplenium kaulfussii Schltdl. forma bipinnatum (Hillebr.) D. D. Palmer,
comb. nov

Asplenium bipinnatum Hillebr. Fl. Hawaiian Isl. 595. 1888. A. lydgatei
Hillebr. Fl. Hawaiian Isl. 596. 1888. A. meiotomum Hillebr. Fl. Hawaiian
Isl. 596. 1888.

This rare and variable form is fully 2-pinnate, with long-stalked pinnae
and short-stalked pinnules that are smaller but of the same character as the
pinnae of 1-pinnate fronds.

Hillebrand originally described A. bipinnatum and noted its relationship
to A. kaulfussii. This isolated form of A. kaulfussii has ovate pinnae with 7
or 8 pinnules on each side. It was found in Niu Valley on Oahu, but has not
been collected for many decades.

Asplenium kaulfussii Schltdl. forma dareoides (Hillebr.) W. H. Wagner,
Contr. Univ. Michigan Herb. 22:160. 1999.

‘This rare form, not recently collected, is characterized by pinnatisect to 2-
pinnate pinnae and linear ultimate segments with tips expanded into 2(4)
obtuse lobes. Fronds of this form have on a few occasions been found on rhi-

Asplenium kaulfussii Schltdl. forma gemmiparum (Hillebr.) D. D. Palmer,
comb. nov.

Asplenium kaulfussii Schltdl. var. gemmiparum Hillebr. Fl. Hawaiian Isl.
992. 1888. A. enatum Brack. U. S. Expl. Exped., Fil. 153 t. 21, f. 1. 1854

PALMER: TAXONOMIC NOTES ON HAWAIIAN PTERIDOPHYTES 99

A. enatum Brack. var. gemmiparum Hillebr. Fl. Hawaiian Isl. 593. 1888. A.
gemmiferum Schrad. var. enatum (Brack.) O. Deg. & I. Deg., Fl. Hawaiien-
sis, fam. 17d. 1959. A. mannii Hillebr. var. gemmiparum Hillebr. FI.
Hawaiian Isl. 595. 1888.

This uncommon, localized, 1-pinnate form bears one or more prolifera-
tions on the adaxial surface on one or a few lateral veins of a single pinna,
or on several pinnae including the apical one. There are forms with very few
proliferations, but the variation in this character seems to be continuous,
and nonproliferous fronds are sometimes found on the same rhizomes as
proliferous ones.

Asplenium kaulfussii Schltdl. forma kaulfussii

This is the typical 1-pinnate form with entire pinnae, frond tips resem-
bling lateral pinnae, and no proliferations. The pinnae tips vary from wide-
obtuse, to long-acute.

Asplenium peruvianum Desv. var. insulare (C. V. Morton) D. D. Palmer,
comb. nov.

Asplenium fragile C. Presl var. insulare C.V. Morton, Amer. Fern J. 37: 116.
1947. A. fragile sensu Hillebr., non C. Presl, Tent. Pterid. 108. 1836. A.
rhomboideum sensu auctt., non Brack. U. S. Expl. Exped., Fil. 16:156, t. 21.
1854.

The name Asplenium peruvianum Desv. is recognized as having priority
for the taxon commonly known as A. fragile C. Pres]. Some authors have rec-
ognized A. rhomboideum Brack. as a synonym of A. triphyllum C. Presl, an-
other fern in a group with close affinities to A. peruvianum.

Asplenium peruvianum var. peruvianum is native to the Peruvian Andes.
The Hawaiian variety is larger and coarser, with thicker rachises, larger
fronds, and a superior basal lobe on almost all the pinnae.

Asplenium peruvianum var. insulare may in fact include two taxa: a
delicate, nonproliferous, longer, narrower, light green form often found in
dark lava tubes, and a coarser, shorter, wider, darker green, usually pro-
liferous form found in more open areas. This taxon is part of a group
of closely related species that needs further study in Hawaii as well as in
Peru.

Christella x intermedia (W. H. Shieh & J. L. Tsai) D. D. Palmer, comb. nov.

Cyclosorus x intermedius W. C. Shieh & J. L. Tsai, J. Sci. Engin. (Natl.
Chung-Hsing Univ.) 24:8. 1987. Christella Xincesta (W. H. Wagner)
Nakaike & Kawabata, J. Nat. Hist. Mus. Inst. Chiba 3(2):145. 1995. Cyclosorus
x incestus (W. H. Wagner) W. H. Wagner, Contr. Univ. Michigan Herb.

100 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

22:156. 1999. Thelypteris x incesta W. H. Wagner, Contr. Univ. Michigan
Herb. 19:79. 1993.

A book in preparation will include, and some older treatments would have
included, this hybrid in the genus Christella, along with the related C. cya-
theoides, C. boydiae, C. dentata, C. parasitica, and C. wailele.

This hybrid had been thought limited to Hawaii when the name Thelypte-
ris Xincesta was published in 1993. However, the name Cyclosorus inter-
medius, identifying this hybrid in Taiwan and published in 1987 clearly has
priority.

Christella wailele (Flynn) D. D. Palmer, comb. nov.

Thelypteris wailele Flynn, Contr. Univ. Michigan Herb. 20:246. 1995. Cyclo-
sorus wailele (Flynn) W. H. Wagner, Contr. Univ. Michigan Herb. 22:156.
1999.

A book in preparation will include, and some older treatments would have
included, this recently described fern in the genus Christella, along with the
related C. cyatheoides, C. boydiae, C. dentata, and C. parasitica.

Dryopteris crinalis (Hook. & Arn.) C. Chr. var. podosora (W. H. Wagner) D.
D. Palmer, comb. nov.

Dryopteris podosora W. H. Wagner, Contr. Univ, Michigan Herb. 20:252.
1995.

Dryopteris crinalis var. podosora may be distinguished from var. crinalis
by sori that are globular and raised on conspicuous, narrow, pale, vascular-
ized, stalks 0.8-1.3 mm long. All other characteristics of this taxon fall with-
in the range of variation seen in var. crinalis.

Dryopteris crinalis var. podosorus is very rare and is restricted to wet,

mossy walls near streams in damp, dark valleys near the Pihea Trail in the
Kokee area of Kauai.

Dryopteris glabra (Brack.) Kuntze var. alboviridis (W. H. Wagner) D. D.
Palmer, comb. nov.

Ae alboviridis W. H. Wagner, Contr. Univ. Michigan Herb. 22:177.

Dryopteris glabra var. alboviridis differs from var.
color, coarser Cutting with usually larger, wider ultimate segments (var. albo-
viridis 3-5 mm wide and var. glabra 2-3 mm wide) with crenate-dentate
margins, and thicker and more leathery texture.

Careful study of this localized fern near the Pihea Trail in the Kokee area

Kauai has shown that while the polar form of the variety

glabra by its light green

of ; magi
is distinct, a

PALMER: TAXONOMIC NOTES ON HAWAIIAN PTERIDOPHYTES 101

continuum of forms blending into D. glabra var. glabra is found within a
few hundred meters of plants typical of var. alboviridis.

Dryopteris glabra (Brack.) Kuntze var. flynnii D. D. Palmer, var. nov.—Type:
Kauai: Waimea District: Kauaikinana Stream, W of Pihea Trail, D. D. Palmer
& J. Obata, 14 Sept 1994, K. Robinson 2139 (holotype BISH).

Dryopteris glabra var. glabrae similis sed differt lamina linearitriangulari
atque plus elongata, stipite angustiore, segmentis ultimis angustioribus, et
glandulis multo numerosioribus.

This variety is similar to D. glabra var. glabra, but differs in having linear-
triangular and more elongated lamina, narrower stipes, narrower ultimate
segments, and lamina covered with many more glands.

Plants delicate, drooping, very glandular, found on wet, dark stream
banks. Rhizomes decumbent bearing sparse, brown, linear-triangular, 5-8
mm long scales. Fronds drooping, to 60 cm long. Stipes thin, 1-2 mm diam.,
stramineous to brown, glabrous, except for a few scattered scales at very
base. Blades linear-triangular, 2-3 times as long as wide, 3—4-pinnate, pale
green, very glandular; rachises thin, mostly 1-1.5 mm diam. Pinnae heavily
clothed with globular, adnate glands, with multicellular, uniseriate hairs
scattered on the veins; ultimate segments mostly less than 1.5 mm wide, ex-
cept for wider anterior basal lobe, rounded to toothed at tips. Sori marginal
to submarginal at tips of ultimate segments. Indusia glandular.

OTHER SPECIMENS EXAMINED.—Hawaii: Kauai: Waimea District: Kokee State Park, Kauaikinana
Stream W of Pihea Trail, in Metrosideros-dominated wet forest with Cheirodendron, Broussaisia,
Cyanea, Melicope, Cibotium, Diplazium, and Athyrium, elev. 1100 m, terrestrial on vertical,

shaded, mossy wet wall, just above stream, 28 July 1992, W. H. Wagner, F. S. Wagner, D. D.
Palmer & T. Flynn 92117 (MICH).

Presently known only from a local population, growing between 1200-
2000 m on the lower parts of shaded, mossy, wet banks along Kauaikinana
Stream west of the Alakai Swamp Trail on Kauai.

The varietal name honors Tim Flynn, collection manager of the herbarium at
the National Tropical Botanical Garden, who discovered this variety in 1992.

Fronds of this delicate and extremely glandular variety stick tightly to the
papers they are dried in and must be removed with care. Its drooping habit,
pale green color, fronds that are 2 to 3 times as long as wide, and its habitat on
shady, wet banks distinguish it from the other varieties of Dryopteris glabra.

Dryopteris glabra var. hobdyana (W. H. Wagner) D. D. Palmer, comb. nov.
Dryopteris hobdyana W. H. Wagner, Contr. Univ. Michigan Herb. 22:177.

102 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

Study of this fern in its habitat on the upper slopes of Haleakala on Maui
has shown that while the extreme manifestation of this variety is quite dis-
tinct, a continuum of plants with features blending into Drvyopteris glabra
var. glabra is found within a few hundred meters of plants typical of var.
hobdyana.

A rhizome of this fern, removed from its habitat at an elevation about
2150 meters on the north slope of Haleakala—where it is exposed to trade
winds, bright morning sun, afternoon cloudiness, and frequent freezing tem-
peratures at night—was transplanted to a sheltered, less exposed, much
warmer climate at Lyon Arboretum in upper Manoa Valley on Oahu. A few
fronds typical of D. hobdyana were left attached to the rhizome. The very
next frond that emerged from this rhizome, and several fronds since, were
typical of var. glabra, suggesting that plants typical of D. hobdyana may be
ecotype forms.

Further study is needed. For the present, this distinctive taxon is assigned
the varietal name D. glabra var. hobdyana.

Hypolepis hawaiiensis Brownsey var. mauiensis (Hillebr.) D. D. Palmer,
comb. nov.

Phegopteis punctata (Thunb.) Hillebr. var. mauiensis Hillebr. Fl. Hawaiian
Isl. 663. 1888.

Hypolepis hawaiiensis var. mauiensis is a fully fertile miniature variety of
H. hawaiiensis. Fertile fronds range from 6-25 cm long with blades that are
1-pinnate-pinnatifid to 3-pinnate. The rhizomes are slender (1-3 mm diam.)
and long-creeping. The stipes, rachises, and costae are well covered with
fine, tan, catenate (chain-like), acute-tipped hairs.

This rare fern occurs only on West Maui, where it was recently collected
at an elevation of 1700 m on Puu Kukui in the Kapunakea Preserve; it was
reported from Mount Eke by Hillebrand in 1888.

Microlepia strigosa (Thunb.) C. Pres] var. mauiensis (W. H. Wagner) D. D.
Palmer, comb. nov.

Microlepia mauiensis W. H. Wagner, Contr. Univ. Michigan Herb. 19:73, f. 6.
1993.

PALMER: TAXONOMIC NOTES ON HAWAIIAN PTERIDOPHYTES 103

Microsorum spectrum (Kaulf.) Copel. var. pentadactylum (Hillebr.) D. D.
Palmer, comb. nov.

Polypodium spectrum Kaulf. var. pentadactylum Hillebr. Fl. Hawaiian Isl.
560. 1888.

The 5-lobed blades of var. pentadactylum distinguish it from var. spec-
trum, which has a triangular, 3-lobed blade. It is found in scattered areas on
Kauai.

Polypodium pellucidum Kaulf. var. acuminatum D. D. Palmer, var. nov.—
Type: Kauai: Waimea District, Koaie Stream in upper Waimea Canyon, 16
Apr 1991, D. D. Palmer, B. Gustafson & staff of NTBG 586 (holotpe BISH).

Polypodium pellucidum var. pellucidum simulans sed differt lamina line-
arideltata, lobis acutis margine crenulatis, venis basalibus pinnarum basa-
lium interdum anastomosantibus, et pagina abaxiali pilis brevibus vestita.

This variety resembles P. pellucidum var. pellucidum, but differs in hav-
ing linear-deltate blades, acute lobes that are crenulate on the margin, basal
veins of the basal pinnae sometimes anastomosing, and the abaxial surface
clothed with short hairs.

Fronds to 55 cm long. Rhizomes long-creeping, 3-5 mm in diam., the
scales light brown to tan, linear-lanceolate, narrowing to a hairlike tip.
Blades linear-deltate, deeply pinnatisect to fully 1-pinnate at the base with
2-several lobes not winged to the rachises; lobes of blades up to 32, long,
narrow, acuminate, with regularly crenulate to crenulate to serrate margins;
false veins absent; fine, round-tipped hairs numerous on the abaxial surfaces
of the lobes, costae, and veins, these 0.1 mm long, 3—7-celled, sparse on the
adaxial surfaces: veins often anastomosing to form a few areolae at the bases
of the lobes, especially lowest lobes; sori medial.

OTHER SPECIMENS EXAMINED.—Kauai: Waimea District: Koaie Canyon 1-2 miles above Lonomea
Camp, elev. 520-580 m, in crevice of rock face above stream, 16 Apr 1991, T. Flynn, D. Lorence,
S. Perlman, K. Wood, D. Palmer & B. Gustafson 4532 (NTBG). Maui: Haleakala, Kaupo Gap near
National Park boundary, adjacent to the trail, elev. ca. 1525 m, 3 Jun 2000, D. D. Palmer with H.
Oppenheimer, T. Ranker, J. Ott, & P. Welton 3162 (Hb. Palmer).

Polypodium pellucidum var. acuminatum differs from var. pellucidum by
having fully 1-pinnate fronds with 2 to several lobes not winged to the rach-
ises; long, narrow, pointed lobes; veins sometimes anastomosing at the bases
of some lower lobes; the lack of “false veins”; and the presence of numer-
ous, minute, 0.1 mm long, 3-celled, round-tipped hairs common over the
abaxial surfaces of the lobes, costae and veins. :

Although Hillebrand’s Polyodium pellucidum var. y (Fl. Hawaiian Isl. 558.
1888) is not a legitimate name, he was the first to recognize this variety.

This little-known variety is not well studied and is known only from
isolated areas, 500-580 m, in the Waimea, Kokee, and Hanalei districts

104 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

of Kauai, and the Kaupo Gap area of Haleakala on East Maui at 1525 m
elevation.

Plants with characteristics between var. pellucidum and var. acuminatum
are found. The minute hairs characteristic of this variety are found in occa-
sional scattered populations of the usually glabrous var. pellucidum. In these
populations the hairs may be scattered or numerous.

ACKNOWLEDGMENTS

Alan Smith, Barbara Parris, and Ken Wilson were very helpful in the preparation of this pa-
per. My frequent phone calls, faxes, and emails to them must have been a burden. Herb Wagner
encouraged my interest in Hawaiian ferns and his tutelage made the writing of this paper possi-
ble. William R. Anderson provided the Latin diagnoses.

American Fern Journal 92(2):105—111 (2002)

Novelties in Pteridaceae from South America

JEFFERSON PRADO
Secgao de Briologia e Pteridologia, Instituto de Botanica Caixa Postal 4005,
01061-970 Sao Paulo, SP, Brasil
ALAN R. SMITH
University Herbarium, 1001 Valley Life Sciences Bldg. University of California,
Berkeley, CA 94720 USA

AssTract.—Three new species (Adiantum squamulosum, Pteris boliviensis, and P. krameri) from
South America are described and illustrated. We also provide a new name for an endemic spe-
cies of Cheilanthes from Brazil and a new record of Pteris for Bolivia.

ResuMo.—Trés novas espécies (Adiantum squamulosum, Pteris boliviensis e P. krameri) da
América do Sul sao descritas e ilustradas. Séo também apresentados um nome novo para
uma espécie endémica de Cheilanthes do Brasil e um novo registro de ocorréncia para Pteris na
Bolivia.

Continuing studies with South American Pteridaceae, especially from
Bolivia and Brazil, indicate that the following taxa are either undescribed,
need a new name, or constitute a new country record. We hereby record
these additions and dedicate this paper to the memory of Warren (Herb)
Wagner Jr., who contributed to our understanding of Pteris, especially in
Hawaii and Florida.

Adiantum squamulosum Prado & A. R. Sm., sp. nov. (Fig. 1).—Type:
BOLIVIA: Depto. Beni: Pcia. Vaca Diez: 13 km E of Riberalta on road to Gua-
yaramerin, then 3 km N on side road, 10°58’S, 66°58’W, 230 m, 27 May 1982,
J. C. Solomon 7818 (holotype MO on 2 sheets; isotypes LPB not seen, UC).
Ex affinitate A. diogoano Glaz. ex Baker, stipitibus et rachibus dense rufos-

quamulosis, pinnulis abaxialiter et adaxialiter rufosquamulosis, squamis

valde ciliatis aut setosis, pinnulis 41—46-jugatis (vs. 21-24-jugatis), indusiis
squamulosis (vs. pilosis) differt.

Plants terrestrial. Rhizomes short-creeping, ca. 0.6 mm in diam., scaly, the
scales somewhat shiny, essentially concolorous, appressed throughout or
spreading distally, varying from medium to dark brown, lanceate, sparsely
denticulate at margins. Fronds monomorphic, 2-pinnate, 73-116 cm long,
the laminae 25-35 cm wide; stipes approximate, 1/2-2/3 the length of the
frond, black, adaxially sulcate, densely scaly, the scales appressed through-
out, concolorous, ferrugineous, 3-4 mm long, linear to narrowly lanceate
with a filiform apex, ciliate at margins and with several processes at the
base; rachises similar to stipes and their indument similar; pinnae oblong-
lanceate, slightly narrowed at their base, tapering at the apex, (8)15-30 x
1.8-3.2 cm, the lateral pinnae (3)4—9 pairs, ascending, alternate, the terminal

AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

106

es:

pene re ans

ide:

YU), fi i Gi hy
Luin sib wk
eA ro
— sg SST at il aS <a —
COR yi at ss «(OS
Sot ius ot
IK Wes i
Saint wee ;
Bil =i ii
IU

uae

Ss

ah

ee

Sy
i, ays
= sce <P> { {)
Zi
OO a elite,
SOMITE) sss Z, Os ;
ea LL Pe Ose sp. LM.

part of fertile frond;

and part of stipe; B. Distal

ules.

Adiantum squamulosum. A. Rhizome

C. Rachis scales; D. Abaxial view of fertile pinn

Pas. 1.

PRADO & SMITH: NOVELTIES IN PTERIDACEAE 107

pinna conform, 1—1.5 times longer than the subtending pinnae, 0.8—1 times
as long as medial pinnae; indument of costae like that of stipes and rachises;
pinnules (11)41-46 pairs per pinna, ca. 2 times longer than wide, charta-
ceous, not articulate, free-veined, without an evident midrib, the proximal
pinnules reduced, somewhat rounded or rhombic, the medial pinnules di-
midiate, trapeziform to oblong and with the acroscopic base truncate, the
sterile apices obtuse to acute, sterile margins denticulate, fertile apices angu-
lar, distal pinnules ca. 1/2 or less as long as the medial pinnules; both pin-
nule surfaces scaly, the scales sparse, ferrugineous, 1.0-1.5 mm long but
otherwise similar to those of the stipes, glands absent, veins slightly promi-
nulous, idioblasts present between the veins; sori oblong, up to 7(9) per pin-
nule; indusia scaly, scales with filiform apices and pectinate bases, entire to
erose, the cells well evident; spores yellowish, trilete, tetrahedral-globose,
with prolonged angles, the surface rugulate.

Adiantum squamulosum is distinguished by having densely scaly stipes
and rachises, large scales on both surfaces of the pinnules, a relatively large
number of pinnule pairs per pinna (up to 46), and scaly indusia. This spe-
cies occurs in partially disturbed, primary, non-inundated forest at low ele-
vations (ca. 230 m) along road margins. Adiantum diogoanum Glaz. ex Baker
is probably the most closely related species but differs in having stipes, ra-
chises, and pinnules glabrescent, fewer pinnule pairs per pinna (up to 24),
and indusia with reddish hairs, rather than scales. It grows in drier, inland
forests along rivers and slopes.

Adiantum squamulosum is known from a single locality in Bolivia, whereas
A. diogoanum has a wide range in Brazil, occurring in the states of Pernam-
buco, Alagoas, Minas Gerais, So Paulo, and Parana. The type was cited by
Smith et al. (1999, p.247) as questionably A. humile Kunze.

Pteris boliviensis Prado & A. R. Sm., sp. nov. (Fig. 2A—C).—Type: BOLIVIA:
Depto. Cochabamba: Prov. Chapare, 22 Feb 1929, 1700 m, J. Steinbach
9327 (holotype UC).

Ad P. lividam Mett. affinis, a qua imprimis frondibus pinnatis (vs. triparti-
tis) et 2 vel 3 areolae inter contiguas costulas (vs. 1 areola) differt.

Plants terrestrial. Rhizomes short-creeping, densely clothed at apex with
light brown ovate-lanceate scales 2-3 mm long, the scale margins entire and
glabrous. Fronds erect, to ca. 52 cm long; stipes ca. 1/2 of frond length, up
to 3 mm in diam., straw-colored, dark brown at base, grooved adaxially,
smooth, glabrous except for a few ovate-lanceate scales at base; laminae
chartaceous, 2-pinnate-pinnatifid at base, with 1-4 pinna pairs per frond, 1-
pinnate-pinnatifid above the base, ending in a broadly based, pinnatifid api-
cal pinna; rachises similar to stipes, alate distally, glabrous; proximal pinna
pair stalked (stalks ca. 0.5 cm long), opposite, deeply pinnatifid, furcate, the
basal basiscopic pinnules 12-16 x 4.5—7.0 cm, the remainder of the pinnae
20-22 x 5-7 cm; median pinnae sessile, subopposite, deeply pinnatifid, de-
current on the rachis, 13-14 5.5—-6.5 cm; distal pinnae sessile, alternate,

AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

PRADO & SMITH: NOVELTIES IN PTERIDACEAE 109

decurrent on the rachis, 8-9 X 2.5—3.0 cm; costae not sulcate on both sides,
glabrous, awns absent adaxially; pinna segments subopposite to alternate,
oblong, the margins entire at the middle to crenate-serrate at the apex, the
terminal segment of each pinna narrowly deltate, acute or acuminate, si-
nuses between segments roundish to biangulate; venation copiously areolate
but free near margins and segment apices, with 2 or 3 costal areoles between
adjacent costules, the costules slightly prominent especially on abaxial sur-
faces. Sori interrupted at sinuses and absent at apex of segments; indusia
greenish when young, margins entire; spores brown to tan, trilete, tetrahe-
dral, the surfaces rugulate with roundish tubercles and a smooth equatorial
flange.

PaRATYPE.—BOLIVIA. Cochabamba: Pcia. José Carrasco Torrico: 137 km antigua carretera Cocha-
bamba-Villa Tunari, 17°06’S, 65°35’W, 1600 m, 18 July 1996, M. Kessler et al. 7393 (paratype
UC; isoparatype LPB not seen).

The two or three areoles between adjacent costules distinguish this species
from Pteris livida Mett., which has only one areole between adjacent
costules. In addition, the fronds are ternate in P. livida and pinnate in P.
boliviensis.

Pteris boliviensis grows in wet forests and cleared forests at ca. 1600—
1700 m.

Pteris krameri Prado & A. R. Sm., sp. nov. (Fig. 2D, E).—Type: GUYANA:
Upper Essequibo Region: Rewa River, near camp 2 at foot of Spider Moun-
tain, forest on light brown sand, 03°08’N, 0°58’W, 16 Sept 1999, M. J.
Jansen-Jacobs et al. 5923 (holotype UC; isotype U on 2 sheets).

A P. altissima Poir. costis aristis adaxialiter carentibus differt.

Plants terrestrial. Rhizomes stout, woody, creeping, densely clothed at
apex with linear-attenuate, bicolorous scales, 3-5 mm long, these with a
blackish to reddish brown central portion and entire to more or less erose,
glabrous, pale margins. Fronds erect, ca. 1.3 m long; stipes ca. 2/3 of frond
length, up to 6 mm in diam., straw-colored, grooved adaxially, smooth,
glabrous except for a few bicolorous scales at base; laminae chartaceous, 2:
pinnate-pinnatifid at base, 3-7 pinna pairs per frond, the median portion
1-pinnate-pinnatifid, fronds ending in a broadly based pinnatifid apical pin-
na; rachises similar to stipes, narrowly alate distally, glabrous or with sparse,
whitish hairs ca. 1 mm long; proximal pinna pair stalked (stalks ca. 1.0 cm
long), opposite, deeply pinnatifid, furcate, the basiscopic pinnules (12-22 x
3.7-6.0 cm, the remainder of the pinnae 29-35 x7.5—10.0 cm; median pin-
nae sessile, subopposite, deeply pinnatifid, 22-30 x 7.5-8.5 cm; distal pinnae

—

Fic. 2. Pteris species.
veins; C. Detail of fertile segments. D-E
lamina showing veins, costa, and costules.

A-C. Pteris boliviensis. A. Habit; B. Detail of a sterile pinnae showing
_ Pteris krameri. D. Median pinnae; E. Adaxial view of

110 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

sessile, alternate, 13-14 x 3.9-4.5 cm; costae not sulcate adaxially, glabrous
or with sparse, minute, whitish hairs ca. 1 mm long, awns absent; segments
subopposite to alternate, long-lanceate, the margins entire to conspicuously
serrate at apex in fertile fronds (sterile fronds not seen), the terminal seg-
ment of each pinna elongate-deltate, acute to long-acuminate or sometimes
caudate, the sinuses between the segments acute to roundish; venation copi-
ously areolate but free near margins and segment apices, with 2 or 3 areoles
between adjacent costules, veins slightly prominent especially on abaxial
surfaces. Sori interrupted at sinuses and absent at segment apices; indusia
greenish or brownish, the margin entire; spores tan, trilete, tetrahedral, the
surfaces rugulate with roundish tubercles and smooth equatorial flange.

Pteris krameri is apparently most similar morphologically to P. altissima
Poir., and both species occur in similar habitats: near river and stream mar-
gins, moist ravines, in rocky soils or silt from big rocks, or on humus-rich
soils in forests. However, P. altissima differs in having an awn at the base of
each costule on the adaxial side of the lamina. Although it has been inad-
equately assessed, the character of presence or absence of awns along the pen-
ultimate axes seems of fundamental importance in assessing relationships in
Pteris, so the apparent similarity of P. krameri to P. altissima may be the
result of convergent evolution. Other New World species that lack awns
adaxially, and that have pinnatifid pinnae or pinnules, are P. angustata
(Fée) C. V. Morton, P. boliviensis Prado & A. R. Sm., P. brasiliensis Raddi, P.
congesta Prado, P. decurrens C. Presl, P. denticulata Sw., P. fraseri Mett. ex
Kuhn, P. lechleri Mett., P. leptophylla Sw., P. limae Brade, and P. pearcei
Baker, all species of South America and especially Brazil (Prado and Wind-
isch, 2000). None of these species seems closely related or similar morpho-
logically to P. krameri. Whether the presence or absence of adaxial awns on
the lamina is a character that circumscribes monophyletic groups is an intrigu-
ing but unanswered question.

Although Pteris krameri is known only from the type gathering, it may
have a wider range in northern South America. The species epithet honors
the late Karl Kramer, who contributed much to the study of fern systematics
and made Pteris one of his specialties; Dr. Kramer also focused especially on
the ferns of the Guianas, the source of this new species.

Brade 13494 (holotype RB, not seen, frag. HB!).

This rather uncommon species can be distinguished by its delicate, terete
stipes with uniseriate, jointed hairs, laminae proximally with 1-3 pairs of
free pinnae and distally pinnatifid. It is endemic to southeastern Brazil

PRADO & SMITH: NOVELTIES IN PTERIDACEAE 111

(Minas Gerais). The new name proposed below honors Alexander Curt
Brade, who described this species in Notholaena (Brade, 1940).

SPECIMENS STUDIED.—BRAZIL. Est. Minas Gerais: Estrada Diamantina—Curvelo, Williams & Anderson
8466 (UB); Santana do Riacho, Serra do Cip6, Parque Nacional da Serra do Cip6, Cachoeira da Farofa,
CFSC 10240, Prado et al. (SPF); Idem, Serra do Cipé, Sena s.n. (RB, HB); Idem, Schwacke 14520
(BHCB); Ouro Preto, Itacolomi, Schwacke 9906 (RB); Lima Duarte, Krieger s.n. (BHCB); Mun.
Gouveia: Rib. do Tigre, Hatschbach 27841 (MBM, UC).

Pteris bakeri C. Chr.

According to Tryon and Stolze (1989. p. 80), Pteris bakeri C. Chr. is en-
demic to Peru and occurs in forests at middle to high elevations, 2300-3000
m. This is the second collection outside of Peru; Arbeléez (1996) also re-
ported this species from the Colombian Andes.

Pteris bakeri can be distinguished by fronds ca. 1 m long, laminae decom-
pound (5-pinnate at the base), very small ultimate segments 1-1.5 mm wide,
free veins, and spiculate, scaly axes abaxially.

SPECIMENS STUDIED.—BOLIVIA. Depto. Cochabamba: Pcia. José Carrasco Torrico: 116 km antigua
carretera Cochabamba-Villa Tunari, 17°08’S, 65°38’W, 2350 m, Kessler et al. 7064 (LPB not seen,
UC); Idem, 123 km antigua carretera Cochabamba-Villa Tunari, 17°08’S, 65°37’W, 2100 m, Kess-
ler et al. 7117 (LPB not seen, UC).

ACKNOWLEDGMENTS

We thank curators of the Missouri Botanical Garden (MO) and Utrecht (U) Herbaria for loans
of types of Adiantum squamulosum and Pteris krameri, respectively. We also thank Sra. Emiko
Naruto for preparing the illustrations.

LITERATURE CITED

ARBELAEZ, A. A. L. 1996. La tribu Pterideae (Pteridaceae). Flora de Colombia 18:10—106.

BRaADE, A. C. he Filices novae Brasilianae VI. Anais Reuniaéo Sul-Amer. Bot. 2:5—-11.

PRADO, J. a _G. Wmnoiscu. 2000. The genus Pteris L. (Pteridaceae) in Brazil. Bol. Inst. Bot.
(Sao ee 13:103-199.

SmitH, A. R., M. Kessier, and J. Gonzates. 1999. New records of pteridophytes from Bolivia.
Amer. Fern J. o 244-266.

Tryon, R. M. and R. G. STOLzE. 1989. er ele of Peru. Part II. 13. Pteridaceae-15. Denn-
staedtiaceae. an Bot., n.s. 2

American Fern Journal 92(2):112—118 (2002)

Is Gametophyte Sexuality in the Laboratory a Good
Predictor of Sexuality in Nature?

ToM A. RANKER and HEATHER A. Houston
University Museum and Department of Environmental, Population,
and Organismic Biology, 218 UCB, University of Colorado, Boulder, CO 80309-0350 USA

than what is found among natural populations, although this may be true for any lab-based
study regardless of growth medium. Thus, we suggest caution in the use of agar as a growth
medium, and the use of laboratory conditions in general, for studies of fern gametophyte sexual
development.

The gametophytes of homosporous ferns possess the ability to be either
unisexual (antheridiate or archegoniate) or bisexual. In addition, they may
undergo developmental sequences involving changes in sexuality over time
due to genetic and/or environmental factors (Klekowski, 1969a; Greer &
McCarthy, 1999). Because gametophytes of most species are easy to grow in
culture, many studies have been conducted on cultured gametophytes to

antheridiogen production and response (e.g., Stevens & Werth, 1999), and
populational genetic load/isolate potential (Peck et al., 1990). Numerous
studies have employed data from lab-cultured gametophytes to make inferen-
ces about mating systems operating in nature (e.g., Soltis & Soltis, 1990, and
references therein; Chiou et al., 1998: Li & Haufler, 1999). Although many
studies have examined gametophytes that were cultured on mineral-enriched
agar, several have employed various natural substrates (Rubin & Paolillo, 1983;

tion in common: that the patterns and processes observed in the laboratory
are indicative of what is occurring in nature. At least one study, however, has
demonstrated significant differences in the sexual expression of gametophytes

RANKER & HOUSTON: GAMETOPHYTE SEXUALITY—GOOD PREDICTOR 113

collected from nature versus those cultured on mineral-enriched agar.
Schneller (1979) found that 79.9% of gametophytes of Athyrium filix-femina
(Woodsiaceae) collected from the wild were sexual, whereas those cultured
were only 60.1% sexual. The proportions of gametophytes that were asexual,
antheridiate, archegoniate, or bisexual differed between the two populations:
wild-collected were 21.1% asexual, 49.6% antheridiate, 20.6% archegoniate,
and 8.6% bisexual; lab-cultured were 39.9%, 30.5%, 27.3%, and 2.3%, respec-
tively. These two distributions differ statistically (2 x 4 contingency table,
lab vs. field by sexual category, x7 = 100.55, df = 3, p < 0.001; our analy-
ses). That study demonstrates that, at least for gametophyte populations of
A. filix-femina, the sexual expression of gametophytes cultured on mineral-
enriched agar was not a good indicator of the sexual expression of gameto-
phytes occurring in the wild population studied. Such comparisons are
crucial to assess the veracity of data collected from artificially cultured
gametophytes for inferring patterns and processes occurring in nature.

Ranker et al. (1996) studied the sexual expression of lab-cultured gameto-
phytes of two species of the endemic Hawaiian genus Sadleria, S.
cyatheoides Kaulf. and S. pallida Hook. & Arn., grown on mineral-enriched
agar. That substrate was chosen primarily for the sake of convenience and
the ease of gametophyte manipulation. Observations were made on a total of
5,749 gametophytes from 26 sibships of S. cyatheoides and 3,032 gameto-
phytes from 15 sibships of S. pallida (see Ranker et al., 1996, for explicit
methods). All gametophytes were sampled from multigametophyte cultures.
Although the two species differed statistically in the exact proportions of
gametophytes that were antheridiate, archegoniate, or bisexual, the overall
pattern among sexual gametophytes was the same in the two species: a small
proportion was antheridiate (grand means of 9.7% and 10.0% from
S. cyatheoides and S. pallida, respectively), a larger proportion was arche-
goniate (84.7% and 81.2%), and a small proportion was bisexual (5.6% and
8.8%). The predominance of unisexual gametophytes was consistent with
analyses of sporophyte-population surveys of allozymic variation from which
Ranker et al. (1996) inferred that sporophytes of both species primarily arose
via intergametophytic matings.

Another aspect of the study of Ranker et al. (1996) involved assessments
of the ability of Sadleria spp. gametophytes to produce and respond to an-
theridiogen (Dépp, 1950; Naf, 1979). Based on an excess of antheridiate ga-
metophytes in treatment vs. control cultures, Ranker et al. (1996) concluded
that an antheridiogen system was operating in these species. These results
are at least superficially incongruent with the data from the sexual expres-
sion studies described above; that is, one might expect to find a greater pro-
portion of antheridiate gametophytes from multi-gametophyte cultures, due
to the action of antheridiogen, than was observed in either species.

We conducted field studies on the sexual expression of natural popula-
tions of Sadleria spp. gametophytes. Our primary goal was to compare field-
collected data to the lab-based data of Ranker et al. (1996) to determine if
the laboratory data were an accurate reflection of what is occurring in nature.

114 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

In particular, we were interested in comparing gametophyte populations
cultured in the laboratory to those sampled in the field for: 1) the ratio of
asexual to sexual gametophytes; 2) the ratio of unisexual to bisexual
gametophytes, among sexual gametophytes and, 3) the ratio of antheridiate
to archegoniate gametophytes, among unisexual gametophytes.

MATERIALS AND METHODS

Five populations of Sadleria spp. sporophytes were located on the Hawaiian
Island of Kauai in November, 1999. Three populations were mixtures of
S. cyatheoides and S. pallida (populations CP-1, CP-2, and CP-3), one popu-
lation was pure S. squarrosa (population S-1), and one population was pure
S. unisora (population U-1). Populations CP-1, CP-2, CP-3, and U-1 were
from the Hanalei District and population S-1 was from the Waimea District;
exact localities are available from the first author. Samples of earthen
or bryophyte-covered substrate approximately 10 cm by 10 cm by 1-2 cm
(deep) were collected with a kitchen spatula from the vicinity of Sadleria
sporophytes. Five to 10 of such samples were collected at each site. Samples
were transported in plastic bags to the herbarium of the National Tropical
Botanical Garden. Each sample was then inspected under a binocular
dissecting microscope for the presence of Sadleria gametophytes, which
could be distinguished from those of other species of ferns by the presence
of distinctive glandular trichomes that are present on both gametophytes and
sporophytes. Sadleria gametophytes were gently removed from the substrate,
rinsed in water, and categorized under a compound microscope as belonging
to one of four classes of sexuality: asexual, antheridiate, archegoniate, or
bisexual. Only gametophytes with mature archegonia and antheridia were
scored as bisexual.

RESULTS

We found and recorded observations from a total of 200 gametophytes
across the five field populations surveyed, with individual sample sizes
ranging from 26 to 67 (Table 1). All gametophytes sampled possessed notch
meristems.

ASEXUAL vs. SEXUAL GAMETOPHYTES.—The percent ratio of asexual to sexual ga-
metophytes in individual field populations ranged from 4:96 to 41:59 and,
summed across populations, that ratio was 20:80 (Table 1). The results
summed across populations were significantly different from what was
observed in the laboratory, where nearly the opposite percent ratio (74:26,
asexual:sexual) was obtained across all sibships of both species (2 x 2
contingency table of asexuality/sexuality vs. lab/field, y° = 286.6, df = 1,
p < 0.001). Similarly, the results summed across just those field populations
likely to be mixtures of gametophytes that were conspecific with those cul-
tured (CP populations) also exhibited a percent asexual:sexual ratio (18:82)

RANKER & HOUSTON: GAMETOPHYTE SEXUALITY—GOOD PREDICTOR 115

TaBLE 1. Sexuality of sampled gametophytes.

Population N' Asexual? —- Sexual”_~— Antheridiate* Archegoniate* Bisexual”
Field results:
49 10 (20) 39 (80) 34 (87) 1 (3) 4 (10)
CP-2 26 1 (4) 25 (97) 9 (36) 13 (52) 3)(72)
CP-3 67 14 (21) 53 (79) 41 (77) 9 (17) 3 (6)
S-1 27 11 (41) 16 (59) 15 (94) 1 (6) 0
U-1 31 3 (10) 28 (90) 27 (96) 0 1 (4)
Total 200 39 (20) 161 (80) 126 (78) 24 (15) 14 (7)
Laboratory results’:
a oides 5749 4375(76) 1374 (24) 133 (10) 1164 (85) 77 (5)
S. pallida 3032 2089 (69) 943 (31) 94 (10) 766 (81) 83 (9)
Total 8781 6464(74) 2317 (26) 227 (10) 1930 (83) 160 (7)

* Sample size.

2 Number (percent of total rounded to nearest whole percent).

3 Number (percent of sexual gametophytes rounded to nearest whole percent).

4 Data from Ranker et al. (1996), summed across sibships.

that was significantly different from that of laboratory gametophytes (y* =
220.9, df = 1, p < 0.001). The percent ratio of asexual:sexual summed

across laboratory sibships of S. cyatheoides was 76:24 and that in S. pallida
was 69:31

UNISEXUAL vs. BISEXUAL GAMETOPHYTES.—Among the sexual gametophytes
sampled, all field populations exhibited a predominance of unisexual game-
tophytes (Table 1). Field populations did not differ statistically from labora-
tory populations in the ratio of unisexual to bisexual gametophytes, both
summed across all field populations (y” = 0.001, p > 0.05) and summed
across just the CP populations (%* = 0.462, p > 0.05).

ANTHERIDIATE VS. ARCHEGONIATE GAMETOPHYTES.—As with the ratios of asexual
to sexual gametophytes, field-collected and lab-cultured gametophytes ex-
hibited nearly opposite percent ratios of antheridiate to archegoniate gameto-
phytes. The overall percent ratio in the field was 84% antheridiate to 16%
archegoniate, among CP populations that ratio was 79:21, and that in the lab-
oratory was 11:89. These field and laboratory proportions were significantly
different from each other in total (x2 = 584.2, df = 1, p < 0.001) and when
comparing only CP populations to laboratory populations (y* = 397.6, df =
1, p < 0.001).

DISCUSSION

ted populations of gametophytes
d interpretation of data from lab-
of other taxa as well.

The results of our study of field-collec
have important implications for the use an
cultured gametophytes of Sadleria spp. and, possibly,
ASrxuAL vs. SEXUAL GAMETOPHYTES.—The striking difference between labora-
tory and field gametophytes in the relative proportions of asexual-to-sexual

116 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

gametophytes suggests that the laboratory conditions employed by Ranker
et al. (1996) were a poor mimic of the natural environment in terms of simply
allowing or forcing gametophytes to become sexual. Those laboratory condi-
tions were similar to those that have been employed by numerous other
investigators in studies of fern gametophytes. Thus, some factor or combina-
tion of factors in our laboratory cultures appears to have been inhibiting the
development of sexual organs. One potential impact on inferences resulting
from the laboratory data is that they would lead to an underestimate of the
sexual reproductive potential of a population or species relative to those
based on field observations, although Ranker et al. (1996) did not explicitly
make any such inferences.

UNISEXUAL vs. BISEXUAL GAMETOPHYTES.—In both laboratory and field popula-
tions of gametophytes there was a predominance of unisexual gametophytes
among those that were sexual. These results are similar to those from A
filix-femina (Schneller, 1979) and from Blechnum spicant (Blechnaceae;
Cousens, 1979, 1981). Among sexual field-collected gametophytes of B.
spicant, 21% were bisexual and 79% were unisexual (Cousens, 1981).
Cousens (1979) agar-cultured gametophytes of B. spicant as isolated gameto-
phytes and at moderate and high densities. Among sexual isolates, only 8%
were bisexual and among both groups of multi-gametophyte cultures, 20%
exhibited bisexuality. In terms of assessing the likelihood of unisexuality vs.
bisexuality and, thus, the relative likelihood of intergametophytic mating
vs. selfing, gametophytes cultured on mineral-enriched agar appear to behave
in a manner consistent with development of gametophytes in nature.

ANTHERIDIATE VS. ARCHEGONIATE GAMETOPHYTES.—Among gametophytes of
Sadleria spp. and A. filix-femina (Schneller, 1979), there were significant
differences in the ratio of antheridiate-to-archegoniate gametophytes in the
laboratory vs. in the field [using Schneller’s data from A. filix-femina, we
analyzed a 2 x 2 contingency table of antheridiate vs. archegoniate by lab
vs. field; y7 = 29.07, df = 1, p < 0.001]. In both cases, laboratory popula-
tions were generally mostly archegoniate and field populations were mostly
antheridiate. These results are similar to those of Rubin and Paolillo (1983)
where unisexual gametophytes of Onoclea sensibilis (Woodsiaceae) that were
agar-grown were disproportionately archegoniate whereas those that were
soil-grown were disproportionately antheridiate. Also, in field populations
of B. spicant unisexual gametophytes were predominantly antheridiate
(Cousens, 1981). The relative proportions of antheridiate vs. archegoniate
gametophytes in laboratory populations of B. spicant, however, varied with
density (Cousens, 1979), Among isolated gametophytes, all were archegoniate
and all but one (98%) were archegoniate when cultured in moderate density
(only one was antheridiate). By contrast, when cultured at high density, 96%
of unisexual gametophytes were antheridiate and only 4% were archegoniate,
similar to gametophytes collected from nature, Cousens attributed these
differential patterns to a greater likelihood of antheridiogen effects at higher
densities in the laboratory.

RANKER & HOUSTON: GAMETOPHYTE SEXUALITY—GOOD PREDICTOR 117

Laboratory studies of S. pallida, S. cyatheoides (Ranker et al., 1996),
A. filix-femina (Schneller, 1979), and B. spicant (Cousens, 1979) have demon-
strated that gametophytes of all of these species can produce and respond to
the male-inducing pheromone antheridiogen, as has been shown for many
other species of ferns (e.g., Naf, 1979; Schneller et al., 1990). Thus, as was
suggested for B. spicant, it is possible that antheridiogen has a greater effect
in natural populations of Sadleria spp. and A. filix-femina than on agar-
cultured populations. The possibility that an agar-based medium directly in-
hibits antheridia formation and/or promotes archegonia formation, however,
cannot be ruled out (see Rubin & Paolillo, 1983).

Similarly, the impact of potential differences in the age structure of natural
and cultured populations on sexual expression should also be considered. In
laboratory cultures, gametophyte populations typically are established
from single sowings of spores, resulting in nearly even-aged populations.
In nature, gametophyte populations would presumably be composed of
mixed-age individuals. Because gametophytes do not respond to antheridio-
gen after they reach the notch-meristem stage and because most gametophytes
in an even-aged stand would reach that stage essentially simultaneously, few
cultured gametophytes would become antheridiate due to antheridiogen
response. That would be particularly true at low to moderate densities. By con-
trast, in a mixed-aged natural stand, antheridiogen-producing gametophytes
may often already be present in a location when new spores arrive, thus
causing newly produced gametophytes to become antheridiate. Thus for ex-
perimental studies of the mating systems of species in which an antheridio-
gen system is controlling sexuality in nature, and depending on the goals of
a study, it is important to establish laboratory conditions that allow for ef-
fective antheridiogen-mediated interactions.

In conclusion, we found that for generally inferring mating systems
operating in nature (i.e., relative likelihood of intra- vs. intergametophytic
mating), agar-based laboratory studies of gametophytes would lead to the
same conclusions as would observations of field-collected gametophytes in
Sadleria spp. For detailed studies of gametophyte sexuality and develop-
ment, however, an agar-based medium produced significantly different
results than what we found among natural populations of gametophytes.
Whenever an even-aged population is created in the laboratory, however,
this may be true on any growth medium. Thus, as have others before us, we
suggest caution in the use of agar as a growth medium, and laboratory condi-
tions in general, for studies of fern gametophyte sexual development.

ACKNOWLEDGMENTS

We dedicate this paper to the memory of Herb Wagner for his constant inspiration, encourage-
ment, support, and friendship. This study was funded by a grant from the University of Colorado
Graduate School. Permission to collect specimens was granted by the State of Hawaii, Division
of Land and Natural Resources. We are grateful to Tim Flynn, Steve Perlman, Ken Wood, and
Diane Ragone of the National Tropical Botanical Garden for generously providing assistance in

the field and logistical support.

118 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

LITERATURE CITED

Cuiou, W.-L., D. R. FARRAR, and T. A. RANKER. 1998. peri be morphology and reproductive
biology in i ean Schott. Canad. J. Bot. 76:1967—-19

Cousens, M. I. 1979. Gametophyte ontogeny, sex expression, and sco load as measures of
merch Pie divergence in Blechnum spicant. Amer. J. Bot. 66:116—132.

. 1981. Blechnum spicant: habitat and vigour of optimal, argina and disjunct popula-
tions and field observations of gametophytes. Bot. Gaz. 142:25-

Dopp, W. 1950. Eine die Antheridienbildung bei Farnen fordernde ees in den Prothallien

9—159.

Dyer, A. ng pies The culture of fern gametophytes for experimental investigation. Pp. 253-305
in A. F, Dyer, ed. The Experimental Biology of Ferns. Academic Press, London.

a ee K. 1993. The influence of soil topography and spore-rain density on gender expression
in ee populations of the homosporous fern Aspidotis densa. Amer. Fern ie

mans . on McCartuy. 1999. Gametophytic plasticity among four species of ferns with con-
trasting ecological iy cotinine Int. J 9-886
Haurter, C. H. and D. E. Soxtis. 1984. Obligate outcrossing in a s easeeuoiorciia fern: field confir-
mation ofa a laboratory sear a Amer. J. Bot. 71:878-881.
KLEKOWSKI, JR., E. J. 1969a. oe biology of the pteridophyta. II. Theoretical considera-
tions. Bot. J. Linn. Soc. 62:347-359.
. 1969b. ecole sd of the pteridophyta. III. A study of the Blechnaceae. Bot. J.
Linn. vps
Li, J. W. an os pale 1999. Genetic variation, breeding systems, and cis of diversifi-
cation in Hawai sige ogg (Polypodiaceae). Syst. Bot. 24:339-35
Linpsay, S. . F. Dyer. 1996. Investigating the phenology of secre development: an
alt ae rien Pp. 633-650 in J. M. Camus, M. Gibby, and R. J. Johns, eds. Pteri-
ology in Perspective. Royal Botanic Gardens
NAr, U. 1979. Antheridiogens and antheridial adtalcnsaides: Pp. 435-470 in A. F. Dyer, ed. The
Experimental Biology of Ferns. Academic Press, London.
Pancua, E., S. Linpsay, and A. Dyer. 1994. Spore germination and gametophyte development in
three species of Asplenium. Ann. Bot. 73:587—-593.
Peck, J. H., C. J. Peck, and D. R. Farrar. 1990. ct ea se Ppa studies and the distribu-
tion of pteridophyte populations. Amer. Fern J. 8
Ranker, T. A., C. E. C. GeEMMiLL, P. G. Trapp, A. HAMBL LETON, ate ss Ha. 1996. Population genetics
and reproductive ee! of lava-flow colonising species of Hawaiian Sadleria (Blechna-
ceae). Pp. 581-598 in J. M. Camus, M. Gibby, and R. J. Johns, eds. Pteridology in Perspec-
tive. Royal Botanic “Na a Kew
Rusin, G. and D. J. PAoLi1o, Jr. 1983. Scud development of Onoclea sensibilis on agar and soil
media without the addition of antheridiogen. Amer. J. Bot. 70:811-815.
SCHNELLER, J. J. 1979. cn a investigations on the Lady Fern (Athyrium filix-femina). PI.
Syst. Evol. 132:255-2
. H. HAUvFLEr, vied “ A. Ranker. 1990. Antheridiogen and natural gametophyte popula-
ian Amer. Fern J. 80:143-152,
Soxtis, P. S. and D. E. =e 1990. Evolution of i
lies. Pl. Sp. Biol. 5:1-
TEVENS, R. D. and C. R. sina 1999. Interpopulational comparison of dose-mediated antheridi-
ogen response in Onoclea sensibilis. Amer. Fern J. 89:221-231.

nbreeding and outcrossing in ferns and fern-al-

American Fern Journal 92(2):119-130 (2002)

Additional Support for Two Subgenera of
Anemia (Schizaeaceae) from Data for the Chloroplast
Intergenic Spacer Region trnL-F and Morphology

J. E. SkoG
Biology Department, George Mason University, Fairfax, VA 22030
E. A. ZIMMER
Laboratory for Molecular Systematics, Smithsonian Institution, Washington, DC 20560
J. T. MICKEL
New York Botanical Garden, Bronx, NY 10458

Asstract.—An analysis of morphological data for 13 species with 33 characters and molecular
data for 14 species from the chloroplast DNA intergenic spacer region trnL-F indicates that spe-
cies of the genus Anemia fall into two well-supported subgenera, Anemiorrhiza and Anemia. In
addition, one species of the genus Mohria appears to belong within Anemia. Although further
study is required, these data support the relationships suggested by a previous study of fossil
and extant representatives of the genus.

The fern genus Anemia Sw. (Schizaeaceae) comprises about 120 species
distributed mainly within the tropics and subtropics. Most of the species are
found within the New World, with only about 12 in Africa and one in India.
No monograph of the whole genus has been produced, although subgenus
Coptophyllum (Mickel, 1962), subgenus Anemiorrhiza (Mickel, 1981), and
spores of subgenera Coptophyllum and Anemia (Hill, 1977, 1979) have been
studied. Since those works were produced, several new species have been
described (Mickel, 1982, 1984, 1985), and a study of some of the Cretaceous
fossils within the genus completed (Skog, 1992). In addition, the spores of
the family Schizaeaceae have been described for modern and fossil represen-
tatives (van Konijnenburg-van Cittert, 1991, 1992). Other fossil representa-
tives of the family have been described from Mesozoic and Cenozoic time
periods and are summarized in Skog (2001) and Collinson (2001). There is
clearly a need for a new revision of the genus and integration of all morpho-
logical data from fossil and living species with data from new molecular
studies. We began a collaborative study in 1999. This paper reports some re-
sults indicating that the chloroplast sequence data are consistent with the
fossil phylogeny reported earlier by Skog (1992).

The Schizaeaceae, considered to be a basal family of leptosporangiate
ferns, includes the genera Lygodium, Schizaea, Actinostachys, Mohria, and
Anemia. Fossil records of the family extend back to the Jurassic (Skog,
2001). The position of this family is not clear; it generally falls among sev-
eral clades, including the Hymenophyllaceae, Cyatheaceae, Schizaeaceae,
Matoniaceae, aquatic ferns, and more derived ferns (Raubeson & Stein, 1995;
Pryer et al., 1995; Pryer et al., 2001). However, there is strong support from

120 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

the morphological and molecular evidence to date that the genera within the
Schizaeaceae form a well-supported clade (Pryer et al., 1995; Pryer, 1999;
Wikstrém et al., 2000, pp. 149-150). A molecular study using the chloroplast
gene rbcL has been done for the Schizaeaceae and this study included sev-
eral species of Anemia (Wikstrém et al., 2000).

MATERIAL AND METHODS

Species from which DNA was isolated and sequenced, as well as the
voucher information are presented in Table 1. Morphological characters
(Table 2, 3) for the preliminary analysis were derived from the descriptions
and characters cited in the literature (Mickel, 1962, 1967, 1981, 1982; Skog,
1992; Tryon & Tryon, 1982).

Three subgenera (Anemiorrhiza, Coptophyllum, and Anemia) were repre-
sented in the samples. Both herbarium specimens and living material dried
in silica gel were sources for the DNA. More recent herbarium specimens
provided sequence data, but many older specimens yielded little, if any,
DNA. Other taxa included were one species of Mohria, which is often sug-
gested as congeneric with Anemia (Mickel, 1962), two species of Lygodium,
one species of Schizaea, and a species in the Hymenophyllaceae (Cardio-
manes reniforme) as the outgroup. The Hymenophyllaceae is also considered
fairly basal among the ferns and has been shown to be more basal than the
Schizaeaceae in analyses using morphology and rbcL chloroplast data (Pryer
et al., 1995, 2001).

Total DNA was extracted from 15-20 mg of leaf tissue dried in silica gel
(then frozen in liquid nitrogen) with the DNEasy Plant Mini kit from Qiagen,
following the manufacturer’s protocol. Double-stranded DNA amplifications
were performed in a 50 ul volume containing 5 pl MgCl, 4 ul dNTP, 5 ul
buffer, 2 ul primers, 1 ul BSA, 5 ul DNA, 0.5 pl TAQ, and 25.5 ul distilled
water to volume. Following an activation step of 10 min at 94°C for the en-

nutes before being held at 4°C. A few species were sequenced for the cpDNA
gene rbcL using the primers previously published for fern sequences (Hasebe
et al., 1995). The trnL (UAA)-trnF (GAA) intergenic spacer was amplified

Sequence data were generated for both strands of PEG (polyethylene gly-
col) purified PCR product using the ABI PRISM dye terminator cycle se-

quencer (Applied Biosystems, Inc.): 4 min at 96°C; then 25 cycles of 10 sec
at 95°C, 0.5 sec at 50°C, 4 min at 60°C; followed by 1 min at 96°C; and then

SKOG ET AL.: ADDITIONAL SUPPORT FOR TWO SUBGENERA 121

T . List of species (with subgenera of Anemia in paraentheses), voucher specimens, and
localities for the material used in the molecular analysis.

Species (Subgenus) Specimen Locality

Anemia adiantifolia (Anemiorrhiza) Bradley 30582, GMUF Bahamas, Andros Isl.
(L.) Sw.

A. cicutaria (Anemiorrhiza) Kunze Bradley 30691, GMUF Bahamas, Andros Isl.
ex. Spreng.

A. jaliscana (Anemia) Maxon Mickel 1689, NY Mexico, Jalisco

A. munchii (Anemia) Christ Mickel 6874, NY Mexico, Oaxaca

A. semihirsuta (Anemia) Mickel Mickel 1120, NY Mexico, Oaxaca

A. hirsuta (Anemia) (L.) Sw. Prado 1064, NY Brazil, Sao Paulo

A. phyllitidis (Anemia) (L.) Sw. Mickel 1121, NY Mexico, Oaxaca

A. underwoodiana (Anemia) Maxon Greuter & Rankin 24997, B Dominican Republic

A. villosa (Coptophyllum) Humb. & Salino 1771, NY Brazil, Sao Paulo
Bonpl. ex Willd.

Mohria cafforum (L.) Desv. NYBG 136/97 (living coll.) South Africa

Lygodium microphyllum (Cav.) R. Br. NYBG 1245/89B (living coll.) unknown

L. flexuosum (L.) Sw. NYBG 1281/76B (living coll.) unknown

Schizaea elegans (Vahl) Sw. H. Tuomisto 12744, TUR Peru, Loreto

Cardiomanes reniforme (Forst.) C. Presl A.R. Smith 2606, UC New Zealand, No. Isl.

Long Ranger gel. The resulting chromatograms were edited with Sequencher
version 3.1 (Gene Codes, Inc.) and the final consensus sequences were ex-
ported for alignment to Clustal X. All sequences were aligned manually with
the aid of Se-Al version 1.0a1 (Rambaut, 1996) multiple sequence editor fol-
lowing initial alignment in the program Clustal X. Alignment of the rbcL da-
ta was guided by the sequences of Pryer et al. (1995). The trnL-F sequences
were aligned by eye within the genus and between genera using the editing
program Se-Al. Sequences were submitted to GenBank (Accession numbers
AF448922-AF448935), and the alignment used in phylogenetic analyses will
be posted at www.l|ms.si.edu.

All phylogenetic reconstruction analyses were conducted using version
4.08b of PAUP* (Swofford, 2001). Phylogenetic reconstruction under maxi-
mum parsimony was conducted using both the heuristic search algorithm
with TBR branch-swapping, MULPARS and ACCTRAN options active, as
well as the Branch and Bound search option in PAUP*. For molecular data,
characters were assigned equal weights at all nucleotide positions; for mor-
phological characters, equal weighting was also used. Robustness of cladistic
linkages was evaluated with 1000 bootstrap replicates.

RESULTS

Results from the few species sequenced for rbcL were in agreement with
the study done by Wikstrém et al. (2000). After their abstract was published,
we dropped this gene from our study and accepted their discussion of a phy-
logeny based on that gene. Their topology has not yet been published. The
plastid DNA sequence trnL-F was chosen because it has been shown to have

AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

TABLE 2. Character names and states used in the morphological analysis.

1. rhizome habit

12.

spat cell shape

24.

apical plate cells

0: scarab horizontal sodiametric 0: thick-walled
1: creeping [ at ate Ls ein
2: sich te upright 13. stomata 25. spore shap
2. stele 0: attached 0: tetrahedral
0: dictyostele 1: floating 1: round
1: solenostele 14. laminar hair 2: bilateral, elongate
3. axillary pockets 0: uni- or ltl 3: tetraglobose (rounded)
0: absent 1: multic 26. spore ornamentation
1: present 2: ies 0: narrow sulci, wide muri
4. arrangement of leaves 15. laminar trichomes 1: wide sulci, narrow muri
0: polystichous 0: glandular 2: granulate
1: distichous 1: nonglandular, pointed 27.

a
n
=¢
ao)
®
Q
°
=
o
=

1: variable or med. brown
6. trichomes on rhizome

2: nonglandular, broad

. fertile pinna position
rizontal

rect
fl ne apnaeigg

i)
fos)

groove ornamentation
: smooth

1: granulate

. ridge ornamentation
th

0: smoo

0: scales 17. 1: granulate
1: hairs long (4-10 mm) chains erile 2: spinose
2: hairs short (1-3 mm) Neon ate 29. spore size
7. rhizome trichome color 18. fertile iy length : <50um
: 0: shorter than sterile blade 1: 50-60
1: dark brown or maroon 1: exceeding sterile blade 2: 60-80
2: yellowish brown 2: ca. equal to sterile blade 3: 385
8. stipe size 19. fertile pinna differentiation 30. spore ridge —
0: slender (0.6 mm) 0: undifferentiate 0: parallel,
1: medium (1.0 mm) : basal pair differentiated 1: coarsely anne
2: stout (2-3 mm) 2: fronds fully dimorphic 2: finely reticulate
frond division 3 22 pairs differentiated 31. fertile “eon on pinna

pinnate

20.

0: all fert
4: i all ih sterile

2: at tips Sa ste
. Stipe sha

N
fo
—_
ind
i")
|
=
w
N

10 21. fertile pinna insertion : terete
0: several ni (4-8) 0: sessile 1: somewhat flattened
1: many (10-15 +) 1: stalked 2: flattened
2: none 22. veins 33. indument on stipe
11. pinna rai shape 0: free 0: hairs scattered
: truncate 1: anastomosing 1: abundant orange hairs
1: as 23. sporangia shape 2: abundant white hair:
0: oval, oblong 3: abundant stiff black hairs
1: spherical 4: subglabrous

relatively high and even substitution rates among the plastid loci within an-
giosperms (Richardson et al., 2000) and it has been successfully used for
phylogenetic analyses among species within a genus of angiosperms
(Molvray et al., 1999). It had not been used for many analyses within the
ferns, but has been used for a study in a eusporangiate fern family (Hauk
et al., 1996), and we have had some success in a study of the Osmundaceae
(in progress). Within that region, only sequences between the “e” and “f”

SKOG ET AL.: ADDITIONAL SUPPORT FOR TWO SUBGENERA

ria, Actiniostachys, and Schizaea. Character coding is given in Table 2.

Tapte 3. Character state matrix for ten species of Anemia and one each of Moh

Species abbreviations for Anemia use the subgeneric designations given in Table 1.

18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33

6010 11 32 13 14 13 16 77

O20

a
1

12 3°4 5 6

Res (9G 0 ee 9 ae

Taxa
Ar. adiantifolia

Oo Oe Ha a A tO

°

oO

o

o

o

_ oO

o

©

An. phyllitidis

ow

i=)

o

o

i=)

So

o

o

ct

o

Oo

-

o

ise)

i)

o

o

1 80: 2 3

As. pennula

124 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

primers were amplified successfully. The region amplified between primers
“c” and “‘d” in all angiosperms studied in the Laboratory for Molecular Sys-
tematics (Smithsonian Institution) to date did not work, even with optimiza-
tion attempts using a Stratogene Robocycler. We suspect that either an
accelerated rate of evolution in that subregion of the cpDNA has resulted in
the primers’ failing to anneal or that the inversion and rearrangements of the
chloroplast genome, reported for ferns by Raubeson and Stein (1995), will
account for the amplification difficulties. The latter may be more probable,
as we successfully amplified the trnL-F “‘c-d’’ region for the Osmundaceae
(Bauder, Skog & Zimmer, unpubl.), but not for the higher fern genus Elapho-
glossum (Skog et al., 2001, p. 87).

Within the Schizaeaceae, the trnL “e-f’ spacers had a GC content from
36-40%; the outgroup Cardiomanes GC content was 32%. These numbers
were similar to those obtained in Zimmer’s previous work on the basal
angiosperm families Winteraceae and Canellaceae (Karol et al., 2000). For
the taxa studied, there was significant length variation among the sequences.

Within the Schizaeaceae, the trnL sequences ranged in length from 387
base pairs (bp) (Schizaea elegans) up to 525 bp (Anemia cicutaria); in the
outgroup Cardiomanes reniforme, the trnL region sequenced was 483 bp
long. Within previously described subgenera of Anemia, the range of spacer
lengths was much more limited. For subg. Anemia, Spacers were 494-506 bp
long; for subg. Coptophyllum the single trnL region sequenced from Anemia
villosa was 515 bp long; for the two species of subg. Anemiorrhiza, A. adianti-
folia at 524 bp was a single nucleotide shorter than A. cicutaria. Mohria
cafforum, sometimes proposed as being nested within Anemia, had a trnL
“e-f” region of 493 bp. The two Lygodium species were 481 and 489 bp long.

Given the degree of variation in spacer lengths among the Anemia species
and among taxa at the family level in the Schizaeaceae, we had to infer a
number of indel events when aligning the sequences. The overall alignment
required gaps to be inserted into the raw sequences to a final length of 803 po-
sitions. As would be expected from the raw sequence lengths, most of the gaps
that were inferred differentiated Anemia (including Mohria) from the other
two genera of Schizaeaceae and that family from the outgroup Cardiomanes.

alignment used in subsequent phylogenetic inference will benefit from a much
broader sampling of taxa across Anemia and within Schizaea and Lygodium.
However, the relatively lower degree of length variation within and between
Anemia and Lygodium suggests that the general phylogenetic patterns pre-
sented below will not be strongly affected by increased taxon sampling.

In the raw aligned data set for just the trnL‘‘e-f” region, 372 characters were
constant, 180 characters were parsimony-uninformative, and 251 characters

gions could be aligned with Canellaceae outgroup genera for a total length of
990 characters and where 913 characters were constant, 60 characters were

SKOG ET AL.: ADDITIONAL SUPPORT FOR TWO SUBGENERA 125

parsimony uninformative, and only 17 were parsimony-informative (Karol
et al., 2000). The high degree of length variation and large number of parsimony
informative characters seen with just the smaller PCR product from the trnL
region suggest that additional chloroplast spacer regions will be extremely
useful in delineating species relationships in the Schizaeaceae.

Both the Heuristic search and the Branch and Bound search options in
PAUP* produced one most parsimonius tree of length 704 steps, with a
consistency index of 0.875 and a retention index of 0.804. This tree is pre-
sented in Figure 1, with branch lengths given in Figure 1a and bootstrap
values in Figure 1b. The same topology is obtained when the gap characters
were ignored, Cardiomanes was excluded from the analysis, and Schizaea
and Lygodium were set as outgroups. In addition, we used our data for trnL
from the various genera (set as outgroups) within the Osmundaceae
(Osmunda, Todea, Plenasium, Leptopteris) in the analysis and obtained the
same topology for the tree. Maximum likelihood analyses of the data, using
either the default options in PAUP* (Tn/Tv ratio = 2) or a more complex
HKY85 +G+I model, where base frequencies and the proportion of invariant
sites were calculated from the data, also yielded a single tree with a topology
identical to that obtained in the parsimony analysis (data not shown).

The morphological analysis is incomplete. To date, 56 species of the genus
have been coded for 64 characters; however, the data were pruned to only
13 species and 33 characters for this paper (Tables 2, 3), which were ob-
tained from species descriptions in the literature (Mickel, 1962, 1967, 1981,
1982; Skog, 1992; Tryon & Tryon, 1982). These 13 species were chosen be-
cause we also had material for DNA extraction. We present a brief outline of
the results from the pruned morphological analysis, as it suggests several in-
teresting hypotheses to be tested (Figure 2). A single most parsimonious tree
was obtained from this reduced data set, shown in Figure 2a with branch
lengths and Figure 2b with bootstrap values. Eighteen characters were syna-
pomorphic for various clades. Subgenus Anemiorrhiza forms a distinct clade
within the genus (66% bootstrap). The traditional subgenera Anemia and
Coptophyllum cluster together with 52% support for a clade of these subge-
nera and Mohria. There is strong support for Mohria to be included within
the genus Anemia (98%). The one species of Mohria and the one species of
subg. Coptophyllum form a clade with 69% support. Obviously the expanded
morphological data set is necessary, as more species of subg. Coptophyllum
and Mohria are needed to determine if the relationship receives continued
support from both morphological and molecular data. At the moment, only
morphological data indicate this relationship. If these relationships are to be
resolved, additional molecular data is also needed for the rest of the species
of Mohria and additional species of Anemia.

DISCUSSION

In their published abstract, Wikstrém et al. (2000) noted that a maximum
parsimony analysis of 30 living species of Schizaeaceae indicated that Schi-

126 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

An. jaliscana

An. munchii

An. semihirsuta
An. hirsuta

Ac. villosa

An. underwoodiana
An. phyllitidis

Mo. cafforum

Ar. adiantifolia

Ar. cicutaria

Sc. elegans

. Ly. flexuosum

- Ly. microphyllum

Ca. reniforme

2a

>

Ar. adiantifolia
Ar. cicutaria
Ar. wrightii
Ac. villosa
Mo. cafforum
An. hirsuta

An. munchii

An. phyllitidis

An. underwoodiana

An. jaliscana

An. semihirsuta

Sc. elegans

As. pennula

An. jaliscana

An. munchii

An. semihirsuta
An. hirsuta

Ac. villosa

An. underwoodiané
An. phyllitidis

Mo. cafforum

Ar. adiantifolia

Ar. cicutaria

Sc. elegans
Ly. flexuosum

Ly. microphyllum

Ca. reniforme

2b
” Ar. adiantifolia
66 Ar. cicutaria
Ar. wrightii
An. hirsuta
98 An. munchii

51
An. phyllitidis

An. underwoodiana
52 =f Ac. villosa
Mo. cafforum

An. jaliscana

An. semihirsuta

eC Sc, elegans

cerns f\ pennula

SKOG ET AL.: ADDITIONAL SUPPORT FOR TWO SUBGENERA 127

zaea and Lygodium were monophyletic. Anemia was paraphyletic to Mohria,
as was subg. Coptophyllum to subg. Anemia. They noted that Anemiorrhiza
was a sister-group to a clade with all the other species of Anemia and Moh-
ria. They stated that within the family there was a long branch leading to
the species of Lygodium. The outgroup used in their analysis was not
mentioned. The actual tree topology from their study has not yet been pub-
lished and was not included in their abstract.

Our data are not completely consistent with their study. The trnL tree
agrees with Wikstrém et al. (2000) in the placement of taxa within Anemia
(Figure 1). However, Schizaea forms a strongly supported monophyletic
clade with Anemia (79% bootstrap), and only the two Lygodium species are

n a separate long branch relative to the outgroup Cardiomanes. Mohria
clearly falls within Anemia (100% support), and basal to the clade of subge-
nera Anemia and Coptophyllum (77% support). The single species of subg.
Coptophyllum falls within subg. Anemia, and this clade plus Mohria are sis-
ter to subg. Anemiorrhiza, which has 100% support on our tree. We believe
that subg. Coptophyllum and subg. Anemia should be combined into a single
subg. Anemia. There are no good morphological characters to support the
separation of these two subgenera, and the molecular data do not support
their separation either. Furthermore, we suspect that when additional spe-
cies and characters are included in the analysis, subg. Coptophyllum and
Anemia will form a strong single clade within the genus. There is, however,
strong support for subg. Anemiorrhiza as a monophyletic taxon within an
expanded genus Anemia (including Mohria).

As Mickel (1962) noted, Mohria is congeneric with Anemia. Traditionally
the separation of these two genera was based mainly on the possession of
scales by Mohria, but, as noted by Skog (1992), these scales have filiform tips
similar to the trichomes found in Anemia and the trichomes on the leaves of
the two genera are identical. Mohria bears sporangia on all pinnae of the fer-
tile frond, but the confinement of the fertile pinnae to the basal pair is not
always definitive. Some species of Anemia have dimorphic fronds (Mickel,
1984). According to van Konijnenburg-van Cittert (1991, 1992), the spores of
Mohria show a hollow area in the ridges of the exospore that is not found in

—

ious tree derived from DNA sequences trnL ‘‘e-f” region using
r maximum parsimony settings in PAUP*. Branch lengths are
Bootstrap consensus tree for the molecular
bove the lines leading to the lineages. Spe-

Fic. 1a. The single most parsimon
the branch-and-bound option unde
given above the lines leading to each lineage. Fig. 1b.
data; bootstrap support values above 50% are given a

cies are those listed in Table 1. The abbreviations are: Ac = Anemia subg. Coptophyllum, An =
Anemia subg. Anemia, Ar = Anemia subg. Anemiorrhiza, Mo = Mohria, Sc = Schizaea, Ly =
Lygodium, and Ca = Cardiomanes.

Fic. 2a. The most parsimonious tree based on the preliminary morphological data set, using a.
same options as above. Branch lengths are given above the lines leading to each lineage. Fig. 2 :
Bootstrap consensus tree for the morphology data set. Bootstrap values are given above the line.
The abbreviations are those used in Figure 1, plus As = Actinostachys.

128 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

e spores of Anemia subg. Coptophyllum and Anemia. She also noted
(1992) that hollow ridges occur in subg. Anemiorrhiza. Mickel (1962, Plate
IV) discussed and illustrated three types of ridge structure: a solid ridge with
no internal differentiation, a medulla within the ridge containing a spongy
network, or a simple medulla, which may be a small portion of the ridge as
in Mohria and in some species of Anemia. We see few substantial arguments
to maintain Mohria and Anemia as separate genera, although we await the
outcome of expanded analyses that include more than a single species of
Mohria.

Hill (1977, 1979) suggested that spore morphology might be an important
character within Anemia because the spore morphology was a conservative
character. Even in our preliminary analysis of morphological characters, we
see that the spore morphology of ridges, grooves, and bacculae will help to
delimit taxa, and that more species require examination for the critical char-
acteristics of the spores. Some characters that should be included are, for ex-
ample, elaborations at the angles of the spore outline, orientation of the
ridges, width of the striations, cross ridges between the ridges, and ornamen-
tation on the ridges and between the ridges.

The trnL data support the phylogeny previously inferred from the morpho-
logical study of fossil and modern species (Skog, 1992). No outgroup was
used in that analysis. At the time, no fossil representatives were known for
the Hymenophyllaceae earlier than the appearance of the Schizaeaceae; char-
acters from the fossils known were few, leading to a matrix with many miss-
ing characters; and relationships within the primitive fern families were not
stable or easy to discern. However, the analysis of the species within Ane-
mia and the fossils attributed to the genus or closely aligned to it indicates
that there is strong support for the subg. Anemiorrhiza, that subg. Coptophyl-
lum is paraphyletic to subg. Anemia, that these form a sister group to Ane-
miorrhiza, and that Mohria and two fossil species form another clade that is
not placed consistently in either of the other two clades and was better sup-
ported as a third group (Skog, 1992).

Interesting questions remain concerning Anemia. Subgenus Anemiorrhiza
is consistently diploid, whereas the other subgenera have various levels of

the greatest current morphological diversity for the genus. However, the fam-
ily Schizaeaceae first appears in the northern hemisphere in the Mesozoic,
as does Anemia (Skog, 2001), suggesting biogeographic questions might be
addressed when the phylogeny of the genus is better known. There are also

SKOG ET AL.: ADDITIONAL SUPPORT FOR TWO SUBGENERA 129

morphologically intermediate between the fossil species, which had com-
plete fertile fronds, interspersed fertile pinnules, or basal fertile pinnules.
Whether these extant species with nonextended fertile pinnae form a grou
or are dispersed throughout the genus will be of benefit in the understanding
of the development of the fertile fronds and the fertile pinnae. In our pre-
liminary morphological trees for the expanded data set, species with nonex-
tended fertile pinnae are currently scattered throughout the tree.

Based upon our data, we support the phylogeny of Anemia suggested pre-
yap # by fossil data (Skog, 1992) and rbcL data (Wikstrém et al., 2000, pp.
149-150). We support only two of the three subgenera of the current genus
yon p Oe Se and Anemia (including subg. Coptophyllum). We be-
lieve that Mohria should be placed in Anemia.

ACKNOWLEDGEMENTS

This paper is dedicated to the memory of Warren Herb Wagner (1920-2000). Professor Wagner
taught Skog the modern ferns and encouraged her study of fossil ferns. He was the mentor, advi-
sor, and friend of Mickel. We also thank Youngbae Suh for DNA sequencing instruction and
Molly Nepokroeff and Ken Karol for advice on sequence editing and alignment, as well as phy-
logenetic analyses. We are indebted to reviewers Alan Smith, Don Farrar, and Tom Ranker for
their helpful improvements to the manuscript.

LITERATURE CITED

CoLLINSON, M. 2001. Cainozoic ferns and their distribution. Brittonia 53:173-235.

Hasese, M., P. G. Wotr, K. M. Pryer, K. Uepa, M. Ito, R. Sano, G. J. Gastony, J. YOKOYAMA, J. R.
Manuart, N. Murakami, E. H. CRANE, C. H. HAUFLER, and W. D. Hauk. 1995. Fern phylogeny
based on rbcL nucleotide sequences. Amer. Fern J. 85:134—181.

Haux, W. D., C. R. Parks, and M. W. Cuase. 1996. A comparison between trnL-F intergenic
nai and rbcL DNA sequence data: an example from Ophioglossaceae. Amer. J. Bot.

7126.

Hi, 5 R. 1977. Spore morphology of Anemia subgenus pein Amer. Fern J. 67:11-17.

. 1979. Spore morphology of iat subgenus Anemia. Amer. Fern J. 69:71-79.

Karot, K., Y. Sux, G. Scuatz, and E. A. Zimmer. 2000. Molecular evidence for the position of
iit in the Winteracess: paca from nuclear ribosomal and chloroplast gene
spacers. Ann. Missouri Bot. Gar d. 87:414—-432.

Kovac Crrrert, J. H. A. vAN, 1991. Diversification of spores in fossil and extant Schi-
ae. Publ. Syst. Assoc., spec. vol. 44:103—-118 ie ;

"1992 The evolutionary development of schizaeaceous spores in situ. Courier Forschung-
sinst. Senckenberg. 147:109-117.

MicxzL, J. T. 1962. A monographic study of the fern genus Anemia, subgenus Coptophyllum.
Iowa State Coll. J. Sci. 36:349-482.

———. 1967. The phylogenetic position of Anemia colimensis. Amer. J. Bot. 54:432—-4

—. 1981. Revision of Anemia subgenus Anemiorrhiza a Brittonia 33: eicue

———. 1982. The genus pom Veins) in Mexico. Brittonia

———. 1984. New Tropical American Ferns. Amer. Fern L 74: 111-119.

—. 1985. Three new anemias from northern South America. Amer. Fern : 75:33-37. :

Mo vray, M., P. J. Kores, and M. W. Cuase. 1999. Phylogenetic relationships within Korthalsella
(Viscaceae) based on nuclear ITS and plastid trnL-F sequence data. Amer. J. Bot 86:249-
260

130 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

Pryer, K. P. 1999. Phylogeny of marsileaceous ferns and relationships of the fossil Hydropteris

pinnata reconsidered. Int. J. Pl. Sci. 160:931-954
SMITH, and J. E. Skoc. 1995. Phyiopunetia apse ge et ma extant ferns based on
hie are from morphology and rbcL sequences. Am ee —282.

. SCHNEIDER, A. R. SMITH, R. CRANFILL, P. G. Wore, 18 ae and S. R. Sires. 2001.
Sete and een ts are a monophyletic group and the oaest living hens to seed
plants. Nature — 618-621

— A. 1996. Se-Al Sécuimics Alignment Editor Version 1. 0 alpha 1. Department of

Zoology, Papin of ace South Parks Road, Oxford OX1 4JD, England.

Rauseson, L, and D. B. STEIN. 1995. Insights into fern evolution from mapping chloroplast
genomes. Tle. Fern J. a 193- 204.

RICHARDSON, J. E., M. F. Fay, Q. C. B. Cronk, D. Bownam, and M. W. Cuase. 2000. A a eon
analysis Rhamnacoae using rbcL and trnL-F plastid DNA sequences. Amer. J. Bo
87:1309-13

Skoc, J. E. 1992 ann Lower Cretaceous ferns in the genus Anemia (Schizaeaceae), Potomac
Group of Virginia, and relationships within the genus. Rev. Palaeobot. Palynol. 70:279-

2

—.. 2001. Pag cad of Mesozoic leptosporangiate ferns related to extant ferns. Brittonia
53:236—

24) Ens em R. Moran, and E. A. Zimmer. 2001. Phylogeny of the fern genus Elaphoglos-
sum based on two chloroplast Program/Botany 2001, Botanical Society of America.

ruman State pit Sapa:

Sworrorp, D. 2001. PAU Lh analysis using parsimony, version 4. 08b. Laboratory of
Molecular ce Smithsonian Institution, Washington, D.C., and Sinauer, Sunder-
land,

TABERLET, P., L GrELLy, G. Panton, and OUVET. 1991. Universal primers for amplification of
three non-coding regions of cloopiast DNA. PI. Molec. Biol. 17:1105-1109.

Tryon, R. M. and A. F. Tryon. 1982. Ferns and allied plants with special reference to tropical

merica. Springer-Verag ey Yo

WikstROM, N., P. KENRICK, and J. C. VocEL. 2000. Schizaeaceae: a phylogenetic approach. Abstr
6th Conf. Int’l. Ong Palaeobot., july 31—-August 3, 2000, Hebei, China. see taeear ae
Committees of Palaeobotanical Society and Botanical Society, China.

American Fern Journal 92(2):131-149 (2002)

Intrafamilial Relationships of the Thelypteroid Ferns
(Thelypteridaceae)

ALAN R. SmiTH and RAYMOND B. CRANFILL
University Herbarium, 1001 Valley Life Sciences Building #2465, University of California,
Berkeley, California 94720-2465

ABSTRACT.—Data from three chloroplast genes (rps¢4 + trnS spacer, + trnL spacer; 1350 base
pairs) for 27 of the recognized segregates show the Thelypteridaceae to be monophyletic and sis-
ter to an unresolved alliance of blechnoid, athyrioid, onocleoid, and woodsioid ferns. The family
comprises two primary lineages, one phegopteroid, the other thelypteroid (including cyclosor-
oid). The phegopteroid lineage (Macrothelypteris, Pseudophegopteris, and Phegopteris) includes
those elements that are the most dissected, lack adaxial grooves on the frond axes, and are gen-
erally morphologically the most distinct elements within the family. Within the thelypteroid-
cyclosoroid lineage, three predominantly north-temperate subgroups, including Thelypteris s.s.,
form a free-veined clade that is in turn sister to the rest of the family. All segregates possessing
x=36 (Cyclosorus sensu Smith, with predominantly anastomosing veins) form a strongly sup-
ported clade. Those groups with dysploid base chromosome numbers (x= 27, 29 St, 32,33;
34, 35) form a series of smaller clades basal to Cyclosorus s.]. Although our sampling is not
yet sufficient to favor one classification over another, recognition of an intermediate number of
genera may be the most reasonable taxonomic course.

ww
2

Since its taxonomic separation from the dryopteroid ferns as a distinct
group, about 60 years ago, Thelypteridaceae has been treated as a natural
group comprising nearly 1000 mostly tropical species. Although generally
recognized as a natural monophyletic group, there is a wide divergence of
views about generic circumscription. Morton (1963) placed all species in a
single genus Thelypteris. Ching (1963) outlined a classification that accepted
18 Asian genera, including Hypodematium, which is generally excluded
from the family by other workers, even by Ching (1978). Ching (1978) later
added two newly described genera to Thelypteridaceae: Craspedosorus
(which we regard as a synonym of Leptogramma, often included in Stegno-
gramma sensu Iwatsuki, 1963); and Trichoneuron (regarded by us as belonging
to Lastreopsis, a dryopteroid, definitely not thelypteroid). Both of these genera
are monotypic and poorly known. Still later, Shing (1999), subsumed Amphi-
neuron under Cyclosorus s.J. and removed Trichoneuron from the family.
Iwatsuki (1964) recognized three genera in the family, Stegnogramma and
Meniscium (each with four sections), and Thelypteris, the last comprising 14
subgenera and several additional sections; Iwatsuki (1964:23) regarded two of
his subgenera of Thelypteris (Haplodictyum and Cyrtomiopsis) as probably
‘generically distinct.’ Holttum (1971, 1982) characterized 25 genera in the
Old World but did not explicitly address New World groups. Pichi Sermolli
(1977:335-337), largely following Holttum, accepted 32 genera. In the most re-
cent classification, Smith (1990) adopted an intermediate view, recognizing

132 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

five genera. Most of the paleotropical segregate genera (Holttum, 1969, 1972,
1973a, 1974, 1975, 1976a, 1976b, 1977, 1981; Holttum & Grimes, 1979;
Iwatsuki, 1963), and several of the neotropical ones (Maxon & Morton, 1938;
Smith, 1971, 1980), have been recently revised or monographed, making this
one of the best known fern families morphologically, cytologically, and
distributionally. Little, however, is known about relationships among these
segregates.

The goals of this work are to provide a phylogenetic hypothesis for the
Thelypteridaceae based on molecular evidence from the 25-30 groups com-
monly recognized within the family. A second goal is to confirm or refute
the hypothesis of monophyly for the family. We make no attempt in this pa-
per to incorporate a rigorous morphological data set to contrast with the mo-
lecular data set, although we find it useful, and we hope informative, to
comment on certain characters that are often used to distinguish genera or
groups of genera, in light of the molecular results. In this paper, we choose
to concentrate on the higher-level relationships in the family, and this initial
approach necessarily involves only one or, in a few cases, two species per
group (genus, subgenus, section). A study directed to the generic subdivision
of the family would need to incorporate a minimum of two and ideally at
least three species per group, so as to address the monophyly of the individ-
ual genera, subgenera, or sections (depending on classification employed).

MATERIALS AND METHODS

TAXON SAMPLING.—As a basis for sampling, we used the classification of Holt-
tum (1971), as modified by Holttum (1982) for Old World groups. For New
World groups, we sampled from groups recognized by Morton (1963). A total
of 30 ingroup taxa were sampled (Table 1). These represent 19 of the 22
Malesian genera recognized by Holttum (1982; only Ampelopteris, Amphi-
neuron, and Cyclogramma are missing from our analysis), 25 of the 32
genera of Thelypteridaceae recognized by Pichi Sermolli (1977: only
Ampelopteris, Amphineuron, Cyclogramma, Glaphyropteris, Haplodictyum,
Menisorus, and Stegnogramma s.s. are missing), and 13 of the 20 Chinese
genera recognized by Ching (1978; only Ampelopteris, Amphineuron, Cras-
pedosorus (regarded by us as a synonym of Leptogramma), Cyclogramma,
Mesopteris (regarded by us as a synonym of Sphaerostephanos), Stegno-
gramma s.s., and Trichoneuron are missing). All neotropical groups, with
the exception of Glaphyropteris s.s. (which we regard as a subgroup of
Steiropteris; Smith, 1980), were also sampled. Recent phylogenetic analyses
of the Polypodiales (higher leptosporangiate ferns) by Hasebe et al. (1995)
and by Cranfill (unpubl. data) indicate that the thelypteroid ferns belong to
this order, and that the most closely related groups (families in some classifi-
cations) are an unresolved alliance of blechnoid, athyrioid, deparioid, ono-
cleoid, and woodsioid ferns. Sixteen representatives from these families
were used as outgroups.

SMITH & CRANFILL: INTRAFAMILIAL RELATIONSHIPS 133

DNA ExtRACTION, AMPLIFICATION, AND SEQUENCING.—We utilized silica gel-dried
leaf material and, in a few cases, leaf material removed from herbarium
specimens as sources for genomic DNA, which was extracted using DNEasy
Plant Mini DNA extraction kits from Qiagen Corporation. Amplification was
performed using Amplitaq Gold DNA taq polymerase produced by Perkin
Elmer Corporation. Ongoing work by Cranfill previously demonstrated the
phylogenetic utility of the chloroplast gene rps4 (Cranfill, 2000a, 2000b) and
the trnS and trnL-F intergenic spacer regions (Cranfill, unpubl.), and so data
from these three markers were collected for phylogenetic analysis. We also
considered the inclusion of rbcL, but chose to exclude it from our study for
reasons of time and economics. A preliminary analysis showed that rbcL
data provided more or less the same phylogenetic information at roughly the
same taxonomic levels as rps4.

The rps4 region was amplified using the primers provided by Nadot et al.
(1995), yielding an amplicon of approximately 650 bp. The intergenic spacer
region between rps4 and trnS was amplified using the reverse compliment of
Nadot et al. (1995), reverse rps4 primer R1 and the novel reverse primer trnS
R (5’-TAC-CGA-GGG-TTC-GAA-TC-3’), yielding an amplicon of approxi-
mately 400 bp. The intergenic spacer at the 3” end of trnL-F was amplified
using primers e and f from Taberlet et al. (1991), yielding an amplicon of ap-
proximately 350 bp. Amplification was accomplished using standard thermo-
cycling protocols, with annealing temperatures of 48°C-54°C for rps4 and
45°C-48°C for the two intergenic spacer regions, using an MJ Research PTC-
200 Peltier thermal cycler. Direct sequencing of each PCR product was con-
ducted in both directions for each marker using the BigDye cycle sequencing
kits produced by Applied Biosystems Inc. (ABI) using an ABI PRISM 377
DNA automated sequencer in the Molecular Phylogenetics Laboratory of the
University of California, Berkeley.

DNA sequences were manually aligned. Alignment of the coding rps4 se-
quences was unambiguous. The two intergenic spacer regions were aligned
with more difficulty, and all ambiguously aligned regions were removed
from the analyses. Gaps were treated as missing, while unambiguous and
phylogenetically informative indels were treated as binary characters and
coded at the end of the data matrix.

Maximum parsimony analyses were performed with PAUP* (Swofford,
1999). Combinability of data from the three markers was investigated using
the partition homogeneity test (Farris et al., 1995), as implemented in PAUP*,
the results of which supported data combination. Heuristic searches were

sultant trees were rooted with appropriate outgroups identified previously in
the rbcL phylogeny by Hasebe et al. (1995), and confirmed by Cranfill from
an unpublished five-gene molecular phylogeny of the Polypodiales. In addi-
tion to computing a strict concensus of trees obtained from our searches, we

TABLE 1. Sources of material of ingroup and outgroup species providing rps4, trnL spacer, and trnS a. sequences for this study. All seq data
reported in this paper are newly generated. Parenthetical RBC numbers are DNA extraction num

Species Locality Voucher/Source Herb. GenBank rps4-trnS' GenBank trnL spacer
Amauropelta oe Costa Rica: San José: Las U. C. Bot. Gard. 57.0002 UC AF425162 AF425125
(Willd.) er ubes
Chingia Jonglestinn Fiji: Viti Levu: Tomaniivi Game 99/270 (RBC 818) UC AF425163 AF425126
(Brack.) Holttum (Mt. Victoria)
pei arida (D. Don) Malaysia Cranfill BF-22 (RBC 718) UC AF425164 ~
Holttu
Christel hispidula cult., unknown source N. Y. Bot. Gard. 477/77A UC AF425165 AF425127
ne.) C. F. Reed = Cranfill s.n. (RBC
577)
Christella (Pelazoneuron) _U.S.A.: Florida: Grey- U. C. Bot. Gard. 78.0268 UC AF425166 AF425128
augescens (Link) Pic nolds Park, S end West = Cranfill s.n. (RBC
e, Miami 785)
Coryphopteris seemannii Fiji: Viti Lev Tomaniivi Game 95/147B (RBC 817) UC AF427096/ AF425129
Holttum (Mt. Victori AF427097
Cyclosorus Agree cult. by B. Parris, New Cranfill s.n. (RBC 707) UC AF425167 AF425130
(Willd. Zealand
Picton ae Taiwan Cranfill TW-004 (RBC UC AF425168 AF425131
6
clphyroptridops Taiwan Cranfill TW-197 (RBC UC AF425169 AF425132
erubescens (Hook.) 640)
Ching
Goniopteris Naa Mexico: Oaxaca: near Mickel 5799 = U. C. Bot NY AF425170 AF425133
(Bory) Chin xtepec Gard. 78.0266 (RBC
oo tottoides Taiwan Cranfill TW-79 (RBC 638) UC AF425171 AF425134
Bers edie New Zealand Cranfill s.n. (RBC 673) UC AF425172 -
torresiana re
hing
Meniscium sp. Panama: Canal Zone: A. Smith 2633 (from P. UC AF425173 AF425135

across street from Sum-
mit Zoo

Hammond) (RBC 786)

(2002) Z MHEIWON 26 ANNTIOA “TYNYNOl NYAA NVOMANV

TABLE 1. Continued.
Species Locality Voucher/Source Herb. GenBank rps4-trnS' GenBank trnL spacer
Mesophlebion Malaysia Cranfill BF-59 (RBC 703) UC AF425174 AF425136
—cesiaai (Blume)
Mtatholyptrs dayi Malaysia Cranfill CH-35 (RBC 745) UC AF425175 AF425137
(Bedd.) Holttu
Nannothelypteris Philippines: Mt. Makil- Price 670 (RBC 826) UC AF425176 -
aoristisora (Harr.) ing, Laguna
Holttum
Oreopteris limbosperma Great Britain: South Crabbe et al. 11830 (RBC UC AF425177 =
(All.) Holub Somerset, 4 km SW of
Nether Stowey
Parathelypteris U.S.A.: California: Plu- U. C. Bot. Gard. 60.0707 JEPS AF425178 AF425138
nevadensis (Baker mas Co., ca. 7 mi SE of (RBC 464)
Holttum eddia
Phegopteris connectilis cult., Mickel garden, New — Cranfill s.n. (RBC 575) UC AF425179 AF425139
(Michx t York
Phegopt cult., Unknown source N. Y. Bot. Gard. 685/76 UC AF425180 -
decursivepinat (H. C. = Cranfill s.n. (RBC
Hall) F 576)
hie cambio archboldiae Fiji: Viti Levu: Serua, A. C. Smith 9348 (RBC UG AF425181 -
(Copel.) Holttum between Waininggere 829)
and Waisese Creeks
Pneumatopteris ecallosa Malaysia Cranfill CH-25 (RBC 639) UE AF425182 AF425140
(Holttum) Holttum
Pronephrium simplex China: Hong Kong: U2, = Gard. 79.0293 UC AF425183 AF425141
(Hook.) Holttum Victoria Peak, (RBC 580)
Victoria Isl.
Pseudocyclosorus iwan Cranfill TW-29 (RBC 713) UC AF425184 AF425142
esquirolii (H. Christ)
shing

SdIHSNOILV' Tae TVITINVAVALLNI “THANVYO 8 HLIDNS

SEL

TABLE 1. Continued.
Species Locality Voucher/Source Herb GenBank rps4-trnS' GenBank trnL spacer
Pseudophegopteris aurita —_cult., Parris garden, New Cranfill s.n. (RBC 238) UC AF425185 -
(Hook.) Ching Zealand, orig. from Ja-
pan
Sphaerostephanos Malaysia: Bangi Fern Cranfill BF-24 (RBC 641) UC AF425186 AF425143
penniger (Hook.) Garden
Holttum
Sphaerostephanos Taiwan: Taibei Botanical Cranfill TW-231 (RBC UC AF425187 -
taiwanensis (C. Chr.) Garden 709)
Holttum ex Kuo
Steiropteris leprieurii French Guiana: Granville 13264 (RBC UC AF425188 ~
(Hook.) Pic. Serm. Montagnes de la Tri- 828)
nité, Bassin de la Mana
Thelypteris palustris N. Y. Bot. Gard. (wild) Cranfill s.n. (RBC 574) UC AF425189 AF425144
Schott
Trigonospora ciliata cult., unknown source N. Y. Bot. Gard. acc. = UC AF425190 AF425145
(Benth.) Holttum Cranfill s.n. (RBC 582)
Outgroups
Acystopteris japonica Taiwan Cranfill s.n. (RBC 590) UC AF425150 AF425121
(Luerss.) Nakai
Asplenium cristatum Costa Rica: Puntarenas: U. C. Bot. Gard. 90.2234 UC AF425146 -
Las Alturas, 31 km = Cranfill s.n. (RBC
from San Vito 042)
Asplenium nidus L. cult. origin Cranfill s.n. (RBC 309) UC - AF425118
Athyrium filix-femina cult., Mickel garden, New — Cranfill s.n. (RBC 356) UC AF425152
(L.) Roth ex York
Mert. s./.
Cystopteris protrusa cult., Mickel garden, New — Cranfill s.n. (RBC 369) UC AF425148 AF425120
(Weath.) Blasdell York
Deparia lancea (Thunb. cult., source unknown, U. C, Bot. Gard. 71.0038 UC AF425153 AF425123
ex Murray) Fraser-Jenk. native to SE Asia = Cranfill s.n. (RBC
1
Didymochlaena cult., unknown source Cranfill s.n. (RBC 432) UC AF425161 _

truncatula (Sw.) J. Sm.

9ET

(Z00z) Z WAAIWAN 26 AWN'TOA “TVNUNOL NAA NVORIANV

TABLE 1. Continued.

Species Locality Voucher/Source Herb GenBank rps4-trnS! GenBank trnL spacer
Gymnocarpium cult., Parris garden, New Cranfill s.n. (RBC 240) UC AF425149 -
oyamense (Baker) Zealand, origin from
in apan
Homalosorus cult., Mickel garden, New Cranfill s.n. (RBC 597) UC AF425154 AF425124
pycnocarpos (Spreng.) York
Pic. Serm.
Hypodematium crenatum —_ Japan, sent by M. Kato Hyashi s.n. (RBC 768) TI AF425151 AF425122
(Forssk.) Kuhn
Lorinseria areolata (L.) C. U.S.A.: S. Carolina: U. C. Bot. Gard. 82.2087 - AF425155 -
Presl Orange Co., along = Cranfill s.n. (RBC
Rte. 176 170)
Matteuccia struthiopteris _—_cult., original source North Carolina Bot. Gard UC AF425158 -
(L.) Tod. unknown = Cranfill s.n. (RBC
460)
Onoclea sensibilis L. U.S.A.: Massachusetts Weed s.n Bot. UC AF425159 -
Gard. 80.0321 (RBC
005
Onocleopsis hintonii F. Mexico: Oaxaca, sent by Mickel s.n. IND AF425160 -
Ballard Gaston
Sadleria cyatheoides U. S.A.: Hawaii: Maui: Ornduff 8506 = U. C. UC AF425156 -
Kaulf. between Olinda and Bot. Gard. 78.0380
Wai Kamoi stream (RBC 003)
Stenochlaena milnei Philippines (“New U.C. Bot. Gard. 55.0076 UC AF425157 -
uinea”’ on label): Sa- (RBC 0.37)
mar, from spores
Woodsia polystichoides cult., Mickel garden, New — Cranfill s.n. (RBC 551) UC AF425147 AF425119
York

' rps4—trnS sequence data repre

sent one contiguous sequence with rps4 at the 5’ end and the trnS spacer at the 3’ end. Sequences for Cory-
phopteris are for rps4 alone (AF427096), trnS spacer alone (AF427097), and trnL alone (AF425129).

SdIHSNOILV Tau TVITINVAVELLNI “THANVYD 8 H.LINS

ZEl

138 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

also explored clade stability using bootstrap resampling (Felsenstein, 1985),
as implemented in PAUP*.

RESULTS

support, including a clade comprising Dictyocline and Leptogramma (100%)
and a clade comprising Goniopteris and Christella augescens (78%) (Fig. 1).

Hasebe et al. (1995), who sampled four representatives in their global analy-
sis: Amphineuron opulentum (Kaulf.) Holttum, Christella acuminata (Houtt.)
Holttum, Parathelypteris beddomei (Baker) Ching, and Thelypteris palustris

SMITH & CRANFILL: INTRAFAMILIAL RELATIONSHIPS 139

PSEUDOCYCL ei ORUS

75 Sirintenbedtratnd P
SPHAEROSTEPHANOS
56 cian
CHRISTELLA
67 pepsin elaine be
PLEISIONEURON
PRONEPHRIUM

C
78 CHRISTELLA AU
93 GONIOPTERIS
MENISCIUM

89 AMAUROPELTA
98 PARATHELYPTERIS
69| 76 CORYPHOPTERIS
ETATHELYPTERIS

68
HOMALOSORUS
ce TIUM
ae
— ASPLENIUM
10

nious trees drawn as a phylogram, combined analysis utilizing
acer, and frnL spacer. Branch lengths are proportional to the
h. a me for clades, if specter than 50%, is indi-

teps; CI = 5; RC = 0.63. See Table 1 for

full names of species and additional voucher data.

Fic. 1. One of 26 most parsimo
the chloroplast genes rps4, trnS sp

140 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

Schott (listed here under their segregate generic names). The first two of
these species belong to cyclosoroid genera, the last two species to amauro-
peltoid genera or Thelypteris s.s. (Fig. 2). No representatives of the phegopte-
roid clade have been included in previous analyses, and no other global or
family level molecular surveys have provided any additional evidence on re-
lationships either outside or within the family. Our study clearly shows that
Thelypteridaceae, in the sense used by nearly all modern authors, is mono-
phyletic, with high (98%) bootstrap support (Fig. 1).

Holttum (1971) hypothesized a relationship between Thelypteridaceae and
Cyatheaceae, a kinship that is now effectively refuted on the basis of ample
morphological (Smith, 1990) and molecular evidence (Hasebe et al., 1995;
Pryer et al., 1995; Wolf et al., 1999). Pichi Sermolli (1977) postulated a close
relationship of Thelypteridaceae to Aspleniaceae, and indeed this relation-
ship appears relatively close, although more remote than with several other
large clades. Our analysis, and work by Cranfill (unpubl.), clearly show that
the thelypteroid ferns are most closely related to a large terrestrial clade
comprising the athyrioid, woodsioid, blechnoid, and onocleoid ferns (Fig. 1).
The exact interrelationships of these immediate outgroups are still not well
supported in our analysis or any other published molecular analysis.

INTRAFAMILIAL RELATIONSHIPS.—Using morphological and cytological evidence,
several students of the family have postulated intrafamiliar relationships
within Thelypteridaceae, most notably Loyal (1963), Smith (1971), and Pichi
Sermolli (1977). All of these studies invoked morphological and cytological
evidence, and offered somewhat different and conflicting pictures of rela-
tionships. Loyal (1963) depicted two main evolutionary lines within the fam-
ily, one line with species groups having a chromosome base number of 36,
with species having one or more pairs of veins from adjacent pinna segments
united below the sinus. The second evolutionary line in Loyal’s scheme
comprised species mostly with dysploid (stepwise progression in the basic
chromosome set) chromosome numbers (27 to 35) and free veins. Smith’s
(1971:46) scheme was an attempt to update Loyal’s tree, and showed basi-
cally a free-veined evolutionary line, with the phegopteroid ferns terminat-
ing this branch, and three shorter anastomosing- or connivent-veined lines
having x=36. These included an Old World line, a New World line, and
Stegnogramma s.].

Pichi Sermolli (1977:441) hypothesized a basal dichotomy in the family
leading, along one line, to the evolution of several free-veined genera
(phegopteroid ferns plus Metathelypteris), followed by other free-veined
elements (Parathelypteris, Coryphopteris, Oreopteris, and Thelypteris), and
ultimately giving rise to several free-veined and connivent-veined New
World species groups (Amauropelta, Steiropteris, and Glaphyropteris). As a
second evolutionary line, Pichi Sermolli derived all of the anastomosing-
veined Old World genera. Although Holttum (1971, 1982) revised nearly all
of the Old World genera, and commented extensively on relationships, he
never presented a formal phylogenetic treatment. Holttum (1971) believed

SMITH & CRANFILL: INTRAFAMILIAL RELATIONSHIPS 141

that certain Old World cyclosoroid genera (Pronephrium, Nannothelypteris,
Stegnogramma, Sphaerostephanos, and Pneumatopteris) formed a natural
group. He also postulated that Mesophlebion, Chingia, and Glaphyropteri-
dopsis were related, and that Coryphopteris and Parathelypteris formed an
isolated group. According to Holttum (1982:386), Cyclosorus s.s. and Ampel-
opteris (with united veins) and Thelypteris s.s.(with free veins), all with
wide distribution and similar aquatic habitat, formed another closely related
group. One of Holttum’s most important contributions was the recognition
that the three phegopteroid genera formed a related group (Holttum, 1969,
1971).

Although elements of all of these precladistic and largely intuitive trees
have some credence, the molecular data provide a unique new hypothesis of
relationships. The phegopteroid genera (Phegopteris, Macrothelypteris, and
Pseudophegopteris) appear to be the sister group to all other taxa in the fam-
ily (Figs. 1, 2). Within the phegopteroid clade, Macrothelypteris appears as
the sister group to Pseudophegopteris and the two Phegopteris species. All
nodes within the phegopteroid clade have high bootstrap support (Fig. 1).

Following the origin of Thelypteris s.s., which is shown to be basal in the
thelypteroid—-cyclosoroid clade, the free-veined thelypteroids arise, with
greater or lesser bootstrap support for various nodes, and no significant sup-
port at all for some of the nodes at the crown of one of the main branches
(Fig. 1). Thelypteris s.s. is shown to be only distantly related to Cyclosorus
s.s., thus refuting the close relationship suggested by Holttum (1982:386).

The cyclosoroids, from Steiropteris on up, form a large, mostly unresolved
or poorly resolved clade toward the tip of the tree (Figs. 1, 2). This topology
supports the contention of Pichi Sermolli that, in general, those elements in
the family with anastomosing veins and x= 36 are evolutionarily derived in
the family. In the tree shown, Dictyocline and Leptogramma appear as sister
genera, a result that supports all recent classifications of the family. Most
authors, including Iwatsuki (1963), Holttum (1982), and Smith (1990) unite
these two segregates in the same genus or subgenus.

The lack of resolution of various genera within the cyclosoroid clade
(Figs. 1, 2) may be an artifact of inadequate sampling or an indication of in-
sufficient variation in the genes sampled; more likely, it also reflects weak
distinctions between and among genera within this clade. Several hybrids
are known between species in different cyclosoroid genera (Viane, 1985;
Quansah & Edwards, 1986; Sledge, 1981; J. Game, unpubl. data). Many of the
cyclosoroid generic segregates seem little more than one- or few-character
genera, often containing exceptional species that do not fully conform in
their diagnostic features (Smith, 1990). We note in this regard that the three
species of Christella sampled in our study do not form a monophyletic
group, and in fact the single New World species sampled, Christella auges-
cens (Link) Pic. Serm., is well separated from the other two species. This
ecies (Smith, 1971; Smith, 1990) has

sometimes been treated in a different subgenus, subg. Pelazoneuron, in con-

trast to the largely Asian subg. Christella.

142 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

CHARACTER EvoLuTION.—Below we discuss briefly several characters often ap-
plied in circumscription of groups (genera, subgenera, sections) of Thelypter-
idaceae, expecially those characters that appear to correlate to a significant
extent with the results presented here.

Venation.—All students of the family have recognized the importance of
venation in the classification and subdivision of the thelypteroid ferns and
have utilized that character, whether free, connivent at the sinus, or anasto-
mosing in various ways, as a primary key character and feature delimiting
genera or infrageneric taxa within the family (e.g., Christensen, 1907, 1913,
1920; Holttum, 1971, 1982; Tryon & Tryon, 1982:432—453; Smith, 1990). Our
results show that the basalmost clades in the family, from the phegopteroid
genera through Amauropelta and Parathelypteris (Fig. 3) are all free-veined
or at least have veins that run to the sinus (but do not anastomose). In con-
trast, most of the groups belonging to the cyclosoroid clade (Fig. 3, shaded
boxes) have veins that are either connivent at the sinus, or anastomose at an
acute angle below the sinus, or unite at an oblique angle below the sinus
with an excurrent vein to the sinus. In some groups, there are multiple series
of anastomoses between the costa and pinna margin, as in the neotropical
groups Meniscium and Goniopteris, or in the paleotropical groups Prone-
phrium, Sphaerostephanos, and Dictyocline. Vein fusion is known to have
arisen independently in many leptosporangiate fern lineages, and no doubt
has also evolved many times within Thelypteridaceae. Smith (1990:271)
pointed out that half of the 20 subgenera of Cyclosorus s.]. have both free-
veined and anastomosing-veined members. The evolutionary trend toward
increased vein anastomosing, however, is unmistakable, and lends support
to the overall topology of the tree. Many of the outgroups we used in this
study are also free-veined, but certain ones, such as Onoclea and Lorinseria,
also have evolved anastomosing venation.

Anastomosing or reticulate venation in Thelypteridaceae is generally of a
simple or very repetitive nature: at most there are simple (unforked), excur-
rent free veins within areoles, as in Meniscium. This pattern of vein reticula-
tion, termed ‘“‘intersegmental” by Wagner (1979), is so common in man
thelypteroid groups that Wagner termed the venation pattern “‘thelypteroid”’;
if the pattern is repeated several times, Wagner (and others) have termed it
“meniscioid” (after the genus Meniscium) or “goniophlebioid” (similar to a
venation pattern in the genus Goniophlebium, a member of the Polypodia-
ceae). The various venation types in the family have been discussed and il-

forked veinlets in the areoles.

Adaxial sulcation.—Holttum (1960) was one of the first to discuss in depth
the importance of sulcation, or adaxial grooving, as applied to the system-
atics of ferns. He soon recognized that all three phegopteroid genera, as well

SMITH & CRANFILL: INTRAFAMILIAL RELATIONSHIPS 143

2 PSEUDOCYCLOSORUS

AISUNOBO1DAD
AA TAHL

AION LAA ISHS.

x=34

ACN. AOOsHa

rooted on Asplenium, and Asplenium through Cystopteris represent outgroups used in the anal-
ysis. Familial and subfamilial groups are indicated at the right. See Table 1 for full names of
species and additional voucher data.

Fic. 3. Strict concensus tree of the 28 most parsimonious trees found in the combined parsi-
mony analysis, based on the chloroplast genes rps4, trnS spacer, and trnL spacer. Thelypteroid
taxa having connivent or anastomosing veins are indicated by shaded boxes; chromosome base
numbers for various groups are shown along the right. See Table 1 for full names of species and

additional voucher data.

as Metathelypteris, lacked adaxial grooves along the costae (Holttum, 1969,
1982). This contrasts with all other thelypteroid ferns, which have the costae
adaxially sulcate. As Holttum (1960) noted, some genera of higher leptospor-
angiate ferns (e.g., Dryopteris and Polystichum) have the grooves of one axis
continuous with the grooves of the axes of greater (and often lesser) orders,
but continuous grooving is unknown in Thelypteridaceae.

The grooving of many of the outgroups used in this study has not been
sufficiently studied, but most appear to be adaxially grooved, including both
Cystopteris and Gymnocarpium, the apparent closest relatives. .

Blade dissection—With rare exceptions, blades of all cyclosoroid and
thelypteroid genera in Figure 2 are pinnate-pinnatifid or less divided. —.
few species that have more divided fronds (e.g., Thelypteris (Amaurope ta)
pteroidea (Klotzsch) R. M. Tryon) are clearly derived, and barely 2-pinnate.
A few members of some cyclosoroid clades have simple or merely pinnatifid

144 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

blades, e.g., certain species of Meniscium, Goniopteris, Pronephrium, and
Sphaerostephanos. In contrast, the phegopteroid genera have blades that are
bipinnate-pinnatifid (or even more divided), or the blades are bipinnatifid or
tripinnatifid (Phegopteris).

Dimorphism.—Frond (blade) dimorphism is nowhere strongly pronounced
in the family, as it is in many other fern families, but does occur on a subtle
to moderate scale (subdimorphism) in many (but not all) members of the cy-
closoroid clade, e.g., Christella, Goniopteris, Meniscium, Pronephrium, and
Sphaerostephanos. Dimorphism is nearly totally lacking among members of
the phegopteroid clade, as well as most members of the thelypteroid clade;
exceptions to this are found in the subdimorphic group Thelypteris s.s., and
subtly in Coryphopteris and Parathelypteris. There appears to be a positive
correlation between degree of vein anastomosing and blade dimorphy in the
family, i.e., those groups that have significant vein reticulation also are more
likely to be subdimorphic. Hemidimorphism (fertile-sterile differentiation on
separate parts of the same leaf; Wagner, 1977) is essentially unknown in
Thelypteridaceae. As pointed out by Wagner (1979), dimorphy appears to be
a character that is valuable mainly at the species level, not at generic rank,
and the evolution of dimorphy (or subdimorphy) has occurred indepen-
dently many times in the ferns and in the Thelypteridaceae. It is doubtful
that this character is of any help in delimiting clades within this family.

Among outgroups used in this study, most have monomorphic fronds, but
members of the Blechnaceae (e.g., Lorinseria and Stenochlaena) are often,
but not always (e.g., Doodia, Sadleria, a few Blechnum species), strongly
dimorphic, as are members of the onocleoid ferns (Onoclea and Matteuccia
in our sample).

Spore morphology.—Spores of thelypteroid ferns are generally monolete
and kidney-bean-shaped, with a relatively thick, folded, cristate, reticulate,
or echinate perispore. Spore surveys of thelypteroid genera suggest that spore
ornamentation is, in some cases, correlated with the segregate taxonomy

in Amauropelta. Species of Stegnogramma, in the cyclosoroid alliance, have
distinctive echinate spores, but these are not unlike many species of Sphaer-

pteris and Meniscium, as well as in the paleotropical group Mesophlebion
(Tryon & Lugardon, 1991:401), and these similarities (as well as others) sug-
gest a possible close relationship. Spores of Trigonospora are trilete, the only

SMITH & CRANFILL: INTRAFAMILIAL RELATIONSHIPS 145

such spores in the family; this is unquestionably a derived condition in
Thelypteridaceae, and not likely an indication of relationship to the Jurassic
fossil Aspidistes, as suggested by Holttum (1982) and others. No doubt, there
is much more evidence of a taxonomic nature to be gleaned from a thorough
and comparative study of spore morphologies in the family.

Biogeography.—Most of the segregate genera recognized in thelypteroid
ferns are restricted either to the New World or to the Old World. In the
tropics, exceptions are Cyclosorus s.s., Christella (predominantly Old World)
and Amauropelta (predominantly New World). In north-temperate areas,
Oreopteris, Thelypteris, and Parathelypteris are amphi-oceanic, occurring in
both North America and Eurasia.

The cyclosoroid clade (Fig. 2) is entirely tropical or subtropical in its cur-
rent distribution; few members extend above 25° north latitude. Contrast-
ingly, of the basal thelypteroid segregates, Thelypteris s.s. (except the
segregate T. confluens (Thunb.) C. V. Morton also occurs in south temperate
regions), Oreopteris, and Parathelypteris are mostly found north of 25° north
latitude, and Metathelypteris also has many temperate representatives.
Amauropelta and Coryphopteris are the only largely tropical segregates. In
the phegopteroid alliance, Phegopteris itself is north-temperate in distribu-
tion, in both eastern Asia and North America, while Macrothelypteris and
Pseudophegopteris are restricted essentially to the Old World tropics and
subtropics (M. torresiana (Gaudich.) Ching is widely naturalized in the New
World). Several outgroups sampled are cosmopolitan (e.g., Asplenium, Cys-
topteris, Woodsia); others are more restricted (e.g., Lorinseria and Sadleria).

We note that the phegopteroid clade and also the basal clades of the the-
lypteroid lineage are, for the most part, Laurasian in distribution. This distri-
bution perhaps indicates an east Asian origin, north of the Tethys sea, for
the family. Although both Pseudophegopteris and Macrothelypteris are con-
fined to the Old World (in their native distribution), the other phegopteroid
and most of the basal thelypteroid genera are found in both the Old World
and New World, and most are decidedly northern in their general distribu-
tion. Only higher in the thelypteroid lineage, in particular the cyclosoroid

nearly confined to the Cen
the historic inclusion of thelypteroid ferns in the distantly related dryopte-
roid assemblage. The best known and earliest examples of thelypteroid fos-
sils appear to be from Eocene strata of North America, and Europe
(Collinson, 2001:209-212). Further study and re-evaluation of identifications
in light of modern systematic knowledge of extant members is needed to de-
termine whether the fossil evidence adds support to a Laurasian origin for
the family.

Chromosome numbers.—
are characterized by having a
tatives of all of the cyclosoroid genera s
on x = 36, and many species in most 0

Most of the segregate genera of Thelypteridaceae
single base chromosome number. All represen-
ampled (Fig. 3) are known to be based
f these segregates have been sampled.

146 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

Chromosome base numbers are known for all groups except Nannothelyp-
teris, which has been combined with Pronephrium sect. Dimorphopteris by
Smith (1990) as suggested by Holttum (1982:538). Of the phegopteroid gen-
era (Fig. 3), Phegopteris is consistently x = 30, with all three species
counted; Pseudophegopteris is x = 31, with eight species counted; and
Macrothelypteris is also x = 31, with four species counted. The situation in
the free-veined thelypteroid segregates is more complex. There appears to be
a dysploid series of numbers that characterizes the various groups (Fig. 3:
Lovis 1977). Thelypteris s.s. is consistently x = 35; Oreopteris is x = 34 (all
three species counted); Coryphopteris species are x = 32 and 33 (Smith,
unpubl.), but with few counts available; reports for Metathelypteris are
mostly x = 31, 34, 35, and 36 (five species counted); Parathelypteris is
variously x = 27, 31, 32, 34, with the first two numbers being most
commonly reported; and Amauropelta is consistently x = 29, with ca. 25
species counted. Whether the multiple numbers for some of the segregates
indicate unnaturalness in the existing classifications or accurately reflect
chromosomal variation among (and even within) closely related species
remains to be determined. It does appear, however, that there is some
chromosomal instability within the basal thelypteroid group of genera.

SUMMARY AND CONCLUSIONS

Although taxon sampling is still only preliminary, we conclude that the
family Thelypteridaceae is monophyletic, with a high degree of certainty. It
excludes a few genera that have sometimes been included in the family by
Ching (e.g., Hypodematium; Ching, 1963; Trichoneuron; Ching, 1978) but ex-
actly coincides with the circumscription given by Iwatsuki (1964), Holttum
(1971), Pichi Sermolli (1977), and Smith (1990). Although our sampling is
not yet sufficient to favor one of the many Classifications (Morton, 1963;
Iwatsuki, 1964; Holttum, 1971, 1982; Ching, 1963, 1978; Pichi Sermolli,
1977; Smith (1990) of the family over another, our analysis suggests that rec-
ognition of an intermediate number of genera may be a reasonable taxonomic
course. The three phegopteroid genera appear to be basal in the family and
clearly sister to the remaining genera and subgenera.

ACKNOWLEDGMENTS

ial of Cyclosorus, material which was to serve as

SMITH & CRANFILL: INTRAFAMILIAL RELATIONSHIPS 147

We thank the following individuals for help in procurring either living or silica-gel-dried materi-
al for sequencing: John Game, Philip Hammond, Barbara Joe Hoshizaki, Masahiro Kato, John T
Mickel, and Barbara Parris. Gerald Gastony kindly provided DNA sample of Onocleopsis. We
also thank individuals and staff at the New York Botanical Garden and the University of Califor-

collection data, and providing for the collection of vouchers. Thanks also to numerous individu-

als who provided field assistance in Malaysia and Taiwan, where Cranfill collected many of the
Asian vouchers and DNA material cited in Table 1. The National Science Foundation provided
support to both Smith (DEB-9616260) and Cranfill (DEB- 0073036).

LITERATURE CITED

CHING, : C. 1940. On natural classification of the family ‘‘Polypodiaceae”. Sunyatsenia 5:

pre A coasted of the family Thelypteridaceae from the mainland of Asia. Acta
Phytotax. Sin. 8:289—
. 1978. The Chinese gel families and genera: ‘e paeecr arrangement and historical ori-
Acta Phytotax. Sin. 16(3):1-19 and 16(4):16—
Cont C. 1907. Revision of the American species - Dryopteris of oe group a 5 opposita.
el. Danske Vidensk. Selsk. Skr., Naturvidensk. Math. Afd., ser. 7, 4:247—
——. aats, A monograph of the genus Dryopteris. Part. I. The rae pone pinata
bipinnatifid species. Kongel. Danske Vidensk. Selsk. eg Naturvidensk. Math. Afd., ser. 7,

. 1920. A enol of the genus Dryopteris. Part. II. The tropical American bipinnate- ap
compound species. Kongel. Danske Vidensk. Selsk. Skr., Naturvidensk. Math. Afd., ser. 8,

6:1-13
COLLINSON, “4 E. 2001. Cainozoic ferns and their distribution. Brittonia 53:173-235.
CranFiLL, R. 2000a. Has ecological > aemimaignn driven a fundamental phylogenetic split in the

Polypodiales? Amer. J. Bot. 87(6, Suppl.):9
. 2000. Phylogenetic utility of plastid acta protein S4 (rps4) in land plants. Amer. J.
Bot. 87(6, Suppl.):121.
sear a S., M. KALLeRsJO, A. KLUGE, — C. Butt, 1995. Constructing a significance test for in-
ongruence. Syst. Biol. 44:570-57
PaLsenstet, J. 1985. Confidence aes on n phylogenies: An approach using the bootstrap. Evolu-
783-791

Hasese, M., P. G. Siem K. M. Pryer, K. UEDA, — R. Sano, G. J. GasTONY, : YOKOYAMA,
J. R. MANHART, N. MURAKAMI, Ec H: sunt _H. Haurier, and W. D. Haux. 1995. Fern
phylogeny based on rbcL nucleotide sequences. Amer. Fern J. 85:134—181.

Houtrum, R. E. 1960. Vegetative characters ; eal the various groups of ferns included

in soars in — s ai Filicum, and other ferns of similar habit and sori.
9. Studies in the se vaiskepisadlioee The genera Phegopteris, Pseudophegop-
aia por Macrothelypteris. Blumea 17:5—32.

. 1971. Studies in = family Thelypteridaceae III. A new system of genera in the Old
World. Blumea 19:

. 1972. Studies in ie fa
20:105—126

. 1973. Studies in the family
Blumea 21:293-325.

. 1973b. The family Thelypteri
Crabbe & B. A. Thomas, eds.
Press, London

. 1974. Studies in the family

mily Thelypteridaceae IV. The genus Pronephrium Presl. Blumea

Thelypteridaceae V. The genus Pneumatopteris Nakai.

daceae in the Old World. Pp. 173-189 in A. C. Jerm
The Phylogeny and Classification of the Ferns. Arcane

Thelypteridaceae VII. The genus Chingia. Kalikasan 3:13-28.

148 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

- 1975. Studies in the family Thelypteridaceae VIII. The genera Mesophlebion and Plesio-
neuron. Blumea 22:223-250.
1976a. bein in the family Thelypteridaceae X. The genus Coryphopteris.

. 197 eb. satis in ve arene Thelypteridaceae XI. The genus Christella Léville, sect.
agora Kew Bull. 31:

—.. 7. Studies in Pet family Thelypteridaceae XII. The genus Amphineuron
shee Blumea 23:205-218.

- 1981. The genus Oreopteris (Thelypteridaceae). Kew Bull. 36:223-226.

. 1982. oo Flora Malesiana, ser. II. Pteridophyta 1(5):331-560. Martinus
Nijhoff, The Hague

—, = J. W. Grimes, 1979. The genus Pseudocyclosorus Ching (Thelypteridaceae). Kew Bull.

499-516.
Iwatsuxi, K. 1962. On the venation of the cana ferns. Acta Phytotax. Geobot. 20:219-227,
——. 1963. Taxonomic studies of Pteridophyta VII. A revision of the genus Stegnogramma

emend. Acta Phytotax. Geobot. 19:112-126.

. 1964. Taxonomy of co thelypteroid ferns, with special reference to the species of Japan
and adjacent regions. III. Classification. Mem. Coll. Sci. Kyoto Imp. Univ., ser. B, Biol.:
31:11—40.

Lovis, ie D. 1977. Evolutionary patterns and processes in ferns. Pp. 239-440 in R. D. Preston &
H. W. Woolhouse, eds. Advances in Botanical Research. Academic Press, London

LoyaL, D. S. 1963. Some evolutionary trends in family Thelypteridaceae with partionlar refer-
ence to lene yee species. Mem. Indian Bot. Soc. 4:22—29,

Maxon, W. R. and C. V. Morton. 1938. The American species of Dryopteris, subgenus Menis-
cium. Bull. Tetrey Bot. Club 65:347-376.

Morton, C. V. 1963. The classification of Thelypteris. Amer. Fern J. 53:149-154.

Napot, S., G. Birrar, L. Carrer, R. LaCroix, and B. LEJEUNE, 1995. A phylogenetic analysis of
monocotyledons based on the ‘deraplaat gene hed using parsimony and a new numerical
henetics method. Molec. Phylogen. Evol. 4:25

PICHI SERMOLLI, R. E. G. 1977. Tentamen Se genera in taxonomicum ordinem redi-
gendi. Webbia 31:313-512

Pryer, K. M., A. R. Soir, an dj. E. Skoc. 1995. Phylogenetic yeaa of extant ferns based
on evidence from morphology and rbcl. sequences. Amer. Fern J. 85:205—282

Quansak, N. and D. S. Epwarps. 1986. A natural bigeneric fern hybrid in Thelypteridaceae from
Ghana. Kew Bull. 41:805-809.

SHING, K. 1999. Flora Reipublicae Popularis Sinicae, vol. 4(1). Pteridophyta: Hypodematiaceae,
Thelypteridaceae. Science Press.

SLEDGE, W. A apo The Thelypteridaceae of Ceylon. Bull. Brit. Mus. (Nat. Hist.), Bot. 8:1-54

SmitH, A. R. 19 = ess “ si neotropical species of Thelypteris section Cyclosorus.
Univ. Calif. pre Bot.

- 1980. Taxonomy of a spters os Steiropteris, including Glaphyropteris (Pterido-

phyta). Univ. Calif. Publ. Bot.

. 1990. ed econ Pp. pores n K. U. Kramer & P. S. Green, vol. eds., The Fami-

of Vascular Plants. ar" I. Pteridophytes and Gymnosperms. Springer-

Verlag, hatin
Sworrorp, D. L. 1999. Phylogenetic analysis using parsimony (*and other methods). Version 4,

eta 8. Sinauer Associates, Sunderland, MA

- GIELLY, ou, and J. Bouver. 1991. Universal primers for amplification of
three non-coding regions of chloroplast DNA. Pl. Molec, Biol. 17:1105—1109.
‘ B. LuGAaRDON. 1991. Spores of the Pteridophyta. Springer- -Verlag, New Y

and A. F. Tryon. 1982. Ferns and Allied Plants with Special Reference to Tropical
America. pgelleane: New York.

ViANE, R. L. L. 1985. A n W species and a new Sa hag of Thelypteris (Pteridophyta) from the
Ivory Coast. Bull. hog Roy. Bot. Belg. 118:4

SMITH & CRANFILL: INTRAFAMILIAL RELATIONSHIPS 149

WAGNER, us - o 1977. Fertile-sterile leaf dimorphy in ferns. Gard. Bull. Straits Settlem.
30:2

os Retculate veins in the systematics of modern ferns. Taxon 28:87-95.

WoLr, P. G., S. D. Swes, M. R. Wuite, M. L. Martines, K. M. Pryer, A. R. SmirH, and K. UEDA
1999. Baraat ee of the enigmatic fern families Hymenophyllopsidaceae
and ears: Epa vidence from rbcL nucleotide sequences. PI. Syst. Evol. 219:263-270.

Woop, C. C. 1973. Spore frneren in the Thelypteridaceae. Pp. 191-202 in A. C. Jermy, J. A.
Crabbe & : A. Thomas, eds. The Phylogeny and Classification of be Ferns. hectic
Press, London.

American Fern Journal 92(2):150-160 (2002)

Two New Species of Moonworts
(Botrychium subg. Botrychium) from Alaska

Mary Ciay STENSVOLD
USDA Forest Service, Alaska Region, 204 Siginaka Way, Sitka AK 99835
DONALD R. FARRAR
Department of Botany, Iowa State University, Ames, IA 50011
CINDY JOHNSON-GROH
Department of Biology, Gustavus Adolphus College, St. Peter, MN 56082

AssTRact.—Botrychium tunux and Botrychium yaaxudakeit, new species of moonworts currently
known only from southern Alaska, are described and illustrated. These ferns are distinguished

B. lunaria, with which they have been confused, by allozyme data and their morphological
characteristics. Ploidy levels of B. tunux (diploid) and B. yaaxudakeit (tetraploid) are inferred
from allozyme patterns. A key to Alaskan moonworts is presented.

Before 1995, little attention was paid to Alaskan moonworts, Botrychium
subg. Botrychium. Information gained from rare plant surveys since then has
improved our knowledge of moonwort abundance, distribution, and relation-
ships. Seven species of moonworts are now known in southern Alaska:
Botrychium ascendens W.H. Wagner, B. lanceolatum (S.G. Gmel.) Angstrém,
B. lunaria (L.) Sw., B. minganense Vict., B. pinnatum H. St. John, the new
species B. alaskense W.H. Wagner & J. R. Grant, and two additional new
species described here, B. tunux and B. yaaxudakeit.

During a 1980 USDA Forest Service rare plant survey, several Botrychium
identified as B. lunaria (M. C. Muller 3806, ALA, TNFS) were collected in
sandy upper beach meadows near Yakutat, in southeastern Alaska. In 1986,

. H. Wagner and F. S. Wagner (1986) described a new moonwort,
Botrychium ascendens, and re-identified some of the 1980 specimens as
B. ascendens; other specimens on the herbarium sheets remained B. Junaria.
The inclusion of Alaska in the range of B. ascendens in Volume 2 of the

Flora of North America (W. H. Wagner, 1993) was based on the 1980 Muller
collection.

STENSVOLD ET AL: TWO NEW SPECIES OF MOONWORTS 151

Farrar’s analysis identified B. minganense, B. lunaria, and two moonworts
previously unknown to science. These new moonworts, a diploid and an
allotetraploid, resemble B. Junaria most closely in morphology, but are
distinct genetically and morphologically from all other known moonworts.
To our knowledge the diploid has not been previously collected. Previous
collections of the tetraploid had been included in B. Junaria.

Genetic comparison of B. tunux with B. lunaria at 18 allozyme loci
yielded a value for Nei’s (1978) genetic identity (GI) of 0.5102 (Farrar, un-
publ. data). This represents a level of genetic similarity only slightly greater
than that between B. Junaria and B. simplex (GI=0.4646) and is significantly
less than the similarity between B. Junaria and its sister taxon, B. crenula-
tum (GI=0.7015). Absence of fixed heterozygosity at any enzyme locus
implies that plants of B. tunux are diploid. Presence of fixed heterozygos-
ity and a spore size significantly larger than that of B. Junaria implies that
plants of B. yaaxudakeit are tetraploid. Allozyme patterns also show that
one ancestral diploid parent of B. yaaxudakeit is North American B. lunaria;
the other ancestral diploid is probably not among the known North Ameri-
can diploid species. Preliminary investigation of European B. Junaria indi-
cates that B. yaaxudakeit may be the result of ancient hybridization between
genetically distinct North American and Eurasian genotypes of B. Junaria.

Key TO THE SPECIES OF Botrychium Susc. Botrychium OF SOUTHERN ALASKA

iy rhs on ca cans hk es ae he ER EE PPS Oe OERET CED RY Has os
3(2). Pinna bases cuneate; pinna apices acute .....---+++++- see rr rrr rt eects
3. Pinna bases obtuse to truncate; pinna apices rounded : . pinnatum
4(1). Basal pinnae narrowly fan-shaped to oblong with the blade spanning an arc less
than 90°; pinnae remo
4. Basal pinnae broadly fan

FS ko oisclind scl css SEK Gilg Te OUR OAT LE ES eh iS entel an veie tes
-shaped with the blade spanning an arc greater than 90°; pin-
n ere Se. er Seat Oh OF eee MAE Fe 6
5(4). Trophophores sessile; pinnae strongly ascending, their margins dentate to lacerate; spor-
TTC sc scans soe oe es ascendens
cending, oblong to narrowly fan-shaped,
lly lacking on the trophophore pinnae
by rilece alegre eeuaeanrs B. minganense
6(4). Basal pinnae asymmetrical with the basal portion expanded; mature sporophore
stalks shorter than or equal to the trophophores; basal sporophore branches spread-
Seip and twined: plants Q-idiene Bill. ove «1s onrss nits dats seee es TS Ree
6. Basal pinnae generally symmetrical; mature sporophore stalks longer than the
trophophores; basal sporophore branches ascending and straight; plants 8-25 cm i

Bey en a ee ee ee a

falls 2 Ree ade Ginna Sees

7(6). Pinnae strongly overlapping one another an
greater than 180°; basiscopic inner margins 0

45 (43-48) pm in longest diameter ...----+-+-++errrrrrrss

d the rachis; basal pinnae spanning an arc
f the basal pinnae strongly recurved; spores
B. yaaxudakeit

152 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

7. Pinnae approximate to somewhat overlapping, not overlapping the rachis; basal pinnae

spanning an arc of less than 180°; basiscopic inner margins of the basal pinnae linear to
moderately concave; spores 36 (34-39) um in longest diameter ............. B. lunaria

Botrychium tunux Stensvold & Farrar, sp. nov.—Type: U.S.A.: Alaska:
Yakutat area, ca. 5 km W of Yakutat, on the Phipps Peninsula ca. 1.5 km
SE of Point Carrew, just N of the mouth of the Ankau River, ca. 1.5 m elev.,
25 Jul 2001, M. C. Stensvold & D. R. Farrar 7936 (holotype ISC; isotypes
ALA, NY, TNFS, US, WTU). Figs. 1c, 2c, 3b, 3c.

Plantae supraterraneae 6-12 (ave. 9) cm altae; stipites communes 0-3 (ave.
1.5) cm longi. Trophophora flavovirentes coriaceae oblongae. Paria pinnarum
plus minusve perpendicularia ad rhachim, separata vel leviter imbricata;
pinnae basales 7-20 (ave. 13) mm longae, 7-18 (ave. 12) mm latae, sessiles,
flabellatae, angulo ad basim 120-180°, saepe asymmetricae, parte basali
expanso, margine externa integra vel interdum lobata, sinibus rotundatis.
Stipites sporophora 2.5-5 (ave. 4) cm longi, trophophoria aequantes vel
breviores. Sporae 38—42 jum diametro.

12 (ave. 9) cm tall with a common stalk 0-3 (ave. 1.5) cm long. Tropho-
phores yellow-green, leathery; stalks 0-1 cm long; blades 2.5-7 (ave. 4) cm
long, 2-4 (ave. 3) cm wide, narrowly ovate to ovate, once pinnate. Pinna
pairs 4-6, more or less perpendicular to the rachis, separated to slightly
overlapping, not overlapping the rachis. Basal pinnae 7-20 (ave. 13) mm
long, 7-18 (ave. 12) mm wide, sessile, fan-shaped spanning an arc of 120—
180°, often asymmetrical, with the basal portion expanded; basiscopic inner

major veins entering the pinna base, 45-55 veins ending at the margins.
Sporophores 5-10 (ave. 7) cm long; sporophore stalks 2.5-5 (ave. 4) cm
long and shorter than or equaling the length of the trophophore; sporangia-
bearing portion erect, 1—2-pinnate, broadly ovate in outline, branches 4-6,
ascending to spreading, especially the lowermost, which are often twisted
such that the sporangia project downward; sporangia partially embedded
in the distally thickened sporangiophore branches. Spores 38—42 (ave. 40)
tm in longest diameter, released later than those of B. lunaria. Apparently
diploid.

This species is morphologically most similar to B. lunaria, from which it
can be distinguished by its short stature, short common stalk, frequently
stalked, ovate trophophore, asymmetrical pinnae with their basiscopic side
expanded, entire margins commonly cleft by shallow incisions with rounded
sinuses, and sporophore stalks seldom exceeding the height of the tropho-

.
>

generally short-stalked or unstalked trophophore. Its fan-shaped pinnae are

153

STENSVOLD ET AL: TWO NEW SPECIES OF MOONWORTS

Ayal

b yy
beh
7

<a
a

oS

.

=
S
N

eae
ade
Bae
CO
OH Ce). ar
OO pc:

Ep ton ee
tay

*
Pa
+

STENSVOLD

showing the differences in stature and mor-

phology. Fig. 1a. B. yaaxudakeit. Fig. 1b. B. lunaria. Fig. 1c. Botrychium tunux.

Fic. 1. Illustrations of three species of Botrychium

154 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

generally symmetrical, or if they are asymmetrical, the acroscopic side is
larger, and if cleft, the clefts have acute sinuses, and the sporophore stalks at
maturity usually exceed the height of the trophophore (Figs. 1b, 2b).

ParaTyPes.—U.S.A.: Alaska: Yakutat area, E shore of the Phipps Peninsula
ca. 1.5 km SE of Point Carrew, just N of the mouth of the Ankau River,
Stensvold 6854,7280 (TNFS), Farrar 99-07-23-2 (ISC); Blacksand Spit, off
the mouth of the Situk River, Stensvold 7949 (ISC), Farrar 01-07-27-1 (ISC);
Yakutat forelands, ca. 68 km SE of Yakutat, on NW-SE-oriented sandspit
separating the Pacific Ocean from the mouth of the Akwe River, ca. 3 km W
of the junction of the Akwe and Ustay Rivers, V. Harke s.n. (TNFS), Stensvold
7305, 7770 (both TNFS), Farrar 99-07-24-2, 99-07-24-6 (both ISC); Yakutat
forelands, ca. 80 km SE of Yakutat, Dry Bay, NW side of the mouth of the
Alsek River, Farrar 99-07-25-2 (ISC).

Botrychium tunux is known from four locations in the Yakutat area: Point
Carrew near the village of Yakutat, Blacksand Spit off the mouth of the Situk
River, Akwe beach, and the northwest shore of Dry Bay, all on the Yakutat
forelands, an area 80 km long extending between Pt. Carrew and Dry Bay in
habitats adjacent to saltwater but not influenced by saltwater inundation.
Searches for B. tunux in similar habitat on the Copper River Delta barrier
islands, some of the Malaspina Glacier forelands, Cross Sound, Icy Strait,
Glacier Bay and Lynn Canal areas were unproductive, although other
Botrychium species were found. We believe potential habitat for both of the
new species extends along the Alaska coastline for 250 km in either direc-
tion from Yakutat. Because the area is frequented by storms and accessible
only by aircraft capable of beach landings, this coastline is botanically
underexplored. A single collection of six plants from the Nutzotin Moun-
tains ca. 13 km northwest of Chisana (ca. 305 km. NNW of Yakutat) differ
somewhat genetically and morphologically from the coastal B. tunux, but
may be a conspecific population long isolated from coastal plants.

In the Yakutat area, B. tunux grows in habitats ranging from open sand on
dunes and upper beaches to well drained upper beach meadows with sandy
substrates (Figs. 3b-d). The plants May occur as scattered individuals or in
loosely associated groups. Vascular plants most commonly associated with
B. tunux include Achillea borealis, Festuca rubra, Fragaria chiloensis, Gentia-
nella amarella, Leymus mollis, Lupinus nootkatensis, Oxytropis campestris,

squarrosus. The moonworts B. ascendens, B. lunaria, B. minganense, and
B. yaaxudakeit also grow in these meadows. Vascular plant and bryophyte
cover can range from a trace to 100%, with the percent cover and species
number increasing landward. These vegetation assemblages are early seral

deposition or isostatic rebound and tectonic uplift. Picea sitchensis and Alnus
viridis colonize the meadows and develop into Picea sitchensis forests

STENSVOLD ET AL: TWO NEW SPECIES OF MOONWORTS 155

"
3

Fic. 2. Pressed specimens of Botrychium cl
of the new species near Yakutat, Alaska. F
Fig. 2c. B. tunux.

. \
‘

b
Cc

ipped just above ground level, from the type locality
ig. 2a. Botrychium yaaxudakeit. Fig. 2b. B. Junaria.

156 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

eS

Fic. 3. Botrychium yaaxudakeit and Botrychium tunux at the type locality at Ankau Beach near
Yakutat, Alaska. Fig. 3a. Two plants of B. yaaxudakeit. Fig. 3b. A single plant of B. tunux.
Fig. 3c. B. tunux (left) and B. yaaxudakeit (right) showing the typical stature of the two species.

—
"a

‘ig. 3d. Sparsely vegetated upper beach sand habitat of B. yaaxudakeit and B. tunux, with a
Sitka alder (Alnus viridis subsp. sinuata) beach-forest ecotone and a young Sitka spruce (Picea
sitchensis) forest in the background.

(Fig. 3d), while new meadows develop on recently formed land to the seaward
(Shephard, 1995).

Related plants in the Nutzotin Mountains were widely scattered on a
sparsely vegetated southwest-facing slope in the Gold Hill vicinity at an
elevation of about 1615 m on alpine scree slopes. Vascular plants at this
location include Erigeron caespitosus, Senecio cymbalaria, Arnica frigidia,
Silene uralensis subsp. uralensis, Anemone drummondii var. lithophila,
and Elymus trachycaulus subsp. violaceus. The moonworts B. ascendens

STENSVOLD ET AL: TWO NEW SPECIES OF MOONWORTS 157

and B. minganense also grow on this scree slope. Further study of plants
from this and similar inland sites is needed.

In reviewing herbarium material and literature, we found no prior collec-
tions for B. tunux. Neither Coville (1895) nor Stair and Pennell (1946) men-
tion any species of Botrychium in their papers describing the flora of the
Yakutat area. According to their reports, they likely visited the type locality.
During the 1899 Harriman Alaska Expedition, Coville and Kearney collected
B. lunaria in Yakutat Bay (Merriman, 1910), and in 1904 C. V. Piper col-
lected what appears to be B. yaaxudakeit at the type locality. We suspect
that B. tunux may not have been collected because of its dwarfed and mal-
formed appearance relative to the larger B. Junaria and B. yaaxudakeit, with
which it grows.

The epithet tunux was chosen by the elders of the Yakutat Tlingit people
and honors Tunux, the Tlingit warrior who initiated the 1805 attack on
the Russian fort located near the type locality. As a result of the battle, the
Russians were removed from Yakutat and never returned. Tunux was in the
Eagle Moiety and the Teikweidi (Brown Bear) Clan. The word Tunux is
pronounced with a guttural “x” similar to the German ‘‘ach,” making the
name sound like “‘toonook,” with a soft ‘“k.”

Botrychium yaaxudakeit Stensvold & Farrar, sp. nov.—TypE: U.S.A.: Alaska:
Yakutat area, ca. 5 km W of Yakutat, on the Phipps Peninsula ca. 1.5 km
SE of Point Carrew, just N of the mouth of the Ankau River, ca. 1.5 m el-
ev., 25 Jul 2001, M. C. Stensvold & D. R. Farrar 7937 (holotype ISC; iso-
types ALA, NY, TNFS, UC, US, WTU). Figs. 1a, 2a, 3a, 3c.

Plantae supraterraneae 8-25 (ave. 18) cm altae; stipites communes 1-5
(ave. 2.5) cm longi. Trophophora viride nitidae coriaceae ovatae. Paria pin-
narum plus minusve perpendicularia ad rhachim, valde imbricata; pinnae
basales 7-30 (ave. 18) mm longae, 9-32 (ave. 20) mm latae, breviter petio-
latae, late flabellatae, angulo ad basim 180-250°, plerumque symmetricae;
margine interna basiscopica valde recurvata; margine externa integra vel
undulata vel interdum denticulata, interdum fissurata sinibus angulatis.
Stipites sporophorora 5—9 (ave. 7) cm longi, excedentes trophophoria. Sporae
43-49 um diametro. ,

Rhizomes erect, unbranched, their apex 2-4 (ave. 5) cm below the soil sur-
face, bearing 5-24 (ave. 13) fleshy roots; gemmae absent. Above-ground
plants 8-25 (ave. 18) cm tall, with a common stalk 1-5 (ave. 2.5) cm long.
Trophophores green, leathery; stalks 0-0.5 cm long; blades 1.75-11 (ave. 7)
cm long, 1.25—6 (ave. 4) cm wide at the base, oblong to ovate, once pinnate.
Pinna pairs 4-7, angled toward the apex, strongly overlapping one another,
the anterior portion overlapping the rachis. Basal pinnae 7-30 (ave. 18) mm
long, 9-32 (ave. 20) mm wide, short-stalked, fan-shaped, spanning an arc of
of 180° to 250°, usually symmetrical; basiscopic inner margin strongly
recurved; outer pinna margins entire to undulate, occasionally denticulate or

158 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

occasionally shallowly cleft with angular sinuses; veins dichotomous, 3—5
major veins entering the pinna base, 90-120 veins ending at the margins.
Sporophores 8-18 (ave. 14) cm long; sporophore stalks 5-9 (ave. 7) cm long
and longer at maturity than the length of the trophophore; sporangia-bearing
portion erect, 1-2 pinnate, broadly lanceolate to narrowly ovate in outline,
the branches 6-8, basal branches ascending and not twisted. Spores 43-48
(ave. 45) jum in longest diameter, released earlier or at about the same time
as those of B. Junaria. Apparently tetraploid.

This species is morphologically most similar to B. Junaria, from which it
can be distinguished by its taller stature, shorter common stalk, more ovate
trophophore, and strongly overlapping pinnae that also overlap the rachis.
Its basal pinnae are often short-stalked with a span of over 180° and have
strongly recurved basiscopic inner margins (Figs. 1a, 2a). Botrychium lunaria
is shorter in stature (12 cm), has a longer common stalk (3.5 cm), oblong to
narrowly ovate, usually unstalked trophophores, and pinnae that generally
do not strongly overlap one another or the rachis, are sessile with a span of
less than 180°, and have slightly concave basiscopic inner margins (Figs. 1b,
2b). The spores of B. yaaxudakeit are significantly larger than those of
B. lunaria.

PaRATypPes.—Alaska, Yakutat area, E shore of the Phipps Peninsula ca. 1.5
km SE of Point Carrew, just N of the mouth of the Ankau River, M. C. Muller
3808, Stensvold 7272, (TNFS), M. Shephard 22 (TNFS), Stensvold 6854-2
(TNFS), Farrar 99-07-23-3 (ISC); Ankau River, 31 Aug — 1 Sep 1904, C. V.
Piper 4565 (US); Yakutat forelands, ca. 68 km SE of the town of Yakutat, on
a NW-SE-oriented sandspit separating the Pacific Ocean from the mouth of
the Akwe River, ca. 3 km W of the junction of the Akwe and Ustay Rivers,
Stensvold 7305 and 7773 (both TNFS), Farrar 99-07-24-7 (ISC); Yakutat fore-
lands ca. 80 km SE of Yakutat, NW side of the mouth of the Alsek River,
well drained upper beach meadow, Farrar 99-07-25-3 (ISC); Glacier Bay, near
Rush Point, Stensvold 7532-1 (ISC): Glacier Bay, N of Gloomy Knob at the
outlet of Vivid Lake, Stensvold 7554 (ISC); Glacier Bay, near the mouth of
Tarr Inlet, Stensvold 7669-1 (ISC); roadside along the Richardson highway
11.5 km NW of the Salcha River crossing, at the junction with Johnson Road,
24 Jun 1999, W. H. Wagner s.n. (ISC).

Glacier Bay, and Lynn Canal areas resulted in B. yaaxudakeit discoveries at
three sites in Glacier Bay. The Yakutat habitats are described in the B. tunux

STENSVOLD ET AL: TWO NEW SPECIES OF MOONWORTS 159

loosely associated groups. Vascular plants most commonly associated with
B. yaaxudakeit in Glacier Bay include Achillea borealis, Fragaria chiloensis,
Festuca rubra, Honckenya peploides, Lupinus nootkatensis, Leymus mollis,
Gentianella amarella, Heracleum lanatum, Moehringia lIateriflora, Oxytropis
campestris, Rhinanthus crista-galli, Rubus stellatus, and Taraxacum spp.
Bryophytes commonly found with B. yaaxudakeit include Rhytidiadelphus
squarrosus, Ceratodon purpureus and Racomitrium canescens. The moon-
worts B. ascendens, B. lunaria, and B. minganense also grow in these mead-
ows. Vascular plant and bryophyte cover was 100%, except at the Tarr Inlet
site where plant cover was 30%. A single plant collected in south-central
Alaska near Fairbanks by W. H. Wagner (24 June 1999) was confirmed by
isozyme electrophoresis to be B. yaaxudakeit. This plant grows in roadside
vegetation along the Richardson highway.

In reviewing herbarium material, we found prior collections of B. yaaxu-
dakeit that had been identified as B. Junaria. C. V. Piper collected what
appears to be B. yaaxudakeit at the type locality: Ankau River, 31 Aug-1
Sep 1904, C. V. Piper 4565 (US 421289). Collections made at the type loca-
lity in more recent years include: Yakutat ca. 3 mi W of town on the eastern
Phipps Peninsula just N of the Ankau River, forest, sandy beach ecotone, 17
Jul 1980, M. C. Muller 3808 (TNFS) and Yakutat, in the back-beach meadow
at Point Carrew, 11 Jul 1993, Shephard 22 (TNFS). Two specimens from
Glacier Bay National Park and Preserve may also be B. yaaxudakeit, Glacier
Bay, Beardslee Islands, 3 Aug 1921, J. P. Anderson 1155 (US), and bedrock
knob north of Vivid Lake, 10 Jul 1990, K. Bosworth, s. n. (Glacier Bay
National Park and Preserve Herbarium).

The elders of the Yakutat Tlingit people chose the epithet, yaaxudakeit, to
honor Yaa Xu da Keit’. He was the leader of an ancient clan living at the
base of the mountain now known as Mt. St. Elias, and purchased Yakutat for
the Copper River people. Yaa Xu da Keit’ was from the Raven Moiety and
the Kwaashkikwaan (Humpback Salmon) Clan. Yaa Xu da Keit’ is pro-
nounced with a guttural “x” and “k” and a pinched ‘“‘t,”” making the name
sound like ‘“‘yaa khoo da kayt”’.

AACKNOWLEDGMENTS

thank David Wagner for his analysis of the Ya
and logistical assistance, Lynn Clark for the
the figures, Elaine Abraham, George Ramos an
support.

Latin descriptions, Anna Gardner for preparing
d David Ramos for their thoughtful advice and

LITERATURE CITED

Covitte, F. V. 1895. Botany of Yakutat Bay, Alaska, with a field report by F. Funston. Contr. U.

S. Natl. Herb. 3:325-350.
Merriman, C. H., ed. 1910. Harriman Alaska Series, vol. V, Cryptogamic Botany. Smithsonian

Institution, Washington, DC.

160 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

Nel, M. 1978. Estimation of average heterozygosity and genetic distance from a small number of
individuals. Genetics 89:583-590.

SHEPHARD, M. E. 1995. Plant community ecology and classification of the Yakutat Foreland,
Alaska. Tech. Rep. R10-TP-56. Juneau, AK: U.S. Department of Agriculture, Forest Service,
Alaska gr 206 pp.

Stair, L. and F. W. PENNELL. 1946. A collection of plants from Yakutat, Alaska. Bartonia 24:

9-21.
Wacner, W. H. Jr., and F. S. Wacner. 1986. Three new species of ee (Botrychium subg.
Botrychium) oudunite in western North America. Amer. Fern J. 76:3
and 1993. Ophioglossaceae C. Agardh. Adder’s-tongue nee ly. Pi 85-106 in Flora
of North Ainevica Editorial Committee, eds. Flora of North America, North of Mexico, vol.
2: Pteridophytes and Gymnosperms. Oxford University Press, New York.

American Fern Journal 92(2):161—163 (2002)

Isoétes < herb-wagneri, an Interspecific Hybrid of
I. bolanderi x I. echinospora (Isoétaceae)

W. Cart TAYLOR
Botany Department, Milwaukee Public Museum, Milwaukee, WI 53233

Asstract.—Isoétes Xherb-wagneri, the interspecific hybrid between Isoétes bolanderi and
I. echinospora, is described. The epithet commemorates Warren (Herb) Wagner for the direction
and help he provided to the author.

It can be difficult to find one’s niche, but Herb Wagner made it easy for
me. I met Herb during an American Fern Society foray at the University of
Michigan Field Station near Pellston, Michigan in August 1977. At the time,
Herb and I had just finished collaborating on a paper about Dryopteris
x leedsii. Although I had seen him at several gatherings, we had never for-
mally met. I recall our first meeting. For some reason, I thought Herb would
recognize me because we had corresponded by phone and mail about
D. x leedsii. I spotted him standing by the pay phone at the field station the
first evening we assembled for the field trip. I approached, smiled, and gave
him a cheery “Hi, Dr. Wagner’’. Herb paused, stared directly at me, and then
said, “excuse me, but who the h__ _ are you?”’ Stammering, I mentioned the
D. Xleedsii paper and my name. Herb immediately smiled, shook my hand
energetically, and then paid me a wonderful compliment. “I thought you
were much older,” he said. To my mind, of course, he was implying that,
based on our correspondence, he thought I had the wisdom and knowledge
of a much older person. During the two-day field trip Herb gave me lots of
attention and opened my eyes to reticulate evolution. He reinforced his
words with enthusiastic hunts for Dryopteris hybrids. Each time I found a
hybrid, Herb was right there to excitedly praise and encourage me. I was over-
whelmed by his interest and enthusiasm for ferns and for me. By the end of
the foray, I knew I would be studying pteridophyte evolution. The following
summer, I told Herb about my fascination with Isoeétes and how much |
enjoyed hunting them. A short while later he told me, ‘I want you to spend
the rest of your life studying quillworts. I'll help you.” Never had I received
such clear direction. After that Herb created all kinds of opportunities for
me to study Isoétes. He made contacts for me. He told others about my inter-
est in the genus. Herb even arranged for a helicopter to take me to the sum-
mit of Mount Eke on the Hawaiian Island of Maui to collect a quillwort he
and I would later name and describe.

The following Isoétes hybrid is named to honor Warren
whose direction and help I will always be grateful.

(Herb) Wagner, for

162 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

3cm

Fic. 1.
and its parents. A-C. Megaspores. A. I. bolanderi. B. L x herb-wagneri. C. I. echinospora.
D-F. Microspores. D. I. bolanderi. E. I. x herb-wagneri. F. I. echinospora. G. I. bolanderi (left),

Megaspores, microspores, and a living plant in cultivation of Isoétes herb-wagneri

I. x herb-wagneri (center), I. echinospora (right)

Isoétes x herb-wagneri W. C. Taylor, hybr. nov. Type: United States: Monta-
na: Ravalli Co.: from the bed of the upper of Twin L Lakes, near edge of the
autumn level, about 12 ft below the summer level, Lost Horse Canyon, 12
October 1940, Fred A. Barkley & Reuben A. Diettert 4804 (US 1786885). A

TAYLOR: AN INTERSPECIFIC HYBRID 163

mixed collection of three taxa annotated by W. Carl Taylor as (a) I. bolanderi,
(b) I. echinospora, (c) I. bolanderi x echinospora, (d) sterile plant. Fig. 1

Planta hybrida inter Isoétem bolanderi et I. echinosporam, aliis charac-
teribus inter parentes intermedia, sporis male formatis.

Isoétes x herb-wagneri has been collected only from the type locality at
Twin L Lakes. Water is drawn from the adjoining lakes during the summer
months, so that by fall, when water levels are low, Isoétes plants are exposed
along the margins of the lakes, rooted in a gravelly, clay soil. These exposed
plants are a mixture of I. x herb-wagneri and I. bolanderi. Isoétes echino-
spora occurs slightly deeper in the lakes, still submerged in the fall, and
scattered in stands where the mineral soil is combined with more organic
sediments that have accumulated in places.

Isoétes < herb-wagneri can be recognized by its irregular megaspores that
vary in size, shape, and surface ornamentation (Fig. 1B). In contrast, the
more uniform spores of I. bolanderi and I. echinospora are tuberculate or
echinate, respectively (Fig. 1A, C). Isoétes x herb-wagneri has characters
intermediate between its parents. For example, the uniform microspores of
I. bolanderi and I. echinospora (Fig. 1D, F) are brown or gray, respectively.
The irregular microspores of I. < herb-wagneri (Fig. 1E) are a combination of
brown, light brown, and gray. In addition, the leaves of I. bolanderi taper
abruptly to a bristle-like apex, whereas those of I. echinospora taper gradu-
ally to the apex. The leaf apices of I. x herb-wagneri are intermediate for this
character (Fig. 1G).

Paratype:—United States: Montana: Ravalli Co.: Upper Twin L Lake, 9
October 1989, David Bilderback s.n. (MIL).

American Fern Journal 92(2):164—170 (2002)

Botrychium alaskense, a New Moonwort
from the Interior of Alaska

WARREN HERB WAGNER, JR.
Department of Biology, University of Michigan, Ann Arbor, MI 48109-1048
JASON R. GRANT
Laboratoire de botanique évolutive, Institut de botanique, Faculté de Sciences, Université de
Neuchatel, Case Postale 2, CH-2007 Neuchatel, Switzerland

Asstract.—Botrychium alaskense W. H. Wagner & J. R. Grant is described as a new species from
the interior of Alaska. It is an allotetraploid of B. Junaria (L.) Sw. x B. lanceolatum (S.G. Gmel.)
Angstr., the same species parentage that gave rise to the morphologically and isozymically dis-
tinct B. pinnatum H., St. John.

Outside of North America, only seven species of Botrychium are currently
recognized: Botrychium boreale Milde, B. lanceolatum (S.G. Gmel.) Angstr.,
B. lunaria (L.) Sw., B. matricariifolium (Déll) A. Braun, B. multifidum (S.G.
Gmel.) Rupr., B. simplex E. Hitchc., and B. virginianum (L.) Sw. Each of
these species is also recorded in North America, although the identity of
B. boreale-like plants in North America is in question, and likely represent
either B. pinnatum H. St. John or B. alaskense W.H. Wagner & J.R. Grant. In
addition, North America has more than 25 endemic species, nearly all of
which have been recognized in the past three decades, and some of which
are still undescribed. A surprisingly large number of these occur in the cor-
dilleran region of western North America.

The majority of the North American endemics have been overlooked appa-
rently because of their inconspicuous appearance, their common occurrence
in unexpected, disturbed habitats, or their general rarity. We here describe
another new species, this one from the interior of Alaska.

Botrychium alaskense W. H. Wagner & J. R. Grant, sp. nov., Figs. 1-2.
A B. pinnato H. St. John tropophoro oblongo-deltato, saturate viridibus,
pinnis usque 4—6-jugis irregulariter decompositis, inter se approximatis vel

distantibus, lobulis plerumque oblongis angulatis, basi anguste (per 40—-90°)
cuneatis diversa.

—

Fic. 1. Three species of Botrychium from northwestern North America. Bottom row left
(3 specimens), Botrychium pinnatum (W. H. Wagner 99128 et al., MICH). Bottom row right
(3 specimens), B. lanceolatum (W. H. Wagner 99120 et al., MICH). Remaining (18 specimens),
B. alaskense (W. H. Wagner 99005 et al., MICH).

WAGNER & GRANT: BOTRYCHIUM ALASKENSE 165

166 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

TYPE: U.S.A. ALASKA. BIG DELTA QUAD: Vicinity of Fairbanks: Salcha
River, 12 km. up the river from its crossing of the Richardson Hwy, at mile
323.1, 64°29'195”N, 146°39’629’W, 23 June 1999, W. H. Wagner 99005 with J.
R. Grant, F. S. Wagner, A. Gilman, P. Zika, H. W. Grant and W. L. Grant (holo-
type MICH; isotypes ALA, B, BM, F, G, GH, ISC, MO, NY, P, S, U, UC, US,
WTU)

Trophophore stalk 3-10 mm long, up to 2 times the length of the tropho-
phore rachis; lamina bright green, leathery, oblong-deltate, 1-pinnate, up to 6
cm long and wide; pinnae up to 6 pairs, horizontal to slightly ascending, ap-
proximate to distant, the basal and supra-basal pair not or slightly more dis-
tant than the supra-basal and 3rd pairs, the basal pinnae equal to or slightly
longer than the supra-basal pair, broadly lanceolate to broadly trullate, un-
lobed (small plants) to deeply lobed (large plants), with up to 5, narrowly
oblong pairs of lobes, these up to 10 mm long, 3 mm wide, asymmetrically
truncate, broadly (2-3 mm) attached and confluent to the rachis, shallowly
concave at the basiscopic base; lobe tips irregularly crenate-dentate. Sporo-
phore 2-pinnate, with 3 major branches. Chromosomes evidently tetraploid.

PARATYPES: U.S.A. ALASKA. ANCHORAGE QUAD: Black Spruce Camp-
ground, Fort Richardson Military Reservation northeast of Anchorage, dis-
turbed floodplain with graminoids and forbs among young cottonwood and
alder, 1 August 2001, D. Farrar 01-07-01-4. BIG DELTA QUAD: Salcha River,
12 km up the river from its crossing of the Richardson Highway at mile
323.1, 64°29’N, 146°45’W, July 1996, J. R. Grant 96-02625b (ALA); Salcha
River, 12 km up the river from its crossing of the Richardson Highway at
mile 323.1, 64°29’N, 146°45’W, 23 June 1999, J. R. Grant 99-03572 with
W. H. Wagner, F. Wagner, P. Zika, A. Gilman, H. W. Grant, & W. L. Grant
(ALA); Between Harding Lake and the Salcha River on the Richardson Hwy
(Alaska Hwy. 2), opposite Harding Rd. [the entrance to the Harding Lake
Recreation Area], 64°26’N, 146°54’W, 255 m, 26 June 1999, W. H. Wagner
99119 with J. R. Grant, F. S. Wagner & P. Zika (MICH): Between Harding
Lake and the Salcha River on the Richardson Hwy (Alaska Hwy. 2), opposite
Harding Rd., 64°26’N, 146°54’W, 255 m, 26 June 1999 J. R. Grant 99-03584
(ALA); Salcha River, 12 km up the river from its crossing of the Richardson
Highway at mile 323.1, 64°29’N, 146°45’W, July 2000, J. R. Grant 00-3828
(ALA); Between Harding Lake and the Salcha River on the Richardson Hwy.,
opposite Harding Rd., 64°25’N, 146°53’W, 12 July 2000, J. R. Grant 00-3833
(ALA); Between Harding Lake and the Salcha River on the Richardson Hwy.,
opposite Harding Rd. 64°26’N, 146°54’W, 12 July 2000, J. R. Grant 00-3837
(ALA); Mile 323.1 Richardson Hwy where it crosses the Salcha River, dry
disturbed area in pioneer vegetation along roadside, 64°28’N, 146°55’W, 24
June 2001, J. R. Grant 01-4093 (ALA); Between Harding Lake and the Salcha

—_—

se = Botrychium alaskense W. H. Wagner & J. R Grant (drawn from Grant 00-3833 and Grant
-3837.

WAGNER & GRANT: BOTRYCHIUM ALASKENSE

FRASER
BO VOR oo
Cd
6 ae
Se
con

Botrychium

alaskense

168 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

TaBLE 1. Comparison of full-sized trophophores (and sporophores) of Botrychium pinnatum and
B. alaskense

Character B. pinnatum B. alaskense
Trophophore outline Narrowly ovate Broadly ovate-deltate
Abruptly narrowed Gradually narrowed
Pinna overlap when Mainly near pinnae bases Mainly submedial on pinnae,
present commonly absent
Lowest pinna length Mostly 2.0 cm or less Mostly 2.5 cm or less
Lowest pinna width Mostly 1.0 cm or less Mostly 1.5 cm or less
Lobe apices Strongly rounded Angular
Pinna base angle 120°-140° °-~90°
Contracted pinna base Less than 0.5 mm, broadly adnate _— At least 1 mm, narrowly adnate
Veinlet interval Nearly 1 mm Mostly 0.5 mm or less
Sporophore length 0.5—1.5 trophophore length 0.4-1.0x trophophore length
Low, widely separated Occasional Prevalent

sporophore branches

River on the Richardson Hwy., opposite Harding Rd., 64°26’N, 146°54’W, 26
June 2001, J. R. Grant 01-4096 (ALA). LAKE CLARK QUAD: North Rim of
the Western Amphitheater of the Northern Telaquana Badlands, Lake Clark
National Park, in a basin excavated by wind, south facing slope at 5 degrees,
vegetation 50% vascular, 50% lichen, 15 August 2001, P. Caswell s.n. LIVEN-
GOOD QUAD: White Mts., limestone ridge in a grassy slope, 800 m, 15 July
1953, O. Gjaerevoll 593 (ALA). MT. McKINLEY QUAD: Kantishna Mining
District, 63°31’N, 150°56’W, 900 m, Wickersham Dome, moist herbaceous,
NE-facing alpine slopes, 26 June 1990, C. J. Parker 2294 (ALA). NENANA
Q : Tamarack Inn, mi. 298, Parks Hwy. [Alaska Hwy. 2], 7 miles S of
Nenana, 64°28’N, 149°03’W, 25 June 1999, open field with B. lanceolatum
(rare), B. Junaria, and B. minganense, W. H. Wagner 99113 with J. R. Grant,
F. S. Wagner, & P. Zika (MICH); J. R. Grant 99-03578 (ALA). TALKEETNA
QUAD: Chedotlotina Glacier, Denali National Park, with netleaf willow,
mountain avens, polar willow, dwarf scrub meadow, 20 August 2001, M.
Duffy MD-01-258A. TANANA QUAD: Manley Hot Springs area, 21 km E of
Manley, 65°05’N, 150°19’W, roadside ditch near Extensive Survey Stand No.
138, 16 June 1973, J. Foote 3049 (ALA).

Herbarium records from western Alaska and Siberia may also represent
Botrychium alaskense, but require further study to confirm their identity.
The new species may be distinguished from its closest relatives by the fol-
lowing key:
+; Trophophore deltate; sporophore stalk 1/3-1/4 as long as the sporophore; pinna pairs 3 or

4; basal pinna pair 2x as wide as the adjacent pinna pair; basiscopic pinna margin straight
to slightly concave and upswept

=

WAGNER & GRANT: BOTRYCHIUM ALASKENSE 169

2(1). Trophophores mostly ovate-deltate; pinnae somewhat irregularly incised, bright green,
) i t to approximate, their lobes mostly narrowly oblong, angular; pinna

bases forming an auele of Ca, 40°90" ooo noes ctaagact ts coc Autos, fae! us ps B. alaskense
2. Trophophores mostly ovate; pinnae very regularly incised, dark (olive) green, broadly

oblong, mostly approximate, their lobes mostly broadly ovate, rounded; pinna bases

truncate to cordate, forming an angle of ca. 120°-140°..............4, B. pinnatum

From plants analyzed using enzyme electrophoresis, it was determined
that B. alaskense is an allotetraploid of B. Junaria x B. lanceolatum (D. Farrar,
pers. comm.). That would not be a particular surprise, except that B. pinnatum
has the same parentage. Farrar’s preliminary results indicate that B. alaskense
may have arisen through hybridization between B. Janceolatum x Eurasian
B. lunaria, whereas B. pinnatum originated as B. lanceolatum x American
B. lunaria.

Most of the localities where moonworts have been found in interior Alaska
include Botrychium alaskense; the others observed to date have only B. Ju-
naria and/or B. minganense. All sites with B. alaskense studied by one or
both of us are in open, non-forested areas. They include revegetating sand-
bars along the Salcha River, a mowed field or lawn ca. 30 years old along
the Parks Highway, and roadsides along the Richardson Highway.

The immediate vicinity of Fairbanks has produced seven species: B. alas-
kense, B. lanceolatum, B. lunaria, B. minganense Victorin, B. multifidum, B.
pinnatum, and a new species, B. yaaxudakeit Stensvold & Farrar. A possible
eighth species with affinities to B. alaskense may be another undescribed
species that is currently being investigated by the junior author. As presently
known, the Fairbanks area has the largest number of species of Botrychium
in Alaska. The most common are B. lunaria and B. minganense, and the
least common is B. multifidum. Botrychium lunaria was the only species in
several localities, while B. minganense appeared as the only species in one
locality. B. Junaria and B. minganense grew together in all localities except
the foregoing. B. alaskense, B. lunaria, B. minganense, and B. lanceolatum
were found growing together in three localities, and all seven species were
found growing together in one locality.

In three of the sites, the individuals of B. alaskense are estimated at well
over 200. The plants are obvious because of their unusually large size, char-
acteristic trophophore shape, and shining, bright green color. Growing with
B. alaskense are young woody plants or herbs. Associated woody plants in-
clude Arctostaphylos uva-ursi, Betula neoalaskana, Picea glauca, Populus
balsamifera, P. tremuloides, Rosa acicularis, Rubus idaeus, Salix alaxensis,
S. bebbiana, Shepherdia canadensis, an
prominent associated herbs that can site:
even when B. alaskense is absent include Taraxacum officinale, Fragaria vir-
giniana, Potentilla norvegica, Cornus canadensis, and Galium boreale. Par-
ticularly, the presence of strawberries nearly always indicates the presence
of moonworts.

Botrychium alaskense alw
In naturally disturbed sites,

ays grows in somewhat recently disturbed areas.
‘t is found on re-vegetating sandbars and along

170 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

new oxbow lakes. In man-induced disturbances, populations are found in
abundance growing in weedy, infrequently-mowed fields or lawns, or in the
ditches and associated edges of the major highways of the interior of Alaska.

The phenology of B. alaskense is intermediate between B. Junaria and B.
lanceolatum. Botrychium lunaria sporulates first and is completely finished
before the sporangia of B. lanceolatum have begun to ripen. Botrychium
alaskense sheds its spores after B. Junaria has finished and before those of B.
lanceolatum.

ACKNOWLEDGMENTS

Henry W. and Wyan L. Grant, and Peter Zika. Rupert Barneby is thanked for preparing the Latin
diagnosis and for reviewing the manuscript. Bobbi Angel skillfully prepared the line-drawings.

American Fern Journal 92(2):171-178 (2002)

A New Name for an Old Fern from North Alabama

James E. WarkIns, JR." and DoNALD R. FARRAR
Department of Botany, Iowa State University, Ames, lowa 50011

Asstract.—The varieties of Thelypteris pilosa have been recognized as the sole New World

Two regionally sympatric morphotypes, varying from deltate to lanceolate fronds, occur through-
out Mexico and have been described as var. major and var. pilosa, respectively. A more distinct
type, described as var. alabamensis, is endemic to north Alabama rockhouse habitats and has
been reported from only a single county. Data on ecology, spore morphology, gametophyte biol-
ogy, and gross frond morphology support the elevation of T. pilosa var. alabamensis to specific
status under the proposed name of T. burksiorum.

The year 2002 marks the 160th year since the publication of Martens and
Galeotti’s Mémoire sur les fougéres du Mexique. This monumental work re-
presented the first attempt to catalogue the ferns of Mexico and thus intro-
duced several new fern species to science, including Thelypteris pilosa,
which they discovered in the state of Oaxaca (Martens & Galeotti, 1842,
p. 27). This taxon has been recognized as the sole New World member of the
obscure, largely Old World Thelypteris subg. Stegnogramma. Over one hun-
dred years later, Crawford (1951) discovered a single population of T. pilosa
in Alabama and initiated the first examination of the New World representa-
tives of subg. Stegnogramma. Using a limited set of morphological charac-
ters, Crawford recognized three varieties, two of which occur throughout
Mexico, Guatemala, and Honduras, and the third consisting of the Alabama
plants. He applied the name T. pilosa var. pilosa to a lanceolate morphotype,
which happens to correspond with the line drawing rendered by Martens
and Galleotti, and T. pilosa var. major to a deltate morphotype. The more
distinct and disjunct Alabama plants, characterized by their much smaller
fronds and obtuse pinna apices, he recognized as T. pilosa var. alabamensis.
This segregation of the Mexican varieties has been challenged by Smith
(1981, p. 236), while Iwatsuki (1964) recognized Crawford’s original vari-
eties. It remains unclear whether these taxa are sufficiently distinct to be rec-
ognized as varieties, warrant elevation to specific rank, or are simply
extremes within a morphological cline. Because both varieties major and pi-
losa are common, occur sympatrically in Mexico, and intergrade morphologi-
cally throughout their distribution in Mexico, the question of their genetic

1 Current address: Department of Botany, University of Florida, P. O. Box 118526, Gainesville,

FL 32611-5826.

172 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

distinction remains unresolved. Here, we discuss the unique morphology
and biology of Crawford’s third variety, T. pilosa var. alabamensis, and re-
cognize this taxon as a distinct species.

Thelypteris burksiorum J. E. Watkins and D. R. Farrar, sp. nov.—TYPE:

.S.A., Alabama: Winston County, in fissures of Pottsville sandstone, on

west fork of Sipsey River 5 miles E of Double Springs, Watkins 797 (holo-
type NY; isotypes ISC, UC, UF, UNA). (Fig. 1).

Thelypteris pilosa (M. Martens & Galeotti) Crawford var. alabamensis Craw-
ord, Amer. Fern J., 41:19. 1951. Leptogramma pilosa var. alabamensis
(Crawford) Wherry, So. Fern Guide 346. 1964. Stegnogramma pilosa var.
alabamensis (Crawford) K. Iwatsuki, Amer. Fern J., 54:142. 1964. TYPE.—
U.S.A. Alabama: Winston County: In fissures of Pottsville sandstone, on W
fork of Sipsey River 5 miles E of Double Springs, Crawford 241 (holotype
UNA; isotypes UNA, US).

Ex affinitate T. pilosa (Martens & Galeotti) Crawford a qua imprimis differt
plantae epipetricae. Frondes lanceolatae, 3.5-20 cm. longae; pinnae 0.5-1.5
x 0.5—-0.7 cm; laminae, pinnae, et segmenta apice rotundatae.

Rhizomes short-creeping, with lanceolate, reddish brown scales covered
with acicular hairs. Fronds fasciculate; stipes 1.5-8 cm long; laminae ever-
green, 3.5-20 cm long, 1-3.5 cm wide, often covered with acicular hairs on
the veins and laminar tissue, pinnate, lanceolate, tapered to the base, the
apex acute and pinnatifid; sori elongate, exindusiate; sporangia with acicular
hairs; spores papillate, never spinulose; gametophytes cordate, regularly pro-
liferating, with acicular and copious glandular hairs on the thallus surface.

The species is named in honor of Dr. Robert and Mrs. Mary Burks, who
have been voices of conservation in Alabama for decades and who aided in
the establishment of the Bankhead National Forest.

Several excellent descriptions of the frond morphology of T. burksiorum
(Smith, 1993, pp. 217-218) and of T. pilosa exist (Mickel & Beitel, 1988,

387; Moran & Riba, 1995, p. 193). Sporophytes of T. burksiorum
differ most notably from T. pilosa in their smaller size and abrupt apices. A
statistical analysis of 23 quantitative morphometric frond characters found
the Alabama plants of T. burksiorum to fall outside the range of variation
of the Mexican varieties of T. pilosa (Watkins, 2000). Following is a more
detailed description of the unique spore and gametophyte characteristics of
T. burksiorum.

Spore MorpHo.ocy.—Thelypteris burksiorum exhibits a spore morphology
outside the range of variation of the Mexican material examined (Watkins,
2000). The surface architecture can be described as papillate-
verrucose (Fig. 2E, F). The blunt tubercles of the perispore that give this
appearance vary in size with an average length of 0.62 mm (N=15). These
outgrowths were never observed to converge and form connections, as
occurs In some specimens from Mexico. The monolete laesura is usually

WATKINS & FARRAR: A NEW NAME FOR AN OLD FERN 173

ght. Frond. Middle right. Detail of pinna
showing linear, exindusiate sori an acicular hairs on the veins as and laminar tissue. Lower
right. Capsule of a sporangium with acicular hairs. Left. Habit of an individual plant growing on

Fic. 1. Sporophytes of Thelypteris burksiorum. Top Ti

smooth on its surface and free of such ornamentation. The perispore surface
nular on and between tubercles. Mean spore diameter of T.
burksiorum along the greatest axis is 47 mm (N=100), ranging from 44 to 54
mm; along the smallest axis it is 30 mm, ranging from 27 to 36 mm. Mean
spore diameter of T. pilosa var. pilosa along the greatest axis is 46 mm
(N=120), ranging from 39 to 49 mm; along the smallest axis it is 26 mm,
ranging from 25 to 31 mm. Meiotic squashes yielded a base chromosome
number of N=36, which has been reported to be the base number of Thelyp-
teris subg. Stegnogramina (Smith, 1993, p. 217).

is slightly gra

and gametophyte morphology. Secretory hairs begin
along the margin (Figs. 2A, 3A),
cushion. By two months, the thallus surfa
secretory hairs. Development of acicular hairs (Fig. 2A) begins after further
maturation (approximately 2.5 months). These are produced sparsely over

174 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

WATKINS & FARRAR: A NEW NAME FOR AN OLD FERN 175

the archegonial cushion, rarely on the wings, and never along the margins.
The gametophyte margins have a ragged-irregular appearance (Figs. 2B-D,
3B), and the surface topography is smooth. In multispore cultures, most ga-
metophytes were without gametangia, but a quantitative determination of
gametangium abundance was made difficult by the development of copious
gametophytic proliferations described below. For this reason, determining
the exact sequence of gametangial development was impossible. Archegonia
and swimming sperms (when placed in solution) were visible as early as 2.5
months. All gametophytes that developed gametangia, including those col-
lected in the wild, were hermaphroditic when mature. Attempts to induce
cross-fertilization between T. burksiorum and the varieties of T. pilosa failed
to produce sporophytes (Watkins, 2000).

An unusual form of asexual reproduction occurs in gametophytes of T.
burksiorum that does not occur in gametophytes of Mexican T. pilosa. Iso-
lated gametophytes, as well as those in multispore culture, begin to produce
proliferations on the dorsal side of the thallus early in development (Fig. 2B,
arrows). The proliferations usually begin as filamentous branches that even-
tually broaden and flatten to form strap-shaped outgrowths (Fig. 3B). Many
of these outgrowths eventually take on the appearance of new heart-shaped
thalli with pluricellular meristems (Fig. 2C). With experimental severance
and transfer to new media, these developed into typical gametophytes that
also produced additional proliferations similar to the parent gametophyte.
Through proliferation, a single isolated gametophyte, in one year, increased
to a size of 3.5 cm and formed a mass of meristematic thalli on its dorsal
surface. After 1.5 years of growth, the original thallus still persisted and con-
tinued to proliferate (Fig. 2D). Although young, developing proliferations do
not produce gametangia while still attached to the parent gametophyte, it
seems likely that these proliferations can eventually become independent,
producing clonal populations with sufficient gametangia to promote sexual
production of sporophytes.

Tue Unusual Hapirat oF THELYPTERIS BURKSIORUM.—After Crawford’s original
discovery of T. burksiorum in 1952, the taxon went basically unnoticed until
bridge-building in the 1960's supposedly destroyed the type and only known
locality. It was later rediscovered by Jack Short and John Freeman (1978),
and subsequently studied intensively by Gunn in the early 1990's (Gunn,
d by the authors in 1998 and 1999. The species is
17 populations, all of which occur on a

fr * . . .
4-mile segment of the Sipsey Fork of the Black Warrior River in Winston

,

and ornamentation.

176 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

Fic. 3. Gametophytes of Thelypteris burksiorum. A. Distribution of acicular and glandular hairs
on the thallus. B. Thalli showing asexual proliferations in the basal region

County, Alabama. This area is characterized by sheer, towering sandstone
cliffs composed of Pottsville Sandstone, upon which all known populations
of this plant occur (Fig. 4). All plants grow in crevices and rough surfaces
on the roofs and floors of sandstone rockhouses formed along these cliffs.
Although these rockhouses take on many forms, in the classic sense they are
semicircular recesses extending under cliff overhangs (Farrar, 1998; Walck
et al., 1996). The key to understanding the microclimates of these intriguing
habitats is that they behave much like caves in providing temperature and
moisture moderation. In most cases, this creates areas that are always warm-
er than the surrounding area during the winter and cooler during the sum-
mer (Farrar, 1971, 1998). Those rockhouses that have the ability to maintain
relatively constant microclimates, combined with a constant supply of water,
have been shown to harbor a unique assemblage of tropical bryophytes
(Sharp, 1937), tropical pteridophytes (Farrar, 1998), and endemic temperate
angiosperms (Walck et al., 1996). It seems probable that Thelypteris burksio-

WATKINS & FARRAR: A NEW NAME FOR AN OLD FERN

Fic. 4. Habitat of Thelypteris burksiorum, which is found only in crevices of i ee sand-
stone along the Sipsey Fork of the Black Warrior River in Winston County, Alabam

rum, like other tropical members of the rockhouse floras, are relictual from a
broader pre-Pleistocene distribution of the species.

ACKNOWLEDGEMENTS

The authors thank John Mickel, Gary Landry, and Boone Halberg for their field grog
Mexico, as well as Alan Smith and Paul Wolf for their insightful ricgeiraaae = . wer
tates Forest Service and the United States Fish and Wildlife spain oe ee zs pean incalcu-
lect T. burksiorum in the Bankhead National Forest; and Marion Licther (USF* oe the s pcbna,
lable field assistance in this area, Anna Gardner oo ng si ie aa 4 = their helpful
and Jonathan Wendel, Mark Widrlechner, David Lellinger, and a % ay ae Kevin sas of
comments on the manuscript. We also thank Willard Bawers, ip piety ha slaleaas
Alabama Power Company for providing for logistical support for field c

wn

178 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

LITERATURE CITED

Crawrorp, L. C. 1951. A new fern for the United States. Amer. Fern J. 41:15-20

Farrar, D. R. 1971. The Biology of Ferns with Asexually Reproducing Gametophytes in the East-
em pa States. Ph.D. Dissertation, University of Michigan, Ann Arbor, MI.

The tropical “eels of rockhouse cliff formations in the eastern United States. J.
eis Bot. Club 125:9

Gunn, S. C. 1990. Sensitive ee of the Bankhead National Forest: survey of selected compart-
ments of the northern one-third. Unpublished Report to U.S. Forest Service. 68 pp. Mont-
gomery, AL.

———. 1991. An update on the status of Thelypteris pilosa var. alabamensis. Unpublished
Report to U.S. Fish and Wildlife Service. 19 pp. Jacksonville, MI.

- 1994. A survey for the occurences and habitats of the federally oi threatened

cp streak-sorus fern, Thelypteris pilosa (Mart. and Gal.) Crawford var. alabamensis
n the erat National Forest. Unpublished Report to U.S. Forest kee 37 pp. Mont-
pai

. 1996. eas streak-sorus fern Pe chan pilosa var. alabamensis) recovery plan. 27

pp. U.S. Fish and Wildlife Service, Atlanta

IwarsukI, K. 1964. An American Species of mane Amer. Fern J. 54:141-143.

Martens, M., and H. Gateorti. 1842. Mémoire sur les fougéres du Mexique et considérations sur
la — botanique de cette contrée. Nouv. Mém. Acad. Roy. Sci. Bruxelles 15:

Saat ‘ T.; ne y M. BEITEL. 1988. Pteridophyte Flora of Oaxaca, Mexico. Mem. New York
Gard., 48:1-568.

ae e C., and R. Rwa, eds. 1995. Flora Mesoamericana, vol. 1. Universidad Nacional Autén-
oma de México, México,

SuarpP, A. J. 1937. Tropical bryophytes in the southern Applachians. Ann. Bryol 11:141-144.

Sort, J. W., and J. D. FREEMAN. 1978. Rediscovery, yeaa and phytogeographic affinities
of Leptogramma pilosa in yin aly Amer. Fern J. 68:

SmiTH, A. R. 1981. Flora of Chiapas, part 2. Pteridophytes. at. Academy of Sciences, San
Fransci

1958, ‘Thelypesdacone Ching ex Pichi-Sermolli—Marsh Fern Family. Pp. 206-222 in

Flows of North ca Editorial Committee, eds. Flora North America North of Mexico,
vol. 2. Oxford Univ. ae New York.

Watck, J. L., J. M. BAskin, and C. C. BASKIN. 1996. Sandstone rockhouses in the eastern United
States, with particular reference to the ecology and evolution of endemic plant taxa. Bot.

Rev. 62:311

Watkins, J. E e - Origin and Taxonomic Affinities of Thelypteris pilosa (Martens and
Galeotti) ome (Thelypteridacese), M.S. Thesis, Iowa State University, Ames

American Fern Journal 92(2):179-183 (2002)

Continued Pteridophyte Invasion of Hawaii

KENNETH A. WILSON
Natural History Museum of Los Angeles County, 900 Exposition Boulevard,
Los Angeles, CA 90007

ABSTRACT.—Two new alien pteridophytes have become established in the Hawaiian Islands since
1996, bringing the total of naturalized alien ferns to 32. Also, established alien species continue
to spread onto new islands.

Warren Herb Wagner, Jr. was the first to publish a comprehensive report
of the pteridophytes naturalized in the Hawaiian Islands (Wagner, 1950).
According to this publication, a total of 21 alien species had become estab-
lished in the Hawaiian Islands. By 1996 an additional nine naturalized species
had been discovered (Wilson, 1996). This paper reports on two newly
discovered alien species that have become established, three new records
documenting the spread of naturalized species into additional islands, a
collection recording the earlier presence of a species than was previously
known, and one revised identification of a widespread alien fern. The current
status of naturalized ferns and fern allies in Hawaii is shown in Table 1.
In the table the species are arranged by the date of their first collection in
Hawaii, as determined by the earliest available herbarium collection.

Azolla filiculoides Lam.—In 1943 Fosberg reported Azolla filiculoides
to have become fully naturalized in taro patches and irrigation ditches on
Oahu (Fosberg 1943), after having earlier been deliberately introduced into
the Islands as part of a mosquito abatement program. Wagner (1950) reported
having collected it on Oahu, as well as on Maui. In 1996, I reported the
species to be found from flooded areas on all of the islands except Hawaii,
although I speculated that it was also to be found there. A collection by
Clyde Imada made in June 1999 from the island of Hawaii now documents
its presence there, where he found it growing in a taro farm in Waipio Valley
(Imada 99-16, BISH).

Blechnum appendiculatum Willd.—The Blechnum species that grows in
Hawaii has been known as B. occidentale L. since its occurrence was first re-
ported. Recent studies, however, have shown that the rachises of B. occiden-
1 surface, whereas those of B. appendiculatum
Blechnum appendiculatum also differs in
d darker rhizome scales than does B. occidentale
lso Hoshizaki & Moran, 2001, p. 216). The
(syn. B. glandulosum
Kaulf. ex Link). Both species are natives of tropical America. ae

Lygodium japonicum (Thunb.) Swartz—A population of Lygodium japoni-
cum has been known to be growing north of Hilo, where it was first col-

180 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

TABLE 1. Naturalized ferns and fern allies in Hawai'i arranged by year of their first collection.
(x = Species recorded on the island

Species Year Kauai O’ahu’ Molokai Maui Hawai’i
Pteris vittata L 1887 x x x x
ae dinitertes ee ) 1887 x x x x x
t. John
ha gramma austroamericana Domin 1903 x x x x x
Adiantum raddianum C. Presl 1907 x x x x“ mt
Macrothelypteris torresiana 1908 x se - x x
(Gaud. in
Pityrogramma calomelanos (L.) Link 1908 x x x x x
Phlebodium aureum (L.) J. Smith 1909 x“ x - oe x
Diplazium esculentum (Retz.) Swartz 1910 x x = x x
Blechnum appendiculatum Willd. 1917 x x x x x
Ceratopteris thalictroides (L.) Brongn. 1919 x x = S =
'vmatosorus grossus 1919 x x x
(Langsd. & Fisch.) Brownlie
Adiantum hispidulum Swartz 1923 x x x x x
ny ves ‘score sea ) 1923 x x x x x
J tex C
es pain 4) Fosb. 1926 x D4 x sx x
Cyrtomium falcatum (L. f.) C. Pres] 1928 x x x x
Cheilanthes viridis (Forssk.) Swartz 1928 x x - x ~
Lygodium japonicum (Thunb.) Swartz 1936 - x - - x
gi oan aes (Cav.) 1936 x x * 4 x
Chr. *
pes palin 1937 x x x x x
Deparia petersenii (Kunze) M. Kato 1938 * x x x x
Selaginella kraussiana (Kunze) A. Br. 1938 = pd = x x
Angiopteris evecta (G. Forster) Hoffm. 1950 = x ~ x x
Lindsaea ensifolia Swartz var. 1969 x x a * x
ensifolia
Selaginella stellata Spring 1969 - ~ - - x
Ti 1970 x - x x
W. J. Hook. ex F. Muell.) Domin
Bok incisa ree 1972 x x - - x
Adiantum ‘Edwi 1981 oe = = x =
Nephrolepis hist (G. Forst.) 1983 x = = 4 =
Presl ‘S
Adiantum rigottaie Swartz 1987 = = = x =
Platycerium bifurcatum (Cav.) 1991 _ x ze x x
C. Chr. subsp. bifurcatum
Marsilea crenata C. Pres] 1995 - x es ms a
Salvinia molesta D. S. Mitchell 1999 - x _ _ x
Selaginella umbrosa 2000 = = z 5 x

Lemaire ex Hieron.

lected in 1936 and where it has persisted. Until recently this species is not
known to have spread to other sites in Hawaii nor to have become particu-
larly damaging, as it is in the southeastern United States. The status of this
species in Hawaii has changed significantly in the last few years. In 1998
George Staples collected it on Oahu, where he found it growing on the

WILSON: CONTINUED PTERIDOPHYTE INVASION OF HAWAII 181

grassy hills above Heeia State Park (Staples 1166, BISH, LAM). Reports have
reached me that populations of this species have been seen growing on the
island of Hawaii on the hill opposite the Astronomy Department building of
the University of Hawaii, Hilo (Palmer, pers. com.), and on Oahu along the
Ulu Paina trail, adjacent to Temple Valley, Kaneohe, as well as on another
site in the same vicinity (Waters, pers. com.). In 2000, plants were found in
Kaneohe, volunteering in gardens among cultivated and native plants (Gagne
s.n., BISH #664545, LAM).

Marsilea crenata C. Presl—A well established population of this aquatic
was found growing in the demonstration taro patches maintained at the
Hawaii Nature Center, Makiki Valley, Honolulu, Oahu. The presence of
this population was first recorded on 18 May 1994 in the journal maintained
by the Nature Center by Marie Bruegmann. The first collection made at this
site was by D. Palmer (Palmer 2275, BISH #650412) on 1 June 1995. This
taro patch was recently neglected and the water was allowed to drain, but
M. crenata has persisted in the marshy area, although it does not grow so
luxuriantly. Renewed attention to the demonstration taro patch will certainly
stimulate the growth of this fern. Marsilea crenata is native to the Philip-
pines, Malaysia, Indonesia, Australia, where it grows in shallow fresh water
and in drying mud. It is often found in rice paddies. Plants are rooted in the
mud, and the leaflets float on the water surface or the erect petioles extend
above the surface of the water and hold the leaflets well above the surface.
Sporocarps apparently do not develop on the submerged plants, but have
been found on a few plants growing along the margins of the taro patch.

Salvinia molesta D. S. Mitchell—This floating aquatic was discovered
growing on Oahu in Enchanted Lake, Kailua, and in Lake Wilson, Wahiawa,
in April 1999. The earliest collection I have seen of S. molesta in the wild
was made on 1 April 1999 from the canal and upper end of Kaelepulu Pond,
adjacent to Kiukee Place, Enchanted Lakes area, Kailua (J. Cook S.n., BISH).
This fern is grown in garden ponds and aquaria and is known to be in culti-
vation in the Islands. The first collection documenting its occurrence in cul-
tivation in Hawaii was made on 2 July 1991 on Oahu (Wilson & Staples s.n.,
BISH #606163, LAM). It is also known to be in cultivation on the island of
Hawaii. Salvinia molesta readily escapes from cultivation and invades
bodies of fresh water where it grows aggressively, rapidly becomes a serious
pest. Numerous collections were made on Oahu in April 1999, both in the
Enchanted Lakes area, Kailua, and in Lake Wilson, Wahiawa, where it has
become naturalized. On 15 April 1999 the Honolulu Star-Bulletin reported
on planned eradication efforts of this noxious, weedy, aquatic fern. Salvinia
molesta is native to southern South America (Brazil, Paraguay, and Uruguay)
and has become widely naturalized in Africa, Asia, Australia, and New
Zealand.

This species has recently also b
from Texas to Florida, where it is
systems (Jacono, 1999). It has been declared a Fe
such, it is illegal to own, cultivate, transport, or se

een found in the southern United States,
becoming a serious threat to the aquatic
deral Noxious Weed and, as
ll. Every effort needs to be

182 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 2 (2002)

made to control its growth and spread. Salvinia molesta is sexually sterile
but reproduces rapidly by fragmentation and is capable of doubling in
volume in two to three days to form extensive mats that clog waterways and
irrigation canals, block passage, obstruct irrigation pumps, prevent light from
reaching aquatic plants, reduce the oxygen content of the water, and seri-
ously degrade the quality of the water (Thomas & Room, 1986).

Selaginella stellata Spring—This species was reported to have been first
collected in Hawaii in 1990, but an earlier collection has come to my atten-
tion that documents its presence in Akaka Falls State Park, Hawaii, on 30
August 1969 (Lehto & Lehto 16429, RSA).

Selaginella umbrosa Lemaire ex Hieron.—This species of Selaginella was
discovered at Akaka Falls State Park by Kay Lynch on 15 June 2000 (Lynch
s.n., LAM). It is found in scattered locations in the area and is not yet wide-
spread. Selaginella umbrosa, characterized by having a bright red, un-
branched, erect lower stem and a blade-like, flabellate, much branched,
bright green upper portion. It is known to be in cultivation in the nearby
Hawaii Tropical Botanical Garden. In Hawaii this species is not infrequently
found in cultivation, often grown as a ground cover. It is native to from
Mexico and Guatemala and south to Colombia. Widely cultivated, it has
naturalized in tropical areas including northern Queensland, Australia.

DISCUSSION

The Hawaiian ecosystem continues to be challenged by the invasion of
new species of pteridophytes as well as the spread of already present natu-
ralized species. In the six years since the last report on the alien ferns and
fern allies in Hawaii (Wilson, 1996), three additional species have been
found to be naturalized. Marsilea crenata is a known weed in rice paddies
in southeastern Asia. In Hawaii it has been found only in one area in a
small, cultivated taro plot. There is no information about how it came to be
introduced, but it has not been reported to be in cultivation. Its spread in
Hawaii is to be anticipated.

Salvinia molesta made its appearance in 1999 and is already well estab-
lished in different lakes on Oahu. Manual eradication efforts were begun
immediately and must be continued if its growth and spread are to be con-
trolled. Biological control projects have not been reported, but these would
involve the introduction of insects to the Islands with possible unpredictable
consequences. Moran (1992) discussed the biology of this fern as well as the
biological control methods that have been used in efforts to prevent its
continued spread.

Selaginella umbrosa is not reported to present special problems where it
is naturalized. It is widely cultivated, and its continued spread may be an-
ticipated. Cultivated plants in botanical gardens and home gardens provide a
rich source for the spread of additional species into the native ecosystem.

Lygodium japonicum is beginning to spread particularly on the wind-
ward coast of Oahu, but apparently also on the island of Hawaii. In southern

WILSON: CONTINUED PTERIDOPHYTE INVASION OF HAWAII 183

Alabama and Florida it is weedy and develops a dense growth over the vege-
tation, thereby shading the underlying plants. Control efforts should begin as
soon as possible to protect the Hawaiian flora against a similar growth pat-
tern. Azolla filiculoides is now documented to be present on the island of
Hawaii. This species is now known to be present on all of the large islands.

The Hawaiian ecosystem continues to be challenged by preridophytes as
well as seed plants, and continued efforts to control and eradicate these in-
vaders must be continued.

AACKNOWLEDGEMENTS

I wish to thank George Staples, Daniel D. Palmer, William J. Hoe, Clyde Imada, Brian Aikins,
and Brad Waters for continually keeping me apprised of new and a pteridophyte dis-
coveries in the Islands. They have notified me of new sightings and have sent me specimens,
photographs, and newspaper clippings. David Johnson very kindly sintiaca the identification
of Marsilea crenata. The herbaria of the B. P. Bishop Museum (BISH), Natural History Museum
of Los Angeles County (LAM), and Rancho Santa Ana Botanic Garden (RSA) have kindly made
their facilities and collections available to me.

LITERATURE CITED

FosBerG, F. R. 1943. Notes on ties of the Pacific islands—III. Bull. Torrey Bot. Club 70:386-387.

Hosuizakl, B. J., an . 2001. Fern Grower’s Manual, Revised and Expanded Edition.

imbe ore Porta d, ny

JACONO, ee 1999. Salvinia molesta (Salviniaceae), new to Texas and Louisiana. Sida 18:
927—

Moran, R. e 1992. The Story of the molesting Salvinia. Fiddlehead Forum 19:26—

Tuomas, P. A., and P. M. Room. 1986. Taxonomy and control of Salvinia earning ‘Natne 320:
581-5

Wacner, W. H., Jr. 1950. Ferns naturalized in Hawaii. Occas. Pap Bernice Pauahi Bishop Museum
20(8):95—-121.

Wi.son, K. A. 1996. Alien ferns in Hawai’i. Pacific Sci. 50:127-141.

ae i ee ee eee nee

Botanical Garden Libr.

Wi

5648

Authors are encouraged to submit manuscripts pertinent to pteridology for pub-
lication in the American Fern Journal. Manuscripts should be sent to the Editor.
Acceptance of papers for publication depends on merit as judged by two or more
referees. Authors are encouraged to contribute toward publishing costs; however,
the payment or non-payment of page charges will affect neither the acceptability
of manuscripts nor the date of publication.

Authors should adhere to the following guidelines; manuscripts not so prepared
may be returned for revision prior to review. Submit manuscripts in triplicate
(xerocopies acceptable), including review copies of illustrations and originals of
illustrations. After review, submission of final versions of manuscripts on diskette
(in PC- or Mac-compatible formats) is strongly encouraged. Use standard 8% by
11 inch paper of good quality, not “erasable” paper. Double space manuscripts
throughout, including title, authors’ names and addresses, short, informative ab-
stract, text (including heads and keys), literature cited, tables (separate from text),
and figure captions (grouped as consecutive paragraphs separate from figures).
Arrange parts of manuscript in order just given. Include author’s name and page
number in upper right corner of every sheet. Provide margins of at least 25 mm
all around on typed pages. Do not submit right-justified copy, avoid footnotes,
and do not break words at ends of lines. Make table headings and figure captions
self-explanatory. Use S.I. (metric) units for all measures (e.g., distance, elevation,
weight) unless quoted or cited from another source (e.g., specimen citations). For
nomenclatural matter (i.e., synonymy and typification), use one paragraph per
basionym (see Regnum Veg. 58:39-40. 1968). Abbreviate titles of serial publi-
cations according to Botanico-Periodicum-Huntianum (Lawrence et al., 1968,
Hunt Botanical Library, Pittsburgh) and its supplement (1991). References cited
only as part of nomenclatural matter are not included in literature cited. For shorter
notes and reviews, omit the abstract and put all references parenthetically in text.
Use Index Herbariorum (Regnum Veg. 120:1—693. 1990) for designations of her-

a.
Illustrations should be proportioned to fit page width with caption on the same
page. Provide margins of at least 25 mm on all illustrations. For continuous-tone
illustrations, design originals for reproduction without reduction or by uniform
amount. In composite blocks, abut edges of adjacent photographs. Avoid com-
bining continuous-tone and line-copy in single illustrations or blocks. Coordinate
ng of figures (and of tables) with order of citation in text.

sequence and numberi :
Explain scales and symbols in figures themselves, not in captions. Include a scale
and reference to latitude and longitude in each map. 3
Proofs and reprint order forms are sent to authors by the printer. Authors should
send corrected proofs to the editor and reprint orders to the printer. Authors will
be assessed charges for extensive alterations made after type has been set. |
For other matter of form or style, consult recent Issues of American Fern Jour-
nal and The Chicago Manual of Style, 14th ed. (1993, Univ. Chicago Press,
Chicago). Occasionally, departure from these guidelines may be justified. Amis
are encouraged to consult the editor for assistance with any aspect of manuscript

tion. : ee :
P pee eaees than 32 printed pages may be sent to the Editor of Pteridologia
(Memoir Editor, see cover 2).

QE I
AMERICAN a
FERN nee

J O U R N A L i Bios ate

QUARTERLY JOURNAL OF THE AMERICAN FERN SOCIETY

Differentiation of Eastern hae American Athyrium filix-femina Taxa: Evidence From
Allozymes and S
Carol Lynn *Kelloff Judith Skog, Laura Adamkewicz and Charles R. Werth 185

A New Hybrid Polypodium Provides In Cc th
scouleri and its Sympatric Conkenets TJ. Hi ildebrand, phleicena H.
Haufler, James P. Therrien, Cathy Walters and Philip Hammond 214

Comparative retiiig of — of Olfersia alata and Olfersia cervina (Dryop-
teridaceae) iceto Mendoza, Blanca Pérez-Garcia and Ram6én Riba 229

Shorter Notes
(ibe hesperium in the Wallowa Mountains of Oregon
eter F. Zika, Edward R. Alverson, Warren H. Wagner and Florence S. Wagner 239
A Binomial for the Hybrid Polypodium of Eastern North America Allison W. Cusick 240
Lycopodium lagopus New in West Virginia Joan Eiger Gottlieb 241

Marsilea mutica in Virginia
avid A. Knepper, David M. Johnson, and Lytton John Musselman 243

3,8-Di-C-arabinosylluteolin, a New Flavonoid From Pteris vittata —_ Filippo Imperato 244

Reviews
Pteridophytes of Upper Katanga (Democratic Republic of Congo)
James D. Montgomery 247

The Illustrated Flora of Illinois. Ferns. 2nd ed. R. James Hickey 248

The American Fern Society

Council for 2002

CHRISTOPHER H. HAUFLER, Dept. of Botany, University of Kansas, Lawrence, KS 66045-2016.
President
TOM RANKER, University Museum, Campus Box 265, University of Colorado, Boulder, CO tee ars
e President
W. CARL TAYLOR, 800 W. Wells St.. Milwaukee Public Museum, Milwaukee, WI 53233-1478,
Secretary
JAMES D. CAPONETTI, Dept. of Botany, University of Tennessee, Knoxville, TN 37916-1110.
Treasurer
GEORGE YATSKIEVYCH, Missouri Botanical Garden, P. O. Box 299, St. Louis, MO 63166-0299.
Membership Secretary
JAMES D. MONTGOMERY, Ecology III, 804 Salem Blvd, Berwick, PA 18603-9801.
Back Issues Curator

R. JAMES HICKEY, Botany Dept., Miami University, —— OH 45056. Journal Editor
DAVID B. LELLINGER, U.S. National Herbarium MRC-1
mithsonian Institution, Washington, DC 20560-0166. Memoir Editor
CINDY JOHNSON-GROH, Dept. of ot os wggiias Adolphus College,
800 W. College Ave., St. Peter, MN 56082- Bulletin Editor

American Fern Journal
EDITOR

R. JAMES HICKEY

Miami University, game: OH 45 45056
ph. (513) 529-6000, e-mail: hickeyrj @ muchio.edu

ASSOCIATE EDITORS
pacecesabl GASTONY 33.35. “— f Biology, Indiana University, Bloomington, IN 47405-6801
PHER H. HAUFLER .... Dept. of Botany, University of Kansas, Lawrence, KS 66045-2106
Sopeer C: ns New York Botanical Garden, Bronx, NY 10458-5126
JAMES H. PECK Dept. of Biology, University of Arkansas—Little Rock,
2801 S. University Ave., Little Rock, AR 72204

“

‘American Fern Journal” (ISSN 0002-8444) is an illustrated quarterly devoted to the general
su of ferns. It is owned by the American Fern Society, and published at The American Fern Society,
© Missouri Botanical caearye! P. O. Box 299, St. Louis, MO 63166-0299. Periodicals postage paid at
St Louis, MO, and additional e

eames — Seek issues, sa 6 months (domestic) to 12 months (foreign) after the date of issue,
ack issues should be addressed to Dr. James D. Montgomery, Ecology III, 804 Salem

Blvd. Berwick, PA 18603-9801.
Changes of address, dues, and applications for membership should be sent to the Membership

General inquiri ing ferns should be addressed to the Secret
Subscriptions ete rs to U.S.A., Canada, and Mexico; $45.00 to wea in the world (—$2.00
agency fee, if applicable); sent free to members of the American Fern Society (annual dues, $25.00 +

$7.00 mailing i beyond U.S.A., Canada, — anaes life membership, $300.00 + $140.00
a boa beyond U.S.A., Canada, and Mexic
umes are available for most a as vena issues or on microfiche. Please contact the
See sti Curator for prices and evita bili
: Send address changes to SCENES FERN JOURNAL, Missouri Botanical Garden, P. O.
Box 299, St. Louis, MO 63166-0299.
FIDDLEHEAD FORUM
The editor of the Bulletin of the Anroce Fern Society of aime contributions from members and
non-members, including miscellaneous notes, offers to a or purchase materials, personalia,
horticultural notes, and reviews of non-technical books on fi
SPORE EXCHANGE
Ms. Denia Mant, 12616 Ebbetson Ave., Downey, CA 90242-5050, is citar oo exchanged
and lists of sent on request. http://amerfernsoc.org/sporexy.htm
GIFTS AND BEQUESTS

Gifts and bequests to the Society enable it to expand its services to members and to others interested
in ferns. ak

ms other gifts are always welcomed and are tax-deductible.
hould b dd: | the Secretar ¥-

American Fern Journal 92(3):185-—213 (2002)

Differentiation of Eastern North American
Athyrium filix-femina Taxa: Evidence
From Allozymes and Spores

Caro. LyNN KELLorr’ and JupirH SkoG
Department of Biology, George Mason University, Fairfax, VA 2
US etanel Herbarium, Smithsonian Institution, NHB 166, ie uct. se 20560

LAURA ADAMKEWICZ
Department of Biology, George Mason University, Fairfax, VA 22030
CHARLES R. WERTH
Department of Biological Sciences, Texas Tech University, Lubbock, TX 79409

Asstract.—Athyrium filix-femina (Lady Fern) comprises a complex of homoploid (n = 40) taxa,
distributed over much of the northern hemisphere and extending into South America, whose
evolutionary relationships are poorly understood and whose taxonomic treatment is problematic.
The A. filix-femina complex of North America comprises as many as four taxa with overlapping
ranges and provides an especially suitable context for exploring patterns and processes of diver-
gent evolution and its taxonomic consequences in ferns. We addressed differentiation of two
eastern North American taxa distinguished on the basis of growth form, frond shape, and spore

Southern Lady Fern respectively). Although, these two taxa have been long perceived as closely
related, they have been known to intergrade and recombine to form a hybrid zone in their rela-
tively n narrow region of overlap. This a ia is supported by the data from the present study.

: LM,
SEM, and TEM, and/or allozymes (16 loci coding 10 enzymes). The two taxa exhibited highly
distinct perispore surfaces: all A. angustum individuals had papillose surfaces, whereas most
A. asplenioides individuals were rugose with a reticulum of inflated folds. Spores from the
northernmost A. asplenioides population sampled (Shirley, NJ) showed varying degrees of inter-
mediacy suggestive of introgressive hybridization with A. angustum. Levels of allozyme poly-

orphi in p tio means: P= ) = =0.1

from 0.008 to me for individual loci (0.255 across loci) with especially high values for Idh-1,
Pgi-2, Pgm-2, and Tpi-2. Hierarchical Fs; analysis indicated that differences between the two
taxa (Fxy=0.216) accounted for most allele fre equency divergence among populations

clusters each comprising populations of one taxon. Populations of A. angustum and A. asple-
nioides were joined within their clusters at $=0.938 and S=0.945 respectively, while the two
taxon clusters were joined at S=0.848. The spore and isozyme data indicate substantial diver-
gence between A. angustum and A. asplenioides, suggesting that they merit distinction at the
rank of subspecies or species. Additional study of populations in their region of sympatry is re-
quired to determine the nature and extent of hybridization.

MISSOURI BOTANICAL

’ Author for correspondence.

nav 0 4 2002
GARDEN LIBRARY

186 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 3 (2002)

Athyrium filix-femina (L.) Roth (Lady Fern) comprises a wide-ranging
complex of divergent, homoploid (n = 40) taxa of allopatric or parapatric
distribution, variously divided and treated at different ranks (species, sub-
species, or variety) by different authors. At least three taxa occur in North
America, which Butters (1917) treated as separate species ‘amply distinct
from each other,” although he considered A. filix-femina from western North
America as conspecific with the “true” A. filix-femina of Europe. Wherry
(1961) followed Butters (1917) in recognizing as distinct species the western
A. filix-femina (specifying var. sitchense Ruprecht), the northeastern
A. angustum (Willd.) Presl., and the southeastern A. asplenioides (Michx.)
A. Eaton, although noting that they “‘intergrade to such an extent as to defy
any simple classification.” Lellinger (1985) treated these taxa as subspecies
of a single globally distributed species, i.e., western A. filix-femina ssp. cyclo-
sorum (Rupr.) C. Chr., northeastern A. filix-femina ssp. angustum (Willd.)
Clausen and the southeastern A. filix-femina ssp. asplenioides (Michx.)
Hultén. More recently, Kato (1993), stating that “the delimitation and infra-
specific classification of A. filix-femina need detailed study,” treated these
taxa at varietal rank, recognizing four North American varieties: in the west,
northern A. filix-femina var. cyclosorum Ruprecht (= var. sitchense), a more
southerly A. filix-femina var. californicum Butters (which Butters, 1917 had
noted as distinctive but treated as a variety of A. filix-femina), and in the
east, northern A. filix-femina var. angustum (Willd.) G. Lawson, as well as a
southern A. filix-femina var. asplenioides (Michx.) Farwell.

hese differences in classification reflect neither strongly divergent view-
points among authors as to the nature of species versus infraspecific taxa, nor
differing insights following acquisition of compelling data. Rather, they reflect
the best judgment of individual authors in the face of a lack of critical informa-
tion as to character consistency, degree of intergradation, and genetic relation-
ships among taxa in a group noted for a high degree of variability (Schneller
and Schmid, 1982). As a contribution toward elucidating the nature of the taxa
composing the A. filix-femina complex, the present investigation addressed
the distinctness of the two parapatric taxa of eastern North America, A. filix-
femina var. angustum and A. filix-femina var. asplenioides (referred to hence-
forth informally as bare epithets angustum and asplenioides respectively).
Characters differentiating these taxa, provided by Butters (1917) and by and
large echoed in more recent treatments (Lellinger, 1985; Kato, 1993), include
rhizome habit, leaf shape, and notably color and surface features of the spores
(Table 1). The goal of the present study was to evaluate more fully the degree
to which these putative taxa are distinct by 1) characterizing fine features of the
spores using electron microscopy, and 2) surveying allozyme variation across
a north-to-south transect through the eastern portion of the range of both taxa.

MATERIALS AND METHODS

CoLLecTIONs.—All observations were made on plants collected by CLK and
CRW from 17 localities, nine A. angustum and eight A. asplenioides (Table 2).

TABLE 1. Comparison of morphological features of the two putative wi ha North American taxa in the Athyrium filix-femina complex. Compiled
from taxon descriptions in Butters (1917), Lellinger (1985), and Kato (1993).

Character

A. angustum

A. asplenioides

Rhizome orientation
cales:

Stipe length relative to
lamina

Lamina shape
Frond base

Frond apex
Widest portion of lamina

Pinna attachment
Pinna shape

nna apex
Sterile v. fertile frond
Pinnules of fertile fronds
Indusium length
Indusium margin*
Sporangial stalks
Spore color

Spore surface
Mean spore dimensions

erect or ascending, more condensed
sie lanceolate,
=—10 * 1.5-2 mm,
est to dark brown
up to half the lamina length

elliptic or rhombic
gradually tapered to an acute
obtuse base
acute to acuminate
near or just below middle
short-stalked or sessile
pense i usually widest
ee nd not parallel-sided
e to ac
pestien acorn aah the segments

narrowly lanceolate and acute
tending to be shorter than in asplenioides,
to 1.1 mm

irregularly dentate, and/or ciliate with
eglandular hai

bearing glandular hairs or less often
sec ondary sporangia

yellowish

sparsely agg

ascending to creeping, more extended
lanceolate, 3-9 X 2-3 mm, bronze to light
brown or brown

equaling lamina length

narrowly deltoid lanceolate, ovate-lanceolate to lanceolate
slightly reduced and truncate at base

acute, acuminate, or more-or-less caudate
d
oblong-lanceolate to lanceolate, nearly parallel-sided
te
not tending toward dimorphism
oblong or linear-oblong and obtuse
tending to be longer than in angustum, up to 1.3 mm
ciliate aes glandular or non-glandular hairs
as long as indusial width
Gecunaats bearing glandular hairs
brownish-yellow to dark-brown or black

wrinkled or reticulated exospore, sometimes nigrescent
36,0 % 25:5

“ Descriptions of the indusium margin in the two taxa are inconsistent in the literature. The table entry is a tentative consensus.

WOIYWAHLVY NVOMANV HLYON NYSLSVS “TY LY AAOTTAN

Z8L

TABLE 2, Collection localities for Athyrium filix-femina populations examined in this study. Localities are ordered from north to south.

Population Collector(s) and Year No. of isozyme No. of spore
Designation collection number Collected Locality samples samples
A. filix-femina var. ne
Mt. Sainte Hilaire C. Werth s 1997 Province du reeiie bakin hiking 26 th
trail through for
Mont Ste.-Hilai
Mast Landing EB. Kass sD. 1987 Maine, Soubeiead County, _ 0 4
Landing Sanctuary, Freepor
North Hudson C. A pai and 1997 New York, Essex County, Cie 26 10
C. Caplen s. n. woods and stream

bank, 11.3 mi W of North Hudson
and I-87, exit 387
to Blue Ridge

Schenevus C. Werth and 1997 New York, Otsego County, woods 11 it
C. Caplen s. n. near edge E side I-88,
ear Schene
Barnet C. Werth, D. Conant, 1997 Vermont, Caledonia County, ditches 72 20
and C. Caplen s. n. and woods edge along dirt
road in vicinity of Conant farm,
near Barnet
Ithaca CL Kelloff 560 1988 New York, Tompkins County, fen 0 20

Binghamton CL Kelloff 559 1988 New York, Broome County, floodplain 0 20
of Chenango R.
1.5 miles N of Binghamton
Ralston-1 CL Kelloff 569 1988 & Pennsylvania, Lycoming County, 20 58
1995 flood plain along Route
14, 7 miles north of Ralston
Ralston-2! CL Kelloff 570 1988 Pennsylvania, Lycoming County, ax 78

ood plain along Route
14, 3 miles south of Ralston

881

(z00Z) € MWASINON 26 ANNTIOA “TVNYNOlL NASA NVOMSNV

TABLE 2. Continued.

Population Collector(s) and Year No. of isozyme No. of spore
Designation collection number Collected Locality samples samples

A. filix-femina var. asplenioides
Shirley CL Kelloff 1289 1997 New Jersey, Elmer County, floodplain 40 27
along County Road
611, 1.5 miles east of jct. NJ State
Route 77, in Shirle

Breathed Mountain CL Kelloff 25 1987 = Virginia, Lena County, 0 6
eathed Mountains
George Mason CL Kelloff 101 1988 ie Fairfax ae creekside on 0* 30
eorge Mason University campus
n Fairfax
Hopewell CL Kelloff 565 1988 & Virginia, Fauquier County, floodplain 30 42
1995 along Route 629, 3.3 miles N
e 601 near Hopewell
Mountjoy Store CL Kelloff 564 1988 & Virginia, Stafford County, bog at 30 63
1995 ountjoy Store, se ms
of routes 611 a
Pond Drain C. Werth, 1997 & Virginia, Giles aera north-facing 50 0
E. Potter, and 1998 wooded slope along rd.
K. Sciaretta s. n. to White Pine Lodge and above

2}
ain

Pond Drain, NW
rt. 613 and Mountain Lake

Pipevine Hollow CL Kelloff s. n. 1987 Tennessee, Davidson County, 0 tS
Pipevine Hollow
Sandy Run Swamp CL Kelloff 1285 1997 North Carolina, Onslow County, creek 32 18

bank and roadside ditch E side of
Sandy Run Swamp along Haws
Run Road, just W of Padgett,

ca. 60 air km NE of Wilmington

"Initial isozyme data were reported for 78 individuals from Ralston-2 (Kelloff 1990), but this population was not resampled during 1995 and iso-
zyme data are not included in the present report (see text for further explanation).

WOIYAHLV NVOIAWNV HLYON NYALS Va “TV LY JAOTTEN

681

190 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 3 (2002)

Of these, seven were examined only for spores, while for ten (five of each
taxon), population samples for isozyme electrophoresis were obtained as well.
At most localities, plants were collected in a transect and individuals num-
bered successively. Voucher specimens for localities collected by CLK are
deposited at the George Mason University Herbarium (GMUF), those collected
by CRW at the E. L. Reed Herbarium of Texas Tech University (TTC).

Spore MorPHoLocy.—Spores were collected from fresh samples gathered from
16 of the localities listed in Table 2. The spores were examined for surface
characteristics of the perispore using light microscopy (LM; 409 individuals),
and scanning electron microscopy (SEM; one to two individuals from ten
populations, 13 from the Shirley, NJ population). For exine ultrastructure,
transmission electron microscopy (TEM) was used to examine four individ-
uals, 2 of each taxon. For LM, spores were mounted by gently tapping a
pinnule bearing mature sporangia over a glass slide and then adding a few
drops of Permount’~ mounting medium and a glass coverslip. Spores were
examined under a standard brightfield optic system.

For SEM, air dried spores were affixed to glass coverslips with ethanol,
mounted on SEM stubs, and sputter coated with gold or carbon/gold palla-
dium using the Hummer VII sputtering system. The spores were then exam-
ined using either a Hitachi S-530 or a LEO 440 scanning electron
microscope. From each SEM sample at least 100 spores were examined.

For TEM, each spore sample was embedded in 1% agar as a pellet and
fixed in sodium veronal-acetate buffered potassium permanganate (Dawes,
1971). The sample pellet, cut up into 1.0 mm cubes, was dehydrated in a
graded series of ethanol. After dehydration, the spores were infiltrated with
a series of propylene oxide and Spurrs plastic embedding medium in a
labeled BEEM capsule and polymerized in an oven at 70°C for 16 hours. The
blocks were then thin-sectioned with a Dupont diamond knife, the sections
picked up on #200 mesh grids and examined with a JEOL 100C transmisson
electron microscope. Post-staining of the spore sections was not necessary
when fixed with the buffered potassium permanganate (Dawes, 1971).

IsoZYME ELECTROPHORESIS.—Fronds to be analyzed electrophoretically were
collected in the field and kept refrigerated in plastic bags until homogenized
within a few days. For each sample, approximately 1 g leaf tissue was
ground using eight drops of ‘‘microbuffer’’ (Werth, 1985) enhanced with 5%
PVP-40T and 1% 2-mercaptoethanol. The tissue was ground in a porcelain
spot plate using a small test tube as a pestle and sand to facilitate grinding.
The extract was either absorbed onto filter paper wicks and used immedi-
ately or stored at —78°C for up to four months before being thawed and
absorbed onto wicks. Starch gel concentration varied from 11% to 13.5%
depending on the properties of individual starch lots.

The following eleven enzymes were analyzed using the buffer systems in-
dicated: on lithium hydroxide (Selander et al., 1971)—glutamate oxaloace-
tate transaminase (GOT), hexokinase (HK), and leucine aminopeptidase
(LAP); on system number 6 (Soltis et al., 1983)—phosphoglucose isomerase

KELLOFF ET AL.: EASTERN NORTH AMERICAN ATHYRIUM 191

(PGI), phosphoglucomutase (PGM), and triose-phosphate isomerase (TPI); on
morpholine-citrate pH 8.2 (Werth, 1991)—aldolase (ALD), isocitrate dehydro-
genase (IDH), malate dehydrogenase (MDH), 6-phosphogluconate dehydro-
genase (6PGDH), and shikimate dehydrogenase (SKDH). Gels were run until
the bromophenol blue marker migrated 10 cm for lithium-hydroxide and
number 6, or 12 cm for morpholine-citrate. Staining followed Werth (1985)
as modified for the zymecicle technique (Werth, 1990). Isozyme banding pat-
terns were readily interpreted as allelic variation for single gene loci from
whole leaf extracts (cf. Gastony and Darrow, 1983) using standard principles
(Wendel and Weeden, 1989; Murphy et al., 1990). Data were analyzed using
the BIOSYS-1 computer program (Swofford and Selander, 1981), entered into
the program as individual multilocus genotypes. Occasionally, sets of two
or three successive samples from the field possessed identical genotypes. As
Athyrium filix-femina is known to occasionally form clones up to nearly
20 m in extent (Sciaretta et al/., ms), such sets were assumed to comprise sin-
gle genets and each set was represented by only a single entry in the data set.

Initial isozyme data were obtained from four populations, two of each
taxon (data reported in Kelloff, 1990). During 1995, three of these popula-
tions (Ralston-1, PA, Hopewell, VA, and Mountjoy Store, VA) were revisited
and re-sampled, and an expanded isozyme data set obtained; this was aug-
mented by sampling of additional populations during 1997 and 1998. The
initial data were largely consistent with the more recent results; however,
only the latter are reported here to avoid including inadvertently re-sampled
individuals.

RESULTS

Spores, Light Microscopy.—Observation of spores through the light micro-
scope (LM) revealed features largely consistent with previous reports of dif-
ferences between the two taxa (Butters, 1917; Lugardon, 1971; Schneller,
1989). Spores of both A. angustum (Fig. 2a, b) and A. asplenioides (Fig. 3a, b)
were monolete with an unbranched longitudinal proximal laesura, ovoid to
ellipsoidal in polar view, and possessed a perispore (terminology of Erdtman,
1969). Although most of the surface details were obscure when viewed under
LM, the difference between the two taxa was readily seen in their outline. The
spores of A. angustum had a somewhat smooth surface (Fig. 2a, b), while those
of A. asplenioides had an uneven surface (Fig. 3a, b).

Spore color has been described as differing between A. angustum with yel-
lowish spores, and A. asplenioides with yellowish-brown to blackish spores
(Table 1). This characterization evidently is based on appearance of mass
spores under reflected light as viewed unaided or with a dissecting micro-
scope. When viewed under LM using transmitted light, spores of both taxa
appeared yellowish, with the exception of occasional A. asplenioides spores
that displayed an especially dark or blackish reticulation overlaying the yel-
low surface of the spore.

192 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 3 (2002)

hirley
Breath Mountain
topewell

~~- George Mason
~~~Mountjoy Store

_— Sandy Run Swamp

A. angustum
[| A. asplenioides

0 500 Miles

Fic. 1. Map showing overlapping ranges of two Athyrium taxa, angustum and asplenioides, in
eastern North America and location of populations studied.

SCANNING ELECTRON Microscopy.—Observations of spores under SEM were
consistent with those of Tryon and Lugardon (1991) and Schneller (1989) in
revealing that the perispore of A. angustum (Fig. 2c—f) differs markedly from
that of A. asplenioides (Fig. 3c—f). The perispore of A. angustum was papil-
lose, its surface completely and evenly covered with minute verruciform
(i.e., wart-like) projections. On some spores of A. angustum, thin, irregular
laminae of perispore material appeared as occasional “flakes.” In contrast,
the perispore of A. asplenioides was rugose, characterized by a network of
muri (i.e. ridges/walls) that meandered and joined to surround lacunae. The
surface of the lacunae completely lacked the projections as in A. angustum.

TRANSMISSION ELECTRON Microscopy.—Under TEM the similarities and differ-
ences between the structural and sculptural features of the spore wall of the
two taxa were clearly distinct. Both A. angustum and A. asplenioides (Figs.
2f and 3f) possessed a thick external exospore (Ee). Although not evident in
these sections, the fern exospore is composed of feuillettes, i.e., fused plates
(Lugardon, 1976, 1979; Tryon, 1986). TEM revealed that the sculptural ele-
ments composing the surface details seen under SEM were not derived from

KELLOFF ET AL.: EASTERN NORTH AMERICAN ATHYRIUM

Fic. 2. Spores of rey cate angustum.
of perispore. C, D, E. Spores

A, B. Spores viewed under LM. Note smooth aq
viewed under SEM. Note papillate perispore sur
viewed under TEM. Note low perispore (p), external exospore (ee), and internal exospore

ace. F oe wal

(ei).

the exospore, as for example in the spores of Trichomanes or Protomarattia
(Tryon, 1986), but instead from the perispore, the outer stratum circumjacent
to the external exospore. The perispore of both taxa fit tightly to the external
exospore. This stratum in the spores of A. angustum (Fig. 2f) was a shallow,

pearance

193

7—el lu

Spores of Athyrium asplenioides. A, B. Spores viewed under LM. Note rugose, reticulate
appearance of perispore. C i. Spores viewed under SEM
perispore surface. F. Spore wall viewed under
exospore (ee), and faint internal exospore (ei).

Note reticulating inflated folds of
TEM. Note inflated fold in perispore (p), external

fairly uniform layer. In A. asplenioides spores, what appeared to be muri
under the SEM (Figs. 3c—e) were more clearly seen under TEM to be inflated
folds in the perispore because the columellae, characteristic of true muri
were lacking (Fig. 3f). Beneath the external exospore on A. angustum was

AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 3 (2002

KELLOFF ET AL.: EASTERN NORTH AMERICAN ATHYRIUM 195

a thin endospore or internal exospore (Ei), a layer that is deposited shortly
before germination of the spore (Tryon, 1986). A remnant of this layer was
also visible in A. asplenioides.

TAXON FIDELITY OF SPORE MORPHOTYPES AND EVIDENCE FOR HysRIDIZATION.—AI]
populations of A. angustum sampled for this study exhibited the smooth,
micropapillate spore morphotype, and all populations of A. asplenioides
except one exhibited the rough, rugose spore morphotype. Reliance on light
microscopy, which may fail to reveal the subtle detail of the perispore, and
on color, which varies in A. asplenioides despite the overall similarity of
perispore in this taxon, may have led to some confusion in the previous lit-
erature. For example, Liew (1971) concluded that most individuals of both
A. angustum and A. asplenioides produced both kinds of spores, entirely in-
consistent with our observations.

The exceptional A. asplenioides population was the northernmost popula-
tion sampled, Shirley, NJ. At this site the spores had perispore sculpturing
characteristics of both A. angustum and A. asplenioides. While none of the
plants from this site possessed A. angustum-type spores, spore arrays of the
13 plants examined under SEM possessed unique and variable morphologies
that suggested an influence from A. angustum (Fig. 4). The greatest degree of
A. angustum influence was exhibited by six individuals in which prevailing
perispore morphotype was very irregular in appearance, combining an array
of sculptural elements that were difficult to characterize (Fig. 4a, b). These
included irregularly shaped papillae, briefly-extending isolated muri, and a
preponderance of irregular small lamellae reminiscent of the ‘“‘flakes’’ seen
in A. angustum. This prevailing morphotype is readily interpreted as inter-
mediate between the highly distinct morphotypes of A. angustum and A. as-
plenioides, seeming to average their disparate surface features and indicating
that these individuals may be hybrids between the two taxa. Many of the
highly variable spores of these putative F1 hybrids were not entirely inter-
mediate, but rather showed tendencies toward either the smooth A. angus-
tum type or the rough A. asplenioides type (Fig. 4a, c-g). Spores that fully
resembled parental types were discovered in the spore arrays of these six in-
dividuals, although these extremes were decidedly rare. The spores of these
putative hybrids were not abortive, as is typically the case in interspecific
fern hybrids, but rather appeared normal, i.e., well-filled, and were found to
be viable as evidenced by successful germination (unpublished data). More-
over, the seven other individuals in the Shirley, NJ, population examined
under SEM possessed spore arrays exhibiting various degrees of intermediacy
but strongly skewed toward the A. asplenioides morphology. These are
hypothesized to include backcrosses between first-generation hybrids and
A. asplenioides (two individuals) and later generation backcrosses (five indi-
viduals).

ALLOZYMES, GENERAL.—The eleven enzymes assayed were coded by 17 inter-

pretable loci of which only one (Pgi-1) was invariant across all ten popula-
tions. The remaining sixteen loci (Ald, Got, Hk, Idh-1, Lap, Mdh-1, Mdh-2,

196

AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 3 (2002)

: esi snioides viewed
> variable

Athyrium angustum and ,
id. A. Array of spore

B. Close-up of ‘“‘fla

Fic. 4. Spores of putative hybrids between
SEM. A-G, spores of putative first-generation hybr lote
morphology and ap exomean of “flaky’’ intermediates. ky”
spore morphotype. C-G. Examples . spore morphotypes occurring in putative F-1 hybrid,
ing from A mecaestics tee: ‘e to; = oo s-like (G). H— spores of putative backcross
rt stween Fs hybrid and A. asple ots 2s. H. Array of spore bie Sa exhibited by the puta-
ve backcross. Note prevalent of A. asple aaa s-like spores. I. Close-up of A. asplenioides-like
spore from backcross. Note irresolute nature of inflated folds and iach ncy tomeed ‘flakiness’
spore morphotypes putative backcross. Note variation from
) A. asplenioides-

under
intermediate
vary-

J-M. Examples of occurring in
“flaky” intermediate (J) tc ike (M).

june

KELLOFF ET AL.: EASTERN NORTH AMERICAN ATHYRIUM 197

Mdh-3, Mdh-4, Pgi-2, Pgm-2, 6-Pgd-1, 6Pgd-2, Skdh, Tpi-1, and Tpi-2) were
variable in at least one population, and the frequencies of the alleles were
computed (Table 3). There was a strong tendency for all populations to share
the more common alleles and some of the less frequent ones as well. Most of
the loci were weakly polymorphic, a single allele predominating in all popu-
lations of both A. angustum and A. asplenioides. In contrast, at the four
most polymorphic loci (Idh-1, Pgi-2, Pgm-2, and Tpi-2), population allele fre-
quencies were similar within A. angustum and A. asplenioides respectively,
but strikingly different between the two taxa (Table 3).

The most contrasting locus was Idh-1 for which alleles Idh-1° and Idh-1°
prevailed in A. angustum at frequencies ranging from 0.545 to 0.676 and
0.296 to 0.455 respectively, while allele Idh-1“ was the most frequent in
A. asplenioides with frequencies ranging from 0.828 to 0.950. At Pgi-2 both
taxa shared Pgi-2" as their most frequent allele; however, Pgi-2° was repre-
sented in all A. angustum populations at much higher frequencies, ranging
from 0.154 to 0.450, than in populations of A. asplenioides for which the fre-
quency of this allele ranged from 0.000 to 0.106. At Pgm-2, both taxa shared
Pgm-2° and Pgm-2° as principal alleles, the former prevailing in A. angus-
tum populations at frequencies ranging from 0.856 to 1.000, the latter pre-
vailing in A. asplenioides populations at frequencies ranging from 0.580 to
0.857. At Tpi-2 both taxa shared three prevalent alleles, with Tpi-2* being of
substantial frequency in all populations. Allele Tpi-2° occurred at high fre-
quencies, ranging from 0.355 to 0. 550, in A. angustum populations whereas

pi-2° was infrequent in this taxon, occurring at frequencies from 0.000 to
0.100. Conversely, Tpi-2° often was the most frequent allele in A. asple-
niordes populations, occurring at frequencies from 0.316 to 0.606, while Tpi-
2” was rarer, occurring at frequencies from 0.000 to 0.255. The contrasting
allele frequency trends for these four loci were highly consistent among pop-
ulations within each taxon. Exceptions to these frequencies trends were in
the northernmost sampled A. asplenioides population, Shirley, from south-
ern New Jersey for the characteristically A. angustum alleles Idh-1" and Pgi-
2° and likewise for Tpi-2” in the highest elevation (1300 m) population of
A. asplenioides sampled, Pond Drain, from the mountains of southwestern
Virginia. Moreover, five of the six Shirley, NJ, individuals hypothesized to
be first-generation hybrids on the basis of spore morphology were hetero-
zygous for A. angustum and A. asplenioides 1 marker alleles for at least three
of the four most divergent loci, i.e., Idh-1°", Pgi-2"°, Pgm-2°© (scored for
only two of these individuals) and Tpi-2°°. No other individuals in the entire
data set possessed this genotype combination.

GENETIC VARIATION.—Genetic variation was quantified for each population
and the species as a whole by computing three standardly used indices. Val-
ues for percent loci polymorphic (P) ranged from 23.5% to 47.1%, with a
mean of 36.48%; for mean number of alleles per locus (A), the range was 1.5
to 2.5, mean 1.97; and for mean expected heterozygosity (Hg), the range was
0.112 to 0.147, mean 0.129 (Table 4). The mean values for these indices in

TABLE 3. Allele frequencies for 17 isozyme loci in ten populations of Athyrium filix-femina in eastern North America.
angustum asplenioides
Mt. Ste. Barnet North Schenevus’ Ralston Shirley Mountjoy Hopewell Sandy Run Pond Drain

Locus Allele Hilaire QE VT Hudson NY NY PA NJ Store VA Vv Swamp NC V.

Ald A. 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 0.969
B 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.021
G 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.010
(N) 9 i 1 48

Got A 0.000 0.033 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
B 1.000 0.951 1.000 1.000 1.000 0.955 1.000 0.917 1.000 0.980
C 0.000 0.016 0.000 0.000 0.000 0.045 0.000 0.083 0.000 0.010
D 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.010
(N) 5 6

Hk* A 0.000 0.000 0.000 0.000 0.150 0.000 0.017 0.000 0.000 0.016
B 0.960 0.965 1.000 0.955 0.800 0.868 0.967 0.967 0.893 0.952
C 0.040 0.023 0.000 0.045 0.050 0.105 0.017 0.017 0.089 0.032
E 0.000 0.012 0.000 0.000 0.000 0.025 0.000 0.017 0.000 0.000
F 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.018 0.000
(N) 25 43 19 11 10 19 30 30 28 ot

Idh-1* A 0.019 0.007 0.000 0.000 0.000 0.828 0.950 0.833 0.859 0.920
B 0.596 0.676 0.596 0.545 0.650 0.172 0.017 0.133 0.094 0.060
C 0.385 0.296 0.385 0.455 0.350 0.000 0.000 0.000 0.000 0.020
D 0.000 0.021 0.019 0.000 0.000 0.000 0.000 0.000 0.000 0.000
F 0.000 0.000 0.000 0.000 0.000 0.000 0.033 0.033 0.047 0.000
(N) 26 11 20 32 30 30 32 50

Lap A 0.000 0.000 0.000 0.000 0.000 0.018 0.000 0.000 0.000
B 0.100 0.038 0.063 - 0.175 0.000 0.036 0.000 0.031 0.016
C 0.850 0.885 0.750 - 0.550 1.000 0.946 1.000 0.938 0.935
D 0.050 0.058 0,188 - 0.275 0.000 0.000 0.000 0.016 0.032
E 0.000 0.019 0.000 - 0.000 0.000 0.000 0.000 0.016 0.000
F 0.000 0.000 0.000 - 0.000 0.000 0.000 0.000 0.000 0.016
(N) 0

86L

(z00z) € MASINON 26 ANNTIOA “TYNUNO! NUdA NVOIMSNV

TABLE 3. Continued.
angustum asplenioides
Mt. Ste. Barnet North Schenevus Ralston Shirley Mountjoy Hopewell Sandy Run Pond Drain

Locus Allele Hilaire QE VT Hudson NY NY PA NJ Store VA VA Swamp NC VA

Mdh-1 A 0.981 0,950 0.942 1.000 1.000 0.939 0.967 1.000 0.969 1.000
B 0.019 0.020 0,058 0.000 0.000 0.061 0.033 0,000 0.000 0.000
Cc 0.000 0.030 0,000 0.000 0.000 0.000 0.000 0,000 0.031 0.000
(N) 26 50 26

Mdh-2 A 1,000 0.990 1.000 1.000 1.000 1,000 0.933 0.950 1.000 1.000
B 0,000 0,010 0.000 0.000 0.000 0.000 0.067 0.050 0.000 0.000
(N) 11 20 33 15 20

Mdh-3 A 1.000 1.000 1.000 1.000 1.000 0.985 0.967 1.000 0.953 1.000
B 0,000 0.000 0.000 0.000 0.000 0.015 0.033 0.000 0.047 0.000
(N) 33 50

Mdh-4 A 0.960 0.920 0.962 1,000 1.000 1.000 0.967 0.950 1.000 1,000
B 0.040 0.070 0.038 0,000 0,000 0.000 0.033 0.050 0.000 0.000
C 0.000 0.010 0.000 0,000 0.000 0.000 0.000 0.000 0.000 0,000
(N)

Pgi-1 A 1.000 1.000 1.000 1,000 1.000 1.000 1.000 1.000 1.000 1,000
(N) 11 10 13 25 24 27

Pgi-2 A 0.000 0.000 0,000 0,000 0.000 0.015 0.033 0.000 0.000 0.000
B 0.000 0.014 0.019 0,000 0.000 0.000 0.017 0.133 0.000 0.000
Cc 0.019 0.00 0,000 0,000 0,000 0.000 0.000 0.017 0.000 0.010
D 0.000 0.000 0.000 0.000 0.000 0.045 0.100 0.000 0.078 0.010
E 0.692 0.715 0.808 0.500 0.550 0.742 0.800 0.850 0.906 0.930
F 0.019 0.014 0.000 0.000 0.000 0.045 0.017 0.000 0.000 0.020
G 0.269 0.250 .154 0.409 0.450 0.106 0.033 0.000 0.016 0.030
H 0,000 0,000 0.000 0,000 0.000 0.045 0.000 0.000 0,000 0.000
I 0,000 0,000 .019 0.000 0.000 0.000 0.000 0.000 0.000 0.000
J 0,000 0,000 0.000 0.091 0.000 0.000 0.000 0.000 0.000 0.000
(N) 26 te 26 33

Pgm-2 A 0.024 0.000 0,000 0.000 0.000 0.000 0.000 0.083 0.000 0.020
B 0,857 0.856 0.906 1.000 0.975 0.143 0.300 0.100 0.222 0.390
C 0.119 122 0.094 0.000 0.025 0.857 0.683 0.800 0.704 0.580

WOIYAHLY NVOTYIWNV HLYON NYALSVS “TV LY AAOTIAN

661

TABLE 3. Continued.

0072

angustum asplenioides
Mt. Ste. Barnet North Schenevus’ Ralston Shirley Mountjoy Hopewell Sandy Run Pond Drain

Locus Allele Hilaire QE VT Hudson NY NY PA NJ Store VA VA Swamp NC VA
D 0.000 0.022 0.000 0,000 0.000 0.000 0.017 0.017 0.074 0.000
E 0,000 0,000 0.000 0,000 0.000 0.000 0.000 0.000 0.000 0.010
(N) 21 45 16 i ba | 20 21 30 30 27 50

6Pgd-1 A 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
B 1.000 1.000 1.000 1,000 1.000 0.957 1.000 1.000 1.000 1.000

0,000 0.000 0,000 0.000 0.000 0.043 0.000 0.000 0.000 0.000

(N 11 38 11 6 23

6Pgd-2 A 0,000 0.000 0.000 0,000 0,000 0.015 0.033 0.017 0.017 0.021
B 1.000 1.000 1.000 1.000 1.000 0.985 0.950 0.967 0.983 0.947

0.000 0,000 0.000 0.000 0,000 0.000 0.017 0.017 0.000 0.021

D 0.000 0.000 0.000 0.000 0.000 0,000 0.000 0.000 0.000 0.011
(N 26 69 21 11

Skdh A 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.067 0.000 0.000
B 0.981 0.955 0.942 0.955 1,000 1,000 1.000 0.933 1.000 0.990
Cc 0.019 0.036 0.058 0.045 0.000 0,000 0.000 0.000 0.000 0.010
D 0.000 0.009 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
(N) 26 56 26 11 20

Tpi-1 A 0.947 1.000 0.981 0.900 1.000 0.970 1.000 1.000 0.984 1.000
B 0.053 0.000 0.000 0.100 0.000 0.030 0.000 0.000 0.016 0.000
Cc 0.000 0.000 0.019 0.000 0.000 0,000 0.000 0.000 0.000 0.000
(N) 19 47 26 5 33

Tpi-2* A 0.579 0.601 0.404 0.400 0.450 0.318 0.440 0.420 0.406 0.429
B 0.368 0.355 0.538 0.500 0.550 0.076 0.100 0,000 0.063 0.255
C 0.053 0.036 0.058 0.100 0.000 0.606 0.460 0.580 0.531 0.316
E 0.000 0.007 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
(N) 19 69 26 5 10 a0

(z00Z) € WAAINAN 26 SWN'IOA *TVNYNOL NYA NVORANV

KELLOFF ET AL.: EASTERN NORTH AMERICAN ATHYRIUM 201

these eastern North American Athyrium populations were somewhat greater
than the means for ferns (P=36.0%, A=1.65, H=0.109) obtained by averaging
across 32 taxa (Li and Haufler, 1999), which are in turn similar to the means
for angiosperms (P=34.2%, A=1.53, Hp=0.113; Hamrick and Godt, 1989).
Mean values across A. angustum populations for P (34.12%) and A (1.84)
were slightly lower than those for A. asplenioides (mean P=38.84%, mean
A=2.1), while the mean value in A. angustum for Hy=0.138 was greater than
that of 0.119 in A. asplenioides. This indicates that while A. asplenioides
possesses greater allelic diversity, frequencies of alleles within loci are more
evenly distributed in A. angustum populations.

Values for mean observed heterozygosity (Ho) were also computed for
comparison to Hg. Values of Ho were very similar to (although tending to be
slightly greater than) those for Hg, suggesting a tendency toward random
mating.

COMPARISON TO HARDY-WEINBERG.—Genotype proportions for each polymorphic
locus in each population were compared to Hardy-Weinberg expected values
by computing the fixation index F, and the statistical difference of F from 0
was evaluated using the chi-square test, with pooled genotype classes if the
number of alleles exceeded 2 (Table 5). The test was considered valid if
two of the three genotype classes were represented by expected values > 5.
Of 42 validly tested loci, 40 conformed to Hardy-Weinberg expectations.
Moreover, 45 out of 51 non-valid tests also indicated conformance to Hardy-
Weinberg values, despite the tendency for non-valid tests to indicate false
non-conformance due to low expected values. Thus, mating was inferred to
approximate random mating via predominant outcrossing between gameto-
phytes, as appears to be the general case in most ferns (Soltis et al., 1988)
including Athyrium species (Schneller, 1979).

GENETIC RELATEDNESS OF POPULATIONS AND TAxXA.—The degree of genetic diver-
gence among populations (Table 6) was quantified by computing F-statistics
(Wright, 1965, 1978), including hierarchical analysis (Wright, 1978). Values
for Fsy computed among all populations varied among loci from a low of
0.019 for 6Pgd-2 to high values of 0.468 for Pgm-2 and 0.460 for Idh-1,
the latter two loci having exhibited the greatest allele frequency difference
between A. angustum and A. asplenioides. The value across loci of Fs;r=
0.255 indicated very substantial differentiation among populations. This val-
ue is very high in comparison to other fern species examined, exceeding, for
example, computed values for mean Fs of 0.024 among populations of Poly-
stichum munitum ranging from Oregon to Idaho (Soltis et al., 1987), Fs;=
0.152 among Pteridium aquilinum populations ranging from Massachusetts
to Florida (Speer et al., 1998), and Fs;=0.100 to 0.248 in various species of
Dryopteris ranging widely across eastern North America (Werth, ms.).

To evaluate the contribution of differences between A. angustum and A.
asplenioides to overall population differentiation, hierarchical F-statistic
analysis (Wright, 1978) was carried out (Table 6). For most individual loci,
as well as for the combined values across loci, the variance between the two

TaBLe 4. Estimates of genetic variation at 17 loci in ten populations of Athyrium filix-femina s. 1. (standard errors in parentheses).

Mean sample size

Mean no, of alleles

t loci

Mean heterozygosity

ardy-Weinberg

H
Expected (Hp)

aa per locus per locus (A) Re asia (P) Observed (H,)
A. angus
Mt. Ste. a QE 21.4 (1.4) 1.9 (0.2) 35.3 0.150 (0.056) 0.140 (0.047)
Barnet, VT 52.5 (3.4) 2.5 (0.3) 41.2 0.134 (0.042) 0.141 (0.042)
North Hudson, NY 21.7 (1.5) 1.8 (0.2) 41.2 0.133 (0.046) 0.136 (0.047)
Schenevus, NY 9.1 (0.8) 1,5:(0,2) 2a.0 0.145 (0.071) 0.126 (0.055)
Ralston, PA 15.5 (1.4) 1.5 (0.2) 29.4 0.150 (0.061) 0.147 (0.057)
ean + 1.84 34.12 0.142
A. asplenioides
Shirley, NJ 28.2 (2.0) 2.0 (0.3) 35.3 0.127 (0.041) 0.127 (0.040)
Mountjoy Store, VA 22.8 (2.1) 2,2:(0:3) 41.2 0.133 (0.052) 0.123 (0.042)
Hopewell, VA 25.7 (1,8) 1.9 (0.2) 47.1 0.125 (0.038) 0.118 (0.036)
Sandy Run Swamp, NC 28.9 (1.5) 2.0°(0.2) 85.0 0.104 (0.039) 0.117 (0.041)
Pond Drain, VA 44.5 (2.3) 2.4 (0.3) 35.2 0.118 (0.049) 0.112 (0.046)
ean 29.62 ypil 38.84 0421 0.119
Mean across all 26.83 1.97 36.48 0.131 0,129

populations

* A locus is considered polymorphic if the frequency of the most common allele does not exceed 0.95.

> Unbiased estimate (Nei, 1978

c0Z

(z00Z) € MAGWON 26 ANN TIOA “TVNUNOL NSA NVOMENV

TABLE 5. ye 280 of se fixation index F as a measure of the conformance of population genotype ratios to those saa under Hardy-Weinberg
ulation, and tested for statistical difference from 0 (i.e
ci with more than two alleles and therefore more than three genotypic
ts were se ang valid only if two of the three genotypic classes were represented by an expected values > 5. Results fro
ackets. Values marked as ‘“‘ns’ be phen ne sane ‘a eae to Hasty eres expected proportions (p >

equilibriu

classes, Chi-square tes

F was c

omputed for each polymorphic locus in each popu
Weinberg cone using the chi-square test. Pooling was carried out for loc

non-valid tests are provided in
and therefore are not statistically different from 0. Values marked w

conformance to Hardy-

< 0.05 (one asterisk),

0.05),

p = 0.01 (two asterisks), or p < 0.001 (three asterisks).
Mt. Ste. Mountjoy Sandy Run
Hilaire Barnet N Hudson Schenevus _ Ralston Shirley e Hopewell Swamp Pond Drain
Locus QE VT NY NY PA NJ VA VA

Ald [—0.025 ns]
Got 040 {[—0.048 ns] —0,091 ns {[—0.015 ns]
Hk {[-0.042 — {—0.028 ns] —0.048 ns} [0.403 ns] —0.124ns [-—0.026 ns] [—0.026 ns] 0.266 ns {—0.039 ns]
Idh —0.00 102 ns 0.148 ns {0.267 ns] —0.208 ns [0.653***] -—0.163 ns 0.002 —0.070 ns
Lap [10, = on —0.087 ns 0.216 ns 0.154 ns {[—0.043 ns] [0.216***] [-—0.046 ns]
Mdh-1 {—0.020 ns .376** [-0.061 ns] {[-0.065 ns] [—0.034 ns] [-0.032 n
Mdh-2 {-0.010 n [-0.071 ns] [—0.053 ns]
Mdh-3 [-0.015 ns] [—0.034 ns] [0.650***]
Mdh {1.000***]} 0.192 ns [—0.040 ns {[—0.034 ns] [—0.053 ns]
Pgi-2 —0.375 0.021 ns -—0.190 ns [—0.266 ns] —0.212 n —0.054 ns 0.136 156 ns —0.088 ns —0.048 ns
Pgm-2 0.240 ns 0.120 ns [—0.103 ns] {—0.026 ise -0.167 ns —0.280 ns 0.028 ns 0.259 ns —0.057 ns
6Pgd-1 [1.000***]
6Pgd-2 {-0.015 ns] [—0.040 ns] [—0.026 ns] [—0.017 ns] -—0.038n
Skdh {[—0.020 ns} —0.039 ns [—0.061 ns] [—0.048 ns| {—0.071 ns] {[-—0.010 aa
No. tests 3[5] 8[2] 4[5] o[5] 3[3] 5[5] 27] 5(5] 5[3] 5[5] Total: 40[45]

showing

conformance

to HW'
No. of tests 1[(1] 1[0] o[0]} o[1] o[o] o[1] o[1] o[o] 0[2] o[0} Total: 2[6]

showin

8
non-conformance
to HW!

* number of non-valid tests in brackets.

WOIYAHLVY NVOTHYNV HLYON NYSLSVa YTV LY AAOTISN

£02

TABLE 6, i statistic analysis (Wright, 1965, 1978), including enc wage ie 1978), for 16 polymorphic loci across Athyrium filix-
femin pulations in eastern North America. Statistical differe evaluated using contingency chi-square analysis. For
biseared tal analysis, Athyrium populations were assigned to their ccs a eens or asplenioides). Differences between values of Fs; and
Fiocality/total are attributable to differences in computational method (discussed in Swofford and Selander, 1981).

Hierarchical F-statistic Analysis (variance
@

F Statistics mponents in parentheses)

FOZ

Total Limiting
Variance

Locus Fis Fer Fst Focality/Total FLocality/Taxon Fraxon/Total ria

Ald —0,025 —0.002 0.022 ns 0.012 (0.00007) 0.014 (0.00009) —0.002 (—0.00002) 0.00623
Got —0.060 =O:017 0.041 ns 0.028 (0.00109) 0.030 (0.00116) —0.002 (—0.00007) 0.03895
Hk 0.116 0.160 0.049*** 0.021 (0.00273) 0.031 (0.00398) —0.010 (—0.00125) 0.12816
Idh-1 0.043 0.485 0460*** 0.452 (0.29124) 0.002 (0.00073) 0.451 (0.29050) 0.64377
Lap 0.077 0.204 O137*** 0.113 (0.02383) 0.092 (0.01886) 0.024 (0.00497) 0.21012
Mdh-1 0.036 0.060 0.025 ns 0.008 (0.00037) 0.014 (0.00068) —0.006 (—0.00031) 0.04939
Mdh-2 —0.059 =—0:013 0.044** 0.017 (0.00042) 0.016 (0.00040) 0.001 (0.00001) 0.02501
Mdh-3 0.302 0.321 0.028 ns 0.007 (0.00013) 0.002 (0.0003) 0.005 (0.00010) 0.01889
Mdh-4 0.208 0.230 0.029 ns 0.010 (0.00046) 0.014 (0.00067) —0.005 (—0.00021) 0.04724
Pgi-2 —0.143 0,012 OF 15" ** 0.096 (0.03914) 0.040 (0.01551) 0.058 (0.02363) 0.40775
Pgm-2 —0.001 0.469 0.468""* 0.459 (0.23433) 0.045 (0.01293) 0.434 (0.22140) 0.51042
6Pgd-1 1.000 -000 0.039 ns 0.018 (0.00016) 0.023 (0.00020) —0.004 (—0.00004) 0.00866
6Pgd-2 —0.032 —0;013 0.019 ns 0.011 (0.00037) 0.000 (0.00000) 0.011 (0.00037) 0.03325
Skdh —0,052 —0.019 0.031"** 0.013 (0.00061) 0.011 (0.00054) 0.001 (0.00006) 0.04782
Tpi-1 =0,070 —0.020 0.047 ns —0.003 (—0.00014) 0.000 (0.00001) —0.004 (—0.00' ) 0.04269
Tpi-2 —0.214 =0,029 6:159""* 0.137 (0.08853) 0.013 (0.00763) 0.125 (0.08091) 0.64843
Combined across loci —0.052 0.217 0.255""* 0.238 (0.68333) 0.028 (0.06343) 0.216 (0.601990)

(Z00Z) € WAAWON 26 ANN'TIOA “TVNYNOlL NAA NVORMANV

KELLOFF ET AL.: EASTERN NORTH AMERICAN ATHYRIUM 205

taxa with respect to the total (Fxyy=0.216) was an order of magnitude greater
than variance among populations within taxa (Fxy=0.028). Thus, differences
between taxa explained most of the variance among populations with respect
to the total (Fxy=0.238). This result is consistent with and explained by the
large allele frequency differences at the four most polymorphic loci (Idh-1,
Pgm-2, Pgi-2, and Tpi-2) between populations of different taxa as opposed to
populations of the same taxon.

Values for Nei’s Genetic Identity, I, (Nei, 1978) and Rogers’ Genetic Simi-
larity, S, (Rogers, 1972) were computed for each pair of populations (Tables
7 and 8). Values for these indices were consistently higher between popula-
tions of the same taxon, ranging from I=0.990 to 1.000 and S=0.930 to
0.975, than between populations of different taxa, ranging from I=0.875 to
0.938 and S=0.803 to 0.881 (Table 8).

Populations were clustered using the Unweighted Pair-group Method with
Averaging (UPGMA) based on both S and I. The two indices resulted in very
similar dendrograms that differed only in the association among some of the
A. asplenioides populations; only the dendrogram based on S is illustrated
(Fig. 5). Two clusters, each comprising all the populations of one taxon,
were joined at S=0.849. Within the A. angustum cluster, the two northern-
most populations Mt. Ste. Hilaire, QUE and Barnet, VT were placed as most
similar, joined at S=0.975, and to this cluster the other three A. angustum
populations were joined successively in order from north to south; the
southernmost A. angustum population Ralston-1, PA joined the A. angustum
cluster at S=0.938. The topology of this A. angustum cluster was identical
in the dendrogram based on I (not shown). In A. asplenioides, the most simi-
lar populations based on S=0.966 were the southernmost population Sandy
Run Swamp, NC and the next most southeastern occurring population
Mountjoy Store, VA, and to these were joined the Pond Drain, VA popula-
tion from the mountains of southwestern Virginia at S=0.958 to form a
subcluster. A second subcluster, comprising the two northernmost A. asple-
nioides populations Shirley, NJ and Hopewell, VA, joined the more southern
subcluster at S=0.947. The topology of the A. asplenioides cluster differed
somewhat in the dendrogram based on I (not shown) indicating that the geo-
graphic ‘‘signal”’ in the A. asplenioides data is weaker than in the A. angus-
tum data.

DISCUSSION

The A. filix-femina complex, distributed across four continents and com-
prising as many as four North American taxa with overlapping ranges, pro-
vides an especially suitable context for exploring patterns and processes of
divergent evolution and its taxonomic consequences in ferns. The two east-
ern North American taxa, A. angustum and A. asplenioides, long have been
perceived as close relatives separable by distinctive characters that are con-
sistent within the vast northern and southern areas they respectively occupy,

206 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 3 (2002)

LE 7. x of pairwise values for Rogers (1972) genetic similarity (above diagonal) and Nei
heer ae genetic identity (below diagonal).

Population

al 2 3 4 5 6 7 8 5 10

1. Mt. Ste. Hilaire, QE - 0.975 0.962 0.953 0.939 0.853 0.865 0.847 0.862 0.881
2. Barnet, VT 1.000 - 0.956 0.938 0.930 0.851 0.864 0.853 0.858 0.879
3. North Hudson, NY 0.999 0.997 - 0.944 0.943 0.844 0.856 0.840 0.852 0.873
4. Schenevus, NY 1.000 0.996 0.995 - 0.942 0.841 0.842 0.831 0.843 0.863
5. Ralston, PA 0.993 0.990 0.994 0.993 - 0.817 0.821 0.803 0.825 0.843
6. Shirley, NJ 0.915 0.913 0.906 0.900 0.883 - 0.943 0.951 0.959 0.932
7. Mountjoy Store, VA 0.924 0.921 0.916 0.908 0.892 0.996 - 0.950 0.966 0.959
8. Hopewell, VA 0.912 0.911 0.903 0.893 0.875 0.998 0.997. - 0.951 0.937
9. Sandy Run Swamp, VA 0.924 0.921 0.916 0.905 0.891 0.998 1.000 0.998 - 0.956
10. Pond Drain, VA 0.938 0.936 0.934 0.923 0.911 0.990 0.998 0.990 0.996 -

but that intergrade and recombine to form a hybrid zone in their relatively
narrow region of overlap. This perception is supported by the data from
the present study. The spores of the two taxa show striking and consis-
tent differences in the perispore sculpturing, a low papillate perispore in
A. angustum versus a rugose perispore in A. asplenioides, and frequencies
of allozymes exhibit strong differences between as compared to within the
taxa. An abrupt shift in allele frequencies at the four most polymorphic loci
(Idh-1, Pgm-2, Tpi-2, and Pgi-2) of Athyrium corresponds to the geographic
boundary between A. angustum and A. asplenioides, i.e., between northern
Pennsylvania and southern New Jersey in our sample. Although virtually all
alleles were shared between the two taxa, some alleles that predominated or
were frequent in one taxon were nearly absent in the other, e.g., Idh-1“ of
A. asplenioides, Idh-1° of A. angustum, and Pgi-2° of A. angustum. In other
cases, alleles were more extensively shared between the taxa but at very
different frequencies, e.g. Idh-1", Pgm-2°, Pgm-2°, and Tpi-2" (Table 3).
UPGMA analysis of Athyrium resulted in two distinct taxon clusters joined
at a substantially lower similarity value (S=0.849) than that joining popula-
tions within their respective taxon clusters (S=0.938 for A. angustum;

TABLE 8. Means of pairwise values for iy va Similarity (S) and Nei’s Genetic Identity (I) for
eae within and between Athyrium taxa. Ranges are given in parentheses. Each category
was represented by ten pairwise comparisons.

Taxon combination I S
angustum—angustum 0.996 0.948
(0.990—1.000) (0.930—0.975)
asplenioides—asplenioides ; 0.951
(0.990—1.000) (0.932-0.966)
angustum-asplenioides 0.911

(0.875—0.938) (0.803-0.881)

KELLOFF ET AL.: EASTERN NORTH AMERICAN ATHYRIUM 207

Mt. Ste. Hilaire, QUE
Barnet, VT
North Hudson, NY angustum

ei Schenevus, NY

Ralston-1, PA
Shirley, NJ —
[ i Hopewell., VA
i Mountjoy Store, VA asplenioides
= Sandy Run Swamp NC
Pond Drain, VA
Rogers’ Similarity

. 5. Dendrogram resulting from UPGMA analysis based on pairwise values of Rogers’ Simi-
larity (Table 7).

S=0.947, for A. asplenioides). These spore and allozyme differences for the
most part are consistent among populations of each taxon, yet there is evi-
dence of introgressive hybridization in their region of overlap as discussed
below.

The distinctness of these taxa and the consistency of the character differ-
ences specified by Butters (1917) frequently have been questioned by state-
ments indicating that the taxa intergrade extensively (e.g., Benedict, 1934;
Weatherby, 1936; Fernald, 1946; Shaver, 1954; Wherry, 1961). However,
such statements seem anecdotal in that they are not accompanied by docu-
mentation of character combinations in specimens. The most extensive
specimen-based data set is that of Liew (1971, 1972), who carried out a phe-
netic analysis of North American Athyrium sensu lato based on 170 speci-
mens scored for 99 characters. Liew indicated that cluster analysis based on
paired similarity coefficients separated each Athyrium taxon, including
A. angustum and A. asplenioides, into its own cluster, but the summary
dendrograms presented leave it uncertain as to whether some of the speci-
men placements were inappropriate or equivocal.

In the present study, the degree of differentiation between A. angustum
and A. asplenioides may appear exaggerated by the omission of localities
where the two taxa co-occur. This omission was neither intentional nor an
oversight, rather it resulted from a failure to discover such localities despite
a substantial effort to do so. A search for co-occurring populations in south-
ern Pennsylvania resulted in finding only a few isolated individuals of
A. asplenioides in this highly agriculturalized and urbanized region, and
no sizable populations from which allele frequency data could be obtained.
All individuals in populations sampled for the present investigation from
Quebec south to northern Pennsylvania were readily assignable to A. angus-
tum on the basis of leaf and spore morphology. Similarly, all individuals in

208 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 3 (2002)

populations from North Carolina north to northern Virginia were assignable
to A. asplenioides. However, a genetic influence from A. angustum is sug-
gested by the high frequency of characteristically A. angustum alleles Idh-1”
and Pgi-2° in the northernmost A. asplenioides population Shirley, NJ, as
well as Pgm-2” and Tpi-2” in the highest elevation A. asplenioides popula-
tion Pond Drain, VA. Evidence of introgressive hybridization is augmented by
the intermediate spore morphologies encountered at the Shirley, NJ, locality.
No individuals assignable to A. angustum were found at this site, but it is
possible that A. angustum spores could have migrated from populations fur-
ther north and effected hybridization (Wagner, 1943). The prevalence of fully
intermediate spores in six of the individuals, five of which are heterozygous
for taxon marker allozymes, suggest that these are first-generation hybrids.
The skewed array of spore morphologies of other individuals suggests that
they are backcrosses and provides evidence that hybridization has gone
beyond the first generation resulting in introgression. Thus, spore and iso-
zyme data combine to support previous morphology-based suggestions that
the region of overlap between the taxa, and possibly higher elevations in the
southern Appalachians as well, could represent a hybrid zone resulting from
secondary contact between recently diverged sister taxa (Benedict, 1934;
Shaver, 1954; Sciarretta et al., ms).

Narrow hybrid zones in which formation of fertile hybrids and backcrosses
result in intergradation between divergent taxa occur between numerous spe-
cies or subspecies pairs of animals and angiosperms (reviewed by Arnold,
1997), and occur in a few agamosporous ferns (Gastony and Windham,
1989). Hybridization between fern species usually results in spore abortion
due to abnormal meiosis, and introgressive hybridization between homo-
ploid taxa as divergent as A. angustum and A. asplenioides is decidedly rare
in ferns, with only three cases having been documented previously: (1)
swarms of fully fertile hybrids involving Pteris quadriaurita and P. multia-
urita, first generation as well as backcrosses, occur in disturbed forests of
Sri-Lanka (Walker, 1958); (2) extensive hybridization among three species of
the tree fern genus Alsophila resulted in a complex swarm of fertile hybrids
in Puerto Rico, a scenario hypothesized to give rise to new species through
allogamous allohomoploidy (Conant and Cooper-Driver, 1980; Conant, 1990);
(3) morphological and molecular data provide evidence of hybrid swarms
between Polystichum munitum and P. imbricans of northwestern North
America (Mayer and Mesler, 1993; Mulleniex et al., 1998). Of these, the situa-
tion in Polystichum most closely parallels that of the Athyrium taxa angus-
tum and asplenioides. The two Polystichum taxa share alleles at most
enzyme loci, but are differentiated with respect to frequencies at three loci
resulting in a lower magnitude of genetic identity between the taxa (I=
0.842) than within each taxon (means: I=0.974 for P. imbricans, 1=0.957 for
P. munitum; Soltis et al., 1990, 1991). Additionally, the Polystichum taxa hy-
bridize introgressively, although they do not form a hybrid zone as in Athy-
rium; rather hybridization occurs at various localities across a broad area of
sympatry between the two taxa (Mayer and Mesler, 1993; Mulleniex et al.,

KELLOFF ET AL.: EASTERN NORTH AMERICAN ATHYRIUM 209

1998). Maintenance of distinction between these two Polystichum species
most likely results from a combination of their partial intersterility and di-
versifying selection imposed by the very different habitats—forests versus
exposed cliffs—occupied by P. munitum and P. imbricans respectively.

The nature, frequency, and geographic extent of hybridization between
A. angustum and A. asplenioides remain uncertain and merit further
research that combines field, herbarium, molecular, and breeding studies. It
is critical to determine with greater precision and full documentation the
degree to which these two taxa maintain versus blend their macromorpho-
logical, micromorphological, and allozymic differences where they coexist. It
is unknown whether there is preferential mating within taxa, whether taxon
characters tend to remain associated in the face of hybridization or are com-
pletely recombined, and whether there exists a cline for allozyme frequen-
cies and morphological characters within the overlap region. There is a need
for intensive exploration for coexisting A. angustum and A. asplenioides pop-
ulations in the areas of their overlap from the eastern seaboard through the
midwest, and in the southern Appalachians where the existence of cryptic
taxa has been hypothesized (Wagner and Wagner, 1966). Isozyme studies of
these populations should be coordinated with critical analyses of morphologi-
cal character combinations obtained from numerous specimens and with
experimental crosses that can quantify the propensity of the taxa to hybridize.

Beyond the taxonomic significance of hybridization, the unprecedented
formation of normal intermediates between spores of such divergent mor-
phology provides an opportunity to gain insight into the genetics underlying
spore morphology (Schneller, 1989). The variability of spore morphology
within individuals of the putative primary hybrids from the Shirley, NJ site,
which includes expression of parental types, indicates that inheritance is
polygenic rather than a simple one-or-two-gene inheritance mechanism, and
that the spore genotype determines or at least influences perispore pheno-
type. Contrasting observations and inferences were obtained by Schneller
(1989), who reported the formation of normal intermediate spores in experi-
mentally produced hybrids between A. angustum and A. asplenioides, but
found that all spores from a sporangium were of the same type, implicating
sporophytic determination of the perispore morphology. Explanations that
can account for these differing observations include the possibility that the
Shirley, NJ, plants were not true F1 hybrids, or that there may be a maternal
effect that varies as a polymorphism. Clearly, further studies of the inheri-
tance of spore morphology are in order.

TAXONOMIC CONCLUSION: AT WHAT RANK SHOULD A. ANGUSTUM AND A. ASPLE-
NioipES BE RECOGNIZED?—Over the second half of this century, the nature of
pteridophyte species has been clarified significantly by application of tech-
nological advances in cytology and molecular systematics in combination
with detailed morphological and field studies (Manton, 1950; Wagner, 1963;
Haufler, 1987, 1989; Paris et al., 1989; Conant, 1990). Nonetheless, definitive
ranking of closely related divergent taxa with overlapping ranges, such as

210 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 3 (2002)

the two Athyrium taxa considered here, remains challenging due to the un-
settled controversy as to the nature and definition of species (e.g., Mayr,
1992; Davis and Nixon, 1997; Baum, 1998; DeQueroz, 1998) as well as to the
unpredictable nature of reproductive interactions between diverged taxa ex-
periencing secondary contact (Arnold, 1997). The rank assigned to A. angus-
tum and A. asplenioides has varied considerably in floristic treatments
published in this century. While Small (1938) and Wherry (1948, 1961) fol-
lowed Butters (1917) in treating these taxa as separate species, other authors
have tended to treat them as infraspecific taxa, either subspecies (Lellinger,
1985) or varieties (Fernald, 1950; Mickel, 1979; Cody and Britton, 1989;
Gleason and Cronquist, 1991; Kato, 1993).

Spore and isozyme data combined indicate that populations of these two
taxa are significantly divergent, exhibiting greater differences than ordinarily
encountered within single species of ferns thus far studied. However, the
fertility of hybrids (Schneller, 1989) provides a potential for introgression
between the two taxa. The preliminary evidence that hybridization and
introgression do in fact occur indicates that A. angustum and A. asple-
nioides would be considered conspecific under species definitions as differ-
ent as the Biological Species Concept (Mayr, 1942) and the Phylogenetic/
Diagnostic Species Concept (Cracraft, 1983; Davis and Nixon, 1992). None-
theless, in practice numerous pairs of taxa that form hybrid swarms or zones
are treated as distinct species (Grant, 1981; Arnold, 1997).

The occurrence of northern and southern infraspecific taxa in eastern
North American Athyrium is paralleled in Pteridium aquilinum L., which
comprises northern and southern varieties Jatiusculum (Desv.) Underw. and
pseudocaudatum (Clute) Heller, respectively, and which shows north to
south clines in allele frequencies near the overlapping taxon boundary
(Speer et al., 1998). However, the pattern of allozyme variation in Pteridium
differs from that in Athyrium in that the most abrupt shift in allele frequen-
cies occurs within the range of P. latiusculum, genetic identities between
populations of the two Pteridium varieties ranged substantially higher (I=
0.916 to 0.999) than those of the two Athyrium taxa (I=0.875—0.938, mean
=0.911), and UPGMA clustering failed to separate the varieties (Speer et al.,
1998). On the basis of the lack of genetic differentiation between P. latiuscu-
lum and P. pseudocaudatum, variety was indicated as the highest rank at
which to recognize them (Speer et al., 1998). In contrast, the more substan-
tial differentiation between the Athyrium taxa angustum and asplenioides
and the consistent characters uniting a vast number of individuals north and
south of their hybrid zone suggest that the taxa should be ranked at least at
the level of subspecies. Ranking at the species level would not be inconsis-
tent with the treatment of such taxa in the broader plant literature.

AACCKNOWLEDGMENTS

The authors thank John Kress and Vicki Funk for facilitating portions of this research in vari-
ous ways; Mark Strong, Cynthia Caplen, Kim Sciarretta, Erin Potter and Heather Hartless for

KELLOFF ET AL.: EASTERN NORTH AMERICAN ATHYRIUM 211

help with sng Se Aig Nowicke for advice on preparation and interpretation of electron
sae sane: Walter R. Brown and Susann Braden for help with the use of the SEM; Stanley
Yank for bec on light microscopy; Cheryl Roesel, Lori Croisatiere, Thao Le, and Leigh
Ann aida for assistance with electrophoresis; Greg McKee for carrying out germination
tests; and David Lellinger for comments on the manuscript. This research was supported by a

eorge Mason University Research Fellowship awarded to CLK, by NSF Equipment Grant

the Structure and Evolution of Terrestrial Ecosystems awarded by the U. S. National Herbarium
and NSF Grant DEB-9220755 awarded to CRW. A portion of this study is based on the MS thesis
of CLK completed at George Mason University.

LITERATURE CITED

ARNOLD, M. L. 1997. Natural hybridization and evolution. Oxford University Press, NY.
Baum, D. A. 1998. Individuality and the existence of species through time. Syst. Biol. 47:
64

BENEDICT, R. C. 1934. Can dens readily distinguish the northern and southern lady fern spe-
cies. eI Fern J. 24:117-11

Butters, F. K. 1917. Taxonomic and geographic studies in North American ferns. I. The genus
Athyrium pee the North American ferns allied to Athyrium filix-femina. Rhodora 19:
169-207

Copy, W. J. AND D. M. Brirron. 1989. Ferns aha fern allies of Canada. Publication 1829/E.
Resea eck picages pearance Canada, Ott

Conant, D. S. 0. Observations on the reproductive biology of Alsophila species and hybrids
ee a aak Ann. Missouri ae sas :290-296

AND G. COOPER-DRIVER. 1980. togamous silokiomoplosdy in Alsophila and Nephelea
ieijattienceand a new hypothesis is speciation in homoploid homosporous ferns. Amer. J.
Bot. 67:1269-1288.

CracrarT, J. 1983. Species concepts and speciation analysis. Curr. Ornith. 1:159-187.

Davis, J. I. AND K. C. Nixon. 1992. Ph eating genetic variation, and the delimitation of phylo-
genetic come Syst. Biol. 41:421-4

Dawes, C. J. 1971. Biological techniques in mee microscopy. Barnes and Noble, New York.

Say

sa Pp
erlocher, eds. Endless forms: Species and Secaceeiarie Oxford

Ss)
aa

thy versity Press, Oxford, Englan

ERDTMAN, G. 1969. Handbook of palynology. Hafner, New York.

dicing M. L. 1946. Some trivial American forms of lady fern. Rhodora 48:389—391.

950. Gray’s manual of botany: a handbook of the flowering plants and ferns of the cen-
aa ne northeastern United States and adjacent Canada. American Book Company,

Gastony, G. J. AND D. C. Darkow. 1983. Chloroplastic and cytosolic isozymes of the homosporous
e

. Species sin in eo the treatment and defini-

tion of amnesonous epoctoe. Amer. Fern J. 7

Gueason, H. A. AND A. Cronquist. 1991. Manual of be cane plants of northeastern United
States and adjacent Canada. 2d ed. The New York Botanical Garden, Bronx, NY.

Grant, V. 1981. Plant speciation. Columbia University Press, N

Hamrick, J. L. anp M. J. W. Gopt. 1989. Allozyme diverary } in plant species. Pp. 43-63 in
A. H. D. Brown, M. T. Clegg, A. L. Kahler, and B. S. Weir, eds. Plant population genetics,
breedi ing and ete resources. Sinauer Associates, Sunderland, MA.

Haur.er, C. H. 1987. Electrophoresis is modifying our concepts of evolution in homosporous
pteridophytes. Amer. J. Bot. 74:95—-966.

———. 1989. Species concepts in the pteridophytes: summary and synthesis. Amer. Fern J.

212 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 3 (2002)

Kato, M. 1993. Athyrium. Pp. 255-258 in FNA Editorial Committee, - Flora of North America,
me : sige wnbal and gymnosperms. Oxford Univ. Press,
iste C L. 0. A cones iss of the eastern North American va fern, Athyrium filix-
femina Hes panies: M. S. Thesis, George Mason University, Fairfax
LELLINGER, ch a 1985. A field manual of the Lone in fern-allies of the ate States and Canada.
sonian Institution Press, Washington
Li, + AND cee H. HAvuFLer. 1999. Genetic variutio n, cearte systems, and oe of diversifica-
mn in io lyorcca Polypodiam (Polypodiaceae). Syst. Bot. 24:339-35
Liew, Ky Ss: . Studies on North American Athyrium (L.) Roth. Ph.D. Tau: Columbia
ene

. 1972. Numerical taxonomic studies on North American Lady Ferns and their allies. Tai-
nia 17:190-221.
ions B. 1971. Contribution a la connaissance de la morphogenése et de la structure me parois
ages chez les filicinées isosporées. Thesis of Université Paul Sabatier de Toulou
976. Sur la structure fine de l’exospore dans les devers groups de Sat udicie
ero (microspores et isospores). Linn. Soc. Series 1:231-2
— Sur la formation du sporoderme chez Psilotum triquetram Sw. (Psilotaceae). Grana

165;
Manton, ‘4 pn Problems of cytology and evolution in the pteridophytes. Cambridge University
Press, London
Mayr, E. he Stomakies and the origin of species. Columbia University Press, NY.
. 1992. A local flora and the biological species concept. Amer. J. Bot. 79:222-238.
orp

ag
. 1998. Molecular ones at hybrid swarms in
the ie enue Polystichum aie Syst. Bot. 23:421-4

Murpny, R. W., J. W. Sires, Jr., D. G. BUTH, AND . HAUFLER. 1990. sn Isozyme electro-
phoresis. Pp. pees in D. Hillis and C. Maite, eds. Molecular Systematics. Sinauer,
Sunderland, MA.

Nel, M. 1978. Estimation of Pig oe and genetic distance from a small number of
individuals. Genetics 89:583—
Paris, C. A., F. S. WAGNER, AND in “ Wacner, JR. 1989. Cryptic species, species delimitation,
and taxonomic practice in the homosporous ferns. Amer. Fern J. 79:46—54.
Pete J. S. 1972. Measures of genetic seep and genetic Rieti Studies in Genetics,
University of Texas Publication 7213:145-15
SCHNELLER, J. J. 1979. a a investigations | on the lady fern (Athyrium filix-femina). P.
ma Evol, 132:255-
989. Remarks on heer regulation of — wall pattern in intra- and interspecific
crosses of 2 . Bot. J. Linn. Soc. 99:115—
ND B. W. Scumip. 1982. lnvetigation on the inspec variability in Athyrium filix-
‘eats L) Roth Bull. Mus. Natl. H oe B. 3—4:215-—
SELANDER, R. K., M. H. Situ, S. Y. YANG, W

E. JOHNSON, wien 'B. GENTRY. 1971. Biochemical
ii ta and systematics in the genus Peromyscus. I. variations in the old-field
90.

7 Tennessee with the fern allies exchndad. George Peabody College for

Teachers, Nashville, IN.

SMaLL, J. K. 1938. Ferns of the southeastern United States. The Science Press Printin
Lancaster, PA.

Soitis, D. E., C. H. Haurter, D. C. Darrow, anv G. J. GASToNny. 1983. Starch gel electrophoresis of
ferns: A compilation ‘of grinding buffers, gel and electrode buffers, and staining schedules.
Amer. Fern J. 73:9—27.

Soxtis, P. S., ALVERSON. 1987. Electrophoretic and morphological confirma-
tion of interspecific hybridization between Polystichum kruckebergii and P. munitum.
Amer. Fern J. 77:

g Company,

KELLOFF ET AL.: EASTERN NORTH AMERICAN ATHYRIUM 213

Soxtis, D. E., P. S. Soitis, AND P. G. Wor. 1990. Allozymic divergence in North American Poly-
stichum (Dryopteracoan) Syst. Bot. 15:205-215.
Soxtis, P. S., D. E. Soitis, AND K. E. Housincer. 1988. Estimates of shy ares aa selfing and
interpopulational te me in ng aneen ferns. Amer
DP. 1. Allozymic and chloroplast DNA ia Hil of Polyploiy i in
Polstican pence irraneret) I. The origins of P. californicum and P. scopulinum. Sys

SPEER, W. I., C. R. WertuH, AND K. W. Hiu. 1998. Relationships between two infraspecific taxa of
ache poh ate tDennstaedicene II. Isozyme evidence. Syst. Bot. 23:313-326.

SworrorpD, D. AND R. SELANDER. 1981. BIOSYS-1. A FORTRAN program for the wae olrelig
analysis of a onttiey ne data in population genetics and systematics. J. Hered. 72:281—283.

Tryon, A. F. 1986. Stasis, diversity and function in spores based on an electron annem sur-
vey of the Pteridophyta. Pp. 233-249. in S. Blackmore and I. K. Ferguson, eds. Pollen and

spores: form and function. No 12. Academic Press, London

AND B. LuGaRDON. 1991. Spores of the pteridophyta. ea wall structure, and diversity

based on electron microscope studies. Springer-Verlag, N

Wacner, W. H., Jr. 1943. Hybridization by remote control. Amer. Fern J. 33:71-73.

bier A biosystematic survey of United States ferns - coat Se ti eh fi Amer. Fern J.

53:1-

D . WaGNER. 1966. pla ale of sig ety Lake area, Giles Co., Virginia: bio-
ae lige: 1964-6 anea 31:1
Wa tker, T. G. 1 Dect in some conn peas L. Evol. 12:82-92.
WEATHERBY, C. ‘ a A list of varieties and forms of the ferns of ana North America. Amer.

WenpeL, J. F. anp N. F. WEEDEN. 1989. Visualization and interpretation of plant isozymes. Pp.
5-45 in D. E. Soltis and P. S. Soltis, eds. Isozymes in plant biology. Dioscorides Press,
Portla oa OR.

WERTH, . as 1985. Implementing an isozyme laboratory at a field station. Va. Je Sci. 36:53-76.

0. rani Pre-prepared frozen isozyme assays. Isozyme Bull. 09.

WHERRY, a T. 1948. Remarks on the American Lady Fern. Amer. Fern J. 38:155-158.

61. ies fern guide. Northeastern and midland central states and adjacent Canada.
‘dilate and Co., Garden City

Wricut, S. 1965. The Beiasryp ere of ee ee structure by F-statistics with special regard to
systems of mating. Evol. 19

. 1978. Evolution and tha onan of populations, as 4. Variability within and among
netarel populations. University of Chicago Press, Chic

American Fern Journal 92(3):214—228 (2002)

A New Hybrid Polypodium Provides Insights
Concerning the Systematics of Polypodium scouleri
and its Sympatric Congeners

T. J. HILDEBRAND, CHRISTOPHER H. HAUFLER, JAMES P. THERRIEN,
and CatHy WALTERS
Department of Ecology and Evolutionary Biology, University of Kansas, Lawrence,
Kansas 66045-2106
PuiLtip HAMMOND
University Herbarium, University of California, Berkeley, CA 94720

AsstRACT.—With its thick, leathery leaves, reticulate venation, and large sori, Polypodium scou-

leri, located in a narrow band along the Pacific coast of North America, is the most distinctive
member of the cosmopolitan P. vulgare species complex. Although early studies based on mor-

and trnL DNA sequences with isozymic analyses suggested that P. scouleri originated relatively
recently and is closely allied to and sympatric with P. californicum and P. glycyrrhiza. Consis-

The Polypodium vulgare L. complex (Polypodiaceae) first drew attention
with the publication of cytological counts for several European and North
American members (Manton, 1950). Controlled crosses among members of
the complex followed (Shivas, 1961), and resulted in further taxonomic stud-
ies founded on interbreeding boundaries and subtle morphological distinc-
tions. A major organizational leap in the circumscription of North American
Polypodium L. species was the definition of eastern and western complexes
(Lloyd and Lang, 1964). In their search to identify parental lineages for allo-
polyploid members, Lloyd and Lang grouped the complexes on the presence
(eastern) or absence (western) of sporangiasters. Although sporangiasters
were later shown to constitute synapomorphies in several western species
(P. amorphum Suksdorf, P. saximontanum Windham, and P. sibiricum
Siplivinsky [Haufler and Windham, 1991]), recognizing the importance of

ese unique soral features within American polypodiums was insightful.
Additionally, Lloyd and Lang (1964) hypothesized that the progenitors of the
allotetraploid P. californicum Kaulfuss were P. glycyrrhiza D. C. Eaton (2n)

HILDEBRAND ET AL.: A NEW HYBRID POLYPODIUM 215

and diploid P. californicum Kaulfuss. It was not until much later, however,
that Whitmore and Smith (1991) described the tetraploid cytotype as a dis-
tinct and reproductively competent species, P. calirhiza S. A. Whitmore &
A. R. Smith.

Research explorations of morphology (Haufler and Windham, 1991;
Whitmore and Smith, 1991; Haufler et al., 1993), cytology (Haufler and
Wang, 1991), isozyme variation (Haufler et al., 1995b), chloroplast DNA
restriction site analysis (Haufler et al., 1995a), and DNA sequence data
(Haufler and Ranker, 1995) have further defined relationships within the
Polypodium vulgare complex. Yet a review of the above reveals the exclu-
sion of P. scouleri Hooker & Greville from all but two studies (Haufler et al.,
1993; Haufler and Ranker, 1995).

Confined to a narrow distribution along the Pacific Coast of North America,
and tentatively recognized as a member of the western complex, the dis-
tinctive morphology of P. scouleri clearly separating it from its congeners
precluded the formulation of accurate hypotheses about phylogenetic rela-
tionships. The overview by Haufler et al. (1993, pp. 315-323) in their treat-
ment of Polypodium in Flora of North America (Fig. 1) allied P. scouleri
with other western members, and provided a detailed morphological de-
scription. In addition, the investigation of rbcL sequence data suggested a
sister taxon relationship with P. glycyrrhiza that was supported by isozyme
profiles (Haufler and Ranker, 1995). Haufler and Ranker (1995) hypothesized
that the particularly distinctive morphological features of P. scouleri may
have evolved through adaptive response to environmental stress. That study
did not consider another close relative, P. californicum, and, because only
diploids were included, did not incorporate allopolyploid P. calirhiza.

Recently, leaves resembling P. scouleri but having some atypical features
were collected at a site in California. At the same time, leaves of P. calirhiza
and more typical P. scouleri were obtained. The present study was designed
to examine the atypical leaves morphologically and use molecular ap-
proaches to determine whether these plants originated through hybridization
between P. scouleri and other members of the western complex. Several mor-
phological features of suspected hybrid leaves were investigated and com-
pared with features of typical P. scouleri and P. calirhiza leaves. Starch gel
electrophoresis was used to reveal isozyme marker alleles and characterize
the suspected hybrid and parental lineages.

MATERIAL AND METHODS

Stupy ArEA.—Located within the confines of highly populated San Francisco
County are the natural vegetation preserves of Tank Hill and Mt. Sutro
(Fig. 2). The Open Space Program of the San Francisco Recreation and Parks
Department retains ownership of these sanctuaries. Land stewards manage
vegetation of the preserves, emphasizing enhancement and restoration of
native species as well as the removal of invasive exotic plant populations.

216 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 3 (2002)

P. scouleri

P. californicum

P. glycytthiza
P. hesperium
P. amorphum

P. saximontanum

P. sibiricum

Fic. 1. Kinship of diploid and tetraploid members of the Western Polypodium vulgare complex

of P. sibiricum, which is circumboreally distributed in North America and Asia. Shaded ovals
represent sterile backcrosses (3n) in the complex. The dashed line between P. scouleri and
P. calirhiza indicates the subject of the present study.

A dense forest of mature, non-native cypress (Cupressus macrocarpa
Hartw. Ex Gord.) and eucalyptus (Eucalyptus globulus Labill.) was planted on
Mt. Sutro as early as 1870 and now covers much of the hill. Polypodium
species on Mt. Sutro flourish beneath the forest canopy, nestled among rocks
and at the bases of trees as hemiepiphytes. Leaves of Polypodium calirhiza
and P. scouleri were collected from the east face of Mt. Sutro (Site 1, Fig. 2;
Table 1).

Eucalyptus and cypress were also planted on Tank Hill, but, in contrast to
Mt. Sutro, they are sporadic on Tank Hill, and primarily at lower elevations.
The summit of Tank Hill is dominated by large tracts of rocky fields and
ledges of Franciscan radiolarian chert. In the fields and on the ledges are
Polypodium populations exposed to the harsh sun and buffeted by gusting
winds. Native species associated with ferns in this predominantly open
habitat include: Nootka reed grass (Calamagrostis nutkaensis (Presl) Steud.),
yarrow (Achillea millefolium L.), coast barberry (Berberis pinnata Lag.), soap
plant (Chlorogalum pomeridianum (DC.) Kunth. var. divaricatum (Lindl.)
Hoov.), and seaside daisy (Erigeron glaucus Ker.). Among the boulders and
crevices near the crest of Tank Hill, leaves of Polypodium calirhiza and

HILDEBRAND ET AL.: A NEW HYBRID POLYPODIUM 217

San Francisco
County

Fic. 2. Collection localities on Mt. Sutro and Tank Hill, San Francisco County, California.

P. scouleri, as well as the suspected hybrid, were collected (Site 2, Fig. 2;
Table 1). To characterize the amount of electrophoretically detectable genetic
variation across the range of P. scouleri, population samples were obtained
from four additional California populations. Representative specimens of all

218 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 3 (2002)

TaBLE 1. California sites from which Polypodium plants were collected. All specimens are
deposited in the McGregor Herbarium, University of Kansas (KANU). N = number of individuals
for each population.

Collection Site (county) Species (ploidy level: x = 37) N Voucher

~

Mt. Sutro (San Francisco) calirhiza (4x)
Mt. Sutro (San Francisco) P. scouleri (2x)

woo
as
a
Sf
Dm OD
Sse
Q 8
55
a8
Go
NR hd&S
—~ me
On

Tank Hill (San Francisco) P. calirhiza (4x) 12 Hildebrand 3219
Tank Hill (San Francisco P. scouleri (2x) 7 Hildebran
Tank Hill (San Francisco) P. calirhiza X scouleri (?) 5 Hildebrand 3220 & 3221
Fern Canyon (Trinity) P. scouleri (2x) 20 ~=Therrien s.n.
Point Reyes National P. scouleri (2x) 15 Therrien s.n.
Seashore (Marin
Trinidad (Humboldt) P. scouleri (2x) 6 Therrien s.n.
Fort Ross (Sonoma) P. scouleri (2x) 3. ‘Therrien s.n.

collections (Table 1) were pressed and deposited at McGregor Herbarium of
the University of Kansas (KANU).

MeETHODS.—Leaves were removed from live material in the field, placed in
plastic bags, and shipped on ice. Upon arrival, bags were transferred to 4°C
storage where they were kept until preparation for electrophoresis. At least
one sample of plant material for each leaf was prepared immediately upon
receipt in the laboratory. Most leaves, particularly of P. scouleri and the sus-
pected hybrid, retained their fresh appearance in storage for considerable
time (up to one month). Preparations from fresh plant material stored for
longer periods yielded banding patterns comparable to that prepared imme-
diately, indicating extended retention of enzymatic activity.

Plant material was prepared by crushing in phosphate-PVP buffer (Soltis
et al., 1983) followed by absorption of the homogenate into filter paper wicks
and storage of the wicks at —80° C. Freezing prepared fresh material allowed
storage of samples for several months with no loss of enzymatic activity.

Selection of enzymes and the systems best suited for their survey was
based on previous studies of Polypodium (Haufler et al., 1995a, 1995b) and
other fern genera (Haufler, 1985b; Haufler et al., 1990: Soltis et al., 1990;
Pryer and Haufler, 1993). Banding patterns were obtained by electrophoresis
on 12.4% starch gels for the following enzymes: aldolase (ALD), fructose 1,6-
bisphosphatase (FBP), glyceraldehyde 3-phosphate dehydrogenase (G-3PDH),
hexokinase (HK), isocitrate dehydrogenase (IDH), phosphogluconate dehy-
drogenase (PGDH), triosephosphate isomerase (TPI), and phosphoglucoiso-
merase (PGI). Bands were resolved best for ALD and IDH on system 11
(Haufler, 1985b) whereas only the 7.5 pH version of the morpholine/citrate
system (MC) (Clayton and Tretiak, 1972) revealed clear bands for G-3PDH.
Both system 11 and MC resolved bands for FBP, MC and system 8 (Haufler,
1985a) for MDH, and MC and system 6 (Soltis et al., 1983) for PGDH. System
8 also revealed bands for HK, LAP, and, in addition to system 6, for PGI. TPI
bands were revealed only with system 6. Digital images were obtained of all

HILDEBRAND ET AL.: A NEW HYBRID POLYPODIUM

TABLE 2.
and P. calrhiz

219

or aan of character states identified for Polypodium scouleri, calirhiza X scouleri,

Character

Polypodium scouleri

Polypodium
calirhiza X scouleri

Polypodium calirhiza

Frond texture
Rhacis scales

Pinna venation

Guard cells
- shape
- mean length:
uum (range)
peponinno cell
argins
Paces ate
stomata density
Spore length:
um (s.d.)

- distribution

- size

stiff, leathery

broadly ovate,
tapering to ca 3
cells in width; not
occurring in pairs

regularly
anastomosing,
forming one row
of areoles

round

35 (33-38)

smooth

150/mm?

61.3 (+ 8.4)
round or oblong;

closest to midrib

generally > 3 mm

stiff, leathery
narrowly ovate to
sear lanceolate;

forming areoles
round
38 (35-43)
distinctly lobed
70/mm*
malformed
round or oblong; both
shapes on

individual plants
variable

herbaceous

lanceolate to
lanceolate-ovate;
3-6 cells wide,
tapering to 1-3
cells; not occurring
in pairs

free, no areoles

med

elliptic
56 (48-63)

distinctly lobed
40/mm*

Bib (= 71)
oblong

midway between
costa and margin
generally < 3 mm

variable, 14 mm, to
sent

gels using a Nikon CoolPix 950 camera and visualized with Adobe Photo-
shop 5.5 software for Macintosh.

Morphological characters documented as useful for delimiting Polypodium
species (Haufler et al., 1993; Whitmore and Smith, 1991) were examined on
leaves from both parental species and the hybrid. These included leaf tex-
ture, sorus diameter, pinna venation, spore length, and rachis scale width. In
addition, lower surface (abaxial) epidermal peels were produced to measure
sizes and characterize shapes of guard and subsidiary cells, as well as differ-
ences in stomatal density (Table 2).

RESULTS

MorpHoLocy.—The leaves of Polypodium scouleri differ from those of other
members of the genus (although leaves of P. calirhiza may become somewhat
thickened in exposed coastal environments) by their leathery texture, and

220 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 3 (2002)

mean = 61.5 mean = 61.3
70 | s.e.= 7.1 se. = 8.4

mean = 55.3
s.e. = 6.2

Spore length (micrometers)

45 =

P. calirhiza P. scouleri FP. scouleri

(CA) (CA) (OR)
n= 2 = 2 n=
x = 205 x = 189 “2 55

Fic. 3. Comparisons of spore length from plants gathered at the Mt. Sutro and Tank Hill Poly-
podium populations, and from a P. scouleri population from Oregon. Vertical bars represent +/—
one standard error. Expected mean spore lengths (Haufler et al., 1993) for P. calirhiza and P. scou-
leri are > 58 wm and < 53 um, respectively. n = number of plants sampled; x = number of spores
measured.

their well-defined aeroles produced by anastomosing venation. Other dis-
tinctive morphological features include individual leaf segments that are
greater than 12 mm in width, soral diameters of more than 3 mm, and rachis
scales that are large, pale reddish brown and broadly triangular, tapering to a
point less than three cells wide (Table 2). The pair of guard cells surrounding
the stomata on the lower epidermal surface is circular in outline; individual
guard cells average 35 jm in length, and are adjoined by smooth-margined
subsidiary cells. Stomata density is approximately 150 stomata/pm’.
Tank Hill and Mt. Sutro P. scouleri populations had a mean spore length of
61.3 um (Fig. 3).

Polypodium calirhiza plants have herbaceous leaves lacking the leathery
texture of P. scouleri and are, in general, of smaller stature. Pinna venation
is free or weakly anastomosing with some to many segments lacking aeroles.
Leaf segments seldom exceed 12 mm wide, and soral diameters are less than
3 mm. Rachis scales, as in P. scouleri, are a translucent, pale reddish brown,

HILDEBRAND ET AL.: A NEW HYBRID POLYPODIUM 221

but, in contrast to P. scouleri, are lanceolate, only a few cells wide proxi-
mally, and narrow to only one to three cells distally. Epidermal peels
showed elliptic paired guard cells (average length: 56 wm) and subsidiary
cells with distinctly lobed margins. In striking contrast to that observed for
P. scouleri stomata, stomata density is approximately 40 stomata/mm‘’.
Spores of the California P. calirhiza plants averaged a mean length of
61.5 um.

The putative hybrid individuals showed both hybrid vigor (Charlesworth
and Charlesworth, 1987) and dwarfing of leaves. On nearby Mt. Davidson, a
more sheltered locale, P. scouleri produced leaves greater than 70 cm in
length and lush in appearance. Polypodium calirhiza morphotypes vary with
exposure, and increasingly open areas produce plants of more diminutive
stature.

Morphological character states observed for hybrid leaves were either i) a
combination of discrete features inherited without change from each parent,
or ii) an additive blending of parental traits resulting in intermediate mor-
phological states (Table 2). An exception was the consistently leathery, stiff
texture of hybrid leaves, a phenotype that resembles only the P. scouleri
parent. Leaves of the hybrid never had the herbaceous texture of the P. cali-
rhiza lineage.

Pinna venation of hybrid leaves incorporates features of both parents: most
veins are free, but occasional anastomoses and areoles occur. Soral size and
development are extremely variable on hybrid leaves and of three general
categories. Sori were 1) as large or slightly larger than those typical of P.
scouleri (> 3 mm), 2) smaller and resembling P. calirhiza sori (< 3 mm), or
3) entirely undeveloped. Translucent, reddish brown scales occur along the
rachis of hybrid leaves, but, in contrast to the deltate and lanceolate parental
scales (P. scouleri and P. calirhiza, respectively), rachis scales on hybrid
leaves are narrowly triangular. In addition, adjacent rachis scales are often
fused (from the base) for approximately one third their length. Paired guard
cells were orbicular in outline, as observed for P. scouleri, whereas subsid-
iary cells showed the distinctive, lobed margins of P. calirhiza. The average
length of guard cells (38 ym), and stomatal densities of approximately 70
stomata/mm are intermediate between parental lineages. All spores on
hybrid plants were shrunken and malformed, suggesting inviability.

MoLEcULAR.—Previous isozymic work by Haufler et al. (1995b) investigated
P. californicum and P. glycyrrhiza (the progenitors of P. calirhiza) and iden-
tified electrophoretic markers for each diploid member, based on sampling
the infraspecific variation from eleven populations. In the present study, iso-
zyme profiles of the Tank Hill and Mt. Sutro individuals of P. scouleri were
verified as representative for the species by sampling other populations
(Table 1). In contrast to the genetic variability detected for other Polypodium
diploids, P. scouleri isozymes were monomorphic across all populations and
for all enzymes sampled. Of the ten enzyme systems considered, four (HK,
PGI, PGDH, and MDH) yielded reproducible, well resolved banding patterns

222 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 3 (2002)

that discriminated individuals representing parental species and hybrids.
Allozymes (allelic variants within loci) are distinguished from isozymes
(products from different loci) by assuming that models of gene product com-
partmentalization (Gastony and Darrow, 1983) are applicable to Polypodium
enzymes.

The following results were obtained for each of the applicable enzyme sys-
tems. Gels stained for the monomeric enzyme hexokinase (HK) showed a
slow-migrating allele in P. calirhiza samples in contrast to the faster allele
present in P. scouleri. The hybrid expressed both alleles, one from each
parental species (Fig. 4a). Two isozyme loci were resolved for phosphogluco-
isomerase (PGI) and phosphogluconate dehydrogenase (PGDH). The faster
migrating isozymes (Pgi 1, Pgdh 1) for both enzymes were monomorphic for
all samples investigated. In contrast, banding patterns for the slower migrat-
ing isozymes (Pgi 2, Pgdh 2) were more variable for each enzyme (Fig. 4b &
4d). Polypodium calirhiza possessed a slower migrating allozyme for Pgi 2,
whereas P. scouleri revealed a faster allozyme. The hybrid exhibited an addi-
tive banding pattern for dimeric phosphoglucoisomerase, possessing both
allelic variants (Fig. 4b). A similar pattern was observed from gels stained for
phosphogluconate dehydrogenase (PGDH). The polymorphic locus (Pgdh 2)
expressed only the faster migrating allele in P. calirhiza, both allelic variants
in P. scouleri, and an additive banding pattern in the hybrid (Fig. 4d). Three
isozymes were revealed for malate dehydrogenase of which two (Mdh 2,
Mdh 3) were monomorphic across all samples. The polymorphic locus (Mdh
1) produced bands that represent a fast allele for P. calirhiza and a slower
migrating allele present in P. scouleri samples. In addition, both intra- and
inter-locus heterodimers were formed during electrophoresis and were subse-
quently revealed by staining for malate dehydrogenase. Bands produced by
the formation of both intra- and inter-locus heterodimers further supported
the hybrid pattern of additive banding from parental species (Fig. 4c).

DISCUSSION

Hybridization between species is a frequent phenomenon in plants and is
especially common in pteridophytes (Wagner, 1968). The clarity and preci-
sion of species recognition and the accuracy of phylogenetic hypotheses can
be enhanced by identifying and characterizing naturally occurring hybrids.
Especially in groups such as the Polypodium vulgare complex, where spe-
cies differences are particularly subtle, unrecognized interspecific hybrids
that usually blend features of the parental individuals can appear to bridge
gaps between otherwise distinct species.

Zymograms were exceptionally informative for delimiting P. scouleri and
P. calirhiza, and for unequivocal verification of hybrid leaves collected on
Tank Hill and Mt. Sutro. Typically, additive banding patterns indicate the
presence of both parental alleles and are observed in hybrids (Crawford,
1990; Murphy et al., 1996). The additive banding patterns visualized on gels

HILDEBRAND ET AL.: A NEW HYBRID POLYPODIUM 223

Pgi2

Mdh1

Mdh2

Mdh3

Pgdh2

Fic. 4. Representative gels stained for enzymes from bis seca —— included in
present study. P. calirhiza = ca; P. calirhiza x scouleri = casc; P. scouleri = sc. eee
HK); B. ee (PG I); C. malate phen (MDH); wad D. phosphogluco-
nate dehydrogenase (PGDH). Sampling in B. & C. corresponds to species as labeled in D. See text
for interpretation of banding patterns.

224 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 3 (2002)

stained for enzymes HK, PGI, PGDH, and MDH combine and confirm paren-
tal contributions from the P. scouleri and P. calirhiza genomes to the hybrid.

The difficulty encountered in developing equivalently distinctive morpho-
logical characterizations of the hybrid individuals requires further discus-
sion. Hybrids are often recognized initially because they have morphological
peculiarities that can signal the amalgamation of two distinct genomes.
Wagner (1962) reviewed deviations from morphological symmetry and their
role as indicators of hybrid origins in ferns. For genera studied (Asplenium,
Cystopteris, Cheilanthes, Osmunda, Polystichum, Pteris, Woodsia), he devel-
oped three broad conclusions regarding hybrid morphology: First, hybrid
structures form symmetrically, but blend traits from parental lineages. If
large differences occur between parental lineages, hybrids tend toward irreg-
ular or asymmetric development. Second, when asymmetric development
does occur, it is retained in hybrids, and may be useful in identifying the
hybrid individual. Third, the discovery of morphological irregularities
should key investigators to the possibility of hybrid origins and further study
of possible parental lineages. Thirty years of further investigation of these
fern genera (e.g., Moran, 1982; Murakami et al., 1999; Yatskievych et al.,
1988; Mickel, 1979; Haufler et al., 1990), in addition to other pteridophytes
(e.g., Montgomery, 1982; Palmer, 1998; Pryer and Haufler, 1993; Tyron,
1968) support the conclusions on hybrid morphology summarized by
Wagner. Likewise, our study reported intermediate, blended traits in P. cali-
rhiza X scouleri that are consistently expressed in all leaves, and that help
to resolve the definition of the hybrid and the identification of parental spe-
cies. In comparison, other character states were not intermediate between pa-
rental lineages, or were so highly variable that they could not contribute to
the definition of P. calirhiza < scouleri (Table 2).

The loss of reticulate venation found in Asplenium and Polystichum
(Wagner, 1962), and other Polypodium hybrids (Whitmore and Smith, 1991)
is also observed in P. calirhiza < scouleri. Whitmore and Smith (1991) in-
vestigated other members of the western P. vulgare complex, and revealed a
loss of vein anastomosis following hybridization. They observed only 0-33%
of the veins per pinna in Polypodium calirhiza anastomose, whereas paren-
tal species P. californicum has weakly to fully anastomosing venation (5—
100%) and P. glycyrrhiza produces pinna with entirely free venation. The
primarily free venation occurring in P. calirhiza x scouleri provides further
evidence for a propensity toward less reticulated venation whenever ge-
nomes differing in this character are present.

Position of fertile pinnae on the leaf (terminal, mid-, lower), often a combi-
nation of parental features in fern hybrids (e.g., Osmunda hybrids, Wagner,
1962), was not a useful diagnostic character for P. calirhiza X scouleri. Poly-
podium scouleri and P. calirhiza fertile segments tend to be positioned ter-
minally in the former, and may comprise all but the lower 1-3 segment pairs
in the latter. Nonetheless, large variation occurs on parental leaves, particu-
larly in P. calirhiza. Likewise, hybrid leaves show great variation ranging
from few, often terminal segments with sori, to all segments producing sori.

HILDEBRAND ET AL.: A NEW HYBRID POLYPODIUM 225

However, the large variation in soral maturation on segments aptly indicates
hybridization, with sori found in all developmental stages, albeit with mal-
formed spores.

Fertile vein development and soral position frequently aid in delimiting
fern hybrids. For example, Polystichum lonchitis produces fertile veins prog-
ressing from the midrib to the margin with sori midway and “dorsal” upon
the vein whereas P. acrostichoides has fertile veins terminating at sori half-
way between the margin and the midrib. A combination of parental traits
occurs in fertile vein development of the hybrid P. acrostichoides x lonchitis
with some sori terminal on fertile veins, other sori dorsal, and some sori
lacking development (Wagner, 1962). In contrast, no distinction in soral
position between P. scouleri and P. calirhiza, as well as the hybrid (when it
occurs), is observed. Sori are terminal on fertile veins of parental species
and P. calirhiza < scouleri. Close observation of parental species failed to
determine differences in the manner by which fertile veins terminated, and,
although P. calirhiza veins appear to end in a more reduced and less club-
like form than those of P. scouleri, this difference may merely result from
differences in leaf texture. Leathery hybrid leaves produce fertile veins more
closely resembling P. scouleri, but, again, this may be an artifact of similar
leaf textures. Sori do occur closest to the costa in P. scouleri whereas they
are located midway between the costa and margin on P. calirhiza pinnae.
The hybrid is highly variable producing sori both near the costa or midway
between it and the pinna margin.

Leaf outlines of hybrids often blend those of parental individuals (Wagner,
1962), but this morphological feature is not transitional in P. calirhiza X
scouleri leaves. Texture, leaf outline, pinna width and apex, as well as sinus
angle and depth are all commonly useful for hybrid identification, but are
not useful in delimiting P. calirhiza X scouleri. The resemblance of hybrid
leaves to P. scouleri may be the most significant factor contributing to past
difficulties in recognizing hybrid populations.

An unexpected discovery of the present study regards the average spore
length for sampled P. scouleri plants from California. Published average
spore length (Haufler et al., 1993) for P. scouleri is less than 53 um, whereas
P. calirhiza spores exceed 58 1m. Indeed, P. calirhiza spores from Tank Hill
and Mt. Sutro measured well within expected values whereas P. scouleri
spores exceeded the expected average (Fig. 3). In an effort to explain the
increased spore size, P. scouleri spores were harvested and measured from
plants in Oregon (Hildebrand #3214, KANU), and, with an average of 55.3
uum, fell within the expected range.

Three explanations may account for the larger than average spores
obtained from the California P. scouleri plants. Certainly, contamination of
spore samples may have occurred. Although spores were removed directly
from sporangia, P. calirhiza spores from dried material may have contami-
nated the P. scouleri plants measured. Two alternative possibilities are more
difficult to assess. Environmental factors could account for the increase in
average spore length found in California populations of P. scouleri. Spore

226 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 3 (2002)

sizes of Isoetes species have been found to be strongly affected by environ-
mental parameters including temperature, solar radiation, and elevation (Cox
and Hickey, 1984). Finally, the increased average spore length observed in
plants of P. scouleri from California could be explained by an increase in
chromosome number. Although not always correlated to ploidy level, in-
creases in spore size may indicate the formation of both auto- and allopoly-
ploids. For example, average spore length was found to increase with ploidy
level by a multiplier of 1.26 in species of neotropical Polystichum (Barring-
ton et al., 1986). The increase in average spore length for sampled California
P. scouleri plants (ca. 15%) may correspond to an increased chromosome
number via autopolyploidy.

To further explore this possibility, guard cells, often positively correlated
with ploidy level (Barrington et al., 1986), were compared between Oregon
and California P. scouleri plants. Guard cell size was previously determined
to predict allopolyploidy in the western P. vulgare complex. Barrington et al.
(1986) found tetraploid P. calirhiza guard cells average 1.12 and 1.2 times
larger than progenitors P. californicum and P. glycyrrhiza, respectively.
Nevertheless, differences in guard cell length between California and Oregon
P. scouleri populations were not observed, and decreased the possibility of
an autopolyploid event in P. scouleri from sampled California populations.

Early cytological investigations by Manton (1951) of three plants identified
as P. scouleri from Point Reyes, California revealed n pairs and n univalents
at meiosis (vs. typically unpaired chromosomes in triploids) and suggested a
closer genetic relationship between Polypodium tetraploids and diploids in
western North America. It now seems most likely that the California plants
Manton investigated were not P. scouleri, but P. calirhiza x scouleri. No
cytological study has been completed for P. scouleri plants from Mt. Sutro
and Tank Hill populations, and efforts to re-locate the Point Reyes hybrid
populations have been unsuccessful.

The present study provides morphological and molecular evidence for the
hybridization of Polypodium scouleri and P. calirhiza in California, and
helps to secure the placement of the former in the western P. vulgare com-
plex of North America. Questions remain regarding the California P. scouleri
populations that merit further investigation. Future cytological studies of
these populations, in conjunction with spore length measurements from
fresh material, may aid in clarifying any remaining ploidy level conun-
drums.

Polypodium calirhiza x scouleri

Stem stout (5-15 mm diameter), occasionally whitish, acrid to slightly
sweet-tasting. Rhizome scales uniformly brown to weakly bicolored with
pale margins, lanceolate to lanceolate-ovate, symmetric, with occasional
teeth on erose margins. Blades to 39 cm in length with a stout petiole to
3 mm in diameter. Lamina stiff and leathery, ovate-lanceolate, pinnatifid,
usually widest at or just above the base, to 20 cm wide: sparsely scaly to

HILDEBRAND ET AL.: A NEW HYBRID POLYPODIUM 227

abaxially glabrescent; rachis sparsely puberulent adaxially. Rachis scales,
concolorous, pale reddish brown, lanceolate-ovate; often adjacent, fused
(from the base) to one third their length. Segments oblong to linear, usually
more than 12 mm wide, with rounded apices, sparsely crenulate margins,
midribs adaxially sell Venation primarily free but with some anastomo-
ses and irregularly formed aeroles. Sori oval to circular, but with widely
varying development, usually closer to midrib than margins, 1-4 mm in di-
ameter, producing malformed spores. Sporangiasters absent.

ACKNOWLEDGMENTS

We thank R. Cranfill for initial field recognition of atypical leaves, and A. R. Smith for his pre-
liminary cytological examination of collected material.

LITERATURE CITED

BARRINGTON, D. S., C. A. Paris AND T. A. RANKER. 1986. Systematic inferences from spore and sto-
mate size in ferns. Amer. Fern J.

CHARLESWORTH, D. AND B, CHARLESWORTH. 1987. Inbreeding depression and its evolutionary conse-
quences. Ann. Rev. Ecol. Syst. 18:

Cayton, J. W. AND D. N. TRETIAK. 1972. ke ae buffers for pH control in starch gel electro-
gan J. Fish. Res. Board of Canada 29:1169-1172.

Cox, P. A. aNp R. J. Hickey. 1984. Convergent megaspore evolution and Isoetes. Amer. Naturalist
i nei

Crawrorp, D. J. 1990. Plant molecular systematics: macromolecular approaches. John Wiley &
Sons. New fee New Yor

Gastony, G. J. AND D. C. Darrow. 1983. arg ere and cytosolic isozymes of the homosporous

Athy ae -femina L. Amer. J. Bot. 70:1409-1415.

preeny C. H. 1985a. Enzyme variability and sas of evolution in Bommeria (Pteridaceae).
Syst. Bot. 10: aC 104

. 1985b. Pteridophyte evolutionary biology: the electrophoretic approach. Proc. Roy. Soc.

Edinbur gh 86:3
eres po RbcL sequences provide apa insights among sister spe-
cies ste iis forn genus lige um. Amer. Fern J. 85:361—-374.

D. E. Sottis aNp P. S. Sotis. 1995a. Phylogeny af tha Polypodium vulgare complex: in-
hts oe sae nan restriction site data. Syst. Bot. 20:110-11
R. Wanc. 1991. Chromosomal analyses and the contest of allopolyploid
Polpodium cate (Polypodiaceae). Amer. J. Bot. 78:624-629.
WINDHAM. . New species of North American gi and Polypodium,
ae enonidents on their reticulate relationship. Amer. Fern J. 81
, M. D. WinpHaM, F. A. LANG AND S. A. WHITMORE. 1993. ee. Pe 315-323 in Flora
of North America Editorial Committee. Eds. Flora of North America north of Mexico.
Oxford Ache Press, New Yor
M. INDHAM AND E. W. ee “1995b. Reticulate evolution in the Polypodium vulgare
compl = dd Bot. 20:89-10
WINDHAM AND T. A. te 1990. an analysis of the Cytopteris tennes-
seensis = complex Ann. Missouri Bot. Gard. 77:314—329.
Lioyp, R. M. anp F. A. Lanc. 1964. The Priel tg vulgare complex in North America. Brit.
Fern Gaz. es 168-17
Manton, I. 1950. Polpoodiud vulgare. Chapter 8 in Problems of trad and evolution in the
Pteridophyta. Cambridge University Press, Cambridge, Great Bri
——. 1951. The cytology of Polypodium in America. Nature 167: ns

228 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 3 (2002)

eer P 1979. The fern genus Cheilanthes in the continental United States. Phytologia
—437.

eee J. D. 1982. Dryopteris in North America. Part II. The hybrids. Fiddlehead Forum

anada. Eee Fern J. 72:5—

MurakaMI, N., S. Nocami, M. WATANABE AND K. IWATSUKI. esi i ioe of Aspleniaceae
inferred from rbcL nucleotide sequences. Amer. Fern J. 89:232—

Murpuy, R. W., J. W. Sires Jr., D. G. Bur AND C. H. HAuFLer. ou sel isozyme electropho-
resis. Pp. 51-120 in D. M. Hillis, C. Moritz, and B. K. Mable. Eds. Molecular Systematics
(2nd Edition). Sinauer Associates, Inc., Sunderland, MA

Pater, D. D. 1998. Cibotium x heleniae hyb. nov. (C. chamissoi < C. menziesii; Cyatheaceae):
a eres occurring hybrid from Oahu, Hawaii. Amer. Fern J. 88:150-154.

Pryer, K. M. AND C. H. HAur.er. 1993. Peete and chromosomal evidence for the allote-
ape ere of the common oak fern Gymnocarpium dryopteris (Dryopteridaceae). Syst.
Bot. 18 72

Suivas, M. G. 1961 disesag to the cytology oo apa of Polypodium in Europe and
America. II. Taxonomy. J. n. Soc., Bot. 58:2

So.tis, D. E., C. H. HAvuF.er, D. C. as ow AND G. J. i ASTONY. 1983. Starch gel electrophoresis in
erns; a counjeilatien: of grinding buffers, gel and electrode buffers, and staining schedules.
Amer. Fern J. 7 27

SOLTIS, S., D. E. SO $ AND P. G. Wo r. 1990. oe divergence in North American Polysti-

m(Dryoperdacoae Syst. Bot. 15:205-215

Tyron, “ F, Comparisons of sexual and inant: races in the fern genus Pellaea.
Rhodora o “a

Wacner, W. H. 1962. Tragulae morphological development in hybrid ferns. Phytomorphology

12:87-100.
. 1968. Hybridization, taxonomy, and evolution. Chapter 9 in V. H. Heywood. Ed. Modern
nuethods in _ taxonomy. Academic Press, London.

Wuitmokg, S. A. AND A. R. SmiTH. 1991. Recognition of the noe Polypodium calirhiza (Poly-
podiaceae), in western North America. Madrofio 38:233-24
YATSKIEVYCH, G., D. G. STEIN AND G. J. GASTONY. 1988. pet lata DNA evolution and Plasto
. rhea (Dryopteridaceae) and related fern genera. Proc. Natl. Acad. Sci. U.S.A.
89-25

American Fern Journal 92(3):229-238 (2002)

Comparative Research of Gametophytes of Olfersia
alata and Olfersia cervina (Dryopteridaceae)

ANICETO MENDOzA, BLANCA PEREZ-GarciA, and RAMON Ripa!
Departamento de Biologia-Botdnica Estructural y Sistematica Vegetal, Universidad Autonoma
Metropolitana—Iztapalapa, Apdo. Postal 55-535, 09340, México, D. F. Fax: (55) 58 04 46 88

AsstTract.—The prothallial development of gametophytes of Olfersia alata and Olfersia cervina

s
(Dryopteridaceae) is described and compared. Spores are monolete, ellipsoid, and with broadly

winged perispore. Germination is Vittaria-type and the prothallial development is Aspidium-
type. Adult gametophytes are cordiform-spatulate to cordiform-reniform, with marginal and

These two species share some features with some species of Arachniodes, Cyrtomium, Dryo
ris, Phanerophlebia, and Polystichum, such as type of germination and prothallial development
and trichomes. They differ from Didymochlaena truncatula, which has prothallial development
of the Adiantum-type and lacks trichomes on the sexual phase.

The genus Olfersia Raddi (Dryopteridaceae), has two species: Olfersia
alata C. Sanchez & Garcia Caluff and Olfersia cervina (L.) Kunze. Olfersia
alata is endemic to Cuba; its main characteristics are all sterile pinnae have
decurrent bases, and fertile leaves which are smaller and have fewer pinna
pairs than the vegetative leaves. It grows in mountainous mesophytic forests,
between 350—400 m (Sanchez et al., 1991). Olfersia cervina is widely distrib-
uted in the tropics, from Southern Mexico (Chiapas, Oaxaca, Veracruz), to
Southeastern Brazil and the West Indies. In this species the bases of the ster-
ile pinnae are not decurrent onto the rachis, and pinnae are short-petiolu-
late. It grows between 450-1000 m in damp tropical forests, on rocky and
very shady banks (Moran, 1986, 1995; Riba and Pérez-Garcia, 1999). Both
taxa are usually terrestrial, rarely hemiepiphytic, with a short trailing rhi-
zome. Leaves are markedly dimorphic, sori are exindusiate and linear to ob-
long, and spores are monolete, echinulate with a broad perispore.

This paper complements existing information about the morphogenesis of
the gametophytic phase of dryopteriod ferns and, particulary, focuses on
gametophytes of Olfersia. We hope to contribute in this way to the knowledge
of the sexual phase of Mexican ferns.

MATERIALS AND METHODS

Spores of O. cervina were collected from living plants from the following
site: Lote 69 in “Los Tuxtlas” Biological Station, between Laguna Azul and
Laguna Seca, Mun. of San Andrés Tuxtla, in the state of Veracruz, Mexico;
vouchers are in UAMIZ (AMR 202 and AMR 251). Spores of O. alata were

230 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 3 (2002)

collected by Carlos Sénchez and L. del Risco (77825) near Farallones de Moa,
Farallon Redondo and La Escondida, Mun. Moa, Holguin province, Cuba;
the voucher is in HAJM.

Fertile pinnae were kept in paper bags until spores were shed. Subse-
quently, the spores and shed material were passed through a sieve (with
pores 0.074 mm in diameter) in order to eliminate traces of sporangia and
indusia. Spores of each species were sown at an average density of 150-200
spores per cm’ in two small pots with a mixture of black soil and organic
matter and in 30 Petri dishes, 5 cm in diameter, containing Thompson’s
solution of mineral salts and agar on a sterile nutrient medium (Klekowski,
1969).

Petri dishes and pots were kept inside transparent plastic bags in order to
avoid contamination and desiccation, with a photoperiod of 12 h light/dark-
ness, with artificial light (75 Watt lamps, daylight) and a temperature of 23-
25° C (Mendoza et al., 1999a, 1999b). Two dishes were kept in darkness in
order to determine photoblastism. After 100 days, none of the spores grown
in darkness had germinated.

All pictures of microscopic material were taken from living material grown
in the laboratory.

RESULTS

Spores of both species are monolete, nearly spherical, with a light brown
perispore. Spores of Olfersia alata measure (64) 73 (83) x (49) 53 (55) um,
including the winged perine around the spore; the perispore measures (15)
16 (20) jm wide (Fig. 1). Spores of O. cervina measure (44) 48 (51) < (37) 39
(40) um, also including the winged perine which measures (5) 6 (8) um wide
(Fig. 2). Spores of O. alata are larger than those of O. cervina primarily due
to the size of the perispore. These measurements were obtained from an
average sample of fifty spores per species.

Germination is Vittaria-type (Nayar & Kaur, 1971) in both species. In
Olfersia alata germination began 20-23 days after spores were sown, whereas
in O. cervina it began 8-12 days after sowing. Gametophytes of both species
first develop a rhizoid, which is short, hyaline, and without chloroplasts.
The first prothallial cell is short and oval; division begins in this cell with a
transverse wall and ultimately forms a short germ-filament, 2—4 cells long.
This filament eventually ends in an apical trichome. During this stage of
development, the spores retain their coat (Figs. 3-5).

In O. alata, the prothallial plate begins to develop approximately 25 days
after spore germination from intercalary cells of the filaments which undergo
longitudinal divisions (Figs. 6-7). In some cases, the terminal cell of the fila-
ment, after producing a trichome, will divide longitudinally in such a man-
ner that the trichome is placed over one of the daughter cells, which will
remain inactive until other cells develops into a gametophytic plate. This
plate, from which a meristematic cell will emerge, is usually asymmetric

MENDOZA ET AL.: GAMETOPHYTES OF OLFERSIA

phases of Olfersia. 1. Spor
cervina (8 days). 4. O.

hallial

and filamentous and laminar
s of germination. 3.

prot

1-9. Spores, germination,
Spore, O. cervina. 3-4. I Initial stages
(22 Sie 5 7. Germ filament. 5-6. O. alata (22 days). 7. O. cervina ye days). 8-9. Young lami-
9. O. cervina (35 day e = cover of spores, cp =
trichome, zm cietitiaiathe zone.

nar phases 8. O. alata (38 days
cell, cr = rhizoid cell, p = perispore, t =

232 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 3 (2002)

Fics. 10-16. Laminar arp tae and secretory, unicellular, capitate trichomes of Olfersia.

10-11. O. cervina (41 and 100 days). 12-13. O. alata (95 and 138 days). 14-16. Trichomes. 14.
O. cervina (175 days). = O. alata (149 days). 16. O. cervina (175 days). a a = archegonia, cse =
extracellular secretion cover, t = trichome.

with an apex that continues to change and form a notch. Finally, after 90—
100 days, the prothallial plate becomes cordate, the so-called Aspidium-type
prothallial plate development (Nayar & Kaur, 1969; Figs. 8-10). Afterwards,
a cushion bearing the gametangia forms and the adult gametophyte is cordi-
form-spatulate with many marginal and superficial trichomes. The formation

MENDOZA ET AL.: GAMETOPHYTES OF OLFERSIA 233

Fics. 17-20. Adult gametophytes and gametangia of Olfersia. 17-18. Adult phases of O. alata

(109-158 days). 19-20. Gametangia. 19. Antheridium, O. cervina (246 days). 20. Mouths of

archegonium, O. alata (138 days). ba = mouth of archegonium, cb = basal cell, cm = ring cell,
= opercular cell.

of the prothallial plate in O. cervina takes less time, beginning on day 15.
The pattern of prothallial development is the same as in O. alata. The first
adult cordiform gametophytes are completely differentiated 60-80 days after
the spores were sown (Figs. 12-13).

Trichomes are unicellular, capitate, and secretory (Fig. 14). In O. alata
they measure approximately 36 um long by 23 um wide at the base. The api-
cal third of the trichome is globose, 17 1m high by 24 um wide, with a thin
cover of extracellular secretion ca. 3 pm thick (Fig. 15). Olfersia cervina tri-
chomes are 34 pm long by 20 ym wide at the base and the apical third of the
trichome is globose, 21 pm high by 26 pm wide, with an extracellular secre-
tion 8 ym thick (Fig. 16). These measurements are from mature trichomes,
found at the middle basal region of the gametophytes.

The gametangia are typical of leptosporangiate homosporous ferns. They
begin differentiating between days 120-244 in O. cervina while in O. alata
they develop between days 100-150. Antheridia are distributed on the lower
surface of the plate on the basal half of the cushion (Fig. 19). Antheridia are
globose and consist of a basal cell, a ring cell, and an opercular cell. These
three cells surround the androgenous cell.

In both species, the necks of the archegonia, have four tiers of neck cells.
Archegonia are found on the central region of the plate, on the cushion, and

TasLeE 1, Comparison of different stages of the prothallial development of Olfersia alata, and O. cervina with other genera and species of
Dryopteridaceae.

Type of prothallial

Type of Filamentous evelopment and
Spores germination phase adult form Trichomes Antheridium Archegonium
'*Arachniodes —_ Monolete, Vittaria Long filaments Aspidium, Unicellular, 3 cells 4 rows of cells,
with perispore, -5 cells), cordiform capitate, with each row with
measuring 3 with an apical gametophytes with a thin coat of 4-6 cells
42 «um trichome lacerate margins extracellular
secretion 2 :m
*Cyrtomium Monolete with Vittaria Long filaments Aspidium, Unicellular, 3cells 4 rows of cells
perispore, (2-5 cells), cordiform capitate
measuring 32 with an apical = ga metophytes, with
45 um trichome lacerate margin
*Didymochlaena_ Monolete with Vittaria Short filaments Adiantum, Absent 3-4 cells 4 rows of cells,
perispore, (2-3 cells), cordiform- throughout each row with
measuring 30 apical tri- reniform development 4-6 cells
37 um chome absent gametophytes with
entire
argins
*Dryopteris Monolete with Vittaria_ Long filaments Aspidium, Unicellular, 3cells 4 rows of cells,
perispore, (2-5 cells), po capitate each row with
measuring 36 with an heii ifor 4-5 cells
51 pm trichom gametophytes with
lacerate margins
Olfersia alata Monolete with a Vittaria Short filaments Aspidium, Unicellular, 3cells 4 rows of cells,
broad1 (2—4 cells), spatulate- capitate with each row with
winged with an apical _—_cordiform an extra- 4-5 cells
perispore, trichome gametophytes cellular
measuring 53 with entire secretion coat
73 pm margin i

vEC

(200z) € WAGON 26 SNNTIOA “TVNYNOl NYaA NVOMANV

TABLE 1.

Continued.

Type of prothallial

Filamentous development and
Spores germination phase orm Trichomes Antheridium = Archegonium
O. cervina Monolete with Vittaria Short filaments Aspidium, Unicellular, 3 cells 4 rows of cells,

i 2—4 cells), spatulate- apitate wi each row with
perispore, with an apical _cordifor extra- 4-5 cells
measuring 39 trichome iia with cellular
48 pm ee margins secretion coat

8 :m thick
'Polystichum Monolete with Vittaria Long filaments | Aspidium, Unicellular, 3-4 cells 4 rows of cells
perispore, cordiform appilate
measuring 34 without gametophytes with — capitate
45 pm trichomes lacerate margins secretors and
non secretors
*Phanerophlebia Monolete with Vittaria Long filaments Aspidium Unicellular, 3-4 cells 4 rows of cells,
perispore, —6 cells), PRE HN -cordiform capitate, wit each row with
measuring 25 with an apical gametophyes with an extra- 4-6 cells
33 pm trichome lacerate margins cellular
secretion coat
5 :m thick

' Chandra & Nayar 1970, * Mendoza et al. 1999a, 1999b,” Pérez-Garcfa et al. 1999. * Mendoza, unpublished.

VISHAITO AO SHLAHdOLAWVD “TV LY VZO0NEW

Sez

236 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 3 (2002)

near the meristematic zone. The necks are oriented toward the basal region
of the gametophytes (Figs. 13, 20). Two hundred days after sowing the
spores, the young sporophytes had not yet formed.

DISCUSSION AND CONCLUSIONS

There is literature dealing with the morphology of the gametophytic phase
of ferns closely related to Olfersia, of the Dryopteridaceae (both Old and
New World), e.g., Arachniodes, Cyrtomium, Didymochlaena, Dryopteris, and
Polystichum (Atkinson, 1973; Chandra & Nayar, 1970; Cousens, 1975; Kaur,
1977; Mendoza et al., 1999a, 1999b; Pérez-Garcia et al., 1999: Stokey &
Atkinson, 1954).

Spores of O. alata average 73 X 53 «um, including the winged perispore;
spores of O. cervina average of 48 X 39 um. Spores of O. alata seem much
larger, but in reality, if the perispore is not considered, the spores are 38 X
23 um, and the winged perine is 16 um wide or more in its widest part.
Spores of O. cervina are 39 X 31 wm and the perispore is approximately 6
jum wide at its widest point and tends to be more spherical, which is an in-
dication that the spores of O. alata are a little smaller than those of O. cervi-
na. (Figs. 1, 2).

Both species share the same germination pattern, Vittaria-type, which is
the most common type in ferns. In this type, the rhizoid develops first after
the formation of a wall perpendicular to the polar axis of the spores. Eventu-
ally, the first prothallial cell divides by means of the formation of a perpen-
dicular wall thus giving rise to two cells. The apical cell then divides again,
giving rise to a short filament 2-4 cells long. The time for germination differs
between these two species; spores of O. cervina germinate faster (8-12 days)
compared to spores of O. alata (20-22 days).

Prothallial development in both species is of the Aspidium-type in which
the germ filament commonly ends in a trichome, and the prothallial plate
is formed by the activity of the intercalary cells of the filament. The adult
gametophyte develops faster in O. cervina (60-80 days) than in O. alata
(90-100 days).

Trichomes differ in size and in the thickness of the extracellular secretion;
the longest ones, belongins to O. alata (36 X 23 um), have a thinner extra-
cellular secretion (3 um), whereas trichomes of O. cervina are slightly shorter
(34 X 20 um) and have a thicker (8 um) extracellular secretion.

Olfersia alata and O. cervina share features with the following dryopterid
genera: Arachniodes, Cyrtomium, Dryopteris, Phanerophlebia, and Polysti-
chum (Atkinson, 1973; Chandra and Nayar, 1970; Cousens, 1975; Mendoza
et al., 1999b; Pérez-Garcia et al., 1999). These genera all have monolete
spores with perispore, a Vittaria-type germination pattern and an Aspidium-
type prothallial development. However the two Olfersia species differ from
the rest in the shape of their trichomes, which are short and wider at the
base, capitate, with a globose apex, and with a dense extracellular secretion.

MENDOZA ET AL.: GAMETOPHYTES OF OLFERSIA 237

The gametophyte margins are entire in Olfersia, compared with the lacerate
margins of species of the other genera. These other genera also have longer,
capitate trichomes, with very thin extracellular secretions distributed on the
lacerate margins and on the surfaces of the plate (Table 1).

Olfersia alata and O. cervina, together with the above mentioned taxa,
share some features with Didymochlaena truncatula, such as the monolete
spores and Vittaria-type germination. This last species differs from the rest
in that is has a prothallial development of the Adiantum-type, characterized
by a differentiation of an apical meristematic cell during the early stages of
the plate’s formation. The gametophytes of Didymochlaena, are completely
glabrous throughout their development, in contrast to the other species of
Dryopteridaceae mentioned.

d on our results, we conclude that Olfersia alata and O. cervina share
characteristics such as the Vittaria-type germination pattern, Aspidium-type
prothallial development, and unicellular capitate trichomes with a uniform
extracellular secretion. These same characteristics are characteristic of spe-
cies of Arachniodes, Cyrtomium, Dryopteris, Phanerophlebia, and Polysti-
chum (Table 1). The most common feature of all of these genera is the
development of an apical trichome during the filamentous stages of prothal-
lial development. Gametophytes of Didymochlaena truncatula differ from
these genera in having prothallial development of the Adiantum-type and
lacking trichomes. Finally, with the exception of Didymochlaena truncatula
we did not find important differences among the different taxa of the Dryo-
pteridaceae.

ACKNOWLEDGMENTS

This paper is part of an MSc (Plant Biology) thesis, Morfogénesis de la fase sexual de pterido-
fitas mexicanas, familia et aan eb Adee Sgn of the sexual phase of Mexican Pteri-
dophytes, family Dryopteridaceae”), written by the first author, developed under the supervision

his photographic assistance, and the anonymous reviewers for their suggestions and criticisms.

LITERATURE CITED

ATKINSON, L. M. 1973. The Gametophyte and Family Relationships. In: A. C. Jermy, J. A. Crabbe
& B. A. Thomas (Eds.). The Phylogeny and Classification of The Ferns. J. Linn. Soc. Bot.
a No.1, 67:73-90.

CHANDRA, P. & B. K. NAyar. 1970. Morphology of Some hg ge shane i. ny Gametophytes

Arachniodes, Cyrtomium and Polystichum. J. Linn. Soc. B 3:265—
Cousens, M. I. 1975. Gametophyte Sex Expression in Some ones ot ciehoni Amer. Fern. J.
0:13-—27.

Kaur, S. 1977. Morphology of a yee and Juvenile Sporophytes of Some Species of Dry-
opteris. Proc. Indian Acad. S$ —171.

— E. J., JR. 1969. eae pe of the Pteridophyta. II. A study of the Blechna-
ceae. J. Linn. Soc. Bot. 62:361-377

238 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 3 (2002)

MEnpoza, A., B. PEREZ-Garcia & R. Ripa. 1999a. Morfologia y anatomia del gametofito de Didymo-
chlaena oe (Dryopteridaceae). Rev. Biol. Trop. 47:87—93
. 1999b. Morfogénesis = la fase sexual del lisleaho Arachniodes denticu-
ta Dryopteridacen) Rev. Biol. Trop. 47:791—797.
mers 2 G. 6. The neotropical fern ie ski in Fern J. 76:161-17
1995. achat meri Pages 210-226. In: R. C. Moran & R. Riba (eds.). Ae Mesoameri-
cana, Vol. 1: Psilotaceae a Salviniaceae. Instituto re Biologia, Universidad Nacional

o; DF
Nayar, B. K. & S. Kau ee is of prothallial development in homosporous ferns. Phyto-
ee, 19: ote
971. Cametopytes of homosporous ferns. Bot. Rev. 37:295-396.
PE eR B., (l MENpoza, I. REYES & R. Ripa. 1999. Morfogénesis de la fase sexual de seis es-
pecies geet de helechos del género Dryopteris Fries aie araigeh Rev. Biol. Trop.

47:63-—
Ripa, R. & B. Vee oe pts Dryopteridaceae. Flora de México, Consejo Nacional de la Flora
e México oe
SANCHEZ-VILLAVERD oe ciA-CALUFF & C. ZAVARO-PEREZ. 1991. Nueva onciigee cubana del
género Olfersia ect eesalie esate ey Fontqueria 31:229-23
Story, A. G. & L. R. ATKINSON. 1954. The gametophyte of Didymochlaena hee Desv. Phyto-

morphology 4:310-315

American Fern Journal 92(3):239-—246 (2002)

SHORTER NOTES

Botrychium hesperium in the Wallowa Mountains of Oregon.—The
Wallowa Mountains of northeastern Oregon boast the greatest fern diversity
in the state. We reported 47 taxa in the range (Zika & Alverson, Amer. Fern
J. 86: 61-64. 1996), which included 14 taxa of Botrychium. A number of ele-
ments from the Rocky Mountains are found in Wallowa County, to which
we can now include Botrychium hesperium (Maxon & R. T. Clausen) W.H.
Wagner & Lellinger, an addition to the Oregon flora (Wagner & Wagner, Ophio-
glossaceae in Flora of North America, Vol. 2, Oxford Univ. Press, 1993).

Botrychium hesperium is restricted in the Wallowa Mountains to a narrow
elevational band in the Lostine River drainage, between 1535-1660 meters,
where steep canyon walls shade the valley floor from direct sunlight early
and late in the day. It is found in mesic meadows or forest edges, in full sun
or partial shade, at all aspects, but only on gentle slopes or flats on the valley
floor. It has yet to be located on steep slopes at higher elevations. The forests
are primarily Pinus contorta Dougl. ex Loud., with low wet areas dominated
by Picea engelmannii Parry ex Engelm. Associated herbs include: Achillea
millefolium L., Agoseris aurantiaca (Hook.) Greene, Antennaria rosea Greene,
Calamagrostis rubescens Buckl., Carex concinnoides Mack., C. geyeri Boott,
C. hoodii Boott, Elymus glaucus Buckl., Erigeron compositus Pursh, Festuca
occidentalis Hook., Fragaria virginiana Duchesne, Gentiana amarella _L.,
Hieracium albiflorum Hook., Linnaea borealis L., Sedum stenopetalum Pursh
subsp. stenopetalum, Senecio pseudaureus Rydb., and Viola adunca Sm. It
grows with Botrychium lanceolatum (Gmel.) Angstr. subsp. Janceolatum.,
B. minganense Victorin, B. paradoxum W. H. Wagner, B. pedunculosum
W. H. Wagner, and B. pinnatum St. John. The sites are valley bottom Quater-
nary surficial deposits, locally reworked by the Lostine River or small tributar-
ies. Adjacent slopes are sedimentary bedrock in the Triassic/Jurassic Hurwal
Formation. In places the upper west wall of Lostine Canyon is granite, and
the east wall is pure limestone of the Martin Bridge Formation. It is possible
that all or most of the Botrychium populations are influenced by basic or cir-
cumneutral groundwater percolating through calcareous glacial till or morai-
nal debris. It may be no coincidence that the richest diversity and greatest
abundance of Botrychium species are found in the calcareous canyons of the
Wallowa Mountains, rather than in the granitic or volcanic basins.

We are aware of four extant populations of Botrychium hesperium in the
Wallowas. The Oregon range of the species is included in ca. 5.5 km of river
valley. The total known number of plants at this time is less than 100, and
they face threats from fire suppression, pack animal grazing, wood-cutting,
and recreation-associated activities, despite the fact that most or all plants
are within the Lostine River Wild and Scenic River corridor, a part of the
Eagle Cap Ranger District of the Wallowa-Whitman National Forest.

240 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 3 (2002)

Collections of Botrychium hesperium were first made in 1981 (W. H. Wagner
81130 MICH), with later collections in 1991 (Zika & Alverson 11295 WTU),
1992 (Zika & Alverson 11794 WTU), 1993 (Wagner et al. 93047 MICH) and
1996 (Zika & Alverson 12908 OSC). We were puzzled by these plants for
many years, and thought they might represent an undescribed taxon, related
to B. hesperium, but with slightly angular upper pinnae and shorter basal
_pinnae. This was a false impression, based in part on the large Wallowas
plants growing in sheltered or partly shaded sites, and based on a limited
sample of B. hesperium from Oregon and elsewhere. To get a better idea of
variation in B. hesperium, we studied large living populations in Montana,
Arizona and Colorado. Finally, as we saw more Oregon plants, we con-
cluded they were part of the natural variation of B. hesperium, united by
their grayish-green color, exaggerated and asymmetrical basal pinnae, broad
rounded upper pinnae, and ample sporophores.

We are pleased to acknowledge our funding sources for fieldwork: the
Native Plant Society of Oregon, the Oregon Natural Heritage Program, and
the Wallowa-Whitman National Forest. We are grateful for specimens and
discussions of B. hesperium, provided by Peter Root, Peter Lesica, Kathy
Ahlenslager, and Don Farrar.—PrTrer F. ZIKA and Epwarp R. ALVERSON,
Herbarium, Dept. of Botany and Plant Pathology, Oregon State University,
Corvallis, OR 97331, and Warren H. Wacner (deceased) and FLORENCE S.
Wacnrr, Department of Ecology and Evolutionary Biology, University of
Michigan, Ann Arbor, MI 48109.

A Binomial for the Hybrid Polypodium of Eastern North America.—Two
species of Polypodium (Polypodiaceae) occur in eastern North America, the
diploid P. appalachianum Haufler & Windham and the tetraploid P. virginia-
num L. These species hybridize, producing a sterile triploid recognized by
its abortive spores and intermediate morphology. The differences between
these three taxa are well described by Haufler and Wang (Amer. J. Bot.
78:624—629. 1991) and Haufler and Windham (Amer. Fern J. 81:7—23. 1991).
The triploid hybrid so far has been found only on the Appalachian Plateau
where P. appalachianum and P. virginianum are sympatric. The hybrid
has been documented so far in Ontario, Canada and eight states: New
Hampshire, New Jersey, North Carolina, Ohio, Pennsylvania, Tennessee,
Vermont, and Virginia (Evans, Research Div. Monograph 2. Virginia Poly-
technic Inst. and State Univ., Blacksburg, VA. pp. 117-146. 1970; Haufler &
Wang, op. cit.; Montgomery, Bartonia 59:113-117. 1996). Kentucky and West
Virginia can be added to this distribution, based upon specimens at OS and
WVU, respectively. The hybrid likely will be documented in other states
and provinces as well. Indeed, the triploid may prove rather frequent, as
shown for New Jersey and Pennsylvania by the work of Montgomery cited
above.

SHORTER NOTES 241

It seems appropriate and practical that this widespread hybrid have a
binomial. Perhaps providing this taxon with an epithet may raise botanists’
awareness of this taxon and spur future discoveries and understanding of
this hybrid.

Polypodium Xincognitum Cusick, fAybr. nov.—Holotype: Ohio, Meigs
County, sandstone exposures on mesic slope above Leading Creek, Co Rt 10,
0.25 mi (0.02 km) SW of Twp Rt 27, N of Dexter, Sect 6, Salem Twp, 6 Aug
1985, Cusick 24620, OS; Isotypes, MICH, MU, NY.

Hybrida e Polypodium appalachianum et P. virginianum exorta, aliis char-
acteribus inter parentes media, sporis abortivus.

My research was supported in part by the Division of Natural Areas and
Preserves, Ohio Department of Natural Resources.—ALLISON W. Cusick, Divi-
sion of Natural Areas and Preserves, Ohio Department of Natural Resources,
1889 Fountain Sq. Ct., F-1, Columbus OH 43224.

Lycopodium lagopus New in West Virginia.—West Virginia is a southern
outpost for many boreal species (e.g. Larix Jaricina in Preston County) that
were stranded in the state’s highlands and arctic-like bogs following the last
glacial retreat (P.D. Strausbaugh and E.L. Core, Flora of West Virginia, Mor-
gantown WV, Seneca Books, 1997). Along the Allegheny Front, elevations
reach 1482 m (Spruce Knob) and there are ten peaks over 1430 m. Lycopodium
lagopus (Laestradius ex C. Hartman) G. Zinserling ex Kuzeneva-Prochorova,
(Fl. Murmansk Obl. 1:80, 1953), generally more northern in its distribution,
was recently located here as well. A small, but thriving population grows on
the site of a coal strip mine, now used as a cross country ski trail in Blackwater
Falls State Park, Tucker County, at an elevation of about 1070 m. Its sister spe-
cies, L. clavatum, is also here in abundance, but the two lycopods remain dis-
tinct; L. lagopus features single strobili on slender peduncles, a more compact
growth habit, more appressed and shorter leaves, and sporophylls that taper
gradually to a hair tip.

Lycopodium lagopus (formerly L. clavatum var. monostachyon Hooker and
Greville) goes by the apt common name “‘one-cone club-moss”’ (Flora of North
America, New York, Oxford Univ. Press, 1993). It shares many characters
with the common club moss, L. clavatum, e.g., general growth and branching
patterns, stalked strobili, and hair-tipped leaves, but L. clavatum has multi-
ple strobili (typically two) on most of its peduncles, spreading and longer
leaves, and sporophylls that end abruptly in hair tips. No hybrids are docu-
mented between these closely related species, nor, for that matter, between
any species in the genus Lycopodium s.s. This is in sharp contrast to the
many hybrids described since 1956 within the related genera Lycopodiella,
Huperzia, and Diphasiastrum (J. Eiger, Biol. Rev. City Coll. 18:17-—22, 1956;
Flora of North America).

As a boreal plant, L. lagopus occurs from Alaska to Newfoundland, Green-

242 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 3 (2002)

land, Scandinavia, and northern Eurasia. In the contiguous 48 states it has
been reported from Maine, Michigan, Minnesota, Wisconsin, New York,
New Hampshire, Vermont, and now West Virginia. Michigan would be the
closest known neighbor of the West Virginia population, about 650 km dis-
tant (Flora of North America, cited above). The West Virginia site is the flat
top of an abandoned coal strip mine within Blackwater Falls State Park, near
the town of Davis. The park was established in 1937, and mining operations
within its borders ceased prior to that. Since that time a remarkable assem-
blage of plants has reclaimed the coal spoils piled along the mine highwall.
In cool ravines there is Tsuga canadensis, while on the open, sunny, coal-
strewn areas Picea rubens, Acer rubrum, Rhododendron maximum, Kalmia
latifolia, and Vaccinium species dominate the woody flora. Three orchids are
prominent among the herbaceous plants- Cypripedium acaule, Platanthera
clavellata, and Spiranthes cernua. Many grass, sedge, and Sphagnum species
grow in the boggy soils near two ponds at the intersection of the Dobbin
House and Woodcock ski trails in the area of L. Jagopus. An impressive
group of fern allies has also reclaimed this disturbed, acidic habitat, includ-
ing Lycopodiella inundata on moist soil near the ponds. Diphasiastrum digi-
tatum, D. tristachyum, and their hybrid D. Xhabereri are abundant on
exposed tailings. Lycopodium obscurum, L. dendroideum, and L. hickeyi al-
so occur along the wooded edges. And, the aforementioned L. clavatum is
found in nearly all surrounding habitats. Several Dryopteris species and
Pteridium aquilinum (with some rare, fertile colonies) are also common in
the immediate area. Asplenium montanum grows on granite rocks along the
Pase trail nearby, and Vittaria appalachiana (Appalachian gametophyte) is
spreading under a sandstone ledge near the prominent waterfall for which
the 683 hectare (1688 acre) park is named.

The Lycopodium lagopus colony consists of about a dozen long rhizomes,
three occurring on exposed soil adjacent to the Woodcock Trail and the rest
in a protected area about 6 m into low spruce woods beyond the trail. The
colony is probably clonal and is quite fertile, nearly all upright shoots bear-
ing characteristic single strobili when the site was surveyed in mid-July,
2001. A voucher specimen of one fertile, upright shoot was collected for de-
posit with the herbarium of the Carnegie Museum in Pittsburgh, PA (sheet
No. 494379, CM). The origin of the ‘‘one-cone club-moss” here is uncertain,
but it is hardly the only disjunct, rare pteridophyte known in West Virginia.
The western species Asplenium septentrionale and Cheilanthes eatonii
(C. castanea) occur on shale in Hardy and Monroe Counties (W. H. Wagner,
Jr., Ann. Missouri Bot. Gard. 59:203-217, 1972).—JoAN EIGER GOTTLIEB, 2310
Marbury Road, Pittsburgh, PA 15221.

SHORTER NOTES 243

Marsilea mutica in Virginia.—Of the six species of water-clover, Marsilea,
described in Flora of North America, five are native and M. quadrifolia L. is
introduced in much of the northeastern United States. Johnson (pp 332-333,
in FNA ed. comm., Flora of North America, vol. 2, 1993) suggested that M.
quadrifolia has been deliberately planted as a curiosity because many local-
ities are artificial bodies of water. A seventh species, the paleotropical Marsi-
lea minuta L., has been collected in Florida (Burkhalter, Sida 16:544—-549,
1995). We document here the apparent establishment of a previously un-
reported water-clover in southeastern Virginia, M. mutica Mettenius, which,
unlike most species in the genus, can be readily identified in sterile condi-
tion by its two-toned leaves (Fig. 1). This station represents the first docu-
mented occurrence of this Australasian species in North America.

A small but vigorous population was discovered on 20 October 2001 grow-
ing in shallow water (<1.0 m depth) and adjacent muddy shores for 10 m
along Cooper’s Ditch, a canal dug through non-tidal forested wetlands in
the early 1980’s as a stormwater in the city of Chesapeake. Plants were re-
stricted to shallow water and associated with Utricularia sp., Hydrilla verti-
cillata (L.f.) Royle, Hydrocotyle verticillata Thunb., Myriophyllum pinnatum
(Walter) BSP, and filamentous algae. Like other species in the genus, floating
leaves were much larger than leaves on terrestrial plants (Fig. 1). Rhizomes
were rooted. No sporocarps were present. Collection data follows: Virginia:
City of Chesapeake, NE side of Hillwell bridge across Cooper’s Ditch, Co-
ordinates: 36°42'01”N, 76°13'32”W, 20 October 2001, L. J. Musselman and
D. A. Knepper, 2001-36 (ODU).

Winter hardiness in this species is not known. Nothing is known about
the invasiveness of this species and the population should be monitored to
measure its persistence and its spread.—Davip A. Knepper, U S Army Corps
of Engineers, Fort Norfolk, Norfolk, VA 23510-1096; Davin M. JOHNsoN,
Department of Botany-Microbiology, Ohio Wesleyan University, Delaware,

Fic. 1. Habit view of Marsilea mutica.

244 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 3 (2002)

OH 43015; and Lyrron JoHN MussELMAN, Department of Biological Sciences,
Old Dominion University, Norfolk, VA 23529-0266.

3,8-Di-C-arabinosylluteolin, a new flavonoid from Pteris vittata.—In spite
of the fact that fern flavonoids are of chemotaxonomic interest, little is
known of the distribution of these compounds in some fern families (e.g.
Pteridaceae). Previous work on the flavonoids of Pteris vittata L. (Pterida-
ceae) has led to the identification of an anthocyanin (luteolinidin 5-O-gluco-
side) by Harborne (Phytochemistry 5:589-600, 1966). In addition, acid
hydrolysis of extracts of this fern has led to the identification of kaempferol,
quercetin, leucocyanidin and leucodelphinidin by Voirin (Ph. D. thesis, Uni-
versity of Lyon, p. 151, 1970). More recently 3-C-(6'’’-acetyl-cellobiosyl)-
apigenin (Amer. Fern J. 89:217-220, 1999) and 6-C-cellobiosylisoscutellarein
8-methyl ether together with quercetin 3-O-glucuronide and rutin (Amer.
Fern J. 90:42—47, 2000) have been identified by Imperato and Telesca. Three
kaemperol glycosides (3-O-glucoside, 3-O-glucuronide and 3-O-(X",X’-di-
protocatechuoyl)-glucuronide) together with quercetin 3-O-(X",X"-di-protoca-
techuoyl)-glucuronide have been found in this fern by Imperato (Amer. Fern
J. 90: 141-144, 2000).

In the present paper a new C-glycosylflavone (identified as 3,8-di-C-arabi-
nosyl-luteolin (I)) and 6-C-arabinosyl-8-C-glucosylluteolin (II) have been iso-
lated from Pteris vittata L. growing in the Botanic Garden of the University
of Naples. This fern has been identified by Dr. R. Nazzaro (University of
Naples); a voucher specimen (149.001.001.01) has been deposited in the Her-
barium Neapolitanum (NAP) of the University of Naples.

Flavonoids (I and II) were isolated from an ethanolic extract of aerial parts
of Pteris vittata L. by preparative paper chromatography in BAW (n-butanol-
acetic acid-water, 4:1:5, upper phase), 15% HOAc (acetic acid) and BEW
(n-butanol-ethanol water, 4:1:2.2). Further purification was carried out by
Sephadex LH-20 column chromatography eluting with methanol.

Color reactions (brown to yellow in UV+NHs), ultraviolet spectral analysis
in the presence of usual shift reagents (Amax (nm) (MeOH) 258, 272, 348;
+NaOAc 283, 322 (sh), 402; +NaOMe 266, 281, 340 (sh), 407 (increase in in-
tensity); +AICl; 277, 299 (sh), 331 (sh), 426; +AICl,/HCl 280, 300 (sh), 358,
388) and chromatographic behaviour (Ry values on Whatman No 1 paper:
0.15 in BAW; 0.31 in 15% HOAc; 0.12 in HO) suggested that flavonoid (I)
may be a flavonoid glycoside with free hydroxyl groups at positions 5, 7, 3’
and 4’. Since treatment with 2N HCl (2 hr at 100°C) failed to produce an
aglycone, flavonoid (I) may be a C-glycosylflavonoid. Electrospray mass spec-
trum (ESMS) showed a pseudomolecular ion at m/z 573 [((M+H)+Na]*, an
ion at m/z 595 [(M+H)+2 Na]* and an ion at m/z 1123 [(M x 2)+ Na+H]*
which corresponds to a dimer. These data suggest that flavonoid (1) is a di-C-
pentosylluteolin. 'H NMR spectrum (DMSO-d,) showed signals at 6 3.11—
3.91 (ten sugar protons, m), 5 4.61 (1H, d, J=8 Hz, anomeric proton), 6 4.68
(1H, d, J=8 Hz, anomeric proton), 5 6.27 (1H, s, H-6), 6 6.93 (1H, d, J=8.8

SHORTER NOTES 245

Fic; 1.

Hz, H-5’), 6 7.39 (1H, dd, J=2.0 and 8.8 Hz, H-6’) and 6 7.40 (1H, d, J=2 Hz,
H-2'). These data suggest that flavonoid (I) is a luteolin 3,8-di-C-pentoside.

Wessely-Moser isomerization (3N HCl; 3 hr at 100°C) gave a mixture in
which flavonoid (I) and four isomers were detected by paper chromatography
in BAW; these isomers were not present in sufficient amount to allow charac-
terization. 'H NMR spectrum (DMSO-dg) of the above mixture showed a sin-
glet at 6 6.56 (H-8) and a singlet at 6 6.26 (H-6) confirming that flavonoid (I) is
a 3,8-di-C-glycosylflavone. Since flavonoid (I) gave at least four isomers, ara-
binose may be attached at C-3 and/or C-8 because C-arabinosylflavones on
acid treatment undergo pyranose-furanose isomerization and a-linkage-f-link-
age isomerizatiion of C-glycosidic link as described in a review of Chopin et
al. (pp. 449-503 in J.B. Harborne and T. J. Mabry eds., The Flavonoids: Ad-
vances in Research, Chapman and Hall, London, New York, 1982). Treatment
of flavonoid (I) with 2,2-dimethoxypropane and 6N HCl-dioxan in dry di-
methylformamide according to Jarman and Ross (J. Chem. Soc. (C):199-203,
1969) gave a diisopropilidene derivative ([M]* at m/z 630 in EI-mass spec-
trum); hence arabinose is attached at C-3 and C-8 of flavonoid (I) since this
isopropilidenation is specific for C-galactosyl and C-arabinosyl residues in
mono- and di-C-glycosylflavones as described in the above review by Chopin
et al. FeCl, oxidation of flavonoid (I) gave L-arabinose. The above results
show that flavonoid (I) is 3,8-di-C-arabinosylluteolin (Fig. 1), a new natural
product; this is the first report of a 3,8-di-C-glycosylflavone from ferns.

3,8-Di-C-glycosylflavones were found for the first time in plants in 1985 by
Matsubara et al. (Nippon Nogeikayaku Kaishi 59: 405-410, 1985) who iso-
lated apigenin 3,8-di-C-glucoside and diosmetin 3,8-di-C-glucoside from Cit-
rus sudachi peelings; subsequently these two flavonoids have been found
also in Citrus sinensis peelings (Agric. Biol. Chem. 50: 781-783, 1986) and
the former flavonoid has been found also in Citrus junos peelings (Nippon
Nogeikayaku Kaishi 59: 683-687, 1985) by Kumamoto et al.

Flavonoid (II) was identified as luteolin 6-C-arabinoside-8-C-glucoside by
ultraviolet spectral analysis with usual shift reagents, treatment with 2N HCl
(which failed to give an aglycone), ESMS (which gave a pseudomolecular

246 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 3 (2002)

ion at m/z 603 ([(M+H)+Na]*), FeCl, oxidation (which gave D-glucose and
L-arabinose) and "H NMR spectrum (DMSO-d,). Assignment of D-glucose to
C-8 was based on doublings of signals in 'H NMR spectrum (two signals
were observed for H-3 and H-6’) since this feature is due to the presence of a
C-linked hexose at C-8 as described in a review by Jay (pp. 57-93, in J.B.
Harborne ed., The Flavonoids, Advances in Research since 1986, Chapman
and Hall, London, 1994). Flavonoid (II) is a new fern constituent; it has pre-
viously been found in bryophytes (Blepharostoma tricophyllum and Pleuro-
zia conchifolia) and in angiosperms (Lespedeza capitata, Glycine max and
Astrantia major) as described in the review by Jay.

The author thanks MURST (Rome) for financial support. Mass spectral data
were provided by SESMA (CNR, Naples).—Fiuirro Imperato, Dipartimento di
Chimica, Universita della Basilicata, I-85100 Potenza, Italy.

American Fern Journal 92(3):247 (2002)

REVIEW

Pteridophytes of Upper Katanga (Democratic Republic of Congo), by Jan
Kornas, Anna Medwecka-Kornas, Francois Malaisse and Malgorzata Matjasz-
kiewicz. 2000. Botanical Papers 35, Institute of Botany Jagiellonian Univer-
sity, Kracow, Poland.

This fairly standard pteridophyte flora covers the Upper Katanga, which is
the southeastern part of the Democratic Republic of Congo (formerly Zaire).
The area is bordered by Tanzania, Zambia and Angola in central Africa. The
work is the outgrowth of collections by two Belgian botanists in Katanga as
identified by the Polish pteridologist Jan Kornas, author of many articles on
African ferns. The paper was completed by Anna Medwecka-Kornas after
the death of Dr. Kornas in 1994.

The 180-page book includes a description of the area (climate, geology,
soils and vegetation), a list of species by family and genus (alphabetically),
distribution maps, and remarks on the pteridophyte flora. The description of
vegetation types is brief but complete, with correlation to soils and climate.
There are, unfortunately, no keys or descriptions for the species; however,
for each species, distribution, relative abundance, herbarium citations and
habit are given. Taxonomic notes are useful for some species where there are
questions or problems.

The final section of the work is a brief discussion of the taxonomic compo-
sition of the Pteridophytes to adjacent regions of Africa. The flora includes
183 species in 60 genera. Asplenium is the largest (27 species), as expected
in this part of the world, followed by Thelypteris (12), Selaginella (11), and
Pteris and Trichomanes (9 each). Although some species are relatively com-
mon, more than 50 species have been found at only one or two stations.
Katanga has a richer pteridophyte flora than some other adjacent regions, but
not as rich as tropical East Africa. The publication of this flora is important
because the destruction of forests and political unrest in the region may
prevent gathering of additional information in the near future.—JAmes D.
Montcomery, Ecology III, Inc., 804 Salem Blvd., Berwick, PA 18603.

American Fern Journal 92(3):248 (2002)

REVIEW

The Illustrated Flora of Illinois. Ferns. 2nd ed., by Robert Mohlenbrock.
Southern Illinois University Press, Carbondale. 240 pp. $39.95.

Local, illustrated fern floras provide a wonderful service by providing a
handy, complete, and easy to use reference without the ‘“‘chatter” of species
not found in the area. These floras are designed to provide a streamlined,
comprehensive view of the pteridoflora. Without a doubt, Mohlenbrock’s
first edition accomplished this goal and more. The Introduction provided a
brief account of the history of Illinois fern collecting, a brief account of pteri-
dophyte morphology, and a very nice discussion of habitats. Add to this the
combination of functional keys, state maps, detailed species accounts, and
excellent illustrations and you have a recipe for success. Having said this,
the second edition is a considerable disappointment. All of the new material
is packed into a 45 page Appendix. The new taxa are keyed out separately
from all the rest in a section entitled “Key to the Additional Ferns of IIli-
nois’’. For a plant not covered in the first edition, therefore, one must first
attempt to identify the species in the main part of the book and then go to
the Appendix and start over.

Much of the Appendix consists of a “do-it-yourself” editing experience, in
which you look at information provided and then go into the main part of
the text and fix it up. As an example, on page 180 you are told that the name
for Lycopodium porophilum Lloyd & Underwood (see p. 24) is Huperzia po-
rophila (Lloyd & Underwood) Holub and that you should add dots to the
map on page 26 to include Brown, Carroll, Cumberland, JoDavies, Johnson,
Lake, Lee, Randolph, Rock Island, Schuyler, and Will counties. To make
things worse, for any genus in which new taxa have been added there are
new keys that should be utilized, so please make note not to use the ones in
the first 173 pages. The new illustrations while adequate for identification,
in most cases lack the detail and precision of those found in the previous
edition. While the second edition does provide up to date information, it
seems unreasonable in this day of electronic wizardry that the changes
found in the Appendix could not have been incorporated into a cohesive w-
hole before publication—R. James Hickey, Botany Department, Miami Uni-
versity, Oxford, Ohio 45056.

tanical Garden Libr,

WA WIM If

iy

Authors are encouraged to submit manuscripts pertinent to pteridology for pub-
lication in the American Fern Journal. Manuscripts should be sent to the Editor.
Acceptance of papers for publication depends on merit as judged by two or more
referees. Authors are encouraged to contribute toward publishing costs; however,
the payment or non-payment of page charges will affect neither the acceptability
of manuscripts nor the date of publication.

Authors should adhere to the following guidelines; manuscripts not so prepared
may be returned for revision prior to review. Submit manuscripts in triplicate
(xerocopies acceptable), including review copies of illustrations and originals of
illustrations. After review, submission of final versions of manuscripts on diskette
(in PC- or Mac-compatible formats) is strongly encouraged. Use standard 8% by
11 inch paper of good quality, not “erasable” paper. Double space manuscripts
throughout, including title, authors’ names and addresses, short, informative ab-
stract, text (including heads and keys), literature cited, tables (separate from text),
and figure captions (grouped as consecutive paragraphs separate from figures).
Arrange parts of manuscript in order just given. Include author’s name and page
number in upper right corner of every sheet. Provide margins of at least 25 mm
all around on typed pages. Do not submit right-justified copy, avoid footnotes,
and do not break words at ends of lines. Make table headings and figure captions
self-explanatory. Use S.I. (metric) units for all measures (e.g., distance, elevation,
weight) unless quoted or cited from another source (e.g., specimen citations). For
nomenclatural matter (i.c., synonymy and typification), use one paragraph per
basionym (see Regnum Veg. 58:39—40. 1968). Abbreviate titles of serial publi-
cations according to Botanico-Periodicum-Huntianum (Lawrence et al., 1968,
Hunt Botanical Library, Pittsburgh) and its supplement (1991). References cited
only as part of nomenclatural matter are not included in literature cited. For shorter
notes and reviews, omit the abstract and put all references parenthetically in text.
Use Index Herbariorum (Regnum Veg. 120:1—693. 1990) for designations of her-
bairia.

Illustrations should be proportioned to fit page width with caption on the same
yage. Provide margins of at least 25 mm on all illustrations. For continuous-tone
illustrations, design originals for reproduction without reduction or by uniform
amount. In composite blocks, abut edges of adjacent photographs. Avoid com-
bining continuous-tone and line-copy in single illustrations or blocks. Coordinate
sequence and numbering of figures (and of tables) with order of citation in text.
Explain scales and symbols in figures themselves, not in captions. Include a scale
and reference to latitude and longitude in each map.

Proofs and reprint order forms are sent to authors by the printer. Authors should
send corrected proofs to the editor and reprint orders to the printer. Authors will
be assessed charges for extensive alterations made after type has been set.

For other matter of form or style, consult recent issues of American Fern Jour-
nal and The Chicago Manual of Style, 14th ed. (1993, Univ. Chicago Press,
Chicago). Occasionally, departure from these guidelines may be justified. Authors
are Satie to consult the editor for assistance with any aspect of manuscript
prepara’

5 aaa louse than 32 printed pages may be sent to the Editor of Pteridologia
(Memoir Editor, see cover 2).

PTERIDOLOGIA ISSUES IN PRINT

The following issues of Preridologia, the memoir series of the American
Fern Society, are available for purchase:

1. Wagner, David H. 1979. Systematics of Polystichum in Western North

America North of Mexico. 64 pp. $10.00 postpaid.

2A. Lellinger, David B. 1989. The Ferns and Fern-allies of Costa Rica,
Panama, and the Chocé (Part 1: Psilotaceae through Dicksoniaceae). 364 pp.
$32.00 postpaid.

3. Lellinger, David B. 2002. A Modern Multilingual Glossary for Taxonomic
Pteridology. 263 pp. $28.00 postpaid.

Send your order with a check or money order to: American Fern Society,
Inc., c/o U.S. National Herbarium MRC-166, Smithsonian Institution, Washing-
ton, DC 20560.

AMERICAN FERN JOURNAL ON MICROFICHE

Volumes 1—61 of the American Fern Journal are available as archival quality,
silver positive microfiches. Single volumes or the entire run may be purchased.
The fiches are easily read with 10X or greater magnification (using a dissecting
microscope and transmitted illumination or a fiche reader). Silver negative micro-
fiches of vols. 1-50 are also available. The price is $4.00 per volume or $244.00
per set of 61 volumes, postpaid.

Send your inquiry or order with a check or money order to: American Fern
Society, Inc., c/o Dr. James D. Montgomery, Ecology III, Inc., R.D. 1, Box 1795,
Berwick, PA 18603.

VISIT THE AMERICAN FERN SOCIETY’S
WORLD WIDE WEB HOMEPAGE:
http://www.amerfernsoc.org/

OK |
A395

AMERICAN os
FERN number
JOURNAL pe

QUARTERLY JOURNAL OF THE AMERICAN FERN SOCIETY

The Morphological and Genetic Distinctness of Botrychium minganense and B. crenula-
tum as Assessed by Morphometric Analysis and RAPD Markers
Linda M. Swartz and Steven J. Brunsfled 249

Reproductive Behavior of Cloned Gametophytes of Pteridium aquilinum (L.) Kuhn
Forbes W. Robertson 270

A New Population of Aleutian Shield Fern scp gare aleuticum C. Christens.) on Adak
Island, Alaska andra Looman Talbot and Stephen S. Talbot 288

Shorter Note
Trichomanes ribae (Hymenophyllaceae), a New Filmy Fern from Costa Rica and Panama
Leticia Pacheco 294
Referees for 2002 296

Index to Volume 92 (2002) 297

The American Fern Society

Council for 2002

CHRISTOPHER H. HAUFLER, Dept. of Botany, University of Kansas, Lawrence, KS 66045-2016.
President
TOM RANKER, University Museum, Campus Box 265, University of Colorado, Boulder, CO 80309-0265.
Vice President
W. CARL TAYLOR, 800 W. Wells St., Milwaukee Public Museum, Milwaukee, WI 53233- : id
retary
JAMES D. CAPONETTI, Dept. of Botany, University of Tennessee, Knoxville, TN 37916-1 : 10.
Treasurer
GEORGE YATSKIEVYCH, Missouri Botanical Garden, P. O. Box 299, St. Louis, MO 63166-0299.
Membership Secretary
JAMES D. MONTGOMERY, Ecology III, 804 Salem Blvd, Berwick, PA 18603-9801.
Back Issues Curator

R. JAMES HICKEY, Botany Dept., Miami University, Oxford, OH 45056. Journal Editor
DAVID B. LELLINGER, U.S. National Herbarium MRC-166,
mithsonian rene ee DC 20560-0166. Memoir Editor
gto JOHNSON-GROH, Dept. of Biology, Gustavus Adolphus College,
OO W. College Ave., St. Peter, MN 56082-1498. Bulletin Editor

American Fern Journal
EDITOR

R. JAMES HICKEY Botany Department,
ami University, Oxford. OH 45056
ph. (513) Sy 6000, e-mail: hickeyrj@muohio.edu

ASSOCIATE EDITORS
GERALD J GASTONY (2... — of Biology, Indiana University, Bloomington, IN 47405-6801
CHRISTOPHER H. “spariakacnie Dept. of Botany, University of Kansas, Lawrence, KS 66045-2106
ROBBIN C. MORA New York Botanical Garden, Bronx, NY 10458-5126
JAMES H. PECK Dept. of tag University of Arkansas—Little Rock,

1 S. University Ave., Little Rock, AR 72204
The “American Fern Journal” (ISSN 0002-8444) is an illustrated quarterly devoted to the een
stanky of ferns. It is owned by the American Fern Society, and published at The American Fern Society,
% Missouri Botanical Garden, P. O. Box 299, St. Louis, MO 63166-0299. Periodicals postage aid at
St. is, MO, and additional entry.
Claims for missing issues, made 6 months (domestic) to 12 months (foreign) after the date of issue,
and orders for back issues should be addressed to Dr. James D. Montgomery, Ecology II, 804 Salem
Blvd. oe PA 18603-9801.
Changes of address, dues, and applications for membership should be sent to the Membership

General inquiries concerning ferns should be addressed to the Secretary.

Subscriptions $35.00 to U.S.A., Canada, and Mexico; $45.00 i oo in the world (—$2.00
agency fee, if applicable); sent free to members of the American Society (annual dues, $25.00 +

.0O0 mailing surcharge beyond U.S.A., Canada, and Mexico; Pa eens. $300.00 + $140.00
mailing surcharge beyond U.S.A — ae and Mexico
Back vol

yon
olumes are available for most years as printed issues or on microfiche. Please contact the
Back Issues Curator for prices as; availability.
POSTMASTER: Send address changes to AMERICAN FERN JOURNAL, Missouri Botanical Garden, P. O.
Box 299, St. boas MO 63166-0299
FIDDLEHEAD FORUM
Bloc: editor of the Bulletin of the American Fern Society oo contributions from members a
members, including miscellaneous notes, offers to exchan one x purchase materials, aE
feat ceeak notes, and reviews of non-technical books on fern
SPORE EXCHANGE
ia Mandt, 12616 Ibbetson Ave., Downey, CA 90242-5050, is ae. Spores exchanged
and ae a available spores sent on request. _http://amerfernsoc.org/spo mil
GIFTS AND BEQUESTS
Gifts and bequests to the Society enable it to expand it ices to members and to others interested
in ferns. Back is Reems ik die 1 4 + ot. -o ‘. 1 4 - rs deductible

o"
ig ees *

oe 2 1. 2% ae ets g ie

MISSOURI BOTA: -AL

JAN 1 4 2003
American Fern Journal 92(4):249—269 (2002) GARDEN LIBRARY

The Morphological and Genetic Distinctness
of Botrychium minganense and B. crenulatum
as Assessed by Morphometric Analysis
and RAPD Markers

Linpa M. Swartz! and STEVEN J. BRUNSFELD”
Department of Forest Resources, University of Idaho, Moscow, ID 83844-1133

AsstTract.—Two species of Botrychium subgenus Botrychium (moonworts, Ophioglossaceae),
Botrychium minganense Victorin and B. crenulatum W. H. Wagner, can sometimes be confused in
the field, even by experts, because of their reduced morphology. Botrychium minganense can
imitate B. crenulatum, which is more rare. They are afforded different degrees of protection on
Federal lands, making the distinctness of these species a question of management, conservation,
and systematic interest. The purpose of this study was to compare a morphometric analysis of
these two species with an analysis of DNA markers from the same individuals, and to assess their
distinctness under each method. Collections were made in Washington, Oregon, Idaho, and
Montana from seven auger i - B. ehidpessdess and 18 populations of B. minganense. Each
plant was measured, emph g cited by authors in the original species descriptions.
Canonical variate analysis ser on SAS separated the samples into two oe groups with
32% overlap. RAPD genetic markers revealed more genetic variation than has previously been
UPG

be confirmed or ruled out with markers from one or two RAPD primers. Both B. crenulatum and B.
Junaria have been suggested as possible diploid parents of tetraploid B. minganense. All RAPD
markers absent in B. crenulatum but present in B. minganense were also present or polymorphic in
B. lunaria, supporting B. /unaria as a possible parent. One very small population of B. minganense
showed a monomorphic RAPD profile, consistent with inbreeding, but all other populations had
multiple genotypes. Some plants of B. minganense clustered most closely with plants from
populations up to 400 km away, suggesting that variation may be introduced into populations by
occasional colonization by spores from distant sources.

Species of Botrychium subgenus Botrychium (moonworts, Ophioglossaceae)
are an enigmatic part of the temperate flora, notable for their small size,
reduced morphology and difficult identification. Several moonwort species in
North America are listed as sensitive or rare because of small and/or few
known populations. Small populations are more vulnerable to extirpation,
whether from natural stochastic events or human activities. Understanding the
threat to species requires accurate information on numbers of individuals and

" Current address: USDA Forest a Hiawatha National Forest, St. Ignace Ranger District,
1798 West Highway 2, St. Ignace, MI 4
* Corresponding author.

250 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 4 (2002)

populations. If species intergrade morphologically, questions can arise not
only about actual numbers of individuals and populations but also about
species boundaries and the genetic distinctness of each species.

All moonworts are relatively small and bear a single leaf with a fertile
segment (sporophore) and sterile segment (trophophore) each season from an
underground bud. They are notoriously hard to find, especially in thick
vegetation. As increasing emphasis has been focused on rare plants in recent
decades, more concentrated searches have extended the known ranges of
common Botrychium species and provided material from which 13 new
species have been described since 1980. Several of these are endemic to
western North America: B. crenulatum W. H. Wagner, B. echo W. H. Wagner,
B. lineare W. H. Wagner, B. montanum W. H. Wagner, B. paradoxum W. H.
Wagner, B. pedunculosum W. H. Wagner, B. pumicola Coville, and B.
pinnatum H. St. John (the latter two described in 1900 and 1929, respectively).
Distinguishing moonwort species in the field often depends on subtle
differences in phenology, color, texture, proportions of the parts of the single
leaf, and dissection of the pinnae. Such species, poorly morphologically
differentiated but evolutionarily distinct, have been called cryptic species
(Stebbins, 1950; Paris, Wagner, and Wagner, 1989; Hauk and Haufler, 1999).
Although species differences may be subtle, some species are also quite
variable among regions, among sites, and even within the same site (e.g.,
Wagner and Lord, 1956). One of those species is B. minganense Victorin.

This study was initiated in response to the practical need to distinguish
between B. crenulatum and B. minganense. These two species have been
confused in the western United States by many botanists (Zika, 1992).
Although both have been listed as “sensitive” in the past by National Forests
in the Pacific Northwest Region (Region 6) and the Northern Region (Region 1),
B. minganense has been delisted in Region 6 in response to the discovery of
many more populations, while B. crenulatum retains its official status as rare.
Species designation affects management options where B. crenulatum occurs.
The documented distribution of B. crenulatum is the mountain states of the
American west (Arizona, California, Idaho, Montana, Oregon, Nevada, Utah,
Washington, and Wyoming), whereas B. minganense is widespread in the
western mountains and across northern North America (Wagner and Wagner,
1993).

Botanists have employed both lumping and splitting approaches to the
confusing variability of B. minganense. Botrychium minganense has been
interpreted by many authors as a variety of B. Junaria (L.) Sw. (see Wagner and
Lord, 1956 for discussion). In Flora of the Pacific Northwest (Hitchcock and
Cronquist, 1973), only five moonworts are recognized, and the taxon to which
B. minganense keys is called B. Junaria var. onongadense (Underw.) House.
Cronquist said that B. minganense is “...morphologically scarcely separable
from diploid var. onongadense...” and considered it conspecific with B.
Junaria (Gleason and Cronquist, 1991). Botrychium minganense is a currently
accepted taxon (International Taxonomic Information System database http://
www. itis.usda.gov/plantproj/itis, April 15, 2000; Kartesz, 1994). In the most

SWARTZ & BRUNSFELD: MORPHOMETRIC ANALYSIS OF BOTRYCHIUM SPECIES 251

recent treatment of North American moonworts (Wagner and Wagner, 1993), B.
minganense is reported to be sometimes misidentified as B. dusenii of South
America. It is also easily confused with B. Junaria (Wagner and Lord, 1956;
Farrar, 1998), B. ascendens (Zika, 1992; Farrar, 1998), B. pallidum (Zika, 1992),
B. spathulatum (Zika, 1992), and B. crenulatum (Wagner and Lord, 1956;
Lellinger, 1985; Wagner and Devine, 1989; Zika, 1992; Farrar, 1998). Zika (1992)
described B. minganense as “‘treacherously variable’’. Just as B. minganense
was recognized as an independent species from the more widespread and
common B. lunaria, so too were B. pallidum and B. spathulatum formerly
confused with B. minganense. Both Wagner and Wagner (1988), and Wagner
(1994) have suggested that B. minganense may represent a species complex.

Unlike B. minganense, B. crenulatum is more constant in form when well
developed, but as with any moonwort, the identity of small plants can be
ambiguous. Botrychium minganense can approach the form of B. crenulatum
closely. Wagner and Wagner (1981) state that some of the collections on which
the original description of B. crenulatum was based were originally identified
as B. lunaria var. minganense.

Botrychium crenulatum is diploid (2n = 90, F. S. Wagner, 1993), whereas B.
minganense is tetraploid (2n = 180 Wagner and Lord, 1956; but see Hauk and
Haufler, 1999). Many fern species, however, have races with different ploidy
levels (e.g. Asplenium trichomanes, Wagner et al., 1993), and ideally,
additional evidence of genetic differences would be employed to separate
species (for discussion, see Gastony and Windham, 1989).

Molecular techniques are well suited to clarify problems of cryptic species.
Hauk (1995) used rbcL sequences in a phylogenetic analysis of 20 species of
Botrychium subgenus Botrychium. Hauk found that four samples of B.
minganense (from Michigan, Colorado, and Ontario) shared identical
sequences, along with B. paradoxum and B. Xwatertonense, and lacked the
single synapomorphies that distinguished the simplex and campestre
subclades of the “‘simplex-campestre” clade. Botrychium crenulatum formed
a separate clade with B. /unaria, identical in sequence to the United States B.
lJunaria sample, and well separated from the “simplex-campestre” clade by
a total of nine substitutions.

In contrast to the rbcL data, which did not distinguish B. crenulatum from B.
Junaria, isozymes differentiated B. crenulatum from all others (Farrar, 1998:
Hauk and Haufler, 1999). Among the sampled diploids B. crenulatum was
most similar to B. /unaria, but their genetic identity (Nei, 1978) was only 0.53
(Hauk and Haufler, 1999). Botrychium minganense possessed the highest
variability of the western moonworts (Hauk and Haufler 1999, Farrar 1998), but
neither study inferred the variation to be indicative of species-level
differentiation within B. minganense.

Random Amplified Polymorphic DNA (RAPD) markers, a type of genetic
fingerprint, have revealed a level of genetic variation useful for distinguishing
populations and sometimes species, and typically possess more variation than
isozymes (for reviews, see Bachmann, 1997; Crawford, 1997). RAPD has been
particularly useful in assessing variation in rare plants because, as a PCR-based

252 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 4 (2002)

technique, it requires only small tissue samples, fresh or dried. DNA markers
such as RAPD may provide important information for critically assessing
morphometric analyses in taxonomically confusing groups. Morphometric
analysis can suffer from a circular logic in which a taxon exhibits a certain
range of morphologic variation because of assumptions made in the
assignment of specimens to that taxon. Assigning specimens based on genetic
markers can provide more robust morphometric insights, as has been shown in
a variety of studies (e.g., Hardig et al., 2000).

Thus, the goals of this study are 1) to determine the genetic distinctness of B.
minganense and B. crenulatum, on the basis of RAPD markers 2) to document
patterns of genetic variation within B. minganense and B. crenulatum, on the
basis of RAPD markers, and 3) to assess quantitatively the morphological
differences between plants of B. minganense and B. crenulatum classified on
the basis of genetic markers.

METHODS

Collections.—Samples were collected from seven populations of Botrychium
crenulatum and 18 populations of B. minganense in the states of Washington,
Oregon, Idaho, and Montana (Table 1). Within this region populations were
chosen to include a full range of habitats and geography. Plants with
morphology intermediate between the two species were collected when
found, and small plants were collected as well as large, well-developed ones to
represent a full spectrum of the morphology found in each population. Plants
were collected throughout the spatial extent of each site. Sample sizes are
given in Table 1. In addition, two populations of B. Junaria and one of B.
simplex E. Hitchcock (both subgenus Botrychium) were collected to provide
a larger sample of species level molecular comparisons. The ecological
associations of B. minganense and B. crenulatum were quite different in
different parts of their ranges. Botrychium crenulatum in Washington is
sometimes found in somewhat wetter and more open habitats than B.
minganense, but the large populations sampled for this study were all growing
under a Thuja plicata/mixed conifer canopy on subirrigated ground. By
contrast, the Goofy Springs, Oregon, population was growing in heavy
graminoid cover in an opening on seepy ground; the Stewart Creek, Montana,
site was a wet mowed roadside and ditch; and at Lapover Ranch, Oregon (the
one site on private land), B. crenulatum and B. minganense were growing
together in grass cover under Pinus contorta with no spring evident. In
Washington, B. minganense was found almost exclusively under riparian
Thuja plicata stands with depauperate understory, but one sampled
population (Mill Gate) came from an herbaceous mountain meadow. The
association with Thuja plicata stands may be an artifact of the circumstance
that moonworts have mainly been searched for in association with proposed
timber sales. In Oregon, the sampled B. minganense populations were all
under forest canopy open enough to support a luxuriant shrub and/or herb
layer, except Dusty, which was from a wet meadow.

SWARTZ & BRUNSFELD: MORPHOMETRIC ANALYSIS OF BOTRYCHIUM SPECIES

TABLE 1.

segregated under more tha
occurred at a single site, each species is listed as a
minganense, c = B. crenulatum, L = B. lu

253

Collection locations of Botrychium used in this study. ania eng from some sites were
n one collection number.

Where more than one analyzed —

population “ae an “aries letter
naria). Vouchers are deposited in I

Site Site
Species (no. of plants) abbreviation Location Voucher
B. minganense Watson Point (2) Watson OR, Wheeler Co. Swartz 387
wery Trail (6) Flowery WA, Stevens Co. Swartz 393
Kelsey Creek (6) Kelse MT, Lincoln Co Swartz 394
Rock Bottom (7) Rock ID, Boundary Co Swartz 398
Deer (7) Deer ID, Boundary Co Swartz 399
Wenatchee Ford WenT WA, Chelan Co. Swartz 401A
TSHE (7
Wenatchee Ford WenR WA, Chelan Co. Swartz 401B
PA (5
Devil’s Club Devil WA Chelan Co. Swartz 402
Creek (7)
Mill Gate (17) Mill, MillIB WA, Chelan Co. Swartz 403,
Swartz 453
Aladdin 1 (6 Aladdin1 WA, Stevens Co. Swartz 414
Bulldog Cabin (5) Bulldog WA, Stevens Co Swartz 420
Poison Springs (7) Poison OR, Grant Co Swartz 425
m Hodgson Creek (7) mHodgson WA, Ferry Co. Swartz 466
m Rd. #9576 (6) mRd9576 WA, Ferry Co. Swartz 468
ManleyX Swartz 486
m Manley Creek (15) mManley WA, Ferry Co. Swartz 470
La Grande 32 (6) aG32 OR, Union Co Swartz 504
Shady Camp (7) Shady OR, Wallowa Co Swartz 506
Dusty (6) Dusty OR, Union Co. wartz/
Riley 508
Swartz/
Yanskey 509
B. crenulatum — Goofy Spring (7) Goofy OR, Crook Co. Swartz 38
Stewart Creek (5) Stewart MT, Flathead Co. Swartz 396
Okanogan Cabin (7) OKCabin WA, Okanogan Co, Swartz 404
Aladdin Aladdin WA, Stevens Co. Swartz 427
Blowdown (7)
Deadman Creek (5) Deadman WA, Ferry Co. Swartz 445
c Hodgson Creek (7) cHodgson WA, Ferry Co. Swartz 467
Lapover Ranch (10) apove OR, Wallowa Co. Swartz 507
B. lunaria L Rd. #9576 (5) LRd9576 WA, Ferry Co Swartz 469
L Manley Creek (7) LManley WA, Ferry Co Swartz 471
B. simplex La Grande LGMead OR, Union Co. Swartz 505
Meadow (5)

Plants were collected by snipping them off at ground level to avoid
disturbing the roots and the next year’s below-ground bud. This procedure is
not believed to have a significant negative impact on survival (Johnson-Groh
and Farrar, 1996; Montgomery, 1990). Where possible, plants were collected
after they had shed spores. Plants were pressed, individually numbered,
color photocopied, and digitally imaged before grinding for DNA extraction.

254 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 4 (2002)

The color photocopies are deposited in the University of Idaho Herbarium
(ID) as facsimile vouchers, along with additional collections from the same
populations.

Morphometric analysis.—As a quantitative approach to capturing morpho-
logical subtlety, a morphometric analysis of characters that can be scored from
herbarium specimens was made. Characters cited by authors in the original
species descriptions were used whenever possible. Some characters that are
valuable to botanists in the field, such as color, texture, or folding of pinnae,
could not be scored because they are distorted or destroyed by pressing and
drying. Forty-one different measurements or ratios were recorded for each
plant. Measurements were made using a Panasonic WV-CD20 video camera
and Mocha image analysis software (SigmaScan Pro version 3.0 Jandel
Scientific).

Analysis.—Canonical discriminant analysis identifies one or more canonical
variables that are linear combinations of multiple measured characters. These
canonical variables can show the greatest morphological differences between
groups. Statistical calculations were performed using the SAS CANDISC
procedure, SAS Release 6.11 (SAS Institute Inc.), on the same samples of B.
minganense and B. crenulatum used for genetic analysis, excluding three
plants that were browsed. Two very unusual plants of Botrychium minganense
also were excluded from the morphological analysis. One was extremely large,
and the other had only rudimentary peg-like pinnae. One plant (cLapover.09)
was excluded because it displayed an additive RAPD profile, and thus was pos-
sibly a hybrid. Preliminary one-way ANOVA showed that the means of many
characters were significantly different between the species at the alpha =
0.05 level, including the ratio of trophophore width to the width of its axis
(reflecting the tendency of pinna margins to be decurrent on the rachis);
average angle of the margins of the four basal-most pinnae (degree of fanning);
ratio of the length of the space between the first two pinnae pairs to greatest
pinna width (a measure of overlapping of pinnae); ratio of greatest pinna width
to least pinna width; ratio of length to width of trophophore; ratio of length to
width of sporophore; ratio of pinna width to length; length of the sporophore;
average angle made by the four basal-most pinnae with the rachis; total height
(ground level to tip of sporophore); length of trophophore; length of
trophophore stalk; and length of gap between first two pinna pairs. Measure-
ments of these characters are illustrated in Figure 1. All pinna measurements
were made from the same pinna for each plant, one of the largest pair. In the
largest plants the lowest pinnae are sometimes partly transformed into
sporangial branches. In that case one of the largest untransformed pair was
chosen. The ratios of 1) trophophore width: trophophore axis width and
2) maximum:minimum pinna width were log-transformed, and the length
of sporophore was square root-transformed to bring them closer to a normal
distribution. The distribution of all variables used was judged to be within the
limits of robustness of the procedures (K. Steinhorst, pers. comm.). Variances
were compared between species groups for each character to see that they were
equal, or if not, the variance of the larger group did not exceed that of the

SWARTZ & BRUNSFELD: MORPHOMETRIC ANALYSIS OF BOTRYCHIUM SPECIES 255

Fic. 1. Measurements made for characters that were significantly different between Botrychium
minganense and B. crenulatum in morphometric analysis. a. Length of common stalk. b. Length of
sporophore stalk. c. Length of sporangia-bearing part of sporophore (b+c = length of sporophore,
a+b+c = total height. These and any other curved lengths were traced directly on the image of the
plant). d. Length of longest sporophore branch (2d = sporophore width). e. Length of trophophore
stalk. f. Distance between centers “ fast are pena pairs. g. Balance of length of trophophore
(e+f+g = length of trophophore). h inna (for average of four basal-most pinnae).
i. Angle at which pinna meets axis of rachis i average of four basal-most pinnae). j. Greatest
width of rachis. k. Trophophore width. |. Least width of largest pinna m. Greatest width of largest
pinna. n. Length of largest pinna.

smaller group by more than a factor of 2.5, a conservative level chosen for
unequal sample sizes. In general, variances were greater for B. minganense.
RAPD analysis.—DNA was isolated from 10 mg samples of each pressed
plant. For those plants that were less than 10 mg, the whole plant was used.
Plants were ground on ceramic well plates with liquid nitrogen, ground further

256 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 4 (2002)

with 600 ul 70°C CTAB buffer, and transferred to 1.5 ml tubes. The grinding
buffer and subsequent isolation procedures followed Stewart and Via (1993),
with the following modifications: the homogenate was incubated at 70°C for 30
minutes before the chloroform extraction, the precipitated DNA pellet was
washed with 1 ml cold 76% ethanol with 10 mM NH,AC, and the dry pellet
resuspended in 50 ul TE. Of several tested, this protocol was the least likely to
yield gummy residues coprecipitating with the DNA, which was a problem
with some samples. The residue, when it occurred, was removed by
centrifugation before quantifying the DNA with a fluorometer. DNA was
amplified (Williams et al., 1990) in 25 ul reactions containing 1x buffer
(Promega M190A), 0.1 mM of each deoxynucleotide, 2 mM MgCl, 0.00005 %
bovine serum albumin, 5 pmols 10-mer primer (Operon), 10 ng genomic DNA,
and 0.5 units Taq DNA polymerase (Promega), overlaid with 25 ul mineral oil.
Samples were amplified in an MJ Research PT 100 thermocycler (44 cycles of
1 min at 94°C, 1 min at 36°C, 2 min at 72°C, with a final 5 min at 72°C). Products
were electrophoresed in 1.5% agarose gels, visualized by UV illumination after
staining with ethidium bromide, and imaged with Alphalmager v. 3.2
software. Populations were divided among multiple PCR runs, and a sub-
sample was run multiple times to confirm repeatability of each band chosen.
Bands were scored manually by comparison to standard size markers. Bands
are designated by the name of the primer with the approximate size in base
pairs as a subscript, e.g. B-11575.

Primer screening.—Primers were screened against two samples each of B.
minganense and B. crenulatum. Twelve primers (A-11, B-11, B-12, C-6, C-8, C-
9, C-10, C-11, D-11, D-16, D-20, X-1) showing the best well-spaced bands
polymorphic in one or both species were selected for the final data set. One
hundred ninety-four plants were scored manually for presence or absence of
74 RAPD bands each. As more species and populations were added, fewer
primers and bands within primers could be used because some new bands
were Close to the position of old bands or amplified with different intensity,
making them difficult to score. Therefore, the scoring is conservative and
reflects only minimum differences among all populations, whereas man
additional differences that are not included in the data set are readily apparent
among individuals and populations in the same gel.

Cluster analysis of RAPD data.—UPGMA cluster analysis was performed on
RAPD data with NTSYSpc version 2.02 (Rohlf, 1997) using simple matching
and Jaccard metrics.

RESULTS

Morphometric analysis.—Optimal separation of the two species on a mor-
phological basis requires consideration of multiple characters at once. The six
variables in Table 2, when analyzed together, provided the greatest separation
of the groups in this data set. The canonical variate analysis tested the null
hypothesis that there are no differences between the two species based on the
chosen variables. This hypothesis was rejected at the p = 0.0001 level, with

SWARTZ & BRUNSFELD: MORPHOMETRIC ANALYSIS OF BOTRYCHIUM SPECIES 257

E 2. Correlations and coefficients of the six variables that provided the greatest
discrimination between Botrychium minganense and B. crenulatum in Canonical Discriminant
Analysis.

Pooled Within Pooled Within-Class
Canonical Structure Canonical Coefficients
Variable CAN1
AVANGMAR 0.50 0.65
AVANGPIN 0.16 0.40
WIDLEN 0.20 -0.18
MLSPORO —0.18 —0.94
NTWIDAX 0.51 0.51
NPINMAMI 0.41 0.52
AVANGMAR ge angle of pinnae margins; AVANGPIN=average angle of pinnae with rachis;
PWIDLEN=ratio of pins width: pinna mip ESO square root transformed length of
sporophore; NTWIDAX=log-transformed t width of trophophore

axis; NPINMAMI=log-transformed ratio of greatest pinna width: least pinna width.

F = 35.52 and degrees of freedom of numerator 6 and denominator 164.
Canonical scores may be computed by taking the original value for each plant
on each measurement, multiplying it by the respective canonical coefficient
from CAN1 (Table 2), and adding all these products plus a constant adjustment
for the means. Scores graphed by species form two overlapping groups, with
the mean for B. crenulatum at 1.81 and the mean for B. minganense at —0.71
(Fig. 2). CAN1 scores of 55 of the total of 171 plants, or 32%, fell in the 0 to 2
range where species identity is ambiguous. Scores of 92 plants of B.
minganense out of 123 (75%) fell in the 0 to —4 range, and 23 plants of B.
crenulatum out of 48 (49%) scored from 2 to 4, where each had a high
probability of correct species identity. Only one B. minganense had a CAN1
score above 2

Canonical variates can be interpreted in terms of those variables that
contribute the most to the separation of the groups. Although canonical
variates are artificial and must be interpreted with caution, they can be
identified in terms of their correlations with the original individual variables
(Johnson and Wichern, 1992). These ‘‘within” structure coefficients indicate
how closely a variable and the canonical variate are related, or the extent to
which they carry the same information (Klecka, 1980). The ratio of
trophophore width:trophophore axis width had the highest correlation, 0.51,
followed by average angle of pinna margins, 0.50, and pinna maximum
width:minimum width, 0.41. The within-class correlation for pinna width:
length was 0.20, average angle of pinnae to rachis 0.16, and length of
sporophore —0.18.

Another way of looking at the contributions of each individual variable
within classes is by comparing coefficients that have been transformed so their
standard deviations are equal to 1. These standardized coefficients then
measure the relative contribution of each variable to the canonical variate
score. As a relative measure, the standardized coefficient of each variable will
change depending on the contribution of other variables. If two variables share

ses

sini

pes

poeta

fa

ium minganense

E Botrychium crenulatum
Spp.

0 Botrych
Ei Both

eee
S38 Bas

penoanmons

se sais i etree
a8 a pach

a

iables (Table 2) of 123 Botrychium

1c var

ically.

ity was confirmed genet

ies ident

the standardized co-
if one variable was not
y could also be

t the other. The

>

correlated)

la

but

’

The

ise.
ially cancels ou

Tl

part

10nS

iate correlat

lvar

% a res

‘ fons
i ee 2
ae ss sooiects
ee 20 ee : poet ie
tart i as asia

se pee coat
He see anima

rise enrncncinese ice

inunnnens ae pee prcnnny te nee
SRR inna sees uusuanaannanenamnanmnmmninsaa me omen EERIE REE ENE TET

Boe enaier 3 spaonons PEERED EDIE SEE

See
Stone

ighly
etween them

AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 4 (2002)

258

see

PON MN N08

se

oe

iene

aisaasasanssensainty

incon

sia
pee se

pi

ee eh
nod
. i
eee
phone z
Raaeanart Sues

apananie
oe saat

pies at

sia
a Be

ice
Sie

——

eee

ee ae si

Bs

Aouanbe1J

sa ti

ee

sy

es

Se Saal

ix morphometr

ing s
d 48 B. crenulatum plants whose spec

Plot of CAN1 scores us

minganense an

Fig. 2.

ion (are h

some of the same informat
efficient value will be partly d

ided b

1V

t of the other would

so that one
ts, by contrast, are simple b

1en

used, the standardized coeffic

larger but have oppos

>

ite sign

within structure coefficien

SWARTZ & BRUNSFELD: MORPHOMETRIC ANALYSIS OF BOTRYCHIUM SPECIES 259

(Klecka, 1980). The standardized coefficient for length of sporophore, —0.94,
had the highest absolute value among the variables. This was the measurement
reflecting total size of the plant that contributed most to separation of the
species. When it was included, other direct measures of size, such as total
height, had small absolute values. Average angle of pinna margins, 0.65, and
ratio of pinna maximum:minimum width, 0.52, are each related to the fanning
of the pinnae in different ways, and they do not cancel each other out. The
pooled within-class standardized canonical coefficient for ratio of ttophophore
width:trophophore axis width was also high, 0.51, and average angle of
pinnae, 0.40, and ratio of pinna width:length, —0.18, were lower.

RAPD diagnostic bands.—Botrychium crenulatum was most differentiated
(Table 3), set apart by seven bands (B-111 400, C-6g25, C-8g50, C-81700, C-91180, C-
104275, and D-11,975) that did not occur in the other three species. One band (B-
114975) was present in B. crenulatum, absent in B. minganense and B. simplex,
and polymorphic in B. Junaria. One band (C-9;909) was present in B.
crenulatum, absent in B. minganense, and polymorphic in B. Junaria and B.
simplex. Bands not present in B. crenulatum included six that were present in
all individuals of B. minganense. Of these, three (C-11¢75, D-16775;, and D-20g99)
were polymorphic in B. Junaria and present in all B. simplex, two (B-11575, C-
9690) were polymorphic in B. Junaria and absent in B. simplex, and one (D-
114225) was present in B. Junaria and polymorphic in B. simplex. Four bands
not present in B. crenulatum were present at high frequency (0.97—0.99) in B.
minganense. Three of these (D-11375, D-16400, D-165109) were polymorphic in B.
lunaria and absent in B. simplex, and one (D-11;399) was polymorphic in both
B. lunaria and B. simplex. No bands in the sampled plants were unique to B.
minganense or B. lunaria, and one band (C-6459) was seen only in B. simplex.
Bands common to all four species were not scored.

Clustering —UPGMaA clustering of the RAPD data using a simple matching
metric resulted in four well-defined species groups (Fig. 3). The B. crenulatum
cluster was most distinct. The B. simplex cluster and the B. Junaria cluster
grouped together, and the “simplex lunaria’”’ cluster associated most closely
with the B. minganense cluster. Use of a Jaccard metric, discounting 0/0
matches, produced relationships conforming to those discussed below, except
that the B. Junaria cluster associated most closely with the B. minganense
cluster, and the B. simplex cluster grouped with the ‘‘minganense lunaria”
cluster (dendrogram not shown). The B. Junaria and B. simplex clusters are
displayed in Fig. 4.

Botrychium crenulatum.—Within the B. crenulatum group (Fig. 5), the
largest cluster contained all the plants from Washington except two. This
Washington cluster contained four subgroups within which plants had
identical profiles. One subgroup included five OKCabin plants, and three
contained plants from Deadman, cHodgson, and/or Aladdin. All B. crenulatum
populations were polymorphic. Associated with the Washington cluster was
a cluster that contained samples from Montana (Stewart) and Oregon
(Lapover), plus two genetically distinct plants from the Hodgson population
from northeastern Washington, which includes both B. minganense and

260 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 4 (2002)

TaBLE 3. Diagnostic RAPD bands.

Band name B. crenulatum B. minganense B. lunaria B. simplex
B-Ligyst# 0 1 P fy)
B-111075** p 0 Pp 0
B-114400*** 1 0 0 0
C-6450 0 0 0 1
C-6395*** 1 0 0 0
C-8g50* ** 1 0 0 0
C-84700* ** 1 0 0 0
C-Secot# 0 1 P 0
C-94000* 1 0 P |
C-9, 180* i 0 0 0
C-104275*** 1 0 0 0
C-11675tV 0° 1 F 1
-11g00* 1 0 P 1
D-11g75t# 0 P(1) P 0
D-114975*** 1 0 0 0
D-114225fO 0 1 1 P
D-11) 3007 0 P(1) is P
D-16490t# te P(1) ly 0
D-165;0t# 0 P(1) P 0
D-16775tV 0° 1 P(1) 1
D-20g90fV 0 1 P 1

1 = present, 0 = absent, P = polymorphic, P(1) = polymorphic, present very high frequency.
*** Present in B. crenulatum, absent in B. minganense, B. lunaria, B. simplex.

** Present in B. crenulatum, absent in B. minganense, polymorphic in B. Junaria, absent in B.
simplex.

* Present in B. crenulatum, absent in B. minganense, polymorphic in B. Junaria, present or
polymorphic in B. simplex.

¢ Present in B. minganense, absent in B. crenulatum.

+ Polymorphic at high frequency in B. minganense, absent in B. crenulatum.

# Polymorphic in B. Junaria, absent in B. simplex.

V Polymorphic in B. lunaria, present in B. simplex.

O Present in B. lunaria, polymorphic in B. simplex.

“ Also present in Lapover9.

B. crenulatum. One of these individuals, ‘‘m’Hodgson.05, was classified in
the field as B. minganense, but clearly groups genetically with B. crenulatum.
The most distinct group was formed by Oregon plants. Goofy was the only
population to form an exclusive cluster. Goofy and Lapover grouped together,
but some members of Lapover also clustered with Stewart, the Montana
population, in the mixed cluster. cLapover.09 was the most dissimilar member
of the B. crenulatum cluster, and in fact displayed three bands otherwise found
only in B. minganense, B. lunaria, or B. simplex (C-11¢75, D-16499, and D-16,75)
as well as all the bands displayed only in B. crenulatum.

Botrychium minganense.—In the B. minganense group (Fig. 6), three
populations formed exclusive clusters: Watson (Oregon), Poison (Oregon),
and Devil (Washington). All others formed mixed groups. Thirteen plants from
the Idaho Panhandle and neighboring Montana populations (Rock, Deer, and
Kelsey) had identical profiles. Some members of each of those populations

SWARTZ & BRUNSFELD: MORPHOMETRIC ANALYSIS OF BOTRYCHIUM SPECIES 261

| B. crenulatum

| B. minganense

| B. lunaria

=

B. simplex

eo Se T ma aes T is T | ae, T T T a oo | a amas |
0.72
Similarity coefficient
Fic. 3. Species clusters from UPGMA dendrogram of RAPD data from a sg of 194 plants of four
Be B,

moonwort species: Botrychium crenulatum, 128 B. minganense, 12 B. lunaria, and six
simplex. The scale represents the similarity coefficient between iar a s.

clustered with other groups. One population, Aladdin1, was monomorphic.
Members of WenR and WenT, which grew adjacent to each other, each grouped
in a separate larger cluster. Mill, from the Washington Cascades, and Manley,
from northeastern Washington, were particularly diverse: each had members in
four different larger clusters, and other members that were highly divergent.

DISCUSSION

Morphometrics.—The cryptic moonwort species Botrychium minganense
and B. crenulatum can be separated by canonical variate analysis into two
partially overlapping groups. Plotting the CAN1 scores of 171 plants whose
identity had been genetically confirmed showed that 32% fell in the zone of
overlap where the two species could not be separated using the characters
scored. The characters that contributed most to the separation were measures
related to pinna shape, proportions of the trophophore, and size. Pinna
fanning, as reflected in the pinna shape characters, is emphasized in
descriptions of B. crenulatum (Wagner and Wagner, 1981; W. H. Wagner,
1993). Average size of B. minganense is larger than that of B. crenulatum (mean
height of sampled plants 84 mm and 73 mm respectively), but each can be less
than a centimeter tall (pers. obs.). The ratio, width of trophophore: maximum
width of trophophore axis, is a complex character that combines elements of
several differences between the species, including length of pinnae, angle
at which the pinnae meet the axis, and tendency of pinnae to be decurrent on
the trophophore axis or of the axis to be flattened. These characteristics have
not been emphasized in taxonomic descriptions of B. minganense and

262 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 4 (2002)

LRd9576.01

s ead.
sLGMead.05

aaa
0.44 0.58 0.72
Similarity coefficient
Fic. 4. Detail of Botrychium lunaria and B. simplex clusters from UPGMA dendrogram of RAPD
data (Fig. 2). Plants are labeled by population abbreviation (Table 1) and individual number within
population. Leading letter signifies species (L = B. Junaria, s = B. simplex). The scale represents
the similarity coefficient between clusters.

B. crenulatum, although the trophophore of B. minganense has been described
as narrow (Wagner and Wagner, 1993).

The statistical analysis of morphology was limited compared to field
identification. Some useful morphological characters of live plants could not
be used in an analysis of herbarium specimens. The characters that could not be
captured include phenology, color, texture, and many aspects of plant habit,
such as cupping of the pinnae. Other useful characters, such as the number
of pinna pairs, or number of crenulations on pinna margins, are not nor-
mally distributed and therefore violate the requirements for canonical variate
analysis.

The 32% ambiguity rate in the morphological analysis contrasts with the
correct field identification of all but seven of the 171 analyzed plants.
However, in the field, individual plants are not independently identified, as is
the case with the statistical analysis. Field botanists generally examine the
range of variation at a site and make an identification on the basis of a group of
typical plants. Some well-developed plants will show characters that smaller
ones lack. This is also true of herbarium identification. In fact, consulting
botanists request collections of about a dozen plants for identification of
moonwort species (Wagner, 1992; Wagner and Wagner, 1993; Zika et al., 1995).
This study generally supports the assumption that similar plants associated at
one site belong to the same species. In the genetic analysis, none of the
populations identified in the field as containing B. minganense and not B.
crenulatum, or B. crenulatum and not B. minganense, contained sampled
individuals of the other species. However, even experienced botanists can
occasionally be misled (example given in Farrar, 1998). Although we sought
mixed populations for this study, only three of 23 (Hodgson, Manley, Lapover)
had mixed B. minganense and B. crenulatum populations.

SWARTZ & BRUNSFELD: MORPHOMETRIC ANALYSIS OF BOTRYCHIUM SPECIES 263

cGoofy.01 |

2
RARE
uosdIO

cDeadman.06
cHodgson.03
c :

cHodgson.07

|
& > Ser ht oss 556 g
apeooes vA
ee CERES REE RO
BG S566s55e5S
PP oor Ss! ee: ge
2 Seal eae he
UO}FUTYSE AY

Aladdin.
cLapover.09
aA ons 8 ae ee T T T T T T T 1 aL T T 1
0.44 0.58 0.72 0.86 1,00
Similarity coefficient

Fic. 5. Detail of Botrychium crenulatum cluster from UPGMA dendrogram of RAPD data (Fig. 3).
Plants are labeled by population abbreviation (Table 1) and individual number within population.
Leading letter signifies species as identified in the field (c = B. crenulatum, “‘m” = B. minganense
as identified in the field). The scale represents the similarity coefficient between clusters.

RAPD analysis: Interspecific variation.—In contrast to the morphometric
results, all sampled plants, with one exception (cLapover.09), grouped clearly
by species based on RAPD markers. Primers C-10 and/or D-11, run with
a known B. crenulatum sample, would be sufficient to confirm or rule out the
identification of B. crenulatum. Separation of B. minganense from B. lunaria
by means of RAPD markers requires more primers; because neither has unique
bands and they are separated in the similarity analysis by differences in

264 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 4 (2002)

mWatson.01 mF lowery.02
ie b anes 5
a mM G3 mbower “
é ock.
mS Ab) mWenR.28
mLaG32.02 mManley.03
mShady.04 mManley.05
mShady.02 mFlowery.06
mShady.03 mFlowery.08
mShady.06 mF lowery.09
mMiullB.01 mFlowery.07
Sena net
i ulldo
mese33 10 mBulldog 07
mr ae33 05 Wen 39
aG32. e
mAladdin1.01
mataddint 03 mBulldog.o4
inl. ulldo
mAladdin1.04 Idog
matadgm!.t mRd95 1
inl.06 mRd9576.02
mHodgson.04 mRd9576.03
mHodgson.02 ey
mManley 01 mRd9576.05
mi jodgson. 7 mKelsey.02
mHodgson.08 mKelsey.03
mHodgson.09 mKelsey.05
mHodgson. | 1 mRock:01
mManley. mRock.02
mManley. mRock.03
mManley. | mRock.
mManl mRock.05
mManleyX.05 mDeer.03
mPoison.01 mDeer.04
mipowson.O¢ mieer.5
‘olson. mDeer.

Al mPoison.03 mDeer 08
mPoison.06 mKelsey.06
mPoison.09 mKelsey.08
mPoison. 10 mManleéy. 1
mMuill. 04 or. 06
mM mes 07

all. elsey.
mMill.07 msrady 01
mDustyR.03 i mMill 13
mDustyR.01 mRock,06
mDustyR.02 mRd9576.04
mDuse 04 mManiey. 02
mDustyY O1 mManley 07
LS ELA eee ee ee mMillB.03
0.72 0.86 1.00 mMfaniey, 16
Similarity coefficient mWenT 02
mWenT .04
mWenT.06
mMill.06
B mWenT.05
mWenT.07
mWenT .03
mMill.09
mMill. 11
. mMill. 05
mMill.08
mMill. 10
A mMill. 14
mMill.01
: mDevil.03
B. minganense mDevil 04
peal t
evil.
mDevil.07

B mDevil.08

mDevil. 11
0.72 0.86 1.00 0.72 0.86
Similarity coefficient Similarity coefficient

Fic. 6. Detail of Botrychium minganense cluster (c) and subclusters (a, b) from UPGMA
dendrogram of RAPD data (Fig. 3). Plants are labeled by population abbreviation (Table 1) and
individual number within population. Leading letter signifies species as identified in the field (m =
B. minganense). The scale represents the similarity coefficient between clusters.

SWARTZ & BRUNSFELD: MORPHOMETRIC ANALYSIS OF BOTRYCHIUM SPECIES 265

frequencies of bands. All B. simplex sampled had one band that appeared to be
unique to the species, but more populations must be sampled to verify this as
a species marker. The assignment of plants to species based on RAPD markers
agreed with their classification in the field, with one exception. That plant,
‘“‘m’’Hodgson.05, had a CAN1 score of 0.88 in the morphometric analysis
(Fig. 2) and fell in the range of minganense/crenulatum overlap. It was
field-identified as B. minganense, but had the RAPD pattern of B. crenulatum.
It was a small plant with non-crenulated pinnae, from a mixed population of
B. minganense and B. crenulatum.

RAPD analysis: Distribution of variation within species.—RAPDs provide
greater evidence of variability in species of Botrychium than isozymes have.
Hauk and Haufler (1999) reported 14 isozyme genotypes among 252 plants of
B. minganense, and one genotype among nine plants of B. crenulatum. In this
study, there were 100 RAPD genotypes among 128 individuals of B.
minganense, and 28 genotypes among 48 plants of B. crenulatum. Within five
populations of B. minganense (Watson, Bulldog, mHodgson, LaGrande32, and
Dusty) and one of B. crenulatum (Deadman), no two sampled individuals had
the same RAPD profile.

The population showing the highest genetic similarity among individuals
was B. minganense—Aladdin1. This small population of about ten plants
growing in approximately 9 m*, was sampled heavily because it was
morphologically ambiguous, and could not be identified to species in the
field (W. H. Wagner, Jr., pers. comm.). The plants were small, light green, and
displayed rather broadly fanned pinnae. Three of the sampled plants had
CAN1 scores in the 0 to 2 range where the scores of species groups overlapped,
and three had slightly negative scores in the ‘‘minganense” range. This was the
only population that lacked within-population variation (scorable or unscor-
able) on the gels, and might be described as clone-like. All other sampled
Botrychium populations were larger, and had more than one RAPD profile.

Most individuals from the Deer (Idaho), Rock (Idaho), and Kelsey (Montana)
populations of B. minganense had the same RAPD profile. These populations
were located within approximately 50 km of each other. However, proximity
was not a good predictor of genetic similarity across all populations. The twin
B. minganense populations from Wenatchee Ford, in Washington, grew about
30 m apart and were reported as one population on the Wenatchee Forest
sensitive plant sighting form. They were kept separate in the analysis because
one group was growing in deep shade in a riparian zone under Thuja plicata
and Tsuga heterophylla, and the other was under Rubus parviflorus on
a roadside. The Tsuga group clustered with the Mill population from
a mountain meadow on the Wenatchee Forest, but the Rubus group clustered
with Flowery and Bulldog, shaded forest sites from northeastern Washington,
approximately 225 km away. No association between ecological sites and
RAPD genotype is evident in this or other clusters.

Although some populations, such as Goofy (B. crenulatum) and Devil (B.
minganense), showed low similarity to any other, most populations had
members in more than one cluster. For example, some B. minganense plants

266 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 4 (2002)

from Mill clustered with plants from Manley in northeastern Washington,
whereas others clustered with Shady and LaGrande32 in northeastern Oregon,
about 400 km from the Mill site.

Genetic variation within B. minganense did not suggest any coherent genetic
groups that might be associated with its morphological variability. Hauk and
Haufler (1999) reported more isozyme variability within B. minganense than
any other polyploid sampled. Given that RAPDs are revealing more variability
than isozymes, additional sampling from across the species range may reveal
genetic patterns within the species.

Genetic structure of populations.—The contrasting patterns of genetic
similarity may result from processes that hinder or promote genetic isolation.
Two important factors influencing the structure of genetic variation in plants
are breeding system and dispersal of propagules. The breeding system of
moonworts is not known from experimental investigations, because they have
not been cultivated successfully. The most direct evidence comes from the
allozyme work of Hauk and Haufler (1999) on other species of subgenus
Botrychium. Low variability within populations hampered their inferences of
breeding systems, but they attributed the low frequency of heterozygotes found
in four populations of diploid moonworts (B. simplex and B. lanceolatum) to
inbreeding. Electrophoretic studies on Botrychium species in subgenus
Sceptridium (McCauley et al., 1985; Watano and Sahashi, 1992) and subgenus
Osmundopteris (Soltis and Soltis, 1986) reported extremely high levels of
inbreeding. Outcrossing may be hindered by the underground gametophytes of
this genus (Tryon and Tryon, 1982), although moonwort hybrids have been
reported (Ahlenslager and Lesica, 1996; Wagner, 1980, 1991; Wagner et al.,
1984; Wagner, Wagner, and Beitel, 1985; Wagner and Wagner, 1988),
demonstrating at least occasional outcrossing. The number of allopolyploids
also documents that outcrossing is an important evolutionary process in
subgenus Botrychium (Hauk and Haufler 1999). In our study, the lack of
genetic diversity in the RAPD profiles of the small Aladdin1 population is
consistent with inbreeding.

Genetic variability within populations of an inbreeding species could be
increased by immigration of propagules from distant sources, and occasional
outcrossing. Fern spores are light and can travel long distances, as ferns
colonize remote islands (Tryon, 1970; Tryon, 1986; Ranker et al., 1994). Tryon
(1970) presented evidence that 800 km is not a significant barrier to the
migration of a fern flora. Tryon and Tryon (1982) characterized Ophioglossa-
ceae in particular as a colonizing group.

Because of the dominant inheritance of RAPD markers (Bachmann, 1997;
Crawford, 1997), these data do not provide unequivocal insights into the
breeding system of moonworts. Although the variability detected in this study
may not have been predicted based on isozyme studies, it is not inconsistent
with a high dispersal rate and a largely inbreeding mating system.

Ancestry of B. minganense.—The rbcL sequence of tetraploid B. minganense
did not match that of any known diploid (Hauk, 1995). On the basis of the
match between the hypothetical isozyme profile of the non-chloroplast parent

SWARTZ & BRUNSFELD: MORPHOMETRIC ANALYSIS OF BOTRYCHIUM SPECIES 267

of B. minganense and the isozyme profile of B. crenulatum, Hauk and Haufler
(1999), proposed B. crenulatum as the most likely candidate for that parent.
The RAPD data, however, did not support this relationship, because B.
crenulatum showed seven unique bands absent in B. minganense. An earlier
hypothesis (F. S. Wagner, 1993) based on morphological data, proposed B.
lunaria and B. pallidum as the parental diploids. The RAPD evidence is
consistent with a close relationship between B. minganense and B. Junaria,
because neither had bands that did not occur in the other. More genetic
evidence is needed to clarify the origins of B. minganense.

cLapover.09.—The identity of one plant from the Lapover site in the Lostine
River Valley, Oregon, was uncertain when it was collected. It combined the
color and luster of B. crenulatum with rounded, broad-based pinnae otherwise
seen only in B. minganense. The RAPD profile of this plant included all seven
diagnostic B. crenulatum bands, plus three characteristic B. minganense bands
including C-1157; and D-16,75 (also polymorphic in B. Junaria and present in
B. simplex), and D-16499 (polymorphic in B. Junaria and not present in B.
simplex). Eight bands documented in all sampled B. minganense were not
present in the plant. cLapover.09 appears to be a hybrid, both because of
intermediate morphology and mixed markers. The three “‘minganense”’
markers could also have come from B. Junaria or B. simplex, but these species
were not recorded from the site, whereas B. minganense and B. crenulatum
were present. Other moonwort species recorded at the site were B. ascendens
and B. lineare, for which we have no RAPD data. Neither of these moonworts
typically has rounded pinnae. Because not all of the diagnostic B. minganense
bands were present, cLapover.09 does not appear to be an F, hybrid between B.
minganense and B. crenulatum, but is more likely a backcross or later
generation hybrid derivative.

CONCLUSIONS

Although many plants of B. minganense and B. crenulatum could not be
reliably distinguished by canonical variate analysis of morphology, all
sampled plants, except an apparent hybrid, could readily be assigned to
species on the basis of RAPD profile. This supports the distinctness of the two
species although their morphologies intergrade.

Although breeding system cannot be inferred from dominant markers such
as RAPDs, higher levels of variability were detected within populations and
species than might be predicted from previous genetic data, which suggested
a high level of inbreeding. Thus codominant DNA markers such as micro-
satellites might be a productive avenue for further research into breeding
system and evolutionary processes in Botrychium.

ACKNOWLEDGMENTS

Thanks to Kathy Ahlenslager, Elroy Burnett, Leslie Ferguson, Joann Harris-Rode, Jerry Hustafa,
Kirk Larson, Mick Mueller, Mark Mousseaux, Scott Riley, Faye Streier, Jim Vanderhorst, Kari

268 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 4 (2002)

Yanskey, and Gene Yates for help with collecting. For illuminating the mystery of the moonworts,

improved by the comments of two reviewers, W. Hauk and R. Moran, but any errors are the
responsibility of the authors. This study was supported by the C. R. Stillinger Trust, University of
Idaho.

LITERATURE CITED

AHLENSLAGER, K. and P. Lesica. 1996. pipe rem of Botrychium oso and its putative
parent species, B. hesperium and B. paradoxum. Amer.

BACHMANN, 28 1997. Nuclear DNA markers in pers biosystematic emer Opera Bot. 132:
137-14

CRAWFORD, 2 J. 1997. Molecular markers for the pea of genetic variation within and between
uae of rare plants. Opera Bot. 132:1

Farrar, D. R. 8. Isozyme cies her er moonwort ferns (Botrychium subgenus
panei aa their interspecific relationships in northwestern North America. Draft final

Gastony, G. J. and M. D. WinpHaM. 1989. Species concepts in tat the treatment and
definition of agamosporous species. Amer. Fern J. 79:65—
GLeaAson, H. A. and A. Cronquist. 1991. Order ees. tee 13, in Manual of vascular
lants of northeastern United States and Adjacent Canada, second edition. The New York
ork.

Haroic, T. M., S. J. BRUNSFELD, R. S. Fritz, M. Morcan and C. M. Orians. 2000. Morphological and
molecular evidence for hybridization and introgression in a willow (Salix) hybrid zone. Mol.
Ecol. 9:9-24.
Hauk, W. D. 1995. A molecular assessment of relationships Bi oa species of Botrychium
ieee Botrychium (Ophioglossaceae). Amer. Fern J. 8
an ER. 1999. Isozyme variability g CTY} i rol Bonychdum subgenus
—633.

=

Hircucock, C. L. and A. Cronquist. 1973. Botrychium. Pp. 44-45, in diana of the Pacific Northwest,
an Illustrated Manual. vag 3 of poucsa: jars Press, Seat

JOHNsoN, R. A. and D. W. WIcHERN. 1992. sith multivariate oe analysis, third edition.
Prentice Hall, Englewood Cliffs, New Jerse

JOHNsON-GroH, C. L. and D. R. Farrar. 1996. Effects of leaf loss on moonwort ferns, Botrychium
subgenus Botrychium. Amer. J. Bot. 83(6 suppl):127.

wane co T. 1994. A synonymized checklist of the vascular flora of the United States, Canada, and

eenland, second edition, Vol. 1-Checklist. Timber Press, Portland.
R. 198

es
ee D. B. 1985. Ophioglossaceae, Botrychium. Pp. 101-115, in A field manual of the
ferns and oe -allies of the United States and Canada. Shalarcias Institution Press, Wash-

ian See D. “" D. P. Wuirtier and L. M. REILLY. 1985. Spe ag rate of self-fertilization
i Tape ap tg aati dissectum. Amer. J. Bot. 7

Ricestan LD 0. Survivorship and sa changes in he eens of Botrychium
dissectum in ois Pennsylvania. Amer. Fern J. 80:173-182.

Net, M. 1978. Estimation of rc ai ct and genetic distance from a small number of
individuals. Genetics 89:583

Paris, C. A., F. S. WAGNER and W. a eke Jk. 1989. Cryptic species, — delimitation, and
taxonomic Les in the PRO LNCR RE ferns. Amer. Fern J. 79:46—

Ranker, T. A., S. K. FLoyp, M. D. WinpHaM and P. G. Trapp. 1994. Sane biogeography of
ple adiontem-nigram (A le A acm in North America an gid mplications for
speciation theory in homosporous pteridophytes. Amer. J. Bot. 81:776-78

SWARTZ & BRUNSFELD: MORPHOMETRIC ANALYSIS OF BOTRYCHIUM SPECIES 269

Rout_r, F. ? 1997. NTSYS-pc, Numerical Taxonomy and Multivariate Analysis System, ver. 1.8.
Exeter Software, http://www. oe e.com

SAS ae 6.11, SAS Institute Inc., ome ay, N

Soxtis, D. E. and P. S. Sottis. 1986. E for inbreeding in the fern Botrychium
virginianum (Ophioglossaceae). Amer. J. Bot. 73:588-592.

Sressins, G. L. 1950. Variation and Evolution in Plants. Columbia University Press, New
Work, Nu ¥:

Stewart, C. N. Jr. and L. E. Via. 1993. A rapid CTAB DNA megs a useful for RAPD

Tryon, R. M. and A. F. Tryon. 1982. Ophioglossaceae. Pp. 25-39, in Ferns and dillivad plate seth
oii reference to tropical America. Springer-Verlag, New York.
Wacner, D. H. 1992. Guide to the species of Botrychium in Oregon. United States Department of
Agriculture oe Service Report.
Leaf morphology of the Botrychium lunaria group in Washington and Oregon. United
rte ‘Departmen of Agriculture Forest Service Report.

WAGNER, F. S. 3. Chromosomes of North American ie ne and moonworts (Ophioglossa-
ceae: Bouya) Cont. Univ. ieee ant Herb. Pg
WAGNER, W. RB. Jr. 980. A probable new I t [ life li x simplex,

from central seu Michigan Bot. 19
9 examples of the moonwort iar Botrychium matricariifolium xX simplex
4.

o
=
Z
®

. Ophioglossaceae. Adder’s-Tongue Family. Pp. 98-103, in J. C. ener ed. The
Jepson manual eer plants of California. University of California Press, Berke
He . 1989. Moonworts (Botrychium: a a in the iemiosvill area,
Butte and Tehama counties, California. Madrono 36:131-—
and L. P. Lorp. 1956. The morphological and aa distinctness of Botrychium
minganense and B. lunaria in Michigan. Bull. Torrey Bot. Club. 83:261-280.
_R. C. Moran and C. R. WertH. 1993. Aspleniaceae Newman. Spleenwort Family. Pp. 228—
229, in Flora of North America Editorial Committee, eds. Flora of North America north of
Mexico, vol. 2. Oxford University Press, New York.
and F. S. WacNer. 1981. New species of moonworts, Botrychium subg. Botrychium
sis a emcee ee font North America. Amer. Fern J. 71:20—30
and . 1983. Two moonworts of the Rocky aeons Botrychium hesperium and
a new ica = omer confused with it. Amer. Fern J. 7

and 8. Detecting Botrychium hybrids in the jee es elite region. Michigan Bot.
27:75—80.
and . 1993. Ophioglossaceae C. Agardh Adder’s-tongue family. Pp. 85-105, in Flora

of North America Editorial Committee, eds. Flora of North America north of Mexico, vol. 2.

Oxford ane ig’ Press, New York.

. BeITEL. 1985. Evidence for a hybridization in eo aon

with shee mycoparasitic gametophytes. Proc. Roy. Soc. Edinburg 73-

HaurF.er and J. K. pee 1984. A new brent ne od ea
«. (Ophioslossacoae Botrychium). Can. J. Bot. 62:629-634

Watano, Y. and N. SaHAsHI. 1992. roar eareie inhreeding and its Sage consequences in
a oer fern genus, Sceptridium —— Syst. Bot. 17:486—502.

Wituams, G. K., A. R. Kuseuk, K. J. Livak, J. A. Raratskt and S. V. Tincey. 1990. DNA poly-
morphisms amplified by arbitrary primers are useful as pete markers. Nuc. Acids Res. 18:
6531-6535.

Z1KA, P. F. 1992. The results of a survey for rare Botrychium species (moonworts and grapeferns)
July-September 1991 in the Wallowa-Whitman National Forest (draft). Oregon Natural
ergs Program report to United States Department of Agriculture Forest Service, Portland.

R. B B. N. Newuouse. 1995. Grapeferns and moonworts (Botrychium,
zhioalcsoe) in the Columbia Basin. United States Department of Agriculture Forest

Service Report.

American Fern Journal 92(4):270—287 (2002)

Reproductive Behavior of Cloned Gametophytes
of Pteridium aquilinum (L.) Kuhn.

Forses W. ROBERTSON
41 Braid Farm Road, Edinburgh EH10 6LE, U.K.

ABsTRACT.—Spores from single fronds of three different taxa of Pteridium aquilinum (L.) Kuhn were
collected at different sites in Scotland, England and Sri Lanka. Gametophytes developed from
these spores were treated to produce arrays of genetically identical clones. Sporophyte formation
was determined when such clones from th different tophyt erived from the same

The breeding behavior of bracken gametophytes presents some unresolved
problems. Thus, Wilkie (1956) produced experimental evidence for genetic
incompatibility in bracken by recording the frequency of sporophyte formation
in combinations of clones from different gametophytes derived from single
fronds. The clones were produced by harvesting the prothalli which
proliferated from the margins of sectioned gametophytes. The results, in
clones prepared from three Scottish populations of bracken, could be plausibly
reconciled with the occurrence of two mating types in each population. Since
combinations of clones between populations were cross-compatible, it
appeared that a single locus, multi-allele system was present so that each
sporophyte would be heterozygous for dissimilar alleles. It was also noted that,
although well defined, the apparent incompatibility was not absolute since
a low, variable frequency of sporophytes occurred among putatively in-
compatible combinations.

On the other hand, Klekowski (1972) reported that single, isolated
gametophytes of bracken from different localities display wide variation in
sporophyte formation, ranging from zero to nearly 100%, with most samples
from different sites exceeding 30%. Self-fertilization in such gametophytes is
the rule and incompatibility is conspicuously absent. The differences in
frequency of sporophyte formation per sample were attributed to differences in
the frequency of recessive sporophytic lethals for which the parent plants were
heterozygous. He also noted that Wilkie’s findings could be explained if the
populations concerned carried balanced lethals, whereby each parent plant
would be a double heterozygote for recessive sporophytic lethals at two loci
linked in repulsion. Haploid gametophytes would carry one or the other of the

ROBERTSON: CLONED PTERIDIUM GAMETOPHYTES 271

lethals and thereby present the appearance of two mating-types. If the lethality
of either homozygous combination were incomplete i.e., the lethals were
leaky, apparent cases of breakdown in the incompatibility system, inferred by
Wilkie (1956), could be accounted for. Klekowski also indicated the need for
further study of Scottish populations that might prove atypical, a suggestion
that prompted the present study.

he experiments described here were designed to discover whether the
appearance of two ‘“‘mating-types’’, whatever their origin, could be detected in
bracken populations from Scotland, England and Sri Lanka. It is particularly
important to discover whether or not evidence from the British and Sri Lankan
populations, geographically separated and belonging to different subspecies,
leads to the same conclusions.

MATERIAL AND METHODS

SAMPLE SiTES.—Seventeen spore samples were obtained from single fronds
collected at the sites indicated in Table 1. Ten Scottish spore collections were
obtained from Clunie Dam (CD1 to CD5), Black Hill (BH3 and BH11), Temple
(T1), Rubery Reservoir (R1) and Edgelaw Reservoir (E1). Spores were also
obtained from one English site and two Sri Lankan sites: Farr’s Inn and
Bambarakanda Falls. Five samples (SL1 to SL5) were collected at the former
and one (SL6) at the latter site.

Three taxa are included in these collections. Both Pteridium aquilinum (L.)
Kuhn ssp. aquilinum and ssp. fulvum (Kuhn) Page & Mill. are included in the
Clunie Dam samples. CD2, CD3, and CD4 belong to ssp. fulvum and CD1, and
CD5 to ssp. aquilinum. The stand of ssp. fulvum is roughly triangular with
sides of approximately 20 m (Page and Mill, 1994). Three fronds CD2, CD3, and
CD4 were collected approximately 10 m apart from the west side of the stand.
The ssp. aquilinum fronds, CD1 and CD5 were collected adjacent to,
respectively, the north and south sides of the stand of ssp. fulvum, which is
surrounded by sporophytes belonging to ssp. aquilinum. The Black Hill site
refers to a roughly circular, isolated stand of ssp. aquilinum surrounded b
Calluna moor. Two fronds were collected 60 m apart. The English population
was from a scattered distribution of aquilinum.

From the Sri Lankan populations of ssp. revolutum (Kuhn) Wu Zheng-yi &
Raven, which is common in upland areas and the only subspecies in the
island, five fronds were collected over a 15 m distance within a fairly
continuous stand bordering one side of a road. The other site, (SL6) is several
miles away from Farr’s Inn and at a lower elevation by some 1500 m.

CuLTURE OF GAMETOPHYTES.—Spores were collected overnight by inverting
fertile fronds on paper. The spores were washed three times by centrifugation
in sterile water. Single drops of suspended spores were transferred by micro-
pipette to petri dishes with 1% agar ( Sigma A7002) made up with Knop’s
solution and sterile water (Wilkie, 1956). All cultures were kept at 20°C under
a standard fluorescent strip light, except for occasional periods under daylight

272 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 4 (2002)

TaBLE 1. Locations of the different single-frond spore collections from three sub-species of
Pteridium aquilinum (L.) Kuhn. Nomenclature for British samples follows Page (1982) and Page
and Mill, (1994), and for the Sri Lankan samples Wu Zheng-yi and Raven (1999). Map references
for British samples refer to the U.K. Ordnance Survey, Landranger Series.

Site Identification of spore collections Map reference
ssp. aquilinum

ie Dam CD1, CD5 GR NN 915 592

Black Hill BH3, BH1 GR NT 184 636
Rubery Reservoir R1 GR NT 311 571
Edgelaw Reservoir El GR NT 306 581
Temple 11 GR NT 312 583
Hutton-le-Hole a GR SE 700 890

ssp. fulvum
Clunie Dam CD2, CD3, CD4 GR NN 915 592
ssp. revolutum

Farr’s Inn Sit) SP2,/SU4. SLAUSES 80 E496 E 49
Bambarakanda Falls SL6 80 E516N 46

and ambient temperature, which applied equally to all cultures within a set of
comparisons.

The frequency of sporophyte formation by gametophytes was recorded
under the following conditions. When thalli had grown to about 0.5 cm
diameter, a random sample was transferred individually to small compart-
ments, 2 cm square and 1.7 cm deep, in plastic boxes made up of 25 such
compartments, each provided with sterile, washed sand moistened with
Knop’s solution. The appearance of a sporophyte was attributed to self-
fertilization. Pairs of such gametophytes, whose members were from different
sites, were also kept under similar conditions. Any sporophytes which
appeared in the latter comparisons may have arisen by selfing or crossing
between gametophytes.

A different kind of experiment was carried out with cloned gametophytes
derived from single thalli. This entailed the combining of the cloned
gametophytes in pairs, either according to a regular scheme described below,
or randomly combined within or between different spore samples.

To produce clones, young spore derived gametophytes were either treated
for five minutes with 0.5 M KCl (Dyer, 1979) and then washed with water or
they were cut into segments. Most, but not all, gametophytes treated either way
and kept thereafter on Knop’s agar substrate produced many small thalli
around the margins. These small thalli were removed and grown to produce
arrays of genetically identical clones. For each sample of spores from a given
collection, 25 randomly chosen gametophytes were used to produce clones.
Either method of producing clones led to the same conclusions. Treated
gametophytes differed in the rate of formation of daughter clones. When
twenty or more clones became available, for at least nine or ten treated
gametophytes, the clones were removed to set up the experiments described

ROBERTSON: CLONED PTERIDIUM GAMETOPHYTES 273

Taste 2. Diallel combinations of cloned gametophytes. The numbers 2 and 4 refer to the potential
maximum number of sporophytes for combinations of clones from, respectively, the same or
a different gametophyte. Combinations between identical clones occur once, but twice between
clones from different gametophytes.

Clone numbers

6

—
Co

NP PB] w&
ee
ee a

NP bP
NPP PPP PIN
ee oe)
Se a co)

POON OO FW DH eR

ee
o

below. In the case of CD4, for reasons explained later, a particular test was
repeated with sets of clones that had developed later.

To compare the behavior of cloned gametophytes they were transferred in
appropriate pairs, when about 0.5 cm in diameter, to individual compartments
of the plastic boxes under the conditions noted above. After ten to fourteen
days they were irrigated with aerated tap water. This was repeated at intervals
until two to three weeks had elapsed without the further appearance of
sporophytes, when the experiment was terminated. The presence of spor-
ophytes was determined by inspection with a low-power binocular micro-
scope. To avoid possible damage due to handling and to avoid the risk of
contamination, the occurrence of archegonia and antheridia was not followed
during these tests. As will be apparent later, that in no way detracts from the
significance of the evidence but points to an obvious subject of future enquiry.

ANALYTICAL PROCEDURE.—Genetic incompatibility has been claimed to account
for the reproductive behavior of cloned gametophytes. An obvious way to
check this is to set up a N X N diallel mating system whereby members of the
arrays of clones derived from the same frond are combined in pairs in all
possible ways, including combinations between sister clones. This results in
the mating scheme illustrated in Table 2. For convenience it can be collapsed
into the indicated triangular form. The lower diagonal position (diagonal slots)
refers to the pairing of sister clones while all the other positions refer to
combinations between the arrays (e.g., 1 X 2, 2 X 1). Since a gametophyte,
cloned or otherwise, has the capacity to produce a single sporophyte, the
maximum number of sporophytes expected in the diagonal slots is two
whereas four are expected for all the other, duplicate combinations shown in
the diagram. Interpretation of the reproductive behavior of clones depends on
the nature of the departure from the numerical distribution shown in Table 2
when they are combined in such a diallel scheme.

274 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 4 (2002)

It is first necessary to consider how the presence of a simple genetic
incompatibility system would affect the distribution of sporophytes in a diallel
test. In the case of two, equally frequent haploid mating types (+ and —) where
only the heterozygote (+/—) will give rise to a sporophyte, the results of
combining gametophytes from a heterozygous individual can be represented as:

+ = + -
O 2] + or, more succinctlyas O 4] +
2 O! - ot =

Extending the same sort of diagrammatic representation to the simplest
hypothesis of diallel combinations in which the gametophyte clones in
individual arrays are all (+) or all (—), the results can be ordered to display the
characteristic pattern shown in Table 3a. Note there is a rectangular set of
positions, representing the heterozygotes, with a maximum number of four
sporophytes. All the other positions will fall into one of the two triangles that
represent either the (++) or the (~—) homozygotes; these do not produce
sporophytes. In a diallel test of this kind the practical task is to see whether the
order of the paired gametophytes can be arranged to display the characteristic
pattern of sporophyte production.

Exactly the same kind of pattern can be generated if the parent individual is
heterozygous for two recessive sporophytic lethal genes at two loci linked in
repulsion, in which case only the double heterozygote will give rise to
a sporophyte. This balanced lethal situation (Table 3b) leads to the same
pattern of sporophyte production as the case of simple incompatibility,
provided the linkage is complete. For either hypothesis the model assumes
equal numbers of the alternative genotypes among the gametophytes which
give rise to the clones used in any diallel test. In practice there will be chance
variation about the 1:1 ratio and this will lead to corresponding departure from
the precisely symmetrical pattern of the theoretical distribution illustrated in
Tables 3a and 3b. Where it is necessary to test for departure from a 1:1 ratio the
Chi-Square test has been used.

RESULTS

SPOROPHYTE PRODUCTION IN NON-CLONED GAMETOPHYTES.—Young gametophytes
were removed from the agar plate at random and allowed to develop either on
their own (Table 4) or in the proximity of another gametophyte derived from
a different frond of the same or a different taxon (Table 5). In the first situation
the comparisons include samples from ssp. aquilinum (CD1, CD5, E1, R1, and
T1), from ssp. fulvum (CD2, and CD4) and from ssp. revolutum (SL1, SL3, SL4,
SL5, and SL6). Among these isolated gametophytes, the frequency of
sporophyte production ranged from 0.16 to 0.76, with an average of 0.43.
These data are consistent with the variation reported by Klekowski (1972). In
the second design, the combination of gametophytes from different sources
resulted in a much higher frequency of sporophyte production. Almost all
(0.95) of a total of 598 such combinations produced at least one and often two

ROBERTSON: CLONED PTERIDIUM GAMETOPHYTES 275

TABLE 3. The potential maximum numbers of sporophytes produced by crossing, in all possible
ways, clones from eight different gametophytes derived from a sporophyte heterozygous for either:
a) (+) and (—) genotypes or b) balanced lethals linked in repulsion. It is assumed that alternative
haploid genotypes are equally frequent in each situation. The diagonals refer to the single
combinations of identical clones; all other combinations occur twice.

a) gametophytes derived from a sporophyte heterozygous for (+) and (—) genotypes

Clone numbers and genotype

Af 2 3 4 5 6
+ a + + - - - =
0 0 0 0 4+ 4 4 4 ab 1
0 0 0 a + 4 4 + Z
) 0 4 4 4 4 + 3
0 4 4 4 4 + 4
0 0 0 0 = 5
0 0 0 = 6
0 0 = f
0 = 8

b) gametophytes derived from a sporophyte heterozygous for balanced lethals linked in repul-
sion

Clone numbers and genotype

1 2 3 4 5 6 7 8
at! = is a a os = Lan

0 0 0 4 4 4 4 +1 1

0 0 0 4 4 4 4 +1 2

0 0 4 4 4 4 cad 3

0 4 4 4 4 +1 4

0 0 0 0 1+ 5

0 0 0 fe 6

0 0 1+ 7

0 1+ 8

sporophytes, whether the combinations were within or between taxa (Table 5
The lower production of sporophytes in isolates suggests a high incidence of
recessive, sporophytic lethals at different loci. Given full penetrance of
sporophytic lethals and heterozygosity in the source sporophytes for one, two,
or three different, independently assorting lethals, frequencies of respectively
0.5, 0.25 and 0.125 are expected in random samples of isolated, selfed
gametophytes. All the tests referred to in Table 4 can be reconciled with
heterozygosity for either one or two lethals except T1 and SL5 in which the
proportion of combinations with a sporophyte significantly exceeds 0.5 (

01). However, for environmental or genetic reasons, not all lethals may be
fully expressed when homozygous. Lethals may be “leaky” so that sporopyhte
occurrence is higher than would otherwise be predicted, possibly so in T1 and
SL5. The high frequency of sporophyte production with gametophytes from
different sources (Table 5) is also consistent with the occurrence of recessive
lethals at different loci.

276 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 4 (2002)

TABLE 4. Sporophyte frequency in single, isolated, non-cloned gametophytes derived directly
from spores. N refers to the number and + and 0 to those with or without a sporophyte.

Origin N + 0 % fertile

ssp. aquilinum

CD1 50 14 36 0.28

CD5 25 15 10 0.60

E1 99 50 49 0.51

R1 100 55 45 0:55

Hig 99 66 33 0.67
ssp. fulvum

CD2 50 12 38 0.24

CD4 26 16 9 0.64
ssp. revolutum

SL1 25 rs 18 0.28

SL3 25 5 20 0.20

SL4 25 4 21 0.16

SL5 25 19 6 0.76

SL6 25 7 18 0.28

CLONED GAMETOPHYTES OF SssP. FULVUM.—Tables 6 and 7 show the diallel
combinations for the Clunie Dam samples derived from the stand of fulvum
(CD2, CD3, and CD4). The original gametophytes used to produce clones were
arbitrarily assigned numbers e.g., CD1 to CD25 to identify the clones derived
from a particular gametophyte. These are the axes in Tables 6 to 9 and 11 to 13.
For each analysis, the combinations have been arranged to best compare the
observed distribution of sporophytes with the patterns illustrated in 3a and 3b
of Table 3.

The distribution of sporophyte production in Table 6 for pair combinations
can be explained by the presence of two genotypes for which the parent frond
was heterozygous. When members of a gametophyte pair belong to the same
genotype, sporophyte formation does not occur but does so when they differ in
this respect. All sister clone pairs, derived from the same gametophyte (the
diagonal slots), failed to produce sporophytes and the pattern of presence or
absence of sporophytes corresponds to the pattern in both Tables 3a and 3b.
Thus, for CD2 clone numbers 1, 4, 5, 6, 15, 2, and 9 did not produce
a sporophyte or only rarely did when paired with a member of that set. Similar
results are seen for the members of the other set, clone numbers 3, 10 and 14.
However, when members of different sets were combined at least one and often
three or four sporophytes were produced. Occasionally, one or two
sporophytes occur where, according to either the incompatibility or balanced
lethal model, none are predicted. The same pattern is encountered with CD3.
The two categories or classes of clone include numbers 7,9, 20 and 21 on the
one hand, and numbers 4, 1, 9,12, 16 and 17 on the other, with the same
qualifications as noted for CD2.

The diallel test with CD4 was carried out twice (Table 7). In the first test,
performed at the same time as the tests with CD2 and CD3, all nine series of

AMERICAN Pace
FERN 2002
JOURNAL

QUARTERLY JOURNAL OF THE AMERICAN FERN SOCIETY

Editor
R. James Hickey
Botany Department,
Miami University,
Oxford, OH 45056
hickeyrj@muohio.edu

Associate Editors
Gerald J. Gastony, Department of Biology, Indiana University,
Bloomington, IN 47405-6801

Christopher H. Haufler, Department of Botany, University of Kansas,
Lawrence, KS 66045-2106

Robbin C. Moran, New York Botanical Garden,
Bronx, NY 10458-5126

James H. Peck, Department of Biology,
University of Arkansas—Little Rock, Little Rock, AR 72204

Table of Contents for Volume 92
(A list of articles arranged alphabetically by author)

ADAMKEWICZ, Li. (Se6'@ i. sEPIOREY <6 5 ccs cus aia ieke Ge hk bie de RR dcr
PE ee os nn had Sek ek i oe oR Res wR ee Bk bnew
Romie, ed: fee a, A ee eh ce FG Se Ke Re Ao ww en bdednwe

Cuiou, W.-L., D. R. FARRAR, & T. A. RANKER. The Mating Systems of Some Epiphytic Poly-
PP ates aici te aie in oe a PP eae Gued GOW awed Bs

Cg Bh COs I os eed 6 3b ae dw be eo ee na ks
Cusick, A. W. A Binomial for the Hybrid Polypodium of Eastern North America........
EAE i TO, BUR ose hg i ee wd Rw oh da eo ie nad hwaeweees
FARRAR, D. R. Obituary: Warren H. Wagner, Jr. (1920-2000) ...............-.0000.
PROGR, Be Eee i) hse sw en Ke o a5 oe Pup aken ae vor PE GRA

Gort igs, J. E. Lycopodium lagopus New in West Virginia .....................-..
GRANT J. R. (see W. H. WAGNER, Jr.)

HILDEBRAND, T. J., C. H. HAUFLER, J. P. THERRIEN, C. WALTERS, & P. HAMMOND. A New
ybrid Polypodium Provides Insights Concerning the Systematics of Polypodium
scouleri and its Sympatric Congeners

Houston, H. A. (see T. A. RANKER)

JOHNSON-GROH, C., C. RIEDEL, L. SCHOESSLER, & K. SKOGEN. Belowground Distribution and
bundance of Botrychium Gametophytes and Juvenile Sporophytes

sommiscet-tsmon, C, (are A ©. SrenevOn ny). os kc ok wb ee dk hl Sok es veo ken
KELLoFF, C. L., J. SkoG, L. ADAMKEWwICcz, & C. R. WERTH. Differentiation of Eastern North
American Athyrium filix-femina Taxa: Evidence From Allozymes and Spores

KNEPPER, D. A., D. M. JOHNSON, & L. J. MUSSELMAN. Marsilea mutica in Virginia

LELLINGER, D. B. Bibliography of Warren Herbert Wagner, Jr.

LELLINGER, D. B. Additions to the Fern Flora of Saba, Netherlands Antilles

BRELTAGER, TP, TD, Ga TRAIN ok i 2 4 GWA hin 444.64 bla ok 6 da uA one een wk,
MENDOZzA, A., B. PEREZ-GarciA, & R. RIBA. Comparative Research of Gametophytes of
Olfersia alata and Olfersia cervina (Dryopteridaceae)

MIcKEL, J. T. (see J. E. SkoG)

Montcomery, J. D. Pteridophytes of Upper Katanga (Democratic Republic of Congo) .. .

Morrow, A. C. & R. R. DuTe. Crystals Associated with the Intertracheid Pit Membrane of
es Wet Pictnl PPCM HURT 6.6 se 60a Haw Wn 649 Wh OR OME RRS Ce AO
MUSSELMAN, L. J. (see D. A. KNEPPER)

PAaCcHECO, L. Trichomanes ribae (Hymenophyllaceae), a New Filmy Fern from Costa Rica and
Panama

PALMER, D. D. Taxonomic Notes on Hawaiian Pteridophytes....................0-.
PERE GARCIA, .B (see A. NENDOAA) is ow. inca cua Jadhede oie. oe ew eins ae xin ee

PRADO, J. & D. B. LELLINGER. Adiantum argutum, an Unrecognized Species of the A. lati-
FOE PM at icunk nnn accuse bpadens CAO p oe eo eat be ones renee

PrApo, J. & A. R. SmitH. Novelties in Pteridaceae from South America ..............

RANKER, T. A. & H. A. Houston. 7 acu Sexuality in the Laboratory a Good Pre-

mate te Dena ey i, SUE gous 55. veka cdc oda ee eee eeews oe eed wane or
RANEER 1 A2(SEe Wists. OHIOU jiixal saca wie bd Oe ee ee es aha ee ae
RST ER oi Sees ee VIL POA ano Riri es DID Fee ena Ally bats SOL OD ack Mine
Te FS Es I 5 6 a 6 aw eG Pine PGW NEM a oie oa Hewisede bh des

ROBERTSON, F. W. Reproductive Behavior of Cloned Gametophytes of Preridium aquilinum
(L.) Kuhn

SANCHEZ, C. A New Filmy Fern from the Dominican Republic .....................
SCHOESSLER: L. ((see' ©. JOHNSON*GROH) 14.2 oeo oo dd Ped Boos See beh hee icks
DHBPRIEUESy Ei (hee- ds MAN) ahs ou Seas a oon al ei ae eee eee ote wos

SIMAN, S. E. & E. SHEFFIELD. Polypodium vulgare Plants Sporulate Continuously in a Non-
seasonal Glasshouse Environment .................000c ce eecccceeccuee

Skoa, J. E., E. A. ZIMMER, & J. T. MICKEL. Additional Support for Two Subgenera of Anemia
(Schizaeaceae) from Data for the Chloroplast Intergenic Spacer Region trnL-F and
MOL PHOlOR yao a seins sects us aes lO is BAe eho GA hue ie

PROG (See Ge Pew ROIs tn oad toting etl Materiel (Gl ott, tet ee a
SROGEN, ic. (see © JOHNSON=GRON) oi ons ha tatiana Shag abo seh sae

Situ, A. R. & R. B. CRANFILL. Intrafamilial Relationships of the Thelypteroid Ferns (Thel-
PONE CAG CAE) octet ain hate wre Hoe Ad nln id vied, AME eds Gs Oe Wie he har aoa

Sea Re (see 5, PRADO io:04 eh oee BER ee 8 hale wee be eaw eked baubce

STENSVOLD, M. C., D. R. FARRAR, & C. JOHNSON-GROH. Two New Species of Moonworts
(Botryvechian subs. Boirychiun) fom Alaska. sans 2 ec Sere sas CONES awe ee

Swartz, L. M. & S. J. BRUNSFLED. The Morphological and Genetic Distinctness of Botrych-
ium minganense and B. crenulatum as Assessed by Morphometric Analysis and
RAPE ARES oad nsed eg es ero ota ea omnes ts creek eens taal ates eae

TALBOT, S. L & S. S. TaLBor. A New Population of Aleutian Shield Fern (Polystichum aleu-
own C..Christens:) on Adak Island, Alaska s20% occancatcucsedseeane tees

PALBOE. Gc. (See Sr ls PAEBON) 6 4 cahe crys enabling Hodcln as. cee em AWE sense metas od

TAYLOR, W. C. Isoétes 3 herb-wagneri, an Interspecific Hybrid of . bolanderi 3 I. echino-
Cd | ee ee ee ee ee re. eee eee et

TPHERRIEN IP sei 0 it DEH RAINES Win oa.c to feed trea as toe lee oe. hdteae
"WAGNERY Ec ciy (S60 Eby CIR AN ore Gits 1) pie yrs ete as Sst ees Sa wk os Bets Oak

WAGNER, JR., W. H., & J. R. GRANT. Botrychium alaskense, a New Moonwort from the In-
MAE en Eo ei ks eg ae eo oo we 1k a a

WAGNER, W. H. (see P. F. ZIKA)

WALTERS, C. (see T. J. HILDEBRAND)

WarkIns, JR., J. E. & D. R. FARRAR. A New Name for an Old Fern from North Alabama ....
WERTH, C. R. (see C. L. KELLOFF)

WILSON, K. A. Continued Pteridophyte Invasion of Hawaii

ZIKA, P. F., E. R. ALVERSON, W. H. WAGNER, & F. S. WAGNER. Botrychium hesperium in the
Wallowa Mountains of Oregon

ZIMMER, A. (see J. E. SKOG)

Volume 92, Number |, January—March, pages 1—38, issued 7 March 2002

Volume 92, Number 2, April—June, pages 39-184, issued 30 July 2002

Volume 92, Number 3, July-September, pages 185-246, issued 25 October 2002
Volume 92, Number 4, October-December, pages 247-304, issued 30 December 2002

2239
119

ROBERTSON: CLONED PTERIDIUM GAMETOPHYTES 277

TaBLE 5. Frequency of at least one sporophyte per pair (+) when non-cloned gametophytes from
different locations are combined in pairs. In all the Clunie Dam (CD) pairs one gametophyte
belongs to aquilinum and the other to fulvum. The Sri Lankan pairs were derived from putatively
different individuals of revolutum, while in the remaining combinations (SL3 and BH11) one
gametophyte belongs to aquilinum and the other to revolutum. N refers to the number of pairs.

—

Pair combinations N ~ ) % fertile

ssps. aquilinum and fulvum
CD1 and CD2

174 170 4 0.97
CD1 and CD3 100 85 15 0.85
CD5 and CD2 Z 25 a) 1.00
CD5 and CD4 124 122 2 0.98
ssps. revolutum and revolutum
SL1 and SL3 25 25 0 1.00
SL1 and SL4 25 25 0 1.00
SL3 and SL5 25 22 3 0.88
SL3 and SL6 25 24 el 0.96
SL4 and SL6 25 22 3 0.88
ssps. aquilinum and revolutum
11 and SL 50 48 2 0.95
Total 598 568 30 0.95

clones failed to produce sporophytes or did so only very rarely. As noted
above, although the model assumes two equally frequent classes among the
gametophytes, derived from the single frond that gave rise to the clones,
chance will cause variation about a 1:1 ratio in a random sample. However, it
seems improbable, but not impossible, that nine gametophytes would belong
to a single genotype.

The test with CD4 was repeated with a second series of clones which had
developed later. This second test is in accord with the data from the previous
CD2 and CD3 tests, suggesting that the first test with CD4 was non-
representative. In the second test, one genotype included clone numbers 10,
14, 23, and 25 while the other included clone numbers 24, 1, 8, 11, 13, and 15.
To identify which of the two genotypes is the one represented in the first test,
clone numbers 18 and 19 of the first test were combined with all but one of
the different clones used in the second test. Table 8 indicates that all the
clones used in the first test belong to the same genotype as clone numbers 10,
14, 23, and 25 of the second test. Although unlikely, these data support a
departure from a 1:1 ratio of genotypes among the cloned gametophytes of the
first test.

It was noted earlier that CD2, CD3, and CD4 were derived from a single stand
of ssp. fulvum. It is therefore of interest to ascertain the genetic comparability
among them. Clones belonging to the alternative genotypes of respectively CD2
and CD3, CD2 and CD4, and CD3 and CD4 were combined in pairs. To
compensate for the occasional shortage of replicates, clones not represented in
the original test were included e.g., CD3 number 18 and CD4 number 3. One
combination was lost due to algal infection. Table 9 indicates identity of the

278 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 4 (2002)

TaBLE 6. Diallel tests with the Clunie Dam samples (CD2 and CD3) of cloned gametophytes of ssp.
fulvum. The clone numbers have been arranged to reveal the pattern of combinations that do or do
not produce sporophytes, as in Tables 3a and 3b.

CD2 clone numbers

zi 4 5 6 15 2 9 3 10 14
0 0 0 0 0 L 2 1 | 3 t
0 0 0 0 x | 1 a 4 4 4
0 0 0 0 0 ca 3 2 a
0 0 0 0 3 3 2 6
0 di 0 4 2 eS: 15
0 1 4 3: 4 2
0 4 3 4 9
0 0 0 3
0 3 10
0 14
CD3 clone numbers
7 19 20 aA | 4 1 9 12 16 17
0 0 Z Z ve 2 3 3 3 3 7
0 0 0 a a 2, 3 2 2 19
0 0 al Zz Zi 2 4 4 20
0 3 97 | a 4 2 ZA
0 0 0 0 0 0 4
0 0 0 0 0 ‘l
0 1 0 0 9
0 0 0 12
0 0 16
0 17

two genotypes in the three sets of cloned gametophytes. Pairings within
genotype failed to produce sporophytes or did so only rarely so whereas
pairing between genotypes often yielded the maximum of two sporophytes. By
cross referencing, the total number of clones listed in the tests described in
Tables 6 to 9 can be assigned to two genotypes comprising 24 and 16 samples
respectively. This is not significantly different from a 1:1 ratio (y2 = 1.6, p>
0.1). This evidence makes it likely that the stand of ssp. fulvum, from which
CD2, CD3, and CD4 were collected, constitutes a single individual. This
conclusion is also consistent with the results of randomly combined, paired,
cloned gametophytes either from the same or different fronds from the stand of
ssp. fulvum (Table 10, Sections i and ii). Among these pairs within fronds
there is a 1:1 ratio of combinations that produce sporophytes and those that fail
to do so. The same is generally true for combinations of clones derived from
different fronds, except for a statistically significant excess of pairs which
produce sporophytes when clones belonging to CD2 and CD3 were combined
(y* = 13.5, p < 0.01). This is in sharp contrast to the combinations of cloned
gametophytes between taxa, Sections iii and iv of Table 10. In these
combinations, there is a consistently high incidence of sporophyte formation
and often the maximum is produced.

ROBERTSON: CLONED PTERIDIUM GAMETOPHYTES 279

TABLE 7. The repeat diallel test with the Clunie Dam sample (CD4).

CD4 i)
2 6 16 17 18 19 20 Pa | 22
0 0 1 0 0 0 0 0 1 2
0 0 0 0 0 0 1 0 6
0 0 0 0 0 0 0 16
u | 0 0 0 0 0 17
0 0 1 0 0 18
0 0 0 0 19
0 0 0 20
0 0 21
0 22
CD4 ii)
10 14 23 25 24 1 8 11 13 15
0 0 10) 0 4 3 3 3 2 3 10
0 0 0 3 + 3 3 3 4 14
0 1 2 4 4 3 3 e 23
0 4 3 4 2 4 1 25
0 0 Z 0 0 0 24
0 0 0 0 0 1
0 0 0 0 8
0 0 0 gta
0 0 13
0 15

CLONED GAMETOPHYTES OF SSP. AQUILINUM—These experiments involve two
samples collected adjacent to and on opposite sides of the stand of ssp.
fulvum. Pairing among clones of either CD1 or CD5 were carried out in the
same manner as described for fulvum and the results are shown in Table 11. In
the CD1 diallel combinations the situation is similar to that already seen in
fulvum. The diagonal slots are all zero. It appears that clone numbers 1, 3, 7,
11, 14 and possibly 12 belong to one genotype and 2, 5, 8, and 9 belong to the
other. In CD5, although there is again evidence for the presence of two
genotypes, there are more exceptions to the predicted occurrence of
sporophytes. Thus, three of the ten pairings of sister clones give rise to one
or two sporophytes. Clone numbers 2, 5, 13, 17, and 10 probably belong to one
genotype and clone numbers 11, 12, 20, 4 and 6 to the other.

The two Black Hill samples (Table 12) also differ to some extent from each
other. In BH3, clone numbers 1, 3 and 9 appear to belong to one category and
numbers 4, 6, 12, 14, and 19 to the other; clone numbers 15 and 20 are
exceptions. In their case sister clone pairing leads to the appearance of either
one or two sporophytes in the diagonal slots. Also both numbers 15 and 20
produce sporophytes when paired with a member of either of the two sets: (1,
3, and 9) or (4, 6, 12, 14, and 19). In the balanced lethal model, this could occur
if linkage is incomplete so that recombination in the parent sporophyte
produces gametophytes which do not carry either of the sporophytic lethals.
Alternatively, interactions with genes at other loci might be responsible for

280 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 4 (2002)

+ £ + CTV\A

ABLE 8, g i identity 4
clone numbers 18 or 19 of the first test with the CD4 clones used in the second diallel test. Each
combination was represented by a single pair so the maximum predicted number of sporophytes

CD4 clones of second test

10 14 23 25 1 8 5G 13 15 CD4 clones of first test
0 0 0 0 7) 2 i z 2 18
0 0 0 0 Z i | 1 a | Zz 19

allowing sporophytes to develop when they would not otherwise do so. BH11
behaves like fulvum with two categories of cloned gametophytes that
distinguish, respectively, clone numbers 11, 21, 25, 22 and 18 from clone
numbers 24, 14 and 9. In the small York sample with five sets of clones there is
again evidence of two categories, numbers 4, 5 and 16 and numbers 1 and 22.

CLONED GAMETOPHYTES OF SSP. REVOLUTUM.—The diallel results for SL1, SL2, SL3,
and SL4 are shown in Table 13. All the diagonal slots are empty, except for one
exception in both SL2 and SL3, and there is the now familiar pattern of two
alternative sets of clones. For SL1, one set or genotype includes clone numbers
3, 4, 5, 6, 8, and 10 and the other includes clone numbers 1 and 2. In SL3 the
alternative sets comprise clone numbers 2, 3, 4, 6, 7, 8, 9, and 15, on the one
hand, and clone numbers 1 and 5 on the other. In SL4 the alternative sets or
genotypes are clone numbers 1, 3, and 4 and clone numbers 5, 2, 6, 7, 8, and 9.
None of these distributions depart significantly departs from a 1:1 ratio.

SL2 is inconsistent and illustrates the kind of exception referred to in CD5
and BH3. Thus, six of the clones fall into two categories, numbers 3, 4, and 5
and numbers 1, 7, and 9. Clone numbers 2 and 8 are exceptions. These clones
produce sporophytes when combined with a sister clone or a clone belonging
to either set (Table 13).

DISCUSSION

Three features of these experiments are of particular significance. Firstly,
with very few exceptions, sister clone pairs fail to produce sporophytes.
Pooling the data over all tests, only 12 sporophytes formed out of a total of 118
pairs of sister clones. Such infertility is not due to loss of potential to develop
the hermaphroditic condition on the part of cloned gametophytes since the
pairing of such clones derived from different spores collected at different sites
or from different taxa regularly led to a high and often maximum rate of
sporophyte production (Table 10).

Secondly, the evidence from the diallel experiments indicates that clones
derived from a single frond belong to one of two classes such that, sprorophyte
formation depends generally on the joint presence of a cloned gametophyte
from each class (Tables 6 to 9 and 11 to 13). It is assumed that the difference
between classes is genetic.

ROBERTSON: CLONED PTERIDIUM GAMETOPHYTES 281

TABLE 9. Tests of genetic identity among clones of the Clunie Dam samples of fulvum (CD2, CD3,
and CD4). Cloned gametophytes are combined in pairs. One combination (1 X 16) was lost due to
fungal infection.

CD2 clones
14 10 3 15 9 6 5 4 1 2
0 0 1 1 2 1 af 2 u 1 1
zt 1 0 z Zz 4 Zz L 1 1 9
0 1 0 1 2 1 Hi uN 1 A | 12
0 0 0 1 P 2 A 1 - e 16
@) 0 0 2 2 1 al 1 2 1 18 CD3 clones
0 BI ut 1 Zz 7) 2 2 if Z 4
i | HF 0 0 < lt 0 0 0 Z
aI Z at 0 0 0 0 0 0 el 19
aT 1 1 0 0 0 0 0 0 0 20
2 Z 2 0 0 0 0 0 0 0 21

CD2 clones
14 10 3 15 9 6 5 4 7 2
0 0 0 1 7 2 4 2 z y a
0 1 0 2 2 1 2 2 1 v4 6
0 0 0 a | 2 AE 1 2 y 2 16
0 0 0 1 2 2 1 1 2 1 tT?
0 ) L, 2 1 1 1 2 2 1 18 CD4 clones
0 0 0 Zz 1 g | z | 2 Z ? 19
0 | 0 | P4 i! : 3 a z 1 20
0 0 0 Z 1 1 Z 2 Z, a | 21
0 1 0 1 1 2 2 Z t 1 22

CD3 clones
- Z 19 20 yt | yi 9 +2 16 17
1 0 1 1 1 0 0 0 0 0 2
2 2 if ‘s | 0 0 4 0 0 6
A 2 1 0 1 0 0 0 0 0 16
0 Z Z Z t 0 0 0 0 0 47
1 2 1 2 z 0 0 0 0 (8) 18 CD4 clones
bi 1 0 2 1 0 0 0 0 0 19
0 1 Z 2 1 0 At 0 0 0 20
- Z 1 2 a 0 0 0 0 0 21
Ps 1 0 1 a‘. 0 0 0 0 0 22

Thirdly, when young, non-cloned gametophytes are isolated, self-fertilisa-
tion regularly occurs and the variable lack of complete fertility may be ascribed
to the incidence of sporophytic lethals for which the parent plant is het-
erozygous (Table 4). Thus, cloned and non-cloned gametophytes appear to
differ dramatically in their capacity for self fertilisation. It was this contrast
that prompted Klekowski (1972) to invoke the hypothesis of balanced lethals
and suggest the possibility of atypical behavior in the populations studied by
Wilkie (1956). Since the behavior of the cloned gametophytes is essentially the
same in samples belonging to different taxa, or derived from geographically

282 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 4 (2002)

TABLE 10. Frequency of formation of at least one sporophyte in random combinations of pairs of
cloned gametophytes derived from the same or different spore samples.

Sporophyte production

Samples N + 7 Frequency

i. fulvum—Clunie Dam

CD2 50 24 26 0.48

CD3 50 24 26 0.48

CD4 50 30 20 0.60

Total 150 78 72 0.52
ii. fulvum—Clunie Dam

CD2 and CD3 50 38 12 0.76

CD2 and CD4 50 22 28 0.44

CD3 and CD4 49 28 21 0.56

Total 149 88 61 0.59
iii. aquilinum and fulvum—Clunie Dam

CD5 and CD2 25 25 0 1.00

CD5 and CD4 25 24 1 0.96

CD1 and CD5 25 21 4 0.84

Total 79 69 5 0.92
iv. aquilinum and revolutum

BH3 and SL4 50 48 2 0.96

remote sites, it is likely that it holds for the Pteridium complex generally under
the conditions provided in these experiments. At first sight the appearance of
multi-allelic incompatibility looks like the obvious interpretation of the results
of the diallel tests and this interpretation certainly cannot be excluded,
although it encounters the embarrassing evidence for self fertilisation on the
part of isolated, single non-cloned gametophytes. If gametic incompatibility
does account for the reproductive behavior of cloned gametophytes it appears
necessary to infer that the process of cloning has altered physiology or
development to uncover an incompatibility system which is not normally
expressed in non-cloned gametophytes.

However, it is necessary to enquire whether the apparent contradiction in
the behavior of cloned and non-cloned gametophytes might be resolved within
the framework of what is known about the reproductive behavior of
gametophytes. Naf (1958) concluded that gametophytes form antheridia in
response to antheridogen secreted into the medium by other, more rapidly
growing gametophytes. If this external stimulus is absent, antheridia do not
form although archegonia do. Hence, if a gametophyte develops from a single
spore in isolation, it will be unable to undergo self-fertilisation. This appears to
hold generally although exceptions may occur, especially after long periods of
isolation.

It may be assumed that a cloned gametophyte behaves the same way and
similarly requires an external stimulus to produce antheridia. The minimum
requirement is the presence of another gametophyte that can provide the

ROBERTSON: CLONED PTERIDIUM GAMETOPHYTES 283

TABLE 11. Diallel tests on cloned gametophytes of Clunie Dam samples (CD1 and CDs) of
aquilinum. One combination (1 X 7) was lost due to algal infection.

CD1 clones
A! 3 7 11 14 AZ Z 5 8 9
0 0 - 0 0 2 = a zZ 4 |
0 0 Z 0 0 4 2 3 2 2
0 0 0 2 1 4 1 4 is
0 0 2 4 4 1 1 a
0 0 3 2 Z z 14
0 2 2 Ps A 12
0 0 0 0 2
0 0 0 a
0 1 8
0 9
CD5 clones
2 5 13 ae 10 Tt 12 20 4 16
0 0 0 zt Zz 3 3 1 it 2
0 0 0 1 3 3 a) | a 5
0 0 2 2 4 a 3 - 13
2 di 3 4 Zz | 3 17
0 2 3 3 = 2 10
0 1 2 0 0 ii
1 | 0 2 12
1 0 Hf 20
0 Zz 4
0 16

stimulus, but not any gametophyte can provide it. The diallel tests suggest that
the stimulus is generally present only when the two gametophytes concerned
are genetically different. Since antheridogen is accepted as the agent which
induces antheridia formation, one might wonder whether, within a species, it
constitutes a single chemical entity or might occur in different forms which
can be recognized by a gametophyte as different from the form it secretes, in
which case only exposure to a different form would allow antheridia formation
and hence the possibility of sporophyte formation. An alternative model could
be envisaged in which the antheridogen is constant but other compounds take
over the role just suggested, except that they would be responsible for
determining whether or not a gametophyte responded positively to the
presence of antheridogen.

At least, this hypothesis may have the merit of resolving the discrepancy
between the behavior of sister clones and isolated non-cloned gametophytes
and removes the need to invoke balanced lethals. Although separated at an
early age, it is likely that the latter have already been primed to produce
antheridia via exposure to the chemical stimuli contributed by genetically
different gametophytes on the agar plate.

number of Wilkie’s (1956) results can be accommodated within this
general scheme. Thus single, isolated gametophytes, derived from single
spores by micro-manipulation, only rarely gave rise to a sporophyte. When

284 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 4 (2002)

TABLE 12. Diallel tests with cloned gametophytes of aquilinum (BH3, BH11, and Y1).

—
i)
=
>
eal
©

o
oO
Oo © ©; 0
NNwWhN
CRN NM WwW] SF
SOON NNNWTOD
FP OONNN DN &
ONRP OR WWD &
Orooowwwwr
bs

BH11

=
lo}
bo
>
=
n=

co)
eos
Soo Ss

eo eo o/*
ee oo o>
COoWWW PF Pw
OR WWNHWS WE

Coco WwWRN WWW! O
i)
ht

Oo
ooo
OW FN] RB
me NN &
=
ior)

such gametophytes from the same frond were combined in pairs, only half of
the combinations produced sporophytes, an observation which suggested a role
for incompatibility, and which is also consistent with the diallel analyses and
the results of randomly combining cloned gametophytes from different arrays
derived from the same frond. The analysis of clone properties, in Wilkie’s case,
followed a procedure that differed from that used in the present experiments.
The capacity to form a sporophyte was ascertained for single cloned
gametophytes, when combined with sperm suspensions prepared from
alternative clones derived from the same frond. Two classes of clone were
detected and sporophyte formation was attributed to the union of genetically
different gametes. However, it seems not unlikely that the antheridia inducing
agent(s) will be included in such sperm suspensions, thereby inducing
antheridia formation that would not otherwise occur and making self-

285

ROBERTSON: CLONED PTERIDIUM GAMETOPHYTES

Diallel tests with cloned gametophytes of revolutum.

TABLE 13.

SL1

SL2

ce)

SL3

SL3

<<}

o

oS

286 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 4 (2002)

fertilization possible. Hence, the assumption of obligatory cross-fertilization is
open to doubt. It would be informative to repeat the experiment, using also
a fraction of the original sperm suspension from which sperm had been
removed by filtration. To distinguish between self- and cross-fertilization
requires genetic markers. It would be feasible to make such a distinction by
comparing appropriate, variable DNA sequences in tissues of sporophytes and
the gametophytes used in such experiments. There remains, however,
a problem in Wilkie’s report of the production of sperm by cloned
gametophytes. According to the hypothesis developed here that would not
be expected. One can only speculate whether the process of producing sperm
suspensions introduces special conditions not encountered in the present
experiments.

It was noted in the diallel tests, that there are occasional instances in which
a cloned gametophyte can produce a sporophyte when combined with any
clone of either of the two classes derived from a single frond. This suggests
there are additional genetic or environmental factors that can influence
antheridia formation or genetic incompatibility, if that is operating. It must
also be asked whether the process of cloning might introduce reactions, apart
from the speculative suggestion noted above, that would not occur in a normal,
uncloned gametophyte. Dopp (1959) in his studies of antheridia and
archegonia formation in Pteridium, noted that when thallus wings were
sectioned, antheridia arose only behind a cut in the area furthest from the
meristem which is responsible for a diffusible inhibitor of antheridia
formation. It is conceivable that clones that develop from different regions of
the original gametophyte may not be equivalent in their capacity to form
antheridia and this could contribute to a reduction in sporophyte number
below the maximum predicted in some instances. Such an effect would
introduce a stochastic element that would tend to obscure underlying
regularity in the recognition of two classes of clone per frond. If present, such
an effect appears comparatively unimportant since the regularity is not
obscured, whereas pairing of clones between taxa or sites, where it should
equally apply, regularly results in high rates of sporophyte formation.

One might also enquire whether the evidence from the experiments
described here is relevant to the natural history of gametophytes in the wild.
Little information exists on this score, although Voeller and Weinberg (1969)
have drawn attention to the natural scarcity of both gametophytes and young
plants. If antheridia formation depends on genetically determined stimuli, of
the kind suggested in the interpretation of the diallel tests, then the behavior of
gametophytes in the wild will be density dependent. If their frequency is low,
as seems to be often the case, then such gametophytes will behave in the same
way as gametophytes isolated from the spore stage: self-fertilisation will be
unlikely, or at least infrequent. At intermediate density, fertility will largely
depend on the joint presence of genetically different classes responsible for
reciprocal induction of antheridia formation and this will tend to promote
variability. One might also wonder how far a similar system might apply more
widely in ferns.

ROBERTSON: CLONED PTERIDIUM GAMETOPHYTES 287

Clearly many problems remain for future study, and the present experi-
ments, with their limited objectives cannot settle the issues raised. In
a different context, there is an unexpected by-product of the diallel analysis.
Systematic combination of cloned gametophytes in such a scheme provides an
empirical way of determining whether a particular stand of bracken is made up
of more than one individual. For example, analysis of the samples collected
from different locations within the Clunie Dam stand of ssp. fulvum (Table 9)
suggested that it comprised a single individual which had spread to occupy
a substantial area.

ACKNOWLEDGMENTS

Thanks are due to Drs. C. N. Page and A. Dyer for general information about ferns and especially
to Dr. Page for bringing the Clunie Dam situation to my notice and for collecting the fronds from
which the spores were obtained. I am indebted to Dr. Dyer for critical comment on the manuscript
and to Professor John Thomson for stimulating discussion.

LITERATURE CITED

Dopp, W. 1959 . Uber eine hemmende und eine fordernde Substanz in den Prothallien von

Dyer : :

KiekowskI, E. J. 1972. Evidence rsa FY og self-incompatibility in the homosporous Fern
Pteridium aquilinum. Evolution & —73.

Nar, U. 1958. On the physiology of id formation in the bracken fern Pteridium aquilinum

Pace, C. N. 1982. The Ferns of Britain and Ireland. Cambridge University Press, Cambridge.
and R. MILL. 1994. Scottish Bracken (Pteridium): new taxa and a new combination. Bot.
J. Scotland 47:139-140.
VoOELLER, B. and E. S. WEINBERG. 1969. gegen, and physiological aspects of antheridia
formation in ferns. Pp 77-93, in J. E nckel (ed.), Current Topics in Plant Science.
¥ as

Wik, D. 1956. Incompatibility 4 in bracken. Heredity 10:247—256.
Wu ZHENG-y1, W. and P. H. Raven. 1999. Flora of China. Missouri Botanical Garden Press. (St.
Louis).

American Fern Journal 92(4):288—293 (2002)

A New Population of Aleutian Shield Fern
(Polystichum aleuticum C. Christens.)
on Adak Island, Alaska

SANDRA LOOMAN TALBOT
Alaska Science Center, U.S. Geological Survey, 1011 East Tudor Road, Anchorage, AK 99503
STEPHEN S. TALBOT
U.S. Fish and Wildlife Service, 1011 East Tudor Road, Anchorage, AK 99503

TRACT.—We report and describe a new population of the endangered Aleutian shield fern
(Polystichum aleuticum C. Christens.) discovered on Mount Reed, Adak Island, Alaska. The new

because it increases the total number of known populations and individuals for the species.

The Aleutian shield fern, Polystichum aleuticum C. Christens., is one of the
most restricted and rare ferns in North America (Smith, 1985); it is listed as an
endangered species (U.S. Department of Interior, 1988). In a circumpolar
assessment of the rare vascular plants of the Arctic, the fern was classified
according to the IUCN Red List threat categories as ‘‘endangered”’ (Talbot et al.,
1999). The species was first collected by W. J. Eyerdam, an assistant to Eric
Hultén, in 1932 on Atka Island, one of the Andreonof Islands located in the
center of the Aleutian Island chain, Alaska (Fig. 1). Eyerdam’s holotype
specimen (W. J. Eyverdam 1086, 5 July 1932) is accessioned at S; isotypes are
accessioned at CAS, DS, and US. The species was first described by
Christensen (1938). Attempts to relocate the original collection site on Atka
in years subsequent to the species’ description were unsuccessful (Smith and
Davison, 1988). Then, in 1975, a population of the species was discovered, this
time on the northeast arm of Mt. Reed, on Adak Island (Fig. 1), also of the
Andreonof Islands of the Aleutian Island chain, Alaska (Smith, 1985). Several
subsequent searches, performed from 1986 to 1988, on various islands of the
Aleutian Island chain (Adak, Kagalaska, Atka and Attu; see Talbot et al., 1995
for review) failed to locate additional populations. Finally, in 1988 and 1993,
we discovered a second and third population, respectively, again on Mt. Reed
(Talbot et al., 1995). Three populations comprising approximately 117
individual fern clumps are known for the species (Tande, 1989; Talbot et al.,
1995).

In August 1999, genetic studies of Polystichum aleuticum were initiated to
assess the species’ relationship to P. lachenense (Hook.) Bedd. of Asia, as
recommended in Talbot et al. (1995). While collecting samples as part of this
genetic research, as well as ongoing systematic monitoring of the three known
populations (Anderson, 1992), we discovered a fourth population of P.
aleuticum, located approximately 142 m below the two populations found on

TALBOT & TALBOT: POLYSTICHUM ALEUTICUM IN ALASKA 289

176°45°W 176°. oe
; 2 | 9 %,
,
b NRE
@ Qs
SW |
P a)
9 ey O 4 Co
2 : OY ?
g pS Lake
2g al : @. Am
— Moun Resd
Sering Sea
5 fea <
Be a Ada
Rp D 0 ~ QQ Island
} “isicchaie

51°48'N

i
176°45'W
Polystichum aleuticum populations
0) Southern site on northeast arm of Mt. Reed

afittu Unimak
ce Aleutian Islands s
@) Northern site on northeast arm of Mt. Reed -

N

® Site on northwest arm of Mt. Reed Adak -* Va \@
"Reps —wRe * ——

@ se aca )

It. Reed

1. Location of the four known populations of the Aleutian shield fern, Polystichum
oa. on Mt. Reed, Adak Island, Aleutian Islands, Alaska.

the northeast arm of Mt. Reed, Adak Island, at an elevation of 338 m. As
was the case for the other three sites found on Mt. Reed, the fourth site was at
the base of a steep rock outcrop on a northeast-facing slope. The slope angles
at this site ranged from 60° to 90°. Notably, the site is located approximately
22 m below the lowest of the three previous populations (Population 3, Fig. 1,
Table 1), in an area considered too treacherous to survey during earlier efforts
due to steep, unstable, slippery slopes. This finding expands the elevational
range of P. aleuticum, placing the species at elevations between 338 m and
525 m (Table 1). Also, notably, the fourth site is located on a northeast-facing
slope. Climatologic records based on observations from 1950 to 1982 indicate
wind direction on Adak is predominantly from the west-southwest, averag-
ing 10.5 knots per hour, from June to November, the period of time during
which the habitat would likely be free of snow (U.S. Department of the Navy,
1989). Thus, the occurrence of all known populations on northeast-facing

TABLE 1. Population characteristics of Polystichum aleuticum and associated geographical variables.
Latitude/
Island Location Pop. # Longitude Elevation (m) Aspect # Individuals Year Relocated
Atka unknown, “within v unknown unknown unknown unknown 1932! N
ie the village e Atha” very rare

Adak Reed, NE A uh 51° 49.640' N 475.5—-525.8 NE 98 1975” xd
176° 41.861’ W

Adak Mt. Reed, NE Arm 2 5 9.491’ N 457.2-469.4 NE 14 1988° ed
176° 41.776’ W

Adak Mt. Reed, NW Arm 3 51° 49.960’ N 360.4 NE 5 1993° Y.
176° 44.141’ W

Adak Mt. Reed, NE Arm 4 51° 49.378' N 338.0 NE 14+ 19994 Y

176° 41.733’ W

1 Christensen (1938).

* Smith (1985)

* Talbot et al. (1995).

* Present study.

062

(Z00z) * YAAWON 26 AWN'TIOA “TVNYNOL NYA NVOMANVY

TALBOT & TALBOT: POLYSTICHUM ALEUTICUM IN ALASKA 291

slopes suggests habitats supporting populations of P. aleuticum are those of-
fering protection from high winds during critical growth and reproductive
periods.

At least 14 clumps constituted the new population; however, due to the
inaccessibility of some of the vertical rock faces, a complete count of
individual clumps was not possible. This number therefore underestimates
the size of the population. Among the 14 clumps counted, five are associated
with rock grottos; the remaining clumps are found on ledges on steep rock
faces. Unlike the other locales, no clumps were associated with herb meadows.
Clumps in the grottos and ledges comprise from five to 30 fronds, with all
clumps containing fronds with sori. Clumps associated with steep rock faces
comprise 12—40 fronds, again with all clumps examined containing fronds
with sori. However, for safety reasons, not all clumps on vertical rock faces
were examined to determine the number of fronds or presence of sori. The
presence of sori on some P. aleuticum fronds suggests this population may be
useful if spores were used for controlled propagation.

We recorded vascular plants associated with the new population; nomen-
clature follows USDA, NRCS (2001). Vascular plants associated with grotto
and ledge clumps were the creeping dwarf shrub Salix rotundifolia Trautv.; the
forbs Achillea millefolium L. var. borealis (Bong.) Farw., Anemone narcissi-
flora L., Arnica unalaschcensis Less., Campanula lasiocarpa Cham., Conio-
selinum gmelinii (Cham. & Schlecht.) Steud., Huperzia chinensis (Christ)
Czern., Lycopodium alpinum L., Pedicularis verticillata L., Platanthera sp. L.
C. Rich., Polygonum viviparum L., Polystichum lonchitis (L.) Roth, Potentilla
villosa Pallas ex Pursh, and Valeriana acutiloba Rydb.; and the graminoids
Carex macrochaeta C. A. Mey., Carex circinata C. A. Mey., Poa sp. (L.) Roth
(viviparous) and Tofieldia coccinea Richards. Vascular plants associated with
vertical rock faces included the forbs Achillea millefolum var. borealis,
Epilobium hornemannii Reichenb., and Viola langsdorfii Fisch. ex Gingins;
and the graminoids Carex circinata, Poa sp. (viviparous) and Tofieldia
coccinea. Comparison of this list of associates from the other sites (Lipkin,
1985; Talbot et al., 1995) indicates a high degree of similarity in species
composition among the four known populations.

Using a hand-held Global Positioning System, we recorded precise geo-
graphic coordinates for this population as 51° 49.578’ N, 176° 41.733’ W. The
discovery of this new population increases the number of known individuals
to 131. We note here that this population was discovered accidentally while
we were disoriented in the fog on Adak, and the area would, under normal
circumstances, have been considered too treacherous to survey. It is very
possible that additional populations of P. aleuticum inhabit Mt. Reed or
nearby mountains, in areas too dangerous to survey without risk of injury, and
we suggest the number of individuals representing P. aleuticum on Adak is
likely underestimated, despite extensive surveys undertaken throughout the
island from the mid-1980s to the mid-1990s.

This discovery is significant because it increases the number of known
individuals by approximately 12%, and the number of known populations

292 AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 4 (2002)

from three to four, thus providing an increased buffer against loss of either
individuals or populations of this rare and endangered species. It also expands
the known elevational range within which future searches for new populations
should be targeted.

Despite this new finding on Adak, however, P. aleuticum continues to be
one of the rarest ferns in North America. While conducting botanical studies
on other Aleutian islands between 1985 to 2001, the second author searched
rocky outcrops similar to those on Adak that support the four known
populations, without finding any new populations. These islands include:
Adugak, Aiktak, Amlia, Buldir, Chagulak, Davidof, Kasatochi, Khvostof, Kiska,
Nizki, and Uliaga islands. Additional surveys by both authors of Attu Island in
2000 and Simeonof Island of the Shumagin Island group, eastern Aleutians, in
1997 also failed to yield discoveries of new populations. Thus, biologists have
searched for P. aleuticum on 16 islands in the Aleutian chain during the past
fifteen years, and have found the fern only on the northern slopes of Mt. Reed,
Adak Island.

ACKNOWLEDGMENTS

Financial support was provided by the Western Area Ecological Services (WAES) and
Department of Refuges, Region 7, U.S. Fish and Wildlife Service, and the Alaska Science Center,
U.S. Geological Survey, Biological Resources Division. We thank Art Davenport of WAES and the
staff of the Aleutian Islands Unit, Alaska Maritime NWR, Adak Island, particularly Jenna Mueller
and Jeff Williams, for field support. Terri Morganson (Division of Realty, U.S. Fish and Wildlife
Service) prepared Figure 1. We thank Judy R. Gust, Dirk V. Derksen, R. James Hickey, and an
anonymous reviewer for valuable comments on the manuscript.

LITERATURE CITED

AnpeERSON, B. L. 1992. Aleutian shield fern (Polystichum aleuticum C. Chr. in Hultén) recovery

lan. Unpublished Report. U.S. Fish and Wildlife Service, Anchorage, Alaska.

CHRISTENSEN, C. 1938. Polystichum aleuticum C. Chr., a new North American species. Amer. Fern J.
28(3):111-112.

LipKIN, R. 1985. Status Report on Polystichum aleuticum C. Chr. Unpublished report, U.S. Fish and
Wildlife Service, Office of Endangered Species, Anchorage, Alaska.

SmitH, D. K. 1985. Polystichum aleuticum from Adak Island, a second locality for the species.
Amer. Fern J. 75(2):72.

Davison, P. G. 1988. Polystichum aleuticum C. Chr. in Hultén. Site survey of Atka
Island, Alaska, 1988. Unpublished field report, U.S. Fish and Wildlife Service, Fish and
Wildlife Enhancement, Anchorage, Alaska.

Ta.sor, S. S., Tabor, S. L., and ScHoFIELD, W. B. 1995. Contribution toward an understanding of
Polystichum aleuticum C. Chr. on Adak Island, Alaska. Amer. Fern J. 85(3):83-88

» YURTSEV, B. A., Murray, D. F., Arcus, G. A., Bay, C., and E.tvepaxk, A. 1999. Atlas of rare
endemic vascular plants of the Arctic. Conservation of Arctic Flora and Fauna (CAFF)
Technical Report No. 3. U.S. Fish and Wildlife Service, Anchorage, Alaska.

TaNnbDe, G. F. 1989. Aleutian shield-fern (Polystichum aleuticum C. Chr.). Field studies for 1989:
establishment of permanent population monitoring plots and habitat characterization.
Unpublished report, U.S. Fish and Wildlife Service, Office of Ecological Services, Anchorage,
Alaska.

TALBOT & TALBOT: POLYSTICHUM ALEUTICUM IN ALASKA 293

USDA, NRCS. 2001. The PLANTS Database, lean 3:1 — //plants.usda.gov). National Plant
Data Center, Baton Rouge, nena 70874-4490, U

U.S. DEPARTMENT OF INTERIOR. 1988. termination of
Final Rule. Federal Register Borat an 626-4630.

U.S. DeparTMENT Or THE Navy, Nava OcEANOGRAPHY COMMAND DETACHMENT. 1989 . Summary of
meteorological prone aay nee 1949-1987 Adak Naval Air Station. U. S. Navy, Adak,
Alaska.

gered status for Polystichum aleuticum.

American Fern Journal 92(4):294—295 (2002)

SHORTER NOTES

Trichomanes ribae (Hymenophyllaceae), a New Filmy Fern from Costa Rica
and Panama.—The Hymenophyllaceae in Costa Rica and Panama are well
known due to the works of Lellinger (Pteridologia 2: 185-228. 1989) and
Pacheco (in G. Davidse, M. Sousa S., and S. Knapp, eds. Flora Mesoamericana.
vol. 1. Psilotaceae a Salviniaceae. Univ. Nacional Aut6noma de México,
México, D. F. Pp. 62-83. 1995). However, as a result of additional studies
during a recent trip to Costa Rica, a new species has been identified.

Trichomanes (Trichomanes) ribae Pacheco, sp. nov.—TYPE. Panama:
Panama, 5-10 km NE of Altos de Pacora, on trail at end road, 750 m, 7
Mar 1975, S. Mori & J. Kallunki 4964, (holotype, MO). (Fig. 1)

Rhizoma repens, 0.1 cm diametro, trichomatibus catenatis; folia remota, 4.8—
11.5 X 2.7—3.8 cm; petioli 0.08—0.5 x 0.05 cm, trichomatibus catenatis; laminae
4.7-11 cm lanceolatae, 2-pinnatifidae, apice pinnatifidae, basi subtruncatae;
rachis alata; pinnae sche can once imbricatae; sori 1-4 per pinnam, in-
volucris in lobis immersi, 0.2—0.2 .2cm, exserto.

Rhizome long creeping, 0.1 cm in diameter with catenate trichomes; leaves
distant, 4.8-11.5 < 2.7—3.8 cm; petioles 0.08—0.5 x 0.05 cm, nonalate, loosely
and deciduously clothed with brown catenate trichomes, similar trichomes on
rachis and veins; blade lanceolate, 4.7—-8.3 cm, 2-pinnatifid, chartaceous, apex
pinnatifid, base subtruncate; rhachis alate, wings 0.07—0.08 cm wide on either
side; pinnae oblong-lanceolate, 11-18 on a side below the apex, 1.2—1.6 X 0.7,
overlapping at in right angles to rhachis, their apices pinnatifid; segments
oblong, 0.15-0.18 x 0.1—0.18 cm, apex obtuse to bifid, plane, margins
complete; venation open, anadromous, pinnate, veins 2-furcate, not reaching
the apices of the lobes; lamina cells almost isodiametric, translucent; sori
lateral on the pinnae, 1—4 per pinna, involucres immersed, 0.2-0.25 < 0.2 cm
campanulate, apex wide-flaring; receptacle exserted.

PARATYPE.—COSTA RICA: Limon; Siquirres, Las Brisas de margen izquierda
de Quebrada Jesus, afluente innominado, Camino a Cerro Tigre. 09° 56’ 40” N;
83° 25’ 15” W, 800 m, 22 Mar. 1996, G. Herrera 8849 & G. Valverde (CR).

Trichomanes ribae belongs to subgenus Trichomanes as evidenced by the
anadromous venation, sori lateral on the pinnae, distant leaves, and 2-
pinnatifid lamina. Its nearest relative is T. rupestre (Raddi) Bosch from which
it differs by shorter leaves, 1-4 sori per pinna, and campanulate immersed
involucres with wide-flaring apex. This species is always epiphytic while
T. rupestre is epipetric or terrestrial, but not epiphytic. This species is dedica-
ted to Ramon Riba y Nava Esparza.

This work was supported by the Instituto Nacional de Biodiversidad, Costa
Rica, Nelson Zamora provided financial and logistic help. I thank Rolando

RANKER ET AL.: GAMETHOPHYTE SEXUALITY

& AG, 4 5d
Spr N GE

SiH A si ? Sieh
NY ie, Ms es eT si “s SOUR 2
= (es SF So DSSS
Aig @, Ve 5 oi 5) Shee
TEAR ig Ms j ; Wee
USE SENS
=. ‘a r,. F Tt
PEGA AAG Mec OE BCI
OE NES Sis BS yk
ee
he J) de cS &
iD oh
z WE Sout
AY SW yas a
$Y B27] le 1g Ys fs
\ Y N le \ 5 } ‘
¥ Ny 5, i SYP a}
ie 2 XS ys i & ») Sh we
Wests Se BP BSS SEL.
gic | ~ & Was
= PO a oe - fF K 4 .
es "9 i 3 J i :
RZ ME ;
PE Ponce M3 is ND ut
a H od Qe oh ee
Pma.y®2 SIRF PRN
"hee eee
oy ge

Fic. 1

Trichomanes ribae. a) habit; b) general view of leaf; c) pinna with campanulate immersed
involucres with wide-flaring apices. Trichomanes rupestre. d) pinna with involucres.

Jiménez for the drawings that illustrate this new species. I am also grateful to

Fernando Chiang for the Latin translation of the species diagnosis.—Leticia
Pacueco. Departamento de Biologia-Botanica Estructural y Sistematica Vegetal,
Universidad Autonoma Metropolitana-Iztapalapa, Apdo. Postal 55-535, 09340
México D. F., México.

American Fern Journal 92(4):296 (2002)

Referees for 2002
All papers submitted to the journal are peer reviewed. Members of the editorial board and the
continued success. The American Fern Society and I extend our thanks to the following reviewers

for their assistance, diligence, and patience in the year 2002. Special thanks are given to Dr. David
B. Lellinger for assuming the task of editing the Wagner Memorial Issue—R. James Piakiey.

DaviD BARRINGTON Davin M. JOHNSON THOMAS RANKER
ALD R. F PAULO LaBIAK CHRISTINA ROLLERI

Davip FRANCKO Joun T. MICKEL Jupy Ski

ERALD GASTONY NICHOLAS MONEY Y SMALL
GARY ER JAMES D. MONTGOMERY ALAN R. SMITH
CHRISTOPHER HAUFLER Rossin C. Moran W. Cari TAYLOR
Kerry, HEAFNER AMES PECK, FLORENCE S. WAGNER
TerRRY A. HILDEBRAND KATHLEEN PRYOR KENNETH A. WILSON
ALFREDO HUERTA R. RABELER GEORGE YATSKIEVYCH

Index to Volume 92

3,8-C-arabinosylluteolin, a New Flavonoid
om Pteris hee
A Binomial for the a Polypodium of
0

astern iertt America, 24
A new Hybrid Polypodiu rovides Insights
ncerning the Systematics of Polypodi-

m scouleri and its Sympatric Congeners,

A New Population of Aleutian Shield F
(Polystichum aleuticum Chri sae on
Asak ot Alaska, 8

Acystopteris 138, 139; japonica, 136, 143
ADAMKEWICZ, L. (see C FF)
Adenophorus pinnatifidus, 97; pinnatifidus

var. pinnaifidus, 97; pinnatifidus var.

rockii, 97; sarmentosus var. rockii, 97
Adiantumi, 23; argut 23-25, 27; diogoano,
‘Edwinii’, 180;

7; tenerum, 180; viviesii, 23
Adams 237, 257
ave, 17

yoncanl aaurantiaca, 239

Alnus shale ‘ees viridis subsp. sinuata, 156

Alsophila, 2

ALVERSON, E aba e P. F, Zika)

ie ioe 142, 144-146, oligocar-
, 14

Apel fe
pense, as rapa 138

: sebicennié 128;

128; semihirsuta, 121, 123, 126; subg
Ane 21. : ; subg. Anemior-
rhiza, 119- , 124, 125, 127-129; subg

nderwoodiana, 121, 123
it 21. eon 124, 126: are 123, 126

Anemone drummondii var. lithophila, 156;
narcissifl

139, 143, 145, 224; bipinnatum, 98;
pei 136, 143; enatum, 98; enatum

mmiparum, 99; fragile, 99; fragile

var. insulare, 99;

Hate 99; septentrionale, ig tricho-

S, 42, 45, 251; triphyllum, 9
Penny major,
Athyrium 8 01, 139; angustum, 185-213
naked 185-213; filix-femina, 113,

rum, i cen ssp. sitchense, 186;
Chinn ae
Azolla filiculoides, ee 180

BARRINGTON, D. S. (see G. J. Gastony)
erberis pinnata, 21

Betula alleghaniensis, 87; neoalaskana, 169

Blechnaceae, 144

Blechnum, 144, 179; appendiculatum, 179, 180;
gl meprer 179; occidentale, 179; spi-
cant, 116, 117

ain tricophyllum 246

Botrychium, 10, 18, 41, 43, 81-83, 85-91, 150,
153-155, 157, 164, 169; alaskense, 164—

164; campestre, 83-90; crenulatum, 151,

249-263, 265, 267; dissectum, 11, 14;

250; mormo, 83, 90; Hh 08

um, 25 ? + pemadasara, 250 pet

bg. Scepteri-

» 152, 150-158; virgin-
ianum, 11, Th. 83-88, 90, 164;
Xwatertonense, 251; ya — 157,

83, 85-87, 150, 151 153- 15

Botryychium hesperium in the . Moun-
tains of oe egon, 239

Broussaisia,

BRUNSFELD, S. : ck. Swartz)

Calamagrostis nutkaensis, 216; rubescens, 239
Campanula lasiocarpa,
isin sce angus Nip 67, 68-70, 73-
phyllitidis, 67-70, 72-76
Caneaea , 124
iomanes, 124, 125, 127; reniforme, 120,
121, 124, 126
Carex ae EO aT 46; concinnoies, 239;
i, 239; macrochaeta, 291

80
Cheilanthes, 224; bradei, 110; castanea, 242:

is Sap 291; eatonii, 242; venusta, 110;
8

Ghokeaenen 4

Chingia, 139, ng oe 134, 143

Cuiou, W-L., D. FARRAR, and T. A. RANKER.
The eer Systems of Some Epiphytic
olypodiceae.

Choral pea ee var. divaricatum,

Sesame 100, 139, 141, 144, 145; acuminata,
138; arida, 134, 143; augescens, 134, 138,
141, 143; boydiae, 100; cyatheoides, 100;

AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 4 (2002)

da 100, hispidula, 134, 143;
Xincesta, 99; x apr 99, 97; para-
sie 100; wailele, 100,

Cibotium, 101

Peco ch sinensis, 245; junos, 245

Comparative Research at Gametophytes of
Olfersia alata - Olfersia cervina (Dry-
Bhs gape e), 2

NT, D. S. (see G. } er

Coryphoptrs, an, 144-146; seemannii,
13
B. (see A. R. Smiru)
Cupressus macrocarpa, 216
A. W

CUusIcK, Psedns for the Hybrid
De rth America, 240

Cyanea, 101
Cyathea arborea, 94; cooperi, 180; muricata, 94
14

9, 141, 142, 145; Xincestus,
es Kiaclovsardlings 99, 100; interruptus,

ule,
Cyrtomium, 229, 234, 236, 237; falcatum, 180
Costoptrs 138, 139, 143, 145, 224; protrusa,
, 143

Deparia lancea, 136; 143; petersenii, 180
Dictyocline, 138, 139, 141, 142; griffithii, 134,

Didymochlaena, 138, 139; truncatula, 136, 143,

: , 236,
Differentiation of Eastern North American
pers filis-femina xed Evidence
neg and Spores, 1
Biersidonns
rao ae digitatum, 242; habereri,
2; tristachyum,
Diplazium, 101, 139: cristatum, 95; esculentum,
nal

Doodia
Dryopteridaceae, 142

sora, 100,
alboviridis, 100, 97, ; glabra var
nii, 101, 97, 102; glabra var. hobdya-
na, 101, 102; rg 101; 102;
x leedsii, 161; podosora,
Dute, R. R. (see A. C. Morr ek

INDEX TO VOLUME 92

Eicer, J. E. Lycopodium lagopus New in West
irginia, 241
Elaphoglossum, 124
lymus glaucus, 239; trachycaulus subsp. vio-
ceus, 15
Epilobium hornemannii, 289
Equisetum, 43
Erigeron caespitosus, 156; compostus, 239;
Jaucus, 21

Eucalyptus globulus, 216

Farrar, D. R. Obituary: Warren H. Wagner, Jr.

RRAR, D. see W-L Cuiou)
Festuca déeideablas 239; rubra, 154, 159
agro chiloensis, 154, 159; virginiana, 169,

Galium boreale, ro

Gastony, G. J. S. BARRINGTON and D
CONANT. ital Rolla Milton Tryon, Jr.
1916-2001),

=
lis =]
Soa
5
25
EA

4
Goniopteris, 138, 139, 142, 144; poiteana, 134,
3

Grammitis, 46
Grant, J. R. (see W. H. WAGNER, JR.)
ait — 139, 143; ovamense, 137

Hammonb, P. (see T. J. HILDEBRAND)
Haplodictyum, 132
HAUuFLeR, C. “| (see T. J. HILDEBRAND)
Heliconia,
eracleum arn
Hickey, R. = Review: Ms Flora of Illinois.
Ferns. 2nd ed.,
ces abiloram, a
ILDEBRAND, T. J., C. H. HAUuFLER, J. P. ne RRIEN,
Watters, and P. HamMonp. A New
Hybrid Polypodium Prariiee Insights
Concerning the Systematics of Polypodi-
um scouleri and its Sympatric Congeners,
214

malosorus, 1 137; pyenocarpos, 137
2

Huper :
Hydrilla a 243

Hydrocotyle verticillata, 243
e

, 128

coides, 20; integri-
a ta seo 20, 21;

oe. a ge 139, 146 crenatum,

anal hawaiiensis, 102; hawaiiensis var.
mauiensis, 102, 97

Illustrated Flora of Illinois. Ferns. 2nd ed.
Re 24

evie
IMPERATO, Fr. 3,8-C-arabinosylluteolin, a New
Flavonoid from Pteris vittata, 244

Intsia, 10

Isoétes, 163; bolanderi X echinospora, 163;
olanderi, 161-163; echinospora, 162-
63; Xherb-wagneri, 162-163, 161

Johnson, D. M. (see D. A. Knepper)

Johnson-Groh, C. (see M. C. Stensvold)

Johnson-Groh, C., C. Riedel, L. Schoessler, and
K. Skogen. Belowground Distribution and
Abundance of Botrychium Gametophytes
and Juvenile Sporophytes, 80

Kalmia lIatifolia, 242

KELLoFF, C. L., J. Skoc, L. ADAMKEWiIcz and C. R.
Wert. Differentiation of Eastern North
American Athyrium filis-femina Taxa:
sage from Allozymes and Spores, 185

Knepper, D. OHNSON and L,
aeetoee Mestion mutica in Virginia,

Larix laricina, 241

Lastreopsis, 131

ectotypification: Adiantum argutum, 23

Kea D. B. Additions to the a Flora of
a, Netherlands Antilles, 9

ing D. B. tia a of =a Herbert

ie D. 'B, ak Prado

Le oe 131, 132, 138, 139, 141; pilosa,
tottoides, 134, 143

nia ce 125; Pa ls 31

Lorinseria, 139, 142, 144, 145; areolata, 137

upinus nootkatensis, 154, 159

Lycopodiella, 241

Lycop diiu [pi 289; clavatum, 241, 242:
clavatum var. monostachyon, 241; den-
droideum, 242; =— 242; lagopus,
241-242; obscurum, 2

300

Lycpodium lagopus New in West Virginia, 241
Lygodium, 119, 120, 124, 125, 127; flexuosum,
121, 126; japonicum, 179, 180, 182; micro-
phyllum, 121, 126

Macrothelypteris, pe 139, 141, 144-146; tor-
resiana, 134, 143, 145; 180

Marsilea Pete a 0-182; —— 181, 182;
mutica 243; quadrifolia,
rsilea pasate in Virginia, Sr

Matoniaceae, 119

Matteuccia, 139, 144 struthiopteris, 137

01

_
—

PEREZ- at wl R. Ripa.
arative Research of Gametophytes

Olfersia sae and Olfersia cervina (Dry-
opteridaceae), 229

Meniscium, 131, 134, 139, 142-144

Meniso 132

Me seatesa Pe 139, 141, 144 crassifolium, 135,

Mesopteris, 132
Meta thes 138-140, 143, 145, 146; dayi,

MICKEL, i x a oc)

Microgramma heterophylla, 67-70, 73-76

Microlepia mauiensis, 102; strigosa, 102; stri-
osa var. mauiensis, 102, 97

Microsorum spectrum var. pentadactylum,

103, 97
Siloingin ge pi 159
Mohria, 119, 120, 123, 125, 127 128-129;
ene

cafforum, 121, 28, i.
Montcomery, J. Review: ophytes of
Upper Katanga eee ratte ic of

Congo), 2
Morrow, A. a ee Crystals
Associated with the Intertracheid Pit
Membrane of Woody Fern Botrychium
multifidum, 1
MussELMAN, L. J. (se a A. KNEPPER)
Myriophyllum pinnatum, 243

oo 139, 141, 146; aoristisora,
143
Nephroeps 94; falcata, lai hirsutula, 180;
multi

Pe on abineitte 7a

Obituary: Rolla Milton Tryon, Jr. (1916-2001), 1
Olfersia alata, 229-238; cervina,

Onoclea, 139, 142, 144; sensibilis, 116, 137

nocleopsis 139; hintonii,
Oreopteris, 139, 140, 145, 146; limbosperma,
135, 143

AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 4 (2002)

wings 125, 224; cinnamomea, 30, 31, 36;
claytonia

Osmundaceae, a 125

Osmundopteris, 83

Oxytropis campestris, 154, 159

PacHEco, L. Trichomanes ribae (Hymenophylla-
ceae), a New Filmy Fern from Costa Rica

axonomic Notes on Hawaiian
Pteridophytes, 97
sider 138-146; beddomei, 138; neva-

» ABO,
hibakes verticilata, 291
ne

—76, 18
Phymatosorus grossus, 180; ie: 67-
Picea shnicn, 169; rubens, 242; sitchensis, 154,
156

Pinus contorta, 252
a austroamericana, 180; calomela-

Side tiens clavellata, 242; sp., 291

Platycerium bifurcatum, 180

Plenasium, 125

Plesioneuron archboldiae, 135, 139, 143

Pleurozia conchifolia,

Pneumatopteris, 139, 141, 144; ecallosa, 135,
sei

Poa sp., 291
paba titi, 10, 14

Polygonum viviparum, 291
olypodiaceae, 142
Polypodiales, 133
olypodium, 32; rphum, 32, 214, 216;

sheeleaaee ;

net a ; oe e dun var
ecmnine 97, 104; pellucidum
var. ““gamma’’, 103; pellucidum var. pel-
lucidum, 75, 103, 104; rockii, 97; sax-
imontanum, 214, 216; scouleri, 214-228;
sibiricum, 214, 216; spec penta-

actyium, 103; spectrum var. spectrum,

INDEX TO VOLUME 92

sain Riley 240; vulgare, 32-36,

15222,
Pavers oo 143, , (226; 229, 235—2397;
acrostichoides, 225; ae cern x lon-
itis, 225; aleuticum, 288-293; i

ns, 208, 209;
lonchitis, 225, 291; munitum, 201, 208,
Populus balsamifera, 169; canescens, 46;

grandidentata, 46; tremuloides, 169
Potentilla norvegica, 169; villosa, 2
R and D. B. LELLINGER. Adiantum
argutum, an Unrecognized Species of the
A; a Group,
. R. Situ. Novelties in Pter-
ica,
Pronephrium, 139, 141, 142, 144, 146; simplex,
135, 143
Protomarattia, 193

seudocyclosorus 139; gost 135, 143

sarge eg Lge 9, 141, 144-146;
, 136,
Preceny LZ

Pteridaceae, 105
Pter. ium 31, 36, 37; aquilinum, 30, 201, 210,
287; aquilinum ssp. aquilinum,

: 9

ri poe um

aa 271, ni

; pearcei, 110;
= ig a 208; peioanfite, = vitiate,
-24

Racomitrion canescens, 154, 159

RANKER, T. A. (se OU

Ranker, T. M. and H. z ae Is Gameto-
phyte Senile in the Laboratory a Good

edictor of Sexuality in Nature? 112

Reproductive Behavior of Cloned Gameto-
phytes of Pteridium aquilinum (L.) Kuhn.,
270

Rhinanthus crista-galli, 154, 159

Rhododendron maximun, 242

agra squarrosus, 154, 159
A, R. (se ENDOZA

a Oe a a JoHNSON-GrRoH)

Ropertson, F. W. Reproductive Behavior of
oe sat a of Pteridium aqui-

m (L.) K
Rosa ae
wise a a a te 263; stellatus,

sea 113-117, 139, 144, 145; * hereon
3-115, 117, 137; pallida, 113-115,
, 114; unisora, 114
Salix alozesis 169; bebbiana, 169; rotundifo-

29
pee ar he 180-182
SANCHEZ, Cc. New age Fern from the
Dominican Republic
Schizaea, se 1 aes 127; elegans, 121,
a 124, 126
chizaeaceae, 120; 124, 125, 128

SCHOESSLER, e C. JOHNSON-GROH
edum eater tie ssp. Pp ago
ia 182; kraussiana, 180; sellata, veal
mbrosa, 180

eens ples 156; as 239

Silene uralensis 7 pees ensis, 156

SIMAN, E. and E ea Polypodium
vu oe Plants ‘Sporulate Continuously
ina do Seasonal Glasshouse Environ-

SKOG, 1 peers . KELLOFF)

BKOG; Ae oe esis and J. T. MICKEL.
Additional aes for Two Subgenara of
Anemia (Schizaeaceae) from Data for the
Mcnues Intergenic Spacer Region
trnL-F and Morphology, 119

SKOGEN, K. (see C. JoHNSON-GROH)

H, A. R. (see J. PRapo)
ae and R B. CranFILu. Intrafamilial
Relatigastaps ef the o_o roid Ferns
(Thelypteridaceae), 1

Spharotephanes a5, a 141, 144;
penn 136, 137, 143; iehutincinniin,
13

Sphagn

242
gi > apis a en 140, 141, 144; pilosa,
ions 132, 139-141, 144; leprieurii, 136,

143
Stenochlaena 139, 144; milnei, 137

302

a = ie D. R. Farrar and C. JoHNson-
o New Species of Moonworts
eee subg. Botrychium) from Alas-

a, 150
Swartz, L. M. and S. J. BRUNSFELD. The Mor-
eal: ne Genetic Distinctness of

tum as Assessed by Morphometric Anal-
ysis and RAPD Markers, 249

TaALBoT, S. L. and S. S. Ta.sor. A New
Population of Aleutian Shield Fern (Poly-
stichum aleuticum C. Christens.) on Asak

6

9
‘soétes eee gneri, an In-
terspecific Hybr id of I. bolanderi x I.
echinospora (Isoétaceae), 161
Tectaria incisa, 180
The Morphological and Genetic pesuinaes of
Botrychium minganense an nula-
tum as Assessed by sai pr
RAPD Markers, 2
Thelypteridaceae, 131, 132, 138, 140, 142-145
Thelypteris 131, 138-141, 144-146; burksio-

Tofieldia coccinea, 291
Torreya yunnanesis, 10

AMERICAN FERN JOURNAL: VOLUME 92 NUMBER 4 (2002)

pads 48, 193; krausii, 95; membrana-
ceum, 95; punctatum subsp punctatum,
ten e€ 294-295; rupestre, 292

aie ge euged be enophyllaceae), A

Film n from Costa Rica and

Panama, 294

Trichoneuron, 131, 6

Trigonospora 139, a cline 136, 143

Tsuga canaden

Tsuga haere ale 265

Utriculuaria, 2
Vaccinium 242; uliginosum, 169
satan acutiloba, 291

iola adunca, 239; langsdorfii, 291
he 48; appalachiana, 242
Vittaria-type, 229, 230, 234, 235, 236, 237

Wacner, F. S. (see P. F. Zika)

-»» & J. R. Grant. Botrychium
alaskense. a New eceiait from the
Interior of Alaska, 16

Wa ter, C. (see T. J. Hons

Watkins, J. E., Jk. and D. R. Rarrar. A New
seed - an old Fern from North Alaba-

dene Pa i (see C. L. KELLOFF

WiLson, K. A. Continued Pteridophyte Invasion
oO waii, 179

ica, 124

Woodsia 139, 145, 224; polystichoides, 137

a P. F., E. R. ALVerson, W. H. WAGNER, jr.

FP. S. Wacner. Botrychium hesperium

“i Wallowa “pa of Oregon, 239
tila E. A. (see J. E. Skoc)

g -

xara - : -
o oa 2 i

Tate
a ae —_ iN, ey

- * yj

oS
.

i OeOme) a Sis 7
= “ou a S 3 | ; _

on. ae lay 7 ales eae Ht in
i ae. ; oe 7 : -

Ee c= Se : ee

a
7
oe)

- i, OD ihe ;
| - - p _ othe \J97 7 ;
: -
- ai) = |
| SS ! : oo _ ; : 5
=a) i i o ad 7 5 leg a. a oo 7
7 Br a "2 °
a x - eal es 7

- oF am cs Bia, Gy,
oe oe eh © om

A ” raat on sare) ey ——- Penagle, ay 4... 4
: 2

= Ripe icc ae YJ ; a a ach aren. Oia
‘ a he oe Oe
i +) : 7 4 : : -
K i ' i ff 7
ri - ia = ; ce, 7 ay 7 bis
ra 7 7 z
= © he Ul! PAR in Bia 0 we syed 2 a,

ak a a aera ss Te

att 0° ees ee Saclaiinseeiliael nm. % a is 1

eu i Y a : : i
Se Yee ee ee eS ia. |
Saree Vcr ae eA S30 9S Ret ow Fee ? : : -
ie, ae i pe f se " - me -
- ¥ 5 7 _ re : % :

4 a S.

: ‘ ra = ie <r To
+ ken it .
4 a ai fey im 7 i te ee ee a i
hs ‘ = ow ,
—_ hae ob a —s bitin al iiss ~~ 7 eel ne.
a i aS Peete a ok ai 7
- Paahe a ehemr 7 ee a ao re ies & _ wat he
= 1 gerade pen Paa. Vehtien 2 fet tent iv |
af | ou Bare a ane 7 ee —
ot - i te rae % ene nde a.
iar ie “eg: .
_ ; A Sa
7 wi) v b a > Fs
one 7 "
q _—— . & S
a on 7 7 = 7
Yen mis 2 a as oo |
: " 5 +
2B a
: a5) |
_ : ! in x
i!
ae > ae a < nt :
ie = ee a h WA ee
; = a te - i 7 ON ee 7 oe i bi
; tg Non 1 aes con ae a a Tis Vin Sas 7 =
7 >. e a - ve : : ws 7 2 “i 7 a 7 i 7
ae oh tc : 7 ot bask : ; 7 Fan ee 7 - ne a at) ‘2 — a
7 7 > poo YS m , 7
ae ‘1 oe bala Mae Pas _ : ‘piey i oe ne mee - aes. =
: ho Seen < a ee ave a ee Ve ec 7 wot aren | ‘
f a oe : Sy eee Ol.
' 7 bag a > G
= x
- /
=i. - ae yee
: ere See ae. en eae
pea fae a 7 _ ee See! i
> as Sa, hate ie ie ; ae —_ _

=

ve ane
Mee

on rl ete

ee To ere : Pr sas me’. 2." : io: ag | be ee
tn a ae : ae

. .
_ 7 | ee 7 :
> 7 - = 7
aes a la ae jen ee ar be stiine

7 ra

i Hi If) ti

INFORMATION FOR Au : Ml

Authors are encouraged to submit manuscripts pertinent to pteridology for pub-
lication in the American Fern Journal. Manuscripts should be sent to the Editor.
Acceptance of papers for publication depends on merit as judged by two or more
referees. Authors are encouraged to contribute toward publishing costs; howev:
the payment or non-payment of page charges will affect neither the cane
of manuscripts nor the date of publication.

Authors should adhere to the following guidelines; manuscripts not so prepared
may be returned for revision prior to review. Submit manuscripts in triplicate
(xerocopies acceptable), including review copies of illustrations and originals of
illustrations. After review, submission of final versions of manuscripts on diskette
(in PC- or Mac-compatible formats) is strongly encouraged. Use standard 8% by
11 inch paper of good quality, not “erasable” paper. Double space manuscripts
throughout, including title, authors’ names and addresses, short, informative ab-
stract, text (including heads and keys), literature cited, tables (separate from text),
and figure captions (grouped as consecutive paragraphs separate from figures).
Arrange parts of manuscript in order just given. Include author’s name and page
number in upper right corner of every sheet. Provide margins of at least 25 mm
all around on typed pages. Do not submit right-justified copy, avoid footnotes,
and do not break words at ends of lines. Make table headings and figure captions
self-explanatory. Use S.I. (metric) units for all measures (e.g., distance, elevation,
weight) unless quoted or cited from another source (e.g., specimen citations). For
nomenclatural matter (i.e., synonymy and typification), use one paragraph per
basionym (see Regnum Veg. 58:39—40. 1968). Abbreviate titles of serial publi-
cations according to Botanico-Periodicum-Huntianum (Lawrence et al., 1968,
Hunt Botanical Library, Pittsburgh) and its supplement (1991). References cited
only as part of nomenclatural matter are not included in literature cited. For shorter
notes and reviews, omit the abstract and put all references parenthetically in text.
Use Index Herbariorum (Regnum Veg. 120:1—693. 1990) for designations of her-

a.

Illustrations should be proportioned to fit page width with caption on the same

. Provide margins of at least 25 mm on all illustrations. For continuous-tone
illustrations, design originals for reproduction without reduction or by uniform
amount. In composite blocks, abut edges of adjacent photographs. Avoid com-
bining continuous-tone and line-copy in single illustrations or blocks. Coordinate
sequence and numbering of figures (and of tables) with order of citation in text.
Explain scales and symbols in figures themselves, not in captions. Include a scale
and reference to latitude and longitude in each map.

Proofs and reprint order forms are sent to authors by the printer. Authors should
send corrected proofs to the editor and reprint orders to the printer. Authors will
be assessed charges for extensive alterations made after type has been set.

For other matter of form or style, consult recent issues of American Fern Jour-
nal and The Chicago Manual of Style, 14th ed. (1993, Univ. Chicago Press,
Chicago). Occasionally, departure from these guidelines may be justified. Authors
are encouraged to consult the editor for assistance with any aspect of manuscript
preparation.

Papers longer than 32 printed pages may be sent to the Editor of Preridologia
(Memoir Editor, see cover 2).

PTERIDOLOGIA ISSUES IN PRINT

The following issues of Pteridologia, the memoir series of the American
Fern Society, are available for purchase:

1. Wagner, David H. 1979. Systematics of Polystichum in Western North
America North of Mexico. 64 pp. $10.00 postpaid.

2A. Lellinger, David B. 1989. The Ferns and Fern-allies of Costa Rica,
Panama, and the Choco (Part 1: Psilotaceae through Dicksoniaceae). 364 pp.
$32.00 postpaid.

3. Lellinger, David B. 2002. A Modern Multilingual Glossary for Taxonomic
Pteridology. 263 pp. $28.00 postpaid.

Send your order with a check or money order to: American Fern Society,
Inc., c/o U.S. National Herbarium MRC-166, Smithsonian Institution, Washing-
ton, DC 20560.

AMERICAN FERN JOURNAL ON MICROFICHE

Volumes 1—61 of the American Fern Journal are available as archival quality,
silver positive microfiches. Single volumes or the entire run may be purchased.
The fiches are easily read with 10 or greater magnification (using a dissecting
microscope and transmitted illumination or a fiche reader). Silver negative micro-
fiches of vols. 1-50 are also available. The price is $4.00 per volume or $244.
per set of 61 volumes, postpaid.

Send your inquiry or order with a check or money order to: American Fern
Society, Inc., c/o Dr. James D. Montgomery, Ecology III, Inc., R.D. 1, Box 1795,
Berwick, PA 18603.

VISIT THE AMERICAN FERN SOCIETY’S
WORLD WIDE WEB HOMEPAGE:
http://www.amerfernsoc.org/