Skip to main content

Full text of "American fern journal."

See other formats 


AMERICAN 
FERN sages 
JOURNAL 


QUARTERLY JOURNAL OF THE AMERICAN FERN SOCIETY 


Editor 
Alan R. Smith 
Department of Botany, University of California, Berkeley, CA 94720 


Associate Editors 
Gerald J. Gastony, Department of Biology, Indiana University, 
Bloomington, IN 47401 


Christopher Haufler, Department of Botany, University of Kansas, 
Lawrence, KS 66045 


David B. Lellinger, U. S. National Herbarium NHB-166, Smithsonian Institution, 
Washington, D.C. 20560 


Terry R. Webster, Biological Sciences Group, University of Connecticut, 
Storrs, CT 06268 


The American Fern Society 
Council for 1987 
FLORENCE S. WAGNER, Dept. of Botany, University of Michigan, Ann Arbor, MI 48109. 


President 
JUDITH E. SKOG, Biology Dept., George Mason University, ey. a 22030. Vice-President 
W. CARL TAYLOR, oo Public Museum, sarees WI5 Secreta 


Ty 
JAMES D. CAPONETTI, Dept. of Botany, University of Tennessee, >t neville. TN 3 ssh a 
DAVID S. BARRINGTON, “a of Botany, University of ak Burlington, VT 0 
Re corde lances 
JAMES D. MONTGOMERY, Ecology III, R.D. 1, Berwick, PA 1 Back Issues Curator 
ALAN R. SMITH, Dept. of Botany, University of California, behets — 94720. heed Editor 
DAVID B. LELLINGER, Smithsonian Institution, Washington, DC 2056 Memoir Editor 
DENNIS Wm. STEVENSON, Dept. of ——e Sciences, Barnard se 
Columbia — New York, NY ‘f Fiddlehead Forum Editor 


The can Fern Journal” (ISSN 0002-8444) is an illustrated quarterly devoted to the general 
study of fens It is owned by the American Fern Society, and published at the Pringle Herbarium, 
pen vt of Vermont, Burlington, VT 05405, and printed by Allen Press, Inc., 1041 New Hampshire 

.. Lawrence, KS 66044. Second-class postage paid at Burlington, VT, and additional entry point. 
rders for back issues should be addressed to Dr. James D. Montgomery, Ecology III, R.D. 1, 
Berwick, PA 18603. Back volumes 1910-1978 $5.00 to $6.25 each; single back numbers of 64 pages or 
less, $1.25; 65-80 pages, $2.00 each; over 80 pages, $2.50 each, plus shipping. Back volumes 1979 et 
seq. $8.00 each; single back numbers $2.00 each, plus shipping. Ten percent discount on orders of 
six volumes or more. 
Subscriptions $12.00 gross, $11.50 net if paid through an agency (agency fee $0.50); sent free to 


members of the American Fern Society (annual dues, $10.00 + $4.00 mailing surcharge beyond 
U.S.A., Canada, and Mas life membership, $200 
Fiddlehead Forum 
The editors (Dennis Wm. and Jan W Wassmer Stevenson) nj tributi f; b d 


non-members, including miscellaneous notes, offers to pipe or purchase materials, personalia, 
horticultural notes, and reviews of non-technical books on ferns. 
Spore Exchange 
Mr. Neill D. Hall, 1230 Northeast 88th Street, Seattle, WA 98115, is Director. Spores exchanged 
and lists of available spores sent on request. 


Gifts and Bequests 


in ferns. Botanical books, bock is issues of the journal, and cash or other — are pe ag welcomed, 
and are tax-deductible. Inquiries should be addressed to the Secr retary 


fore ys 4} eL e 


Table of Contents 


(A list of articles arranged alphabetically by author) 


ALVERSON, EDWARD R. {see Soltis, Pamela. clal) = ee... 
Bere, Joseph: M, [see Mickel) ee 
GASSA DE Pazos, LILIANA A. isee OO CLBLL i een 

oe HUGH Wo isee Gravina) ee ae ee ees 
USICK, ALLISON W., A binomial for a common hybrid Lycopodium ...................... 
— ns, A. M., review of Field guide to the ferns and other iets of Georgia Peete 
FREEMAN, JOHN D., Terrestrial Psilotum in east-central Alabama .................-....... 
GRAYUM, MICHAEL H., and HUGH W. CHURCHILL, An eee to the pteridophyte flora of 
Finca La eo Costa Rica. = 306 ee a es ee 

Hauke, RICHARD L., The expanded adaxial epidermis of Equisetum rhizome sheath teeth 
e ochreole of Equisetum: A prophyllar sheath ..................-...-.4.055->: 
Fiogriceact: BARBARA JOE, review of Encyclopaedia of ferns............-...-.....-..-0... 
Kuarg, P. K., and RAMA SHANKAR, Variations in petiolar structure of Hypodematium crenatum 
eae Davip B., The disposition of Trichopteris (Cyatheaceae) ...................... 
. Nomenclatural notes on some ferns of Costa Rica, Panama, and Colombia.—III .... 

on $. Reteten (ene Rite etal) 
MICKEL, JOHN T., A new fern from western Mexico and its bearing on the taxonomy of the 
ehallanthnid tems ee ee: 
— and JosEPH M. BEITEL, Notes on x Pleopodium and Pleopeltis in tropical America .. 
Montcomery, JAMES D. (see Root) ...-. ... 2-5-2 eee eee 
Moran, Rossin C., review of pao and allied plants of Victoria, Tasmania and South Australia 
(ge Gmaith) 2 a eee 
NAUMAN, CLIFTON E., Additions to the fern flora of the Bahamas......................-.. 
o species of Adiantum new to Florida .............---- +++ ++ esses e rete e eee 
PacHEco, tenes inne Riba tM) eee 
Ponce, MARTA Monica (see Sota et al.) .......-.---- 0-5-0 s eect 
Power, Martua S., and JupITH E. Skoe, Ultrastructure of the extrafloral nectaries of Pteridium 


ilin 

Price, M. ee reviews of A monograph of the fern genus Pyrrosia ee, and The 
Pyrrosia species formerly referred to Drymoglossum and Saxiglossum ........... 

_ review of Illustrations of pteridophytes of Japan, volume : ae 

_ review of Illustrations of pteridophytes of Japan, v Voumes 2. 

Revere ULrike, Growth patterns of gemmlings of Lycopodium ne Wed 

Ripa, RAMON, LETICIA PACHECO, and ESTEBAN MarTiNEz S., New records of pteridophytes from 


Root, Peter G., Botrychium hn CS EO ee 
SCHEELE, CORNELIA (see Wollenweber et al.) ......-.-.------5 222005 e cert ersten ees 


Situ, ALAN R., and Rossin C. Moran, New combinations in Megalastrum (Dryopteridaceae} 
. review of Index of Thelypteridaceae ........----- +++ +12 sheet ttt 
—_, review of A key to the genera of New Zealand ferns and allied plants ............ 
Souris, Douc.as E. (see Soltis, Pamela, et al.) ......--.------ 2-2 eh rt 
SoLtis, PAMELA S., Douctas E. Sottis, and EpwarD R. AL , Electrophoretic and mor- 
onfirmation of interspecific Ss ae inetas Secuen Polystichum krucke- 

beciis Fees 
Sora, ELias DE LA, MARTA Monica Ponce, and LILIANA A. CassA DE Pazos, Chromosome 

inners of conse forma from Angonting ... eee 


STOLZE, ROBERT G., A new species of Danaea from Peru ................................ 
5) Rene Dit Cmenwired Wi Pate 6 a 
; ene menuly of Hymennpnylhin crstaum «6 sk 
TRYON, ALICE F., review of ioe palynologica pteridophytorum Italiae ......._... 
i cradl bhicameosnterccage. 1. 6h ee) es ee ee 
WHITTIER, DEAN P., Germination of s igamTgeme: BOE GR ee 
WINDHAM, MicHaeEL D., Argyrochosma, a new genus of cheilanthoid ferns......... 
WOLLENWEBER, ECKHARD, CORNELIA SCHEELE, aa ALICE F. TRYON, Plavousiis and spores of 
latyzoma microphyllum, an endemic fern of Australia......................_... 


Volume 77, Number 1, pages 1-36, issued 24 March 1987 
Volume 77, Number 2, pages 37-72, issued 22 December 1987 
Volume 77, Number 3, pages 73-108, issued 3 March 1988 
Volume 77, Number 4, pages 109-144, issued 3 May 1988 


American nA oF 
Fern Number 4 
Journal 


QUARTERLY JOURNAL OF THE AMERICAN FERN SOCIETY 


Editor 
Alan R. Smith 
Department of Botany, University of California, Berkeley, CA 94720 


Associate Editors 
Gerald J. Gastony, Department of miraait Indiana University, 
Bloomington, IN 474 


Christopher Haufler, Department of pone University of Kansas, 
Lawrence, KS 6604 
David B. Lellinger, U. S. National Herbarium NHB-166, Smithsonian Institution, 
Washington, DC 20560 


Terry R. Webster, ee Sciences pei University of Connecticut, 
Storrs, CT 0628 


The American Fern Society 


Council for 1988 
JUDITH E. SKOG, Biology Dept., George Mason University, Fairfax, VA 22030. President 
DAVID B. LELLINGER, Smithsonian Institution, Washington, DC 20560. Vice-President 
W. CARL TAYLOR, Milwaukee Public Museum, Milwaukee, W153233. Secretary 


JAMES D. CAPONETTI, Dept. of Botany, University of Tennessee, Knoxville, TN 37916. Treasurer 
DAVID S. BARRINGTON, Dept. of Botany, University of Vermont, Burlington, VT 05405. 
Records Treasurer 
JAMES D. MONTGOMERY, Ecology Ill, R.D. 1, Berwick, PA 18603. Back Issues Curator 
ALAN R. SMITH, Dept. of Botany, University of California, Berkeley, CA 94720. Journal Editor 
DAVID B. LELLINGER, Smithsonian Institution, Washington, DC 20560. Memoir 
JOHN T. MICKEL, New York Botanical Garden, Bronx, NY 10458. Fiddlehead Forum Editor 
The “American Fern Journal” (ISSN 0002—8444) i illustrated quarterly devoted to the g ] 
study of ferns. It is owned by the American Fern Society, and published at the Pringle Herbarium, 
University of Vermont, Burlington, VT 05405-0086, and printed by A-R Editions, Inc., 315 W. 
Gorham St., Madison, WI 53703. Second-class postage paid at Burlington, VT, and additional entry 


point. ; 

Orders for back issues should be addressed to Dr. James D. Montgomery, Ecology Hl, R.D. 1, 
Berwick, PA 18603. Back volumes 1910-1978, $5.00 to $6.25 each; single back numbers of 64 pages 
or less, $1.25; 65—80 pages, $2.00 each; over 80 pages, $2.50 each, plus shipping. Back volumes 1979 
et seq. $8.00 each; single bach bers $2.00 each, plus shipping. Ten percent discount on orders of 
six volumes or more. 

Subscriptions $20.00 gross, $19.50 net if paid through an agency (agency fee $0.50); sent free to 
members of the American Fern Society (annual dues, $15.00 + $4.00 mailing surcharge beyond 
U.S.A., Canada, and Mexico; life membership, $300.00). 

Fiddiehead Forum 
The editor 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 ferns. 


Spore Exchange 
Mr. Neill D. Hall, 1230 Northeast 88th Street, Seattle, WA 98115, is Director. Spores exchanged 
and 12,4 - cp pe SE + re 


Gifts and Bequests 
ifts and bequests to the Society enable it to expand its services to members and to others 
interested in ferns. Botanical books, back issues of the Journal, and cash or other gifts are always 
welcomed, and are tax-deductible. Inquiries should be add d to the Secretary. 


Table of Contents 
(A list of articles arranged es by author) 
BoucHER, PAUL F, Rediscovur yofG ryopteris in Arizoon ... <4: .........5: 71 


CINQUEMANI, DIANE M., MILDRED E. Faust, and DONALD J. LEopotp, Periodic censuses 
(1916—1986) of Phyllitis scolopendrium var. americana in central New York State 


Deak Giese Sle pri oe Wir eb 6 Wea dials ol Pare Web huale eee aie WS Re 37 
DARNAEDI SIE (600 RAIN) se ee ea ee ee ee a 77 
FAUST, Mitonen & (see Cinguemanl) 2. 6. ed er een eae 37 
Gas — = LD J. ve Pellaea glabella arena Electrophoretic evidence for the 

I sie 44 
GRAYUM, MICAHEL H. (see Smith) Ce eas eas Sec aN a la eek 105 
HAMILTON, RoBERT G., The significance of spore banks in natural populations of Athyrium 
Pyenocurpon and A. thelypterioides. 6.61 i ee ee eee 96 
HIcKEY, R. JAMES, Isoétes pallida, a new species from Mexico. ......... 00.0. cece eens 35 
KaTo, Masanrro, and DEpY DARNAEDI, Taxonomic and ities hic relationships of 
Diplazium flavoviride, D. pycnocarpon, and ig pnd aes EE a ane ee 77 
JOHNSON, used M., Marsilea scalaripes, a new member ¢ tion Clemys from th 
Asian DOpIGS 6 ee ea 68 
review of Monograph of the Neotropical fern g Polybotrya (Dryopterid ) 

Oe ee ye eee ee ee ee ee 76 
LELLINGER, nol B., Some new species of Campyloneurum and a provisional key to the 

OTIS se ae ee a ss eS ee ee te bes ere cae ee ees cee 14 
LEOPOLD, enh (see CINGHeIaahy) oe ee ee 37 
LLoyp, Sue M., ma neae studies on the probability of selfing by protandrous 

MI ke ie fee ne 0a we whe er 117 
LUEBKE, Ne ie ~ Taylor) ee ee 6 
Munro, DEREK ti f Aspleni ta-muraria, with Pellaea atropurpurea and 

abella, in eastern Ontario. SE ee A ee ee 136 
PECK, yi sea a ie a es ee ee es es 73 
Peck, JAMEs H., and CAROL LJ. PECK, Seven Clubmosses 1 new to Arkansas SSS SEE Rada s rai 73 
SHARP, ., Teview of ICO hay eae 75 
SMITH, ALAN R., review of Liebmann’ s Mexican ferns: His ao a translation of his 

‘Mexicos Bregner,’ and a reprinting of the original work ..............-...40055 75 

and MicHaEt H. Grayum, Cyathea stolzei x aos a distinctive tree fern hybrid 

irom Coste Ries a ee ec eee ks 105 
TayLor, W. Cart, and Net T. Luesxe. Isoétes x hickeyi: A naturally occurring hybrid 

lietween L. echinospord end 1. macrospord .. ..« . <5 oo ee cb eset ce ee 6 

TumnNee Mervin D. (see White) ... . 6 2. kn cc ccc re eee cee ecb renews erences 86 
Wacwun, WH. fr. (G00 ZO) ooo a ir ae ere es ie eh anne ee eee cokes 122 
WuHite, RICHARD A., and MELVIN D. TURNER, Calochlaena, g f dick ioid f 86 
MITTIER, DEAN P., Dark-grown Psilotum. ...... 0... 6. e ene e een eaee 109 
ZIKA, PETER F., The status of A OY og a ee oes ee tek 72 
Zou, XIAOMING, and W. H. WAGNER, Jr., A preliminary review of Botrychium i Chi ak. 122 


Volume 78, Number 1, pages 1—36, issued 2 September 1988 
Volume 78, Number 2, pages 37—76, issued 30 November 1988 
Volume 78, Number 3, pages 77-108, issued 1 December 1988 
Volume 78, Number 4, pages 109-140, issued 10 February 1989 


AMERICAN 
FERN 


JOURNAL 


QUARTERLY JOURNAL OF THE AMERICAN FERN SOCIETY 
Ultrastructure of the Extrafloral Nectaries of Pteridium aqui. 


Notes on x Pleopodium and Pleopeltis in Tropical Am 


Martha S. Fewer and Judith E. Skog 
Flavonoids and Spores of Spree microphyllu 


1 
John T. ‘Mickel and Joseph M. Beitel 16 
, an En “ern of 
ckhard Welleenhice: Cacnaiie Scheele, and Alice F. Tryon 28 
A New Species of Danaea from Peru Robert G. Stolze 33 
Reviews 
Announcement: 1987 AIBS Meeting—Call for Papers 
Information for Authors 


The American Fern Society 
Council for 1987 


FLORENCE S. WAGNER, Dept. of Botany, University of Michigan, Ann Arbor, MI 48109. 


President 
JUDITH E. SKOG, Biology Dept., George Mason University, Sages = 22030. Vice-President 
W. CARL TAYLOR, Milweukes Public Museum, Milwaukee, WI Secretary 


JAMES D. CAPONETTI, Dept. of Botany, University of Tennessee, pyre ™ ies a 
DAVID S. BARRINGTON, Dept. of Botany, University of Vermont, Burlington, VT 054 
cso iaener 
JAMES D. MONTGOMERY, Ecology III, R.D. 1, Berwick, P Back Issues Curator 
ALAN R. SMITH, Dept. of Botany, University of California, ic. CA 94720. Journal Editor 
DAVID B. LELLINGER, Smithsonian Institution, Washington, DC 20560. Memoir Editor 
DENNIS Wm. STEVENSON, Dept. of Biological Sciences, Barnard College, 
Columbia University, New York, NY 10027 Fiddlehead Forum Editor 


Madcdss Fa0k Jounal 
EDITOR 


PURER. SOE Dept. of Botany, University of California, 

Berkeley, CA 94720 

ASSOCIATE EDITORS 
GERALD J. Ape Dept. of Biology, Indiana University, Bloomington, IN 47401 
CHRISTOPHER Pe Dept.. of Botany, — of Kansas, 
e, KS 66045 

DAVID B. LELLINGER oe . U.S. Nat'l Herbarium NHB-166, Smithonian Institution, 
: ashington, DC 20560 
‘TERRY R. WEBSTER ____ Biological Sciences Group, University of Citar pias CT 06268 


2 The “American Fern Journal” (ISSN 00 an illustrated quarterly devoted to the general 
: porn of ferns. It is owned by the American Fern Gociaty. and cabisad at the Pringle Depvonianes 

boven of ‘Vermont, yen site VT 05405. Second-class postage paid at Burlington, VT, a 
Claims ie missing i issues, th 12 ths (f the date of issue, 

wre et sel James D. “shes Bae TH: RD. i; 


and spplication for membership should be sent to the Records Trea- 


agency {agency fee $0.50); sent free to 
annual dues, $10.00 + $4.00 mailing surcharge beyond 
- $2.00 ea eS spi Oto 6.25 each; single back umbers of 6 less, $1.25: 65-80 
tak. open arco rire Back volumes 1979 et seq. $8.00 each: 
h, plus shipp ~ - — discount on orders of six volumes or 


RICAN Fem oom, Dept of Botany. University of 


BESSOUR! BATANICAL 


APR 1 1987 


American Fern Journal 77(1):1-15 (1987) 
GARDEN LIBRARY 


Ultrastructure of the Extrafloral Nectaries of 
Pteridium aquilinum 
MARTHA S. POWER 
M.S. 927, United States Geological Survey, Reston, VA 22092 
JUDITH E. SkoG 
Department of Biology, George Mason University, Fairfax, VA 22030 


The presence of nectaries on the stipe and frond distinguishes Pteridium aqui- 
linum (L.) Kuhn from most other vascular cryptogams. The largest and most 
obvious nectaries are located on the stipe at the base of the lower pinnae. Nec- 
taries on the stipe decrease in size toward the apex of the frond. The smallest 
nectaries are located on the abaxial surface of the frond, scattered on pinnae 
axes. 

Described first in 1877 by Darwin, the nectaries of Pteridium have undergone 
further microscopic examination (e.g., Lloyd, 1901; Liittge, 1961; Schremmer, 1969; 
Page, 1982). Lloyd (1901) noted that cells in the glandular tissue are smaller and 
contain more protoplasm than adjacent ground parenchyma tissue. These cells 
have vacuolated cytoplasm and thin cell walls. The anatomical differences noted 
by Lloyd have provided the basis for subsequent reviews without further elu- 
cidation. Fahn (1979a), when reviewing differences between structured and 
non-structured nectaries, cited the nectaries of Pteridium as non-structured. 
Structured nectaries can be identified macroscopically and their secretory cells 
differentiated microscopically whereas non-structured nectaries are basically 
unmodified tissue that secretes nectar through stomates. We find that nectaries 
of Pteridium which exude nectar through stomates can be distinguished both 
macroscopically and microscopically. Macroscopically, they are distinct protu- 
berances, differing in color from the rest of the stipe and lacking trichomes. 
Microscopically, these nectaries are composed of layers of nectariferous tissue, 
distinctly specialized compared to ground parenchyma. Thus they can be defined 
as structured nectaries. 


MATERIALS AND METHODS 


Two populations of Pteridium aquilinum yielded the specimens in this study: 
one from Reston, Virginia (Power 7734 in George Mason University Herbarium) 
and one from Mountain Lake Biological Field Station of the University of Vir- 
ginia in Pembroke, Virginia (vouchers in Mountain Lake Herbarium). Secreting 
and non-secreting nectaries from stipes were excised and fixed in FAA (for- 
malin, acetic acid, alcohol), or glutaraldehyde according to procedures outlined 
by Warmbrodt and Evert (1974a). Directions for dehydrating and embedding 
followed those in Mokotoff (1978). Tissue was embedded in Spurr’s medium 
(Spurr, 1969). Thick (5-10 um) sections for light microscopy were mounted on 
glass slides and stained with either safranin-fast green or toluidine blue; thin 


2 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 1 (1987) 


sections (0.6-0.7 um) for transmission electron microscopy (TEM) were placed 
on grids and viewed with a JEOL 100C microscope. 

Secreting and nonsecreting nectaries were prepared for scanning electron 
microscopy (SEM) by fixing 5 mm sections in FAA or glutaraldehyde and de- 
hydrating in an absolute alcohol series. Prior to fixation some nectaries were cut 
in two with a razor blade. They were then dried in a critical point drier. All 
nectaries were mounted on stubs and coated with gold-palladium, prior to view- 
ing with either a JEOL 35C microscope or a Hitachi 5530 microscope. 


RESULTS 


Morphologically, nectaries appear as smooth protuberances, raised 0.1-1.0 mm 
above the surface of the stipe. Located on the abaxial surface of the stipe at the 
base of each pinna or pinna pair, they are approximately 1-4 mm in diameter 
and vary in color from dark green (secretory) to brown (nonsecretory). 

Transverse sections of the stipe show that the nectaries are distinguished from 
other tissues by specialized secretory parenchyma cells between the endodermis 
and the epidermis. The nectariferous tissue appears to be divided into three 
regions: a basal layer adjacent to a meristele, a broader cortical region, and an 
epidermal region. These combined regions measure approximately 0.3 mm from 
meristele to nectary surface. The tissue is 0.4 mm wide and is bordered on either 
side by cortical ground parenchyma cells which are large, irregular in outline 
and highly vacuolated. Figure 1 shows an overview of a nectary shortly after 
cessation of secretion; Figure 2 depicts a secretory nectary. 

The basal region next to the meristele ranges from 3-7 cells wide and abuts 
the outer walls of the endodermis (Figs. 1 and 2). The cells are smaller than the 
surrounding cortical parenchyma cells and appear cuboidal. The granular cy- 
toplasm is filled with organelles and several large vacuoles (Fig. 3). 

Adjacent to the basal region is an area composed of specialized parenchyma 
cells, the cortical secretory parenchyma. These cells are small and isodiametric, 
compared to the larger elongated ground parenchyma cells which compose most 
of the cortex and pith (Figs. 1 and 2), and contain densely staining cytoplasm 
filled with small vesicles but no large vacuoles. Nuclei appear large compared 
to the volume of the cell; each nucleus contains a nucleolus. In contrast, cyto- 
plasm in ground parenchyma cells is more diffuse, often parietal, and hence 
stains very little. Nuclei occupy a smaller volume in these larger cells and are 
thus often excluded from the plane of section. 

The cortical secretory parenchyma cells are connected by plasmodesmata which 
are often found within primary pit fields and simple pits if any secondary cell 
wall has been laid down (Fig. 4). The cells contain much endoplasmic reticulum 
(ER), some with enlarged cisternae, and dictyosomes. Fibrillar inclusions similar 
nig ena bemietion —— Duos cat gap fig. 11) can be seen outside the 
proning el wadongedrel bi cisternae (Fig. 5). Figure 6 shows a plasmo- 
tains a relatively large (0.27 um) tiie A sets re ce cca oral 
A small segment of a nucleus is visible in one cell ve omega rt reas 

. Near the nuclear envelope is 


POWER & SKOG: PTERIDIUM NECTARIES 3 


Fics. 1 and 2. Pteridium nectary anatomy. Fic. 1. Cross-section of stipe at cessation of secretion. 
Stomate is open, but epidermal cell walls have been thickened. Cortical secretory parenchyma (cs), 
basal region of parenchyma (b), the endodermis (e), and cells in the pericycle region (p) can be easily 
distinguished. Fic. 2. Cross-section of a secretory nectary. The cortical secretory cells (c) are distinct. 


+ AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 1 (1987) 


Fics. 
— . es . — of secretory nectary cells. Fic. 3. Transmission electron microscope 
si 1 a 2 _ aapmerger “ the basal region showing vacuoles {v) with some fibrillar 
é os : i : 
: astids and mitochondria (m). Fic. 4. SEM view of a cortical secretory 


chyma cell in split section showin imary pi : 
be seen {arrow}. ss pit field. Several openings for plasmodesma can 


POWER & SKOG: PTERIDIUM NECTARIES 5 


Fic.5. TEM ofa hasvbinai - a nine snereery. cell. The vacuole is at the right. Cytoplesn contain 
much ER with the outer 

and some fibrous rhatorial (f) is ‘outside the membrane. Some cisternae contain porible Sitrous 
material. 


6 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 1 (1987) 


a strand of endoplasmic reticulum, the cisterna of which is enlarged (Fig. 6). The 
other cell contains several mitochondria near the plasmodesma. Vesicles have 
formed adjacent to the endoplasmic reticulum. No plastids are visible in these 
cells. 

Cortical secretory parenchyma is distinguished during its functional period by 
inclusion of many globules stained brilliantly red with safranin under light mi- 
croscopy (Fig. 2). These globules are also visible in vascular parenchyma and 
basal parenchyma although they are smaller. The intercellular spaces in this 
cortical region anastomose into substomatal chambers. 

The epidermal region of the nectary is composed of 1-3 cell layers (Fig. 7). 
Each cell is oblong to rectangular and contains a nucleus, chloroplasts, vesicles, 
ER, and dictyosomes. The outer epidermal cells are covered by a thin cuticle on 
their outer walls (Fig. 8). The epidermal cells of the nectary are isodiametric 
(Fig. 9), whereas the epidermal cells of the stipe are elongated and almost linear. 
Stomates are much more abundant on the nectary compared to the stipe area 
(Fig. 9). Guard cells of the stomates are elevated above the epidermis (Figs. 1, 9, 
and 10). The guard cells are covered by a thin cuticle that projects into cuticular 
ledges (Fig. 1). The cuticle extends into the stomate and lines the substomatal 
chamber (Fig. 7). No subsidiary cells are present. 

During the crozier stage when the nectaries are actively secreting, stomates 
are open and covered with globular secretions (Figs. 9 and 11). With cessation 
of function, the stomates become occluded (Fig. 10) and the cell walls of the 
epidermis become thickened and ridged. 

In addition to these three differentiated regions, the adjoining meristele under- 
goes certain modifications. The part of the meristele next to the basal region 
enlarges with the formation of two or three layers of cells inside the endodermis 
(Figs. 1 and 2). These cells are somewhat larger than surrounding pericycle cells, 
cuboidal, and densely staining with very granular cytoplasm, large nuclei, and 
no vacuoles. The neighboring pericycle cells are somewhat smaller and variable 
in size, vacuolated, and form only one cell layer. These modified pericycle 
parenchyma cells appear quite similar to the basal cells in light microscopy (Fig. 
2). The endodermis also undergoes modification in this region. Large vacuoles 
are evident in both light microscopy (Fig. 2) and TEM. No Casparian strip is 
evident in this region during the crozier stage (secretory phase) of development. 
Only after the frond has expanded fully and nectaries have ceased functioning 
does the Casparian strip begin to form. 

The vascular tissue, composed of sieve cells, xylem tracheids and vessels, and 
parenchyma cells (Figs. 1 and 2), is modified in content but not in structure. Light 
microscopy reveals the presence of safranin-stained globules within the sieve 


All sieve cells contain refractive spherules, located both centrally and periph- 
erally in the lumen (Figs. 12 and 13). Bounded by single unit membranes, the 


spherules range in size from 0.1 to 0.7 um. These spherules tend to aggregate 


POWER & SKOG: PTERIDIUM NECTARIES 


Fics. 6 and 7. Ultrastructure of secretory nectary cells. Fic. 6. TEM of plasmodesma strand con- 
necting cortical secretory cells. Within the plasmodesma are multivesicular bodies {right arrow) and 
ER (left arrow). Upper cell shows portion of nucleus (n). Fic. 7. SEM of the epidermal region around 


stoma. Two or three layers of cells in the epidermal region can be seen. The cuticle layer extends 


down into the stomate (arrow). 


AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 1 (1987) 
8 


100 mia 


e epidermal layer showing thin cuticle 
9. SEM of nectary surface showing isodia- 
the elongated cells and very few stomates 


Fics. 8 and 9. Epidermal layer of nectary. Fic. 8. TEM of th 
{arrow} and vacuoles with electron dense material. Fic. 
metric cells and more numerous stomates as opposed to 
of the stipe. 


POWER & SKOG: PTERIDIUM NECTARIES 9 


Fics. 10 and 11. Scanning views of nectary epidermis. Fic. 10. SEM of nectary epidermis at ces- 
sation of secretion with raised guard cells and trichomes of non-secretory region at left. Fic. 11. 
Open stomate with globular secretory material. 


10 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 1 (1987) 


near the sieve pores which connect the sieve cell to the surrounding paren- 
chyma. 

The vascular parenchyma cells stain more densely than the larger sieve cells 
(Fig. 12). The dense cytoplasm of the parenchyma contains plastids, mitochon- 
dria, endoplasmic reticulum, vesicles, and vacuoles. The vesicles and vacuoles 
contain a fibrillar material similar to that in the sieve cells (Fig. 13). Dictyosomes 
are not evident in these cells. 

Vascular parenchyma and sieve cells are interconnected by plasmodesmata, 
which join the parenchyma to the sieve pore areas of the sieve cell (Warmbrodt 
& Evert, 1974a and 1974b). ER may line the interior of these cytoplasmic con- 
nections. Figure 12 indicates the location of some of the plasmodesmata. 

The developmental stage of the frond determines the structure and functioning 
of the nectaries of Pteridium aquilinum. The nectaries differentiate and function 
only during the crozier stage. With maturation of the frond, the nectaries cease 
secreting and some lysigenous degradation is noted in the tissue (Fig. 1). Sec- 
ondary cell walls are formed in most cells, fewer globules and more tannins are 
seen in the secretory cells, a Casparian strip forms in the endodermal cells, and 
the stomata become closed (Fig. 10). 


DISCUSSION 


Studies of the ultrastructure of nectaries and the possible function of the cells 
have emphasized angiosperms. It is interesting to compare data derived from 
angiosperms with these observations of the Pteridium nectary. In angiosperm 
nectaries vascular tissue is generally separated from the secretory cells by non- 
glandular or subglandular parenchyma (Durkee, 1983). Characteristics of these 
cells include abundant mitochondria, well-developed vacuoles, large numbers 
of plasmodesmata, less dense cytoplasm, and less developed ER than the secre- 
tory cells. In Pteridium the cells of the basal region are well vacuolated, have 
granular cytoplasm, many mitochondria, and plastids, and are tightly packed in 
contrast to the surrounding parenchyma cells. Thus they possess most of the 
characteristics of subglandular parenchyma cells. 

Although there is no specialized vascular tissue supplying the nectary in Pte- 
ridium, the vascular strand of the meristele is only a few cells distant from the 
secretory cells. Furthermore, the cells of the pericycle and endodermis are highly 
modified in the region of the nectary. These cells are similar to the basal paren- 
chyma but lack the extensive development of vacuoles. The Casparian strips of 
the endodermal cells are not developed. Such changes in this region of the 
meristele may facilitate transport of the pre-nectar (phloem sap) from the phloem 
to the secretory parenchyma of the nectary. Durkee (1983) reported that a com- 
mon component of the parenchyma cells in the phloem of Passiflora nectaries 
is membrane-bound fibrillar material (proteinaceous) and that rough ER is abun- 
dant. Vesicles containing fibrillar material and abundant ER can be seen in the 
vascular parenchyma cells of Pteridium (Fig. 13). 

According to Fahn (1979b), the secretory cells of some angiosperm nectaries 
are characterized by increased numbers of mitochondria, increased amount of 


POWER & SKOG: PTERIDIUM NECTARIES 11 


Fic. 12. TEM section of the vascular tissue with xylem element (x), vascular parenchyma (p} and 

sieve cells (s) containing refractive spherules (r). Plasmodesmata are seen between vascular paren- 

chyma and sieve cells (arrow). Many small vesicles are present bordering the lumen of the sieve 
1 


12 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 1 (1987) 


- 13. TEM of sieve cell above and vascular parenchyma cell below, with fibrillar material in 
vacuoles and in a large membrane-bound organelle between the cytoplasm and the cell wall. E 


doplasmic reticulum can be seen in parenchyma cell (arrow). Refractive spherules (r) are obvious in 
the sieve cell. 


POWER & SKOG: PTERIDIUM NECTARIES 13 


ER with swollen cisternae, vesicles, and also, in others, numerous Golgi bodies. 
These characteristics were all noted in the cortical region of the Pteridium nec- 
tary. In addition, the distinct differences between the cortical cells of the nectary 
as opposed to the surrounding cortex of the stipe (Figs. 1 and 2) indicate that 
they are indeed modified and are probably secretory in function. 

The large nucleus which appears to fill most of each cortical secretory cell is 
almost obscured by the large number and size of safranin-stained red globules. 
These globules are similar to those seen in sieve cells and vascular parenchyma 
and represent an increase in size of 2-4 times, possibly indicating modification 
of the secretion in these cells. 

Once the secretion is modified, it may be packaged and/or transported by 
ultrastructural organelles in the cortical secretory parenchyma before it is re- 
leased into a substomatal chamber. Ultrastructural organelles involved in this 
transport of secretory material may include plasmodesmata, endoplasmic retic- 
ulum, and multivesicular bodies. Plasmodesmata not only link parenchyma cells 
to sieve cells in vascular tissue, but they also interconnect the cortical secretory 
parenchyma cells. They permit symplastic transport to occur throughout nectar- 
iferous tissue (Liittge, 1971). 

Durkee (1983) stated that ER is a notable feature in glandular cells, and the 
ER cisternae sometimes are swollen and vesicles are numerous. In Pteridium 
cortical secretory parenchyma, some ER cisternae are swollen with fibrillar in- 
clusions and many small vesicles are present. A form of packaging and secretion, 
noted by Fahn and Rachmilevitz (1975), involves vesicles that may bud off from 
the ends of the cisternae. As vesicles approach the plasmalemma, the mem- 
branes may fuse, releasing the fibrillar contents to the outside of the cytoplasm. 

Multivesicular bodies are aptly named for their structure because their func- 
tion remains unclear. Noted in plasmodesmata (Fig. 6), they may function in 
formation or transport of pre-nectar (Rachmilevitz & Fahn, 1973). Fahn (1979a) 
noted that multivesicular bodies are often present in active secretory cells and 
summarized several functions suggested for their role in the secretory process. 

Other characteristics of secretory cells also observed in the cortical secretory 
parenchyma of Pteridium include numerous mitochondria, few plastids, reduced 
vacuoles, and dictyosomes adjacent to the wall. 

With the exception of stomates, the epidermal cells covering the surface of 
the nectary do not appear to be involved directly with nectar secretion. No 
globules appear within the cells when viewed in light microscopy. Viewed in 
scanning electron microscopy, however, secretory stomates that are open are 
covered with a globular material. Closed stomates have none of the material. 
Since stomates are much more abundant in the epidermis of the nectary than in 
the rest of the stipe, they appear to be involved with more than respiration and 
gas exchange. Stomates, described as modified when they are coupled with 
secretion, usually remain open (Fahn, 1979a). However, closed stomates do occur 
on nectary surfaces. 

The ultrastructure of the nectaries of Pteridium aquilinum indicates that most 
of the organelles and cell types associated with vascularized angiosperm nec- 


14 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 1 (1987) 


taries are present in these fern nectaries. The Pteridium nectaries are definitely 
structured nectaries, contrary to Fahn’s (1979a) description of them as undiffer- 
entiated and Page’s (1982) statement that they are composed only of small, iso- 
diametric parenchymatous cells with dense protoplasmic contents. 

Field observations by Page (1982) indicate a secretory period from six days to 
about four weeks. There are corresponding anatomical changes that occur in the 
nonfunctional nectary following cessation of secretion. These changes in cell 
structure appear to occur from the epidermal region centripetally (Fig. 1). Epi- 
dermal cell walls become thickened first and the stomata are occluded (Fig. 10). 
In the cortical secretory parenchyma the globules decrease and tannins become 
more obvious. These cells also begin to form secondary cell walls. Lignification 
occurs in most cells and the endodermis forms a Casparian strip in the cells 
within the nectary region. 

The anatomy of Pteridium nectaries correlates with field observations and 
with the reported examples of other structured nectaries. Durkee (1983) noted 
that research studies of nectaries should be concerned with ultrastructural detail, 
the nature and function of subglandular tissue, and the vascular supply to nec- 
taries in regard to phloem physiology. The presence of modified tissue in the 
nectary and simple sieve cells in the phloem would appear to make Pteridium 
an excellent tool for physiological studies of phloem and nectary activity. 


ACKNOWLEDGMENTS 
Support for the SEM in this study was provided by NSF Grant BSR-8511148. Mr. Jan Endlich 
assisted in the preparation of the photographs; Ms. Effie Shaw (USGS) assisted in preparations of 


Figures 10 and 11. This study was done as partial completion of the requirements for a Master of 
Science Degree at George Mason University. 


LITERATURE CITED 
Darwin, F. 1877. On the glandular bodies on Acacia sphaerocephala and Cecropia peltata serving 


as food for ants, with an appendix on the nectar-glands of the common brake fern, Pteris 
aquilina. J. Linn. Soc., Bot. 15:398-409. 


of nectaries in relation to nectar secretion. Amer. ]. Bot. 66:977- 


and T. RACHMILEVITZ. 1975. An autoradiographical study of nectar secretion in Lonicera 
japonica Thunb. Ann. Bot. (London) 39:975-976. 
Lioyp, F.E. 1901. The extra-nuptial nectaries in th 


CA ch 


brake, Pteridium quili Science, 


Lorre: a. 1961. a die 7 t d ektare 1 4 5 
E, : Uber g Nekt: nd den Mechanismus seiner Sekretion. 


a . : porate “at function of plant glands. Annual Rev. Pl. Physiol. 22:23-44. 

, G. FE. ; tron microscopy laboratory techni : > Li 

: feces ry techniques. Monroe, New York: Library 

Pace, C. N. 1982. Field observations on the nectaries of b 
Bri 


racken, Pteridi ili : aa 
rit. Fern Gaz. 12:233-240 um aquilinum, in Britain. 


POWER & SKOG: PTERIDIUM NECTARIES 15 


RACHMILEVITZ, S and A. rie Hastie Ultrastructure of nectaries of Vinca rosea L., Vinca major L. 
and C eck cv. Valencia and its relation to the mechanism of nectar secre- 
ns ‘Agn. Bot. (London) = 1-9 
gop F. 1969. E Nektarie en. Beobachtt an Salix eleagnos Scop. und Pterid- 
um aquilinum (L.) Kuhn. Oesterr. Bot. Z. 117:205- 222. 
SPURR, a. R. 1969. A low viscosity epoxy resin embedding medium for electron microscopy. J. 
ltrastruct. Res. 26:31-43. 
WaARMBRODT, R. D. and R. F. Evert. SA gr Structure ae ec ngleciaasgs of the sieve element in the 
stem of Lycopodium lucidulum er. J. Bot. 6 
. 1974b. Structure of the auadac eae in ts stem som Lycopodium lucidulum. Amer. 
J. Bot. 61:437-443. 


American Fern Journal 77(1):16-27 (1987) 


Notes on XPleopodium and Pleopeltis in 
Tropical America 


JOHN T. MICKEL and JOSEPH M. BEITEL 
New York Botanical Garden, Bronx, NY 10458 


Interpretation of the polypodioid ferns has been a classic problem for pteri- 
dologists. The groupings are, with some exceptions, fairly clear, but whether 
they should be treated as subgenera of Polypodium or as distinct genera is a 
matter of considerable disagreement. There seems to be growing acceptance of 
their recognition as distinct genera (Smith 1981, 1985; Tryon & Tryon, 1982; Lel- 
linger, 1985; Mickel & Beitel, in press), although some authors continue to rec- 
ognize Polypodium in a broad sense (Proctor, 1977, 1985; Stolze, 1981). 

Some of the splinter genera in America (e.g., Campyloneurum, Niphidium) 
stand well apart from Polypodium s.s., but others are apparently very closely 
allied to Polypodium. This is especially so in regard to Pleopeltis, some members 
of which hybridize with species of Polypodium s.s. The best known example of 
this is Polypodium x leucosporum (=Polypodium lanceolatum x P. thyssanole- 
pis), which was pointed out by Vareschi (1969) and described in detail by Wagner 
and Wagner (1975), who treated it and its parents within Polypodium. Recently, 
Anthony and Schelpe (1985) described a similar case in southern Africa, in which 
Pleopeltis macrocarpa (Bory ex Willd.) Kaulf. (=Polypodium | latum) 
with Polypodium polypodioides var. ecklonii (Kunze) Schelpe. The authors gave 
a hybrid intergeneric name to their hybrid species, x Pleopodium simianum 
Anthony & Schelpe. 

In our studies on the ferns of Oaxaca, Mexico (in press), we have seen spec- 
imens from various parts of Mexico that involve several members of these two 
genera. Most of the hybrids have binomials under Polypodium. The purpose of 
this paper is to make several new combinations under x Pleopodium and Pleo- 
peltis, and to clarify the putative parentage of the hybrids. 


A. x Pleopodium leucosporum (Klotzsch) Mickel & Beitel, comb. nov.—Polypo- 
dium leucosporum Klotzsch, Linnaea 20:404, 1847.~Pleopeltis leucospora 
(Klotzsch) Moore, Index Fil. 77 ~Lepicysti 
( ; ul. 77. 1857.—Lepicystis leucospora (Klotzsch) Diels, 
in Engler & Prantl, Natur. Pflanz. 1(4):324. 1899..“TypE: Venezuela [‘“Colom- 
bia”], Moritz 306 (B!; isotype NY!}). 


raph 963; Wagner & Wagner, 1975; Walker, 1966, 
1973), resulting in theory in at least four types of crosses: diploid PI. macrocarpa 
with two cytotypes of P. thyssanolepis and tetraploid Pl. macrocarpa with two 


MICKEL & BEITEL: PLEOPODIUM AND PLEOPELTIS 17 


TABLE 1. Comparison of x Pleopodium leucosporum and its Parents. 


macrocarpa leucosporum thyssanolepis 
lade simple irregularly lobed pinnatifid 
Rhizome scales erose i fimbriate 
eins pleopeltid intermediate goniophlebioid 
Blade scales round, 0.3-0.4 mm diam. intermediate lanceolate, 0.6-1.5 
mm long, with 
elongate tip 
over 1 mm apart intermediate close to overlapping 
Soral scales black-centered lacking lacking 
ores ormal abortive normal 


cytotypes of P. thyssanolepis. All of these hybrids have abortive spores but differ 
in the contributions of the two parents, explaining the considerable variation of 
this hybrid (Wagner & Wagner, 1975; Table 1). 

Polypodium thyssanolepis is quite common in Mexico, but there does not seem 
to be any true Pleopeltis macrocarpa in Mexico, at least in Oaxaca and north. 
(Pleopeltis macrocarpa is distinguished by a combination of characters: black- 
centered soral scales, scattered fimbriate laminar scales, and black-centered, 
non-comose rhizome scales with conspicuous lumina.) Rather, Weatherby’s (1922) 
varieties (our species) take its place and most of them are involved in crosses 
with species of Polypodium. 


/2. xPleopodium tricholepis Mickel & Beitel, hybr. nov. (Fig. 1C-E).Tyee: 
Mexico, Oaxaca, Distrito Etla-Cuicatlan, 39 km N of Rte 190 past Telixtla- 
huaca, mixed oak-juniper forest along stream banks, 6200’, 8 Oct 1969, Mick- 
el 3873 (NY!; isotype UC!). 


Planta inter Pleopeltim mexicanam et Polypodium thyssanolepidem hybrida 
a parentibus frondium divisione intermedia et sporis abortivis differt. [Gr., thrix, 
hair, and lepis, scale, referring to the ciliate (comose) scales and also a combi- 
nation of the parental species epithets (Polypodium trichophora = Pleopeltis 
mexicana)]. 

Rhizome creeping, 1.5-2 mm diam.; rhizome scales deeply fimbriate, dimor- 
phic; rhizome scales 1.5-2 mm long, bicolorous with brown-black center and 
narrow pale brown margin, occasionally with long brown hairs from central 
point; scales at base of stipe 2-3 mm long, pale brown with or without short dark 
central streak: fronds distant to nearly clumped; stipe ca. of frond length, 
castaneous to atropurpureous, densely clothed with bicolorous scales, reddish 
brown with pale brown laciniate margin, round (0.5-0.8 mm wide) to lanceolate 
(1.5-2 mm long); blade irregularly pinnatisect, 16-18 cm long, 5-6 cm wide, 
broadest at base; pinnae or lobes 4-6 pairs, 5-7 mm wide; abaxial surface with 
scattered to dense, deeply fimbriate scales, mostly lanceolate (1.0-1.5 mm long), 
some round (0.8-1.0 mm wide), with brown center grading to whitish margin; 
midrib dark with lanceolate scales, 1.5-2 mm long; adaxial surface with sparse, 
deeply lacerate, lanceolate scales; sori round to slightly ovate, surrounded by 


AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 1 (1987) 


18 


MICKEL & BEITEL: PLEOPODIUM AND PLEOPELTIS 


TABLE 2. Comparison of x Pleopodium tricholepis and its Parents. 
mexicana tricholepis thyssanolepis 
Blade ple irregularly lobed pinnatifid 
Rhizome scales dimorphic dimorphic monomorphic 
densely co occasionally comose non-comose 
margin denticulate fimbriate fimbriate 


0.8-1 mm long (1.5-2 


1.5-2 mm long (2-3 mm 


3-4 mm long 


mm at stipe base) ase 
Stipe flattened slightly flattened terete 
Abaxial laminar sparse scattered to dense dense 
scales 


round, 0.3-0.5 mm diam. 


some round, mostly lan- 
ceolate, 1-1.5 mm 
long 


lanceolate, 0.6-1.5 mm 
long 


Hydathodes lacking present present 

Soral scales present, round, erose rare, lanceolate, fim- lacking 
briate 

Spores normal abortive normal 


laminar scales: soral scales rare, peltate to cordate, lanceolate, 0.8 mm long with 
pale brown point of attachment and lighter brown fimbriate margin; spores abor- 
tive. 

Distribution.—Epiphytic in oak-juniper woods. Known only from the type col- 
lection. 

Discussion.—This specimen, exhibiting hybrid characters of abortive spores 
and irregular blade lobes, appears to represent a hybrid between the pinnatisect 
species Polypodium thyssanolepis (Fig. 1F-H) and a simple-bladed species of 
Pleopeltis (see Table 2). Although P. conzattii was the only species of Pleopeltis 
found at the same locality, the presence of round scales and the absence of 
black-centered rachis scales makes that species unlikely as a possible parent. 
The few soral scales are not strongly and deeply bicolorous as one would expect 
in hybrids involving P. polylepis, P. crassinervata, P. astrolepis, and P. interjecta. 
Pleopeltis mexicana (Fig. 1A, B), with lightly colored soral scales, long stipe, and 
occasional tufts of long hairs on the rhizome scales, is probably the other parent. 

Another specimen (Mexico, DF, Angostura, 2600 m, Lyonnet 3414, US) has 
rhizome scales monomorphic and larger (2-2.5 mm long vs. 1.5-2), blade scales 
darker, farther apart (sparse), and smaller (0.8-1 mm long vs. 1-1.5), pinna pairs 


— 
Fic.1. Hybrids of x Pl li j their putative p ts. A, B. Pleopeltis mexicana (Mickel 4357, 
NY, Oaxaca). A. Habit. B. Abaxial blade scale. C-E. x Pleopodium tricholepis (Mickel 3873, NY, 


holotype). C. Habit. D. Abaxial blade detail. E. Abaxial blade scale. F-H. Polypodium thyssanolepis 
(Mickel 7065, NY, Oaxaca). F. Habit. G. Abaxial blade detail. H. Abaxial blade scale. I, J. Pleopeltis 
polylepis (Mickel 7065, NY, Oaxaca). I. Habit. J. Abaxial blade detail. K, L. * Pleopodium bartlettii 
(Bartlett 10286, US, holotype). K. Habit. L. Abaxial blade detail. M, N. Polypodium polypodioides 
var. aciculare (Mickel 1019, NY, Oaxaca). M. Habit. N. Abaxial blade detail. 


20 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 1 (1987) 


TABLE 3. Comparison of x Pleopodium bartlettii and its Parents. 


polylepis bartlettii polypodioides 
Blade simple irregularly lobed pinnatifid 
Rhizome scales rarely comose non-com non-comose 


Laminar scales: 


Abaxial round round & lanceolate round & lanceolate 
dark red-center red-brown center dark brown center 
0.5-1 mm diam. 0.1-0.3 mm diam. 0.1-0.3 mm diam. or 0.4- 
0.6 mm lon, 
Adaxial round, erose round base with long tip, minute base with long tip 
some round 
Soral scales round round lacking 
Spores 37-45 um (42.2) 37-42 um (38.5) 37-42 um (39.7) 


fewer, and soral scales dark-centered (vs. pale brown-centered). Tentatively we 
consider it as a variant of xP. tricholepis, but conceivably it may represent the 
hybrid of Pleopeltis interjecta x Polypodium thyssanolepis. 


“3. xP um bartlettii (Weath.) Mickel & Beitel, comb. et stat. nov. (Fig. 1K, 
L).—Polypodium bartlettii Weath. (pro sp.), Amer. Fern J. 25:56. 1935.—TyPE: 
Mexico, Tamaulipas, vicinity of San José, on tree trunks, Bartlett 10286 (US!; 
isotype fragment GH!). 


Hybrid between Pleopeltis polylepis (Roemer ex Kunze) Moore (Fig. 11, J) and 
Polypodium polypodioides (L.) Watt (Fig. 1M, N) (Table 3), as suggested by 
Weatherby (1935), who noted similarities with P. leucosporum; he also thought 
it might be an odd form of P. lanceolatum. Reproductive status unknown. 

Distribution.—Known only from the type collection. 


4. xPleopodium fallacissimum (Maxon) Mickel & Beitel, comb. et stat. nov. (Fig. 
2F, G).—Polypodium fallacissimum Maxon (pro sp.), Contr. U.S. Natl. Herb. 
17:567. 1916.—TypE: Mexico, Coahuila, San Lorenzo Canyon, 6 mi SE of 
Saltillo, Palmer 426 (US!; isotype NY!). 


Hybrid between Pleopeltis erythrolepis (Weath.) Pic. Ser. (Fig. 2B-E) and 
Polypodium guttatum Maxon (Fig. 2H-J) (see Table 4), as suggested by Thomas 
Wendt [note on paratype at US and pers. comm. as polylepis x guttatum; he 


considered erythrolepis as a variety of polylepis (Wendt, 1980)]. Reproductive 
status unknown 


Distribution.—Southern Coahuila. 


Discussion.—The paratype, Palmer 425, is smaller and less lobed, and is prob- 
ably just a smaller form of Ple 


MICKEL & BEITEL: PLEOPODIUM AND PLEOPELTIS 21 


description Fée says there are scales among the sporangia, suggesting that Pleo- 
peltis is involved. : 

In our work on Oaxacan pteridophytes (in press), we are treating the tradi- 
tional varieties of Polypodium lanceolatum (Pleopeltis macrocarpa) as distinct 
species, necessitating two new combinations under Pleopeltis. 


Pleopeltis interjecta (Weath.) Mickel & Beitel, comb. nov.--Polypodium pelta- 
agf4 tum Cav. var. interjectum Weath., Amer. Fern J. 34:17. 1944.<Pleopeltis 
: macrocarpa var. interjecta (Weath.) A. R. Smith, Amer. Fern J. 70:26. 1980.— 
Type: Guatemala, Chimaltenango, 2700 m, Standley 60957 (F). 

Pleopeltis mexicana (Fée) Mickel & Beitel, « comb. nov. (Fig. 1A, B).Drynaria 
mexicana Fée, Mém. foug. 8:97. 1857.—SyNTyPEs: Mexico. Veracruz, Gal- 
eotti 6321; Puebla, Schaffner 179; Popocatepetl, Schaffner 292 (P?). 


a 


“Polypodium lanceolatum L. var. trichophorum Weath., Contr. Gray Herb. 65: 
8. 1922.TypE: Mexico, lava fields near Eslaba, 8000’, Pringle 11797 (GH!; 
isotype US!). 


The type of Pleopeltis (P1. angusta) is pinnate, but as commonly construed the 
genus is comprised mostly of species with undivided fronds. The genus is dis- 
tinguished by its peltate scales in the sorus. Polypodium (Pleopeltis) percussum 
Cav. seems to lack them, even in the very young sori, and might better be 
considered a Microgramma. Conversely, some species often treated as Pleopeltis 
have peltate soral scales, but in other respects seem disparate, e.g., Pleopeltis 
munchii (Christ) A. R. Smith. 

One species usually treated as a Polypodium, P. fallax, has peltate soral scales, 
and therefore we are placing it in Pleopeltis. Another interesting feature of this 
species is its comose rhizome scales, which are present in nearly all species of 
Pleopeltis in Mexico. Such scales are also common in Pecluma and only occa- 
sionally in Polypodium, e.g., P. adelphum Maxon. 


.<4 Pleopeltis fallax (Schlecht. & Cham.) Mickel & Beitel, comb. nov. (Fig. 28-U).— 
Polypodium fallax Schlecht. & Cham., Linnaea 5:609. 1830.—TyPE: Mexico, 
[Veracruz], Misantla, Schiede & Deppe 758 (BI, photo BM!; isotype LE!). 

Pleopeltis fallax apparently hybridizes with two other species of Pleopeltis. (If 
Pl. fallax is maintained in Polypodium, then these hybrids would fall into x Pleo- 
podium.) 

“Pleopeltis x sordidula (Maxon & Weath. in Weath.} Mickel & Beitel, comb. et 
stat. nov. (Fig. 2P-R).—Polypodium sordidulum Maxon & Weath. in Weath. 
(pro sp.), Amer. Fern J. 17:92. 1927.<TypE: Mexico, Veracruz, epiphytic in 
coffee trees near Cordoba, Spence 114 (GH!; isotype US!). 

Since the original description of this species included both this and the fol- 
lowing hybrid, we are redescribing Pl. xsordidula [hybrid between Pleopeltis 
astrolepis (Liebm.) Fourn. (Fig. 2L-O) and Pl. fallax (Fig. 25-U)] here in a more 
restricted sense. 

Rhizome creeping, 0.5-0.8 mm diam.; rhizome scales 0.1-0.3 mm diam., round, 


22 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 1 (1987) 


ri 
ary INGOT, 
OTR 


, SL Petey 
OTE RES? 
- 32. 
45 iiss a; 


3 


MICKEL & BEITEL: PLEOPODIUM AND PLEOPELTIS 


TABLE 4. Comparison of x Pleopodium fallacissimum and its Parents. 


erythrolepis fallacissimum guttatum 
Blade simple irregularly lobed pinnatifid 
Rhizome scales margin cells lengthwise margin cells toward margin cells toward 
margin margin 
sclerotic sclerotic-clathrate sclerotic-clathrate 
comose non-comose non-comose 
Stipe flattened flattened terete 
Laminar scales: 
Abaxial dense to scattered sparse 
round & lanceolate deltate-ovate deltate-ovate 
non-clathrate subclathrate clathrate 
emarginate marginate marginate 
brown center dark only at attachment _no dark center 
Adaxial aera lanceolate lacking 
fimbria denticulate 
dark cee no dark center 
Hydathodes lacking present present 
“Stretch marks” present present lacking 
(epidermal lines) 
Soral scales frequent lacking 


rare 
clathrate center 
light brown margin 


dark brown center 
brown to light brown 
margin 


bicolorous, with dark brown to black center with narrow whitish margin, comose 
from central point with 0.5-0.7 mm reddish brown hairs, margin entire; fronds 
cose stipe %-% of frond length, flattened, atropurpureous, scales api 
und with comose center; blade irregularly pinnatifid, 4-7 cm long, 0.5-1.0 c 
Bee linear-lanceolate, with acuminate tip; abaxial lamina with sparse Bait 
scales, some lanceolate (0.5-0.7 mm long), some round (0.1-0.3 mm diam.), with 
semi-clathrate center and tan to whitish tan margin; midrib dark with scattered 
lanceolate scales with dark sclerotic center; lateral veins obscure; adaxial lamina 
with sparse scales similar to abaxial scales except more deeply cut; soral scales 


— 
Fic. 2. Hybrids of x Pleopodium and pe aod. ~—_ putative parents. * Pleopeltis crassiner- 
vata (Hallberg 1389, NY, Oaxaca), habit. & Gentry 22975A, NY, 
Chihuahua). B. Habit. C. Abaxial blade an Dz. Abexial blade scale. E. Adaxial blade scale. F, G. 
x Pleopodium fallacissimum (Palmer 426, NY, isotype). F. H Habit. G. Abaxial blade scale. H-]. Poly- 
podium guttatum (Rzedowski 24040, NY, Edo. Mexico). H. Habit. I. Abaxial blade detail. J. Abaxial 
blade scale. K Purpus 5798, US, holotype), habit. L-O. Pleopeltis oe 
(Mickel 5165, NY, Oaxa ca). L. Habit. M. Abaxial blade scale. N. Abaxial blade detail. O. —— 
e. P-R. Pleopeltis x sordidula (Copeland 149, US). P. Habit. Q. Stipe detail. R. Rhizom 

S-U. Pleopeltis fallax (Mickel 6468, NY, Oaxaca). S. Habit. T. Abaxial blade detail. U. iad blade 
scale. 


AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 1 (1987) 


<, 


24 
TaBLE 5. Comparison of Pleopeltis x sordidula and its Parents. 
astrolepis sordidula fallax 
Blade simple irregularly lobed pinnate-pinnatifid to 
pinnate-bipinnatifid 
Rhizome scales bicolorous icolorous concolorous 
black t ] black center/whitish black 


rown margin 
ovate, 0.3-0.5 mm long 


margin 
round, 0.1-0.3 mm diam. 


round, 0.3-0.5 mm 
diam. 


comose comose c 
Stipe short, flattened short, slightly flattened long, terete 
Rachis scales black sclerotic brown with sclerotic clathrate 
center 
Laminar scales: 
Abaxial scattered sparse rse 
lanceolate (0.5-1 mm) to lanceolate (0.5-0.7 mm) _ lanceolate (0.5-0.8 
round (0.3-0.5 mm} ‘o round (0.1-0.3 mm) mm) 
bicolorous bicolorous concolorous 
dark brown center/pale _ brown center/pale rown 
brown margin brown to whitish mar- 
gin 
non-clathrate semiclathrate clathrate 
Adaxial as abaxia as abaxial lacking 
elongate elongate round 
Spore size 44-51 um (47.1) 38-45 wm (41.8) 37-45 um (40.1) 


extremely rare, round to lanceolate (0.3-0.5 mm long), dark brown to brown 
conus with light brown fimbriate margin; sori slightly oval; spores appearing 
n 


ormal. 

Additional collections: Mexico. Veracruz: Rio Blanco, 4800’, Fisher 37 (US); 
Metlac [near Fortin de las Flores], 900 m, Copeland 149 (US). 

Polypodium sordidulum was based on three specimens from Veracruz. Maxon 
and Weatherby considered it “a local offshoot of P. astrolepis.” They believed 
it to be similar to P. astrolepis except for the lobing, subterete stipe, and orbicular 
sori (although the figure shows them as oblong). 

The type specimen and one of the two paratypes (Fisher 37) appear to rep- 
resent the hybrid combination of Pleopeltis astrolepis x Pl. fallax: the other 
paratype (Purpus 5798) differs in several characters from this taxon. These dif- 
ferences, which may be due to its presumed origin as the hybrid of Pleopeltis 

ax and PI. crassinervata, will be discussed under the next hybrid 

Pleopeltis x sordidula (see 
between a simple-bladed parent (PI. astrolepis) and a more divided parent (PI. 
fallax), with blade scale distribution (on both abaxial and adaxial surfaces, al- 


MICKEL & BEITEL: PLEOPODIUM AND PLEOPELTIS 


TaBLE 6. Comparison of Pleopeltis x melanoneuron and its Parents. 


crassinervata melanoneuron fallax 
Blade simple irregularly lobed pinnate-pinnatifid to 
pinnate-bipinnatifid 

Rhizome scales lanceolate lanceolate round 

1-1.5 mm long 0.5-0.8 mm long 0.3-0.5 mm diam. 

bicolorous bicolorous concolorous 

dark brown center/wide dark brown to black black 

pale brown margin center/narrow pale 
argin 

comose comose 
Rachis scales non-comose n comose 

lanceolate lanceolate round 


Laminar scales: 


Abaxial scattered sparse sparse 
lanceolate (0.8 mm long) lanceolate (0.5 mm long) _ lanceolate (0.3) 0.5-0.8 
and round (0.5 mm and round (0.5 mm mm long 
i diam.) 
bicolorous bicolorous concolorous 
brown center/pale brown center/pale clathrate 
brown margin rgin 
Adaxial as abaxial as abaxial lacking 
Lateral veins dark, evident dark, evident obscure 


trolepis, semiclathrate center as in the clathrate scales of PI. fallax). The short, 
slightly flattened stipes are intermediate between the short, strongly flattened 
stipe of Pl. astrolepis and the terete stipe of Pl. fallax; the bicolorous rachis scales 
are also intermediate between its two parents. 


Pleopeltis x melanoneuron Mickel & Beitel, hybr. nov. (Fig. 2K).—Tyee: Mexico, 
Veracruz, Zacuapan, Jan 1912, Purpus 5798 (US!; isotype UC). 


Planta inter Pleopeltim crassinervatam et P. fallacem hybrida a parentibus 
frondium divisione rhizomatisque squamis intermedia differt et a P. x sordidula 
nervis nigris laminaeque apice acuta abstat. (Gr., melos, black, and neuron, vein, 
referring to the dark secondary veins at the base of the blade.) 

Rhizome creeping, 0.5-0.8 mm diam.; rhizome scales 0.5-0.8 mm long, lanceo- 
late, with dark brown to black center and narrow pale brown margin, comose 
from central point with 0.5-0.8 mm reddish brown hairs, margin fimbriate; fronds 
distant; stipe ca. % of frond length, round, atropurpureous; blade irregularly 
pinnatifid, 4.3-5.5 cm long, 1.0-1.8 cm wide, oblanceolate, with acute tip, lateral 
lobes with irregular teeth at tips; abaxial and adaxial surface with sparse fim- 
briate scales, some lanceolate (0.5 mm long), some round (0.1-0.3 mm diam.}, 
with brown center and pale brown margin; midrib dark with scattered lanceolate 
scales, 0.8-1.0 mm long, similar in color to laminar scales, lateral veins evident, 
blackened, especially at base; soral scales extremely rare, peltate, 0.1 mm wide, 
light brown with fimbriate margin; sori slightly oval; reproductive status uncer- 


26 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 1 (1987) 


tain, specimen with juvenile sporangia and mature sporangia open with no spores 


esent. 

Additional collection: Mexico, Veracruz, Altotonga, 1938, M. B. Foster 14 (US!). 

Discussion.—The type of Pl. x melanoneuron was originally a paratype of 
Polypodium sordidulum. However, it differs in several characters that point to 
a separate origin as the hybrid of Pleopeltis fallax (Fig. 25-U) and PI. crassiner- 
vata (Fig. 2A) (see Table 6). It resembles Pl. x sordidula in its irregular blade 
division and both lanceolate and round scales on the adaxial and abaxial laminar 
surfaces. The evident lateral veins, blackened at their bases, differ significantly 
from the obscure lateral veins of Pl. x sordidula and point to Pl. crassinervata, 
with its black lateral veins, as its simple-bladed parent. The rhizome scales of 
Pl. xX melanoneuron are definitely lanceolate, bicolorous, and larger than those 
of PI. x sordidula, again pointing to PI. crassinervata with its longer, lanceolate 
rhizome scales with wider pale margins than in Pl. astrolepis. The stipe in PI. 
x sordidula is flattened (from Pl. astrolepis with its strongly flattened stipe), 
whereas in PI. x melanoneuron the stipe is terete (since both PI. fallax and PI. 
crassinervata have terete stipes). The rachis scales in Pl. x melanoneuron are 
lanceolate and non-comose (similar to Pl. crassinervata), whereas in PI. x sor- 
didula they are round and comose (both PI. fallax and PI. astrolepis have comose 
rachis scales, round in the former and lanceolate and round in the latter). The 
oblanceolate blade of PJ]. x melanoneuron is acute at the apex and has scales 
with dark, non-clathrate centers, whereas PI. x sordidula has a linear-lanceolate 
blade with an acuminate apex and scales with dark, semiclathrate centers. 


ACKNOWLEDGMENTS 


We thank Thomas Wendt, Michael Nee, and W. H. Wagner Jr. for useful discussions, and Rupert 
Barneby for his kind assistance with the Latin diagnoses. Figures 1A, B, F-H, M, N, 2A, L-N, $-U 
were drawn by H. Fukuda; the other figures and the plate layouts were prepared by Bobbi Angell. 
i Pee and publication were supported in part by the Joann Wolfe Ward Memorial Research 

und. 


LITERATURE CITED 


ANTHONY, N. C. and E. A. SCHELPE. 1985. x Pleopodium—a putative intergeneric fern hybrid from 
Africa. Bothalia 15:555-559. 

Evans, A. M. 1963. New chromosome observations in the Polypodiaceae and Grammitidaceae. 
Caryologia 16:671-677. 

LELLINGER, D. B. 1985. A field manual of the ferns & fern-allies of the United States & Canada. 
Washington, D.C.: Smithsonian Institution Pr. 


MIcKEL, J. T. and J. M. Berret. Pteridophyte flora of the state of Oaxaca, Mexico. Mem. N.Y. Bot. 
Gard. (in press). 
Proctor, G.R. 1977. Flora of the Lesser Antilles. Vol. 2. Pteridoph i i 
re npn years I phyta. Jamaica Plains, Massachu- 
PLL co. Ferns of Jamaica. London: British Museum (Natural History). 
MITH, A. R. 1981. Pteridophytes. In Flora of Chiapas, Part 2, ed. D. E. B isco: 
biliinal Mele seas . Breedlove. San Francisco: 


. 1985. Pteridoph i i i i 
Ast ophytes of Venezuela, an annotated list. Berkeley, California: published by 


MICKEL & BEITEL: PLEOPODIUM AND PLEOPELTIS 27 


Stouze, R. G. 1981. Ferns and fern allies of Guatemala. Fieldiana, Bot. n.s. 6:i-vii, 1-522. 

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

VARESCHI, V. 1969. Flora de Venezuela. Vol. 1. Helechos. Caracas: Instituto Botanico. 

Wacner, W. H. Jr. and F. S. WAGNER. 1975. A hybrid polypody from the New World tropics. Fern 
Gaz. 11:125-135. 

Wa LkeER, T. G. 1966. A cytotaxonomic survey of the pteridophytes of Jamaica. Trans. Roy. Soc. 
Edinburgh 66:169-237 + 5 
. 1973. Additional cytotaxonomic notes on the pteridophytes of Jamaica. Trans. Roy. Soc. 
Edinburgh 69(5):109-135. 

WEATHERBY, C. A. 1922. The group of Polypodium lanceolatum in North America. Contr. Gray 
Herb. 65:3-14. 

———.. 1935. On certain Mexican and Central American ferns. Amer. Fern J. 25:52-59. 

WENDT, T. 1980. Notes on some Pleopeltis and Polypodium species of the Chihuahuan Desert 
region. Amer. Fern J. 70:5-11. 


1987 AIBS MEETING—CALL FOR PAPERS 


The American Fern Society and the Botanical Society of America will meet 
with the American Institute of Biological Sciences at Ohio State University, 
Columbus, 9-13 August 1987. Members of the American Fern Society wishing 
to present a paper or poster may obtain abstract forms from the Program Chair- 
man: Dr. Christopher H. Haufler, Department of Botany, University of Kansas, 
Lawrence, KS 66045. 

A Fern Foray to the Hocking Hills will take place on Sunday, 9 August 1987. 
Cost: $21 per person, includes box lunch. Trip is limited to 45 people and will 
leave from the 12th Avenue (south side) of the Ohio Union at 8:00 a.m. and 
return at 6:00 p.m. The pteridophytes of the unglaciated areas south of Columbus 
will be studied in their natural habitat and will include especially those of the 
sandstone cliff faces and grottos. Hybrids will be seen in several genera: As- 
plenium x trudellii, Cystopteris x wagneri, and Dryopteris x neo-wherryi, as 
well as the remarkable grotto gametophytes of Vittaria and Trichomanes. A 
number of threatened and endangered species on the Ohio list will be observed, 
but collecting of only common species will be permitted. Leaders are R. James 
Hickey, W. H. Wagner Jr., and Charles R. Werth. For more information, contact 
Wagner at (313) 764-1484. 


American Fern Journal 77(1):28-32 (1987) 


Flavonoids and Spores of Platyzoma microphyllum, 
an Endemic Fern of Australia 


ECKHARD WOLLENWEBER and CORNELIA SCHEELE 
Institut fir Botanik der TH, D-6100 Darmstadt, West Germany 
ALICE F. TRYON 
Gray Herbarium, Harvard University, Cambridge, MA 02138 


This is the first report of flavonoid aglycones in the Australian endemic fern, 
Platyzoma microphyllum R. Br. These are deposited on the leaves as filamentous 
crystals from glandular trichomes (Barthlott & Wollenweber, 1981). Such lipo- 
philic deposits have been identified on leaves of several genera of the Pterida- 
ceae. They occur in almost all species of Pityrogramma, many in Notholaena, 
some in Cheilanthes, and a few in Pellaea and Pterozonium (Wollenweber, 
1978). They have been designated as “wax” or “ceraceous indument” in taxo- 
nomic literature and as “farinose exudate” or merely as “farina” in phytochem- 
ical literature. The term “wax” is incorrect as this material is not a true wax in 
chemical terms. Extensive studies of the chemical composition of such exudates 
have revealed that they are usually a mixture of flavonoid aglycones (Wollen- 
weber, 1978), sometimes with considerable amounts of terpenoids (Wollenweber 
et al., 1982; Arriaga-Giner & Wollenweber, 1986). As part of our continuing stud- 
ies of the flavonoid excretions (Wollenweber, 1985) we have made a detailed 
analysis of Platyzoma microphyllum. The species is remarkable for the incipient 
heterosporous condition (A. Tryon, 1964: A. Tryon & Vida, 1967; Duckett & Pang, 
1984} and production of filamentous male gametophytes as well as largely ar- 
chegoniate spatulate gametophytes. 


MATERIAL AND METHODS 


Specimens were collected at Nicotine Creek, southwest of Mareeba, Queens- 
land, Australia (R. & A. Tryon 7342, GH; Wollenweber Herbarium, Darmstadt). 
The plants were growing in dry Eucalyptus scrub, an unusual habitat for ferns 
(Fig. 1). There were many large clumps growing among sparse grass, on sand 
that was damp below the surface. A xeric habitat is often characteristic of other 
species of the Pteridaceae with flavonoid aglycones. 


tified by co-chromatography with markers (for details of experimental routine 
procedures see, e.g., Wollenweber et al., 1985). 


WOLLENWEBER ET AL.: PLATYZOMA 29 


RESULTS 


The spores are of two sizes, with the larger ones 162-190 um and usually 16 
per sporangium. The small spores are 65-105 wm and usually 32 per sporangium. 
The size differences are apparent in the sample illustrated (Fig. 3). A granulate 
deposit often covers the reticulate surface in both spore types. There are a series 
of parallel ridges in the equatorial region of the large spores (Fig. 4) that are 
absent in the small ones. The development of morphologically distinct gameto- 
phytes from the two spore types and origin of heterospory were examined by 
Duckett and Pang (1984). 

The leaves have indeterminate growth with two rows of coriaceous, pouch- 
like pinnae adjacent to the rachis (Fig. 2). Two- or three-celled, capitate glands 
are abundant on the pinnae, which usually have accumulations of yellowish 
exudate on the surface. Freshly collected specimens have a characteristic scent 
and, when pressed, leave an oily stain on paper. The odor and stain might be 
due to the presence of essential oils of some kind. The flavonoid aglycones that 
cause the yellow color of the leaf exudate were analyzed. 

Two major flavonoids were obtained in crystalline form. They were found to 
be 2',6’-dihydroxy-4’-methoxy chalcone (Fig. 5a) and 2',6’-dihydroxy-4’,5’-di- 
methoxy chalcone (pashanone, Fig. 5b). As minor components, we identified the 
flavanone pinocembrin-7-methyl ether (pinostrobin, Fig. 5c) and the flavanols 
galangin-7-methy] ether (izalpinin, Fig. 5d), galangin-3,7-dimethyl ether (Fig. 5e), 
kaempferol-7-methy] ether (rhamnocitrin, Fig. 5f), kaempferol-3,7-dimethyl ether 
(kumatakenin, Fig. 5g), and kaempferol-3,7,4’-trimethy] ether (Fig. 5h). Some fur- 
ther trace constituents could not be identified, due to lack of material. 

The 2',6’-dihydroxy-4’-methoxy chalcone is a typical component of yellow or 
orange leaf exudates in Pityrogramma species (except P. triangularis; Wollen- 
weber & Dietz, 1980). Several species of Cheilanthes and Notholaena also pro- 
duce this chalcone (Wollenweber, 1982a). The rare 2’,6’-dihydroxy-4’,5’-dime- 
thoxy chalcone, pashanone, occurs jointly with the former chalcone as major 
constituents of the leaf exudate of Onychium siliculosum (Desv.} C. Chr. (Wol- 
lenweber, 1982b). Pinocembrin-7-methyl ether is assumed to be an artifact, de- 
rived from the first mentioned chalcone by cyclization. The methylated flavonols 
(Figs. 5d-h) are common constituents of the ceraceous indument in Cheilanthes 
and Notholaena species and in some varieties of Pityrogramma triangularis 
(Smith, 1980). They were also encountered, in trace amounts, in the thin epicu- 
ticular layer on glaucous leaflets of several Pellaea species (Wollenweber, 1982a). 


DISCUSSION 


The systematic position of this fern has long been in question, due to its un- 
usual leaf morphology, incipient heterosporous condition, and two forms of ga- 
metophyte. It was initially placed in the Gleicheniaceae by Robert Brown in 
1810 or considered a separate family, Platyzomataceae, in the Gleicheniales 
(Nakai, 1950). It was treated as subfamily Platyzomatoidecae in the Polypodiaceae 


30 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 1 (1987) 


Fics.1-4. Platyzoma microphyllum. Fic. 1. Three 
background, near reeba, Queensland, Australia. Fic. 
pinnae, bar = 
face of _ sig 
(GH), bar = 


clumps . the plants with Eucalyptus trees in the 


art of filiform leaf with pouch-like 
m. Fic. 3. Small spores and three large ones below, bar = 500 um. Fic. 4. Proximal 


ie sola ridges in the equatorial region. Queensland, pliner Brass & White 64 


WOLLENWEBER ET AL.: PLATYZOMA 31 


HCO OH a HCO OH 
“one Cre 
B H,CO b 
OH O 


OH O 


H3CO O H5CO 0 
a ae 
OR OR’ 
OH O OH O 


d :R=H f -R'=R=H 
e :R=CH, g :R'=CH3,R=H 
h :R'=R*=CH, 


Fic. 5. Flavonoid aglycones that cause the yellow color of the leaf exudate in Platyzoma micro- 
phyllum: a—2',6’-dihydroxy-4’-methoxy chalcone, b—pashanone, c—pinostrobin, d—izalpinin, e— 
galangin-3,7-dimethy] ether, f—rhamnocitrin, g—kumatakenin, h—kaempferol-3,7,4’-trimethy] ether. 


(A. Tryon 1961, 1964), and in the tribe Platyzomateae in the Pteridaceae (R. & 
A. Tryon, 1982). Flavonoid aglycones that are excreted and accumulated on the 
leaf surface have so far been reported for representatives of the genera Pityro- 
gramma and Pterozonium (tribe Taenitideae), for Cheilanthes, Notholaena, Pel- 
laea, Negripteris, Sinopteris, and Onychium (tribe Cheilantheae), and for Adian- 
tum (tribe Adianteae) (Wollenweber, 1979). 

Several characteristics of the plants as the medulated protostele, sporangium 
morphology, and especially the chromosome report of n = 38 suggest connections 
with the Schizaeaceae (A. Tryon & Vida, 1967). The incipient heterosporous 
condition, known in several unrelated families as the Selaginellaceae, Marsile- 
aceae, and Salviniaceae, has undoubtedly developed independently in Platy- 
zoma. The presence of flavonoid aglycones, such as those forming the lipophilic 
leaf exudate in Platyzoma microphyllum, are thus far known only in the Pteri- 
daceae. This supports its treatment in that family, while the unusual morphology, 
chromosome number, and heterosporous condition support recognition of the 
species in a separate tribe. 

ACKNOWLEDGMENTS 

Phytochemical studies were supported by Deutsche Forschungsgemeinschaft grant Wo 231/3-3 to 
Wollenweber. Work on spores is supported by a National Science Foundation Grant BSR 84-07046 
to A. F. Tryon and R. Tryon. 


32 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 1 (1987) 


LITERATURE CITED 


— F. J. and E. WOLLENWEBER. 1986. 6a-Acetoxy-168, 22-dihydroxyhopan 24-oic acid, 
rpene from the fern Notholaena candida var. copelandii. Phytochemistry 25:735-737. 
BARTHLOTT, W. and E. WoLLENWEBER. 1981. Zur Fein hemie und taxonomischen Signifi- 
kanz epicuticularer Wachse und ahnlicher Sekrete. “Heo. Subtrop. Pflanzenwelt (Akad. 
Wiss. Lit. Mainz) 32:35-97. 
Duckett, J. G. and W. C. Panc. 1984. The origins of heterospory: a comparative study of the sexual 
haviour in the fern ches basta microphyllum R. Br. and the horsetail Equisetum gigan- 
teum L. J. Linn. Soc., Bot. 
Nakal, T. 1950. A new aon: a Cech Bull. See Sci. Mus. 29:1-71. 
SmirH, D. M. 1980. Flavonoid analysis of the P. g plex. Bull. Torrey Bot. 
Club 107: 134- 145. 
Tryon, A. F. 1961. Some new aspects of the fern Platyzoma microphyllum. Rhodora 63:91-102. 
——. 1964. Platyzoma—A Queensland fern with incipient heterospory. Amer. J. Bot. 51:939- 


nn G. Vipa. 1967. Platyzoma: A new look at an old link in ferns. Science 156:1109-1110. 
TRYON, M. and A. F. Tryon. 1982. Ferns and allied plants, with special reference to tropical 
America. New York: Springer-Verlag 
sperm as E. 1978. The distribution und ag constituents of the farinose exudates in 
mnogrammoid ferns. Amer. Fern J. 68: 
: ae Einige Neufunde externer sated i bei amerikanischen Farnen. Flora 268:138- 


. 1982a. Flavonoid aglycones as constituents of epicuticular ewe in ferns. In The plant 
cuticle, ed. D. F. Cutler et al. Linn. Soc. Symp. Ser. 10:215-22 
1982b. The pte Car of — in the farinose eo of the fern Onychium 
siliculoeam. Phytochemistry 21:1462-1464. 
1985. isphck ame sate on cheilanthoid ferns. Taxon 34:356-357 
a V. H. Dietz. 1980. Flavonoid patterns in the farina of auldanbenk and silverback 
ferns. — Syst. nag 8:21-33. 
G. SCHILLING, J. FAVRE-BONVIN, and D. M. SMITH. 1985. hopes from che- 
mick pee ae the scktbeck fern, Pityrogramma triangularis. Phytochemistry 24:965-971. 
————., P. Ruepi, and D. S. SeIGLER. 1982. Diterpenes of Cheilanthes argentea. 25 gare ac: 
1283-1285. 


American Fern Journal 77(1)}:33-35 (1987) 


A New Species of Danaea from Peru 


ROBERT G. STOLZE 
Department of Botany, Field Museum of Natural History, Chicago, IL 60605 


Current studies for the “Pteridophytes of Peru,” in which I am collaborating 
with Dr. Rolla Tryon of Harvard University, have already yielded some fasci- 
nating discoveries worthy of special note. The first of these to come to my atten- 
tion was a new species of Danaea, which is hereunder described. 


{10% Danaea oblanceolata Stolze, sp. nov. (Figs. 1-3).<Type: Peru, Dept. Pasco (as 
Junin), Cahuapanas, on Rio Pichis, Killip & Smith 26777 (US, frag. F). 


Folia sterilia 40-50 cm longae, 13-18 cm latae, ad apices prolificantia; petiolus 
1-2-nodosis; pinnae 10-12-jugae, late oblanceolatae, usque ad 11 cm longae et 
2.8 cm latae, ad apicem acuminatae et serratae abrupte terminantes; venae ple- 
rumque simplices, autem interdum geminae vel 1-furcatae. Pinnae fertiles cerca 
12-jugae, usque ad 8 cm longae et 1 cm latae, ad apicem obtusae. 

Sterile leaves 40-50 cm long, 13-18 cm broad, apical segment replaced by a 
proliferous bud; petiole with 1-2 nodes, moderately to abundantly scaly; rachis 
narrowly alate; pinnae 10-12 pairs, mostly short-stalked, oblong to (more com- 
monly) broadly oblanceolate, larger ones 7-11 cm long and 2-2.8 cm broad, 
inequilateral at base, narrow and rounded to cordate basiscopically, broader and 
cuneate acroscopically, terminating abruptly in an acuminate and serrate apex, 
abaxial surface amply provided with minute, dark brown scales; veins commonly 
simple, but sometimes paired at origin or forked. Fertile pinnae 12-14 pairs, 
larger ones 7-8 cm long and 0.8-1 cm broad, short-stalked, the apex obtuse. 

Terrestrial in dense forests, 0-500 m, thus far known only from Peru: Depts. 
Pasco and Ucayali. 

Additional collections: PERU. Pasco: Oxapampa, Palcazu Valley, Iscozacin, 
R. Foster 9466 (MO), 10049 (F). Ucayali: Vicinity of Aguaytia, Croat 20938 (MO). 

This is perhaps most closely related to D. alata Sm., of the West Indies and 
Venezuela, especially in that both species have predominantly simple veins. 
However, in D. oblanceolata pinnae are fewer and relatively shorter and broad- 
er, and most of them are broadest well above the middle, where the margins 
then curve abruptly to a short-acuminate apex. In D. alata, as in all members of 
the D. moritziana complex, pinnae are broadest at or near the middle, from 
whence they taper gradually to a moderately acuminate or attenuate apex. Dan- 
aea moritziana Presl, found rather frequently in Venezuela, Colombia and Peru, 
is further distinguished from D. oblanceolata in the predominantly forked veins 
(only occasionally simple or paired at base). Moreover, it is possible that the 
Central American D. cuspidata and one or two West Indian species are synon- 
ymous with D. moritziana. A number of taxa with pinnae under 2.5 cm broad 
were separated by Underwood (Bull. Torrey Bot. Club 29:669-679. 1902), merely 
on the degree of forking and spacing of veins, an apparently inconsistent char- 


om . 
SA ] 
eth 
Sr 


cami 


Fics. 1-3. Danaea oblanceolata. Fic. 1. Habit, sterile leaf, showing apical proliferous bud. Fic. 2. 
Portion of rachis and three fertile pinnae. Fic. 3. Section of sterile pinna, abaxial side, with predom- 
inantly simple veins, amorphous scales on costa and veins. 


R. G, STOLZE: DANAEA 35 


acter correlated rarely or not at all by other features; hence, a comprehensive 
revision of the genus is needed to clarify the taxonomy. 

In all the specimens of D. oblanceolata thus far examined, a proliferous bud 
has been found at the apex of the sterile leaves. This character has been ob- 
served in three of the other five species of the genus in Peru; it is frequently 
found in D. humilis Moore and D. trichomanoides Moore, and occasionally in 
D. moritziana. Apical proliferations are also found in several other West Indian 
and Central American species. Although this is an interesting feature, it is usually 
not a diagnostic one, for in most species with which I am familiar, it is not fully 
constant. 


REVIEWS 


‘“‘A monograph of the fern genus Pyrrosia (Polypodiaceae),” by P. Hovenkamp. 
1986. xiii + pp. 1-280 including 6 pp. of photos and 37 figs. “The Pyrrosia species 
formerly referred to Drymoglossum and Saxiglossum (Filicales, Polypodi- 
aceae),” by W. J. Ravensberg and E. Hennipman. 1986. pp. 281-310, 4 figs. Leiden 
Botanical Series, Vol. 9. Available from E. J. Brill, P.O. Box 9000, 2300 PA Leiden, 
Netherlands, 120 guilders, approx. $53.33. ISBN 90-04-08065-1. 

Pyrrosia is among the most abundant of Old World epiphytes in both individ- 
uals and species. In his 1947 Genera Filicum, Copeland estimated 100 Pyrrosia 
species without including Drymoglossum, a number now reduced to a conser- 
vative 51 by Hovenkamp, Ravensberg, and Hennipman. 

Although combined in a single volume, the monograph has two sections with 
different authors; the pagination is continuous, and all species are included in 
one key, but the indices to collections and taxonomic names are not integrated. 
The much smaller second portion by Ravensberg and Hennipman treats six 
species of four different affinities, grouped only because the six were often re- 
ferred to Drymoglossum and Saxiglossum, both reduced to Pyrrosia. The main 
treatment dealing with 45 species, by Hovenkamp, also has detailed sections on 
morphology, phylogeny, and biogeography. Altogether it is a very thorough and 
impressive accumulation of information. The nomenclature, descriptions, and 
distributions are authoritative and very useful. 

However, the chapter on phylogeny was not easy for me to understand, and 
I found the premise, that Pyrrosia and Platycerium are sister genera, unlikely. 
Such would require that an immediate ancestor of Platycerium was also the 
ancestor of all extant Pyrrosia. A more attractive hypothesis is that Pyrrosia is 
considerably older than the highly specialized Platycerium, and had already 
diversified into plural species we would unhesitatingly classify as Pyrrosia if 
extant today, and that one of these early Pyrrosia species was the source of 
Platycerium. 

Perhaps the sister genus misconception has led to further misconceptions. I 
believe the immediate ancestor of Pyrrosia was very similar to other polypods, 
sharing a creeping rhizome with internal sclerenchyma strands, peltately at- 


36 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 1 (1987) 


tached paleae, fronds articulate to phyllopodia, veins anastomosing with regular 
areoles, hydathodes present, etc. But instead of allowing for a common origin 
with other Polypodiaceae, Hovenkamp (pp. 100-102) presents a case for a rela- 
tionship of Pyrrosia with Dipteris, a genus with a radically different ground- 
plan and with only a very remote affinity, at best, to the polypods. 

On the basal branch of the Pyrrosia cladogram (p. 95) is the P. africana group 
consisting of two African species that completely lack rhizome sclerenchyma, 
have +irregular venation, pseudopeltate paleae, vestigially articulate fronds, no 
woolly rays of stellate hairs, a total absence of hydathodes in one, the other with 
a unique modification of the annulus. These two species appear to me to be 
reduced derivatives of the P. porosa group which they closely resemble in over- 
all morphology and which contains two other African species with reduction 
characters transitional to these in rhizome sclerification and laminar indument. 

A further argument against the P. africana group being relictual in Pyrrosia 
is phytogeographical. The genus would probably have reached tropical America 
if, as hypothesized, it had an early origin in what is now Africa; by far the 
greatest diversity in the genus is now in the eastern Himalayas, southeast Asia, 
and Malesia. 

However, it is only because Hovenkamp gives us so very much information 
that it is easy to debate with him about the phylogeny of his genus. This is a 
positive, not a negative aspect, and not a detraction from the great value of the 
work as a whole.—M. G. Price, Herbarium, North University Building, Univer- 
sity of Michigan, Ann Arbor, MI 48109. 


“Index of Thelypteridaceae,” by J. W. Grimes and B. S. Parris. 1986. iv + 50 
pp. Royal Botanic Gardens, Kew. Available from B. Parris, Royal Botanic Gar- 
dens, Kew, Richmond, Surrey TW9 3AE, England. £7.25 (incl. postage and han- 
dling). ISBN 0 947643 03 6. 

This is an alphabetical listing by specific epithet (accepted names and syn- 
onyms) of all species of Thelypteris s.]., with citation of original genus and current 
disposition of the name as recognized by authorities in the group. There is also 
a selected bibliography of 65 references to major literature concerning the family. 
No new combinations are included. The list will be useful primarily to curators 
of herbaria who wish to arrange their specimens according to recent reclassifi- 
cations, primarily by Holttum, and also as a reference for specialists and floris- 
ticians working with the family. I found the work to be relatively free of errors, 
but a brief search uncovered several overlooked basionyms and incorrect attri- 
butions of current names.—ALAN R. SmitH, Department of Botany, University of 
California, Berkeley, CA 94720. 


INFORMATION FOR AUTHORS 


Authors are encouraged to submit manuscripts pertinent to pteridology for 
publication 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 pre- 
pared may be returned for revision prior to review. Submit manuscripts in trip- 
licate (xerocopies acceptable), including review copies of illustrations. Do not 
send originals of illustrations until they are requested. Use standard 81% by 11 
inch paper of good quality, not “erasable” paper. Double space manuscri) 
throughout, including title, authors’ names and addresses, 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. 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 mea- 
sures (e.g., distance, elevation, weight) unless ‘quoted or cited from another source 
(e.g., specimen citations}. For nomenclatural matter {i.e., synonymy and typifi- 
cation}, use one paragraph per basionym (see Regnum Veg. 58:39-40. 1968). Ab- 
breviate titles of serial publications according to Botanico-Periodicum-Huntian- 
um (Lawrence, G. H. M. et al., 1968, Pittsburgh: Hunt Botanical Library). 
References cited only as part of nomenclatural matter are not included in lit- 
erature cited. For shorter notes and reviews, put all references parenthetically 
in text. Use Index herbariorum (Regnum Veg. 106:1-452. 1981) for piu acesnaees 
of herbaria. | ee 

Illustrations should be proportioned to fit page width ith caption on the > same : 

page. Provide margins of ‘at least 25 mm on all trati Fe 
soaker ai design originals for reproduction without reduction « or ey a uniform : 
amount. In composite blocks, abut edges of adjacent photographs. Avoid com- 
opy in single illustrations or blocks. Coordinate a 


sequence and numbering of figures (and of tab oo 
Explain scales and symbols in figures themselves, not i in n captions. Include — 
scale and ref i and longitude in each map. a re 
Proofs and reprint order forms are sent to authors by the platen. Authors 


should send corrected proofs to the editor and reprint orders to the printer. © : 


Authors will be assessed omens for extensive aliqrations made wep dane — : 
been set. ae 
For other matters val on or ee oonk recent issues of American Fern ae 
_ Journal and The Chicago manual of style, 18th ed. (1982. Chicago: Univ. Chicago 
Dr, from these 5 . Aut 
atte ~ 


AMERICAN 
FERN — 


Volume 77 


April-June 1987 


JOURNAL 


QUARTERLY JOURNAL OF THE AMERICAN FERN SOCIETY 


Argyrochosma, a New Genus of Cheilanthoid Ferns Michael D. Windham 37 
Electrophoretic and Morphological pa eT of Interspecific Hybridization between 
Polystichum siggy eae and P. m 
amela S. Soltis, D baie E. Soltis, and Edward R. Alverson 42 
Growth Patterns of Gemmlings of Lycopodium lucidulum Ulrike Reutter 50 


The Expanded Adaxial Epidermis of Equisetum Rhizome Sheath Teeth 
Richard L. Hauke 58 


Schizaea pusilla Discovered in Peru Robert G. Stolze 
Shorter Notes 
Numbers of Some Ferns Argentina 
Elias R. de la Sota, Marta Ménica Ponce, and Liliana A. Cassa de Pazos 56 
Botrychium pinnatum in Colorado Peter G. Root and James D. Montgomery 68 : 


ee aa Chiapas, Mexico 
Ram6n Riba, Leticia Pacheco, and Esteban Martinez Ss. 


Additions to the Fern Flora of the Bahamas oe Clifton E Nauman es 
Reviews 48, 65 
Announcement: 1988 AIBS Meeting 72 


The American Fern Society 
Council for 1987 
FLORENCE S. WAGNER, Dept. of Botany, University of Michigan, Ann Arbor, MI 48109. 
President 
JUDITH E. SKOG, Biology Dept., George Mason University, footers — 22030. Vice-President 
W. CARL TAYLOR, Milwaukee Public Museum, Milwaukee, WI 5. Secretary 
JAMES D. CAPONETTI, Dept. of Botany, University of Tennessee, =i ee ™N agg Siigeccr: 
DAVID S. BARRINGTON, Dept. of Botany, University of Vermont, Burlington, che 
oe Treancrex 


JAMES D. MONTGOMERY, Ecology III, R.D. 1, Berwick, PA 1860: A ck Issues Curator 
ALAN R. SMITH, Dept. of Botany, University of California, bake, CA 94720. Journal Editor 
VID B. LELLINGER, Smithsonian Institution, Washington Memoir Editor 


DENNIS Wm. STEVENSON, New York Botanical Garden, ae pone 104 
=e Forum Editor 


American Fern Journal 
EDITOR 
Ore Dept. of Botany, University of California, 
Berkeley, CA 94720 
GERALD] GASTONY Dept. of Biology, Indiana University, Bloomington, IN 47401 
SAREE PAUELER 0 Dept. of Botany, University of Kansas, 
Lawrence, KS boa 
DAVID BD LELLINGER 2-2 U.S. Nat'l Herbarium NHB-166, reer Insti 


ashington, pea ae 
_ TERRY R. WEBSTER | . Biological Sciences Group, University of ie Storrs, CT 06268 


_ The “American Fern Journal” (ISSN a. is an illustrated — devoted to the general 
study of ferns. It is owned by the American Fern Society, and published the Pringle Herbarium, 
University of heen Burlington, VT 05405. Second-class postage ar at Burlington, VT, and 
__ additional entry poi 
ousecn: for missing os made e months (domestic) to 12 months (foreign) after the date of issue, 


d orders for bi dressed to Dr. James D. Montgomery, Ecology III, R.D. 1, 


; rae ie ae we. — 
dues, and applications f bership should be sent to the Records Trea- 
oS 4 £. L 1) , pe | J Pe 


eral inquiries 
a ‘Sisters $12.00 -~ $11.50 net if paid through an an agency (uawiey fee $0.50); sent free to 
: “members of, f the gan encerin Te pest annual does $10.00 + $8.00 mailing surcharge beyond 
oe as volu : . 1 7 L of 64 
: cefckore se es chmeene ack volumes 1979 et seq, $8.00 cach: 
single back numbers $2.00 each, plus shi pping on orders of six volumes or — 


mi 3 Ls Dept. of Botany, University of . 
_idtonead Forum 


Se ecient spa : 


: Soha a Director. Sores exchanged a 
ailable spores sent on request. _ . ’ ce 


DEC 29 1987 


American Fern Journal 77(2):37-41 (1987) 
GARDEN LIBRARY, 
Argyrochosma, a New Genus of Cheilanthoid Ferns 


MICHAEL D. WINDHAM 
Department of Botany, University of Kansas, Lawrence, KS 66045 


The group of fern species related to Notholaena nivea (Poir.) Desv. has long 
been a source of taxonomic contention. In the 183 years since N. nivea was first 
described (as Pteris nivea Poir.), members of this alliance have been assigned to 
no fewer than ten genera, including Acrostichum, Cheilanthes, Gymnogramma, 
and several others no longer in use. Detailed morphological and anatomical 
investigations leave no doubt that these species are cheilanthoid ferns in the 
strict sense (Tryon & Tryon, 1982), and most recent authors place this group in 
either Notholaena or Pellaea. Rolla Tryon, Maxon, and others who favor place- 
ment in Notholaena (as typified by N. trichomanoides) emphasize two characters 
that seem to indicate an affinity to that genus: 1) the presence of a farinose 
indument on the abaxial leaf surface in most species and 2) the absence of a 
pseudoindusium formed by a modified leaf margin. Those who classify the N. 
nivea complex with Pellaea (i.e., Prantl, Christensen, and Morton) stress simi- 
larities in spore ornamentation, sporangial distribution and leaf architecture. 
Despite disagreement over the relative importance of different morphological 
traits, both groups of pteridologists seem to recognize the N. nivea complex as a 
natural alliance, and the sectional name Argyrochosma (proposed by J. Smith in 
1841) has often been applied to it regardless of generic assignment. 

Copeland and Weatherby recommended a different approach to the taxonomic 
problems raised by the N. nivea group. In Genera Filicum, Copeland (1947, p. 
70) stated that “The group placed under Pellaea in Christensen’s Index, p. XL, 
as sect. Argyrochosma, typified by P. nivea (Poir.) Prantl, has no proper place 
in the genus. .. . It seems to be a proper generic entity, without a name as such.” 
In a letter to Morton dated March, 1949 (quoted in Morton, 1950, pp. 249-250), 
Weatherby was even more specific. He stated that “this is one of the two groups 
(and the better of the two) which I can see clearly as a segregate genus. If to the 
group of N. nivea, N. dealbata, N. fendleri, et cetera, you add N. Jonesii, N. 
Lumholtzii, Pellaea microphylla, and P. formosa, you get a coherent and, I think, 
natural group, which as a genus, should bear the name Argyrochosma {J. Smith).” 
Unfortunately, neither Copeland nor Weatherby lived to complete his study, and 
subsequent authors have not followed their recommendations. 

Ongoing biosystematic investigations of Pellaea, Notholaena, and the N. nivea 
complex (Windham, unpubl.) serve to reinforce the idea that Argyrochosma is 
a natural (monophyletic) group worthy of generic recognition. Tryon and Tryon 
(1982) pointed out that sections Argyrochosma and N otholaena (the latter typified 
by N. trichomanoides) are distinct in terms of rhizome scales, leaf architecture, 
sporangial distribution, and spore morphology. These taxa also show consistent 
differences in chromosome number, gametophyte morphology, chemical com- 
position of the farinose indument, and patterns of variability at conservative 
enzyme loci (Windham, 1986). Each of these characters indicates a close rela- 


38 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 2 (1987) 


tionship between Argyrochosma and Pellaea sect. Pellaea, suggesting that these 
may be sister groups in a phylogenetic sense. However, the species of Argyro- 
chosma exhibit a chromosome base number (X = 27) unique among cheilanthoid 
ferns, and the two groups are easily distinguished using a combination of mor- 
phological features including rhizome scales, segment size and leaf dissection, 
nature of the leaf margin, and the occurrence of a farinose indument. The unique 
chromosome base number of Argyrochosma is here interpreted as a synapo- 
morphy supporting the monophyletic origin of the group, and I concur with 
Copeland and Weatherby that Argyrochosma should be accorded generic rank 
among the cheilanthoid ferns (included in the family Adiantaceae). 

My comparative study of Pellaea, Notholaena, and Argyrochosma is nearing 
completion, but this detailed work is unlikely to be published prior to deadlines 
established for several new floristic manuals. Therefore, to facilitate the treatment 
of Argyrochosma in the Flora of North America, Flora of California, and The 
ferns and fern allies of Arizona (Windham & Yatskievych, in prep.), I herewith 
describe the genus and make nomenclatural combinations for the sixteen species 
currently included in it. 


Argyrochosma (J. Smith) Windham, stat. nov.—Notholaena sect. Argyrochosma ~ | 


J. Smith, J. Bot. (Hooker) 4:50. 1841.—Lectotype (chosen by Christensen, 1906, 
Ind. Fil., p. XL): Pteris nivea Poir. [Argyrochosma nivea (Poir.) Windham]. 


Rhizome compact, short-creeping, erect to more or less decumbent, bearing 
scales and many fibrous roots. Rhizome scales thin, light brown to reddish brown, 
concolorous, up to 1 cm long, linear to lanceolate with an acuminate tip, entire 
to minutely denticulate. Leaves monomorphic, clustered, up to 40 cm long. Petiole 
terete to very shallowly sulcate, shorter than to slightly longer than the lamina, 
with a single vascular bundle: castaneous, atropurpureous, or blackish in color; 
glabrous, glaucous, sparsely ceraceous, or bearing small widely scattered scales 
or trichomes. Lamina imparipinnate (Fig. 1), linear-lanceolate, ovate or deltate; 
bipinnate to pentapinnate at the base, with up to 15 pairs of subopposite or 
alternate pinnae; coriaceous or (rarely) herbaceous, the upper surface glabrous 
or sparsely ceraceous, the lower glabrous or usually densely white (rarely yellow) 
ceraceous. Rachis similar to the petiole but occasionally flexuous. Pinnae lan- 
ceolate to deltate, generally remote, divided into numerous ultimate segments. 
Segments small, oblong to roundish or cordate, entire to shallowly lobed, pet- 


sporangia for much of their length (Fig. 2). Sporangia with 64 or 32 spores. Spores 

trilete, light to dark brown, with cristate (Fig. 3) or rugose surfaces. Gametophytes 

usually cordate with wide notches, symmetrical and glabrous (lacking farina- 
. 4). 


Distribution.—A strictly American genus of approximately 20 species occu- 
pying rupestral or (rarely) terrestrial habitats from near sea level to an elevation 
of 4200 m in the Andes. Ranging from Missouri, Wyoming, and California to 
Chile (including the Juan Fernandez Islands), Argentina, and the highlands of 
southeastern Brazil. There is a large geographic gap between the North and South 


M. D. WINDHAM: ARGYROCHOSMA 39 


% A the ‘ i’ 
“i Apa, 
al . 


Pe a. 
if 


Fics. 1-4. Characteristics of the genus Argyrochosma. 1. Whole plant of Argyrochosma nivea (Cor- 
rell & Smith P743, GH) showing habit and imparipinnate leaf architecture, x ¥. 2. Leaf from the 

individual sl ing S} gi listributed al gtk veins for much of their length, x 2.3. Typical 
cristate spore of A. formosa (Windham et al. 551, KANU}, x 750. 4. Meiotic chromosome squash of 
A. delicatula (Windham et al. 482, KANU) showing 27II, x 1000. 


American elements of the genus (only A. incana is found in Central America 
and the West Indies}, with the greatest diversity of species occurring in the 
highlands of central and northern Mexico. 


ENUMERATION OF SPECIES (AND Major SYNONYMS} 


1) Argyrochosma chilensis (Fee « Remy) Windham, comb. nov.—Cincinalis 
chilensis Fee & Remy in Gay, Hist. Chile (Bot.) 6:497. 1853.—Notholaena 
chilensis (Fee & Remy) Sturm—Pellaea chilensis (Fee & Remy) C. Chr. 


40 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 2 (1987) 


| 2) Argyrochosma dealbata (Pursh) Windham, comb. nov.—Cheilanthes deal- 
bata Pursh, Fl. Amer. Sept. 2:671. 1814.—Notholaena dealbata (Pursh) Kunze— 
Pellaea dealbata (Pursh) Prantl. 


Although this diploid taxon shows a strong morphological resemblance to the 
agamosporous triploid A. limitanea (Tryon, 1956), isozyme analyses indicate that 
the species, in its present form, was not involved in the origin(s) of that polyploid. 


3) Argyrochosma delicatula (Maxon & Weath.) Windham, comb. nov.—No- 
tholaena delicatula Maxon & Weath., Contr. Gray Herb. 127:7. 1939. 


Preliminary isozyme data indicate that this is quite distinct from A. incana and 
should be maintained as a separate species despite the existence of a few inter- 
mediate individuals discussed by Maxon and Weatherby (1939) and Wollenweber 
(1984). 


4) Argyrochosma fendleri (Kunze) Windham, comb. nov.—Notholaena fen- 
dleri Kunze, Farnkr. 2:87, t. 136. 1851.—Pellaea fendleri (Kunze) Prantl 


5) Argyrochosma formosa (Liebm.} Windham, comb. nov.—Allosorus formosus 
Liebm., Kongel. Danske Vidensk. Selsk. Skr., Naturvidensk. Afd., V. 1:220 
(Mex. Bregn. 68). 1849.—Notholaena formosa (Liebm.) R. Tryon—Pellaea 
formosa (Liebm.} Maxon. 


6) Argyrochosma incana (Presl) Windham, comb. nov.—Notholaena incana 
Presl, Rel. Haenk. 1:19, t. 1, fig. 2. 1825. 


Variations in flavonoid chemistry (Wollenweber, 1984) and spore size (Wind- 
ham, unpubl.) within this widespread taxon suggest that it may include at least 
two distinct species. 


7) Argyrochosma jonesii (Maxon) Windham, comb. nov.—Notholaena jonesii 
Maxon, Amer. Fern J. 7:108. 1917.—Pellaea jonesii (Maxon) Morton. 


As currently defined, this species includes diploid and tetraploid cytotypes 
which tend to occupy different portions of the range. The type specimen, collected 
in Inyo County, California, probably represents the tetraploid. 


a limitanea (Maxon) Windham, comb. nov.—Notholaena lim- 
itanea Maxon, Amer. Fern J. 9:70. 1919.—Pellaea limitanea (Maxon) Morton. 


Both the typical form and var. mexicana (Maxon) Broun are agamosporous 
triploids which probably arose through different polyploidization events (Wind- 
ham, unpubl.). If and when the diploid progenitors are identified, it may be 
necessary to recognize these varieties as distinct species. 


9) Argyrochosma lumholtzii (Maxon & Weath.) Windham, comb. nov.—No- 
tholaena lumholtzii Maxon & Weath., Contr. Gray Herb. 127:16. 1939. 


10) Argyrochosma microphylla (Mett. ex Kuhn) Windham, comb. nov.—Pellaea 
Sec Mett. ex Kuhn, Linnaea 36:86. 1869.—Notholaena parvifolia R. 
ryon. 


M. D. WINDHAM: ARGYROCHOSMA 41 


11) Argyrochosma nivea (Poir.} Windham, comb. nov.—Pteris nivea Poir., En- 
cycl. 5:718. 1804.— Notholaena nivea (Poir.) Desv.—Pellaea nivea (Poir.) Prantl. 


Although the typical form and var. tenera (Hook.} Griseb. both reproduce by 
means of agamospory, recent chromosome counts indicate that var. nivea is 
triploid while at least some individuals of var. tenera are diploid (Windham, 
unpubl.). The status of all four infraspecific taxa comprising A. nivea will need 
to be reassessed when additional genetic data become available. 


12) Argyrochosma pallens (Weath. in R. Tryon) Windham, comb. nov.—No- 
tholaena pallens Weath. in R. Tryon, Contr. Gray Herb. 179:78. 1956. 


13) Argyrochosma palmeri (Baker) Windham, comb. nov.—Notholaena palmeri 
Baker, Hook. Icon. Pl. 17: t. 1678 & text. 1887. 


14) Argyrochosma peninsularis (Maxon & Weath.) Windham, comb. nov.— 
Notholaena peninsularis Maxon & Weath., Contr. Gray Herb. 127:15. 1939. 


15) Argyrochosma pilifera (R. Tryon) Windham, comb. nov.—Notholaena pili- 
fera R. Tryon, Contr. Gray Herb. 179:79. 1956. 


16) Argyrochosma stuebeliana (Hieron.) Windham, comb. nov.—Pellaea deal- 
bata var. stuebeliana Hieron., Hedwigia 48:225, t. 12, fig. 15. 1909.— Notho- 
laena stuebeliana (Hieron.) R. Tryon 


ACKNOWLEDGMENTS 


I thank Dr. Christopher Haufl d e, Theresa, for their hel d t tl hout 
my work on Argyrochosma. Financial sietslite from the University of Kansas ‘General Research 
Fund and the Pteridological Section of the Botanical Society of America (Edgar T. Wherry Award) 
is gratefully acknowledged. 


LITERATURE CITED 


CopeLaNp, E. B. 1947. Genera Filicum. Waltham, Mass.: Chronica Botanica Co. 

Maxon, y R. and C. ie WEATHERBY. 1939. Some species of Notholaena, new and old. Contr. Gray 
erb. 127:3-1 

Morton, . V. 1950. tsike on the ferns of the eastern United States (concluded). Amer. Fern J. 40: 


-252. 
TRYON, : ae 1956. A revision of the American species of Notholaena. Contr. Gray Herb. 179:1- 


and A. F. Trron. 1982. F d allied plants, with special ref to tropical America 
New York: Springer-Verlag. 
WINpDHAM, M. D. 1986. Reassessment of the phylogenetic relationships of Notholaena. Amer. J. Bot. 


73:742 (atietiars) 
WoLLENWEBER, E. 1984. Exudate flavonoids of Mexican ferns as chemotaxonomic markers. Rev. 
tinoamer. Quim. 15-1:3-11. 


American Fern Journal 77(2):42-49 (1987) 


Electrophoretic and Morphological Confirmation 
of Interspecific Hybridization between 
Polystichum kruckebergii and 
P. munitum 


PAMELA 8. SOLTIs and Douctas E. Soutis 
Department of Botany, Washington State University, Pullman, WA 99164-4230 
EDWARD R. ALVERSON 
Department of Botany and Plant Pathology, Oregon State University, Corvallis, OR 97331 


The genus Polystichum (Dryopteridaceae) comprises from 160 (Tryon & Tryon, 
1982) to 175 (Copeland, 1947) species and is nearly worldwide in distribution. 
Substantial morphological diversity exists within Polystichum (Barrington, 1985), 
and considerable taxonomic confusion has characterized Polystichum since its 
description in 1799. Systematic problems in Polystichum stem largely from ex- 
tensive hybridization and allopolyploidy (Knobloch, 1976; Tryon & Tryon, 1982; 
Mickel in Barrington, 1985; D. Wagner in Barrington, 1985; W. Wagner in Bar- 
rington, 1985), which tend to obscure species boundaries and make morphological 
comparisons difficult. Furthermore, the production of interspecific sterile hybrids 
has created additional taxonomic confusion (Manton, 1950; Nakaike, 1973; W. 
Wagner, 1973; Daigobo, 1974; D. Wagner, 1979). 

A prime example of the reticulate evolution typical of the genus (D. Wagner, 
1979, in Barrington, 1985; W. Wagner in Barrington, 1985) is found in the Polys- 
tichum complex from western North America. This group consists of five diploid 
species, three once-pinnate (P. imbricans, P. lonchitis, and P. munitum) and two 
highly dissected (P. dudleyi and P. lemmonii). There are five tetraploids, P. 
andersonii, P. braunii, P. californicum, P. kruckebergii, and P. scopulinum; all 
but P. braunii incorporate various combinations of the diploid genomes discussed 
above (W. Wagner, 1973: D. Wagner, 1979). All of these species are restricted to 
western North America except P. braunii and P. lonchitis, which are circumboreal 
in distribution, and P. scopulinum, which also occurs sporadically in eastern 
North America. [Following D. Wagner (1979), P. lemmonii is considered distinct 
from the South American P. mohrioides.] Polystichum setigerum 


monii X scopulinum, and P. munitum x scopulinum (W. Wagner, 1973), and P. 
braunii x lonchitis in Europe (Sleep & Reichstein, 1967). In addition, P. braunii 
has hybridized with the eastern North American P. acrostichoides (Thompson 
& Coffin, 1940; Barrington, 1986), and hybrids between P. lonchitis and P. ac- 


SOLTIS ET AL.: POLYSTICHUM HYBRID 43 


rostichoides have also been reported (W. Wagner & Hagenah, 1954). Furthermore, 
both P. braunii and P. lonchitis have hybridized with the European species P. 
aculeatum and P. setiferum (reviewed by Sleep & Reichstein, 1967). In this paper 
we provide morphological and electrophoretic evidence for hybridization be- 
tween P. munitum, a diploid, and P. kruckebergii, an allotetraploid whose pre- 
sumed diploid progenitors are P. lemmonii and P. lonchitis (W. Wagner, 1973). 

The putative hybrid plants under study are readily recognized as unusual; they 
possess moderately incised pinnae, which in nature are twisted relative to the 
axis of the rachis. The plants are medium-sized, with leaf blades of mature plants 
ranging from 35-42 cm by 5-6 cm; the plants are generally larger than the small, 
rock-dwelling polystichums, such as P. kruckebergii, but smaller than the large, 
forest-dwelling taxa, such as P. munitum. Close examination of the sori revealed 
abortive sporangia, as well as irregularly sized and shaped spores; many of the 
spores were shriveled and presumably abortive. Thus, the plants displayed fea- 
tures characteristic of interspecific hybrids (W. Wagner, 1968). 

The putative hybrid between P. munitum and P. kruckebergii originated under 
an unusual set of geological circumstances. Typically, P. munitum and P. krucke- 
bergii are not sympatric; P. munitum usually inhabits moist woods at lower 
elevations, and P. kruckebergii usually occurs at higher elevations, often on 
ultramafic substrates. The putative hybrids occur with P. kruckebergii and P. 
munitum at an elevation of approximately 1065 m on the lower slopes of Devil's 
Thumb in the Coal Creek drainage on the west side of the Cascades in Snohomish 
County, Washington. The plants grow at the base of a large glacial erratic of 
peridotite (an ultramafic mineral) that originated higher up on the mountain slope. 
This population of P. kruckebergii and the putative hybrids occur in an Abies 
amabilis-Tsuga heterophylla forest, at a much lower elevation than is typical of 
P. kruckebergii. This habitat and elevation are more typical of P. munitum in 
this region. Polystichum lonchitis was also present on a non-ultramafic outcrop 
100 m away from the hybrid site and, instead of P. munitum, could conceivably 
be the once-pinnate parent of the putative hybrids. Polystichum lonchitis was 
therefore included in this study. 

The origin of an interspecific hybrid can be extremely difficult to determine 
with certainty when only morphological criteria are employed, particularly in a 
plant group known for its phenotypic plasticity and environmentally induced 
variability (Barrington, 1985). Therefore, we utilized genetic markers in addition 
to morphological analyses to determine the origin of the putative hybrids. 


MATERIALS AND METHODS 


Morphological data were obtained from pressed fronds of P. munitum, P. 
kruckebergii, and the putative hybrids collected at the hybrid locality. Vouchers 
were deposited at WS. es 

We electrophoretically examined the five putative hybrid plants, five individ- 
uals of P. munitum, and four individuals each of P. kruckebergii and P. lonchitis 
from the hybrid locality. Leaf material was collected in the field and stored in 


44 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 2 (1987) 


plastic bags with wet paper towels under refrigeration until electrophoresis was 
conducted. 

Electrophoretic procedures generally followed those of D. Soltis et al. (1983). 
Leaf tissue was prepared using the tris-HCI grinding buffer-PVP solution of D. 
Soltis et al. (1983); three grams of PVP were used per 25 ml of grinding buffer. 
Starch gel concentration was 12.5%. 

The following enzymes were examined: aspartate aminotransferase (AAT), 
fluorescent esterase (FE), leucine aminopeptidase (LAP), malate dehydrogenase 
(MDH), phosphoglucoisomerase (PGI), phosphoglucomutase (PGM), shikimate 
dehydrogenase (SkDH), and triosephosphate isomerase (TPI). AAT, FE, LAP, 
PGI, and TPI were resolved on a modification of gel and electrode buffer system 
8 of Soltis et al. (1983); the gel buffer was composed of 0.033 M tris, 0.005 M citric 
acid, 0.004 M lithium hydroxide, 0.030 M boric acid, PH 7.6, and the electrode 
buffer was composed of 0.039 M lithium hydroxide, 0.263 M boric acid, pH 8.0. 
MDH, PGM, and SkDH were resolved on gel and electrode buffer system 9 of 
D. Soltis et al. (1983). Staining for all enzymes followed D. Soltis et al. (1983), 
except LAP, which followed Soltis and Rieseberg (1986). 

The genetic control of electrophoretic banding patterns was readily interpreted 
based on the known subunit structure and subcellular localization of the enzymes 
(Gottlieb, 1981, 1982). Loci were numbered sequentially, with the most anodally 
migrating locus designated 1, the second most anodal locus 2, and so on. Similarly, 
allozymes were denoted alphabetically, with the fastest migrating allozyme des- 
ignated a, the second fastest allozyme b, and so on. 


RESULTS 


Morphology.—Table 1 compares P. kruckebergii, P. munitum, and the putative 
hybrids for 11 distinguishing morphological features. In many cases, the putative 


pinnae from the plane of the rachis were all closer to the condition of P. krucke- 


non-costal infralaminar scales are noteworthy in that scales representing both of 
the parental phenotypes, as well as intermediate conditions, could be found on 
a single pinna. 

Spores are one feature in which hybrid intermediacy was not consistently 
expressed. Polystichum munitum has small, yellowish, translucent spores that 
average 33 um in length. Spores of the tetraploid P. kruckebergii are much larger, 
averaging 42 um in length; they are dark brown in color and are only slightly 
translucent. Spores of both species are distinctly monolete and relatively uniform 
in size. In contrast, spores of the hybrid are generally spherical, when not shriv- 
eled and malformed. It is noteworthy, however, that many of the spores taken 
from the hybrid plants were large and well formed. These large, spherical spores 


SOLTIS ET AL.: POLYSTICHUM HYBRID 


TABLE 1. 


Morphological Characteristics of P. kruckebergii, P. munitum, and Their Hybrid. 


P. kruckebergii 


hybrid 


P. munitum 


1) Ratio of pinna 
2) Shape of lowest 
innae 
3) Pinna orientation in 
nature 


4) Dissection of pinnae 


5) Teeth of the pinna 
margins 

6) Frond texture 

7) Indusium margin 


8) Non-costal infra- 
laminar scales 


9) Spore shape 


10) Spore color 


11} Mean exospore 
length (S.D.) 


approximately 2:1 


triangular with 
nearly equilat- 
eral sides 

rotated relative to 
the axis of the 
rachis 

pinnae fully pinna- 
tifid, at least the 
first 1-3 lobes 


teeth tending to be 


preadi 
somewhat soft 
more or less entire 
nearly linear, most- 


ly with few, short 
basal projections 


monolete, uniform 


dark brown, some- 
what translucent 
43 um (5 um) 


approximately 3:1 


triangular to ovate- 
lanceolate 


rotated relative to 
the axis of the 


rachis 

pinnae dissected 
4-2/3 way to the 
costa 


teeth tending to be 
incurved 


erose, with occa- 
sional cilia 


projections 
spherical, or shrun- 
ken and irregu- 
lar, often vari- 
able in size and 
outline 
dark brown, 


opaque 
42 wm (8 um) 


approximately 6:1 
lanceolate 


plane relative to the 


axis of the rachis 


pinna margins 
toothed, some- 
times incised, but 
not dissec’ 

teeth mostly strongly 
incurved 

rm 
long ciliate 


lanceolate, with an 
id base, 
the basal projec- 
tions often as long 
as the body of the 
scale 


monolete, uniform 


yellowish-brown, 
sluce 


33 wm (2 pm) 


may be unreduced “mitospores” similar to those described by Morzenti (1962). 
Morphological characteristics of the putative hybrids and the circumstances un- 
der which the hybrids occur clearly suggest that these plants originated by hy- 
bridization between P. kruckebergii and P. munitum. 

Electrophoresis.—Eleven loci were interpreted: Aat, Fe-1, Lap, Mdh-1, Pgi-2, 
Pgm-1, Pgm-2, Skdh, Tpi-1, Tpi-2, and Tpi-3. The observed enzyme bands mi- 
grated anodally for all enzymes. 

To document hybridization, electrophoretic investigations of hybridization re- 
quire genetic markers differentiating the putative parental species. The three 
possible parental species, P. munitum, P. kruckebergii, and P. lonchitis, possess 
different genotypes for Fe-1, Lap, Pgm-1, Pgm-2, and Tpi-3 (Figs. 1-3). At Lap, 


46 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 2 (1987) 


Pe RRR ELL 1 ae LK KK KHH HA HM M 
Fics. 1-4, Starch gels in Polystichum. In all figures, numbers and letters in margins designate loci 


and alleles, respectively. K = Polystichum kruckebergii; M = P. munitum; L = P. lonchitis; H = 
hybrid. 1. LAP. 2. Interpretive drawing of Fig. 1. 3. TPI. 4. PGI. 


all individuals of P. kruckebergii possessed the fixed heterozygous genotype Lap- 
ac (Figs. 1, 2). Polystichum lonchitis exhibited only allele Lap-c, and plants of 
P. munitum displayed Lap-b, Lap-d, or both of these alleles (Figs. 1, 2). The five 
putative hybrids possessed one of the following genotypes: Lap-abc (three in- 
dividuals) or Lap-acd (two individuals), clearly combining the genotypes of P. 
munitum and P. kruckebergii (Figs. 1, 2). 

At Fe-1, three of the putative hybrids possessed allele Fe-1b, the allele found 
in all P. kruckebergii individuals examined. Another putative hybrid possessed 
the genotype Fe-1bc, combining allele b of P. kruckebergii and allele c which 
was found in all individuals of P. munitum from the hybrid site. One putative 
hybrid did not display FE activity. All individuals of P. lonchitis exhibited allele 
d. Data from Fe-1 therefore provide limited support for the hypothesis that the 


SOLTIS ET AL.: POLYSTICHUM HYBRID 47 


the parentage of the hybrids. Polystichum kruckebergii can be distinguished 
rom P. munitum and P. lonchitis at Tpi-1 and Tpi-2, but the latter two species 
cannot be differentiated (Fig. 3). The tetraploid P. kruckebergii exhibited fixed 
heterozygosity at Tpi-3 (Fig. 3). In addition, a five-banded pattern was observed 
in the more anodal zone of activity for TPI (Fig. 3). However, it is unknown 
whether this represents fixed heterozygosity at Tpi-1 or Tpi-2 or both of these 
loci. The putative hybrid could not be distinguished from P. kruckebergii at any 
of the TPI loci (Fig. 3). 

The loci Aat, Mdh-1, and Skdh were monomorphic for all plants examined. 
Although the three possible progenitors were differentiated at Pgm-1 and Pgm- 
2, the banding patterns were not clearly resolved in the putative hybrids. 


DISCUSSION 


Th phological and allozymic data presented herein confirm the occurrence 
of hybridization between Polystichum munitum and P. kruckebergii. The pu- 
tative hybrids are morphologically intermediate between the proposed parental 
species for several characters. Furthermore, the hybrids clearly combine the 
alleles of P. munitum and P. kruckebergii at Lap. The data obtained for all other 
loci examined are also consistent with this hypothesis of hybridization. The data 
for Lap and Fe-1 rule out the possibility that P. lonchitis is one of the parental 
species. 

In the five hybrid individuals examined we detected three different genotypes 
across all loci. This indicates that hybridization between different P. munitum 
and P. kruckebergii individuals occurred at least three times in this mixed pop- 
ulation; the several hybrid individuals were not the result of vegetative repro- 
duction following a single hybridization event. 

Because the tetraploid P. kruckebergii is thought to have P. lemmonii and P. 
lonchitis as its diploid progenitors (W. Wagner, 1973; D. Wagner, 1979}, the ge- 
nomic composition of the hybrids presumably comprises three distinct genomes: 
P. munitum, P. lonchitis, and P. lemmonii. This combination could also result 
from a hybrid between P. lonchitis and P. scopulinum, provided that at least 
some populations of P. scopulinum have the parentage P. lemmonii x munitum 
8. s., as proposed by W. Wagner (1973). More recently, however, D. Wagner (1979) 
suggested that the once-pinnate diploid progenitor of P. scopulinum was P. im- 
bricans rather than P. munitum. Further allozymic investigations should provide 
additional information regarding evolutionary relationships in this species com- 
plex. 

The detection of yet another interspecific hybrid in Polystichum is not sur- 
prising, particularly given the widespread occurrence of hybridization within the 
genus (Knobloch, 1976; Tryon & Tryon, 1982; Mickel in Barrington, 1985; W. 
Wagner in Barrington, 1985). It is noteworthy that five of the seven interspecific 
Polystichum hybrids previously reported from western North America involve 
P. munitum as a parental species. This may be due to the more widespread 
distribution of this species, relative to other western North American Polystichum 
species. However, the frequency with which P. munitum hybridizes with other 


48 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 2 (1987) 


species may also reflect the breeding system of this species. Polystichum munitum 
is almost completely outcrossing; intragametophytic selfing estimates in popu- 
lations of P. munitum ranged from 0 to 3% (P. Soltis & D. Soltis, 1987). This 
outcrossing breeding system provides the potential for interspecific hybridization 
whenever P. munitum occurs in the same locality as other Polystichum species. 
Other Polystichum species are also outcrossing. For example, in P. imbricans 
sional selfing estimates ranged from 0 to 17% (D. Soltis and P. Soltis, 
1987). Dat f P. kruckebergii and other Polystichum 
species are not yet ‘available; ‘however, the frequency of hybridization in this 
genus suggests that high outcrossing rates may be typical of other species of 
Polystichum. 


ACKNOWLEDGMENTS 


e thank Don Armstrong of Vancouver, B.C., for providing us with directions for relocating this 
ae population, of which he made the original discovery By also thank Dave Barrington and 
Dave Wagner for extremely helpful comments on the manuscri 


LITERATURE CITED 


BarRRINGTON, D. S. 1985. The present evolutionary and taxonomic status of the fern genus Polysti- 
chum: The 1984 Botanical Society of America Pteridophyte Section Symposium. Amer. Fern 
J. 75:22 
1986. ee morphology and cytology “ pr anehen: x potteri hybr. nov. (= P. acrosti- 
chek iss x P. braunii). Rhodora 88:297-3 
CALLAN, A. D. 1972. The Shomer ania ued of a ss pt Porystichem munitum x californicum 
hybrid from Douglas County, O 
munitum. M.S. Thesis. suibiicer Susie College, Ashland, Ore 
CopeLanp, E. B. 1947. Genera filicum. Waltham , Massachusetts ntact etait: 
Daicoso, S. 1974. Chromosome numbers of the ‘te genus Polystichum. J. Jap. Bot. 49:371-378. 
GOTTLIEB, e Ke 1981. Electrophoretic evidence and plant populations. Prog. Phytochem. 7:1-45. 
—— Conservation and duplication of i isozymes in plants. Science 216:373-380. 
KNOBLOCH, t w 1976. Pteridophyte hybrids. Publ. Mus. Michigan State Univ. Biol. Ser. 5:277-352. 
Manton, I. 1950. Problems of cytology and evolution in the Pteridophyta. Cambridge: Cambridge 
eres ess. 
M. 


Vai ieties of P oly stichum 


Geis Gs on i section Metapo- 
ystichum at Gobo-sawa, Pref. Chiba. Bull. Nat. a Was Tokyo 16:437-457. 
Seep, A. and T. REICHSTEIN. 1967. Der Farnbastard Polystichum x meyeri hybr. nov. = Polystichum 
raunii (Spenner) Fée x P. lonchitis (L.) Roth und seine Cytologie. Bauhinia 3:299-314. 
Sortis, D. E., C. H. HAuFuer, D. C. DaRRow, and G. J. GAsToNny. 1983. Starch gel electrophoresis of 
ferns: A De oveag of grinding buffers, gel and electrode buffers, and staining schedules. 


Amer. Fern J. 7 
and L. H. Rissenens. 1986. epee in AV agiecs menziesii: Genetic insights from 
enzyme electrophoresis. Amer. J. Bot. 7: 


and P. S. Sottis. 1987. iS cae and roeng systems of homosporous Pteridophyta: A 
re-evaluation. Amer. Naturalist 13 
Sotis, P. + and D. E. aa 1987, Seek ke o a and estimates of gene flow in the homo- 
sporous fern Polystichum munitum. Evolution 41: 620-629 
Megirage R. H. and R. L. pease 1940. A natural hybrid bitwood sighs ston braunii (Spenner) 
ee and P. acrostichoides (Michx.} Schott. Amer. Fern J. 30:8 


SOLTIS ET AL.: POLYSTICHUM HYBRID 49 


TryYON, R. M. and A. F. TrYon. 1982. Ferns and allied plants with special reference to tropical 
merica. New York: Springer-Verlag. 

Wacner, D. H. 1979. Systematics of Polystichum in western North America North of Mexico. 
Pteridologia 1:1-64. 

Wacner, W. H., Jr. 1968. Hybridization, taxonomy, and evolution. Pp. 113-118 in Modern methods 
in plant taxonomy, ed. V. H. Heywood. London: Academic Press. 
. 1973. Reticulation of holly ferns (Polystichum) in the western United States and adjacent 
Canada. Amer. Fern J. 63:99-115 
and D. J. HAGENAH. 1954. A natural hybrid of Polystichum lonchitis and P. acrostichoides 
from the Bruce Peninsula. Rhodora 56:1-6. 


REVIEW 


“Iconographia palynologica pteridophytorum Italiae,” by E. Ferrarini, F. 
Ciampolini, R. E. G. Pichi Sermolli, and D. Marchetti. Webbia 40:1-202. 1968. 

This collaborative work, in Italian, by three biologists at the University of Siena 
and R. E. G. Pichi Sermolli at the University of Perugia, has resulted in an 
impressive series of scanning electron micrographs of spores of the 65 species of 
Italian Pteridophyta. More than 500 SEM’s, assembled in 71 plates, include 3-8 
figures and surface details for each taxon. The conservative nature of spores is 
evident in the similar details of surface and elators of the nine species of Eq- 
uisetum. Similarities of surface contours in many genera with monolete spores 
suggest there may be broad, general alliances between these genera. 

An extensive glossary, illustrated by fine drawings, includes terms as nexine 
and sexine, not applicable to spores. The binary key to genera, based on spore 
morphology is not easily used. Differences of a few microns in the headings 
leading to Phyllitis and Asplenium or to species of Polypodium are difficult 
choices. The heading “Spore senze perina” leading to Polypodium unfortunately 
is an inaccuracy that undoubtedly persists from light microscope observations in 
which the perine was not evident. Scanning and transmission microscope work 
on spores show, with few exceptions, that spores of all Filicineae have perine, 
or perispore. 

In addition to descriptions of the spores, the text consists of comments on 
nomenclature and the cytology of the species. In light of the emphasis on cytology 
it is disappointing to find the SEM magnifications are inconsistent. Differences 
in size of spores of Asplenium ruta-muraria, shown at x 1080 for the autotetra- 
ploid subsp. ruta-muraria, and at x 750 for the diploid subsp. dolomitica, cannot 
be readily visualized. : 

This atlas of spores will be particularly valued as a reference for comparison 
of surface morphology of these 65 species with that of spores from other regions. 

The fine details depicted in SEM spore studies such as this require special 
attention to reproduction of the micrographs for publication. This unfortunately 
has contributed to the cost of this volume, which at 130,000 lira, is somewhat 
more than one hundred dollars.—ALICE F. TRYON, Harvard University Herbaria, 
Cambridge, MA 02138. 


American Fern Journal 77(2):50-57 (1987) 


Growth Patterns of Gemmlings of 
Lycopodium lucidulum 


ULRIKE REUTTER 
Institut fiir Systematische Botanik und Pflanzengeographie, Im Neuenheimer Feld 328, 
6900 Heidelberg, Federal Republic of Germany 


Knowledge of branching patterns in Lycopodium is often based on herbarium 
specimens of mature plants or parts of plants. Comparative studies of the archi- 
tecture and chronological events have received little attention. Most information 
concerns the mode of dichotomy or stelar structure in plant parts of different age 
(Ogura, 1972). Primack (1973) compared shoot growth patterns in five species of 
Lycopodium. 

Investigations of the ontogeny of young plants and their changes as they reach 
maturity are limited by the difficulties in collecting gametophytes and young 
sporophytes of different age. There are few references to the growth pattern of 
vegetatively propagated plantlets and their chronological development (Bruch- 
mann, 1898; Troll, 1937, for L. selago). 

The present study of Lycopodium lucidulum Michaux is concerned with the 
method of elaboration of a mature plant from a single gemma. As L. lucidulum 
grows and branches very slowly in time, plants of different age had to be com- 
pared to analyze ontogenetic changes. 


MATERIALS AND METHODS 


Lycopodium lucidulum populations were studied in the understory of mixed 
hardwoods at Harvard Forest, Petersham, Massachusetts. These occur in old 
forests dominated by white pine and red maple and particularly in wet undis- 
turbed forests close to small streams. 

Measurements and drawings of old individual plants were done in the field: 
whole smaller clumps were removed and studied in the laboratory. To analyze 
the growth in length within a definite period, eight erect axes were marked with 
ribbons at a measured distance from the apex. From this fixed point the additional 
growth of the apices was noted at weekly intervals over a period of 5 months (19 
April to 13 September, 1986) (Fig. 1). An additional estimate of the yearly growth 
was provided by a single axis (marked in 1978 by K. Esseichick) which showed 
a similar average to that of the 8 more recently marked axes. 


RESULTS 


Deterministic shoot dynamics.—Lycopodium lucidulum has erect axes that 
duplicate themselves through equal dichotomy (isotomy) of the apices. Mature 
shoots of L. lucidulum lack definite strobili but have alternating vegetative and 
fertile zones along the stem (Fig. 2). Microphylls or vegetative leaves are longer 


U. REUTTER: LYCOPODIUM GEMMLINGS 51 


os 

3 

3 

oa ee 
T 


+ 
‘ 


wu 
n 
r 


Additional Growth 


— 
4 
a 


1 74°55 bo) 89 ee Ske ek TT OOD RO Mieka 
Fic. 1. Shoot extension in Lycopodium lucidulum (additional growth in length per week) during 
one growth period (19 Apr 1986 to 31 Sept 1986). Bars represent standard deviation (n = 8). 


(12 mm) and wider (2 mm) than the sporophylls (fertile leaves), which are 5 mm 
long and less than 1 mm wide. There is an abrupt transition between these 
different types with only 1-2 whorls of an intermediate size. 

The shoot complexes have a certain “individual” size because of simultaneous 
additional growth at the distal green end and rotting at the rhizomatous creeping 
end (Table 1). Erect distal axes retain a uniform height because older parts recline 
to a prostrate position with yellowing microphylls indicating loss of chlorophyll 
(‘transition zone’’) (Table 1). 

The lower portion of older, now rhizomatous axes becomes covered by ac- 
cumulated leaf litter within which root develpment is pronounced. Rhizome 
length may be arbitrarily measured from the point of the youngest visible root 
that is just penetrating the stem cortex to the oldest persistent part of the axis. 

Roots are initiated endogenously. Some root primordia can be found near the 
apex, associated either with two branches of a dichotomy or within the zone of 
the main yearly growth increment, indicated externally by larger vegetative mi- 
crophylls. Root initials are much delayed in their further development. Addi- 
tional, possibly adventitious, root primordia can be found in older parts of the 
shoot. The region where root extension is most pronounced seems to be the 
transition zone. Here the root meristems develop and grow about 1-2 cm inside 
the cortex in proximal direction. They break through the cortex at irregular 
distances from their origin and fix the now horizontal part of the axis to the 
ground. 

Field observations, growth measurements, and dissection of winter buds sug- 
gest that the small sporophylls are preformed late in the vegetative period of the 
previous growing season but remain enclosed in the apical bud during the winter. 
Dissection of a bud in early spring shows the sporophylls bearing immature 
sporangia. The large vegetative microphylls appear later during the main growth 
period in April to June, ie., they are newly formed. 


52 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 2 (1987) 


| 


25 mm 


2 3 


Lycopodium lucidulum. 2. Mature shoot (fragment) with alternating zones of sporophylls 
and larger microphylls. Gemmaphores of the recent growth period with gemmae. 3. Gemmling with 
gemma still attached. 


Fics. 2, 3. 


In this period of vegetative growth, the plant also produces gemmae. Axes that 
are to branch also will undergo isotomy at this time. The age of a shoot can be 
estimated from fluctuation in leaf size since a single vegetative and fertile zone 
develops each year. Growth measurement of some shoots during the 5 month 
monitoring period confirmed the seasonal cycle (Fig. 1). The decrease of the 
growth rate in June is probably related to the unusually high rainfall and low 


U. REUTTER: LYCOPODIUM GEMMLINGS 53 


TABLET. A ge GrowthM ts of a Mature “Individual” of Lycopodium lucidulum; n = 15. 
Mean Range 

Total length of “individual” 31.9 + 6.5 21-47 cm 

Height of erect green axis 12.8 + 2.5 9-19 cm 

Length of transition zone 46+ 1.4 2-7 cm 

Length of rhizome 145 +72 4-33 cm 
Age of “individual” 13.4 + 2.8 10-20 years 
Additional growth 24.2517 2.1-2.7 cm/year 
Number of dichotomies Ne We ap 2-5 
Time interval between two dichotomies 4g 17 2-7 years 


temperature during this period. Case (1943) described a similar zonation pattern 
for L. selago, although this species shows a different development in time. Spo- 
rangia are initiated in early summer and need about two years for ripening; 
gemmae are laid down in mid-summer and are shed in fall of the following year. 

Equal branching of the apical meristem occurs only in the green, erect shoots. 
The number of terminal axes (x) follows the simple relationship 

x= 2 

where n is the number of previous dichotomies. This regular pattern can be found 
in many gemmlings (i.e., young plants derived from gemmae; Stevenson, 1976). 
In older plants the regular system is often disturbed by damage and consequent 
loss of one or more shoots. No deterministic mechanism for the development of 
additional branches as a response to damage was observed, so that a lost branch 
is irreplaceable. The total number of branches (x) on the plant therefore dimin- 
ishes from the ideal number according to the relation 


oe me eee 
(for an injury at the a-th dichotomy) or 
er 2 ee 


(for an injury at the a-th and b-th and . . . dichotomy). There is no evident change 
in growth form, rate of extension, or branching frequency of one axis of a di- 
chotomy, compared to undamaged shoot pairs, if the corresponding one is dam- 
aged. Very occasionally lateral gemmae that fail to become detached can replace 
a lost main axis, but this is not a deterministic feature of growth. 

Gemmae development.—No plants were observed developing from gameto- 
phytes. The following information therefore relates only to plants developed from 
gemmae. 

Gemmae are produced on older shoots by structures (‘‘gemmaphores’’) that 
have the position of a microphyll but show all features of a shoot (Stevenson, 
1976). Typically, the gemmaphore initiates 3 pairs of microphylls that remain on 
the old stem after gemma separation. The distal part of this gemmaphore develops 
another 3 pairs of microphylls, one pair of them large and fleshy. This fleshy pair 
forms a heart-shaped structure (the gemma sensu stricto) that separates from the 


54 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 2 (1987) 


gemmaphore by means of an abscission layer. A root primordium develops on 
the proximal end of the gemma before abscission occurs. 

Gemmae are shed in fall and become buried with leaf litter. In the following 
spring the gemmae develop an axis and microphylls that differ in shape from 
those of the adult plant (Fig. 3). The length of these first axes depends mostly on 
the thickness of the covering litter. The covered parts bear few scalelike micro- 
phylls. On average the exposed axis of the gemmling in the first season reaches 
a length of 1.5-3.0 cm with green leaflike structures, which are rounder and about 
half the size of the adult microphylls (5 <x 3 mm). They are arranged in an 
increasing phyllotactic spiral. Often there are one or two endogenous roots in 
addition to the first developed gemma root. Usually the two fleshy microphylls 
become lignified and persistent so that they can be recognized on plants up to 
18 years old. This allows the reconstruction of the development and branching 
pattern of the plant from its first dichotomy. 

The gemmling undergoes its first dichotomy at the beginning of the second 
year (Fig. 4). Subsequent dichotomies occur either in the next vegetative period 
or in later years. Usually the two daughter branches of a dichotomy show the 
same branching pattern, so that young plants with few branches have a very 
symmetrical shape (Fig. 5). In comparison with the adult stages there is a mea- 
surable tendency in older plants to prolong shoot growth and reduce branching 
frequency, expressed in the increasing distance between two successive dichot- 
omies (Fig. 6A). The delay of branching can be recognized by the seasonal change 
in leaf size, which is first shown by the plant in its second or third year. Although 
no sporangia are produced on plants of this age, one can distinguish larger from 
smaller microphylls, the latter equivalent in shape and size to sporophylls. Annual 
growth periodicity is the same as in the adult plant; little variation results from 
different climatic or nutritional circumstances (Fig. 6B). 

The internal angle between two daughter branches of one dichotomy decreases 
from a range of 90°-180° in the first dichotomy to 20°-30° in the dichotomies of 
the mature shoot (Fig. 6B)..The planes of the first and second dichotomy, inde- 
pendent of the angle between the daughter shoots, regularly are arranged in an 
interplanar angle of 90°, described as “cruciate” by Troll (1937) for Lycopodium 
selago. Subsequent inclinate orientation of dichotomy planes (Ollgaard, 1979) 
shows interplanar angles with less than 30°, which means that all descendants 
of each branch resulting from the first dichotomy are arranged with their apices 
in a half circle. This variation of the branching angles, together with the increasing 
shoot growth increments results in the arrangement of the erect shoots in “growth- 
rings” or “fairy-rings” (Fig. 5). The largest diameter of a plant observed with an 
attached basal gemma was about 50-60 cm. In older stages, the original orientation 
towards a center of origin is less easily recognizable. 

Gemmilings bear the first sporangia at an age of 6-7 years. Daughter gemmae 
appear about 2 years later. This change to fertility seems independent of the 
frequency of previous dichotomies. With increasing age and diameter the oldest 
parts of the plant decay and the distal shoots become separated. The persistent 
fragments, consisting usually of 2-8 connected shoots resulting from the most 
recent 1-3 dichotomies, have several functional apical meristems at their distal 


U. REUTTER: LYCOPODIUM GEMMLINGS 55 


Gemma 30mm 
O 
Gemma 
6A 68 
aST 150 
14 2% 
w 
13t a 
Pye | wu 120 
Le c 
ay Sai ‘ an 
pi g 6 - 
2316 > 100 3 
a ec ee > 
> ot “oe = 
oe E 
= 8 Nd “z 80 oe 
& 5 S ec 70 > 
tn =. @ Ps 
°o 6 +60 @ 60 
E ae £ 
aS 5S 50 > 
eb ib 60} se 
is . = 3 
$, SF eee eee 3 
22 2q@ <— 20 € 
mn 
4 1 p< 10 TL 
1 2 3 4 5 6 7(Number of 1 2 3 4 5 6 7 {Number of 
di dichotomy! 


Corresponding age of shoot interva 
deviation (n = 10). 6B. Annual growth ‘ 
circles}. Angle between the sister branches of dichotomies of increasing 
represent standard deviation (n = 10). 


56 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 2 (1987) 


ends. In this way new “individuals” (ramets) are recognized (Fig. 2). The branch- 
ing pattern of each single fragment shows no specific relation to other fragments 
and is the same as described for the mature plant in Table 1. The variation 
observed is large. The unusual synchrony of dichotomies therefore occurs only 
in “young” stages of development. 


DISCUSSION 


Branching, by multiplying meristems, is used to define the growth units out of 
which a modular organism is composed (Harper & Bell, 1979). Lycopodium lu- 
cidulum, as a clonal plant, repeats a simple pattern: prolongation of each axis 
for a certain distance, and division of the apical meristem into two equal portions, 
which each grow the same length before dichotomizing again. 

The plant (population of fragments) therefore shows an intrinsic exponential 
growth within a constant time unit. However, in contrast to the continuous pop- 
ulation growth described for other plants (Harper & White, 1974), L. lucidulum 
produces gemmae as an additional vegetative method of multiplying. These fol- 
low different rules of growth per time unit. The analysis of a plant growing from 
a gemma shows a controlled change of the modular units, which is expressed in 
the time interval between two dichotomies. As it ages, the gemmling shows at 
least two different growth models. It might therefore be possible to find a third 
developmental pattern in plants developed from gametophytes. Although some 
similarities exist in pattern of phyllotaxis and microphyll shape (Bruce & Beitel, 
1979; Bruchmann, 1899, for L. selago) between the first axes of both sexually and 
asexually initiated shoots, there is no reference to the further development of 
young sporophytes in the literature. 


Lk ity to repair damage, but 
compensates for this disadvantage by the development of gemmae. This could 
be interpreted as a derived pattern imposed on an ancestral determinism. 


U. REUTTER: LYCOPODIUM GEMMLINGS 


ACKNOWLEDGMENTS 


This study was made possible by a pepe tock of the DAAD to the author at Harvard Forest, 
aa. Massachusetts. I thank Dr. P. B. Tomlinson for providing facilities and also for suggesting 
this study. I much appreciate his patience, ine and encouragement during the course of this work 
and during my stay at Harvard Forest. 


LITERATURE CITED 


BELL, A. 1984. Dynamic morphology: a contribution to plant population ecology. Pp. 48-65 in 
ee on plant population ecology, ed. R. Dirzo and J. Sarukhan. Sunderland, Mass.: 


BRUCE, J. c: beg M. res 1979. A community of Lycopodium gametophytes in Michigan. Amer 
ern J. 69:33- 
ea H. nea “bak die Prothallien und die Keimpflanzen mehrerer europdischer Lyco- 
n. Go 


Case, I. M. niet Nan in the development of fertile and sterile zones in Lycopodium selago. 
New Phytol. 42:93-97 

Harper, J. L. AND A. D. BELL. 1979. The population dynamics of growth form in organisms with 
modular construction. Pp. 29-52 in Population dynamics, ed. R. M. Anderson et al. The 20th 
Symposium of The British Ecological Society, London 
and J. WHITE. 1974. The demography of plants. Annual Rev. Ecol. Syst. 5:419-463. 

Ocura, A. Y. 1972. Comparative anatomy of vegetative organs of the pteridophytes. Berlin: Ge- 
briider Borntraeger. 

QMLLGAARD, B. 1979. Studies in Lycopodiaceae II: The branching patterns and infrageneric groups 

ycopodium s.|. Amer. Fern J. 69:49-61. 

PRIMACK, R. B. 1973. Growth patterns of five species of Lycopodium. Amer. Fern J. 63:3-7. 

STEVENSON, D. W. 1976. Observations on the phyllotaxis, stelar morphology, the shoot apex and 
gemmae of Lycopodium lucidulum Michaux (Lycopodiaceae). J. Linn. Soc., Bot. 72:80-100. 

TROLL, W. 1937. Litman Morphologie der héheren Pflanzen. Bd. 1, Teil 1:465-483. Berlin: 
Gebriider Borntraeg 

Witce, |. H.: 1972. Esead | on I. General spore patterns and the generic segregates of Lyco- 
podium. Amer. Fern J. 62:65-79. 


American Fern Journal 77(2):58-63 (1987) 


The Expanded Adaxial Epidermis of Equisetum 
Rhizome Sheath Teeth 


RICHARD L. HAUKE 
Department of Botany, University of Rhode Island, Kingston, RI 02881 


Plants of Equisetum have an extensive rhizome system that produces nodal 
buds that grow into either aerial stems or branches of the rhizome. Certain species 
also have tuber buds. Both the aerial stems and the rhizomes of Equisetum have 
nodal sheaths bearing a crown of teeth. The sheath is interpreted as a whorl of 
fused leaves, with the teeth representing the free tips of the individual leaves. 
The buds consist of a series of superposed sheaths that overarch the shoot apex. 
Subsequent intercalary meristematic activity at the base of each nodal sheath 
separates the nodal sheaths and produces the internodes. 

The sheath teeth of the subterranean buds have a distinctive adaxial epidermis, 
a feature rarely noted in the literature. Francini (1942) compared rhizome and 
aerial buds of E. ramosissimum, and described trichomes developing from the 
inner epidermis of the rhizome sheath teeth and growing together to help the 
teeth form a cap protecting the apex. She also described mucilage production 
from the abaxial epidermis of the outer sheaths and pointed out how only one 
internode at a time elongates in the rhizome. Sachs (1882, p. 401) included a 
figure of a longitudinal section through an underground bud of Equisetum arvense 
showing an elaboration of the adaxial surface of the sheath teeth, but he did not 
describe this feature. Sadebeck (1902, p. 532) copied Sachs’ figure, but did not 
mention the structure of the adaxial surface of the teeth. The purpose of this 
study is to describe more completely this little noted anatomical feature of Eq- 
uisetum, and to speculate on its function. 


MATERIALS AND METHODS 


Rhizomes of Equisetum hyemale L., E. arvense L., and E. telmateia Ehrh. 
subsp. braunii (Milde) Hauke were excavated, washed, and their buds removed. 
Potted plants of E. x schaffneri Milde, E. scirpoides Michx, and E. diffusum Don 
were depotted and rhizome terminal buds removed. The buds were killed and 
fixed in FAA for 24 hours and then stored in 70% EtOH. Sources of material are 
given on Table 1. 

The outermost sheath was dissected from the bud. The clean bud was dehy- 
drated in a tertiary butyl alcohol series, embedded in Paraplast (56-57°C), sec- 
tioned on a rotary microtome at 15 ym (longitudinal sections) or 20 um (transverse 
sections}, stained with safranin-fast green or safranin-toluidine blue (Berlyn & 
Miksche, 1976), and mounted in diaphane. Photos were taken on Kodak Plus-X 
film with a Nikon Microflex UFX camera mounted on a Zeiss microscope. 

Buds used for scanning electron microscopy (SEM) were dehydrated in an 
ethanol series that included two changes of 100% ethanol and critical point dried 


R. L. HAUKE: EQUISETUM 59 


TABLE 1. Sources of Material of Equisetum Rhizome Buds. 


Species Locality Voucher’ 

E. arvense Rhode Island, Galilee Hauke 516 

E. arvense Rhode Island, Kingston Hauke 514 

E. diffusum Greenhouse, Univ. Rhode Island (original- Hauke s.n. 
ly from near Simla, India) 

E. hyemale var. affine Rhode Island, Galilee Hauke 515 

E. x schaffneri Greenhouse, Univ. Rhode Island (original- Hauke 205 
ly from near San Jose, Costa Rica} 

E. scirpoides Greenhouse, Univ. Rhode Island (original- Hauke 490 
ly from near Rutland, Vermont) 

E. telmateia subsp. braunii California, Oakland, Claremont Canyon Hauke C1 

? All in KIRI. 


using CO, substitution and a Tousimis Samdri-PVT-3B critical point drier. They 
then were mounted on aluminum stubs with double-sided tape, coated with gold- 
palladium in a Hummer II coater, and observed with a Cambridge $4 scanning 
electron microscope. 


RESULTS 


The rhizome buds of Equisetum are different from the aerial stem buds in 
being sharply pointed (Fig. 1) and in having fewer, superimposed sheaths. Rather 
than the simultaneous development of several internodes, one internode elon- 
gates at a time. Roots develop at each node, but branches (rhizome or aerial) 
develop only occasionally. Numerous nodal buds are initiated, each with a root 
apical initial and a shoot apical initial. Frequently the root apical initial is func- 
tional, while the shoot apical initial generally is dormant or aborts. Occasionally 
the shoot apical initial becomes active, and the bud subsequently enlarges and 
erupts through the nodal sheath. Abundant mucilage, apparently produced by 
the abaxial epidermis, is associated with the rhizome buds, and is particularly 
obvious when the out t sheath is dissected from the bud. Those buds destined 
to produce aerial shoots will possess these same features as they develop at the 
rhizome node, but change as they grow upward. 

The sheath is initiated as a nodal ring, from the upper edge of which a crown 
of teeth is initiated, grows rapidly upward, and arches over the shoot apex. Then 
the nodal ring grows up to prod heath isting of a whorl of segments, 
one under each tooth, joined laterally by commissures (Hauke, 1985). The adaxial 
epidermis of the teeth of the rhizome sheath becomes greatly inflated as the teeth 
elongate and arch over to form a conical cap over the apex (Fig. 3). At first the 
apical ends of the outer walls of the epidermal cells expand and form papilliform 
trichomes (Fig. 2), while later the entire cell enlarges and a vertical flange is 
formed on the adaxial face of the rhizome sheath teeth (Fig. 4). Since the sheath 
teeth arch inward to form a convex cone the vertical flanges press together to fill 
the space above the next inner sheath. 


AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 2 (1987) 


I] 


ae 


? 
' 


— 
— 
—— 
oe 
—_ 
—— 
— 
—— 
—— 
ws 
oe 
— 
ae 
— 
_ 
es 
— 
— 
— 
— 
— 
— 
— 
— 
— 
— 
nae 
—— 
— 
— 
—— 
— 
— 
— 
—_ 
sae 


Seco 


Fics. 1-3. Rhizome buds in Equisetum. 1. E. x schaffneri. 2. E. telmateia, SEM of young sheaths 
with distinct papilliforme trichomes: scale = 50 um. 3. E. arvense, longisection of bud with overarching 
sheaths showing expanded adaxial epidermis (arrows); scale = 50 um 


R. L. HAUKE: EQUISETUM 


7 


SEM of sheath with flanges of expanded 
200 um. 5. E. hyemale, SEM of well-developed trichomes at 
_6. E. scirpoides, longisection of sheath apex with papillae; 
heaths with expanded adaxial epidermis {arrows}. Scale 


Fics. 4-7. Rhizome sheaths in Equisetum. 4. E. arvense, 


cells on sheath teeth (arrows); scale = 
apices of young sheaths; scale = 200 um 
scale = 50 um. 7. E. arvense, transsection of s 
as in Fig. 6. 


62 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 2 (1987) 


Differences exist in the degree of initial outgrowth of the apical end of the 
epidermal cells, so that in some species there is a more trichome-like appearance 
of the adaxial epidermis than in others. Equisetum hyemale (Fig. 5) shows the 
most trichome-like appearance of the adaxial epidermis, followed by E. x schaff- 
neri, E. telmateia (Fig. 2), E. diffusum, E. arvense (Fig. 3), and E. scirpoides (Fig. 
6). In the last named species, the epidermal outgrowths are mere papillae. The 
initial apical expansion is most pronounced in the acropetal cells, and there is a 
transition in epidermal cell size from highly expanded to normal as one moves 
from apex to base of a sheath tooth in all species studied (Fig. 3). Thus, in a 
transverse section showing concentric sheaths, the outer sheaths show less epi- 
dermal inflation than the inner ones (Fig. 7). The earliest {i.e., outermost and 
subsequently lowermost) sheaths on buds destined to become aerial shoots also 
show rhizomatous features. 


DISCUSSION 


My observations are consistent with Francini’s (1942) description of the ex- 
panded inner epidermis of Equisetum rhizome sheath teeth. She described the 
sheath teeth as producing trichomes, which curve upward toward the center of 
the apex, intertwine, and fill the conical space. According to her, they begin to 
form in the second sheath from the apex (second youngest sheath), become quite 
long in the third, and interweave along the axis and develop thickened walls in 
the fourth sheath. At first I thought that Francini’s description of the expanded 
epidermis was erroneous, because the whole epidermal cell appeared expanded, 
rather than having a trichome grow from it. The reason for the initial discrepancy 
between Francini’s observations and mine then became obvious, because I could 
see that initially there was a distinct outgrowth from the upper end of the epi- 
dermal cell, followed by a more general expansion of the whole cell. To describe 
them as trichomes is somewhat misleading. The outgrowths in E. scirpoides, E. 
arvense, and E. diffusum are more accurately described as papillae, which Uphof 
(1962) defines as being unicellular, slightly elongated, and little differentiated. 
Since the structures in E. telmateia, E. x schaffneri, E. hyemale, and probably 
E. ramosissimum are more elongated but still unicellular and little differentiated, 
they should be referred to as papilliform trichomes. The term “papillenhaare” 
of Eckhart, or papillate hair, is used for trichomes with a papillose surface (Uphof, 
1962) and cannot be used in this case. 

The sharply pointed bud ends, expansion of one internode at a time, mucilage 
production, and expanded adaxial epidermis of Equisetum rhizome buds prob- 
ably all relate to the growth of rhizomes through the ground, and resultant forcing 
of the apical bud through the soil. The pointedness of the bud could help penetrate 
the soil. The expansion of one internode at a time would permit a node to complete 
its development and initiate roots relatively close to the growing point. French 
(1984) pointed out that the rhizomes of Equisetum delay outgrowth of roots and 
buds from elongating regions of the main axis. This reduces friction and prevents 
damage to these organs. Also, according to French, a single distal region of 


R. L. HAUKE: EQUISETUM 63 


elongation better provides the force for soil penetration than would several active 
intercalary meristems separated by distances up to several centimeters in length. 

The younger sheaths remain enclosed and protected by an older, more indu- 
rated sheath for a longer time. The mucilage possibly facilitates the passage of 
the apical bud through both the oldest enclosing sheath and the soil. The mucilage 
may also protect the tissues from soil microorganisms. Finally, the expanded 
adaxial epidermis may, by filling the conical space formed by the enclosing teeth, 
maintain the sharply pointed form of the sheath as it pushes through the soil, 
then yield to the next sheath pushing through it. 

The rhizome has an apical meristem protected by young leaves. The growing 
rhizome, subjected to stresses similar to those of a growing root, requires a struc- 
ture analagous to the root cap for protection against these stresses. In the case of 
Equisetum, the shoot apical bud of a rhizome functions in a similar manner to 
a root apex. To prevent their being injured by elongation of the root, root hairs 
do not develop until elongation ceases. Similarly, Equisetum nodal roots develop 
after elongation of the internode below them ceases, and elongation of one 
internode at a time permits this to occur close to the apex. Mucilage is associated 
with root apices, and Equisetum rhizome buds produce copious mucilage. The 
root apex forms a root cap, which is continually parting the soil to permit the 
apex to penetrate, and is continually being replaced by the apex. Similarly, 
Equisetum rhizome apices have a cap of superposed sheaths. These sheaths are 
sharply pointed to penetrate the soil, are strengthened by the expanded adaxial 
epidermis, and are continually being grown through by the apex, as well as 
continually being replaced. Thus the expansion of the adaxial epidermis of the 
rhizome sheath teeth of Equisetum is probably part of the structural adaptation 
that permits the rhizome to grow successfully in the soil. 


LITERATURE CITED 


BERLYN, G. P. and J. P. MIKSCHE. 1976. Botanical microtechnique and cytochemistry. Ames: lowa 
State University Press. 

FRANCINI, E. 1942. La struttura dell’apice del rizoma in confronto alla struttura dell’apice del fusto 
aereo in “Equisetum ramosissimum” Desf. Nuovo Gior. Bot. Ital. 49:337-357. 

FRENCH, J. C. 1984. Occurrence of intercalary and uninterrupted meristems in Equisetum. Amer. J. 
Bot. 71:1099-1103. 

Hauke, R. L. 1985. Ontogeny of the commissure of Equisetum. Amer. Fern J. 75:111-119. 

Sacus, J. 1882. Textbook of botany. Translated by S. H. Vines. 2nd Ed. Oxford: Clarendon Press. 

SADEBECK, R. 1902. Equisetaceae. Pp. 520-548 in Die natiirlichen Pflanzenfamilien, 1(4), ed. A. 
Engler and K. Prantl. Leipzig: W. Engelman. 

Upnor, J. C. TH. 1962. Plant hairs. Handbuch der Pflanzenanatomie 4{5). Berlin-Nicholassee: Ge- 
briider Borntraeger. 


American Fern Journal 77(2):64-65 (1987) 


Schizaea pusilla Discovered in Peru 


ROBERT G. STOLZE 
Department of Botany, Field Museum of Natural History, Chicago, IL 60605 


Most American pteridologists are familiar with the rare “Curly Grass Fern,” 
Schizaea pusilla Pursh, and are aware that it occurs only in scattered localities 
in northeastern North America. Thus far it has been found (with luck) in south- 
ern New Jersey and Long Island, New York, in Nova Scotia and Newfoundland, 
and on the Bruce Peninsula of Ontario (the latter unconfirmed since the original 
report, possibly erroneous). Yet recently it has been found on a mountainside in 
Central Peru! Specimens were collected by Dr. Robin Foster, Research Associate 
at Field Museum: Peru, Dept. Pasco, Oxapampa, Cordillera Yanachaga, Cerro 
Pajonal, 12 km SE of Oxapampa, 75°20’W, 20°35’S, 2700-2800 m, Foster 9065 (F, 
GH, MO, US, USM); shrubland on white sandstone, spongy sphagnum humus 
up to 2 m deep except where burned; plant colonizing sandy, wet landslide on 
steep slope; fertile frond brown, growing with Drosera and dwarf Xyris. 

When Foster presented specimens to me for identification, I remarked that 
this looked like a new record for Peru, S. fistulosa Labill. var. australis (Gaud.) 
Fosb., and set it aside for further study. Later, during research on Schizaea in 
Peru, it became clear that the specimens were instead S. pusilla: in the size of 
the plant, in the number, shape and pubescence of the fertile segments, and as 
to the character of the spores. The latter were examined by Dr. Alice Tryon on 
a duplicate collection at GH, who explained (pers. comm.) that spores of this 
species are quite distinctive in size, shape, and dense pitting of the surface. (For 
further discussion of the distinctive types of spores in Schizaea, see Selling, 1944, 
and Tryon & Tryon, 1982, pp. 80-82.) The only way the Foster specimens differ 
from the North American collections is that sterile leaves are not conspicuously 
curled but are straight to merely flexuous. It is likely that some will consider this 
single character sufficient basis for specific or infraspecific distinction: however, 
blade habit can be somewhat variable, even in the Northern plants in which 
sterile leaves are occasionally flexuous, rarely straight. Actually S. pusilla is not 
so closely related to the Old World S. fistulosa and its variety australis from Chile 
and Argentina, which differ in the lack of pubescence on the sporangiophore 
and in the erose-lacerate margins of the fertile ultimate segments. It has greater 
affinity to S. incurvata Schkuhr (Surinam to Venezuela, Peru, northern Brazil), 
especially in the abundance of long, flexuous trichomes on the subentire fertile 
segment margins and among the sporangia. Schizaea incurvata, however, is a 
much larger fern, with once- to twice-forked fertile leaves, 2-4 times as many 
fertile segments, and unpitted spores. 

The discovery of S. pusilla in Peru prompts interesting speculation as to the 
true distribution of the species. Should such a “disjunction” really be so sur- 
prising? May we not hypothesize that it is to be found in numerous other local- 
ities, throughout the neotropics, for example? One has only to attempt a search 
for it, even in reported stations in the United States, to discover how difficult it 


R. G. STOLZE: SCHIZAEA PUSILLA 65 


is to locate. Robin Foster reports that his collection was made quite by accident, 
when he slipped and fell traversing the landslide area, only to find this strange 
little fern virtually “under his nose.” Even then, only careful scrutiny produced 
a dozen more plants. How many more inconspicuous ferns can we expect to 
encounter in the tropics—either by luck or through assiduous search—such as 
species of Hymenophyllaceae and Ophioglossaceae? Even the larger Schizaea 
incurvata has been reported but once from Peru. Obviously, awareness of special 
soil requirements or habitat preferences can increase the likelihood of finding 
certain rare ferns. North American collection reports inform us that S. pusilla 
usually is found “in wet, sandy areas” and “in sphagnous bogs”; and the Foster 
collection proved no exception to this. That is not to say the fern is not found in 
other, e.g., drier or rocky habitats, but according to Selling (p. 81, 1944): “It is 
distinctly a species which prefers moisture and acid soils.” By their very nature, 
general and mass collecting expeditions will yield but a small percentage of such 
fortunate discoveries; but special awareness and advance preparation can raise 
the odds! 


LITERATURE CITED 


SELLING, H. 1944. Studies in the recent and fossil eo. of Schizaea, with particular reference 
o their — characters. Acta Horti Gothob. 16: 
TRYON, ‘ M. AND A. F. TRYON. 1982. Ferns and pee sa with special reference to tropical 
pare New York: Springer-Verlag. 


REVIEW 


“Illustrations of Pteridophytes of Japan, Volume 4,” edited by S. Kurata and 
T. Nakaike with the cooperation of the Nippon Fernist Club. 1985. x + 852 pp. + 
folding map. University of Tokyo Press. Yen 13,000. ISBN 4-13-061064-3. 

The hundred pteridophytes depicted in volume four of this ongoing series bring 
us to the halfway point in coverage of the perhaps 800 Japanese taxa. The format 
of previous volumes, all previously reviewed (Amer. Fern J. 72:11, 1982; 72:48, 
we 74:6, 1984) is faithfully followed. 

genera wholly or partly contained in this volume are Angiopteris, 
Boban Cyathea, Dryopteris (31 species and varieties), Equisetum, Gymnocar- 
pium, Plagiogyria, Pteris, Tectaria, and Woodsia. I noted an error in the spelling 
of Pteris dispar Kunze, a species whose two cytotypes were intensively studied 
and mapped by N. Nakato in J. Jap. Bot. 56:200-205, 1981. The tetraploid is 
northern, with larger spores and smaller scales than the diploid. Some authors 
have treated dispar as a variety of Pteris semipinnata L.—M. G. Price, Herbar- 
ium, North University Building, University of Michigan, Ann Arbor, MI 48109 


American Fern Journal 77(2):66-72 (1987) 


SHORTER NOTES 


Chromosome Numbers of Some Ferns from Argentina.—This contribution is a 
part of fern projects in progress, namely the systematic revision of the Argentine 
Thelypteridaceae (by M. Ponce) and a study of the Microgramma squamulosa 
group in the Neotropics (by E. de la Sota and L. Cassa de Pazos). 

Material of Thelypteris was fixed in situ; Microgramma was fixed from plants 
from Tucuman under cultivation. All material was fixed in Newcomer fluid 
modified by Hunziker (Kurtziana 3:151-156, 1966). Traditional cytological pro- 
cedures were utilized for Thelypteris. Root tips of Microgramma squamulosa 
were pretreated with 0.002 M 8-hydroxyquinoline for about 3 hours at room 
temperature. Both types of material, somatic and gametic, were stained with 
Feulgen and hematoxylin. Photomicrographs were taken with a Zeiss microscope 
and a Zeiss-Ikon camera. Voucher specimens are deposited in LP and SI. 

Results can be summarized as follows: 

Microgramma squamulosa (Kaulf.} Sota (Fig. 1,A-D). 2n = 74, root tips; 2n = 
37 II, meiosis regular. Argentina, Tucuman, Ciudad de Tucuman, Jardin Fun- 
dacion Miguel Lillo, Legname s.n. (LP); under cultivation in Villa Lugano, ciudad 
de Buenos Aires. 

Thelypteris (SAmauropelta) stierii (Rosenstock) Reed (Fig. 1F) 2n = 29 II, 
meiosis regular. Argentina, Salta, Depto. Anta, Parque Nacional “El Rey,” picada 
al Chorro de los Loros, Rio La Sala, Ezcurra et al. 425 (SI). 

Thelypteris (§Goniopteris) abbiattii Reed (Fig. 1E). 2n = 36 II, meiosis regular. 
Argentina, Buenos Aires, Isla Martin Garcia, Barrio Chino, Tur et al. 1825 (LP, 
SI) 


At present, chromosome numbers of nine species of M icrogramma are known. 
Most are based on x = 37, but Smith and Foster (Fern Gaz. 12:321-329, 1984) and 
Evans (Caryologia 16:671-677. 1963), reported n = 36 for Microgramma vacci- 
niifolia (Langsd. & Fisch.) Copel., while Sota and Cassa de Pazos (Bol. Soc. Argent. 
Bot. 19:69-73, 1980) gave n = 37 for the same species. The gametic and somatic 
numbers reported here for M. squamulosa and before for M. vacciniifolia support 
the hypothesis that these taxa are the parents of M. mortoniana Sota, as was 
previously presumed (de la Sota, Amer. Fern J. 63:61-64, 1973). 

Additional counts on Microg iniifolia are desirable. That taxon and 


Meouiawvic. 


M. squamulosa appear to be part of a complex. Attention should also be paid to 
the seasonality of spore production. This might help explain strong meiotic ab- 
normalities noticed by Smith and Foster (I.c.) for Microgramma lindbergii (Mett. 
ex Kuhn) Sota and M. vacciniifolia. Thus, autumnal spores of M. mortoniana 
showed a low germinative power and a high content of oily components (de la 
Sota & Cassa de Pazos, l.c.). 

This contribution was partially supported by Consejo Nacional de Investiga- 
ciones Cientificas y Técnicas, Buenos Aires, Argentina (CONICET), grant no. 
17745A/84. Results were given during the Session of Cytology of the XX Jornadas 


SHORTER NOTES 67 


Ben ai ‘Cones of hol. sconseg Fors, A, root ig — in co squamulosa, showing 


eiosis in M squamulosa, sh owing 
37 bivalents; D, d th E,d f met li Thel ested abot 
showing 36 bivalents; F, se lin Thelypteris stierii, showing 29 bivalents. Bar scale pm. 


68 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 2 (1987) 


Argentinas de Botanica, ciudad de Salta, 16-20 September 1985. We acknowledge 
the economic help given by CONICET and the technical assistance of Victor H. 
Calvetti.—ELias R. DE La Sorta, Division Plantas Vasculares, Facultad Ciencias 
Naturales y Museo, UNLP, Paseo del Bosque, 1900 La Plata, Argentina; MARTA 
Monica PONCE, Instituto de Botanica Darwinion, Labardén 200, C.C. no. 22, 1642 
San Isidro, Argentina; LILIANA A. CassA DE Pazos, Instituto Fitotécnico Santa 
Catalina, Facultad de Agronomia, UNLP, C.C. no. 4, 1836 Lavallol, Argentina. 


Botrychium pinnatum in Colorado.—Western North America has been found 
to be a center of distribution for the moonworts, Botrychium subg. Botrychium, 
with 13 species and several hybrids (Wagner & Wagner, Amer. Fern J. 73:53-62, 
1983; Wagner & Wagner, Amer. Fern J. 76:33-47, 1986). The mountains of Colorado 
have proven to be rich in both species and localities. It is possible to find four 
or even five species growing together in a roadside meadow in the subalpine 
zone, as was demonstrated in the 1984 Annual Fern Foray on Mt. Evans, led by 
the Wagners and ourselves. Although the state has been the scene of intense 
botanical activity, most attention has been focused on the state’s spectacular 
alpine flora. Botrychiums have been collected only incidentally until recently, 
when they became the subject of study. 

On 30 July 1986, while traveling north on US Rt 550 just south of Coal Bank 
Pass, near the southern boundary of San Juan County, one of us (J.D.M.)} noted 
that we were passing what could be good Botrychium habitat. We stopped and 
examined a small dry drainage channel at the edge of an Engelmann spruce 
forest. Elevation was approximately 3045 m (10,000 ft). In a short time approxi- 
mately twelve plants were found of a Botrychium. These were identified as B. 
pinnatum St. John, using the keys in Lellinger (A field manual of the ferns & fern 
allies of the United States & Canada, 1985) and Wagner and Wagner (1986). 
Specimens have been identified also by W. H. Wagner Jr. (pers. comm.). Silhou- 
ettes of this collection, Montgomery & Root 86-279, are illustrated in Figure 1. 
The similarity in vegetative blade and pinna shape and margin with the illus- 
trations in Wagner and Wagner (1983, fig. 1, e-g) is striking and also confirms our 
identification. These plants were growing in a relatively dense growth of low 
herbs in contrast to the more open habitat such as roadsides where Colorado 
moonworts are usually found. In drier gravel near the roadside one plant of B. 
lanceolatum and two of a possible B. lanceolatum x pinnatum hybrid were 
found. The possible hybrids have been sent to W. H. Wagner Jr. for further study. 
Specimens of B. pinnatum have been deposited at COLO and MICH. The col- 
lection has also been reported to the Colorado Natural Areas Program and is 
being considered for listing under plants of special concern for the state. 

Weber (Rocky Mountain flora, 1976) listed B. boreale Milde, a Eurasian species 
formerly confused with B. pinnatum (Wagner & Wagner, 1983) as occurring in 
Colorado. This record is apparently based on a specimen (Willard & Porsild 6062, 
COLO) from Rocky Mountain National Park in Larimer County. Recent exam- 
ination of this specimen has shown that it is probably B. hesperium (Maxon & 


SHORTER NOTES 69 


Fic. %. -Silh 


ouettes of leaves of Botrychium pinnatum from Coal Bank Pass, Colorado (Montgomery 
& Root 86-279). 


Clausen) Wagner & Lellinger. The specimens collected from San Juan County 
are thus the first verified record of B. pinnatum in Colorado. This represents a 
range extension to the south from the range reported by Lellinger (1985) as Alaska 
to the mountains of Montana, northern Nevada, and Oregon. Ranges of most 
moonworts are subject to revision as these small and inconspicuous plants become 
better understood and pteridologists become more familiar with their habitats. 
We observed additional potential habitats for moonworts on both Coal Bank Pass 
and Molas Divide. The San Juan Mountains are a geologically complex area of 
southwestern Colorado which is largely inaccessible and has not been thoroughly 
explored botanically; future examination of this area by pteridologists is certainly 
warranted.—PETER G. Root, Kathryn Kalmbach Herbarium, Denver Botanic Gar- 
dens, 909 York St., Denver, CO 80206, and JAMES D. MONTGOMERY, Ecology III, 
Inc., R.D. 1, Berwick, PA 18603. 


New Records of Pteridophytes from the State of Chiapas, Mexico.—As a result 
of intensive field work for the Flora Mesoamericana Project, several species of 
pteridophytes must be added to those already known for the State of Chiapas. 
Some of them were expected (Smith, Flora of Chiapas, part 2: Pteridophytes, 


70 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 2 (1987) 


1981; Breedlove, Listados floristicos de México, IV, Flora de Chiapas, 1986) 
because they grow at low elevations in the Eastern Highlands and in the eastern 
part of the Central Plateau of the state. These floristic associations are continuous 
with associations in the Petén region of Guatemala. Five out of ten new records 
are filmy ferns, which are easily overlooked because of their small size and 
because often they grow mixed with other species of Hymenophyllaceae. 

In this paper information is given to distinguish the species from related species 
already reported from Chiapas. Identifications have been made or verified by 
Ramon Riba and Leticia Pacheco. 

Actinostachys germanii Fee—Mpio. Ocosingo, Crucero Corozal on the road 
Palenque-Boca Lacantun, 180 m, semi-evergreen seasonal forest (periodically 
inundated), 6 Nov 1985, E. Martinez S. 14891, MEXU, UAMIZ. It differs from 
Schizaea by the sterile leaf which is simple and linear, similar to a small slender 
grass; sporangiophores are digitate. The plant is inconspicuous and is easily 
overlooked in the field. 

Grammitis minuscula (Maxon) Copel.—Mpio. La Trinitaria, Lagos de Mon- 
tebello, 1450 m, pine-oak-Liquidambar forest, 7 Aug 1984, L. Pacheco 1300, UAM- 
IZ. It differs from the other Chiapan species of Grammitis by the entire elliptical 
leaves and round sori. The only difference between Pacheco 1300 and the original 
description is that the specimen is a little shorter, probably due to the degree of 
exposure of the plant. 

Lindsaea portoricensis Desv.—Mpio. Ocosingo, 10 km S$ of Ejido Benemeérito 
de las Américas, on road to Flor de Cacao, Marqués de Comillas, 120 m, semi- 
evergreen seasonal forest, 9 Dec 1984, E. Martinez S. 9508, MEXU, UAMIZ. This 
differs from the other species of Lindsaea in Chiapas by the brownish red, 
abaxially terete stipe, pinnae strongly ascendent, almost vertical and touching 
each other, and pinnules subrectangular, 1.5-2 times as long as wide. 

Schizaea poeppigiana Sturm—Mpio. Ocosingo, Crucero Corozal on the road 
Palenque-Boca Lacantin, 180 m, in semi-evergreen seasonal forest (periodically 
inundated), 8 Jan 1986, E. Martinez S. 15618, 15673, MEXU, UAMIZ. This differs 
from Schizaea elegans by the dimorphic leaves, the fertile leaves not foliose, its 
fertile axis strongly recurved at maturity; sterile leaves repeatedly furcated, the 
divisions not or very slightly expanded and joined only at the base. The species 
has been reported only for the Bahamas, Greater Antilles and from Costa Rica 
to the Guianas and Peru. 

Thelypteris falcata (Liebm.) Tryon—Mpio. Ocosingo, 2 km S of Crucero Corozal, 
on road Palenque-Boca Lacantun, 180 m, semi-evergreen seasonal forest (pe- 
riodically inundated), 21 Sept 1984, E. Martinez S. 7675, MEXU, UAMIZ. It 
differs from T. standleyi by its sporangial stalks without hairs and secondary 
veins nearly straight, not arcuate or subsigmoid. 

Trichomanes ekmanii W. Boer—Mpio. Pichucalco, 6 km N of Pichucalco (by 
air), 200 m, old secondary growth in tropical rain forest, wet creeks, 18 Feb 1985, 
A. Espejo 1440 & S. Hernandez, UAMIZ. This differs by the small blades with 
a continuous submarginal false vein and without marginal hairs; cross-veins 
wanting; involucres immersed, without lips and not dark-edged. 

Trichomanes godmanii Hooker—Mpio. Ocosingo, Crucero Corozal on road 


SHORTER NOTES 71 


Palenque-Boca Lacantun, 180 m, semi-evergreen seasonal forest, 23 Feb 1985, E. 
Martinez S. 11101, MEXU, UAMIZ. This differs from T. ekmanii by the abundant 
false veins, these parallel and perpendicular to the true veins, the venation 
appearing reticulate. 

Trichomanes holopterum Kunze—Mpio. Ocosingo, Crucero Corozal on road 
Palenque-Boca Lacantun, 220 m, semi-evergreen seasonal forest (periodically 
inundated), 6 Nov 1985, E. Martinez S. 15009, MEXU, UAMIZ. This differs from 
T. crispum by the winged petiole, shorter blades, broadly winged rachis, and 
few (5-10) segments on a side. The species has been reported previously only 
for the Greater and Lesser Antilles. 

Trichomanes membranaceum L.—Mpio. Ocosingo, 6 km NE from Pichucalco 
(by air), 220 m, old secondary growth in tropical rain forest, 18 Feb 1985, A. Espejo 
1441 & S. Hernandez, UAMIZ. The species can be easily separated from the 
others of subg. Didymoglossum by the simple or slightly lobed blade, false veins 
parallel to the true veins, and by the paired orbicular scales in the margin of the 
blade. 

Trichomanes tuerckheimii Christ—Mpio. Ocosingo, Nuevo Veracruz, 33 km W 
of Rio Chixoy, road to Chajul, Marqués de Comillas, tropical deciduous forest, 
10 Jan 1986, E. Martinez S. 15908, MEXU, UAMIZ. It is distinguished from T. 
pinnatum by the long-creeping rhizome, distant subsessile leaves, false veins 
few, located near the margin and parallel to the true veins, and by “prehensile” 
trichomes on veins, midribs, and margins in the abaxial face of the blade, which 
cause the lamina to adhere to the surface of the host plant.—RAMON RIBA and 
LETICIA PACHECO, Depto. de Biologia, C.B.S., Universidad Autonoma Metropol- 
itana-Iztapalapa, Ap. Postal 55-535, México, D. F. 09340; and ESTEBAN MARTINEZ 
S., Depto. de Botanica, Instituto de Biologia, UNAM, México, D. F. 14020. 


Additions to the Fern Flora of the Bahamas.—The publication of Correll and 
Correll’s (1982) Flora of the Bahama Archipelago has significantly improved our 
understanding of the Bahamian flora and, as is the case with any new flora, has 
stimulated the discovery of corrections and new taxa. As part of a study of the 
phytogeography of the Bahamian archipelago, I encountered several specimens 
that represent additions to the flora. 

Hypolepis repens (L.) Presl.—This taxon represents a new genus and species 
for the Bahamas. It occurs on the islands of Grand Bahama (Correll & Kral 42936, 
FTG), New Providence (Correll & Correll 48320, FTG), North Andros (Correll & 
Proctor 47828, FTG), and San Salvador (Correll & Wasshausen 46861, FTG). All 
four collections are sterile, and this has apparently confounded identification by 
D. S. Correll who annotated two of the sheets as young plants of Pteridium 
aquilinum var. caudatum. The Grand Bahama collection consists of very young 
plants (less than 1.5 dm tall) and would not be placed as H. repens except for 
the presence of other more readily identifiable specimens for comparison. The 
occurrence of the species in the Bahamas is not especially surprising in light of 
its wide range in the West Indies and the Florida peninsula (cf., Proctor, Flora 
of the Lesser Antilles. Vol. 2. Pteridophyta, 1977). The plants occur in forested 


72 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 2 (1987) 


areas (coppice, pineland, or Casurina thickets) in or around ponds or depressions. 
The lack of apparently suitable habitats (based on populations outside of the 
Bahamas) may account for the species’ restriction to larger islands where suitable 
habitats are more likely to exist. Additionally, the occurrence near the more 
populated portions of the islands and the fact that the collections are more recent 
than 1974 suggest that H. repens may represent an escaped cultigen. This will 
be a difficult proposition to test given the species’ wide geographic range and the 
probability that cultivated plants may also have originated in Florida, Cuba, or 
Hispaniola. 

Polypodium polypodioides (L.) Watt var. michauxianum Weatherby.—Correll 
and Correll (1982) reported only the typical variety for the Bahamas, but several 
specimens of var. michauxianum have been discovered. The variety is here 
reported as new to the flora. It occurs only on Great Abaco (Correll et al. 42651, 
FTG; Correll & Meyer 44540, 44590, 44609, all FTG). All of the collections were 
from ledges, rocks, or logs in dense coppice vegetation. 

Pteris x delchampsii Wagner & Nauman.—This hybrid between P. vittata L. 
and P. bahamensis (Agardh) Fée occurs on the island of Grand Bahama (Correll 
50507, FTG). The collection is from a single clump in a shaded sinkhole in open 
pinelands, a habitat typical of Florida populations. The plant’s occurrence in the 
Bahamas was to be expected since both parental taxa also occur there. 

Thelypteris hispidula (Decne.) Reed var. versicolor (R. St. John) Lellinger.— 
This species is here reported as new to the Bahamian flora. It occurs on North 
Andros (Hill 3032, FTG) and Mangrove Cay (Popenoe 228, FTG). Both collections 
are from sinkholes in habitats dominated by pine. Suitable habitat exists else- 
where in the Bahamas and this species is expected to spread.—CLIFTON E. NAvu- 
MAN, Fairchild Tropical Garden, 10901 Old Cutler Road, Miami, FL 33156. 


Announcement: 1988 AIBS Meeting 


The annual meeting of the American Fern Society will be held in conjunction 
with the American Institute for Biological Sciences on August 14-18, 1988, at the 
University of California, Davis. Persons wishing to present talks should contact 
Dr. Judith Skog, Biology Dept., George Mason University, Fairfax, VA 22030, 
before 15 January 1988 for information. 

A Fern Foray to Bear Valley, Point Reyes Peninsula, will take place on Sat- 
urday, 13 August 1988. A second trip will be to the University of California 
Botanical Garden and the Tilden Park Botanical Garden on Sunday, 14 August. 
Many native ferns, including Blechnum spicant, Dryopteris expansa, Polypo- 
dium spp., Aspidotis spp., and Pellaea spp. will be seen during the two days. 
Trips will leave from Davis and be restricted to 45 people. No collecting will be 
permitted. Leaders are Alan Smith, Don Macneill, and Tom Lemieux. For more 
information, contact Smith at (415) 642-7890. 


INFORMATION FOR AUTHORS 


Authors are encouraged to submit manuscripts pertinent to pteridology for 
publication 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 pre- 
pared may be returned for revision prior to review. Submit manuscripts in trip- 
licate perciieurs acceptable), including review copies of illustrations. Do not 
send originals of illustrations until they are requested. Use standard 81 by 11 
inch paper of good quality, not “erasable” paper. Double space man 
throughout, including title, authors’ names and addresses, text (including hoteles 
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. 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 mea- 
sures (e.g., distance, elevation, weight) unless quoted or cited from another source 
{e.g., specimen citations). For nomenclatural matter {i.e., synonymy and typifi- 
cation), use one paragraph per basionym (see Regnum Veg. 58:39-40. 1968). Ab- 
breviate titles of serial publications according to Botanico-Periodicum-Huntian- 
um (Lawrence, G. H. M. et al., 1968, Pittsburgh: Hunt Botanical Library). 
References cited only as part of nomenclatural matter are not included in lit- 
erature cited. For shorter notes and reviews, put all references parenthetically _ 
in text. Use Index herbariorum (Regnum Veg. 106:1-452. 1981) for —— : 
of herbaria. 

Illustrations should be proportioned to fit page width with caption on the same > 

e. Provide margins of at least 25 mm on all illustrations. For continuous-tone 
illustrations, design originals for reproduction without reduction or by a uniform — 
amount. In composite blocks, abut edges of adjacent photographs. Avoid com- 

d 42. 


sequence and numbering of figures (and of tables) with order of citation in text. _ 
Explain 


copy in single illustrations or blocks. Coordinate __ 


scales and symbols in figures themselves, not in a Include c 


scale and reference to latitude and longitude in each map. © : 
Proofs and reprint order forms are sent to authors by the printer. ‘anes co 
should send corrected proofs to the editor and eprint orders: to | the printer. 
Seva will be assessed pores h for extensive ; ns made after ty has 
soo ‘ie shashions ot evi or style, conmkt recent iaeues of American Fem | 
_ Journal and The Chicago manual of style, 13th ed. (1982. Chicago: Univ. Ch ah 
Press). Occasionally, departure | from these guidelines may be justified. Authors a 
| eee nuscript — 


AMERICAN a 
FERN Mn 
JOURNAL 


QUARTERLY JOURNAL OF THE AMERICAN FERN SOCIETY 


An Introduction to the Pteridophyte Flora age Finca La Selva, Costa Rica 
ichael H. Grayum and Hugh W. Churchill 73 


The Disposition of Trichopteris (Cyatheaceae) David B. Lellinger 90 
Germination of Helminthostachys Spores Dean P. Whittier 95 
Shorter Notes 

A Binomial for a Common Hybrid Lycopodium Allison W. Cusick 100 


Nomenclatural Notes on Some Ferns of Costa Rica, Panama, and Colombia.—IiI 
David B. Lellinger 101 


Terrestrial Psilotum in East-Central Alabama John D. Freeman 102 
Reviews 106, 107, 108, 108 : 


The American Fern Society 
Council for 1987 


FLORENCE 8S. WAGNER, Dept. of Botany, University of Michigan, Ann Arbor, MI 48109. 


President 
JUDITH E. SKOG, Biology Dept., George Mason University, Fairfax, VA 22030. Vice-President 
W. CARL TAYLOR, Milwaukee Public Museum, Milwaukee, WI 53233. Secretary 


JAMES D. CAPO! . of Botany, University of Tennessee, Knoxville, TN . Sota 
DAVID S. BARRINGTON, Dept. of Botany, University of Vermont, Burlington, od 054 
cords Srecuapee 


JAMES D. MONTGOMERY, Ecology III, R.D. 1, Berwick, PA 18603. aa es Curator 
ALAN R. SMITH, Dept. of Botany, University of California, paqek - 94720. pap Editor 
DAVID B. LELLINGER, Smithsonian Institution, Washington, DC 2 Memoir Editor 
DENNIS Wm. STEVENSON, New York Botanical Garden, Bronx, casi 1045 
prea Eace Forum Editor 
American Fern Journal 
EDITOR 
Mire eee Dept. of Botany, eine of regains 
eley, CA 9472 
ASSOCIATE EDITORS 

GERALD J.GASTONY ............. Dept. of Biology, Indiana University, Bloomington, IN 47401 
CPIRETC PIER CIAUIPIER Dept. of Botany, University of Kansas, 
wrence, KS 66045 

DAVID B. LELLINGER ............... U.S. Nat'l Herbarium NHB-166, Smithsonian Institution, 


ashington, DC 20560 

TERRY R. WEBSTER .... Biological Sciences Group, University of Connecticut, Storrs, CT 06268 

The “American Fern Journal” (ISSN 0002-8444) is an illustrated quarterly devoted to the general 

_ study of ferns. It is owned by the American Fern Society, Se ee erbarium, 

University of Vermont, Burlington, VT 05405. Second-class postage paid at Burlington, VT, and 
additional entry point. 


Oe ae a . ra Re gt 


to 12 months (foreign) after the date of issue, 
and orders for back issues should be addressed to Dr. utes D. Montgomery, Ecology III, R.D. 1, 
Berwick, PA 18603. 
| Ce ee eeeeenesbly shalt be eet tthe Heconds Trow- 
surer. 

Subecriptons $1200 gros, 81.90 net if paid thscooh an apaney teats fee $0.50); sent free to 
_ members of the American Fern ee Sie (eel oes tran + sues - eee Cores Seed 
ae U.S.A, Canada, and Mexico; life membership, $200. 
_ Back volumes 


4070. fa 


cS less, $1.25; 65-80 
: ce opaeen meni ‘Back volumes 1070 et sex $8.00 each; 
_ single back ping. Ten percent discount on orders of six volumes or 


— 


ms totes 


ser ani and tr ne ee 


American Fern Journal 77(3):73-89 (1987) 


An Introduction to the Pteridophyte Flora of 
Finca La Selva, Costa Rica 


MICHAEL H. GRAYUM 
Missouri Botanical Garden, P.O. Box 299, St. Louis, Missouri 63166 
HuGuH W. CHURCHILL 
Department of Botany, University of Vermont, Burlington, Vermont 05405 


Finca La Selva, now more properly known as the La Selva Biological Station, 
is the flagship field station of the Organization for Tropical Studies (OTS) in Costa 
Rica. A comprehensive floristic inventory of La Selva was initiated in 1979 (Ham- 
mel & Grayum, 1982), and a technical flora has recently begun to appear in fascicle 
form in the journal Selbyana (see Wilbur, 1986). The recent completion of the 
pteridophyte treatments for this flora (Grayum & Churchill, 1988a, 1988b; here 
summarized in Appendix 1) permits the present informal introduction and anal- 
ysis. 


THE SITE 


Finca La Selva is located in the Caribbean lowlands of Costa Rica, in the zone 
where the volcanic Central Cordillera begins to rise up out of the swampy coastal 
plain. The Station occupies 1533 hectares at the confluence of the Rios Puerto 
Viejo and Sarapiqui. The elevation on the property ranges from about 35 m 
(where the rivers meet) to 130 m on the higher ridges toward the South Boundary. 
Recent alluvial soils occur along the rivers and lower portions of the major 
quebradas (creeks), but volcanically derived soils predominate elsewhere. The 
major habitat at La Selva is upland primary forest; however, several other habitats 
are represented, including swamp forest, open marshes, streamsides and riparian 
forest, alluvial forest, secondary forest, and weedy areas (see Appendix 1). 

La Selva lies in one of the wettest parts of Costa Rica, receiving an average of 
about 3900 mm of rain per year. Although there isa slight dry season from January 
through April, it is not nearly so pronounced as that experienced on Costa Rica’s 
Pacific slope during these months. 

More detailed descriptions and additional information can be found in nu- 
merous other sources (e.g., Grayum, 1982; Hartshorn, 1983; Hammel, 1986; Wilbur, 
1986 


HISTORY OF PTERIDOPHYTE COLLECTING AT LA SELVA 


The earliest known pteridophyte collections from on or very near the site now 
known as La Selva were made during the “Golden Period” of Costa Rican natural 
history (see Gomez & Savage, 1983), around the turn of the century. Pablo Biolley 
and Henri Pittier, Swiss botanists based in San José, visited the ‘“‘confluence of 
the Rio Puerto Viejo and Sarapiqui” at least twice, in April, 1892 and@aigjgpt 


MAR 9 1988 


GARDEN Lipsrapy 


74 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 3 (1987) 


1893, making general plant collections. Among the ferns collected by Biolley and 
Pittier were two species, Pteris quadriaurita and Pityrogramma tartarea, that 
have not since been recollected from La Selva. 

During most of the first half of the present century, natural history languished 
in Costa Rica for a variety of reasons (Gomez & Savage, 1983), and little or no 
botanical fieldwork was undertaken in the La Selva vicinity. The next pterido- 
phyte collections on record from La Selva are those of Edith Scamman, a fern 
specialist associated with Harvard’s Gray Herbarium, who visited Costa Rica in 
1951, 1953, 1955, and 1956, amassing a total of about 1400 specimens (Tryon & 
Tryon, 1968). Herbarium labels indicate that Scamman collected ferns at La Selva 
(then the property of Dr. Leslie R. Holdridge) on at least one occasion, from 18- 
28 February 1955; however, the exact number of specimens collected at the site 
is difficult to establish from this source. 

By 1967 Finca La Selva had become OTS property, and in the summer of that 
year OTS conducted a field course on the biology of tropical pteridophytes in 
Costa Rica (Mickel, 1967). This course was coordinated by Dr. John T. Mickel, 
and included Drs. Elias de la Sota, Warren H. Wagner, Jr., A. Murray Evans and 
David B. Lellinger as full- or part-time faculty. The list of participants in the 
course, including “ten outstanding students ... selected from nine universities 
across the United States,” now reads like a ““Who’s Who in American Pteridology.”’ 
During a few days at La Selva in mid-August, this group produced the first 
checklist of La Selva pteridophytes (OTS, 1967), accounting for 89 of the 173 
species presently known. De la Sota (1971) later published a list of the 56 epiphytic 
species then known from La Selva. 

During the interval from 1967 to 1979, minor pteridophyte collections were 
made at La Selva by individuals too numerous to mention, including many of 
those involved in the 1967 course. The Costa Rican pteridologist Luis Diego Gomez 
visited the site on many occasions, making several significant collections and 
compiling a checklist, circulated by OTS during the mid- and late 70s, that 
enumerated 100 pteridophyte species. From March to September of 1978, the 
pe author made numerous collections of pteridophytes and other groups at La 

elva. 

The “Flora of La Selva” project was officially inaugurated in 1979, and in the 
summer of that year the first author worked as the first collector for that project. 
Intensive general collecting was pursued at La Selva from that time until about 
1985, and collecting continues, albeit spottily, to the present. A number of col- 
lectors have been involved in this effort, chief among them Barry Hammel, Robert 
L. Wilbur, James Folsom, Tim McDowell, Brian Jacobs, Isidro Chacon, Damon 
Smith, and John Sperry. The first set of this material is deposited at DUKE. 

In March, 1985, June Barcock prepared approximately 100 fern specimens 
during the visit of “Operation Raleigh” to La Selva. This collection, deposited at 
Kew, is significant in that it was made in a poorly explored area of the property; 
moreover, some specimens ined from the canopy by means of a platform 
pelanipe’ by the group. In addition to several rare species, two species new to 

e flora (Trichomanes crispum and Blechnum polypodioides) were collected. 


GRAYUM & CHURCHILL: LA SELVA PTERIDOPHYTES 75 


TABLE 1. Total Number of Pteridophyte Species in Selected Floras. 


Place Area (km?) Pterid. spp. Authority 

California 411,013 86 Grillos, 1966 

exas 692,405 107 Correll, 1956 
Chihuahua 245,612 126 Knobloch & Correll, 1962 
Florida 151,670 135 Lakela & Long, 1976 
Zambia 752,614 146 Kornas, 1979 
La Selva 15 173 Grayum & Churchill, 1988a, b 
Hong Kong 1,034 175 Edie, 1978 
Puerto Rico 8,897 263 Kepler, 1975 
Australia 7,686,849 416 Jones & Clemesha, 1981 
Jamaica 10,962 579 Proctor, 1985 
Chiapas ca. 74,000 609 Smith, 1981 
Costa Rica 50,900 ca. 900 Grayum, unpubl. data 


THE PTERIDOPHYTE FLORA: GENERAL REMARKS 


At the present time, 173 species of pteridophytes in 45 genera are known from 
Finca La Selva, ranking this tiny site respectably high among several much larger 
political units (see Table 1) selected more or less randomly on the basis of ease 
of extraction of comparable data. Seven additional species, including one ad- 
ditional genus (Dicranopteris), have been collected immediately adjacent to La 
Selva and have been considered likely hypotheticals and treated in full for the 
Flora. Two further species, Antrophyum lineatum and Grammitis turrialbae, are 
vouchered for La Selva by a single correctly identified specimen each; however, 
both are mid- to high-elevation species and label mixups are strongly suspected. 

Particularly surprising is the comparison of the La Selva pteridoflora with that 
of Florida (Table 1), the most tropical of the continental 49 states and about 10,000 
times the size of La Selva, but with 21.5% fewer pteridophyte species. 

In view of the thoroughness with which La Selva has been botanically inven- 
toried, the above figures probably represent a reasonably close approximation 
of the actual totals. Clearly, however, they are not exact, nor can we ever hope 
to make them so. Although the area of La Selva is quite small, the terrain is 
rugged and complex. Large areas of the reserve are inaccessible by trail, and 
most of these have been only cursorily explored. Certain parts of the so-called 
“Western Annex,” added to La Selva in 1982, remain completely unexplored. 

More intensive collecting along creeks and on ridges in the more remote and 
relatively poorly explored southern portions of La Selva will certainly yield new 
fern records (as demonstrated by the Barcock effort discussed previously). This 
seems especially likely when one considers the highly localized distribution 
patterns characteristic of many plant species in the La Selva rain forest, contrary 
to what one might predict of this superficially uniform environment. For example, 
33 (about 19%) of the 173 pteridophyte species presently known from La Selva 
are classed as “very rare” (known from just one or two collections; see Appendix 
1), including such conspicuous ferns as Metaxya rostrata and Thelypteris gigan- 


76 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 3 (1987) 


tea; an additional 45 species (26% of the total) are considered rare (known from 
3-5 collections). Thus about 45% of the La Selva fern flora consists of species 
unlikely to be encountered by the short-term visitor (interestingly, roughly the 
same fraction unaccounted for by the 1967 course). These figures are surprising 
in view of the “notorious” ease with which fern spores are dispersed in the air 
(Smith, 1972; Gentry & Dodson, 1987); however, comparable numbers for the 
likewise air-dispersed Orchidaceae at La Selva may be higher yet (J. Atwood, 
pers. comm.). Hammel (1986) reported that about 75% of the La Selva species in 
six angiosperm families with more conventional seed-dispersal modes “are known 
from a few individuals or from a few small populations.” 

The inherently dynamic nature of natural floras p ts an additional obstacle 
to a precise inventory of the La Selva pteridophytes. Certain habitats seem es- 
pecially subject to ostensibly temporary colonization by basically extralimital 
species, pteridophytes and spermatophytes alike. Such habitats include recently 
disturbed or weedy sites in primary or secondary forest and, especially, the banks 
of the Rios Puerto Viejo and Sarapiqui, regularly disturbed by landslides and 
diaspore-bearing floodwaters. The distribution at La Selva of species such as 
Blechnum fraxineum, Dennstaedtia cicutaria, and Thelypteris resinifera suggests 
ephemeral occurrence in the latter habitat. This may also be the case with Pit- 
yrogramma tartarea and Pteris quadriaurita, not recollected in the La Selva 
vicinity for nearly a century. 

The air-dispersed diaspores of pteridophytes perhaps render the La Selva 
pteridophyte flora more dynamic than the spermatophyte flora. This is suggested 
by recent discoveries of species new to the flora in heavily collected sites, e.g., 
Stigmatopteris longicaudata along the Quebrada El Taconazo and, most dra- 
matically, Ophioglossum reticulatum in the laboratory clearing. 

A particularly striking feature of the La Selva fern flora that is also probably 
attributable to the high vagility of air-dispersed plants is the dearth of undescribed 
species. In fact, just one La Selva pteridophyte species (an Elaphoglossum) is 
believed to be new to science (although a few others, such as Diplazium pactile 
and D. striatastrum, were described within the last decade). Similarly, only a 
single La Selva orchid species (out of 113) is apparently undescribed (J. Atwood, 
pers. comm.). This is markedly in contrast with the situation for spermatophytes 
in general: for example, 12 (about 10%) of the 123 species in the subset of six 
angiosperm families analyzed by Hammel (1986) were described by Hammel 
himself, and over 50 new species of seed plants have been described from La 
Selva altogether (Wilbur, 1986). 

e La Selva pteridoflora also includes but two apparent country records 
(Danaea grandifolia and Bolbitis aliena), contrasting with the spermatophyte flora 
(Grayum & Hammel, 1982) in this respect as well. 

Before proceeding with the analysis of the La Selva pteridoflora, an important 
taxonomic qualification is necessary. The number of pteridophyte genera attrib- 
uted to La Selva and (to a lesser extent} the number of species depends on one’s 
iexonoune viewport: Pteridophyt toriously intractable as regards familial 
and generic classification, For the purposes of the La Selva flora, rather broad 
circumscriptions were opted for in most cases. Without narrowing these very 


GRAYUM & CHURCHILL: LA SELVA PTERIDOPHYTES 77 


TABLE 2. Most Diverse Pteridophyte Genera at Three Neotropical Sites. 


LS' BCI RP 
Genus Spp. Genus Spp. Genus Spp. 
Thelypteris 18 Polypodium 13 Thelypteris 11 
Polypodium 17 Trichomanes 9 olypodium 7 
Trichomanes 14 Adiantum 9 Adiantum 6 
Asplenium qi Thelypteris 8 Asplenium 5 
Selaginella 9 Selaginella 6 Diplazium 5 
ectaria 9 Asplenium 6 Tectaria 4 
Diplazium 8 Pteris 4 Selaginella 3 
Adiantum 7 Elaphoglossum 3 Trichomanes 2 
Elaphoglossum 6 Elaphoglossum 3 
Stigmatopteris 3 


1 Acronyms and data sources: LS = La Selva (Grayum & Churchill, 1988a, b); BCI = Barro Colorado 
Island (Croat, 1978); RP = Rio Palenque (Dodson & Gentry, 1978). 


radically, well over a dozen pteridophyte genera (e.g., Lycopodiella, Huperzia, 
Trichipteris, Megalastrum, Peltapteris, Gochlidium, Olfersia, Campyloneurum, 
Phlebodium, Microgramma, Niphidium, Pecluma, Pleopeltis, Christella, Go- 
niopteris, Macrothelypteris, Meniscium, Ananthacorus) could be added to the 
La Selva total of 45. An additional species could be appended, raising the total 
to 174, were Asplenium auritum recognized as distinct from A. cuspidatum (both 
entities occur at La Selva). 


COMPOSITION OF THE FLORA 


Of the 173 La Selva pteridophyte species, 160 (in 43 genera) belong to the 
Divison Polypodiophyta and 127 (in 35 genera) to the broadly-circumscrib 
Polypodiaceae, including three weedy species introduced from the Old World 
(Nephrolepis multiflora, Thelypteris dentata, and T. torresiana). Polypodiaceae 
thus reigns as the largest family in the La Selva vascular flora, surpassing the 
largest angiosperm family, Orchidaceae, which comprises 113 species {and two 
fewer genera; Atwood, 1986). Furthermore, ferns are conspicuously more abun- 
dant than orchids (from an understory perspective, at any rate). Pteridophyte 
species account for about 9% of the La Selva vascular flora (based on an estimated 
total of 1900 species; see Wilbur, 1986}, which falls in the range of 5-10% given 
by Smith (1981) for wet tropical forests. 

The most diverse pteridophyte genera (hypotheticals excluded) at La Selva are 
Thelypteris, Polypodium, and Trichomanes, which would rank among the ten 
largest genera of La Selva vascular plants based on data presented by Hammel 
& Grayum (1982). Table 2 lists the largest genera of pteridophytes at La Selva 
and at two other neotropical sites for which data are available. Pertinent infor- 
mation regarding all three sites is presented in Table 3; additional data may be 
found in Croat (1978), Dodson and Gentry (1978), and Hammel (1986). All three 
sites are in the lowlands, at roughly the same elevation. Barro Colorado Island 


78 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 3 (1987) 


Relevant Abiotic Characteristics and Total Number of Pteridophyte Species for Three 
Neotropical Sites. 


recip. 
Site Size (ha) Elev. (m) (mm/yr) Temp. (°C) No. spp. 
i 167 150-220 2980? 20-28 83° 
BCI 1560 25-165 2750 21-32 104 
LS 1533 37-130 3900 24 173 


1 RP = Rio Palenque; BCI = Barro Colorado Island; LS = La Selva. 
? Datum from Dodson, et al. (1985). 

* Incorporating updates from A. Gentry (pers. comm.). 

* Mean annual temperature. 


(BCI) in Panama is about the same size as La Selva; however, the Rio Palenque 
reserve in Ecuador is an order of magnitude smaller. Both of the last-named two 
sites are drier and more seasonal than La Selva, BCI markedly so. 

Atwood (1986) judged that ‘La Selva is most similar generically to Rio Palen- 
que” in its orchid flora (as compared with BCI and Volcan Mombacho, Nica- 
ragua); Hammel (1986), on the other hand, found La Selva to “share more genera 
and species” of six angiosperm families with BCI than with Rio Palenque. As 
regards pteridophytes, La Selva shares the same number of genera (34) with BCI 
and Rio Palenque. Five additional genera (Botrychium, Anemia, Lygodium, Max- 
onia, and Ceratopteris) occur at Rio Palenque, whereas BCI boasts nine genera 
not known from La Selva (Lygodium, Schizaea, Cnemidaria, Acrostichum, Cer- 
atopteris, Maxonia, Triplophyllum, Dictyoxiphium, and Salvinia). 

Twenty-eight pteridophyte species are shared among all three of the sites (see 
Appendix 1). Sixty-five of the 104 pteridophyte species at BCI (about 62.5%) are 
shared with La Selva, while Rio Palenque shares 48 species (57.8% of the total 
83). Thirty-seven of the species shared by La Selva and BCI do not occur at Rio 
Palenque. More surprising, perhaps, is the fact that 20 species are shared by La 
Selva and Rio Palenque but do not occur at the geographically intermediate BCI 
station. Moreover, just four species (Ceratopteris pteridoides, Asplenium laetum, 
Maxonia apiifolia, and Saccoloma elegans) are common to BCI and Rio Palenque 
but absent from La Selva. 

Similarly ambiguous results are obtained when Sorensen’s coefficient of sim- 
ilarity (Sarmiento, 1975) is calculated (at the species level) for these three sites. 
Comparison of the La Selva and BCI pteridofloras yields the highest figure, 47; 
the coefficient for La Selva/Rio Palenque is 38. However, when Rio Palenque is 
compared with BCI, the value is lower still (32). 

We believe that the La Selva pteridoflora is probably more similar to that of 
Rio Palenque, but that this relationsh ipi hat ob ed by the much smaller 
size of the latter site and the consequent absence of many species that may occur 
(or have occurred) in the near vicinity. The latter factor casts a long shadow over 
absolute-number comparisons among the three sites. La Selva and BCI are com- 
parable in size and elevation, and the fact that La Selva has 65% more pteri- 
dophyte species than BCI must be explainable in terms of the significant differ- 


GRAYUM & CHURCHILL: LA SELVA PTERIDOPHYTES 79 


ences in amount and distribution of precipitation. Gentry (e.g., 1982) has repeatedly 
emphasized the importance of precipitation in explaining phytogeographical pat- 
terns, and noted that the Rio Palenque flora is more like that of Central American 
wet forests than that of nearby but more seasonal forests in western Ecuador. 
Lellinger (1975) had previously arrived at essentially the same conclusion re- 
garding the Choco pteridophyte flora. 

Although the total annual precipitation at Rio Palenque is not much higher 
than that at BCI, the effects of the dry season at the former site are somewhat 
mitigated by prevailing cool and cloudy conditions (A. H. Gentry, pers. comm.). 


GROWTH HABIT 


The only aspect of growth habit that we investigated was the distribution of 
epiphytic vs. terrestrial (or epipetric) species. We might have predicted, for ex- 
ample, that the higher diversity of such groups as pteridophytes and Araceae 
(Grayum, 1982) at La Selva, as compared with BCI, was due to conditions at the 
former site being more conducive to epiphytes in general on account of the higher 
precipitation (de la Sota, 1971; Gentry & Dodson, 1987). In fact, epiphytes account 
for nearly twice as large a percentage of the total vascular flora at La Selva than 
at BCI (Gentry & Dodson, 1987). Surprisingly, however, the percentage of epi- 
phytic species in the La Selva pteridoflora (41.9%) is not significantly different 
from that at BCI (40.4%), nor in all probability from that at Rio Palenque (35.97%). 
Thus, whereas many seed plant taxa become more predominantly epiphytic in 
wet forest as opposed to moist forest, these preliminary data suggest that pteri- 
dophytes apparently do not. The dramatic increase in epiphytic pteridophytes 
with increasing elevation, discussed in a later section, is perhaps due to some 
factor other than absolute precipitation; indeed, the average annual rainfall at 
the cloud-forest reserve at Monte Verde, Costa Rica, is actually considerably 
lower than at La Selva (G. S. Hartshorn, pers. comm.). 


DISTRIBUTION OF LA SELVA PTERIDOPHYTES BY HABITAT 


Over half of the 123 spermatophyte species in Hammel’s (1986) study occurred 
in upland primary forest. This is also the favored pteridophyte habitat at La Selva, 
the corresponding figure being 50.0% (86 species). The most common and char- 
acteristic species for each La Selva habitat are listed below, along with the 
percentage of the total pteridoflora represented there (these figures are not ad- 
ditive, since many fern species occur in two or more habitats). 

Upland Primary Forest.— Danaea wendlandii, Trichomanes collariatum, T. 
godmanii, Cyathea multiflora, Adiantum obliquum, Asplenium cirrhatum, A. cus- 
pidatum, Elaphoglossum palmense, E. peltatum, Grammitis linearifolia, G. ser- 
rulata, Hypolepis hostilis, Lomariopsis fendleri, Polybotrya caudata, P. cervina, 
Polypodium triseriale, Pteris pungens, Saccoloma inaequale, Salpichlaena vol- 
ubilis, Thelypteris curta, T. lingulata (50.0% ). 

Alluvial Forest.—Selaginella eurynota, S. flagellata, Adiantum tetraphyllum, 


80 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 3 (1987) 


Dennstaedtia bipinnata, Polypodium percussum, Tectaria nicotianifolia, T. rivalis 
(26.87%). 

Riparian Habitats.—Selaginella atirrensis, S. umbrosa, Cyathea trichiata, 
Blechnum occidentale, Tectaria mexicana, Thelypteris angustifolia, T. balbisii, 
T. francoana, T. torresiana (21.2%). 

Secondary Growth.—Selaginella arthritica, Alsophila cuspidata, A. firma, An- 
trophyum lanceolatum, Bolbitis portoricensis, Cyclopeltis semicordata, Dicrano- 
glossum panamense, Diplazium striatastrum, Nephrolepis biserrata, Polypodium 
aureum, P. ciliatum, P. phyllitidis, P. lycopodioides, Tectaria incisa, Thelypteris 
nicaraguensis (19.4%). 

wamp Forest.—Cyathea ursina, Adiantum tetraphyllum, Diplazium lindber- 
gii, D. macrophyllum, Polypodium occultum, P. sphenodes, Tectaria athyrioides 
(13.4%). 

Quebradas.— Selaginella umbrosa, Trichomanes diaphanum, T. krausii, 
Asplenium otites, A. repandulum, Bolbitis nicotianifolia, Tectaria plantaginea 
(12.37%). 

Weedy Areas.— Adiantum latifolium, A. petiolatum, Hemionitis palmata, Hy- 
polepis repens, Pityrogramma calomelanos, Thelypteris dentata, T. nicaraguen- 
sis (6.7%). 

Open Marshes.— Nephrolepis biserrata, Thelypteris serrata (1.2%). 

Additional data can be found in Appendix 1. 


DISTRIBUTION OF LA SELVA PTERIDOPHYTES IN CosTA RICA 


Dogma has it that the pteridoflora of lowland tropical sites is boring and not 
very diverse, comprising mainly common and widespread lowland species (Wag- 
ner & Gomez, 1983). This is not strictly the case at La Selva, which is relatively 
wet and cool due to its location at the foot of the Central Cordillera. This proximity 
to the mountains makes La Selva an atypical lowland site when compared to 
many other parts of the lowland tropics. 

Based on data culled from the MO and CR herbaria and summarized by 
Grayum and Churchill (1988a, b), only 30.1% of the species in the La Selva 
pteridoflora are restricted (in Costa Rica) to the lowlands {i.e., elevations of less 
than 1000 m); 10.6% are restricted to elevations of less than 500 m; 55.2% range 
into mid-elevations (1000-2000 m); and 14.5% to higher elevations (above 2000 
m). Six species (Lycopodium cernuum, Selaginella flagellata, Asplenium cuspi- 
datum, A. serra, Blechnum occidentale, and Elaphoglossum peltatum) ascend to 
3000 m or higher. 

Only 17.9% of La Selva pteridophyte species are confined, in Costa Rica, to 
the Caribbean slope (not prisingly, this total includes nearly 75% of the species 
restricted to below 500 m). 

Since we know of no comprehensive enumeration of pteridophyte species for 


GRAYUM & CHURCHILL: LA SELVA PTERIDOPHYTES 81 


elevation cloud forests (Gentry & Dodson, 1987), where such predominantly epi- 
phytic pteridophyte genera as Lycopodium, Hymenophyllum, Blechnum, Ela- 
phoglossum, and Grammitis have many species in Costa Rica. Wagner and Gomez 
(1983) maintain that 70% of the total Costa Rican pteridophyte flora consists of 
epiphytic species, compared with only 41.9% at La Selva. However, it has also 
been suggested (Gentry & Dodson, 1987) that this increase in epiphytic species 
with elevation is countered by a corresponding decrease of terrestrial species. 
Mainly terrestrial genera such as Selaginella, Danaea, Adiantum, and Tectaria 
are perhaps most diverse in the humid lowlands. Trichomanes, though largely 
epiphytic, has many more terrestrial species in the Costa Rican lowlands than 
at higher elevations (however, the reverse may be true of other genera, such as 
Elaphoglossum). Tree ferns are often associated with cloud forests; for example, 
Lee et al. (1987) report that tree-fern diversity in the Monte Verde area is highest 
in “montane rainforest” (1535-1610 m) and declines at lower elevations. However, 
the total number of tree-fern species in their entire study area (7) is one less than 
the La Selva total (including Metaxya). 

At the country level, it may indeed be true that pteridophytes in general are 
most diverse at mid-elevations, as is apparently the case in Panama (Lellinger, 
1985). Still, it is perhaps of interest in this connection that the estimated percentage 
of pteridophytes in the La Selva vascular flora (9%) is identical to that in the 
total Costa Rican flora (9.0%, based on 900 pteridophyte species out of an esti- 
mated 10,000 vascular plant species). Dogma also has it that endemism and in- 
sularity increase as elevation increases. Thus, we might predict that pteridophyte 
diversity at a particular mid-elevation site would not necessarily exceed that at 
La Selva, even given that pteridophytes were more diverse at mid-elevations on 
the country level. Floristic data from a mid-elevation cloud forest locale such as 
the Monte Verde reserve in Costa Rica would shed much light on this situation. 

Elevation is not, of course, the only factor affecting fern distribution. In a 
previous section of this paper, the important role of precipitation was alluded to. 
Other factors are not well understood. For example, Finca El Bejuco, a small 
biological field station only about 7 km away from La Selva and at approximately 
the same elevation, harbors numerous pteridophyte species never collected from 
La Selva (among them Dicranopteris flexuosa, Trichomanes ankersii, T. poly- 
podioides, T. punctatum, T. galeottii, Cyathea stolzei, Elaphoglossum backhou- 
sianum, Thelypteris leprieurii, and Triplophyllum funestum; A. R. Smith, in litt.). 
This is perhaps not very surprising, inasmuch as localized distributions of fern 
species within the boundaries of La Selva itself have already been disc 
Local climatic or edaphic factors or chance vagaries of dispersal may be respon- 
sible for these phenomena. 


BIOGEOGRAPHICAL AFFINITIES OF THE LA SELVA PTERIDOFLORA 


The following analysis is based on data presented in Grayum and Churchill 
(1988a, 1988b). 

About 83.6% of La Selva pteridophyte species are here considered widespread, 
that is, having neither their northern nor southern limit of distribution in Costa 


82 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 3 (1987) 


Rica. About 7% of these (10 species) are pantropical, occurring also in Asia and/ 
or Africa. Approximately */ of the New World species range into the West Indies, 
the remainder being strictly continental. Twenty-six (92.9%) of the 28 more nar- 
rowly distributed species reach their northern limit in Costa Rica, whereas but 
two (7.1%) apparently reach their southern limit here (Pseudocolysis bradeorum 
and Tectaria rufovillosa). 

Not a single La Selva pteridophyte species is known to be endemic to Costa 
Rica. In contrast, Hammel (1986) reported that about 13% of the angiosperm 
species in his study “appear to be endemic to Costa Rica.” Endemism is in general 
low in ferns as compared with spermatophytes (Smith, 1972); however, Lellinger 
(1975) reported 17 endemic species from the Chocd—probably owing to the fact 
that this region is biogeographically, rather than politically, defined. 

Thus, the La Selva pteridoflora comprises mostly very wide-ranging species, 
yet clearly shows evidence of southern rather than northern affinities. Southern 
affinities were also evident in the six angiosperm families analyzed by Hammel 
(1986), who commented that such a relationship “should come as no surprise.” 
This is certainly the case with pteridophytes, especially considering the previously 
discussed connections between the pteridofloras of Central America and the 
Choco region. 

Endemism is very low or nonexistent among La Selva pteridophytes, which 
again probably relates to air-dispersal of fern diaspores. 


PHENOLOGY 


Phenological studies on ferns in general are rare (Wagner & Gomez, 1983), and 
on La Selva ferns rarer still. Despite their abundance and diversity at La Selva, 
pteridophytes have been the subject of exceedingly few ecological or demo- 
graphic studies. Herbarium data suggest that Selaginella eurynota is seasonally 
reproductive, producing strobili mainly from November to January. Ongoing 
studies by J. Sharpe (pers. comm.) have shown that all five La Selva Danaea 
species exhibit pronounced seasonality in the maturation of their fertile fronds. 
On the other hand, Moran (1986) suggested that Olfersia (Polybotrya) cervina 
may be continuously reproductive at La Selva. The La Selva Biological Station 
is obviously an excellent place for studies of this sort, which would help fill a 
significant void in our understanding of pteridophyte reproductive biology. 


ACKNOWLEDGMENTS 


We thank Drs. Alwyn H. Gentry, David B. Lellinger, John T. Mickel, Robbin C. Moran, and Alan 
R. Smith for critically reviewing this paper. The following authorities provided determinations for 


(Ctenitis, Lastreopsis, Stigmatopteris, Thelypteris). We gratefully acknowledge these contributions, 
well as general determinations of La Selva material by Dr. David B. Lellinger. All final taxonomic 
and nomenclatural decisions are our own, however, along with the responsibility for any errors or 
misjudgments. 
We are further indebted to Drs. Barry E. Hammel, E. Hennipman, Pablo Sanchez V., Rolla M. 
Tryon, Melvin Turner, and Robert L. Wilbur for valuable advice and assistance. 


GRAYUM & CHURCHILL: LA SELVA PTERIDOPHYTES 83 


LITERATURE CITED 


tTwoop, J. T. 1986. Flora of La Selva, Costa Rica: Orchidaceae. Amer. J. Bot. 73:749 
ne ELL, D. S. 1956. Ferns and fern allies of Texas. Renner, Texas: Texas Nesaoich Foundation. 
Croat, T. B. 1978. Flora of Barro Colorado saloded Stanford: Stanford University Press. 
Dopson, C. H. and A. H. Gentry. 1978. Flora of the Rio Palenque Science Center. Selbyana 4:1- 


_ and F. M. VALVERDE. 1985. La flora de Jauneche. Quito: Banco Central de 


Ecuador. 

Epiz, H. H. 1978. Ferns of Hong Kong. Hong Kong: Hong Kong University Press. 

GENTRY, A. H. 1982. Phytogeographic patterns as evidence for a Choco refuge. Pp. 112-135 in 
Biological diversification in the tropics, ed. G. T. Prance. New York: Columbia University 
Press 


and C. H. Dobson. ned a and biogeography of neotropical vascular plants. Ann. 
Missouri Bot. Gard. 74:2 Ee 
GOMEz, . hee and J. M. SAVAGE. pet Searchers on that rich coast: Costa Rican field biology, 1400- 
0. Pp. 1-11 in Costa ical natural history, ed. D. H. Janzen. Chicago: University of 
endl Press 
Grayum, M. H. 1982. “The aroid flora of Finca La Selva, Costa Rica: a lowland wet forest locality. 
Aroideana 5:47-59. 
] H. W. CHURCHILL. 1988a. The vascular flora of La Selva Biological Station, Costa Rica. 
Lycopodiophyta. Selbyana (in press}. 
and —————. 1988b. The vascular flora of La Selva Biological Station, Costa Rica. Poly- 


sidiehye: momar ess). 
GRILLOs, ss J. 1966. Ferns and fern allies of California. Berkeley: University of California Press. 
HAM mee 1986. Characteristics and phytogeographical analysis of a subset of the flora of La 
va (Costa Rica). Selbyana 9:149-155. 
and M. H. GrayuM. 1982. emey) report on the flora of La Selva field station, Costa 
Ann. Missouri Bot. Gard. 69: 
Hien: G. S. 1983. Plants: paaphne ey ok 418-183 in Costa Rican natural history, ed. D. H. 
zen. Chicago: University of Chicag' 
Jones, D. oe nd S. C. CLEMESHA. 1981. pci sen and fern allies. Sydney: A. H. & A. W. 


Re 
KEPLER, A. K. 1975. Common ferns of Luquillo Forest, Puerto Rico. San Juan: Inter American 
University Press. 
KNosiocu, I. W. and D. S. CorrELL. 1962. Ferns and fern allies of Chihuahua, Mexico. Renner, 
Texas: Texas Research Foundation. 
Korna§, J. 1979. Seen hors ecology of the pteridophytes in Zambia. Warszawa: Panstwowe 
Wydawnictwo Na 
LaKELA, O. and R. W. Lona. 60a Ferns of Florida. cg Banyan Books. 
LEE, M. A. B., P. A. BuRROwES, J. E. FAUTH, and J. C. KoELLA. 1987. The distribution of tree ferns 
along an altitudinal gradient in Monteverde, a Rica. Brenesia (in p: 
LELLINGER, D. B. 1975. A phytogeographic analysis of C Chocé pteridophytes. Fern Gas. 11:105-114. 
———. 1985. The distribution of Panama’ s pteridophytes. Pp. 43-47 in The Sooke and ees 
history of Panama: La botanica e historia natural - eae, eds. W. G. D’Arcy and M 
Correa A. Monogr. Syst. Bot. Missouri Bot. Gard. 1 
MICKEL, < T. 1967. An advanced course in pteridophyte in Costa Rica. Amer. Fern J. 57: 
145-161. 
Moran, R. C. 1986. The neotropical fern genus Olfersia. Amer. Fern J. 76:1 
OTS (ORGANIZATION FOR TROPICAL STUDIES). 1967. Biology of tropical ridges Unpublished 


neues 2: 233-251. 
Smit, A. R. 1972. Comparison of fern and flowering plant distributions with some evolutionary 


interpretations for ferns. Biotropica 4:4- 


B4 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 3 (1987) 


. 1981. Pteridophytes. Pp. 1-370 in Flora of Chiapas, Part 2, ed. D. E. Breedlove. San 
Francisco: California Academy of Sciences. 
Sora, E. DELA. 1971. El epifitismo y las Pteridofitas en Costa Rica. Nova Hedwigia 21:401-465. 
TRYON, R. M. and A. F. TRYON. 1968. Edith Scamman (1882-1967). Amer. Fern J. 58:1-4. 
Wacner, W. H. and L. D. Gomez. 1983. Pteridophytes (helechos, ferns). ~ 311-318 in Costa Rican 
natural history, ed. D. H. Janzen. Chicago: University of Chica 
Witsur, R. L. 1986. The vascular flora of La Selva Biological nar Seah ae Rica. Introduction. 
Selbyana 9:191. 


APPENDIX 1. CHECKLIST OF THE PTERIDOPHYTES OF LA SELVA 


Ab 
Species Habit Habitat dance Distr. 


Division Lycopodiophyta 
Family Selaginellaceae 
TEl 


Selaginella anceps C. Presl 3 
Bebuine ie rthritica Alston Tl SAd 4 BCI 
ti Hie Tl R 4 
Selosinelia bombycina att Ee 2 
Selagi ta A. Brau dw 4 
Selaginella flagellata Spring TLEUI Ad 4 BCI 
Selaginella oaxacana Sp 3 
Selaginella silvestris Aspl. Tl Ad 1 RP 
Selaginella umbrosa Lemaire ex Hieron. ELTI RQ a 
Family Lycopodiaceae 
ycopodium cernuum L. Tl 1 BCI 
Lycopodium dichaeoides Maxon Ce P 1 
Lycopodium dichotomum Jacq. Ce PA z BCI 
Lycopodium linifolium L. CUe ASPw 3 RP 
Division Polypodiophyta 
Family Ophioglossaceae 
Ophioglossum reticulatum L. Te Ww 1 BCI 
Family Marattiaceae 
Danaea cuspidata Liebm Te P 1 
Danaea elliptica J. E. Smith Te Pr 2 
Danaea grandifolia L. Underw. Te PwrZ 1 
Danaea nodosa ([L.) J. E. Smith Te Pw 3 BCI 
Danaea wendlandii Reichb. f. Te P 4 
Gleicheniaceae 
Dicranopteris pectinata (Willd.) L. Underw. R (1) BCI, RP 
Gleichenia bifida (Willd.) Spring Rd 2 
Hymenophyllaceae 
Hymenophyllum brevifrons Kunze Cl P 1 BCI 
Hymenophyllum hirsutum (L.) Sw LI P 2 
Hy. yllum maxonii C ex Morton Ul P 3 
H hyllum polyanthos (Sw.) Sw GLI P 3 
Trichomanes angustifrons (Fée) W. Boer Ul PA 2 
beanie ankersii i Parker ex Hook. & Grev. Ul Pr (1) 
Trichom Ih Bosch HI P 4 RP 
Pecunia crispum L. Ce P 1 


GRAYUM & CHURCHILL: LA SELVA PTERIDOPHYTES 


APPENDIX 1. CONTINUED 


Abun- 
Species Habit Habitat dance Distr. 
Trichomanes curtii Rosenstock Ul Ls 1 
Trichomanes diaphanum H.B.K. UEl QPw 3 
Trichomanes diversifrons (Bory) Mett. Te PrA 2 BCI 
Trichomanes ekmanii W. Boe Ul P 2 BCI 
Trichomanes elegans Rich. Te Pw 2 
Trich nes godmanii Hook UI P 4 CI 
Trichomanes krausii Hook. & Grev. UEl Qs 3 BCI, RP 
Trichomanes membranaceum L. UEl Pw 3 Pp 
Trichomanes osmundoides DC. Te R (1) 
Trichomanes pinnatum Hedwig Te rr 1 BCI 
Trichomanes rigidum Sw. Te P 1 
Trichomanes tuerckheimii Christ Ul Pr i 
Metaxyaceae 
Metaxya rostrata (H.B.K.) C. Pres! Ts Pr 1 BCI 
Cyatheaceae 
Alsophila cuspidata (Kunze) Conant Ta SP 4 BCI, RP 
Alsophila firma (Baker) Conant Ta S 4 
Cyathea microdonta (Desv.) Domin Ta RdS 2 BCI 
Cyathea multiflora J. E. Sm Ta Pw 4 
Cyathea schiedeana (C. Presl) Domin Ta Z 1 
Cyathea trichiata (Maxon) Domin Ta R 3 BCI, RP 
Cyathea ursina (Maxon) Lellinger Te Z 2 
Polypodiaceae 
Adiantum latifolium Lam. Tl WRdPw 4 RP 
Adiantum macrophyllum Sw. Ts 2 RP 
Adiantum obliquum Willd. Tsl id 4 BCI 
iantum oe saison ee Desv. Tls WRd 4 BCI, RP 

Adia ook. Ts R 1 I 
Adbeiten pares his H. & B. ex Willd. Tsl ZA 3 BCI, RP 
Adiantum wilsonii Hook. Tl R 1 
Anetium citrifolium (L} Spl Ul PZ 3 BCI, RP 
Antrophyum cajenense ([ peers Spreng. Us Ly 2 
Antrophyum lanceolatum (L.) Kaulf. Us SP 3 
Antrophyum lineatum (Sw.) Kaulf. Us P : 
Asplenium abscissum Willd. Ee Q 1 
Asplenium cirrhatum Rich. ex Willd. LEe P 3 

Asplenium cuspidatum CUe PSA 4 BCI 
Asplenium falcinellum ine UCe r 3 BCI 
Asplenium formosum Willd. LEe R 2 
Asplenium holophlebium Baker Ul Pw - 
Asplenium otites Link Ee Q : 
Asplenium pteropus Kaulf Ue Pw 1 BCI, RP 
Asplenium re um Kunze Es Q 3 
Asplenium serra Langsd. & Fisch. UCs 4 3 
Asplenium serratum UCe P 5 BCI, RP 
Blechnum fraxineum wit Te R 1 
Blechnum occidentale L. Te Rd 3 
Blechnum polypodioides Raddi Te Pd 1 


86 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 3 (1987) 


APPENDIX 1. CONTINUED 


Abun- 
Species Habit Habitat dance Distr. 
Bolbitis aliena (Sw.) Alston TEs QPw t 
Bolbitis nicotianifolia (Sw.) Alston Es 3 BCI, RP 
Bobitis nicotienifolia ew. ) — HI PwZ 2 
) Hennipma TEs QSA 3 BCI 
Ctenitis sloanei (Poeppig ex Sprengel) eas TLe RAd 1 BCI 
Ctenitis subincisa (Willd.) Ching Te SAw 2 RP 
Cyclopeltis semicordata (Sw.) J. Smith Te Ss 4 BCI, RP 
Dennstaedtia bipinnata (Cav.) Maxon Tl AdS 3 
Dennstaedtia cicutaria (Sw.) T. Moore Tl R 4 BCI, RP 
Dennstaedtia obtusifolia (Willd.) T. Moore Ts Pw 2 RP 
ene glossum panamense (C. Chr.) L. D. 
omez Ue SA 3 BCI 
ne ae truncatula (Sw.) Fe Smith Te Aw 3 RP 
Diplazium cristatum (Desr.) Alsto Te Aw 1 RP 
pcupiuel grandifolium (Sw.) i Te R + BCI 
plazium ingens Christ Te Z 1 
eieee lindbergii (Mett.) Christ Te Zz 3 
aarraan pia sa fea Kunze Te R 2 
macrophyllum Desv. Te ZPw 3 
areas pactile Etter Te ye 2 
Diplazium striatastrum Lellinger Ta SAZ 4 RP 
Elaphoglossum amygdalifolium (Mett.) Christ Ul Aw Zz RP 
Elaphoglossum herminieri(Bory ex Fée)T. Moore Cs P 3 BCI, RP 
Elaphoglossum latifolium (Sw.) J. Smith Gs P 8 
Elaphoglossum palmense Christ CUs Bs 4 
Elaphoglossum sath ia ) Urban Cl P a 
Elaphoglossum sp. no’ Us Pw 3 
— nears sts ) Steudel Ce PA 4 
rammi serrulata (Sw w.) Sw CLI PA 3 
rammitis Seymour Us Pp s 
Hecistopteris pumila (Sprengel) J. Smith CH P 2 
Hemidictyum marginatum (L.} C. Presl Te R 7 BCI, RP 
Hemionitis palmata L. TLe WAd 4 
ivealge hostilis (Kunze) C. Presl a4 Pd 4 
Hypolepis repens C. Presl Tl WAd 4 
Lastreopsis exculta (Mett.) Tindale subsp. 
guatemalensis (Baker) Tindale Ts R 3 
indsaea lancea (L.) Bedd. var. lancea Ts Pr 1 
Lindsaea quadrangularis Raddi subsp 
ata Kramer Ls Prw 1 
Lomariopsis fendleri D. Eaton HI P 4 BCI 
Lomariopsis j japurensis (Martius) J. Smith H] SAd 3 RP 
Lo aL. Tes RAwSw 3 
Nephrolepis biserrata (Sw. ) coma TULe SWZMPd 4 BCI 
Mort TUe § 2 
N ephrolepis pendula Seay Te Smith Ue A 3 BCI, RP 
Nephrolepis rivularis (Vah . ex Krug UCe Pp 2 
Oleandra articulata (Sw.)} o es ULCI P 2 
Pityrogramma calomelanos (L.} Link Te WRd 3 BCI, RP 


Pityrogramma tartarea (Cav.} Maxon Te Rd 


— 
on 
ae) 


GRAYUM & CHURCHILL: LA SELVA PTERIDOPHYTES 87 
APPENDIX 1. CONTINUED 
Abun 
Species Habit Habitat dance Distr. 
Polybotrya alfredii Brade HUI P 1 
Polybotrya caudata Kunze HI P 4 BCI 
Polybotrya cervina (L.) Kau LTs Pd 4 
Polybotrya osmundacea H. : B. ex Willd. HTl ZP 3 RP 
Polypodium angustifolium Sw. Cs Z 2 RP 
Polypodium aureum L. UCs SP 3 
Polypodium ciliatum Willd. Cul SA 4 BCI, RP 
Polypodium crassifolium | ee Us RZQ Z CI 
Polypodium dissimile L. Us SAQ 3 
Polypodium furfuraceum Schlecht. & Cham. Cs PA 3 
Polypodium hygrometricum Spli UE]! SR 2 BCI 
Polypodium loriciforme nstock UCI PS o 
Polypodium lycopodioi Cl SPA + BCI 
Polypodium maritimum Hieron. Cl AP 3 BCI, RP 
Polypodium occultum ist Us ZA 2 BCI 
Polypodium pectinatum L UCs AS 3 BCI 
Polypodium percussum GUI AP 4 BCI, RP 
Polypodium phyllitidis L. Us SA 4 
Polypodium sororium H. & B. ex Willd. Ul Pp 3 
Polypodium sphenodes Kunze ex Klotzsch Ul ZQP z 
Polypodium triseriale Sw. cl P 3 BCI 
ee bradeorum (Rosenstock) 
LD Ul QSZR 2 
Pteris eis Poiret Tes ZAws 3 BCI, RP 
Pteris propinqua Agardh Te AdW 2 BCI 
Pteris pungens Willd. Te P 3 BCI 
Pteris quadriaurita Retz. Te Rd (1) 
accoloma inaequale (Kunze) Mett Te Pd 4 
Salpichlaena volubilis (Kaulf.) J. Smith Tv P 4 
Shemaners longicaudata Lapeshiige ) . oe, Te Q a 
Te Z 1 
Tectaria athyrioides (Baker) C. Chr. Tes Zz 3 
Tectaria brauniana (Karsten) C. Chr. Ts Pw 2 
ectaria draconoptera (D. Eaton) Copel. Te Pw K 
Tectaria incisa Cav. TEe SAwQR 4 BCI, RP 
Tectaria mexicana (Fée) Morton TEs RAd 4 
Tectaria nicotianifolia (Baker) C. Chr Tl AS 3 BCI, RP 
Tectaria plantaginea (Jacq.) Maxon Es Q 2 
Tectaria rivalis (Mett. ex Kuhn) Te A 2 
ctaria rufovillosa (Rosenst Te P 1 
arian angustifolia (Willd.) Proctor TLs R 2 RP 
The isii ae Ching TLe R 4 BCI 
thera biolleyi (Christ) Proctor Te S 2 
Thelypteris curta (Christ) Reed Te A 3 
Thelypteris decussata (L.) Procto Te Pdw 2 
Thelypteris dentata (Forssk.} E. St. John TLs WQ 4 BCI, RP 
Thelypteris falcata (Liebm.) R. Tryon TLs PwS 2 
helypteris francoana (Fourn.) Morton TLe R 3 RP 
Thelypteris ghiesbreghtii (Hook.) Morton Tl s 3 
TLEs QZ 1 RP 


Thelypteris gigantea (Mett.} Morton 


AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 3 (1987) 


88 
APPENDIX 1. CONTINUED 
Abun- 
Species Habit Habitat dance Distr. 

Thelypteris hispidula (Decne.} Reed Te S 2 

Thelypteris leprieurii (Hook.) R. Tryon TEe R (1) 
bear lingulata (C. Chr.) Morton Ts P 4 
Thelypteris nicaraguensis (Fourn.} Morton TLEe SQW 4 BCI 
Thedy odie poiteana (Bory) Proctor Tl Ww 2 BCI, RP 
Thelypteris resinifera (Desv.} Proctor Te R 1 RP 
Thelypter rata (Cav.) Alston 1ks RMQ 2 BCI 
Thelypteris torresiana (Gaud.)} Alston Ts d 4 BCI, RP 
Thelypteris urbanii (Sodiro) A. R. Smith Te Aw Z 
Thelypteris villana L. D. Gm ETe Q (1) 
Vittaria costata Kun Us SP 2 BCI 
Vittaria lineata La. “ Smith CUEe AP 3 BCI 
Vittaria stipitata Kunze CUe PA 3 


LEGEND FOR APPENDIX 1 


Habit.—Two types of information are encoded here: the substrate is indicated by a capital letter, 
and the nature of the rhizome by a lower-case letter. 
C—Cano ti fi i cl to the ground, as in light gaps or pejibaye trunks; may 
be ais haeo encountered on branchfalls. 
E—Epipetric; growing on rocks or, occasionally, concrete. 
H—Hemiepiphytes; species which begin life on the ground, or very close to it, and at maturity are 
characteristically appressed-climbers on trunk bases {the connection with the ground commonly 
ing lost 


being ; 
L—Log-dwellers; species that are typically found growing on rotting logs or, occasionally, timber 
(epiphytes may persist on fallen trees for some time, and many terrestrial species are occasionally 


species. 
U-Undersory epiphytes; including mainly trunk epiphytes, having no close association with the 
und. 


peer species with a distinct, self-supporting trunk of at least 0.5 m in height. 
e —erect; ies with a short, erect rhizome, the fronds closely approximate in a rosette (most 
terrestrial ferns fit this description). 
1 —long-creeping rhizome, the fronds arising rather distantly. 
s —short-creeping rhizome, the fronds more or less approximate and often clustered at the apex. 
v —vining; ferns with an erect rhizome, climbing high into adjacent foliage by means of the greatly 
elongated frond axes (at La Selva, only a single species falls into this category). 


For either of the two subcategories of “Habit” (as well as for the “Habitat” subcategories) two or 
more alternative conditions may be vapsaareeel at . othe specien, These : are always listed in order 
of prevalence: th sp canopy 
epiphyte, but sometimes occurs in the point similarly, “‘Tls” would be used to Satie a 
terrestrial species with rhizomes that are long-creeping, or sometimes short-creeping. 


ecscae bless category regularly Fae - ae Waxes aad srenetienet a second. been broad habitat 
ve specified 


Z 


na the use of an appropriate Sie ten =a 


A—Alluvial forest; forest which exists on alluvial soils, especially along the major rivers and in the 
vicinity of the Quebrada (Q.} Leonel, zr near the lower reaches of the Q. El Salto, Q. El Sura 
(including the Arboretum) and Q. Saba 


GRAYUM & CHURCHILL: LA SELVA PTERIDOPHYTES 89 
M—Marshes; open, treeless, wet sites (a very poor fern habitat at La Selva). 


P—Primary forest; comparatively well-drained slopes and plateaus in upland primary forest, the 
habitat that prevails over most of La Selva 


Ws ale ae small, usually rocky, tastahovina a i and similar sections of larger streams (e.g., 
. Esquina, and the Q. El Salto beyond line 2 

PaaBeve de ey of the larger rivers, as well as se foes portions of the Q. El] Salto, Q. Sabalo 
and Q.E 


S ree oe Anclodes abpnd pond sage prniatons es formerly cleared jand in various 


fobect! the two subcategories are not al ly de ‘ ed. 
W—Weedy land; i oo or pia alindihood clearings, groves, pastures, etc. The 
pejibaye grove, the station clearing and first-year successional stri 


mples. 
Z—Swamp forest; pain’ swampy areas, as exem po icine by most of Plot II; many of the species of 
this habitat also occur along sl 


d —disturbed areas, such as light gaps, " piidelides or road 
r —ridges or hilltops, generally toward the back of the 
w—wet or damp, low-lying sites, often near streams or jrincsie, or on seeps. 


Abundance.—The estimates of abundance are relevant only within the context of the habitat of 
the fern ae consideration. 


1—very rare; known from only 1-2 collections. The casual observer should not expect to encounter 
any of thes 


2—rare; awe from 3-5 collections. The casual observer might expect to encounter some of the 


raparenant widely scattered but 
good. chance ¢ porte ay of these. 
ht habitat, will be hard-p dt k th 
( | hypoth species not known from ‘La Selva proper, but collected in as adjacent 
ually the ‘Cloud Forest Ridge,” or ae - _ Sal san 
eerie species indicated for La Selva by properly llections believed to i t 
label errors as to locality. 


common, or else locally abundant. The casual observer 


4 


Distribution. 


BCI—species shared with Barro Colorado Island, Panam 
RP —species shared with the Rio Palenque Science Calle Ecuador. 


American Fern Journal 77(3):90-94 (1987) 


The Disposition of Trichopteris (Cyatheaceae) 


DAvID B. LELLINGER 
Department of Botany, Smithsonian Institution, Washington, DC 20560 


The delimitation of genera and families has been a persistent problem in fern 
taxonomy, and the Cyatheaceae sensu stricto is no exception. Christensen (1905- 
06) adopted clearly artificial genera (Cyathea, Hemitelia, and Alsophila) based 
on complete (totally surrounding the sorus), partial, and absent indusia. He in- 
cluded Lophosoria and Metaxya in Alsophila. The latter two satellite genera are 
only distantly related to the major genera of the family, and nowadays are often 
placed in one or two families of their own. 

Holttum (1963) proposed a single genus Cyathea for the Flora Malesiana region 
with two very distinct subgenera, Sphaeropteris and Cyathea. Holttum has main- 
tained (1981, p. 466) that the ‘‘only subdivision of the genus clearly definable is 
that between subgenus Sphaeropteris and the rest.”’ This indicates that Alsophila 
and Cnemidaria are less distinct from Cyathea than all three are from Sphae- 
ropteris, which is confirmed by the lack of hybrids with Sphaeropteris. In study- 
ing the species of the Flora Malesiana region, Holttum came to the fundamental 
conclusion, among many, that indusium type is not an important generic char- 
acter, for within a few species it varies widely. 

Tryon (1970) divided the Cyatheaceae sensu stricto on the basis of scale char- 
acters, indusium presence or absence, and venation. He adopted the genera 
Sphaeropteris (scales conform), Alsophila and Nephelea (scales non-conform 
and setate), Trichopteris (scales non-conform and non-setate, laminae free-veined, 
and sori exindusiate), Cyathea (scales non-conform and non-setate, laminae free- 
veined, and sori indusiate), and Cnemidaria (scales non-conform and non-setate 
and laminae net-veined. According to Holttum and Edwards (1983, p. 179), this 
classification has assorted closely related species into Cyathea, Sphaeropteris, 
and Trichopteris. It is apparent that genera based on these characters are not 
natural. 

Working from Tryon’s (1970) generic concepts, I have found it possible to define 
readily recognizable and coherent genera in the Cyatheaceae sensu stricto by 
including in Cyathea the genus Trichopteris and the New World species con- 
sidered to be Sphaeropteris, except for the S. horrida group. I accept the genera 
Sph teris, Alsophila (including Nephelea), Cyathea (including Trichopteris), 


and Cnemidaria. 

Occasional hybrids occur within Alsophila and Cyathea and between Cnemi- 
daria and Cyathea. This is evidence of a greater degree of relationship than with 
Sphaeropteris, but in my opinion should not alone be the basis for adopting an 
inclusive Cyathea (either excluding or even including Sphaeropteris), for the 
characteristics of Alsophila and Cnemidaria are sufficiently different from those 
of Cyathea to distinguish the genera readily, and intergeneric hybrids in ferns 
are by no means rare and sometimes occur between large and supposedly sep- 
arate genera (e.g., Dryopteris x Polystichum). 


D. B. LELLINGER: TRICHOPTERIS 91 


Sphaeropteris is almost entirely an Old World genus. Only S. horrida (Liebm.) 
Tryon and five allied species occur in the New World tropics. I agree with Holttum 
and Edwards (1983, pp. 161-162) in this delimitation. The other New World 
species that were placed in Sphaeropteris belong to Cyathea. In its restricted 
sense, this group was monographed by Tryon (1971). The stipe base scales of 
Sphaeropteris are distinct from those of all other New World Cyatheaceae genera 
in having cells of the same thickness and color from the central portion of the 
scales to the margin, except for some spreading, spinelike, often darker setae 
along the margins. The spores of Sphaeropteris bear flattened projections that 
are unique among New World Cyatheaceae (Gastony & Tryon, 1976). The laminae 
of Sphaeropteris are mostly 2-pinnate-pinnatifid (as in most Cyathea species), 
the pinnule segments are often falcate, and the abaxial surface is often pale and 
bears whitish scales on the abaxial surface. The scales of Cyathea poeppigii 
(Hook.) Domin [syn. Sphaeropteris elongata (Hook.) Tryon] and C. myosuroides 
(Liebm.) Tryon mimic those of Sphaeropteris, but are subtly different in their 


ginal cells and scales with erose margins, elongate marginal cells and scales with 
fringelike margins, or, in a few cases, scales bearing marginal spinelike processes 
(see key couplet 1). The structure of th les is a valuable cl ter in assessing 
the relationship of species within the genus. The indusia vary from membrana- 
ceous and complete (sphaeropteroid) to firm and saucer-shaped (cyatheoid) to 
scalelike (hemitelioid) to absent. 
Cnemidaria is an exclusively New World tropical genus that includes 25 species. 
It was monographed by Stolze (1974). The species of Cnemidaria have low, erect 
caudices and lack the tall, treelike trunks typically found in the other genera. 


92 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 3 (1987) 


The veins are regularly anastomosing or nearly so, as stated in the key. The 
laminae are only pinnate-pinnatifid, with relatively low lobes or segments that 
have large, open, V-shaped sinuses between them. The abaxial surface of the 
laminae and its axes tend to be glabrous or nearly so with little variety in in- 
dument, unlike the foregoing genera. The spores are distinctive in being smooth 
(Stolze, pers. comm.) and in having three equatorial pores. 

The species that have been placed in Trichopteris belong to several evolu- 
tionary lines and appear to be related to different species or species groups in 
Cyathea. 

Cyathea atrovirens (Langsd. & Fisch.) Domin and C. dichromatolepis (Fée) 
Domin appear to have given rise to the series C. miersii (Hook.) Domin, C. 
elegantula Domin, and C. corcovadensis (Raddi) Domin, the type of the generic 
name Trichopteris. All have similar stipe base scales with contorted marginal 
cells and bullate and plane scales on the abaxial costae or costules. However, 
C. corcovadensis has the least lobed pinnae or pinnules and the fewest hairs on 
the adaxial surface of the costae and costules, a common correlation in “Tri- 
chopteris.”’ 

Cyathea sipapoensis (Tryon) Lellinger and C. marginalis (Klotzsch) Domin 
appear to be related to C. macrocarpa (Presl) Domin. All have similar stipe base 
scales that are concolorous, whitish, and with little cellular differentiation. 

Two groups of “Trichopteris” species share rather strongly bicolorous rhizome 
scales with an elaborated, fringe-like margin. Almost all of these have conform 
apices, although not all have the apices articulate, another common correlation 
in “Trichopteris.” I have not found likely ancestral species elsewhere in Cyathea 
for these groups, which may indicate their relative remoteness within the genus. 
In the pinnate group, C. speciosa Willd. appears to be least specialized and more 
like other species of Cyathea in having non-conform lamina apices; the related 
species are C. cyclodium (Tryon) Lellinger, C. stolzei A. R. Smith ex Lellinger, 
and C. williamsii (Maxon) Domin. All are subarborescent. In the bipinnate group, 
C. petiolata (Hook.) Tryon may be least specialized because it has non-conform 
apices and is clearly arborescent (caudices to 5 m); C. conformis (Tryon) Stolze 
and C. intramarginalis (Windisch) Lellinger also are arborescent, but have con- 
form lamina apices. The remaining species are all subarborescent, with caudices 
usually less than 1 m long. Cyathea intramarginalis is related to CG. dissimilis 
(Morton) Stolze; both share the character of I-forked veins, a reduction from the 
usual pinnate branching of the vein groups that terminates in C. akawaiorum 
Edwards, which has mostly simple veins. The other species of this group are C. 
impar Tryon and C. steyermarkii Tryon. 


KEY TO THE GENERA OF NEOTROPICAL CYATHEACEAE SENSU STRICTO 
1. Cells of the stip ] tirely unif. (th ie ad oe 
processes excepted), the cells along the margin not different from the 
central ones; spinelike processes present at the apex and margins of the 
stipe scales, these regular, antrorse, usually distant, and often dark; basal 
basiscopic vein of each vein group always arising from the costa; spores 


D. B. LELLINGER: TRICHOPTERIS 93 


bearing flattened, scalelike projections; indusia sphaeropteroid ....... . 
Oe oe eee Fel ae ee ee ee Sphaeropteris 
Cells of the stipe scales not entirely uniform, the cells along the margins 
slightly to markedly different in size, shape, wall thickness, or orientation 
from the central ones; spinelike processes absent (except in Alsophila, 
with strongly bicolorous scales and in Cyathea poeppigii, myosuroides, 
and senilis, with thinner-walled scales along the margins and irregular, 
often spreading, usually approximate spinelike processes), but approxi- 
mate, thin, long, fil tous p Pp tin pecies of Cyathea; 
basal basiscopic vein of each vein group not arising from the costa; spores 
bearing hairlike projections, granular deposits, or ridges, or nearly smooth; 
indusia cyatheoid, hemitelioid, sphaeropteroid, or absent. 
. Stipe scales provided with a dark (rarely pale) apical seta and some- 
times lateral setae, the scales strongly bicolorous, the central band often 
several cells thick at the base; peripore of spores ridged ... Alsophila 
Stipe scales lacking apical or lateral spinelike processes {approximate 
or rarely distant, lateral, rather thin setalike processes at entirely right 
angles to the axis of the scale present in a few species of Cyathea, 
especially in the C. swartziana group), the scales concolorous to weakly 
bicolorous, the central band usually only 1 cell thick at the base, the 
apex filamentous to round; perispore of spores not ridged. 
Basal veins of each vein group forming regular areolae along the costae 
or the basal veins connivent to the base of the sinus or occasionally 
meeting the sinus just above the base; plants not arborescent; spores 
nearly smooth, bearing a single pore at or near the equator on each of 
the 3 sides; laminae mostly pinnate-pinnatifid with shallowly lobed 
pinnae and the sinuses between the lobes broadly V-shaped........ 
ee ee er ra a Cnemidaria 
3. Basal veins usually free and not connivent or anastomosing (except a 
transverse costal vein joining adjacent pinnate vein groups in C. pe- 
tiolata and williamsii); plants ar t or someti barb t 
spores lacking equatorial pores; laminae mostly 2-pinnate-pinnatifid 
with deeply lobed pinnules and the sinuses between the lobes narrowly 
LgbANGd cs oii ee oe eee Cyathea 
All of the species of the New World that had been placed in Sphaeropteris 
and most that had been placed in Trichopteris have valid names in Cyathea. 
For the few that do not, I wish to make the following combinations: 


= 


i) 


te 


- 


99/0 Cyathea axillaris (Fée) Lellinger, comb. nov.—Phegopteris axillaris Fee, Gen. _ 
9909 Fil. 243. 1852, effectively a nom. nov. based on Polypodium axillare Raddi, --*' On§ 
1819, non Aiton, 1789. : 
@9)2Cyathea barringtonii A. R. Smith ex Lellinger, nom. nov.—Alsophila cordata— 79// 
Klotzsch, Linnaea 20:441. 1847, non Cyathea cordata (Desv.) Mett. ex Diels, i} 
in Engl. & Prantl, 1899. . 
49/5 Cyathea cyclodium (Tryon) Lellinger, comb. nov.—Trichopteris cyclodium Tryon, — 77/ 
Rhodora 74:446. 1972. 


94 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 3 (1987) 


4qi% Cyathea demissa (Morton) A. R. Smith ex Lellinger, comb. nov.—Alsophila de- 
44/¢ missa Morton, Fieldiana, Bot. 28(1):7. 1955. 
9909 _ dombeyi (Desv.) Lellinger, comb. nov.—Alsophila dombeyi Desv. Mém. 
64 Y Soc. Linn. Paris 6:320. 1827.) 
Qqro Gratien gardneri (Hook.) Lellinger, comb. < —Alsophila gardneri Hook. Sp. 
Fil. 1:40. 1844. 
21085 Cyathea nanna (Barrington) Lellinger, comb. nov.—Trichopteris nanna Barring- 
2\0%6 ton, Rhodora 78:3, t. 1, f. 3, 4. 1976. 
492 Cyathea pauciflora (Kuhn) Lellinger, comb. nov.—Alsophila pauciflora Kuhn, 
992) Linnaea 36:156. 1869. : 
4924 Cyathea rufa (Fée) Lellinger, comb. nov.—Alsophila rufa Fée, Crypt. Vasc. Brés. 
99221:165. 1869. 
1425 Cyathea tryonorum (Riba) Lellinger, comb. nov.—Alsophila tryonorum Riba, 
672! Rhodora 69:66. 1967. 
2\0? } Cyathea venezuelensis A. R. Smith ex Lellinger, nom. nov.—Trichopteris s a 
2\08% ermarkii Tryon, Rhodora 74:446, f. 11, 12. 1972, non Cyathea steyermarkii 
Tryon, 1972. 


I thank D. S. Barrington, D. S. Conant, R. E. Holttum, J. T. Mickel, G. R. Proctor, 
A. R. Smith, and R. G. Stolze for their helpful comments on various aspects of 
this classification. 


LITERATURE CITED 


BARRINGTON, D. S. 1978. A revision of the genus Trichipteris. Contr. Gray Herb. 208:3-93. 
CHRISTENSEN, C. 1905-1906. Index Filicum. Copenhagen: Hagerup 
Conant, D. S. 1983. a revision of the genus Alsophila thesia) in the Americas. J. Arnold 
Arbor. 64:333-38: 
Gastony, G. J. 1973. po revision of the fern genus Nephelea. Contr. Gray Herb. 203:81-148 
and R. M. Tryon. 1976. Spore morphology in the Cyatheaceae. II. The genera Lophosoria, 
Metaxya, Sphaeropteris, Alsophila, and Nephelea. Amer. J. Bot. 63:738-758. 
Ho.trum, . ra 1963. Cyatheaceae. FI. Males. II, 1:65-176. 
. The tree-ferns of Africa. Kew Bull. 36:463-482, t. 15, 16. 
an a 5 J. Epwarps. 1983. The tree-ferns of Mt. Roraima and neighbouring areas of the 
Guayana Highlands with comments on the family Cyatheaceae. Kew Bull. 38:155-188. 
Stouze, R. — 1974. A taxonomic revision of the genus Cnemidaria (Cyatheaceae). Fieldiana, Bot. 
1-98. 


TRYON, e 1970. The classification of the Cyatheaceae. Contr. Gray Herb. 200:3-53. 
——————. 1971. The American tree ferns allied to Sphaeropteris horrida. Rhodora 73:1-19. 
———.. 1976. A revision of the genus Cyathea. Contr. Gray Herb. 206:19-98. 
WINDIscH, P. G. 1977. i of the genus a Rasa a erenaiee with a revision of the 
neotropical exindusiate species. Bot. Jahr bye. aed 
. 1978. pe ar (Cyatheaceae): The sy f the g f Sph pteris hirsut 


- 


Mem. New York Bot. Gard. 29:2- 


American Fern Journal 77(3):95-99 (1987) 


Germination of Helminthostachys Spores 


DEAN P. WHITTIER 
Department of General Biology, Vanderbilt University, Nashville, TN 37235 


For nearly a century the spores of the Ophioglossaceae have been sown, but 
few investigators h ucceeded in inating them (Boullard, 1963). Campbell 
(1895, 1907) was the most successful in germinating spores of the Ophioglossaceae 
on soil. He reported the early stages of germination of spores of Botrychium 
virginianum (Campbell, 1895) and of three species of Ophioglossum (Campbell, 
1907). An earlier report on spore germination in B. ternatum by du Buysson (1889) 
remains in question (Whittier, 1981) because the gametophytes that developed 
had essentially the same morphology as gametophytes of leptosporangiate ferns. 

More recently, spores of Botrychium and Ophioglossum have been germinated 
in axenic culture (Whittier, 1972, 1981; Gifford & Brandon, 1978). These spores 
germinated in the dark on nutrient media containing minerals and sugar. In some 
cases, the spores took 3-4 months to germinate under these conditions (Whittier, 
1981). 

The mature gametophyte of Helminthostachys, the third genus of the Ophio- 
glossaceae, has been described from nature (Lang, 1902; Nozu, 1961; Pant et al., 
1984). It usually has a long, thick, cylindrical axis with numerous short lobes at 
its base. The lobes have apical meristems and appear to be short branches (Pant 
et al., 1984). The basal region of the gametophyte contains the mycorrhizal fungus 
and the cylindrical portion bears the gametangia. Spore germination and early 
stages of gametophyte development are unreported for Helminthostachys. Since 
germination and gametophyte development in Botrychium and Ophioglossum 
have been studied with the techniques of axenic culture, the aim of this study 
was to use these techniques to determine the conditions that promote germination 
and early gametophyte development in Helminthostachys. 


MATERIALS AND METHODS 


Fertile spikes of Helminthostachys zeylanica (L.) Hooker were obtained from 
plants grown in greenhouses at the New York Botanical Garden, Longwood 
Gardens, and the University of Massachusetts. The spores were sown within two 
weeks of their collection from the fertile spikes. 

The techniques of Whittier (1973) were employed. Spores were sown on 15 ml 
of nutrient medium in culture tubes with a diameter of 20 mm. The tubes had 
screw caps which were tightened to reduce moisture loss. The nutrient medium 
was composed of a modified Moore’s solution of mineral salts, minor elements, 
FeEDTA, and 0.6% agar. A liter of the modified mineral salt solution contained 
100 mg NH,CI; 100 mg MgSO,-7H,O; 40 mg CaCl,; and 100 mg K,HPO,. The 
medium was supplemented with 0.2% glucose and had a pH of 5.3 after auto- 
claving. The spores were cultured at 24 + 1°C in light at an intensity of 1400 lux 
from cool white fluorescent lamps or in darkness. 


96 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 3 (1987) 


Fics. 1-6. Germination and early gametophyte development of Helminthostachys. All bars = 25 
xm. 1. Spore with cristate spore coat. Three arrows indicate ends and point of intersection of two 
fel ae a: sy 2 _— mh ° c Sag 1 . : 


8 I gol sp 3. Two-celled gametophyte 
with spore coat attached. Arrow indicates cell wall separating proximal and distal cells. 4. Two-celled 
+ , Se ssh y PR BS ‘ j ra 5 4 1 Ps £ g 


| iar’ dela t - Pf r bg oS t t wes 
celled gametophyte with mucilage (arrow) on surface of proximal cell. 


en 


=. 


RESULTS 


The hydrated spores of Helminthostachys are almost spherical and have an 
average diameter of 34 um. The spore coat is cristate (Fig. 1) which makes ob- 
serving the contents of the spores difficult. A major portion of the storage materials 
appears to be lipids since crushing the spores releases large amounts of oil which 
can be stained with Sudan IV. The triradiate ridge is difficult to observe (Fig. 1) 
because it is indistinct and small and all surfaces of the spore have the cristate 
ornamentation. The arms of the triradiate ridge are short with an average length 
of 10 um. 

After 8 months, spores on the nutrient medium in the dark began to germinate. 
Spores that were kept on the nutrient medium in the light for over a year failed 
to germinate. 

Rupturing of the triradiate ridge initiates germination. The spore cracks open 
and the enlarging cell protrudes. This original protrusion has a small diameter 


D. P. WHITTIER: HELMINTHOSTACHYS SPORES 97 


initially (Fig. 2) because of the small-sized triradiate ridge. The ornamented spore 
coat hinders observation of the first cell division in the enlarging cell. However, 
as the protrusion enlarges, a cell wall can be seen at the base of the protrusion 
(Fig. 3). The first division of the enlarging cell is perpendicular to the polar axis 
of the spore. This division produces a proximal cell (near the triradiate ridge), 
the protruding cell, and a distal cell (away from the triradiate ridge), which 
initially remains inside the spore coat. 

The enlarging two-celled gametophyte, especially the distal cell, ruptures the 
spore coat more completely (Figs. 3, 4). Even after the tears in the spore coat 
extend to its distal surface, the spore coat can remain attached to the distal cell 
or its derivatives (Fig. 4). With additional enlargement of the young gametophyte 
the spore coat is sloughed off (Fig. 5). 

The second cell division of the young gametophyte occurs in the distal cell 
and is parallel to the polar axis of the spore (Fig. 6). A small globular gametophyte 
is produced by additional cell divisions occurring in cells derived from the distal 
cell. The proximal cell, at least in the early stages of gametophyte development, 
remains undivided. The surface wall of the proximal cell is thicker than other 
cell walls in the young gametophyte (Fig. 6). In addition to the thicker wall, there 
is an accumulation of mucilage on the surface of the proximal cell (Fig. 6, arrow). 


DISCUSSION AND CONCLUSIONS 


The dark requirement for the germination of spores of Helminthostachys is 
the same as for spores of Botrychium and Ophioglossum in axenic culture. How- 
ever, the time necessary for germination of Helminthostachys spores, 8 months, 
is longer than for other spores of the Ophioglossaceae that have been germinated 
in axenic culture (Whittier, 1981). 

Helminthostachys spores have smaller triradiate ridges than comparable-sized 
spores of Botrychium (e.g., B. dissectum forma obliquum). The arms of the tri- 
radiate ridge are 10 um long for Helminthostachys and 17 um long for B. dis- 
sectum. The small triradiate ridge is probably responsibl the narrow diameter 
of the original protrusion of the enlarging cell during earlier germination. In 
Helminthostachys the protrusion is only about half the diameter of the spore. 
The protrusion in Botrychium, which has the larger triradiate ridge, has essen- 
tially the same diameter as the spore. 

The pattern of early divisions in the young gametophytes of Helminthostachys 
is essentially the same as has been found for Ophioglossum and Botrychium 
(Campbell, 1907; Whittier, 1981). The first division is perpendicular to the polar 
axis of the spore producing proximal and distal cells. The second division occurs 
in the distal cell and is parallel to the polar axis of the spore. In Botrychium and 
Ophioglossum, the subsequent divisions in the cells derived from the distal cell 
produce the gametophyte (Whittier, 1981). 

The proximal cell in the early stages of gametophyte development, as with 
Botrychium and Ophioglossum (Whittier, 1981), remains undivided. The thick- 
ened surface wall of the proximal cell and the secretion of mucilage through 
areas of this wall are characteristic for the Ophioglossaceae. Mucilage production 


98 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 3 (1987) 


by the proximal cell has been reported for Botrychium (Melan, 1985) and ob- 
served in Ophioglossum (Whittier, unpublished). 

The gametophytes of the Ophioglossaceae are slow growing (Boullard, 1963) 
and often live for more than one year (Jeffrey, 1897; Lang, 1902; Bruchmann, 1904; 
Campbell, 1911). The subterranean habitat appears to be a more protected site 
for these long-lived gametophytes than the soil surface. More water should be 
available for growth below the soil surface. Bruchmann (1906), St. John (1949), 
and Foster (1964) have reported finding living gametophytes in the soil during or 
after drought conditions. The subt abitat appears to provide more stable 
environmental conditions throughout the year which allows many of these ga- 
metophytes to be perennial. 

The subterranean habitat is also important to the gametophytes of the Ophio- 
glossaceae because they are dependent on an association with fungi in the soil 
for organic nutrients. In addition, the mycorrhizal fungus is also probably re- 
sponsible for increasing the absorptive surface for uptake of mineral nutrients. 
The dependency on this mycorrhizal association occurs in early gametophyte 
development. The spores do not appear to contain enough reserves to support 
extended gametophyte growth because these young gametophytes undergo only 
a few cell divisions if not infected by fungi in soil (Campbell, 1907) or if no 
carbon energy source is available in the nutrient medium in axenic culture 
(Whittier, 1973). The initiation of gametophyte development in the soil would 
appear toi tk p ibility th g inati gsp dy gg toy 1 yt 
will be in close proximity to mycorrhizal fungi in the soil when organic nutrients 
are first required for development. 

Evidence from this and other studies (Whittier, 1973, 1981) carried out with 
axenic techniques suggests that the spores of the Ophioglossaceae germinate only 
in the dark. Even short exposures of low light intensities are sufficient to prevent 
germination in Botrychium (Whittier, 1973). Presumably the spores sift or per- 
colate down into the soil before germination can occur. The prevention of ger- 
mination by light would be an effective mechanism to insure that germination 
will occur only after the spores are buried in the soil. The time delay, weeks or 
months, for germination in the dark probably also is important in assuring that 
the spores percolate to appropriate depths before germination. Since the game- 
tophytes are adapted for the subterranean habitat, it is important for early ga- 
metophyte development to occur there. 


ACKNOWLEDGMENTS 
° thank John T. Mickel (New York Botanical Garden), Donald G. Huddleston (Longwood Gardens), 
Edward L. Davis {TIni it fA + eS <sL (TInt . ChMA 1 ta) 


for supplying thé spores used in this study. This study was supported in part io the Vanderbilt 
University Research Council. 


}andR 


LITERATURE CITED 


BOULLARD, B. 1963. Le gametophytes des Ophioglossacees. Considerations biologiques. Bull. Soc. 
Linn. Normandie 4:81-97. 


D. P. WHITTIER: HELMINTHOSTACHYS SPORES 99 


BRUCHMANN, H. 1904. Ueber das Prothallium und die Keimpflanze von Ophioglossum vulgatum L. 
ot. Zeitung, 2. Abt. 62:227-247. 
. 1906. he das Prothallium und die Sporenpflanze von Botrychium lunaria Sw. Flora 
96:203-2 
BUYSSON, sts DU 9. Monographie des cryptogames d’Europe. II. Filicinees. Rev. Sci. Bourbonnais 
Sue 2:153-164. 
CAMPBELL, 1p et 1895. The structure and development of mosses and ferns. 1st ed. New York: 
Macmilla 
1907. didted on the Ophioglossaceae. Ann. Jard. Bot. tg II. 6:138-194. 
=o 91 he eusporangiatae. Carnegie Inst. Washington Pub. 
Foster, D. B. 1964. The gametophytes and embryogeny of five species . Borrychikiin: Ph.D. thesis, 
Cornell Univers Ithaca, New York. 
GirForD, E. M. and D. D. BRANDON. 1978. Gametophytes of Botrychium multifidum as grown in 
enic Ae age Fern J. 68:71-7 
JEFFREY, E 2 1897. The gametophyte of Rolryehiuns virginianum. Trans. Roy. Canad. Inst. 5:265- 


LANG, we 1 1902. On the ere of Ophioglossum pendulum and Helminthostachys zeylanica. 
Ann. Bot. (London) 16 
MELAN, M. A. 1985. spate aN and physiological changes in spores of Botrychium dissectum 
forma obliquum during germination. Ph.D. thesis, Vanderbilt University, Nashville, Ten- 
ee. 


Nozu, _ ae. The ga sapesergt hg Helminthostachys zeylanica and Ophioglossum vulgatum. 
spur ape! 11:1 

Pant, D. D., TIYAL, snl m R. Misra. 1984. Gametophytes of Ophioglossaceae. Phyta 
Mon Bee se 

St. JOHN, E. P. 1949. 2g valutél of the Ophioglossaceae of the eastern United States. Quart. J. 
Florida Acad. Sci. 12:207-219. 


WHITTIER, P. 1972. sepetrnigy of Botrychium dissectum as grown in sterile culture. Bot. Gaz. 
ee) 133: 
. The effect of ight ee other factors on spore g tion in Botrychium dissectum 


py J. Bot. 51: 1791- et 
981 


paievin in axenic culture. Amer. For: 1. 72: 13- 19. 


Zrvz7se 


American Fern Journal 77(3):100-105 (1987) 


SHORTER NOTES 


A Binomial for a Common Hybrid Lycopodium.—The taxonomic status of Ly- 
copodium porophilum Lloyd & Underwood (Lycopodiaceae) recently has been 
clarified by Waterway (Syst. Bot. 11:263-276, 1986). She presented convincing 
evidence for considering this taxon a species distinct from the similar clubmosses, 
L. lucidulum Michx. and L. selago L., with which it formerly had been confused. 
Hybrids between L. porophilum and L. lucidulum frequently are encountered. 
These hybrids are characterized by sterile spores and morphological characters 
intermediate between those of the parents. Waterway provided a helpful chart 
outlining the distinguishing features of these three taxa (p. 272). 

The geographic range of the hybrid is wholly sympatric with the distribution 
of L. porophilum from Ohio and North Carolina west to Wisconsin, Iowa, and 
Arkansas. Although sterile, the hybrid reproduces readily by vegetative gemmae. 
On the unglaciated Appalachian Plateau in Hocking County, Ohio, the hybrid is 
common in habitats intermediate between those of its parents. Lycopodium po- 
rophilum typically is restricted to the upper portions of massive sandstone ex- 
posures; L. lucidulum grows on talus and soil at the foot of the cliffs and on 
alluvial terraces of small streams. The hybrid occurs both with L. porophilum 
and also on lower portions of the same rock faces. In the mesic coves of southeast 
Ohio, populations of the hybrid frequently are more extensive and vigorous than 
colonies of either parent. The hybrid plants usurp the space on the cliffs and 
ledges which is available to L. porophilum. This preemption of its habitat ap- 
parently is a significant factor in the present rarity of L. porophilum in Ohio. I 
have not seen the hybrid growing in soil with L. lucidulum. 

Since this hybrid clubmoss is so common and widespread, it seems useful to 
provide it with a binomial to replace a cumbersome formula. 


Lycopodium xbartleyi Cusick, hybr. nov.—Type: Ohio, Hocking Co., mossy 
sandstone boulders above Keifel Rd, 0.3 mi NE, jct of Big Pine Creek Rd, 
Sect 7, Benton Township, 18 March 1987, Cusick 26204 (OS; isotypes MICH, 
MU). 


Hybrida e Lycopodio lucidulo et L. porophilo exorta, aliis characteribus inter 
parentes media, sporis abortivis. 

The epithet honors Floyd Bartley (1884-1974) of Circleville, Ohio, who collected 
thousands of plant specimens—including this hybrid—from southeast Ohio from 
the 1930s through the 1960s. Stuckey (Ohio J. Sci. 75:209-210, 1975) summarized 
Bartley’s contributions to our knowledge of the Ohio flora. 

‘Thanks to Kerry Barringer, Brooklyn Botanic Garden, for checking the Latin 
diagnosis.—ALLISON W. Cusick, Division of Natural Areas and Preserves, Ohio 
Department of Natural Resources, Columbus, OH 43224. | 


SHORTER NOTES 101 


Nomenclatural Notes on Some Ferns of Costa Rica, Panama, and Colombia.— 
IlJ.—This is a continuation of the series begun a few years ago (Amer. Fern J. 
67:58-60. 1977; 75:31. 1985) to record changes of names pertinent to ongoing 
floristic projects. 


992+ Arachniodes ochropteroides (Baker) Lellinger, comb. nov.—Nephrodium och- 
992 Yopteroides Baker, Ann. Bot. (London) 5:325. 1891.Type: Jamaica, Fox’s 
Gap, Apr 1886, Hart (K; isotype IJ). 


This species, A. leucostegioides (C. Chr.) Ching, and A. macrostegia (Hook.) 
Proctor seem more closely related to Arachniodes than they do to Polystichopsis, 
where the first two had been placed by Morton (Amer. Fern J. 50:152, 155. 1960). 

Polystichopsis seems to me to be separable from Arachniodes on the basis of 
the long, stiff, straight, colorless hairs that have cells much longer than wide and 
that are borne on the axes of the ovate-acuminate to narrowly triangular, usually 
2-pinnate or sometimes 3-pinnate laminae. Arachniodes lacks long hairs (except 
for A. ochropteroides, which has long, slightly lax, pale tan hairs that have cells 
only slightly longer than wide and with usually obvious cross-walls; the hairs are 
reminiscent of those in the genus Ctenitis), and many species are quite glabrous; 
the laminae of most are relatively broader and are 4- or 5-pinnate. The genus 
Polystichopsis includes P. chaerophylloides (Poir.) Morton, P. lurida (Jenm. ex 
Underw. & Maxon) Morton, P. muscosa (Vahl) Proctor, and P. pubescens (L.} 
Morton. 


1539/ Cyathea nigripes var. brunnescens (Barr.) Lellinger, comb. nov.—Trichopteris 
/4is@ nigripes var. brunnescens Barr., Rhodora 78:4, f. 5-6. 1976.“TyPE: Colombia, 
Dept. El Valle, Rio Yurumangui, 5-50 m, Cuatrecasas 16155-C (US!; isotype 

GH). 


| 493% Cyathea stolzei A. R. Smith ex Lellinger, comb. nov.—Trichopteris pinnata Stolze, '< 


Amer. Fern J. 74:103, f. 2. 1984, non Roxb. ex Clarke, 1874.Type: Panama, 
Pcia. Colén, Santa Rita Ridge road 21-26 km from the Transisthmian High- 
way, 500-550 m, Knapp 5881 (MO; isotype F). 


q928 Cyathea ursina (Maxon) Lellinger, comb. nov.—Alsophila ursina Maxon, J. Wash. - ‘*” 


Acad. Sci. 34:48. 1944.—TypeE: Belize, Stann Creek Distr., Stann Creek Valley, 
Antelope Ridge, Gentle 3197 (US!; isotype MICH). 
21090 Pecluma ptilodon var. caespitosa (Jenman) Lellinger, comb. nov.—Polypodium 
33 25pectinatum var. caespitosum Jenman, Bull. Bot. Dept., n.s. 4:125. 1897.<TyPE: 
Jamaica, St. Andrew Parish, Old England, 4000 ft, Jenman (NY). 
272° Phlebodium pseudoaureum (Cav.) Lellinger, comb. nov.—Polypodium pseu- 
9929 doaureum Cav., Descr. Pl. 247. 1802.<TypE: Without locality, Nee (MA), 
examined by Christensen (Dansk Bot. Ark. 9(3):12. 1937). 


The correct specific epithet, as pointed out long ago by Christensen, must be 
pseudoaureum. Although I have not seen the type specimen, the species is so 
distinct that Christensen could scarcely have misidentified it. This species usually 
has been called Polypodium aureum var. areolatum (Humb. & Bonpl. ex Willd.) 
C. Chr. or Phlebodium aureum var. areolatum (Humb. & Bonpl. ex Willd.) Farw. 


102 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 3 (1987) 


However, it is surely an independent species, differing in range and ploidy from 
the true Phlebodium aureum (L.) J. E. Smith. Phlebodium pseudoaureum is widely 
distributed throughout tropical America, whereas P. aureum appears to be absent 
from Central America and from Colombia to Bolivia. 


Tectariaceae Lellinger, fam. nov.—Type: Tectaria Cav. 


Rhizomata stipitesque ad basin squamosi, squamosis angustis saepe fibrillosis 
concoloribus non lanceolatis vel ovatis bicoloribusque. Rhachides fuscae teres 
vel sulcatae, sulcis continuis non interruptis per sulcos costarum, plerumque 
saltem leviter pilosae, pilis multicellularibus pl que catenatis, aliquando gla- 
brae vel squamosae. 

This family is established for the genus Tectaria, its close allies, including the 
genera Ctenitis, Aenigmopteris, Ataxipteris, Psomiocarpa, Lastreopsis, Atalo- 
pteris, Pleocnemia, Pteridrys, Heterogonium, Camptodium, and Stenosemia, and 
its more distant allies, Pleuroderris, Dictyoxiphium, Hypoderris, and Amphi- 
blestra. 

The name Tectariaceae replaces in part the illegitimate name Aspidiaceae, 
which is based on the illegitimate generic name Aspidium. Under Art. 18.1 of 
the present Code, such a family name cannot be conserved because it is based 
on an illegitimate generic name. The names Hypoderriaceae Ching and Dic- 
tyoxiphiaceae Ching are not validly published because they lack a Latin descrip- 
tion, according to Pichi Sermolli (Webbia 25:273. 1970). I do not believe either 
has received a Latin description, and there is no provision in the Code to validate 
a family description on the basis of a validly described monotypic genus. There 
appear to be no other families based on these generic names. Ching applied his 
names to monotypic families. In contrast, the name Aspidiaceae has been used 
in an exceedingly broad sense far beyond my concept of Tectariaceae, for instance 
by Copeland (Gen. Fil. 100-154. 1947). It would be confusing to adopt it or Hy- 
poderriaceae or Dictyoxiphiaceae for my concept of Tectariaceae.—Davip B. 
LELLINGER, Department of Botany, National Museum of Natural History, Smith- 
sonian Institution, Washington, DC 20560. 


Terrestrial Psilotum in East-Central Alabama.—On 27 October 1986, plants of 
Psilotum nudum (L.) P. Beauv. (Psilotaceae) were discovered in Lee County, 
Alabama, at the southern extreme of the Piedmont Plateau in a mixed pine- 
deciduous woodland south of Loblockee Creek, near County Highway 11, about 
five miles north of Loachapoka. The population represents another extension of 
the known range of the species more than 240 kilometers inland from the Gulf 
Coast and is apparently a new state record. An effort to determine the extent of 
the population was made by several students and myself during the following 
week. More than 100 plants, usually in small patches of 5-10 aerial shoots/m?, 
were located within an area comprising ca. 10 hectares, outside of which no 
additional plants were observed. Eight specimens representing the size range of 
shoots were transplanted to containers and moved to the Botany Greenhouses at 
Auburn University so that comparisons between greenhouse-protected plants 
and those in the field population could be made during the onset of winter. These 


SHORTER NOTES ‘an 


{TT PE HST RE ST RRA NO AS 
Let opa Lt meiua See PY AIT RS SEO 
HH : EEE TIS cay Pee eo ES 

Ce TEESE gl ee nr 
a Ca patayiu snnaisaaer tees gt sen uate 
Be eerie erdayistar aansaiianst cca Gries 
Ape a erate etn anny cess ecay a 
m1 ts rama S B Benes eitnt 
SU SeRRnnn ATS eee etre eae a 
panapeasestiacget cect Gynmenee™ < 
LAS AP a HE et 
OIC D IE ECT 
Pape cage AR 

YY ar an Et x 

EE i 


aid 
3 
8 


Fic. 1. County distribution of Psilotum nudum in the southern United States excluding peninsular 
Florida (shaded). 


observations were continued through June 1987 when the first aerial shoots of 
Psilotum emerged in the field. Since this new locality differs markedly from most 
others previously reported for this species, this note will describe the habitat and 
discuss some aspects of this and other terrestrial populations while further studies 
are in progress. 

Distribution of Psilotum nudum in the southern United States, where it reaches 
its northernmost limit in North America, is shown by county dots (except in 
peninsular Florida) in Figure 1. The other distinctly inland sites in addition to 
the Alabama locality reported above are: Freestone County, Texas (Lodwick, 
Amer. Fern J. 65:62. 1975); Lincoln Parish (Rhodes, Sida 3:525. 1970) and Ouachita 
Parish, Louisiana (Thieret, Louisiana Ferns and Fern Allies, pp. 32-33. 1980); and 
Darlington County, South Carolina (Radford et al., Manual of the Vascular Flora 
of the Carolinas, p. 3. 1968). The map is based upon these and various other 
sources (Jones et al., Sida 3:359-364. 1969; Clewell, Guide to the Vascular Plants 
of the Florida Panhandle, pp. 51-52. 1985; Snyder & Bruce, Field Guide to the 
Ferns and Other Pteridophytes of Georgia, p. 254. 1986) as well as personal 
observations in northern Florida. With statements such as “‘various habitats, epi- 
phytic, epipetric, or terrestrial ....” (Radford, loc. cit.) or “In soil, humus, moss 
mats, or rotten wood in low to mesic woods” (Thieret, loc. cit.), certain authors 
generally acknowledge that Psilotum may grow in soil, but many imply that the 
species is primarily either epiphytic or grows on fallen trees, tree stumps, or 
palmetto bases (Lellinger, A Field Manual of the Ferns and Fern Allies, p. 49. 
1985; Mickel, How to Know the Ferns and Fern Allies, p. 180. 1979). Most of the 
Psilotum sites in northern Florida, Louisiana, Texas, South Carolina, and now 
Alabama, too—i.e., the populations at the northern limits of the species distri- 
bution—are indeed terrestrial. Due to floristic similarities apparent between one 


104 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 3 (1987) 


Louisiana locality (Rhodes, loc. cit.) and the one described below, one might 
expect additional populations to be found elsewhere, if not commonly, throughout 
the entire Southeast. 

The Alabama site consists of formerly cultivated land that was abandoned to 
natural succession about 45 years ago (based upon ring counts determined for 
some of the largest trees and information from local residents). Shallow tillage 
furrows still mark the surface of the hillside that now is forested by young 
hardwoods and scattered pines. The larger pines were harvested for pulpwood 
about 18-19 years ago. The property on which most of the Psilotum plants were 
found was purchased by this writer in 1977. Although the variety of plant life at 
the site has been the object of my continued scrutiny for nearly a decade, no 
Psilotum specimens were observed there until 1986. Dominant overstory hard- 
wood species are sweetgum (Liquidambar styraciflua), yellow-poplar (Lirio- 
dendron tulipifera), and northern red oak (Quercus rubra). Understory species 
include dogwood (Cornus florida), black walnut (Juglans nigra), red mulberry 
(Morus rubra}, and Florida maple (Acer saccharum subsp. floridanum). The major 
low shrubs are Carolina buckthorn (Rhamnus caroliniana) and brook euonymus 
(Euonymus americanus); woody vines include Japanese honeysuckle (Lonicera 
japonica), greenbrier species (Smilax glauca and S. rotundifolia), and muscadine 
grape (Vitis rotundifolia). Perennial herbs observed during various seasons over 
the past ten years include Chasmanthium sessiliflorum, Trillium underwoodii, 
T. cuneatum, Uvularia perfoliata, U. sessilifolia, Galium uniflorum, Botrychium 
virginianum, B. biternatum, and Asplenium platyneuron. Plants of C. sessiliflo- 
rum, G. uniflorum, and the ferns dominate the herbaceous ground cover during 
the fall throughout the area where Psilotum was found. 

The substrate on which the woodland described above has developed is typical 
of the Alabama Piedmont: red clays formed from gneissal granite. Shallow topsoil 
only a few centimeters thick overlays the red, rocky soil, and a heavy humus 
layer often 3-6 cm thick is present. Rotting pines killed by dark beetles several 
years ago are scattered on the forest floor, partially covered by leaf litter and 
humus. The underground portions of the Psilotum plants are located mainly 
within the leaf litter and uppermost organic soils. 

Aerial stems of field specimens of Psilotum were still rather small (only 10 cm 
long or less) when a mild frost occurred in November 1986. Shoots of the largest 
specimens were branched in the typical dichotomous pattern six or more times, 
but several were only 2-3 cm tall and were branched once or not at all. The 
length of stem increments between branch points varied considerably from plant 
to plant, and these differences remained evident in and diagnostic for certain 
greenhouse specimens in growth occurring subsequent to transplantation. Most 
transplanted specimens remained green and apparently healthy, but three of the 
eight potted plants wilted and died within a week of being moved. The surviving 
ones showed no evident shock, continued to grow without loss of vigor, and 
attained heights exceeding 15 cm during the winter and spring of 1986-87. They 
remained sterile through the first week of October 1987. In the field, most aerial 
stems exposed above the leaf litter turned brown, softened, and dried out during 
November 1986 before any “killing” frost occurred. The pattern of deterioration 


SHORTER NOTES 105 


in aerial stems commonly was from the base up, the tips of some specimens 
remaining green well into December, when the first subfreezing temperatures 
of the season occurred. 

Underground stems (or rhizomes) in specimens selected and dug for vouchers 
(deposited at AUA and MICH) were branched in the coralloid pattern typical of 
Psilotum. Withered and dried aerial stems, apparently remnants from the 1985 
season, were observed to originate from both aerial and underground stems at 
distances ranging from 2 to 20 mm behind the location of the stem of the 1986 
season. One branched rhizome with an axis 12 cm long had several other erect 
lateral “stubs” where stems of earlier years evidently once were borne. No living 
aerial stems were observed to bear sporangia in 1986, and the withered stems 
for the previous year also showed no sporangia. Since the fall of 1986 was 
unusually mild, it would seem that sporulation probably could occur only very 
rarely (if ever) at this location. The extensive area covered by the field population, 
its development in an area that formerly was cultivated, the number of plants 
present, and their perennial sterility (ensured by annual die-back of aerial stems) 
stand as a combination of factors suggesting that propagation by some asexual 
means may be occurring at this site. 

The habitat type in which this population of Psilotum now exists is rather 
common in Alabama and other southeastern states. That this species may occupy 
other similar sites yet only infrequently be detected may be the actual case, but 
factors accounting for this possibility warrant some discussion. Due to their small 
size, the shoots are very inconspicuous. At the Alabama locality, they are made 
even more obscure because of close superficial resemblance to plants of Smilax, 
Euonymus, and Galium that are abundant, also green, short, stubby, and branched 
due to browsing by white-tail deer. Furthermore, it seems that the inland popu- 
lations consist only of vegetative plants even as late as November. If sporulating 
plants should develop, they would appear only very late in the fall (or winter) 
when few persons familiar with Psilotum would likely chance upon them while 
conducting routine field research. Our 1986 census of the population revealed 
numerous vegetative shoots, but more plants were located during August and 
September of 1987 within the same area that had been closely checked a year 
earlier. In one case, in Hawaii, Psilotum grows as a non-green rhizomorph deep 
within dark lava tubes (Wagner, pers. comm.) without forming typical aerial stems 
or sporangia, so the species seemingly could live primarily as rhizome forms in 
other parts of its range, too. Only further demographic studies and observations 
can test this hypothetical possibility concerning the scarcity of inland collections 
of Psilotum from woodland habitats. In the meantime, features of the habitat of 
the only known Alabama population of Psilotum suggest that it may occur in 
{and that we should search for) other inland and upland sites. 

I acknowledge with gratitude the assistance of Alvin Diamond, Susan Scott, 
Daureen Miller, Harland Hendricks, and Charlotte Tanner in locating plants of 
Psilotum at this locality and in determining the apparent limits of the population. 
I also thank Herb Wagner for his helpful comments and encouragement in 
preparing this report.—JOHN D. FREEMAN, Department of Botany and Microbi- 
ology, Auburn University, AL 36849. 


American Fern Journal 77(3):106-108 (1987) 


REVIEWS 


“Field Guide to the Ferns and Other Pteridophytes of Georgia,” by L. 
Snyder, Jr., and J. G. Bruce. 1986. 270 pp. Athens: University of Georgia seat 
ISBN 0-8203-0838-2 (cloth, $25.00), ISBN 0-8203-0847-1 (paperback, $12.50). 

This is an up-to-date flora of the ferns and fern allies of Georgia, patterned 
after its forerunner, the 1951 Ferns of Georgia, by McVaugh and Pyron. Much 
new has been added to the knowledge of Georgia ferns, and there could probably 
still be a few more county dots added, but this looks like 1986 state-of-the-art. 
Each taxon is illustrated with a full page line drawing, often with detailed insets; 
many are from the McVaugh and Pyron illustrations, some are new. The illus- 
trations are representative and adequate for identification in most cases, espe- 
cially with the ferns. Each taxon also has the usual description, range, and habitat, 
and a county dot map on the facing page for easy reference. An introductory 
section outlines fern morphology, physiographic regions of the state, and how to 
use the book. There is a combination morphology-habitat outline to genera. This 
substitutes for the conventional dichotomous key, which is omitted. I don’t think 
this will be helpful, but beginners will find their way by making good use of the 
illustrations, and advanced fern enthusiasts probably bypass most generic keys 
anyway. 

This is a careful treatment. Mr. Snyder, retired and not a trained biologist, 
thought a few years ago that it would be fun to learn a fern or two. I think he 
has graduated from hobbyist to semipro. Dr. Bruce knows the subject academi- 
cally and thoroughly. They have made a good team. There are the usual nit- 
pickings which one can inevitably make with a flora full of records and descrip- 
tions. I doubt that the descriptions or illustrations are adequate for Asplenium 
resiliens, A. heteroresiliens, and A. heterochroum. They are close and subtle; 
one may also have to resort to number of spores per 32 in A. resiliens 
and A. heteroresiliens and 64 in A. heterochroum. Pilularia is also known from 
Tennessee. In Asplenium and Cystopteris the polyploid species are referred to 
as “fertile hybrids” of named parents, which is an oversimplification, and con- 
fusing when “‘sterile hybrids” are also listed. These ‘fertile hybrids’’ would be 
better considered as “‘fertile species of hybrid origin,” or as ‘‘allopolyploid species 
derived from named parents” as was done for Dryopteris celsa. Lastly, there is 
no discussion of the arrangement of genera. They are placed in a more or less 
phylogenetic sequence, except for the sa af Polypodium “up front’ be- 
tween Osmunda (O 1 Lygodium S }, and the blechnoid 
ferns with elongate sori with tecinione: and Athyrium; these are placed de- 
scriptively with genera having similar so 

One of the high points of the book is oe treatment of the bog clubmosses, a 
special research interest of Dr. Bruce and an especially confusing group on the 
Georgia Coastal Plain. The line drawings help but the photographic plates are 
all one could ask to help unravel the three species and three hybrids of the 
Lycopodium al appressum—prostatum com 


mplex 
Those interested i in the pteridophytes of this part of the wot will be glad to 


REVIEWS 107 


see how much has changed since 1951.—A. M. Evans, Botany Department, Uni- 
versity of Tennessee, Knoxville, TN 37996. 


“Ferns and allied plants of Victoria, Tasmania and South Australia,” by Betty 
D. Duncan and Golda Issac. 1986. 258 pp. Melbourne Univ. Press. ISBN 0-522- 
84262-3. Available hardcover only, exclusive U.S. distributor: International Spe- 
cialized Book Services, Inc., 5602 N.E. Hassalo Street, Portland, Oregon 97213- 
3640. $20.00 + $2.25 postage. 

This book is a field guide for naturalists, gardeners, and professional botanists. 
The introductory chapter makes the book understandable to the layman since it 
covers the fern life cycle, morphology, taxonomy, and how the dot distribution 
maps were prepared. In the following taxonomic chapters, the Polypodiaceae 
sensu latissimo is divided into twelve families, and genera are treated within the 
families. The genera are often narrowly defined, e.g., Christella and Pneuma- 
topteris are separated from Thelypteris; Polyphlebium, Macroglena, and Ap- 
teropteris are separated from Hymenophyllum and Trichomanes. There are no 
keys to the families, but the book has two keys to the genera: 1) an illustrated, 
dichotomous key, and 2) a foldout tabular key that resembles a sideways den- 
drogram. The species treatments describe 130 pteridophytes in 53 genera. Syn- 
onyms are not listed unless widely used or controversial, as in Holttum’s vs. 
Tryon’s classification of tree ferns. The species descriptions are brief, followed 
by a list of the most important field characters. The discussions deal with habitats, 
variation, and other features of the plants, and here the authors show that they 
have considerable field experience as evidenced by their original observations 
about the ecology of ferns in the region. Each species treatment ends with notes 
on cultivation and a statement of world range. Following the taxonomic chapters 
is a nine page, illustrated chapter on growing ferns (by C. J. Goudey and R. J. 
Hill). The authors emphasize growing ferns from spores, special requirements of 
epiphytic or rupestral ferns, and problems with insects and other pests. A com- 
prehensive bibliography, glossary of terms, and index to names conclude the 
book. 


The book’s greatest strength is its superb illustrations. Almost all the species 
have black and white photographs showing sori, leaf cutting, and, in many cases, 
habit and habitat. In addition, eight color plates, each with six pictures, are 
interspersed throughout the text. All the photographs are sharp, detailed, and 
well reproduced. I congratulate Bruce Fuhrer, the photographer, for his fine work. 
In addition to the photographs, line drawings are given and often show details 
of scales, hairs, or other features difficult to capture on film. The dust jacket is 
an attractive watercolor showing ten species of the region. 

The book has, however, two weaknesses. First, although it covers Victoria, 
Tasmania, and South Australia, dot distribution maps are given only for Victoria. 
Second, cytological information is lacking—| table b h inf ti 
is important in fern classification. Nevertheless, the book is well written, beau- 
tifully illustrated, and reasonably priced. I recommend it to everyone from down 


108 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 3 (1987) 


under and anyone unfamiliar with the Kangaroo fern family.—RosBIN C. Moran, 
Missouri Botanical Garden, Box 299, St. Louis, MO 63166-0299. 


“Illustrations of Pteridophytes of Japan, Volume 5” edited by S. Kurata and T. 
Nakaike with the cooperation of the Nippon Fernist Club. 1987. x + 818 pp. + 
folding map. University of Tokyo Press. Yen 14,000. ISBN 4-13-061065-1. 

The fifth volume, with another hundred taxa of Japanese ferns, is as splendid 
as the preceding four, all of which have recevied reviews in this Journal [for 
review of preceding volume, see Amer. Fern J. 77:65. 1987]. Volume five is 
principally devoted to Hymenophyllaceae, Blechnaceae, Vittariaceae, Adiantum, 
Arachniodes, Cyrtomium, and Lindsaea. 

The three beautiful color plates on two pages of frontispiece, a feature of each 
of the volumes, here depict species having reddish fronds, at least when young. 
The red pigments are presumably anthocyanins, once thought to be absent in 
ferns. 

Names of several included species are subject to nomenclatural change, es- 
pecially in Hymenophyllaceae. Arachniodes hasseltii has been placed in Dryop- 
teris subg. Nephrocystis (H. Ito) Fraser-Jenkins [Bull. Brit. Mus. (Nat. Hist.)}, Bot. 
14:197. 1986], and Ctenitis maximowicziana in the genus Dryopsis Holttum & 
Edwards [Kew Bull. 41:179. 1986]. All plants named Adiantum diaphanum Blume 
from Japan, S. China, and the Philippines should be called A. setulosum J. Smith, 
since the Javan species which includes Blume’s type is very distinct, characterized 
by glabrous indusial flaps.—M. G. Price, Herbarium, North University Building, 
University of Michigan, Ann Arbor, MI 48109. 


“A key to the genera of New Zealand ferns and allied plants,” by P. J. Brownsey 
and T. N. H. Galloway. 1987. 31 pp. National Museum of New Zealand Miscel- 
laneous Series no. 15. Available from the Librarian, National Museum, Private 
Bag, Wellington, New Zealand. NZ $5.95 (incl. postage). ISSN 0110-1447. 

This publication provides an illustrated key for identifying the 66 genera of 
ferns and fern allies occurring in the New Zealand botanical region. Some 92 
species (out of a total of 211) are illustrated by line drawings, with every genus 
represented by at least one illustration. Technical terms have been kept to a 
minimum, and all are defined in the illustrated glossary. 


INFORMATION FOR AUTHORS 


Authors are encouraged to submit manuscripts pertinent to pteridology for 
publication 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 pre- 
pared may be returned for revision prior to review. Submit manuscripts in trip- 
licate (xerocopies acceptable), including review copies of illustrations. Do not 
send originals of illustrations until they are requested. Use standard 814 by 11 
inch paper of good quality, not “erasable” paper. Double space manuscripts 

throughout, including title, authors’ names and addresses, text {including heads 
and keys}, literature cited, tables (separate from text), and figure captions (groupe 
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. 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 mea- 
sures (e.g., distance, elevation, weight) unless quoted or cited from another source 
(e.g., specimen citations). For nomenclatural matter (i.e., synonymy and typifi- 
cation), use one paragraph per basionym (see Regnum Veg. 58:39-40. 1968). Ab- 
breviate titles of serial publications according to Botanico-Periodicum-Huntian- 
um (Lawrence, G. H. M. et al., 1968, Pittsburgh: Hunt Botanical Library). | 
References cited only as part of nomenclatural matter are not included in lit- _ 
erature cited. For shorter notes and reviews, put all references parenthetically — 
in text. Use Index herbariorum (Regnum Veg. 106:1-452. 1981) for designations 
of herbaria. oe 

a eee ee c 
page. Provide margins of at least 25 mm on Tl ill a 
illustrations, design o 
amount. In Composite | locks, abut edges of adjacent phatahs ‘Avoid co com- _ 
locks. Coordinate 


es ms 


7 


conga scales and ile in figures themselves, me — ; is. Inchude @ 


sequence and | bering | Ir 5 (a d of tables rithondor of caion nist 


AMERICAN 
FERN 
JOURNAL 


Volume 77 
Number 4 


October-December 1987 


QUARTERLY JOURNAL OF THE AMERICAN FERN SOCIETY 


A New Fern — Western Mexico and its Bearing on the Taxonomy of the Cheilanthoid 
Fern 


The Ochreole of Equisetum: A Prophyllar Sheath 


New Combinations in Megalastrum (Dry 


Variations in Petiolar Structure of Hypodematium crenatum 
P. K. Khare and Rama Shankar 131 


The Identity of Hymenophyllum cristatum 


Shorter Note 
Two Species of Adiantum New to Florida 


Review 
Referees, 1987 
Index te Volume 77 


Information for Authors 


John T. Mickel 109 


Richard L. Hauke 115 


) 
Alan R. Smith and Robbin C. Moran 124 


Robert G. Stolze 137 


Clifton E. Nauman 141 


The American | Fern Society 


r 1987 
FLORENCE S. WAGNER, Dept. of —- . of Michigan, Ann Arbor, MI 48109. 
President 
JUDITH E. SKOG, Biology Dept., George Mason University, pct sn 22030. Vice-President 
W. CARL TAYLOR, Milwaukee Public Museum, Milwaukee, WI Secret 


ary 
JAMES D. CAPONETTI, Dept. of Botany, University of Tennessee, a a ™ mipoge A presi! 

DAVID S. BARRINGTON, Dept. of Botany, University of Vermont, t, Burlington, VT 0 
Sinan Pveunager 


JAMES D. MONTGOMERY, Ecology III, R.D. 1, Berwick, PA 1860 Back Issues Curator 
ALAN R. SMITH, Dept. of Botany, University of California, Pek. CA 94720. Journal Editor . 
DAVID B. LELLINGER, Smithsonian Institution, Washington, DC 20560. Memoir Editor _ 


DENNIS Wm. STEVENSON, New York Botanical Garden, ee NY 10458. 
Fiddlehead Forum Editor 


American Fern Journal 
WEAN RSI 8 eS Dept. of Botany, —— of California, 
: . keley, CA 94720 
ne ASSOCIATE EDITORS 

GERALD] GASTONY ..... Dept. of Biology, ci Sgeiene Bloomington, IN 47401 
ogame SURUIPUE 2 a of Botany, University of Kansas, 
Bee wrence, KS 66045 
e vAVED. B. LELLINGER i ne a ae U.S. Nat'l Herbarium NHB-166, Smithsonian Institution, 


oe ashington, DC 20560 
Se — R. WEBSTER . Biological Sciences Group, University of Connecticut, Storrs, CT 06268 
_ The “American Fern Journal” (ISSN 0002-8444) is an illustrated quarterly devoted to the general 
‘study of ferns. It is owned by the American Fern Society, and published at the Pringle Herbarium, 
a University of Riser, Burlington, VT 05405. Second-class postage paid at Burlington, VT, and 
Claims f 


le 6 ths (f ft e date of issue, 


and ordars for bck aes shouldbe addroed to Dr James D. Montgomery, eckers ED 1 
Berwick, PA 18603. 
nd applications for membership should be sent to the Records Trea- 
 surer. 
rr . tT : 3: 33 : the 


tions $12.00 gross, $11.50 net if paid through an agency baseuent fee $0.50); sent free to 
‘members of the American Fern Society (annual dues, $10.00 + $4.00 mailing surcharge beyond 

U.S.A., Canada, and Mexico, wm; life membership, mabe 
- Back volumes 1910 k t f 64 pages or less, $1.25; 65-80 
pon Shinada elaepaag sich pt shiping Back volumes 1979 et seq. $8.00 each; 
Ten percent discount on orders of six volumes © or 


nges to AMERICAN FERN JOURN t, Dept of — er of 
-Fiddlehead Forum 


oes, and reviews of f non-technical books on ferns. 


oci able it to ¢ xpani nd | teeter 


ee 3 
exchange or fra cine We aea personalia, : 2 


American Fern Journal 77(4):109-114 (1987) 


A New Fern from Western Mexico and its 
Bearing on the Taxonomy of the 
Cheilanthoid Ferns 


OHN T. MICKEL 
New York Botanical Garden, Bronx, New York 10458 


Throughout the history of pteridology, and in taxonomy in general, it has been 
stressed that the reproductive parts are conservative and taxonomically most 
useful. In the ferns the sori have been the focal point of taxonomic scrutiny 
because they seem to be dependable indicators of relationship, especially at the 
generic and higher levels. The vegetative parts generally vary much more and 
are taxonomically less dependable. Linnaeus recognized 15 genera based largely 
on sorus form. The general form of the sorus was examined more closely as years 
went on, and it was found that the same general sorus type was often reached 
in different ways. There was heavy reliance on the sorus until well into the 
present century. The sorus position was vital to Bower (1923-1928) in his con- 
struction of major phyletic lines in the ferns. The recent prevailing philosophy 
among pteridologists has been to use an assortment of characters, with modern 
studies extending to finer microscopic detail of morphological features (especially 
with the SEM) and chemical examination. 

Thus, Acrostichum, originally encompassing all ferns with sporangia spread 
across the back of the fertile frond, now contains but three species, with other 
groups having acrostichoid sori, such as Elaphoglossum, Bolbitis, Lomariopsis, 
and Polybotrya, being recognized as having no close ties with Acrostichum. 
Similarly, species belonging at one time to Polypodium, with dorsal, round, 
exindusiate sori, included species now placed in Grammitis, Thelypteris, Cy- 
athea, Lophosoria, Dryopteris, and others. 

In the Adiantaceae (s.1.) there are two basic sorus types. Several genera have 
the sorus at the end of veins with a distinct, differentiated marginal flap, or false 
indusium, recurved to protect the sorus. This includes such genera as Pteris, 
Cheilanthes, Pellaea, Doryopteris, Cryptogramma, and Llavea. 

On the other hand, “gymnogrammoid” sori, with the unprotected sporangia 
along the veins, are found in Pityrogramma, Anogramma, Hemionitis, Bom- 
meria, Antrophyum, Hecistopteris, Jamesonia, Eriosorus, Pterozonium, Conio- 
gramme, and others. 

During a study of the ferns of western Mexico, the author came across a plant 
(Figs. A-C) that appeared to be a hybrid between two species that represent these 
two extremes of figuration: Hemionitis subcordata (D. Eaton ex Davenp.) 
Mickel with gymnogrammoid sori (Figs. F, G) and Cheilanthes skinneri (Hook.} 
Tryon & Tryon with marginal sori (Figs. D, E). 

The specimen has characters of the presumed parents (Table 1), both of wnich 
were present at the type locality. (Selaginella pallescens was the only additional 
pteridophyte collected there.) Most notably, the plant is intermediate in dissection 


sayss0URt BOTANICSS 


110 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 4 (1987) 


Fics. A-C. Cheilanthes gryphus (McVaugh 15908). A, Habit. B, Rhizome scale. C, Pinnule. D, E. 
Cheilanthes skinneri (Mickel 727, ISC, Guerrero). D, Pinna. E, Pinnule. F, G. Cheilanthes subcordata 
{Mickel 6241, NY, Oaxaca). F, Pinna. G, Close-up of pinna. 


J. T. MICKEL: CHEILANTHES 


111 


TABLE 1. A Comparison of Characters in Cheilanthes subcordata, C. gryphus, and C. skinneri. 


Character 


C. subcordata 


C. gryphus 


C. skinneri 


Rhizome stout, short- 
reeping 
Rhizome scales bicolor- 
ous 
Stipe tan, sulcate- 
winged 


ng 
Rachis sulcate-winged 
Blade dissection 


Pinnae, length of petio- 
lule 

Pinnae, number of pairs 
Pinnae shape 


Pinnae basiscopically 


Lower blade surface 
sparsely pubescent 


eins 

Proximal portion of low- 
er surface of pinna 
midvein 

Pinna margin 


Sori 


Spores 
Spore size 


x 
x 


Xx 


1-pinnate 


1-3 mm 


lance-ovate, acute 


X, pale 


many anastomoses 
black 


not modified, nar- 
row, revolute 


along distal % to 
all of vein 

cristate 

31.5 wm 


(nearly 2-pinnate vs. 1 
are limited to the distal 1% of the veins (vs. most 0 


pinnate-pinnatifid 
t innate 


4-5(-10) mm 


5-7 
lanceolate, acu- 


X, pale 


few anastomoses 
+ black 


irregularly modi- 
fied, medium 
width and high- 
ly modified in 


places 
along distal “% of 


ei 
cristate 
30.1 uhm 


2-4-pinnate-pinna- 


d 
(6-)10-25 mm 
8-12 
lance-deltate, acu- 
minate 


X, green 
free 


green 


highly modified, 
discrete or con- 
tinuous, revolute, 
b 


at vein tips 


cristate 
31.3 pm 


-pinnate and 3-pinnate in the presumed parents), the sori 
f the vein length and vein ends), 


and the veins are casually anastomosing (vs. free and netted). On the other hand, 
it seems to be fertile. Chromosome 


the few spores present are well formed, so 
studies by Michael Windham have shown it to 
Ranker (in litt.) show th 


litt.). Furthermore, isozyme studies by 


the intermediate not to be cumulative of the presum 
expected of hybrid plants. Thus, 
parents and intermediate in several o 
the plant is a fertile diploid species ra 


however, on reproduction in this species group. 


e presumed parents of the 
configuration that it had not previously 
allied. They had been placed by all taxonom 
even in different families. Pichi Sermolli (1977), for example, 


plant in question are so 


be diploid, n = 30 II (Ranker, in 
e compounds of 
ed parents as would be 
although it is found only with its presumed 
utstanding characters, it is concluded that 
ther than a hybrid. More work is needed, 


distinct in their sorus 
been suspected that they were closely 
ists in separate genera and by some 
placed Hemionitis 


112 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 4 (1987) 


in the Hemionitidaceae and Cheilanthes in the Sinopteridaceae. Holttum (1947) 
suggested, however, that Pityrogramma and Hemionitis, although with gymno- 
grammoid sori, belonged with the cheilanthoid group rather than with the other 
gymnogrammoids, and Tryon and Tryon (1982) placed Bommeria and Hemionitis 
with the cheilanthoid ferns (tribe Cheilantheae), leaving Pityrogramma with Ja- 
mesonia and allies in tribe Taenitideae. On the basis of habitat, range, rhizome 
scales, and spore morphology, I would concur with this relationship, and the 
discovery of this plant suggests an even closer relationship than was previously 
supposed. 

Neither of the presumed parents is typical of its genus. Cheilanthes skinneri 
has until recently resided in the genus Pellaea s.]. It was recently transferred to 
Cheilanthes (Tryon & Tryon, 1981), although it is not typical of that genus either. 
Hemionitis subcordata has been placed in Coniogramme and Gymnopteris, the 
latter in turn later combined with Hemionitis (Mickel, 1974), but it is not a typical 
member of that genus either, differing from most species by the subglabrous 
blade, crested spores and different chemistry (Mickel, 1974; Giannasi & Mickel, 
1 


Apparently, characters other than sorus position should be used to tie this small 
group of species together. Characters they have in common include bicolorous 
rhizome scales, blade indument of hairs blending to narrow scales, and perhaps 
most significantly segments and pinnae decurrent onto the rachis, making a single 
sulcus on the stipe. If indeed this is a natural and distinct group (or even genus), 
other species that may belong here include Cheilanthes lozanii (Maxon) Tryon 
& Tryon, which agrees in rhizome and blade indument but not decurrent seg- 
ments, and C. bolborrhiza Mickel & Beitel, ined., which agrees in its decurrent 
segments but has concolorous rhizome scales and glabrous lamina. 

If Cheilanthes skinneri and Hemionitis subcordata do belong to the same 
genus, as some common characters suggest, we do not at this time know what it 
would be. Because of the drastically different sori, Cheilanthes skinneri cannot 
be a Hemionitis and Hemionitis subcordata does not fit well in Cheilanthes. 
Circumscription of a distinct new genus to accommodate them both is premature 
until we know more about the variation of the characters in other cheilanthoid 
ferns, especially since there seem to be no other species that share the common 
characters of C. skinneri and H. subcordata. 

Because of the evidence against the plant representing an intergeneric hybrid 
between Cheilanthes and Hemionitis, I am treating this small group within Chei- 
lanthes, recognizing that it may eventually deserve generic rank. This requires 
a transfer of Hemionitis subcordata (D. Eaton ex Davenp.) Mickel: 

4 Cheilanthes subcordata (D. Eaton ex Davenp.) Mickel, comb. nov.—Gymno- 

4%>\ gramma subcordata D. Eaton ex Davenp. in Rose, Contr. U.S. Natl. Herb.5: 
138, t. 16. 1897. 

44%? Cheilanthes gryphus Mickel, sp. nov. (Figs. A-C). Tyre: Mexico, Colima, low 
mountain summits 7 mi N of Santiago, road to Durazno, Jalisco, via the bridge 
over Rio Cihuatlan, deciduous llands, 200 ft, 30 Jul 1957, McVaugh 15908 
(MICH). 


Inter Cheilanthem skinneri et C. subcordatam notulis nonnullis intermedia. 


J. T. MICKEL: CHEILANTHES 113 


Rhizoma horizontale compactum; paleae lineares bicolores; lamina pinnato-pin- 
natifida; pinnae 5-7 jugae obtuse 2-4-lobae; nervi hinc inde anastomosantes; 
sporangia secus venas distaliter per 2-4 mm seriatim disposita; sporae tetrahedro- 
globosae cristatae. 

The specific epithet derives from L., gryphus, griffin, a hybrid between a lion 
and an eagle, hence an archetype of the union of incompatibles. 

Rhizome horizontal, compact, 3-7 mm diam.; rhizome scales 2-3 mm long, 
linear, extremely slender, bicolorous, with a central, lustrous, dark castaneous to 
atropurpureus streak and narrow, tan margin; fronds to 46 cm tall, approximate; 
stipe stramineous, grayish-brown at base, with very slender, mostly bicolorous 
scales at base, otherwise glabrous, deeply 3-grooved, about equalling the blade 
in length; blade broadly lanceolate or moderately deltate, pinnate-pinnatifid to 
nearly bipinnate at base; pinnae 5-7 pairs, each lanceolate to deltate, acuminate, 
with 2-4 pairs of broad obtuse lobes, the basiscopic lobes ca. twice as long as the 
acroscopic ones; upper surface glabrous, lower surface with sparse straight hairs, 
which grade into very sparse, narrow, bicolorous scales on the rachis and pinna 
rachises: veins mostly free but with occasional anastomoses (2-5/pinna); sporan- 
gia running along the terminal 2-4 mm of the veins, indusium lacking, blade 
margin not reflexed nor modified; spores tetrahedral, crested. 


Paratype: Mexico, Nayarit, S of Tepic along Hwy 200 near km post 24 between Colonias and El 
Refilén, mixed deciduous forest in moist, shaded gully, 8 Jul 1985, Ranker 799 (KANU). 


Several other fern genera or generic groups include taxa with highly disparate 
soral dispositions. Tectaria, with round, dorsal sori, is closely allied to Cionidium, 
which has extramarginal sori. Polypodium, with round, dorsal sori, is close to 
Dicranoglossum and Neurodium with linear, submarginal sori. Polystichum, with 
round, dorsal, indusiate sori, has a splinter Plecosorus, in which the indusium is 
lost but the sorus is protected by a differentiated margin (false indusium) as in 
Cheilanthes. Asplenium has elongate, dorsal sori, whereas Diellia and Loxo- 
scaphe have sori in subterminal cups. Perhaps the most dramatic example of 
diverse sorus types is seen in Tectaria and its close relative Dictyoxiphium, which 
has linear, marginal sori, and the intergeneric cross, x Pleuroderris michleriana, 
as described in detail by Wagner et al. (1978). 

Within the Adiantaceae itself there are some transitional states between mar- 
ginal and gymnogrammoid sori, although they are not as dramatic as that found 
in Cheilanthes gryphus and allies. In Adiantum the sori, although protected by 
the marginal flap, are located along veins under the false indusium. In the mar- 
ginal sori of Pellaea the sori are, in fact, somewhat elongate along the veins, 
sometimes a millimeter or two, and in one species, P. bridgesii, the sori are medial 
along the veins for 1-2 mm (Wagner et al., 1983). Haufler (1979) has pointed out 
the extreme variation of sorus length in Bommeria. 

The case of Cheilanthes gryphus suggests that in this complex, at least, sorus 
pattern is of little taxonomic significance. We need to find new sets of characters 
for the generic segregation in the whole family, e.g., rhizome scales, blade in- 
dument, axis grooving, gametophyte morphology, breeding systems, and chem- 
istry. All characters must be examined critically without heavy reliance on tra- 
ditional characters. 


114 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 4 (1987) 


ACKNOWLEDGMENTS 


The drawings were prepared by Bobbi Angell. I am grateful to W. H. Wagner Jr., Thomas Ranker, 
Christopher Haufler, and Timothy Reeves for their suggestions and helpful information, to Michael 
Windham for the chromosome count of Cheilanthes gryphus, and to Rupert Barneby for his help 
with the Latin diagnosis. 


LITERATURE CITED 


Bower, F. O. 1923-1928. The aces 3 vols. Cambridge: Cambridge Univ. Pre 

see D. E. and J. T. Mic 1979. Systematic implications of sien cicueeiie in the fern 
us Hemionitis eae Brittonia 31:405-412. 

HAUFLER, io H. 1979. A biosystematic revision of Bommeria. J. Arnold Arbor. 60:445-4 

Ho.trum, R. E. 1947. A revised classification of leptosporangiate ferns. J. Linn. Soc., ate 53:123- 


MICKEL, J. T. 1974. A redefinition of Hemionitis. Amer. Fern J. 64 


PicH! SERMOLLI, R. E. G. 1977. Tentamen pteridophytorum genera in pat sabathes ordinem redi- 
endi. Webbia 31: 313-512 
Tryon, R. M. and A. F. TRYON. 1981. Taxonomic and nomenclatural notes on ferns. Rhodora 83: 
pened? ne 


1982. Ferns and allied plants with special reference to tropical America. 
ee York: Springer-Verlag 
Wacner, W. H., Jr., F. S. WAGNER, and L. D. Gomez P. 1978. The singular origin of a Central 
American a muses michleriana. Biotropica 10:254-264 
———, A. R. Smitu, and T. R. Pray. 1983. A aoe brake hybrid, Pellase bridgesii x mucronata, 
and its systematic significance. Madrofio 3: 


American Fern Journal 77(4):115-123 (1987) 


The Ochreole of Equisetum: A Prophyllar Sheath 


RICHARD L. HAUKE 
Department of Botany, University of Rhode Island, Kingston, Rhode Island 02881 


The genus Equisetum is unique in its mode of branching. The branches are 
borne at the nodes, but they develop exogenously from the top of the nodal ring, 
alternating with the sheath segments or leaves. The branch buds develop between 
leaves rather than being axillary to the leaf, which is the arrangement found in 
seed plants, or with no relation to the leaves, as in ferns. It is obvious that the 
branches are associated with the node below them, rather than the nade above, 
because if the number of leaves at a node changes, the number of branches is 
the same as the number of leaves below it. Although exogenous in origin, the 
branch bud is enclosed by the leaf sheath and must grow out through its base. 
Where it emerges, a fringe of sheath tissue is left around the branch base as a 
collar. The base of the branch is surrounded by a sheath-like structure, which 
differs from the regular nodal sheaths in being usually scarious, brownish in 
color, and unvascularized, and in lacking stomata and a subjacent internode. 

This structure was described at length by Duval-Jouve (1864, as “gaine basi- 
laire”) who followed Vaucher (1821) in considering it a protective structure for 
the branch bud, analogous to the bud scales of phanerogams. Milde (1867, p. 155) 
thoroughly discussed the ‘‘Asthiille (Ochreola).” He pointed out that it is always 
different in color from the branch sheaths, with short, blunt teeth. According to 
Milde, the ochreole does not grow from the branch but rather develops from the 
inner cylinder of the stem. He claimed it lacked commissures so was not of fused 
leaves, but of a single, toothed leaf. The outer (abaxial) angle is different from 
the others in being thick, with a vascular bundle, chlorophyllous cells, and scat- 
tered stomata. Thus the ochreola represents a bract with a midrib, in the axil of 
which the branch develops. Milde emphasized that the ochreole does not ter- 
minate an internode, as do the branch and stem sheaths. It belongs not to an 
internode, but rather to the whole branch. He also called attention to the similarity 
between this and the sheath subtending the juvenile plant as it develops from 
the embryo. On a later page in the same publication Milde (1867, p. 379) corrected 
and expanded his earlier description. The Asthiille, unlike a bract, belongs to 
the branch rather than the main axis, and is the first leaf whorl on the branch. 
He compared it to the first leaf (the prophyll) on most monocot branches. 

Janczewski (1876) described the initiating bud as forming a vegetative cone 
and a basal portion. As the vegetative cone develops, the upper part of the basal 
portion is elevated to form a rim around the vegetative cone, and this rim rep- 
resents the first sheath of the branch. Janczewski found Milde’s interpretation 
of this as a bract subtending the branch very odd (“bien singuliere”). DeBlock 
(1923) described the ochreole as I nd becoming brown, with no trace 
of stomata, chlorophyll, or vascular tissue. Johnson (1937) in describing hyda- 
thodes in Equisetum leaves, commented on their absence on the ochreole, which 
is thin and usually devoid of vascular tissue. Even when vasculated, the ochreole 


116 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 4 (1987) 


lacks hydathodes according to Johnson. I find no mention of the ochreole in 
standard morphology texts, such as Eames (1936), Smith (1938), Bierhorst (1971), 
Foster and Gifford (1974), or Sporne (1962). 

This paper is intended to determine the accuracy of descriptions of the ochreole, 
to resolve discrepancies among them, and to compare the ochreole to the prophyll 
of seed plants. 


MATERIALS AND METHODS 


Buds of Equisetum arvense (Hauke 514, collected 1 Nov 1977) and E. telmateia 
(Hauke C1, collected 17 Nov 1980), and young shoots of E. diffusum (Univ. Rhode 
Island greenhouse), and E. x schaffneri (Univ. Rhode Island greenhouse) were 
embedded in paraffin, serially sectioned, stained with safranin and toluidine, 
and studied with light microscopy. Stem nodes with branches were removed from 
herbarium sheets of all of the branched species of Equisetum, E. ramosissimum 
(Hauke s.n.), E. myriochaetum (Hauke 238), E. giganteum (Hauke 392), E. syl- 
vaticum (Hauke 13), E. pratense (Hauke 40), E. bogotense (Hauke 228), E. palustre 
(Hauke 51), E. fluviatile (Hauke 43), E. arvense (Hauke 45), E. telmateia (Hauke 
C1), E. diffusum (Hauke s.n.), and E. xlitorale (Hauke 44), all in KIRI. These 
were soaked in detergent to hydrate, cleared with 10% aqueous KOH, stained 
with tannic acid and ferric chloride, then dehydrated in ethanol. The ochreoles 
were removed from the nodes, mounted on slides in diaphane under a cover 
slip, and examined for vasculation, commissures, stomata, and other features 
expected in sheaths. Photographs were made on Kodak Plus-x film using a Nikon 
Microflex UFX camera mounted on a Zeiss microscope. 

Hydrated nodes of E. bogotense, E. sylvaticum, and E. diffusum from the 
sources cited above were dehydrated with ethanol, critical point dried using 
liquid CO, in a Samdri-PVT-3B dryer, mounted on copper boats, gold-palladium 
coated with a Hummer II coater, and photographed on Polaroid film using a 
JEOL 1200 EX electron microscope in the scanning mode. 


RESULTS 


Ochreoles, or basal sheaths of Equisetum branches, although all relatively 
membraneous, vary among the different species of Equisetum. Equisetum dif- 
fusum has the most complex, well-developed ochreole (Fig. 1), with prominent 
teeth (Fig. 2), commissures (Fig. 3), vascular strands and hydathode stomata (Figs. 
4, 5) [considered to be such because they overlie the veins and are only on the 
adaxial surface; aerating stomata do not lie over the veins and are found only 
on the abaxial surface of the sheath (See Johnson, 1937)]. Equisetum ~ litorale 
has prominent teeth, commissures, and vascular strands, but lacks stomata or 
hydathodes (Figs. 6, 7). Equisetum sylvati has prominent teeth with long hairs, 
a feature seen only in this species and stomata, but no vasculation, and obscure 
commissures (Figs. 8, 9). Equisetum arvense has prominent teeth, and E. palustre 
has prominent commissures. The other species have more rudimentary ochreoles 
(Fig. 10), with only E. bogotense showing papillae in the center of most abaxial 


R. L. HAUKE: EQUISETUM OCHREOLE 


Fic 1. Node with branches sd ochreoles. 2. SEM of ochreole 
with basal collar (b), and commissure (arrow). 3. SEM of commissur 


-3. Equisetum diffusum ochreole. 


‘ P t LAg @ 8 
PPPON TORO REE? ee 


ee 
da 


I h i ascular strana 


E. diffusum. 4. Och 


Fics. 4-7, Cleared ochreoles. Figures 4-5. 
and adaxial hydathode stomata. 5. As in Figure 4 but focused on abeiial epidermis. Figures 6 and 7. 
. Vascular strand. 


Equisetum ~ litorale. 6. Teeth, commissure, and cells overlying vascular strand. 7. 
118 


R. L. HAUKE: EQUISETUM OCHREOLE 119 


& * Be - © 2g < 
io | 100vn MEMO 


Fics. 8-11. Ochreole of Equisetum. Figures 8 and 9. SEM of Equisetum sylvaticum. 8. Whole ochreole 
with teeth bearing hairs and with basal collar (arrow). 9. Stomate (s) and a tooth hair (h). Figures 10 
and 11. Equisetum bogotense. 10. SEM of base of branch with ochreole and basal collar {arrow}. 11. 


Clearing showing papillae on surface of ochreole. 


120 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 4 (1987) 


cells (Figs. 10, 11). All species studied show a basal collar formed from the outer 
tissues of the nodal sheath through which the branch has erupted (Figs. 2, 8, 10, 
18). 

The development of the ochreole was similar in the four species studied from 
sectioned material, E. diffusum, E. arvense, E. telmateia and E. x schaffneri. The 
branch is initiated by a prominent cell from the upper part of the nodal ring of 
the parent stem, between teeth (Fig. 12). This divides to form a pyramidal apical 
initial cell and several flank cells (Fig. 13). From the basal flank cells a ring forms 
(Fig. 14), and apical growth of this ring produces a basal sheath (Fig. 15). Also 
from the basal part of the branch is initiated a root apex (Fig. 15). As the branch 
bud develops further, the basal sheath grows longer but remains mostly 2 cell 
layers thick, and additional nodal rings and sheaths are initiated (Figs. 16, 17, 
18). The nodal rings other than the basal one develop intercalary meristems (Figs. 
17, 18). As the branch grows longer, it erupts through the sheath enclosing it (Fig. 
18). The well-developed ochreole of E. diffusum has four segments, of which the 
two abaxial ones are thicker, with vascular strands, and the two adaxial ones are 
simpler, only 2 cell-layers thick (Figs. 19, 20). 


DISCUSSION 


The ochreole of Equisetum possesses a sufficient number of sheath-like fea- 
tures—teeth, commissures, vascular strands, stomata, and hydathodes—to suggest 
its identity as the first sheath on the branch. This basal sheath is much reduced, 
and no one species of Equisetum has an ochreole that displays all of these sheath 
characteristics. The early ontogeny of the ochreole is also indicative of its sheath 
nature, with the exception of the lack of formation of an intercalary meristem 
and resultant subjacent internode. If DeBlock (1923) and Johnson (1937) had 
looked at ochreoles of additional species of Equisetum, they would have found 
vasculation, stomata, and hydathode stomata. 

The similarity of the ochreole to the sheath subtending the juvenile plant 
developing from the embryo, as pointed out by Milde (1867), is dramatically clear 
if Figs. 15, 16 above are compared with fig. 140G in Smith (1938). Duval-Jouve 
(1864) analogized the basal sheath to the bud scales of seed plants. Milde (1867) 
compared it to the primary sheath of the embryo (the closet thing to cotyledons 
in this non-seed plant) and to the first leaf of a monocot branch. Both of these 
comparisons suggest the possibility that the ochreole should be considered a 
prophyll. 

The ochreole does appear to fit the concept of prophyll, though that concept 
has never, to my knowledge, been applied to the ochreole previously. Goebel 
(1905) and Arber (1925) described the prophyll as the first leaf produced by a 
lateral branch, which usually differs from subsequent leaves on that branch. 
Normally, there are two prophylls in dicots and one in monocots. According to 
Goebel, they function to protect the bud. Guédés (1979) compared prophylls to 
cotyledons. “They, as it were, are the cotyledons to the branch” (p. 16) and ‘“The 
main shoot first bears the cotyledons. In the same way, axillary shoots first form 
their prophylls. Monocotyledons have one prophyll only. Dicots have two...” 


R. L. HAUKE: EQUISETUM OCHREOLE 121 


Fics. 12-15. Longitudinal sections of young branches. Figures 12 and 13. Equisetum arvense. 12. 
Branch initial cell (arrow). 13. Branch apical dome with pyramidal apical cell (arrow). Figures 14 and 
15. Equisetum diffusum. 14. Apex initiating ochreole (arrow). 15. Apex with ochreole and initiating 
root (arrow). 


122 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 4 (1987) 


—— 40pm 


f 


R. L. HAUKE: EQUISETUM OCHREOLE 123 


below them, so that they are at the very base of the branch 

The ochreole of Equisetum represents the first whorl of leaves on a branch, 
which differs from the others in being reduced and lacking a subjacent internode 
It resembles the first leaves formed by the embryo, in a sense the cotyledons. All 
aspects of the prophyll, as seen in seed plants, are present in the ochreole of 
Equisetum, so I interpret it as a prophyllar sheath. 


(p. 127). Guédés described some plants as having prophylls without internodes 
(p. 64). 


LITERATURE CITED 
ARBER, A. 1925. rhe gate Bates Cambridge: University Press. 
BieRHorsT, D. W. rphology of vascular plants. New York: The Macmillan Co. 
DEBLOCK, L. 1923. pelea a l’étude des Equisétacées. Thesis. Faculty of Medicine & Pharmacy, 


ee ee J. 1864. Histoire naturelle des eso de France. Paris: Bailliere et Fils. 


EamMEs, A. J. 1936. pomeonesiac of vascular plants. Lower groups. New York: McGraw Hill. 
Foster, A. S. and E FFORD. 1974. Comparative Sank of vascular plants. 2nd ed. San 
Francisco: ae 


GoeBeL, K. 1905. Organography of plants especially of the Archegoniatae and emer’: i. 
pecial organography. Translated by J. B. Balfour. Oxfo rd: Clarendon Press 

GuEbDEs, M. 1979. Morphology - seed plants. Vaduz: J. Cramer. 

JANCZEWSKI, er A. 1876. le développ t des bourg 1 les Préles. Mém Soc 
Sci t. Cherbourg 20:69-108. 

JOHNSON, M. * 1937. Hydathodes in the genus Equisetum. Bot. Gaz. (Crawfordsville) 98:598-608. 

Mipe, J. 1867. Monographia Equisetorum. Nov. Actorum Acad. Caes. Leop. -Carol. German. Nat. 
Cur. 32(2). 


SMITH, G. M. 1938. Cryptogamic botany. Vol. II. Bryophytes and pteridophytes. New York: McGraw 


SPORNE, K. . 1962. The morphology of pteridophytes. London: Hutchinson and Co. 
VAUCHER, J. P. 1821. Monographie des Préles. Mém. Soc. Phys. Genéve. 1:329-391 


~— 

Fics. 16-20. Equisetum diffusum branches with ochreoles. Figures 16-18. Longitudinal sections. 16. 
Apex with ochreole and one nodal sheath. 17. Apex with ochreole and two nodal sheaths with 
intercalary meristems (arrow). 18. Branch erupting through stem sheath, with basal collar (b), ochreole 
(o) and five nodal sheaths with intercalary meristems ems. Figures 19 and 20. Cross-sections. 19. B Branch 
with ochreole adjacent to stem with sheath. 20. Abaxial segment of ochreole with vascular strand 
{arrow}. 


American Fern Journal 77(4):124-130 (1987) 


New Combinations in Megalastrum 
(Dryopteridaceae) 


ALAN R. SMITH 
Department of Botany, University of California, Berkeley, California 94720 
ROBBIN C. MORAN 
Missouri Botanical Garden, P.O. Box 299, St. Louis, Missouri 63166 


One of Carl Christensen’s many important contributions to fern taxonomy was 
demonstrating that Dryopteris, as defined around the turn of the century, was a 
vast, unrelated assemblage of ferns (Christensen, 1913, 1920). He indicated this 
diversity by dividing Dryopteris into 11 subgenera, using characters such as 
pinnule arrangement, hair type, and scale type that were not widely employed 
or appreciated by his predecessors and contemporaries. Ever since his work on 
Dryopteris, the trend has been to raise Christensen’s subgenera to the generic 
rank or to treat them as subgenera in genera other than Dryopteris. In some 
cases, even his informal species groups have received generic status. This paper 
deals with one such grouping: the D. subincisa group of Dryopteris subg. Ctenitis. 

Nowadays, all pteridologists recognize Christensen’s subgenus Ctenitis as a 
distinct genus, Ctenitis. Within Ctenitis s.]., one of Christensen’s informal groups, 
comprising the species allied to Ctenitis protensa, was recently established as 
the genus Triplophyllum by Holttum (1986a). He provided evidence that Triplo- 
phyllum is more closely allied to Tectaria than to Ctenitis. Another informal 
group, that of Dryopteris subincisa, was made a section of Ctenitis by Tindale 
(1965) and a genus by Holttum (1986b), as Megalastrum. We agree with Holttum 
that Megalastrum deserves generic status, as it is phenetically and cladistically 
distinct from Ctenitis. Furthermore, we are not aware of any species that is 
ambiguous with regard to placement in either Megalastrum or Ctenitis. 

The best characteristic to distinguish Megalastrum from Ctenitis is the type of 
hair present on the adaxial surface of the axes. In Megalastrum, these hairs are 
coarse, whitish, multiseptate, with pointed tips, and antrorsely strigose or spread- 
ing (Fig. 1D). When dried, these hairs either remain terete or, if the cell walls 
collapse laterally, they all collapse in the same plane. In Ctenitis (and all other 
genera of tectarioid ferns), hairs of the above type are lacking. Rather, Ctenitis- 
hairs are present, i.e., hairs that are fine, usually blunt, reddish, erect to spreading 
(never antrorsely strigose), which when dried have cells that often collapse in 
planes perpendicular to the adjacent cells thereby imparting a catenate appear- 
ance to the hair, i.e., adjacent cells appear twisted relative to each other (Fig. 
1C). Megalastrum may have Ctenitis-hairs on the margins and/or abaxial surface 
: the lamina, but never on the adaxial surface of the axes, as in all other tectarioid 
erns. 

Another important difference in indument is that Ctenitis always has unicel- 
lular, cylindrical, glandular hairs on the stalks of the sporangia (Holttum, 1986b). 
These glandular hairs may also occur on the margins of the indusia or on the 


SMITH & MORAN: MEGALASTRUM 125 


f Ctenitis sl i(P S gel) C. Morton and Megalastrum acrosorum. 


Fic.1. Compari (Poeppig preng 

A, distal half of a proximal pinna, C. sloanei. B, Distal half of a proximal pinna, M. acrosorum, note 
position of the basal basiscopic lobes on the costa. C, typical Ctenitis-hairs from adaxial surface of 
the costa, C. sloanei. D, hairs from the adaxial surface of the costa, M. acrosorum. E, pinnule bases 
of C. sloanei; note the basiscopic veins arising from the costules and vein tips reaching the margin 
(pinna apex to the right). F, pinnule bases of M. acrosorum; note the basal basiscopic veins arising 
from the costa and clavate vein tips (pinna apex to the right). A, C, E: Killip 2678 (MO). B, D, F 


Moran 3354 (MO). 


126 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 4 (1987) 


abaxial surface of the lamina appressed to the veins. Megalastrum however, 
lacks such glandular hairs. 

Megalastrum further differs from Ctenitis by a subtle characteristic of the 
venation and cutting. In Ctenitis, the basal basiscopic vein on the distal pinnules 
arises from the costule (Fig. 1E); however, in Megalastrum this vein arises from 
the costa (Fig. 1F). Concomitantly, in Megalastrum the basal basiscopic lobe of 
the distal pinnules appears broadly adnate to the costa (Figs. 1B, F). In Ctenitis, 
the basal basiscopic lobe may be decurrent on or only partially adnate to the 
costa, but never broadly adnate (Figs. 1A, E). 

The veins in Megalastrum end behind the margin in conspicuous clavate tips 
(these are best seen on the adaxial surface). The veins in Ctenitis extend to the 
margin or may end behind the margin, but they never terminate in conspicuous 
clavate tips (cf. Figs. 1E, F). 

Ithough more samples need to be examined, evidence from spore morphology 
supports the distinction between Ctenitis and Megalastrum. Ctenitis spores vary 
from rugose to coarsely echinate with few, stout spines (Tryon & Tryon, 1982). 
Two kinds of spores occur in Megalastrum: finely echinate (spinulose) and cris- 
tate. The echinate type, present in M. inaequalifolium, M. lanuginosum (from 
Africa), and M. villosum, is the closest to Ctenitis of the two types, but this type 
differs from Ctenitis by having much more numerous, narrower, and shorter 
spines. Megalastrum lanuginosum and M. villosum have filiform connecting 
strands between the bases of the spines (Holttum, 1986b, plate 3), an apparently 
unique feature among tectarioid ferns. The cristate spore type, present in M. 
kallooi (Jermy & Walker, 1985, fig. 11) and M. pulverulentum (Tryon & Tryon, 
1982, fig. 67.24), has long, low, narrow, erose-margined, semiparallel wings. This 
spore type may be unique among tectarioid ferns, and could be formed by the 
coalescence of individual spines to form erose wings (Tryon & Tryon, 1982). 
Although distinct in their extreme forms, the echinate and cristate spore types 
are not fundamentally different, as evidenced by the intermediate spores of M. 
connexum, which have short, spinose wings connected at the base by filiform 
strands (Tryon & Tryon, 1982, fig. 67.26). 

Although the foregoing character differences argue for distinguishing Mega- 
lastrum from Ctenitis solely on the basis of phenetic distance, there are also 
cladistic reasons for doing so. Lastreopsis is more closely related to Ctenitis than 
is Megalastrum, because Lastreopsis has Ctenitis-hairs on the adaxial surface of 
the axes and has unicellular, cylindrical, glandular hairs on the sporangial stalks 
and other parts of the plant (Tindale, 1965). Thus, since all present-day pteri- 
dologists recognize Latreopsis, it would be inconsistent to subsume a more distant 
group, Megalastrum, in Ctenitis. Given the phenetic and cladistic arguments in 
favor of recognizing Megalastrum as distinct, we make the following combina- 
tions for the neotropical taxa. No doubt there are also a number of undescribed 
species, especially in South America. These are best dealt within a more critical 
revision of Megalastrum. Ranges are given in general terms, mostly following 
Christensen (1913, 1920). 


Megalastrum Holttum, Gard. Bull. Straits Settlem. 39:161. 1986.—TyPE: Mega- 
lastrum villosum (L.) Holttum. 


_ 1p 00¥ 


SMITH & MORAN: MEGALASTRUM 127 


“425 Megalastrum abundans (Rosenstock) A. R. Smith & R. C. Moran, comb. nov.— 
Dryopteris abundans Rosenstock, Hedwigia 46:133. 1906.—Ctenitis abundans 2 yo4d 
(Rosenstock) Copel. Brazil. 

419* Megalastrum acrosorum (Hieron.) A. R. Smith & R. C. Moran, comb. nov.— 

43% Nephrodium acrosorum Hieron., Bot. Jahrb. Syst. 34:446. 1904.—Ctenitis | 390 
: acrosora (Hieron.) Copel. Costa Rica to Colombia. 

1¢%* Megalastrum adenopteris (C. Chr.) A. R. Smith & R. C. Moran, comb. nov.— 
Dryopteris adenopteris C. Chr., Kongel. Danske Vidensk. Selsk. Skr., Na- 
turvidensk Afd., ser. 8. 6:85. 1920.—Ctenitis adenopteris (C. Chr.) Ching. 2°?" 
Southern Brazil, N. Argentina. 

qa Megalastrum andicola (C. Chr.) A. R. Smith & R. C. Moran, comb. nov.—Dryop- 

943" teris andicola (C. Chr., Kongel. Danske Vidensk. Selsk. Skr., Naturvidensk 
s Afd., ser. 8. 6:88. 1920.—Ctenitis andicola (C. Chr.) Ching. Andes. - '$/// 
995° Megalastrum aripense (C. Chr. & Maxon) A. R. Smith & R. C. Moran, comb. 
nov.—Dryopteris aripensis C. Chr. & Maxon, J. Wash. Acad. Sci. 14:143. geese 
1924.—Ctenitis aripensis (C. Chr. & Maxon) Lellinger. Trinidad. 735 ‘ 

995 | Megalastrum atrogriseum (C. Chr.) A. R. Smith & R. C. Moran, comb. nov.— 
Dryopteris atrogrisea C. Chr., Kongel. Danske Vidensk. Selsk. Skr., Natur- 777 7 
vidensk Afd., ser. 8. 6:70, fig. 15. 1920.—Ctenitis atrogrisea (C. Chr.) Ching. ‘“° 9 
Central America. 

945 = Megalastrum bidecoratum (Lellinger) A. R. Smith & R.C. Moran, comb. nov.— 
Ctenitis bidecorata Lellinger, Proc. Biol. Soc. Wash. 98:373. 1985. Costa Rica. 774° 

4153 Megalastrum biseriale (Baker) A. R. Smith & R. C. Moran, comb. nov.—Poly- 744) 
podium biseriale Baker, Syn. Fil. 309. 41867.—Ctenitis biserialis (Baker) Lel- 
linger. Panama, Andes. 

‘454 Megalastrum canescens (Kunze ex Mett.) A. R. Smith & R. C. Moran, comb. 
nov.—Phegopteris canescens Kunze ex Mett., Abh. Senckenberg. Naturf. 1% 42 
Ges. 2:314. 1858.—Ctenitis canescens (Kunze ex Mett.) C. Morton. Brazil. 242*Y 
445 S Megalastrum connexum (Kaulf.) A. R. Smith & R. C. Moran, comb. nov.—Poly- 
4}59 podium connexum Kaulf., Enum. Fil. 120. 1824.—Ctenitis connexa (Kaulf.) 7757 
Copel. Southern Brazil to Paraguay. 
9966 var. lateadnatum (Christ) A. R. Smith & R. C. Moran, comb. nov.—Phegopteris 
lateadnata Christ, Annuaire Conserv. Jard. Bot. Geneve 3:36. 1899. Paraguay. 
795+ Megalastrum crenulans (Fée) A. R. Smith & R. C. Moran, comb. nov.—Aspidium mY 
crenulans Fée, Crypt. Vasc. Brés. 1:139, tab. 47, fig. 1. 1869.—Ctenitis cren- ' 700 
: ulans (Fée) Ching. Southern Brazil, Paraguay, Venezuela. oe 
445¢ Megalastrum eugenii (Brade) A. R. Smith & R. C. Moran, comb. nov.— Dryopter _ 
eugenii Brade, Rodriguésia 13:298, tab. 2. 1940.—Ctenitis eugenii (Brade) 2ySy6 
Brade. Southern Brazil. 
4989 Megalstrum grande (C. Presl) A. R. Smith & R. C. Moran, comb. nov.—Poly- é 
podium grande C. Presl, Del. Prag. 1:171. 1822.—Ctenitis grandis C. Presl) 2 44 
Copel. Southern Brazil.—Holttum (1986) excluded this species from Mega- 
lastrum because it lacks hairs on the costae adaxially, but - all other r —— 
it belongs here: clavate vein tips, venation and cutting, spinulose spores with 
a tendency for the spines to coalesce into subparallel ridges, and toothed 
scales, Holttum’s observation that the scales of Ctenitis grandis are funda- 


94943 


34 Ye 
Bigs 


S 


128 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 4 (1987) 


mentally different from those of Megalastrum in that the teeth are the out- 
growths of a single cell (vs. two or more cells in Megalastrum) is incorrect; 
many of the teeth in M. grande are formed at the base from parts of two 
adjacent cells. Certainly, M. grande is more closely related to other species 
of Megalastrum than to any other tectarioid genus, either Old World or New 


World. 
442 Megalastrum haitiense (Brause) A. R. Smith & R. C. Moran, comb. nov.—Dryop- ° 


teris subincisa (Willd.) Urban var. haitiensis Brause, Ark. Bot. 17(7):67. 1922.— 
Ctenitis haitiensis (Brause) Lellinger. Haiti. * © 
- Megalastrum hirsutosetosum (Hieron.) A. R. Smith & R. C. Moran, comb. nov.— 
Dryopteris hirsutosetosa Hieron., Hedwigia 46:343, tab. 6, fig. 16. 1907. Andes. 
A8T> Mogalas trum honestum (Kunze) A. R. Smith & R. C. Moran, comb. nov.—Poly- 
-' podium honestum Kunze, Linnaea 9:49. 1834.—Ctenitis honesta (Kunze) R. 
& A. Tryon. Andes. 
1974 Megalastrum inaequale (Kaulf. ex Link) A. R. Smith & R. C. Moran, comb. nov.— 
Polypodium inaequale Kaulf. ex Link, Hort. Berol. 2:107. 1833.—Ctenitis 
_ inaequale (Kaulf. ex Link) Copel. Brazil. 
“\/7> Megalastrum inaequalifolium (Colla) A. R. Smith & R. C. Moran, comb. nov.— 
74°” Polypodium inaequalifolium Colla, Mem. Reale Accad. Sci. Torino 36:49. 
1836.—Ctenitis inaequalifolia (Colla) Ching. Juan Fernandez. 5‘)! 

40) Megalastrum kallooi (Jermy & T. Walker) A. R. Smith & R. C. Moran, comb. 
nov.—Ctenitis kallooi Jermy & T. Walker, Bull. Brit. Mus. (Nat. Hist.), Bot. 
13:267, fig. 11. 1985. Trinidad. 

» Megalastrum lasiernos (Sprengel) A. R. Smith & R. C. Moran, comb. n 

94¢5 Polypodium lasiernos Sprengel, Syst. Veg. 4:61. 1827.—Ctenitis thee 

o (Sprengel) Copel. Brazil. 

497° Megalastrum leptosorum (C. Chr.) A. R. Smith & R. C. Moran, comb. nov.— 
Dryopteris leptosora C. Chr., Ind. Fil. 274. 1905. Andes. 9964 


44 Megalastrum lunense (Christ) A. R. Smith & R. C. Moran, comb. nov.—Aspidium “ 
lunense Christ, Bull. Herb. Boissier, sér. 2. 6:55. 1906.—Ctenitis lunensis | 


(Christ) Lellinger. Costa Rica, Andes? 


9944 


(920 Megalastrum macrotheca (Fée) A. R. Smith & R. C. Moran, comb. nov.—Phe- 74 is 
gopteris macrotheca Fée, Mém. Fam. Foug. 11:56. 1866.—Ctenitis macro- *""** 


theca (Fée) Ching. Lesser Antilles. 
949° Megalastrum mollicomum (C. Chr.) A. R. Smith & R. C. Moran, comb. nov.— 


Dryopteris mollicoma C. Chr., Kongel. Danske Vidensk. Selsk. Skr., Na- q 
turvidensk. Afd., ser. 8. 6:75. 1920.—Ctenitis mollicoma (C. Chr.) Ching. > 


Andes. 

“92. Megalastrum palmense (Rosenstock) A. R. Smith & R. C. Moran, comb. nov.— 
Dryopteris subincisa var. palmensis Rosenstock, Feddes Repert. Spec. Nov. 
Regni Veg. 22:11. 1925.—Ctenitis palmensis (Rosenstock) Lellinger. Costa 
Rica, Panama. 

age? Megalastrum pansamalense (C. Chr.) A. R. Smith & R. C. Moran, comb. nov.— 

‘720 Dryopteris pansamalensis C. Chr., Kongel. Danske Vidensk. Selsk. Skr., ? 

Naturvide 


4970 
3121 


720 


nsk. Afd., ser. 8. 6:72. 1920. Scns pansamalensis (C. Chr.) Lel- =} | 3-\ 


aa linger. Southern Mexico to Costa Ric 
“* Megalastrum platylobum (Baker) A. R. ‘Eeatth & R. C. Moran, comb. nov.— 


SMITH & MORAN: MEGALASTRUM 129 


?' Polypodium platylobum Baker, Syn. Fil. 307. 1867.—Ctenitis platyloba (Ba- 
oo. ker) C. Morton. Andes, Venezuela. 

1°"'* Megalastrum pleiosoros (Hook. f.) A. R. Smith & R. C. Moran, comb. nov.— 
45 Polypodium pleiosoros Hook. f., Trans. Linn. Soc. London, ser. 2. 20:166. 

1847.—Ctenitis pleiosoros (Hook. f.) C. Morton. Galapagos. 
1220 Megalastrum pulverulentum (Poiret in Lam.) A. R. Smith & R. C. Moran, comb. 
2/22 nov.—Polypodium pulverulentum Poiret in Lam., Encycl. Meth. 5:555. 1804.— 
Ctenitis pulverulenta (Poiret) Copel. Greater Antilles, S. Mexico to N. Ar- 


gentina. 
494% var. heydei (C. Chr.) A. R. Smith & R. C. Moran, comb. nov.—Dryopteris 
;42- karsteniana var. heydei C. Chr., Kongel. Danske Vidensk. Selsk. Skr., Na- 
turvidensk. Afd., ser. 8. 6:77. 1920.—Ctenitis pulverulenta var. heydei (C. 
Chr.) Stolze. Guatemala. 
| galast kutchii (Lellinger) A. R. Smith & R. C. Moran, comb. nov.—Ctenitis 
9926 skutchii Lellinger, Proc. Biol. Soc. Wash. 98:375. 1985. Costa Rica, Panama. 
(002° Megalastrum spectabile (Kaulf.) A. R. Smith & R. C. Moran, comb. nov.—Poly- 
437 podium spectabile Kaulf., Enum. Fil. 121. 1824.—Ctenitis spectabilis (Kaulf.} 
Kunkel. Chile. 

2/25! yar. phillippianum (C. Chr.) A. R. Smith & R. C. Moran, comb. nov.— Dryopteris 
spectabilis var. phillippiana C. Chr., Kongel. Danske Vidensk. Selsk. Skr., 
Naturvidensk Afd., ser. 8. 6:70. 1920.—Ctenitis spectabilis var. phillippiana $4 b 
(C. Chr.) Rodriguez. Chile. 
galast 1 issimum (Sodiro) A. R. Smith & R. C. Moran, comb. nov.— 

39@® Nephrodium squamosissimum Sodiro, Anal. Univ. Quito 10(66):12. 1894. : 
[Crypt. Vasc. Quit. 256. 1893.]—Ctenitis sq issima (Sodiro) Copel. Andes. 2 ya5P 

24.955 var. bogotense (Hieron.) A. R. Smith & R. C. Moran, comb. nov.—Dryopteris 

_ _ 4984 subincisa var. bogotensis Hieron. Hedwigia 46:349. 1907. 

/°¢C2 NMegalastrum subincisum (Willd.) A. R. Smith & R. C. Moran, comb. nov-— | 
#2@ Polypodium subincisum Willd., L. Sp. Pl. ed. 4, 5:202. 1810.—Ctenitis sub- © ~~ 
incisa (Willd.) Ching. Antilles, Mexico to Bolivia. 

'0° > Megalastrum umbrinum (C. Chr.) A. R. Smith & R. C. Moran, comb. nov.— 

999° Dryopteris umbrina C. Chr., Kongel. Danske Vidensk. Selsk. Skr., Natur- 
vidensk Afd., ser. 8. 6:81. 1920.—Ctenitis umbrina (C. Chr.) Ching. Southern ~ 85 P 
Brazil, Paraguay. _ 
/ voy Megalastrum vastum (Kunze) A. R. Smith & R. C. Moran, comb. nov.—Poly- 
7474| podium vastum Kunze, Linnaea 9:50. 1834. Costa Rica to Bolivia. 
(0°05 Megalastrum villosulum (C. Chr.) A. R. Smith & R. CG. Moran, comb. nov.— 
4992. Dryopteris villosula C. Chr., Kongel. Danske Vidensk. Selsk. Skr., Natur- 7, 
vidensk Afd., ser. 8. 6:89. 4920.—Ctenitis villosula (C. Chr.) Ching. Bolivia. 2¥*° 
000(¢Megalastrum villosum (L.) Holttum, Gard. Bull. Straits Settlem. 39:161. 1986. 
4993 Polypodium villosum L., Sp. Pl. 1093. 1753.—Ctenitis villosa (L.) Copel. Great- °° / 


er Antilles. : b 
'0°°7 Megalastrum wacketii (Rosenstock ex C. Chr.) A. R. Smith & R. C. Moran, comb. 
49?4 nov.—Dryopteris wacketii Rosenstock ex C. Chr., Kongel. Danske Vidensk. 
3260 Selsk. Skr., Naturvidensk Afd., ser. 8. 6:64. 4920.—Ctenitis wacketii (Rosen- 
stock ex C. Chr.) Ching. Southern Brazil. 

\0% 8 Megalastrum "iain (Christ & Rosenstock) A. R. Smith & R. C. Moran, comb. 


42a 
A 
es oe a es 


[O06] ng 


130 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 4 (1987) 


as nov.—Dryopteris yungensis Christ & Rosenstock, Feddes Repert. Spec. Nov. 
Regni Veg. 5:234. 1908. Bolivia. 


LITERATURE CITED 


CHRISTENSEN, C. 1913. A monograph of the genus Dryopteris, part I. The tropical American pin- 
natifid-bipinnatifid species. Kongel. Danske Vidensk. Selsk. Skr., Naturvidensk Afd., ser. 7. 
10:55-282. 

. 1920. A monograph of the genus Dryopteris, part II. The tropical American bipinnate- 
Pee aria species. Kongel. Danske Vidensk. Selsk. Skr., Naturvidensk Afd., ser. 8. 6:3- 


Taare E. 1986a. Studies in the fern-genera oy se Tectaria Cav. V. Triplophyllum, a new 
genus of Africa and America. Kew Bull. 41:2 
1986b. Studies in the fern-genera allied to Sore Cav. VI. A conspectus of genera in 
the Old World omeirphen as related to Tectaria, with descriptions of two genera. Gard. Bull. 
traits Settlem. 39:153-1 


JERMY, A. C. and T. G. WALKER. pl Cytotaxonomic studies of the ferns of Trinidad. 3. a 
new species and hybrids and a new combination. Bull. Brit. Mus. (Nat. Hist.), Bot. 1 
a 276. 


TiNDALE, M. D. 1965. A monograph of the genus Lastreopsis Ching. Contr. New South Wales Natl. 
Herb. 3:249-339, +23 plates. 

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


Statement of Ownership, Management, and Circulation 


Publication title and number: American Fern Journal (0002-8444). Date of filing: 
9 March 1988. Frequency of issue: quarterly. Annual subscription price: $20.00. 
Office of publisher: Pringle Herbarium, Dept. of Botany, University of Vermont, 
Burlington, VT 05405-0086 USA. Editor: Alan R. Smith. The American Fern 
Journal is wholly owned by the American Fern Society with no bond holders. 
urpose, function, and nonprofit status of the Society and its tax exempt 
status for Federal income tax purposes remains the same as in past years. The 
press run for each issue is 1300, of which (just prior to the filing date) 1110 copies 
were mailed as paid circulation, and 15 were mailed as free distribution, leaving 
175 copies for office use and back-issues sales. I certify that these statements are 
correct and complete.—Davip S. BARRINGTON, Records Treasurer of AFS 


American Fern Journal 77(4):131-136 (1987) 


Variations in Petiolar Structure of 
Hypodematium crenatum 


P. K. KHARE and RAMA SHANKAR 
Department of Botany, University of Allahabad, Allahabad, India 


The importance of leaf traces in fern taxonomy has been discussed in detail 
by Tansley (1907, 1908), Sinnott (1911), Davie (1918), Ogura (1972), and others. 
Ching (1936) differentiated between Thelypteris and Dryopteris on the basis of 
petiolar characters. Bower (1914) used these characters along with others in es- 
tablishing the relationships between Blechnum and allied genera. Sen (1964), 
Keating (1968), and Lucansky (1974a, 1974b) also used petiolar characters in 
working with dennstaedtioid, cyatheoid, dicksonioid, and pteroid ferns as did Bir 
(1962) and Kato (1972, 1975) in differentiating members of Athyriaceae. Recently, 
Lin and DeVol (1977, 1978) have given a key, based entirely on petiolar characters, 
to 170 species of Taiwan ferns and tried to establish relationships between various 
families of extant ferns. The above literature and many other references show 
that petiolar characters, particularly vascular pattern, are relatively constant with- 
in a species. In order to test constancy in Indian ferns, we have tried to study 
materials of the same taxon collected from diverse localities (see Khare, 1984; 
Khare & Shanker, 1984, 1986). In the process we noted that Hypodematium 
crenatum (Forssk.} Kuhn (Aspidiaceae of Copeland, 1947) showed considerable 
variation among the petioles of plants collected from different sites. A detailed 
study of their petioles was then begun. 


MATERIALS AND METHODS 


Plants of H. crenatum were collected from Pachmarhi (Draupadi Dweep, Han- 
dikho, and Kalakund) in Madhya Pradesh province in Central India and Almora 
in Uttar Pradesh province in Western Himalayas. These sites are about 1200 
kilometers apart at an elevation of about 1000 meters. Plants from both places 
grow along road sides in shade in moist soil. Petioles were serially cut along their 
length and fixed in FAA. Hand or microtome sections were cut at transverse, 
longitudinal, and paradermal planes at 8-10 um thickness and stained with saf- 
ranin-fast green. At least ten samples from each locality were examined. Vouchers 
are deposited in the Herbarium of Botany Department of Allahabad University. 
Clearings of petioles were made by treating them in ten percent aqueous sodium 
hydroxide solution and chloral hydrate and subsequently staining them with 
safranin. To confirm the presence of lignin, cutin, tannin, starch grains, and 
phlobaphenes special microchemical and histochemical tests were conducted, 
e.g., lignin and cutin were tested by obtaining red color with phloroglucinol and 
sudan IV, tannin and starch grains by blue color with ferric chloride and iodine 
solution, respectively (see Johansen, 1940). Phlobaphene was detected by its nat- 
ural yellowish brown color as suggested by Reeve (1951). 


132 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 4 (1987) 


RESULTS 


Petioles of Hypodematium crenatum collected from both sites are yellowish 
green. Petiole bases are covered with large yellowish brown linear-lanceolate 
scales made up of uniformly thick-walled cells (Fig. 1J). Multicellular, uniseriate, 
elongate hairs with tapering ends occur over the entire petiole in plants of Pach- 
marhi (Fig. 1K). However, hair density increases gradually towards the petiole 
apex. Petioles of plants from Almora are glabrous. Petioles from both localities 
have a very shallow adaxial groove that becomes obscure distally; for most of 
their length petioles appear terete. In surface view the epidermal cells of a petiole 
appear narrow, elongate, and polygonal, with thick anticlinal walls and, occa- 
sionally, simple circular pits on their outer surface (Fig. 11). Usually petioles are 
devoid of stomata, although plants collected from Almora sometimes show a few 
anomocytic stomata surrounded by 5-7 epidermal cells on both sides of the groove 
(Fig. 2]). 

In transverse section petioles of plants from both sites are bounded by epi- 
dermal cells that appear small, thin-walled, and covered by a thin cuticle. The 
bulk of the petiole is composed of ground tissue consisting of a thick-walled 
lignified outer zone (3-4 layers) and several layers of thin-walled parenchymatous 
inner zone. Cells of the latter zone are usually filled with starch grains (Fig. 21). 
Petioles of plants collected from both localities show two vascular strands at the 
base that are circular in plants from Pachmarhi (Figs. 1B, C, G) and slightly 
elongate in those from Almora (Figs. 2B, C, G). Each vascular strand is composed 
of a pericycle (2-3 layers) of thin-walled parenchyma cells and several layers of 
phloem surrounding the centrally placed xylem. The xylem is hooked in plants 
of Pachmarhi (Figs. 1B, C, G) but ribbon-shaped in those of Almora. The vascular 
strands are surrounded by a single-layered endodermis whose cells are usually 
filled with a dark phenolic compound phlobaphene. The two strands in the plants 
from Almora remain separate up to about the middle of the petiole (Fig. 2A). 
However, the bundles gradually become oval and the xylem become hippocam- 
pus-shaped (Figs. 2C, G). Both strands then gradually fuse with each other and 
form a single vascular strand for the remaining length of the petiole (Fig. 2A). 
During the fusion, first the endodermis (Fig. 2D) and at a slightly higher level the 
pericycle (Fig. 2E) and ultimately the phloem and xylem bundles of the two 
strands fuse, forming an almost W-shaped xylem strand with its distal arm turned 
inwards and surrounded completely by phloem cells. The vascular strand at this 
stage becomes somewhat U-shaped (Figs. 2F, H). In plants of Pachmarhi the two 
strands at the base fuse in the swollen basal part of the petiole (Figs. 1A, D, E), 
and thus there is a single strand for almost the entire petiole. This strand appears 


= 


Fic. 1. Hypodematium crenatum from Pachmarhi. A, diagrammatic representation of course of 
vascular strand in the petiole. B, diagrammatic transection of petiole base showing two separate 
vascular strands. C-F, st ges showi gf : f the two 1 a Ses oo fp stints 
S he structural details of C and F respectively. I, epidermis of petiole showing pits. J, scale from 
petiole. K, hair from petiole. In C-F, line indicates endodermis (end), larger dots pericycle (per), finer 
dots phloem (phl), and hatched areas xylem (xyl). 


KHARE & SHANKER: HYPODEMATIUM PETIOLE 


ae, enue: 
> 


C/ 


a a TN 


— oy 
DG 


— 


. a 
v5) 


134 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 4 (1987) 


ce TEES 
oh 


F 
¥ 
Ba 
: 
‘ 
i 


Oy ai 


0 aq 


I> 9 
28 


6: 
Re 
ats 
Or 
OTR y 
FO 
Se 


py 
AY 

we 
( 


— 


J 
ET Lo 
“are; 
* en8! 
> 


+ 


Taw 
~ae 
4,5 << 
2 


wer iee 
%.e 
rjaae? 


Lo ares 


KHARE & SHANKER: HYPODEMATIUM PETIOLE 135 


broadly triangular with V- or Y-shaped xylem surrounded by phloem (Figs. 1F, 
H). The xylem in both materials is mesarch with several protoxylem points located 
near the free arms of xylem strands. Protoxylem elements have annular and 
spiral tracheids and metaxylem cells have scalariform tracheids. Phloem consists 
of sieve cells and parenchyma. Cells of phloem and pericycle are sometimes 
filled with tannin. 


DISCUSSION 


In the past Mehra and Loyal (1956), Nayar and Bajpai (1970), and Loyal (1972) 
have vaguely described the vascular pattern in the petioles of Hypodematium 
crenatum and stated the presence of double strands. These authors have, how- 
ever, neither described the details of vascular strands nor the changes they 
undergo at different levels. Our observations on the petioles of this taxon collected 
from two localities show structural as well as external differences. In other char- 
acters, e.g., rhizome, frond, epidermis, sori, and spores, plants from the two areas 
appear identical. Recently, Vasudeva and Bir (1982) have studied the cytology of 
this species from Pachmarhi and found the chromosome number (n = 41) identical 
to plants from Western Himalayas (Mehra & Loyal, 1956). Present observation 
on H. crenatum thus appears an exception to the view regarding the consistency 
of petiolar characters and represents, as far as we know, the first report of such 
variation within a species. 


ACKNOWLEDGMENTS 


We are grateful to Professor D. D. Pant, former Head of the Botany Department of Allahabad 
University, for valuable suggestions and to Dr. R. D. Dixit, Regional Botanist, Botanical Survey of 


LITERATURE CITED 
Bir, S. S. 1962. Taxonomy of the Indian members of family ‘Aspleniaceae’. Bull. Bot. Surv. India 


4:1-16. 

Bower, F.Q. 1914. Studies in the phylogeny of the Filicales IV. Blechnum and allied genera. Ann. 
Bot. (London) 28:363-431. 

Cuinc, R. C. 1936. A revision of the Chinese and Sikkim-Himalayan Dryopteris with reference to 
some species from neighbouring regions. Bull. Fan Mem. Inst. Biol. 6:237-252. 

CopeLanp, E. B. 1947. Genera filicum. Waltham, Massachusetts: Chronica Botanica. 

Davi, R. C. 1918. A comparative list of fern pinna traces with some notes on the leaf trace in the 
ferns. Ann. Bot. (London) 32: 235-245. 


i 

Fic.2. Hypodemati tum from Almora. A, diagr ic rey tation of f vascular 
strands in the petiole. B, diagrammatic transection of petiole base showing two vascular strands. Cc- 
F, stages showing fusion of the two vascular strands. G, H, structural details of C and H, respectively. 
I, epidermis and ground tissue of petiole in transection showing starch grains. J, petiole epidermis 
showing a stoma. In C-F, line indicates endodermis (end), larger dots pericycle (per) finer dots phloem 
(phl) and hatched areas xylem (xyl). 


136 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 4 (1987) 


JOHANSEN, D. A. 1940. Plant microtechnique. New York: McGraw-Hill. 
Kato, M. 1972. The cigrag structure and its taxonomic significance in the Athyriaceae. Acta 
Phytotax. Geobot. 25:79-91. 
975. On the Ley position of Athyrium mesosorum. Acta Phytotax. Geobot. 27: 


56-60. 

KeatTInc, R. C. 1968. Trends of — in the stipe anatomy of Dennstaedtia and related 
genera. ae? Fern J. 58 

Kuare, P. K. 1984. On the stipe i a ind capillus veneris L. and A. venustum Don. Pp. 21-25 
i in Developmental and wore aspects of plant structure and function. Allahabad. 

: ; tiolar structure of some ferns. Pp. 257-263 in Proc. Vth 

tas Geenienel. Conf., et (1983), Spl. Publ. 

—_—— ia 1986. On the petiole structure of some Adiantum species. Geophytology 16: 


Lin, B. L. and C. E. DEVoL. 1977. The use of stipe characters i in fern taxonomy I, Taiwania 22:91-99. 
197 


nd The use of stipe characters in fern taxonomy II, Taiwania 23:77-95. 
LoyAL, D. s. is Morphology of Hypodematium crenatum (Forsk.) Kuhn. Amer. Fern J. 62:88-92. 
Lucansky, T. W. Comparative studies of the nodal and vascular anatomy in the neotropical 


: parative studies of the nodal and vascular anatomy in the neotropical Cy- 
icone II. Squamate genera. Amer. J. Bot. 61:472-480. 
Menra, P. N. and D. S. Loyat. 1956. Some observations on fio cytology and anatomy of Hypo- 
dematium crenatum (Forsk.) Kuhn. Curr. Sci. 25:3 
NAYAR, se K. he N. Bajpal. Seg A reinvestigation of the ate of Hypodematium crenatum. 
Fern J. 60:107-118. 
OcuRA, mo anil Co lavas a ciently of vegetative organs of the pteridophytes. Berlin-Stuttgart: 
Gebruder, Borntraeger 
REEVE, % M. 1951. bistochenteal tests for polyphenols in plants. Stain Technol. 2 
SEN, U. 1964. Importance of anatomy in phylogeny of tree-ferns and their sete pig Bot. Soc. 
6-39. 
SINNOTT, E. W. 1911. The evolution of the filicinean sai trace. Ann. Bot. (London) 25:167-191. 
TANSLEY, A. G. 1907. Lectures on the evolution of the filicinean lancgennes system. New Phytol. 6:25- 
35, 53-68, 109-120, 135-147, 148-155, 187-203, 219- a 253-2 
1908. Lectures on the evolution of the filicinean sevens ‘eke: New Phytol. 7:1-16, 


VasuDEVA, S. M. and S. S. Bir. 1982. Chromosome numbers and evolutionary status of ferns and 
fern allies of Pachmarhi Hills (central India). Aspects of Plant Sciences 6:119-181. 


American Fern Journal 77(4):137-140 (1987) 


The Identity of Hymenophyllum cristatum 


ROBERT G. STOLZE 
Field Museum of Natural History, Chicago, Illinois 60605 


While examining types for the “Pteridophyta of Peru” I encountered evidence 
of general confusion as to the identity of the Andean Hymenophyllum cristatum 
Hook. & Grev., as well as to its type specimen and early illustrations. The pro- 
tologue calls attention to the principal characters that separate it from related 
species: “It is distinguished no less by the serrated or spinulose crested midrib 
and rachis, than by the large and conspicuous involucres.” The accompanying 
illustration clearly depicts the numerous foliar toothlike processes (lamellae) that 
spring from the abaxial side of veins and costae, perpendicular or oblique to the 
plane of the lamina. Also shown is a subglobose receptacle, quite distinct from 
the usual elongate to turbinate receptacles seen in most related species. Morton 
(1968) based his section Buesia on “... having some accessory wings not in the 
plane of the frond” in addition to the subglobose to clavate receptacles. Within 
the section he included H. cristatum and five other species. Section Buesia in 
turn is included within the subgenus Hymenophyllum which is characterized 
by the essentially glabrous lamina and conspicuously serrate ultimate segments. 

There is a type folder at the British Museum marked “Hymenophyllum cris- 
tatum H. & G.,” with a pencilled question mark preceding the printed “Type 
Specimen.” On the sheet (Fig. 1) are two separate specimens, neither of which 
are H. cristatum, but are instead H. fucoides (Sw.) Sw., a common species from 
the Neotropics. This species is a member of subg. Hymenophyllum, sect. Ptych- 
ophyllum, according to Morton (1968) and is quite distinct from H. cristatum in 
that it lacks any sort of lamellae on the costae or veins and the receptacles and 
indusia are elongate. The upper specimen is labelled Spruce 5422., “H. pedi- 
cellatum Kze.?, in Andibus Ecuadorensibus”; the lower is annotated “Herb. John 
Smith, H. fucoides Sw.” Pinned beside Spruce 5422 is an exact copy of t. 148, 
from the Icones Filicum (Hooker & Greville, 1829). This illustrates H. cristatum, 
complete with enlargements of the subspherical indusium, globose receptacle 
and conspicuous lamellae. Glued beside the lower plant is a label containing 
exact data for the type collection of H. cristatum (Jameson, Mt. Cayambe, etc.). 
One can only assume that the drawing and Jameson label were inadvertently 
affixed to the wrong sheet. At any rate, neither specimen can be the type of the 
latter, label and attached illustration notwithstanding. 

In a loan from Kew I discovered a type folder marked “H. fucoides” that 
contains a sheet (fig. 2) with two different collections, both from Hooker’s her- 
barium. The upper left, marked “H. fucoides, Quito, Jameson,” is not a type, but 
it is properly identified to species, as it has (among other characters) ovate, 
subentire indusia and the veins and costae lack lamellae. The right and lower 
plants, marked ‘‘8(2?)9, Hymenophyllum, Forests of the Andes,” have subspher- 
ical sori, toothed indusia, many conspicuous lamellae on veins, and somewhat 
flexuous rachises (annotated by Morton as H. cristatum). Flanking a leaf of the 


AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 4 (1987) 


138 


Loy? ee ‘4 Piiricg., 
wong hapbenambyy 
éreg 


Wher ~eh ie ae 


mippeeepoacy Signe gs 
Je me Rtg mig Gree ie 


ae 


R. G. STOLZE: HYMENOPHYLLUM CRISTATUM 139 


latter are two detailed pencil sketches, one illustrating the nearly subglobose 
receptacle and deeply toothed indusium, the other depicting the lamellae on the 
veins. Since this is apparently a Jameson specimen from Ecuador, it is very likely 
to be a part of the type collection of H. cristatum, and should be considered an 
isotype. 

The holotype (E), part of the Greville Herbarium, simply contains two leaves 
of H. cristatum, with no mixture of other species. The label, apparently in Jame- 
son’s hand, indicates they were collected on Mt. Cayamhe, east of Quito, at 15,000 
ft., on trunks of trees. Above the determination, someone has lightly pencilled 
‘=H. fucoides,” which is a further indication of the general confusion concerning 
the identities of these two distinct species. 

Another unfortunate product of this confusion has been the drawing (t. 63) 
used to illustrate H. fucoides in Hooker's A century of ferns (1854). Although the 
habit sketch and fig. 1 of this plate are somewhat representative of H. fucoides, 
fig. 2 is questionable, and fig. 3 is clearly inaccurate. Fig. 1 correctly illustrates 
the ovoid shape of the indusium with its denticulate margin and veins devoid of 
lamellae. However, the indusium shown in fig. 2 is nearly spherical, and fig. 3 
shows the indusium to be laciniate and subspherical, and the veins of segments 
bearing a number of conspicuous lamellae. Evidently the artist produced this 
illustration from a mixed collection of the two species, and the solution to this 
riddle may be found on the sheet that contains the probal le isotype of H. cristatum 
from Kew, mentioned above. The detailed sketch alongside the specimen is 
almost exactly reproduced (in mirror image) as Hooker's t. 63, f. 3. So apparently 
this sheet with the mixed collections of H. cristatum and H. fucoides is the very 
one used by the artist to produce the illustration for t. 63 in Hooker's A century 
of ferns. Unfortunately, it was also used to illustrate H. fucoides in Proctor’s 
Ferns of Jamaica (1985), when only the habit sketch and fig. 1 correctly apply. 
Thus, the identities of both species have been confused, as to type collections as 
well as to illustrations. It is not surprising then that some authors have placed 
two very different species in synonymy. 

A key delineating chief differences between the two species, as well as data 
on types, synonyms, and the distribution of H. cristatum are provided below: 


Lamellae lacking; sori elongate, indusia margins entire to dentate (or laciniate 
in var. pedicellatum); receptacle elongate (sometimes exserted); lamina 
usually compact, rarely more than 20 cm long; Neotropics .........-. 

H. fucoides (Sw.)} Sw. 


ee eens eee ee 
pig et ie A rer nel meme aCe Ore insane ae Sie VEY ay OLE Pe Nea ee nS Se Eee ed hah alee estos Hates 


— 


roa £1.f LI bon See ey 


Fics.1and2. Hymenophyllum types. 1, Spurious type speci (BM) J phy 
with copy of original illustration of H. cristatum pi ed alongside upper specimen of H. fucoides, 
and piece of Jameson's original label glued next to lower specimen of H. fucoides. 2, Sheet (K) 
containing probable isotype of Hymenophyllum cristatum and another specimen (upper left) of H. 
fucoides. 


140 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 4 (1987) 


er exserted; lamina greatly elongate, often to 30 cm long; Ecuador (also 
tein) ek nee H. cristatum Hook. & Grev. 


Hymenophyllum cristatum Hook. & Grev., Icon. fil. 2, t. 148. 1829.—Hymeno- 
phyllum fucoides Sw. var. cristatum (Hook. & Grev.) Hieron., Hedwigia 45: 
227. 1906.—Buesia cristata (Hook. & Grev.) Copel., Univ. Calif. Publ. Bot. 19: 
296. 1941.—TypE: Ecuador, Jameson s.n. (holotype E!; probable isotype K’). 
Hymenophyllum sodiroi C. Chr., Index fil. 368. 1905. —Hymenophyllum pen- 
dulum ., Anal. Univ. Quito 6 (46):236. 1892 [seors.: Crypt. Vasc. Quit. 31. 
1897], not Son. 1833.—TypeE: Ecuador, “Crece en los bosques del volcan El 
Corazon”, 3000 m, Sodiro s.n. (not located). 


Hieronymus proposed H. cristatum asa variety of H. fucoides, citing collections 
of Stiibel as examples: 41h, 42, 206B, 236B (Colombia); 812, 845 (Ecuador). He 
claimed that these specimens agreed well with the original drawing of H. cris- 
tatum, calling attention to the more luxuriant leaves (up to 30 cm long), the sharply 
dentate indusia margins, and the thick receptacle. However no mention is made 
of lamellae on the veins or elsewhere. I have not seen the Stiibel specimens, and 
it is not clear from Hieronymus’ discussion whether they represent true H. cris- 
tatum. 

Although I have not located the type of H. sodiroi, the original description 
seems to fit H. cristatum in every way, and I consider the two synonymous. 

Distribution.—Apparently confined to Ecuador, unless the Stiibel collections 
mentioned above add Colombia to the range. 


ie sical —EcCuADOR. Cone ape Espejo, ae. bap eaten ve i (UC). Imbabura: 


L , Cayambe, C 503 Volcan Pichincha, 
Balslev et al. 1735 (UC). Tungurahua: SW asite des lah 3600-3800 m, Satscks 827 (F); 
Paramo of Minza, 3800 m, Penland & Summers 396 (F, US); E slope of Volcan Tungurahua, 3500 m, 


Rimbach 14 aay UC, US); Cordillera Orientalis, Cubillin, 3500 m, Rimbach 16 (UC, US). 


Drs. David Lellinger (US) and Alan Smith (UC) have been most cooperative 
in examining specimens at their respective herbaria, to confirm the identification 
of duplicates of those specimens seen at F and to report additional collections 
not seen by me but cited above. 


LITERATURE CITED 


Hooker, W. J. 1854. A century of ferns. London: William Pamplin. 

and R. K. GREVILLE. 1829-1832. Icones filicum, vol. 2. London 

Morton, C. V. 1968. The genera, subgenera, and sections of the Symenspliytlecose: Contr. U.S. 
Natl. Herb. 38:153-214. 

Proctor, G. R. 1985. Ferns of Jamaica. London: British Museum (Nat. Hist.). 


SHORTER NOTE 


Two Species of Adiantum New to Florida.—Adiantum tetraphyllum Humb. & 
Bonpl. ex Willd. was discovered in Dade County, Florida, by A. Kesler in early 
1987 from a single plant in Costello Hammock (Nauman et al. 1881, FTG). This 
represents the first occurrence of the species for Florida. The plant is growing 
in dense tropical hammock on exposed oolitic limestone. There is no evidence 
nearby (e.g., dumping) to suggest it has escaped, nor is the species cultivated in 
commercial or private nurseries in the immediate area. As a result, this plant 
appears to represent a natural range extension for the species. Outside of Florida, 
the species occurs in the West Indies and tropical America from Mexico south- 
ward to Brazil and Bolivia (Proctor, Ferns of Jamaica, 1985). It can be distin- 
guished from the other Florida species by dull, scaly stipes and rachises and 
sessile or subsessile basal pinnules. 

The Asian A. caudatum L. was found by C. Delchamps in Everglades National 
Park (A. Herndon, pers. comm.) as an escape on exposed limestone; these pop- 
ulations are now extirpated. The species has been rediscovered in the “Whis- 
pering Pines” area of Cutler in Dade County on exposed limestone in an open 
somewhat disturbed pineland (Nauman 1880, FTG). The plants have apparently 
escaped from cultivation nearby and at the time of this writing have disappeared 
from the original collection site. Since this species is common in cultivation, it is 
likely to escape again and be encountered by collectors. It can be distinguished 
from other Florida species by its 1-pinnate leaves and elongate proliferous leaf 
tips. Acknowledgment is given to R. Hammer and A. Herndon for bringing these 
species to my attention.—Clifton E. Nauman, Fairchild Tropical Garden, 10901 
Old Cutler Road, Miami, Florida 33156. 


REVIEW 


“Encyclopaedia of Ferns,” by David L. Jones. 1987. xvii + 433 pp., 251 color 
plates, 154 b/w illustrations, 78 line drawings. Timber Press, 9999 S.W. Wilshire, 
Portland, Oregon 97225. $50.00 ISBN 0 88192 054 1. 

This most attractive book with its many excellent photographs and drawings 
is written primarily for fern growers. However, with the need to maintain live 
plants for hand to save through botanical gardens the rare and endang d 
species, this book should have an even wider appeal. Part one, the Introduction, 
Structure and Botany of Ferns, takes about one fourth of the book. It includes a 
very informative chapter on the economic importance of ferns, with a list of the 
ferns that are eaten in Asia and the Pacific areas. The chapter on structure is 
well illustrated for clarity. Under the discussion of root types, fleshy fern roots 
were not mentioned. Under stipes, the significance of different tissue patterns as 
seen in cross-sections might have been included. Classifications and life cycles 


142 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 4 (1987) 


of pteridophyte orders and some families are discussed in detail. Save for As- 
pidiaceae, families to which commonly known ferns belong are regretfully omit- 
ted. Life cycle details are easily accessible in botany texts, but descriptions of 
common fern families are not; it seems that an opportunity has been missed in 
omitting them. 

The next quarter of the book deals with the culture of ferns on a how-to-do 
basis. General and specialized propagation practices, pest and disease control, 
and hybridization are thoroughly covered, clearly written, and amply illustrated. 
A number of the horticultural materials suggested for use such as marl, brown 
coal, basic slag, and blisters, will be unfamiliar or difficult to obtain in the United 
States, while materials successfully used in the United States such as diazinon 
for fungus gnats and metalaxyl (Subdue) for root rot are not mentioned and 
perhaps not available in Australia. 

The remaining half of the book lists approximately 700 species of ferns, each 
followed by a short text. Some species are arranged by taxonomic relationships, 
some by similarities in growth habit, and some by similarities in cultural require- 
ment. This unusual arrangement necessitates a catch-all category entitled ‘“Mis- 
cellaneous ferns.” The text after each species gives the common name, synonyms, 
country of origin, frond dimensions and division, hardiness, habit, and brief 
comments on cultural needs and uses. Because the descriptions are brief and 
only select ferns are illustrated, identification of unknown ferns will be mainly 
incidental. Some of the drawings could have had more diagnostic value if they 
were less reduced in printing, and some could have been edited with a more 
critical eye, as in places where peltate indusia should be reniform or vice versa. 
Emphasis is given to semi-hardy and warmer-climate species in the photographs 
and illustrations; however, hardier species are covered adequately in the text. 
Some of the species listed as hardy probably would not survive in the colder 
parts of the United States and northern Europe. Though one is tempted to mention 
the genera and species omitted, I am more impressed by the inclusion of so many 
diverse species. The latest name changes for a number of species have been used 
with cross-references to most of the well known synonyms. With such a large 
number of entries it is inevitable some misplacements will appear, such as cul- 
tivars under improper species or names under improper pictures. The book 
concludes with a list of ferns for various landscape purposes, a glossary, bibli- 
ography, and list of fern societies and study groups. 

The problems I have pointed out are mostly small in comparison to the scope 
of this book. The many high quality photographs and the thorough and detailed 
discussion with a hands-on-approach to growing ferns will inspire advanced fern 
growers to pursue their hobby even more avidly, and will serve to introduce 
beginning growers to the fascinating world of ferns.—BARBARA JOE HOSHIZAKI, 
Mildred Mathias Botanical Garden, University of California, Los Angeles, Cal- 
ifornia 90024. 


American Fern Journal 77(4):143-144 (1987) 


Referees, 1987 
I thank the Associate Editors and ae listed below for their valuable assistance in the review 
process. Their evaluations of manuscripts submitted to American Fern Journal have aided authors, 
made my job easier, and contributed fe ee quality of our journal.—ALaNn R. SmIr 


Ed Alverson Jane Kotenko Ralph Seiler 
David S. Barrington Karl U. Kramer Douglas E. Soltis 
L. Earl Bishop Nels Lersten Aura E. Star 
James D. Caponetti Robert M. Lloyd Dennis W. Stevenson 
David S. Cona Terry W. Lucansky Robert G. St 
Michael I. Cousens . Lugardon R. Dale sate 
T. A. Lumpkin R. E. Toia 


Donald R. Farrar 
Ernest M. Gifford 


R. James Hickey Racieauial Ollgaard David H. Wagner 

G. S. Hicks Cathy A. Paris Warren H. Wagner, Jr 
Leslie G. Hickok V. Raghavan I. Watanabe 

Judith Jernstedt Thomas Ranker Dean Whittier 

Suzanne Koptur David Seigler Michael D. Windham 


Index to Volume 77—1987 


Classified entries: botanical names spot names in boldface); nae eA ae pet 


key words from titles; reviews (grouped, list by last 
of authors, followed by titles of articles or eoameg to first authors, are = hate ks in 


Table of Contents, iii-iv. 


. 


Actinostachys germanii, Costa Rica, La Selva pteridophytes, 73 


Adiantum: A. el o A. tetraphyllum, = Ctenitis, 124 
141 Cyathea, 90; C. axillaris, 93; C. barringtonii, 93 
AIBS meeting, call for papers: 1987, 27; 1988, 72 um, 93; C. demissa, 94; C. dombeyi, 
94 ri, 94; C. nanna, 94; C. nigripes 


Allozymes, in Polystichum, 45 
Alsophila, 90 r. brunn 
Arachniodes ochropteroides, 101 sali 94: C. stolzei, 101; C. tryonorum, 94; C. 


hosma, 37; A. chilensis, 39; A. dealbata, 
40: A. delicatula, 40; A. fendleri, 40; A. for- 


40; ess A. pallens, 41; A. 
A. 41; A. pilifera, 41; A. stue- 


beliana, 

Boge La cs Spm 81 

Botrychium 

Branching ae ae 

Cheilanthes: C. gryphus, 112; C. ok 109; C. 

subcordata, 112 

Cheilanthoid ferns, 37, 109 

Chr umbers: Argyrochosma delicatu- 
la, 39; Microgramma squamulosa, 66; Thelyp- 
teris abbiattii, 66; Thelypteris stierii, 66 

Cnemidaria, 90 


ursina, 101; C. venezuelensis, 94 
Cyatheaceae, 90 
Danaea oblanceo 


lata, 33 
Dinelbation, La Selva pteridophytes, 73 
1 Tr rae | + Le | Se =n. DeilAah 


dum in Alabama, 102 
Electrophoresis, in Polysti 

uisetum: ochreole, 115; pm sos teeth, 58 
Extrafloral nectaries, Pteridium. 
Fern Foray: 1987, 27; 1988, 72 
Flavonoids: Platyzoma, 28 
Flora, Finca La Selva, 73 
Gametophytes, Helminthostachys, 95 

Gemmilin, ings, i in a 50 


a, 
Helminthostachys Sank spore germina- 
tion, 95 


144 AMERICAN FERN JOURNAL: VOLUME 77 NUMBER 4 (1987) 


Heenionitis sbcordate, 109 
an a2. 100: 
Pleopeltis, 16; "x Pleopodium, 16; Polypodium, 
16; Polystichum, 42 
Hymenophyllum: H. cristatum, 137; H. fucoides, 
137 


lypodematium cic petiole structure, 131 
Hypolepis repens 
La Selva, Costa mnt pteridophyte flora, 73 
Lastreopsis, 
Lindsaea portoricensis, 70 
Lycopodium: L. x bartleyi, 100; L. lucidulum, 50, 


8; M. palmense, 128; 
M. ense, 128; M. roe 128; M. 
pleiosoros, 129 129; 


mum var. bogotense, 129; Msubincium 129; 
M. umbrinum, 129; M. vastum, 129; M. vil- 
losulum, 129; M. ellneui 129; Ae aarti 


: ense, 129 
Morphology: Equisetum, epidermis of rhizome 
sheath teeth, 58; Equisetum, ochreole, 115; Hy- 
ematium, petiole structure, 131; Lycopo- 
— lucidulum, growth patterns, 50; Pteridi- 
extrafloral nectaries, 1 
Notholena 37; sect. Argyrochosma, 38 
phioglossaceae, tickets tiation spore ger- 
mination, 95 
Pecluma ptilodon var. caespitosa, 101 
Pellaea, 37 
enology, La Selva Oana 82 
Phlebodium: P. aureum, 1 pseudoaureum, 


ag aaa . 
Pleopeltis: erythrolepis, . is pie 21;-F. 
e P,. leucospora, 16; P. mexicana, 
21; P. x melanoneurun, 25; P. polylepis 20; P. 
x 21 
Pi lium: P. bartlettii, 20; P. fallacissi 
20; P. leucosporum, 1 lepis, 17 
Polypodium, 20; P. Ta a var. michaux- 
ianum, 72 


Po lystichopsis, 101 
Polysti 98% P. kruckebergii, 42; P. munitum; 42; 


hybrid, 4 
Pailonaen nudum, 102 


caudatum, 
Pteris orem 
Range extensions: =a Hypolepis repens, 
ichaux- 


chium pinnatum, 68; 
audatum, 141; = estan, 14 
eaciaaee 1987, 1 
Reviews: cues P. J. and T. N. H. Galloway, 
A key to the genera of New Zealand ferns and 
allied plants, 108; Duncan, B. and G. Issac, Ferns 
and allied plants of Victoria, Poneala and 
South Australia, 107; Ferrarini, E., et al., 


genus Pyrrosia (Poly- 


lustrations of pteridophytes of lapatk, Volume 
4, 65; Kurata, S. and T. Nakaiki, eds., Ilustra- 
tions of phciidunhiane of Japan, Volume 5, 108; 
Ravensberg, W. J. and E. Hennipman, The Pyr- 
rosia apackon imaaie ions to Drymoglos- 
r, L. H., Jr. and 

J. G. Bruce, Field ra to paces and other 
Sa elie Georgia, 106 

Schizaea: poeppigiana, 70; pusilla, 64 

Sphaeropteris, 90 


res: ochosma, 38; Helminthostachys, 
germination, 95; Megalastrum, 124; Platyzo- 
ma, 28 


Tectariaceae, 102 
Thelypteris: T. falcata, 70; T. hispidula var. ver- 
sicolor, 72 
Trichomanes: T. ekmanii, 70; T. godmanii, 70; T. 
opterum, T. membranaceum, 71; T. 


Ultrastructure, extrafloral nectaries in Pteridi- 


um, 


are en 
— 5) epa 


INFORMATION FOR AUTHORS 


Authors are encouraged to submit manuscripts pertinent to pteridology for 
publication 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 pre- 
pared may be returned for revision prior to review. Submit manuscripts in trip- 
licate {xerocopies acceptable), including review copies of illustrations. Do not 
send originals of illustrations until they are requested. Use standard 8¥ by 11 
inch paper of good quality, not “erasable” paper. Double space manuscripts 
throughout, including title, authors’ names and addresses, 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. 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 mea- | 
sures (e.g., distance, elevation, weight) unless quoted or cited from another s : 
(e.g., specimen citations). For nomenclatural matter {i.e., synonymy and typifi- 
cation), use one paragraph per basionym (see Regnum Veg. 58:39-40. 1968). Ab- 
breviate titles of serial publications according to Botanico-Periodicum-Huntian- 
um (Lawrence, G. H. M. et al., 1968, Pittsburgh: Hunt Botanical Library). 
References cited only as part of nomenclatural matter are not included in lit- 
erature cited. For shorter notes and reviews, put all references parenthetically 
in text. Use Index herbariorum (Regnum Veg. 106:1-452. 1981) for designations 


OUULLO 


scale and reference to latitude and longitude in each map. 
Proofs and reprint order forms are sent to authors by the printer. Authors 
Authors will be assessed charges for extensive alterations made after type has 


eG 


American 
Fern Number 1 


January—March 1988 


Journal 


QUARTERLY JOURNAL OF THE AMERICAN FERN SOCIETY 


Ceradenia, a New Genus of Grammitidaceae L. Earl Bishop 1 
Isoétes x hickeyi: A Naturally O g Hybrid bet I. echinosp d 
crospora W. Carl Taylor and Neil T. Luebke 6 
eS es ig : DavidB.Lellinger 14 
Isoétes pallida, a New Species from Mexi R. James Hickey 35 
Cover 3 


Information for Authors 


S s Vermont Burlington VT 0405-006, 


The American Fern Society 
Council for 1988 


JUDITH E. SKOG, Biology Dept., George Mason University, Fairfax, VA re President 
DAVID B. LELLINGER, Smithsonian Institution, Washington, D.C. ee Vice-President 
W. CARL TAYLOR, Milwaukee Public Museum, Milwaukee, WI5 Secret 


ary 
JAMES D. CAPONETTI, Dept. of Botany, University of Tennessee, Cole TN — Treasurer 
DAVID S. BARRINGTON, Dept. of Botany, University of Vermont, Burlington, VT 0 

ican Treasurer 


JAMES D. MONTGOMERY, Ecology III, R.D. 1, Berwick, PA 18603 Back Issues Curator 
ALAN R. SMITH, Dept. of Botany, University of Berkeies - 94720. Journal Editor 
DAVID B. LELLINGER, Smithsonian Institution, eee - 205 Memoir Editor 
JOHN T. MICKEL, New York Botanical Garden, Bronx, NY 1 a" iddlehead Forum Editor 
American Fern cual 
EDITOR 

ALAN. SHATTER os a ss ee Dept. of Botany, pagans of California, 

erkeley, CA 94720 

. ASSOCIATE EDITORS 

GERAD GASTONY ...........% Dept. of Biology, Indiana University, Bloomington, IN 47401 

CRSP APL ce ee ee ee Dept. of Botany, University of Kansas 
Lawrence, KS 66045 
DAVID B. LELLINGER be eee a eae U.S. Nat’l Hecbeciun NHB-166, Smithsonian Institution, 
Washington, DC 20560 

TERRY R. WEBSTER ..... Biological Sciences Group, University of Connecticut, Storrs, CT 06268 
The ‘American Fern J 1” (ISSN 0002-8444) i illustrated th 


study of ferns. It is owned by the American Fern Society, and saree at the atone Her! barium, 

University of Vermont, Burlington, VT 05405-0086. Second-class postage paid at Burlington, VT, 
and additional entry point. 

3 eo Ss 409 (i, ff. ss ilk ee 

and orders for back issues should be addressed to Dr. James D. Montgomery, Ecology Ill, R.D. 1, 
Berwick, PA 18603 
. Changes of eddreas, dues, and applications for membership should be sent to the Records 

reasurer. 


LL vk. jj. : res ec 


a 
c..b rea fon nn 


$19.50 net if paid through an agency (agency fee $0.50); sent free to 
members of the American Fern Society (annual dues, $15.00 + $4.00 mailing surcharge beyond 
U.S.A., Canada, and Mexico; life membership, $300.00). 

Back volumes 1910-1978 $5.00 to $6.25 each; single back numbers of 64 pages or less, $1.25; 
 BO-O0 pages. $2. oo each; over 80 pages, $2.50 each, plus — Back volumes 1979 et seq. ‘$8. 00 


oe 


ec... 3 het 


s to AMERICAN FERN JOURNAL, Dept. of a University of 


The editor tributi including miscellaneous . 
‘notes, offers to to exchange or purchase materials, ass cng horticultural notes, and reviews of 
oe non-technical books ca feree. 
_Mr.Nei Ti Lalli i Ne aoe S Sata node tT " Se 7. g dand 


American Fern Journal 78(1): 1-5 (1988) 


Ceradenia, a New Genus of Grammitidaceae 


L. EARL BISHOP # oc2o2ss 
Herbarium, Department of Botany, University of California, Berkeley, California 94720 


William J. Hooker’s generic concepts for ferns essentially dominated the 
taxonomic community for the latter half of the nineteenth century. We now see 
that his genera, based almost solely on soral and venational characters, were 
nearly as artificial in conception as those of Linnaeus’ sexual system for 
flowering plants. With the recognition in this century that some superficially 
similar groups such as the thelypterids and dryopterids were not at all closely 
related, the Hookerian dream of easy and obvious fern genera was permanently 
undermined. Today, even in complex groups, there is still resistance to the 
elucidation of genera based on nontraditional characters. Although 
conservatism is generally a cardinal virtue in taxonomy, the maintenance of very 
large, complex g | cases in which natural and discrete 
groupings of more manageable size can be demonstrated and defined by 
multiple, correlating characters. 

In my paper on Cochlidium (Bishop, 1978), I suggested that the only rational 
alternative to recognizing radically revised generic concepts in the 
Grammitidaceae is the inclusion of all species of the family into a single genus. 
Workers in at least two subsequent publications have adopted this approach for 
New World species (Tryon & Tryon, 1982; Proctor, 1985). The necessarily 
detailed studies to support more workable and hi hic g i pts in th 
family have proceeded rather slowly. But both Parris, from her studies of 
Indo-Malaysian species, and I, in my investigation of Neotropical groups, have 
become convinced that well defined, natural genera can be delimited among 
these ferns (Parris, 1984, 1986). As to the ease of generic recognition, those who 
wish to constitute the entire family Grammitidaceae as a single genus are 
recognizing a genus based on spore and sporangial characters, whereas the 
natural groupings within the family are to be founded on the more easily 
observed features of trichomes and general morphology. 

Among the larger grammitid ferns, the presence or absence of hydathodes 
seems to be a conservative character. In the Neotropics, the great majority of 
anhydathodous species probably constitute an allied group. The only 
Neotropical anhydathodous grammitids not part of this greater alliance are 
Grammitis graminea, G. turquina, and those few species of Grammitis sensu 
stricto in which the hydathodes are reduced to the point of extinction. In 
addition to lacking hydathodes, ferns of the alliance in question have 
concolorous scales, generally coriaceous fronds, and rather weak laminar setae. 
Another interesting character found among these ferns is the tendency for the 
base of the rachis or the distal stipe to be geniculate. This character is found 
among the hydathodous grammitids only in Grammitis asplenifolia and its 
immediate relatives. 


MISSOURI BOTANICAL 


SEP 13 1988 


2 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 1 (1988) 


This anhydathodous alliance falls quite distinctly into three natural groups. 
Glyphotaenium and another small but striking species group related to G. 
mathewsii are being dealt with elsewhere, while the third is here described as 
the new genus Ceradenia. This genus is most easily characterized by the 
Each paraphysis consists of a uniseriate 
unbranched hair of which the distal 2-5 cells bear globose or obpyriform, waxy 
glands. These glands are conspicuously white in most species, but ina few they 
are tan or yellowish. Although the genus name is derived from the waxy 
appearance of these glands, the actual chemical identity of the substance is 
unknown. In young sori the glands completely cover the developing sporangia 
and can be quite apparent even to the unaided eye. In fact, they have been noted 
in the protologue descriptions of several species, but no author has mentioned 
them in connection with species other than his own. Among old sori the glands 
lose their opacity and are rather difficult to detect except with microscopic 
mountings. In addition, it should be noted that field y of in 
fluid and, possibly, high heat in drying also cause ‘the glands to become clear. 
With moderate magnification and careful observation, however, the paraphyses 
will be quite evident in younger sori, even though they will not be strikingly 
white. Apart from the paraphyses, Ceradenia differs from Glyphotaenium in 
lacking setae conspicuously disposed around the sorus. Most species of 
Glyphotaenium show such circum-soral setae, and while no Ceradenia has 
these, a single rather rare species does have receptacular setae, so that this 
differentiation must be made. 


eradenia is a rather | isi bout 55 species. Approximately 
two dozen species await description, but at this time I merely transfer those 
epithets which I t valid described species. The species fall 


quite clearly into two groups, here ranked as subgenera, that differ primarily in 
the rhizome organization. Although this difference is based on the internal 
symmetry of the rhizome, with care and some experience this distinction can be 
discerned by external examination of the stipe insertions. Subgenus Ceradenia 
is characterized by a radically symmetrical rhizome. All but one species bear 
waxy-glandular laminar trichomes and the stipe is short or lacking. Subgenus 
Filicipecten shows a dorsiventrally organized rhizome. The lamina lacks 
waxy-glandular trichomes and the stipe is generally 0.5—3 times as long as the 
lamina. These two subgenera more or less correspond to the groups of 
Ctenopteris curvata and C. meridensis of Copeland (1955). However, he 
included in these groups various species that are not Ceradeniae (or even 
Grammitidaceae), and other species of Ceradenia are scattered through his other 
groupings. »3? 
rb | 
Ceradenia L. E. Bishop, gen. nov. (Fig. 1}—Type: Polypodium curvatum Sw. 
Hinc turma major filicum neotropicalium cogitur quarum aliquot vulgariores 
multae rarae sunt. Rhizoma breve (per unam speciem elongatum) interne 
siphonostelam amphiphloicam endoderme interna praebens, externe paleis 
concoloribus integris aut ciliatis nitentibus vel subnitentibus ex cellulis inflatis 
vel subinflatis constantibus, radialiter aut dorsiventraliter componitur. Lamina 


L. E. BISHOP: CERADENIA 


) 


Fic. 1. Exemplary species of Ceradenia. Subgenus Ceradenia, A-D. A. C. jungermannioides. B.C. 
pruinosa. D.C. discolor. D.C. curvata. Subgenus Filicipecten, E-G. E. C. semiadnata, portion of 
frond. F. C. brunneoviridis. G. C. kalbreyeri. 


4 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 1 (1988) 


plerumque pinnatifida pinnatave praeter modum simplex aut bipinnata, setas 
(pilos simplices quorum muri cellulares secondarie spissantur) pilos hyalinos 
simplices vel ramosos et/aut pilos glandulosos paraphysum similes gerens, 
venis maximam partem 1-furcatis nonnunquam simplicibus ordinatim liberis 
etsi per occasionem conjungentibus, hydathodis caret. Sori mediales vel 
submarginales ad venulam acroscopicam siti setis circum soros carentes, 
sporangia capsulis 140-250 x 120-185 pm annulis suis ex 9—13 cellulis 
constantibus, sporas hemisphaericas vel subtetraedricas frequenter binucleatas 
25—55 pm in diametro longiore includentia, etiam paraphyses quarum consistit 
unaquaeque ex pilo longo cellularum hyalinarum quarum 2-5 distalis glandulis 
globosis vel obovoideis opacis niveis vel eburneis gerunt, praeterea per unam 
speciem setas pro paraphysibus continent. Chromosomata n=37 per illam 
speciem solam pro qua numerus decretus est. 


Subgenus Ceradenia 
Rhizome radially symmetrical, with overlapping leaf gaps (dictyostelic);. 
lamina bearing waxy-glandular trichomes, these similar to the paraphyses, but 
without the long hyaline stalk (no glandular laminar hairs in C. 
jungermannioides); stipe lacking or, if present, less than half the length of the 
lamina; spores 25—40 pm. 


Ceradenia albidula (Baker) L. E. Bishop, comb. nov.—Polypodium albidulum 
Baker in Martius, Fl. bras. 1, 2:598. 1870. 

eradenia capillaris (Desv.) L.E. Bishop, comb. nov.—Polypodium capillare 
Desv., Ges. Naturf. Freunde Berlin Mag. Neuesten Entdeck. Gesammten 

Naturk. 5:316. 1811. 

“Ceradenia curvata (Sw.) L. E. Bishop, comb. nov.—Polypodium curvatum Sw., 
J. Bot. (Schrader) 1800, 2:24. 1801. 

-Ceradenia discolor (Hook.) L.E. Bishop, comb. nov.—Polypodium discolor 
Hook., Icon. pl. 4:pl. 386. 1841. 

Ceradenia farinosa (Hook.) L.E. Bishop, comb. nov.—Polypodium farinosum 

Hook., Icon. pl. 10:pl. 947. 1854. 

-Ceradenia fragillima (Copel.) L. E. Bishop, comb. nov.—Ctenopteris fragillima 
Copel., Philipp. J. Sci. 84:470. 1955 [1956]. 

Ceradenia fucoides (Christ) L.E. Bishop, comb. nov—Polypodium fucoides 
Christ, Bull. Herb. Boissier II, 5:2. 1905. 

Ceradenia herrerae (Copel.) L.E. Bishop, comb. nov.—Ctenopteris hererae 
Copel., Philipp. J. Sci. 84:467. 1955 [1956]. 

—Ceradenia jungermannioides (Klotzsch) L. E. Bishop, comb. nov.—Polypodium 
jungermannioides Klotzsch, Linnaea 20:373. 1847. 

eradenia nubigena (Maxon) L.E. Bishop, comb. nov.—Polypodium 
nubigenum Maxon, Contr. U.S. Natl. Herb. 17:599. 1916. 

€eradenia pearcei (Baker) L. E. Bishop, comb. nov.—Polypodium pearcei Baker 
in Hook. & Baker, Syn. fil. 508. 1874. 

“Ceradenia pilipes (Hook.) L. E. Bishop, comb. nov.—Polypodium pilipes Hook., 
Icon. pl. 3:pl. 221. 1840. 


L. E. BISHOP: CERADENIA 5 


Ceradenia podocarpa (Maxon) L.E. Bishop, comb. nov.—Polypodium 
podocarpum Maxon, Smithsonian Misc. Collect. 56, 24:2. 1911. 

_€eradenia pruinosa (Maxon) L. E. Bishop, comb. nov.—Polypodium pruinosum 
Maxon, Proc. Biol. Soc. Wash. 52:117. 1939. 


Subgenus Filicipecten L.E. Bishop, subgen. nov.—Type: Polypodium 
kalbreyeri Baker — 26% / 0047 
Rhizome dorsiventrale, solenostelicum; lamina glandulis albidis carens; 
stipes lamina 0.5—3plo longior; sporae 35-55 pm. 


/€eradenia brunneoviridis (Baker ex Jenman) L.E. Bishop, comb. nov.— 
Polypodium brunneoviride Baker ex Jenman (as P. brunneo-viride), J. Bot. 
19200. 1677. 

-Ceradenia fendleri (Copel.) L.E. Bishop, comb. nov.—Ctenopteris fendleri 
Copel. Philipp. J. Sci. 84:463. 1955 [1956]. 

_€eradenia kalbreyeri (Baker) L. E. Bishop, comb. nov.—Polypodium kalbreyeri 
Baker in Thurn, Timehri 5:215. 1886. 

€eradenia knightii (Copel.) L.E. Bishop, comb. nov.—Ctenopteris knightii 
Copel., Philipp. J. Sci. 84:419. 1955 [1956]. 

-€eradenia kookenamae (Jenman) L.E. Bishop, comb. nov.—Polypodium 
kookenamae Jenman in Thurn, Timehri 5:215. 1886. 

,€eradenia longipinnata (Copel.) L.E. Bishop, comb. nov.—Ctenopteris 
longipinnata Copel., Philipp. J. Sci. 84:459. 1955 [1956]. 

-€eradenia margaritata (A.R. Smith) L.E. Bishop, comb. nov.—Grammitis 
margaritata A. R. Smith, Proc. Calif. Acad. Sci.40:221. 1975. 

—Ceradenia mayoris (Rosenstock) L. E. Bishop, comb. nov.—Polypodium 
mayoris Rosenstock, Mém. Soc. Sci. Nat. Neuchatel 5, 2:53. 1914. 

-€eradenia melanopus (Grev. & Hook.) L.E. Bishop, comb. nov.—Polypodium 
melanopus Grev. & Hook. (as P. melanopum), Bot. Misc. 3:384. 1833. 

_€eradenia meridensis (Klotzsch) L.E. Bishop, comb. nov.—Polypodium 
meridense Klotzsch, Linnaea 20:380. 1847. 

,Geradenia nudicarpa (Copel.) L. E. Bishop, comb. nov.—Ctenopteris nudicarpa 
Copel., Philipp. J. Sci. 84:462. 1955 [1956]. 

Ceradenia semiadnata (Hook.) L.E. Bishop, comb. nov.—Polypodium 
semiadnatum Hook., Icon. pl. 10:pl. 948. 1854. 

€eradenia spixiana (C. Martius ex Mett.) L. E. Bishop, comb. nov.—Polypodium 
spixianum C. Martius ex Mett., Abh. Senckenberg. Naturf. Ges. 2:57. 1856. 


LITERATURE CITED 


Bisuop, L. E. 1978. Revision of the genus Cochlidi (G itid : }. Amer. Fern J. 68:76—94. 

CopELAND, E. B. 1955 [1956]. Ctenopteris in America. Philipp. J. Sci. 84:381-471. — 

Parris, B.S. 1984. Another intergeneric hybrid in Grammitidaceae: Ctenopteris longiceps X 
Grammitis sumatrana. Fern Gaz. 12:337—340. : ae 

1986. Grammitidaceae of peninsular Malay Singapore Kew Bull. 41:491—517. 

Procror, G. R. 1985. Ferns of Jamaica. London: British Museum [Natural History]. 

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


American Fern Journal 78(1): 6—13 (1988) 


Isoétes Xx hickeyi: A Naturally Occurring Hybrid 
between I. echinospora and I. macrospora 


W. Cart TAYLOR and NEIL T. LUEBKE 
Botany Section, Milwaukee Public Museum, Milwaukee, WI 53233 


Speculation about the occurrence of interspecific hybrids in Isoétes began as 
early as 1896 when Dodge noted that some New England species of Isoétes 
“‘intergrade at times.’’ Jeffrey and Hicks (1925) and Jeffrey (1937) attributed 
variation in size and shape of megaspores, lack of protoplasmic content in 
microspores, and lagging chromosomes during microsporogenesis to 
hybridization in Isoétes from Nova Scotia. Boom (1980) artificially crossed 
several Isoétes species. Subseq y, he proposed six interspecific hybrids f 
southeastern United States (Boom, 1982). However, Kott and Britton (1983) 
believed that there was insufficient evidence to support interspecific 
hybridization in northeastern North American Isoétes, and that either 
ethological or sterility barriers exist among most taxa. Data are presented here to 
document the natural occurrence of an interspecific hybrid in Isoétes. 

Hybrids between Isoétes echinospora and I. macrospora were discovered in 
Neva Lake, Langlade County, Wisconsin. These plants possess megaspores that 
are more variable in size, shape, and surface ornamentation than typical 
megaspores of I. echinospora and I. macrospora. The origin of I. echinospora X 
macrospora is supported by evidence from spore morphology and viability, 
chromosome counts made from root tip squashes, and electrophoresis of leaf 
enzymes. 


Isoétes <x hickeyi W. Taylor & N. Luebke, hyb. nov.—T PE: Wisconsin, Langlade 
County, Neva Lake, 15 Oct 1983, Taylor 5010b (MIL). 

Plantae inter I. echinosporum et I. macrosporum interpositae. Megasporae 
forma et amplitudine variabiles, plerumque normales hemisphaerio proximali 
angulari, aliquando globosae hemisphaericae seu libramentiformae, ad 1000 pm 
longae, ornamento paginae echinato seu reticulato. Microsporae plerumque 
aprotoplasmicae. Chromosomatum numerus 2n = 66. 

Known only from the type locality, from several collections. Eight plants of I. 
x hickeyi (Taylor & Luebke 5157) are in culture at MIL and will be distributed to 
herbaria. One plant is at BM. The epithet for this hybrid honors Dr. R. James 
Hickey, who has encouraged and assisted us in our study of Isoétes in North 
America. 


MATERIALS AND METHODS 
Spore morphology.—Megasporangia of I. x hickeyi were opened and their 
contents dispersed in petri dishes containing demineralized water. Megaspores 
were examined and photographed through a dissecting microscope. Fifty 


TAYLOR & LUEBKE: ISOETES x HICKEYI 7 


megaspores of I. x hickeyi and twenty-five megaspores each of I. echinospora 
and I. macrospora were air-dried and placed in concavity slides; their greatest 
diameters, including ornamentation, were measured using a compound 
microscope fitted with an ocular micrometer. Spore surfaces of I. x hickeyi, I. 
echinospora, and I. macrospora were examined using an Hitachi S-570 scanning 
electron microscope (SEM). Air dried spores were mounted on copper tape that 
had been glued to stubs with colloidal graphite. Mounts were sputter coated with 
140 nm of gold. Megaspores were photographed at 100 x and microspores at 
1000x. 

Hydrated microspores of I. x hickeyi, I. echinospora, and I. macrospora were 
mounted on slides in Aqua-Mount (American Scientific Products). Two 
hundred twenty-five randomly selected microspores of I. x hickeyi were 
examined for the presence of an ellipsoid endosporium and protoplasm. 

Pe 4} . 


-Ellipsoid endosporia, filled with protoplasm, longest 
diameters in twenty-five microspores each of I. x hickeyi, I. echinospora, and I. 
macrospora. 


Spore viability.—Megaspores of I. x hickeyi, I. echinospora, and I. 
macrospora were cultured following the procedures described by Taylor and 
Luebke (1986) for germinating spores of aquatic Isoétes. Three replicate cultures 
of I. echinospora and I. macrospora and eight replicate cultures of I. x hickeyi 
were made. Each culture dish contained 30 or more megaspores. After 
vernalization at 2°C for 100 days, spores were incubated at 18°C with a 12L:12D 
photoperiod. Cultures were scored for germination after 100 days of incubation. 

Chromosome counts.—Chromosome counts of I. x hickeyi were made from 
root tip squashes. Roots up to 1 cm long with white apices were collected 
between 8 and 9 AM and pretreated for 3 hours in a saturated solution of 
paradichlorobenzene in the dark at ca 20°C, then fixed in a 3:1 solution of 95% 
ethanol and glacial acetic acid. Following a protocol similar to that of Wiley 
(1971, p. 59), each root was hydrolyzed for 30 minutes in a 3:1 solution of 95% 
ethanol and concentrated HCl, neutralized for 30 minutes in 95% ethanol, 
stained for 35 minutes in Wittman’s hematoxylin, and destained for 3 minutes in 
glacial acetic acid. Finally, the apical 0.5 mm was removed and placed in a drop 
of Hoyer’s medium where it was macerated with a brass rod and squashed. 
Chromosome figures were photographed at 1060 x . 

Electrophoresis.—Leaves from three plants each of I. echinospora, I. 
x hickeyi, and I. macrospora from Neva Lake were crushed and ground in a 
buffer solution. The resulting mixtures were absorbed onto filter paper wicks 
and subjected to horizontal starch gel electrophoresis. Gels were stained for 
phosphoglucomutase (PGM) and triosephosphate isomerase (TPI). 
Electrophoretic procedures and composition of grinding buffers, gel and 
electrode buffers, and staining methods follow Soltis et al. (1983). Phosphate 
grinding buffer was used for PGM, and Tris-HCl grinding buffer was used for TPI. 
PGM was resolved on electrode and gel buffer system 5 in a 12% gel and TPI on 
system 7 in a 13% gel. Electrophoresis was conducted at 4°C and at a constant 
current of 40 mamp for PGM and 35 mamp for TPI. Wicks were removed from the 
gels after 15 minutes. PGM was run for 6.5 hrs. and TPI for 8.5 hrs. Gel slices were 


8 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 1 (1988) 


incubated in substrates in the dark at 37°C for 1 hr., rinsed in distilled water, and 
photographed. 


RESULTS 


Spore morphology.—Megaspores of I. echinospora and I. macrospora 
megaspores are typically globose and uniform in size but those of I. x hickeyi are 
variable in shape and size. Most megaspores of J. x hickeyi have a flattened, 
angular proximal hemisphere, others are subglobose to nearly hemispherical, 
whereas a few appear as tiny wedges or large dumbbell-shaped structures (Fig. 
1). Megaspores of I. x hickeyi average 465 + 126 ym across but range from less 
than 100 pm to over 1 mm in length. Megaspores of Neva Lake I. echinospora 
average 435 + 26 um indiameter, 390 pm to 490 pm, whereas those 
of Neva Lake I. macrospora average 630 e 47 pm in n diarheter, ranging from 550 
um to 720 pm. 

Differences between the microspores of I. x hickeyi and those of I. 
echinospora and I. macrospora are seen in the contents and the shape of the 
enclosing endosporium located within the perine (Fig. 2). Whereas the 
endosporium of I. x hickeyi microspores is variable in shape, often shriveled, 
and usually contains little or no protoplasm (Fig. 2a), the endosporium of I. 
echinospora and I. macrospora microspores is ellipsoid and filled with 
protoplasm (Figs. 2b, c). Only about 10% of I. x hickeyi microspores possess an 
ellipsoid endosporium and only about one half of these contain any protoplasm 
(see microspore at lower left in Fig. 2a). Ellipsoid endosporia of I. x hickeyi 
microspores average 34 ym in length, intermediate between those of its putative 
parents, I. echinospora and I. macrospora, which average 26 pm and 41 pm in 
length, respectively. 

Megaspore ornamentation of I. x hickeyi appears to combine ornamentation 
patterns of both I. echinospora and I macrospora (Fig. 4). Megaspores of I. 
echinosporaare echinate (Fig. 4a), whereas those of I. macrospora bear branched 
to anastomosing ridges and a band of papillae encircling the distal side of the 
equitorial ridge (Fig. 4b) (Kott & Britton, 1983). Megaspores of I. x hickeyi bear 
spines and short ridges that occasionally branch and anastomose (Figs. 4c, d). 
They also possess a band of papillae along the distal side of the equatorial ridge. 

Mature microspores of I. echinospora are spinulose (Fig. 5a) and those of I. 
macrospora are papillate (Fig. 5b). Surface ornamentation of I. x hickeyi 
microspores varies from spinulose to tuberculate (Figs. 5c, d). 

Spore viability —Megaspores of I. x hickeyi did not germinate in culture, 
whereas I. echinospora and I. macrospora megaspores did germinate to form 
megagametophytes (Table 1). Megaspores were scored as having germinated if 
they opened to expose gametophytic tissue that developed archegonia. 

Chromosome counts.—Root tip cells of I. x hickeyi are hexaploid 
(2n=6x=66) (Fig. 3). Isoétes echinospora is a diploid (2n=2x=22) and I. 
macrospora is a decaploid (2n = 10x = 110) (Kott & Britton, 1980). Neva Lake I. 
echinospora and I. macrospora conform to these published counts. 

Electrophoresis—Enzyme banding patterns for PGM and TPI are shown in 
Figure 6. The PGM zymogram shows four bands of activity (Fig. 6a). Isoétes 


TAYLOR & LUEBKE: ISOETES x HICKEYI 9 


Fics. 1-3. Spores of Isoétes x hickeyi and parents and I. x hickeyi wags cisggger es Fic. 1. Incident 
00u.m 


light photomicrograph of I. x hickeyi megaspores in water; bar = = 2 ee 
photomicrographs of Isoétes microspores from Neva Lake; a. I. x hickeyi; b. I mane : : 
macrospora; bar = 50 wm; (e) endosporium; (p ) Lagans Fic. 3. Brightfield mlannemrnaniatan 0 

x hickeyi chromosomes in root tip squash; bar = 10 pm 


echinospora expresses only the least swan band. Isoétes x hickeyi also has this 

not appear in I. macrospor 
ap pee a isodtes x hickeyi also fas these three bands but they do not 
appear in Isoétes echinospora. The TPI zymogram shows five bands a powiad 
Fig. 6b). Isoétes echinospora expresses only the most anodal ra — es 
macrospora has three, less anodal bands. Isoétes X hickeyi expresses live 7 . 
Four of these bands appear to correspond with bands possessed by - . 
echinospora or I. macrospora. Isoétes x hickeyi has an additional band tha 
occurs between the two most anodal bands for each putative parent. 


10 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 1 (1988) 


Fic. 4. SEM photomicrographs of Isoétes megaspores from Neva Lake; a. I. echinospora; b. I. 
macrosporg; c. and d. I. x hickeyi; bar = 250 um 


DISCUSSION 


Spore size and ornamentation are often primary characters used to distinguish 
species of Isoétes. Using a 10x hand lens, it is easy to recognize the globose, 
echinate megaspores of I. echinospora and the ridged to reticulate spores of I. 
macrospora. In contrast, the megaspores of I. x hickeyi are often seen as flattened 
or irregular in form, variable in size, and ornamentated with closely set spines or 
ridges. 

Species of Isoétes produce viable spores which are globose in form and 
uniform in size. Variable, nonviable spores characterize interspecific hybrids of 
many pteridophytes (Taylor et al., 1985, Wagner et al., 1986). Megaspores of I. 
x hickeyi are variable in form and size and do not germinate in culture. 
Microspores of I. x hickeyi were not cultured, but a lack of contents indicates 
most are essentially nonviable. 


TAYLOR & LUEBKE: ISOETES x HICKEYI 11 


Fic. 5. SEM photomicrographs of Isoétes microspores from Neva Lake; a. I. echinospora; b. I. 
macrospora; c. and d. I. x hickeyi; bar = 25 pm. 


Isoétes x hickeyi (2n=66) has the expected chromosome number for an 
interspecific hybrid between I. echinospora (2n=22) and I. macrospora 
(2n=110) where their haploid gametes (n=11 and n=55, respectively) have 
fused. 

Enzyme electrophoretic profiles of I. x hickeyi are additive for I. echinospora 
and I. macrospora and support the hypothesis of hybrid origin for I. x hickeyi. 


TABLE 1. In Vitro Megaspore Germination of Neva Lake Isoétes Taxa. 


No. megaspores No. gametophytes % germination 
cultured formed 
I. x hickeyi 271 0 0 
I. echinospora 90 87 98 
I. macrospora 91 91 100 


ee aa eee 


a2 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 1 (1988) 


Fic. 6. Starch gels stained for PGM and TP after electropt is of Neva Lake Isoétes leaf enzy 

PGM; b. TPI; lanes 1—3 I. echinospora (ec), lanes 4—6 I. * hickeyi (ec x ma), lanes 7—9 I. macrospora 
(ma). Arrowhead indicates a band that possibly represents a heterodimeric enzyme present only in 
the hybrid. 


Such enzyme characters are valuable in detecting hybrids in pteridophytes 
(Haufler, 1985). PGM is a monomeric enzyme and TPI is dimeric (Harris & 
Hopkinson, 1976). The PGM profile for I. x hickeyi is additive for all bands 
expressed by its putative parents. The TPI profile for I. x hickeyi is also additive 
for all bands expressed by its putative parents but it possesses an additional band 
that probably rey ts a heterodimeri ymef ts coded by 
genes contributed by each parent. For both PGM and TPI there appears to be fixed 
heterozygosity in the decaploid, I. macrospora. This is not surprising given the 
high ploidy of this species. From the results obtained, it is impossible to 
determine with certainty that the bands of enzymatic activity seen on the gels are 
-coded for by alleles at a single locus or by multiple loci. 

Dosage effects in I. x hickeyi also support the hypothesis of hybrid origin. For 
both PGM and TPI the bands in I. xhickeyi corresponding to those of I. 
macrospora are more intense than those corresponding to I. echinospora. This 
would be expected in a hybrid having five doses of one parent’s genes and only a 
single dose of the other parent’s genes. 

Data from spore size, shape, ornamentation, and viability in addition to 
chromosome counts from root tip squashes and electrophoresis profiles of PGM 
and TPI in leaves indicate that I. x hickeyi is a naturally occurring hybrid 
between I. echinospora and I. macrospora. 

Isoétes x hickeyi is known only from Neva Lake, a small, glacially formed, 
soft-water lake with a sand, gravel, and muck bottom. Plants of I. x hickeyi were 
found less than 1 m deep, on the east side of the lake, near a public boat launch. 
Disturbance from boating, swimming, and wading occurs at the site. Isoétes 
macrospora is abundant where I. x hickeyi was found. Plants of I. echinospora 
are also present at the collection site, but they are more numerous in shallower 
water, close to shore. Isoétes echinospora and I. macrospora are the only species 
of Isoétes known within a 200 km radius of Neva Lake. Both species are common 
in the lakes of northern Wisconsin. Their ranges are sympatric over much of 
eastern North America where they often grow together in the numerous lakes 
and ponds of this region. 


i¢£ 
Irom 


TAYLOR & LUEBKE: ISOETES x HICKEYI 13 


ACKNOWLEDGMENTS 
We are grateful to David Lellinger for providing us with a Latin description of I. x hickeyi. We 
thank Susan Brahm, Donald Britton, Christopher Haufler, James Hickey, James Montgomery, Clive 
Jermy, Robbin Moran, Thomas Ranker, Jovanka Ristic, Herb and Florence Wagner, Michael 


Windham, Charles Werth various ways. We are also grateful to 

AA ily res! i £ Pf gtl University of Wi s an Az] 1 Cl Laboratory 

available to us and to Gretchen D for her kind g ity whicl de this study possible. 
LITERATURE CITED 


Boom, B. Me ig Intersectional yrange in a Amer. Fern J. 70: 

United States. iota 47: na . 

Bares 1896. Th f df llies of New England. Binghamton: W. N. Clute 

Kott,L. S. and D. M. BRITTON. 1980. Chromosome numbers for Isoétes in dee North 
America. Canad. BE cei 58:980—984. 

and ———_—_. 3. Spore morphology and taxonomy of Isoétes in northeastern North 

Fey, Canad. 4 ee 61:3140—3163. 

Harris, H. and D. A. HOPKINSON. 1976. er of enzyme electrophoresis in human genetics. 
Amsterdam: North-Holland Pu 

HAUvFLER, C. H. 1985. Pteridophyte 
Soc. Edinburgh 86B:315-—323. 

JEFFREY, E. C. 1937. The cytology of a het Isoétes. Bot. Mag. (Tokyo)51:203—209. 

—_____ and. C. Hicks. 1925. The reduction division in relation to mutation in plants and 
Amer. Naturalist 59:410—426. 

Soxtis, D. E., C. H. HAUFLER, D. C. DARROW, gins - ngeoeabt 1983. Peden Ant electrophoresis of 
f schedules 


biology: the electrophoretic approach. Proc. Royal 


Pf 


i igen lamba eatin Sa es inating sf d growing sporlings of aquatic Isoét 
er. Fern J. 76:21—24. 
—_——, and M.B. SMITH. 1985. eat and hybridization in North American 


Qui illworts. Proc. Royal Soc. Edinburgh 86B:259—263. 

Wacner, W. H. Jr., F. S. WAGNER, and W. C. TAYLOR. agetls Detecting abortive spores in herbarium 
specimens of sterile hybrids. Amer. Fern J. 76:129—140. 

Witey, H. L. 1971. Microtechniques, a laboratory guide. New York: Macmillan. 


16 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 1 (1988) 


excurrent veinlet may even branch and rejoin, forming a narrowly rhombic 
areola parallel to the main lateral veins. 

It is possible that the type of venation found in C. phyllitidis, which has rather 
narrow laminae, is an intermediate stage in the development of the more loosely 
organized venation found in C. latum and in other Campyloneurum species with 
very wide laminae; a succession of fronds on developing plants of C. latum 
should be observed to see if this is true. If so, this may account for the confusion 
between the two species in Central and South America, where the somewhat 
inconstant venation has caused some authors to consider the two species to be 
only one. That decision may prove to be correct, in which case C. phyllitidis (all 
with constant, phyllitidis-type venation) ranges from Florida through the 
Antilles and the Guianas to southern Brazil, and C. latum (with both latum- and 
phyllitidis-type venation) ranges from Mexico through the Andean countries to 
Paraguay and Uruguay. 

Species with fronds usually less than 4(6) cm wide have branched, 
areola-forming, slightly or not prominulous main veins usually without 
transverse veins (angustifolium-type venation). The main veins are often 
sinuous. Because angustifolium-type venation is sometimes seen in 
depauperate fronds of species usually having latum-type venation, 
angustifolium-type venation appears to be the basic condition in the genus, and 
it is reminiscent of other polypod genera like Microgramma, which could be 
ancestral to Campyloneurum. 


GEOGRAPHY 


Campyloneurum is basically neotropical (Fig. 1). Only 3 species are known 
from North America north of Mexico and only 7 from Paraguay, Uruguay, Chile, 
and Argentina. More species are known from the Antilles (10), the Guianas and 
Brazil (19), and Central America (19). The greatest species richness occurs in the 
Andean countries (27 in Venezuela—Colombia—Ecuador; 27 in Peru and 
Bolivia). Narrow endemics, those species occurring in a single region of Fig. 1, 
are most prevalent in Central America (32% of the species) and in Venezuela— 
Colombia—Ecuador (19% of the species). Wide endemics, those species 
occurring in two adjacent regions of Fig. 1, are most prevalent in Peru—Bolivia 
(33% of the species) and in Venezuela—Columbia—Ecuador (30% of the 
species). 

The species as I construe them generally have coherent ranges, which is 
additional evidence that they are biologically valid entities. Of the ten new 
species and combinations published in this paper, 4 are narrow endemics (2 
from southern Central America and 2 from Brazil), and 4 are wide endemics (2 
from the Andean countries, 1 from southeastern Brazil to Paraguay, Uruguay, 

d northeastern Argentina, and 1 from Costa Rica to Ecuador). Only two species 
range more widely, C. irregulare (Costa Rica; Colombia to Bolivia) and C. 
fuscoquamatum (French Guiana: Colombia to Bolivia). 


NEw SPECIES AND COMBINATIONS 


Campyloneurum centrobrasilianum Lellinger, sp. nov. (Figs. 2, 8).—“TyPE: 
Brazil, Est. Minas Gerais, Vicosa, Kuhlmann 1898 (US). 


D. B. LELLINGER: CAMPYLONEURUM 7 


Fic. 1. Number of Campyloneurum species (and single-region endemics in parentheses) in the 
various regions of the New World. 


Ab C. angustipaleato in squamis rhizomatis lanceolatis, subrepandis, ca. 5—8 
cellulis lati lulis sinuatis differt. 

Bhicaies sounanie 2—2.5 mm in diam., black, not alan ananae 
phyllopodia 0.5—1.5 mm long, 1.25 mm in diam.; rhizome scales = ee 
concolorous, brown throughout or slightly paler at the margins, — a e to 
rather broadly lanceolate, 2-3 mm long, 0.5-1 mm wide, attentuate “ t be 
with overlapping auricles at the base or appearing peltate, the = a 
rarely with a few short cilia, the cells clathrate with clear, s ight y iri : 
lumina, those above the base 2—2.5(3) times longer than wide. Stipes aig cm 
long, ca. 1 mm wide, slightly flattened, not sulcate, stramineous. — ne 
attenuate at the base and apex, at least slightly revolute at the — a 
cm long, 3—6 mm wide (or sometimes appearing narrower when the margins a 


18 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 1 (1988) 


H 
é 


tae 


| 


Bit 
ta 


A 
Vive SQ 
ihe 


Figs. 2—7. Rhizome scales of Campyloneurum species with cellular detail. Fic. 2. C. 
centrobrasilianum, Irwin et al. 13057 (US). Fic. 3. C. cooperi, Cooper 6053 (US). Fic. 4. C. 
fuscosquamatum, Stork & Horton 9452 (US). Fic. 5. C. inflatum, Wurdack 868 (US). Fic. 6. C. 
irregulare, Holdridge 1577 (US). Fic. 7. C. wacketii, Vélio 80 (US). 


D. B. LELLINGER: CAMPYLONEURUM 19 


strongly inrolled due to desiccation), olive green adaxially, paler green with the 
blackish veins slightly visible abaxially; midrib stramineous or olive green, 
bearing a few broadly lanceolate, subclathrate, pale brown scales up to 1 mm 
long and 0.5 mm wide; main lateral veins absent, the veinlets in 1(2) series of 
areolae with 1 excurrent unbranched fertile veinlet in each areola and marginal 
free veinlets present; sori ca. 1.5 mm in diam. 

Discussion.—This species is closely related to C. angustipaleatum (Alston) M. 
Meyer ex Lellinger and also to other members of the genus with Vittaria-like 
fronds, such as C. austrobrasilianum (Alston) de la Sota and C. aglaolepis 
(Alston) de la Sota. 

Paratypes: BRazit. Distr. Fed.: Gallery forest ca. 25 km SW of Brasilia, 1000 m, Irwin et al. 13057 

ias: Gall f tca. 30k doP. 


(F, GH, US).G m Nof Alt ca. 1250 m, Irwinet al. 33063 (F, UC, 
US). 
C cooperi Lellinger, sp. nov. (Figs. 3, 9).<T yee: Costa Rica, Pcia. 


Cartago, Cartago, 4250 ft, Cooper 6053 (US832717; isotype US829104). 
Ab C. rigido J. Smith in stipitibus longi brevioribus et squamis rhizomatis 
cellulis non sinuatis differt. 

Rhizomes short-creeping, 4-6 mm in diam., black, slightly pruinose, the 
phyllopodia 2-3 mm long, 1.5—2 mm in diam.; rhizome scales spreading, 
concolorous, brown, broadly lanceolate, 2-3 mm long, 1-1.5 mm wide, 
attenuate at the apex, auriculate with overlapping auricles at the base, the 
margins entire, the cells clathrate, with translucent, yellowish lumina, those 
above the base up to 2 times longer than wide. Stipes obsolete or nearly so (up to 
1 cm long), ca. 1.5 mm wide, slightly flattened, not sulcate, stramineous or pale 
brown. Laminae linear, attenuate to a round apex, more gradually attenuate to 
the base, usually strongly revolute at the margins, 15-50 cm long, 10-16 mm 
wide, yellowish-green and shiny adaxially, yellowish-green and less shiny with 
the blackish veins rarely visible abaxially; midrib paler or darker than the lamina 
surface, bearing a few broadly lanceolate, clathrate, brown scales up to 2mm 
long and 1 mm wide; main lateral veins absent, the veinlets in 2 series of areolae 
with 1 excurrent unbranched fertile veinlet in each areola and marginal free 
veinlets present; sori ca. 2.5 mm in diam.; receptacular paraphyses and 

ersistent sporangium stalks present. 
: Disacsion "This very ie species seems to be most closely related to C. 
rigidum J. Smith from southeastern Brazil. Presumably it is a pendent epiphyte. 
Unfortunately, the locality data are all inexact. Probably it grew on the southern 
and southeastern slopes of Volcan Irazu, an area now heavily devoted to 
agriculture. Plants could remain in forests in the Tapanti area. Campyloneurum 
cooperi has been confused with C. angustifolium, but has medium brown, rather 
than dark brown, rhizome scales and especially the basal cells of the scales are 


isodiametric, rather than distinctly elongate. 
Paratype: Costa Rica. Cartago: Turrialba, Wercklé in 1901-1905 (US). 


~“Campyloneurum densifolium (Hieron.) Lellinger, comb. nov.~Polypodium 
angustifolium f. densifolium Hieron. Bot. Jahrb. Syst. sacar: ws 
“densifolia.” “LECTOTYPE (chosen here): Ecuador, Pcia. Azuay, i 


wuts srates He 1898 TAMILIA: Yolypodivesse 
2542678 PROVEDEMOIA: Vigoss - Mina 
DATA: 29 de woventbro de 1935 

— OLR. T0. Koen 


Fic. 8. Holotype of C. centrobrasilianum. Fic. 9. Holotype of C. cooperi. 


git 


3582717 


ae ee 


OS PANT COCTMODENIES : 
Correcix foaw f € a, Caw Costa ® 


(8861) Ll WAAWON 82 AWN'TIOA “TYNUNO[ NYA NVORMANV 


D. B. LELLINGER: CAMPYLONEURUM 21 


Occidental de Cuenca, near Las Yerbas Buenas, 2500—2900 m, Lehmann 


Discussion.—Unlike C. irregulare Lellinger, with which this species can be 
confused, C. densifolium has mostly or entirely lax, rather adherent rhizome 
scales that are broadly ovate and peltate at the base, and its rhizomes are not 
pruinose. These characters also distinguish it from C. amphostenon, which has 
narrow, lanceolate rhizome scales and pruinose rhizomes. Its laminae are 
regularly 1-3 cm wide (many of them are wider than are those of C. irregulare), 
conspicuous hydathodes are present on the adaxial surface of the laminae, and 
the plants are usually epipetric or terrestrial, although sometimes they are 
epiphytic. 

Representative specimens: VENEZUELA. Mérida: 25 km from Mérida along El Valle road, ca. 2800 
m, Breteler 4663 (US). Cotomsia. Magdalena: Sierra de Perija, 11 km ENE of Manaure, 47 km E of 
Valledupar, 2700 m, Grant 10838 (COL, US). Norte de Santander: Road from Pamplona to Toledo, 
crossing the divide between R. La Teja and R. Mesme, 2800-3000 m, Killip & Smith 19858 and 19915 
(both GH, US). Santander: Mountains E of Las Vegas, 3000-3300 m, Killip & Smith 15809 (GH, US). 
Cundinamarca: Low ridge E of Calle 55, Bogoté, 8800-9000 ft, Little & Little 7825 (GH, US). 
Antioquia: Paramo de Sonson, ca. 2750 m, de Garganta F. 2049 (US). Valle: E slope of Cordillera 
Central, basin of R. Bugalagrande, Cuchilla de Barragén between Las Azules and Las Violetas, 3100 
m, Cuatrecasas 20797 (F, US). Cauca: Mt. Purace, Canaan, 3100-3300 m, Killip (GH, US). Narino: 
Cordillera Oriental, I iwat fR. Guap lak I blo of Caballo Rucio. 3000 m, Ewan 16555 
(US). Ecuapor. Carchi: 18 km W of Julio Andrade on road to El Carmelo, 3200 m, Boom, Balslev & 

. ] Pee a aie me a HS 213 


bm 


Luteyn 1402 (US). Pichincha: Km. 28 of old Quito—Sto g 2. 0m, Dodson 


of R. Matadero), 18—20 km W of Cuenca, 9800-10300 ft, Camp E-4227 (UC, US). Loxa: Muletrack 
between Amaluza and Palanda, W slope of C. Amarillo, 3000 m, Oellgaard & Balslev 9610 (GH). 
Napo: Papallacta, ca. 3100—3300 m, Harling et al. 10329 (GH, US). Peru. Cajamarca: Las Tres Cruces 


Puma-urcu ESE of Chachapoyas, 2700-3000 m, Wurdack 706 (US). S. Martin: Tarapoto, 750 m, LI : 
Williams 6127 (F). La Libertad: Laguna de los Ichus, 3600 m, Lopez & Sagdstegui 3236 (GH). come 
Huaccana, cerro E of Tupe, 2830 m, Cerrate 1062 (GH). Junin: Queros, Soukup 6170 (GH). Pasco: 
Between Huariaca and C. de Pasco, 3800-3900 m, Ferreyra 9502 (GH). Huancavelica: Yacuhuamay, 
6 km below Surcubamba, 2200 m, Tovar 3680 (GH). Cuzco: Chincheros, 3500—3600 m, Davis et al. 
1649 (F). Bourvia. La Paz: Sorata, R. S. Williams 2635 (US). Province unknown: R. Sanjana, 3200 m, 
Herzog 2391 (US). 


a ; tum Lellinger, sp. nov. (Figs. 4, 10).<Type: Peru, 
Den. Huanuco, Tingo Maria on the Rio Huallaga, Stork & Horton 9452 
US; isotypes F, GH, UC). oe : : 
ie brite in squamis ce longioribus angustioribusque integris 
concoloribus vel leviter bicoloribus fuscis differt. 
Rhizomes very long-creeping, 1.5—2.59 mm in diam., pale green akira a 
blackish in age, not pruinose, the phyllopodia 0—1 mm long, ca. 2mm = iam., 
rhizome scales strongly spreading, concolorous or slightly bico dona 
blackish-brown, sometimes paler at the margin, acicular-lanceolate 1 - : 
dilated base, 3—3.5 mm long, 0.75—1 mm wide at the base, ca. 0.1mm is above 
the base, attenuate at the apex, nearly orbicular with short, non-overlapping 


22 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 1 (1988) 


auricles at the base, the margins entire, the cells subclathrate with narrow, 
almost occluded lumina, those above the base 2—4 times longer than wide. 
Stipes variable with respect to the lamina length, (1)2—10(18) cm long, 1-2 mm 
wide, flattened or slightly sulcate, stramineous. Laminae narrowly elliptic to 
narrowly lanceolate, caudate or occasionally acute at the apex, acute (in broad 
laminae) to acuminate (in narrow laminae) at the base, plane or scarcely revolute 
at the cartilaginous margins, 20—40 cm long, 2—8 cm wide, rather dark olive 
green adaxially, paler olive green abaxially; midrib pale brown or often blackish, 
bearing a few scales like those of the rhizome except about 1/3 as large; main 
lateral veins 4—7 mm distant, prominulous; transverse veins flush or within the 
lamina tissue, more visible from the adaxial than the abaxial side, always bearing 
2 fertile included veinlets and lacking an intermediate indi — marginal 
free veinlets absent; sori 1-1.5 mm in diam., paraphyses abse 

Discussion.—This species was formerly confused with C. ite (Aublet) C. 
Presl, largely because both species have a thin, very long-creeping rhizome and 
because some lamina forms of C. fuscosquamatum resemble those of C. repens. 
This species is an epiphyte or perhaps a hemiepiphyte on shrubs and small trees, 
mostly at low elevations. | 

Paratypes: C i de la Matarena, trail between R. Guéjar and Cafio Guapayita, 
ca. 500-600 m, Idrobo & ieosugrs 817 (GH, US). Caqueta: 20 km SE of Garzon, 7800 ft, Little 9379 
(US). Ecuapor. Imbabura: Vic. of Laguna de la Virgen on ridge S of R. Clavadero, E of Cayambe Peak, 
8750 ft, Wiggins 10459 ose Pichinch dillera Occidental, Coraz6n Peak and Pass, 2800 m, Ewan 
16419 (US). Peru. S. Martin: Road to R. Tocache, 400 m, Schunke V. 3590 (F, UC, US). Loreto: Supte 
River N of Tingo Maria, 600 m. Stork & Horton 9568 (F, GH, UC, US). Junin: E of Quimiri Bridge near 
La Merced, 800-1300 m, Killip & Smith 23896 (F, GH, US). Ayachucho: R. Apurimac Valley, near 
Kimpitiriki, 400 m, Killip & Smith 22861 (F, US). Cuzco: Ca. 4 km NE of Hac. Luisiana and Apurimac 
River, ca. 670 m, Dudley 11501 (GH, US). Botrvia. La Paz: Road to Tipuani, Hac. Simaco, 1400 m 
Buchtien 5257 (GH, US); Cochabamba: Antahuacana, 160 km NE of Cochabamba, 750 m, Bucitien 
2154 (UC, US), 2155 (US), and 2158 (GH, US). FreNcH Gu1ana. Foot of Mt. Galbao, 10 km W of Saiil, 
de Granville 8500 (CAY, NY, P, US, Z). 


Campyloneurum inflatum M. Meyer ex Lellinger, sp. nov. (Figs. 5, 11) TYPE: 
Colombia, Depto. Cauca, Cordillera Occidental, W slope of Cerro de 
Munchique, basin of R. Tambito, 2000—2500 m, Arbelaez & Cuatrecasas 
6244 (US; isotypes COL not seen, F). 

Ab C. sublucido in laminis maioribus non lucidis et soris maioribus differt. 

Rhizomes long-creeping, 2—2.75 mm in diam., greenish to stramineous but 
turning black in age, not pruinose, the phyllopodia 1-2 mm long, ca. 3 mm in 
diam.; rhizome scales dense, spreading, persistent, concolorous or slightly 

bicolorous, golden turning brown with golden margins in age, lanceolate from a 

slightly dilated base, 4-5 mm long, 1 mm wide, attenuate at the apex, with 

overlapping auricles at the base or appearing peltate, the margins bearing low 
teeth formed by the extension of end walls of adjacent marginal cells, the cells 
clathrate with translucent, yellowish lumina, those above the base 1—2 times 

longer than wide. Stipes 3—4 cm i , 9—22 cm long, 1.5—2 

mm wide, slightly suleste at least at the apex, stramineous. Laminae elliptical, 

udate at the apex, acute or subacute at the base, slightly revolute 

a“ the: hinaiity cartilaginous margins, 15—31 cm long, 5.5—10.5 cm wide, pale 


Bt 
Pelypatiun repwns ssil, 
Dagan Pew fotine Lacy 


WHT ERaieY OF CaLEreREA 
owasit Retsaeet duectee Biegmdtshon ta We kates, 1ent 60 
Pia, Beastia Ctecha, Hiss, ARAMRENIA 
Plaga, 


Por. age Bummody oo trie on Bho Basi lage; 
pial 


Pers, 
et, be, He ets Stark, Get, Barton 


Fic. 10. Holotype of C. fuscosquamatum. Fic. 11. Holotype of C. 


inflatum. 


webaao even 


rs eer ez 


ntae ee meg 
Arbela 
J.Cuatrecasas 


MERBAMO AX 


Campy loneurum 
e 


verreo eTATee 


Epifite, remeer. 


rie 


wuce; Cordillers 

re "inner ve, "vert. Ve 
white, £hO0-e 
“4 vies 


2 


chrysopodum (Ki ot 
Blanca Lede 


ONAL COL 


‘Oe 


1985, 


Ta AE a Mh Liaw 


mana 


Ooeidentels Corre 
- Bors lieve del 


16+ VE 1-1 93S 


WNYNANOTAdN VD ‘AAONITTAT “A “GC 


€Z 


24 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 1 (1988) 


green on both surfaces; midrib stramineous, bearing a few minute scales (less 
than 1 mm long) like those of the rhizome, with some of the scales reduced 
almost to hairs; main lateral veins prominulous on both surfaces, more strongly 
so on the abaxial surface; transverse veins all occult, bearing 2(3) excurrent 
fertile veinlets, the intermediate sterile excurrent veinlet absent, incomplete, or 
complete (especially toward the margin of the frond); marginal free veinlets 
absent; sori ca. 3 mm in diam.; sporangium stalks persistent; paraphyses absent. 
Discussion.—This species has been found in high, montane rainforests in 
thwestern Columbia and northern Peru. It is unusual in having large laminae 
with the stipes about equalling the laminae in length. 
Paratype: PERU. Amazonas: 2—4 km WSW of Pomacocha, 2200—2400 m, Wurdack 868 (US). 


Campyloneurum irregulare Lellinger, sp. nov. (Figs. 6, 12) Type: Ecuador, 
Pcia. Pichincha, vicinity of Quito, 3000 m, Holdridge 1580 (US). 

Ab C. amphostenone in squamis rhizomatis cellulis contortis et ab C. 
densifolio in squamis rhizomatis lanceolatis patentibus differt. 

Rhizomes rather short-creeping, (2)3—5 mm in diam., usually brown, strongly 
pruinose, the phyllopodia 1-3 mm long, 2 mm in diam.; rhizome scales dense, 
spreading, persistent or not, concolorous, medium to dark brown or rarely 
yellowish, lanceolate from a slightly dilated base, 3-5 mm long, 0.5-1.5 mm 
wide, attenuate at the apex, with overlapping auricles at the base or appearing 
peltate, the margins entire, the cells mostly contorted, clathrate with translucent, 
yellowish lumina, those above the base 1—4 times longer than wide. Stipes 
0.25—1 cm distant along the rhizomes, (2)3—15(21) cm long, 1.25—2 mm wide, 
sulcate, the lamina margins prolonged toward the base of the stipe as narrow alae 
or lateral sulcae, stramineous. Laminae linear, attenuate at the apex and base, 
slightly to greatly revolute at the narrowly cartilaginous margins, (6)10—37(50) 
cm long, (0.5)1—2(3) cm wide, pale green adaxially, slightly paler abaxially, 
midrib usually stramineous, sometimes pale green, rarely blackish abaxially, 
bearing a few scales like those of the rhizome but about half as large; main lateral 
veins if present flexuous, poorly or not developed, the areolae in (1)2—3 series 
with 1 or 2 series fertile; marginal free veinlets absent or nearly so; hydathodes 
prominent on the adaxial surface of the laminae; sori 1.5—2.5 mm in diam.; 
sporangium stalks persistent; paraphyses absent. 

Discussion.—This species has a range similar to that of C. densifolium 
(Hieron.) Lellinger, which it also resembles to some extent. The rhizomes of C. 
irregulare are pruinose and bear rather firm, spreading, lanceolate rhizome 
scales with overlapping auricles; the laminae are often narrower than those of C. 
densifolium and less often have conspicuous hydathodes. The plants are mostly 
epiphytic, although they may be terrestrial or epipetric, especially at high 
elevations. 

Paratypes: Costa Rica. Alajuela: Hills of Santiago nr. S. Ramén, 1200-1300 m, Brenes 14209 
(GH). S. José: S. Jerénimo, 1500 m, Wercklé (US). S. José & Cartago: Al Int i H ca 
15-20 km SE by road from El Em ca. 2600 m, Lellinger 1172 (US). Cartago: Cartago, ca. 4750 ft, 
Scamman 6163 (GH). PANAMA. Chiriqui: Volcan de Chiriqui ca. 7.3 mi from Boquete, Armond 533 
(F). Cotomsta. Magdalena: SE slopes of Sierra Nevada de Sta. Marta: basin of R. Donachui: 
Cancurtia, 2400-2650 m, Cuatrecasas & Castaneda 24730 (US). Santander: Vicinity of Vetas, 


WOYNANOTAUN V9) ‘UAONITTAT “A “Cd 


st 7 
(SISOS | 
% 

hp Oe sa 


UNITED STATES maTionAL MyusEuM 


Raeensia’ 


fin 7 ; a 
ial was ¢ BS oy fn. ~£ £6: / : 
. aC ae oe : gor . wah pre Pine biyecinnd Mabe! 

bel. Bianca Leta, 1984 a 8 : 2 Dinan Pippine 
apron ww Su; Foe : ASF es bt f Seale 
6 aaquetifell % . So wine ah at 4, friode £2 a dove Jeaw 
. sensnbein. isacing eto ve ml 


= a 


Fic. 12. Holotype of C. irregulare. Fic. 13. Holotype of C. wacketii. on 


26 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 1 (1988) 


3100-3250 m, Killip & Smith 17258, 17315, 17900 (all GH, US). Boyaca: Sierra Nevada de Cocuy: 
near R. S. Pablin, ca. 3800 m, Grubb & Guymer P15a (US). Cundinamarca: Ca. 5 km SW of Bogoté on 
road to Usme, ca. 2800 m, Smith & Idrobo 1319 (GH, UC, US). Cauca: Alto de Batanes, ca. 2900 m 
Core 892 (US). Narino: Slope of V. Galeras above Pasto, 3300 m, Correll Co484 (UC, US). Ecuanon. 
Carchi: Vic. of El Angel, 10000-12000 ft, June 1921, Popenoe (US). Imbabura: Trail SW of Ibarra, 
2350 m, Mexia 7402 (F, UC). Pichincha: Between La Magdalena and Chillogallo, 2800 m, Firmin 454 
(US). Cotopaxi: First ridge W of Latacunga, 3650 m, Humbles 6289 (UC). Tungurahua: Vic. of 
Ambato, Rose 22370 (US). Chimborazo: Vic. - Huigra, Rose 22270 (GH, US). Loja: Vic. of Tablén de 
Ona, Rose 23114 (US). Peru. Cajamarca: Cerr Cumba Mayo, 3100 m, Sanchez Vega 36 (GH, US). La 
Libertad: Above Cachicadan, 2900 m, Stork & - Hoitan 9969 (F, UC, US). Ancash: Conzuzo, Pampas, 
3800 m, Lépez M. 1158 (US). Lima: R. Blanco, ca. 3600 m, Asplund 11299 (US), 12000 ft, Macbride & 
Featherstone 716 (F). Junin: Huancayo, 3317 m, Soukup 3143 (F, GH, US). Cuzco: Hills of 
Saxaihuamén, 3500 m, Herrera 204 (US), 3600 m, Herrera 2371 (F). Puno: Vic. of Lake Titicaca, 
Moho, 3125 m, Shepard 54 (US). Bottvia. La Paz: Pongo, Tate 106 (US). 


Campyloneurum major (Hieron. ex Hicken) Lellinger, comb. nov.—Polypodium 
phyllitidis f. major Hieron. ex Hicken, Rev. Mus. La Plata 15:272. 
1908.—SynTypEs (all Argentina: Pcia. Misiones): Arroyo Nacanguazii, 
near Puerto Tamaren, 12 Feb 1883, Niederlein (B); Ruins at Candelaria, 20 
Feb 1883, Niederlein (B); and ‘‘Bei der Plantage El Primer Misionero von 
Hernandez, Puck and Fernandez,’ Neiderlein 237 (B). 

Discussion.—This species has been confused with C. phyllitidis, from which 
it differs in having distinctly creeping rhizomes, suborbicular to broadly ovate, 
subappressed rhizome scales, and more complex venation. It is known from 
southern Brazil (coastal Rio de Janeiro, Sado Paulo to Rio Grande do Sul), 
Paraguay, Uruguay, and Argentina. aga poke i pects is absent from 
this region, and all specimens so identified are C. ma 


Representative specimens: Brazit. Rio de Janeiro: Cabo Frio Island, Segadas-Vianna et al. 
“Restinga I’ 743 (US). Sao Paulo: Alto da Serra, Biological Station, 800—900 m, Smith 1891 (US). 
Parana: Jaguariahyva, Dusen 10057 (US), 17354 (US). Sta. Catarina: Blumenau, Haerchen [Ros. Fil. 
Austrobras. Exs. 115] (US). Rio Grande do Sul: Porto Alegre, Lindman [First Regnell Exped. Fil. 
imp (US). UruGuay. Tacuarembé: Gruta Helechos, Herter P]. Urug. 1231 [Herb. Hert. 3739] (GH, 

C, US). Paracuay. Concepcion: Between the R. Apa and the R. Aquidaban, Fiebrig 5079 (GH, US). 
uaira: Colonia Independencia, Lourteig 1980 (US). Misiones; Santiago, Estancia “La Soledad,” 
asics 9548 (UC, US). Paraguari: NE of Cerro Sto. Tomas, near Paraguarf, Balansa 2884 (US). 
GENTIN. i urucuya, Estancia ‘‘Santa Teresa,’’ Pedersen 1803 (US). Misiones: 
oom Ss; fonacia. Salto Tabay.. Crisci 236 (US). 


Campyloneurum remotifolium (Hieron.) Lellinger, comb. nov.—Polypodium 
angustifolium f. remotifolium Hieron., Bot. Jahrb. Syst. 34:531. 1904, as 
“remotifolia.”-“TypE: Colombia, Depto. Tolima, Mt. Ruiz, 3000 m, Ma 
1882, Schmidtchen (B not seen); and Depto. Cauca, Paramo de las Delicias, 
3200—3600 m, Lehmann 4439 (B not seen; isosyntype US). 

Discussion—This species differs from C. angustifolium (Swartz) Fée in 
having fronds about twice as long and with stipes more than twice as long. Itis is 


largely restricted to paramos, unlike C. ang In aspect it 
C. amphostenon, which also grows at — elevations, but is more delicate. 
Representative specimens: VENEZUELA. Mérida: Sra. de la Culata, W flank of Los Adobes, paramo 


Los Conejos, 3500—3650 m, Ruiz-Terdn 6998 og gorwoiegy Valle: Cordillera Occidental, Los 


D. B. LELLINGER: CAMPYLONEURUM 27 


Farallones, 3500—3600 m, Cuatrecasas 17969 (GH, US). Ecuapor. Imbabura: Lago S. Marcos, 
Cayambe, 11200 ft, Cazalet & Pennington 5374 (UC, US). Pichincha: Tambillo, 2700 ft, Mille in 1918 
(US). Azuay: N and NW of Paramo del Castillo, ca. 6—8 km NNE of Sevilla de Oro, 10000-11200 ft, 
Camp E-5150 (US). Napo: C. Cubill4n, 3400-3800 m, Rimbach (Ros. Fil. Ecuad. Exs. 2) (UC, US). 
PERU. Junin: Km. 129-130 of the Concepcién-—Satipo road, ca. 10000 ft, Saunders 1068 (GH). 
Ayacucho: E massif of Cordillera Central opposing the Cordillera Vilcabamba between Tambo S. 
Miguel, Ayna, and Hacienda Luisiana, ca. 3400-3600 m, Dudley 12029 (US). Cuzco: Tres Cruces, 
upper edge of Parque Nac. de Mani, 1-13 km NW of Paucartambo-Pilcopata Road, 3330-3500 m, 
Gentry et al. 23450 (US). Bottvia. Cochabamba: Between Llantas and Aduana, 3100 m, Steinbach 
9557 (UC, GH). Sta. Cruz: Yungas de S. Mateo, Comarapa, 2000 m, Steinbach 8442 (GH). 


Campyloneurum wacketii Lellinger, sp. nov. (Figs. 7, 13)<T pr: Brazil, Est. S. 
Paulo, Rio Grande, Wacket (Ros. Fil. Austrobras. Exs. 213) (US; isotypes 


GH, : 

Ab C. brevifolio (Link) Link et C. coarctato (Kunze) Fée in laminis anguste 
elliptico-oblanceolatis ad basin sensim decrescentibus differt. 

Rhizomes long-creeping, blackish, 3—5 mm in diam., the phyllopodia 2-4 mm 
long, ca. 5mm in diam.; rhizome scales spreading, weakly bicolorous, somewhat 
crisped, dark brown with paler margins, lanceolate (the smaller scales broadly 
so), 1.5-3 mm long, 0.5 mm wide, acute at the apex, peltate at the base, the 
margins entire or slightly repand, the central cells clathrate, with transparent 
lumina, those above the base 2—4 times longer than wide, the marginal cells 
scarcely clathrate, with translucent lumina, often worn away in age. Stipes 
obsolete (1-3 cm long), 2-4 mm wide, distinctly sulcate adaxially, stramineous. 
Laminae narrowly elliptic-oblanceolate, acute or acuminate at the apex, 
attenuate at the base, scarcely revolute at the narrowly cartilaginous margins, 
60—90 cm long, (4)5—6.5(7) cm wide, olive-green and dull adaxially, somewhat 
paler abaxially; midrib olive-stramineous, glaborus and strongly keeled 
abaxially; main lateral veins slightly flexuous, prominulous on both surfaces; 
transverse veins subarcuate, subflexuous, not prominulous, regularly with 2 
fertile excurrent veinlets, the intermediate sterile veinlet absent, partial, or 
complete, especially toward the margins of the frond lamina; marginal free 
veinlets absent; hydathodes minute, often inconspicuous; sori 1-1.5 mm in 
diam.; sporangium stalks persistent; paraphyses absent. 

Discussion.—This species seems closely related to C. brevifolium (Link) Link 
and to C. coarctatum (Kunze) Fée in having rather thick rhizomes and laminae 
that are long-tapered at the base with nearly obsolete stipes. These species differ 
from C. phyllitidis and its allies in having long-creeping rhizomes with 
somewhat distant stipes. Specifically, the rhizome scales of C. wacketii are 
lanceate and their cells are elongate and very narrow, whereas those of Li. 
phyllitidis are ovate and their cells are nearly isodiametric. Their primary 
transverse veins are arcuate, rather than angular or nearly straight. 

Paratypes: Brazit. Rio de Janeiro: Paineiras, Pedra do Beijo, Carauta 285 (F). Sao Paulo: Sa. 
Arariba, Brade 8443 (UC); Iguape, Morro das Pedras, Brade in 1924 (US); Ubatuba, Instituto 
Oceanogréfico, Vélio 80 (US). Parana: Ilha do Mel, Hatschbach & Guimaraes 256880 (UC). Sta. 
Catarina: Joinville, Schmalz 13 (F); Antonio Carlos, Biguassu, Reitz 274 (US); Brusque, Mata 
Hoffmann, Reitz 3091 (US); Brusque, Peterstrasse, Reitz 3166 (US); Itajai, Morro da Ressacada, Reitz 
& Klein 2325 (US); Itajai, Luis Alves, Reitz & Klein 2378 (US); S. Francisco do Sol, Gauiva, Pérto do 
Palmital, Reitz & Klein 5832 (US). 


28 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 1 (1988) 


Campyloneurum wercklei (Christ) Lellinger, comb. nov.—Polypodium 
wercklei Christ, Bull. Herb. Boissier, II. 5:7. 1905.—LEctotTyPE (chosen 
here): Costa Rica, without locality, Wercklé (P-Hb. Christ not seen). The 
other syntype is: Costa Rica, Pcia. S. José, R. Sucio, 800 m, Lehmann 1741 
(B not seen, P-Hb. Christ not seen, US; see discussion below). 

Rhizomes very long-creeping, (1)1.5—3 mm in diam., usually greenish, not 
pruinose, the phyllopodia 1-2 mm long, 2—3 mm wide in diam.; rhizome scales 
dense, subappressed, often not or only the scale bases persistent, slightly 
bicolorous, medium brown at the center but slightly paler toward the margins, 
lanceolate, 1-2 mm long, ca. 0.75 mm wide, subacute to attenuate at the apex, 
peltate at the base, the margins entire or shallowly toothed, the cells not to 
slightly contorted, clathrate with mostly yellowish lumina, those above the base 
2—3 times longer than wide. Stipes 1—4 cm distant along the rhizomes, 4-18 cm 
long, (0.75)1—2(2.5) mm wide, slightly sulcate, stramineous or brownish 
(especially in Ecuadorian specimens). Laminae narrowly elliptic to narrowly 
lanceolate, acuminate to caudate at the apex, acute to acuminate at the base, the 
narrow, cartilaginous margins plane or nearly so, 10-26 cm long, 2—7 cm wide, 
medium green adaxially, paler abaxially; midrib stramineous or pale brown, 
glabrous; main lateral veins straight or slightly flexuous, prominulous; 
transverse veins flexuous, sometimes slightly prominulous, with 2 fertile 
excurrent veinlets in each areola; marginal free veinlets few, short; hydathodes 
prominent on the adaxial surface of the laminae; sori 1.5—2 mm in diam.; 
s ium stalks persistent; paraphyses absent. 

Discussion.—This species was formerly confused with C. sphenodes (Kunze) 
Fée, which I believe is restricted to northeastern Colombia and Venezuela. It 
differs from C. sphenodes and from CG. macrosorum Fée in having appressed, 
subdeciduous rhizome scales and thinner, greenish rhizomes. The Ecuadorian 
specimens, which are disjunct from the Central American ones, differ slightly in 
having generally larger (or at least wider) laminae and rhizome scales that are 
more strongly toothed. It is possible that they represent a separate subspecies. 

Lehmann 1741 was thought by Christ to be sterile material of Polypodium 
wercklei, which it clearly is not, and so it should not be chosen lectotype of this 
species. In order to fix the application of this name and of P. falcoideum Kuhn ex 
Hieron., I choose Lehmann 1741 (US; isolectotypes B not seen, P not seen) as 
lectotype of the latter species. 

Representative specimens: Costa Rica. Guanacaste: Los Ayotes near Tilaran, 600—700 m, 
Standley & Valerio 45407 (US). Puntarenas: 1—4 km SW of biological field station at Finca Wilson, 5 
km S of S. Vito de Java, 1200-1400 m, Mickel 3122 (US). Alajuela: NW of Zarcero, ca. 2 km W of 
Zapote on road to Sta. Elena, ca. 1200 m, Lellinger 1357 (CR, F, MO, US). Heredia: Yerba Buena, NE of 
S. Isidro, ca. 2000 m, Standley & Valerio 49225, 50227, 50251 (all US). S. José: La Palma, 1459 m, 
Tonduz 12609 (US), 1550 m, Tonduz 12424 (US). Cartago: El] Mufeco, 5000 ft, Stork 2670 (UC, US). 

ANAMA. Bocas del Toro: 10—15 mi S from th of Changuinola River, Lewis et al. 994 (GH, K, UC, 

US). Chiriqui: N of S. Felix on C. Colorado copper mine road, 5000—5500 ft, Mori & Kallunki 5860 

(MO, US). Coclé: El Valle, Gonzalez 32 (PMA, US). Ecuapor. Guayas: Riobamba, 2500 m, Rimbach 

85 (UC, US). Pichincha: Mt. Atacazo, 2800 m, Mille 27 (US). Chimborazo: Cafion of R. Chanchan, ca. 

5 km N of Huigra, 5000-6500 ft, Camp E-3365 (US), E~3381 (UC, US). Azuay: Between Cruz Pamba 

and Loma de Canela, region of R. Sadracay, N of Molleturo, 2315-2500 m, Steyermark 52959 (US). 


D. B. LELLINGER: CAMPYLONEURUM 29 
KEY TO CAMPYLONEURUM 


1. Main lateral veins often flexuous, not or scarcely prominulous abaxially, 
borne at ca. a 45—60°(rarely 75°) angle to the midrib, often lacking arcuate 
transverse veinlets; laminae linear or rarely lanceolate or narrowly ob- 
ERC ee ee a ee, 29 

1. Main lateral veins straight and parallel, prominulous abaxially (except in 
C. occultum), borne at ca. a 75° angle to the midrib, always bearing arcuate 
transverse veins with 2—several unbranched, included veinlets; laminae (or 


pinnae) lanceolate, oblanceolate, or broadly to narrowly lanceate .......... 2 
vis) Demnee Ne. ee es 4 
UE tc A es a wie eel es 3 
3(1). Pinnae 2.5—4.5 cm wide; sori 2 between the main lateral veins 

(Martinique, Venezuela, Colombia, Brazil) ...... C. decurrens (Raddi) C. Pres] 
3(1). Pinnae 6—10 cm wide; sori (2)3—4 between the main lateral veins 

(Panama to Venezuela & Bolivia) ................. C. magnificum T. Moore 
4(2). Stipes spaced evenly along the long-creeping to very long-creeping (or 

Varery shor cvoeming) HisOiee oe an ee as ke 13 


4(2). Stipes approximate, clustered near the apex of the cgi hs ae 
PRSROINOS OS rr oe oc eek ce ees 
5(4). Laminae usually less than Gcm wide -.... 6... cco ee ees ; 
5(4). Laminae usually at least 6 cm wide (to 4 cm wide in C. phyllitidis sine 
DOG Ra eee sen eee nay se 
6(5). Venation regular, with 3 free, acroscopic included veinlets, the is 
2 fertile, the middle one sterile and prolonged, sometimes forming a pair of 
areolae; sori in 2(3) regular rows between the main lateral veins (Florida, the 
Bahamas; Mexico to Bolivia; Venezuela to the Guianas and C Brazil) ......... Vv 
COE a Eee oo ee ee Ce PS daa hed we C. phyllitidis (L.) C. Pres] 
(6)5. Venation irregular, with varying included veinlets and areolae; sori in 
2—3(4) irregular rows between the main lateral veins ..............-.++5+ 7 
7(6). Transverse veins not prominulous on the abaxial surface of the 
laminae, often hidden. Laminae subcoriaceous, firm (Greater & Lesser Antilles; 


Mexico to Bolivia; Venezuela to Trinidad & NW Brazil) ....C.latum T. Moore 
7(6). Transverse veins decidedly prominulous on the abaxial surface of the 
TG ee ore wwe ee he ee ee 8 


8(7). Laminae coriaceous, long-decurrent at the base (Ecuador to Bolivia) . . 
Pe ye C. pascoense R. Tryon & A. Tryon 
8(7). Laminae membranaceous-papyraceous, short-decurrent at the base 

A 


(Bolivia & Argontina) i. oo cee es tucumanense (Hieron.) Ching 
9(5). Main lateral veins abaxially not prominulous, straight or not ..... 12 
9(5). Main lateral veins abaxially prominulous We CO go ka 10 


10(9). Stipes ca. 1/4 as long as the laminae; sori ca. 1 mm in diam.; rhizome 
scales slightly bicolorous with a yellowish central area (Costa Rica; Colombia & 
Pras ke C. multipunctatum (Christ) Lellinger 

10(9). Stipes 1/3—1/2 as long as the laminae; sori ca.2 mmindiam ..... 11 


30 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 1 (1988) 


11(10). Laminae repand, lanceate, widest at the middle (Mexico & 


Rs kin os hh es ica co ss ea he We C. tenuipes Maxon 
11(10). Laminae entire, narrowly Epgeciote, widest below the middle 
I kg ws ee thie YD nitidissimum (Mett.) Ching 


12(9). Laminae narrowly oblanceolate, uae at the base, usually 
strongly white-dotted on the adaxial surface, (5)6—10 times longer than wide; 
stipes 1—4(5) cm long (Mexico to Costa Rica) .............. C. xalapense Fée 

12(9). Laminae mostly elliptic to oblong, acute or acuminate at the base, not 
white-dotted on the adaxial surface, (3)4—5(6) times longer than wide; stipes 
(2)4—14 cm long (Florida, Greater Antilles, Mee to Colombia, Venezuela,/ 
Trinidad, the Guianas & C Brazil) .............. ostatum (Kunze) C. Presl 

13(4). Laminae broadly to narrowly paces gee more gradually 
toward the base than the apex) or linear (tapered about equally toward the base 
and apex); stipes usually less than 1/3 as longasthe laminae ............. 19 

13(4). Laminae broadly to narrowly elliptic or elliptic-lanceate, tapered 
about equally at the base and the often caudate apex; stipes usually 1/3—2/3 as 


Oi nh ise Lc a 14 
14(13). Rhizome scales usually weakly bicolorous, pale reddish—brown in 
mass; laminae lanceate, mostly at least 2.5 times longer than wide ........ 17 


14(13). Rhizome scales concolorous, usually tan to golden in mass, 
markedly spreading (brown and readily deciduous and the rhizomes often 
greenish in C. brevifolium); laminae broadly elliptic or elliptic-lanceate, mostly 
2—2.5 times longer than wide (ca. 3 times in C. brevifolium) ............. 15 

15(14). Laminae 2—5 cm wide, 4—14 cm long, especially the adaxial surface 
shiny; sori ca. 1-2 mm in diam. (Costa Rica & Panama) .. C. sublucidum Christ 

15(14). Laminae 8—12 cm wide, 30—40 cm long, the surfaces not shiny . 16 

16(15). Laminae acute at the base; sori ca. 3 mm in diam. (Colombia & Peru) 
did os Va oe eet ee ues cae Gok Eu se C. inflatum M. Meyer ex Lellinger 

16(15). Laminae acuminate at the base; sori ca. 1-2 mm in diam. (Costa Rica 
i Venesucia & Bolivia... C. brevifolium (Link) Link 

17(14). Rhizome scales appressed, usually not persistent, except for some of 
the scale bases, not reddish in mass; rhizomes (1)1.5—3 mm in diam., greenish or 
turning black in age (Costa Rica & Panama; Ecuador) ..................-005 
ghee ee oe C. wercklei (Christ) Lellinger 

17(14). Rhizome scales spreading, persistent, often reddish in mass; 
mene (2)3—4.5 mm in diam., not greenish 

18(17). Rhizomes 2—4 mm in diam., stramineous; rhizome scales 
concolorous, pale reddish-brown in mass, strongly clathrate, slightly iridescent, 
the lumina translucent (Venezuela & Colombia) ... . C. sphenodes (Kunze) Fée 

18(17). Rhizomes (2)3—4.5 mm in diam., dark fe or blackish; rhizome 
scales weakly bicolorous, medium brown, often with paler margins, clathrate, 
not iridescent, the lumina yellowish (W Venezuela & Colombia)... 5 eis 
Fo ee Sr Re amar RS ne Sis Oe ae C. macrosorum Fée 

19(13). Rhizomes very long-creeping, 1.5—2.5(3) mm in diam.; laminae 
herbaceous, flexible 


see 
I a CR A A A A ee ie ae ee i Ue i ale ee ee ea ae ge ge ce ee ae 


D. B. LELLINGER: CAMPYLONEURUM 31 


19(13). Rhizomes 3—9 mm in diam.; laminae papyraceous to subcoriaceous, 
20(19). Laminae acute-acuminate at the base, abruptly tapered to a long, 
narrowly alate stipe, widest near the base. Rhizome scales acicular, blackish 
(Venezuela to Bolivia & Brazil) ................. C. coarctatum (Kunze) Fée 
1 


Eruguay, Paraguay)... ok as C. major (Hieron. ex Hicken) Lellinger 
21(20). Rhizome scales spreading, lanceolate, the central cells longer than 


8 

22(21). Stipes ca. 2 cm long; laminae narrowly elliptic-oblanceolate: 
rhizome scales spreading, the central cells 3—4 times longer than wide (SE 
DPRTA eh GG hg cus cn C. wacketii Lellinger 
22(21). Stipes 4-15 cm long; laminae linear (or rarely narrowly elliptic in C. 
cochense); rhizome scales lax and appressed, the central cells 1-2 times longer 
TM 23 
23(22). Rhizomes 5—8 mm in diam.; main lateral veins prominulous; 
rhizome scales dark brown in mass. Rhizome scales auriculate, with curved, 
overlapping auricles (W Venezuela and Colombia to Ecuador) .............. 
WONDA ee eee CO ays SUNIL Sorry ae C. cochense (Hieron.) Ching 
23(22). Rhizomes 3—-4(6) mm in diam.; main lateral veins usually immersed 

in the lamina tissue; rhizome scales reddish-brown in mass (SE Brazil & 
MON i a ew eenses a un C. fallax Fée 
24(19). Main lateral and transverse veins all hidden in the lamina tissue: 
abaxial surface of the laminae sparsely pilosulous, the hairs pale or reddish, ca. 
0.25 mm long (S Mexico to Venezuela, Bolivia & Brazil) ................... 
«teenage Leeks Cheese ee CUPS sre C. occultum (Christ) L. D. G6mez 
24(19). Main lateral veins and often the transverse veins prominulous; 


abaxial surface of the laminae glabrous ..............0..e0eceeeeeeees 25 
25(24). Apex of the rhizome scales narrow, acicular, the scales usually 
SPYOOUING ic 0 oe ee ee ee. 27 
25(24). Apex of the rhizome scales, wide, obtuse to round, the scales always 
BT OSSO fas eas oe Pa ee, 26 
26(25). Stipes 0-3 cm long; rhizome scale cells ca. isodiametric, not 
comtorted (Peru & Bolivia) .. . 2... 566.5. 6458) C. ophiocaulon (Klotzsch) Fée 


26(25). Stipes 5-15 cm long; rhizome scale cells 2—3 times longer than 
wide, contorted (SE Brazil, Paraguay, NE Argentina) ...................... 
ee Wea a ae ae ee aes a C. herbaceum (Christ in Schwacke) Ching 
27(25). Laminae (2)2.5—.35(5) cm wide, acuminate at the base; rhizome 
scales biauriculate, with an expanded base (Mexico to W Venezuela & Peru) .. . 
a a ee a eo es C. serpentinum (Christ) Ching 
27(25), Laminae (2.5)3—8 cm wide, mostly cuneate at the base; rhizome 
scales peltate, little expanded at the base ......50.0... 000000 e ee eens 28 
28(27). Rhizome scales weakly bicolorous, pale brown, with broad, pale, 


32 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 1 (1988) 


slightly erose-toothed margins, ca. 3 mm long, 0.75—1 mm wide (Jamaica, Puerto 
Rico, Lesser Antilles; S. Mexico to Bolivia; Venezuela to the Guianas & C Brazil) 
ek ee aes A 8 Ue C. repens (Aublet) C. Pres] 
28(27). Rhizome scales essentially concolorous, dark brown, sometimes 
with very narrow, paler, entire margins, ca. 4—5 mm long, 0.5—0.75 mm wide 
regen Giinna: Clemiid (9 BOVia) 4. Gi alee oe 
es ie vase EE Chie Ws cee ae wl eo Se C. fuscosquamatum Lellinger 
29(1). Rhizomes (1)1.5-8 mm in diam., short- to long-creeping; laminae 
lineee Gr rarely Narrowly oplanceclate ........-... 0... ek. 33 
29(1). Rhizomes ca. 1 mm in diam., very long-creeping, often with reddish, 
spreading scales; laminae narrowly lanceate or rarely narrowly lanceolate, acute 
Or spmelinigs ACuMUnaie Al The bese: 2c... i. be ee a eink. 30 


30(29). Stipes 1/4-1/2 aslongasthelaminae ......:...62.-....-.5.. 31 
31(30). Rhizome scales twisted and involute, indistinctly clathrate 
(Hispaniola; Ecuador to Bolivia & Brazil) ...... C. vulpinum (Lindman) Ching 
31(30). Rhizome scales plane, distinctly clathrate .................. 32 
32(31). Rhizome scales lanceolate from a peltate base, the cells of the base 
contorted; stipes 1/4—1/3(1/2) as long as the laminae (Costa Rica & Panama) .... 
ONS Re C. falcoideum (Kuhn ex Hieron.) M. Meyer ex Lellinger 
32(31). Rhizome scales nearly linear from a narrow, peltate base, the cells of 
the base only slightly contorted; stipes (1/3)1/2 as long as the laminae (Venezuela 
Wai... C. chrysopodum (Klotzsch) es 
33(29). Laminae not Vittaria-like, at least 1 cm wide ................ 
33(29). Laminae Vittaria-like, 0.25—0.75(1) cm wide, ca. 30 cm lon Ng Vie a 
34(33). Rhizome scales gray-brown in mass, the cell walls thin, the scales 
strongly clathrate, slightly iridescent (Peru to Argentina & Brazil) ........... 
ee ee ee ee emcees C. aglaolepis (Alston) de la Sota 
34(33). Rhizome scales pale to dark brown in mass, the cell walls thick, the 
scales clathrate, slightly iridescent only in C. centrobrasilianum .......... 35 
35(34). Rhizome scale cells contorted, slightly iridescent (C Brazil) ...... 
pi nek SM wale ie nisi a ead 4 eee ews, ba C. centrobrasilianum Lellinger 
35(34). Rhizome scale cells not contorted, not iridescent ............ 
36(35). Rhizome scales rather broadly ovate and lax, the cells about as long 
as wide. Laminae commonly revolute and curled when dry or at maturity 
(Mexico to El] Salvador) .......... C. ensifolium (Willd.) J. Smith (see note 1) 
36(34). Rhizome scales more or less lanceolate, firm, and spreading, the 
cells ba least twice as long as wide 37 
36). Rhizome scales lanceolate from a peltate base; laminae usually 
AOE revolute when dry or at maturity (iropical America) 2 2..0.00. ec: : 
a ra ee Re ee i ae C. angustifolium (Swartz) Fée 
37(36). Miso scales acicular or lanceolate from a cordate base; laminae 
usually plane and straight when dry or at maturit 38 
38(37). Rhizome scales acicular from a dilated, cordate base with 


OO Me REL Nf a SO eR gh AR A te ae eh ee ae eg ae 


CS EE Oe ee I a OE OR ke a ee 


D. B. LELLINGER: CAMPYLONEURUM 33 


asymmetrically overlapping auricles, ca. 3 cells wide at the midpoint (Peru & 

BouVieh oo ig) 362 C. angustipaleatum (Alston) M. Meyer ex Lellinger 

38(37). Rhizome scales narrowly lanceolate from a cordate base with 

asymmetrically overlapping auricles, ca. 5 cells wide at the midpoint (Peru, 
] 


Bolivia & Brazil}. 3. 2.2. ca. 2c ons C. austrobrasilianum (Alston) de la Sota 
39(33). Rhizome scales firm, spreading, usually narrowly lanceolate; 
mizomes usually pruinose <3 2.2 ee . 44 


39(33). Rhizome scales lax, often appressed or subappressed, ovate (except 
lanceolate in C. cooperi and cubense); rhizomes not pruinose (slightly so in C. 


COO i ice eG ee 40 
40(39).-Laminne firm, shiny, linea? < 200601655 ese. ee Ae 42 
40(39). Laminae flexible, dull, narrowly lanceate or oblanceolate ...... 41 


41(40). Laminae tapered very gradually toward the base, narrowly 
oblanceolate, the main lateral veins slightly sinuous, at a 45—60° angle to the 
midrib; rhizomes 2—3 mm in diam. (Cuba, Jamaica, Hispaniola) ............. 

piey Bi a ae Ca ee ae C. cubense Fée 
41(40). Laminae tapered about equally toward the base and apex, narrowly 
lanceate; main lateral veins flexuous, at ca. a 75° angle to 
2 mm in diam. (Peru to Argentina, Paraguay & SE Brazil) ...............+45. 
Pun ere OE a a C. lapathifolium (Poiret) Ching 

42(40). Laminae medium green, caudate at the apex; rhizomes 
long-creeping. Rhizomes ca. 5 mm in diam., the scales pale brown throughout 
(Venezuela Be Bolivia) 2 2o FE ee C. densifolium (Hieron.) Lellinger 

42(40). Laminae yellow-green, acute to acuminate at the apex; rhizomes 
SUOM-CROUDING fs Sik sac ne CR a  k ig eee ee oe ees 43 

43(42). Laminae acuminate at the base, the stipes 2-6 cm long; rhizomes 
2—4 mm in diam., the scales dark brown with paler margins as Brazi Tapas ea 
Pe ae ere ee Ce cadet Vanes o8eenees rigidum J. Smith 

43(42). Laminae attenuate at the base, the stipes eri or nearly so; 
rhizomes 3—5 mm in diam., the scales medium brown (Costa Rica) ........-. 

ONG e Ss alee ot Ss Sn i vk C. cooperi Lellinger 

44(39). Rhizome scales stramineous (rarely pale brown), not clathrate, very 
long and narrow and filiform at the easily broken apex, the margins sparsely and 
roe serrulate-toothed (Venezuela to Bolivia) ..... C. chlorolepis Alston 

44(39). Rhizome scales pale to dark brown, subclathrate or clathrate ... 45 

45(44). Rhizome scales linear-lanceolate, 5-10 times longer than oade 

mane 2. S(4) warn 1 CAIN 9k kod ie arin eae SA? ewe en be en oe e eee 48 
45(44). Rhizome scales lanceolate or ovate-lanceolate, 2-5 times longer than 
wide; rhizomes 3—5 mm in diam ..... 6... .0 cee ete cee ewe eee eee nees 46 

46(45). Rhizome scales broadly ovate, ca. 2 times longer than wide; laminae 
usually not revolute, the midribs sparsely scaly (Bolivia & Argentina OA) 2 ies 

pe de ee ce es el ee a oo ee es C. lorentzii (Hieron.) Ching 

46(45). Rhizome scales lanceolate, 3—5 times longer than wide; laminae 
often revolute, the midribs usually not scaly ......----++++eseeerrrreee 47 

47(46). Central cells of the rhizome scales nearly straight, occasionally 


34 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 1 (1988) 


contorted (the scales typically 5—7 mm long, 1.5—2 mm wide, concolorous); 
lateral veins prominulous (Cuba, Jamaica, Hispaniola; Mexico to Venezuela & 
UVR ss b,c Oe? C. amphostenon (Kunze ex Klotzsch) Fée 
47(46). Central cells of the rhizome scales strongly contorted; lateral veins 
immersed in the lamina tissue, usually obscure (Costa Rica to Bolivia)........ 
be ey ee a eee Oe a ee C. irregulare Lellinger 


48(45). Rhizome scales peltate, lacking auricles .................... 50 
48(45). Rhizome scales auriculate, the auricles often overlapping and the 
Wenien Berrie Oe pelt he er ee ks ne 49 


49(48). Rhizome scales lanceolate with an expanded base, brown or pale 
brown, the basal auricles decidedly overlapping, the scales often appearin 
peltate (Colombia to Bolivis). 22... ci e ce 2 C. solutum (Klotzsch) Fée 

49(48). Rhizome scales linear to narrowly lanceolate, dark reddish-brown, 
the basal auricles not or only slightly overlapping (Peru & Bolivia)........... 

ee Vee es Ce RN a C. asplundii (C. Chr. in Aspl.) Ching (see note 2) 

50(48). Stipes mostly 1-3 cm long, ca. 0.25 cm distant; rhizome scales 
brown in mass, typically 3-4 mm long, 0.5—0.75 mm wide, usually bicolorous, 
the margins pale (Florida; Greater Antilles: Guadeloupe; Mexico to Bolivia & 
We ae a: C. angustifolium (Swartz) Fée 

50(48). Stipes mostly 8-25 cm long, 1-1.5 cm distant; rhizome scales dark 
reddish-brown in mass. Rhizome scale attachment point very large (W 
Venezuela & Colombia to Peru) .......... C. remotifolium (Hieron.) Lellinger 


Note 1. This name is based on a specimen ostensibly from Peru, collected by Nee (B-Hb. Willd. 
19610). All th i T} fthi : £ ae ee ‘+. of ot j 1 1 


from South America. It is known that the Malespina Expedition stopped in Guatemala and at 
Acapulco, fr here Nee j yed to the Mexi pital. Thus it is possible that the collection is 


J 


mislabelled. 


Note 2. Th li C. leucor. hizon (Klotzsch) Fée may apply to this species if it is lectotypified 
on the one Peruvian syntype. But the name is based on mostly Venezuelan syntypes, and if it is 
lectotypified on a Venezuelan syntype, the name probably does not apply to this central Andean 
species. The name C. asplundii, although later, b lied without doubt to thi i 


2. 


LITERATURE CITED 


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

WAGNER, W. H., JR. and D. R. Farrar. 1976. The Central American fern genus Hyalotricha and its 
family relationships. Syst. Bot. 1:348—362. 


American Fern Journal 78(1): 35-36 (1988) 


Isoétes pallida, a New Species from Mexico 


R. JAMES HICKEY 
Botany Department, Miami University, Oxford, Ohio 45056 


During preparation of the Isoétes treatment for the Flora Mesoamerica it has 
become evident that a collection of Isoétes, from Oaxaca, Mexico, formerly 
recognized as a disjunct population of the Venezuelan I. triangula (Hickey, 1985) 
represents a new species. 


Isoétes pallida Hickey, sp. nov.—T PE: Mexico, State of Oaxaca, 4 mi W of Costa 
Rica, rain ponds along Hwy 190, sandy soil with scattered boulders, 3 Aug 
1965, Kral 25320 (holotype MO!; isotypes MICH! VDB!). 

A I. triangula sporangiis non-pigmentiferis et macrosporis minoribus, a I. 
cubana pustulis megasporarum acutis, numerosioribus differt. 

Corm globose, two- or three-lobed; roots dichotomous. Leaves 20—45, to 300 
mm long, 4.5—8.0 mm wide at the base, 0.5—1.5 mm wide at mid-length, stiffly 
erect; alae 30—80 mm long [15—22 (37)% of the total leaf length], 1-1.5 mm wide 
at the sporangium, chartaceous and nearly transparent, each apex attenuate; 
subula trigonal, stramineous to bright green, the apex attenuate; peripheral 
fibrous bundles distinct; stomates present; scale leaves present, phyllopodia 
absent. Sporangia 3.6—4.0 mm long, 2.5—4.0 mm wide, ovate, tan, unspotted, 
basal. Vela absent. Ligules not seen. Labia 0.9—1.5 mm high, 1.0—1.2 mm wide, 
depressed-ovate to widely depressed-ovate, submembranaceous, tan, 
erose-entire to mucronate, the mucro to 0.45 mm long. Megaspores 390-460 
(x=426) pm diam., white, pustulate, with 70 or more pustules on the distal 
surface, these subacute and evenly distributed and equally well developed on all 
surfaces, rarely anastomosing; equatorial and proximal ridges distinct, 
triangular in cross-section, the equatorial ridges rarely scalloped; microspores 
not seen. 2n= 44. 

Paratype: Mexico. Oaxaca, Mpio. Miltepec, 11.8 mi NW of Zanatepec along Hwy 190, dominant 
plant in shallow rain pools in open ‘savanna’ vegetation, soil a thick gray clay, 19 Oct 1986, Hickey & 
Russell 962 (GH, MU, NY). 

Isoétes pallida resembles members of the I. triangula species complex of 
central Venezuela, northern Brazil, and Guyana, but differs in having an 
unpigmented sporangium and somewhat smaller megaspores (perhaps a 
reflection of ploidy differences). It is also extremely close, morphologically, to I. 
cubana of Cuba, Belize, and the Yucatan peninsula, Mexico. It differs from that 
species in ch ber (I. cubana is a diploid with 2n = 22), the number 
of pustules per megaspore, and slight differences in the shape of the pustules, 
which are more rounded in I. cubana (Figs. 1 and 2). Preliminary allozyme data 
(Hickey, unpublished) support a close evolutionary relationship between I. 
cubana and I. pallida. Both of these species belong to a larger complex, the I. 
panamensis alliance, which is characterized by distinctly trigonal (sharply 
3-angled) subulae and prominent scales or phyllopodia. The panamensis 
alliance is a large, and as yet poorly understood group found from Paraguay to 


36 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 1 (1988) 


Fics. 1 AND 2. Scanning electron micrographs of Isoétes megaspore surface morphology. Fie. 1. I. 
cubana, Wright s.n. (MO). Fic. 2. I. pallida, Kral 25320 (MICH). Bar = 200 um. 


southeastern and western North America. The group includes several widely 
dispersed and well known species such as I. butleri, I. nuttallii, I. orcuttii, I. 
melanopoda, I. flaccida, I. howellii, and I. mexicana. 

Isoétes pallida grows in exposed, low scrub habitat of southwestern Mexico 
very close to the Oaxaca-Chiapas border. To date, only two collections have been 
made, both probably from the same locality. The paratype material was collected 
from a small ephemeral rain pool approximately 30 feet in circumference. The 
maximum water depth of the pool at the time of collection was 12 inches but had 
obviously dropped considerably in the previous weeks. Plants were found in 
equal abundance on the muddy shore and in the deepest parts of the pool. It 
seems likely that at least some of the plants were completely submerged before 
the water level dropped. 

This species, like many other tropical Isoétes, is strongly seasonal. The Kral 
collection, made in August, contains numerous, well-developed, intact 
megasporangia. The October collection of Hickey and Russell contains only 
sporangial remnants and scattered megaspores indicating that the aesitvation 
process had commenced. In addition, when split lengthwise, specimens of the 
October collection showed that the apical meristem had already begun initiating 
a series of scales. Such scales are hypothesized to function as protective organs 
for the meristem during periods of drought and aestivation (Hickey, 1986a). 

This research supported by National Science Foundation Grant BSR 860672. 

LITERATURE CITED 
Hickey, R. J. 1985. Revisionary studies of neotropical Isoétes. Ph.D. dissertation. The University of 
Connecticut, Storrs. 
- 1986a. The early evolutionary and morphological diversity of Isoétes, with descriptions of 
two new neotropical species. Syst. Bot. 11:309-321. 
86b. Isoétes megaspore surface morphology: nomenclature, variation, and systematic 


importance. Amer. Fern J. 76:1—16 


INFORMATION FOR AUTHORS 


Authors are encouraged to submit manuscripts Pertinent 1 to pteridology for 
publication in the American Fern Journal. } t to the Edi- 
tor. 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 ac- 
ceptability of manuscripts nor the date of publication. 

Authors should adhere to the following guidelines; ia not so O pre- 
pared may be returned for revision prior to review. Subm 
cate (xerocopies acceptable), including review copies af illustrations. Do not 
send originals of illustrations until they are requested. Use standard 81/2 by 11 
inch paper of good quality, not ‘‘erasable” paper. Double space manuscripts 
throughout, including title, authors’ names and addresses, text (including heads 
and keys), literature cited, tables (separate from Paice and figure captions 


uscript in order just given. Include author’ s name and page number i in 1 upper 
tlea 


right corner of every shee j st 25 mmal 
- a a eentiy and Sipe not break words at wea of lines. Make table head- 
ings and figure caption Use S.I. (metric) units for all measures 


(e.g., distanee Satan gli unless quoted or cited from another source 
(e.g., specimen citations). For nomenclatural matter (i.e., synonymy and typi- 
fication), use one paragraph per basionym (see Regnum Veg. 58:39—40. 1968). 
Abbreviate titles of serial publications according to nner eae baiiwcortadake 
Huntianum (Lawrence, G. H. M. etal., 1968, Pit 
References cited only as part of nomenclatural matter are not eed in litera- 
ture cited. For shorter notes and reviews, put all references parenthetically in 
text. Use Index herbariorum (Regnum Veg. 106:1—452. 1981) for designations of 
herbaria. 


> ae a: * Fel ee & : +i, ™ 
ld d to fit Pape wittil 


page. Provide margins of at least 25 mm onall 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. Coordi- 
nate sequence and numbering | of ‘figures (and of tables) with order of citation in 


text Explain y ig cap tions. I Include a 
scale and refer to latitud 5 tn ouch aint : 
Proofs and reprint order nas 6 are sett t to authors by the haa Authors 
should send corrected ioe to the editor and p 3 the printer. Au- 
thors will be re ade aft type h + ten : 
For other ff ty] lt sf American Fer Jour- ea 
nal and The chicas manual of style, 13th ed. (1982. Chicago: ope ce 
).O pa idelines may be justified. : ee 
nco 1 ult the editor for assistance then ape! ofan : 


se bed oan ay beth oro Pea ‘ ia 


American oe 
Fern Number 2 

April-June 1988 
Journal 


QUARTERLY JOURNAL OF THE AMERICAN FERN SOCIETY 


fa4n4sc 400c) «-f DL-~- lin L Beoont Wl, 


York State Diane M. Cinquemani, Mildred E. Faust, and Donald J. Leopold 37 


£. Pa 3 Le ac £41. 


Gerald J. Gastony a4 


TL. Doll Leh nile 7 Cl * pmo 
SiC £ 


Agamosporous Taxa anda Revised Taxonom 


Marsilea scalaripes, A wT RA L £ Ag. *3 a: or y c. sh 
Asian Tropics : David M. Johnson 68 
Shorter Notes 
Rediscovery of Gy pium dryopteris in Arizona Paul F. Boucher 71 
wh Cs. a f Aspl ry Seisio t Oreg Peter F. Zika 72 
Seven Clubmosses New to Arkansas James H. Peck and Carol J. Peck 73 
Reviews 75, 76 


Information for Authors Cover 3 


The American Fer! Fern Society . 


JUDITH E. SKOG, Biology Dept., George Mason University, Fairfax, ne pees President 
DAVID B. LELLINGER, Smithsonian Institution, Washington, Dz. sn ep Gaaneis 
W. CARL TAYLOR, Mil Milwaukee, WI5 Secretary 
JAMES D. CAPONETTI, Dept. of Botany, aroma of Tennessee, ae T™N sap pests 
DAVID S Dept. of Botany, University of Vermont, Burlington, VT 054 

bots Treasurer 
JAMES D. MONTGOMERY, Ecology Ill, R.D. 1, Berwick, PA 1860 Bie Issues Curator 
ALAN R. SMITH, Dept. of Botany, wserassninse, of California, a i 94720. Journal Editor 
DAVID B. LELLINGER Washington, = - 2056) Memoir Editor 
JOHN T. MICKEL, New York Botanical Garden, Bronx, NY 10458. — iddlehead Forum Editor 


American Fern Soe 
EDITOR 


PA ee i ee Dept. of Botany, University of California, 
Berkeley, CA 94720 
ASSOCIATE EDITORS 
GERALDJ.GASTONY ............ Dept. of Biology, Indiana University, Bloomington, IN 47401 
RAI ee PL i ek ss Dept. of Botany, University of ine 
Lawrence, KS 6604 
PEt U.S. Nat’l Herbarium NHB-166, soointiwoey ae, 
ashin 


8 ngton, DC 20560 

TERRY R. WEBSTER ..... Biological Sciences sage University of se Storrs, CT 06268 

oe ‘ TL, t 1 (ISSN 0002. RAAA a1 PRN Wa 7 3 hg Ba el ] 

Hees of ferns. It i 2ed by the American Fern Society, and ebhies at the Pringle Herbarium, 

University of dears eo VT 05405-0086. Second-class postage paid at Burlington, VT, 
a. entry poin 

ac eo ee 22 a oe res ive oe fs 
nd orders for b hould be addressed to Dr. James D. Montgomery, Ecology III, R.D. 1, 


Berwick, PA 18603. 
e — of — dues, and applications for membership should be sent to the Records 
reasurer 
Sibecitons $20.00 gross, $19. 50 net if paid through an agency oe fee $0.50); sent free to 
arpa : the American Fe Fern Society ees Lees $15.00 + $4.00 mailing surcharge beyond 
Back ee | 1910-1978 $5. 00 to $6. - ch single back numbers of 64 pages or less, $1.25; 
65 Bes, k volumes 1979 et seq. ae: 00 


L L 


co 00 pelt 4 Le 
or : £ se oO Ss 4 heedinteeaaens 
ae POSTMASTER: Send address changes to A FERN J , Dept. of Botany, University of 
: —— Burlington, VT 05405-0086. 
iy =~ ‘ ions f and non-members, including miscellaneous 
ers to exchange or "purchase materials, "pemsonali, hot cultural notes, and reviews s of 
-technical books on 

ao - Mr. Neil D Hall 1230 Nor stile WA 98115 .. Piinartar a tne a 

Tats of available spores sent on request. . ee 
ie drs cay ices to members and to others 


American Fern Journal 78(2): 37—43 (1988) 


Periodic Censuses (1916-1986) of Phyllitis 
scolopendrium var. americana in Central 
New York State 


DIANE M. CINQUEMANI, MILDRED E. Faust, and DONALD J. LEOPOLD 


1 c 


State University of New York, Col tal Science and Forestry 


Syracuse, New York 13210 


This report presents seventy years of census data for the central New York 
populations of Phyllitis scolopendrium (L.) Newman var. americana Fern., the 
American hart’s-tongue. Phyllitis scolopendrium was first discovered in North 
America near Syracuse, New York, in 1807 (Maxon, 1900). Currently, 
populations occur in two rather distinct areas: along the Niagara Escarpment in 
Ontario (Soper, 1954), on the Upper Peninsula of Michigan (Futyma, 1980), and 
in central New York (Faust, 1969); and in sinkholes in eastern Tennessee 
(McGilliard, 1936) and northern Alabama (Short, 1979). Phyllitis 
scolopendrium is listed as endangered in each state where it occurs; it is 
currently under review for listing as threatened or endangered in the United 
States by the U.S. Fish and Wildlife Service. 

The largest number of P. scolopendrium individuals in the United States 
occurs in the combined populations within central New York, in Onondaga and 
Madison counties. There are currently about 15 locations for P. scolopendrium 
in New York (Faust, 1960). Six locations, or substations, occur at Clark 
Reservation State Park in Jamesville, New York. The substations at Clark 
Reservation are particularly important because most of the P. scolopendrium in 
New York exist here; in 1966, approximately 96% of all mature P. scolopendrium 
in New York (1228 total) occurred here (Faust, 1969). 

The substations at Clark Reservation are also unique in that a census of the 
population at each substation was begun in 1916 by M. Hunter (1922). 
Populations at each substation were censused again in 1920 by Hunter. The next 
census was done in 1936 by M. Faust with assistance from the Syracuse 
Botanical Club and botany graduate students at Syracuse University. Since 
1936, the population at each substation at Clark Reservation has been 
inventoried about every five years. The last census that was reported in the 
scientific literature was made in 1956 (Faust, 1960). The objective of this paper 
is to report results of inventories made about every five years since 1956, 
including the most recent census, undertaken by the first author in 1986. 
Seventy years of monitoring data are also examined in relation to potential 
habitat requi ts and tion measures. 


Stupy AREA 


Clark Reservation State Park is located in north-central Onondaga county, 

about 8 km southeast of Syracuse. In 1915, Mary Clark Thompson gave to the 

state of New York 43.7 ha as a memorial for her father, Myron H. Clark, who was 
MISSOURI BOTANICAL 


38 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 2 (1988) 


governor of New York in 1855. In 1926, Clark Reservation was established as a 
state park and was acquired by the New York State Museum as part of a 
preservation program for notable geologic features. Today, the park covers about 
121.4 ha over a landscape that varies from gently rolling, to extremely steep 
grades in the plunge basins and glacial channels. Elevations range from 174 to 
235 m. There are over 400 species of vascular plants within the park, including 
about 30 fern species (Egler, 1943; Faust, unpublished). 


DESCRIPTIONS OF SUBSTATIONS 


The physical characteristics of each substation are probably the dominant 
influence on the extent, size, and vigor of each P. scolopendrium population. 
Therefore, a brief description of each substation follows. This species is 
generally restricted to northeast-facing slopes on talus. Locations of particular 
substations at Clark Reservation can be furnished upon a request to the authors. 

Substation I is enclosed on three sides by ravine walls; there is no wall to the 
east. Because of this opening, the ravine is less protected from the wind and 
slopes tend to be dry. The slope varies from 65 to 70% where P. scolopendrium is 
found; the aspect ranges from N 28° W to S 71° E. Mature, vigorous plants are 
often in small clumps beneath canopy openings. Loose talus covers the slopes, 
making this site susceptible to rockslide damage. 

Substation II is enclosed on all sides by ravine walls, the east and west walls 
are lower in height than the north and south, and the south wall is much higher 
than the north. The south and west slopes remain moist throughout the year. P. 
scolopendrium is found on aspects ranging from N 06° W to S 77° E and slopes 
from 55 to 74%; approximately 10 mature and immature plants are found ona S 
10° E aspect and 71% slope. Most P. scolopendrium are found at midslope; the 
upper slope consists of a rock outcrop and the lower slope is overgrown with 
Tsuga canadensis and Taxus canadensis. 

Substation III, the largest area occupied by P. scolopendrium in Clark 
Reservation, also has the largest population in New York. It is enclosed entirely 
by ravine walls: the southwest-facing wall, in contrast to many of the other 
ravines, is lower in height than the northeast-facing slope on which P. 
scolopendrium occurs. This latter slope remains moist throughout the year. 
Individuals are both clumped under small canopy openings, or scattered 
throughout the ravine beneath the closed canopy. The slope ranges from 45 to 
68%, and aspects from N 24° E to S 83° E, where P. scolopendrium is located on 
mid to lower slope positions. Where Taxus canadensis is located on the lower 
slope, P. scolopendrium is found at midslope only; the upper slope is mostly 
rock outcrops. 

Substation IV is a relatively dry ravine, and is formed by north- and 
south-facing slopes. There is no barrier to the winds from the east and west, 
which tends to dry the slopes. Phyllitis scolopendrium is found at midslope 
positions at aspects that vary from N 07° W to N 17° W, and slopes from 52 to 
72%. The upper slope consists of a rock outcrop; the lower slope is covered by 
Taxus canadensis. 


CINQUEMANIET AL.: PHYLLITIS 39 


Substation V supports the most vigorous individuals of P. scolopendrium in 
Clark Reservation, and also the largest number of ferns per unit area. This 
population is located within a glacial plunge basin. Slopes supporting the fern 
vary from 59 to 64%, and aspects from N 61° E to S 86° E. Phyllitis 
scolopendrium is found at midslope, in one very large patch, under a canopy 
opening. Various herbaceous species form a subcanopy over 1.0 m high above 
the ferns in the summer; this feature is not as pronounced in other substations in 
Clark Reservation. High humidity is maintained throughout the year in this 
ravine; the basin floor soils usually remain moist. 

Substation VI has contained the smallest number of P. scolopendrium in Clark 
Reservation since censusing began. This is the driest substation, mainly because 
it faces Little Green Lake and not another ravine wall as do the other substations. 
This feature leaves the site very exposed to wind and subsequent drying. The 
fronds of P. scolopendrium at this substation are usually small and not lustrous, 
compared to the more vigorous individuals at other substations. The ferns occur 
at midslope, over an area of about 8 m’, making this the smallest substation of 
those in Clark Reservation. The slope is 66% and has an aspect of N 50° E. The 
ferns are found almost exclusively in the deep crevices between boulders, or 
small cave-like pockets within the slope, both of which allow protection from 
the wind, and maintenance of constant high humidity. 


METHODS 


The 1986 census of P. scolopendrium populations in Clark Reservation State 
Park was done according to the sweep method used in previous studies (Faust, 
1960). Under this method, volunteers begin at the bottom of a slope on which the 
ferns are growing and slowly walk up the slope, counting individual ferns as 
they are seen. The sweep is initiated at the bottom of the slope since P. 
scolopendrium plants are sometimes hidden in crevices on the lower sides of 
rocks. Censusing requires careful coordination among volunteers so that no 
ferns are overlooked, or counted more than once. Because P. scolopendrium is 
evergreen, spring and fall censusing is ideal as herbaceous species do not cover 
the individual ferns during these periods. Leaf litter is generally moved aside 
during the sweep to reveal any covered ferns. 

Each fern encountered during the sweep is placed in one of three categories: 
(1) mature (fronds usually more than 15.0 cm long; mature sporangia present on 
the adaxial surface of the fronds); (2) immature (fronds usually between 2.5 and 
15.0 cm; sporangia absent); or (3) sporelings (fronds less than 2.5 cm long). 


RESULTS 


Census data (Fig. 1) document an increase in mature and immature 
individuals at all sites, since 1966. However, prior to 1966, there was a decrease 
in the mature and immature P. scolopendrium. The three least protected (i.e., to 
wind dessication and abrupt climatic events or extremes) substations, 1, IV, and 
VI, all show large decreases in their populations during the mid-1950s and 


40 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 2 (1988) 


4007 SUBSTATION | 500 7 SUBSTATION IV 
400 -« 
300 | 
300 ad 
200 « 
200 af 
100 - 
100 
0 0 
1900 1920 1940 1960 1980 1900 1920 1940 1960 1980 
e 1000 SUBSTATION II 8007 SUBSTATION V 
s ~” 
Beit all 
aod 
>m 8004 
= 400 - 
Oo 0- 
8 
200 4 
200 ~ 
= 
pao 
= 0 
1900 1920 1940 1960 1980 1900 1920 1940 1960 1980 
1200 - 100 - 
SUBSTATION III SUBSTATION VI 
1000 80 - 
800 = 
60 ~ 
600 4 
40 4 
400 
200 a 
0 
1900 1920 1940 1960 1980 1900 1920 1940 1960 1980 
Year 


Fic. 1. Census data (1916-1986) for mature and immature (CJ) Phyllitis scolopendrium var. 
americana and sporelings (®) in central New York. Note the variations in scale for number of 
individuals. 


mid-1960s; the moister substations, III and V, show smaller decreases during 
these years. Substation II, which is also moist, decreased greatly in population 
size during this period. 

Between 1956 and 1962, the number of mature and immature ferns at 
substation I dropped 89% from 287 to 32: from 1967 to 1986, the population 
increased 502% from 51 to 307 mature and immature individuals (Fig. 1). The 
population at substation II had a 255% increase from 1952 (272 mature and 


CINQUEMANIET AL.: PHYLLITIS 41 


immature plants) to 1956 (965), and then a 93% decrease by 1961 (68). Between 
1967 and 1986, this population increased 67%, from 215 to 358 mature and 
immature plants. Substation IV had 175 mature and immature ferns in 1956; in 
the following census, 1962, the population had decreased 65% to 61 mature and 
immature. In 1967, 51 plants were reported. A 259% increase in this population 
from the 1967 data occurred by the 1986 census. The population at substation VI 
decreased by 89% from 1952 (84 mature and immature fern) to 1957 (9). By 1961, 
this population had increased 578% to 61 mature and immature fern. By 1966, 
only 13 individuals, a 79% decrease, were found; in 1986, 26 were reported. 
Substations III and V, the most mesic ones, had smaller population decreases 
during the mid-1950s and 1960s. Substation III data show a 47% decrease 
beginning with the 1952 census (699 mature and immature fern) and continuing 
to the 1956 census (372). This population increased 19% from 1956 to 1967 (444 
mature and immature fern), and 84% from 1967 to 1976 (818); as of 1986, 1089 
mature and immature ferns (a 33% increase over 1976 data) were reported. 
Substation V, which contains the most vigorous ferns and the greatest density, 
had a 33% decrease in population size between 1951 (553 mature and immature 
plants) and 1957 (370), and a 46% increase between 1957 and 1961 (540). A51% 
decrease occurred from 316 mature and immature plants in 1971, to 156 in 1977. 
By 1986, the population size had increased 322% from the 1977 data. 
Sporeling counts vary considerably from census to census, and may not 
indicate accurate trends. However, sporeling counts are often positively 
correlated to mature and immature fern counts at the same substation. 


DISCUSSION 


Sporeling counts, although somewhat useful in indicating favorable 
conditions for germination, are not included in the discussion of site totals 
because of the potentially high error involved in censusing this size-class; they 
are easily missed by those doing the censusing. 

Variations in census figures are most likely attributable to historical/climatic 
events in a given year. For example, a possible cause of the 89% reduction in 
population size of substation | during the mid-1950s and mid-1960s is the 
drought that occurred in central New York and the northeast U.S. during this 
period (Namias, 1966). Since substation I is one of the drier ravines, any dry 
period would be expected to affect the area. Populations in all other ravines also 
decreased during this period. The drier substations (I, IV, and VI) had larger 
percent decreases than the moister substations (III and. V). Substation II, 
although moist, had a large percent decrease at this time. 

Substations I, IV, and VI, which are more exposed to wind than substations II, 
II. and V, contain the smaller P. scolopendrium populations. The individuals at 
these former locations tend to have smaller and fewer fronds than those at the 
moist substations, and often are found in rock crevices, where the loss of 
moisture to wind is lessened. Substation VI, for example, the driest of those in 
Clark Reservation, has the least vigorous and fewest number of P. 
scolopendrium; almost all are found in rock crevices. The population size 


42 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 2 (1988) 


remains small here, existing only under a canopy opening which lets in light and 
precipitation, enabling growth. Adjacent forested areas do not support growth of 
this fern. 

Growth of P. scolopendrium in the moister substations, II, III, and V, does not 
seem to be as limited as in the drier sustations. Substation V, for example, 
contains tk jividuals in Clark Reservation. Site characteristics 
of this basin that maintain constant humidity are protection from the wind by 
ravine walls, an herbaceous subcanopy that maintains moisture at ground level, 
and a basin floor which remains moist throughout the year. This population 
showed the smallest decrease (33%) in the number of mature and immature ferns 
during the mid-1950s. From 1957 to 1966, the population increased 56%, while 
most other substations decreased. 

Conifer cover seems to affect the growth of P. scolopendrium; the fern is rarely 
found beneath conifers. The effect that Tsuga canadensis has on soil and 
vegetation has been well-documented (Daubenmire, 1931; Lewin, 1975; Rogers, 
1978). Whether T. canadensis has much of an effect on lowering soil pH which 
leads to cation depletion on soils derived from limestone bedrock is not known. 
The dense canopy of T. canadensis greatly reduces the limited light on the 
north-facing slopes to which the fern is restricted. The amount of snow that 
accumulates under conifer canopies as compared to adjacent hardwood areas 
may also be an important factor. The soils exposed under those conifers tend to 
freeze during the winter months; adjacent soils, under hardwood canopies, 
insulated by deep snow cover, do not freeze. The ability of P. scolopendrium to 
obtain moisture from the soil during the winter may be affected by the amount of 
snow insulation. 

In 1925, about 400 mature P. scolopendrium were transplanted to Substation 
IV from a nearby site that was soon to be quarried for limestone. Although these 
transplants initially flourished (Faust, unpublished field notes), many did not 
survive beyond the early 1930’s. If recovery plans are eventually developed for 
this species, this experience suggests that site conditions for this fern are very 
specific. We are currently trying to assess habitat variables to which this species 
is especially sensitive. 

In summary, groupings of P. scolopendrium are found under canopy openings 
where more sunlight and precipitation reach the ground. In adjacent canopied 
areas, ferns are scattered. Rock crevices and small cave-like openings within the 
slope help to maintain a constant humidity near ground level. Humidity is 
maintained over entire slopes by the arrangement of ravine walls, which 
prevents drying by the wind. The largest Clark Reservation populations are 
found in humid, well-enclosed ravines. Fluctuations in population numbers in 
these areas due to extreme environmental conditions, such as drought, are 
smaller than in less humid ravines. Populations at all substations have 
increased in recent years, probably due to a lack of extreme climatic conditions. 

After further habitat study is completed here, conservation measures may be 
suggested to maintain P. scolopendrium populations at current levels or 
possibly even increase population size. One method might be to prevent an 
Increase in conifer cover by the removal of conifer saplings. This may likewise 


CINQUEMANTIET AL.: PHYLLITIS 43 


be done with hardwood saplings, to maintain canopy openings. By developing 
propagation techniques, population numbers may be increased. Unwanted 
visitation to the Clark Reservation populations does not seem to be a problem for 
the survival of P. scolopendrium since the ravines in which it is found are 
difficult to access. 


ACKNOWLEDGMENTS 


We thank all of the individuals who have assisted in making the ay a of these 
ee =o volunteers from the Council of Park Friends. Ww ank S. Cle mants, - 
Gilm n. C Par oo % 8 fal? 

Ww. Pipe: ay his technical assistance. 


LITERATURE CITED 


ee iher er R. F. 1931. agp favoring the persistence of eastern hemlock in Indiana. Butler 
Univ. Bot Stud. 2 Zi 

EGLER, F. E. a 1 fl i lant Clark R tion, Onondaga County, New 
York. Se York Sag College of Forestry Tech. Publ. No. 61, N.Y.S. College of Forestry, 
Syracuse, N.Y., 

Faust, M. hs fi 960. mead of Hart’s-tongue fern in Central New pinied Amer. Fern J. 50:55-62. 

—_—_. . Conservation of the Hart’s-tongue fern in North Amer a. Biol. Cons. 1:256—257. 

— R = 1980. The distribution and ecology of Phyllitis Seeaanpanee in Michigan. Amer. 

se 70: 81-87. 


922. t stat ] in New York State. Amer. J. Bot. 9:28-36. 
LEWIN, D.C. is poe The. vegetation of the ravines of the southern Finger Lakes, New York region. 
r. Midl. Naturalist 91:315—342. 
Maxon, W. wk 1900. On the occurrence of the Hart’s-tongue in America. Pp. 30—46 in Fernwort 
Papers, ed. M. Slosson. Binghamton, New York: Willard N. Clute & Co. 
McGILLIARD, E. 1936. The Hart’s-tongue in Tennessee 1878-1935. Amer. Fern J. 26:1 
Namias, J. 1966. Nature and possible causes of the Northeastern United States are gi 
962 


ey Gao 1 4s Vetsord ¢, it 


Rocerrs, R. $. 1978. Forests dominated by hemlock (T: 
and past settlement history. Canad. J. Bot. = 843-854. 

SHORT, J. W. 1979. Phyllitis scolopendrium newly d ered in Alabama. Amer. Fern J. 69:47-48. 

Soper, J. H. 1954. The Hart’s-tongue fern in Ontario. es Fern J. 44:129-147 


American Fern Journal 78(2): 44—67 (1988) 


The Pellaea glabella Complex: Electrophoretic 
Evidence for the 
Derivations of the Agamosporous Taxa and 
a Revised Taxonomy 


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


Pellaea glabella Mett. ex Kuhn is a species of small ferns usually occurring in 
crevices of limestone rocks and cliffs across the United States and southern 
Canada. This species has been considered a complex of three varieties (Tryon, 
1957; Tryon & Britton, 1958), one reproducing sexually and the other - 
agamosporously (i.e., sp spores an 
initiate new sporophytes without: fertilization): ‘Pellaea glabella var. 
occidentalis (E. E. Nelson) Butters comprises sexual diploid plants from the west 
central United States and Canada (Fig. 1,OCC). Variety simplex Butters consists 
of agamosporous tetraploids in the western U.S. and Canada (Fig. 1, SIM). 
Variety glabella consists of an agamosporous tetraploid race in the eastern and 
east central United States and adjacent Canada (Fig. 1, GLAa) and a more 
recently discovered sexual diploid race (Wagner et al., 1965) known only from 
Missouri (Fig. 1,GLAs). Pellaea atropurpurea (L.) Link is a broadly distributed 
agamosporous triploid species that has been proposed as one progenitor of the 
agamosporous taxa of P. glabella. Its distribution overlaps much of the range of 
the P. glabella complex, stretching from northern Guatemala, through Mexico, 
the west-central to eastern United States, and southeastern Canada, with 
disjunct stations in southern Alberta and British Columbia and in northern 
Saskatchewan (Tryon, 1957, 1972; Rigby, 1968). 

Several hypotheses have sought to explain the evolutionary origins of the 
tetraploid agamosporous varieties of P. glabella. ‘Tryon (1957) proposed that 
they arose as all rom ferti of a haploid egg of sexual 
P. glabella var. occidentalis by the triploid sperm of agamosporous P. 
atropurpurea [agamosporous taxa can act as male parents in crosses with 
archegoniate gametophytes of sexual taxa, and in such cases agamospory is 
always inherited (Walker, 1962)]. She hypothesized that one such hybridization 
could have given rise to var. glabella in the east and another to var. simplex in the 
west. Lellinger (1985) accepted this hypothesis for the origin of var. simplex 
(which he recognized as an independent species, P. suksdorfiana Butters). 
Alternatively, Tryon hypothesized (1957, p. 140) that they arose as 
autopolyploids (not involving interspecific hybridization) from sexual diploid 
var. occidentalis. Several years later, Wagner et al. (1965) discovered a sexual 
diploid race of Pellaea glabella var. glabella in Missouri and found it 
indistinguishable from agamosporous tetraploid var. glabella except in its 
chromosome number and number of spores per sporangium. They concluded 
that this new sexual race made unlikely the origin of agamosporous var. glabella 


G. J. GASTONY: PELLAEA GLABELLA COMPLEX 45 


Fic. 1. Geographic ranges of the four taxa in the Pellaea glabella complex. SIM = P. glabella var. 
simplex, OCC = P. glabella var. occidentalis, GLAa = agamosporous race of P. glabella var. gla- 
bella, GLAs = sexual race of P. glabella var. glabella. 


as a hybrid between P. atropurpurea and P. glabella var. occidentalis. Instead, 
they suggested that the agamosporous tetraploid was more likely an 
autotetraploid derivative of the sexual diploid of the same variety. The latter 
view was accepted by Lellinger (1985). Several other hypotheses are made 
possible by the presence of two sexual taxa in this complex. Pellaea 
atropurpurea might have sired the agamosporous Faces through hybridization 
with the respective sexual entities, or the sexual entities might have initiated 
respective agamosporous varieties via autopolyploidy, or the agamosporous 
varieties might have arisen via crosses between variant genotypes of the two 
sexual entities. Rigby and Britton (1970) stressed the need for a more thorough 
biosystematic study of these taxa in order to clarify the evolutionary 
relationships among them. 


; ae 1 * f£ 1 age Geer re a j 4 -2 


of polyploids because genes coding isozymes in progenitor taxa are additive in 
offspring resulting from crosses and because alleles coding allozymes are 
expressed co-dominantly (Roose & Gottlieb, 1976; Gottlieb, 1982; Haufler et al., 
1985: Werth et al., 1985; Gastony, 1986). Enzyme variants coded by the genes of 


46 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 2 (1988) 


progenitor taxa can therefore be used as markers to trace the ancestry of phyletic 
derivatives. This paper uses taxon-specific banding patterns of six polymorphic 
enzymes to test both the hypothesized role of P. atropurpurea in the origin of the 
agamosporous varieties of P. glabella and the origins of those varieties from the 
respective sexual taxa of this complex. 


MATERIALS AND METHODS 


The taxonomy and nomenclature of Tryon (1957) are provisionally followed 
throughout this paper up to the final section entitled Taxonomic Conclusions, 
where they are revised to reflect insights gained in this study. Following Walker 
(1979), the term ‘“‘agamosporous” (Love & Léve, 1975) is used here in place of 
“apogamous” or “apomictic’” for fern taxa in which the production of unreduced 
spores and apogamous sporophyte initiation occur regularly in the life cycle. 
The geographic ranges of P. glabella (Fig. 1) and P. atropurpurea are based on 
Tryon (1957, 1972), Tryon and Britton (1958), Wagner et al. (1965), Rigby (1968), 
Rigby and Britton (1970), Brunton and Lafontaine (1974), and Brunton (1979). 
Live population samples of the sexual and agamosporous varieties of P. glabella 
were collected from throughout their ranges, and P. atropurpurea was sampled 
overa large part of its distribution (Table 1). Voucher specimens are deposited at 
IND. 


I r. £ om: . . x ; . 
TABLE 1. Locality ata for Cited P. lat f Pell S Examined EF] } t ll 
y 


P. glabella var. simplex. $1—British Columbia, Pavillion Lake, ca. 24 mi NW of Cache Creek, lime- 
stone outcrops along Rte. 12, Gastony 82-17; $2—Washington, Douglas Co., ca. 7 mi NE of Pali- 
sades, Rainy Cave, Rooster Rock Ranch, Gastony 62-19; S3—Washington, Grant Co, oe Bar, 
ca. 10 mi from Quincy, in shallo t Bar condo- 
miniums, Gastony 82-20; $4 —Colorado, Garfield Co., ca. 14 mi NE of Rifle, limestone cliffs 1 mi 
beyond fish hatchery, Gastony eee S5---Arizone, Coonan ., Flagstaff, limestone outcrops 

js eee * s t 7 te Wil- 


along Rio de Flag, Gastony 82-25; S6 
liams Lake to Riske Creek, E side ee Fraser werk  Gastony pies 
P. atropurpurea. A1—Arizona, Cochise t Huack Military Installation, 
near mouth of Scheelite Canyon, eto ctrne 82-271; A2—Arkansas, Benton Co.,S. of Bella Vista 
and NE of Rte. 71, along old country road, Yatakievych & McCrary 83-60. A3—Missouri, Pulaski 
Co., Gasconade River at Rte. 7, between Richland and Lequey, on boulders beside dirt road along 
nie. Gastony. 82-27A; AM-Missourl. Jefferson Co., 2 mi E. of Antonio on Highway M, disturbed 
Gastony 83-33; A5—West Virginia, Pendleton Co., lime- 
stone outcrops, Smoke Hole recreational area, Gastony 83-48; A6—West Virginia, jortersgn a : 
Shepherdstown, cliffs along: River Road, Gastony 83-49; A7—Virginia, Giles Co. 
near New River Cave at G in’s Ferry, We erth s. $.n, = Gastony: 62-30; Se Centre 
Co., ca. 3 mi NE of State College, NW-faci stony 
82-28; A9—New Jersey, Sussex Co., Andover, limestone outcrops along abandoned | RR in id ¢ 
Rte. 206, Gastony 83-53; A10—Maryla sitet Ww 
River, Gastony 83-51; A11—Vermont d “ E of Burling RR cutb 
tween Rte. 15 and Winooskis River gorge, Csiteay 83-54A; Me ase Gatineau Co:, 3 mi S of 
Poltimore, SW-facing cliff on Chamberlin property, Gastony 83-58; A13—Arkansas, Benton Co., 
Bella Vista, trout farm catfish pool #3, on S- and W-facing limestone cliffs near grotto E of pool, 
Yatskievych & McCrary 83-61; A14—Maryland, Washington Co., along the Potomac River at the 
Rte. 34 bridge SW of Sharpsburg, Gastony 83—50; A15—Missouri, Jefferson Co., river bluffs 10 
miles W of DeSoto, near junction of road H and Big River, Gastony 83-34; A16—Texas, Davis Co., 


i 


G. J. GASTONY: PELLAEA GLABELLA COMPLEX 47 


TABLE 1. continued 


Madera Canyon, hillside above rest area on Rte. 118, Gastony 87-8; A17—Nuevo Leon, México, 
Canon de San Francisco, W of Rte. 85, ca. 35 km SE of Monterrey, dry streambed across road from 
pumping station, Gastony 87—-11-7. 

P. glabella var. glabella, agamosporous tetraploid. Gi—Ontario, — limestone walls of old 
quarry on reformatory property, Gastony 83-60; G2—Minnesota, Winona Co., Whitewater State 
Park, limestone outcrops near Chi imney — and Inspiration Point, Gastony 82-1; Mis 
Dent Co., Montauk State Park, limesto nds Werth s.n. = Gas- 
tony 82-43; G4—New Jersey, Hinder ae Bloomsbury, on moi ortar between stones of RR bridge, 
Gastony 83-52; G5—Vermont, Chittendon Co., Ethan Allen Park N of Burlington, limestone out- 
crops, Gastony 83-55; G6—Quebec, Papineau Co., 2 mi SW of Poltimore, SW-facing cliff, Gastony 

3-57; G7—Texas, Potter Co., limestone sgsiaycs neat Rte. 287 CORES, of John Rey Creek, Gas- 
tony selene Sons Centre Co., W-facing limestone cliffs 


of Big Creek, Gastony 82-28; G9—Virginia, Giles Co., limestone cliffs near New 
River at Goodwin’ s Ferry, Werth s.n. = Gastony 82-29; G10—Missouri, Jefferson Co., bluffs along 
Big River on road Y, across road from Bethlehem Church, ca. 4 mi SE of Grubville, Gastony 83-35; 
G11—AMissouri, Shannon Co., limestone outcrops SW of a at } Rte. 18 crossing of | Current River, 
Gastony 83-40; G12—Kentucky, Lyon Co., Buzzard Ro 
Gastony 83-42; G13—Mary land, Washington Co., lies cliffs near Dam 5 along Potomac 
River, Gastony, 83-51; G14—Missouri, Franklin Co., limestone cliffs from Fisher Cave to river, 
eraenes State Park, E of Sullivan, Gastony 83-36. 
bella var. occidentalis. 01—South Dakota, Pennington Co., 3.5 mi N of Deerfield, Flag Mt., 

rock crevices in open woods at top, Gastony 83-63; O2—S South Dakota, Pennington Co., 6 mi S of 
Deerfield, S- to SW-facing limestone cliffs along Rte. 291, Gastony 83-62; 03—South Dakota, 
Lawrence Co., S- and SE-facing high cliffs above Hanna campground, ca. 2 mi = of Cheyenne 
Crossing, Gastony 83-65; O4—Wyoming, CrookC 
jct. Rte. 863 and [90 at Beulah, Gastony 83-67; oN Washakie se cliffs ae huge boul- 
ders lining Leigh sien da off Jane 16, —— 83-70; O6 

alon: 343 from jet. with Ree. 14, ‘puta 83-69; O7—Wyoming, 


Bighorn Co Bighorn Nat. Forest, li 2 mi. NW of Tyrrell Ranger Station, Gastony 
83-71; 08—Utah, saacein Ga., gcc Nat. Forest, ca. 15 mi seat of Maeser, limestone outcrops in 
Brownie Canyon, Gastony 83—72 


P. glabella var. Gietetia, sexual diploid. M1—Missouri, Carter Co., SW- facing cliffs along Current 
River, S of Van Buren, Gastony 83-41; M2—Missouri, Jefferson Co., river bluffs 10 miles W of De- 
Soto, near jct. of road H and Big River, Gastony 83-34; M3 Missouri, Jefferson Co 
River on road Y, across road from Bethlehem Church, ca. 4 mi SE of Grubville, Gastony 83-35: 
M4— Missouri, Dent Go., limestone bluffs along Meramec River between Howe and Sligo, Gastony 
83-39: M5—DMissouri, Pulaski Co., Gasconade River at Rte. 7, between Richland and Lequey, on 
limestone cliff faces, Gastony 82—-27C. 


P. atropurpurea ~ P. glabella var. occidentalis, tet loid hybrid. HAOQ1—Alberta, E-facing cliffs 
at S end of Barrier Lake S of Seebe, Gastony 82-15 HAO2—Wyoming, Crook Co., limestone out- 
crops along Rte. 863, 9.1 mi S of jct. Rte. 863 and 190 at Beulah, Gastony 83-67. HAO3— 
kota, Pennington Co., ca. 10 mi W of Rapid City, li p g Rte. 44, Gas- 
tony 83-64. 

P. atropurpurea x sexual P. glabella\ var. jcatoenes ahs poe hybrid. seaman Dent Co., 
li between Howe and Sligo, Gastony 83— 


So 


Enzyme designations, electrophoretic procedures, and recipes for extraction 
buffer, gel and electrode buffers, and enzyme assays are as in Soltis et al. (1983). 
The Tris-HCl extraction buffer was used with PVP 40,000 at 2.5 9/25 ml. For all 
cited populations (Table 1) an average of 30 sporophytes were analyzed for 14 
enzymes on each of the gel/electrode systems in Soltis et al. (1983) and on the 
Poulik discontinuous Tris-citrate and continuous Tris-citrate II systems of 


48 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 2 (1988) 


Selander et al. (1971). Some of these enzymes did not resolve adequately on any 
of these systems. Of the enzymes that were well resolved, only those (PGM, PGI, 
TPI, SkDH, AAT, MDH) providing taxon-specific genetic markers for Pellaea 
atropurpurea and the P. glabella complex are considered here, because only 
these supply the most unequivocal data about the genomic ancestry of the 
derivative agamosporous tetraploid taxa. These six enzymes are coded by at 
least 10 loci (the genetic basis of the AAT patterns is uncertain), and in all cases 
the bands migrated anodally. When more than one isozyme was present for an 
enzyme, the most anodal isozyme was designated as “‘1,’’ the next most anodal as 
“2,” etc. Thus, isozymes and their coding genes are designated as in the 
following example: MDH-1 (the isozyme), mdh-1 (the coding gene), MDH-2 (the 
isozyme), mdh-2 (the coding gene). 

The enzymes discussed and the gel-electrode systems (Soltis et al., 1983) on 
which they were resolved are: PGM, PGI, TPI on 7 (except in Fig. 8, TPI on 8); 
SkDH on 2; AAT on 10; MDH on 9. All gel slices were incubated in stains in the 
dark at 37°C. The number 7 gel was subjected to electrophoresis at 35 mamp for 
6.5 hr until a bromphenol blue (BPB) marker dye reached 13.2 cm from origin; 
the number 10 gel, 35 mamp for 6 hr, BPB at 13.3 cm; the number 2 gel, 30 mamp 
for 6.5 hr, BPB at 13.8 cm; the number 9 gel, 35 mamp for 6 hr, BPB at ca. 13 cm. 
In all cases 13.2% starch concentration was used, electrophoresis was conducted 
at 4° C, and wicks were removed from the gels after 10 min. of electrophoresis at 
the indicated amperages, except that the numbers 2 and 10 gels were kept at 25 
mamp until the wicks were removed. 

Genetic interpretations of enzyme banding patterns in the sexually 
reproducing taxa are based on gametophytic segregational analyses as in 
Gastony and Gottlieb (1982). The banding patterns of agamosporous fern taxa do 
not segregate because of preferential pairing of sister chromosomes derived from 
the endomitotic division preceding meiosis in Dépp-Manton sporogenesis 


—_—_—ooP 
Fics. 2-12. apcacmanabae banding pees of the agen glabella complex and P. atropurpurea. 
In all figures, the ano , except that the gel was cut two 


cm anodal to origin in ‘Fig 7 and one cm anodal to origin in Fig. 11. S = P. glabella var. simplex, 
O = var. pecidentaiis. G= = agamosporons var. glabella, M = sexual var. glabella, A = P. 
atropurpurea. All d. See text for interpretation 
and discussion of the patterns. Millimeter bar in Fig. Z cepeseen ts a prediction factor of 0.5 for Figs. 
2-12. Fics. 2, 4, 6 have identical plant Papen as follow (acronyms of source populations as in 
Table 1): lane 1 = $1; 2 = S2;3 = $3;4 = S4:5 = S5:6 = eenetogg 2;@ = A3:9 = A4; A 
11 = A6; 12 repeats S3; 13 repeats A4; i: = G1;15 = A7;16 = A8;17 = A9;18 = A10;19 = A11; 
20 = A12;21 = G2; 22 = G3;23 = G4; 24 = elma es Fic. 3:lane1 = S1;2 = $5;3 = O1 
pooled gametophytes; 4 repeats 1; 5 = O5 pooled gametophytes; 6 repeats 1; 7 = G2; 8 = G1; 
9 = M1; 10 = M1; 11 = M2; 12 repeats 1. Fic. 5: lane 1 = S6:2 = $2:3 = S4; 4 = $5; 5 = O1; 6 
repeats 4;7 = M4;8 = M5; 9 = M2; 10 = M1; 11 repeats 4; 12 = G7;13 = G8;14 = G4; 15 repeats 
7; 16 = O03; 17 = G9; ap repens & cyt Ga; 20 = oe pa8 = G10; 22 = G6. Fic. 7: lane 1 = O2; 
2 = 08; 3, 4, 8,9 = si lanes 2, 6, 10; 5 repeats 1; 6 repeats 2; 
7 = O07: 10 repeats 2; 11 repeats 7. Fic. 8:lane1 = $1; 2 = $5:3 = G2; 4 = G1; 5 repeats 1; 6 repeats 
2;7 = M4;8 = 04;9 = 06; 10 repeats 7;11 = M4;12 = M2: 13 repeats 8; Sah an 15 repeats 2; 
16 repeats 9. Fic. 9: lane 1 = M1; 2 = O5. Fic. 10:lane1 = M2;2 = G2:3 = O7;: S: 4; 
6 = S5. Fic. 11: lane 1 = 05; 2 = S6; 3 = O8. Fic. 12: lane 1 = S6; pis a O5; 4 = 08; 
5 = 05;6 = G5;7 = G4. 


G. J. GASTONY: PELLAEA GLABEL LA COMPLEX 49 


50 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 2 (1988) 


(Gastony & Gottlieb, 1985). Thus even heterozygous patterns are fixed in the 
agamosporous taxa examined here, and their genetic interpretations are inferred 
by comparison with the co-migrating patterns of sexual taxa subjected to side by 
side electrophoresis on the same gels. 


RESULTS 


The three agamosporous taxa each have unique banding patterns for the 
monomeric enzyme PGM (Fig. 2). This enzyme occurs as two isozymes, one in 
the cytosol and a second in the chloroplasts (Gottlieb, 1981; Weeden, 1983; 
Gastony & Darrow, 1983). As shown in Fig. 2, var. simplex (lanes S) expresses a 
single PGM-1 band. Pellaea atropurpurea (lanes A) expresses several 
two-banded patterns at PGM-1, all with one of the bands at the position indicated 
by the arrow in Fig. 2. In lane 6, the second PGM-1 band is co-migratory with the 
single PGM-1 band in var. simplex, but in most cases the second band occupies 
one of two slower positions (lanes 7-11, 13, 15-20). Tetraploid P. glabella var. 
glabella (lanes G) is two-banded at PGM-1, with a relatively strong band 
co-migratory with the single band in var. simplex and a slower weaker band 
apparently co-migratory with the invariant PGM-1 band (arrow) in P. 
atropurpurea. Banding intensities at PGM-1 in tetraploid P. glabella var. 
glabella suggest three doses of the allele coding the faster allozyme and a single 
dose of the allele coding the slower allozyme. Note that the triploid genotypes 
coding observed PGM-1 patterns in sampled populations of P. atropurpurea 
could not be added to any haploid genotype to yield the tetraploid PGM-1 
banding patterns of vars. simplex or glabella. 

At PGM-2 (Fig. 2), tetraploid var. simplex (lanes S) has a two-banded pattern 
whose unequal staining sntensinies suggest three doses of the allele coding the 
stronger and faster single dose of f the allele coding the weaker and 
slower allozyme. The tightly ‘ova lecded pattern of tetraploid var. glabella 
(lanes G) also suggests three doses of the allele coding the slightly slower 
allozyme that comigrates with the stronger allozyme of var. simplex (compare 
lanes 12 and 14), and a single dose of the allele coding the slightly faster 
allozyme. Pellaea atropurpurea (lanes A) is single-banded, implying that its 
triploid genotype is fixed for the allele coding the same allozyme that is 
dominant in the other two taxa (compare lanes 12-14). Thus all three taxa 
appear to share three doses of the same allele at PGM-2, so that this locus cannot 
be used to exclude P. atropurpurea from the ancestry of the tetraploid varieties. 

Representative banding patterns of PGM in agamosporous varieties simplex 
and glabella are compared to the patterns of the sexual taxa in Fig. 3. In var. 
simplex (lanes S), single-banded PGM-1 and two-banded PGM-2 are comigratory 
with the apparently homologous allozymes of sexual var. occidentalis (lanes O, 
bath lanes representing randomly selected pooled gametophtyes derived from 
). Thus all PGM bands in var. simplex are 
found in var. occidentalis and can therefore be accounted for by loci and alleles 
inherited entirely from var. occidentalis. Lanes G show the two-banded patterns 
for both PGM-1 and PGM-2 in agamosporous var. glabella. The faster (more 
intense) band at PGM-1 is found in sexual var. glabella (lanes M), but this is not 


G. J. GASTONY: PELLAEA GLABELLA COMPLEX 51 


instructive since it is common to both agamosporous and both sexual taxa of P. 
glabella. The slower (less intense) band at PGM-1 in agamosporous var. glabella 
distinguishes that taxon from var. simplex which lacks it. That slower band has 
not been found in any sporophyte of var. occidentalis examined but has been 
observed in two sporophytes of sexual var. glabella (e.g., lane 9). At PGM-2, 
agamosporous var. glabella (lanes G) is two-banded. The slower band is foundin 
all glabella taxa, including sexual var. glabella (lanes 10, 11). The faster band 
has not yet been found in sexual var. glabella or in var. occidentalis. Its coding 
allele may now be either very rare in the sexual progenitor or lost from it, leaving 
it as an orphan allele in the tetraploid, similar to the orphan alleles encountered 
by Werth et al. (1985) in their study of polyploid Asplenium taxa. 

The dimeric enzyme PGI exhibits two isozymes, one active in the cytosol anda 
second in the chloroplasts (Gottlieb, 1981; Weeden, 1983). PGI-1, the putatively 
chloroplastic isozyme (Gastony & Darrow, 1983), did not resolve into bands (Fig. 
4) and will not be discussed. PGI-2 (Fig. 4) is single-banded in P. glabella var. 
simplex (lanes S), P. atropurpurea (lanes A), and certain populations of 
agamosporous P. glabella var. glabella (lane 23). In most populations of the 
latter, PGI-2 has a three-banded pattern of asymmetric intensities (lanes 14, 21, 
22, 24, 25), suggesting that these tetraploids have 3 doses of the allele coding the 
slower allozyme and a single dose of the allele coding the faster allozyme, with 
the central heterodimeric band nearly equalling the slow homodimeric band in 
intensity. The single PGI-2 band of var. simplex is co-migratory with the single 
band or the most cathodal band of var. glabella (compare lanes 12, 14), but the 
nearly co-migratory single band of P. atropurpurea (lane 13) is slightly slower. 
This lesser migration of PGI-2 in P. atropurpurea is subtle, but it has been 
confirmed many times in gels with various juxtapositions of these taxa. 
Although this distinction might not be convincing of itself, it gains strength from 
its correlation with distinctions based on other enzymes. Thus inasmuch as 
PGI-2 in P. atropurpurea does migrate to a position slightly cathodal to the 
analogous PGI-2 bands in the P. glabella complex, the triploid PGI-2 genotype of 
P. atropurpurea cannot be a constituent of the genotypes of the agamosporous 
tetraploids. 

Figure 5 shows that the invariant one-banded pattern for PGI-2 in var. simplex 
(lanes S) is identical to the invariant one-banded pattern of this isozyme in 
sexual var. occidentalis (lanes O). In hundreds of specimens of var. occidentalis 
from throughout its range, all sporophytes expressed this same single band for 
PGI-2. Figure 5 also shows that this same band is found in sexual var. glabella 
(most lanes M) and agamosporous var. glabella (lanes G). The more anodal 
homodimeric allozyme seen in those populations of agamosporous var. glabella 
characterized by a three-banded PGI-2 pattern (lanes 17, 19, 21, 22) is frequently 
observed in sexual var. glabella (e.g., lanes 8, 9, 18) but never observed in 
hundreds of sporophytes throughout the range of var. occidentalis, suggesting 
that the allele for this allozyme in agamosporous var. glabella came from sexual 
var. glabella. i : 

Dimeric TPI is also expressed as cytosolic and chloroplastic isozymes in 
plants (Gottlieb, 1981; Pichersky & Gottlieb, 1983). One zone of TPI activity 


52 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 2 (1988) 


appears as a non-segregating, asymmetrically stained, three-banded pattern in 
homozygous plants of nearly all fern species examined. This is usually the more 
anodally migrating zone (TPI-1), and this three-banded pattern has been 
attributed to either a single gene duplication or a post-translational modification 
(Gastony & Darrow, 1983; Gastony & Gottlieb, 1985; Haufler, 1985; Haufler et al., 
1985; Gastony, 1986; Haufler & Soltis, 1986; Soltis, 1986; Soltis & Soltis, 1986). 
Post-translational charge modification of some of the coded TPI-1 protein is the 
more probable explanation (see Appendix, p. 66). When the locus coding this 
pattern is heterozygous, it codes 10 enzyme bands (four homodimers and six 
heterodimers), but these usually appear as a four-banded to seven-banded 
pattern depending on the number of overlapping or nearly overlapping bands 
and on the faintness of the anodal homodimers. 

From Fig. 6 it is clear that var. simplex (lanes S) is fixed for a heterozygous 
five-banded pattern at TPI-1 (the slowest band being quite faint) whereas 
agamosporous var. glabella (lanes G) and P. atropurpurea (lanes A) are invariant 
for their respective asymmetrical three-banded patterns. Several side-by-side 
comparisons among these TPI-1 patterns on different gels have shown that the 
bands of P. atropurpurea are not co-migratory with those of var. simplex nor with 
those of var. glabella. Therefore the triploid TPI-1 pattern of P. atropurpurea 
could not be added to any haploid TPI-1 pattern to yield the tetraploid TPI-1 
patterns of P. glabella vars. simplex and glabella. 

Sporophytes of var. occidentalis homozygous at tpi-1 express either of two 
three-banded patterns (Fig. 7, lanes 5, 7) while the comparable heterozygote 
(lane 6) sh several additional heterodimeric bands (seven of the ten expected 
bands are fairly easily distinguished in this gel that was run extra long to 
maximize band separation), and lanes 3, 4, 8 and 9 show segregation of the 
constituent homozygous 3-banded patterns in single gametophytes derived from 
the heterozygous sporophyte in lanes 2, 6, and 10. Similarly, sporophytes of 
sexual var. glabella homozygous at tpi-1 express either of two three-banded 
patterns (Fig. 8, lanes 10, 12), while a sporophtye het ygous at this locus (lane 
11) expresses a five-banded pattern on this gel system (because of comigration of 
some of the bands). The slower three-banded TPI-1 patterns of var. occidentalis 
and sexual var. glabella are identical (Fig. 9; Fig. 10, compare lanes 1, 3). As seen 
in Fig. 8, lanes 7-10, however, the faster three-banded TPI-1 pattern of sexual 
var. glabella is not identical to that of var. occidentalis, but rather all three bands 
of sexual var. glabella {lanes 7, 10) are slightly slower than their counterparts in 
var. occidentalis (lanes 8, 9). 

The characteristic TPI-1 banding patterns of var. simplex and agamosporous 
var. glabella inconsistent with a derivation from P. atropurpurea are consistent 
with derivations from var. occidentalis and sexual var. glabella. All sampled 
populations of agamosporous var. glabella [except the outlier in Texas (Fig. 1)] 
are characterized by a three-banded pattern at TPI-1 (e.g., Fig. 10, lane 2). These 
bands are identical to those of the slower three-banded TPI-1 pattern common to 
sexual var. glabella (Fig. 10, lane 1) and sexual var. occidentalis (Fig. 10, lane 3). 
Thus the data from TPI-1 are consistent with an origin of agamosporous var. 
glabella from either of these sexual taxa. As seen in Fig. 8, lanes 14, 15, var. 


G. J. GASTONY: PELLAEA GLABELLA COMPLEX 53 


simplex exhibits five bands, of which the three anodal bands align perfectly with 
those of var. occidentalis (lanes 13, 16) and in lanes 5, 6 do not align with those 
from sexual var. glabella (lane 7). The slower TPI-1 bands in var. simplex are 
faint as would be expected if three doses of the faster pattern of TPI-1 from var. 
occidentalis were combined with a single dose of the slower pattern of TPI-1 
shared by the sexuals, as seen by comparing lanes 4, 6 of Fig. 10. Under this 
interpretation, the slowest strong TPI-1 band in var. simplex (lane 6) should be 
the heterodimer between the cathodal homodimers of the faster and slower 
three-banded TPI-1 patterns. Figure 11 shows that the slowest strong TPI-1 band 
in var. simplex (lane 2) does correspond to that heterodimeric band in the 
heterozygous TPI-1 pattern of diploid var. occidentalis (lane 1) which has equal 
doses of the tpi-1 alleles. That this band is truly heterodimeric in var. 
occidentalis is established by comparing lanes 5—7 of Fig. 7 where the 
heterozygote is flanked by respective homozygotes. 

At TPI-2 (Fig. 6), var. simplex (lanes S) and agamosporous var. glabella (lanes 
G) exhibit the same asymmetrical three-banded pattern, suggesting that they are 
fixed for three doses of the allele coding the faster | 1i d one dose of the 
allele coding the slower homodimer. The population of var. glabella 
represented by the second-last lane is strongly banded between TPI-2 and TPI-1 
where other populations exhibit only a very faint band taken to reflect 
post-translationally modified protein. The TPI-2 pattern of P. atropurpurea 
(lanes A) is invariant and probably best interpreted as intensely single-banded 
with the faint anodal band representing post-translationally modified protein. 
The strong TPI-2 band of P. atropurpurea is co-migratory with the slower 
(fainter) homodimeric band of vars. simplex and glabella on several 
gel-electrode systems. The genotype coding this locus in triploid P. 
atropurpurea sampled in this study, however, could not combine with any 
haploid genotype to produce the relative staining intensities of the TPI-2 bands 
of tetraploid vars. simplex and glabella. The validity of this reasoning can be 
established by examining the TPI-2 patterns of true hybrids between P. 
atropurpurea and the sexual taxa. Because such hybrids would be tetraploid 
with three genome doses contributed by the triploid sperm of P. atropurpurea, 
their morphological intermediacy should be strongly skewed toward this 
triploid parent. The relative intensities of the bands of the additive enzyme 
patterns of true hybrids should also reflect the 3:1 atropurpurea: sexual-parent 
gene dosages. Occasional tetraploid plants (verified by chromosome counts) 
matching these morphological expectations have been encountered in Alberta, 
Wyoming, South Dakota, and Missouri where the ranges of these taxa overlap. 
As an example, composite Fig. 13 presents TPI banding patterns of these hybrid 
sporophytes (lanes H) in comparison with the patterns of parental P. 
atropurpurea (lanes A) and var. occidentalis (lanes O). Figure 13 also compares 
the TPI patterns of these true hybrids (lanes H) with those of var. simplex (lanes 
S) and agamosporous var. glabella (lanes G). The maternal parent (not shown) of 
the hybrid from Missouri (lane 7) is sexual var. glabella with the faster allozyme 
at both TPI-1 and TPI-2. The observed enzyme additivity and staining 
intensities of these true hybrids are in precise agreement with expectations and 


54 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 2 (1988) 


do not match the relative band intensity patterns of agamosporous tetraploid 
vars. simplex and glabella. 

Both tpi-2 alleles coding the homodimeric TPI-2 bands in var. simplex (Fig. 8, 
lanes S) and agamosporous var. glabella (Fig. 8, lanes G) are found in sexual var. 
glabella (Fig. 8, lanes M) and in var. occidentalis (Fig. 12, lanes O; compare lane 
3 with var. simplex in lanes S and with agamosporous var. glabella in lanes G). 
Because both agamosporous taxa and both sexual taxa carry the same tpi-2 
alleles, data from this locus are not useful in defining specific progenitor- 
derivative pairs, although they are consistent with the autopolyploid derivations 
indicated by other isozymes. 


Ce | 
GGGMMGGGE 


G. J. GASTONY: PELLAEA GLABELLA COMPLEX 55 


For the monomeric enzyme SkDH (Fig. 14), var. simplex (lanes S) expresses a 
very tightly grouped two-banded pattern in which the two contiguous bands are 
of equal intensity. Agamosporous var. glabella (lanes G) is also tightly 
two-banded and the bands are aligned with those in var. simplex, but the more 
anodal is about three times more intense than the cathodal. In P. atropurpurea 
(lanes A), SkKDH is almost always expressed as two well separated bands, the 
exception seen here being the population from Arizona (lane 6) where a single 
band is expressed. The more cathodal of the two bands in P. atropurpurea 
appears to be co-migratory with the more cathodal band in vars. simplex and 
glabella. In most two-banded populations of P. atropurpured: the cathodal band 
is more intense than the anodal (lanes 7—11, 13, 15, 18) suggestive of atwo to one 
gene dosage effect in this triploid taxon. In four of the populations analyzed 
(lanes 16, 17, 19, 20), however, the band intensities are just the opposite, with the 
anodal band approximately twice the intensity of the cathodal. Agamosporous 
vars. simplex and glabella lack the anodal band of most populations of P. 
atropurpurea and their cathodal allozyme is less intense than would be expected 
if they inherited it from triploid P. atropurpurea with the single-banded pattern 
in lane 6. Thus SkDH further decreases the probability that the genome of P. 
atropurpurea is a component of the genomes of the agamosporous tetraploids. 
Both bands of var. simplex are found in var. occidentalis where their tightly 
grouped two-banded pattern in heterozygotes has been shown to segregate in 
gametophytes. Only the stronger, more anodal of the two bands in 
ne anpeaen var. glabella has been observed in sexual var. glabella. 

dimeric enzyme with one to four isozymes reported from plants 
(Gottlieb, 1981, 1982). The distinctive three-banded patterns of var. simplex 
(Fig. 15, lanes S) and var. glabella (lanes G) sharply differentiate them, although 
their most cathodal bands are co-migratory (lanes 13-17). Three AAT patterns 
are expressed by the 12 populations of P. atropurpurea analyzed for AAT, two 


—— 


f OF poll 5% 


Fics. 13-20. Electrophoretic bandi tt f the Pell labell l d P. atropurpurea. 
In all figures th lei 1 tt d ics atthe bottom pt tl ttl g l t0.5cm 
anodal t to origin in Figs. 16,17. S = P. esate var. simplex, O = var. occidentalis,G = agamospor- 
ous var. glabella, M = sexual var. glabella, A = tf atropurpurea. See tent for interpretation and dis- 
cussion of the patterns. Millimeter bar in Fig. 1 0. 
In composite Fic. 13: lane 1 = A1;2 = A4;3 = A12: 4 = A14;5 = HAO1; 6 repeats 5; 7 = HAM1; 
3; 9 = HAO?2; 10 = O3; 11 = O05; 12 = O5: 13 = $2; 14 = S3; 15 = S4; 16 = G12; 
17 = G8; 18 = G13. Fic. 14:lane1 = S1; 2 = $2;3 = $3;4 = S4;5 = S5;6 = A1;7 = A2;8 = A3; 
9 = A4: 10 = AS; 11 = AG; 12 repeats S3; 13 repeats A4; 14 = G1; 15 = A7; 16 = A8; 17 = AQ; 
18 = A10;19 = A11;20 = A12;21 = G2; 22 = G3; 23 = G4; 24 = G5; 25 repeats G1. Fic. 15: lane 
1 = $1; 2 = S2:; 3 = S5; a= S4 snes ae ee B19: ae A7: 8 = A12; 9 = G2; 10 = G3; 11 = G4; 
12 = G6; lanes 13—18 test band 13 repeats 2; 14 repeats 9; 15 repeats 2; 
16 repeats 10; 17 repeats 3; 18 repeats 7. Fic. 16: lane 1 = O05; 2 = O07; 3 = O05; 4 = O5; 07; 
isin rhatanbey a :9 = 05;10 = 05;11 = O5. Fic. 17:lane1 = S1;2 = $2:3 = S5;4 = O6; 
= O6;6 repeats 1;7 = 06; 8 repeats 1;9 = O6; Lac argh 11 = G2;12 = G1;13 = G5;14 = M1; 
os 16 repeats 11; 17 = G12; 18 = bail = G4. Fic. 18: lane 1 = 01; 2 = M5; 3 = M5; 
4 = M5; 5 = O1. Fic. 19: lane 1 = $1; pa 3 = G4; 4 = G1; 5 = A15; 6 = A16; 7 = A17; 
8 = A7; 9 = S5; 10 = S4; 11 = Gi2; 12 = or Fic. 20: lane 1 = M1; 2 = M1; 3 = G5; 4 = G6; 
§ = G11;6 = G7: 7 = M1;8 = M65. 


56 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 2 (1988) 


different three-banded patterns (lanes 5, 8) and a more commonly encountered 
pattern consisting of a single strong band (lane 6) sometimes with a less intense 
band anodal to it (lane 7). The anodal band in var. simplex appears to align with 
the most cathodal of the three P. atropurpurea bands in lane 5, but otherwise all 
observed patterns of P. atropurpurea differ so strongly from those of the other 
two taxa that the triploid genome coding these patterns in P. atropurpurea could 
not titute three-fourths of the AAT genotype of the tetraploid varieties. Lack 
of variation for this enzyme in the sexual taxa and lack of segregation in 
agamosporous taxa preclude genetic interpretation of AAT banding patterns in 
this complex. Both sexual taxa exhibit the same two strong bands and these align 
with the most anodal and cathodal bands of var. simplex, whose central band 
may be heterodimeric. In addition, fainter, more anodal bands are evident in 
some plants of sexual var. glabella. However these do not align with the most 
anodal band usually visible in agamosporous var. glabella, so that the allele or 
gene coding that band is apparently orphaned. The unclear interpretation of 
AAT isozymes in this complex decrease the effectiveness of AAT for purposes of 
this paper, except that the AAT patterns of P. atropurpurea and the 
agamosporous tetraploids are strongly differentiated and strengthen arguments 
from other enzymes against the involvement of P. atropurpurea in the 
derivations of the tetraploids. 

MDH is a dimeric enzyme usually expressed as 3 or more isozymes in plants 
(Gottlieb, 1981). In var. occidentalis, gametophytic segregation studies (as in 
Gastony & Gottlieb, 1982) demonstrate that the MDH-1 patterns in Fig. 16 are 
specified by four alleles coding allozymes 1a, 1b, 1c, 1d. Other ‘‘bands” seen at 
MDH-1 in this figure are post-translational artifacts. MDH-2 is expressed as 
allozymes 2a (seen in homozygous condition in lanes 3, 4, 6, 7, 10, 11) and 2b 
(homozygously expressed in lane 5). MDH-3 is a fixed (non-segregating) 
three-banded pattern best seen in lanes 3, 4, 6, 7, 10, 11. This pattern is either 
coded by a gene and its modified duplication or by post-translational 
modification of some of the coded protein as in the case of TPI-1. The most 
cathodal (slowest) band of this three-banded MDH-3 pattern is faint. Under the 
electrophoretic conditions expressed in this figure, MDH-2b overlaps the faint 
MDH-3 band when its coding allele mdh-2b is homozygous (lane 5). MDH-2b is 
seen slightly cathodal to this faint MDH-3 band when mdh-2b is heterozygous 
with mdh-2a (and allozyme 2b is therefore less intense; lanes 1, 2, 8, 9) or when 
the duration of electrophoresis is increased. When mdh-2 is heterozygous, the 
heterodimer overlaps the most anodal of the MDH-3 bands (lanes 1, 2, 8, 9). 

Three MDH isozymes also characterize sexual var. glabella. MDH-1 is 
invariantly single-banded (Fig. 17, lanes M), and aligns with allozyme 1c of var. 
occidentalis (lanes 4, 5, 9). Sexual var. glabella expresses one MDH-2 allozyme 
that corresponds to MDH-2b of var. occidentalis (Fig. 17, compare lanes M with 
lane 4) and an additional more cathodal allozyme absent from var. occidentalis. 
This slower MDH-2 allozyme of sexual var. glabella (MDH-2c) is seen in 
homozygous condition in Fig. 18 lanes M and in heterozygous condition (with 
heterodimer) in Fig. 17 lanes M. The three-banded MDH-3 pattern of sexual var. 
glabella is the same as in var. occidentalis (Fig. 18, compare lanes M with lane 5). 


G. J. GASTONY: PELLAEA GLABELLA COMPLEX 57 


The agamosporous varieties of P. glabella are easily distinguished by their 
respective, invariant MDH patterns. Variety simplex (Fig. 17, lanes S) has a fixed 
three-banded pattern at MDH-1, the same fixed three-banded MDH-3 pattern 
seen in the sexual taxa, and a single intense MDH-2b band cathodal to MDH-3. 
Agamosporous var. glabella (Fig. 17, lanes G) has a single intense MDH-1c band 
aligned with the most anodal MDH-1 band of var. simplex, the same 
three-banded MDH-3 pattern as the other members of this complex, and a fixed 
three-banded MDH-2 pattern just cathodal to three-banded MDH-3. The most 
anodal MDH-2 band in agamosporous var. glabella (Fig. 17, lane 11) aligns with 
the MDH-2b band in var. simplex (Fig. 17, lane 10), whereas the most cathodal 
band in lane 11 aligns with MDH-2c of sexual var. glabella (lanes M). 

Sporophytes of P. atropurpurea examined for MDH (Fig. 19, lanes A) express 
the MDH-1 allozyme shared by all taxa of the P. glabella complex and may 
express additional MDH-1 allozymes not relevant to the agamosporous taxa of 
the glabella complex (e.g., Fig. 19, lane 5). At MDH-3, P. atropurpurea expresses 
the same fixed three-banded pattern as all the P. glabella taxa. At MDH-2, P. 
atropurpurea has either a single band (Fig. 19, lanes 6—8) aligning with MDH-2b 
of vars. simplex (lanes S) and occidentalis or three bands (lane 5) aligning with 
MDH-2b, MHD-2c, and their heterodimer in agamosporous var. glabella (lanes 
G). Thus unlike the previously discussed enzymes, MDH evidence alone does 
not exclude a combination of the MDH genes of P. atropurpurea from the 
genotypes of the agamosporous tetraploids of the glabella complex. 

MDH data, however, do strengthen the evidence from other enzymes 
supporting autopolyploid derivations of the agamosporous tetraploids. The 
invariant MDH bands of var. simplex (Fig. 17, lanes S) correspond to a subset of 
the bands of var. occidentalis seen in Fig. 17, lanes 4, 5, viz. MDH-1c, MDH-1d, 
and MDH-2b. Similarly, the MDH bands of agamosporous var. glabella (Fig. 20, 
lanes G) correspond to those of sexual var. glabella seen in Fig. 20, lanes M, viz. 
MDH-1c, MDH-2b, and MDH-2c. MDH-1d of var. simplex (Fig. 17, lanes S) is 
coded by an allele that is expressed in var. occidentalis (Fig. 17, lane 5) but that 
has not been found in hundreds of analyzed sporophytes of sexual var. glabella. 
Similarly, MDH-2c in agamosporous var. glabella (Fig. 20, lanes G) is coded by 
an allele expressed in sexual var. glabella (Fig. 17, lanes M; Fig. 20, lanes 2, 7, 8) 
but not found in hundreds of analyzed sporophytes of var. occidentalis. 


DISCUSSION 


C titi th the origins of agamosporous taxa in the P. 
glabella cacti can be tested by analyzing the taxon-specific electrophoretic 
enzyme phenotypes of the derivative and potentially progenitor taxa. This study 
tests hypothesized allopolyploid derivations in which P. atropurpurea is one of 
the progenitor taxa and autopolyploid derivations from the respective sexual 
taxa in the complex. Present hypotheses of allopolyploidy propose that either 
(1) P. atropurpurea crossed with sexual P. glabella var. occidentalis giving rise to 
both var. simplex and the agamosporous race of var. glabella or (2) P. 


58 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 2 (1988) 


atropurpurea crossed with P. glabella var. occidentalis giving rise to var. 
simplex and with the sexual race of var. glabella giving rise to the agamosporous 
race of var. glabella. Both allopolyploid hypotheses involve a triploid sperm 
from agamosporous P. atropurpurea and a haploid egg from one of the sexual 
taxa in P. glabella, and in both cases the offspring should be agamosporous 
tetraploids, as var. simplex and the agamosporous race of var. glabella are. 
However, because P. atropurpurea would contribute three genome doses via its 
triploid sperm versus a single genome dose via the haploid egg of the female 
parent, allopolyploidy also requires the enzyme banding patterns of P, 
atropurpurea to be strongly expressed in the tetraploid offspring. As detailed 
above and summarized below, the electrophoretic results obtained clearly 
preclude the involvement of P. atropurpurea in the origins of the tetraploids. 

At PGM-1, the apparently invariant band in P. atropury is not expressed at 
allin var. simp! d is not exp d in var. glabella at the intensity expected if 
this triploid PGM-1 genotype were a constituent of the tetraploid genome of var. 
glabella. Although the tetraploid varieties differ from each other and from P. 
atropurpurea in their PGM-2 banding patterns, the patterns themselves do not 
preclude the involvement of the P. atropurpurea PGM-2 genotype in the 
genomes of vars. simplex and glabella. This is because the single strong PGM-2 
band of P. atropurpurea aligns with a comparably strong band in the 
agamosporous tetraploids. All PGM bands of var. simplex are found in var. 
occidentalis and the only observed source for the slower PGM-2 band of var. 
simplex is var. occidentalis. All PGM bands of agamosporous var. glabella are 
found in sexual var. glabella except for the orphaned anodal PGM-2 band, and 
the only observed source for the weak cathodal PGM-1 band of agamosporous 
var. glabella is sexual var. glabella. 

At PGI-2, the single band of P. atropurpurea is only slightly slower than the 
comparable band in the tetraploids, but this pattern has been observed in 
numerous gels using plants from a very broad geographic range. The band from 
P. atropurpurea is not found in the tetraploids and therefore P. atropurpurea has 
not contributed its PGI-2 genotype to them. Because var. occidentalis and sexual 
var. glabella both produce the PGI-2 allozyme found in var. simplex and in 
one-banded populations of agamosporous var. glabella, PGI itself is not 
sufficient to determine which sexual entity may have been the progenitor of var. 
simplex and those populations of var. glabella. These PGI-2 data are consistent, 
however, with an origin of var. simplex from var. occidentalis as suggested by 
PGM-2 and with an origin of agamosporous var. glabella from sexual var. 
glabella as suggested by PGM-1. The three-banded PGI-2 pattern of 
agamosporous var. glabella includes a more anodal allozyme (and heterodimer) 
for which the only observed source is sexual var. glabella. 

The TPI-1 allozyme patterns of the three agamosporous taxa each migrate to 
different positions, and the constituent bands are not aligned from taxon to 
taxon. As with PGM-1 and PGI-2, the electrophoretic mobility of TPI-1 in P. 
atropurpurea indicates that the triploid TPI-1 genotype of that species is not a 
constituent of the tetraploid genotypes of the agamosporous varieties. At TPI-2, 
tL SRF a Ly eek ge fuk. Be 4 41 Aes 


kK i ie Ae + 
lel £420 VELL ith Litto 


AUT LGAAG aGictt 


tr 


G. J. GASTONY: PELLAEA GLABELLA COMPLEX 59 


what would be expected if the genotype of P. atropurpurea comprised 
three-fourths of the genomes of the tetraploid varieties. The TPI patterns of 
Figure 13, in which true hybrids are compared to the taxa under discussion, 
confirm this and reject eg aerate involving triploid P. atropurpurea in the 
derivations of var. simplex an var. glabella. At TPI-1, the slower 
three-banded pattern common ‘to var. occidentalis and sexual var. glabella 
characterizes agamosporous var. glabella and is combined with three doses of 
the faster three-banded pattern of var. occidentalis to yield the heterozygous 
TPI-1 pattern of var. simplex. Thus TPI-1 data are consistent with an autoploid 
derivation of agamosporous var. simplex from sexual var. occidentalis and with 
an autoploid origin of agamosporous var. glabella from sexual var. glabella or 
from sexual var. occidentalis. Because TPI-2 patterns are identical in the 
agamosporous tetraploids and because the requisite coding alleles are found in 
both sexual taxa, TPI-2 itself is not definitive in establishing 
progenitor-derivative lineages although it is consistent with interpretations 
based on other enzymes. 

Pellaea atropurpurea cannot contribute to the SkDH patterns of the tetraploid 
varieties because they lack the anodal band of most P. atropurpurea 
populations. Hybridization involving the more cathodal single-banded SkDH 
pattern of P. atropurpurea (as in Fig. 14, lane 6) would yield a tetraploid in which 
the cathodal band has three times the intensity of the anodal band, a pattern not 
observed in vars. simplex and glabella. Both allozymes of var. simplex are 
observed in var. occidentalis, but only the stronger more anodal band of 
agamosporous var. glabella has been observed in sexual var. glabella, the faint 
cathodal band being orphaned or too infrequent for detection in samples 
available from the sexual populations 

Although genetic interpretations of the AAT patterns in this complex are not 
presently available, observed patterns of triploid P. atropurpurea could not 
combine with the pattern of any haploid genotype to yield the AAT phenotypes 
of tetraploid vars. simplex and glabella. Thus AAT at least provides an 
additional independent data set inconsistent with allopolyploid hypotheses 
involving P. atropurpurea. 

For MDH-2 and MDH-3, various sporophytes of P. atropurpurea express the 
same bands as the agamosporous tetraploids. For MDH-1, P. atropurpurea 
expresses the same band as agamosporous var. glabella and the major band 
expressed by var. simplex. Thus MDH data do not exclude P. atropurpurea from 
the ancestry of th ids as other enzyme data do. However, 
MDH data are consistent with durbvations of the agamsporous tetraploids from 
the sexual diploids alone. Moreover, based on these extensive survey data of the 
four taxa within the P. glabella complex, the allele coding the cathodal MDH-1 
allozyme of var. simplex (MDH-1d) could only have come from sexual var. 
occidentalis and the allele coding the cathodal MDH-2 allozyme of 
agamosporous var. glabella (MDH-2c) could only have come from sexual var. 
glabella. Even without the corroborating data from PGI, PGM, and TPI above, 
these MDH data suggest that agamosporous tetraploid var. simplex is an 
autopolyploid derivative of sexual diploid var. occidentalis and that 


60 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 2 (1988) 


Fic. 21. Sporophytes representing the P. glabella complex and P. atropurpurea. Acronyms identify- 
ing populational sources are as in Table 1. Millimeter bar represents reduction by 0.25. 


agamosporous tetraploid var. glabella is an autopolyploid derivative of sexual 
diploid var. glabella. 

Thus the agamosporous tetraploids do not express the electrophoretic 
phenotypes expected for at least six isozymes (PGM-1, PGI-2, TPI-1, TPI-2, 
SKDH, AAT) if P. atropurpurea were one of their progenitors. Therefore P. 
atropurpurea could not have crossed with sexual var. occidentalis to give rise to 
var. simplex to the west and var. glabella to the east as proposed in one of Tyron’s 
(1957) hypotheses and as accepted for var. simplex (as P. suksdorfiana) by 
Lellinger (1985). Nor are the observed electrophoretic patterns those expected if 
P. atropurpurea sired the agamosporous taxa through respective hybridizations 
with the two sexual taxa now known in the P. glabella complex. Furthermore, as 
seen in Fig. 21, the morphologies of agamosporous tetraploid vars. glabella (G14) 
and simplex (S2) are not what should be expected if the triploid genome of P. 
atropurpurea (A3) were combined with the haploid genome of var. occidentalis 
(04, 06) or sexual var. glabella (M2) to produce them. The hypotheses involving 
P, atropurpurea in the derivation of the tetraploid agamosporous varieties of P. 
glabella are therefore rejected. 

An alternative hypothesis for the derivations of the agamosporous taxa in the 
glabella complex is that they arose through hybridizations between sexual var. 
glabella and var. occidentalis. Under this hypothesis, the electrophoretic 
differences observed between the agamosporous taxa would be attributable to 


G. J. GASTONY: PELLAEA GLABELLA COMPLEX 61 


TABLE 2. Occurrences In the Sexual Taxa of the Allozyme Bands That Characterize the Agamospor- 
ous Taxa of the Pellaea glabella Complex. 


PGM-1 PGM-2  PGI-2 TPI-1 TPI-2 SKDH MDH-1 MDH-2 


simplex 
anodal band! both? both both occi? both both both both 
cathodal band occi both both occi occi 

glabella 
anodal band both orbo* glab® both both both both both 
cathodal band _ glab both both both orgl® glab 


a } he ee OF shes * 
10r the Omy Dald for UlIS 1IsOzyme 
2T. me £ ras 44] llal is ai: sila oe Sa ag & 


1 
a 
Ll 
1 


a 
se a 
me 


is found only in var. occidentalis. 
is orphaned t f d in eit 
is found only in sexual var. glabella. 


BA Se gnme Ee 4 


The allele coding this band is found in both agamosporous taxa and in var. occidentalis but is or- 
phaned from sexual var. glabella. 


independent origins of these taxa from sexual individuals of differing 
genotypes. The electrophoretic data presented here cannot directly reject this 
hypothesis. For each of the isozymes summarized in Table 2, the genes coding at 
least one of the bands of the agamosporous tetraploids could be derived from 
either sexual variety. Thus var. simplex could be a hybrid of the two sexual taxa 
as long as var. occidentalis contributed the allele coding the cathodal allozyme 
of PGM-2, SKDH, and MDH-1 and the anodal allozyme of TPI-1. Similarly, 
agamosporous var. glabella could be a hybrid of the two sexual taxa as long as 
sexual var. glabella contributed the allele coding the cathodal allozyme of 
PGM-1 and MDH-3 and the anodal allozyme of PGI-1. 

Two lines of evidence, however, argue against this hypothesis. One is the 
gross morphology of the sporophytes of these taxa. The agamosporous 
tetraploids simply do not look like crosses between the sexual taxa. Instead, as 
seen in Fig. 22, tetraploid var. simplex (S2) looks like a robust form of diploid 
var. occidentalis (04) and tetraploid agamosporous var. glabella (G14) looks like 
a robust form of diploid sexual var. glabella (M2). Wagner et al. (1965), for 
example, were unable to find morphological distinctions between the two 
reproductive types of var. glabella except in the number of spores per 
sporangium. 

The second line of evidence concerns which sexual taxon is the source of the 
allozymes unique to the respective agamosporous taxa. In all cases in which var. 
simplex and agamosporous var. glabella are distinguished from each other by an 
alloyzme that they do not share and that is not found in both sexual progenitors 
(Table 2), the distinguishing allozyme of var. simplex always comes from var. 
occidentalis and the distinguishing allozyme of agamosporous var. glabella 
always comes from sexual var. glabella. If the agamosporous taxa were hybrids 
of the two sexual taxa, one should expect a more random distribution of the 

inctive allozy f the al unique to sexual var. occidentalis 


J OU Vs 


62 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 2 (1988) 


= 3 


Fic. 2 Sporophy ites 


trating tl taxa in the P. glabella complex. sr ms iden- 
tifying reba ocd sources are as in Table L Millimeter bar represents reduction by 0.3 


should occur in agamosporous var. glabella and some of the allozymes unique to 
sexual var. glabella should occur in agamosporous var. simplex. This is not 
observed. On the contrary, the morphological and electrophoretic data are 
consistent with the hypothesis that var. simplex is an autopolyploid derivative 
of var. occidentalis and agamosporous var. glabella is an autopolyploid 
derivative of sexual var. glabella. This study accepts the latter hypothesis and 
invokes the mechanism of unreduced spores (Gastony, 1986) to account for the 
tetraploidy of the agamosporous derivatives. 


TAXONOMIC CONCLUSIONS 


The taxonomic interrelationships of P. atropurpurea and the three P. glabella 
taxa known before 1965 have long been problematic, as indicated by the 
synonymies in Tryon’s revision (1957). Morphological, ploidal, and ecological 
differences justifying taxonomic distinction between these taxa have been 
summarized and discussed by several authors (Rydberg, 1900; Butters, 1917, 
1921a, 1921b; Tryon, 1957; Tryon & Britton, 1958; Rigby & Britton, 1970; 
Brunton, 1979; Lellinger, 1985). Enzyme electrophoretic data presented above 
exclude P. atropurpurea from the phylogeny of the P. glabella taxa and provide 
new insights into the progenitor-derivative relationships of the four taxa now 
known in the complex. These phyletic relationships should be reflected in a 
revised taxonomy of this group. The evolutionary lineage of var. occidentalis 
with its autopolyploid agamosporous derivative var. simplex shows substantial 
morphological discontinuity with the lineage of sexual var. glabella and its 


G. J. GASTONY: PELLAEA GLABELLA COMPLEX 63 


autopolyploid agamosporous derivative (Figs. 21, 22), and this correlates with 
the electrophoretic discontinuity detailed above. These two lineages should 
therefore be recognized as separate species. At specific rank, the two sexual 
progenitors must be known as Pellaea glabella and P. occidentalis. 

In addition to their ploidy differences, the sexual and agamosporous taxa of P. 
glabella in this narrow revised sense are reliably distinguished at the 
morphological level only by the number of spores per sporangium and the sizes 
of cells influenced by ploidy. Electrophoretically detected genotypes of the 
agamosporous tetraploid are a subset of the genetic variants found in the 
progenitor sexual diploid. In living condition, the sexuals tend to have a more 
bluish tint than the more greenish tinted tetraploids, and the leaf segments of the 
tetraploids tend to be broader and more triangularly shaped than those of the 
sexual diploids. However, in common greenhouse culture none of these 
distinctions is consistently reliable. Moreover, both the diploids and the 
tetraploids survive equally well under mon greenh cultural conditions, 
reflecting their similar physiological-ecological amplitude. Thus, the 
agamosporous taxon exhibits neither the substantial morphological 

li tinuity nor the genetic di tinuity from its sexual progenitor that would 

be implied by separating them at the species level. The electrophoretic 
distinction of the true-breeding agamosporous tetraploid derives simply from 
gene dosage effects associated with autopolyploidy and banding pattern fixation 
caused by lack of genetically significant segregation in agamosporous 
sporogenesis. The distinction between these taxa therefore seems best 
recognized at the level of varieties. 

In the case of P. occidentalis as well, the ag p totetraploid is fixed 
for a subset of the sexual’s alleles, much as any one population of the sexual 
diploid might encompass only part of the combined variation of all the diploid 
populations. As in the preceding case, this degree of genetic continuity between 
sexual progenitor and agamosporous derivative argues against distinction 
between these taxa at the rank of species. In this case, however, there is slight 
morphologial discontinuity (perhaps partially attributable to a gigas effect in the 
tetraploid), so that the diploids and tetraploids have long been distinguished 
taxonomically at the rank of variety (Butters, 1917; Tryon, 1957) or even species 
(Butters, 1921a, 1921b; Brunton, 1979; Lellinger, 1985). The morphological 
discontinuity described by the cited authors and the physiological differences 
indicated by the tetraploid’s prolonged survival under greenhouse cultural 
conditions soon lethal to the diploids have an underlying, although 
unquantified, genetic basis. These differences indicate a greater degree of 
genetic distinction between the infra-specific taxa of P. occidentalis than 
between those of P. glabella. This greater degree of divergence within P. 
occidentalis should be reflected in the taxonomic rank assigned to its 
infraspecific taxa. Thus in contrast to varietal distinction within P. glabella, the 
nearly disjunctly distributed infraspecific taxa of P. occidentalis are 
distinguished at the slightly higher rank of subspecies. — 

The foregoing considerations are reflected in the following revision of 
taxonomic relationships within the P. glabella complex sensu Tryon (1957) and 
Wagner et al. (1965). 


64 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 2 (1988) 


1. Pellaea glabella Mett. ex Kuhn, Linnaea 36:87. 1869. 
1a. Pellaea glabella var. glabella (Fig. 1, GLAa; Fig. 22,G14). 

Although I have not examined the type (Missouri, Kimmswick near St. Louis, 
presumably B), its provenance and the more common occurrence of the 
tetraploid promote the assumption that it represents the 32-spored 
agamosporous tetraploid. See Tryon (1957, p. 141) for full synonymy. 
1b. Pellaea glabella var. missouriensis Gastony, var. nov. (Fig. 1, GLAs; Fig. 22, 

M2).—T ype: Missouri, Jefferson Co., river bluffs 10 miles W of DeSoto, near 
jct. of road H and Big River, Gastony 83-34-47 (IND; isotypes GH, MICH, 
MO); paratype (from same locality): Farrar s.n. (MICH). 

A varietate glabella varietas haec distinguitur sporarum numero hic 64 per 
sporangium illic 32, chromosomatum numero hic 29 IJ illic 58 II. 

The varietal epithet for this 64-spored, sexually reproducing, diploid variety is 
based on its discovery by Donald Farrar in Missouri (Wagner et al., 1965) and on 
the restriction of its present known range to the small area of Missouri (Fig. 1, 
GLAs) encompassing the localities cited in Table 1 and in Wagner et al. (1965). 
Occasional pentaploid hybrids between the two varieties of P. glabella were 
discovered in populations from Missouri during this study. These are 32-spored 
and their electrophoretic phenotypes combine those of the parental varieties. 


2. Pellaea occidentalis (E. E. Nelson)Rydb., Mem. New York Bot. Gard. 1:466. 


1900 
2a. Pellaea occidentalis subsp. occidentalis (Fig. 1, OCC; Fig. 22, 04 and 06). 
This is the taxon known as Pellaea glabella var. occidentalis (E. E. Nelson) 
Butters in Tryon (1957, p. 139) where full synonymy can be found. 


2b. Pellaea occidentalis subsp. simplex (Butters) Gastony, comb. et stat. nov. 
(Fig. 1 SIM; Fig. 22, S2).—Pellaea glabella var. simplex Butters, Amer. Fern 
J. 7:84. 1917. 
This is the taxon known as Pallaea glabella var. simplex in Tryon (1957, p. 
142) where full synonymy can be found. 


ACKNOWLEDGMENTS 


Research was supported by NSF grants BSR 8206056 and BSR 8516666. I thank Anne L. Terry, 
Christopher J. Scelonge, and especially Valerie R. Savage for careful k as laboratory technicians, 
George Yatskievych and Charles Werth for population samples, David Barrington, Donald Britton, 
Ralph Brooks, Daniel Brunton, Roy Clarkson, James Montgomery, Steve Hill, Alice Tryon, Charles 
Werth, and Michael Windham for help in locating populations, and Gordon Goldy and the Arthur 
Chamberlin family for permission to collect representative specimens on their properties. 


LITERATURE CITED 


ALTSCHULER, M. I. and D. ScHwartz. 1983. Minor electrophoretic forms of maize alcohol 
ydrog their in vitro production by B ptoethanol. Maydica 28:401—410. 

BRUNTON, D.F. 1979. Taxonomy, distribution, and ecology of the cliff-brake ferns (Pellaea: 
Polypodiaceae) in Alberta. Canad. Field-Naturalist 93:288-295. 

—__——and J. D. Larontarne. 1974. The distribution of Pellaea in Quebec and eastern Ontario. 
Naturaliste Canad. 10:937—939., 

BrYAN, F. and D.E. Soitis. 1987. Electrophoretic evidence for allopolyploidy in the fern 

olypodium virginianum. Syst. Bot. 12:553-561. 


G. J. GASTONY: PELLAEA GLABELLA COMPLEX 65 


BuTTERS, F. K. 1917. Pellaea atropurpurea (L.) Link and Pellaea glabella Mett. ex Kuhn. Amer. Fern 


.1921a. , new western fe orieng se dscns pia Saige J. 11:39-40. 
————.. 1921b. A ican f Ill. Pellaea glabella 
and its western segregates. Amer. Fern J.11 
Gaston, ee u 1986. nega pill evidence for . Be of fern species by unreduced spores. 
r. J. Bot. 73:1563 
pin DG. Pace fey Chik dea and cytosolic isozymes of the homosporous fern 
Athyrium filix-femina. Amer. J. Bot. 70:1409-1415 
and OTTLIEB. 1982. Evidence for genetic heteeokyanaity in a homosporous fern. 
Agee J. Bot. 69:634—637 
bh ——_—. 1985. Genetic variation in the homosporous fern Pellaea andromedifolia. 
mer. J. Bot. 24; 2-207 
sophia. L. D. 1981 t tic evid 1 pl ti sdikesoveses! 7:1-46. 
——. 1982 


sg 


Giga ion dd ti f isozvmes in 1 plants 2 Sai 216: 


HAUFLER, C. H. 1985. Enzyme rine Ht and modes of evolution in Bommeria ea Syst. 
a 10:92—104. 
M.D. WInpHaM, D. M. BRITTON, and S. J. RoBINSON. 1985. Triploidy and its evolutionary 
sini cance in Cystopteris protrusa. Canad. J. Bot. 63:1855—1863. 
and D.E , SOLUS. 1986. peeaiaee bia suggests a see oe poe ferns with high 
roc. Natl. Acad. U.S.A. 8 393 
LELLINGER, D. B. 1985, A Deld manual of ie diy & fern allies of ie ie States & Canada. 
Institution P: 
Love, A. and D. LOVE. 1975. Plant Chromosomes. oa z: J. Cramer 
Soe - and L. D. Gorriies. Evidence for dcpnnaticn of the -strictaral genes coding plastid 
d cytosolic isozymes of triose phosphate isomerase in diploid species of Clarkia. 
Palace 105:421—436 
RicBy, S. 1968. An fiivnatigatlon of Pellaea glabella Mett. ex Kuhn and Pellaea atropurpurea (L.) 
ink and their relationships. Master’s Thesis University of Guelph, Guelph, Ontario, 
sa 
and D. M. BRITTON. 1970. The distribution of Pellaea in Canada. Canad. Field-Naturalist 
84:137-144. 
Roose, M. L. and L. D. GotriieB. 1976. Genetic and biochemical consequences of polyploidy in 
Tragopogon. Evolution 30:818—830. 
RypBERG, P. A. 1900. Catalogue of the flora of Montana and the Yellowstone national park. Mem. 
New York Bot. Gard. 1:1-492. 
SELANDER, R. K.. M. H. Smit, S. Y. YANG, W. E. JOHNSON, and J. B. GENTRY. 1971. IV. Biochemical 
polymorphism and systematics in the genus Peromyscus. I. Variation in ex old-field 
mouse (Peromyscus polionotus). Studies in genetics VI. Univ. Texas Publ. 7103:49-90. 
Soxtis, D. E. 1986. Genetic evidence for diploidy in Equisetum. Amer. J. Bot. 73: nani 
———,, C.H. Haurter, D.C. DARROW, | and G. J. GasToNY. 1983. sesame gel electrophoresis of 
ferns: A compilation of 8 
Amer. Fern J. 73:9—27. 
and P. S. Souris. 1986. Electrophoretic evidence for inbreeding in the fern Botrychium 
virginianum (Ophioglossaceae). Amer. J. Bot. 73 
TRYON, A. F. 1957. . revision of the fern iad Pellaea section Pellaéa: Ann. Missouri Bot. Gard. 
44:1 


rele Spores, chromosomes, and relations of the fern Pellaea atropurpurea. Rhodora 
ae D. ie ered 1958. Cytotaxonomic studies of the fern genus Pellaea. Evolution 


WacNeR. W.H., D. R. FARRAR, and K. L. CHEN. 1965. A new sexual form of Pellaea glabella from 
ogni fae Fern J. 55: 1vt- ink 
WALKER, T.G. 1 the fern genus Pteris L. Evolution 16:27—43. 
. 1979. eee Restoran of — Ge 87-132 in 1 The experimental biology of ferns, ed. A. F. 
Dyer. New York: Academic Pres 


66 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 2 (1988) 


fm ig eM a 


WEEDEN, N. ie 1983. Plastid isozymes. Pp. 139-156 in I. ] ing, part 
s. S. D. Tanskley and T. J. Orton. Amsterdam: Elsevier Science Publishers B.V 
WERTH, C. R. ‘Sl I. GUTTMAN, and W. H. EsHBauGH. 1985. Electrophoretic evidence of eaticadute 
evolution in the Appalachian Asplenium complex. Syst. Bot. 10:184-192 


APPENDIX: Interpreting the Non-Segregating Three-banded 
TPI-1 Pattern in Ferns 


A m4 pegs oe +h, 


2), one zone of TPI activity in ferns is commonly expressed as a fixed, 
asymmetrically stained, pa -banded peter. This pattern usually occurs as he mare snodally 
migrating 20 zone of any (TPE-1) )and 


= 
£. <. pa & ] 


or from | protein. 
ree : 5; ee Ce 41 . Jo gs 1 +1 Rae eae Pama 3 | 
The thine banded TPI-1 p p y the gene dup YI g 
sg : : Satan bb? 2 Loare 
way. The original tpi-1 gene! lerg (possibly jem) dupl d the duplicat 
Ls ad 4 PRA L + ae t . ae, ee | 
subsequently mutated. Thusthe p ded by the original f f the g i the f 
L rae REE 4 2 i ¥s r. b+ 4 t ] hel 3: ff a ] ca% a : 7 ; 3 and 
by k 8 gs gely g 
rs I ry j + 1 1 4 j I 4+ ae ip oa + I ++ j } 8 i Soca - 
4h mt j 3 LI } +1 


of the three bands is anny stained, the most cathodal i is eee stained, and the intensity of the 
central interlocus heterodimeric band is roughly equal to that of the normally stained cathodal 
homodimer. 


1 oe a 7 a2 2 j j re eae 8 : rie & 
l to devel lleli iant wi the original gene d Ifan allelic variant o 
the original gene does arise, hould t the all Jed b ll 
l rm 


+L, 
d ad y Jy Ste variant GilGit lu migrate t wu 
j ] } 


midway | between a and the protein coded by the eas gene. This, however, is not what we 
observe in ferns. ead, when the locus encoding TPI-1 is heterozygous, it codes two distinct 
three-banded oe plus cae Peron ReTE: ni — nics ten bands), put 
over rlapping or mathe overlapping bands and on the von of the anodal homodimers. This i is 
illustrated in Fig. 7, where lanes 5 and 7 show two variant three-banded TPI-1 patterns from 

rophytes of sexual diploid P. glabella var. occidentalis and lanes 6 and 10 show the pattern in 
another sporophyte in which at least seven bands are distinguishable at TPI-1. Application of the 
gametophytic segregation technique of Gastony and Gottlieb (1982) shows that this multi-banded, 
heterozygous, sporophytic pattern segregates into faster (lanes 3, 8) and slower (lanes 4, 9) 
three-banded patte rns in individual gametophytes derived from the sporophyte in lanes 7 and - 
Similar of TPI-1 patterns have been observed in sexually reproducing var. glabella (Fig. 
8, lanes 10-1 19) and i in Cystopteris protrusa (Haufler et al., 1985; Haufler & Soltis, 1986), indicating 


ret - a Se 45 a | 2 2 A rt £al.t 


Oe t=) I © | , it ti This is 
ause when an allelic variant of the original gene occurs, that variant allele is also expressed as a 


duplication, then whenever an allelic variant of the original gene arose, the duplicate gene would 
also have to generate a new allele, since each observed allelic segregant occurs as a three-banded 
pattern. It is simply improbable that each allelic variant of the original gene would always be 
associated with a new SE variant of Led aie cee gene in ver. occidentalis and i in. sexual var. 
glabella. It is even mo species 

Cystopteris protrusa en the allelic variants also occur as mires! banded patterns (Haufler « et al., 
1985; Haufler & Soltis, 1986). Analogous three-banded allelic variants may also underlie the 
complex cathodal TPI ners of diploid Polypodium virginianum (Bryan & Soltis, 1987). It is 
unlikely that independent 
_ the alternative ‘explanation = the three-banded TPI-1 pattern is that a gene which = be 
gous or a protein naturally susceptible during 


protein synthesis or post-translationally. When the gene is homozygous, the coded monomeric 


G. J. GASTONY: PELLAEA GLABELLA COMPLEX 67 


[24 


protein and a charge-modified f These monomers combine randomly 

to form the dimeric quaternary structure of the functional enzyme. The most anodal band of the 
three-banded pattern consists of less active, homodimeric, protein; the most cathodal band of the 
triplet consists of normally active, homodimeric protein; and the central band of the triplet 
represents hetrodomeric protein. Generally with dimeric enzymes, when two different monomeric 


homodimer:heterodimer:homodimer staining {ntansities. In the case of TPI-1 in the P. glabella 
complex and Cystopteris protrusa, because one of the monomers is less active than the other, their 
random association results in a ratio of activity oa is not 1:2:1 but rather approximately 


tpi-1, two allelically coded and two charge-modified forms of those monomers would be 
produced, and heterodimeric functional enzymes would be formed by all dimeric combinations of 
the four monomers. This would yield 10 enzyme bands if none were overlapping. The migration 
distances and staining intensities of the novel heterodimeric bands are predictable from the 


of the heterodimers. In the case of var. occidentalis, the predictions agree with observations, as in 
Fig. 7 lane 6. Thus the three-banded TPI-1 patterns in ferns are herein accepted as specified by a 
single tpi-1 >i a portion of whose protein product is regularly affected by post-translational 
charge modifica 
This neneaiere of these fixed three-banded TPI patterns 2 ee is — by a parallel 

situation with dimeric alcohol d 83). In maize, in 
vivo anodal minor ADH bands associated with prolonged storage of gen a could also be 
produced in vitro by treating enzyme extracts with §-mercaptoethanol. Experiments based on this 
observation lead Altschuler & Schwartz (1983) to conclude that the anodal minor ADH bands are of 
post-translational origin. Although some agent (or combination of agents) other than 
B-mercaptoethanol may cause post-translational modification of this TPI isozyme in ferns, it is 
noteworthy that B-mercaptoethanol (as 2-mercaptoethanol) is a component of the three grinding 

uffers most commonly used in electrophoresis of _— costes et sie ives) In the P. glabella 
complex, however, three-banded forms of TPI-1 p 
the grinding buffer. in maize ADE, wpe electrophoretic mobility of anodal minor bands was shee 
correlated wi 1 that o 
major band seo sikaeed | tlie: mobility of the minor bands to the same degree. This i is precisely shea la is 
observed with TPI in the ferns. 


American Fern Journal 78(2): 68—71 (1988) 


Marsilea scalaripes, A New Member of Marsilea 
section Clemys from the Asian Tropics 


Davin M. JOHNSON 
New York Botanical Garden, Bronx, New York 10458 


A specimen of Marsilea collected by E.J.H. Corner in Kedah, Malaysia, in 
1941 has been the basis for reports of the primarily Neotropical M. polycarpa 
Hook. & Grev. from the Asian tropics (Holttum, 1954; Launert, 1968). Both of 
these authors observed, however, that this specimen probably represented a 
distinct species. My revision of the New World species of Marsilea (Johnson, 
1986) afforded me an opportunity to study abundant material of Marsilea 
polycarpa and then to compare it to a fragment of Corner S.F.N. 38109 in the 
herbarium of the British Museum. Even from this small sample it was apparent 
that the specimen was of a species distinct from M. polycarpa, and a loan of the 
full sheet of the collection from the herbarium of the Singapore Botanic Gardens 
(SING) confirmed this. I provide here a full description of this new species, and 
discuss its relationship to the other members of the small but well-defined sect. 
Clemys of Marsilea. 


Marsilea scalaripes D.M. Johnson, sp. nov. (Fig. 1A-E)—Type: Malaysia, 
Kedah, 1941, Corner S.F.N. 38109 (SING!; fragment BM!). 


A Marsilea polycarpa Hook. & Grev. sporocarpiis ovoideis 3.14.5 mm longis, 
sporocarpio infimo 2.1—4.7 cm supra basim petioli portato, a M. crotophora D. M. 
Johnson sporocarpiis aspectu frontali tereti, 2.0-2.5 mm latis, ab ambobus 
pedunculo 3.1—7.8 mm longis recedit. 


Rhizome long-creeping, 1.0—1.5 mm thick, pale brown to stramineous, 
densely pubescent at apex, soon glabrate proximally, with internodes 1.25.0 cm 
long; nodal shoots suppressed, budlike; roots at nodes and 1—2 on internodes. 
Land leaves with canaliculate petioles 13.7-15.0 cm long, 0.7—1.3 mm thick, 
curved to straight, glabrescent; leaflets 16-17 mm long, 13.5-14.5 mm wide, 
sparsely pubescent adaxially, glabrous abaxially, lateral margins straight or 
slightly concave, terminal margin entire. Floating leaves with flexuous 
canaliculate petioles 9.8-11.7 mm long, 0.8-1.2 mm thick, glabrous; leaflets 

4-16 mm long, 12-14 mm wide, glabrous, with hydropoten uniformly 
distributed abaxially. Fertile leaves bearing 4-9 sporocarps on unbranched 
straight or slightly curving, weakly to strongly ascending peduncles 3.1—7.8 mm 
long, 0.5—0.6 mm thick, thinly pubescent to glabrate, in a row beginning 2.1—4.7 
cm above the petiole base and extending distally for 6-19 mm, the bases of the 
peduncles expanded and thus almost confluent at base. Sporocarps 3.1—4.5 mm 
long, 2.0-2.5 mm wide, and 2.0 mm thick, ovoid, terete in cross-section (i.e., 
frontal aspect), dark brown, thinly covered with matted hairs; raphe and teeth 


D. M. JOHNSON: MARSILEA SCALARIPES 69 


11cm] 


is ) at Hh 


Fic. 1. Marsilea scalaripes D. M. Johnson. A, st p in longit g tion i 
one wall. B, sporocarp, enlarged. C, habit, floating leaves (leaflets reconstructed). D, habit, land 
leaf. E, sporocarp-bearing portion of land leaf, enlarged. All based on Corner S.F.N. 38109 (SING). 


70 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 2 (1988) 


absent; sporocarp veins ca. 15—16, connected by a zigzag lateral vein 1/3 distance 
from the midrib. Sori 11, attached to a sorophore that tapers to a blunt tip, with 
1—3 megasporangia and an undetermined number of microsporangia per sorus. 
Megaspores 560—650 wm long and 480-500 wm wide, with a papilla 50-90 pm 
long. Microspores ca. 40 per sporangium, 35-50 ym in diameter [probably 
immature]. 

The specific epithet is from Latin signifying “ladder-foot,”’ alluding to the 
series of more or less parallel runglike sporocarp peduncles of the new species. 

Marsilea scalaripes is probably closest to M. polycarpa but differs in the high 
position of its sporocarps, longer peduncles, and longer sporocarps. The high 
position of the sporocarps on the petiole is like that of M. crotophora, a 
Neotropical species with the lowest sporocarp attached 1.1—7.5 cm above the 
petiole base; the sporocarps of that species are, however, pear-shaped, ovate in 
cross-section, and borne on strongly recurved peduncles only 1.7—2.7 mm long. 

Sporocarp length and shape are more reliable as taxonomic characters than is 
peduncle length. The peduncles on floating leaves of Corner S.F.N. 38109 were 
4.7—-7.8 mm long (x = 5.70+0.99, n= 10), while those of the land leaves were only 
3.1-5.2 mm long (x=4.08+0.62, n=8). Sporocarp length, in contrast, was 
3.1—4.5 mm on the floating leaves (x = 3.85+0.49, n=10), and 3.1-4.0 mm on the 
land leaves (x=3.50+0.26, n=8). This further supports the point made 
elsewhere (Bhardwaja, 1967; Johnson, 1986) that peduncle length is 
environmentally plastic. 

Marsilea scalaripes, although only known from a herbarium specimen, 
appears to be, like other members of sect. Clemys, a strictly aquatic species. The 
type specimen has samples of floating leaves and of land leaves, both of which 
are fertile. Only the Neotropical M. deflexa A. Braun (Johnson, 1986) and the 
African M. berhautii Tardieu (Kornas, 1983) are known to produce sporocarps 
abundantly i in water. Another character correlated with an aquatic life cycle in 
Marsilea is th simple plates in root metaxylem elements. 
Marsilea scalaripes, along with M. deflexa and M. crotophora of sect. Clemys, 
lacks these plates; all other species of Marsilea investigated to date have them 
(White, 1961; Mehra & Soni, 1971; Bhardwaja & Baijal, 1977; Johnson, 1986). 

The members of Marsilea sect. Clemys and their distributions now stand as 
follows: 1) M. deflexa, Mexico to South America; 2) M. berhautii, western 
tropical Africa, 3) M. crotophora, Mexico, Central America, and western South 
America; 4) M. scalaripes, Malay Peninsula; and 5) M. polycarpa, Mexico to 
seal South America, Greater Antilles, and the Society Islands of the South 

acific. 


ACKNOWLEDGMENTS 
Gardens se cae: specimens raga for my study, and W.R. Anderson of the. University of 
Michigan Herbarium for requesting and transferring specimens sent on loan. My visit to English 


herbaria was made Lawes by a Rac kham Dissertation Grant from the University of Michigan. I 
thank N. A. Murray and R 


tat 


D. M. JOHNSON: MARSILEA SCALARIPES 71 


LITERATURE CITED 


Buarpwayja, T. N. 1967. Pedicel attachment in Marsilea diffusa var. approximata in relation to 
habitat factor. Trop. Ecol. 8:17—21. 
, and J. Baijal. 1977. Vessels in rhizome of Marsilea. Phytomorphology 27:206—208. 
Ho.LtTTuM, R. E. 1954. A revised flora of Malaya. Vol. 2, Ferns of Malaya. Singapore: Government 


Printing Office. 

JoHNsoN, D. M. 1986. Systematics of the New World species of Marsilea (Marsil ). Syst. Bot. 
Monogr. 11:1-87. 

Korna§,J. 1983. Pteridophyta collected i th Nigeri Jnorthern Cameroon. Acta Soc. Bot. 


Poloniae 52:321—335. 

LAUNERT, E. 1968. A monographic survey of the genus Marsilea Linnaeus, I: the species of Africa 
and Madagascar. Senckenberg. Biol. 49:273—315. 

MeuRA, P.N., and S.L. Sont. 1971. Morphology of the tracheary elements in Marsilea and 
Pilularia. Phytomorphology 21:68-70. 

Waite, R. A. 1961. Vessels in the roots of Marsilea. Science 133:1073—1074. 


SHORTER NOTES 


Rediscovery of Gymnocarpium dryopteris in Arizona—During the summer of 
1986, biologists for the Coconino National Forest conducted riparian surveys in 
the East Clear Creek drainage of the Colorado Plateau. A high priority was placed 
on the development of an accurate plant species list since none was available for 
the area. Many new distribution records were recorded. Identification was 
accomplished with the help of taxonomists from the Deaver Herbarium at 
Northern Arizona University (ASC), Flagstaff, Arizona. 

A plant collection of particular interest was the Oak Fern, Gymnocarpium 
Dryopteris (L.) Newm., which was collected in Dane Canyon, along a perennial 
stream, approximately 19 kilometers (12 miles) SSE of the Blue Ridge Ranger 
Station in Coconino County. Two plants were discovered, and at the time of 
collection both plants were fertile. The plants were on a sandstone ledge 
adjacent to the creek bottom. The exposure was north-facing in a steep, narrow 
canyon. The associated vegetation was riparian shrub with a coniferous 
overstory. A portion of the plant (Boucher 500, ASC 48519) was collected on 17 
August 1987. 

The new locality for this widespread, primarily boreal species is only the 
second recorded station for Arizona. The first collection was made from Bonito 
Creek, in the White Mountains of Apache County, Arizona 23 July 1912, by L.N. 
Goodding. Goodding 1222 (ARIZ 6964, 6965) and Goodding 1252a (ARIZ 6968), 
are housed at the University of Arizona Herbarium, Tucson, Arizona. Recent 
attempts to relocate this original population have proven unsuccessful. At 
present, the only living plants known from Arizona are the two plants located in 
Dane Canyon.—Paut F. Boucuer, USDA Forest Service, Coconino National 
Forest, 2323 East Greenlaw Lane, Flagstaff, Arizona 86004. 


72 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 2 (1988) 


The Status of Asplenium viride in Oregon—Asplenium viride Huds. was 
discovered in Oregon in 1908 by William C. Cusick (Mason, G., Guide to the 
Plants of the Wallowa Mountains of Northeastern Oregon. Museum of Natural 
History, Univ. of Oregon, Eugene, 2nd ed., 1980). The label data on Cusick’s 
sheet (Cusick s.n., ORE) seem to pinpoint the locality: ‘‘Granitic cliffs, source of 
the Imnaha River, 9000 feet alt.” A pencilled annotation on the original label 
asks: “limestone cliffs?” Since this was the only historical record for green 
spleenwort in Oregon, I was interested in relocating the population, which had 
not been seen since 1908 and was believed to be ‘‘apparently extirpated” from 
Oregon (Kagan, J, S. Yamamoto & C. Levesque, Rare, Threatened & Endangered 
Plants and Animals of Oregon. Oregon Natural Heritage Data Base, Portland, 
1987.) 

The source of the Imnaha River is in the remote high country of the Eagle Cap 
Wilderness in the Wallowa Mountains of northeastern Oregon. Four main forks 
converge to form the Imnaha, which flows east through the Wallowa-Whitman 
National Forest. The headwaters encompass roughly 15 square kilometers of 
alpine and sub-alpine terrain. At least six peaks and many ridges rise above 2740 
meters (9000 feet) in the watershed. The bedrock is mostly pale grey or white 
limestone, but some outcrops of a white granite (granodiorite) are also exposed. 

In the summer of 1987 I examined the various headwaters of the Imnaha. | 
found a small population of Asplenium viride in the cirque at the head of the 
Middle Fork of the Imnaha River (T 5 S, R 45 E, S 9 NW 1/4 NW 1/4, Wallowa 
County). The site is at 45° 8’ 57” N, 117° 13’ 23” W, and is 1.5 km SSE of Sentinel 
Peak. The ferns were on a large, white Triassic limestone deposit in the Martin 
Bridge formation (Walker, G. W., Reconnaissance Map of the Oregon Part of the 
Grangeville Quadrangle, Baker, Union, Umatilla and Wallowa Counties, 
Oregon. U.S. Geological Survey, Denver, 1979). A rocky bowl and a line of 
north-facing cliffs are found on the south side of the stream between 2280 and 
2320 meters. The steeply angled limestone is shaded in the noon sun of August. 
Among the cool crevices, vertical cracks and overhanging ledges were 47 rosettes 
of A. viride. 

The site is near the treeline, and some scrub conifers grow around the cliffs. 
Green spleenwort was growing with Campanula rotundifolia L., Cystopteris 
fragilis (L.) Bernh., Pellaea breweri D. Eaton, Polystichum lonchitis (L.) Roth, 
Potentilla fruticosa L., Salix vestita Pursh, Saxifraga oppositifolia L., 
Thalictrum occidentale Gray, Valeriana sitchensis Bong., and Zigadenus 
elegans Pursh. 

Though speculative, it is possible the 1987 station could be the same as the 
1908 station. But this raises two questions. First, why is the 1908 report from 
460 meters higher? Cusick’s herbarium labels show he frequently overestimated 
elevations in the Wallowas. The poor maps of his day would account for this. 
Second, how could Cusick confuse limestone with granite? Both rocks are 
common and similarly colored in the Wallowas. Cusick may simply have 
guessed wrong about the substrate. 

The apparent rarity of green spleenwort both in 1908 and 1987 is 
circumstantial evidence that the two stations could be the same. Cusick 


SHORTER NOTES 73 


botanized in the Wallowas for years (he lived nearby) and only collected the 
species once. The recent search of each of the four forks of the Imnaha River 
yielded only one site. Spleenwort habitat, shady north-facing limy ledges at 
timberline, is scarce in the Imnaha drainage. Unsuccessful attempts were also 
made to locate A. viride in 1987 in the limy Hurricane Creek, Wallowa River, and 
Frances Lake basins of the Wallowas. 

Green spleenwort has a circumboreal range and is too common in the northern 
United States for federal threatened or endangered status under the Federal 
Endangered Species Act as amended. Green spleenwort can not be considered 
threatened or endangered within Oregon at this time, according to current state 
statutes. The Oregon colony is distant from the nearest roads and hiking trails, 
and therefore remote from their implied threats. Although the U.S. Forest 
Service allows domestic sheep and horse grazing in the Eagle Cap Wilderness, 
the Asplenium is not threatened or endangered by grazing since it grows on 
inaccessible cliffs. However, the solitary population is small and disjunct, and 
thus should remain on the Oregon and Wallowa-Whitman National Forest 
sensitive species lists. Because the A. viride population is small, I limited my 
voucher (Zika 10431 ORE, OSC) to a few fronds from large plants and did not 
uproot rosettes. 

I thank the Oregon Natural Heritage Data Base and the Wallowa-Whitman 
National forest for supporting this field work —PETEr F. Z1xa, District Botanist, 
Bureau of Land Management, Eugene District, PO Box 10226, Eugene, Oregon 
97440. 


Seven Clubmosses New to Arkansas.—Until recently, only Lycopodium 
lucidulum and L. appressum were known to occur in Arkansas (Taylor & 
Demaree, Rhodora 81:503-548, 1979). Lycopodium lucidulum Michx., first 
reported by Bowers and Redfearn (Amer. Fern J. 57:91—92), 81:503-548, 1979), is 
represented in Arkansas by several populations in the Boston Mountains that are 
extreme southwestern disjuncts from the species’ metropolis in northern and 
eastern North American. In contrast, L. appressum (A. Chapman) F. Lloyd & L. 
Underw., is widely dispersed to the south along the Gulf Coastal Plain, but in 
Arkansas, it was known (Taylor & Demaree, Rhodora 81:503—548, 1979) from 
only five southcentral counties (Clark, Hempstead, Ouachita, Saline, and Union 
cos.). During the last three years (1985-1987), collectors encountered four 
Lycopodium species and three interspecific hybrids that are new to the Arkansas 
flora. In addition, they located L. appressum in two more counties (Calhoun and 
Hot Spring cos.). These discoveries significantly increase the number of species 
and distribution of Lycopodium in Arkansas. 

Lycopodium digitatum A. Braun was collected in Arkansas for the first time 
from Benton Co. on 11 Nov 1987. Edwin B. Smith, curator of the herbarium at the 
University of Arkansas at Fayetteville kindly sent the voucher (Mottesheard 55, 
UARK) to us for verification. The plant was located on a slope within 
pine/hardwood forest, 5—7 km ESE of Siloam Springs. No other representative 
from Lycopodium sect. Diphasium is known from Arkansas. This population is 


74 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 2 (1988) 


disjunct 400 km from its nearest populations to the north in Iowa and Illinois 
(Peck, Contr. Milwaukee Pub. Mus. Geol. Biol. 53:1-143, 1982) and to the east in 
Kentucky (Johnson & McCoy, Amer. Fern J. 65:29, 1975). 

Three Lycopodium species and three interspecific hybrids new to Arkansas 
were discovered in 1985 and 1986 by C. Amason, E. Bridges, S. Orzell, C. J. Peck, 
and J. H. Peck. While inspecting eight L. appressum colonies located by Amason 
in Calhoun Co., it was noted that these populations contained additional 
Lycopodium taxa (Orzell & Bridges, Phytologia 64:81—144, 1987). This discovery 
led to a search that located 14 more colonies in Calhoun Co. Analysis of these 
populations resulted in the addition of six taxa to the vascular flora of Arkansas. 
Growing with L. appressum were the species L. carolinianum L., L. 
alopecuroides L., and L. prostratum Harper and their hybrids L. x bruceii 
Lellinger (L. appressum X prostratum), L. x copelandii Eiger (L. alopecuroides 
x appressum), and L. alopecuroides x prostratum. The populations were found 
at disturbed sites with moist sand or clay soil, such as graded timber-access road 
ditches, pine plantation reforestation sites, barrow pit ponds, and gravel pit 
operations. The authors’ voucher specimens are deposited in the herbarium of 
the University of Arkansas at Little Rock (LRU). 

he phenomenon of many taxa of a pteridophyte genus commonly growing 
together, recently known as a “genus community” (Wagner & Wagner, Taxon 
32:53-62, 1983), provides an ideal opportunity to compare and contrast species 
of a genus that are difficult to identify when found growing alone. A difficulty in 
identification exists with respect to members of the L. inundatum complex in 
the Gulf Coastal Plain where they are reported as commonly occurring in genus 
communities (Bruce, Univ. Microfilm 76-9355, 1975). Observations of a 
Lycopodium community in neighboring Kentucky (Johnson & McCoy, loc. cit.; 
Cranfill, Amer. Fern J. 71:97-100, 1981) suggest that L. appressum and its 
hybrids may be more winter-hardy than L. alopecuroides and L. prostratum. In 
that all parents and hybrids were found in Arkansas, we suggest either that the 
colonization process is more recent in Arkansas, or that the residence time of 
winter-sensitive parental species may be more prolonged in Arkansas. In either 
case, the Arkansas populations provide exceptional opportunities for com- 
parative studies of long-term population events among the species and their 
hybrids.—JaMes H. Peck and Carot J. PEcK, Biology Department and Office of 
Testing and Student Life Research, University of Arkansas—Little Rock, Little 
Rock, Arkansas 72204. 


REVIEWS 75 


REVIEWS 


“Bibliografia comentada sobre pteridofitas de México,’ by Ramon Riba and 
Armando Butanda. 1987. 88 pp. Consejo Nacional de la Flora de México, 
Apartado Postal 17-584, Delagacién Miguel Hidalgo, 11410 México, D.F. $10.00 
U.S. (incl. postage). ISBN 968-6144-02-1. 

Fern enthusiasts will find this extensive bibliography listing the publications 
on Mexican pteridophytes for the last 25 years (since publication of An 
annotated bibliography of Mexican ferns, by G.N. Jones, 1966) extremely 
useful. Most citations are followed by a short note indicating their contents. 
Following the Introduction, there is a catalogue of the books and periodicals 
used in the bibliography. The 56 pages of citations are followed by these very 
useful indices: Bibliographies, lists and glossaries; Ethno-botany; Geography: 
States of Mexico; Geography: Countries and regions of America; Systematics; 
Vegetation and Floristic Studies. The seven pages of an index to authors is 
followed by three pages of additional citations. The authors are to be 
commended for making this very valuable tool available to all botanists.—A. J. 
SHarp, Department of Botany, University of Tennessee, Knoxville, Tennessee 
37996. 


“Liebmann’s Mexican ferns: His itinerary, a translation of his ‘Mexicos 
Bregner,’ and a reprinting of the original work,” by John T. Mickel with 
contributions by Rogers McVaugh, Sven Karell, and Henrik Balslev. 1987. 350 
pp. Contr. New York Bot. Gard. 19. Available from Scientific Publications 
Office, New York Botanical Garden, Bronx, NY 10458. $30.50 (U.S.), $31.50 
(non-U.S.), paperbound. USBN 0-89327-324-4. 

Although Liebmann’s (1849) work on the ferns of southern Mexico (Puebla, 
Veracruz, and Oaxaca) is largely outdated taxonomically as a result of modern 
accounts, it is still important from an historical perspective to study his 
collections, itinerary, and published work in order properly to afix names to 
collections. Most of the publication by Mickel et al. is an English translation of 
Liebmann’s work and a reprinting of the original Danish version (Mexicos 
Bregner, en systematisk, critisk, plantgeographisk Underségelse. Kongel. 
Danske Vidensk. Selsk. Skr., Naturvidensk. Afd., V. 1:151-322. 1849). The 
translation contains the modern names (according to Mickel) of the plants 
treated by Liebmann and an index to all names, both present-day and those used 
by Liebmann. What makes this work most useful is a discussion by McVaugh of 
Liebmann’s itinerary and a gazetteer of localities given in Liebmann’s published 
work and on his collections in Copenhagen. Many of these localities are obscure 
(some no longer extant), not found on modern maps, and never again revisited by 
botanists, who nowadays often travel by car rather than by foot. McVaugh’s 
jualed knowledge of the travels of early botanists in Mexico will be essential 
to anyone trying to localize one of Liebmann’s collections. This book will be 
welcomed by all who deal with plants of southern Mexico and by those 
interested in early botanical narrative-—ALAN R. Smitu, Herba D t t 
of Botany, University of California, Berkeley, California 94720. 


76 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 2 (1988) 


“Monograph of the Neotropical fern genus Polybotrya (Dryopteridaceae),”’ by 
R. GC. Moran. 1987. Illinois Natural History Survey Bulletin 34(1): 1-138. ISSN 
0073-4918. Available free of charge from the Illinois Natural History Survey, 
Natural Resources Building, 607 East Peabody Drive, Champaign, Illinois 61820, 
USA. 

From an unlikely source, the Illinois Natural History Survey, comes an 
excellent new work on Neotropical ferns: a revision of the 35 species of the 
hemiepiphytic/climbing-fern genus Polybotrya, by R. C. Moran, currently of the 
Missouri Botanical Garden. The work follows a traditional monograph format, 
with discussions of taxonomic history, ecology, geography, morphology and 
anatomy, and phylogeny preceding the key and descriptions. All species 
accounts include synonymy, full descriptions, citation of published 
illustrations, discussion, and citation of specimens; each account is 
accompanied by a clear and detailed full-page line drawing by the author. Ten 
new species are described. A bibliography, distribution maps, a list of 
exsiccatae, and a general index complete the volume. 

I found that the monograph provided for easy identification of i The 
key is straightforward, once some specialized terms have been mastered by the 
user, and dite characters are illustrated. The key emphasizes characters of 
the sterile leaves (in fact, sterile material is required for many of the leads), and 
you can get to couplet 37 (out of 42) on sterile material alone. When the key did 
not bring a positive determination, I was able to make an identification by 
matching my specimen with an illustration. A listing of species by country (p. 
10) is another handy identification aid. 

In the section on Morphology and Anatomy are recounted some of the unusual 
biological aspects of Polybotrya species, such as the presence of aerophores and 
mucilage on the rhizome and petiole, and presence of nectaries on the rachis 
(previously known for P. osmundacea, but here also reported for P. alfredii). I 
was disappointed, however, that no descriptions were given of Polybotrya 
gametophytes, even from cultured spores. The section on Ecology was brief, 
regrettable in view of the author’s extensive field experience with most of the 
species, and the section on Geography would have been more illuminating if the 
described patterns had been examined in light of the author’s proposed 
phylogeny (p. 31) of the species of Polybotrya. Moran’s phylogenetic 
conclusions concerning Polybotrya contain some surprises: the members of the 
distinctive segregate genus Soromanes are highly specialized polybotryas, the 
terrestrial habit of P. sorbifolia and P. fractiserialis is derived within the genus 
from a hemiepiphytic one, and the closest relative of Polybotrya and its 
monotypic sister genus Olfersia is probably Cyclodium, not Maxonia as many 
earlier authors have proposed. Ample data are presented to provide hours of 
cladistic entertainment. 

There is no reason for anyone interested in the taxonomy of ferns not to own 
this book. As stated on the copyright page, “Single copies of most publications 
[of the Illinois Natural History Survey] are available to anyone requesting 
them.” Write for your copy today.—Davin M. JoHNson, New York Botanical 
Garden, Bronx, New York 10458. 


INFORMATION FOR AUTHORS 


Authors are encouraged to submit manuscripts pertinent to pteridology for 
publication in the American Fern Journal. \ d be sent to the Edi- 
tor. Acceptance of papers for publication depends on merit as sete 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 ac- 
ceptability of manuscripts nor the date of publication. 

Authors should adhere to the following guidelines; manuscripts not so pre- 
pared may be returned for revision prior to review. Submit manuscripts in tripli- 
cate (xerocopies acceptable), including review copies of illustrations. Do not 
send originals of illustrations until they are requested. Use standard 81/2 by 11 
inch paper of good quality, not “erasable’’ paper. Double space manuscri 
throughout, including title, authors’ names and addresses, text (including heads 
vi nye) literature cited, tables (separate from text), and figure captions 

(groupe figures). Arrange parts of man- 
uscript in ae just given. Include author’ zs name and page number i in upper 


right corner of every sheet. P tleast 25 n typed 
pages. Avoid footnotes and an not break words at ends of lines. Make table head- 
ings and figure captions se t Use S.I. (metric) units for all measures 


(e.g., distance, elevation: ses Fe ) unless quoted or cited from another source 
(e.g., specimen citations). For nomenclatural matter (i.e., synonymy and typi- 
fication), use one paragraph per basionym (see Regnum Veg. 58:39—40. 1968). 
Abbreviate titles of serial publications according to Botanico-Periodicum- 
Huntianum (Lawrence, G. H. M. et al., 1968, Pittsburgh: Hunt Botanical Library). 
References cited only as part of nomenclatural matter are not included in litera- 
ture cited. For shorter notes and reviews, put all references parenthetically in 
text. Use Index herbariorum (Regnum Veg. 106:1—452. 1981) for designations of 
herbaria. 


33. ne i+ 4 . dth ee a Po Boa Py are 


age. Provide margins of at least 25 mm on all il 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. Coordi- 
nate sequence and numbering of figures (and of tables) with order of citation in 
text. E n Include a 
scale and reference to latitude and longitude in each map 

Proofs and reprint order forms are sent to authors by the printer. Ashore 
should send corrected proofs to the editor and reprint 0 orders to the printer. Au- 
casts will be extensive al ype has been 


— 
: -_ re es os cr fA RR, Toiir. 


nal and 1 The reset manual zz our 13th ed. (1982. Chicago: Univ. Chicago - 
Press) from om these 


oo Se oe 
Papers longer than 32 printed pages n ~~ cvtothe Baloo Merits : 
o — see cover 2). : eis ae) 


: se gui delines cine ies Authors 
{ to consult the edit any epect Of manneCDE 


American peouries 


Number 3 


Fern July-September 1988 


Journal 


QUARTERLY JOURNAL OF THE AMERICAN FERN SOCIETY 


be ag = DL DD us cree? * a ae Dp 
= SA 


Masahiro Kato and Dedy Darnaedi 


Calochlaena, a New Genus of Dicksonioid Ferns 
Richard A. White and Melvin D. Turner 


The Ci LF 4 tc ae ee, eRe es In. ‘5s ££ Ask * 1. 


Robert G. Hamilton 


C th f I. s ba 2 Tae ..s2. a? is a Sas Hyh si ¢. ge. ai. ws 
Alan R. Smith and Michael H. Grayum 


105 


Information for Authors  Cover3 


The American Fern Society 
Council for 1988 


JUDITH E. SKOG, Biology Dept., George Mason University, Fairfax, is — President 
DAVID B. LELLINGER, Smithsonian Institution, eons a D. ede Vice-President 
W. CARL TAYLOR, Mil waukee, WI5 Secreta 
JAMES D. CAPONETTI, ee . Botany, sore a Tennessee, eS TN hee: Treasurer 
DAVID S of Botany, University of Vermont, Burlington, VT 054 
ea Treasurer 
itgeaes D. MONTGOMERY, Ecology III, R.D. 1, Berwick, PA 1860 Back Issues Curator 
R. SMITH, Dept. of Botany, University of California, Bekele, = 94720. Journal Editor 
iene LELLINGER, Smithsonian Institution, Washington, D.C. 2 Memoir Editor 
JOHN T. MICKEL, New York Botanical Garden, Bronx, NY 10458. a iddlehead Forum Editor 


American Fern Journal 
EDITOR 


eee ed Cee ees Dept. of Botany, University of California, 
Berkeley, CA 94720 


ASSOCIATE EDITORS 
GERALD]. GASTONY --....-.:.5. Dept. of Biology, Indiana University, Bloomington, IN 47401 
ee ee ei ee: Dept. of Botany, University of Kansas, 

Lawrence, KS 66045 
DAVID E. LELLINGER 2 oe ees, U.S. Nat’l Herbarium NHB-166, — — 

Washington, DC 205 
TERRY R. — Deus etiam Sciences Group, University of Connecticut, ae CT eae 

The “A rt. 1” (ISSN 6002 OAAAX: “i 4 Pee 1 fo ee eR } 
study of ferns. It is owned by the American Fern Society, and aaah at the apenas 
University of Vermont, Burlington, VT 05405-0086. Second-class postage paid at cis ema 
and additional entry point. 


, 


Py Sey a ey al Pa s eee Pe Tepes es pa 


: . & 
a = £, K ae 1 uld be pp 


and : be addressed to Dr. James D. Montgomery, Ecology Ill, R.D. 4, 
Berwick, PA 18603. 


Changes of address, dues, and applications for membership should be sent to the Records 
‘Treasurer. 


a 


L [eee 3 _}. : oe +L fal 


* 
c..k con an 


$19.50 net if paid through an agency sip ency fee $0.50); sent free to 


— -s of the American Fern Society (annual di dues, § $15. 00 + $4. 00 mailing surcharge beyond 


ges volumes 1910-1978 $5. 00 to spade ech single back numbers of 64 pages or less, $1.25; 
6 ee acl volumes 1979 et seq, $8.00 
: ; ae L tn AR re +s ie a 7 


Ss th sams ~ 3 r VOIUTNES 


> 


or more 


PosTMASTER: Send add s h to A F lounval Dept of Botany, University of 
ee oes maces : 
Fiddiehead Forum 


ee! ig 7 + i 
gist SUitus 


ncluding miscellaneous 


SS : notes, offers to seme ae or ——_ materials, ee, ‘cticaiecat notes, and reviews of 


i ee ee ee and to others 
Botanical b opcumiopey of the Journal, Sod ee ee - 
e. In es ee ue 


American Fern Journal 78(3): 77—85 (1988) 


Taxonomic and Phytogeographic Relationships of 
Diplazium flavoviride, D. pycnocarpon, 
and Diplaziopsis 
MASAHIRO Kato and Depy DARNAEDI' 


Botanical Gard Faculty of Science, University of Tokyo, 3-7-1 Hakusan, Bunkyo-ku, 
yo 112, Japan 


Diplazium, a worldwide genus of about 400 species, is the largest genus of the 
athyrioid group, although Ching (1964b) placed the majority of species of 
Diplazium in the genus Allantodia. Several monotypic or oligospecific genera, 
such as Callipteris, Dictyodroma, Diplaziopsis, Hemidictyum, and 
Monomelangium, are often segregated from Diplazium and recognized on the 
basis of a few specialized characters (Copeland, 1947; Crabbe et al., 1975; Ching, 
1978). All of these are closely related to and probably derived from 
Diplazium. Diplaziopsis, comprising two species, is defined by its pinnate 


leaves with a conform terminal pinna, thin lamina texture, anastomosing veins, 
and sausage-like sori (Goockad: 1947). Ching (1964a) added two new species 
from China to Diplaziopsis. 


Diplazium flavoviride was described by Alston (1940a, 1940b) from New 
Guinea. He noted “‘[it is] Apparently allied to the North American Athyrium 
pycnocarpon (Spreng.) Tidestr., which as pointed out by Butters (Rhodora XIX, 
p. 178 (1917)), should be referred to Diplazium as D. pycnocarpon (Spreng.) 
Broun.” However, D. flavoviride has not been given attention in later studies of 
the taxonomic and phytogeographic relationships of D. pycnocarpon, which 
was compared only with Diplaziopsis (Tryon & Tryon, 1973; Kato & Iwatsuki, 
1984). Diplazium pycnocarpon is commonly referred to Athyrium by many 
North American pteridologists who otherwise recognize Diplazium as distinct 
from Athyrium (e.g., Morton, 1968; Mickel, 1979; Lellinger, 1985). 

The intercontinentally discontinuous distribution of identical or closely 
related taxa of plants between eastern North America and eastern Asia, 
extending to southern and southeastern Asia in some cases, is a classic and 
well-known example of disjunct distribution patterns (Li, 1952; Graham, 1972), 
referred to as a Tertiary relict disjunction (Wood, 1972). Tryon and Tryon (1973) 
and Kato and Iwatsuki (1984) included Diplazium pycnocarpon and 
Diplaziopsis among the pairs vicariant between the two regions. Boufford and 
Spongberg (1984) note “‘our knowledge and understanding of the eastern North 
American—eastern Asian floristic connection will in large measure reflect and be 
determined by our knowledge of the systematics of the taxa involv 

In this stu tudy w ip of Diplazium flavoviride 
and D. pycnocarpon, and also that between these two species and the segregate 


Pigs Er...k . Ck Ce Din 


eh > ID ‘. ir; } i + LIPI 
e Raya Juanada 22—24, Bogor, aa 


missour! BOTANICAL 


pec 30 1988 


AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 3 (1988) 


Fic. 1. Diplazium flavoviride. A, Seram specimen with erect rhizome and lanceolate leaf, x .42. 
Arrows indicate reduced basal pinnae. B, Cleared fertile pinna, x 5.5 


KATO & DARNAEDI: DIPLAZIUM 79 


genus Diplaziopsis. On the basis of the taxonomic relationships, we discuss the 
phytogeographic relationship of these species. 


MATERIALS AND METHODS 


In this study we used living material of D. flavoviride, D. pycnocarpon, and 
Diplaziopsis cavaleriana (Christ) C. Chr. cultivated in the Botanical Gardens, 
Faculty of Science, University of Tokyo, and also dried and pickled specimens of 
these species and Diplaziopsis javanica (Blume) C. Chr. The material of 
Diplazium flavoviride was originally collected from Seram (Ceram) Island, the 
Moluccas, east of New Guinea. For examination of veins, pinnae were cleared in 
commercial bleach and stained with 1% basic fuchsin solution. Rachises were 


sectioned with a freezing the cross sections of the adaxial 
grooves. For cytological rues Aah sbi tips of D. flavoviride were pretreated with 
0.02M 8—hydroxyquinoline solution, kept in a dark room at 20° C for three hours, 


separated from the root cap, fixed in 45% acetic acid for 2-3 minutes, stained 
with 2% aceto-orcein solution, and squashed. Spore size was measured in 100 
spores from one specimen of each species. For scanning electron microscopy 
(SEM) of spores, untreated spores were coated with gold and observed under an 
Hitachi S700. Voucher specimens of the four species examined are deposited 
in the herbarium of the University of Tokyo (TI). 


RESULTS 


Diplazium pycnocarpon and the species of Di well known ferns, 
and their gross morphologies have been described by many workers (for D. 
pycnocarpon, Morton, 1968; Mickel, 1979; Lellinger, 1985; for Diplaziopsis, 
Sledge, 1962; Ching, 1964a; Ohwi, 1965). Although Diplazium flavoviride is 
relatively little known, its characters were fully described with a photograph by 
Alston (1940a, 1940b). Therefore, the following descriptions concentrate on 
poorly known characters. 

Habitat and habit. — Diplaziopsis species and Diplazium pycnocarpon are 
terrestrial on wet ground or slopes in shade of tropical montane forests at 
1000-2000 m (Diplaziopsis javanica), warm-temperate (Diplaziopsis 
cavaleriana), and temperate (D. pycnocarpon) woods. Diplazium flavoviride is 
a lithophyte growing on mossy limestone rocks in light shade in tropical, mossy 
montane forests at about 2000 m on Seram Island (1500-1800 m in New Guinea). 
The plants have erect or suberect rhizomes bearing 3-5 clustered leaves with 
curved rachises and obliquely downward pointing lamina tips, and many 
stipe-bases not so prominent as in D. pycnocarpon (Johnson, 1986). 

Lamina. — The leaf lamina of all four species is pinnate and thin-herbaceous. 
The pinnae are entire and truncate or subauricled at the acroscopic base (Fig. 1). 
The lower pinnae are somewhat reduced. The lamina is gradually narrowed 
toward the narrow-deltoid, lobed apex in the very similar D. flavoviride and D. 
pycnocarpon (Fig. 2), while it has a terminal pinna conforming to the lateral 
pinnae in Diplaziopsis. 


80 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 3 (1988) 


a 


Fic. 2. Apical part of leafin Diplazium species. A, Diplazium flavoviride. B,D. pycnocarpon. A,B, 
2, 


Rachis. — The rachis as well as the stipe is terete on the abaxial side and 
grooved adaxially in the four species. The groove is V-shaped in cross section in 
Diplazium flavoviride, D. pycnocarpon and Diplaziopsis. It is U-shaped with a 
flat bottom in the Diplazium dilatatum group, the major infrageneric group of 
Diplazium (Kato, 1977), to which D. mettenianum is referred (Fig. 3 


Fic. 3. Cross sections of rachis in Dipl d Diplaziopsis. A, Diplazium flavoviride, x 25. B 
D. pycnocarpon, X 17. C, Diplaziopsis cavaleriana, x 17. D, Diplazium mettenianum, x 14 


’ 


Veins and sori. —Veins are free with usually once-forked lateral veins 
reaching nearly to the barely crenulate pinna-margin in D. flavoviride and D. 
pycnocarpon, although the crenulation is more regular in the latter species (Fig. 
4A, B). In Diplaziopsis the veins are sagenioid-anastomosing, i.e., having 
elongate areoles with no included free veinlets, and vein-endings, whether free 
or areolate, ending well short of the entire or subundulate lamina margin (Fig. 
4C). Sori are elongate and thick with vaulted, more or less thick or thin 
(Diplaziopsis) indusia in all four species. 


KATO & DARNAEDI: DIPLAZIUM 81 


Fic. 4. Venation in Diplazium and Diplaziopsis. A, Diplazium flavoviride, x 2.7. B, D. 
pycnocarpon, X 2.4. C, Diplaziopsis javanica, xX 2.7. 


Spores. — Spores of the four species are monolete and subglobose (Fig. 5). 
They are dark brown in D. pycnocarpon, and brown in the other three species. 
The diameter (excluding perispore) is 30-40 (mean, 35.3) x 25-35 (30.7) umin 
D. flavoviride, 25—36 (31.2) X 23—29 (25.5) ym in D. pycnocarpon, 28—43 (35.0) 
x 20-31 (27.1) um in Diplaziopsis javanica, and 23—33 (30.2) x 20—25 (22.0) 
um in Diplaziopsis cavaleriana. The perispore of all four species is winglike in 
light microscopy, and lophate with convolute or echinate projections on the 
lumina in SEM. The height of the perispore is up to 8 pm in D. flavoviride, 3 1m 
in D. pycnocarpon, 8 pm in Diplaziopsis javanica, and 6m in Diplaziopsis 
cavaleriana. The perispore bears fine reticulate ornamentations on the surface 
in D. flavoviride and D. pycnocarpon (Fig. 5A—D). Such ornamentations are 
rather poorly developed in Diplaziopsis javanica (Fig. 5E, F), while they are 
more finely reticulate or scaly-reticulate in Diplaziopsis cavaleriana (Fig. 5G, 
H). 
Chromosomes. — Diplazium flavoviride has 80 cl 
examined at mitosis (Fig. 6), the same number as that reported for D. 
pycnocarpon (Léve et al., 1977). 


so +} 1] 


TAXONOMIC RELATIONSHIPS 


Compari in almost all characters described above as well as in others such 
as scales, leaf size, and stipes (Alston, 1940a, 1940b) show remarkable 
similarities between D. flavoviride and D. pycnocarpon. They include not only 
such leaf characters as leaf shape, texture, pinna shape, rachis-grooves, veins, 
and sori, but also spore morphology and chromosome number. A recent work 


82 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 3 (1988) 


Fic. 5. SEM photographs of Diplazium and Diplaziopsis spores. A, B, Diplazium flavoviride. C, D, 
D. pycnocarpon. E, F, Diplaziopsis javanica. G, H, Diplaziopsis cavaleriana. A,C, and E magnified 


as in G; B, D, and F magnified as in H. 


KATO & DARNAEDI: DIPLAZIUM 83 


hare Fs 7 
ie ae ta 
. ’ cee 
BOY —s 
V7 NL 4, pall Ponsa 
a eee LEA 
: si : SG 
pr Adal! i 
> * - ~ 
A) {> s@ maz li A Ze 
é io . Me, N 
Oy »4 


Fic. 6. Somatic cl (2n = 80) of Diplazium flavoviride. A, Photograph. B, Interpretative 
drawing. 


(Nakato & Mitui, 1979) showed that Diplazium has chromosome numbers of 
x= 41, and also x=40, rarely. Thus the shared chromosome number is a derived 
character that may suggest a close taxonomic relationship. Among the common 
characters, furthermore, the cross sections of rachis-grooves are useful to define 
the infrageneric groups of Diplazium (Kato, 1977). A combination of those 
characters suggests a close relationship between the two species, as correctly 
noted by Alston (1940a, 1940b). Although D. pycnocarpon is commonly 
referred to Athyrium (Morton, 1968; Mickel, 1979; Lellinger, 1985), it seems to 
be a temperate Diplazium species. Diplazium pycnocarpon has none of the 
characteristic features of Athyrium such as swollen stipe-bases, adaxial spines at 
the bases of costae and costules, pale, cartilaginous edge of lamina, and curved 
sori (Kato, 1977). : 

Differences between the two species include rhizome habit (erect in D. 
flavoviride and creeping in D. pycnocarpon) and dimorphism (not apparently 
dimorphic in D. flavoviride and subdimorphic in D. pycnocarpon). Johnson 
(1986) pointed out that D. pycnocarpon has starch-containing, persistent 
stipe-bases; these are not prominent in D. flavoviride. 

Diplazium flavoviride and D. pycnocarpon are also closely related to 
Diplaziopsis, as pointed out by Tryon and Tryon (1973) and Kato and Iwatsuki 
(1984). This is suggested by their similar pinna shape, rachis-grooving, indusia, 


84 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 3 (1988) 


and spores, although they are different in lamina apex, venation, and the 
chromosome number (Love et al., 1977). While the first two are derived 
characters of Diplaziopsis, the last (x = 41) is shared between it and the bulk of 
Diplazium. The overall similarities lead us to conclude that Diplaziopsis, even 
though quite different from Diplazium in a few obvious characters, should be 
treated as an infrageneric group of the latter genus, as done by Kato (1977). A 
Diplazium from which Diplaziopsis is excluded would be a paraphyletic taxon 
still including species closely related to Diplaziopsis. No nomenclatural or 
taxonomic ranking for this group is given here, pending further work ona natural 
infrageneric classification of Diplazium. 


Puy RELATIONSHIP 


The disjunct distributions between eastern North America and eastern Asia 
are not only demonstrated by many flowering plants (Li, 1952), but also by some 
pteridophytes (Tryon & Tryon, 1973; Kato & Iwatsuki, 1984). Diplazium 
pycnocarpon and the species of Diplaziopsis have been cited as examples of this 
by Tryon and Tryon (1973) and Kato and Iwatsuki (1984). However, the real 
vicarious pair of species are, as discussed above, D. flavoviride of eastern 
Malesia and D. pycnocarpon of eastern North America (Fig. 7). Although such a 
distribution is somewhat aberrant, compared to the common eastern North 
American-eastern Asian distribution pattern, the Diplaziopsis group including 
the two species of Diplazium is widely distributed in eastern, southern, and 
Southeast Asia and in the Pacific islands. A similar distribution pattern is 
shown by another athyrioid group, Deparia sect. Lunathyrium (Kato & Iwatsuki, 
1984), and somewhat similar patterns are found in fl ing plant has 
Astilbe, Gordonia, Illicium, Itea, Lindera and Nyssa (Li, 1952). It seems likely 
that the distribution pattern of the Diplaziopsis group has been affected by the 
same geological and climatic conditions as the eastern North American—eastern 


(oviride 


Fic. 7. Distributions of Diplazium flavoviride, D. pycnocarpon, and Diplaziopsis. 


KATO & DARNAEDI: DIPLAZIUM 85 


Asian distribution patterns exhibited by many vascular plants. As almost all 
fern species of Seram Island are Malesian in affinity (Kato, 1987), this 
amphipacific craps pattern of the Diplaziopsis group is unique. 

We thank Drs. M.G. Price and K. Mitui for reading the manuscript. We also 
thank Ms. T. = Puniediitla for technical assistance, and the director of the 
University Museum, University of Tokyo, for providing facilities for SEM 
observations. This study was in part supported by a Grant-in-Aid (62043018) for 
Overseas Scientific Research from the Ministry of Education, Science and 
Culture, Japan. 


LITERATURE CITED 


ALSTON, f fe G. 1940a. Undescribed ferns from New Guinea. J. Bot. 78:225-229. 
————, 1940b. Undescribed ferns from New Guinea. Nova Guinea, n.s., 4:109-112 
Seerons: os E. and S.A. SPONGBERG. 1984. Eastern Asian—eastern North American 
Se amdiuree ee history from the time of Linnaeus to the twentieth 
ntury uri Bot. Gard. 70:423—439. 
CHING, RC 1964a. On A genus Piplemoptee., _ Acta Phytotax. Sin. 9:31-36. 
1964b. On ly Athyriaceae. Acta Phytotax. Sin. 9:41-84. 
, 1978. The Chinese fern families and genera: systematic arrangement and historical 
origin. Acta Phytotax. Sin. 16(3):1-19, 16(4):16—37. 
Copeland, E. B. 1947. Genera filicum. Waltham, Mass.: Chronica Botani 
Crabbe, J. * A.C. Jermy, and J. T. Mickel 1975. A new generic eae for the pteridophyte 
arium. Fern Gaz. 11:1 
Graham, * (ed.). 1972. Phat a paleofloristics of Asia and eastern North America. 
snp Elsevier. 
Johnson, D. M. 198 is in North American species of Athyrium (Aspleni ). Syst. Bot 


11:26— 

Kato, M. 1977. Classification of Athyri d allied apan. Bot. Mag. Tokyo 90:23—-40 

————., 1987. Floristic richness and phytogeography of haa pteridophytes. Acta Phytotax. 
Geobot. 38:57-62. oe Japanese) 

—— and K. Iwatsuki. 1 lati idophytes t temperat 
North America sat si “Ann. Missouri Bot. Gard. 70:724-733. 

Lellinger, D. B. 1985. A field manual of the ferns and fern-allies of the United States & Canada. 
Washington, D.C.: Smithsonian Inst. Press 

Li, H—L. 1952. Floristic relationships between eastern Asia and eastern North America. Trans. 
Amer ee . Soc. 42: 371 1-429. (Reprinted as: Morri s Arboretum — Edition. 

al on retained. 


1971. W i. Origin 

Léve, A., D. Léve, and im E. G. Pichi Sermolli. 1977. sania atlas of the + aoe 

aduz: J. Cram 

Mickel, J. T. io How - know the ey and fern allies. Dubuque, lowa: Wm. C. Brown 

Morton, C. V. 1968. Pteridophyta. Pp. 1-57 in H.A. Gleason, The new Britton and Brown 
iP aee flora of the northeastern United States and adjacent Canada, Vol. 1. New York: 
H 

Nakato, N. and K. Mitui. 1979. Intraspecific polyploidy in Diplazium subsinuatum (Wall.) Tagawa. 
J. Jap. Bot. 54:129-136. 

Ohwi, J. 1965. Flora of Japan. Washington, D.C.: Smithsonian Institutio 

Sledge, W. A. 1962. The athyrioid ferns of Ceylon. Bull. Brit. Mus. (Nat. Hist), Bot. 2:277-323. 

Tryon, A. F. cips R. M. Tryon. 1973. Thelypteris in northeastern North America. Amer. Fern J. 


76. 
Wood, C. E. ~< 1972. Morphology and phytogeography: the classical approach to the study of 
disjunctions. Ann. Missouri Bot. Gard. 59:107—124. 


American Fern Journal 78(3): 86—95 (1988) 


Calochlaena, a New Genus of Dicksonioid Ferns 


RICHARD A. WHITE AND MELVIN D. TURNER 
Department of Botany, Duke University, Durham, North Carolina 27706 


The genus Culcita C. Pres] contains several species of dicksonioid ferns. 
Before 1922 these species were included in Balantium Kaulf., then regarded as 
typified by Balantium culcita (L’Hér.) Kaulf. (now known as Culcita macrocarpa 
C. Presl). Maxon (1922) demonstrated that Balantium should be typified by B. 
auricomum Kaulf., a synonym of Dicksonia arborescens L’Hér., the type of 
Dicksonia. Balantium is thus a synonym of Dicksonia. Culcita is the correct 
name for the segregate genus typified by Dicksonia culcita L’Hér. ( = Balantium 
culcita). 

Culcita macrocarpa is found in the Azores, Madeira, Spain, and Portugal 
(Fraser-Jenkins & Lainz, 1983). It is closely related to C. coniifolia (Hook.) 
Maxon, of tropical America. Also included in Culcita are several species from 
Australasia and the Philippines. This latter group of species was first associated 
with Balantium when Diels (1899) and Copeland (1908, 1909) added the species 
Balantium stramineum (Labill.) Diels, Balantium dubium (R. Br.) Copel., and B. 
javanicum (Blume) Copel. to that genus. 

Maxon (1922) recognized eight species of Culcita. Differences in soral 
characters led him to create a new subgenus Calochlaena for two of these 
species, the Australian C. dubia (R. Br.) Maxon and an alleged new species from 
Fiji, C. blepharodes Maxon, which we believe is a synonym of C. dubia. 
Although Maxon regarded such species as C. straminea and C. javanica as 
typical Culcita, not Calochlaena, more recent authors have enlarged subg. 
Calochlaena to include all of the Australasian species (Copeland, 1947; Holttum 
& Sen, 1961; Holttum, 1963, 1964). These authors apparently were not aware 
that this was not Maxon’s original subgeneric concept. 

Significant differences between members of the two subgenera of Culcita have 
frequently been reported. These include differences in anatomy (Davie, 1918; 
Holttum, 1963; Holttum & Sen, 1961), in spore morphology (Gastony, 1981), and 
in chromosome number (Manton in Holttum & Sen, 1961; Manton in Holttum, 
1963; Roy & Holttum, 1965). These have prompted suggestions that the genus 
may need to be divided (Holttum, 1963; Roy & Holttum, 1965; Gastony, 1981; 
Tryon & Tryon, 1982). The need for taxonomic reassessment of Culcita became 
evident to us in the course of an anatomical survey of the tree ferns (see White, 
1984). In the present paper we summarize evidence for the generic distinctness 
of the two groups and erect a new genus for the species of subg. Calochlaena. 
Culcita thus becomes restricted to the two species C. macrocarpa and C. 
coniifolia. 


MATERIALS AND METHODS 


Live and preserved plants of Culcita (Culcita) coniifolia, C. (Calochlaena) 
straminea, C. villosa, and C. dubia were studied in detail, and herbarium 


WHITE & TURNER: CALOCHLAENA 87 


specimens of all species of Culcita were examined. Three-dimensional anatomy 
of the shoot was studied by cinematographic analysis of stems cut with a sliding 
microtome (Zimmerman & Tomlinson, 1965). The anatomy of sori, stems, 
petioles, and rachises was investigated using microtome sections of 
paraffin-embedded samples as well as freehand sections of fresh or preserved 
material. Some dried leaf fragments bearing sori were expanded in dilute 
Contrad 70 (Schmid & Turner, 1977) for study. Spores of all species were 
examined with light microscopy, and spores of Cucita villosa and Oenotrichia 
novae-guineae were studied using scanning electron microscopy. 


COMPARISONS 


Organography and Anatomy.—Although plants of C. coniifolia outwardly 
appear to have massive short stems similar to the trunks of other tree ferns, 
sections show that the “‘trunk’’ consists of a relatively small erect stem 
surrounded by persistent petiole bases, hairs, and roots (Fig. 1a, 1b). The 
enlarged petiole bases contain much starch and together form an imbricated 
storage organ. Petiolar buds and branches are lacking, although we have seen 
individuals with dichotomously bifurcated stems. Shoots of subg. Calochlaena 
show a strikingly different pattern. Culcita straminea and C. villosa have stout 
horizontally growing rhizomes with distinct nodes and internodes. The 
rhizomes produce many branches or stolons from lateral buds located on the 
abaxial sides of the leaf bases (Fig. 2a). The somewhat smaller rhizomes of C. 
dubia are essentially similar to those of the other species. The creeping, 
branching rhizome systems allow these plants to form extensive clumps through 
continued vegetative reproduction. References in the literature to arborescent 
Calochlaena (e.g., Copeland, 1909) are probably in error. 

Stems are solenostelic in both subgenera, but the solenosteles are very 
dissimilar. Culcita coniifolia (Figs. 1b, 3) has a relatively small, strongly lobed 
stele, the lobes decurrent from the insertions of the leaf traces. Leaf positions are 
close together and the arrangement of leaf gaps approaches the dictyostelic 
condition. The pith has a sclerenchymatous core from which extend tracelike 
strands of sclerenchyma associated with the leaf traces (Fig. 3). In the leaf axils 
the sclerenchyma strands join the scl hymat ter cortex of the stem. No 
bud or branch traces are found in Culcita coniifolia. The figure of Sen (1968) 
indicates that the stem of C. macrocarpa resembles that of C. coniifolia. 

In contrast to the pattern in subg. Culcita, steles of subg. Calochlaena are 
circular in cross section in the long internodes and show a different pattern of 
lobing at the nodes (Figs. 2b, 2c). Sclerenchyma is present as islets or strands 
scattered throughout the cortex and pith (Fig. 4). Vascularization of branches 
and buds in Calochlaena resembles that in Lophosoria and Metaxya (Lucansky, 
1974). seaabis 

The configuration of the petiolar vascular bundle is very different in the two 
subgenera. Petiolar bundles of subg. Culcita have a simple form with widely 
diverging margins (Fig. 5; Sen, 1968) whereas those of subg. Calochlaena show a 
shallowly lobed, adaxially involute “C’” shape in cross section (Fig. 8). The 
patterns of pinna vascularization also differ sharply in the two groups (Figs. 7, 8; 


88 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 3 (1988) 


branching from buds (b) on abaxial side of leaf bases. 2b, Cross section of internode. Stem is 
solenostelic with islets of sclerenchyma in cortex and pith. 2c, Section at node. Branch trace (bt) 


WHITE & TURNER: CALOCHLAENA 89 


Fics. 3-8. Anatomy of Culcita and Calochlaena. Fic. 3. Culcita coniifolia. Cross section of stem 
howing lobed sol tele. C f scl hyma (s) in pith gives rise to “sclerenchyma traces.” Fic. 

t ] telic with scattered islets of sclerenchyma in 

ole of Culcita coniifolia. Vascular bundle with diverging 
le of Calochl trami Vascular bundle adaxially 

tial ti (distal ight) of rachis of C. coniifolia. Pinna trace 


5 
Caontinn at at f Calochl villosa 


cortex and pith. Fic. 5. Cross section of peti 
daxial i Fic. 6.C ti f I ti 


involute, corrugated. Fic. 7. Seq { gnu 
arises from margin of foliar bundle. Fic. 8. Sections of rachis of Calochlaena villosa. An 
1 ae +} ; trace ale te = 5 CH. 


extramarginal sector of the vascular 


Davie, 1918). In subg. Culcita, pinna traces depart from the margins of the foliar 
bundle (Fig. 7), whereas in Calochlaenaa wide extramarginal sector becomes the 
pinna trace (Fig. 8). 

Prominent mucilage ducts occur in the vascular bundles of petioles and 
rachises of subg. Calochlaena but are not found in subg. Culcita. They are 
essentially identical to structures found in Cibotium, Lophosoria, and Cyathea. 
In contrast, mucilage cells occur throughout the ground tissue of Culcita 
coniifolia; these are not seen in subg. Calochlaena. 


90 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 3 (1988) 


Fics. 9-14. Soral morphology of Culcit d Calochlaena. Fic. 9. Large sori of Culcita coniifolia (on 
left) with those of Calochlaena villosa. Fic. 10. Culcita conitfolia Longitudinal section of sorus. 
an egg recepiacl= is cna with base of inner indusium. Sporangia are long-stalked. Fic. 11. 


any ATT 


ranc 477 (US}}, showing lack of fusion of receptacle and 
indusium. Sporangia are short-stalked. Fic. 12. Abaxial view of sori of Calochlaena straminea 
[Setchell 389 (US)]. Sori covered by bivalved indusia. Fic. 13. Sori of Calochlaena dubia (US 25078). 
uter indusium represented by a lobe of the margin, sorus appears inframarginal. Fic. 
Calochlaena dubia eure s.n. (MICH)]. Mature sori with most sporangia not covered by indusia. 
o = outer indusium; i = inner indusium;r = receptacle. Scale bar in Fig. 9 = 5 mm; scale bars in 
Figs. 10-14 = 1mm 


Sori.—Culcita macrocarpa and C. coniifolia have large bivalved sori (Figs. 9, 
10) in which the receptacle is united with base of the inner (abaxial) indusium 
(Tryon & Tryon, 1982). The vascular bundle of the sorus extends into the 
expanded receptacle (Fig. 10). The sori of Culcita resemble those of Dicksonia 


WHITE & TURNER: CALOCHLAENA 91 


which show a similar but more extensive fusion of the receptacle and indusium 
(Goebel, 1905; Tryon & Tryon, 1982; Churchill, 1983). Members of subg. 
Calochlaena have much smaller sori that exhibit no such fusion of the receptacle 
and the indusium (Figs. 9, 11). In this regard, subg. Calochlaena resembles 
Cibotium (see Tryon & Tryon, 1982) rather than Dicksonia. 

There is a striking diversity of soral morphology within Calochlaena. Most, 
including C. javanica and C. straminea, have typically dicksoniaceous bivalved 
sori (Figs. 11, 12). The sorus of C. dubia, on the other hand, differs sharply from 
those of other dicksonioid ferns. The “outer indusium” is represented only by a 
slightly reflexed but unmodified lobe of the lamina so that the sorus appears to be 
superficial and inframarginal with a single indusial flap (Fig. 13). This 
characteristic formed the original basis of Maxon’s subg. Calochlaena. 
Collections of C. dubia often have sori intermediate between this condition and 
that of the other members of the group. 

The inner indusium encloses the maturing sporangia in C. straminea (Fig. 11, 
12), but the sporangia of other species of Calochlaena are often largely exposed 
(Fig. 14). The inner indusium tends to be lobed or laciniate and commonly has 
cilia or marginal hairs. Inner indusia of some collections of C. javanica have 
trichomes and emergences on their outer surfaces. 

Other differences between the subgenera have been noted in the literature. 
Reported chromosome numbers of n = 66-68 in Culcita macrocarpa (Manton, 
1958) and n = 66 inC. coniifolia (G6mez—Pignataro, 1971) are widely different 
from the count of ca. 55-58 for C. dubia (Manton, cited by Holttum, 1963). 
Gastony (1981) pointed out that the two subgenera are strikingly different in 
spore morphology. Spores of subg. Culcita possess a perine, lacking in 
Calochlaena, and the psilate to microverrucate exine of subg. Culcita differs 
strongly from the sculptured exine of broad spinules found in subg. 
Calochlaena. 

The two subgenera of Culcita appear to be united chiefly by a general 
resemblance in frond form and dissection and similar patterns of channelling of 
the upper sides of leaf axes. Apart from these probably superficial similarities, 
the subgenera share few or no characteristics beyond those common to other 
genera of dicksonioid ferns. In fact, Calochlaena and Culcita are as distinct 
morphologically as any two genera of dicksonioid ferns, and it is likely that each 
is more closely related to other genera of the group than it is to the other 
subgenus. The numerous differences strongly support the conclusion that subg. 
Calochlaena should be elevated to generic rank. 


Calochlaena (Maxon) M.D. Turner & R.A. White, stat. nov.—Culcita subg. 
Calochlaena Maxon, J. Wash. Acad. Sci. 12:458—459. 1922.—T PE: Culcita 
dubia (R. Br.) Maxon 

Large terrestrial ferns. Rhizomes prostrate to suberect, not arborescent, often 
spreading by stout stolonlike branches, stem apices densely hairy with bristly or 

soft hairs, st ] t fscl h 


} = = me 


ic with islets or strand ym in cortex and 
pith. Fronds spirally arranged, large (to 2 m or more), long-petiolate; petiolar 
vascular bundle simple, incurved adaxially, with vascular mucilage ducts, 


92 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 3 (1988) 


pinna traces extra-marginal. Leaf axes and lamina pilose to glabrescent. Lamina 
deltoid, 3-4 pinnate-pinnatifid, dissection anadromous, basal pinnae largest; 
leaf axes adaxially grooved with grooves of large and small axes confluent; 
ultimate segments oblique, larger on acroscopic side, pinnatifid to toothed, 
venation simple. Sori marginal, terminal on veins on monomorphic leaves, 
indusia bivalved, outer indusium cuplike or (in C. dubia) only slightly 
differentiated from the lamina and sorus appearing inframarginal; inner 
indusium membranous, enclosing or only partly covering the sporangia, often 
lobed or laciniate, often marginally ciliate, sometimes with hairs on outer 


; Pp t united with base of inner indusium; ; phy many,on 
receptacle and sporangial stalks, filamentous, contorted, and vermiform. 
Sporangia short-stalked, annuli oblique; spores tetrahedral-trilete with broad 
spinules or tubercles laterally fused into ridges. 

Copeland (1947) commented that the species of Caloch! “too similar’, 
and Holttum (1963) placed two species in synonymy under Culcita javanica. 
The species listed below show consistent differences in some characters, 
although there appears to be much variation within species. Future systematic 
studies may alter these species concepts. We exclude Culcita formosae (Christ) 
Maxon (Dennstaedtia formosae Christ) from Calochlaena. It is a Dennstaedtia, 
apparently D. smithii (Hook.) Moore. 


1. Calochlaena dubia (R. Br.) M. D. Turner & R. A. White, comb. nov.—Davallia 
dubia R.Br., Prodr. 157. 1810—Dicksonia dubia (R. Br.) Gaudich., Voy. 
Bonite, Bot. 367. 1827.—Sitolobium dubium (R. Br.) Brackenr., U.S. Expl. 
exped., Filic. 16:273. 1854.—Balantium dubium (R. Br.) Copel., Philipp. J. 
Sci. 3:301. 1908—Culcita dubia (R. Br.) Maxon, J. Wash. Acad. Sci. 12:458. 
1922. 

Balantium brownianum C. Presl, Tent. pterid. 134. 1836. 

Hemitelia godeffroyi Luerssen, J. Mus. Godeffroy. 3:4. 1873. (fide D. B. Lellinger, 
in litt.) —Type: Australia, Brisbane River, Dietrich 26. 

Culcita blepharodes Maxon, J. Wash. Acad. Sci. 12: 459. 1922.—T ype: Fiji, 
Lomo-Lomo, Wilkes Expedition (US!). 

Range:—Eastern Australia. 

The type of Culcita blepharodes closely agrees with several Australian 
collections of the variable C. dubia. Possibly this specimen was incorrectly 
attributed to Fiji, or it may represent a waif or temporary introduction. We have 
seen no other Fijian specimens of this species, and it is omitted from Brownlie’s 
(1977) flora. 


2. Calochlaena straminea (Labill.) M.D. Turner & R.A. White, comb. 
nov.—Dicksonia straminea Labill., Sert. austro.—caledon. 7., pl. 10. 
1824.—Dennstaedtia straminea (Labill.) J. Smith, Hist. fil. 265. 1875 — 
Balantium stramineum (Labill.) Diels in Engl. & Prantl, Nat. Pflanzenfam. 
1(4):119. 1899. (not Sitolobium stramineum Brackenr.).—Typr: New 
Caledonia, Labillardiére s.n. (P). 


WHITE & TURNER: CALOCHLAENA 93 


Dicksonia torreyana Brackenr., U.S. Expl. exped., Filic. 16. 278, pl. 38, f. 2. 
4 


Range:—New Caledonia, Fiji, Samoa, Vanuatu (New Hebrides), Solomon 
Islands, Admiralty Islands, New Guinea, SE Philippines. 


3. Calochlaena javanica (Blume) M.D. Turner & R.A. White, comb. 
nov.—Dicksonia javanica Blume, Enum. l. Javae. 240. 
1828.—Dennstaedtia javanica (Blume) Christ, Bull. Herb. Boissier, sér. II. 
4:617. 1904.—Balantium javanicum (Blume) Copel., Philipp. J. Sci. 4:62. 
1909.—Culcita javanica (Blume) Maxon, J. Wash. Acad. Sci. 12:456. 
1922—TypPE: Java, Blume s.n. (L; isotype US!). 

Dicksonia copelandii Christ, Philipp. J. Sci., Bot. 2:183. 1907.—Balantium 
copelandii (Christ) Christ, in Copel., Philipp. J. Sci. 3:301. 1908.—Culcita 
copelandii (Christ) Maxon, J. Wash. Acad. Sci. 12:457. 1922.—SYNTYPEs: 
Philippines, Luzon, Baguio, Benguet Province, 1400 m, Loher s.n.; Baguio, 
Benguet Province, Elmer 6025; Bagnen, Lepanto District, 2000 m, Copeland 
1912. 

Balantium pilosum Copel., J. Straits Branch Roy. Asiat. Soc. 63:71—72. 
1912.—Culcita pilosa (Copel.) C. Chr., Index fil. suppl. 3. 57. 1934.—T PE: 
Borneo, Bukit Lawi, Ulu Limbang. 

Range:—Java, Borneo, Philippine Islands. 

We follow Holttum (1963) in including in C. javanica the plants from Borneo 
and the Philippines that have been called Culcita copelandii. Calochlaena 
javanica and C. villosa appear to be closely related. They may eventually prove 
to be a single widespread and variable species. 


4. Calochlaena villosa (C. Chr.) M. D. Turner & R. A. White, comb. nov.—Culcita 
villosa GC. Chr., Brittonia 2:283. 1937.—TypeE: New Guinea, Urunu, Vanapa 
Valley, 1900 m, Brass 4791 (BM; isotypes US!, GH!). 

Range:—New Guinea. 

Christensen noted the small inner valve of the indusium in C. villosa and 
suggested it might be placed with C. dubia in Maxon’s subg. Calochlaena. This 
species is close to C. javanica. Holttum (1964) erroneously labeled this species 
‘pilosa’ on a distribution map (Gastony, 1981). 


5. Calochlaena novae—guineae (Rosenstock) M. D. Turner & R. A. White, comb. 
nov.—Davallia novae-guineae Rosenstock, Repert. Spec. Nov. Regni. Veg. 
5:36. 1908.—Leptolepia novae-guineae (Rosenstock) Alderw., Malayan 
ferns 283. 1909.—Ithycaulon novae-guineae (Rosenstock) Alston, J. Bot. 
77:289. 1939.—Oenotrichia novae-guineae (Rosenstock) Copel., Univ. 
Calif. Publ. Bot. 16:82. 1929—Type: New Guinea, Mt. Gelu, 1700 m, 
Werner (isotype US!). 

Range:—New Guinea. ose 
This distinctive species has long been wrongly placed in the genus 
Oenotrichia (Dennstaedtiaceae). Calochlaena novae-guineae is distinguished 


94 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 3 (1988) 


from C. villosa by its larger, more coarsely divided leaves with pinnae diverging 
at a wide angle, larger sori and sporangia, and larger, differently sculptured 
spores. Few collections have been seen: Millar & Holttum N. G. F. 15835, near 
Wau, Morolee dist., Bulldog Rd., Edie Creek, 6,900 ft.; Marchant 78.363, Wau, 
forming dense thickets on summit of Mt. Kindie. 


RELATIONSHIPS OF CALOCHLAENA 


Culcita (including Calochlaena) has been assigned several different 
systematic positions by different authors. It was put in subfamily Dicksonieae of 
the Cyatheaceae with Dicksonia and Cibotium by Diels (1899). Bower (1926) and 
Christensen (1938) similarly classified it in the Dicksonieae of the 
Dicksoniaceae. Copeland (1947) included all dicksonioid genera in a large, 
heterogeneous family Pteridaceae. Holttum and Sen (1961) assigned Culcita to 
subfamily Thyrsopteridoideae of the Cyatheaceae. Other workers have placed 
Culcita with Thyrsopteris and Cibotium in a separate family Thyrsopteridaceae 
(Mabberley, 1987). Culcita was put in a separate family Culcitaceae by Pichi 
Sermolli (1977). Although it is debatable which familial scheme best expresses 
the phylogenetic relationships of the plants, it is clear to us that Calochlaena 
should not be placed with Culcita in a separate family or subfamily. Even though 
detailed phyl | f the dicksonioid ferns are lacking, Calochlaena 
is evidently only remotely sae to Culcita. Both genera seem isolated among 
the tree ferns, but C imilar to Cibotium and even Lophosoria 
than it is to Culcita. Culcita ‘tas apart from the other genera in shoot anatomy 
and morphology, but the similar fusion of receptacle and lower indusium 
indicates a possible relationship with Dicksonia. Spore characteristics also 
suggest a link between Culcita and Dicksonia (Gastony, 1981). Anatomically, 
neither Calochlaena nor Culcita show any close affinity with Thyrsopteris. 

It has been suggested that Culcita/Calochlaena is an evolutionary link 
between the dicksonioid ferns and Dennstaedtia (e.g., Copeland, 1908, 1909, 
1947; Christensen, 1938), and the genus is included in the Dennstaedtiaceae by 
Duncan and Isaac (1986). We believe that the strong superficial resemblance of 
Calochlaena to some Dennstaedtiaceae does not reflect a close phylogenetic 
relationship. In important details of shoot anatomy and morphology, 
Calochlaena resembles other dicksonioid ferns and differs from Dennstaedtia. 

As noted above, several authors have at various times noted that differences 
between Culcita and Calochlaena suggest that they could be separate genera. 
The addition of evidence from shoot anatomy has made it clear that the two 
groups are as dissimilar as any two genera of dicksonioid ferns. The recognition 
that Culcita pag Calochlaena are two distinct genera will affect any future 
discussions of th ' y, and evolution of the tree ferns. 


eo = : oy 


ACKNOWLEDGMENTS 


We th . K. U. Kramer, R. L. Wilbur, D. B. Lellinger, R. Tryon, and A. Tryon for valuable 
discussions, Dr. M. G. Simpson for technical assistance, Dr. A. Juncosa for material of C. straminea. 


WHITE & TURNER: CALOCHLAENA 95 


lay SO eg +1 fa dutta nf ohn £.3% 


MICH, NY, UC, US. Part of this ic was supported by NSF grant # DEB—7714648. 


AA, DUKE, GH, 


LITERATURE CITED 
Bower, F. O. 1926. The ferns. Vol don: C iversity Press. 
BROWNLIE, G. 1977. The skal flora of Fiji. Vaduz: |. can 
CHRISTENSEN, C. 1938. Filicinae. Pp. 522-550 in Manual of Sabha od F. Verdoorn. The Hague: 
Nijhoff. 


CHURCHILL, H. W. 19823 iy j J 7 4 J: fal ] £ j yyy es Ph D 
epee University of Massachusetts. 
COPELAND, E.B. 1 i Philipp. J. Sci. 3:301—302. 
. 1909. F f the Mal iatic regi - Philipp J. Sci. 4 1-65. 
. 1947. Genera filicum. Waltham, M Botanica 


DaviE, R. 1918. A comparative list of fern pinna-traces, with some notes on the leaf-trace in the 
s. Ann. Bot. (London) 32: 233—245 

DIELs, F. snot Cyatheaceae. Pp. 113-139 in Die Natiirlichen Pflanzenfamilien, Teil I, Abt. 4., ed. A. 
Engler and K. Prantl. Leipzig 

Duncan, B.D. and G. Isaac. 1986. Ferns and mnie plants of Victoria, Tasmania and South 
Australia. Carlton, Victoria: v. Press. 

shri gAt inti C.R., and M. Lafnz. 1983. hat macrocarpa—a new locality in Spain. Fern Gaz. 

2:299-301 


(leon: G.1981.S hol the Dicksoni I. The genera Cystodium, Thyrsopteris, 
and Culcita. _ Amer. I Bot. 68: 808-819 
GOEBEL, K. 1905. Organography of plants. IL. Special organography. trans. ed. by I.B. Balfour. 
Oxford: C Press 
su peantebereat ore L.D. 1971. ira upseead ps Loner della Costa Rica. 1. Ist. Bot. 
Reale Univ. Reale Lab. Crittog. Pavia Atti se 
Ho.ttTu, R.E. 1963. Cyatheaceae. Flora Malesiana “0 Il, 1: a 
. 1964. Distribution of some of the more primitive ferns a Mt. Kinabalu. Proc. Roy. Soc. 
London, Ser. B, Biol. Sci. 161:38—-4 
and U. SEN. 1961. Morphology and classification of the tree ferns. Phytomorphology 


11:406—-42 re 
LUCANSKY, T.W. 19 eprom studies of the nodal and vascular anatomy in the neotropical 


ienaia I. Metaxya and Lophosoria. Amer. J. Bot. 61:464-4 

MABBERLEY, Dt 1987. The ste book. Cambridge: bia University Press. 

MANTON, I to ‘‘Pteridaceae”’. J. Linn. 
neg ee 56:7 
N, W.R. 1922. ake gens Culcita. J. Wash. Acad. Sci. 12: 454-460 

vet ‘Seaiiee R.E.G. 1977. Tentamen pteridophytorum genere: in taxonomicum ordinem 
serie che Bi 31:313-512. 

Roy, S.K., and R.E. Hottrum. 1965. New cytological is for Cystodium and Dicksonia. Amer 


Fern. J. 35-37. 

ScHMIpD, R. and M.D. TurRNER. 1977. Srcing 70, an effective softener of herbarium material for 
anatomical study. Taxon 26:551-5 

SEN, U. 1968. Anatomy of Culcita macroca ene Canad. J. Bot. 46:43—-46. 

TRYON, R. and A. TRYON. 1982. Ferns and allied plants with special reference to tr pical America 
New _— ae erlag. 

Waite, R.A. 4. Comparative development of vascular tissue patterns. Pp. 53-107 in 
Contemporary problems in ‘Hiant anatomy, ed. R. A. White and W.C. Dickison. New York: 
Academi 

ZIMMERMAN, M. = and P.B, ToMLINSON. 1965. nen of the palm Rhapis excelsa. I. Mature 
vegetative axis. J. Arnold Arbor. 46:160—17 


American Fern J 1 78(3): 96—104 (1988) 


The Significance of Spore Banks in Natural 
Populations of Athyrium pycnocarpon and A. 
thelypterioides 


ROBERT G. HAMILTON 
Department of Botany, Ohio University, Athens, Ohio, 45701 


How do homosporous plants survive in variable environments? One 
mechanism available to plants is seed or spore dormancy (Cohen 1966, 1967). 
An observable consequence of seed and spore dormancy is the formation of a 
seed or spore bank (Harper, 1977; Grime, 1979). Work in this area goes back at 
least to Darwin (1859), who assayed a cupful of mud from a stream bank for 
viable seeds. 

Cohen (1966, 1967, 1968) developed models of seed and spore dormancy 
describing the environmental conditions favoring its evolution. He predicted 
that dormancy will evolve when the variance of environmental quality greatly 
exceeds the mean of environmental quality. Under this condition natural 
selection will favor individual seeds and spores that a) have delayed 
germination and prolonged viability, as a means of avoiding unfavorable 
environmental conditions, and b) germinate only in response to those 
environmental signals that correlate positively | with future success. These 
models have served an) a sbesis for a number of the effects 
of seed dormancy on the di tion and abund of seed plants (Leon, 1985; 
Brown & Venable, 1986; Ellner, 1985a,b). To understand the effect of spore 
dormancy on fern ecology, knowledge of spore characteristics and of the benefits 
and consequences of delayed spore germination is needed. 

There are many in vitro studies of the physiological characteristics associated 
with spore dormancy. Prolonged viability has been reported for a wide variety of 
fern taxa (Lloyd & Klekowski, 1970). Differential response to germination cues is 
well known, particularly light quality (Towill & Ikuma, 1973; Mohr, 1956; Sugai 
& Furuya, 1967; Tomizawa et al., 1983), light quantity (Isikawa & Oohusa, 1956; 
Eakle, 1975) and hormones (Schraudolf, 1985; Howland & Edwards, 1979). 
There is little information of spore banks among natural populations of ferns, 
although Leck and Simpson (1987) provided evidence of spore banks among 
naturally occurring pteridophyte populations. In order to expand our 
knowledge of spore banks, I assayed soils from among populations of Athyrium 
pycnocarpon (Sprengel) Tidestrom and Athyrium thelypterioides (Michx.) 
Desv. for dormant spores. 


MATERIALS AND METHODS 


Soil samples were collected from natural populations of Athyrium 
pycnocarpon and Athyrium thelypterioides in Hawk Woods, an oak dominated 


R. G. HAMILTON: SPORE BANKS IN ATHYRIUM 97 


woods in the city of Athens, Ohio (T9N R14W Athens, Ohio quadrangle: 82° 4’ 
30’’ W, 39° 21’ 30’’ N), and from Long Run, a beech/maple woods within Wayne 
National Forest near Glouster, Ohio (T11N R14W sec. 11 Corning, Ohio 
quadrange: 82° 4’ 30’’ W, 39° 32’ 30’’ N) 

Other pteridophytes at the Hawk Woods site were Thelypteris 
noveboracensis, Adiantum pedatum, and Polystichum acrostichoides. These 
species were also present at the Long Run site, along with Dryopteris goldiana, 
Cystopteris fragilis, and Thelypteris hexagonoptera. 

1. j ff: se 


7 h + ; swtintAMm rae oT py 
o ascertain whether storage of soil under influ 


oO 


results, samples from the top 1 cm of the soil were collected on 26 October 1986, 
from populations of Athyrium thelypterioides and of Athyrium pycnocarpon 
in Hawk Woods and stored in a refrigerator at 4° C. Each month, from November 
1986 to April 1987, five 2-gram aliquots were removed from each sample and 
inoculated onto 15 mm petri dishes filled with a nutrient agar medium prepared 
as described by Klekowski (1969). The soil was spread evenly across the surface 
of the agar, forming a thin layer. The inoculated petri dishes were then 
incubated for 30—40 days under continuous incandescent and fluorescent light 
totalling approximately 30 pE/m?sec, at room temperature. Each dish was then 
assayed for the total number of gametophytes present. The results of this assay 
were used as an estimate of the number of viable spores present in the soil. 
Similar samples were again collected from the same populations in April 1987, 
with five 2-gram aliquots assayed as described above. The month of April was 
chosen to compare the effects of overwintering in the soil to storage during that 
time in a refrigerator. None of the above-mentioned locally occurring taxa were 
observed to release spores during the winter or early spring, although a 
systematic study of that specific feature was not undertaken. A final assay of the 
soil collected in October 1986 was completed in January of 1988, specifically to 
determine spore viability. Data were analyzed using th pletely randomized 
design analysis of variance (CRD), and means compared using the least 
significant difference parameter (LSD), as described by Steel and Torrie (1980). 
Nine vertical soil cores were collected from populations of both species from 
Long Run between June 1987 and January 1988, and assayed for the presence of 
viable spores as described above. An 18 inch soil sampling tube with 13/16 inch 
inside diameter was used to collect soil cores. Due to compaction, this corer 
could only be used to sample the soil to a depth of 16 cm. The number of spores 
per gram of soil at each centimenter of depth in each core was estimated using the 
method described above. : 
Assuming the density of the soil to be 1 gram per cm’, an underestimate 
(Hausenbuiller, 1978), and given that there are 10,000 cm? in avolumeim X 1m 
x 10cm, the estimated number of spores in a ‘‘slice” of the soil column 1m x 1m 
x 1 cm, for each cm of soil depth, was the number of spores per gram in the 
sample from that depth x 10,000. The values at all levels were then summed to 
give an estimate of the number of spores in the volume of soil beneath each 
square meter of the soil surface. The mean number of spores at ooch level was 
also estimated, along with the mean number of spores beneath each sq 
of soil surface. This underestimates the number of spores per square meter due 


98 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 3 (1988) 


to the above assumptions and because the entire spore-containing column was 
not assayed where the soil was deeper than 16 cm. 

Following the assay for spores, gametophytes from each culture were 
transferred to soil pots. Fertilization was promoted by regularly flooding the 
surface of each pot. The resulting sporophytes were used to determine the 
species composition at each level. 


RESULTS 


Ungerminated Athyrium pycnocarpon and A. thelypterioides spores were 
found to be present in the soil at both sites. Polystichum acrostichoides and 
Thelypteris noveboracensis were also present in the soil at the Hawk Woods 
site. Only the Athyrium species and Adiantum pedatum were identifiable in 
cultures of soil collected at the Long Run site, as the sporophyte cultures died 
before identification of other species could be made. These species were present 
throughout the soil column. 

There were significant differences among months for the number of spores 
germinating from soil stored under refrigeration (p < .001 for A. pycnocarpon 
p< 0.025 for A. thelypteroides). Comparison of means (Tables 1, 2) revealed that 
for both species, germination increased significantly following one month of 
storage under refrigeration, then dropped significantly by the 5th month of 

orage. For A. pycnocarpon, there was a significant increase in germination 
from the fifth to the sixth month of storage. There was no such increase for A. 
thelypterioides. Germination for month 5 was significantly lower than 
germination for month 1 in both species; however, month 6 was not significantly 
different from month 1. There was no difference in germination between month 
1 and the sample collected in April 1987 in both species, but significantly more 


TABLE 1. Mean Number of Spores Germinating in 2 Grams of Soil from an Athyrium pycnocarpon 
Population. Nov 86 to Apr 87 and Feb 88 are results from soil collected in October 1986. Apr 87A 
is the result from soil collected in April 1987. Data in matrix are differences among means. Lower 
case characters indicate that the difference between means compared is significant: a, p < 0.05; 
b, p < 0.01; c, p< 0.001. 


Nov 86 Dec 86 Jan 87 Feb 87 Mar 87 Apr87 + Apr8s7A Feb 88 
336.6 519.8 308.0 274.8 178.0 299.8 255. 302.4 


Feb 88 Apr 87A Apr 87 Mar 87 Feb 87 Jan 87 Dec 86 


Nov 86 34.2 81.6 36.8 158.6° 61.8 28.6 182.3° 
Dec 86 217.4° 264.8° 220.0° 341.8° 245.0° 211.8° 
Jan 87 5.6 53.0 8.2 130.0° 
Feb 87 27.6 19.8 25.0 96.8? 
124.4> 77.0 121.8 
Apr 87 2.6 44.8 


Apr 87A 47.4 


R. G. HAMILTON: SPORE BANKS IN ATHYRIUM 


TABLE 2. Mean Number of Spores Germinating in 2 f Soil fr Ath thel 
Population. Nov 86 to Apr 87 and Feb 88 are results from soil collected in October 1986. Apr 87A 
is the result from soil collected in April 1987. Data in matrix are differences among means. Lower 
case characters indicate that the difference between means compared is significant: a, p < 0.05; b 
p <0.02;c, p< 0.01; d, p < 0.001. 


Nov 86 Dec 86 Jan 87 Feb 87 Mar 87 Apr 87 Apr 87A Feb 88 
98.4 140.6 126.4 95.4 62.2 79.2 127.4 86.8 


Feb 88 Apr 87A Apr 87 Mar 87 Feb 87 Jan 87 Dec 86 


Nov 86 11.6 29.0 19.2 36.28 3.0 28.0 42.2> 
Dec 86 54.2° 13.2 61.42 78.0° 45,2° 14.2 
Jan 87 39.68 1.0 47.2° 64.22 31.0 
Feb 87 8.6 32.0 16.2 33.28 
Mar 87 24.6 65.29 17.0 
Apr 87 7.6 48.2° 
Apr 87A 40.6° 


A. pycnocarpon spores germinated in the soil stored for 1 month than in soil 


collected in April 1987. The number of spores germinating in 15-month old 
cultures was the same as in fresh cultures, but 


y iOwel than in one 
month old cultures for both species. Basic trends in : germination were the same 
for both species. 

Soil depth at sites sampled ranged from 8 cm deep to more than 16 cm deep. 
Viable spores were present throughout the soil column (Table 3). A bimodal 
distribution is apparent, with peaks in spore numbers at the surface and at a 
depth of 13 cm (Table 3). While some variation existed among cores, spore 
numbers were consistently within the 10° order of magnitude beneath each 
square meter of the soil surface (Table 3). 


DISCUSSION 


Assays of spore numbers in the soil column confirm the presence of a spore 
bank. Spores of both A. pycnocarpon and A. tl of 
the spore bank. Schneller (1979) observed prolonged viability in A. filix -femina, 
which, along with these results, indicates that spore banks are characteristic of 
temperate Athyrium species. 

Significant variation a among months when soil was assayed for viable 
The consistency in the pattern of spore 
germination between the two species, particularly the increases after 1 month 
and 6 months of storage, i tthere mayb 
of the spores assayed. Whatever the cause, assays of soil stored for different 


100 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 3 (1988) 


TABLE 2. Estimated Numbers of —* in Soil Sampled in Wayne National Forest. Estimates are 
r X 1meter X 1 centimeter. Totals are numbers per square 


meter of the. soil iaarkac ts, 


Date sample collected 
Depth 11 June 87 14 June 87 16 Oct 87 16 Oct 87 16 Oct 87 
icm 790,000 580,000 485,000 $30,000 180,000 
2cm 700,000 625,000 207,000 415,000 320,000 
3cm 145,000 185,000 180,000 455,000 360,000 
4cm 350,000 380,000 390,000 485,000 430,000 
5cm 45,000 230,000 375,000 365,000 ; 
6cm 60,000 160,000 195,000 480,000 270,000 
7cm 60,000* 40,000 140, eS 255,000 125,000 
8cm 60,000 50,000 255,000 75,000 
9cm 180,000 30,000 A 155,000 
10cm 225,000 15,000 40,000 190,000 
11cm 425,000 50,000 
12cm 495,000 
13 cm 645, 
14cm 215,000 
15cm 25,000 
16cm 165, 
total 4,585,000 2,295,000 2,002,000 3,640,000 2,224,000 
* sample lost, result estimated. 
Date sample collected 
; Mean 
Depth 1 Jan 88 1 Jan 88 1 Jan 88 1 Jan 88 each level 
1cm 320,000 . 460,000 270,000 - 305,000 480,000 
2cm 465,000 535,000 315,000 325,000 434,111 
3cm 490,000 150, 385,000 165,000 322,444 
4cm 670,000 250,000 205,000 275,000 381,667 
5cm 465,000 5 180,000 195,000 246,111 
6cm 415,000 245, 185,000 190,0 244,444 
7cm 660,000 245,000 415,000 195,000 Ys A 
8cm 230,000 170, 405,000 75,000 
§ 355,000 90,000 385,000 85,000 170,000 
10cm 110,000 10,000 260,000 473,125 
11cm 155,000 100,000 85,000 190,000 167,500 
12cm 115,000 60,000 160,000 85,000 183,000 
13 cm 370,000 230,000 70,000 111,000 285,200 
14cm 65,000 180,000 90,000 55,000 121,000 
15cm 120,000 250,000 95,000 35,000 105,000 
16cm 35,000 100,000 120,000 20, 


: : 000 
+ total ? 5,040,000 $3,350,000 3,625,000 : 2,360,000 4,275,000 


R. G. HAMILTON: SPORE BANKS IN ATHYRIUM 101 


lengths of time cannot be expected to yield homogenous results. Further studies 
may yield data indicating induced spore dormancies such as those described for 
seeds (Baskin & Baskin, 1988). 

Results clearly show a reserve of spores in the soil. Leck and Simpson (1987) 
reported finding 10° spores of Onoclea sensibilis in the soil column beneath one 
square meter of soil in each of a scrub forest, high marsh, and cattail marsh. Since 
this species accounted for about 98% of the fern spores found, the total spore 
densities reported by Leck and Simpson are about 1/1000 of those I have observed 
at the Long Run site. They also reported finding 98.8% of all fern spores in the 
top 2 cm of the soil column, again greatly different from my results. These 
differences suggest that there is significant variation in spore bank 
characteristics among fern taxa. 

Germination strategies are expected to vary as environmental conditions vary 
(Cohen, 1966, 1967). Variation in natural environments occurs between two 
extremes. At one extreme, the environment is constant and of sufficient quality 
to maintain adapted populations. At the other extreme, the environment is 
highly variable, predominantly inhospitable to organisms but occasionally of 
sufficient quality for an organism to complete its life cycle. In favorable 
environments that are constant temporally and spatially, dormancy would not 
be expected. Under these conditions a delay in germination would increase the 
probability of spore death. In variable environments, reproductive success is 
correlated with environmental quality at the time of germination. The cost of 
dormancy (increased spore death) is offset by the benefit gained by avoiding 
germination in low quality environments. When the benefit of dormancy 
exceeds its cost, spores with capacities for dormancy will increase in frequency 
in a population. Models of seed dormancy predict that in many less extreme 
environments, a mixed strategy of bothi diate and delayed germination will 
be most successful (Cohen, 1967; Leon, 1985; Ellner, 1985a). Investigations of 
natural populations have revealed such a mixed strategy. (Cavers & Harper, 
1966; Baskin & Baskin, 1976; Harper, 1977). 

Since seeds and spores represent different stages in the life history of vascular 
plants, models of seed dormancy may be inapplicable to spores. Brown and 
Venable (1986) suggested that models of dormancy should account for its effect 
on each life history stage. This consideration is necessary, since the 
environmental conditions suitable for one life history stage may be lethal for 
another. I suggest the following basic model for determining spore fitness as a 
function of the expected reproductive success of its phenotype. Pteridophytes 
produce spores, which, at the time of production, have some fitness, 


W, = P(g) x P(f) x P({m) x n 


where W, is the absolute spore fitness at the time it is produced, P(g) is the 
probability that it will germinate, P(f) is the probability that a gamete from the 
resulting gametophyte will participate in a fertilization, P(m) is the probability 
that the resulti hyte will reach | maturit is the total 


of spores expected from the mature sporophyte. Fitness is measured as an 
- q: t ce: Bo 42 Ea 
absolute, but it can be easily st d to yield 


102 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 3 (1988) 


The model presented above incorporates the response of a give phenotype at 
each of its life history stages. This model requires an understanding and further 
elaboration of the effect of dormancy on each stage of the pteridophyte life cycle. 
The hypothesis that spore and seed dormancy evolved in response to 
environmental variation is the result of studies of seed plants. Another possible 
role of a spore bank, discussed below, would be unique to pteridophytes. 

The pattern of spore dispersal from ferns is the same as the pattern of seed 
dispersal from angiosperms. Following dispersal, the majority of the spores 
released from a frond are found within a few meters of the frond (Raynor et al., 
1976). This distribution of spores is nonrandom. The soil into which spores are 
dispersed is not a static medium: soil is a very dynamic living ecosystem 
(Adams, 1985; Hausenbuiller, 1978). Small particles within the soil, including 
spores, are moved about by animals. The most active animals within soils are 
earthworms (Darwin, 1881; Adams, 1986). 

Earthworms feed by ingesting the soil within which they move. They retain 
this soil while they digest those materials their digestive systems are capable of 
processing. Earthworm excretions form characteristic structures known as casts 
(Lee, 1985). These casts are excreted both on the surface of and within the soil 
matrix. The net effect of earthworm activity has been studied by observing the 
occurrence of worm casts. 

Lee (1985) summarized investigations that suggest that from 50% to almost 
100% of the aggregates found in the top few centimeters of the soil are earthworm 
casts. Lee calculated that 25% of the Aj soil horizon passes through earthworm 
guts yearly in a temperate pasture. As Darwin (1881) suggested, there is a large 
cumulative effect of earthworm activity. Vimmerstedt (1983) observed that in 
174 days, surface litter moved down into the soil to an average depth of 5.5 cm 
when earthworms were present, but only to 10 mm in soil without earthworms. 
The net effect of earthworm activity is a 55 x increase in the downward rate of 
movement of litter. This movement of soil is critical to any dormant spore, as 
failure to germinate immediately will probably result in its burial by 
earthworms. Earthworm activity also allows for a probability of future 
reemergence on the soil surface. Earthworms, as agents of spore burial and spore 
reemergence, become a significant ecological factor in pteridophyte 
populations. 

worms move horizontally as well as vertically. Unbiased movement of 
earthworms with respect to direction would result in the diffusion of spores 
within the soil matrix. Such diffusion of spores could potentially have a large 
effect on pteridophyte reproductive ecology. The effect of soil diffusion 
depends on a number of factors: 1) the rate of diffusion; 2) the length of time 
ungerminated spores remain viable; and 3) the length of time between spore 
dispersal and spore germination. Spore viability and germination 
characteristics are partly a function of the characteristics of the spore. Rate of 
diffusion is entirely a function of the soil environment. Variation of these factors 
ge. Sore I bability ant ing i g tophyt poputatt Study of 
the effects of spore randomization on pteridophyte reproductive strategies are 
g yw Dterid 2 § 1: . are 


needed t i 


A AMSA CGOD VU 


ae Bak © seo 


regulated. 


R. G. HAMILTON: SPORE BANKS IN ATHYRIUM 103 


LITERATURE CITED 


ADAMS, J. A. 1986. Dirt. vena para ee Press. 

BASKIN, C. C. and J. M. Bask ati gy of herbaceous plant species in a 
temperate region. Ane : ty 75: 286-305. 

BASKIN, J. M. and C. C. BASKIN. 1976. Couataain dimorphism in Heterotheca subaxillaris var. 
subaxillaris. Bull. Torrey Bot. Club. 103:201-206. 

BROWN, 4 ae D.L. VENABLE. 1986. Evolutionary ecology of seed-bank annuals in temporally 

ge environments. Amer. Naturalist 127:31-47 

CAVERS, P. B. and J. L. HARPER. 1966. — polymorphism in Rumex crispus and Rumex 
statin. J. Ecol. 54:367—38 

CoHEN, D. 1966. Optimizing SHEET in a randomly varying environment. J. Theor. Biol. 
12:119—-129. 


. 1967. Optimizing lar tciauaas in a randomly varying environment when a correlation 
may exist between conditions at the time a choice has to be made and the subsequent 
cp prsicy J. Theor. Biol. 16:1-14. 

A Ase model of optimal reproduction in a randomly varying environment. J. 
oo fs 9-228. 
DaRwIn, C. Sib nts origin of species. London: John Murr 
——_. . The formation of vegetable mould sien ‘i action of worms. London: John 


i Tray. 

EAKLE, T. W. 1975. Photoperiodi trol of spore germination of Acrosti Amer. Fern 

65:94—95. 

ELLNER, S. 1985a. ESS germination hpraeio in randomly varying environments. I. Logistic-type 
models. Theor. Pop. Biol. 2 
. 1985b. ESS germination ee in randomly varying environments. II. Reciprocal 
yield-law models. Theor. Pop. Biol. 28:30—116 

GRIME, J. P. 1979. Plant strategies and vegetation Susy’ ib anak eee Wiley. 
ER, J. L. 1977. The biology of plants. New Yor —— cP 

HAUSENBUILLER, R. L. ve Soil science, 2" ed. Dubuiie 


HOWLAND, G.P. and M.E. Epwarps. 1979. Photomorphogenesis of fer gametophytes. Pp. 
394-434 in The experimental biology of ferns, ed. A. F. Dyer. New York: Academic Press. 
IsIkawa, S. and T. OoHusa. 1956. Effects of light upon the germination - spores : rage Il. Two 


& 


light periods of bcp: crassirhizoma Nakai. Bot. Mag. (Tokyo) 69:132. 
KLEKOwWSKI, E.J., JR. 1969. ei abaougo heggne of the Pteridophyta. III. - oy of the 
Blechnaceae. J. Linn. Soc., Bot, 
Leck, M. A.andR. L apenas 1987. ban! Del iver freshwater tidal wetland. Bull. 
Torrey Bot. Club 1 
LEE, K. E. 1985. Ryan us ecology and relationships with soils and land use. New York: 
cademic Press. 
LEON, J. A 1985. Germination strategies. Pp. 129-142 in Evolution: Essays in honour of John 
Maynard Smith, ed. P. J. Greenwood, P. H. Harvey and M. Slatkin. New York: Cambrid, 
LLoyp, R. M: and E. J. _KLEKOWSKI Jr. 1970. Spore germination and — in Pistidephyte: 


Evolutionar y aig inapandeg 2: eganlioe 
Monr, H. 1956 infl 1g der Kei g Fa E t 
Planta 46:534-551. 


Raynor, G. S., E.C. OGDEN, and J. V. Hayes. 1976. Dispersion of fern spores into and within a 
forest. song 78: 473-487. 
SCHNELLER, J. J.1 tigati the lady fern (Athyrium felix-femina). Plant 
acta & Evolution 132: 255-277. 
upoLF, H. 1985. Action and phylogeny of fe ak Pp. 75-80 in Biology of 
eridophytes, ed. A. F. Dyer and C. N. Page. Edinburgh: The Royal Society of of Edi 
STEEL, R. GC. D. and J.H. Torriz. 1980. Principles and procedures of statistics. Toronto: 


raw-Hill. 
Sueal, M. and M. FuRUYA. 1967. Photomorphogenesis s in Pteris vittata. I. Phytochrome mediated 
ti Pl. Cell. Physiol. 8: 737-748. 


£ o 


AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 3 (1988) 


tion and P;, level in 


TOMIZAWA, K., M. SUGAL, and K. MANABE. 1983. 
Lygodium japonicum. PI. Cell aay 24:1043-1048. 


TOwILL, iE R. and H. IkuMa. . Photocontrol of the germination of Onoclea spores. I. Action 
spectrum. se sess ‘s ni 978. 


VIMMERSTEDT, J. P. hworm 
229-240 in p25 ct. ecology, ed. 4. E. Satchell. New York: as and Hall. 


] mining sites in Ohio. Pp. 


American Fern Journal 78(3): 105—108 (1988) 


Cyathea stolzei x ursina, a Distinctive Tree Fern 
Hybrid from Costa Rica 


ALAN R. SMITH 
Herbarium, Department of Botany, University of California, Berkeley, California 94720 
MICHAEL H. GRAYUM 
Missouri Botanical Garden, P. O. Box 299, St. Louis, Missouri 63166 


Recent field work and collections from the Atlantic lowlands of Costa Rica 
have revealed a hybrid between two exindusiate species of Cyathea. This hybrid 
is noteworthy because it combines the characters of two species that are rather 
dissimilar and morphologically distinctive. Both of the parents are acaulescent 
tree ferns of reduced laminar architecture; both are also rather rare and spottily 
distributed, known from relatively few collections, and one was described only 
four years ago (Stolze, 1984). 

The hybrid in question is a cross between Cyathea stolzei A.R. Smith ex 
Lellinger [originally described as Trichipteris pinnata Stolze], from Costa Rica 
and Panama, and C. ursina (Maxon) Lellinger [Trichipteris ursina (Maxon) R. 
Tryon], occuring from Belize to Panama. Both parents and the hybrid were 
found growing together at the same locality: Costa Rica, Pcia. Heredia, Cerros 
Sardinal, ca. 2 km N of Chilamate de Sarapiqui (Finca La Martita), 10° 28’ N, 84° 
04’ W, ca. 80-160 m, Smith et al. 1780 (CR, MO, UC). Cyathea stolzei [Smith et 
al. 1794 (CR, MICH, MO, NY, UC)] was a relatively common fern in the forest 
understory, while C. ursina [Smith et al. 1814 (AAU, CR, MICH, MO, NY, UC)] 
was common in the forest and along trails, especially in somewhat swampy 
sites. A single plant of the hybrid was discovered growing along a trail at the 
edge of the forest. The area was one rather rich in ferns, with about 70 species 
collected, but the only other tree fern species that were found in the area were 
Cyathea trichiata (Maxon) Domin [Trichipteris trichiata (Maxon) R. Tryon], a 
species characterized by its bipinnate-pinnatifid fronds, abundantly hairy axes, 
and trunks several meters tall, and C. multiflora J.E. Smith, also with 
bipinnate-pinnatifid fronds, a small hemitelioid indusium, and trunks 1-2 m 
tall. 

One additional collection of the hybrid has been seen from an adjacent 
property: Costa Rica, Pcia. Heredia, Finca El Bejuco, at southern end of Cerros 
Sardinal (N of Rio Sarapiqui), Chilamate de Sarapiqui, ca. 100 m, 10° 27’ N, 84° 
04’ W, Grayum et al. 4613 (MO). With it also grew the putative parents, C. stolzei 
(Grayum et al. 4608, CR, MO, UC) and C. ursina (Grayum et al. 4621 CR, MO). 
One additional tree fern was found at this site: Cyathea microdonta (Desv.} 
Domin; it is easily distinguished by its bipinnate-pinnatifid fronds and spiny 
rachis and costae. At El Bejuco, C. stolzei grows only on the ridgetops and upper 
slopes; C. ursina occurs only in the swampy stream bottom at the base of the 
ridge. The hybrid is in the intermediate zone—the lower slopes of the ridge. Of 
the two parental species, only C. ursina has been found from the fern-rich (173 


106 


LE1.C 


AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 3 (1988) 


re rere | 


per taxon, are those cited in Fig. 1, all from Finca La Martita. 


C. ursina, and their Putative Hybrid. The plants scored, one 


Character C. stolzei Hybrid C. ursina 
Habitat ridgetops and upper lower slopes ofridge § swampy creek-bottom 
slopes at base of ridge 
Stipe scale shape ovate—lanceolate, lanceolate, 10 x 2mm linear—lanceolate, 
and size 
Stipe and rachis sparse to nearly absent moderately scaly densely scaly 
scale density 
Lamina reduction lower pinnae longest lower 3 or 4 pairs lower 8 pairs 
or 1 pair slightly gradually reduced gradually reduced 
Lamina/stipe ratio 46/31 75/25 62/15 
Lamina dissection pinnae lobed < 1/10 pinnae lobed 1/3—1/2 pinnae lobed ca. 3/4 
their width their width wi 
Lamina texture thick, firm intermediate thin-herbaceous 
No. pinna pairs 8 15-17 28 
Distance between 7cm 4.5cm 3cm 
proximal pinnae 
Pinna shape elliptic intermediate lanceolate 
Ratio of longest to 11/g 12/5—16/5 8/3 
Pinna articulation pronounced intermediate slight 
Costal scales absent few numerous 
Venation lower 2 or 3 converging lower converging lower reaching margin 
below sinus near sinus ove sinus 


total species) area of La Selva, about 7 kilometers distant (Grayum & Churchill, 
1987 


The putative hybrid is intermediate between the two parents in essentially all 
characters, especially including laminar cutting and blade outline, scaliness, 
number of pinna pairs, and venation (Table 1: Fig 1). It is also somewhat larger 
than either of the parents. Unfortunately, spores have been shed from specimens 
of the hybrid, so it has not been possible to determine whether or not they are 
well-formed or viable. Morphologically, the hybrid resembles most closely 
Cyathea phalaenolepis (C. Chr.) Domin, a rare species known only from Pacific 
coastal areas of Colombia and Ecuador. That species can be distinguished by its 
more deeply cut pinnae (incised ca. 3/4 their width), often numerous trichomes 
on the costa abaxially, and lack of reduced basal pinnae. It is tempting to 
speculate that C. phalaenolepis could have arisen through hybridization 


SMITH & GRAYUM: CYATHEA HYBRID 


RR ar ee tage teat 


Fic. 1. Cyathea stolzei, C. ursina, and their putative hybrid. A, frond outline of C. stolzei (Smith et 
al. 1794, UC). B, frond outline of C. stolzei x ursina (Smith et al. 1780, UC). C, frond outline of C. 
ursina (Smith et al. 1814, UC). D—F, pinna outlines of A-C, respectively. H-J, pinna segments of 


A-C, respectively. K—M, stipe base scales of A-C, respectively. 
between C. stolzei and C. ursina and that the hybrid could have become 
stabilized and dispersed by the mechanism of autogamous allohomoploidy, 
described for other Cyatheaceae by Conant and Cooper—Driver (1980). 


108 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 3 (1988) 


The three non-parental species of tree ferns found at the two sites are all very 
unlikely to have participated in the hybridization described herein because they 
are all more dissected (bipinnate-pinnatifid) and belong to different species 
groups within Cyathea (see Barrington, 1978; Tryon, 1976). Furthermore, they 
all have additional characters not shown in the hybrid, e.g., the indusiate 
condition in C. multiflora, the spininess of C. microdonta, and the pubescence, 
rhizome scales, and pale green color of C. trichiata. 

Other hybrids have been reported in this group of Cyathea ( Pichinieas of 
Barrington, 1978). Tryon (1976) listed one putative hybrid between Cnemidaria 
and Trichipteris from Grenada, six hybrids between Cyathea and Trichipteris 
from the Antilles and Central America, and three hybrids between Cnemidaria 
and Cyathea from the Antilles. Conant (1983) reported a number of hybrids in 
the genus Alsophila from the Greater Antilles. We subscribe to the recent 
reclassification of Cyatheaceae by Lellinger (1987) in which Trichipteris (often 
spelled Trichopteris) is reunited with Cyathea. 

Cyathea stolzei, originally known only from the Panamanian type (Stolze, 
1984), is now known from three additional Panamanian collections in addition 
to the Costa Rican ones cited above: Pcia. Panama, Distrito Panama, Cerro Jefe, 
van der Werff & van Hardeveld 6989 (MO, UC); Pcia. Panama, road past Cerro 
Azul, 20 km from jct. with road through Tocumen, Mori & Kallunki 2191 (MO, 
UC); and Pcia. Colon, Santa Rita Ridge road, 8.3 mi E of Transisthmian Hwy., 
McPherson 7464 (MO, UC). 

The range of C. ursina is now known to extend into Panama: Pcia. Coclé, 
Caribbean side of divide at El Copé, 200-400 m, Hamilton & Davidse 2677 (MO); 
Pcia. Coclé, between Cafio Sucio and waterfall at base of Cerro Pife, Sytsma et al. 
2545 (MO). 

We are grateful to Sr. Franco Madrigal and Dr. Thomas Ray for permission to 
collect and field assistance at Finca La Martita and Finca El Bejuco, respectively. 
We also thank Linda Vorobik for the illustrations. 


LITERATURE CITED 


BARRINGTON, D. S. 1978. A revision of the genus Trichipteris. Contr. Gray Herb. 208:1-93. 

Conant, D. S. 1983. A revision of the genus Alsophila (Cyatheaceae) in the Americas. J. Arnold 

Arboretum 64:333-382. 

and G. Cooper—DrIver. 1980. _ Autogamous apg wcthevnt in Alsophila and Nephelea 
h Amer. J. 


(Cyatheaceae): An 
Bot. 67:1269-1288. 

Grayum, M. H. and H. W. CaurcHILL. 1987 [1988]. An introduction to the pteridophyte flora of 
Finca La Selva, Costa Rica. Amer. Fern J. 77:73-89. 

LELLINGER, D.B. 1987 [1988]. The disposition of Trichopteris (Cyatheaceae). Amer. Fern J. 
77:30—94 

SToLze, R. G. 1984. Two new tree ferns from Panama. Amer. Fern J. 74:101—104. 

TrYON, R. M. 1976. A revision of the genus Cyathea. Contr. Gray Herb. 206:19-98. 


F 


co (Memoir Editor, see cove 2). 


INFORMATION FOR AUTHORS 


Authors are encouraged to submit manuscripts pertinent to pteridology for 
publication in the American Fern Journal. Manuscripts should be sent to the Edi- 
tor. 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 ac- 
ceptability of manuscripts nor the date of publication. 

Authors should adhere to the following guidelines; manuscripts not so pre- 
pared may be returned for revision prior to review. Submit manuscripts in tripli- 
cate (xerocopies acceptable), including review copies of illustrations. Do not 
send originals of illustrations until they are requested. Use standard 81/2 by 11 
inch paper of good quality, not ‘“‘erasable’”’ paper. Double space manuscripts 


i e4 


, aut ores’ namesand addresses, text 


and keys), literature. cited, tables (separate from text), and figure captions 


(groupe 


uscript in order just given. Include author s name and page number in upper 
right corner of every sheet. Provid fat least 25 mm all 

pages. Avoid footnotes and do not break wo words at ends of lines. Make table head- 
ings 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 typi- 
fication), use one paragraph per basionym (see Regnum Veg. 58:39—40. 1968). 
Abbreviate titles of serial publications according | to Botanico-Periodicum- 
Huntianum (Lawrence, G. H. M. et al., 1968, Pittsburg} | Library). 
References cited only as part of nomenclatural cathe are not included in litera- 
ture cited. For shorter notes and reviews, put all references parenthetically in 
text. Use Index herbariorum (Regnum Veg. 106:1—452. 1981) for designations of 
herbaria. — 


lj 7 4 > Jal. <1. <2 +L 
WV atatai 


page. Provide margins of at least 25 mm onall 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 pista and line-copy i in single illustrations or blocks. Coordi- 


Par ara SS 5 ‘5 : 
nate sequence g figur es ( ) with order of citation in 
text Explain } ] | eg Fae +} ] : noti ption -Includea 

7 Lae pe Re Al c St.40dn in aarh ma n 


scale and 3 efer ence t 


Proofs and reprint order forms are “sent to che by the printer. Authors 


should send corrected 088 to the editor print orders to the p nter - Au- 3 

_ thors will be a iterations mace atter typ 

nal aed The Chicago manual of style, 13th ed. (1982. ncago: U 

Press). O e guidelines may be justified. Authors 
.d to consult the editor for assist ce whny asa manscit oo 


preparation. 


American 
Fern 
Journal 


Volume 78 
Number 4 
October—December 1988 


QUARTERLY JOURNAL OF THE AMERICAN FERN SOCIETY 


Dark-grown Psilotum 
Dean P. Whittier 109 
gr. a I gts Be sk T.. of S z 
by Protandrous Gametophytes 
Robert M. Lloyd 117 
A Den Review QB. Po) 
Xiaoming Zou and W. H. Wagner, Jr. 122 
Shorter Note 
A Disjunct Station of Aspleni ti ja, with Pellaea atropurpurea 
and P. glabella, in Eastern Ontario 
Derek Munro 136 
Referees, 1988 138 
Index to Volume 78 139 


Information for Authors 


The American Fern Society 
Council for 1988 


JUDITH E. SKOG, Biology Dept., George Mason Uni ity, Fairfax, VA 22030. President 
DAVID B. LELLINGER, Smithsonian Institution, ee. DC 20560 Vice-President 


W. CARL TAYLOR, MilwankneP bli Milwaukee, WI 53233. Secretary 
JAMES D. CAPONETTI, Dept. of Botany, University of Tennessee, Knoxville, TN 37916. Treasurer 
DAVID S , Dept. of Botany, University of Vermont, Burlington, VT 0540 

Records Treasurer 
JAMES D. MONTGOMERY, Ecology III, R.D. 1, Berwick, PA 18603. Back Issues Curator 
ALAN R. SMITH, Dept. of Botany, University of California, Berkeley, CA 94720. _— Editor 
DAVID B. LELLINGER, Smithsonian Institution, Washington, DC 20560 moir Editor 
JOHN T. MICKEL, New York Botanical Garden, Bronx, NY 10458. his = orum Editor 


American Fern Journal 
EDITOR 


MER OMn Dept. of Botany, University of California, 
; Berkeley, CA 94720 


* ASSOCIATE EDITORS 
GERALDJ.GASTONY ............ Dept. of ey: Indiana agli ehe Bloomington, IN 47401 
CHRISTOPHER iat oles EE Se Ses ans Dept. of Botany, University of Kansas, 


Lawrence, KS 66045 
DAVIDE LELLINGER 9.3. U.S. Nat’l Herbarium NHB-166, — eons: 
: : ashington, DC 2056 
TERRY R. WEBSTER .... . Biological Sciences Bie oe of eee Storrs, CT ei 

The ee 8444) ill ly d 1 to th ] 


study of ferns. It is owned by the American Fern Society, and as at the Sua Herbarium, 

University of Vermont, — VT 05405-0086. Second-class postage paid at Burlington, VT, 

and additional entry point 

and orders for back issues should be addressed to Dr. sae D. oe Ecology I, R.D. 1, 

Berwick, PA 18603. 

. Changes of address, dues, and applications for membership should be sent to the Records 
reasurer. 


1 Soh J 3. 2a al c 


ae a 5 1050 net I paid hough an y (agency fee $0.50); * fron in 
members of the American F ety (annual dues, $15.00 + $4.00 mailing surcharge beyond 


_U.S.A., Canada, and Mexico; Reucatcay $300.00 
Back volumes 1910-1978 $5.00 to 36. 25 each; single back numbers of 64 pages or less, $1.25; 
65-80 pages, $2 E $1979 et t seq. ‘$8 .00 


ee en eee a ae nw Pe Sea 


conan a ral 43. 


Vermont, Balingon, VT OS40-0008 


anges to AMERICAN FERN JOURNAL, Dept. of Botany, University of 
Fiddlehead Forum 


COE non-members, including miscellaneous 
te ~— materials, — —— , and reviews of 


Pe al a 


200 Nora oath Stest Seattle, Wexubtic. i Di St hanged and 


American Fern Journal 78(4): 109—116 (1988) 


Dark-grown Psilotum 


DEAN P. WHITTIER 
Department of General Biology, Vanderbilt University, Nashville, Tennessee 37235 


Sporophytes of Psilotum are slow growing plants. This may be one reason 
why Psilotum has been subjected to very few experimental investigations. 
Marsden and Wetmore (1954) found that the apices of aerial stems grown in 
axenic culture often differentiated into rhizome apices. Rouffa (1966) observed 
that Psilotum, which was illuminated with cool white fluorescent light at 
200—400 ft.-c. for extended periods of time, would produce unusual fertile 
appendages. 

Warren H. Wagner, Jr. (pers. comm., 1986) observed unusual plants of 
Psilotum collected in Hawaii by Francis G. Howarth on the floor of a lava tube in 
complete absence of light. These plants had the appearance of short aerial 
rhizomes. The modifications of these plants were apparently caused by their 
growth in the dark, so Wagner suggested that such modified development might 
be produced experimentally by giving greenhouse-grown plants a dark treat- 
ment. The aim of this investigation was to determine whether Psilotum plants 
could grow at all for an extended period in the dark and whether their 
morphology would be altered by such growth. 


MATERIALS AND METHODS 


A plant of Psilotum nudum P. Beauv. growing in a clay pot with a diameter of 
28 cm and a height of 26 cm has been cultivated in the greenhouse at Vanderbilt 
University for over 20 years. This large plant with an extensive rhizome system 
and over 150 aerial shoots was placed in a dark growth chamber in a dark room 
for one year. During the experiment the plant was watered 2 or 3 times a week 
and the temperature varied from 21—27° C. 

After one year in the dark, a sample of the new aerial growth was removed and 
fixed in FAA. The plant was returned to the greenhouse to determine if it would 
continue to grow and if the new aerial growth would return to the typical 
Psilotum morphology. After 4 months in the greenhouse, the new growth was 
examined and fixed in FAA. 

Thin hand sections were used to study the anatomy of the new aerial stems and 
to compare it with aerial stems formed in the light and rhizomes. Toluidine blue 
0 was used as a general stain for the sections (O’Brien et al., 1964). Oil red 0 
(Pearse, 1968) and phloroglucinol (Jensen, 1962) were employed to demonstrate 
lipid and lignin respectively in the sections. The terminology for Psilotum of 
Foster and Gifford (1974) has been employed for the descriptive aspects of this 
study. ; 

Cell sizes are reported with the standard error of the mean. An analysis of 
variance and Duncan’s multiple range test (Li, 1964) were used to determine if 


MISSOUR! BORANICAL 


APR eg 1989 


110 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 4 (1988) 


differences between the means were statistically significant. Means followed by 
different letters are significantly different to the 1% level. 


RESULTS 


The initially green aerial stems remained green for the year in the dark. During 
this period these stems did not undergo any significant growth or unusual 
development. Since observations would have disturbed the soil around the 
plant, no effort was made to study rhizome growth. The main effort was to 
examine any new aerial growth formed in the dark. 

The new aerial stems, which developed in the dark, differed from green aerial 
stems grown in the light (Figs. 1, 2). These aerial stems developed from rhizome 
apices that had turned toward the soil surface. The color of these cylindrical 
stems ranged from white to dark brown. The apices were white, and the stem 
color changed from yellow-brown below the apical region to dark brown in the 
older basal region (Figs. 1, 2). 

The basal portions of the main aerial stems differed from light-grown stems by 
bearing rhizoids (Fig. 3). The rhizoids were short, less than 200 ym long, and 
bicellular. Other portions of the dark-grown stems had stomata and occasional 
rhizoids. The stomata were similar in size and shape to those on green stems; 
however, there were about 1/10 the number per unit area on the dark-grown 
stems. Areas of the dark-grown stems with large numbers of rhizoids lacked 
stomata. 

The apex of the main stem produced enations especially in the later stages of 
stem development (Figs. 1, 2, 3). The mature enations were about the same size 
as enations on green stems; however they were white instead of green. 

The main stems often branched but not dichotomously (Figs. 1, 2). A lateral 
branching resulted from meristematic areas along the stem. The lateral 
meristematic regions were connected by vascular tissue to the stele of the main 
stem. Stems with lateral branches had a modified cylindrical shape; stems 
became slightly fluted for a short distance above the lateral branches. Although 
the cylindrical shape was altered by the branches, the dark-grown stems never 
attained the fluted condition of green aerial stems grown in the light. 

The development of the lateral branches repeated that of the main stems. The 
region of a lateral branch closest to the main stem had a cylindrical shape and 
often bore short rhizoids even though these areas were well above the soil. As 
the lateral branches became longer and increased in diameter, the rhizoids 
became infrequent. The shape of these branches remained cylindrical if no 
additional branching occurred. 

The internal structure of the dark-grown main and lateral stems was similar. 
The basal portions of the stems had protosteles (Fig. 4) and the cortex was 
composed of parenchyma cells. In the basal portions of the older stems, 
occasionally, some of the cortical cells had wall elaborations of phlobaphenes or 
condensed tannins (Fig. 8). The phlobaphene wall elaborations were similar to 
those often found in the cortical cells of Psilotum rhizomes and in the inner 
cortex of older light-grown stems. As long as the diameter of the stem and stele 
remained small, the stele was protostelic (Fig. 4). In most cases, the main and 


D. P. WHITTIER: PSILOTUM 


, 
+ 


* ee 
ia Sent 


Fics. 1-7. The morphology and cst of dark-grown aerial stems of Psilotum. Fic. 1. 
Dark-grown aerial stems. Original green ae to better 

dark-grown stems. Bar = 1cm. Fic. 2. naal regions of dark-grown stems. Arrows s indicate young 
lateral branches. Bar = 1 cm. Fic. 3. Portion of dark-grown stem on short rhizoids an 
enations. Bar = 2mm. Fic. 4. shosparsusegertios gate stem. Bar = 100 pm. Fic. 5. Siphonostele 
with anieiinedah Wh sie of dark-grown stem. B = 100 pm. Fic. 6. Parenchymatous 

dar 


-grown 
Seeeeeneie| 6 em iat woke Oil red O ining. Bar = 


112 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 4 (1988) 


lateral stems increased in diameter above their bases. Under these conditions 
the steles had a larger diameter and became exarch siphonosteles (Fig. 5), with 
isolated groups of xylem cells surrounding a parenchymatous pith. The cortex 
of the siphonostelic regions was composed of parenchyma cells (Fig. 6). Minor 
phlobaphene elaborations were rarely found in the walls of the inner cortex 
cells. There were no sclerenchyma or chlorenchyma cells in the stems grown in 
the dark. 

Because MacDougal (1903), Burkholder (1936), and Kolda (1937) found cell 
size differences between dark-grown and illuminated seed plants, the size of the 
cortical and epidermal cells of dark- and light-grown aerial stems and those of 
rhizomes were measured (Table 1). The length and width of the inner cortex | 
cells of both types of aerial stems were essentially the same (Table 1). However, 
the cortical cells of the rhizome were shorter and wider than the cortical cells of 
the aerial stems. The epidermal cells of the aerial stems, dark- and light-grown, 
were not significantly different at the 1% level (Table 1). The surface cells of the 
rhizome did ieee in size, again shorter and wider, from the epidermal cells of 
the aerial ste 

The pidccaal cells of the dark-grown stems (Fig. 7) were more similar to those 
of the light-grown, green stems (Fig. 13) than those of the rhizomes (Fig. 9). The 
outer wall thickness of the epidermal cells of the dark-grown stems averaged 
about 8 pm. The outer 1 pm of this wall was impregnated with lipid materials 
(Fig. 7, arrow). Surface areas of the dark-grown stems with large numbers of 
rhizoids had epidermal cells with thinner (4 »m) outer walls. The epidermal 
cells of green stems (Fig. 13) had thicker outer walls (14 »m) with lipid material 
in the outer 2 wm of the wall in addition to cuticular material on the surface. The 
outer walls (about 2 wm thick) of the surface cells of the rhizome were only 
slightly thicker than the inner walls of these cells (Fig. 9). Also, these cells had 
only a very thin layer (0.5 pm) of lipid materials impregnated in the outer part of 
this surface wall. 

The steles connecting the lateral branches to the main stems were protosteles. 
If the main stem had a protostele at the level of branching, the protostele 
branched to supply the lateral axis. If a siphonostele was present in the main 
— “igs or two bundles diverged to connect with the protostele of the lateral 

ranch. 

After one year in the dark, the plant was returned to the greenhouse. Some of 
the white apices of the dark-grown stems dried out under the less humid 
greenhouse conditions, but most of them continued to grow in the light. These 
apices, terminal and lateral, gave rise to dichotomously branched, green, fluted 
stems with enations (Figs. 10,11). The large arrows in Figures 10 and 11 indicate 
where the terminal apices of main dark-grown d the normal green 
stems under greenhosue conditions. Smaller arrows show where the apical 
meristems of lateral branches gave rise to the typical green aerial stems of 
Psilotum 

The internal anatomy of these green stems which developed from the 
dark-grown stems had the normal structure for aerial stems (Fig. 12, 13). These 
stems had stomata, a chlorenchymatous outer cortex, a sclerenchymatous 
middle cortex, and a parenchymatous inner cortex (Fig. 13). The steles for the 


D. P. WHITTIER: PSILOTUM 113 


& Pilihapiecs wall deposits 
Fic. 9. Epidermal ¢ celts of a. 


Fics. 8-13. Aspects of dark- and light-grown axes of Psilotum. Fic. 8 
= 100 um. 


arrows indicate where mais apex ‘of dark-grown stem became a green stem. a 
tems. Bars = 1cm. Fic. 


“s a Ld 7 eg os dark-grown 


sioheactrmt of a green stem which developed from a 


eS sah 


12. Si 
stem. ‘Bar = 100 pm. seo 13. Tee wa 
dark-grown stem. Bar = pm. 


TABLE 1. Cell Size Variation in Psilotum Axes 


AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 4 (1988) 


x 


A 


Cell types Green aerial Brown aerial Rhizome 
Cortex 
Length 513.0+9.5a 470.0+16.1a 221.7+6.8b 
Width 53.1+0.9a 56.9+1.2a 111.3 =2.2b 
Epidermis 
Length 256.3+9.2a 245.6+5.1a 131.9+4.7b 
Width 31.3+0:7a 29.1+0.5a 60.5 +2.0b 


larger diameter green stem had the typical exarch siphonostele with groups of 
xylem cells surrounding a sclerenchymatous pith (Fig. 12). The smaller 
diameter tips of the green stems were normal with protosteles and no 
sclerenchyma tissue in the cortex. 


DISCUSSION AND CONCLUSIONS 


This Psilotum plant remained alive and continued to develop during the year 
of darkness. Many factors may have been involved in maintaining this green 
plant in the dark, however, three factors were assumed to be important. The 
demands for organic nutrients should be relatively small because Psilotum 
grows slowly, and the mycorrhizal fungus in the rhizome system should 
continue to supply nutrients even in the dark. Also, this large plant should have 
storage products available for growth. 

Psilotum does not etiolate in the dark. The dark-grown aerial stems show no 
evidence of unusual elongation and, except for the absence of chlorophyll, the 
enations are unchanged. The cortical and epidermal cells in the dark-grown 
stems are essentially the same size as those from the light-grown, green stems. 
The overall brown color of these aerial stems further confirms that etiolation has 
not occurred. 

The dark-grown stems have several charateristics exhibited by Psilotum 
rhizomes. They are brown, cylindrical, and bear rhizoids. In addition, these 
stems branch laterally which is a branching possibility for rhizomes but not for 
green aerial stems. However, if the general morphological appearance and not 
the absence of chlorophyll is considered, the dark-grown stems have more of a 
resemblance to normal aerial stems than to rhizomes. 

The rhizoids on the dark-grown stems are very short and are not obvious to the 
naked eye even in areas with large numbers of rhizoids. Since most of the stem 
surfaces are almost devoid of rhizoids, the dark-grown stems lack the hairy 
appearance of rhizomes which are covered with long rhizoids. It is the 
smoothness of these aerial stems along with the presence of enations which 
make the dark-grown stems appear different from rhizomes. 

Anatomically, the dark-grown stems have characteristics of the normal aerial 
stems. The shift from protostele to siphonostele is similar. Also, the xylem 
configuration in the exarch siphonostele is that of normal aerial stems. The 


D. P. WHITTIER: PSILOTUM 115 


epidermal and inner cortex cell sizes are essentially the same for dark- and 
light-grown stems. Both types of stems have stomata and thickened outer 
epidermal cell walls even though there are fewer stomata and less of an 
epidermal wall thickening for the dark-grown stems. 

The cortex and pith of the dark-grown stems vary from the green aerial stems. 
There is an absence of sclerenchyma cells in the middle cortex and the pith of the 
siphonostelic regions. This would appear to be due to the lack of light during 
development. Most seed plants have a reduction in mechanical tissue if grown 
in the dark (MacDougal, 1903; Burkholder, 1936; Kolda, 1937). Often seed 
plants also have reduced amounts of xylem tissue if grown in the dark (Kolda, 
1937); however, this does not appear to have happened with Psilotum. 

During normal development, rhizome apices give rise to the apices of aerial 
stems which form the above ground portions of Psilotum. The reverse (aerial 
stem apex to rhizome apex) is also known to occur. Marsden and Wetmore 
(1954) showed that the apices of aerial stems could become rhizome apices in 
sterile culture. Bierhorst (1971) described another method for converting the 
aerial stems into rhizome apices. He found that if young aerial stems were buried 
in soil, they would form rhizomes. The apical regions with their enations were 
modified to become rhizome apices. Also, under these conditions rhizoids 
could develop from the immature portions of buried aerial stems. 

None of the green aerial stems became rhizomes during this experiment. The 
plant used in this experiment had completed its annual flush of growth at the 
time it was placed in the dark. Since the development of these green stems was 
essentially complete at the start of the experiment, it would have been difficult 
for rhizomes to form from them. If the experiment had been initiated before the 
development of the green stems was fully determined, there would have been a 
greater possibility for the apices of the young green stems to change their 
development. 

Whether greenhouse-grown Psilotum can be experimentally caused tq form 
other types of aerial growth is unknown at this time. Since the aerial rhizomes of 
Psilotum from the lava tubes are different from the aerial growths of this study, it 
appears that Psilotum can assume more than one form in the dark. Thus, it is 
possible that experimental conditions other than those used in this study could 
cause a different growth habit for greenhouse-grown Psilotum in the dark. 


ACKNOWLEDGMENTS 
I thank W. H. Wagner, Jr. for suggesting the problem. This study was supported in part by the 
Vanderbilt Uni itv Research Council 


LITERATURE CITED 
Breruorst, D. W. 1971. Morphology of vascular plants. New York: Macmillan Co. 
BURKHOLDER, P.R. 1936. Role of light in the life of plants. II. The influence of light upon growth 


and differentiation. Bot. Rev. (Lancaster) 2:97-172. 
Foster, A. S. and E. M. GirForD. 1974. Comparative morphology of vascular plants. 2nd Ed. San 
Francisco: W. H. Freeman and Co. 


116 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 4 (1988) 


JENSEN, W. A. lag oe histochemistry. San Francisco: Ww H. - Freeman and Co. 
KoLpa, - 1937. lie wirk 


Roe a 
achtraglicher Kultur am Lichte. Beih. Bot. Centralbl. 57: sans 380. 
13,3.G. R. 1964. Statistical inference I. Ann Arbor: Edwards Bros., 
MacDouecat, D. T. growth and develor t. Mem 
New Pek Bx Car 2:1-319. 
MARSDEN, M. P. F. and R. H. Wetmore. 1954. In vitro culture of the shoot tips of Psilotum nudum. 
Amer. J. Bot. 41:640-—645. 
O'BRIEN, T. P., N. FEDER, and M. E. vaca tihng Polycl tic staining of plant cell walls by 
O 


Protonlacma 


PEARSE, A.G.E. 1968. Histochemistry, pene and applied. 3rd ed. Vol. 1. Baltimore: 
Williams and Williams C 

RourrFa, A.S. 1966. Induced Paice fertile-appendage aberrations. Morphogenetic and 
evolutionary implications. Canad. J. Bot. 45:855—861. 


American Fern Journal 78(4): 117—121 (1988) 


Experimental Studies on the Probability of Selfing by 
Protandrous Gametophytes 


RoBERT M. LLoypD 
Department of Botany, Ohio University, Athens, Ohio 45701 


Breeding systems have a significant influence on the genetic structure of 
populations (Allard et al., 1968; Baker, 1953; Jain, 1976; Wright, 1969). Many 
species of flowering plants exhibit mechanisms that prevent self-fertilization or 
increase the probability of cross-fertilization. These mechanisms usually 
function in the sporophyte generation and include heterostyly, dichogamy, and 
dioecy. In pteridophytes, the mating system is a function of the gametophyte 
generation. Unlike flowering plants, homosporous pteridophytes are faced with 
a potentially major evolutionary constraint due to the loss of genetic variation in 
progeny after a single generation of intragametophytic selfing. Klekowski (1972, 
1973) hypothesized that homosporous taxa evolved a polyploid genetic system 
coupled with homoeologous chromosome pairing to compensate for the loss of 
variability due to selfing. The evidence that has accumulated to date suggests 
that this type of genetic system is not operative in most species (Buckley et al., 
1985; Haufler, 1985). Analysis of genetic variation of sporophytes in natural 
populations by expression of recessive lethal genes and isozyme marker loci 
indicates that intrag tophytic selfing 1th Iting | ygosity are rare. 

Mating systems in ferns are controlled both intragametophytically and 
intergametophytically. Intragametophytic factors include the temporal and spa- 
cial arrangement of gametangia and the presence of recessive lethal genes 
that prevent zygote development. Variations of gametangial development and 
occurrence include male to hermaphroditic, female to hermaphroditic, initially 
hermaphroditic, and sequentially unisexual. Intergametophytic factors are 
populational phenomena that relate to sex expression (e.g., dioecy, 
antheridiogens) and the density and availability of sperms. Of the 
intragametophytic factors, protandry (male to hermaphroditic gametangial 
sequence) has been hypothesized to have the greatest probability of 
intragametophytic selfing (Lloyd, 1974a). 

The recent discovery of a recessive mutant allele for chloroplast packing in 
Acrostichum danaeifolium Langsd. & Fischer has provided an opportunity to 
test the probability of selfing in gametophytes with a male to hermaphroditic 
gametangial sequence. 


MATERIALS AND METHODS 


Spores were collected from a natural population in Florida, on U. S. Hwy. 41, 
0.7 mile S of State Route 92 (Lloyd 4790b, BHO). Initial screening of 
gametophyte cultures indicated that spores from one sporophyte from this 
population yielded gametophytes of two distinct phenotypes. Spores from this 
sporophyte were sown and gametophytes grown on an Inorganic nutrient 


118 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 4 (1988) 


medium solidified with 1% agar (see Klekowski, 1969, for composition) in 100 x 
15 mm petri dishes under continuous illumination by fluorescent and 
incandescent lamps. About ten days after germination and prior to sexual 
maturation, individual aseptic cultures were established in 60 x 20 mm petri 
dishes by placing gametophytes with different phenotypes immediately 
adjacent as follows: (1) pairs: one mutant (M) with one wildtype (WT); (2) one M 
with two WT; (3) one M with 3 WT; (4) one M with five WT; (5) one M with seven 
WT. In all cultures, wildtype gametophytes were placed equidistant from each 
other, between 0.5 and 1 cm from the side of the mutant gametophyte (e.g., in 
cultures 3—5, WT’s form a triangular to circular pattern around the mutant). The 
number of replicate dishes for cultures 1—5 are, respectively: 45, 20, 15, 17, 16. 
Data from the composite culture reflect random sowing of spores without regard 
to placement (Table 2). The ratio of M:WT gametophytes in composite culture 
was not determined but for the purposes of this report it is assumed to be 1:1 (see 
Warne & Lloyd, 1987). Fertilization was facilitated by the addition of distilled 
water into each culture until much of the gametophyte tissue was covered with 
water, starting seven days following transfer and thereafter every 4 to 7 days for 
55 days. Following fertilization the phenotype of the resulting sporophyte on 
the mutant gametophyte was observed to ascertain origin of the sperm effective 
in fertilization. Gametophyte ontogeny (Table 1) was studied by periodically 
sampling gametophytes grown in composite culture after random spore sowing. 
These gametophytes were mounted in Hoyer’s medium mixed with aceto- 
carmine stain for observation. 


RESULTS 


Genetics of the mutant phenotype.—Gametophytes expressing the mutant 
phenotype h spotted app due to the packing of chloroplasts into one 
end of each cell. This expression is very similar to that of the “‘bar’’ mutation in 
gametophytes from irradiated spores in Osmunda regalis (fig. 4 in Howard 
& Haigh, 1968). Analysis of segregation patterns and expression of the mutant 
indicate that it is due to a single recessive allele (Warne & Lloyd, 1987). 
Sporophytes originating from selfed mutant gametophytes exhibited the same 
abnormal choroplast distribution; sporophytes produced from cross- 
fertilization with sperm from wildtype gametophytes exhibited normal 
chloroplast distribution. 

Sequence of gametangial development.—In composite culture sexual 
development is fairly rapid. In both wildtype and mutant gametophytes, 
antheridia appear prior to archegonia in over 80% of those sampled (Table 1). 
Mutant gametophytes have a fairly prolonged antheridiate st 
prior to becoming hermaphroditic. Wildtype gametophytes have a slight 
rapid growth rate with a corresponding shortened unisexual antheridiate stage. 
Of th tant gametophytes, 6.2% developed both gametangial types by the first 
date of sampling, and fewer than 5% were initially female. Wildtype 
gametophytes exhibited higher frequencies of initially female gametophytes. 


R. M. LLOYD: GAMETOPHYTES 119 


TABLE 1. Sex Expression (%) in Gametophytic Cultures of A. danaeifolium, Mutant and Wildtype 
Phenotypes. DFS = Davs from ; . 4 ity of 2 Ws/ 9 


41 
1 22 prothaili/cm 


oO 


Mutant Wildtype 
DFS Neuter Male Female’ Bisexual Neuter ale Female Bisexual 
34 7i.0 21.9 0 6.2 63.6 30.3 0 6.1 
37 61.6 26.9 3.8 ees 45.2 25.0 9.6 19.4 
41 Z:2 93.5 0 4.3 0 54.8 o.5 HS ed 
53 0 26.6 4.8 66.7 aa a oe cee 


Spatial arrangement of gametangia was similar to that reported previously for 
this species (Lloyd & Gregg, 1975). 

Mating system.—Experiments were designed to test the hypothesis that 
sporophyte ratios obtained following fertilization would result from random 
mating. Genotype frequencies can be predicted with the assumption of random 
mating within the experimental population and the goodness-of-fit to 
expectation may be tested by chi-square (X’) analysis or calculation of exact 
probabilities (Sokal & Rohlf, 1969). The X? is a test of the null hypothesis that the 
observed genotype frequencies in resulting sporophytes do not differ 
significantly from those predicted by random mating. Three aspects of the 
mating system are of importance, proportion (density) of intra- and 
intergametophytic sperm, temporal/spatial arrangement of gametangia, and 
presence of recessive lethal genes (genetic load) which would prevent products 
of selfing from maturing into viable sporophytes. 

To test for the presence of genetic load in this plant, 35 mutant and 35 wildtype 
gametophytes were isolated prior to sexual maturity and allowed to self. In all 70 
gametophytes, viable sporophytes were produced. Therefore, it is concluded 
that such lethal genes are not expressed at stages that would confound the 
experimental results. 

In testing proportion (density) of intra- vs. intergametophytic sperms on 
mating of gametophytes with a male to hermaphroditic gametangial sequence, 
analysis was made of observed vs. expected frequences of mutant vs. wildtype 
sporophytes on mutant gametophytes under the various culture regimes. 
Chi-square analysis (modified with Yates’ correction for size classes fewer than 
five and sample sizes less than 200) and calculation of exact probabilities (for 
sample sizes fewer than 25) indicate that significant departures from expected 
frequencies due to random mating occurred in cultures 1, 2, 3, and the composite 
culture (Table 2). Cultures 1—3 are characterized by a ratio of from 1 to 3 WT to 
1 M gametophyte and deviation from random mating is fairly uniform. In culture 
4 (5WT:1M), the data suggest that a greater number of selfings occur than would 
be expected. In culture 5 (7WT:1M), sporophyte production approaches that 
consistent with random mating. From culture 1 and the composite culture to 
culture 5, as the proportion of WT to M gametophytes increases, there is a 
corresponding decrease in intragametophytic selfing (68—79% to 18% of mutant 


gametophytes). 


120 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 4 (1988) 


Ln oe Fh Sit ge 


TABLE 2. Analysis of O| S ompared t to Random- 
mating Expectations by Chi- -square with Yates’ Corrections (Culture 1 and Cosi or Calculation 
of Exact Probabilities (Cultures 2—5). 


Observed Expected 
Culture Wildtype:Mutant Wildtype:Mutant X? (df1) P 
si 14:31 ea0 62.0 5.68 0.01* 
2 8:12 ge 2 ea ay | O.0t3* 
3 ee (is BATS By sy — 0.004* 
4 122.5 14.225 2.8 — 0.14 
bs 14 ar A — 0.32 
Composite 15°57 So. 736 23.34 0.001* 
* significant 
DISCUSSION 


If we assume that all gametophytes make equivalent number of sperms, in 
cultures 1—5, as intergametophytic sperm density increases the probability of 
intergametophytic mating increases. As these gametophytes express a male to 
hermaphroditic gametangial sequence, the results suggest that gametophytes 
with this gametangial sequence will have an increased probability of selfing 
unless they occur in populations with higher proportions of available 
intergametophytic sperms. In these populations, random mating will occur only 
when gametophyte density and spatial arrangements are such to offset the 
proximal advantage of sperms froma hyte to immediate access to newly 
maturing archegonia on ‘that gametophyte. Therefore, the probability of random 
mating will increase in direct relationship to the increase in density of spore 
deposits. Due to the leptokurtic pattern of spore dispersal exhibited by ferns, the 
highest probability of random mating should be in the immediate vicinity of the 
parental spore source. Although random mating may occur under these 
conditions, inbreeding will most likely result as the vast majority of the spore 
population will originate from a single parent. Gametophytes that occur outside 
of the dense pattern have higher probabilities of selfing and isolated 
gametophytes must be obligately selfed. 

If these results are applicable to natural populations, there should be an 
association between apecies whose gametophytes exhibit the male to 
hermaphroditic sequence < y at marker loci. There is a paucity of 
data from natural populations of ferns both in gametophyte gametangial 
sequences and genetic diversity and it is biased toward temperate taxa from 
larger populations. Most of the evidence to date indicates that populations are 
characterized by substantial amounts of genetic heterozygosity (Haufler, 1985). 
However, in almost all taxa studied, gametangial sequences, when reported, 
have been other than male to hermaphroditic. Two exceptions to this are lava 
flow plants of Nephrolepis palit in Hawaii, which exhibit a male to 

evoid of genetic load, suggesting an origin by 
selfing (Lloyd, 1974b), and species a Ceratopteris, which have very low levels of 
genetic load and a high frequency of individual homozygosity for 


R. M. LLOYD: GAMETOPHYTES 121 


electrophoretic enzyme loci (Warne, 1985). However, gametophyte populations 
in Ceratopteris are diverse in gametangial sequences and may be mediated by an 
antheridiogen. 

If homospory and the genetic homozygosity that result from selfing have been 
an evolutionary constraint in ferns, any adaptation that facilitates crossing will 
be selectively advantageous. It is not unexpected, therefore, that a large number 
of taxa exhibit gametangial sequences other than male to hermaphroditic, 
including those mediated by antheridiogens. As breeding systems are a major 
factor influencing the genetic structure of populations, and as autogamy in 
flowering plants has been suggested to be a derived feature and characteristic of 
taxa in specialized habitats (e.g., colonizers) or with short life-cycles, it can be 
hypothesized that selfing in ferns is also derived and limited to species in which 
genotype coherence has immediate reproductive advantage. 


ACKNOWLEDGMENTS 


It is a pleasure to acknowledge the technical assistance of G. E. Muenchow, T. R. Warne, D. P. 
Buckley, and K. Connolly. This study was supported in part by the National Science Foundation 
(Grant Nos. BMS—7507191 and DEB—7905079). 


LITERATURE CITED 


ALLARD, R. W., S. K. JAIN, and P. L. WoRKMAN. 1968. The genetics of inbreeding populations. 
Advances Genet. 14:55—-131. 

Baker, H. G. 1953. Race formation and reproductive method in flowering plants. Symp. Soc. Exp. 
Biol. 7:114 — 143. 


BUCKLEY, D. P., R. M. Ltoyp, K. M. CONNOLLY, and T. VIERHELLER. 1985. Patterns of inheritance in 
oO 7 "LelinT s ree at 1 5 g SE: 4: 4 2 t sh : Amer. J. 


Bot. 72:920. 

HAUvFLER, C. H. 1985. Pteridophyte evoluti biology: the electrophoretic apy h. Proc. Roy. 
Soc. Edinburgh 86:315—323. 

Howarp, A. and M. V. Haicu. 1968. Chloroplast aberrations in irradiated fern spores. Mutat. Res. 
6:263—280. 

Jaq, S. K. 1976. The evolution of inbreeding in plants. Annual Rev. Ecol. Syst. 7 469-495. 

KLEKOwSKI, E.J. JR. 1969. Reproductive biology of the Pteridophyta. III. A study of the 
Blechnaceae. J. Linn. Soc., Bot. 62:361—377. 
. 1972. Genetical features of ferns as contrasted to seed plants. Ann. Missouri Bot. Gard. 
59:138-151. 
_ 1973. Sexual and subsexual systems in homosporous pteridophytes: a new hypothesis. 
Amer. J. Bot. 60:535-—544. 

Lioyp, R. M. 1974a. Reproductive biology and evolution in the Pteridophyta. Ann. Missouri Bot. 
Gard. 61:318-—331. 
_1974b. Mating systems and genetic load in pioneer and non-pioneer Hawaiian 
Pteridophyta. J. Linn. Soc., Bot. 69:23— 

and T.L. Gregg. 1975. Reproductive biology and gametophyte morphology of 

Acrostichum danaeifolim from Mexico. Amer. Fern J. 65:105—120 

SoxkaL, R. R. and F. J. ROHLF. 1969. Biometry. San Francisco: W. H. Freeman and Co 

Ww ; 


American Fern Journal 78(4): 122-135 (1988) 


A Preliminary Review of Botrychium in China 


XIAOMING Zou’ and W. H. WAGNER, JR. 
Department of Biology and Herbarium, University of Michigan, Ann Arbor, Michigan 48109 


The genus Botrychium has only recently received detailed attention. Since 
Clausen’s monograph in 1938, most publications on this genus have been 
connected with floristic studies. Only Sahashi’s study (1982) endeavored to take 
a broad, monographic approach. As is well known, the genus is difficult 
taxonomically because of subtle characters, variability, rarity, and few and often 

oor speci s of many of the taxa. Ideally the species should be studied in the 
wild as natural populations. Japan, western Europe, and North America are the 
areas best understood. China, on the other hand, has evidently been rather 
poorly collected, and a number of the taxa that were described by the great 
botanist R. C. Ching have been brought into question. For this reason, and in the 
hope of encouraging further investigations of Chinese Botrychium, we have 
prepared this preliminary review. We have maintained the genus intact, having 
recognized no segregate genera, and we have also used a conservative approach 
to species delimitation. 

Through the kindness of Kunghsia Shing and the authorities of the Institute of 
Botany (PE), and the Institute of Forestry and Pedology (SFP), Academia Sinica, 
we were supplied with their Botrychium specimens for study. Dr. Shing kindly 
translated the labels. We also used specimens from the New York Botanical 
Garden (NY). University of Michigan (MICH), United States National Herbarium 
(US), Harvard University (A/GH), and University of California Berkeley (UC). 
Where Clausen had only 13 Chinese specimens to study, we had 191. DeVol 
(1945) listed only three species from east central China. A number of the taxa 
that had been proposed by R. C. Ching were synonymized in keeping with our 
present knowledge of variability in the genus. We extrapolated our knowledge 
of the variations of related New World species to arrive at our conclusions. Most 
of the taxa that have been reduced to synonymy were segregates of two species, 
B. lanuginosum and B. robustum. 

The twelve species that we recognize in China display a diverse array of 
geographical ranges (Fig. 1) correlated with topography, climate, and 
agricultural activities. We did not map the records from the literature because of 
the chances of misidentification. The natural distribution of the species is 
limited to the southeastern half of the moist territory roughly separated by the 
line (Fig. 2) from the Da Hinggan Mountains to the Taihang Mountains to the 
Qingling Mountains, to the eastern and southern edges of the Qing-Zang Plateau 
(Inst. Bot., 1974; Chen et al., 1986; Ching, 1959; and Wu, 1983). An exception is 
B. lunaria, in Xinjiang Province, northwest China (not shown in Fig. 1), where 
the precipitation is higher on the north-faci g slopes of the mountains (Wuetal., 
1980). Within the major distribution area of Botrychium, annual precipitation is 


1 De Ss = 


part t of Forestry, Colorado State University, Fort Collins, Colorado 80523 


ZOU & WAGNER: BOTRYCHIUM IN CHINA 123 


- Ce a | 
B. lunaria @ a . ae 
8. daucifolium @ 


8. boreale B 


p ER 4 oe WE s. nD ms : hues Oh sh 
t ¥ Ui GULIUlS. 


1G. 1: Tg Oh oe Benet eee ot, 


7 


usually greater than 400 mm. Elevation ranges from several hundred meters to 
1000 m in the hilly regions of southeastern China. It increases to 1000-2000 min 
the Loess Plateau and Yunnan-Guizhou Plateau, and finally reaches 4000 m at 
the eastern and southern edges of the Qing-Zang Plateau (Wu et al., 1983). 

The ranges of the species in China are also characterized by disjunctions 
within a given geographical region. This is due apparently to intensive 
agricultural activities. Almost all flat lands in plains, basins, and even in 
mountainous regions have been cultivated for centuries for various crops. Onl 
the mountains were left for non-agricultural usage, but even the slopes of these 


AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 4 (1988) 


Fic. 2. Diagram of Chi Botrychium distribution. Heavy line = western border of all species (B. 
lunaria). A, cold temperate (B. boreale). B, Warm temperate (B. lanceolatum, B. strictum). C, 
East-central subtropical (B. japonicum, B. ternatum). D, Western subtropical (B. lanuginosum to the 
southwest, B. robustum, B. virginianum to the northeast). E, Southern subtropical (B. formosanum, 
B. daucifolium, B. nipponicum). 


were used for tea plantations and for other crops. Most specimens from China 
were collected in the mountains. If there is a disjunction between two mountain 
regions, then there is a disjunction in range. For instance, the highly agricultural 
Sichuan Basin separates the distrib of B. virginianum into two regions, one 
in Hubei province, the other in Yunnan and Guizhou provinces (Fig. 1). 

Seven geographical distribution groups can be recognized on the basis of the 
materials we have studied, two in the temperate zone, four in subtropics, and one 
very widespread. Botrychium boreale appears in the Da Hinggan Mountains in 
far northeastern China. Botrychium lanceolatum and B. strictum together 
comprise a group ranging in central and southern temperate zones. Botrychium 
ternatum and B. japonicum occupy the east-central subtropical areas, whereas 
B. robustum and B. virginianum are scattered in the western subtropical area. 
Botrychium lanuginosum grows at high elevations in Yunnan-Guizhou Plateau 
and in the southeast corner of the Qing-Zang Plateau. To the far south, B. 
formosanum, B. daucifolium, and B. nipponicum occur in the Nanling 
Mountain region—the southern subtropical area. The most interesting 
distribution is that of B. lunaria, which extends all the way from the Da Hinggan 


ZOU & WAGNER: BOTRYCHIUM IN CHINA 125 


Mountains in northeastern China to the south edge of the Qing-Zang Plateau. The 
range of B. lunaria provides the western boundary for the occurrence for all the 
Chinese botrychiums (Figs. 1, 2). Northward and westward from this line, 
grasslands, and sand and gravel deserts prevail, except to the far northwest, 
where we have two records of B. lunaria (see below, species 1). Of the 12 species 
credited to China below, B. boreale, B. lanceolatum, B. formosanum, and B 

nipponicum are the first reports for this country. 

In the enumeration to follow, we have listed only the sy yms that have been 
used for the respective species in China, with some additions where these seem 
necessary for clarification. For illustrations we have used various sources, 
Chinese or otherwise. The collection data have been abbreviated; numbers in 
parentheses following province names indicate numbers of collections seen 
from that province. The discussions under each taxon are aimed at calling 
attention to taxonomic problems, ecology, and human uses. We used C. Y. Wu’s 
“Vegetation of China” to estimate parameters such as rainfall and temperature in 
the different zones. 


Key To Botrychium oF CHINA 


1. Leaf small, 3-15 cm long; leaf primordium glabrous; trophophore (sterile 
blade) pinnate or ternate, 1—2-pinnate 
2. Pinnae fan-shaped, the pinna midrib absent. ............. 1. B. lunaria 
2. Pinnae elongate, the pinna midrib present. 
3. Pinnae rounded at apex, often overlapping; primordial blade 


ee ce a ee 2. B. boreale 
3. Pinnae pointed at apex, usually well separated; ae idl blade 
bending OVOl: . oy icc ee ee he ee ae 3. B. lanceolatum 


. Leaf medium to large, mostly greater than . cm long; ead primordium 
usually hairy; trophophore ternately 2—4-pinna 

4. Sporophore stalk joined to the trophophore ak at or above the base of the 

trophophore blade. 

5. Sporophore stalk joined to the rachis above the base of the trophophore 
blade, AieaiNen variable in positon; leaf very hairy to essentially 
giebrotig. kee ie ee ee et ne 4. B. lanuginosum 

oF Spueephnte stalk joined to the base of the trophorphore blade; leaf nearly 
to entirely glabrous. 

6. Ultimate segments contracted at the base, deeply divided; sporangia 
on 2—3-pinnate sporophore; sporangial cluster deltate. 
ed oe we ee nn ee ete ee 5. B. virginianum 

6. Ultimate segments not contracted, shallowly jobed: sporangia on 

1-pinnate sporophore; sporangial cluster narrowly linear. 
SE a a es Ee a ee ke 6. B. strictum 
. Sporophore stalk joined to the trophophore stalk below be base of the 
trophophore blade. 

7. Leaves sparingly hairy; leaf primordium with many fine hairs; blade 

mostly large, 10—25 cm long. 


i 


rhe 


126 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 4 (1988) 


8. Sporophore stalk joined to the trophophore stalk near the base of the 

Td a is ok ch 7. B. japonicum 

8. Sporophore stalk joined to the trophophore stalk ca. 1/3—2/3 from base 

to trophophore blade. 

9. Trophophore blade up to 3-pinnate, chartaceous; sporophore stalk 
arising ca. 1/2—2/3 of the way from base to the trophophore blade. 
Pee ee re ee eis Bie cm end ccs 8. B. daucifolium 

9. Trophophore blade up to 4-pinnate, herbaceous; sporophore stalk 
arising ca. 1/3—2/3 of the way from base to the trophophore blade. 

he ye A i so ee i ee 9. B. formosanum 
7. Leaves mostly glabrous; leaf primordium glabrous or only the upper part 
hairy; blade mostly smaller, 5-15 cm long. 

10. Terminal pinnules large, differentiated from lateral pinnules, 8—15 
mm long, 5-10 mm wide, elongate, acute; margins finely serrate: 
fronds mostly turning red-brown in winter. .... 10. B. nipponicum 

10. Terminal pinnules smaller, more or less conform with the lateral 
pinnules, 3-10 mm long, 2—8 mm wide; rounded to truncate at the 
apex; margins nearly entire to finely dentate; fronds mostly green in 
winter. 

11. Segments blunt, rounded or subacute; margins entire to shallowly 
and coarsely crenate or dentate: trophophore leathery. 
ee ee 11. B. robustum 

11. Segments mostly acute; margins shallowly to deeply denticulate; 
trophophore herbaceous. .........:......... 12. B. ternatum 


Representative Specimens: Nei Monggol (Inner Mongolia) (2 coll. seen), Da Hinkoeh Mts., Z. 
Wang et al. 1506 (PE). Jilin (2). Changbai Mt., T. N. Liou et al. 1633 (PE). Hebei (6). Chengde, Nan Kai 
430 (PE). Shanxi (3). Taibai Mt., T. N. Liou et al. 2209 (PE). Xinjiang (2). Yili Zhaosu, A. R. Liet al. 
10496 (PE). Sichuan (1). Maerkang, Z. Y. Wu 32726 (PE). Yunnan (5). Zhongdian, K. M. Feng 1689 
(A/GH). Xizang (Tibet) (8). Chayu, Demola Mt., Oing Zang Team 731215 (PE). 

This species is not only the most widespread but apparently the most common 
Botrychium (Fig. 1). This is no surprise, as this plant, the familiar moonwort of 
English-speaking countries, is the most cosmopolitan of all members of the 
genus, completely circumboreal and reappearing in southern South America, 
New Zealand, and Australia. Botrychium lunaria is known in Xinjiang 
Province but without specific locality (PE). It should be noted that this area in 

northwestern China, widely separated to the west from the other localities, is ina 
different drainage system. The moonwort is a very distinctive plant with its 
broadly fan-shaped, simple segments, and it is quite uniform, varying mainly in 
size and in occasionally producing dentate or incised pinnae. 

This species extends from northeastern to southwestern China over many 
degrees of latitude. It requires substantial precipitation, 400-500 mm per year, 
and grows in mountainous regions at elevations of 1300—4000 m. Its various 


ZOU & WAGNER: BOTRYCHIUM IN CHINA 127 


Fic. 3. Silhouettes of selected Chi Botrychi ies. a, B. lanuginosum. b, B. nipponicum. c, 
B. robustum. d, B. boreale. e, B. lanceolatum. 


habitats include semi-moist forest areas or fields: Betula forest, Abies forest, 
grasslands, and open stony pastures. 


2. Botrychium boreale Milde, Bot. Zeit. (Berlin) 15:470, 880. 1857. 
Illustrations: Fig. 3d; Sahashi (1978, p. 52). 
sE ined: NeiM 1 Mongolia). Da Hinggan Mts., Wang et al. 1506 (same 


data for other sheet of B. lunaria, PE). 

Closely related to the much more common B. lunaria, this far northern species 
differs in its ascending, rather pointed pinnae. The occurrence of B. boreale in 
China has evidently not been reported previously. After earlier identifications 
as “B. lunaria,” “B. lunaria var. subincisum,” and “‘B. ramosum”’ (syn. of B. 
matricariifolium), this collection was finally identified correctly by J. Z. Wang 
in 1981. The three specimens of the single collection are entirely typical of the 
species as it occurs across northern Europe, especially in Scandinavia. The 
North American plant confused in the past with it [as B. boreale subsp. 
obtusilobum (Rupr.) Clausen] is a distinct species, B. pinnatum St. John 
oe unpubl.). Thus far, typical B. boreale has not been found in North 
Amer 


128 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 4 (1988) 


This species grows in grasslands scattered with birch trees in the Da Hinggan 
Mountains (Fig. 1) at an elevation of about 1000 m. Average January 
temperatures range from —28 to —38°C and average July temperatures from 16 to 
20°C. Average precipitation is 400-500 mm. No doubt, further exploration in 
this cold region will reveal other populations, these often mixed with B. lunaria. 


3. Botrychium lanceolatum (J. Gmelin) Angstrém, Bot. Not. 68, 1854. 
Botrychium manshuricum Ching, Fl. Reipubl. Pop. Sin. 2:329. 1959.-IsoTyPE: 
Jilin, Changbei Shan, Fusong Xian, N. T. Liou 1663 (PE!). 
Botrychium ramosum Wang Wei et al. in N.T. Liou, Fl. Pl. Herb. Chinae 
Bor.—Orient. 1:21, fig. 23. 1958 (non B. ramosum Asch., 1864). 
Illustrations: Fig. 3e; Tagawa (1959, p. 42). 
Collections E ined: Jilin. Tizhi River, Changbai Mt., T.N. Liou et al, 1663 (another sheet with 
same data at SFP is B. Junaria); W. Wang et al. 2386 (SFP). 


This species is reported here for the first time from China. Evidently very rare 
and local, it is known from only two localities, both on Changbai Mountain (Fig. 
1). Both collections were first identified as B. ramosum Asch. (synonym of B. 
matricariifolium A. Braun), a fairly common amphiatlantic species of western 
Europe and eastern North America. Botrychium lanceolatum has a much 
broader range, being practically circumboreal at high latitudes. It differs from B. 
matricariifolium in its triangular rather than oblong trophophore, its shiny dark 
green rather than dull whitish green color (when alive), and its short, basally 
branched rather than tall, more distally branched sporophore. 


4. Botrychium lanuginosum Wallich ex Hook. & Grev, Icon. fil. 1, pL 79. 1831. 
—Botrypus lanuginosum (Wallich ex Hook. & Grev.) Ching in Shing, 
Glossary terms and names ferns 37. 1982. 

Botrychium yunnanense Ching, Fl. Reipubl. Pop. Sin. 2:329. 1959. —Botrypus 
yunnanense (Wallich ex Hook. & Grev.) Ching in Shing, Glossary terms and 
names ferns 37. 1982.—Type: Yunnan, Dali, Ducloux 55 (PE!). 

Botrychium decurrens Ching, Fl. Reipubl. Pop. Sin. 2:329. 1959. —Botrypus 
decurrens (Ching) Ching in Shing, Glossary terms and names ferns 37. 1982. 
—TyPe: Sichuan, without locality, West China Academy Science 4729 (PE!). 

Botrychium modestum Ching, Fl. Reipubl. Pop. Sin. 2:329. 1959. —Botrypus 
modestus (Ching) Ching in Shing, glossary terms and names ferns 37. 1982. 
—Type: Yunnan, Degqing, Chizhung, K. M. Feng 5668 (PE!). 

Botrychium parvum Ching, Fl. Reipubl. Pop. Sin. 2:330. 1959. —Botrypus 
parvus (Ching) Holub, Preslia 45:277. 1973. —Typk: Guizhow, without 
locality, S. W. Tang 41679 (PE). ; 

Illustrations: Fig. 3a; Ching (1959, pl. 2). 

Representative Specimens: Guangxi (3). Linyin, Miao Mts., R. C. Ching 24772 (PE). Yunnan 
(24). Kong Mt., C. W. Wang 67157 (PE). Sichuan (2). Shimian, J. S. Yin 4672 (PE), Hunan (1). 
without locality, G. Z. He 4856 (PE). Xizang (Tibet) (8). Zhayu, T. P. Yi 79162 (PE). Guizhow (1). 
without locality, S. W. Tang 41679 (PE). 


ZOU & WAGNER: BOTRYCHIUM IN CHINA 129 


Clausen (1938) pointed out that “B. lanuginosum is decidedly variable in 
texture and the amount of pubescence of the leaf.” We might add that this species 
is also extremely variable in size, the fertile leaves ranging from 10 to 75 cm tall. 
The position of the sporophore is likewise variable, usually arising on the rachis 
between the first and second pinna pairs, but sometimes above the second pinna 
pair. One unusual specimen possesses a sporophore at the normal place on the 
rachis plus secondary sporophores on each of the basal costae at the 
corresponding position between first and second pinnule pairs (Yunnan, Yu 
19897). Rarely the sporophore arises between the bases of the two lowest pinnae, 
approximately as in B. virginianum (Fig. 3a). With such variability it is not 
suprising that a number of forms have been described as separate species. Even 
the level of ploidy varies. It is known to be tetraploid (n= 90) and octoploid 
(n=180), and sterile hexaploids are known (Jermy & Walker 1977; Sahashi 
1982). 

Sahashi (1982) reported detailed observations on the habitat of B. 
lanuginosum in Nepal. He found that at lower elevations of 1000-2000 m, it is 
terrestrial, but at higher elevations of 2000+ m, many specimens are epiphytic 
on tree trunks or grow on moss-covered boulders. This is the only Botrychium 
that has been reported to be epiphytic. No specimens have been recorded as 
epiphytes in China. Various notations have been made on our specimens, such 
as open woods, rocky soil, shaded forest floor, and disturbed areas. The plants 
grow on the Yunnan—Guizhou Plateau (Fig. 1) at elevations of 1600—2600 m. 
Average January temperatures in that area are 12—14°C and July temperatures 
28—29°C. Rainfall is roughly 400-500 mm in the eastern Qing-Zang Plateau and 
900—1000 mm in the Yunnan—Guizhou Plateau. 

Ching interpreted his new species B. yunnanense to be intermediate between 
B. virginianum and B. lanuginosum because of the cutting and position of the 
fertile segment just a short distance from the bases of the lowest pinnae. 
However, to us it appears to be just one of many forms of B. lanuginosum. 
Botrychium parvum is just a small form of B. Januginosum. Such low-stature 
plants are commonly found growing together with large ones, and may simply be 
fertile juveniles which are well known in other species of Botrychium. The type 
specimen of B. parvum looks odd partly because the sterile segments and the 
sporophore are still somewhat immature and not fully developed. Botrychium 
decurrens is based upon a single and imperfect leaf with coarse cutting, and the 
normal wings along the upper rachis and costae are somewhat exaggerated. The 
specimen is abnormal, probably badly damaged when found, and possesses only 
one pinna, which the author of the species evidently confused with a whole 
blade. Botrychium modestum was an individual specimen separated from B. 
lanuginosum because of its small size, narrow and remote upper pinnae, and 
other characters known to be highly fluctuating. All of these presumed species 
are well within the expected morphological variability of B. lanuginosum. In 
the future it may be shown that there are specifically distinct elements within B. 
lanuginosum, but this can be justified only on the basis of extensive and critical 
field studies and collections. Highly variable species are well known in the 
genus, such as the widespread B. simplex and the eastern North American B. 


130 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 4 (1988) 


dissectum. No other Chinese species of grapefern is as variable as B. 
lanuginosum. 


5. Botrychium virginianum (L.) Sw., J. Bot. (Schrader) 1800(2):111. 1802.— 
Botrypus virginianus (L.) Holub; C. Y. Wu, Fl. Xizang. 1:34. 1983. 
Botrypus tibeticus Ching in C.Y. Wu, Fl. Xizang. 1:34. 1983.-TypE: Xizang 
(Tibet); Lingzhi, Chinese Medicine Plant Invest. Team 3390 (PE!). 
Illustration: Inst. Bot. (1974, p. 31). 
Representative Specimens: Shanxi (1). Nanwutai Mt., Y. T. Sphien 2025 (PE). Sichuan (2). Fu 
Mt., E. H. Wilson 1033 (PE). Yunnan (4). Weixi, Zhang River, K. M. Feng 3895, 4391 (both PE). Hubei 
(3). Qing County, Wudong Mt., H. C. Chow 2530 (PE). Xi (Tibet) (3). Salwin Kiukiang Divide, G 


oS 


Forrest 21659 (PE). 


This very widespread and familiar circumboreal grapefern is apparently 
frequent in two main areas in China. It does not intergrade with its nearest 
relatives, B. lanuginosum and B. strictum. In Latin America and Antilles, 
however, B. virginianum intergrades with the plant commonly identified as B. 
cicutarium (Savigny) Sw. The type of Botrypus tibetica Ching is merely a very 
large form of B. virginianum. An additional specimen labeled “‘B. laciniatum 
Ching and Kung n. sp.” was evidently not published and was later annotated as 
B. virginianum. 

One wonders whether the similar species B. lanuginosum and B. strictum 
were not confused by Chinese doctors with B. virginianum, as these are often 
confused even by professional botanists. According to Ching (1959) the whole 
plant of B. virginianum is used as a source of medicine. It is purported to give 
relief from poisons, strengthen sexual ability, and aid in recovery from injury, 
especially by snake bite (Inst. Bot., 1974). This species was cultivated in a 
Chinese medicinal plant farm in Xinshan County, Hubei Province (Ching, 1959), 
and, we would assume, was grown elsewhere in the country. 

The range of B. virginianum (Fig. 1) is roughly the same as the much less 
common B. robustum (see below). The two species appear in two places 
separated by the Sichuan Basin, namely Hubei and the Tibetan Regions. 
Botrychium virginianum occurs in valleys to upper mountain slopes and is 
capable of tolerating shade. 


6. Botrychium strictum L. Underw., Bull. Torrey. Bot. Club 20:52. 1902.— 
Osmundopteris stricta (L. Underw.) Nishida, J. Jap. Bot. 27:276. 1952.— 
Botrypus strictus (L. Underw.) Holub; Shing, Glossary terms and names 
ferns 37. 1982. 

Illustration: Tagawa (1959, p. 44). 


I | jitin(6). Antu County, Naitou Mts., T. N. Liou 4253 (PE). Liaoning (1). 
a. Y.C. Zhu et al. 1321 (SFP). Hubei (1). Shennongjia, Sino-Amer. Exped. 1686 


Misidentified as B. virginianum by many authors in the past, this species can 
be readily distinguished from it by characters of the sporophore and the 


trophophore. Although undoubtedly different from it, B. strictum resembles the 
Central and South American B. virginianum var. mexicanum (= B. cicutarium) 


ZOU & WAGNER: BOTRYCHIUM IN CHINA 131 


more closely in certain characters than it does B. virginianum. These include the 
shorter and simpler sporophore, the less divided pinnae, and ap more winged 
costae and costules. As noted above, B t and appears to be 
confluent with B. virginianum, and should probably be merged oe it. 

The whole plant of B. strictum is used as medicine for relief from poisons, and 
for recovery from snake bite injury (Inst. Bot., 1974). 

The range was reported as extending to Shanxi and Yunnan by Ching (1959), 
but all but one of the specimens from China that we have examined are from Jilin, 
in Changbai Mountains. One specimen from Harvard University was collected 
in Hubei. The plants occur on the forest floor at elevations of 600 to 1100 m. The 
species occurs much further north than B. virginianum, but is very rare or absent 
from the Tibetan region of southwestern China. 


7. Botrychium japonicum (Prantl) L. Underw., Bull. Torrey. Bot. Club 25:538. 
1898.—Botrychium daucifolium var. japonicum Prantl, Jahrb. K6nigl. Bot. 
Gart. Berlin 3:340. 1885.—Sceptridium japonicum (Prantl) Lyon, Bot. Gaz. 
(Crawfordsville) 40:457. 1905 

Illustration: Sahashi (1983, p. 243). 

Representative Specimens: Jiangsu(1). Jurong, Mopan Mt., M. B. Deng et al. 3669 (PE). 
Zhejiang(1). Xi Tianmu Mt., H. Migo s. n. (PE). Fujian(1). without locality, O. O. Chang 964 (PE). 
Taiwan(2). Taitotyo, Sinkogun, Kusuhara, T. Suzuki 19618 (MICH). Guangdong(5). Ruyuan, S. P. 
Ko 53622 (PE). Guizhou(3). Fanjing Mt., C. P. Jian et al. 32464 (PE). 


This species, which is very common and widespread in Japan from northern 
Honshu southward, has been reported only a few times from China. Poor 
specimens can be confused with B. formosanum and to a lesser extent with B. 
daucifolium. As might be expected from their respective overall distributions, 
in China B. japonicum occurs considerably to the north of B. formosanum and B 
daucifolium (Fig. 1). Botrychium japonicum is recorded in swamps and wet 
woods, as well as along streams in valleys and along roadsides. Except for 
preferring damp sites, the general characters of the zones in which B. japonicum 
occurs are likely those of B. ternatum, a much more common species. 


8. Botrychium daucifolium Wallich ex Hook. et Grev. Icon. fil. 161. 1829.— 
Sceptridium daucifolium (Wallich ex Hook. et Grev.) Lyon; mab Glossary 
terms and names ferns 37. 1982. 

Botrychium subcarnosum Wallich ex Beddome, Ferns S. India 23, fig, 68, 1863. 

Illustrations: Beddome (1892, p. 470); Sahashi (1981, p. 344). 

Representative Specimens: Yunnan (4). Malipo, Laojun Mt., K. M. Feng 13771 (PE). Gwangxi (1). 
Yao Mts., C. Wang 40529 (PE). Jiangxi (1). Anyun, Jiagang, J. F. Cheng 40248 (PE). Taiwan (1). Taito, 
near Zyonoru, Taito-gun, M. Tagawa 1918 (MICH). 

The group of B. daucifolium, B. formosanum, and B. japonicum (as well as B. 
javanicum Sahashi) involves closely related and variable taxa with distinctive 
but ranges. Poorly developed or sterile specimens are 
ee difficult to ‘identify. We greatly need more and better specimens ¢ of these 
tax 
doncifolict and ‘B. japonicum. ‘Botrychium daucifolium is especially well 


132 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 4 (1988) 


developed in India and Ceylon, ranging eastern and southward to Taiwan and 
the Malay Peninsula. Specimens are easily confused with B. formosanum, 
however, and some locality data may be erroneous. From B. japonicum, B. 
daucifolium differs not only in range, but in its less divided blades with the 
distal segments more adnate, the pinna apices more attenuated and pointed, and 
the tendency of the fertile stalk attachment to be higher, the common stalk 
extending usually to the distal half of the distance between leaf base and the base 
of the trophophore. The chromosomes of B. daucifolium number n = 90; those of 
B. japonicum n=135. The latter number is rare among botrychiums, and is 
interpreted as a hexaploid based on x= 45. 

Botrychium daucifolium grows in the Nanlin Mountain region and in 
northern Taiwan (Fig. 1), at elevations from 600 to 1900 m. The mean annual 
temperature of 20—22° C averages more tropical than that of B. japonicum at 
16—21° C. The difference is more pronounced in winter: 12—14 vs. 5—12° C. 
Judging from label data, B. daucifolium prefers rich and moist sites in valleys, 
slopes, streamsides, and bamboo groves. 


9. Botrychium formosanum Tagawa, Acta. Phytotax. Geobot. 9:87. 1940.— 
Botrychium daucifolium Wallich ex Hook. et Grev., sensu DeVol et al. Fl. 
Taiwan. 1:64. 1975. 

Illustration: Sahashi (1981, p. 344). 


I p (1). Changfeng, Y. Tsiang 43345 (PE). Gwangdong (2). Suni, 
Y. K. Wang 31141 (PE). Gwangxi (1). Yao Mts., C. Wang 40782 (PE). Jiangxi (1). Jiulian Mt., Z. B. 
Yang 2424 (PE). Yunnan (1). Maguan, Gulingjing, S. K. Wu 4166 (PE) 


We have accepted the identification of these specimens on the authority of N. 
Sahashi. He wrote (1982, p. 123) that ‘“This species differs from S. daucifolium 
... in having the blade inserted below the middle of the plant with the longer 


y VOCVY 


range of B. formosanum is especially interesting in this respect, for B. 
daucifolium runs from India and Ceylon to southern China to the Malay 
Peninsula to Taiwan; B. formosanum from southern China and Taiwan and 
north along the Ryukyus to the southern tip of Kyushu; and B. japonicum from 
eastern China and Taiwan, skipping most of the Ryukyus, and then extending 
through much of Japan, except for Hokkaido (Sahashi 1982, p. 164). We suggest 
that for further studies of this complex, efforts should be made in eastern China, 
including Taiwan, to locate mixed populations, in which two or three of the 
species grow together. The distinguishing characters could thus be studied in 
detail (Wagner & Wagner 1983). It seems probable that in Taiwan all three 
species may be found together. Botrychium formosanum is tetraploid with 
n=90, like B. daucifolium. 

The habitat of B. formosanum is probably very similar to that of B. 
daucifolium. In range, B. formosanum (Fig. 1) is evidently similar to that of B. 
daucifolium, but does not extend nearly so far to the west and north. 


ZOU & WAGNER: BOTRYCHIUM IN CHINA 133 


10. Botrychium nipponicum Makino, J. Jap. Bot. 1:5. 1916. 
Illustration: Fig. 3b. 
Collections Examined: Guangxi. Damiao Mts., S. O. Chen 15395 and 16931 (both PE). 


Evidently this species is very rare and local in China, being known at present 
only in Guangxi Province (Fig. 1). It was not recognized as a species, nor even 
cited, by Clausen (1938). Our specimens were identified as B. nipponicum var. 
nipponicum by Sahashi. He has listed numerous localities in Japan (Sahashi 
1982, pp. 192-194, pl. 174). He also records it from near Seoul, Korea. The 
collections above are the first reports for China. It grows in moist sites in forest in 
valleys along slopes. The frond shape is very distinctive; the pinna tips are 
undivided and prolonged, and greatly resemble those of the eastern North 
American B. dissectum f. obliquum. As Sahashi has pointed out, the blades turn 
reddish in the winter. 


11. Botrychium robustum (Rupr.) L. Underw., Bull. Torr. Bot. Club 30:51. 
1903.—Sceptridium robustum (Rupr.) Ching in Shing, Glossary terms and 
names ferns 90. 1982. 

Botrychium longipedunculatum Ching, Fl. Reipubl. Pop. Sin. 2:330. 1959.— 
Sceptridium longipedunculatum (Ching) Ching in Shing, Glossary terms 
and names ferns 90. 1982.-TypE: Yunnan, Kunming, Ducloux 14 (PE!). 

Botrychium sutchuanense Ching, Fl. Reipubl. Pop. Sin. 2:330. 1959.- 
Sceptridium sutchuanense (Ching) Ching in Shing, Glossary terms and 
names ferns 90. 1982.—IsoTyPE: Sichuan, Chenkou, Farges s.n. (PE!). 

Botrychium officinale Ching, Fl. Reipubl. Pop. Sin. 2:330. 1959.—Sceptridium 
officinale (Ching) Ching in Shing, Glossary terms and names ferns 90. 
1982.—PARATYPEs: Sichuan, Nanchuan, Jingfu Shan, C. Pei 7299 and W. P. 
Fang 938 (Both PE!). 

Illustration: Fig. 3c. 
Representative Specimens: Sichuan (5). Danba, Mochige, The 8th team of forest management 

2550 (PE). Yunnan (7). Chienchuan Mekong Divide, G. Forrest 22546 (PE). Liaoning (1). Andong, K. 

Kitsukin 8268 (PE). 


According to Ching (1959), B. multifidum had recently been found in 
northeast China, but no specimens were cited. As we interpret it, B. robustum 
occurs in eastern Russia, Kamchatka, northern Japan and China (south to 33° N), 
and the Aleutians, but true B. multifidum occurs widely in northern North 
America and northwestern Eurasia. In equivalent large fronds, B. robustum is 
more divided than B. multifidum (3—4-pinnate vs. 2—3 pinnate), the segments 
are more overlapping (vs. more remote), more angular (vs. rounded or shallowly 
round-lobed), more lanceolate (vs. more oval), with margins more coarsely 
toothed (vs. entire or finely and shallowly denticulate or crenulate), the tips 
more pointed (vs. rounded to truncate), the veins in dried material are more 
obvious (vs. immersed), and the laminar surfaces somewhat rugulose with 
parallel elongate depressions (vs. plane). (Sahashi and Wagner are currently 
completing a detailed study of the relationships of B. robustum and B. 
multifidum.) 


134 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 4 (1988) 


Botrychium longipedunculatum is apparently a compact sun form of B. 
robustum with an unusually large and well developed sporophore (as is 
common in exposed forms of practically all Botrychiums). The type specimen 
grew ona grassy slope. The other specimen so identified by Ching was found ina 
grassland. Botrychium officinale Ching specimens are more lax and have larger, 
less overlapping segments. The two habitats given are roadside and under 
forest. At best, this is a minor form of B. robustum. Botrychium sutchuanense 
Ching is a form similar to B. longipedunculatum, dwarfed with segments 
somewhat overlapping, presumably due to growth in an exposed habitat. 

Botrychium robustum extends from the Qingling Mountains to the Chang 
Jiang River into the valleys southeast of the Qing-Zang Plateau (Fig. 1), from 
800—4000 m, with an annual precipitation of 600-1000 mm. It occurs in 
disturbed areas, roadside, grassy slopes, and among shrubs. It is used as a 
medicine in China, providing relief from injuries (Ching, 1959). The name “‘B. 
officinale’’ as adopted by Ching probably referred to the medicinal uses of this 
plant. 


12. Botrychium ternatum (Thunb.) Sw., J. Bot. (Schrader) 1800(2): 111. 
1802. pre ternatum (Thunb.) Lyon, Bot. Gaz. (Crawfordsville) 
40:458. 

Illustrations: aa (1959, p. 17); Inst. Bot. (1974, p. 33). 

Representative Specimens: Anhwei (1). Lantien, D. Tsoon 2265 (PE). Jiangsu (2). Jingtan, Mao 
Shan, M. B. Deng et al. 3618 (PE). Zhejiang (1). Mogan Shan, Z. P. Jian et al. 61344 (PE). Fujian(1). 

an, Lianhuafeng, Y. Ling 667 (PE). Jiangxi(10). Pingxiang, Wenjiapo, Jiangxi team 2586 (PE). 
esisih Huayun, L. H. Liu 10093 (PE). Hubei(i). without locality, P. C. Silvestri 3455 (PE). 

Guangxi(1). Zhuang divide, S. H. Zhou 12522 (PE). Sichuan(2). Muchuan, Sichuan economical 

plant team 2147 (PE). Guizhow(9). Chengfeng, Y. Tsiang 4447 (PE). Liaoning(1). Xiuyan, Tangchi, 

W. Wang 1457 (SPF). Guangdong(1). Lohfan Shan, E. D. Merrill 11018 (US). 

This common and widespread eastern Asiatic species is ordinarily quite 
readily distinguished from B. robustum because of its thinner, often delicately 
herbaceous texture, long-stalked pinnae, more strongly pointed, separated 
pinnules, and sharply denticulate margins. Botrychium robustum is generally 
more leathery, has pinnae with shorter stalks, coarser, more parallel-sided, and 

mmonly However, compact sun forms 

ob the two species may ie rather similar. The most distinctive form of B. 

ternatum nai in damp, deeply shady, grassy woods, and has extremely long, 

narrow ted petiolules. The name 
ternatum has at one time or another been applied to practically all of the Chinese 
evergreen grapeferns. 

Although it may be found in dry, open sites, B. ternatum is more commonly 
found on moist floors of dense forest and disturbed woods and frequently in 
bamboo groves. It commonly grows with or near B. japonicum. Its wide range is 
shown i in Figure 1. 


are 


China. It relieves poisoning and 
is used as a treatment for . and to stop nea (Inst. Bot., 1974). 


ZOU & WAGNER: BOTRYCHIUM IN CHINA 135 


ACKNOWLEDGMENTS 


We are indebted to the late R. C. Ching and to K. H. Shing | of the Institute of Botany, Academia 
Sinica, for making this study possible. T ovided us with many t 
data of the specimens which they sent on loan. Michael C. Price gave us much help in various ways 
in carrying out this study. Norio Sahashi of Toho University, Japan, has aided us at all levels. His 
unmatched knowledge of Asiatic and western eae otrychium has been shared with us and 
enabled us to interpret difficult taxonomic problems. We are also grateful to D. B. Lellinger and 
George Yatskievych for their careful reviews and iso comments. This study is part of a 

Botrvchium 


NSF-sponsored project (grant no. BSR-820 


LITERATURE CITED 


BEDDOME, R. H. 1892. Handbook to the ferns of British India, Ceylon and the Malay Peninsula. 
London: W. Thacker 

CHEN, S. P. . ie 1986. Atlas os geo-sci l f Landsat i in China. Beijing: Science 
Pre 


Jv J o Z 


CHING, R. a toe. Flora Annan yi Popularis Sinicae. Vol. 2. Beijing: Science Press 
CLAuSEN, R. T. 1938. A monograph of the Ophioglossaceae. Mem. Torrey Bot. Club 19(2): 1-177. 
DEVOL, C.E. 1975. Cebiadisend at InT. S. Liu etal. (eds.), Flora of Taiwan. Vol. 1. Taipei: Epoch 
Pe ee 
llies of east central China. Mus. Heude Notes Bot. Chin. 7:1-154 + 


INST. Ye hee cap etal. 1974. Flora Tsinlingensis. Vol. 2. Pteridophyta. Beijing: Science Press. 
Jermy, A. C. and T. G. WALKER. 1977. A note on the cytology of Botrychium lanuginosum and the 
occurrence uc the genus in Malesia. Gard. Bull. Straits Settlem. 30:293 
SAHASHI, N. 1978. Morphologic: al pee! taxonomical studies on Cphinglosssios in Japan and the 
vchiu J. Jap. Bot. 53:51—60 
. 1981. Morphological and icon studies on Opeioeiorsates 3 in Japan cad adjacent 
regions te ). J. Jap. Bot. 56:339—347. 
. 1982. ibeeial and taxonomical studies on the oe in Japan and the 
aoe regions. Ph.D. dissertation. Toho University, Jap 
983. Morphological and taxonomical studies on bi ldalodenlex in Japan and the 
ajacent Do (8). New taxa of Sceptridium in Isl. Oshima, the Izu Islands. J. Jap. Bot. 
58:24 


TaGawa, M. loi ches on Formosan ferns. 1. Acta Phytotax. Geobot. 9:87— 
. 1959. Coloured Illustrations of the Japanese Pteridophyta. uae ie Hoikusha 
Publishing Co. : » 

Wace WE eel) A 1983. G tematic tool in the study of 
New World (Ophioglossaceae). Taxon. 32: 51-61. 

Wu,C. Y. oh a. 1980. Vegetation of ‘China, Beijing: Science Press. 

EEAETEA TEES . Flora Xizangica, Vol. I. Beijing: Science Press. 


136 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 4 (1988) 


SHORTER NOTE 


A Disjunct Station of Asplenium ruta-muraria, with Pellaea atropurpurea 
and P. glabella, in Eastern Ontario.—In March 1988, I observed about 200 
plants of Asplenium ruta-muraria L. growing along 800 m of west-facing 
conglomerate rock on the east side of the Rocky Narrows, about 1.5 m above the 
highwater mark of Big Rideau Lake, South Burgess Twp., Leeds Co., Ontario 
(Munro 3110, DAO). These plants grew from 0.5 m to 4.5 m above the highwater 
mark. I also found about 200 plants of Pellaea glabella Mett. ex Kuhn var. 
glabella along about 100 m of this same cliff, growing from 0.3 m to 5 m above the 
highwater mark (Munro 3111, DAO), and about 50 plants of Pellaea 
atropurpurea (L.) Link on the cliffs and on thin grass slopes at the southern end 
of this location. These plants were growing from 3 m to 12 m above the highwater 
mark (Munro 3112, DAO). In June, two sites with an additional 250 plants of 
Pellaea glabella were found 4.5 km and 5.5 km south along the lake on two 
islands (Munro 3114, 3115). 

Wall-rue, Asplenium ruta-muraria, occurs through parts of east and central 
North America including Ontario, Quebec, and Vermont, south through New 
York and Pennsylvania to Alabama and Mississippi and west to Arkansas and 
Missouri (Lellinger, A field manual of the ferns & fern-allies of the United States 
& Canada, 1985). There are also disjunct populations in northern Michigan in 
Chippewa and Keweenaw Counties (Billington, Ferns of Michigan, Bull. 32, 
1952). In Canada, A. ruta-muraria has been known from only two localities in 
Ontario, namely Manitoulin Island and the northern end of the Bruce Peninsula, 
Bruce County. Both locations are in the northern Lake Huron region (Soper, 
Amer. Fern J. 45:97—104, 1955; Dickson & White in Argus & White, eds., Atlas of 
the rare vascular plants of Ontario, Part II, 1983). Asplenium ruta-muraria is 
also found in Quebec along Baie Missisquoi, the northeastern part of Lake 
Champlain. This plant is considered rare for Canada (Argus & White, The rare 
vascular plants of Ontario, Syllogeus No. 14, 1977; Bouchard et al., The rare 
vascular plants of Quebec, Syllogeus No. 42, 1983). 

This new station represents a significant disjunct range for A. ruta-muraria, 
being 230 km west, 375 km east and 195 km north of previously known sites. A 
recent map of the Canadian distribution is given in Cody and Britton (Ferns and 
fern allies of Canada, in press). Asplenium ruta-muraria is a calciphilous fern 
which is commonly epipetric on dolomitic limestone rock. 

Purple cliff-brake, Pellaea atropurpurea, is wide-ranging in North America 
from British Columbia to Quebec, Vermont and Rhode Island, south to Florida 
and west to Wyoming, Colorado, and Arizona (A. Tryon, Rhodora 74:220—241, 
1972). In Canada, several sites are known from southeastern British Columbia, 
adjacent Alberta (Brunton, Canad. Field-Naturalist 93:288—295, 1979). It is also 
found as an outlier along Lake Athabasca, Saskatchewan, in Ontario on 
Manitoulin Island and the northern Bruce Peninsula, Niagara Falls, eastern 
Ontario and western Quebec, and in Montmorency Co., Quebec (Rigby & Britton, 
Canad. Field-Naturalist 84:137-144, 1970; Brunton & Lafontaine, Naturaliste. 


SHORTER NOTES 137 


Canad. 101:937—939, 1974; Cody, Ferns of the Ottawa District, 1978; Dickson & 
White, 1983). This fern is classed as rare in Canada by Dickson and White (1983). 

Smooth clif-brake, Pellaea glabella, occurs through parts of east and central 
North America including Minnesota, Ontario, Quebec, Vermont then south to 
the mountains of Virginia and Tennessee and southwest through Arkansas to 
northern Texas (Lellinger, 1985). In Canada, this fern occurs sporadically along 
the entire length of the Niagara Escarpment in Ontario to Manitoulin Island with 
a few locations in northwestern Ontario and eastern Ontario. There are only two 
geographical locations known for var. glabella in Quebec. These are in Gatineau 
and Richmond counties (Rigby & Britton, 1970; Brunton & Lafontaine, 1974; 
Cody, 1978). This variety of Pellaea glabella is considered rare in Canada (Argus 
& White, 1977). 

The two Pellaea spp. are comparatively more common in eastern Ontario but 
the nearest known stations are about 50 km away in several directions for either 
species. Recent maps of the Canadian distribution of P. atropurpurea are given 
in Cody and Britton (in press). Maps of the distribution of P. glabella for Canada 
and Ontario are found in Cody and Britton (in press) and Brunton and Lafontaine 
(1974). Both species are calciphilous. At this locality, P. atropurpurea plants 
receive at least partial shading from trees or shrubs. On the higher cliff-face 
plants of this taxon grew intimately with A. ruta-muraria along several seams. 
Pellaea atropurpurea exhibited a large and open growth pattern while P. 
glabella grew in fasciculated clusters. Pellaea glabella was found on the more 
exposed and probably drier poriton of the conglomerate rock. Alongone crackin 
the rock, plants of both this taxon and A. ruta-muraria were growing intimately 
together. 

This Big Rideau Lake site is unusual because of the conglomerate rock 
substrate. The nearby rock stratum is Grenville crystalline limestone and quartz 
with belts of rusty weathered gneiss (Wilson & Dugas, Geology of Perth Map 
Area, Map 1089A, 1961). This poorly sorted boulder conglomerate may 
represent the base of the Nepean Formation which is Cambrian in age. The 
conglomerate probably constitutes an anomalous thick section and it lies 
unconformably on crystalline Precambrian rock and is derived from the same. 
The pebbles and boulders are mostly rounded marble with angular fragments of 
quartz, marble and granite pegmatite minerals cemented into the loose matrix. 
Cleavages of calcite crystals are also present and are indicative of quick reburial 
with little chemical action and weathering. This conglomerate cliff lies within 
100 m of a fault-line and probably represents an eroded fault-line scarp (pers. 
comm., Gary Ansell, Geological Survey of Canada, Ottawa). An application of 
15% HCl produced vigorous bubbling, indicative of carbonate. The pH of the 
rooting medium was measured from two substrate samples at 6.9 and 7.1. This 
cliff faces west and would be in the afternoon sun for several hours in the 
summer. However, many of the ferns were either growing in the shade of 
overhanging trees or occurred in cracks where white calcium deposits could be 
seen, evidently from seepage. The loose structure of the conglomerate provides 
an ephemeral habitat for the ferns. As noted above some plants grew within 


138 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 4 (1988) 


0.3 m of the highwater mark and would be washed by wave action on windy 

ays 

The dominant tree species were Pinus strobus, Thuja occidentalis with many 
Juniperus virginiana. Other woody plants included Arctostaphylos uva-ursi, 
Betula papyrifera, Fagus grandifolia, Juniperus communis, Lonicera dioica, 
Pinus resinosa, Ribes sp., Sambucus pubens, Shepherdia canadensis, 
Symphoricarpos albus, Tilia americana, Tsuga canadensis. The forbs included 
Aquilegia canadensis, Carex eburnea, Linaria vulgaris, Minuartia michauxii, 
Oryzopsis asperifolia, Pedicularis canadensis, Poa compressa, Polygonatum 
pubescens, Saxifraga virginiensis, and Viola adunca. Other ferns growing above 
the cliff face included Cystopteris fragilis s. lat., Dryopteris marginalis, and 
Polypodium virginianum. This locality represents an interesting range 
extension for these three fern species. Interpretation of the local geology maps 
suggests that further locations may be found for these rare ferns. This station is 
located on private land but the conglomerate cliff will probably prevent 
extensive human damage to the plants. 

The author thanks G. Baillargeon, P. Catling, and W. Cody of our centre for 
constructive comments.—DEREK Munro, Biosystematics Research Centre, 
Agriculture Canada, Ottawa, Ontario, Canada K1A 0C6. 


Referees, 1988 
I thank the A iate Edit referees listed below for their valuable assistance in the review 
process. Their evaluations of manuscripts submitted to American Fern Journal have aided authors, 
made my j job easier q y of our journal.—ALan R. SMITH 


GORY J. ANDERSON Davin M. JOHNSON Douctas E. SoLtis 
Davin E. BILDERBACK JANE L. KoTENKO PAMELA S. SOLTIS 
L BISHOP BLANCA LEON B DARLENE SOUTHWORTH 
Davip BouFFoRD JOHN T. MICKEL ROBERT G. STOLZE 
D M. BRITTON BRENT MISHLER W. CARL TAYLOR 
Davip S. CONANT RoBBIN C. MoRAN Davin H. WAGNER 
DANIEL CRAWFORD y A. Paris W. H. WAGNER, JR 
Dona Lp R. F. JAMEs H. Peck RICHARD A. WHI 
R. JAMES HICKEY MicHaet G. Price JAMEs W. WALLACE 
Leste G. HicKOK J S. PRINGLE MICHAEL D. WINDHAM 
RICHARD E. HOLTTUM GEORGE PROCTOR GEORGE YATSKIEVYCH 


BARBARA JOE HOSHIZAKI 
P. HOVENKAMP 


Index to Volume 78—1988 


Classified entries: botanical names (new ite’ in n boldface); major subject headings, Pappecitie 


key words from titles; reviews (grouped, listed by 
byt 


of Contents, iii. 


Acrostichum danaeifolium, experimental 
studies in selfing, 1 

Asplenium: A. ruta-muraria, 136; A. viride, 72 

Athyrium: A. pycnocarpon, 77, 96; A. 
thelypterioides, 96 

Botrychium: key to spp. in China, 125; B. 


133; B. lunaria, 126; B. matricariifolium, 128; 
B. modestum, 128; B. manshuricum, 128; B. 
nipponicum, 133; B. officianale, 133; B. 


ternatum, 134; B. virginianum, 130; B. 
yunnanense, 128 

Botrypus tibeticus, 130 

Calochlaena, 86: C. dubia, 92: C. javanica, 93; 
C. novae-guineae, 93; C. straminea, 92; C. 
villosa, 93 


si Abinennalenoes 14; key to species, 29; C. 
Cc 


decurrens, 29; C. densifolium, 19; C. 
ensifolium, 32; C. falcoideum, 32; C. fallax, 
31; C. fuscosquamatum, 21; C. herbaceum, 
31; C. inflatum, 22; C. peace 24.°G, 
lapathifolium, 33; C. latum, 29; C. 
leucorhizon, 34; C. lorentzii, 33; C. 


nili disaimiur:. 30; C. occultum, 31; C. 

ophiocaulon, 31; C. oxypholis, 32; C. 

pascoense, 29; C. phyllitidis, 29; C. 

remotifolium, 26; C. repens, 32; C. rigidum, 
sol 


28; C. xalapense, 30 


139 


tl listed Ipk beti lly SHEER 


Ceradenia, 1; subg. Ceradenia, 4; subg. 
Filicipecten, 5; C. albidula, 4; C. 
runneoviridis, 5; C. capillaris, 4; C. curvata, 
4; C. discolor, 4; C. farinosa, 4; C. fendleri, 5; 
C. fragillima, 4; C. fucoides, 4; C. herrerae, 4; 
jungermannioides, 4; C. kalbreyeri, 5; C. 
knightii, 5; C. kookenamae, 5; C. 
longipinnata, 5; C. margaritata, 5; C. 
mayoris, 5; C. melanopus, 5; C. meridensis, 
5; C. nubigena, 4; C. nudicarpa, 5; C. pearcei, 
4; C. pilipes, 4; C. podocarpa, 5; C. pruinosa, 
5; C. semiadnata, 5; C. spixi 
i biases numbers: Dinlexium flavoviride, 
81; Isoétes x hickeyi, 7; I. pallida, 35 
Cochlidium, 1 
Ctenopteris: C. curvata, 2; C. meridensis, 2 


aoe C. phalaenolepis, 106; C. stolzei, 105; 
ei X ursina, 105; C. ursina, 105 
hs ahs 105 
Dennsta ov 94 
Dicksoniaceae, 86 
Diclactopele: D. cavaleriana, 79; D. javanica, 79 
iplazium: D. dilatatum group, 80; D. 
flavoviride, 77; D. mettenianum, 80; D. 
7 


pycnocarpon, 

Ecology: Phyllitis scolopendrium in New York 
State, 37; spore banks in — 
pycnocarpon and A. thelypterioides, 9 

pln gig Isoétes, 7; Pellaea iii 
com 

nie: experimental studies on selfing 
in Acrostichum danaeifolium, 117 

Genetics, mutants in Acrostichum 


Geographical distribution: Hotivehinin SPP.» ' 
122; Campyloneurum, Genus communities 


Grammitis: G. ce 1; G. graminea, 1; G. 
ohne 4 

Gymnoc m dryopteris, 71 

Hyalo aha anetioides, 14 

Rice ae Cyathea, ne Isoétes, 6 

es, 6; I. cubana, 35; I. echinospora, 6; I. 

ihe 6; I. macrospora, 6; I pallida, 35; I. 
panamensis, 35; I. triangula, 35 


140 AMERICAN FERN JOURNAL: VOLUME 78 NUMBER 4 (1988) 


Life cycle variations: agamospory in Pellaea, 44 
. alopecuroides, 73; 
x prostratum, 73; L. 
appressum, 73; xbrucei, 73; 
carolinianum, 73; L. xcopelandii, 73; L. 
digitatum, 73; L. lucidulum, 73 
Marsilea: M. sect. Clemys, 68; M. berhautii, 70; 
de on 70; M. 


ye pes, 6 
a systems, in ‘Acrestichum a. 


pes: eee 
Morphology: ahi tichum gametophytes, 117; 
Calochlaena, 87; Culcita, 87; Diplazium and 
Diplaziopsis, 79; Psilotum nudum, 
tgs 109 
Niphidium, 14 
pete electrophoresis, 47; P. atropurpure 


Psilotum nudum, 1 
Range extensions: ne eastern Ontario, 
Asplen 
Gcrchiias spp., 
kansas, Lycopodium alopecuroides, 73; L. 
alopecuroides xX prostratum, 73; L. 


Gymnocarpium dryopteris, 71; Oregon: 
Asplenium viride, 72 
Referees, 1988, 138 
Reviews: Mickel, J. T., with R. McVaugh, S. 
ell, and H. Balslev, Liebmann’s Mexican 
ferns: His itinerary, a translation of his 
’Mexicos Bregner,’ and a reprinting of the 
original work, 75; Moran, R. C., Monograph 
of the Neotropical fern genus Polybotrya 


(Dryopteridaceae), 75; Riba, R. and A. 
Butanda, Bibliografia comentada sobre 
pteridofitas de México, 75 

Spore banks, in Athyrium py 
thelypterioides, 96 


, 000; P. atropurpurea x glabella var. 
cteeeanaleht 47; P. glabella complex, 44; P. 


andA 


occidentalis, 44; P. glabella var. simplex, 44; 
P. occidentalis subsp. occidentalis, 64; P. 
occi is subsp. simplex, 64 
Phyllitis scolopendrium var. americana, 37 
Phytogeography: Diplazium flavoviride, D. 
pycnocarpon, and Diplaziopsis, 77; between 
tern North America and eastern Asia, 84 


Dipl : [soétes 


‘echtnospors, 71 ane laae 6; I. macrospora, 


Statement of Ownership, Management, and Circulation 


Publication title and number: American Fern Journal (0002-8444). Date of 
filing: 9 March 1988. Frequency of issue: Sol a Annual subscription price: 
$20.00. Office of publisher: Pringle Herbarium, Dept. of Botany, University of 
Vermont, Burlington, VT 05405-0086 on Editor: Alan R. Smith. The American 
Fern Journal is wholly owned by the American Fern Society with no bond 
holders. The purpose, function, and nonprofit status of the Society and its tax 
exempt status for Federal income tax in past years. 
The press run for each issue is 1400, of which (just prior to the filing date) 1014 
copies were mailed as paid circulation, and 15 were mailed as free distribution, 
leaving 371 copies for office use and back-issues sales. I certify that these 
statements are correct and complete.—Davip S. BARRINGTON, Records 
Treasurer of 


INFORMATION FOR AUTHORS 


Authors are encouraged to submit manuscripts Pertinent to pteridology for 
publication in the American Fern Journal.M 1 be sent to the Edi- 
tor. 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 ac- 
ceptability of manuscripts nor the date of publication. 

Authors should adhere to the following guidelines; manuscripts not so pre- 
pared may be returned for revision prior to review. Submit manuscripts in tripli- 
cate (xerocopies acceptable), including review copies of illustrations. Do not 
send originals of illustrations until they are requested. Use standard 81/2 by 11 
inch paper of good quality, not “erasable” paper. Double space manuscri 
throughout, including title, authors’ names and addresses, text (including heads 
and keys), literature cited, tables (separate from text), and figure captions 


(groupe figures) ‘A man- 
uscript in order just given. Include author’ s name and page number in upper 
right corner of every sheet. Pr f at least 25 mm typed 


pages. Avoid footnotes and do not break words at ends of lines. Make table head- 
ings 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 typi- 
fication), use one paragraph per basionym (see Regnum Veg. 58:39—40. 1968). 
Abbreviate titles of serial publications according to Botanico-Periodicum- 
Huntianum (Lawrence, G. H. M. etal., 1968, Pittsburgh: Hunt Botanical Library). 
References cited only as part of nomenclatural matter are not included in litera- 
ture cited. For shorter notes and reviews, put all references parenthetically in 
text. Use Index herbariorum canes Veg. 106:1—452. 1981) for ——— of 
herbaria. 


- rad , & ¥.3 3. 4: 34. is Lee Pe Lp Saupe 42 +h 


page. Provide margins of at lea: . For, continuous-tone 

illustrations, des ducti ithout or by uniform : 

amount. In composite 1e blocks, ‘abut ‘edges of haut iekusenhe Avoid com- 

bining continuous-tone and line-copy i in single i illustrations or blocks. Coordi- 
dad 


nate sequence an is 5 (and i tables) of citation i in 
text C1.: } fk tn tt ] ti si Include a 
Sealed e PS SC See me i Oe es be eek eae : 

- Proofs and reprint order forms are sent to authors by the printer. jon 

should send corre: p : orders to the printer. Au- 

thors will be assessed charges for extensive alterations t as Se aa ee : 

ae gs ‘ ye ae es scwcd American Fern Jour- oe i 
j “8 


nal and The Phineas manual of style, 13th ed. (1982. Chicago: Univ. Ch ae 
Press). Occasionally, departure from these —— may be arta ativan. : 
are encouraged to consul nsult the ditor f y aspect ot anuscript 
preparation. — ae 
Papers longer than 2 printed pages may been to the Editor of of Pleridologia oe 
Pee Sie, see cover oo oo