AMERICAN FERN JOURNAL Volume 93 Number1 January-March 2003 QUARTERLY JOURNAL OF THE AMERICAN FERN SOCIETY e Gametophyte of Diphasiastrum s Polypodium appalachianum: An Unusual Tree Canopy Epiphyte in the Great Smoky Mountains National Park Harold W. Keller, Paul G. Davison, Christopher H. Haufler and Damon B. Lesmeister Shorter Notes The American Fern Society Council for 2002 CHRISTOPHER H. HAUFLER, Dept. of Botany. University of Kansas, Lawrence, KS 66045-2016. TOM RANKER, I as Box 265, University of Colorado, Boulder, CO 80309-0265. W. CARL TAYLOR, 800 W. Wells St., Milwaukee Public Museum, Milwaukee, WI 53233-1478. Secretary JAMES D. CAPONETTI, Dept. of Botany, University of Tennessee, Knoxville, TN 37916-1110. GEORGE YATSKIEVYCH. Missouri Botanical Garden, P. O. Box 299, St. Louis. MO 63166-0299. JAMES D. MONTGOMERY, Ecology III, 804 Salem Blvd. Berwick, PA 18603-9801. Back Issues Curator R. JAMES HICKEY. Botany Dept . Miami University. Oxford. OH 45056. Journal Editor DAVID B. LHI I IM.i !• '• IRC-166, Smithsonian institution. Washington. DC 20560-0166. Memoir Editor CINDY JOHNSON-GROH, Dept. of Biology. Gustavus Adolphus College, 800 W. College Ave., St. Peter, MN 56082-1498. Bulletin Editor R. JAMES HICKEY American Fern Journal EDITOR Botany Departmeni.Miami University. Oxford, OH 45056, ASSOCIATE EDITORS GERALD J. GASTONY Dept. of Biolouv. Indiana University. Bloomington, IN 47405-6801 GARY K. GREER Dept. of Biolog) WV 25112-1000 CHRISTOPHER H. HAUFLER .... Dept. of Botany. University ot Kansas. Lawrence. KS 66045-2106 JAMES H. PECK msas—Little Rock, 2801 S. University Ave., Little Rock, AR 72204 The "American Fern Journal" (ISSN 0002-K44- is mi stra ed quarterly devoted to the general study of ferns. It is owned by tib at The American Fern Society, r r Missouri Botanical Garden. P. O. Box 299. St. Louis. MO 63166-0299. Periodicals postage paid at St. Louis, MO. and additional entry. and orders for ... Blvd. Berwick, PA 18603-9801. agency fee. if up ul dues, $25.00 + $7.00 mailing surcharge beyond U.S.A.. Canada, and Mexico, life membership. $300.00 + $140.00 mailing surcharge beyond U.S.A., Canada, and Mexico). Back volumes are available Please contact the POSTMASTER: Send address changes to Box 299, St. Louis, MO 63166-0299. AMERICAN FERN JOURNAL, Missouri Botanical Garden, P. O. FIDDLEHEAD FORUM SPORE EXCHANGE Ms Denia Mandt. 12616 Ibbetson Ave., Downey. CA 90242-5050. is Director. Spores exchanged GIFTS AND BEQUESTS in terns Back i!e tax-deductible. An Evaluation of Sceptridium dissectum (Ophioglossaceae) with ISSR Markers: Implications for Sceptridium Systematics i.•• ,,»•!, i icillv. \annus murp hoi unit's li,i\c been mi liulcd within S tliss'-etum. For example, ograph recognized five infraspecific taxa in S. dissectum, of which only the >f variety dissectum and obliquum are currently retained. However, the taxonomic o varieties has been debated. We used ISSR (Inter-Simple Sequence Repeat) markers . dissectum var. dissectum and var. obliquum in 17 Ohio populations. Five ISSR primers generated 69 reproducible loci. In UPGMA analyses and AMOVA, S. dissectum var. dissectum individuals did not cluster separately from var. obliquum individuals, nor did individuals from the same population cluster together. ISSR markers revealed levels of population genetic structure in S. dissectum similar to levels detected by previous isozyme investigations. Our results concur with recent treatments nt S ,//\s,-( turn that do not formally recognize infraspecific taxa, and may bring into question current species circumscriptions in Sceptridium. We illustrate the use of Sceptridium. Species of Sceptridium Lyon, the evergreen grapeferns, are common members of temperate and north temperate habitats, though the genus has a worldwide distribution (Wagner and Wagner, 1983). In North America, the center of species diversity lies east of the Mississippi River to the Atlantic Coast, and from the southern Gulf Coast to the northern coasts of the Great Lakes (Wagner and Wagner, 1993). Within this range, Wagner and Wagner (1993) recognized seven species, and it is not uncommon to find more than one species at a single site (Wagner, 1960a). Most Sceptridium species inhabit a variety of moderately disturbed habitats such as secondary-growth woods, old fields, and grassy slopes, although some species may occur in more undisturbed habitats (Clausen, 1938). Species of Sceptridium, like other members of Ophioglossaceae, generally produce one epigeal leaf per year, which is divided into a sterile trophophore and a fertile sporophore (Clausen, 1938). Unlike some members of the family {e.g., Botrychium s.s.; Wagner, 1990), Sceptridium species do not always produce a sporophore, and under stressful conditions may not produce a trophophore (Wagner, 1960b; Montgomery, 1990; Wagner and Wagner, 1993; Kelly, 1994). The leathery, photosynthetic trophophore persists through the winter, hence the moniker "evergreen" grapefern. Sceptridium species, as well t of Botany, Miami I "niwrsitv. ( MISSOURI BOTANICAL MAY 1 9 2003 GARDEN LIBRARY FERN JOURNAL: VOLUME 93 NUMBER 1 (2003) as other members of the family, possess subterranean, non-photosynthetic, mycoparasitic gametophytes (Wagner et al, 1985; Melan and Whittier, 1989). The subterranean nature of the gametophytes may be associated with high selffertilization rates (Tryon and Tryon, 1982) as has been documented in some Sceptridium species (McCauley et al, 1985; Watano and Sahashi, 1992). S. DISSECTUM (SPRENG.) LYON.— Sceptridium was first recognized as a genus by Lyon (1905) after observing the embryo morphology of Botrychium dissectum var. obliquum (Muhl.) Clute. Lyon (1905) found that the embryo of B. dissectum var. obliquum differed from the embryo of B. virginianum (L.) Swartz by possessing a long suspensor that lacked a pronounced lateral cotyledon and a root that emerges from the basal side of the gametophyte. Further, Lyon (1905) noted that most of Underwood's (1898) ternate Botrychium species, of which B. dissectum var. obliquum was included, had a sporophyll that divided into a trophophore and a sporophore near the rhizome. On this basis, Lyon placed most of Underwood's (1898) ternate Botrychium species in the genus Sceptridium, anticipating each would possess these three characters. Most North American taxonomists have treated Sceptridium as a subgenus of Botrychium (Clausen, 1938; Lellinger, 1985; Wagner and Wagner, 1993). However, other authors have maintained Sceptridium as a separate genus within Ophioglossaceae (Sahashi, 1979; Kato, 1987; Watano and Sahashi, 1992; Hauk, 1996). Of the seven currently recognized North American Sceptridium species, S. dissectum is the most variable morphologically (Wagner, 1960b; Wagner and Wagner, 1993). Commenting on S. dissectum's variability, Wagner (1960b) stated that "Botrychium [Sceptridium] dissectum Spreng. is so outlandishly variable that it has apparently misled botanists in delimiting other, closely related, but more uniform, species correctly..." The diversity of blade morphologies encompassed by S. dissectum has led to taxonomic disagreement over what range of variation should be included within S. dissectum, and what putative segregates deserve recognition as distinct species (Clausen, 1938; Wagner, 1960a; Wagner, 1960b; Wagner, 1961). Sceptridium dissectum was described by Sprengel in 1804 as Botrychium dissectum, and only sporophytes of the more dissected morphology were included in the species by early authors (Sprengel, 1804; Underwood, 1898). Sporophytes possessing relatively less dissected and a more broadly laminated blade morphology were ascribed to B. obliquum Muhl. (Underwood, 1898). Variations on these names existed, for example, Prantl (1884) recognized B. obliquum and B. obliquum var. dissectum (Spreng.) Prantl. Until Clausen's monograph (1938), nomenclatural chaos existed concerning the taxonomic limits of B. dissectum (for a complete list of synonyms see Clausen, 1938). In his monograph of the Ophioglossaceae, Clausen (1938) treated B. dissectum as four varieties and one subspecies: B. dissectum var. typicum [dissectum), var. obliquum (Muhl.) Clute, var. oneidense (Gilbert) Farw., var. tenuifolium (Underw.) Farw., and subspecies B. dissectum ssp. decompositum (Mart. & Gal.) Clausen. Of Clausen's five infraspecific taxa, three have been TAXONOMIC HISTORY OF SCEPTRIDIUM AND BARKER & HAUK: ISSR MARKERS IN SCEPTRIDIUM DISSECTUM 3 elevated to species or subsumed into other taxa. Only varieties dissectum and obliquum (Fig. 1) remain designated as varieties (Clausen, 1938), forms (Wagner, 1960a; McCauley et al, 1985), or not officially recognized but their morphologies mentioned (Lellinger, 1985; Wagner and Wagner, 1993). To have working taxa for analyses and discussion, we followed the nomenclature of Clausen (1938) and considered the two morphologies as varieties. ISSR PCR.—Multilocus DNA markers have become a useful tool for examining relationships among closely related taxa (Gillies and Abbott, 1998; Kardolus et al, 1998; Parker et al., 1998; Campbell et al., 1999; Nkongolo, 1999; Crawford, 2000; Huang and Sun, 2000; Wolfe and Randle, 2001) because they provide numerous characters derived from multiple sites within the genome (Wolfe and Liston, 1998). ISSR PCR (inter-simple sequence repeat polymerase chain reaction) is a multilocus DNA marker system that has successfully examined relationships among closely related taxa (Wolfe et al., 1998; Huang and Sun, 2000; Culley and Wolfe, 2001; Wolfe and Randle, 2001). Highly variable regions flanking microsatellites are amplified by ISSR PCR primers and minute amounts of genetic variation can be detected (Wolfe and Liston, 1998). When compared to similar techniques such as RAPD (random amplified polymorphic DNA) PCR, ISSR loci are more polymorphic (Kojima et al., 1998; Esselman et al., 1999; McGregor et al., 2000) and reproducible, presumably because of longer primer length and higher annealing temperatures (Nagaoka and Ogihara, 1997; Wolfe and Liston, 1998; Wolfe et al, 1998). In the present study, we present an investigation of taxonomic boundaries within Sceptridium dissectum by comparing inter-simple sequence repeat (ISSR) marker patterns of S. dissectum var. dissectum and var. obliquum. We chose ISSR PCR to 1) assess the genetic distinctness of S. dissectum var. dissectum and var. obliquum, 2) examine S. dissectum population genetic structure, and 3) evaluate the utility of ISSR PCR for studying Sceptridium taxa. MATERIALS AND METHODS Individual sporophytes were sampled from 17 Sceptridium dissectum populations in Ohio (Fig. 2, Table 1). Ten S. dissectum var. dissectum and 52 S. dissectum var. obliquum sporophytes were collected. Individuals were selected to represent the range of morphological variation present at each site. Five sporophytes were collected for nine populations, whereas all sporophytes (<5) were collected from eight smaller populations (Table 1). Leaf material from each individual was dried in silica gel for DNA extraction, and the remaining laminar material was pressed. Vouchers were deposited at the Willard Sherman Turrell Herbarium at Miami University (MU). Total genomic DNA was extracted from approximately 100 mg of silica gel dried leaf material using Qiagen's DNeasy Plant Mini Kit (Qiagen Inc., Valencia, CA). Genomic DNA from each sporophyte was quantitated fluorometrically using the PicoGreen dsDNA quantitation reagent (Molecular Probes, Inc., Eugene, OR) and a TD-360 mini-fluorometer (Turner Designs, Sunnyvale, CA). Quantitations were performed according to the manufacturer's protocol (Molecular Probes, Inc., Eugene, OR). Each sporophyte's DNA was quantitated twice, and the mean concentration was calculated. ISSR PCR primers were selected from the University of British Columbia Biotechnology Laboratory (UBC) primer set #9 (Vancouver, BC, Canada: http:// www.biotech.ubc.ca). Ninety ISSR primers in the UBC set were screened using DNA from two Sceptridium dissectum sporophytes (O-lc & O-ld). We selected the five primers that produced the most robust and clear amplification profiles during primer screening (Table 2). The ISSR PCR reaction mixture included one unit of Taq DNA polymerase, IX PCR buffer, 1.5 raM MgCl2, 0.2 mM of each dNTP (all PCR reagents from TaKaRa Shuzo, Co., Ltd, Shiga, Japan) and 0.3 ^M of a single ISSR primer with BARKER & HAUK: ISSR M \ \/ DISSECTUM 10 ng of DNA template in a total volume of 25 (iL. Reactions were performed in Eppendorf Mastercycler Personal thermalcyclers (Eppendorf AG, Hamburg, Germany) using the following temperature regime: 94°C for 60 seconds, then 35 cycles consisting of 45 seconds at 94°C, 45 seconds at 55°C, and 90 seconds at 72°C followed by a final 5 minute, 72°C extension. Each ISSR PCR reaction was repeated twice, with appropriate controls, to ensure consistent ISSR profiles. Using a 1 kb Plus DNA ladder size standard (Gibco-BRL, Life Technologies, Inc., Rockville, MD), PCR products were separated electrophoretically at 80 volts for seven hours on 2% agarose gels in TBE buffer with 0.2 ng/mL EtBr. Bands were visualized on a UV transilluminator and photographed using a Polaroid MP-4 Land camera (Polaroid Corporation, Cambridge, MA). ISSR bands were scored from gel photographs. The relatively high annealing temperature (55°C) helped ensure that ISSR bands were reproducible among reactions. Only clear and consistently reproducible bands were scored. Bands of indistinguishable mobility between lanes were assumed to be homologous, and to represent a single ISSR locus. For each sporophyte, each locus was scored as present or absent ("1" = locus present, "0" = locus absent). Data were compiled into a Nexus data matrix using MacClade 4.0 (Maddison and Maddison, 2000). ISSR loci data were examined with three types of analyses: 1) primer banding profiles, 2) UPGMA (Unweighted Pair Group Method using Arithmetic averages) cluster analyses, and 3) AMOVA (Analysis of MOlecular VAriance; Excoffier et al., 1992). For all analyses we assumed ISSR locus variation was representative of overall genetic \ AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 1 MU. D = S. dissection var. dissectum andO = S. dissectum var. obliqu Populatio ncode O-l 0-2 D-3 0-3 0-4 Location (Cou ntyj Sample size Voucher Licking Coshocton Franklin Franklin Perry 5 Barker #70 5 Barter" #107 0-5 D-6 0-6 0-7 0-8 0-9 O-10 O-ll 0-12 D-13 0-13 D-14 D-15 D-16 D-17 Richland Richland Morgan Hocking Barker #109 Barker #112 Barker #123 Barker #127 Athens Barker #110 Primer banding profiles were analyzed to assess the utility of ISSR PCR in Sceptridium, and to examine the relationship between the two varieties. Banding profiles generated by each primer were examined for the following parameters: 1) variety-specific markers (loci present in >25% of one variety, but in only a few individuals of the other variety; Wolfe era/., 1998], 2) percent of polymorphic loci, 3) number of loci per primer, and 4) number of unique multilocus genotypes per primer. Mean loci and mean multilocus genotypes were also calculated. UPGMA cluster analysis was used to investigate the distinctness of S. dissectum var. dissectum and var. obliquum, and to examine S. dissectum population genetic structure. A phenetic rather than parsimony-based method was used for cluster analyses because we did not verify that all co-migrating loci were homologous or that they sorted independently. Distance matrices were constructed from Dice (1945) and Jaccard (1908) similarity coefficients for UPGMA cluster analysis (Numerical Taxonomy System (NTSYSpc) ver. 2.It; Rohlf, 2000), and were based only on the shared presence of loci. The absence of an ISSR locus is not informative because any number of non-homologous mutations may result in the loss of a band. Coefficients that calculate distance from both presence and absence of loci are, generally, not appropriate for ISSR data analyses (Wolfe and Liston, 1998). Support for UPGMA clusters was calculated (WinBoot; Yap and Nelson, 1996) with 1000 bootstrap i the data (Felsenstein, 1985). BARKER &HAI Primer Sequence (5'-3') Lei ngth (bp) UBC-818 UBC-824 UBC-835 UBC-846 UBC-880 CAC ACA CAC ACA CAC AG TCT CTC TCT CTC TCT CG AGA GAG AGA GAG AGA GYC CAC ACA CAC ACA CAC ART CCA GAG GAG AGG ACA 17 15 AMOVAs were conducted as an alternative assessment of the relationship between the S. dissectum varieties, and of S. dissectum population genetic structure. Distance matrices for AMOVA were generated (Arlequin 2.001, Schneider et al., 2001) as described by Huff et al (1993). The statistical significance of AMOVA results were calculated by a non-parametric permutational analysis of a null distribution for the variance component. To assemble the null distribution of a variance component, individuals are randomly assigned to populations while the number of populations and population sizes are retained from the main analysis (Excoffier et al, 1992). The P-value calculated from the null distribution represents the probability of obtaining a larger variance component than the observed values by chance alone. In biological terms, a small P-value indicates a low probability of identifying more genetic structure than measured in the observed distribution of individuals, and a high probability of recording less genetic structure. Thus, AMOVA P-values only reflect the probability of finding more genetic structure, and do not indicate the biological significance of the observed quantities of genetic structure. In our AMOVAs, null distributions were generated with 1023 permutations of the data (Arlequin 2.001, Schneider et al., 2001). AMOVA was also used to calculate an FST value for the distribution of S. dissectum population genetic variation. For dominant marker data (e.g., ISSR or RAPD), the PST value calculated by AMOVA is a correlation of genotypes rather than individual co-dominant sites, as in isozymes. Further, identical breeding mechanisms were assumed for all S. dissectum populations. Thus, an FST value calculated from dominant marker data may not be directly comparable to PST values generated from co-dominant marker data. Five ISSR primers produced 69 loci (mean = 13.8/primer) with 94% of the loci polymorphic (Table 3). Primer UBC-818 produced the most loci (16), whereas primer UBC-846 produced the fewest (12). The mean number of unique multilocus genotypes distinguished per primer was 38.2 (Table 4). No variety-specific markers were identified. For each primer surveyed, some individuals of Sceptridium dissectum var. dissectum possessed banding profiles identical to those of some var. obliquum individuals. All individuals were distinguished as unique multilocus genotypes using a combination of any AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 1 TABLE 3. Sample size, total number of loc primers from the ISSR survey of S. dissHctun Taxon S. dissectum var dissectum S. dissectum var. obliquum Total Sample si: i'v^ddS SiSE Total # loci 10 62 69 ar. obliquum. % Polymorphic loci s three primers, and using all five primers, the genotypic diversity (# genotypes/ # individuals) for each taxon was 1.0. Our investigation of S. dissectum population genetic structure revealed that individuals from the same S. dissectum population did not cluster closely in UPGMA analyses (Figs. 3 & 4). Most clusters consisted of individuals from different, and sometimes, distant populations. For example, 0-9a, collected in Logan County in west central Ohio, clustered with 0-6d, a specimen from Richland County in north central Ohio, approximately 120 miles away. This pattern was repeated for other individuals from geographically distant sites, but for the clusters supported by bootstrap values >50%, all individuals were from the same population (Figs. 3 & 4). Seven clusters consisted of two individuals from the same population, and, of these, only O-lc + O-ld + O-le and 0-12a + 0-12b were supported by bootstrap values >50% (Figs. 3 & 4). In neighbor joining (NJ) or maximum parsimony (MP) analyses (not presented), populations did not form discrete clusters. The NJ and MP tree topologies were essentially identical to the UPGMA topologies (Figs. 3 & 4), and had equivalent levels of bootstrap support. AMOVA also revealed little genetic structure among S. dissectum populations. Of the total genetic variation detected, among-population genetic variation was 8.49%, whereas within-population was 91.51%. A low level of genetic structure for the S. dissectum populations was also indicated by the FST value (0.085, P < 0.0001, Table 5). The highly significant P-value suggests that our observed distribution of individuals in populations produces nearly the largest amount of genetic structure possible in our data set. In a comparison of the two S. dissectum varieties, neither var. dissectum nor var. obliquum formed discrete clusters in UPGMA analyses (Figs. 3 & 4). Individuals of var. dissectum frequently clustered more closely with members of var. obliquum than with their own taxonomic group. Bootstrap support for all but three clusters in the UPGMA trees (Figs. 3 & 4) was poor. The two varieties failed to form discrete clusters in NJ and MP analyses (not presented) which had almost identical tree topologies and similar levels of bootstrap support. AMOVA revealed little genetic difference between S. dissectum var. dissectum and var. obliquum. The two varieties were only 3.38% genetically different, while they were 96.62% genetically similar (Table 6). The amount of genetic difference identified between the two varieties was close to the largest amount possible in our data set {P = 0.0140, Table 6). BARKER & HA: M DISSECTUM iber of loci and genotypes distinguished for e i were indistinguishable when examined wi1 combined result for both S. dissectum varie UBC-824 UBC-835 UBC-846 DISCUSSION Marker systems used to examine relationships among species or subspecific taxa should provide highly variable loci, and the system should be able to distinguish as many individuals of a single species as possible, in concordance with the organism's breeding system (Avise, 1994). In the present study, ISSR markers distinguished all S. dissectum individuals by any combination of three primers, a result similar to that observed in other studies (Wolfe et al., 1998; Esselman et al., 1999). Of the ISSR loci distinguished in our S. dissectum taxa, 82% (var. dissectum), 92% (var. obliquum) and 94% (species total) were polymorphic (Table 3), values well within the range of ISSR variability when ISSR markers have successfully discriminated taxa at the species level and lower (Wolfe et al, 1998; Culley and Wolfe, 2001; Wolfe and Randle, 2001). For example, Wolfe et al. (1998) demonstrated patterns of diploid hybrid speciation in Penstemon using ISSR markers, and reported percent polymorphic loci values of 72-95% for the seven taxa sampled. Wolfe and Randle (2001) used ISSR markers to examine taxonomic boundaries and relationships in Hyobanche and found that 64-96% of their ISSR loci were polymorphic in four taxa. ISSR markers discriminated between two varieties of Viola pubescens with 100% of ISSR loci polymorphic for the species (Culley and Wolfe, 2001). As values for ISSR variability in S. dissectum were within the range reported from studies that have successfully used ISSR markers to examine taxonomic boundaries and relationships, ISSR markers appear to be an appropriate tool for examining taxonomic boundaries among Sceptridium subspecific taxa and possibly species. DISTRIBUTION OF GENETIC VARIATION.—As assessed by ISSR genotypes, the distribution of genetic variation in S. dissectum was consistent with results from studies of other pteridophyte species, where most genetic variation was distributed within populations (Haufler and Soltis, 1984; Holsinger, 1987; Kirkpatrick et al., 1990; Soltis and Soltis, 1987; Soltis et al., 1988; Soltis and Soltis, 1988; Watano and Sahashi, 1992). Using ISSR PCR, Camacho and Liston (2001) found most genetic diversity partitioned within populations of Botrycbium pumicola, and little among population genetic differentiation. Within Sceptridium, Watano and Sahashi (1992) reported that 81% of isozyme FIG. 3. UPGMA cluster analysis generated from five ISSR primers using a x generated v Bootstrap values >50% are reported above branches. The scale bel coefficient of similarity represented by corresponding branch ] population codes in Table 1, asterisks indica (i.e., a, b, c, d, e) distinguish individuals of the same popula BARKER & H,\: [ FIG. 4. UPGMA cluster a generated from five ISSF algorithm. Bootstrap valui M DISSECTUM f 62 Sceptridium dissection sporophyte s using a distant matrix generated vv are reported above branches. The scali sented by corresponding 1 , asterisks indicate var. . . . , . . ,. , . 5. AMOVA statistics for S. dissectu m population genetic strut ;ture based up on ISSR ma rker profiles. P is the probability of obtaining a larger variance value. TABLE Variance component Among-populations Variance % 1 1 1.240 Fst P 0.085 <0.0001 91.51 allelic genetic diversity was distributed within and 19% was distributed among S. ternatum populations. Based upon similar isozyme data, McCauley et al. (1985) estimated an FST value of 0.090 for S. dissectum var. obliquum populations, similar to our ISSR-based value of 0.085. Sceptridium dissectum genetic variation, as measured by ISSR genotype distribution, is within the range for isozyme allelic distribution reported by Soltis and Soltis (1990) for both outcrossing and inbreeding fern species. Thus, our ISSR data are consistent with previous evidence from ISSR and isozyme studies concerning the distribution of genetic variation. UPGMA cluster analyses revealed most S. dissectum individuals [i.e., genotypes) did not group by population, but, rather, individuals from disparate populations often grouped together. Camacho and Liston (2001) found that cluster analysis of ISSR data did not segregate by population individuals of the related and presumably inbreeding Botrychium pumicola. Other, similar ISSR studies have detected population genetic structure that was evident in cluster analyses (Wolfe et al., 1998; Culley and Wolfe, 2001). Among five populations of Viola pubescens var. scabriuscula Schwein., populations were clearly defined by cluster analysis (Culley and Wolfe, 2001). We generated a similar number of scorable loci as reported by Culley and Wolfe (2001). Thus, if S. dissectum populations are truly differentiated genetically, the amount of ISSR data generated in our study should have revealed it, especially because a high proportion of individuals at each site was collected. Initially, the inability of UPGMA to cluster S. dissectum ISSR genotypes by population may appear contrary to published isozyme studies. Watano and Sahashi (1992) reported only two isozyme genotypes were shared among three populations of S. ternatum, indicating the three populations were markedly dissimilar in genotypes and suggesting individuals within populations should be of similar genotype. However, when Watano and Sahashi (1992) measured allelic diversity, 81% of the genetic variation was distributed within populations. Thus, allelic diversity and genotype distribution in S. ternatum did not produce similar estimates of the distribution of genetic variation. The apparent discrepancy between allelic diversity and genotype distribution in S. ternatum may be a consequence of founder effects, selection, genetic drift (Watano and Sahashi, 1992), or the total amount of variation detected. If more isozyme genotypes had been detected, then estimates of allelic diversity and genotype distribution may have been more similar. Watano and Sahashi (1992) identified only 30 genotypes from 138 S. ternatum individuals, whereas ISSR markers identified 62 genotypes from 62 S. dissectum individuals. Based on ISSR data, populations of congener S. dissectum do not frequently consist of BARKER & HAUK: ISSR MARKERS IN SCEPTRIDIUM DISSECTUM TABLE 6. AMOVA statistics from a comparison of the ISSR profiles of, Variance component individuals of identical or similar genotypes, although this does not exclude the possibility that the populations may contain similar isozyme genotypes, as observed by Watano and Sahashi (1992) in S. ternatum. Further, ISSR ; of genotype distribution in S. dissectum are more similar to overall i of fern isozyme allelic diversity (Soltis and Soltis, 1990) than to the distribution of isozyme genotypes as reported by Watano and Sahashi (1992). Other populations of inbreeding pteridophytes studied previously with isozymes should be surveyed with ISSR PCR to determine if genotype distributions similar to ours can be documented. The rather low partitioning of genetic variation within and among fern populations has been explained by high rates of spore dispersal, rapid colonization of a region with little subsequent genetic differentiation, or both (Soltis and Soltis, 1988; Soltis and Soltis, 1990). In S. dissectum either scenario is possible, and probably a combination of both has contributed to the current distribution of genetic variation. During the latest glaciation event (Wisconsian), ending approximately 15,000 y.a. (Smith and Smith, 2001), S. dissectum may have been restricted to the Southern Appalachians and the Gulf Coast. Recolonization of deglaciated areas with insufficient time for subsequent genetic differentiation of populations may have contributed to the observed distribution of genetic variation. Alternatively, spore dispersal in S. dissectum may be high enough to effectively link the sampled populations to form a large metapopulation, which may account for the present distribution of genotypes. Based on ISSR data, it is impossible to exclude either rapid colonization or spore dispersal as the primary cause of the observed distribution of genetic variation in S. dissectum. Dominant marker systems, such as ISSR PCR, are inappropriate for estimating self-fertilization rates, and inferring an organism's breeding system. Although ISSR markers may provide enough resolution to distinguish many or all individuals in a population, the technique does not provide a measure of true heterozygosity, a requirement for estimating self-fertilization rates (Wolfe and Liston, 1998). As such, we were unable to determine the breeding system of the S. dissectum populations sampled. However, the isozyme studies of McCauley et al. (1985) and Watano and Sahashi (1992) demonstrated that high rates of self-fertilization characterize the breeding system of the Sceptridium species studied. Our results cannot support or refute the results of these isozyme studies. ISSR DATA.—Analyses of ISSR marker data demonstrated no ISSR loci specific to either Sceptridium dissectum var. TAXONOMIC IMPLICATIONS OF 14 AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 1 (2003) dissectum or S. dissectum var. obliquum. If the two taxa were genetically distinct, each taxon should have unique ISSR loci. Moreover, individuals from each variety should form discrete clusters in UPGMA, and this did not occur (Figs. 3 & 4). The lack of genetic distinction between the two varieties is illustrated by relationships between individuals D-6a and 0-3e. Both individuals clustered more closely to each other than to any other individual in the data set, but, morphologically, D-6a represents var. dissectum and 0-3e represents var. obliquum. The AMOVA comparison of genetic variation between the S. dissectum varieties also demonstrated that the taxa were genetically indistinguishable, sharing 96.62% of their ISSR genetic variation. These ISSR results support Tryon's (1936) observations of a few var. dissectum individuals producing var. obliquum fronds (and vice versa) in subsequent seasons. If S. dissectum trophophore morphology truly exhibits such seasonal plasticity, then the clustering of var. dissectum individuals with var. obliquum individuals would be expected, and both ISSR data and Tryon's (1936) observations indicate no genetic distinctness between the two varieties. Although the UPGMA analyses did not separate the two S. dissectum varieties, most groups were poorly supported by bootstrap analysis. The lack of bootstrap support for most groups generated in the UPGMA cluster analyses may be a result of primer to primer variation, e.g., 0-8b clustered with D-13e using primer 835, but with 0-5e using primer 824. The lack of consistent relationships among ISSR primers in S. dissectum may be due to the scoring of non-homologous ISSR loci. Available evidence from other studies (S. Datwyler, Ohio State Univ., pers. comm.) suggests that this is an unlikely source of the inconsistencies observed. Shannon Datwyler (pers. comm.) examined ISSR loci from 30 different Scrophulariaceae species and established estimates of ISSR locus homology. For high frequency bands (present in >6 individuals) in Penstemon, she reported 83% of the bands scored as homologous were homologous as determined by Southern hybridization. In Scropbularia and Hyobanche, 93% and 100% of co-migrating ISSR bands, respectively, were homologous (Datwyler, pers. comm.). In our data set, 97% of scored loci were considered high frequency by Datwyler's criterion, but we have not verified the homology of scored bands. Additionally, in Helianthus and Brassica, Adams and Rieseberg (1998) found that even when 20% of the bands in a RAPD PCR data set were non-homologous, there was negligible effect on species relationships as generated by principal-coordinate-analysis ordination. If these findings in Brassica and Helianthus can be extrapolated to ISSR cluster analyses in Sceptridium, then even a substantial number of nonhomologous bands may have no significant impact on relationships among individuals. Another possible cause for the low UPGMA bootstrap support and primer to primer variation was the nature of ISSR loci variation. In the data matrix containing the calculated genetic similarity values for the S. dissectum ISSR results (not presented), some individuals were equally similar to other individuals, although their banding patterns were all unique. Because UPGMA clusters individuals by seeking combinations of the least different similarity BARKER & HAUK: ISSR MARKERS IN SCEPTRIDIUM values (Avise, 1994), the few sets of identical genetic similarity values may have caused the UPGMA algorithm to make arbitrary decisions between individuals when clustering (Takezaki, 1998), resulting in the production of tie trees during bootstrap analysis. The production of tie trees can lower bootstrap support for UPGMA clusters (Takezaki, 1998), and this may explain the low bootstrap support in the ISSR UPGMA analyses. After close examination, the low bootstrap values for the UPGMA analyses do not discredit the interpretation that the two varieties are not different, but further support this conclusion as genetic similarity values between individuals of the two varieties were frequently equivocal with similarity values between individuals of the same variety. Wagner (1960a) argued that if two putative taxa co-exist over large areas with intergradation in morphological characters between the taxa, then the two entities should not be recognized. Our ISSR data provide evidence that no underlying genetic differentiation correlates with the morphologies of S. dissectum var. dissectum and var. obliquum, and recognition of varieties with formal taxonomic status in S. dissectum is not supported. Because taxonomic designations based upon morphology often imply genetic distinctness (Paris et ah, 1989), formal recognition of var. dissectum and var. obliquum may perpetuate this assumption. Based on the available ISSR evidence and Wagner's (1960a) criteria for varieties, var. dissectum and var. obliquum should not be recognized as formal taxonomic units. More recent classifications that do not formally recognize infraspecific variation in S. dissectum [e.g., Lellinger, 1985; Wagner and Wagner, 1993) reflect more clearly the genetic evidence at hand than do earlier classification systems (e.g., Clausen, 1938). Morphology alone apparently does not accurately depict genetic relatedness among individuals of the highly variable S. dissectum. A logical extension of these data calls into question species level taxonomy in Sceptridium that is based solely on variable morphological characters. ISSR markers have proven useful for examining species level distinctions in angiosperms (Wolfe et ai, 1998; Wolfe and Randle, 2001) and may be useful for examining relationships among Sceptridium species. For example, preliminary ISSR data suggest that S. oneidense (Gilb.) Lyon, a taxon previously included as a variety of S. dissectum, is not genetically distinct from S. dissectum, whereas Botrypus [=Botrychium] virginianus (L.) Michx. is distinct at many loci (Barker and Hauk, unpubl. data). If ISSR markers reveal other Sceptridium species closely related to S. dissectum as genetically indistinguishable, then a critical reexamination of species concepts in Sceptridium is warranted. The large range of morphological variation in Sceptridium species may be the consequence of two different phenomena. First, Sceptridium species possess some of the highest reported self-fertilization rates among vascular plants (McCauley et al, 1985; Watano and Sahashi, 1992). The high selffertilization rate may be a source of morphological variation among selffertilizing lineages through within-lineage fixation of genes controlling laminar characters. For example, Schneller and Holderegger (1997) reported 16 AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 1 (2003) inbred progeny of Athyrium filix-femina (L.) Roth demonstrated "considerable morphological variation" over outcrossed progeny. Another possible explanation for morphological variation in Sceptridium may be phenotypic plasticity affected by various environmental conditions. This may explain the differences observed by Tryon (1936) between S. dissectum var. dissectum and var. obliquum (i.e., individuals producing blades with either morphology in different years), and ISSR markers did not reveal any genetic differences between the two varieties. Combined with other data sources (isozymes, RFLPs, DNA sequences, etc.) ISSR markers should be useful for examining critically species delimitations in Sceptridium, and may contribute to a better understanding of morphological variation in the genus. CONCLUSIONS ISSR markers proved useful for examining infraspecific genetic in S. dissectum by distinguishing all individuals and producing levels of polymorphic loci within the range reported by similar ISSR studies. The low level of population genetic structure detected by ISSR markers in S. dissectum populations was consistent with previous isozyme studies of S. dissectum and other fern species. Morphologies traditionally identified as var. dissectum and var. obliquum did not correlate with ISSR marker variation, and our data do not support the recognition of these as infraspecific taxa. Species boundaries in Sceptridium should be critically examined because morphological distinctions among the species are not always clear. i University Research Foundation, the Anderson Summer Scholi of Denison University, and the Office of the Associate Provost of Denison University f support. We thank Shannon Datwyler for providing unpublished data and constructive the manuscript. We are grateful to George Yatskievych, Randy Small, and an anonymoi for providing helpful advice for improvement of the manuscript. LITERATURE CITED R. P. and L. H. RIESEBERG. 1998. The effects of non-homology in RAPD bands on similarity and multivariate statistical ordination in Brassica and Helianthus. Theor. Appl. Gen. 97:323-326. AVISE, J. C. 1994. Molecular Markers, Natural History and Evolution. Chapman & Hall, New York. CAMACHO, F. J. and A. LISTON. 2001. Population structure and genetic diversity of Botrychium pumicola (Ophioglossaceae) based on inter-simple sequence repeats (ISSR). Amer. J. Bot. 88:1065-1070. CAMPBELL, C. S., L. A. ALICE and W. A. WRIGHT. 1999. Comparisons of within-population genetic variation in sexual and agamospermous Amelanchier (Rosaceae) using RAPD markers. PI. ADAMS, CLAUSEN, R. T. 1938. A CRAWFORD, D. J. 2000. 49:479-501. monograph of the Ophioglossaceae. Mem. Torrev Bot. Club 19:1-177. . CRAWFORD, J. L. WINDUS and A. D. Wc ieri ssp. insperata (Poaceae): compara mrphic DNA (RAPD) I ( i . HAUFLER, C. H. and D. E. SOLTIS. 1984. Obligate W. D. 1996. Phylogenetics of Ophioglossaceae: the v\ ululioniiiy consequences of morphological simplification. Ph. D. Dissertation, University of North Carolina at Chapel Hill, Chapel Hill, NC. HOLSINGER. K. E. 1987. Gametophytic self-fertilization in homosporous plants: development, evaluation, and application of a statistical method for evaluating its importance. Amer. J. Bot. HAUK, J. C. and M. SUN. 2000. Genetic diversity and relationships of sweetpotato and its wild relatives in Ipomoea series Batatas (Convolvulaceae) as revealed by inter-simple sequence repeat (ISSR) and restriction analysis of chloroplast DNA. Theor. Appl. Gen. 100:1050-1060. D. R., R. PEAKALL and P. E. SMOUSE. 1993. RAPD variation within and among populations of outcrossing buffalograss [Buchloe dactyloides (Nutt.) Engelm.). Theor. Appl. Gen. 86:927-934. JACCARD, P. 1908. Nouvelles recherches sur la distributuon florale. Bull. Soc. Vaud. Sci. Nat. 44: HUANG, HUFF, J. P., H. J. van ECK and R. G. van den BERG. 1998. The potential of AFLPs inbiosj stematic s a first application in Solarium taxonomy (Solanaceae). Pi. Syst. Evol. 210:87-103. M. 1987. A phylogenetic classification of Ophioglossaceae. Card. Bull. Sing. 40:1-14. D. 1994. Demography and conservation of Botrychium australe, a peculiar, sparse mycorrhizal fern. New Zealand J. Bot. 32:393-100. KIRKPATRICK, R. E. B., P. S. SOLTIS and D. E. SOLTIS. 1990. Mating system and distribution of genetic variation in Gymnocarpium dryopteris ssp. disjunction. Amer. }. Bot. 77:1101-1110. KOJIMA, T., T. NAGAOKA, K. NODA and Y. OGIHARA. 1998. Genetic linkage map of ISSR and RAPD markers in Einkorn wheat in relation to that of RFLP markers. Theor. Appl. Gen. 96:37-45. LELLINGER, D. B. 1985. A Field Manual of the Ferns & Fern-Allies of the United States and Canada. Smithsonian Institution Press, Washington, D.C. LYON, H. L. 1905. A new genus of Ophioglossaceae. Bot. Gaz. 40:455-458. MADDISON, D. R. and W. P. MADDISON. 2000. MacClade 4: Analysis of Phylogeny and Character Evolution, Version 4. Sinauer Associates, Sunderland, MA. MGCAULEY. D. E., D. P. WHITTIER and L. M. REILLY. 1985. Inbreeding and the rate of self-fertilization in a grape fern, Botrychium dissectum. Amer. J. Bot. 72:1978-1981. MCGREGOR, C. E., C A. LAMBERT, M. M. GREYLING, J. H. Louw and L. WARNICH. 2000. A comparative assessment of DNA fingerprinting techniques (RAPD, ISSR, AFLP and SSR) in tetraploid potato [Solanum tuberosum L.) germplasm. Euphytica 113:135-144. MELAN, M. A. and D. P. WHITTIER. 1989. Characterization of mucilage on the proximal cells of young gametophytes of Botrychium dissectum (Ophioglossaceae). Amer. J. Bot. 76:1006-1014. . 1990. Survivorship and predation changes in five populations of Botrychium l Eastern Pennsylvania. Amer. Fern J. 80:173-182. KARDOLUS, KATO, KELLY, 18 AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 1 (2003) T. and Y. OGIHARA. 1997. Applicability of inter-simple sequence repeat polymorphisms in wheat for use as DNA markers in comparison to RFLP and RAPD markers. Theor. Appl. NAGAOKA, K. K. 1999. RAPD variations among pure and hybrid populations of Picea mariana, P. rubens, and P. glauca (Pinaceae), and cytogenetic stability of Picea hybrids: identification of species-specific RAPD maik<-is l'l SW Fw.I 215:229-239. C. A., F. S. WAGNER and W. H. WAGNER. 1989. Cryptic species, species delimitations, and NKONGOLO, PARIS, P. C, A. A. SNOW, M. D. SCHLIG, G. C. BOOTON and P. A. FUERST. 1998. What molecules can tell us about populations: choosing and using a molecular marker. Ecology 79:361-382. K. 1884. Beitrage zur systematik der Ophioglosseen. Ber. deutsch. bot. Ges. 1:348-353. F. J. 2000. NTSYSpc. Numerical Taxonomy System, version 2.It. Exeter Software, Setauket, PARKER, PRANTL, ROHLF, es on Ophioglossales in Japan and the S.. D. ROESSLI and L. EXCOFFIER. 2001. Ariequin: A software for population genetics data analysis, Ver 2.001. Genetics and Biometry Lab, Dept. of Anthropology, University of Geneva. SCHNELLER, J. J. and R. HOLDEREGGER. 1997. Vigor and survival of inbred and outbred progeny of SCHNEIDER, R. L. and T. M. SMITH. 2001. Ecology & Field Biol* gs, New York. D. E. and P. S. SOLTIS. 1987. Breeding system of the fern Dryopteris expansa: evidence for mixed-mating. Amer. J. Bot. 74:504-509. P. S. and D. E. SOLTIS. 1988. Genetic variation and population structure in the fern Blechnum spicant (Blechnaceae) from western North America. Amer. J. Bot. 75:37-44. and . 1990. Genetic variation within and among populations of ferns. Amer. Fern J. 80:161-172. , and K. E. HOLSINGER. 1988. Estimates of intragametophytic selfing and interpopulational gene flow in homosporous ferns. Amer. J. Bot. 75:1765-1770. SPRENGEL. 1804. Anleit. Kenntn. Gew. 3:172. TAKKZAKI. N. 1998. Tie trees generated by distance methods of phylogenetic reconstruction. Mol. Biol. Evol. 15:727-737. TRYON, R. M. 1936. Botrychium dissectum and forma obliquum. Amer. Fern J. 26:26-30. and A. F. TRYON. 1982. Ferns and Allied Plants with Special Reference to Tropical America. Springer-Verlag, New York. UNDERWOOD, L. M. 1898. American ferns-I: The ternate species of Botrychium. Bull. Torrey Bot. Club 25:521-541. WAGNER, W. H. 1960a. Evergreen grapeferns and the meanings of infraspecific categories as used in North American pteridophytes. Amer. Fern J. 50:32-45. . 1960b. Pel tationin Botrychium subg. Sceptridium in the Northeastern United States. Bull. Torrey Bot. Club 87:303-325. SMITH, SOLTIS, SOLTIS, dissectum. Rhodora 63:164-175. . 1990. Ophioglossaceae, Pp. 193-197 in The Families and Genera of Vascular Plants vol. V. Pteridophytes and Gymnosperms. K. U. Kramer and P. S. Green (eds.). Springer-Verlag, and F. S. WAGNER. 1983. Genus communities as a systematic tool in the study of new world Botrychium (Ophioglossaceae). Taxon 32:51-63. and . 1993. Ophioglossaceae C. Agardh: Adder's-tongue Family, pp. 85-106 in Flora of North America North of Mexico vol. 2: Pteridophytes and Gymnosperms. Flora of North America Editorial Committee. Oxford University Press, New York. and J. M. BEITEL. 1985. Evidence for interspecific hybridisation in pteridophytes with subterranean mycoparasitic gametophytes. Proc. R. Soc. Edin. 86(B):273-281. WATANO, Y. and N. SAHASHI. 1992. Predominant inbreeding and its genetic consequences in a homosporous fern genus, Sceptridium (Ophioglossaceae). Syst. Bot. 17:486-502. BARKER & HAUK: ISSR MARKERS IN SCEPTRIDIUM DISSECTUM WOLFE, A. D. and A. LISTON. 19 1998. Contributions of PCR-based methods to plant systematics and evolutionary biology, Pp. 43-86 in Molecular Systematics of Plants II: DNA Sequencing. D. E. Soltis, P. S. Soltis and J. J. Doyle (eds.). Kluwer Academic Publishers, Boston. , Q.-Y. XIANG and S. R. KEPHART. 1998. Assessing hybridization in natural populations of Penstemon (Scrophulariaceae) using hypervariable intersimple sequence repeat markers. Mol. Ecol. 7:1107-1125. V. and R. J. NELSON. 1996. WinBoot: A Program for Performing Bootstrap Analysis of Binary Data to Determine the Confidence Limits of UPGMA-Based Dendograms. IRRI Discussion The Gametophyte of Diphasiastrum sitchense The spores of Diph lit- dark on a nutrient medium inorganic nutrients and glucose. Dark-grown prothalli develop into white, carrotetophytes with a tapering base, constricted neck, and gametangial cap. The antheridia d sunken, and the archegonia have long necks with numerous neck canal cells. The e has a zone of radially elongated cells that is comparable to the inner mycorrhizal zone strum gametophytes from nature. Although possessing few derived sporophytic 1. sitchense has a typical carrot-shaped, Diphasiastrum gametophyte. The sporophyte of Diphasiastrum sitchense (Rupr.) Holub is considered to be the most basal member of this genus in North America (Lloyd, 1901; MarieVictorin, 1925; Wilce, 1965; Tryon and Moran, 1997). The main reason for this I 1 s the type of leaf and their arrangement on the stem. The leaves of D. sitchense are isomorphic and spirally arranged on terete branchlets compared to the di- or trimorphic leaves and decussate arrangements on flattened branchlets of the remaining North American members of the genus (Wilce, 1965). In addition, an analysis of many characters has shown that D. sitchense has next to the fewest number of derived characters for the genus worldwide (Wilce, 1965). The known gametophytes of Diphasiastrum are from species having sporophytes with many derived characters. The gametophytes of these species are subterranean, mycorrhizal, and carrot-shaped (Bruchmann, 1908; Bruce, 1979; Whittier, 1981). Because gametophytes from the basal members of the genus are unknown, it would be of interest to determine if the gametophyte morphology of D. sitchense is different from those of the species with derived sporophytic characters. This study was carried out to determine the type of gametophyte in D. sitchense using the techniques of axenic culture. It has been over 150 years since this taxon was recognized, however, no gametophytes have been collected from natural areas. For this reason, growing these gametophytes in culture provided an opportunity to determine the structure of this gametophyte. MATERIALS AND METHODS Spores of Diphasiastrum sitchense were obtained from strobili collected during September in King County, Washington and Lane County, Oregon. Vouchers of the King Co. plants are on deposit at VDB and those of the Lane Co. plants [D. H. Wagner #m0732) are on deposit at OSC. The spores were surface sterilized with 20% Clorox (1.1% sodium hypochlorite), following the techniques of Whittier (1973) and were sown on WHITTIER: GAMETOPHYTE OF DIPHASIASTRUM 15 ml of nutrient medium in 20 X 125 mm culture tubes with screw caps that were tightened to reduce moisture loss. The sown spores were maintained in darkness or under a 14 hour photoperiod (50 umol • m 2 • sec l) from Gro-lux fluorescent lamps at 21 ± 1°C. The nutrient medium contained 100 mg NH4C1, 50 mg MgS04 • 7H20, 20 mg CaCl2, and 50 mg K2HP04 as a final concentration per liter. A liter of the medium was completed with 0.25 ml of a minor element solution (Whittier and Steeves, I960), 4 ml of an FeEDTA solution (Sheat et al, 1959), and 5 g of glucose. The medium was solidified with 1% agar and was at pH 5.2 after autoclaving. The gametophytes were fixed with Randolph's Modified Navashin Fluid (CRAF; Johansen, 1940). After fixation, the gametophytes were embedded in paraffin and sectioned by conventional techniques (Johansen, 1940). The sections were stained with HeidenhahVs hematoxylin, safranin O, and fast green. After 5 months in the dark about 0.5% of the spores germinated (Fig. 1). Germination never exceeded 1% with more time in the dark. No spores had germinated in illuminated cultures after 11 months. Young multicellular gametophytes were found after 8 months. These small, globular gametophytes usually had spore coats attached (Fig. 2). At nine months, larger globular gametophytes were transferred to fresh nutrient medium for further growth. Mature gametophytes were obtained 9 months after this transfer. The oldest gametophytes studied were collected 2 years after sowing the spores. Mature gametophytes were white and carrot-shaped (Figs. 3, 4, 5) and the largest found were about 8 mm long. The upper and basal regions of the gametophytes were separated by constricted necks. This constriction (Figs. 3, 4, 5, 6) is the site of the meristematic region (ring meristem) in gametophytes of Diphasiastrum. The more or less conical basal region was covered with numerous rhizoids. The upper region, the gametangial cap, was the site for antheridia and archegonia. The gametangial caps on young gametophytes produced antheridia first, followed by the formation of archegonia. On mature gametophytes, antheridia were in the middle of the gametangial cap surrounded by archegonia. The archegonia were prominent when present. They had long necks usually with 9-12 neck canal cells (Figs. 6, 7). Neck length, from base of egg cell to tip of neck, averaged 274 urn. The antheridia were large and sunken (Fig. 8). The elongated sperm masses averaged 233 um long and 118 um wide. Large numbers of male gametes were formed by each antheridium. Besides being the site for rhizoid formation, the tapering basal regions of Diphasiastrum gametophytes from nature house a mycorrhizal fungus. In axenic culture the gametophytes grow without the mycorrhizal fungus if sugar is available in the nutrient medium. However, the basal regions of the gametophytes from axenic culture did develop some anatomical features found in 22 AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 1 (2003) gametophytes of other species from nature. Sections show elongated cells close to the basal surface of these gametophytes (Fig. 9). These cells are in essentially the same position as the elongated cells of the inner mycorrhizal region of gametophytes of other Diphasiastrum species. Thus, aspects of a mycorrhizal region differentiated in these gametophytes in the absence of a fungus. The gametophytes described for Diphasiastrum are Type II (Bruchmann, 1898). The Type II gametophytes are carrot-shaped with an upper area, the gametangial cap, separated from the tapered basal region by a constricted neck with a ring meristem. The gametangia of these gametophytes are larger than those found in most of the other gametophyte types described for Lycopodium sensu lato (Bruchmann, 1898). The antheridia are massive and sunken into the gametangial cap (Bruce, 1979). The long-necked archegonia have the largest number of neck canal cells reported for any of Bruchmann's gametophyte types. The gametophyte of D. sitchense fits the description for the gametophytes (Type II) of this genus (Bruchmann, 1908; Bruce, 1979; Whittier, 1981; Whittier and Britton 1995). There is nothing unusual about the gametophyte of D. sitchense. It is carrot-shaped with all the described regions present. The antheridia are large, sunken structures in the gametangial cap and are similar in size to those described for D. digitatum (A. Braun) Holub and D. Xhabereri (House) Holub from axenic culture (Whittier, 1981; Whittier and Britton, 1995). The archegonia have long necks with large numbers of neck canal cells and they are similar in length to the archegonia of D. digitatum from soil and axenic culture (Bruce, 1979; Whittier, 1981). The basal region of Diphasiastrum gametophytes from soil have a distinctive three layered mycorrhizal region (Bruce, 1979; Whittier, 1981). The development of a three layered mycorrhizal region did not occur in the gametophytes lacking an endophytic fungus. However, elongated cells form in the basal region of these gametophytes and they are in the correct position for the elongated cells found in gametophytes of D. digitatum (Bruce, 1979) and D. complanatum (L.) Holub (Bruchmann, 1898) from soil. Also, these elongated cells are in the same position as elongated cells in gametophytes of D. digitatum from axenic culture (Whittier, 1981). The tissues of the basal region of D. sitchense are very similar to those in other gametophytes of the genus. ig. 1. Germinating spore, 320X. Fig. 2. t-shaped gametophytes with gametangial ricted necks (white arrowheads), and conical bases bearing rhizoids. Fig. 3. Ga with small gametangial cap bearing mainly antheridia. Arrow indicates two archegonial necks, 7X. Fig. 4. Gametophyte with archegonia. 6X. Fig. 5. Gametophyte with archegonia, 10X. Fig. 6. Constricted neck of gametophyte with young short-necked archegonia (arrowheads) at lower edge of gametangial cap and mature long-necked archegonia (arrows) on gametangial cap, 16X. Fig. 7. Longitudinal section of archegonium, 130X. Fig. 8. Longitudinal section of antheridia, 130X. Fig. 9. Longitudinal section of conical base with elongated cells (arrows), 130X. 24 AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 1 (2003) The sporophytes of D. sitchense and D. veitchii (Christ) Holub, an Asian species, are different from other species of Diphasiastrum. They have almost all basal characteristics for the genus (Wilce, 1965). However, the gametophyte of D. sitchense is normal and typical for the genus. Bruce (1979) had raised the possibility that gametophytes of D. sitchense and D. veitchii might be informative in bridging the structural differences between Type I and Type II gametophytes. This is not the case with the gametophyte of D. sitchense. Gametophytes of other species will have to be examined to determine if an intermediate condition can be found. This study was supported in part by the Vanderbill I fnivers LITERATURE CITED H. 1898. Uber die Prothallien unc Gotha. . 1908. Das Prothallium von Lycopodium complanatum L. Bot. Zeitung, 2. Abt. 66:169-18 JOHANSEN, D. A. 1940. Plant Microtechnique. McGraw-Hill, New York. LLOYD, F. E. 1901. The genus Lycopodium: a criticism. Torreya 1:56-59. MARIE-VICTORIN, FR. 1925. Les Lycopodinees du Quebec. Contrib. Inst. Bot. Univ. Montreal 3:1-11 SHKAT, D. E. G., B. H. FLETCHER and H. E. STREET. 1959. Studies on the growth of excised roots. VII BRUCHMANN. Phytologist 58:128-154. A. F. and R. C. MORAN. 1997. The ferns and allied plants of New England. Massachuset Audubon Society, Lincoln, Mass. D. P. 1973. The effect of light and other factors on spore germination in Botrychiu. TRYON, WHITHER, . 1981. Gametophytes of Lycopodiun i i >mplanatum var. flabelliform as grown in axenic culture. Bot. Gaz. (Crawfordsville) 142:519-524. and D. M. BRITTON. 1995. Gametophytes of Diphasiastrum Xhabereri. Amer. Fern J. 8 and T. A. WILCE, STEEVES. 1960. The induction of apogamy in the bracken fern. Can. J. Bot. 3 J. H. 1965. Section Complanata of the genus Lycopodium. Beih. Nova Hedwigia 19:1-23 Contribution to the Gametophyte Morphology of the Fern Genus Lomagramma J. Sm. in India* SUBHASH CHANDRA and MRITTUNJAI SRIVASTAVA Pteridology Laboratory, National Botanical Research Institute, Lucknow - 226001 (INDIA) RUCHI SRIVASTAVA Botany Department, Lucknow University, Lucknow - 226001 (INDIA) most of the other genera of the lomariopsidoid ferns. The Drynaria-type of gametophyte development is more charac tenstir of the lomariopsidoid ferns. Older prothalli are cordate and naked throughout. Early development seems to be somewhat plastic and perhaps of limited usefulness as a character for systematic purposes. Lomagramma I. Smith is a genus of about 15 species ranging from northeastern India to Tahiti and into tropical America. It is represented by a lone species, Lomagramma sorbifolia (Willd.) Ching, in India, where it is known to occur in Garo Hills and Lakhimpur in the state of Assam (Chandra, 2000). The plants are scandent, large, terrestrial, and shade-loving growing mostly near streams in dense tropical forests. The species is very similar to Lomariopsis Fee and Stenochlaena J. Sm. in habit but has distinctive bathyphylls and anastomosing veins. Christensen (1938) considered the genus Lomagramma as acrostichoid, probably of dryopteroid origin. Holttum (1947, 1949, 1954) for the first time grouped the Lomariopsidoid genera in a separate sub-family Lomariopsidoideae under the family Dennstaedtiaceae. Alston (1956) raised the status of the sub-family to the family level (Lomariopsidaceae), which was later followed by Nayar (1974) and Pichi-sermolli (1877). Ching (1978) segregated Lomagramma and Lomariopsis as an independent group constituting a separate family Lomariopsidaceae (excluding other Lomariopsidoid ferns), possibly derived from Bolbitidaceae. Bower (1923-28) and Holttum (1949) pointed out that the comparative morphology of fern gametophytes could be of significance in understanding evolutionary relationships. According to Stokey (1951, 1960, 1964), Atkinson and Stokey (1964), and Atkinson (1973) comparison of gametophyte structure and their development strengthens our understanding of the relationships among various genera and higher groups. They further indicated that useful data might be found in spore germination pattern, the manner of cell plate 26 AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 1 (2003) development, meristematic region development, and in the type of early prothallial development. Morphologically, the family Lomariopsidaceae is poorly known except for details regarding the sporophyte of Bolbitis and Egenolfia (Nayar, 1950, 1951, 1955, 1956, 1960, Nayar and Kaur, 1964), Elaphoglossum (Bell, 1950, 1951a, 1951b, 1955, 1956) and the rhizome morphology of Lomagramma sorbifolia (Chandra, 1989). Prothallial morphology in the family Lomariopsidaceae is known only for Bolbitis (Nayar, 1960). Egenolfia (Nayar and Kaur, 1964, 1965), and Elaphoglossum and Rhipidopteris (Stokey and Atkinson, 1957). Few details are known about the gametophyte of Lomagramma sinuata (Atkinson, 1973). The present study aims at describing the pathway of prothallus development in L. sorbifolia and comparing that development with that seen in related ferns. MATERIAL AND METHODS The present study is based on material collected from Assam [S. Chandra, LWG 12594). Fresh spores were surface sterilized with sodium hypochlorite (2%) and thoroughly washed with sterilized water. The sterilized spores were sown onto Petri dishes containing Parker and Thompson's nutrient media (Klekowski, 1969) jelled with 1% agar at 5.4 pH. The cultures were maintained at 22 ± 2°C under 600 ft. C. of light from four fluorescent lamps placed horizontally above the culture dishes. All observations on morphology and development of the gametophytes are based on these laboratory cultures. To study cellular structure, the gametophytes were mounted in a 2% acetocarmine solution, which induced partial plasmolysis of the cells rendering the cell outlines clear. Drawings were made using a camera lucida. RESULTS Spores are monolete, planoconvex to somewhat concavoconvex in lateral view, having a granulose exine, devoid of perine (Nayar and Kaur, 1965) and 19 X 27 urn in size (average of 10 readings in each plane of spores selected at random), swelling to 24.5 X 34 urn after acetolysis. They germinate within 1520 days of sowing. At germination an unequal division by a wall perpendicular to the polar axis (parallel to the equatorial plane) of the spore delimits a large, densely chlorophyllous, hemispherical prothallial cell from a small, lensshaped, and very sparsely chlorophyllous rhizoid initial cell next to the proximal pole of the spore (Fig. 1). The rhizoid initial protrudes through the laesural aperture and elongates parallel to the polar axis of the spore as a slender, highly vacuolated rhizoid. Meanwhile, the prothallial cell enlarges, elongating along the equatorial plane of the spore, splitting open the spore-coat at the laesural region and dividing by a wall parallel to the polar axis (perpendicular to the first wall) of the spore in such a way that the rhizoid is attached laterally to the basal one of the two daughter cells (Figs. 2, 3); this basal cell does not take any further part in prothallial development. Its sister CHANDRA ET AL.: GAMETOPHYTES OF LOMAGRAMMA FIGS. 1-29. Stages in the development i,f ,,a»ramma sorbifolia. 1. Spore germination. 2-3. Uniseriate germ filament. 4. Initiation of plate formation by oblique division of the terminal cell. 8-9. Initiation of plate formation by verti inal cell. 5-7 and 10-13. Germ filament showing early formation of apical cell. 14-15. Germ filament showing the FERN JOURNAL: VOLUME 93 NUMBER 1 (2003) cell elongates further along the equatorial plane and divides repeatedly by walls parallel to its basal wall (parallel to the polar axis of the spore) forming a short germ filament composed of short, barrel-shaped, and densely chlorophyllous cells (Fig. 3). Spore germination, thus, is of the typical Vittaria-type as described by Nayar and Kaur (1968, 1971), the first rhizoid elongating along the polar plane of the germinating spore and the germ filament elongating along the equatorial plane and perpendicular to the first rhizoid. However, due to the physical obstruction provided by the spore coat, the emerging germ filament is often slightly deflected from the equatorial plane. When the germ filament is three to five cells long, formation of a prothallial plate is initiated by an abrupt change in the plane of wall formation in the terminal cell and often extending to the penultimate cell. Instead of dividing by walls perpendicular to the long axis of the germ filament, these cells divide by walls parallel to the long axis so that the germ filament at its anterior end becomes two tired. Commonly the wall formed in the terminal cell is oblique (Fig. 4) so that one of the daughter cells is larger with a broader anterior end. Another wall oblique to this wall formed in the larger daughter cell delimits a wedge-shaped apical meristematic cell (Figs. 5-7, 13). In some cases the first division of the terminal cell is parallel (instead of oblique) to the long axis of filament (Figs. 8, 9) followed by an oblique division in this cell to delimit a wedge-shaped apical meristematic cell. Thus, a transverse row of three daughter cells is formed, of which the middle one is wedge-shaped and acts as a meristematic cell (Fig. 10). This type of prothallial development is termed the Adiantum-type (Nayar and Kaur, 1969, 1971). The meristematic cell cuts off a series of narrow daughter cells alternately against its oblique sides and these daughter cells, by successive anticlinal and periclinal divisions, form an expanded, one-cell-thick, obovate prothallial plate (Figs. 18-20). Daughter cells of the meristematic cell grow and divide rapidly so that the anterior region of the prothallus on either side of the meristematic cell progressively extend anterior to the level of the meristematic cell, ultimately making the young prothallus cordate (Figs. 28, 29). A second abrupt change in the plane of cell divisions occurs in the apical meristematic cell when the young prothallus is distinctly cordate. Instead of dividing by walls parallel to its oblique sides, the apical cell divides by a transverse wall, cutting off its wedge-shaped basal region from the larger anterior region, which then divides repeatedly by longitudinal walls to form a plate of 3 or 4 narrow cells. These cells constitute a pleuricellular meristem (Figs. 26, 27) in which all cell divisions are longitudinal. Ultimately a central midrib is established behind the meristem in the median plane of the thallus. The prothallus becomes symmetrically cordate, and has semicircular lateral wings (Fig. 28). Occasionally, the establishment of an apical cell is much delayed. In such cases the first division of the terminal cell is by a vertical wall (parallel to the long axis of the filament instead of oblique) and soon a second wall is laid down at a right angle to the first. A broad spatulate prothallial plate is formed (Figs. 14-16) by divisions of the distal cells of the germ filament by walls CHANDRA ET AL.: GAMETOPHYTES OF LOMAGRAMMA parallel to the long axis and by repeated longitudinal and t in the daughter cells. This type of prothallial development is termed the Drynaria-type (Nayar and Kaur, 1969, 1971). The plate often becomes 5-10 cells wide and broadly ovate but is devoid of any organized meristem (Figs. 17, 21-24). An obconical meristematic cell is differentiated later by two oblique divisions in one of the marginal cells at the anterior end of the prothallial plate (Figs. 25, 26). Finally, a symmetrical cordate prothallus is formed. In a few cases the terminal cell of the germ filament may not participate in the formation of the apical cell, or may be sluggish in doing so. In such cases, the obconical meristematic cell is formed behind the terminal cell by an oblique wall (Figs. 11, 12). Activity of this form of meristematic cell results in a spatulate prothallial plate. The meristematic activity may be restricted to one side of the plate, but ultimately an Adianturn-type, cordate prothallus is formed. Rarely, the germ filament is branched (Fig. 16), with each branch developing into separate gametophyte. The mature prothallus is a typical heart-shaped structure with a prominent apical notch and takes about 128 days to develop from spore. The young gametophytes are entirely naked, being devoid of any hairs (Fig. 29). The rhizoids are hyaline. Until this stage of development, the midrib is undifferentiated and the sex-organs are not formed. DISCUSSION The early gametophyte development in Lomariopsidaceae has been classified primarily as Drynaria-type (Nayar and Kaur, 1971), or rarely the Aspidium-type as in Elaphoglossum (Stokey and Atkinson, 1957)." The Drynaria-type of gametophyte development has been reported in a majority of the genera of Polypodiaceae (Nayar and Raza, 1970; Nayar and Kaur, 1971; Chandra, 1979; Chiou and Farrar, 1997; Perez-Garcia et al, 2001. This type of development is also characteristic of some Athyrioideae, Cheiropleuriaceae, Cyatheaceae, Dipteridaceae, Dryopteridaceae, Gleicheniaceae, Loxsomaceae, Thelypterdiaceae, (Nayar and Kaur, 1971). The present study reveals that spore germination in L. sorbifolia is the typical Vittaria-type of polar germination, while prothallial development is primarily of the Adiantum-type as reported for the Dennstaedtiaceae (Nayar and Kaur, 1969). The Adiantum-type of prothallial development is characteristic of the families Dennstaedtiaceae, Grammitidaceae, Hypolepidaceae, Lindseaceae, Lygodiaceae and Plagiogyriaceae. In addition, it is also found in some genera of Cyatheaceae, Athyrioideae, Adiantaceae {Adiantum, Coniogramme) and occasionally of the families Dryopteridaceae [Didymochlaena), Aspleniaceae (some species of Asplenium), Blechnaceae (some species of Blechnum) and Cheilanthaceae [Doryopteris, some species of Cheilanthes) (Nayar and Kaur, 1969, 1971). Lomagramma sorbifolia is unusual, so far as the development of the gametophyte {Adiantum-type) is concerned, relative to most of the other genera of the Lomariopsidoid ferns. However, it shows similarities with other 30 AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 1 (2003) members of the Lomariopsidoid group, which have a Drynaria-type of development. The Adiantum-type of development has been considered to be more primitive than that of the Drynaria-type. In the Adiantum-type of development, growth and expansion of the prothallus is mainly through the activity of meristematic cells, whereas in the Drynaria-type of development the meristematic cells do not play a very active part in the growth and expansion of the young prothallus, as is the case for the majority of the Polypodiaceae (Nayar 1962, 1963, 1965). Strap-shaped, lobed, or elongated prothallii (Atkinson, 1973), as reported in some of the Lomariopsidaceae [Lomariopsis hederacea, Egenolfia vivipara, Bolbitis repanda and Elaphoglossum cuspidatum), have not been observed in L. sorbifolia. The prothallus is naked throughout as reported for most species of Bolbitis and Elaphoglossum. The spores are bilateral and non-perinate as in Thysansoria (Nayar and Kaur, 1965). However, at least in some cases of L. sorbifolia, besides the Drynaria-type of early gametophyte development, the most common development is of /la'ianfum-type. Nayar and Kaur (1969) consider this an unusual feature among most of the other genera of the lomariopsidoid ferns. This supports the view of Holttum (1947), who considers them possibly to have been derived directly from a dennstaedtioid stock. We are grateful to Professor B. K. Nayar, ex Head, Department of Botany, Call. ut Univ aluable suggestions and advice. We also acknowledge with Pushpangadan, Dii rector, N.B.R.I., Lucknow (INDIA) for constant encouragemen tl.d.ilj. facilities for this * vork. Mrittunjai Srivastava is grateful to Department of Scie & Ttu.hn New Delhi for the award of Research Associate Fellowship. LITERATURE CITED A'KIX"; BELL, A. H. G. 1 956. The sub divisicm of the Polypodiaceae. Taxon 5i:23-25. 73. The Gametophyl e and family relationship. Bot. J Sot;. (Suppl. and A. G. STOKEY. 1964. Comparative morphology of the gametophyte of the homosporom ferns. Phytomorphology 14:51-70. P. R. 1950. Studies in the genus Elaphoglossum Schott. I. Stelar structure in relation to habit Ann. Bot. (London) 14:545-55. (London) 15:333-46. —. 1951b. Studies in the genus Elaphoglossum Schott. III. Anatomy of rhizome and frond Ann. Bot. (London) 15:347-57. . 1955. Studies in the genus Elaphoglossum Schott. IV. The morphological series in th< genus and their p] itian, Pt. I. Ann. Bot. (London) 19:173-99. . 1956. Studies in the genus Elaphoglossum Schott. IV. The morphological series in th( genus ami th rpretation, Pi BOVVER. F. O. 1923-28. The Ferns, (Vol I-III). Univ. ] CHANDRA, S. 1979. Gametophyte morphology of the fern { CHANDRA ETA I LOMAGRAMMA —. 1989. Structure sorbifolia (Willd.) Ching. Acta —. 2000. The Ferns of 1978. The Chi KAUR, sification of ferns. Biol. Rev. 24:267S. 1962. Bolbitis, Egenolfia and related genera. Ph.D . 1974. The family Lomariopsidaceae (Filicopsida) i E. J. 1969. Reproductive biology of the Pteridop J. Linn. Soc. 62:347-359. B. K. 1960. Morphology of two . 1962. Morphology of spores and prothalli of some 123:223-232. . 1963. Contribution to the morphology of some specie KLEKOWSKI, NAYAR, . 1965. The gametophyte and juvenile leaves of the drynarioid ferns. Bot. Gaz. 126:46-52. . 1974. Classification of homosporous ferns part II. Pp 111-201, in B. K. Nayar and S. Kaur, eds. Companion to R. H. Beddomes Handbook to the ferns of British Inda, Ceylon and the Malaya Peninsula. Chronica Botanica, New Delhi. and S. KAUR. 1964. Studies on the fern genera Bolbitis and Egenolfia II. The gametophytes and the Juvenile sporophytes. J. Linn. Soc, Bot. 59:141-153. and F. RAZA. 1970. Prothalli of some Polypodiaceae II. J. Indian Bot Soc. 49:81-86. and S. KAUR. 1965. Spore morphology of the Lomariopsidaceae. J. Palynol. 1:10-26. and S. KAUR. 1968. Spore germination in homosporous ferns. J. Palynol. 4:1-14. and S. KAUR. 1969. Type of prothallial development in homosporous ferns. Phytomorphology. 19:179-188. and S. KAUR. 1971. Gametophytes of homosporous ferns. Bot. Rev. 37:295-396. PEREZ-GARCIA, B., A. MENDOZA, R. RIBA and L. D. GOMEZ-PIGNATARO. 2001. Development of the sexual phase of Pseudocolysis bradeorum (Polypodiaceae). Amer. Fern J. 91:214-226. PICHI SERMQLLI, R. E. G. 1977. Tentamen pteridophytorum in taxonomicum ordineum redigendi. A. G. 1951. A contribution by the gametophyte to the classification of the homosporous ferns. Phytomorphologx 1 19-58 STOKEY, New combinations in the Tropical American Ctenitis (Tectariaceae) Y of Ctenitis (Tectariaceae) from Brazil we detected two n that genus: nb. nov. and Ctenitis laetevirens (Rosenst.) Salino & Morais comb. nov. The first species is lilar to Ctenitis nigrovenia (H. Christ] Copel., but differs mainly by the short-creeping stem and nitoid hairs on the segment margins. Ctenitis laetevirens is related to C. submarginalis (Langsd. 'isch) Ching, but differs by having pinnae long-petiolulate, ctenitoid hairs absent on the segment rgins, the abaxial side of costae, costule and veins, and by having exindusiate sori. Besides the earlier monographs by Christensen (1913, 1920) and the surveys of Brade (1972) and Sehnem (1979), nothing else has been published on taxonomy of the Brazilian species of Ctenitis. Ctenitis is essentially pantropical with 70 to 80 species. About half of these occurs in the Neotropics (Tryon & Stolze, 1991), and 14 to 16 species in Brazil. This genus is closely related to Lastreopsis which can be distinguished by the configuration of the adaxial axes. In Lastreopsis the ridges are continuous with the ridges on the axes of the next order above or below, while in Ctenitis these ridges are lacking or, when present, not continuous onto adjacent axes (Tryon & Stolze, 1991). Many species were removed from Ctenitis and placed in two other genera: Megalastrum and Triplophyllum both described by Holttum (I986a,1986b). The differences between these three genera are well discussed by Smith & Moran (1987) and Tryon & Stolze (1991). In southern and southeastern Brazil the species of Ctenitis often grow in mesic and moist-shaded habitats such as primary and secondary lowland and montane rain forests, from 0 to 1700 meters in elevation. While working on the taxonomy of Brasilian Ctenitis we detected two species that need to be combined in the genus. Ctenitis abyssi (Sehnem) Salino Sr Morais, comb. nov.—Dryopteris abyssi Sehnem, Fl. Ilustr. Catar. 1 (Aspidiaceas): 156. 1979. TYPE: Brazil. Rio Grande do Sul: Sao Francisco de Paula, Taimbe, 17 Feb 1953, Sehnem 6315 (Holotype, PACA!). Fig. 1A-F. mens (Luederwaldt 1814). G. Habit. H. Abaxial side of segments, si position and veins. I. Stem scale. J. Petiole scale. K. Rachis scale. L. Scale of abaxial s M. Ctenitoid hairs (hydrated) on adaxial side of costae. SALINO & MORAIS: NEW COMBINATIONS FERN JOURNAL: VOLUME 93 NUMBER 1 (2003) This species belongs to Ctenitis based on the ctenitoid hairs on the petiole, both sides of the rachis and pinnae, and along the margins of the segments; vein tips not enlarged and terminating at or very near the segment margin; and the presence of a cylindrical glands on the abaxial side of the pinnae. According to Sehnem (1979), Dryopteris abyssi differs from other species of the genus by the membranaceus lamina and narrow linear segments, and is related to Ctenitis nigrovenia (H. Christ) Copel. Ctenitis abyssi is related to C. nigrovenia because of its similar pinnae, segments with serrate margins, and bullate scales on the abaxial side of the rachis and costae. However, C. nigrovenia lacks ctenitoids hairs on the segments margins, has medial to inframedial sori, and the stem is erect to decumbent. Ctenitis abyssi has a shortcreeping stem, medial to supramedial sori, and ctenitoids hairs on the margins of the segments. Ctenitis nigrovenia is found from southern Mexico to Peru (Tryon & Stolze, 1991), but in Brazil occurs only in the Amazonian region. Ctenitis abyssi is a southern Brazil endemic and is known only from the type collection in the Taimbe Cannion region (State of Rio Grande do Sul). It grows on rock at 700 meters in elevation. Ctenitis laetevirens (Rosenst.) Salino & Morais, comb. nov.—Dryopteris laetevirens Rosenst., Hedwigia 56: 368. 1915. Lectotype (designated here): Brazil. Santa Catarina: Hammonia, Aug 1910, Luederwaldt 1380 (SP!). Fig. 1G-M. This species belongs to the genus Ctenitis based on the ctenitoid hairs on the adaxial side of rachis, costae and costules, and the narrow vein tips terminating at or very near the segment margin. According to Rosenstock (1915), this species is related to Ctenitis submarginalis (Langsd. & Fisch) Ching and C. falciculata (Raddi) Ching. From these species, C. laetevirens differs by having the abaxial side of the costae, costules, veins and laminar tissue glabrous, long-petioluled pinnae, and exindusiate sori. Rosenstock (1915) mentioned that Dryopteris laetevirens resembles Ctenitis aspidioides (C. Presl) Copel. which has cuneate pinnae bases, indusiate sori, and leaves that are brown when dried. Ctenitis laetevirens is frequently confused with C. submarginalis a species with ctenitoid hairs on the segment margins, the abaxial side of costae, costules and veins, often indusiate sori, a moderately scaly stem, petiole, rachis, costae, and costules. Ctenitis aspidioides has longpetioluled pinnae and a poorly developed indumentun, as in C. laetevirens, but C. aspidioides has conform apical pinnae and ctenitoid hairs on the segment margins. Ctenitis laetevirens is endemic to the state of Santa Catarina in southern Brazil. It is terrestrial in the Atlantic Rain Forest between 0 and 100 meters in elevation. ADDITIONAL SPECIMENS EXAMINED.—BRAZIL. Santa Catarina: Blumi (Syntype, UC!); Hansa, October 1911, Luederwaldt 1815 (US); Hammo 1814 '(BHCB, NY, SP, SPF, UC); Warrow, 1905, Goeden 49 (NY, UC). '••I- - -. V ',. ' LITERATURE CITED A. C. 1972. O genero Dryopteris (Pteridophvta) no Brasil e sua divisao taxonomic 1(22):191-261. C. 1913. A monograph of the genus Dryopteris, Part I, the tropical . pinnatifid-bipinnatifid species. Kongel Danske Vidensk. Selsk. Skrift. Naturviden Afd., ser. 7., 10:55-282. . 1920. A monograph of the genus Dryopteris, Part II, the tropical American decompound species. Kongel Danske Vidensk. Selsk. Skrift. Naturvidensk. Math. A BRADE, CHRISTENSEN, A. 1979. Aspidiaceas. In: R. Reitz (Ed.), FL Ilustr. Catarinense. Parta 1. Herb. Barbosa Rodrigues. Itajai. A. R. and R. C. MORAN. 1987. New combinations in Megalastrum (Dryopteridaceae). Amer. SEHNEM, SMITH, Polypodium appalachianum: An Unusual Tree Canopy Epiphyte in the Great Smoky Mountains National Park HAROLD W. KELLER Department of Biology, Central Missouri State University, Warrensburg, MO 64093 PAUL G. DAVISON Department of Biology, University of North Alabama, Florence, AL 35632 CHRISTOPHER H. HAUFLER Department of Ecology and Evolutionary Biology, University of Kansas, Lawrence, KS 6604 DAMON B. LESMEISTER Department of Biology, Central Missouri State University ABSTRACT.—The typ podium appalachianum was discovered as a epiphyte 3b to 40 m above ground on a horizontal branch of a champion-size Lirioc tulipifera in the Great Smoky Mountains National Park. Occurring along with tl documentation of P. appalachianum from the tree canopy was an assemblage of n terrestrial mosses, an unusual assortment of collembola (springtails), and a flightless r insect species previously known only from soil a< five features of this habitat may duplicate some ecological conditions usually found only at ground le' the opportunity for translocating an entire community and providing biologists wi on the origin of some epiphytes. Species of Polypodium in North America grow on rocky surfaces, soil, rotted wood on ground sites, and as epiphytes on living trees (Tryon and Tryon, 1982). As currently circumscribed, there are approximately 100 species worldwide (Haufler et ah, 1993). Polypodium appalachianum Haufler and Windham, often called the "Rock Cap" fern because it usually festoons the crowns of large boulders, is one of three Polypodium species native to eastern North America. Polypodium appalachianum has been reported from the eastern Tennessee counties of Blount, Sevier, and Cocke and in the western North Carolina counties of Hayward and Swain, all within the boundaries of the Great Smoky Mountains National Park (GSMNP). Although occasionally epiphytic at the base of tree trunks (Patricia Cox, pers. comm.), discovery of P. appalachianum growing high in the canopy of a champion-sized, living Liriodendron tulipifera L. (Yellow Tulip Poplar) tree in the GSMNP represents an epiphytic microhabitat not previously documented. In this paper, we describe the canopy microhabitat of P. appalachianum and associated mosses, compare the vertical distribution of bryophytes along the main trunk axis with the horizontal branch that supported the fern microhabitat, briefly describe the climbing techniques used to access the tree canopy, and provide a description and photographs of specimens collected from the tree canopy. KELLER ET AL.: POLYPODIUM APPALACHIANUM, AN UNUSUAL CANOPY EPIPHYTE 37 STUDY AREA AND SAMPLING METHODS The GSMNP comprises more than 200,000 ha and serves as a refuge for one of the richest and most diverse biotas in the temperate regions of the world. It also has the largest remaining tracts of old growth forest in eastern United States, estimated at 40,000 ha. As part of a research effort to inventory all of the life forms in the park, the All Taxa Biodiversity Inventory (ATBI) established 20 one-hectare study plots located in major habitats throughout the park. Site selection was based on major forest/vegetation types, elevation and relative accessibility. Two giant Liriodendron tulipifera trees are located 1,021 m above sea level on each side of the Ramsay Cascades Trail approximately 1.61 km from the trailhead in the Tennessee part of the park. This is near but outside the ATBI Ramsay Cascades study site and within the Cove Hardwood-Eastern Hemlock forest type found throughout the Ramsay Prong ravine. These trees were called "majestic Roman columns" by Gove (1994) along with a description of the Ramsay Cascades Trail in a popular hiking trail book. Polypodium appalachianum was collected August 2, 2001, from the canopy of one of the giant Liriodendron tulipifera trees (#307), which measured 169 cm in diameter at breast height and 52.8 m in total height. During the summers of 2000 and 2001 Central Missouri State University students participating in a tree canopy biodiversity study in the GSMNP climbed and collected bark and epiphyte samples from a total of 240 trees representing 35 different species. The climbers used the double rope climbing technique to access the tree canopy. This technique allows the climber to advance from branch to branch in order to reach higher levels of the tree canopy (Counts et al, 2000). Specimens of epiphytes along with bark samples were collected at approximately 3 m increments. Height above ground was measured by an elevation line attached to the climber's harness. The Liriodendron tulipifera sampled and others in the vicinity were covered with epiphytic mosses and liverworts near the trunk bases. Ferns were absent from the vertical trunks. In the eastern United States little is known about the occurrence of bryophytes in tree canopies and we know of no publications reporting canopy occurrences for any of the species reported here. Table 1 lists the bryophytes identified from bark samples collected at several heights above ground on tree #307. The most remarkable species above 30.5 m are the mosses Rhodobryum roseum, Trichostomum tenuirostre, and the liverwort famesoniella autumnalis because these species are usually restricted to the extreme bases of trees. We have seen /. autumnalis as a rarity at 2 m above the ground on vertical tree trunks but never R. roseum or T. tenuirostre, the latter being more commonly found on rock (Crum and Anderson, 1981). The assemblage of species reported in Table 1 is typical of lower tree trunk floras in mesic woods. Their occurrence high above the ground suggests humidity and AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 1 KELLER ET . \ \ \ i \ l SUAL CANOPY I - • . • Campylium chrysophyllum (Brid.) J. Lange Dicranum montanum Hedw. Dicranum viride (Sull. & Lesq.) Lindb. *Lejeunea ruthii (Evans) Schust. *Lejeunea ulicina (Tayl.) Gott. Leucodon brachypus Brid. Orthotrichum sp. Platygyrium repens (Brid.) BSG *Porella platyphylla (L.) Pfeiff. "Radula obconica Sull. Rhodobryum roseum (Hedw.) Limpr. Thuidium delicatulum (Hedw.) BSG Trichostomum tenuirostre (Hook. & TJ that found at tree bases. Bryophyte voucher he University of North Alabama (UNAF). A horizontal branch at 35 m was the site of a microhabitat where mosses and ferns were confined to the upper surface, extending for about 4 m along the branch (Fig. 1). In order of their abundance, the mosses included R. roseum, (Fig. 2), Thuidium delicatulum, Platygyrium repens, and Anomodon attenuatus. These mosses provided a loose, soil-forming mat of humus approximately 10 to 14 cm thick that supported the creeping P. appalachianum rhizomes. Polypodium appalachianum was in several stages of development including infertile and fertile blades, the latter with immature, yellowish green sori and FIGS. 1-5. Polypodium appalachianum and Rhodobryum roseum. 1. Habit of epiphytic appalachianum collected from horizontal branch in the tree canopy; scale bar = 12 mm. 2. Hal view of Rhodobryum roseum., the bright green moss growing in dense patches of terminal i settes, and other mosses forming a thick, humus mat; scale bar = 4.7 mm. 3. Fertile blade with u mature sori on upper one-third of blade; scale bar = 6 mm. 4. Mature sori present on upper portion 40 AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 1 (2003) mature, rusty red sori with, dehisced sporangia (Figs. 1, 3-5). Polypodium appalachian urn was also observed growing on another horizontal branch at 40 m. 1, 3-5).—Plants gregarious, creeping rhizomes embedded in moss and humus; leaves 5 to 16 cm long, blade deltate, 2.3 to 5.3 cm at widest point near the base; ultimate segments thin, flexible, linear to oblong with acute to narrowly rounded apices, broader at base, 0.4 to 0.8 cm wide, margins entire to crenulate, upper surface and midrib glabrous; venation free; sori on distal 1/3 to 1/2 of blade, borne abaxially at tips of single veins, lacking indusia, located midway between margin and midrib of the ultimate segments, 1.5 to 2.0 mm in diameter, circular when immature; sporangiasters more than 40 per sorus, heads densely covered with glandular hairs; spores ovate, with rough ornamentation of low, flattened projections, verrucose, 38 to 42 urn in length, falling within the diploid range. Voucher specimen deposited at the University of Kansas R. L. McGregor Herbarium (KANU). MORPHOLOGICAL DESCRIPTION OF TREE CANOPY POLYPODIUM APPALACHIANUM (FIGS. Polypodium appalachianum was collected on August 2, 2001, falling within the summer and fall seasonal sporulation for this species. According to Haufler and Windham (1991), P. appalachianum is diploid with a chromosome number of 2n = 74, occurring from southeastern Canada, southward along the Appalachian Mountains and eastern seaboard states to Georgia and Alabama. Montgomery (1996) noted the habitat for P. appalachianum as mostly on rocks, boulders, ledges, cliffs, or rocky woods. A few specimens were recorded from tree trunks or bases of trees. Our collection is the first published record of P. appalachianum in the tree canopy. Polypodium appalachianum was previously treated as part of a single polymorphic taxon, P. virginianum, with 2n = 148, now understood to be an allotetraploid having P. appalachianum as one of its progenitor diploids (Haufler et al, 1993). What conditions have developed over time to provide a suitable habitat for P. appalachianum to become established, develop fertile sporophytes, and spread over several meters on just the upper surface of a horizontal branch? Barkman (1958) noted that Polypodium species produce a dense mat of roots with many fine hairs that serve to trap and retain moisture and nutrients. Thick horizontal branches provide a microhabitat that leads to heavy snow cover in winter protecting the epiphytes against frost and desiccation. In addition more dust, sand, and particulate matter accumulate over time to provide a thick humus greater than on the vertical trunk of the tree, thus favoring the establishment of terrestrial moss and fern species. Certainly the size of this tree would suggest a life span of more than 400 years. Litter and moss samples from the fern site were sent to Dr. Ernest C. Bernard at the University of Tennessee. His analysis of these samples for apterygotes indicated that the canopy collembola (springtails) fauna was distinct from that of the ground fauna, with little or no overlap in species composition. In addition to the collembola taxa KELLER ET Ah.: POLYPODIUM APPALACHIANUM, AN UNUSUAL CANOPY EPIPHYTE 41 collected from this site, which will be published elsewhere, the discovery of Acerentulus confinis (Berlese), a proturan, was a puzzling find, because this group had been considered to be strict soil and litter organisms. The proturans have no known capacity for dispersing to the canopy of trees and surviving there. The tree canopy of old growth forests in eastern United States remains largely unexplored for a myriad of different organisms. The discoveries documented here demonstrate that these habitats should not be taken for granted because they may yield insights on the origin of epiphytes. Whereas there is no doubt that special adaptations evolve in some epiphytic species (e.g., Benzing, 2000), our observations provide support for hypotheses that aerial habitats can mimic those on the forest floor (Bohlman et al., 1995) and provide opportunities for remarkable vertical disjunctions. Our results should encourage others to search in treetops to fully inventory and sample the biodiversity that exists in this aerial ecosystem. We thank Keith Langdon from the GSMNP and Jeanie Hilten from Discover Life in Amrru a who provided assistance with equipment, housing and logistics. Damon Lesmeister was the student climber who climbed the yellow poplar tree and discovered P. appalachianum. Special thanks go to Charly Pottorff, a professional arborist, who provided tree-climbing install don and certification for student climbers. Drs. Patricia Cox and Ernest Bernard from the University of TennesseeKnoxville provided valuable information on their research activities in the GSMNP. This research project was funded by the National Science Foundation Small Grant for Exploratory Research, Division of Environmental Biology, Biotic Surveys and Inventories Program, Award #DEB0079058 and Discover Life in America Awards #2001-26 and #2002-17. LITERATURE CITED S., T. MATELSON and N. NADKARNI. 1995. Moisture and temperature patterns of canopy humus and forest floor soils of a montane cloud forest, Costa Rica. Biotropica 27:13-19. J., L. HENLEY, M. SKRABAL and K. L. SNELL. 2000. Tree Canopy Biodiversity in the Great Smoky Mountains National Park. Inoculum 51(6):l-5. CRUM, H. A. and L. E. ANDERSON. 1981. Mosses of Eastern North America. Vol. 1. Columbia BOHLMAN, COUNTS, GOVE, D. 1994. Ramsay Cascades T Hiking Trails of the Smoh HAUFLER, C. H. and M. D. MONTGOMERY, WIND: J. D. 1996. Polypodium appalachianum, P. virginianum, and their hybrid in New ?d Plants, with Special Reference to Tropical SHORTER NOTES An Adiantopsis Hybrid from Northeastern Argentina and Vicinity.—During a recent collecting trip to the Parque Nacional Iguazu, Misiones, Argentina, an unusual specimen of Adiantopsis was collected {Hickey 01-63 et al., MU; Fig. 1). The single plant was found growing with A. radiata (L.) Fee on steep, moist, wooded slopes along the walkway leading to Iguazu Falls. Its leaves were pedate with inequilateral basal pinnae and extended basal basiscopic pinnules (Fig. 1). In Tryon & Tryon [Ferns and Allied Plants with Special Reference to Tropical America, 1982), this plant keyed to A. pedata (Hook.) T. Moore, a species listed as endemic to the Greater Antilles. Comparisons of the Argentinian collection with Caribbean material showed no obvious morphological differences between the two (Fig. 1). Herbarium loans (BM, MO, MU, UC, SI) revealed additional collections of this unusual A. pedata-\ike plant dating back to 1907, and ranging into adjacent areas of Brazil and Paraguay. Among these collections, Biganzoli et al. 168 (SI) was identified by M. Ponce as A. pedata as was Rojas 10451 (BM) by Pena-Chocarro. Hahn 2013 (MO, UC) was annotated by A. R. Smith as Adiantopsis chlorophylla X radiata. Evidence from spores supports Smith's contention for a hybrid origin of the South American plants. Spores of Adiantopsis pedata from the Greater Antilles number 64 per sporangium and are uniform in shape and size. In contrast, the material from South America shows a variable number of spores (52 to 76 per sporangium), most of which are misshapen, suggestive of a hybrid origin. These South American collections, therefore, represent the first known hybrids in Adiantopsis. The enlarged basal basiscopic pinnules, reduced leaf dissection, and shape of the ultimate segments argue strongly for Adiantopsis radiata as one of the parents. The second parent, contributing the pinnate frond architecture, is probably a member of the taxonomically difficult A. chlorophylla (Sw.) Fee complex. Potential taxa include A. chlorophylla, A. perfasciculata Sehnem, and A. occulta Sehnem. Hypotheses including A. perfasciculata and A. occulta as the second parent are supported by their erect rhizomes and densely crowded stipes, characters quite evident in the hybrid. The creeping rhizomes and more distantly attached stipes of A. chlorophylla argue against its involvement, although it is possible that A. radiata may have individually conferred these traits to the hybrid. Preliminary analyses of spore morphology support A. perfasciculata as the second parent. The spores of this species and the hybrid possess elongate, bent spines, characters not seen in the other species. Ancillary support for parentage is derived from geography. Adiantopsis radiata and A. chlorophylla are both widely distributed and often sympatric in the American tropics. In contrast, the hybrid is restricted to Argentina, Brazil and Paraguay and its absence throughout the range of cos argues against a widespread A. chlorophylla as the second parent. SHORTER NOTES and spores of Adiantopsis > from Misiones, Argentina [Hickey 01-63 et al. MU). bar = A. Xaustralopedata [W. H. Hahn 2013, UC). bar = 25 \im. C. A 35655, US), bar = 5 cm. D. Normal spores of A. pedata {Proctor 3 spores of a [Proctor Both A. perfasciculata and A. occulta, however, have ranges nearly identical to that of the hybrid and both have the expected frond architecture predicted for the second parent. Considering the ranges of the hybrid and its putative parents, it is surprising that there has been no reference to the hybrid in Rosentock (Hedwigia 46:57-167. 1907], the various floristic treatments by Sehnem (Pesquisas 3:495-576, 5f. 1959; Pesquisas 13:1-42, lOf. 1961 in : VOLUME 93 NUMBER 1 Adiantopsis xaustralopedata Hickey, Barker, et Ponce, hybr. nov. Fig. 1. A & B. Type.—Paraguay, Depto. Cordillera, Caacupe, semideciduous forest to 20 m tall on fairly steep slope, Enterolobium, Parapiptadenia dominants, soil sandy with some red clay, 25° 20' S, 57° 10' W, 9 Feb 1984, Hahn 2013 (holotype MO, sheets 1 and 2; isotype UC ). Laminae pedatae; pinnae supernae bipinnatae; pinnae basales tripinnatae, praebens pinnulas basales basiscopicas elongatas magnopere. Ab A. pedata sporis abortivus differt. PARATYPES.—Brazil: Rio Grande do Sul, transiens in Ad. pedata, Cameste do Peiraes, 1907, Jurgens 173a (UC). Paraguay: in altplnitie et declivibus "Sierra de Amambay", May 1907/1908, Rojas 10451 (BM); Colonia Indepencia Villarica, 13.11.1945, Teague 453 (BM). Argentina: Misiones, Dep. Iguazu, Parque Nacional Iguazu, Hickey 01-63, Taylor, Strittmatter & Guaglianone (MU). Dep. Cainguas, Predio de la Universidad Nacional de La PLata, valle de arroyo Curia Pirii, 2do. campo con "Urunday", 27° 07' S-54° 58' W, sotobosque, Biganzoli, Peralta, Giallorenzo & Moreno 168 (SI). The authors are indebted to Lara Strittmatter and Rosa Guaglianone for trip arrangements and field assistance. We also thank the Parque Nacionales Administration for allowing botanical collections at Iguazu Falls, and acknowledge the financial support of the W. S. Turrell Herbarium (MU). This work represents a portion of a Master's project on Adiantopsis being conducted by Michael Barker at Miami University.—R. JAMES HICKEY, MICHAEL S. BARKER, Botany Dept. Miami University, Oxford, OH 45056 U.S.A.; MONICA PONCE, Instituto de Botanica Darwinion, Labarden 200, B1642HYD San Isidro, Argentina. Leaf Flavonoids in the Genus Gleichenia (Gleicheniaceae).—As part of a continuing chemotaxonomic study of flavonoids in genera of the Gleicheniaceae by Umi Kalsom (Blumea 40: 211-215. 1995), our attention has turned to Gleichenia, which contains some five species and two varieties. Apart from the genus Dicranopteris, the family has not been extensively investigated and the results of a general flavonoid survey will be presented later. This paper describes the identification of some of the major flavonoids found in the genus Gleichenia. From the viewpoint of flavonoid chemistry, the only major survey of Gleichenia has been that of Wallace et al. (Amer. J. Bot. 70: 207-211. 1983) who found flavonol 3-O-glycosides to be major components in methanolic leaf extracts of 8 species. In addition, some species appear to accumulate traces of chalcone O-glycosides and/or aurone O-glycosides. The purpose of the present research was to determine whether or not other members of the Gleicheniaceae have flavonoid profiles similar to the gleicheniaceous ferns previously studied. For this, the flavonoid profiles of Gleichenia hirta Bl., G. microphylla R. Br., G. longissima Bl. and G. blotiana C. Chr. as interpreted by Piggot [Ferns of Malaysia in Colour, Tropical Press, Sdn Bhd., Kuala Lumpur, 1998) were determined and compared with those of Gleichenia by Wallace and Markham (Amer. J. Bot. 65: 965-969 1978). Leaves from freshly dried plant material collected from various habitats in Peninsular Malaysia were analysed. Voucher specimens of the ferns (collection number: UKY 326-329) have been deposited in the herbarium of the Department of Biology of the Universiti Putra Malaysia. Standard chromatographic procedures (Harborne, J. B. 1967, Comparative Biochemistry of the Flavonoids, Academic Press, London; Markham, K. R. 1982, Techniques of flavonoid Identification, Academic Press, London) were used for examining flavonoids present in direct and acid hydrolysed leaf extracts; the common aglycones were identified by means of Rf values and color reaction in UV light when compared with standard markers. In acid-hydrolyzed extracts, the flavones were recognized by their distinct, dark yellow spots on paper chromatograms in UV light. When fumed with ammonia vapor they became bright yellow. The flavonols appeared yellow in UV light before and after fuming with ammonia. For complete identification of flavonoid glycosides, samples were separated in one-dimensional chromatograms of direct extracts and then the pure flavonoids were identified by UV spectral analysis using standard procedures of Mabry and coworkers [The Systematic Identification of the Flavonoids, Springer-Verlag, New York, 1970). In addition to spectral techniques, flavonoids were identified by PC (Whatman No. 1) co-chromatography of the glycosides and products of enzyme and acid hydrolyses in n-butanol-acetic acid-water (BAW, 4:1:5) and 50% glacial acetic acid (50% HOAc). The aglycones were identified by TLC (Merck) co-chromatography in BAW, forestal (concentrated hydrochloric acid-acetic acid-water, 3:30:10) and 30% HOAc, whereas the sugars were identified by PC co-chromatography in BAW, hbutanol-ethanol-water (BEW, 4:1:2.2) and toluene-n-butanol-pyridine-water (TBPW, 5:1:3:3). Twelve compounds were obtained in a more or less pure state by means of preparative chromatography. All species produce kaempferol and quercetin, while genkwanin and luteolin were present in G. blotiana C. Chr. and G. hirta Bl. and acacetin in G. microphylla R. Br. This is the first report of acacetin and genkwanin in Gleihenia. Acacetin was isolated as the 7-glucoside. The flavonols of Gleichenia leaves were found to be present as 3-glucosides, 3-rhamnoside, 3-rutinoside, 3,4'-diglucosides, 7-glucosides and 7-arabinoside. Quercetin-3-glucoside was identified as a major flavonoid component of all species studied. Quercetin-3-rhamnoside and quercetin-3,4'-diglucoside were isolated from G. longissima and G. blotiana. In addition, G. blotiana accumulates kaempferol-3-methyl ether-7-arabinoside, rhamnocitrin-3-glucoside and kaempferide-7-arabinoside. Kaempferol-3-rutinoside and kaempferol7-glucoside were found in all species except G. hirta, which does not appear to accumulate the kaempferol derivative. The glucosides were observed as minor constituents. Two compounds which are generally rare in ferns, orientin and 46 AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 1 (2003) vitexin, occur in G. microphylla. Previously, Wallace and coworkers (Amer. J. Bot. 70:207-211. 1983) studied the species of Gleichenia from Hawaii and found different flavonoid patterns. They found quercetin-3-rutinoside, quercetin-3-glucoside and kaempferol-3-glucoside, but they found kaempferol3-rutinoside as well. Furthermore, quercetin-3-rutinoside was identified as a major flavonoid component of all species except G. intermedia, Dicranopteris pectinata and Sticherus cunninghamii (it was a minor component in the latter). Quercetin-3-glucoside and kaempferol-3-glucoside were observed as minor constituents of the two species studied. Thus, our findings are not consistent with the flavonoid profiles of the species analyzed by Wallace and co-workers (Amer. J. Bot. 70:207-211. 1983). From a chemotaxonomic viewpoint, the occurrence of kaempferol and quercetin in all species indicates a close relationship among them. However, the presence of acacetin-7glucoside, vitexin and orientin in G. microphylla is of interest, since these compounds have not been found in this family before. The authors thank Universiti Putra Malaysia for financial support.— UMIKALSOM YUSUF, Department of Biology, Universiti Putra Malaysia, Serdang, 43400, Malaysia, KHAIRUDDIN ITAM, Institute of Bioscience, University of Malaya, 50603, Kuala Lumpur, FARIDAH ABDULLAH, IZANA ZAINAL, Department of Biology, Universiti Putra Malaysia, Serdang, 43400, Malaysia and MOHD. ASPOLLAH SUKARI, Department of Chemistry, Universiti Putra Malaysia, Serdang, 43400, Malaysia. The Cycads, by Loran M. Whitelock. 2002. Timber Press, Portland, Oregon. Hardcover [ISBN 0-88192-522-5]. 374 pp. $39.95. It seems likely that anyone with an interest in the ferns and the so called fern allies would also harbor an interest or potential interest in the cycads. There is something about these plants that tug at those same intellectual strings. Perhaps it is their antiquity or their fern-like foliage, or simply it is their underdog status—after all everyone knows that the ferns and cycads have already had their time in the sun and that they are just waiting for the door to close behind them. Whatever the reason, The Cycads, is a book that you will enjoy. This large format, coffee table sized book is impressive, from its magnificent cover photo to 505 color plates and numerous line drawings. Obviously designed for the cycad gardner or horticulturalist, The Cycads also has a home in the library of any pteridologist or morphologist. The book begins with several light chapters on cycad distribution and classification. These chapters are easy to read and while not precisely exhaustive nor entirely reflective of some of our newest concepts, they are informative at an avocational level. Chapter 3 provides a simple, brief overview of the plant body and reproductive structures and closes with a section oh hybridization within the group. Chapters 4 and 5 discuss cultivation and propagation and chapter six discusses conservation. Chapter 8 is a brief overview of cycad ethnobotany and is supplemented nicely with a number of very nice color plates. The majority of the book is dedicated to generic and specific treatments. Each species account gives an in depth description of the organism as well as statements of native habitat and distribution. The strength of these accounts certainly lies in the paragraphs that follow as they supplement earlier discussions on cultivation, morphological variation, conservation status, and a number of varied aspects of the individual species. These treatments are filled with information that has come about through years of experience and study of this amazing assemblage of plants. The Cycads culminates with a number of helpful appendices dealing with various cultivation aspects of the There are few failings with this book and those that I did find are likely best interpreted to my own idiosyncratic desires for a book of this type. I was disappointed not to find a key to genera and species. This book, with so much accumulated data, would certainly have benefited a wider botanical audience with some identification aid. A second aspect that left me wanting was the lack of explicit literature citation within the body of the text. A reader interested in say the pollination biology of the cycads must search a lengthy, 9+ pages of bibliography in hopes to find an appropriate reference.—R. JAMES HICKEY, Botany Department, Miami University, Oxford, OH 45056. infill i III 1753 00310 5035 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 prepared may be returned for revision prior to review. Submit manuscripts in triplicate (xerocopies acceptable), including review copies oi illustrations and originals of illustrations. After review, submission of final versions of manuscripts on diskette (in PC- or Mac-compatible formats) is strongly encouraged. Use standard 8V2 by 11 inch paper of good quality, not "erasable" paper. Double space manuscripts throughout, including title, authors' names and addresses, short, informative abstract, text (including heads and keys), literature cited, tables (separate from text), and figure captions (grouped as consecutive paragraphs separate from figures). Arrange parts of manuscript in order just given. Include author's name and page number in upper right corner of every sheet. Provide margins of at least 25 mm all around on typed pages. Do not submit right-justified copy, avoid footnotes, and do not break words at ends of lines. Make table headings and figure captions self-explanatory. Use S.I. (metric) units for all measures (e.g., distance, elevation, weight) unless quoted or cited from another source (e.g., specimen citations). For nomenclatural matter (i.e., synonymy and typification), use one paragraph per basionym (see Regnum Veg. 58:39-40. 1968). Abbreviate titles of serial publications according to Botanico-Periodicum-Huntianum (Lawrence et al., 1968, Hunt Botanical Library, Pittsburgh) and its supplement (1991). References cited only as part of nomenclatural matter are not included in literature cited. For shorter notes and reviews, omit the abstract and put all references parenthetically in text. Use Index Herbariorum (Regnum Veg. 120:1-693. 1990) for designations of herIllustrations should be proportioned to fit page width with caption on the same page. Provide margins of at least 25 mm on all illustrations. For continuous-tone illustrations, design originals for reproduction without reduction or by uniform amount. In composite blocks, abut edges of adjacent photographs. Avoid combining continuous-tone and line-copy in single illustrations or blocks. Coordinate sequence and numbering of figures (and of tables) with order of citation in text. Explain scales and symbols in figures themselves, not in captions. Include a scale and reference to latitude and longitude in each map. Proofs and reprint order forms are sent to authors by the printer. Authors should send corrected proofs to the editor and reprint orders to the printer. Authors will be assessed charges for extensive alterations made after type has been set. For other matter of form or style, consult recent issues of American Fern Journal and The Chicago Manual of Style, 14th ed. (1993, Univ. Chicago Press, Chicago). Occasionally, departure from these guidelines may be justified. Authors are encouraged to consult the editor for assistance with any aspect of manuscript preparation. Papers longer than 32 printed pages may be sent to the Editor of Pteridologia (Memoir Editor, see cover 2). PTERIDOLOGIA ISSUES IN PRINT The following issues of Pteridologia, the memoir series of the American Fern Society, are available for purchase: i Western North 2A. Lellinger, David B. 1989. The Ferns and Fern-allies of Costa Rica, Panama, and the Choco (Part 1: Psilotaceae through Dicksoniaceae). 364 pp. $32.00 postpaid. Send your order with a check or money order to: American Fern Society, Inc., c/o U.S. National Herbarium MRC-166, Smithsonian Institution, Washington, DC 20560. AMERICAN FERN JOURNAL ON MICROFICHE Volumes 1-61 of the American Fern Journal are available as archival quality, silver positive microfiches. Single volumes or the entire run may be purchased. The fiches are easily read with 10X or greater magnification (using a dissecting microscope and transmitted illumination or a fiche reader). Silver negative microfiches of vols. 1-50 are also available. The price is $4.00 per volume or $244.00 per set of 61 volumes, postpaid. Send your inquiry or order with a check or money order to: American Fern Society, Inc., c/o Dr. James D. Montgomery, Ecology III, Inc., R.D. 1, Box 1795, Berwick, PA 18603. VISIT THE AMERICAN FERN SOCIETY'S WORLD WIDE WEB HOMEPAGE: http ://w ww.amerfernsoc.org/ AMERICAN FERN JOURNAL April-June 2003 QUARTERLY JOURNAL OF THE AMERICAN FERN SOCIETY Moth Larvae-damaged Giant Secondary Colonization by Ants Klaus Mehltreter, Paricia Rojas and Monica Palacios-Ru The Effects of pH, Temperature, Light Intensity, Light Quality, and Moisture Levels t Spore Germination in Cheilanthes feei of Southeast Missouri Sarah L. Nondorf, Melissa Dooley, Maria Palmieri and Lucinda J. Swatzt Germination of Fern Spores in Natural Soils Wen-Hsiung I New Species in Adiantum from Brazil Jefferson Prac New Species and New Lectotypification of Sei The American Fern Society Council for 2003 CHRISTOPHER H. HAUFLER, Dept. of Botany, University of Kansas, Lawrence, KS 66045-2C TOM RANKER, University Museum, Campus Box 265, University of Colorado, Boulde W. CARL TAYLOR, 800 W. Wells St., Milwaukee Public Museum, Milwaukee, WI 53233-1478. JAMES D. CAPONETTI, Dept. of Botany, University of Tennessee, Knoxville, TN 37916-1110. GEORGE YATSKIEVYCH. Missouri Botanical Garden, P. O. Box 299, St. Louis, MO 63 166-0299. Membership Secretary JAMES D. MONTGOMERY, Ecology HI, 804 Salem Blvd, Berwick, PA 18603-9801. Back Issues Curator R. JAMES HICKEY. Botany I * -lord, OH 45056. Journal Editor DAVID B. LELLINGER. I S \... , ... H. ha i MKC-166, 800 W. College Ave., St. Peter, MN 56082-1498. Bulletin Editor American Fern Journal R. JAMES HICKEY EDITOR Botany Department,Miami University, Oxford, OH 45056, ph. (513) 529-6000, e-mail: hickeyrj@muohio.edu ASSOCIATE EDITORS GERALD J. GASTONY Dept. of Biology, Indiana University, Bloomington, IN 47405-6801 GARY K.GREER Biolog) D ndale, MI 49401 CHRISTOPHER H. HAUFLER .... Dept. of Botany, Univereitj oi Kansas. Lawrence, KS 66045-2106 ROBBIN C. MORAN New York Botanical Garden, Bronx. NY 10458-5126 JAMES H. PECK Dept. of Bio .s—Little Rock. 2801 S. University Ave., Little Rock, AR 72204 The -'American Fern Journal" (ISSN 0002-8444) is an illustrated quarterly devoted to the general stud) of terns, It is owned bj tl ::jn Fern Society. % Missouri Botanical Garden, P. O. Box 299, St. Louis, MO 63166-0299. Periodicals postage paid at and orders for > _y III, 804 Salem General inquiries concerning terns should he addressed to the Secretary. Subscriptions $35.00 to U.S.A., Canada, and Mexico; $45.00 to elsewhere in the world (-$2.00 agency fee, if app M| dues, $25.00 + $7.00 mailing surcharge beyond U.S.A.. Canada, and Mexico; life membership. $300.00 + $140.00 ge beyond U.S.A., Canada, and Mexico). Back Issues Curator foi prices and a\ailabilit> POSTMASTER: Send addivBox 299, St. Louis. MO 63166-0299. The editor of the Bulletin of the American Fern Societv SPORE EXCHANGE eal Garden, P. O. MISSOURI BOTANICAL JUN 0 4 2003 GARDEN LIBRARY Moth Larvae-damaged Giant Leather-fern Acrostichum danaeifolium as Host for Secondary Colonization by Ants MONICA PALACIOS-RIOS i Vegetal, Instituto de Ecologfa. A.C., ABSTRACT.—Leaves of the giant leather-fern, Acrostichum danaeifolium, were infested by larvae of an unknown species of moth (microlepidoptera) at a mangrove site on the Gulf of Mexico. During a nine-month observation period these moths infested 87% of the ferns and 41% of their leaves. The damage caused by the moth larvae consisted of galleries bored into the petioles and rachis; however, this did not affect maximum leaf size. The galleries form a microhabitat that later can be colonized by ants. Among ten ant species found, two introduced tramp species, Tapinoma sessile and Wasmannia auropunctata were the most common ones. Because it does not produce domatia or extrafloral nectaries to attract ants directly, the giant leather-fern becomes an involuntary myrmecophyte by harboring ants in the moth-constructed galleries. Interactions between ferns and insects, especially ants, are relatively rare. Ferns do not rely on pollinators and have only few spore dispersers (Tryon, 1985). Thus, interactions with insects are restricted mostly to herbivory (Auerbach and Hendrix, 1980; Hendrix, 1980; Cooper-Driver, 1985). Several herbivores (Balick et al., 1978), and one ant species, Azteca traili subsp. filicis Forel (Gomez, 1974, 1977), have been reported to be specific to ferns. However, in very few cases are the herbivores and ants living within the fern plant. Ferns that offer hollow rhizomes for a symbiotic coexistence with ants (= domatia) are described from two genera: Solanopteris and Lecanopteris. The best-known neotropical ant-fern interaction is described for the epiphytic Solanopteris brunei (H. Christ) Wagner, which is distributed from Costa Rica to Colombia (Tryon and Tryon, 1982). It possesses hollow tubers on the lateral branches of the creeping rhizome (Wagner, 1972). Six ant species inhabit the tubers (Gomez, 1974, 1977). In the paleotropics ant colonies live within the stems of Lecanopteris species (Jermy and Walker, 1975; Walker, 1986; Gay, 1991,1993a, b). Holttum (1977) reported the invariable presence of ants in L. carnosa in Malaysia. It is possible that the ferns benefit from the higher C02 concentration and the mineral supply because these develop part of their roots inside the rhizomes. These roots may take up minerals (i.e. nitrogen) from the accumulated matter and the excreta of the ants. Other ant-fern interactions are related to the possession of extrafloral nectaries, as in some species of 50 AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 2 (2003) Drynaria (Jolivet, 1996), Polypodium (Koptur et al., 1982; Rico-Gray et al, 1998) and Pteridium (Heads and Lawton, 1984). Myrmecophytes are frequent in mangroves and flooded river areas (Jolivet, 1996), the typical habitat of the giant leather-fern Acrostichum danaeifolium Langsd. & Fisch. This fern species possesses no extrafloral nectaries or domatia. However, we observed ants living within the leaf petioles and rachis of this fern. The galleries colonized by the ants seemed to have been excavated by some other herbivore. The objective of our study was to investigate the origin, frequency and seasonality of the galleries in a natural population of the giant leather-fern, the damage caused by the herbivore activity, and the occurrence of ants inhabiting this microhabitat. MATERIALS AND METHODS The study was carried out in the understory of the black mangrove Avicennia germinans (L.) Stearn (Avicenniaceae) of the Biological Station of La Mancha (19°36'30" N, 96°22'40" W), Veracruz, Mexico, within 230 m of a brackish-water lagoon. Normal climatic conditions at this site are hot and humid, with a dry season from November to April, when mean precipitation is less than 45 mm per month. Mean annual temperature for the last 20 years was 24.4°C and the mean annual precipitation measured 1198 mm. We tagged 30 plants of A. danaeifolium and recorded all new leaves produced each month from November 2000 to July 2001. Leaf length of each leaf was measured with a flexible metric tape each month until it reached its maximum length. The occurrence of holes and galleries was recorded. From these data we calculated the monthly leaf production and the mean herbivore infestation rate of the plant population. Temporal changes in leaf production and leaf infestation were analyzed with a repeated measure ANOVA on ranks (SigmaStat 1995). A Mann-Whitney test was used to compare leaf production of infested and not infested plants. A paired t-test was performed to compare the individual means of the maximum leaf length of infested and undamaged leaves. Leaves heavily damaged, as a consequence of the activity of other herbivores or fungi, were excluded from these data. Additionally, each month we collected 20 infested leaves from 20 different and arbitrarily selected plants to identify all ant species living in the rachis and petioles of the fern leaves and to determine their frequency. Invertebrate taxa were identified by the second author and three ant species were identified by Dr. W. P. MacKay of the University of Texas at El Paso. All collections were deposited at the Departamento de Biologia de Suelos, of the Instituto de Ecologia, A.C. (BSIE). Voucher specimens of the giant leather-fern (PalaciosRios 3883-3889) are deposited at the herbarium of the Instituto de Ecologia, A.C. in Xalapa (XAL), and were identified by the first and third author. Only one to three months old leaves of the giant leather-fern Acrostichum danaeifolium showed recent damage by a xylophagous microlepidoptera r AL.: ANTS AND ACRO.s //< III M - Mean leaf productioi FIG. 1. Monthly production o ("moth") larva. The moth larva produced galleries and tunnels in the petioles and rachis, often with some excavated material adhering to the exit holes. Each leaf contained one to several larvae or pupae (K. Mehltreter, pers. obs.). After two to four months, the moths emerged as adults and left the leaves through holes, leaving the microhabitat available for secondary colonization by ants. During the nine-month study period, the moth larvae infested one or more leaves of 26 of the 30 plants (87%). The proportion of infested versus damaged leaves varied considerably between different plants. While four plants had no damage at all, two plants had all new leaves infested. Leaf production of infested plants did not differ from not infested plants (t = 33.5, P>0.05). The plant population produced 244 new leaves during the < servation period. Of these, 41% were infested, 9% were damaged by other herbivores or fungi, and 50% were undamaged. The maximum size of infested and undamaged leaves was not significantly different (t = 1.46, df = 23, P > indicating that leaf damage may not have been detrimental to the p Newly infested leaves were observed during the entire study period (Fig. 1), which indicates the continuous presence of the adult moths and moth Ian The monthly mean infestation did not vary significantly (x2 = 11.8, P > O.i but the monthly mean leaf production did (x2 = 59.2, P < 0.001). Consequen relative infestation rates of new leaves were highest in February (73%), during the dry season when leaf production was low. Ten ant species, seven native and three introduced species (Table 1), in- 52 AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 2 (2003) TABLE 1. Origin, nesting habits and distribution of ant species, living inside Acrostichum Subfamily and species Tapinoma sessile (Say) ( = MpInrlmptruin) sp. auropunctata (Roger) habited 29% of the infested leaves [N = 180). Some leaves were occupied by two or three ant species simultaneously. In most cases we found complete ant colonies varying from a few dozen to up to several hundred individuals. Colonies consisted of eggs, larvae, pupae, workers, winged males, and one to several dealate queens. Six species belong to the subfamily Myrmicinae, and two species to the subfamilies Formicinae and Dolichoderinae. Most of these species are known to have arboreal nesting habits. Two species dominated as inhabitants of the galleries (Fig. 2). Tapinoma sessile (Say) was present in all monthly samples, and Wasmannia auropunctata (Roger) was present in the samples from April to July during the rainy season. Both are exotic tramp species with variable nesting habits. Casual use of these galleries by other invertebrates was also noted: Acarina, Collembola, Diptera [Corynoptera sp., Sciaridae), diplopods, enchytraeids (Oligochaetes), isopods, nematodes and oothecae of cockroaches (Blatellidae). All seemed to colonize the galleries independently or together with ants, especially on older leaves. Of all these other inhabitants, isopods were the most frequently observed. KKETAL.: ANTS AND AC RO XL host-plant specific than most macrolepidoptera. Insects associated with some common fern species show a great degree of specialization (Cooper-Driver, 1985). Therefore, the moth that we studied could be a new species with a specific relationship to A. danaeifolium. It would be very interesting to check the other two fern species of the same genus, Acrostichum aureum L. with pantropical distribution and Acrostichum speciosum Willd. from the paleotropics, for similar herbivores and secondary inhabitants. As we observed leaf infestations during the entire study period, the adult moths seemed to be continuously present. Thus, new galleries were available at all times for secondary colonization by ants. The moth larvae only infested young leaves, which have softer, developing tissues, and therefore may possess a lower degree of chemical defense mechanisms. After four months the fertile leaves died, and after ten months the sterile leaves died (Mehltreter and Palacios-Rios, 2003). The dead leaves withered completely and finally become stunted. The ant colonies moved to another younger leaf of the same or another plant. Subunits of colonies of T. sessile changed every 12.9 days from one site to another (Smallwood, 1982). The two species, T. sessile and W. auropunctata, are exotic tramp species, widely distributed by human activities. They are very adaptable, opportunistic species of temporary, fragmented, species-poor habitats with diverse nesting habits (Clark et al., 1982; Deyrup et al., 2000). Their colonies can be divided into subunits, which occupy different sites and may interchange individuals, as they have several fertile queens (Holldobler and Wilson, 1990). If one plant of A. danaeifolium offers several moth larvae-infested leaves with new galleries, it might be that these are occupied by subunits of the same ant colony. 54 AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 2 (2003) Although beneficial effects for the giant leather-fern were not observed, we cannot exclude this possibility. Whereas the ant W. auropunctata defends the nectar-producing ginger Costus woodsonii Maas (Zingiberaceae) against a seed predator, the dipteran Euxesta sp. (Schemske, 1980), no aggressive or defensive behavior was observed on any A. danaeifolium by colonizing ants. We conclude that the microhabitat of the galleries may be considered to be opportunistic domatia, because the plant does not produce them, and the ants do not build them. Consequently, the giant leather-fern can be considered as an involuntary myrmecophyte, where the ants find only shelter after its leaves are infested by the moth larvae. The microhabitat of the galleries serves as an additional or alternative niche in the mangroves, and could be of importance for two introduced ant species that are not reported for other habitats nearby. This may be the consequence of strong competition with native ant species due to limited amounts of available microhabitats. We thank the staff of the Biological Station of La Mancha for logistical support, Sandra Cardoner for help during fieldwork, Dr. Sergio Ibanez for the dete ids, and Dr. W. P. MacKay for the identification of some ant species. Special thanks to Dr. Ludwig Miiller, Dr. Theresa L. Pitts-Singer and Dr. James P. Pitts and an anonymous reviewer for the revision and constructive comments on the manuscript. Fieldwork was supported by the Institute de Ecologia, A.C., 902-16 and 902-14. LITERATURE CITED M. and S. D. HENDRIX. 1980. Insect-fe species-area association. Ecol. Entomol. 5 M. J., D. G. FURTH and G. COOPER-DRIVE arthropod predation on ferns. Oecologia 35:55-89. BECKER, V. O. 2000. Microlepidoptera. Pp. 453^68 in J. E. Llorente Bousquets, E. Gonzalez Soriano, and N. Papavero, eds. Biodiversidad, taxonomi'a y biogeografi'a de artrdpodos de Mexico: Hacia una suit ito. Vol. II, UNAM, Mexico. CLARK, D. B., C. GUAYASMIN, O. PAZMINO, C. DONOSO and Y. PAEZ DE VILLACIS. 1982. The tramp ant Wasmannia auropunctata: autoecology and effects on ant diversity and distribution on Santa Cruz Island, Galapagos. Biotropica 14:196-207. COOPER-DRIVER, G. A. 1985. The distribution of insects on ferns. Amer. J. Bot. 72:921. DEYRUP, M., L. DAVIS and S. COVER. 2000. Exotic ants in Florida. Trans. Amer. Entomol. Soc. 126:293-326. GAY, H. 1991. Ant-houses in the fern genus Lecanopteris Reinw. (Polypodiaceae): the rhizome morphology and architecture of L. sarcopus Teijsm. & Binnend. and L. darnaedii Hennipman. AI LKHACII. BALICK, H. 1993b. Rhizome structure and evolution in the ant-associated epiphytic fern Lecanopteris Reinw. (Polypodiaceae). Bot. J. Linn. Soc. 113:135-160. L. D. 1974. Biology of the potato-fern Solanopteris brunei. Brenesia 4:37-61. L. D. 1977. The Azteca ants of Solanopteris brunei. Amer. Fern J. 67:31. HEADS, P. A. and J. H. LAWTON. 1984. Bracken, ants and extrafloral nectaries. II. The effect of ants GAY, GOMEZ, GOMEZ, MEHLTRETER ET AL.: ANTS AND ACROSTICHUM 55 S. D. 1980. An evolutionary and ecological perspective of the insect fauna of ferns. Amer. Naturalist 115:171-196. HOLLDOBLER, B. and E. O. WILSON. 1990. The Ants. Belknap Press, Harvard University Press, Cambridge. 732 pp. HOLTTUM, R. E. 1977. Plant life in Malaya, 2nd ed., 3rd impr. Longman, Malaysia, Singapore. 254 pp. JERMY, A. C. and T. G. WALKER. 1975. Lecanopteris spinosa. A new ant-fern from Indonesia. Fern Gaz. 11:165-176. JOLIVET, P. 1996. Ants and Plants. Backhuys Publishers, Leiden. 303 pp. KOPTUR, S., A., R. SMITH and I. BAKER. 1982. Nectaries in some neotropical species of Polypodium (Polypodiaceae): Preliminary observations and analyses. Biotropica 14:108-113. MEHLTRETER, K. and M. PALACIOS-RIOS. 2003. Phenological studies of Acrostichum danaeifolium (Pteridaceae, Pteridophyta) at a mangrove site on the Gulf of Mexico. J. Trop. Ecol. 19:155HENDRIX, V., J. G. GARCIA-FRANCO, M. PALACIOS-RIOS, C. DI'AZ-CASTELAZO, V. PARRA-TABLA and J. A. 1998. Geographical and seasonal variation in the richness of ant-plant t t in Mexico. Biotropica 30:190-200. D. W. 1980. The evolutionary significance of extrafloral nectar production by Costus woodsonii (Zingiberaceae) an experimental analysis of ant protection. }. Ecol. 68:959-967. SIGMASTAT. 1995. Sigmastat, statistical software, version 2. 0. Jandel Scientific Software, San Rafael. SMALLWOOD, J. 1982. Nest relocations in ants. Insectes Sociaux 29:138-147. TRYON, A. F. 1985. Spores of myrmecophytic ferns. Proc. Roy. Soc. Edinb. 86B:105-110. TRYON, R. M. and A. F. TRYON. 1982. Ferns and Allied Plants with Special Reference to Tropical America. Springer, New York. 857 pp. WAGNER, W. H. 1972. Solanopteris hrunei, a little-known fern epiphyte with dimorphic stems. RICO-GRAY, NAVARRO. SCHEMSKE, WALKER, T. G. 1986. The ant-fern Lecanopteris mirabilis. Kew Bull. 41:533-45. The Effects of pH, Temperature, Light Intensity, Light Quality, and Moisture Levels on Spore Germination in Cheilanthes feei of Southeast Missouri SARAH L. NONDORF, MELISSA A. DOOLEY, MARIA PALMIERI, and LUCINDA J. SWATZELL Department of Biology, Southeast Missouri State University, Cape Girardeau, MO 63701 Oheilanthes feei is a xerophytic fern that is broadly distributed throughout the United of the Mississippi. Although it has a broad distribution, it occupies a very narrow 1.0 m from the top of the bluffs. The physiological basis for the fern's restriction to this nment is unclear. In this study, C. feei spores were subjected to a broad range of s, pH, and light intensities, to varied light qualities, and to different moisture levels, cate that C. feei spores can germinate under a wide variety of conditions. However, that spore germination optima and optimal conditions for protonemal growth overlap he disparity in optimum conditions may be a partial basis for the broad distribution niche of C. feei. Cheilanthes feei Moore is a common fern that is widely distributed in North America. Its range extends primarily from southwestern Canada, south to north central Mexico, and east to the Mississippi and Ohio River valleys of Midwestern United States (Mickel, 1979). Although C. feei is common, it is unusual in several ways. First, C. feei is a xerophytic fern. This is somewhat of an oxymoron, since fern gametes are typically free-swimming and most ferns are restricted to moist environments. However, in C. feei and many cheilanthoid ferns, the need for water for reproduction is circumvented by apogamy. Secondly, although C. feei is widely distributed, it occupies a very narrow niche. In southeast Missouri, C. feei typically grows in crevices of limestone bluffs, typically facing south in full sun and approximately 0.5-1.0 m below bluff tops. The basis for this habitat restriction is unclear. However, there are many feasible explanations. It is possible that the fern cannot compete with more vigorous species in mesophytic habitats, but can survive in more xeric environments. Cheilanthes feei, like other Cheilanthes species, is characterized by some adaptations that can reduce water loss, such as numerous trichomes and a small surface area to volume ratio (Hevly, 1963; Gratani et al., 1998). This would be analogous to the saguaro cactus, Cereus giganteus, which is restricted to areas of intense sunlight, since the thick hydrodermis causes it to be light-limited (Darling, 1989). Another possible explanation for habitat restriction in Cheilanthes feei is that the narrow niche that C feei occupies may provide the optimal growth conditions for the fern, so that the incidence of C. feei in other habitats is very low. A final possibility is habitat specificity. Cheilanthes feei may be restricted to its habitat based on unique characteristics of both the fern and its environment. NONDORF ET AL.: SPORE GERMINATION IN CHFJLANTHES 57 This study addresses the physiological basis for the restriction of Cheilanthes feei to limestone bluff crevices. Optimal conditions for spore germination are often a reflection of optimal growth conditions for the entire life cycles of the ferns. Since the fern gametophyte is the most vulnerable stage in the fern life cycle, gametophyte physiology and the necessary condition ranges for growth and development are limiting for ferns. Hence, optimal conditions for gametophyte survival and development are typically congruous with optimal spore germination conditions (Raghavan, 1980; notwithstanding necessary changes in light qualities which spur developmental changes). In addition, environmental conditions that negatively affect spore germination typically reflect the physiological limitations of a fern species. An obvious example is the need for water. Most spores require the presence of water and undergo imbibition prior to germination. The gametophyte stage and the fertilization process also require at least a film of water. Previous studies demonstrate that cheilanthoid ferns, with respect to spore germination requirements, may be physiologically similar to mesophytic ferns. For example, most ferns germinate and develop best at a slightly acidic pH, at 25°C, in moist conditions, under red light (a phytochrome response), and moderate light intensity (100 ^mol-nT2-s_1) (Raghavan, 1980). With respect to pH, temperature, light quality and light intensity, these are also the optimal conditions for spore germination in several cheilanthoid species (Hevly, 1963; Raghavan, 1973). Still, these ferns have very different distributions than C. feei, and optimal spore germination conditions for C. feei may be different as well. Therefore, we examined the effects of pH, temperature, light intensity, light quality, and moisture levels on germination rates of C. feei spores. In addition, we measured the potential water content and porosity of rock substrate in C. feei habitat. MATERIALS AND METHODS PLANT COLLECTION.—Cheilanthes feei sporophylls were collected in the fall and winter of 2000 from Reis Biological Station, Steeleville, MO. To harvest spores, sporophylls were crushed using a mortar and pestle. Cheilanthes feei spores average 67.0 urn in diameter (Knobloch, 1969) and spores were separated from the plant material using a 75.0 urn brass mesh sieve and stored at 4° C in the dark. CULTURE CONDITIONS.—Although this study addresses optimal spore conditions for Cheilanthes feei, there is previously no information available on optimal culture conditions, growth medium contents or osmolality, with the exception of anecdotal information (Siegler, 2002). Therefore, a standard growth medium, Knudson's-C (C-Fern, 2001), that contains salts, iron, phosphate buffer, sugar as an osmoticum, and agar for solidification, was selected. The effects on spore germination of any of these contents, such as sucrose, on C.feei are unknown, and have varying effects on germination in other species (Raghavan, 1989; Sheffield et al., 2001). Therefore, we strove only to make them consistent through all of the treatments. All treatments then, were cul- FERN JOURNAL: VOLUME 93 NUMBER ments were 25°C, pH 5.5. continuous i Temperature Light intensi Light qualit) tured on the same growth medium lot. In addition, in the absence of information on the affects of surface sterilization on germination, we also used a standard procedure (Guiragossian-Kiss and Kiss, 1998). With the exception of the moisture level experiment, spores were surface sterilized in a 7% (v/v) commercial bleach solution with 0.1% (v/v) Triton X100 for 20 min. Spores were then rinsed in sterile ddH20, sown on a modified Knudson's-C medium (C-Fem, 2001; 3.7 mM (NH4)2S04, 4.2 mM Ca(N03)2-4 H20, 1.8 mM KH2P04, 27.0 uM FeS04-7H2G\ 17.5 mM sucrose, 10 uM H3B03, 10 uM MnS04H20, 3 uM ZNS04-7H20, 0.1 uM Na2Mo04-2H20, 0.01 uM CuS04-5H20, 0.01 uM CoCl2-6H20) with 1.2% (w/v) sucrose (35 mM) in 9-cm Petri dishes (Guiragossian-Kiss and Kiss, 1998). All plates were prepared and poured from the same Knudson's-C preparation to ensure a consistent concentration of sucrose. Dark control plates were wrapped in aluminum foil and incubated in the same conditions as other plates within the same experiment. Spores were incubated for 7 d under various experimental conditions, with the exception of the moisture experiment, during which spores were treated for 10 days. When feasible, conditions for each test (Table 1) were maintained at pH 5.5, 25°C, 100 umolm 2s_1 of white light, and saturated (agar medium). However, the parameters of some experiments required different conditions. For example, with respect to light quality, spores were incubated at 33°C and at 75 umolm 2s * to achieve maximum and consistent light intensity in each light quality. VARIABLES: PH, TEMPERATURE, LIGHT INTENSITY, AND LIGHT QUALITY.—To examine the effects of pH on spore germination, spores were sown on Knudson's-C of pH 4.5, 5.5, 6.5, and 8.5. Knudson's-C does not buffer well at pH 7.5 and this pH was not used. To test the effects of temperature on spore germination, spores were incubated at 4°C, 25°C, and 33°C. Light intensity was established by using either GE Halogen Ultra PAR 38 (90 watts) or Sylvania Halogen XTRA PAR 38 (120 watts) and by varying the distances between the light source and the spores. Light intensity was verified with a Li-Cor Quantum/Radiometer/ Photometer, model LI-185. Light intensities were: 0 (dark), 10, 50, 75, 100, 125, NONDORF ET AL.: SPORE GERMINATION IN CHEILANTHES 59 150 umol-rrr2-s \ Treatments were contained within chambers to avoid incidental light and cooled with an electric fan system. Dark treatments were prepared by wrapping Petri dishes in foil after they were inoculated and sealed with Parafilm. Light qualities tested were blue, red, white, green, far red, and dark. These light qualities were established using plexiglass filters (Cadillac Plastics, Southfield, MI). Wavelengths for light filters were measured using Data Logger Pro software and were: white (420-710 nm), blue (420-570 nm), green (500-595 nm), red (600-695 nm), far red (650-705 nm). DARK CONTROL (DARK PLUS VS. TRUE DARK).—During the surface sterilization and sowing process, spores are typically exposed to white light for approximately 45 min and this characterizes spore preparation for all experiments. Even dark treatments receive this white light prestimulus (Dark Plus). To test the affects of this white pre-stimulus, the germination rate in the standard dark control preparation (Dark Plus) was compared with the germination rate of spores that were surface sterilized and sown in 0.04 umolm 2s_1 of white light (True Dark). MOISTURE LEVELS.—To avoid introduction of additional moisture, spores were not surface sterilized and were sown on sterilized filter paper (Whatman #1, 9 cm) that were wetted with Knudson's-C (no sucrose to avoid contamination that would hinder germination, no agar). Moisture levels were: 0, 10, 20, 30, 40, and 50 ul-crrT2. Germination rates were counted at 7 d, but protonema were allowed to develop to the tenth day for observation and measurement. DATA ANALYSIS AND SAMPLING.—To ensure that spores, which germinate quickly, were counted within only a few hours of each other, but to also obtain large sampling, experiments were conducted independently. Testing all variables at one time would sacrifice the integrity of the counts. Spores were sown on 4-10 plates per treatment, depending on the parameters of the experiment. For example, only 4 plates were used in light quality experiments to ensure that all plates were placed within the center of the filters and received the same light intensity. Spores were scored as germinated or non-germinated on a haphazard basis up to 300 spores per plate, depending the number of plates allowed by the parameters of the experiment (n » 400-1200 spores per treatment). Spore germination was counted when the exine had ruptured and protonemal cell extrusion was visible. Because germination/non-germination is a binomial score, the data were transformed using the arc sine transformation method to ensure normality. An analysis of variance (a = 0.05) was then performed to determine significance of differences in results. For clarity and continuity, the transformed means and standard deviations were used in figures. POROSITY, SPECIFIC RETENTION, AND ACTUAL RETENTION.—Rock samples were obtained with special permission from Reis Biological Station (RB), Steeleville, MO, from bluffs along Big River at Mammoth Road (MR), approximately 1.5 mi south of MO Highway H, and from private land in Cedar Hill (CH), MO approximately 1.5 east of Highway 30 along Highway BB. Because collection was destructive, rock sample sets were purposely limited to 10 approximately 2.5 cm3 pieces. Two sample sets were collected from each site. One set was AMERICAN FERN JOURNAL: VOLUME 9 MO 180 JM.. S" I 80_ 40 20 pH8.5 collected from C. feei habitat and one set was collected from the same stratum, but elsewhere on the bluff where C. feei did not inhabit. Porosity was determined as n = 100[l-(Pb/Pd)], where Pb (bulk density) is defined as the original sample oven dried weight (g) divided by the saturated pre-oven dried volume (cm3) and Pd (particle density) is defined as the original sample oven dried weight (g) divided by the mineral matter volume (cm3). Specific i of the substrate was determined as the amount that the substrate a against gravity divided by the total volume (Fetter, 1988). Actual i values were subsequently determined for cm2 planes within the substrate. This was calculated as the amount of water (cm3) retained against gravity divided by the pore space available for water retention and further divided by 10 for comparison with laboratory conditions. VARIABLES: PH, TEMPERATURE, LIGHT INTENSITY, AND LIGHT QUALITY.—Germination occurred in a broad pH range (Fig. 1). Cheilanthes feei spores (n = 400-1200) germinated at the highest rate in acidic pH (pH 4.5 and 5.5). Limestone pH varies, but is basic [ca. pH 8.3; M. Aide, pers. comm.). pH 7.5 was not tested, since Knudson's-C does not buffer well at this pH. Dark controls (Dark Plus; not shown) were 5% or less in all pH. Cheilanthes feei (n = 800-1000) spores germinated optimally at 25°C (Fig. 2). Note that germination also occurs at 33°C and at 4°C. Dark controls (Dark Plus) are not shown. Dark germination rates at 25°C and 33°C were less than 5%. At 4°C in the dark (storage conditions), no germination occurs. Spores (n = 600-1200) germinated under a wide range of light intensities (Fig. 3). All germination rates are low but comparable to expected values at 33°C (25°C is difficult to maintain under the stronger light intensities). The optimal germination rate occurred at 100 umolm 2s_1 and there were significant differences between the highest germination rate (100 NONDORF ET AL.: SPORE GERMINATION IN CHEIL INTHES —I-~ 3 onis25°C. Germination in jmperatures. In the dark coi I. In the dark, C. feei germii umolm s ), other light intensities, and the dark control (Dark Plus] rate. Cheilanthes feei spores germinated under all light qualities, even in the Dark Plus controls (Fig. 4). Germination rates in different light qualities varied greatly, but not significantly. For example, there was no significant difference between white, far red and green. A notable difference was in the far-red treatment in germinated spores. All germinated spores in this treatment were at the 2-cell protonemal stage when scored. This was not observed in any other light quality treatment. Dark controls (Dark Plus) germinated at 7.7%. Preparation methods markedly affected germination rates (Fig. 5). Without the 45 min white light prestimulus (True Dark), germination occurred at levels that 30 T \ or ll * 1Q ¥ ^ O 1 5 " T I - 125 150 1 50 I 75 100 Dark I Plus J 5 - a Light Intensity in umol r Effects of light intensity on s f JOURNAL: VOLUME 93 NUMBER 2 (2003) r 100 I T 1 ' ! White 60 a,Ue Red C.en Far Red Dark Plus -h riable. No significant <]itten>n<.e exists spores germinated in far red light were i any other light quality treata : 590; True Dark: MOISTURE LEVELS.—Although Cheilanthes feei spores germinated in the dark on dry filter paper, germination rates were optimized in the light at 20-50 ul-cm-2 (Fig. 6). There was no significant difference between these light treatments, but there was a significant decrease in germination in the dark controls between 20-50 ul-cm 2 as moisture increased. In addition, there was a substantial difference in protonematal presence and maturity. For reference, 20 ul-cm-2 will support mildew growth and is merely damp to the touch. Microscopically, no liquid stands between the fibers of the filter paper. At 30 ul-cm-2, a film of water coats the fibers. Above 40 ul-cm-2, the filter paper is saturated and water stands between the fibers. Protonema that germinated in 20 ul-cm"2 were approximately 200 um in length when scored and exhibited planar growth, but protonema in 40 and 50 ul-cm-2 were only 100 um in length and still filamentous (Fig. 7). POROSITY, SPECIFIC RETENTION, AND ACTUAL RETENTION.—Reis Biological Station (RB) is farthest from the St. Louis, MO metropolitan area (approximately 66 mi) and is a protected area for biological studies. Mammoth Road (MR) is a rural site (approximately 30 mi from the St. Louis metropolitan area) that overlooks a boat ramp and fishing area. The area is worn by foot traffic. Cedar Hill (CH) is private rural land that is approximately 16 mi outside of the : SPORE GERMINATION I T Dark Dark Plus 1 1 T Dark Treatments mulus on dark germination i Dark Plus) strongly affects the germination rates of dark germination in True Dark scores between 80-100%. metropolitan area and the closest of the sites to the city (St. Louis, MO Average porosity and specific retention increased with distance from the city. Porosity is the amount of pore space in a rock sample compared to the total volume, and is expressed as a percentage (Fig. 8A). Porosity means 120 x I r^i r ran X. hi k_J L_ Moisture Levels in ul«cm to decrease gradually and there is a significant < 2 significant k Plus; gray) at 20-50 ul-cm ! between 20-50 ul cm -. N AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 2 (; (B). Protonemata grown 1. (A). Mean porosities at t moth Road (white), and Reis Binlw-ii ,il an (gray). There was no significant difference between areas within the same stratum that mthesfeei and areas in which C. feei was absent. (B). Specific retention increased porosity and with distance from the metropolitan area. Conversely, variation in specific tion decreases with distance. (C). Actual amount retained in ul-cm 2 was calculated as the ributed equally throughout the pore space and mathematical!]) •plarn- NONDORF ET AL.: £ 20% . IN CHEILANTHES lili-iim C. feei Present C. feei Absent W I• 10% J I£ 6% 1- lIBVi Jm: wl] 66 AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 2 (2003) increased with distance from the St. Louis metropolitan area. There was a significant difference between mean porosity at Reis Biological Station and that at Cedar Hill. However, there was no significant difference between Mammoth Road porosities and porosities measured from samples at Reis Biological Station and Cedar Hill. In addition, there were no differences in the porosities between the areas that contained C. feei and those that did not. Still, there was substantial variation in the data, so that the standard deviations neared or exceeded the mean. This variation was consistently smaller at RB, a protected site. Specific retention values, amounts of water retained against gravity as a percentage of the total volume of the sample, mirrored results for porosity (Fig 8B). There was no significant difference between MR and RB or CH, but a significant difference between CH and RB. Variation in samples was also large and there were no significant differences between specific retentions measured in samples taken from where C. feei was present and where it was absent. Finally, there were no significant differences between actual amounts retained (Fig. 8C). This applied to comparisons between sites and to comparisons between samples taken from where C. feei was present and where it was absent. Actual amounts retained were determined by the amount of water held against gravity divided by actual pore space. These means fell between 20-30 ul-cirT2. The few exceptions, such as at Cedar Hill where C. feei was present, did not vary significantly from the mean. Overall, the average actual retentions (not shown) for all sites was 24.8 ul-cm2. The average actual retention of samples taken from where C. feei was present was 23.9 (ilcm-2 and 25.7 ul-cirT2 from where it was absent. DISCUSSION PARAMETERS FOR CHEILANTHES FEEI DISTRIBUTION.—Based on spore requirements, Cheilanthes feei has the potential to occupy a broad array of environments. There is no particular restriction to any one condition and germination itself is highly variable. These data suggest that C. feei is extremely versatile. First, substrate pH is nonrestrictive. Cheilanthes feei spores germinated in each pH range tested (Fig. 1). Slightly acidic pH promoted slightly better germination rates than at basic pH. Another variable, temperature, also failed to substantially affect germination (Fig. 2). Admittedly, higher temperature, 33°C, inhibited spore germination, but over 35% of the spores still germinated. Furthermore, spores germinated in the cold (under continuous white fluorescent light), although the germination rate was markedly less than at 25°C. Still, C. feei spores do not germinate in the dark at 4°C. A rise in temperature appears to be important in germination, from dark, cold storage to warmer conditions, but the basis for this is unclear. A rise in temperature may promote the expression of hormones prior to germination or the addition of light could increase sensitivity to hormones present (Davies, 1995). Membrane integrity in the cold may also be compromised and inhibit germination (Cuming, 1999), but spores maintain long term viability in storage NONDORF ET AL.: SPORE GERMINATION IN CHEILANTHES 67 and cold leakage is an unlikely issue. Third, light intensity and light quality made little difference overall in spore germination. Optimal conditions for light intensity and light quality are evident. In continuous white light, 100 umolm 2s~\ they germinate well (Fig. 3), but spores still germinate in all light intensities and in dark with a white light prestimulus (dark plus). The optimal condition for light quality is actually in darkness (Fig. 4-5), with no light prestimulus (true dark). Mean rates for germination in True Dark (Fig. 5) treatments were no higher than in white light, but the variation in samples was notably reduced. Finally, C. feei spores "imbibe" and germinate with no visible source of moisture available and in moist or saturated conditions. Clearly, germination is best when the substrate is moistened or saturated (Fig. 6). However, fern spore germination on dry medium is noteworthy. Although negligible, relative humidity and/or unobservable moisture present on the exine were potential sources of moisture. These spores apparently possess the capacity to uptake water for germination with only high relative humidity as a moisture source. Still, these conditions are present within storage, but imbibition does not occur. Once again, a temperature rise is likely required prior to germination. This is followed by imbibition and germination, optimally following burial beneath debris in limestone crevices. Taken as a whole, these data on spore germination rates in various conditions indicate that Cheilanthes feei spores are neither bound by the inability to compete (with regard to germination) in alternative habitats, nor by the inability to survive in mesic habitats due to morphological or physiological adaptations, nor by a requirement for optimal growth conditions. They can germinate under a wide range of conditions and only require a rise in temperature. Although these spores exhibit nearly 100% germination in certain conditions, they germinate adequately under most conditions. Therefore, there is little with respect to spore germination that explains the narrow niche of this fern, only its broad distribution. RESTRICTION OF CHEILANTHES FEEI TO ITS NARROW NICHE.—One remaining explanation for the narrow niche of Cheilanthes feei in southeast Missouri lies in habitat specificity due to substrate moisture level and protonemal moisture requirements. Cheilanthes feei spores can germinate in most moisture levels and do well in saturated conditions, but protonema do poorly in saturated conditions. There is, then, a narrow range of conditions in which C. feei can germinate and development optimally. Cheilanthes feei spores germinated optimally (80-100%; Figs. 1-5) without a light stimulus, at 25°C and pH 5.5, and when moisture levels were between 20-50 ul-cm 2 (Fig. 5). Although spores germinated well between 20-50 |ilcm~2, data from the moisture level experiment on viability of germinated spores and protonema reveal that protonema develop farther in lower moisture levels (20-30 ul-nT2; a film of water coats the substrate fibers) than protonema in cultures with higher moisture levels (40-50 (il-m-2; water stands between substrate fibers). Protonema in lower moisture levels were 200% larger than those in greater 68 AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 2 (2003) moisture levels (Fig. 7). This may be the result of a disparity in germination time or in protonemal vigor. SUMMARY OF GROWTH REQUIREMENTS.—Optimal conditions for C. feei spore germination and subsequent protonemal development may be summarized as shade or complete burial, moderate temperature, in any pH, but with only 20-30 ulm 2 throughout the germination and protonemal stages. The first three requirements are broad and can be fulfilled in many habitats. The latter is the more difficult to secure and is the restricting factor. POROSITY AND MOISTURE RETENTION ARE RESTRICTING FACTORS.—Based on moisture requirements, Cheilanthes feei can only occupy environment types that offer a narrow margin of moisture conditions for germinated spores and growing protonema (20-30 ulm 2). Sedimentary substrates offer a consistent amount of moisture and air space. The amount of moisture retained depends on porosity and specific retention. Porosity is defined as the percentage of sedimentary rock that is actually pore space. The primary determiner of porosity is weathering. Weathering can be induced chemically from reactions within the rock components or from reactions between rock components and pollution. Weathering can also be induced mechanically by wind, rain, ice, etc. Within the Emminence-Potosi Dolomite formation in southeast Missouri, mean porosity (Fig. 8A) increased and variability, which was substantial, decreased with distance from a metropolitan area (St. Louis, MO). Specific retention, the amount of water retained against gravity and expressed as a percentage of the total rock volume, also increased with distance (Fig. 8B). However, variation decreased slightly with distance from the city. Given that the chemical composition of the substrate is relatively consistent, the amount of weathering, possibly pollution-induced chemical weathering, altered the porosity and specific retention. The important consideration for C. feei, however, is not necessarily the porosity or specific retention, but the amount that the rock substrates actually retain within the available pore space. Pore space, concretion, and subsequent blockage of pores are unique for each site during the weathering process, so that distance from a pollutant source would affect the degree of porosity, specific retention, and variability between samples, but result in a mean actual retention that is or is not adequate to support C. feei colonization. In samples taken from C. feei habitat, the actual amount of water retained was mathematically distributed throughout the entire pore space and rendered within one plane as ulcnT2. The means for all three sites ranged between 20-30 ul-cnT2 (Fig. 8C). The actual retained amounts were achieved by a wide range of porosities and therefore, degree of weathering. With few exceptions, actual retention means from C. feei collection sites fell within this narrow margin. The exceptions varied from the means with no significant differences. These data suggest that, within C. feei habitat in southeast Missouri, moisture level requirements, which restrict C. feei to a narrow niche, are satisfied by and are subject to porosity of its limestone substrate. Future studies are needed to determine if these is consistent with alternative substrates in other North American C. feei habitat. •' I , FEEi.—Cheilanthes feei is slow to establish or re-establish after road cuts and mining. Based on data taken from this study, C. feei habitat in southeast Missouri is non-renewable. Mechanical weathering is a long-term process and substrate characteristics cannot be readily mimicked or replaced. Chemical weathering is more rapid. Chemical weathering induced by pollution may open up new C. feei habitat. Unfortunately, chemical weathering may simultaneously destroy existing habitat. Therefore, additional studies on formations across the western United States, southwestern Canada and north central Mexico are imperative to determine whether C. feei habitat in North America is at risk. THE FUTURE OF CHEILANTHES ate University (Grant P sion to collect spores E Mary Bequette of Ce LITERATURE CITED 2001. C-Fern Web Manual: Basic C-Fern Nutrient Medium. http://cfern.b manual/cfmnutrientpreparation.html. A. C. 1999. LEA proteins, pp. 753-780. in P.R. Shewry and R. Casey (eds.j. Set C-FERN. CUMING, M. S. 1989. Epidermis and hypodermis of the saguaro cactus (Cereus giganteus): anatc and spectral properties. Amer. J. Bot. 76:1698-1706. DARLING, Davies (ed). Plant Hormones: Physiology, Biochemistry and Molecular Biology. Klu^ C. W. 1988. Applied Hydrogeology. MacMillan Publishing Company, NY. L., M. F. CRESGENTE, and G. Rossi. 1998. Photosynthetic performance and water i efficiency of the fern Cheilanthes persica. Photosynthetica 35:507-516. H. G., and J. Z. Kiss. 1998. Spore germination in populations of Schizaea pusi in New Jersey and Nova Scotia. Internat. J. Plant Sci. 159:848-852. FETTER, GRATANI, GUIRAGOSSIAN-KISS, USA. V. 1973. Blue light interference in the phytochrome-controlled germination of th spores of Cheilanthes farinosa. Plant Physiol. 51:306-311. V. 1980. Cytology, physiology, and biochemistry of germination of fern spores. Internat Rev. of Cytol. 62:69-118. RAGHAVAN, V. 1989. Developmental Biology of Fern Gametophytes. Cambridge University Press RAGHAVAN, RAGHAVAN, E., G. E. DOUGLAS, S. J. HEARNED, S. HUXHAM, and J. M. WYNN. 2001. Enhancement of fen spore germination and gametophyte growth in artificial media. Amer. Fern }. 91:179-186. D. 2002. Growing xerophytic (arid-loving) ferns, http://amerfernsoc.org/growdry.html. SHEFFIELD, SIEGLER, Germination of Fern Spores in Natural Soils WEN-HSIUNG Ko ABSTRACT.—In the presence of light, the germination rates of spores of Nephi Phlebodium m »ils were similar to those on v or water agar. All the soils tested promoted elongation of rhizoids of N. exaltata and stimul growth of protonemata of C. glaucum. Spores of the fungus Botryodiplodia theobromae germin completely on water or water agar under light or in darkness but failed to germinate on soils u the same conditions. The results suggest that spores of ferns are not sensit soil. Contrary to microorganism es to soil microbiost; to ferns because it would be ;ui• < essful colonization ontain fungal spores in great numbers (Warcup, 1955), as most ve to soil fungistasis which can be overcome by addition of 3 to soil (Ko and Lockwood, 1967; Lockwood, 1977). Microbial quiescence in natural soils was subsequently extended to include actinomycetes and bacteria, and the term soil microbiostasis has been introduced to described the antagonistic phenomenon of soil against fungi, actinomycetes and bacteria collectively (Ko and Ho, 1984). Microbiostasis in natural soils is considered to be caused by nutrient deprivation resulting from microbial activity (Ho and Ko, 1986). Natural soils also contain a great number of fern spores commonly referred to as the spore bank (Hamilton, 1988). When a small amount of soil was placed on a nutrient agar medium and exposed to light, fern spores in the soil readily germinated (Hamilton, 1988). Most fern spores are between 25 and 50 urn in diameter (Page, 1979; Devi, 1981), about the same size as many fungal spores (Walker, 1952). Because germination of fern spores in soil has not been quantitatively compared with that in non-soil medium, it is not known if fern spores are sensitive to soil microbiostasis. To address this question, spore germination of three different fern species on soils collected from three different locations was compared with that on water and water agar. Fungal spores Botryodiplodia theobromae Pat. were used as a control because they are sensitive to soil microbiostasis and like fern spores their germination does not MATERIALS AND METHODS Fertile fronds of the Hawaiian tree fern Cibotium glaucum (J.E. Smith) Hook. & Arn. (Dicksoniaceae), hare's foot fern Phlebodium aureum (L.) J. Smith (Polypodiaceae) and sword fern Nephrolepis exaltata (L.) Schott. KO: FERN SPORE GERMINATION 71 (Nephrolepidaceae) were collected from nature in the Hilo area. Fronds from each species were placed in an uncovered plastic box (13 X 24 X 35 cm) kept on the laboratory bench for air drying to discharge spores. A small quantity of spores was transferred to 5 ml sterile distilled water in a test tube with a pair of forceps. The concentrations of fern spores used ranged from 1.3 x 103 to 1.8 X 103 spores/ml as determined by a Pipetman (West Coast Scientific, Oakland, CA) microliter pipet (Ko et al., 1973). Fungal spores for comparison were obtained by growing B. theobromae on 10% V-8 agar (10% V-8 juice, 0.02% CaC03 and 2% agar) at 24°C under cool white fluorescent light (2,000 lx) for 9 days. Mature pycnidia were transferred to 5 ml sterile distilled water in a test tube, and crushed with a sterile spatula to release pycnidiospores. Spores were separated from crushed pycnidia by sedimentation before use. The concentrations of pycnidiospores used, ranged from 38 X 103 to 75 X 103 spores/ml. Soil samples were collected from farm lands at Hilo (silty clay loam; pH 6.8), Volcano (silt loam, pH 6.8) and Mealani (silt loam, pH 5.3) on the island of Hawaii. Soils were taken from a depth of 0 to 15 cm after surface litter was cleared, sieved through a 2-mm screen and moistened to about 65% field capacity. These soils were stored in polyethylene bags for at least one month to allow microorganisms to exhaust nutrients which might have become available due to soil disturbance (Chuang and Ko, 1988). For testing germination of spores on soil surfaces, approximately 25 g of soil adjusted to about 75% field capacity was placed in a Petri plate (100 mm diam.). It was compressed to form a disk (ca. 60 mm diam.) and the surface was smoothed with a bent spatula. Three drops (ca. 0.15 ml) of spore suspension were added to a sterile polycarbonate membrane (8 um, 47 mm diam.; Nuclepore Co., Pleasanton, CA) laid on each soil disk in the Petri plate. Inoculated plates were incubated at 24°C under cool white fluorescent light (2,000 lx) or in darkness for 5-9 days for fern spores and 12 hr for fungal spores. After incubation, each polycarbonate membrane was transferred from the soil disk to a moistened paper towel to wipe off soil particles on the lower surface of the membrane. The membrane was then placed on the cover of the Petri plate, and germination of spores was observed under a 40X objective. To determine if exogenous nutrients were required for germination, spore germination was similarly tested on polycarbonate membranes floating on the surface of sterile distilled water in Petri plates or directly on 2% water agar. Percentage germination was determined by counting 100 spores in each treatment. For each treatment, two of the longest rhizoids were measured and the average length was recorded. Two replicates were used and all experiments were done at least twice. Nephrolepis exaltata.—In the presence of light, spores of N. exaltata germinated by producing a protonemal cell and an elongating rhizoid. The average percent germination in 5 days on the three different soils separated by polycarbonate membranes was 50%, which was similar to that on distilled - \L: VOLUME 93 NUMBER 2 Dark i.i.j Germination Medium Hilo soil Volcano soil Mealani soil Water agar (%) Rhizoid length Germinate 377.5 (9.5) 304.5 (14.5) 285.5 (4.5) .Jo, 53 z <;lT8,h I 193.5 (9.5) water separated by polycarbonate membrane or on water agar directly (Table 1). The mean length of rhizoids from spores germinated on the soils was 323 urn, about 108% and 67% longer than those on water and water agar, respectively. Without light, all or nearly all the spores of N. exaltata failed to germinate on soils, water or water agar (Table 1). On Mealani soil, 7.5% of spores examined produced a green protonemal cell but no rhizoids after 5 days in darkness. Phlebodium aureum.—The germination pattern of P. aureum spores on soils was similar to that of N. exaltata spores. Under light, P. aureum also geminated by producing a protonemal cell and an elongating rhizoid, and the average germination rate of 57% after 6 days on the three soils was similar to that on distilled water or water agar (Table 2). The average length of rhizoids from spores germinated on the soils was 290 um which was about the same as that on water and 97% longer than that on water agar. In darkness, all or nearly all the spores of P. aureum failed to germinate on soils, water or water agar (Table 2). On Mealani soil, 8.5% of spores tested produced a green protonemal cell, without rhizoids after 6-day incubation without light. Cibotium glaucum.—In the presence of light, spores of C. glaucum germinated by producing an expanding protonema and an elongating rhizoid. The average germination rate on the three soils was 58% after 9-day incubation, similar to that on water or water agar (Table 3). All the soils tested stimulated growth of protonemata. The protonemata on soils consisted of 3 to 5 cells each, 2. Germination of fern spores f Phlebodium aure um on natural soils under 1 ght and in darkness after incubation at 24°C for 6 days. Standard dev lations are given in parenth TABLE Light Germination Dark Rhizoid length Germination Rhi •jjdtag* 8.5 (0.5) • Medium Hilo soil Hi 285.5 (4.5) 265.5 (5.5) KO: FERN SPORE 3. C^rmii lation of fern spores of Cibotium glau darkr less after in, :ubation at 24°C for 9 days. Standard d i under light and in ioL^reTi'ventl parentheses. Light Ililo • Yolra ill (%) 52.0 63.5 57.5 60.0 42.0 (2.C (4.5 (2.5 (3.C (2.C on Dark Rhizoid length EH Germination Rhizoid langtb | o whereas those on water and water agar contained only 1 or 2 cells each. The average length of rhizoids from spores germinated on soils was 286 urn, similar to those on water or water agar. In darkness, none of the C. glaucum spores examined germinated on soils, water, or water agar (Table 3). Botryodiplodia theobromae.—Light had no effect on the germination of fungal spores of B. theobromae, which germinated by producing an elongating germ tube. Nearly all the spores tested geminated on water or water agar after incubation for 12 hr under light or darkness (Table 4). However, under the same conditions spore germination was completely inhibited on the three different soils tested. Spores of the fungus B. theobromae germinated completely on water or water agar with or without light, but remained inactive on soils under the same conditions. This shows that the three different soils used in this study are suppressive to microorganisms, as are most soils (Lockwood, 1977; Ko and Ho, 1984). However, in the presence of light, the germination rates of spores of all three fern species tested on soils were similar to that on water and water agar, indicating that fern spores are not sensitive to soil microbiostasis. The general phenomenon by which germinable spores of microorganisms are rendered static in soils (Bruehl, 1986), therefore, does not appear to apply to spores of 4. Germination of fungal spores of Botryodif and in darkness after incub ationat24°Cforl2days.Sta ndarddeviatic TABLE ,r,-. Siveninparen "'""*• Germim Medium Light Dark 99.5°(0.5) «M Hilo soil Volcano soil 74 AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 2 (2003) Spores of many fungi are nutritionally dependent and require exogenous nutrients for germination, but others are nutrient independent and are capable of germination in nutrient-free water (Ko and Lockwood, 1967). All the nutrient-dependent, and most of the nutrient-independent spores, are sensitive to soil microbiostasis. Only some of the nutrient-independent types can germinate freely on soil (Ko and Lockwood, 1967; Hwang and Ko, 1974). Fern spores appear to be similar to the latter group although a greater range of fern species awaits investigation. The germination rate of each fern species tested on water was similar to that on water agar. Because water agar contains sufficient nutrients for spore germination (Ho and Ko, 1980), the results suggest that fern spores are nutritionally independent. This is in accordance with previous findings of the ability of a number of fern species to germinate on water (Dyer, 1979). Insensitivity of fungal spores to microbiostasis is detrimental to their survival in nature as germ mycelia from the germinating spores will be lysed due to unavailability of organic nutrients for their growth in soil (Ko and Lockwood, 1970). However, this is not the case with fern spores as inorganic nutrients needed for their growth are available in soil. Therefore, ability to germinate freely on soil is advantageous to ferns for their colonization of suitable habitats. Most species of ferns depend on light for germination of spores (Weinberg and Voeller, 1969). When fern spores fall to the ground in scattered masses from sporophytes after maturation, a portion of them will percolate into the pore space of soil and remain quiescent due to the absence of light. This might be an important source of fern spores in the spore bank. A large number of those spores on the soil surface remain ungerminated as shown by the observation that about 50% of spores of all the three species of ferns tested remain dormant on soil even in the presence of light. These spores may be dispersed and buried in soil through the activity of earthworms (Hamilton, 1988; Hamilton and Lloyd, 1991) and become part of the spore bank. Light is inhibitory to spore germination of some fern species (Whittier, 1973; 1977; 1978). In this case, ungerminated spores on the soil surfaces would also become part of the spore bank. All of the test soils appeared to promote elongation of rhizoids of N. exaltata but not P. aureum or C. glaucum. The activation of rhizoid elongation may be due to minerals present in soils. Elongation of rhizoids in the fern Onoclea sensibilis has been shown to be promoted by mental ions (Miller et al., 1983). Minerals in soils may also account for the growth promotion of protonemata of C. glaucum on soils. However, the actual cause of the stimulatory effects of soils on rhizoid elongation and protonemal growth remains to be investiApproximately 8% of N. exaltata and P. aureum spores germinated by producing a green protonemal cell without any rhizoid on Mealani soil in darkness. It is not know what factor in the soil is responsible for such a phenomenon. The fate of these germinated spores on soil after an extended period of time also remains to be investigated. tification of the ferns. LITERATURE CITED nlborne Plant Pathogens. Macmillan Publishing Company, New York. . H. Ko. 1988. Rhizoctonia soioni'-suppressive soils: detection by chlanation. Ann. Phytopathol. Soc. Japan 54:158-163. DEVI, S. 1981. Reference Manual of Fern Spores. Economic Botany Information Service, National Botanical Research Institute, Lucknow, India. DYER, A. F. 1979. The culture of fern gametophytes for experimental investigation. Pp. 253-305, in A. F. Dyer (ed.), The Experimental Biology of Ferns. Academic Press, New York. HAMILTON, R. G. 1988. The significance of spore banks in natural populations of Athyrium pycnocarpon and A. thelypterioidr*. Amer. Fern [. 78:96-104. HAMILTON, R. G. and R. M. LLOYD. 1991. An experimental study on the effects of earthworms on ecological success of fern gametophytes. Amer. Fern J. 81:95-99. Ho, W. H. and W. H. Ko. 1980. Agarose medium for bioassay of antimicrobial substances. Phytopathology 70:764-766. Ho, W. C. and W. H. Ko. 1986. Microbiostasis by nutrient deficiency shown in natural and synthetic soils. J. Gen. Microbiol. 132:2807-2815. HWANG, S. C. and W. H. Ko. 1974. Germination of Calonectria crotalariae conidia and ascospores on soil. Mycologia 66:1053-1055. Ko, W. H., L. L. CHASE and R. K. KIMMOTG. 1973. A microsyringe method for determining concentration of fungal propagues. Phytopathology 63:1206-1207. Ko, W. H. and W. C. Ho. 1984. Soil microbiostasis. Pp. 175-184, in J. Bay-Peterson (ed.). Soilborne Crop Diseases in Asia. Food and Fertilizer Technology Center, Taipei, Taiwan. Ko, W. H. and J. L. LOCKWOOD. 1970. Mechanism of lysis of fungal mycelia in soil. Phytopathology Ko, W. H. and J. L. LOCKWOOD. 1987. Soil fungistasis: relation to fungal spore nutrition. Phytopathology 57:894-901. LOCKWOOD, J. L. 1977. Fungistasis in soil. Biol. Rev. 52:1-43. J. H., T. C. VOGELMANN and A. R. BASSEL. 1983. Promotion of fern rhizoid elongation by metal ions and the function of the spore coat as an ion reservoir. Plant Physiol. 71:828-834. PAGE, C. N. 1979. Experimental aspects of fern ecology. Pp. 551-589, in A. F. Dyer (ed.). The Experimental Biology of Ferns. Academic Press, New York. WALKER, J. C. 1952. Diseases of Vegetable Crops. McGraw-Hill, New York. WARCUP, J. H. 1955. On the origin of colonies of fungi developing on soil-dilution plates. Trans. Brit. Mycol. Soc. 38:298-301. WEINBERG, E. S. and B. R. VOELLER. 1969. External factors inducing germination of fern spores. Amer. Fern J. 59:153-167. WHITTIER, D. P. 1973. The effect of light and other factors on spore germination in Botrychium dissectum. Canad. J. Bot. 51:1791-1794. WHITTIER, D. P. 1977. Gametophytes of Lycopodium obscurum as grown in axenic culture. Canad. J. Bot. 55:563-567. WHITTIER, D. P. 1981. Spore germination and young gametophyte development of Botrychium and MILLER, New Species in Adiantum from Brazil JEFFERSON PRADO Segao de Briologia Institute) de 1 STRACT.—A new species of Adiantum, A. pulcherrimum Prado, is described from the I ests of Rio de Janeiro and Sao Paulo States, and inland forest from Minas Gerais State, B 1 he distinguished by long-creeping rhizomes, stipes with scattered i rs, laminae glaucous abaxially, i i as well as a key for the related s The genus Adiantum in Brazil is represented by ca 59 species, including one described here. Most species occur in primary and secondary forests in the southeastern region of the country, from sea level to 2000 m. In this area ca 62%, 34 spp., of the species known for Brazil have been found. Several recent studies have dealt with Brazilian Adiantum: Zimmer & Prado (1997); Prado (1997); Prado & Palacios-Rios (1998); Prado (2000); Lellinger & Prado (2001); Prado (2001); Prado & Lellinger (2002). This paper is an additional contribution toward a revision of Adiantum in Brazil and treats a new species from the Atlantic forests of Rio de Janeiro and Sao Paulo States, and the inland forest of Minas Gerais State. Adiantum pulcherrimum Prado, sp. nov., Fig. 1. A A. abscisso Schrad., cui affinis, stipitis cum pilis sparsis et pallidis castaneis, segmentis medianis abaxialiter glaucis in apice principaliter longiacuminatis vel acutis, indusiis glabris differt.—Type. Brazil: Rio de Janeiro, Mun. de Mangaratiba, Reserva Rio das Pedras (RPPN-IBAMA), trilha do Cambuca, 16 Aug. 2001, C. Mynssen et al. 356 (holotype: RUSU!; isotypes: MBM!, NY!, RB!, SP!, UC!). Plants terrestrial. Rhizomes long-creeping, 3-4 mm in diam., scaly, the scales somewhat shiny, essentially concolorous, appressed, varying from light to dark brown, lanceate, sparsely denticulate at margins. Fronds monomorphic, 30-80 cm long; laminae 20-50 cm wide, deltate-pentagonal to ovate, 4- to 5-pinnate at base, 2-pinnate distally; stipes 5-8 mm apart, 1/2-2/3 the length of fronds, dark brown to black, adaxially sulcate, hairy, the hairs scattered, appressed throughout or patent, light brown, minute 0.1-0.2 mm long; rachises similar to the stipes in color and indument; pinnae alternate, stalked, oblong-lanceate, slightly decreasing in width at the base and apex, 10-20 X 4-7, the terminal pinna conform, indument of costae like that of stipes; EW ADIANTUM SPECIES FROM BRAZIL FERN JOURNAL: VOLUME 93 NUMBER 2 (2003) median segments mostly dimidiate, lacking costa, glabrous on both surfaces and glaucous abaxially, trapeziform, 1.5-5 cm long, not articulate to stalks (color of stalks passing into segment bases), the stalks slender, 1-4 mm long, the segment margins curved basiscopically, the outer two sides variously biserrate, crenate or shallowly to somewhat deeply lobed, chartaceous, bases of the segments overlaping the rachis, apices mostly long-acuminate or acute, the proximal pairs of segments reduced, somewhat rounded or triangular, the terminal segment wide and rhombic; veins free, flabellately several-times forked, the veins ending in marginal teeth on the sterile segments; sori varying from ellipsoid to curved-oblong, 1-3 mm long, solitary on lobules of the distal and acroscopic margins, up to about 12 per pinnule, indusia dark brown, glabrous, with entire margins; spores tan, surface verrucate. MATERIAL EXAMINED.—BRAZIL. Minas Gerais, Vigosa, Faz. Aguada, alt. 725 m, 16 Sept. 1930, Y. Mexia 5055a (UC). Rio de Janeiro, Mun. Mangaratiba: Reserva Ecologica Rio das Pedras, trilha do Cambuca, 14 Sept. 1996, J. M. A. Braga et al. 3492 (RUSU); Idem, c. 190 m, 6 May 1997, C. Mynssen et al. 97 (RUSU); Idem, 26 Aug. 1998, M. V. Ddria et al. 01 (RUSU); Idem, 13 Aug. 1999, C. Mynssen et al. 292 (RUSU). Sao Paulo, Iguape, Pocinhos, Aug. 1927, A. C. Brade 8501 (UC 2 sheets); Idem, id., Morro das Pedras, Aug. 1927, A. C. Brade 8503 (UC), 8504 (NY, UC). HABITAT.—Growing in secondary forests, at low elevations (0-725 m), forming large populations. Adiantum pulcherrimum can be recognized by its long-creeping rhizomes, stipes with scattered minute and light brown hairs, glaucous laminae abaxially, median segments curved basiscopically, apices mostly long-acuminate or acute, and glabrous indusia. Adiantum pulcherrimum belongs to the Adiantum trapeziforme group, which is characterized by pedate laminae 4- to 5-pinnate at base, becoming 2pinnate distally, ultimate segments trapeziform to asymetrical, rounded to obtuse or acute to acuminate at tips, glabrous or pubescent axes, dark brown to blackish, and mostly oblong sori confined to the distal and acroscopic margins of the segments. The following species of this group are found in Brazil: Adiantum abscissum Schrad., A. curvatum Sw., A. mathewsianum Hook., A. ornithopodum C. Presl ex Kuhn, A. patens Willd., A. pentadactylon Langsd. & Fisch., A. pulcherrimum Prado, and A. trapeziforme L. (cultivated). Adiantum abscissum is the most closely related species to A. pulcherrimum but it differs in having stipes with scales and hairs, rachises densely puberulent adaxially, and more numerous, smaller segments with apices rounded or obtuse. It is more widely distributed in Brazil, occuring in the states of Ceara, Pernambuco, Alagoas, Bahia, Mato Grosso, Goias, Minas Gerais, Espirito Santo, Rio de Janeiro, Sao Paulo, Parana, and Santa Catarina. PRADO: NEW ADIANTUM SPECIES FROM BRAZIL 79 Adiantum cultratum J. Sm. in Hook, is probably another closely related species, but its identity and typification are uncertain. This species was described by John Smith in Hooker (1851: 34) and two specimens were cited: Hab. St. Vincent, in J. Sm. Herb., Macrae s.n.(BM!) and St. Catherine's, Brazil, Armstrong s.n. (not found). According to Proctor (1977) the Macrae specimen at BM represents an unidentified species of the A. trapeziforme group, probably originating from a cultivated plant. It has never been found again in the Lesser Antilles, and should not be considered a member of the local flora. Hoshizaki (1970) also mentioned the need for further study on the correct name for this species. Most likely, A. cultratum is endemic to southeastern Brazil and is cultivated in several countries. Because the material collected by Armstrong has not been found, and because the Macrae specimen is an undesirable lectotype for this taxon, the identity of A. cultratum remains somewhat uncertain. There is no recent collection of this species in Brazil. KEY TO THE SPECIES OF ADIANTUM TRAPEZIFORME GROUP IN BRAZIL ALLIED > abruptly at segment bases i chartaceous; terminal segment of a penultimate division dusia oblong A. Trapeziforme s rigid to subcoriaceous; terminal segment of a penultimate division asymetncallv rhombic; indusia oblong to semilunate ,4. mat . Color of the stalks passing into segment bases 3. Stipes glabrous along median and distal portions 4. Median segments mostly debate to trapeziform with acute to long-acuminate apices A. pentadactylon 4. Median segments quadrangulate to trapeziform with rounded to obtuse apices 5. Both surfaces of the segments glabrous; laminae rigidly chartaceous to subcoria5. Both surfaces of the segments with minute hairs; h divisions trapeziform with mostly long-acuminate or divisions narrow with long-acuminate apices rounded to obtuse apices 7. Median segments ca. 4-5 times longer than I thank Claudine Mynssen for the loan of the holotype and other c jreparing the illustrations, Dr. David Lellinger for his help with Adia ind Dr. Alan R. Smith for sending the paratypes and for helpful and c nanuscript. This work was funded by CNPq (Proc. number 300843/' LITERATURE CITED 80 AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 2 (2003) D. B. and J. PRADO. 2001. The group of Adiantum gracile in Brazil and environs. Amer. Fern J. 91:1-8. PRADO, J. 1997. Flora da Serra do Cipo, Minas Gerais: Pteridaceae-Adiantoideae e Taenitidoideae. Bol. Bot. Univ. Sao Paulo 16:115-118. PRADO, J. 2000. A new species of Adiantum (Pteridaceae) from Bahia, Brazil. Brittonia 52:210-212. PRADO, J. 2001. Adiantum giganteum (Pteridaceae), a new maidenhair fern from Amazonia, Brazil. Fern Gaz. 16:209-212. PRADO, J. and D. B. LELLINGER. 2002. Adiantum argutum, an unrecognized species of the A. latifolium group. Amer. Fern J. 92:23-29. PRADO, J. and M. PALACIOS-RIOS. 1998. Taxonomy and distribution of Adiantum trapeziforme and A. pentadactylon. Amer. Fern J. 88:145-149. PROCTOR, G. R. 1977. Pteridophyta. Pp. 1-414 in R. A. Howard, ed. Flora of the Lesser Antilles. vol. 2. Harvard University Press, Cambridge. ZIMMER, B. and J. PRADO. 1997. Proposal to reject the name Adiantum dissimile (Polypodiaceae, LELLINGER, merican Fern Journal J New Species and New Combinations of Grammitidaceae from Peru ersity of Texas at Austin, Bio Labs Austin, TX 78712, U.S.A. and Museo de Historia Natural, lies 1256, Apartado 14-0434, Lims ALAN R. SMITH rbarium, 1001 Valley Life Science arsity of California, Berkeley, CA £ ABSTRACT.—We describe two new species from Peru, Ceradenia tryonorum and Terpsichoreyoungii K.i.iinmitul i i I. i i ',nii i n is i mi mbci of subg. Ceradenia and ' hydathodes. Terpsichore youngii belongs to the T. taxifolia group. Three n< made: Melpomene youngii, Terpsichore anfractuosa, and T. subscabra. We j Tryon and Stolze (1989-1994) documented almost 1060 species of pteridophytes in Peru; their work greatly facilitates the recognition of new species and new distributional records. Recent botanical explorations in Peru provide interesting fern additions for this rich tropical flora. In the Neotropics, the Grammitidaceae are represented by nine genera: Ceradenia, Cochlidium, Enterosora, Grammitis, Lellingeria, Melpomene, Micropolypodium, Terpsichore, and Zygophlebia. The Peruvian fern flora includes 67 species in these genera (Tryon & Stolze, 1993), with probably another dozen species expected based on recent accounts from neighboring countries (j0rgensen & Leon-Yanez, 1999; Smith et al., 1999). Here we describe two new species, one in Ceradenia and the other in Terpsichore. Ceradenia tryonorum B. Leon & A. R. Sm., sp. nov. (Fig. 1 A-D) TYPE: Peru. San Martin: Province Mariscal Caceres, Parque Nacional Rio Abiseo, near El Tingo, 7°58'S, 77°18'W, 2800 m, 29 June 1999, B. Leon & K. R. Young 3840 (holotype: USM!; isotypes: TEX!, UC!). Rhizomata breve repentia, radialia; paleae densae, stramineae, lanceolatae ca. 4-7 X 1-1.5 mm, non clathratae, apice obtusae vel apiculatae, setiferae. Frondes 10-15 cm longae. Petioli straminei, phyllopodiis instructi. Laminae pinnatifidae, oblongae, pilis mrcatis et pilis glandulosis conspersis vestitis; venae simplices vel interdum furcatae, liberae, adaxialiter in hydathodis terminantes. Sori lineares, superficiales, 3-6 (-10) mm longi. Rhizomes suberect, radially symmetrical, 5 mm thick, densely covered with overlapping, stramineous to light tan, non-clathrate scales, rhizome scales \AL: VOLUME 93 NUMBER 2 LEON & SMITH: NEW GRAMMITIDACEAE FROM PERU 83 4-7 mm X 1—1.5 mm, lanceolate, apices apiculate or obtuse, apical portion with a glandular furcate hair, branches of hair similar in length or one three times more elongated, margins entire or rarely with furcate hairs. Leaves pendant, 10-15 cm long, petiolate; petioles tan to light brown, dull, 2.7-7 cm X 0.4-0.85 mm, articulate, covered with abundant simple to often furcate hairs 0.1 mm long, also with scattered dark brown setae 1-3 mm long, at the base with inconspicuous, black phyllopodia, 0.5-1 mm long. Laminae thin, with spongy parenchyma, 6.5-10 cm X 2-3.5 cm, narrowly ovate, pinnatifid. incised about halfway to rachis, proximal segments not or slightly reduced, laminae bases shortly cuneate, lamina apices acute, segments ascending (55-) 60-75°, 1-1.5 X 0.5-0.7 cm, segment apices obtuse, costae hidden or slightly prominent abaxially, prominulous and of the same color as the lamina adaxially; abundant red-brown setae on both surfaces, setae mostly 1-2 mm long, lamina abaxially with abundant wax-like glandular hairs, adaxially with scattered furcate glandular hairs; veins free, pinnate, 5-8 pairs of veins per segment, ultimate veinlets simple or furcate, basal veins borne from the rachis at the middle of the lamina, veins ending in hydathodes adaxially, these 0.1-0.2 mm long, without calcareous deposits; sori superficial, linear, 3-6 (-10) mm long, extending from costae to margins of segments; sporangia 200-350 X 120142 \im, with 11-14 annulus cells; spores 25 [im diam. Distribution and habitat.—This species is known only from the type locality in northeastern Peru. It grows as an epiphyte in montane forests. The understory included Chusquea scandens Kunth, with about 40% cover. Stature of canopy dominants was often 11-13 m, with emergents to 15 m. Common medium and large trees included Brunellia, Clethra, Freziera, Hedyosmum, Symplocos, and Weinmannia, among others. The species epithet honors Drs. Alice F. Tryon and the late Rolla M. Tryon for their contributions to our knowledge of the Peruvian pteridoflora. Ceradenia tryonorum is characterized by stramineous to light tan rhizome scales, radially arranged leaves, 2.7-7 cm long petioles, pinnatifid laminae, minute wax-like glandular hairs on the abaxial surface of the laminae, adaxial hydathodes, and linear non-sunken sori. In fresh material, the costae are obscure adaxially, but abaxially they are conspicuous in the proximal portion of the leaf. This species has the wax-like glands characteristic of Ceradenia, a genus of approximately 55, mostly neotropical, species (one in Africa and perhaps a few in Madagascar). The glands are a synapomorphy of the genus, and clearly establish the affinities of C. tryonorum. However, a combination of characters makes it difficult to establish clearly the intrageneric affinities and subgeneric position of C. tryonorum. Rhizome morphology and anatomy, together with laminar indument, were the main characters used to circumscribe two subgenera in Ceradenia (Bishop 1988). Species in subg. Filicipecten have dorsiventral and solenostelic rhizomes, lack wax-like laminar glands, and have petiolate laminae, while subg. Ceradenia has radially symmetrical and dictyostelic rhizomes, wax-like glandular laminar trichomes, and short-petiolate or sessile laminae. Species of both subgenera have round or oblong sori, the usual 84 AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 2 (2003) condition in Grammitidaceae. In the totality of its characters, we believe that Ceradenia tryonorum is a member of subg. Ceradenia, but a very atypical one, especially because of the linear sori, and distinct petioles. Ceradenia (Bishop 1988), Enterosora (Bishop & Smith 1992), and Zygophlebia (Bishop 1989) include exclusively anhydathodous neotropical and a few African-Madagascan species. The anhydathodous condition and the presence of a spongy leaf parenchyma indicate a close relationship among these three genera. Ceradenia and Enterosora include species with entire to shallowly pinnatifid to pinnatisect laminae and mostly free veins, whereas Zygophlebia has anastomosing veins and usually more deeply dissected blades. Until now, only one exception to the anhydathodous condition was known in this closely related assemblage: Enterosora asplenioides L. E. Bishop, from Ecuador and Colombia. The hydathodous condition in C. tryonorum is similar to that found in E. asplenioides, which has thin spongy laminae and superficial sori. The shared presence of hydathodes and the linear sori in these two species may reflect deeper relationships between Ceradenia and Enterosora. Bishop (1989), however, interpreted the absence of hydathodes as an ancestral state within the Grammitidaceae. These recently found exceptions may help to understand the evolution of these mostly upper Terpsichore youngii B. Leon & A. R. Sm., sp. nov. (Fig. 2 A-B) Type: Peru. Cusco: near San Lorenzo, 2300-2500 m, 6 July 2000, B. Leon & K. R. Young 4487 (holotype: USM!; isotype: UC!). Rhizomata breve repentia, 2-3 mm diam.; paleae densae, clathratae, margine setis hyalinis ornatae. Frondes 10-25 cm longae. Petioli brunnei. Laminae pinnatisectae vel pectinatae, anguste lanceolatae, pinnis 25-40 jugis pinnarum, abaxialiter dense pilosis; venae simplices, liberae, adaxialiter in hydathodis non calcareas terminantes. Pendant epiphytes. Rhizomes short-creeping, 2-3 mm wide, densely scaly; rhizome scales clathrate, 0.8-1.5 mm X 0.15-0.3 mm, lanceate, apical and marginal hyaline setae present, setae 0.07-0.13 mm long. Leaves 10-25 cm long, petiolate, petioles 3-7 cm X 0.2-0.7 mm, dark brown, dull, hairs 0.5-1.5 mm long; laminae chartaceous, narrowly lanceolate, gradually reduced at both ends, 1-4 cm wide, pinnatisect or pectinate, with 25-40 pairs of pinnae, these ascending 60-75° from rachis, 1-5 proximal pinnae less than half the total length of the longest pinna, gradually reduced to small segments, pinnae linear 1-2 cm X 1-2.5 mm, acute, pinna bases nearly symmetrical, abaxially with numerous red-brown hairs, 0.3-1 mm long (similar to those on the rachis), adaxially glabrous, pinna margins entire or with a few scattered glandular hairs; rachises densely hirsute, hairs 0.5-1 mm long, red-brown, also with scattered, black club-shaped fungi abaxially; veins free, central pinnae with 5-12 pairs of simple veins, adaxially ending in hydathodes lacking calcareous deposits; sori medial, oblong, sporangia without setae. LEON & SMITH: NEW GRAMMITIDACEAi; Terpsichore youngii 86 AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 2 (2003) Distribution and habitat—This species is known from Peru and Bolivia. It grows in forests dominated by Weinmannia, Clusia, Symplocos, Brunellia, Miconia, Myrsine, and Lauraceae, between 2200-3000 m elevation. The epithet honors Kenneth R. Young for his scientific endeavors in Peru. Collections examined.—Bolivia. Cochabamba: Province Ayopaya, 10 km Cocapata-Cotacajes, 16°38'S, 66°41' W, 3000 m, 9 May 1997, Kessleretal. 9401 (LPB not seen, UC); Cochabamba: Province Jose Carrasco Torrico, 5 km de Siberia a Karahuasi, 17°48'S, 64°41'W, 2200 m, Kessler et al. 9059 (LPB not seen, UC). This species belongs to the Terpsichore taxifolia group (Group 2 of Smith, 1993), which is characterized by the presence of club-shaped black fungi of the genus Acrospermum. Possibly, T. youngii is most closely related to T. alsopteris (C. V. Morton) A. R. Sm. Both species have chartaceous, pectinate laminae, with a few reduced proximal pinnae, and hairs on the laminae between veins abaxially. Terpsichore youngii differs from T. alsopteris (C. V. Morton) A. R. Sm. in having strongly clathrate rhizome scales and noncalcareous hydathodes. Terpsichore youngii has also considerably longer hairs on the lamina abaxially. In addition, the hairs on the rachises and laminae in T. alsopteris are less dense. A closer relationship of T. youngii is probably with T. david-smithii (Stolze) A. R. Sm. from Peru and Bolivia. That species agrees with T. youngii in having clathrate scales; however, the setae on the scales of T. david-smithii are darker, less numerous, and stiffer. Terpsichore anfractuosa (Kunze ex Klotzsch) B. Leon & A. R. Sm., comb, nov. Polypodium anfractuosum Kunze ex Klotzsch, Linnaea 20:375. 1847. Grammitis anfractuosa (Kunze ex Klotzsch) Proctor, Rhodora 63:35. 1961. Melpomene anfractuosa (Kunze ex Klotzsch) A. R. Sm. & R. C. Moran, Novon 2:429. 1992.—Type: Venezuela. Merida: Moritz 330 (holotype B, photo F; isotypes B, US!). Distribution and habitat.—Antilles, s. Mexico, Central America, Colombia, Venezuela, Guyana, Ecuador, Peru, Bolivia; epiphytic in cloud forests. Recent molecular work by Ranker et al. (unpubl.) indicates that this species, with black clavate fungi of the genus Acrospermum, groups with Terpsichore pichinchae (Sodiro) A. R. Sm., and hence belongs in Group 2 of that genus (Smith, 1993). This result might have been predicted simply by the presence of the distinctive black fungus on the abaxial rachis, costae, and sometimes within the sori. The presence of this fungus is a synapomorphy of Terpsichore, Groups 2 and 4, and we are unaware of the presence of this peculiar and distinctive fungus in any other grammitids, or any other fern, for that matter. Terpsichore anfractuosa, a rather strongly divergent and distinctive species itself, was placed in Melpomene by Smith and Moran (1992) because of the small, clathrate, entire rhizome scales. These scales are evidently very reduced in this species (and hence difficult to interpret), however, in a few specimens LEON & SMITH: NEW GRAMMITIDACEAE FROM PERU 87 rhizome scales have marginal setae at and near the apex. Some species of Terpsichore also have clathrate scales (e.g., T. david-smithii, T. pichinchensis). Terpsichore anfractuosa is distinguished from other species in Terpsichore (and Melpomene) by root proliferations ("stoloniform roots"; Tryon & Stolze, 1993:99-100) that produce buds and new plants (leading to a colonial habit on trunks and branches of trees), small fronds, and narrowly elliptical laminae (tapering gradually, at the base). KEY TO TERPSICHORE TAXIFOLIA GROUP IN PERU Marginal setae hyaline T. alsopteris (C. V. Morton) A. R. Sm Marginal setae dark colored and rigid. 8. Leaves < 2 cm wide, with setae abaxially and along margins; veins fewer than 5 pairs per pinna T. pichinchensis (Hieron.) A. R. Sm 8. Leaves > 3 cm wide; glabrous or very sparsely setose abaxially; veins more than ita (Klotzsch) A. Terpsichore subscabra (Klotzsch) B. Leon & A. R. Sm., comb. nov. Polypodium subscabrum Klotzsch, Linnaea 20:377. 1847. Grammitis subscabra (Klotzsch) C. V. Morton, Phytologia 22:80. 1971.—Type: Venezuela. Merida, Moritz 332, (holotype B; isotypes BM-photos F!, K!, TEX-LL!). Polypodium jamesonioides Fee, Mem. foug. 7:59, t. 21, f. 4. 1857. Grammitis jamesonioides (Fee) C. V. Morton, Contr. U.S. Natl. Herb. 38:108. 1967. Terpsichore jamesonioides (Fee) A. R. Sm., Novon 3: 487. 1993.—Type: Colombia. Santander, Ocana, Schlim 399 (holotype L; photos F, UC!, US). 88 AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 2 (2003) Distribution and habitat.—Hispaniola, Costa Rica, Panama, Colombia, w. Venezuela, Ecuador, Peru; epiphytic or epipetric, pendant, in paramos and subparamos, dwarf forests. Terpsichore subscabrum was misinterpreted by Stolze (1991) as a Polypodium, thus contradicting Morton's (1971: 80) placement of the species in Grammitis. Stolze excluded P. subscabrum from Grammitis 8.1. and also from Pecluma, and characterized the taxon as having "Petiole subglabrous, with swollen articulation at base. Lamina pectinate, 22 cm long and 1.7 cm broad, axes and tissue scabrous, viscid, trichomes 0.1 mm long, tightly appressed; pinnae to 0.8 cm long, 0.2 cm broad, linear, subacute; spores yellow, monolete." Our examination of the type, however, shows that it clearly belongs to Terpsichore, and not to Polypodium, where Stolze placed it. Within Terpsichore, it belongs to the group of T. lanigera (Group 3 of Smith, 1993). This group of Terpsichore often has monolete spores (Wagner, 1985; Smith, 1993), and some species, particularly T. subscabra, have viscid, appressed glands, an unusual character in grammitids. Melpomene youngii (Stolze) B. Leon & A. R. Sm., comb. nov. Grammitis youngii Stolze, Fieldiana, Bot. 32:97. 1993.—TYPE: Peru. San Martin, Province Mariscal Caceres, Parque Nacional Rio Abiseo, Puerta del Monte, 3600 m, 19 Nov 1985, K. R. Young 1684 (holotype: USM!; isotype: F). Distribution and habitat.—Peru and Bolivia. This epiphytic species with pendant leaves is commonly found in upper montane forests. It appears to be related to Melpomene sodiroi (H. Christ & Rosenst.) A. R. Sm. & R. C. Moran and M. flabelliformis (Poir.) A. R. Sm. & R. C. Moran, because of its glabrous rachis and long-creeping rhizomes. We thank Antoi litior sponsored by a National Geographic Society Expedition Grant. Fieldwork in Rio Abiseo National Park was sponsored by grants from the John D. & Catherine T. McArthur Foundation and the European Community (Project: Monitoring and Modelling the Impacts of Changing Government Policies on Biodiversity Conservation in the Andes). Thanks to Norman Robson for help with Latin descriptions, and to Sandra Knapp, Allison Paul and Harald Schneider for aid with specimens and suggestions on the manuscript. We also than! eir comments. LITERATURE CITED L. E. 1988. Ceradenia, a new genus of Grammitidaceae. Amer. Fern J. 78:1-5. L. E. 1989. Zygophlebia, a new genus of Grammitidaceae. Amer. Fern J. 79:103-118. BISHOP, L. E. and A. R. SMITH. 1992. Revision of the fern genus Enterosora (Grammitidaceae) in the New World. Syst. Bot. 17:345-362. JBRGRNSEN, P. M. and S. LEON-YANEZ (eds.). 1999. Catalogue of the vascular plants of Ecuador. Monogr. Syst. Bot. Missouri Bot. Gard. 75:154-168. MORTON, C. V. 1971. Supplementary notes on Grammitis in Ecuador. Phytologia 22:71-82. BISHOP, BISHOP, LEON & SMITH: NEW GRAMMITIDACEAE FROM I i Klotzsch. Amer. Fern J. R. G. STOLZE, eds. 1989. Pteridophyta of Peru. Part II Dennstaedtiaceae. Fieldiana, Bot. N.S. 22:1-128. d R. G. STOLZE, eds. 1991. Pteridophyta of Peru. Part IV. Dryopteridaceae. Pteridophyta of Peru. Part III. 16. Thelypteridaceae. uid R. G. STOLZE, eds. 199: iceae. Fieldiana, Bot. N.S. ! nd R. G. STOLZE, eds. 199^ ». Fieldiana, Bot. N.S. 234:: Pteridophyta of Peru. Part V. 18. Asploniac:e;n—2 1. Pteridophyta of Peru. Part VI. 22. Marsileaceae—28. v World gramm Lectotypification of Several Names Currently Placed in Diplazium (Woodsiaceae) i Vegetal, Universidad Auto ROBBIN C. MORAN fork Botanical Garden, Bronx, NY 10458-5126, U.S.A. . sechellarum, and D. sikkimense. ' Diplazium is pantropical with an estimated 400 species, the majority of which occur in the tropics of the Old World (Kato and Kramer, 1990). The genus is taxonomically difficult, poorly known, and in need of a monographic study. In a recent study (Pacheco & Moran, 1999), 15 species that had been treated in Diplazium were recognized in Callipteris because they had anastomosing veins and rhizome scales with black-borders and bifid marginal teeth. The type of rhizome scale characteristic of these species, called the "Callipteris scale type," is known only in Callipteris and certain species of Diplazium; it does not occur in other fern genera. Many species of Diplazium, especially in the Old World, have the Callipteris scale type but exhibit free veins. It is unknown whether they form a monophyletic group with species of Callipteris having anastomosing veins. The species lectotypified in the present paper all exhibit the Callipteris scale type but have free veins. Sano et al. (2000) presented preliminary results based on chloroplast rbcL gene sequences for the phylogeny of the tribe Physematieae, which includes Diplazium and Callipteris. Their analysis included four species of Diplazium with the Callipteris scale type, but more species need to be included in future analyses to determine whether the Callipteris scale type defines a monophyletic group. Until phylogenetic studies using DNA sequences confirm that the Callipteris scale type forms a monophyletic group, we refrain from making new combinations in Callipteris for those species of Diplazium with free veins and the unique scale type. The present paper is a result of studies of Diplazium at BM, K, P, UAMIZ, and US. In general, the lectotypes were chosen based on their completeness and how well they agreed with the original protologues. Diplazium atratum H. Christ, Philipp. J. Sci. 2 C: 163. 1907. Athyrium atratum (H. Christ) Copel., Philipp. J. Sci. 3: 293. 1908. Lectotype (here PACHECO & MORAN: LECTOTYPIFICATION IN DIPLAZIUM 91 1100 m, Mar 1906, The other syntype is Foxworthy 714 (P!), which was collected at the same ocality on the same date. We designate Foxworthy 683 as the lectotype )ecause it is the more complete specimen. H. Christ, J. Bot. 19: 67. 1905. Diplazium virescens (H. Christ) Sa. Kurata, J. Geobot. (Kanazawa) 7: 77. (H. Christ) Ching, Acta Phytotax. Sin. 9: 47. 1964. Lectotype (here designated): Vietnam. Annam, vallee du Long-Gianh, 1903, Cadiere 88 (P!). Diplazium crinipes Ching, Bull. Fan Mem. Inst. Biol. 2: 207, tab. 23-24. 1931. Allantodia crinipes (Ching) Ching, Acta Phytotax. Sin. 9: 53. 1964. Lectotype (here designated): China. Hongkong, New Territory, Ma-on Shan, 3 Feb 1907, Matthew s.n. (K!, photos US!, UAMIZ!). The other syntype is: China. Kwangtung: North River, Tei Loy Hap, 23 Nov 1907, Matthew s.n. (K!, photo US). The Matthew s.n. specimen collected on 3 February 1907 is designated as the lectotype because it is the more complete of the two. Diplazium megaphyllum (Baker) H. Christ, Bull. Herb. Boissier 6: 961. 1898. Asplenium megaphyllum Baker, J. Bot. 264. 1890. Allantodia megaphylla (Baker) Ching, Acta Phytotax. Sin. 9: 50. 1964. Lectotype (here designated): China. Tonkin, Forets du Mont-Bavi, 800 m, 21 Jul 1886, Balansa 1836 (P!; isolectotypes: K! fragment BM!). The other syntype is: China, Tonkin, Forets du Mont Bavi, 1888, Balansa 1846 (K!, P!). We designate Balansa 1836 (P!) as the lectotype because it is more complete and, importantly, the petiole scales can be clearly seen. Diplazium polypodioides Blume var. vestitum (C. B. Clarke) K. Iwats., H. Ohba & S. B. Malla, Himalayan PL 1 (Univ. Mus. Univ. Tokyo Bull. 31): 319. 1988. Asplenium polypodioides Mett. var. vestitum C. B. Clarke, Trans. Linn. Soc. London, Bot. Ser. 2, 1: 501. 1880. Lectotype (here designated): India. Darjeeling, 6500 ft., 19 Jun 1884, Clarke 35382 (K!). The other syntype is: India, Darjeeling, 5500 ft., 17 Aug 1869, Baker 8646 (K!). Because Clarke 35382 is more complete, it is designated as the lectotype. Diplazium sechellarum (Baker) C. Chr., Ind. Fil. 238. 1906. Asplenium sechellarum Baker, Syn. fil. 91. 1874. Lectotype (here designated): Madagascar, Boivin s.n. (K!). 92 AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 2 (2003) Two other specimens were cited in the protologue: Seychelles, without locality, Bouton s.n. (K!); and Seychelles, Sep 1871, Home 165 (K!). We choose Boivin s.n. as the lectotype because it best agrees with the protologue. Diplazium sikkimense (C. B. Clarke) C. Chr., Contr. U.S. Nat. Herb. 26: 304. 1931. Asplenium sikkimense C. B. Clarke, Trans. Linn. Soc. London, Bot. 1: 500, tab. 65, fig. 1. 1880. Allantodia sikkimensis (C. B. Clarke) Ching, Acta Phytotax. Sin. 9: 56. 1964. Lectotype (here designated): India. Sikkim, Hooker s.n. (K!). Someone wrote "lectotype" on the Hooker specimen, but we cannot find any previous publication lectotypifying this name. The other syntype was: India, near the Teesta, 500 ft., Clarke s.n. (K). & K.U.KRAMER. V390. Subfamily Athyrioideae. Pp. 130-144 in K. U. Kramer & eds. The d genera of vascular plants. Vol. I. Pteridophytes and Gy Springer-Verlag, Berl L. & R. C. MOR AN. 1999. Monograph of the neotropical species of Call anastomosing veins (Woodsiaceae). Brittonia 51:343-388. 3, R., M. TAKAMIYA, M. ITO, S. KURITA & M. HASABE. 2000. Phylogeny c Physematieae (Dryopteridaceae), based on Chloroplast rbcL g< Phylogenetics and Evolution 15:403-413. KATO.M. PACHECO, SHORTER NOTES Botrychium lanceolatum subsp. angustisegmentum in Ohio.—In the treatment of Ophioglossaceae, (1993, pp. 85-106, in FNA Editorial Committee, Flora of North America North of Mexio, Volume 2. Pteridophytes and Gymnosperms) Wagner and Wagner reported the distribution of the narrow triangle moonwort, Botrychium lanceolatum subsp. angustisegmentum, as encompassing an area extending from Ontario's Lake Superior coastline to eastern Quebec and southern Labrador, south along the Appalachian Mountains to westernmost Virginia and North Carolina and easternmost Tennessee and Kentucky, and extending west to northern Wisconsin and the northwest corner of Minnesota. A disjunction occurs in the northern Rocky Mountains from northwestern Montana to northern British Columbia and the southern Northwest Territories. All but the southwestern corner of Ohio was included in the distribution of the subspecies. However, for 2000-2001 the Ohio Department of Natural Resources listed B. lanceolatum as extirpated because no Ohio collections were documented for a period of over 20 years (Ohio Department of Natural Resources. 2000. Ohio Rare Plant List, http:// www.ohiodnr.com/dnap/heritage/plantlst.html). We report here two Ohio populations of B. lanceolatum subsp. angustisegmentum that confirm the continued presence of the species in Ohio. While examining Botrychium specimens at the University of Michigan Herbarium (MICH) we encountered a 1970 collection [Wagner and D. Demay 70467A) of the species from Cantwell Cliffs in Hocking Hills State Park, Hocking Co., OH. In June of 2000 we searched Cantwell Cliffs for B. lanceolatum subsp. angustisegmentum and found approximately 15 sporophytes growing in a level, beech-maple mesophytic forest immediately adjacent to a small stream. Infrequent disturbance of the site by flooding appears probable. Woody species closely associated included Acer saccharum Marshall, Fagus grandifolia Ehrh., Tsuga canadensis (L.) Carrire, Liriodendron tulipifera L., Lindera benzoin (L.) Blume, and Ulmus rubra Muhl. Herbaceous associates included Asarum canadense L., Cimicifuga racemosa (L.) Nutt., Osmorhiza longistylis (Torr.) DC, and Tiarella cordifolia L. Pteridophytes at the site were Dryopteris intermedia (Muhl. ex Wild.) A. Gray, Osmunda cinnamomea L., Sceptridium dissectum (Spreng.) Lyon, and Thelypteris noveboracensis (L.) Nieuwl. A brief search in June 2001 revealed only eight individuals. This past June (2002) we intensively searched the site and found 69 sporophytes. A voucher specimen [Hauk et al. 626) was deposited at the Ohio State University Herbarium (OSU). Collections between 1970 and 2000 are not known (to us), and re-establishment may explain the current presence of the population. However, it seems more probable that this population has remained intact for at least the last 30 years, and the demography of other Ophioglossaceae species is consistent with this hypothesis 94 AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 2 (2003) (Montgomery, 1990, Araer. Fern J. 76:7; Kelley, 1994, New Zealand J. Bot. 32:393-400; Johnson-Groh, 1997 in Report to Minnesota Dept. Nat. Resources, St. Paul, MN). A second population of 14 B. lanceolatum subsp. angustisegmentum plants was discovered in Ashtabula Co., OH in August of 2001 by James Bissell of the Cleveland Museum of Natural History (CLM). The population was located beneath a rich mixed forest on a river terrace of the Ashtabula River in Sheffield Twp. with a canopy predominately of Acer saccharum and Liriodendron tulipifera and some scattered Tsuga canadensis. A voucher [JKB:2001:110) was deposited at CLM. The physical distance between the Hocking Co. and Ashtabula Co. sites (—180 mi.] and their apparent similarities in habitat suggest that B. lanceolatum subsp. angustisegmentum may occur in similar habitats across portions of Ohio. Thus, the species may be more common in Ohio than our current knowledge indicates, and its small size probably contributes to its oversight by collectors. We thank Jessica Budke, Emily Gerstle, Heather Hawke, and Larkin Kennedy for field assistance. We also thank James Bissell and Jim McCormak for providing information on the Ashtabula Co. population.—WARREN D. HAUK, Department of Biology, Denison University, Granville, OH 43023 and MICHAEL S. BARKER, Department of Botany, Miami University, Oxford, OH, 45056. Hawai'i's Ferns and Fern Allies, by Daniel D. Palmer. 2003. University of Hawaii Press, Honolulu, ix, 325 pp. illus. Hardcover [ISBN 0-8248-2522-5] $60.00. Daniel D. Palmer, longtime resident of Hawaii, and dermatologist by profession, has spent much of his spare time studying the local ferns and has now published the results of these efforts in this exceptionally well prepared and useful guide to the Hawaiian pteridophytes. Amateurs, fern enthusiasts, field biologists, professional botanists as well as all those interested in the Hawaiian biota can now benefit from his work. It has been a long wait. The first and only comprehensive publication on the Hawaiian pteridophytes was published in 1888 by William Hillebrand in his Flora of the Hawaiian Islands. Winifred Robinson, in 1912-1914, published, in four parts, A Taxonomic Study of the Pteridophyta of the Hawaiian Islands that was incomplete, inadequate and proved not to be particularly helpful in the identification of the ferns. Since then, those interested in the Hawaiian ferns and fern allies have had to rely on a series of checklists by various authors, a few published, but many duplicated and distributed informally. Each list is different and it is often difficult to compare listed binomials in one list to those in another. A few illustrated booklets have been published, but these included only a few of the ferns. It has indeed been difficult to identify the local ferns in the absence of a comprehensive, contemporary publication. Palmer has come to our assistance with the publication of this manual. He presents us with a survey of all species recorded on the Islands. A total of 221 taxa are recognized and included in the book, each one is described and virtually each is accompanied by an illustration. Palmer has had to decide which families, genera and species to recognize, and not all fern taxonomists will agree with his decisions, but he provides a clear justification for his choices. A key to the genera of the ferns and one to the genera of the fern allies precedes the alphabetically arranged generic treatments. There is a description for each genus. Each species treatment provides the scientific name, its etymology, whether endemic, indigenous or naturalized, a listing of the published synonyms as well as unpublished names found in the widely circulated checklists (I find this particularly helpful), the vernacular names, followed by a description with the distinguishing characters in bold type. The habitat and distribution is given following the description, as is also a discussion of existing problems. The final paragraph, in bold type, gives a short diagnostic description. Silhouettes and line drawings accompany the species treatment. Many readers will find the "Quick-and-Easy" guide to the genera helpful. Following this tool, the user can reduce the choices of genera to a few that 96 AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 2 (2003) can then be checked against the descriptions and the illustrations. No other manual of Hawaiian pteridophytes has included illustrations of the species as in this publication. These are a valuable addition and a great aid in identification. Family descriptions and keys to the included genera are found in the Appendix. Here, also, is a glossary as well as an illustrated glossary. A list of references and index to scientific and vernacular names concludes the The Hawaiian pteridophyte flora includes 194 species, in 73 genera and 27 families. Of these 161 are native species, and 114 (71%) of them are en demic. There are 33 naturalized species now known to be growing in the Is lands. The high endemicity reflects the isolation of the island group Adaptive radiation into different island environments has led to speciation Variability is common in many Hawaiian species and gives rise to taxonomic problems. The genera Dryopteris and Asplenium serve as prime examples o: this variation. In such cases, Palmer describes, and frequently illustrates, the variation in the species and groups together species that are morphologicall similar and appear to be related. He has brought structure to what has been confusing. Palmer acknowledges the influence of Warren Herb Wagner. Herl was his mentor, encouraged his study, and frequently joined him in the field Palmer traveled extensively, consulted herbaria throughout the world, ex amined type specimens and conferred with fern specialists. This manual re fleets the extensive research done by Palmer, and it is clearly his individua work. Not all the taxonomic problems have been solved, but when more study is needed this is clearly indicated. This work brings together informa tion that can serve as the catalyst for many studies. Hawai'i's Ferns and Fern Allies is a long awaited and much requested man ual of the Hawaiian pteridophytes. Here, in one volume, is a guide to all o the fern and fern allies of the Islands that will be welcomed by professional! and amateurs alike. This manual is well researched, detailed and comprehen sive. It is an essential addition to the library of all those interested in pterido phytes as well those interested in Hawaiian plants and in island floras — KENNETH A. WILSON, Museum of Natural History of Los Angeles County, 900 Exposition Boulevard, Los Angeles, CA 90007. ,__• illiiii 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 prepared may be returned for revision prior to review. Submit manuscripts in triplicate (xerocopies acceptable), including review copies of illustrations and originals of illustrations. After review, submission of final versions of manuscripts on diskette (in PC- or Mac-compatible formats) is strongly encouraged. Use standard 8^2 by 11 inch paper of good quality, not "erasable" paper. Double space manuscripts throughout, including title, authors' names and addresses, short, informative abstract, text (including heads and keys), literature cited, tables (separate from text), and figure captions (grouped as consecutive paragraphs separate from figures). Arrange parts of manuscript in order just given. Include author's name and page number in upper right corner of every sheet. Provide margins of at least 25 mm all around on typed pages. Do not submit right-justified copy, avoid footnotes, and do not break words at ends of lines. Make table headings and figure captions self-explanatory. Use S.I. (metric) units for all measures (e.g., distance, elevation, weight) unless quoted or cited from another source (e.g., specimen citations). For nomenclatural matter (i.e., synonymy and typification), use one paragraph per basionym (see Regnum Veg. 58:39-40. 1968). Abbreviate titles of serial publications according to Botanico-Periodicuin-Hunrianunt (Lawrence et al., 1968, Hunt Botanical Library, Pittsburgh) and its supplement (1991). References cited only as part of nomenclatural matter are not included in literature cited. For shorter notes and reviews, omit the abstract and put all references parenthetically in text. Use Index Herbariorum (Regnum Veg. 120:1-693. 1990) for designations of herIllustrations should be proportioned to fit page width with caption on the same page. Provide margins of at least 25 mm on all illustrations. For continuous-tone illustrations, design originals for reproduction without reduction or by uniform amount. In composite blocks, abut edges of adjacent photographs. Avoid combining continuous-tone and line-copy in single illustrations or blocks. Coordinate sequence and numbering of figures (and of tables) with order of citation in text. Explain scales and symbols in figures themselves, not in captions. Include a scale and reference to latitude and longitude in each map. Proofs and reprint order forms are sent to authors by the printer. Authors should send corrected proofs to the editor and reprint orders to the printer. Authors will be assessed charges for extensive alterations made after type has been set. For other matter of form or style, consult recent issues of American Fern Journal and 77?^ Chicago Manual of Style, 14th ed. (1993, Univ. Chicago Press, Chicago). Occasionally, departure from these guidelines may be justified. Authors are encouraged to consult the editor for assistance with any aspect of manuscript preparation. Papers longer than 32 printed pages may be sent to the Editor of Pteridologia (Memoir Editor, see cover 2). PTERIDOLOGIA ISSUES IN PRINT 1. Wagner, David H. 1979. Systematics of Polystichum in Western North America North of Mexico. 64 pp. $10.00 postpaid. 2A. Lellinger, David B. 1989. The Ferns and Fern-allies of Costa Rica, Panama, and the Choco (Part 1: Psilotaceae through Dicksoniaceae). 364 pp. $32.00 postpaid. 3. Lellinger, David B. 2002. A Modern Multilingual Glossary for Taxonomic Pteridology. 263 pp. $28.00 postpaid. Send your order with a check or money order to: American Fern Society, Inc., c/o U.S. National Herbarium MRC-166, Smithsonian Institution, Washington, DC 20560. AMERICAN FERN JOURNAL ON MICROFICHE Volumes 1-61 of the American Fern Journal are available as archival quality, silver positive microfiches. Single volumes or the entire run may be purchased. The fiches are easily read with 10X or greater magnification (using a dissecting microscope and transmitted illumination or a fiche reader). Silver negative microfiches of vols. 1-50 are also available. The price is $4.00 per volume or $244.00 per set of 61 volumes, postpaid. Send your inquiry or order with a check or money order to: American Fern Society, Inc., c/o Dr. James D. Montgomery, Ecology III, Inc., R.D. 1, Box 1795, Berwick, PA 18603. VISIT THE AMERICAN FERN SOCIETY'S WORLD WIDE WEB HOMEPAGE: http ://w ww.amerfernsoc.org/ AMERICAN FERN JOURNAL Volume 93 Number3 July-September 2003 QUARTERLY JOURNAL OF THE AMERICAN FERN SOCIETY A Karyotype Comparison B« A Re-evaluation of Isoetes s Rapid Gametophyte Maturation in Ophioglossum crotalaphoroides A Modern Multilingual Glossary for Taxonomic Pteridology Index to Distribution Maps of Pteridophytes in Asia, 2nd Editic Deai The American Fern Society Council for 2003 CHRISTOPHER H. HAUFLER, Dept. of Botany. University of Kansas. Laurence. KS 66045-2016. TOM RANKER, University Museum, Campus Box 265, University of Colorado, Boulder, CO 80309-0265. W. CARL TAYLOR, 800 W. Wells St., Milwaukee Public Museum, Milwaukee, WI 53233-1478. JAMES D. CAPONETTI, Dept. of Botany, University of Tennessee, Knoxville, TN 37916-1110. GEORGE YATSKIEVYCH, Missouri Botanical Garden, P. O. Box 299, St. Louis, MO 63166-0299. Membership S'< < retur\ JAMES D. MONTGOMERY, Ecology HI, 804 Salem Blvd., Berwick, PA 18603-9801. Back Issues Curaioi R. JAMES HICKEY, Botany Dept . Miami University. Oxford, OH 45056. Journal Editor DAVID B. LELLINGER, U.S. National Herbarium MRC-166, Smithsonian Institution. Washington, DC 20560-0166. Memoir Editor CINDY JOHNSON-GROH. Dept. of Biology, Gustavus Adolphus College, 800 W. College Ave., St. Peter, MN 56082-1498. Bulletin Editor American Fern Journal EDITOR Botany Department,Miami University, Oxford, OH 45056, ph.(513, ; R. JAMES HICKEY GERALD J. GASTON Y Dept. oi IN 47405-6801 GARY K. GREER Biologv IX tale, MI 49401 CHRISTOPHER H. HAUFLER .... Dept. of Botany, University of Kansas, Lawrence, KS 66045-2106 ROBBIN C. MORAN NewV.rt \i 10458-5 12b 2801 S. l AR 72204 The -American Fern Journal" (ISSN 0002-8444) is an illustrated quarterly devoted to the general '••• - .., !•'. • •-,..•• • ' ••••• .-. " -• • -•• . Montgomery. Ecology III, 804 Salem ip should be sent to the Membership - $140.00 "MASTER: Send address changes to AMERICAN FER: SPORE EXCHANGE 6 Ibbetson Ave.. Downey, CA 90242-5050. is Director. Spores exchanged GIFTS AND BEQUESTS sed to the Secretary. American Fern Journal 93(3):97-l 15 (2003) ^L^ ' ' '""* GARDEN LIBRARY Soil Spore Bank of Ferns in a Gallery Forest of the Ecological Station of Panga, Uberlandia, MG, Brazil MARLI A. RANAL lade Federal de Uberlandig Uberlandia, MG, Brasil Ans i R \i: i .—The soil spore bank of ferns is a biotic component of plant communities, important for regeneration processes, population dynamics, and conservation programs Kmh veai it is enrii In.I when new units are incorporated, and impoverished when the] are Inst bj predation, loss of \ lability, or by germination. Soil was collected in three microhahitats of the gallery forest of the Panga Stream, at four depths, in the wet and the dry seasons In general, independent of the season, dike samples presented lower numbers of viable spores when compared to samples from the 'middle' and edge' of the forest. The number of viable spores and the number of fern species represented decreased with depth. At the end of the dry season, the number of viable spores decreased only in the first centimeters of the soil. Viable spores of thirteen terrestrial species were registered in the soil ot this gallen Forest. The presence of viable spores in the sod alter six months drought and in deeper soil layers shows the existence of a persistent soil spore bank in the gallery forest of the Panga Stream. A diaspore bank is a biotic component of soil where dispersion units in quiescence or dormancy are found. This biological store can be enriched or impoverished each year, when new units are incorporated, or lost by predation, loss of viability, or germination. Therefore, the diaspore bank is a dynamic component that represents a continuous source of dispersion units important for regeneration processes and population dynamics of plant ((immunities. It is this biological and genetic potential in the soil which permits the local survival of the species during unfavorable environmental conditions or disturbances. Most of the information about diaspore banks is related to the soil seed banks of plant communities (Fenner, 1985, 1995; Leek et al., 1989; Baskin and Baskin, 1998). There is little information on the diaspore banks of bryophytes (Carroll and Ashton, 1965; During and ter Horst, 1983; During et al., 1987; Leek and Simpson, 1987) and fern spore banks (Carroll and Ashton, 1965; Wee, 1974; Strickler and Edgerton, 1976; During et al., 1987; Leek and Simpson, 1987; Hamilton, 1988; Lindsay and Dyer, 1990; Milberg, 1991; Dyer and Lindsay, 1992; Milberg and Anderson, 1994; Penrod and McCormick, 1996; Raffaele, 1996; Schneller and Holderegger, 1996). Sometimes the concept of banks must be amplified to include cases like the belowground structure bank of Botrychium, which is formed by gemmae, gametophytes, sporelings, and spores (Johnson-Groh et al., 2002). For Tropical America, where there are about 3000 fern species, there is little information regarding spore banks (Perez-Garcia et al., 1982; Simabukuro et al, 1998, 1999). Viable fern spores are encountered in different kinds of soil under natural vegetation or agricultural crops, with or without sporophytes near the sample 98 AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 3 (2003) site, and in barren soil (Strickler and Edgerton, 1976; During and ter Horst, 1983; Clymo and Duckett, 1986; Leek and Simpson, 1987; Milberg, 1991; Dyer, 1994). These data confirm that fern spore dispersion occurs over long distances as indicated by Conant (1978) and Page (1979), among other authors, and that viability is maintained under natural conditions and during cultivation of the soil at least for non-chlorophyllous spore species. Soil spore banks of ferns are believed to play an important role in the reproductive success of many species, creating numerous opportunities for spore germination and gametophyte establishment after any form of soil disturbance (Lindsay et al, 1992; Dyer, 1994). Moreover, a large spore bank means that many gametophytes, originating from many different sporophytes, could develop at the same time in a limited space after disturbance of the vegetation, increasing the chance for mating of different genotypes (Milberg, 1991). Asexual reproduction by gametophytic gemmae in Trichomanes speciosum Willd. appears to be the principal kind of dispersion of the species in recent times, and the genetic variability may be attributed to sexual reproduction and spore dispersal in historic times under more favourable climatic conditions (Rumsey et al, 1999). For this type of endangered species, with sporophytes extremely rare and vulnerable in the actual European climatic conditions as indicated by the authors, the soil spore bank could participate in the restoration of species heterozygosity. Soil spore banks also play a relevant role in conservation programs (Dyer and Lindsay, 1996), permitting the propagation of rare or endangered species by means of small soil samples collected without environmental disturbances (Lindsay et al, 1992; Dyer, 1994). The purpose of this paper is to characterize the fern soil spore bank for three microhabitats included in the gallery forest of the Ecological Station of Panga, Uberlandia-MG, Brazil. MATERIALS AND METHODS The Ecologial Station of Panga is situated in Uberlandia, State of Minas Gerais, Brazil (19°09'20"-19°11'10" S, 48°23'20"-48°24'35" W, ca. 800 m altitude). Until 1984 the area occupied by the Ecological Station of Panga was a farm with agriculture and cattle breeding as its principal activities. The owners preserved the gallery forest. In 1985 the Federal University of Uberlandia bought the area and the vegetation recovered naturally. Today it is considered a representative area of cerrado for Central Brazil. Its 409.5 ha are occupied by cerrado sensu lato (Schiavini and Araiijo, 1989; Ratter, 1992). Gallery forest, a component of the mesophytic forests of the Ecological Station of Panga, is situated along Panga Stream. The approximately 1.0 hectare area, from which the soil samples were collected, is situated on the left bank of the stream, 900 meters from the main road (Fig. 1). 'Dike', 'middle' and 'edge' are three microhabitats described by Schiavini (1992, 1997) for this gallery forest. The 'dike' is a natural elevation that borders the stream and extends 10 m out from the stream bank. According to Schiavini RANAL: SOIL SPORE BANK IN A GALLERY FOREST on and vegetation map of the Ecological Station of Panga ( (1992, 1997), fluvial sediments are deposited in this area, making the surface higher than 'middle'. Its soil consists of 85.2% sand, 5.5% silt, 9.3% clay, and 2.9% organic material, having good drainage. The 'middle' is a continuous depression adjacent to the 'dike', varying in width from zero to 40 m along of the stream. This microhabitat presents clay hydromorphic soil, consisting of 52.1% sand, 16.4% silt, 30.6% clay, and 9.2% organic material. It is flooded seasonally and saturated with water most of the year. The 'edge' of the forest is approximately 10 m wide. It has a Dark-Red Latosol (Oxisol) and a hydromorphic soil, depending on location and depth, with 75.6% sand, 9.3% silt, 15% clay, and 4.3% organic material. The water table in this microhabitat can vary in depth from just below the soil surface, for most of the year, to more than 0.5 m deep. The region is included in Koppen's climatic system (1948) as Aw; that is, a tropical wet climate with dry winter. The wet season occurs during the summer, from October to March, and the dry season during the winter, from April to September (Fig. 2). In February 1997, September 1997, and September 1998, soil was collected at four depths, in the three microhabitats of the gallery forest. In April 1998, soil was collected at two depths from the 'edge' microhabitat (Table 1). For each collection date, five holes of 40 cm depth and 900 cm2 of opening (soil collection sites), approximately 10 m distant from each other were opened in each microhabitat. Each soil collection site was used only once. Soil of . 5th METEOROLOGICAL DISTRICT (800m) UBERLANDIA - MG [18] 22.2 °C 1599.C different depths was collected by introducing into each hole plastic tubes with a diameter of 2 cm, parallel to the soil surface. After collection, from the bottom to the top of the hole to prevent contamination, each portion of soil was stored in a plastic bag that was labelled and closed immediately. In the laboratory, soil was homogenized manually inside the bags, transferred to quadrangular, transparent, covered plastic boxes (experimental units), and moistened with distilled water. The superficial area of cultured soil was used to calculate the number of gametophytes and sporophytes formed per square centimeter. The number of gametophytes was the criterion used to evaluate viable spores in the soil samples. As indicated in Table 1, for the February 1997 and September 1998 collections, each portion of soil was divided in two sub-portions. Thus, 60 experimental units (boxes containing soil) could be examined daily for counting gametophytes and 60 experimental units were maintained intact for counting sporophytes at the end of the experiments. Culture conditions are presented in Table 1. All cultures were periodically moistened with distilled water and, after two months of culture when gametophytes and young sporophytes presented the first signals of chlorosis, with nutrient solution (Meyer et al, 1963) every 15 days. Sterilized soil controls (10 replicates) were maintained under the same laboratory conditions. Forty days after each collection, when the gametophytes were at least 1 mm wide, the samples were examined daily under a stereomicroscope to count and remove gametophytes. Because gametophytes were removed at a relatively young age, it was possible to take them out without removing soil particles. The gametophytes removed from the soil were subsequently placed on a microscope slide and examined to search for additional germinating spores or gametophytes at the filamentous stage that might have been undetected under the stereomicroscope. The counting was concluded when the cultures were four months old. RANAL: SOIL SPORE BANK IN A GALLERY FOREST 101 Sporophytes were counted between three and four months after the initiation of the experiments, in intact soil of the duplicate cultures collected in February 1997 and September 1998. The criterion for counting sporophytes was the presence of a perceptible crozier when viewed under stereomicroscope. At the end of the experiments young sporophytes were transplanted to bags containing soil and were maintained under greenhouse conditions until the production of fertile leaves when they were collected. The collected sporophytes were prepared and deposited at HUFU and SP. Some specimens of Thelypteris were also deposited at UC and SI. The experimental unit used to calculate the percentage of gametophytes forming sporophytes consisted of two duplicates. As was described above, for February 1997 and September 1998 collections, one duplicate of soil was used for counting gametophytes without replacement, and the other for counting sporophytes at the end of the experiments. Thus, the percentage was calculated as the proportion of sporophytes to gametophytes in the duplicates. Systematic sampling was used to collect soil samples, due to the known differences among the three analysed microhabitats. The experimental units were randomly distributed in laboratory conditions. The number of gametophytes and sporophytes formed per square centimeter of cultured soil, as well as the percentage of gametophytes forming sporophytes, were submitted to the Shapiro-Wilk test. As part of the original and transformed data showed nonnormality, the Mann-Whitney test was used for pairwise comparisons within microhabitats, depths, and collection dates. RESULTS Gametophyte densities on cultured soil ranged from 0.13 to 29.52 gametophytes per square centimeter and, in general, 'dike' presented soil with lower mean numbers of viable spores than the other microhabitats (Table 2). The number of viable spores was higher at 2-4 and 5-7 cm depth than at 1517 and 20-22 cm depth. The 'edge' of the forest showed fewer viable spores at 2-4 cm depth in April 1998 than in February 1997 collection, at the same depth (Tables 2, 3). Data from the April collection showed the existence of viable spores below 20-22 cm. All soil sample controls remained free of gametophytes during the experimental period, indicating no contamination of the cultures. There is seasonality in the size of the soil spore bank of the gallery forest in the first centimeters of soil column as shown in Tables 2 and 3. Soil collected in February 1997, during the wet season, was richer in viable spores than soil collected in September 1997, at the end of the dry season. Soil samples collected at the end of dry season presented statistical differences between consecutive years only for the first centimeters of soil collected in the 'dike'. Soil collected in September 1997 presented lower number of viable spores than that collected in September 1998. As there was high variability among the replicates of the same sample, the statistical test used was not capable of detecting other differences (see the standard error of the means). AMERICAN FERN f IS £ S £ S S333 jjf CM u in in ci CM in in a CM in in a MO CM in m o CM m in a RANAL: SOU S3 2. Gametophytes (mean ± standard error) produced in soil collecte 1 in the gallery forest of the Ecological Station of Panga, Uberlandia, MG. TABLE 'Edge' gem " .10 bB .08 bC B 0.8860 0.8423 0.6585 3.12 ± 0.35 aA 0.27 ± 0.09 bB 0.13 ± 0.05 bB 0.7717 0.9232 0.8346 W 6.76 ± 1.56 aAB 4.20 ± 0.71 aB 1.37 ± 0.37 aC g cm2: gametophxtrs pci square . mtimrt.-r ut tin* < ulliirrd •-< >i I: IT. Sli.i j >ir< i-\Vilk test (a = 0.05), where boldfaced values indicate normality of the studied i ham teristh in the population (P > 0.05); mean followed by the same lower case letter in each line and by the same capital letter in each column, within the same collection date, are not si -ed on the Mann- The number of sporophytes formed on the cultured soil decreased with depth, as was observed also for the number of gametophytes formed (Table 4). Similar numbers of sporophytes were formed in the three microhabitats analyzed. The reproductive success of the viable spores encountered in the soil, calculated as the percentage of gametophytes forming sporophytes, ranged from 0.76% at 20-22 cm depth in soil of the 'edge' of the forest to 63.33% at the same depth in soil of the 'dike', both values registered for February 1997 collection (Table 5). Due to high variability among replicates of the same sample, few statistical differences in relation to depth and microhabitats were detected. The sporophyte frequency per species for soil collected in September 1998 shows that Thelypteris species predominated in the three microhabitats and four depths (Table 6). This genus was better represented than the others, presenting nine species, while Blechnum presented two species and the other genus one species each (Table 7). Sporophytes of 13 terrestrial species were registered in the analysed soil of the gallery forest of Panga Stream (Table 7). Five of these species were found from 2-4 to 30-32 cm depth, in the three microhabitats of the gallery forest (Blechnum brasiliense Desv., Macrothelypteris torresiana (Gaud.) Ching, Pityrogramma calomelanos (L.) Link var. calomelanos, Thelypteris conspersa (Schrad.) A. R. Sm., and T. opposita (Vahl) Ching). The^ September 1998 collection provided more complete information about species composition of the soil spore bank due to the high survival rate of the sporophytes after ' ! : ^, ! :: ' ^. . ' < . M. -, . v::; :-' ments are included Q Table 2. Collection date 'Edge' Depth f'x alue P value CVII ue P val e u a°• 40 u'0753 — — — — t/v lue P value Feb X Sep 1997 15-17 Feb 1997 X Apr 1 2-4 15-17 20-22 P: probability to accept significantly different; i 0.;i71() 0.035 I 81 5 i Ei it HE J;- lii •VE?• ms that the two medians are s g'lit'nta : /': '.T.si'i! - transplanting. Considering the four collection dates, a similar number of species was observed in the three microhabitats of the forest. The number of species decreased with depth (Table 7, September 1998). DISCUSSION The range of viable spores included in soil samples of the gallery forest of Panga Stream was similar to that reported by Dyer and Lindsay (1992) for soil samples collected in Durham, N.C., U.S.A. 'Dike' samples presented smaller numbers of viable spores when compared to the other microhabitats, perhaps as a consequence of the seasonal leaching of this microhabitat. Depending on the rainfall, there is a fast overflow of the stream, washing the litter deposited in the 'dike' toward the 'middle'. Alluvial deposition, consisting mainly of sand, occurs at the same time. Water reflux towards the streambed occurs rapidly, cleaning the sandy soil of the 'dike'. Movement of spores down through the soil probably occurs as the result of the percolation of rain water, rather than by inundation. Preliminary data about the distribution of adult sporophytes in the studied area (personal observation), evaluated using one transect of 190 m2 per microhabitat, with observations in 10 quadrats of 1 m2 per transect, indicated no significant differences between the three microhabitats [W = 0.607, P = 0.7381 for Kruskal-Wallis test). 'Dike' presented 0.9 ± 1.45, 'middle' 0.5 ± 0.97, and 'edge' 0.3 ± 0.48 sporophytes per square meter (mean ± standard deviation). These results indicate that the differences between microhabitats in soil spore bank densities are not a consequence of differential adult sporophyte distribution in the studied area. A decrease in the number of viable spores with increasing depth was also registered by Leek and Simpson (1987) for high marsh, cattail, and shrub forest in a Delaware River freshwater tidal wetland, by Lindsay and Dyer (1990) for RANAL: SOIL SPORE BANK IN A GALLERY FOREST °n W Sep.19 g8 ^ W 0.22 ± 0.13 aB 0.27 ± 0.17 aB 0.92 ± 0.64 aAB 0.18 ±0.11 aB ao5h. 'Middle 'Dike' 0.8478 1.32 ±0.72 0.8327 0.16 ± 0.10 0.7476 0.26 ± 0.18 0.8387 2.29 ± 0.54 0.7425 0.5521 'Edge' W aAB aB aB aA 0.06 ± 0.04 aB 0.09 ± 0.06 aB IV S( .60 aB o.mi:?r, 0.9017 0.7679 0.7612 0.6965 :\.77 i^ .26 aA 0.9465 .44 aAB 0.8179 11.44 ' I .18 aB 0.9642 30 aB 0.6884 sporophyt s per square centimeter of the cultured soil; W: Shapiro-W lues indicate normality of the studied characteristic in the population (P > d by the same lower case letter in each line and by the sam capital Idler in forests near Edinburgh, Scotland, by Dyer and Lindsay (1992) for several places in North Carolina and Scotland, and by Simabukuro et al. (1998, 1999) for areas of cerrado in Sao Paulo, Brazil. This pattern is also similar to that observed in soil seed banks of forest, savanna, and farmlands of tropical regions (Garwood, 1989). According to Fenner (1995), all studies of vertical distribution of seeds in soil indicate that in undisturbed sites the vast majority of seeds are found in the first 2-5 cm of soil, with a notable decline in numbers with depth. Gametophytes and sporophytes developed more slowly on soil collected in the gallery forest of Panga Stream from 15-17 to 30-32 cm depth than in the more superficial layers, although periodically moistened with nutrient solution. Moreover, some sporophytes had anomalous morphology although transplanted to good soil after their formation. These observations indicate that some of the spores located at greater depths, and which germinated under laboratory conditions, could be older than spores included in soil collected from the first centimeters. Anomalies and slow gametophyte development observed for some species when old spores were used for culture in laboratory conditions (Raghavan, 1980) reinforce this idea. Probably the decrease of viable spores observed at the end of the dry season, especially in the first centimeters of the soil, is in part a consequence of death by desiccation. On the other hand, the decrease in the size of the soil spore bank registered in April in relation to February shows that some spores can germinate from February to April when rainfall decreases gradually, but the soil has sufficient water accumulated during the wet season. Although phenology of the fern species of Ecological Station of Panga is unknown, periodic observations indicate that for some species production of new leaves occurs in October-November, at the beginning of the rainy season, and the production of fertile leaves occurs in December-January. Seasonality of FERN JOURNAL: VOLUME 93 NUMBER 3 i. Percentage of for soil collected in the ophvU-: forming sporophytes Df Ecological Station c S •Di k(.- nn riPnth (cm) .l.lt! S 43.72 27.01 42.67 63.33 "<> s ± 11.98 ± 18.51 ± 20.50 ±22.61 W aA aA aA aA 0.9455 0.7094 0.8747 0.7331 %g Feb. 1997 2-4 5-7 15-17 20-22 20.77 10.75 16.98 30.53 ± 2.50 aA ± 1.99 aB ± 11.39 aAB ±20.14 aAB Sep. 1998 2-4 26.00 ± 8.83 aA 0.8945 44.17 ± 6.22 aA 5-7 16.49 ± 8.12 abA 0.6856 45.93 ± 3.95 aA 15-17 9.56 ± 6.04 aA 0.7657 13.33 ± 8.16 aB iv error) calculated .-.dm. ".. 'Edge' g w 0.8940 20.01 ± 2.18 aA 0.9434 17.81 ± 4.89 aA 0.7694 26.85 ± 18.83 aAB 0.7726 0.76 ± 0.47 aB 0.8518 0.8863 0.7365 0.6888 0.9479 25.67 ± 9.16 aA 0.9895 20.90 ± 5.89 bA 0.6839 9.99 ± 6.02 aA 0.9077 0.9273 0.8105 % g: percentage of gametophytes forming sporophytes on surface of the cultured soil; 11: ShapiroWilk test [at = 0.05), where boldfaced values indicate normality of the studied characteristic in the population (P > 0.05); mean followed by the same lower case letter in each line and by the same capital letter in each column, within the same collection date, are not significantly different based fertile leaves was also observed for some species occurring in a mesophytic, semideciduous forest in the State of Sao Paulo, under similar rain distribution conditions (Ranal, 1995). In the gallery forest of Panga, spore dispersal occurs from December (precocious leaves) to March-April (late leaves), depending on the annual rainfall distribution. Thus, the seasonality of the soil spore bank observed for this gallery forest, especially in the first centimeters of soil column, may be a consequence of the seasonality in spore production and of the gradual loss of viability associated with desiccation of the soil that occurs during the dry season. Seasonality in soil spore banks was also registered in a flooded mountain meadow in Patagonia, Argentina (Raffaele, 1996). The soil spore bank of Dennstaedtia punctilobula (Michx.) Moore varied across preand post-dispersal seasons in two undisturbed hardwood forest sites in central Pennsylvania (Penrod and McCormick, 1996). According to Garwood (1989), unpredictable rainfall during the dry season also causes seed death in tropical regions. The distribution of rainfall registered in the region of Uberlandia in 1997 was atypical in relation to former years. In April 149.8 mm of precipitation was registered, while the mean of the previous 18 years was 87.0 mm; in June 105.1 mm was registered while the mean for the same 18-year period was 19.0 mm. Certainly abundant water in the soil, stimulating precocious germination, followed by low precipitation (36.3 mm in May and zero in July and August), was an important cause of the decrease of viable spores in the soil observed in September 1997 in relation to September 1998 for 'dike' of the forest. Moreover, the precipitation registered in 1997 (1811 mm) was higher than the mean of the previous 18 years (1599.64 mm). As a consequence the level of the stream increased, washing the 'dike' more than in 1998. In 1998 the precipitation was 1356.7 ze or species composition of the seed bank f (Garwood, 1989). RANAL: SOIL SPORE BANK IN A GALLERY FOREST soil collected in September 1998, Thelypteris spp. Pityrogramma calomelanos (L.) Pityrogramma calomelanos (L. Lygodium vt nustum Sw. asiliense Desv. fcalomelanos (L. Blechnum brasiliense Desv. Thelypteris spp. Pityrogramma calomelanos (L. Lygodium ve Blechnum b asifae/ise Desv. Although high numbers of viable spores were registered in soil collected in February 1997 at 2-4 cm depth in the 'middle' and in the 'edge' of the forest, only 20% of gametophytes produced sporophytes. These results, obtained in protected laboratory conditions, without biotic and abiotic disturbances, show the importance of the high number of spores produced by sporophytes for fern establishment. It means that the efficacy of viable spores for fern establishment and the role of the soil spore bank in the dynamics of the plant communities can be better inferred by looking at sporophytes formed. The efficacy of soil seed banks can be evaluated directly simply by counting seedlings formed, but in ferns, the gametophytic phase with its peculiar ecophysiological and reproductive characteristics can lead to different results. The 13 species found as viable spores in soil samples of the gallery forest represent about 25% of the 52 fern species currently registered for the Ecological Station of Panga (Prado and Ranal, unpublished data). These species represent an addition to the list of species capable of forming soil spore banks presented by Dyer and Lindsay (1992). The soil seed bank of this gallery forest, evaluated in 1998 and 1999, also presented lower diversity than the actual vegetation, with 17% of species present as viable seeds (Pereira, 1999). According to Fenner (1985), in frequently disturbed habitats the species composition of the seed bank and the vegetation are usually similar, but as the vegetation matures the disparity between the two increases, and in general seed banks have lower diversity than the aboveground vegetation. Several 3 NUMBER 3 (2003) KRN JOURNAL: 11!|l | 111111 i s § sJ s £ £ 3 3 J §a J §a £ > i 8 i% III ill f 5 1 |15 II I S. i-U IIIIII ess; lbs RANAL: SOIL SPORE BANK IN A GAI.LH ll Is I 1 I 111 si ii 1I I I 1 5 l| i 1| U I e| | o I illjl i! i III SM,! I in aitil ||S|I|ri|o|S| imnnfifi ||a|a|a|S| II Is HI * m ||2| | fi I I II J: |! |l|i! I ! i II Ii -•1 • -••- '•- :--==•= Ill :i s ss h= 1 i§ Jil M 1 ilil H = J = § | i •! ••§ "fi I MM i S11 | § I N^>-K> = IS J I Ii I m HI is mm jinii HI 111 mi bh M Jrf u t i RANAL: SOIL SPORE BANK IN A GALLERY FOREST 111 studies of angiosperm population dynamics in the gallery forest of Panga Stream indicated that the seedling bank, with high diversity, is an efficient form of regeneration in this forest (Oliveira and Schiavini, 1999). There is no information about fern population dynamics, but these results indicate that this forest could support only short-term disturbances and needs to be preserved. This kind of information is important to give support to conservation projects for gallery forests that are endangered, although protected by law. Decrease in the number of fern species occurring in the soil column was similar to the observations made for soil seed banks in several soil profiles in forest, savanna, and farmlands, according to a review presented by Garwood (1989) for tropical regions. In agricultural environments, due to the soil movement in relatively short periods, the vertical distribution of spores can be different, as was observed for seeds by Cavers and Benoit (1989). Although studies of soil spore banks are recent in relation to soil seed banks, a comparison of different results is difficult due to diverse methods of collection of soil and counting of viable spores. There is information concerning the frequency of viable spores per hectare (Wee, 1974), per square meter (Milberg, 1991), per square centimeter (Dyer and Lindsay, 1992) and viable spores per volume of soil (Hamilton, 1988). The numbers are mentioned in relation to gametophyte formation (Lindsay and Dyer, 1990; Milberg, 1991; Dyer and Lindsay, 1992; Milberg and Anderson, 1994), but according to Milberg (1991) some authors perhaps had counted the number of sporophytes formed and some of them did not specify their adopted criterion. Considering that one species can produce some exclusively male gametophytes and some exclusively female, which remain unfertilized due to incompatibility or other problems, it would seem more accurate to estimate viable spores by the number of gametophytes formed. High variability in the numbers of gametophytes formed in the soil collected in the gallery forest of Panga Stream shows that deposition of spores in the soil is heterogeneous. A similar condition exists for soil seed banks (Garwood, 1989; Baskin and Baskin, 1998). According to Fenner (1995) the heterogeneity of the horizontal distribution of the seeds, resulting in a high degree of variability between samples, is one of the main problems in obtaining good quantitative data on seed banks. Thus, it seems more appropriate to express the results as gametophytes per square centimeter in relation to cultured soil, without greater extrapolations. The counting of viable spores by means of number of gametophytes formed on the cultured soil is itself a relative measurement. Certainly some of the spores in the samples remains dormant due to the artificial culture conditions that vary between laboratories. This high variability among soil samples due to the heterogeneous horizontal distribution of the dispersion units makes it difficult to detect differences between microhabitats, depths or other factors. Another important point is the timing of observations. During the experimental period of this study, gametophytes were removed from the soil as soon as they reached 1-2 mm. In this manner, few of them died before counting. Soil used to count sporophytes that were maintained intact during 112 AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 3 (2003) three or four months showed several gametophytes in necrosis at the end of the experiment. Certainly, if the counting took place only at the end of the experiments, the number of gametophytes per square centimeter would be different because several gametophytes would be completely decomposed. Moreover, at three or four months of age, several gametophytes presented vegetative growth that made counting difficult because they formed wrinkled and crowded blades. Part of this vegetative growth was observed as young gametophytes formed in the mother tissue. These gametophytes could be separated and counted inadequately as resulting from spore germination when, in fact, they are vegetative growths of the mother gametophyte. On the other hand, the few rhizoids of young gametophytes removed from the cultured soil, method adopted in this study, can drag spores to the soil surface giving rise to an overvaluation of the soil spore bank. These technical problems pointed out mean that all methods used until now can not evaluate the absolute number of viable spores in the soil, but can be used only as an The literature accumulated during these years permits the conclusion that the soil seed bank can consist of a mixture of transient and persistent species (Fenner, 1995). A species is considered to be transient in the seed bank if its seeds do not persist in the soil in a viable condition for more than a year. These seeds depend on regeneration opportunities such as seasonal gap formation to start the germination process. The persistent seed banks usually characterize plant communities that are submitted to frequent and unpredictable disturbances where opportunities for colonization occur at random and the seeds must remain viable in the soil more than one year. Certainly there are intermediate species between these two described types (Fenner, 1995). These ideas were also presented by Thompson and Grime (1979), Simpson et al. (1989), and Bewley and Black (1994). Although there is less information about fern spore banks, analogous characteristics of germination physiology in seeds and fern spores permits the inference that these two types of species can also be found among the ferns. The principal difficulty in establishing these categories for fern species is the insufficient knowledge on the phenology of spore production, the longevity of spores for the majority of species, and the dynamics of the spore movement process through the soil column. A new and dynamic approach, more related to environmental questions, was given by Walck et al. (1996) and adopted by Baskin and Baskin (1998). The authors suggested that these two types of seed banks should be described in terms of germination seasons rather than age per se. Thus, a transient seed bank is composed of seeds that do not live beyond the first germination season following maturation, and a persistent seed bank is composed of seeds that can live until the second germination season or more than this (Baskin and Baskin, 1998). In this sense, data obtained at the end of the dry season for the gallery forest of Panga Stream could give an idea about the size of the persistent soil spore bank of that environment. The low pluviosity characteristic of the dry winter in the region causes a slower plant growth rate and new spore production will occur only in the next wet season. Thus, there is no new RANAL: SOIL SPORE BANK IN A GALLERY FOREST 113 significant addition to the spore stock from April to September and the germination season will occur in October-November, when rainfall starts. As the gallery forest of the Panga Stream presented higher numbers of viable spores and higher numbers of species in the first centimeters of soil column than in deep soil, it appears that this ecosystem is in good conservation status. Nevertheless, its lower diversity than the actual vegetation, typical of preserved environments, indicates that this forest must be protected against anthropic actions. Part of this study was supported by the Fundagao de Amparo a Pesquisa de Minas Gerais (FAPEMIG). Important constructive criticism and suggestions were given by Dr. Adrian F. Dyer and Dr. Paulo Gunter Windisch. Statistical information and suggestions were given by Dr. Denise G. Santana and Paulo Rangearo Peres. The identification of the species was carried out by Dr. Jefferson Prado and confirmation of Thelypteris species by Dr. Alan R. Smith and Dr. M. Monica Ponce. The field work was conducted with the help of Mr. Helio Pereira, Julio C. Franga Resende, Selma A. da Silva, Grace Cardoso, and Flavio Rodrigues Oliveira. Important help in the laboratory work was given by Iona Paula Calabria. Review of the English text was done by Mr. John David Bagnall. Interesting suggestions were given by Dr. Linda Styer Caldas. The author registers her sincere LITERATURE CITED C. C. and J. M. BASKIN. 1998. Seeds: Ecology, Biogeography, and Evolution of Dormancy and Germination. Academic Press, San Diego. J. D. and M. BLACK. 1994. Seeds: Physiology of Development and Germination. 2nd. ed. BASKIN, BEWLEY, E. J. and D. H. ASHTON. 1965. Seed storage in soils of several Victorian plant communities. Victorian Nat. 82:102-110. P. B. and D. L. BENOIT. 1989. Seed banks in arable land. Pp. 309-328 in M. A. Leek, V. T. Parker and R. L. Simpson, eds. Ecology of Soil Seed Banks. Academic Press, London. CLYMO, R. S. and J. G. DUCKETT. 1986. Regeneration of Sphagnum. New Phytol. 102:589-614. CONANT, D. S. 1978. A radioisotope technique to measure spore dispersal of the tree fern Cyathea arborea Sm. Pollen et Spores 20:583-593. DURING, H. J., M. BRUGUES, R. M. CROS and F. LLORET. 1987. The diaspore bank of bryophytes and CARROLL, CAVERS, H. J. and B. TER HORST. 1983. The diaspore bank of bryophytes and ferns in chalk grassland. Lindbergia 9:57-64. A. F. 1994. Natural soil spore banks: can they be used to retrieve lost ferns? Biodiversity and Conservation 3:160-175. DYER, A. F. and S. LINDSAY. 1992. Soil spore banks of temperate ferns. Amer. Fern J. 82:89-123. DYER, A. F. and S. LINDSAY. 1996. Soil spore banks—a new resource for conservation. Pp. 153-160 in J. M. Camus, M. Gibby, and R. J. Johns, eds. Pteridology in Perspective. Royal Botanic Gardens, Kew. FENNER, M. 1985. Seed Ecology. Chapman and Hall, London. FENNER, M. 1995. Ecology of seed banks. Pp. 507-528 in J. Kigel and G. Galili, eds. Seed Development and Germination. Marcel Dekker, New York. GARWOOD, N. C. 1989. Tropical soil seed banks: a review. Pp. 149-209 in M. A. Leek, V. T. Parker, and R. L. Simpson, eds. Ecology of Soil Seed Banks. Academic Press, London. DURING, DYER, AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 3 (2003) I. The significance of spore banks in natural populations of Athyrium 1 A. thelypterioides. Amer. Fern J. 78:96-104. *IEDEL, L. SCHOESSLER and K. SKOGEN. 2002. Belowground distribution and trychium gametophytes and juvenile sporophytes. Amer. Fern J. 92:80-92. wtologia: com un Esi ' I'ierra. Trad. P.R. Hendrichs S. and A. !'. DYEK. 1990. l-'ern spurn hanks: implii ations for gametophyte establishment. Taxonomia, Biogeograffa y Conservacion de Pteridofitos. Soc. d'Hist. Nat. de les [lies Balears IME. Palma de Mallorca: 243-253. S.. N. WILLIAMS and A. F. DYER. 1992. Wet storage of fern spores: unconventional but far more effective! Fern Hort. 285-294. MEYER, B. S.. D. B. ANDERSON and C. A. SWANSON. 1963. Laboratory Plant Physiology, 3"'. ed. Van Nostrand Company, Inc., Princeton. MILBERG, P. 1991. Fern spores in a grassland soil. Can. J. Bot. 69:831-834. MILBERG, P. and L. ANDERSON. 1994. Viable fern spores in OLIVEIRA, P. E. and I. SCHIAVINI. 1999. Plant dispersal in thi Central Brazil. Spatial and temporal patterns. XVI I 1-7, St. Louis, Missouri, USA. (Abstract). PAGE, C. N. 1979. Experimental aspects of fern ecology. Pp. Experimental Biology of Ferns. Academic Press, I PENROD, K. A. and L. H. MCCORMICK. 1 from hardwood forest soils at ' PEREIRA, S. G. 1999. Banco de semen LINDSAY, LINDSAY, B., A. OROZCO-SEGOVIA and R. RIBA. 1 de los Tuxtlas, Ver. Bol. Soc. Bot. Mexico {< E. 1996. Relationship between seed and meadow (Mallin) in Patagonia, Argentina. V RAGHAVAN, V. 1980. Cytology, physiology, and biochemistry of germinati Cytol. 62:69-118. l mata mesofila s Sao Paulo. 3. Fenologia e sobrevivencia dos individuos. Rev. Brasil. Biol. 55:777-787. RATTER, J. A. 1992. Transitions between cerrado and forest vegetation in Brazil. Pp. 417-429 in P. A. Furley, J. Proctor and J. A. Ratter, eds. Nature and Dynamics of Forest-Savanna Boundaries. Chapman & Hall, London. RUMSEY, F. J., J. C. You i. S. [. Rt SM ii. f. A. BARRETT and M. GIBBY. 1999. Population structure and conservation biology of the endangered fern Trichomanes speciosum Willd. (Hymenophyl- PEREZ-GARCIA, RAFFAELE, I. 1992. Estrutura das comunidades arbdreas de mata de galeria da Estagdo Ecoldgica do Panga (Uberldndia, MG). Thesis, Universidade Estadual de Campinas, Campinas. I. 1997. Environmental characterization and groups of species in gallery forests. Proceedings of the International Symposium on Assessment and Monitoring of Forests in Tropical Dry Regions with Special Reference to Gallery Forests. November, Brasilia, Brazil: SCHIAVINI, SCHIAVINI, I. and G. M. ARAUJO. 1989. Consideracoes sobre a vegetagao da Reserva Ecoldgica do Panga (Uberlandia). Sociedade & Natureza, Uberlandia 1:61-66. J. J. and R. HOLDEREGGER. 1996. Soil spore bank and genetic demography of populations of Athyrium filix-femina. Pp. 663-665 in J. M. Camus, M. Gibby and R. J. Johns, eds. Pteridology in Perspective. Royal Botanic Gardens, Kew. SIMABUKURO, E. A., A. BEGOVACZ, L. M. ESTEVES and G. M. FELIPPE. 1999. Fern spore bank at Pedregulho (Itirapina, Sao Paulo, Brazil). Rev. Brasil. Biol. 59:131-139. SCHIAVINI, SCHNELLER, RANAL: SOIL SPORE BANK IN A GALLERY FOREST 115 E. A., L. M. ESTEVES and G. M. FELIPPE. 1998. Analysis of a fern spore bank in Southeast Brazil. Hoehnea 25:45-57. SIMPSON, R. L., M. A. LECK and V. T. PARKER. 1989. Seed banks: general concepts and methodological issues. Pp. 3-8 in M. A. Leck. V. T. Parker and R. L. Simpson, eds. Ecology of Soil Seed Banks. SIMABUKURO, G. S. and P. J. EDGERTON. 1976. Emergent seedlings from coniferous litter and soil in Eastern Oregon. Ecology 57:801-807. K. and J. P. GRIME. 1979. Seasonal variation in the seed banks of herbaceous species in ten contrasting habitats. J. Ecol. 67:893-921. WALCK, J. L., J. M. BASKIN and C. C. BASKIN. 1996. An ecologically and evolutionarily meaningful definition of a persistent seed bank in Solidago. Amer. J. Bot. 83(Suppl. 6):78-79 (Abstract). WEE, Y. G. 1974. Viable seeds and spores of weed species in peat soil under pineapple cultivation. STRICKLER, THOMPSON. A Karyotype Comparison Between Two Closely Related Species of Acrostichum ADRIAN A B. MARCON, IVA C. L. mento de Botanica, Universidade Fede Cidade Universitaria, 50.670-42 ABSTRACT.—Acrostichum aureum and A. danaeifolium are morphologically similar sympatric species which grow in mangrove communities. To evaluate the cytological differences between these species, their karyotypes were analyzed with conventional staining, triple-staining with chromomycin A3 (CMA), distamycin A (DA) and DAPI, silver nitrate, and in situ hybridization with 45S rDNA as probe. Both species have the same chromosome number (2n = 60) with only small differences in chromosome size and morphology. The CMA+ banding pattern revealed four terminal bands in A. danaeifolium and six in A. aureum. DAPI+ bands were not found. The maximum number of nucleoli per interphase nucleus and the number of 45S rDNA sites were meiotically analyzed materials showed 30n with normal chromosome pairing and segregation, except in one plant with a chromosome bridge and fragmei suggested that sympatry and karyotypic orthoselection have c and karyological similarities in such widespread species. Pteridaceae is a large and diverse family of homosporous ferns of almost global distribution. This family comprises 32 genera, 22 of which occur in the Americas. The pantropical genus Acrostichum includes at least three species: the paleotropical A. speciosum Willd., the pantropical A. aureum L. and the neotropical A. danaeifolium Langsd. & Fisch. (Tryon and Tryon, 1982). The last two species are widely distributed in Brazil, occurring mainly as sympatric members of mangrove communities. Acrostichum danaeifolium also may be found isolated on swampy banks far away from the coast. These two species are morphologically very similar, although there are differences between fertile fronds, petioles, and paraphyses. For example, in A. aureum only the distal few pairs of pinnae are fertile, there are abortive pinnae on the petiole, and the paraphyses [i.e., trichomes occurring between sporangia) are globular, whereas in A. danaeifolium the pinnae are fertile from the apex to almost the base of the blades, the petioles have no abortive pinnae, and the paraphyses have laterally extended apices (Adams and Tomlinson, 1979; Proctor, 1985). The chromosome numbers of both species are n = 30 or In = 60 (Manton and Sledge, 1954; Walker, 1966, 1985; Dujardin and Tilquin, 1971; Lovis, 1977), although polyploids have been reported in two populations of A. aureum with 2n = 120 (Kawakami, 1980, 1982; Roux, 1993) as well as an aneuploid with 2n = 119 (Nakato, 1996). The sympatric distribution of A. aureum and A. danaeifolium suggests that marked genetic differences may maintain reproductive isolation between the species, and mediate against hybridization and polyploidy, which are frequent events in the evolution of pteridophytes, especially in homosporous genera MARCON ET AL.: KARYOTYPE COMPARISON IN ACROSTICHUM Joao Pessoa, Paraiba Rio Tinto, Paraiba Areia, Paraiba Bayeux, Paraiba Caja, Paraiba Ipojuca, Pernambuco Joao Pessoa, Parafba Paulista, Pernambuco Cultivated, immatur ABMarcon & GSBar 225/27446 Cultivated, immatur LPFelix 9367/27454 LPFelix 9369/27452 Cultivated, immatur Cultivated, immatui LPFelix 9368/27453 ABMarcon & GSBar (Walker, 1984). However, Lopez (1978) observed the occurrence of morphologically intermediate individuals in the Dominican Republic. In the study reported here, the cytological divergence between A. danaeifolium and A. aureum was investigated using conventional cytogenetic techniques to analyze chromosome number and morphology, fluorochrome staining to identify heterochromatin blocks, silver nitrate staining to detect the maximum number of nucleoli, and in situ hybridization to localize 45S ribosomal DNA (45S rDNA) sites in the genomes of both species. MATERIALS AND METHODS Samples and collection sites are given in Table 1. Part of the collected material was cultivated in the experimental garden of the Department of Botany of the Federal University of Pernambuco, Brazil, and another part of the material was stored as dried voucher specimens in the UFP herbarium for posterior identification. For mitotic analysis, actively growing root-tips were treated with 0.002 M 8hydroxyquinoline at room temperature for 1 h, followed by 23 h at 6°C, then fixed in Carnoy's solution (ethanohacetic acid 3:1) for 2-24 h. For meiotic analysis, young sporangia were fixed directly in Carnoy's for 2-24 h at room temperature. All fixed material was stored in a freezer until needed. 118 AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 3 (2003) Root-tips were washed twice in distilled water for 5 min, after which they were treated with a mixture of 2% cellulase-20% pectinase for 5-6 h at 37°C and hydrolyzed in 5 N HC1 for 30 min at room temperature. The root meristem was isolated, mounted on a microscope slide in 45% acetic acid, squashed under a coverslip (subsequently removed by freezing with liquid nitrogen), dried at room temperature, stained with 1% hematoxylin or 2% Giemsa, and mounted in Entellan (Merck), according to Guerra (1999). For meiotic analysis, sporangia were squashed in 2% acetic-carmine and analyzed. Karyotype formulas were based on measurements of the long and short arm lengths of each chromosome performed on photographs of the best metaphase figures. Chromosome nomenclature, based on centromeric index (short arm/total length x 100), followed the system proposed by Levan et al. (1964), allowing comparison with results of previous authors. In preparation for fluorescent CMA/DA/DAPI staining root-tips were washed in distilled water, treated with a mixture of 2% cellulase-20% pectinase for 56 h at 37°C, and squashed in 45% acetic acid on a microscope slide. The slides were aged for 3 days at room temperature, stained with 0.5 mg/ml chromomycin A3 (CMA) for 1 h, counterstained with 0.1 mg/ml distamycin A (DA) for 30 min, and 2 |ig/ml 4',6-diamidino-2-phenylindole (DAPI) for 30 min, then mounted in a mixture of glycerokMcIlvaine's buffer (1:1) containing 2.5 mM MgCl2 (following Schweizer and Ambros, 1994). CMA detects heterochromatin blocks rich in guanine and cytosine and DAPI detects blocks rich in adenine and thymine, whereas distamycin A is a non-fluorescent DNA ligant that increases the contrast between CMA and DAPI. For silver nitrate staining, root-tips were treated and squashed as described for fluorochrome staining. The silver nitrate staining technique was based on Rufas et al. (1987). A small drop of silver nitrate (50%, w/v, in formalin-water) was placed over the squashed cells, covered with a coverslip, and incubated at 60°C in a moist chamber for ca. 10 min. An rDNA probe isolated from Arabidopsis thaliana (SK18S + SK25S) containing two separately recloned fragments of the 45S rDNA repeat, representing the 18S and 25S rDNA (Untried et al., 1989; Unfried and Gruendler, 1990), kindly supplied by Prof. D. Schweizer, University of Vienna, was marked with biotin-11-dUTP (Sigma, USA) by nick translation. The 5S rDNA probe was obtained from genomic DNA of Passiflora edulis by PCR using the primers 5'-GTG CGA TCA TAC CAG C(A/G)(G/T)TAA TGC ACC GG-3' and 5'-GAG GTG CAA CAC GAG GAC TTC CCA GGA GG-3' (Gottlob-McHugh et al., 1990). The technique was based on Moscone et al. (1996). The probes were added at a final concentration of 1.2-3.0 ng/^1 to a hybridization mixture containing 60% (v/v) formamide, 5% (w/v) dextran sulphate, and 0.1 ug/nl salmon sperm in 2xSSC. The hybridization mixture and the cytological preparations were denatured at 75°C for 10 min and hybridized for 18-20 h at 37°C in a moist chamber. The 45S rDNA probe was detected with mouse antibiotin monoclonal antibody (Dakopatts n° M743) and visualized with rabbit anti-mouse antibody conjugated to tetramethyl rhodamine isothiocyanate (TRITC) (Dakopatts n° R270). The 5S rDNA probe was detected with sheep MARCON ET AL.: KARYOTYPE COMPARISON IN ACROSTICHUM 119 anti-digoxigenin antibody conjugated to fluorescein isothiocyanate (FITC) (Boehringer Mannheim n° 1207741) and FITC-conjugated rabbit anti-sheep (Dakopatts F135, DAKO). The slides were stained with 2 ug/ml DAPI, washed in 2xSSC, and mounted in Vectashield H-1000 (Vector Labs). The slides were examined using a Leica DMLB microscope and the best cells photographed on Kodak ASA 25 Imagelink HQ film for bright-field and Kodak ASA 400 T-MAX film for fluorescence images. Prints were made on Kodak Kodabromide F3. RESULTS The chromosome number observed was 2n = 60 in all the individuals of the two species. The chromosome size and morphology were similar for both species (Fig. la, b). The haploid chromosome complement was formed by lm + 2sm + 19st + 8t in A. aureum and lm + 3sm + 18st + 8t in A. danaeifolium (Table 2). The metacentric pair was the second smallest of the complement in A. aureum, but in A. danaeifolium it was the fifth smallest pair. However, differences in chromosome length between chromosomes of a complement, which determine the ordering, were very small. Satellites were observed in two subtelocentric pairs in both species.The chromosome size exhibited a gradual variation within each complement, ranging from 4.91 to 8.09 urn in A. aureum and from 5.06 to 8.04 um in A. danaeifolium. The average chromosome sizes were 6.35 and 6.40 ^im and the length of haploid complements was 190.57 and 192.12 urn for A. aureum and A. danaeifolium, respectively (Table 2). Silver nitrate staining did not allow the visualization of the nucleolus organizer regions (NORs), although the nucleoli were well defined. In 887 nuclei analyzed of A. danaeifolium, the number of nucleoli varied from one to four, with three being the most common (56.3%). Cells with three or four nucleoli generally exhibited one nucleolus much smaller than the others. In A. aureum 318 nuclei were analyzed, and the number of nucleoli varied between one and six (Fig. lc), with four being the most common (39.6%). Meiotic analysis was performed in three individuals of A. aureum and four of A. danaeifolium, of which two from each species were growing together (Table 1). Both species nearly always showed normal meiosis, with 30 bivalents (Fig. Id, e). In a single plant from a population of A. aureum some meiocytes showed anaphase I and II with a chromosome bridge and fragment (Fig. If). After CMA/DA/DAPI staining, A. danaeifolium exhibited two pairs of subtelocentric chromosomes with a CMA+ band on their short arms, slightly different in size and brightness (Fig. 2a). The same cell stained with DAPI displayed a homogeneous staining, except the CMA+ regions, which became negatively stained. For A. aureum, three chromosome pairs showed a CMA+/ DAPI band, two of them on the short arms of a subtelocentric and a telocentric chromosome pairs and one on the long arms of a subtelocentric pair (Fig. 2c). The band of the short arms of the subtelocentric chromosome pair was smaller AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 3 »fcx t%^f b >v*v #• 1. Mitotic metaphase, nucleolus number, and meiotic behaviour in Acrostichi Giemsa-stained mitotic metaphase of A. danaeifolium (a) and .1 aureum (b). c. Silve nuclei with 2-4 nui d, e, f. C^rmin-stained meiocytes with 30 bi\ A. danaeifolium (d) and A. aureum (e) and a chromosome bridge with fragment in a later i FIG. and sometimes unstable. No DAPI+ bands were seen on the chromosomes of either of the species (Fig. 2b, d). In situ hybridization with 45S rDNA fragments labeled the terminal regions of the short arms of two subtelocentric chromosome pairs of A. danaeifolium, with sites slightly different in size (Fig. 2e). In A. aureum, there were two sit on the short arms of a subtelocentric and a telocentric chromosome pairs ar one on the long arms of a subtelocentric chromosome pair. The site on tl telocentric chromosome pair was the smallest, while the other two were similar size (Fig. 2f). The 5S rDNA probe did not produce any single signal, spite of repeated attempts. Acrostichum danaeifolium and A. aureum are considered diploid taxa with a chromosome base number x = 30 (Lovis, 1977). The chromosome number found in Acrostichum populations from Northeast Brazil agreed with those i FERN JOURNAL: ' a 1 b c d f reported for Sri Lankan (Manton and Sledge, 1954), Jamaican, and Trinidadian (Walker, 1966, 1985) populations, with In = 60 and n = 30 for both species. Dujardin and Tilquin (1971) also reported n = 30 for a sample of A. aureum from Congo. On the other hand, Kawakami (1980, 1982) and Roux (1993) reported In = 120 for A. aureum from the Japanese island of Iriomote and from Natal (South Africa), respectively, and Nakato (1996) found an individual of MARCON ET AL.: KARYOTYPE A. aureum from the Iriomote population with 2n = 119. The latter is the only report of aneuploidy in Acrostichum, although dysploids are known in some other pteridophytes [Walker, 1984, 1985). The karyotypes of A. aureum and A. danaeifolium were similar in total haploid length (ca. 192 am), general symmetry, and chromosome size variation. They differed slightly in the karyotype formula: lm + 2sm + 19st + 8t for A. aureum and lm + 3sm + 18st + 8t for A. danaeifolium. The average chromosome size of the tetraploid A. aureum, described by Kawakami (1980], was 4.92 urn, therefore, much shorter than the average chromosome size in the present diploid sample (6.35 am). Although this observed difference may be due to differential chromosome condensation, it most likely is due to chromosome size reduction observed in most polyploids (Raina et al., 1994; Leitch and Bennett, 1997). In both Acrostichum species, the maximum number of nucleoli and CMA+ blocks were clearly correlated. For A. danaeifolium, there were four CMA+ bands in the metaphase chromosomes and interphase nuclei and up to four nucleoli per nucleus. A similar correlation occurred in A. aureum, with six CMA+ bands and a maximum of six nucleoli per nucleus. The large number of nuclei with a lower number of nucleoli may be related to the tendency of nucleoli to fuse (see, e.g., Moscone et al., 1995). In angiosperms, most nucleolus organizer regions (NORs) are CMA+/DAPI bands (Guerra, 2000), and the same seems to be true in Acrostichum. For other pteridophytes, apparentlv no previous karvological studies have been published using CMA/ DAPI or silver nitrate staining. In angiosperms, NORs are also correlated in number and size with secondary constrictions and sites of 45S rRNA genes. In pteridophytes, sites for 45S rRNA genes have been previously reported only in Osmunda japonica Thunb. (Kawakami et al, 1999) and in Ceratopteris richardii Brongn. (McGrath and Hickok, 1999), without indication about NORs or secondary constrictions. In A. danaeifolium, there were four secondary constrictions and four 45S rDNA sites; in A. aureum there also were four secondary constrictions but six 45S rDNA sites. This difference is probably due to the fact that 45S rDNA sites of a cell are not always active in A. aureum and some sites may be only rarely activated, resulting in a variable number of secondary constrictions. For example, Citrus sinensis (L.) Osbeck has three 45S rDNA sites and a maximum of three nucleoli per cell, but only 2.5% of the metaphases exhibit three secondary constrictions (Pedrosa et al., 1997, 2000). The meiotic analysis of both species of Acrostichum did not show any significant variation, even in sympatric populations. The anaphase bridge observed in a single individual of A. aureum is most probably due to an intraspecific polymorphism for a paracentric inversion. Therefore, in spite of the karyological, morphological, and ecological similarities between both Acrostichum species, there was no morphological or meiotic evidence of interspecific hybridization, as was reported in specimens from the Dominican Republic by Lopez (1978). Our data suggest that cytogenetic differentiation between A. danaeifolium 124 AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 3 (2003) and A. aureum is limited to very small variations in chromosome morphology and structure. Considering that these two species are sympatric throughout a wide geographical region, occupy narrow ecological niches, and probably have a long history of reproductive isolation it is surprising that there are so few cytological differences between them. Probably, the reproductive isolation is based on genie rather than chromosomic barriers and the karyotypic orthoselection, common in many ferns, has conserved the basic karyotype in both species, even at the level of the distribution of heterochromatin and rDNA sites. 3onardo Felix and George Baracho for help in field collectio: Vienna University, for the SK18S + SK25S plasmids, and t imento Cientifico e Tecnologico (CNPq) and the Fundagao c 3 Estado de Pernambuco (FACEPE), for financial help. LITERATURE CITED S. G., M. LEVESQUE, K. MACKENZIE, M. OLSON, O. YARASH and D. A. JOHNSON. 1990. Organization of the 5S rRNA genes in the soybean Glycine max (L.) Merril and conservation of the 5S rDNA repeat structure in higher plants. Genome 33:486-494. M. 1999. Hematoxylin: a simple, multiple-use dye for chromosome analysis. Genet. Mol. GOTTLOB-MCHUGH, GUERRA, GUERRA, M. 2000. Patterns of heterochromatin distribution in plant chromosomes. Genet. Mol. Biol. S. 1980. Karyomorphological studies on Japanese Pteridaceae II—Pteris, Acrostichum, Cheilanthes, Onychium. Bull. Aichi Univ. Educ. (Nat. Sci.) 29:129-150. KAUAKAMI. S. 1982. Karvomorphological studies on Japanese Pteridaceae IV—Discussion. Bull. KAYVAKAMI, \K wii. s. M.. k. K: PND I ind • fluorescent in situ hybridization (FISH) in diploid and artificially produced haploid sporophytes of the fern Osmunda japonica (Osmundaceae). PI. Syst. Evol. 216:325-331. CH, L. J. and M. D. BENNETT. 1997. Polyploidy in angiosperms. Trends PI. Sci. 2:470-476. A., K. FREDGA and A. SANDBERG. 1964. Nomenclature for centromeric position on IN, Revision del genero Acrostichum en la Republica Dominicana. Moscosoa 1: r patterns and processes in ferns. Adv. Bot. Res. 4:229-415. I. and W. A. SLEDGE. 1954. Observations of the cytology and taxonomy of the p1 flora of Ceylon. Philos. Trans., Ser. B 238:127-185. JRATH, J. M. and L. G. HICKOK. 1999. Multiple ribosomal RNA gene loci in the genome of the homosporous fern ( : mad. J. Bot. 77:1199-1202. GONE, E. A., J. LOIOL, F. EHRENDORFER and A. T. HUNZIKER. 1995. Analysis of active nucleolus organizing regions in Capsicum (Solanaceae) by silver staining. Amer. J. Bot. 82:276-287. •CONE. E. A.. M. A. MATZKE and A. J. M. MATZKE. 1996. The use of combined FISHUISII in ig to identify chromosomes conl JTON, ATO, N. 1996. Notes on chromosomes of Japanese pteridophytes (4). J. Jap. Bot. 71:163-167. MARCON ET AL.: KARYOTYPE COMPARISON IN ACROSTICHUM A., M. GUERRA and W. SOARES-FII.HO. 1997. An hierarchy of organizer regions in Citrus sinensis (L. ) Osbeck. Cytobios 92:43-51 A., D. Schweizer and M. Guerra. 2000. Cytological heterozvgozitv sweet orange [Citrus sinensis (L.) Osbeck]. Theor. Appl. Genet. 100:361-3 PROCTOR, G. R. 1985. Ferns of Jamaica—A Guide to the Pteridophytes. British PEDROSA, PEDROSA, S. N., A. PAKIIJA, K. K. KOI L, S. S. SALIMATH. M. S. BISTII, V. RAJA and T. N Associated chromosomal DNA changes in polyploids. Genome 37:560-564. Roi x. \. 1'. 1993. A new cytotype for Acrostichum aureum. Bothalia 23:75. RITAS, J. S.. J. GIMENEZ-ABIAN, }. A. SU)A and C. GARCIA DE LA VEGA. 1987. Chromosome RAINA, icuwi :. D and P. F. AMBROS. 1994. Chromosome banding. Pp. 97-112 in: J. R. Gosden (ed. Methods in Molecular Biology, Vol. 29. Humana Press, Totowa. I'KYON, R. M. and A. F. TKYOV 1982. Ferns and Allied Plants—with Special Reference to Tropin America. Springer-Verlag, New York. JNFRIED, 1. and P. GRUENDLER. 1990. Nucleotide sequence of the 5.8S and 25S rRNA genes and of th internal transcribed spacers from Arabia s/s ti ,: i m i N'm 1. Acids Res. 18:4011. ixiKiMi, I., I . STOCKER and P. GRUENDLER. 1989. Nucleotide sequence of the 18S rRNA gene fror Arabidopsis thalianu Co 10. Nucl. Acids Res. 17:7513. VALKER, T. G. 1966. A cytotaxonomic survey of the pteridophytes of Jamaica. Trans. Roy. Soc Edinburgh 66:169-237. WALKER. T. G. 1984. Chromosomes and evolution in pteridophytes. Pp. 103-141 in: A. K. Sharm and A. Sharma (eds.). Chromosome F\ roups II. CRC Press, Boca Rator /VAI.KER, T. G. 1985. Cytotaxonomic studies of the ferns of Trinidad 2. The cytology and taxonomi implications. Bull. Brit. Mus. (Nat. Hist.), Bot. 13:149-249. A Re-evaluation of Isoetes savatieri Franchet Argentina and Chile Botany Department, Miami University, Oxford, OH, 45056 CECELIA MACLUF edra de Palinologia, Facultad de Ciencias Naturales y Museo, Universidad N; Plata Paseo del Bosques s/n, 1900 La Plata, Argentina W. CARL TAYLOR Botany Department, Milwaukee Public Museum, Milwaukee, WI regions of South America to the central Andes of Chile and Argentina. An am material supports the recognition of two taxa, a southern /. savatieri and a more northern /. chubutiana, from central Chile and Argentina. The latter taxon is hexaploid and described here as a new species. The morphology of these species suggest that they are sister species resulting from divergence following a polyploidy event. These species, and several other species pairs, provide the best and, to date, only examples of allopatric divergence in polyploid Isoetes is a nearly cosmopolitan genus of aquatic to sub-aquatic perennial lycopsids. Estimates of species number has ranged from 60 (Pfeiffer, 1922) to 150 (Tryon & Tryon, 1982). Recent systematic work in North America (e.g., Brunton & Britton, 1997, 1998; Caplen & Werth, 2000, 2000b) and South America (Small & Hickey, 2001; Hickey, 1994), however, indicates that even 150 is likely to be an underestimate. The actual number of species worldwide is probably closer to 350. There are several reasons for this large disparity. Despite a long history of systematic and morphological interest, the genus is poorly collected (Hickey et al., 1989) and only sporadically studied. Aside from the classic 19th Century works of Baker (1880) and Motelay & Vendryes (1882) there has only been one modern systematic treatment of the genus worldwide (Pfeiffer, 1922). Pfeiffer's monograph stands as the classic treatment of the genus despite a number of significant but unavoidable flaws. Most significant among these is the lack of adequate Neotropical collections examined during the study. Pfeiffer, working out of the Missouri Botanical Garden, relied almost exclusively on specimens housed at F, GH, MO and US. She did not examine the many important collections held in Europe and Latin America and, as a result, was unable to develop a full appreciation of the diversity of the genus as it occurs in South America. South America appears to be the center of both morphological and taxonomic diversity for Isoetes (Hickey, 1990). The richness of the South American flora was first evidenced in the work of Ulrich Weber (1922). In a revision of the South American species, he recognized 18 taxa, 11 of which he described as new. Weber's work, while certainly not complete or entirely HICKEY ET AL.: RE-EVALUATION OF ISOETES SA \ 'A TIEBI 12 7 accurate, stands in contrast to the work of Pfeiffer which recognized only seven species for all of South America. The next significant work on South American Isoetes was that of Fuchs-Eckert (1982) in which he recognizes 75 species. In an overly conservative work, Hickey (1985) recognized 47 South American species but has since accepted considerably more (Hickey, 1994; Small & Hickey, 2001). This paper adds to our knowledge of the genus in South America by describing a new species, allied to Isoetes savatieri Franchet. This new taxon was first recognized as distinctive by Fuchs-Eckert (1982) but was never validly published. We compare both species and continues a discussion (Hickey et al., 1989) on speciation in the genus. Isoetes savatieri Franchet, Bull. Soc. Bot. France 31:395. 1884. Calamaria savatieri (Franchet) Kuntze, Rev. Gen. PL 2:828. 1891-1893. Isoetes lechleri var. savatieri (Franchet) L. D. Gomez, Brenesia 18:5. 1980.—TYPE: Argentina, Puerto Bueno, 15 Feb 1877, Savatier s.n. (Holotype: P), ex char. Figs. 1-3. Corm globose to somewhat laterally elongate, 9-12 mm wide, 8-10 mm high, 2-lobed; roots dichotomous, arising synchronously within the continuous circumbasal fossa. Leaves 6-22, stiffly erect or slightly recurved, brittle, 42-163 mm long, 7-14 mm wide at the base, 1.8-5.0 mm wide at mid length; alae hyaline and chartaceous proximally, dark green and membranaceous distally, 1.0-3.5 mm wide at the sporangium, 12-65 mm long (extending up for 29-65% of the leaf length), each apex obtuse; subula terete, dark green, the apex short acuminate; fibrous bundles absent; stomates absent; scale leaves and phyllopodia absent. Sporangium circular to elliptic, hyaline, tan, concolorous, 3.0-9.5 mm long, 3.5-6.2 mm wide, basal. Velum incomplete, extending (0.5)1.5-2.5 mm down from the top of the sporangium. Ligule deltate to widely ovate, hastate-auriculate, delicate and ephemeral, 2.5-3.0 mm long, 1.8-3.3 mm wide. Labium inconspicuous, represented by a low, entire or scalloped ridge, light green, membranaceous, 40-60 urn high, 70-100 um wide. Megaspores white to off-white, frequently lustrous, 370-580 (x = 479) urn in diameter, rugulate or rarely tuberculate, girdle sparsely ornamented; equatorial and proximal ridges straight, distinct, as high as broad. Microspores light gray, 35.0-46.3 (x = 39) um long, 25.0-33.8 (x = 29) um wide, laevigate. Chromosome number unknown. DISTRIBUTION.—Endemic to the low coastal regions around Tierra del Fuego in Chile. ECOLOGY.—The limited ecological data suggest that this species is typically found below 200 m. The plant is apparently an obligate aquatic, inhabiting the shallows of streams and lake margins. Vegetative reproduction is frequent and is accomplished by the production of cortical gemmae. This species produces megaspores and microspores January through April. Isoetes savatieri is characterized by an acuminate leaf apex, a hastateauriculate ligule, a partial velum, and a minute labium. It differs from /. FERN JOURNAL: VOLUME 93 NUMBER 3 chubutiana in leaf shape, in particular leaf width and apex shape, and to a lesser extent in spore morphology (Table 1). Isoetes savatieri has a blunt apex with a distinct acumen, whereas /. chubutiana generally has a more tapering apex and less obvious acumen. Isoetes savatieri has broad, short leaves; the range in leaf width is 2-5 mm, with a mean and mean of 3 mm. The leaf width/ : RE-EVALUATION ( Alae apex shape Megaspore surface morphology obtuse rugulate to rarely tuberculate attenuate sparsely to densely rugulate, to cristatt Megaspore size (|.im) Microspore surface length ratio ranges from 9 to 47 with a mean of 24. In /. cubutiana, the leaves are narrower; leaf width ranges from 1.5-2.1 mm with a mean of 1.7 mm and a mode of 1.5 mm. The leaf length/width ratio ranges from 19 to 98 with a mean value of 62. Megaspores of J. savatieri are rugulate except in some individuals of Donat 380 where the spores are tuberculate. Some of the spores of this collection have the tubercles confluent to form short muri, approaching a rugulate condition. The megaspores of/, chubutiana are more variable; they range from sparsely to densely rugulate to cristate and finally to reticulate. Although there is a tendency in the latter species to produce a greater number of leaves and to have larger megaspores, these differences are not reliable enough for identification purposes. The microspores of I. savatieri are laevigate, whereas in /. chubutiana they are sparsely to densely echinate. In all other characters the two taxa are virtually indistinguishable. Isoetes savatieri is geographically separated from I. chubutiana by some 1000 km. The megaspores of Isoetes savatieri are about the same size as those of the hexaploid Isoetes chubutiana suggesting they are the same ploidy level (Small & Hickey, 2001; Troia, 2001). SI'IUMIAS EX.WIIM ii.—CHILE. Borge s.n. (GH, NY, US); Lagc (S); Region Riesco; 22 1971, Pisano 2938 (HIP); arrojado a la playa por las olas, 26 Jan 1973, Pisano 3871a (HIP); Isla Rennel Norte, Canal Smyth, 74°12'S, 51°54'W, river bottom, rocky bottom w/ ferric accumulation, submerged+/-0.3 m, Transecta Botdnica de Patagonia Ausmil li.ll (Fill1 -2): < nnsolidated organic mud rich in iron. 7 ra \tagpnia Austral 1205 (HIP). 7-10. Isoetes chubutiana. 7. Proximal and equatorial view of microspores (Ta\ scale bar = 10 urn. 8. Equatorial view of megaspore with smooth girdle and retate sur {Taylor 6171, LP); bar = 100 um. 9. Root tip squash of 2n = 66 (Taylor 6171, MIL) squash of 2n = 66 (Taylor 6168, MIL). FIGS. Isoetes chubutiana Hickey, Macluf & W. C. Taylor, spec, nov.—TYPE: Argentina: Gob. Rio Negro, Lago Hess, 10 Jan 1945, Meyer 8077 (holotype: LIL; isotypes: NY, UC). Figs. 4-10. /. valdiviensis H. P. Fuchs, nom.nud., Proc. Kon. Ned. Akad. Wetensch. C85:255. 1982. /. Meyeri Fuchs, nom. nud., Proc. Kon. Ned. Akad. Wetensch. C85:231, 241, 242, 255. 1982. Based on: Argentina; Gob. Rio Negro, Cascado del Rio Manso, 30 Jan 1945, based on Meyer 8238 (LIL!). Cormus globosus usque lateraliter elongatus, bilobatus, 4-23 mm latus, 3-10 mm elatus; radices dichotomae, e fossa singulari circumbasali exoriente. Folia 9-30, rigide erecta vel raro recurva distale, fragilia, 40-280 mm longa, 6.0-9.0 HICKEY ET AL.: RE-EVALUATION OF ISOETES SAVATIERI ou w i (circles) and /. - 132 AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 3 (2003) mm basi lata, 1.5-2.2(3.0] mm medio lata; alae proximale hyalinae et chartaceae, distale atrovirides et membranaceae, 11-55 mm longae (18-30[45]% per foliae longitudinem ascendentes), apicibus attenuatis; subula teres, atroviridis, apice longe acuminato; fasciculi fibrosi peripheric!, stomata, squamae et phylllopodia absentia. Sporangium circulare usque ellipticum, hyalinum, concolor, 2.8-6.7 mm longum, 2.8-5.7 mm latum, basale. Velum incompletum, descendens ad 0.7-2.7 mm. Ligula deltata usque late ovata, cordata usque hastata, viridi-nigra, tenella atque fugax, 1.5-3.CM-mm elata, 1.7-2.3 mm lata. Labium inconspicuum usque absens. Megasporae albae usque cretaceae, saepe nitidae, 460-750 (x = 595) (.im diametro, rugulosae usque rugulosae-cristatae vel reticulatae; zona non dissimilis usque laevis; cristae aequatoriae proximalesque rectae, distinctae, non altiores quam latae. Microsporae cinereae usque brunneae, 33.8-41.3 (x = 39) jim longae, 26.2-33.8 (x = 30) um latae, sparse usque dense echinatae. Chromosomatum numerus 2n = 66. Corm globose to somewhat elongate laterally, 4-23 mm wide, 3-10 mm high, 2-lobed; roots, dichotomous, arising synchronously within the continuous circumbasal fossa. Leaves 9-30, stiffly erect or more rarely recurved distally, brittle, 40-280 mm long, 6.0-9.0 mm wide at the base, 1.5-2.2(3.0) mm wide at mid length; alae hyaline and chartaceous proximally, dark green and membranaceous distally, 1.2-3.0 mm wide at the sporangium, 11-55 mm long (extending up for 18-30(45)% of the leaf length), each apex attenuate; subula terete, dark green, the apex long acuminate; fibrous bundles absent; stomates absent; scale leaves and phyllopdia absent. Sporangium circular to elliptical, hyaline, tan, concolorous, 2.8-6.7 mm long, 2.8-5.7 mm wide, basal. Velum incomplete, extending 0.7-2.7 mm down from the top of the sporangium. Ligule deltate to widely ovate, chordate to hastate, black, delicate, ephemeral, 1.5-3.0+ mm high, 1.7-2.3 mm wide. Labium inconspicuous to absent. Megaspores white to off white, often lustrous, 460-750 (x = 595) |xm in diameter, rugulate, rugulate-cristate, to reticulate, girdle undifferentiated to smooth; equatorial and proximal ridges straight, distinct, as high as broad. Microspores light grey to dark brown, 33.8-41.3 (x = 39) ^m long, 26.2-33.8 (x = 30) [im wide, sparsely to densely echinate. Chromosome number 2n = 66. DISTRIBUTION.—Endemic ECOLOGY.—Isoetes to the central Andes of Chile and Argentina. chubutiana grows at elevations of 750 to 1300 m as a submerged aquatic in the shallows of streams and lakes. Collections from November, January, February, March and April have megaspores and microspores. Collections from May have only microspores. The absence of collections from the rest of the year precludes further statements about yearly phenology. Isoetes chubutiana, like a number of species of the central and south central Andes, reproduces asexually by the production of cortical gemmae. In technical characters of the sporangium, velum, and ligule, /. chubutiana is indistinguishable from /. savatieri. It can be distinguished from it however by leaf form, as described under /. savatieri, and to a lesser extent by megaspore morphology. HICKEY ET AL.: RE-EVALUATION OF ISOETES SAVATIERI 133 The names I. valdiviensis and I. meyeri were published by Fuchs-Eckert (1982) without latin or english descriptions as part of an enumeration of South American species. In addition, Diem 1105 (GH) is annotated as the type of the unpublished "herbarium" name Isoetes chilensis. This plant and those annotated as I. Meyeri and 7. valdiviensis are best accommodated within I. chubutiana. Megaspore and microspore ornamentations are extremely variable in this otherwise uniform taxon. The most common megaspore type in the southern portion of the range is rugulate while the more northerly collections are typically reticulate. In all other South American species with reticulate megaspore ornamentation the microspores are laevigate. In this species, however, the microspores generally have an underlying or dominant echinate surface pattern. The presence of reticulate megaspores in this species provides additional evidence of convergence in spore morphology and seriously undermines the recognition of a Section Terrestres [sensu Fuchs, 1982; = Reticulatae of Pfeiffer, 1922), a section currently based almost exclusively on the presence of reticulate megaspores. PARATYPES.—CHILE. Aysen: Lago Gral, Paz, 15 Apr 1943, Maldonado 288 (LP); Chile chico, a orillas del algo, 3 Dec. 1946, Castillo s.n. (CONC); Ventisquero Soler, 24-25 Mar 1967, Seki 581II [CONC). Llanquihue: La Turbina, Payne, en orillas del Rio Payne, despues del Salto Chico, 22 Feb 1974, Pisano 4304 (HIP); Puerto Varas, Puella, Rigi, 125 m, 41°06'S, 72°02'W, Mar 1967, Zollitsch 298 (CONC). Osorno: Isla de Rupanco, hidrofito crece a poca profundidad (1 m) orillas, 15 Mar 1978, Godoy s.n. (SGO); Isla de Rupanco, aquatica, se desarrola a 1 m de profundidad en el lago Rupanco, 7 Mar 1979, Godoy 3 (SGO); Isla de Rupanco, acuatica, de desarrolla a 1 m de profundidad y en orillas sobre arena, 7 Mar 1979, Godoy s.n. (SGO); Lago Rupanco, Rio Pecaderos, 8 Dec 1945, Rudolph 43,676 (CONC); Lago Puyehue, (Isla Fresia), costa en sur y vuelta (en el agua), 5 Feb 1954, Levi Heins 1744 (CONC). ARGENTINA. Neuquen: Depto. Minas, Lagunas Epu-Lauquen, Aduana Vieja, sumergida en las orillas de las lagunas, +/- 50 cm de profundidad, 1300 m, 15 Jan 1964, Roelcke et al. 10871 (BAA, SI); extremo norte de la laguna Varvarco Campos, orillas, 2 Feb 1970, Boelcke et al. 14336 (BAA), 14337 (MU, SI); Puerto Manzano, 13 Feb 1934, Burkart 6499 (BM, SI); Lago Espejo y correntoso, 1 m profundidad, 16 Apr 01, Meier s.n. (LP); Lago Lacar, playa cerca a San Martin de Los Andes, 1 Mar 1966, Burkart & Troncoso 26447 (SI, UC); Lago Totoral, bei niedrigem wasserstand beinahe ausserhalb des Wassers, 900 m, 22 Feb 1970, Diem 3379 (L, M, NY); Quetrihue, en aguas tranquilas, 1 m bajo el agua, 30 May 1942, Diem 646 (SI); Punta Quethihue, en playas inundadas, formando comunidades puras y numerosas, 770 m, 8 Mar 1959, de la Sota 2167 (LIL). Rio Negro: Lago Nahuel Huapi: Puerto Panuelo, a 1-2 m de profundidad en las aguas, 15 Feb 1934, Burkart 6548 (SI); Puerto Panuelo, Feb 1911, Hauman 1 (LIL); Parque Nacional Nahuel Huapi, E side of Lago Guillelmo, plants firmly anchored in sandy humus among rocks, submerged 0.75-1.0 m, elev. 840 m, 41°22.3'S, 71°29.7'W, 17 Mar 2001, Taylor 6168 (LP, 134 AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 3 (2003) MIL); Parque Nacional Nahuel Huapi, S side of Lago Mascardi, plants firmly anchored in sandy humus among rocks, submerged 0.75-1.0 m, 41°21.35S, 71°34.3'W, 822 m, 17 Mar 2001, Taylor 6171 (LP, MIL); Lago Frias, 1-2 m bajo agua en extensas colonias enterreno arcuosa-arculloso, 800 m, 1 Nov 1947, Diem 1105 (GH). Cascado del Rio Manso, 30 Jan 1945, Meyer 8238 (LIL). Chubut: Lago Futalaufquen, 14 Jan 1945, Castellanos 114242 (AA); Lago Futalaufquen, Reserva Nacional de los Alerces, 27 Mar 1949, Pedersen 302 (C, S); Lago Verde, Parque Nac. Los Alerces, sumergida en el rio, 25 Feb 1950, Soraimo 4287 (BAA). An interesting aspect of both species is the sporadic occurrence of irregular spores. In Donat 380 [I. savatieri) the megaspores show a high degree of size dimorphism whereas the microspores show ca. 70% spore abortion. Borge s.n. [I. savatieri) contains megaspores with occasional tetraradiate meiotic scars (Fig. 3), often an indication of meiotic irregularity, but shows only 1-2% microspore abortion. In /. chubutiana, a plant from Castellanos 11424 has very irregular megaspores, both in size and shape, yet has perfectly normal microspores. Such situations are frequent in other species of the genus, for example, in occasional specimens of /. storkii Palmer from Cerro de la Muerte of Costa Rica and in /. Luetzelburgii Weber of Brazil. More comparable is the situation in the Isoetes lechleri Mett. complex of Peru and Bolivia (Hickey, 1994). In that complex both of the currently recognized members, /. lechleri and /. herzogii Weber, appear to be tetraploid and, like the two members of the I. savatieri complex, reproduce asexually by means of cortical buds. Members of the /. lechleri complex are notorious for their high rate of megaspore abnormalities, reminiscent of meiotic irregularity. Hickey (1994) hypothesized that this phenomenon was the result of polyploidy followed by differential gene silencing and, through subsequent out-crossing, the accumulation of reciprocal gene silencings (Werth and Windham, 1991). The spore abortion seen in /. savatieri and /. chubutiana is likely to be of similar origin. The hexaploid /. chubutiana is probably the result of stabilization through polyploidy of a sterile triploid, with the triploid springing from a hybridization event between a tetraploid and a diploid. A number of features suggest that the tetraploid parent was a member of the I. lechleri complex. That complex and the members of the /. savatieri complex share cortical gemma production and a turgidly brittle leaf habit, both unusual features in the genus. In addition, they share a similar habit, similar spore morphology, and have nearly contiguous ranges. The rugulate /. herzogii is more likely involved than the laevigate /. lechleri. The diploid parent might be /. boliviensis Weber of Bolivia and Peru or I. alcalophila Halloy (interpreted here as including /. escondidensis Halloy) of northern Argentina. Isoetes hieronymii Weber is another possibility, being found in northern Argentina and proximate to the range of the Isoetes lechleri complex, but its chromosome number is not yet known. Taylor and Hickey (1992) discussed the mechanisms of speciation in Isoetes and have noted two predominant patterns. The first is characteristic of lowland diploid taxa and involves allopatric divergence. The second is allopolyploidy. , typically aquatics of temperate or HICKEY ET AL.: RE-EVALUATION OF ISOETES SAVATIERI 135 tropical upland regions. Little attention has been given to the origin of lowland polyploids nor has convincing evidence been provided for divergence subsequent to a polyplod event. Perhaps the best potential example of polyploid divergence was to be found in the Isoetes riparia Engelm. complex of eastern North America (Proctor, 1949). This assemblage of tetraploids includes several specific and subspecific segregates that have variously been elevated in rank or subsumed since their initial descriptions. The two best known segregates are /. saccharata Engelm. and /. canadensis (Engelm.) A. A. Eaton. Recently, however, it has been shown that this /. riparia complex is polyphyletic and consists of a number of similar but phylogenetically distinct polyploids sharing some but not all ancestors (Caplen & Werth, 2000a, 2000b). In South America, there are several good candidates for divergence following polyploidy. Within the Isoetes lechleri complex, there is a tremendous amount of inter-populational differentiation, and a number of specific segregates have been proposed (Fuchs-Eckert, 1982; Hickey, 1985). Hickey (1994) argued that most of these segregates are best accommodated in a more inclusive I. lechleri. However, the populations from central and southern Bolivia form a cohesive assemblage distinct enough from the northerly I. lechleri to be recognized at the specific level as /. herzogii. Similarities in morphology, identical chromosome number, spore abortion, and cortical gemmae in /. lechleri and /. herzogii argue strongly for a divergent rather than an independent origin for these tetraploids. Likewise, similarity in morphology, spore ornamentation, spore size, and geography supports an allopatric-divergence model for the I. savatieri-chubutiana polyploid pair. These examples then represent the best evidence to date for allopatric speciation in polyploid Isoetes. The combination of a high incidence of polyploidy in the genus (58.1%, Troia, 2001) and the rarity of allopatric polyploid speciation is surprising. It suggests that Isoetes, which appears to date back to the earliest Triassic (Grauvogel-Stamm and Lugardon, 2001), as Isoetites, has persisted through geologic time primarily as basic diploids. It further suggests that either polypoidy is a relatively recent process in the genus and/or that polyploids are ephemeral taxa that position themselves temporarily in vacant niches. Support for both of these models comes from the modern distribution of polyploids: most polyploids being found in fairly recent habitats such as temperate, glaciated regions, areas affected by such glaciations, or high altitude paramos and lakes. These models are also supported by allozyme studies: polyploid Isoetes appear to retain fixed heterozygosity and show little or no evidence of extensive diploidization (Caplen and Werth, 2000a, 2000b), suggesting relatively recent origins. The preponderance of data concerning Isoetes evolution suggest that polyploids are of little consequence from a divergence standpoint, but certainly do evolve by way of additional rounds of polyploidy. Morphological, geographic and cytological data from the /. savatieri-I. chubutiana and the /. lechleri-I. herzogii pairs provide an arena to test whether there is a third model of speciation, divergence following polyploidy, occurring in the genus. Other wide-ranging tetraploids such as the lowland /. panamensis Maxon & C. V. Morton, which ranges from Central America to 136 AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 3 (2003) Paraguay, and the /. triangula complex of Mexico, Venezuela, Brazil, and French Guiana (Stolze and Hickey, 1983; Hickey, 1985; Hickey, 1988) should be studied for additional examples of this evolutionary model. The authors thank Lara Strittmatter and R( t LJ f for tl P t d d { t support, and assistance. Neil Luebke ot the botany Department at Milwaukee Public Museum produced the chromosome figures included in this paper. Thanks also to Cecilia Ezcurra, Centra Regional Universitario, Bariloche, who assisted greatly with the support of the W. S. Turrell Herbarium (MU). LITERATURE CITED J. G. 1880. A synopsis of the species of Isoetes. J. Bot., London 18:65-70. D. R. and D. M. BRITTOV 1997. \ 'urinaria, sp. nov.; Isoetaceae), a new pteridophyte from the eastern United States. Rhodora 99:118-133. BRUNTON, D. R. and D. M. BRITTON. 1998. Isoetes microvela (Isoetaceae), a new quillwort from the BAKER, BRUNTON, C. A. and C. R. WERTH. 2000a. Isozymes of the Isoetes riparia complex, I. Genetic variation and relatedness of diploid species. Syst. Bot. 25:235-259. C. A. and C. R. WERTH. 2000b. Isozymes of the Isoetes riparia complex, II. Ancestry and relationships of polyploids. Syst. Bot. 25:260-280. FUCHS-ECKERT, H. P. 1982. Zur heutigen Kenntnis von Vorkommen und Verbreitung der sudamerikanischen /soefes-Arten. Proc. Ned. Akad. Wetensch. C85:205-260. GRAUVOGEL-STAMM, L and B. LUGARDON. 2001. The Triassic lycopsids Pleuromeia and Aimalepis: relationships, evolution, and origin. Amer. Fern J. 91:115-149. HICKEY, R. J. 1985. Revisionary studies of Neotropical Isoetes. Ph.D. dissertation, The University of Connecticut, Storrs. HICKEY, R. J. 1998. Isoetes pallida, a new species of Isoetes from Mexico. Amer. Fern I. 78:35-36. HickM . R. ). 1990. Studies of Neotropical Isoetes L. I. Euj rms, Ann. Missouri Bot. Card. 77:239-245. HICKEY, R. f. 1994. Family 28. Isoetaceae. in R. M. Tryon & R. G. Stolze. Pteridophyta of Peru. Part VI. 22. Marsileaceae-28. Isoetaceae. Fieldiana, Bot. n.s. 34:1-123. HlCKEY, R. I..C. W. TAvmiund N.T. LUEBKE. 1989. The species concept in Pteridophyta with special reference to Isoetes. Amer. Fern J. 79:78-89. MOTELAY, L. & A. VENDRYES. 1882. Monographie des Isoetaceae. Actes Soc. Linn. Bord. 36:309^106, CAPLEN, CAPLEN, N. A. 1922. Monograph of the Isoetaceae. Ann. Missouri Bot. Card. 9:79-232. G. R. 1949. Isoetes riparia and its variants. Amer. Fern J. 39:110-121. R. L. & R. J. HICKEY. 2001. Systematics of the northern Andean Isoetes karstenii complex. Amer. Fern J. 91:41-69. STOLZE, R. G. and R. J. HICKEY. 1983. Isoetaceae. in, R. G. Stolze. Ferns and fern allies of Guatemala. Part III. Marsileaceae, Salviniaceae, and the fern allies. Fieldiana, Bot, n.s. 12:1-91. TAYLOR, W. C. and R. J. HICKEY. 1992. Habitat, evolution, and speciation in Isoetes. Ann. Missouri Bot. Gard. 79:613-622. TROIA, A. 2001. The genus Isoetes L. (Lycophyta, Isoetaceae): synthesis of karyological data. Webbia PFEIFFER, PROCTOR, SMALL, R. M. & A. F. TRYON. 1993. Ferns and Allied Plants with Special Reference to Tropical America. Springer-Verlag, New York. U. 1922. Zur Anatomie und Systematik der Gattung Isoetes L. Nova Hedwigia 63:219-262. C. R. and M. D. WINDHAM. 1991. A model for divergent allopatric speciation of allopolyploid pteridophytes resulting from silencing of duplicate gene expression. Amer. Nat. 137:515-526. TRYON, WEBER, WERTH, . < •' 11 ..ft ' : - Rapid Gametophyte Maturation in Ophioglossum crotalophoroides . TN 37235-1565 Ai K —W th most species of the Qphio are slow and some species have perennial gametophytes. A few species, including O. crotalophoroides, appear to have gametophytes that mature rapidly. To determine how fast the gametophytes of this species mature, they were grown in axenic culture. The early sequence of cell divisions following germination is die same as tor other species ol the ()phioglossaceae. The formation of mucilage on the proximal cell of the \oimg gametophyte and on the rhizoids of older gametophytes has also been reported for other members of the family. The spores of O. crotalophoroides have the second fastest germination, 8 days, for this family. Gametophytes of this species grow faster than gametophytes of two Botrychium species. The gametangia form on smaller gametophytes of O. crotalophoroides than on those of Born, hium. Rapid spore germination, rapid gametophyte growth, and smaller gametophyte size at maturity all contribute to the formation of sexually mature gametophytes in 6.5 months. This is the fastest gametophyte maturation reported Gametophyte development in the Ophioglossaceae is sluggish (Boullard, 1963). The spores typically take a long time to germinate (Raghavan, 1989) and the growth of the gametophyte after germination is slow (Nayar and Kaur, 1971). Some species have perennial gametophytes (Campbell, 1911; Pant et al, 1984) and it can be a matter of years before sexual reproduction occurs (Bruchmann, 1904). Culturing gametophytes of the Ophioglossaceae under axenic culture conditions does not appear to accelerate their development because it took 22 months for gametophytes of Botrychium dissectum Spreng. to become sexually mature (Whittier, 1972). Although gametophyte development in a majority of the species in this family takes a long time, a few species appear to mature more rapidly. Campbell (1907) concluded that Ophioglossum moluccanum Schlect had annual gametophytes. Gametophytes oi Helminthostachys zeylanica (L.) Hook, and Ophioglossum nudicaule L. are reported to be short lived by Nozu (1961) and Mesler et al. (1975) respectively. It also appears that Ophioglossum crotalophoroides Walt, has rapid gametophyte development because Mesler (1976) found mature gametophytes one year after spores were released into pots under greenhouse conditions. What causes accelerated gametophyte maturation in some species of the Ophioglossaceae has never been examined. A study on gametophyte development in O. crotalophoroides presented an opportunity to examine rapid maturation in this group. This investigation was carried out to determine how fast gametophytes of this species become sexually mature and, if possible, what accelerates gametophyte maturation. \AL: VOLUME 93 NUMBER 3 Spores of Ophioglossum crotalophoroides were obtained from plants in Alabama and Louisiana. Vouchers of the sporophytes are on deposit at the Vanderbilt University Herbarium (VDB). The spores were usually sown within a month of their collection. To reduce the incidence of contamination, the spores were wetted and stored in water for 24 hours before surface sterilization. They then were surface sterilized with 20% Clorox (1.1% sodium hypochlorite) by the method of Whittier (1964), collected on sterile filter paper, suspended in sterile water, and sown on 15 ml of nutrient medium in culture tubes (20 X 125 mm) with screw caps that were tightened to reduce moisture loss. Most of the cultures were maintained at 21 ± 1°C in the dark, but a few had exposures to light (50 umol-m 2s-1) from Gro-lux fluorescent lamps for 14 of every 24 hours. The basic nutrient medium contained 100 mg MgS04-7H20, 40 mg CaCl2, 100 mg K2HP04, and 100 mg NH4C1 or 100 mg NH4N03 per liter. The medium was completed with 0.5 ml of a minor element solution (Whittier and Steeves, 1960), 8 ml of a FeEDTA solution (Sheat et ah, 1959) and 2 g of glucose. The medium was solidified with 1.0% agar and was at pH 5.5 after autoclaving. Any modifications to the basic nutrient medium are presented with the results. The sample size for calculating the average sizes of gametophytes or gametangia was 30. For determining the percentage of spore germination 500 or more spores were examined. Early stages of gametophyte development were cleared and stained with acetocarmine-choral hydrate and drawn with a camera lucida for study (Whittier, 1981). Mucilage formation on the proximal cell and rhizoids was demonstrated by alcian blue staining (Whittier and Peterson, 1984). For the later developmental stages, the gametophytes were fixed with Randolph's Modified Navashin Fluid (CRAF) (Johansen, 1940). After fixation, the gametophytes were embedded in paraffin, and sectioned by conventional techniques (Johansen, 1940). The sections were stained with Heidenhain's hematoxylin, safranin O, and fast green. For scanning electron microscopy, the gametophytes were fixed overnight on ice in a 1:1 solution of 4% glutaraldehyde and 10% acrolein in 0.1 mol/L Hepes buffer (pH 6.8) (Whittier and Peterson, 1984). The gametophytes were postfixed in 1% osmium tetroxide in 0.1 mol/L Hepes buffer (pH 6.8) at room temperature for 1 hour. They were then treated with 1% aqueous thiocarbohydrazide for 30 minutes after the osmium postfixation. The gametophytes were refixed with 2% osmium tetroxide in water for 1 hour and then dehydrated in a graded acetone series. All specimens were critical point dried and coated with gold-palladium before observing with a Hitachi 4500 or 370 scanning electron microscope at 10 or 15kV. OBSERVATIONS The earliest germination occurred 8 days after the cultures were placed in the dark. After 3 weeks in the dark 40% of the spores had germinated. With WHITTIER: GAMETOPHYTE longer dark periods, up to 96% of the spores germinated. Spores maintained for a year in cultures that were illuminated for 14 of every 24 hours did not germinate. Shorter periods of illumination were also sufficient to stop germination. Daily exposures to 15 minutes of light prevented germination in a 28 day experiment. The spore coat cracked open at the triradiate ridge (Fig. 1) to initiate germination. Shortly after the spore coat ruptured, the spore divided perpendicular to its polar axis (Fig. 2). A proximal cell (near the triradiate ridge) and a distal cell (away from the triradiate ridge) were formed. As the two cells expanded the proximal cell bulged out of the spore coat forcing its lobes apart. The distal cell remained inside the spore coat and continued to divide. The second division was more or less parallel to the polar axis of the spore and divided the distal cell into two cells (Fig. 3). The third division, which was usually perpendicular to the polar axis of the spore, occurred in one of the two distal cells (Fig. 4). The fourth division occurred in the other of the two distal cells and its plane was usually perpendicular to the plane of the third division (Fig. 5). The divisions in the 5-celled and larger gametophytes were not followed because more variation existed in the later sequence of divisions and the shape of the young gametophytes made them difficult to follow. At about the 5-celled stage the proximal cell was fully extended beyond the spore coat (Fig. 6). Once this happened, mucilage, which stained for acid mucopolysaccharide with alcian blue, was secreted at the exposed end of this cell (Fig. 7). When fully secreted it took the shape of a triangular ring. The production of mucilage was not dependent on the availability of sugar because it formed at the same stage on gametophytes grown on a medium lacking sugar. AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 3 " wB , *4m WHITTIER: GAMETOPHYTE MATURATION IN OPHIOGLOSSUM CROTALOPHOROIDES 141 Shortly after the 5-celled stage, the gametophyte became free of the spore coat. With additional cell divisions, a small spherical or globular gametophyte was formed (Figs. 8, 9). It was at the spherical stage that rhizoids began to develop (Fig. 9). Regardless of the gametophyte age the rhizoids of O. crotalophoroides secreted mucilage that stains with alcian blue (Fig. 10). The small globular gametophytes grew into short cylindrical gametophytes. At 100 days the gametophytes on the average were 0.3 mm long and 0.2 mm wide. By the time the gametophytes were 0.6 mm long their apical regions had expanded to a width of 0.4 mm. At this stage the gametophytes were conical or teardrop shaped (Figs. 11, 12, 14). The basal regions of these teardrop-shaped gametophytes had numerous rhizoids. The apical regions lacked rhizoids and the youngest of these gametophytes lacked gametangia. Antheridia first formed 4.5 months after sowing the spores (Fig. 11). Gametophytes with 1-3 antheridia averaged 0.7 mm in length and 0.5 mm in width. The antheridia were almost completely sunken into the gametophyte tissue but are recognized at the surface by slightly raised areas (Figs. 11, 12). The antheridial jacket is two cells thick except at the opercular cell (Fig. 13). Although slightly longer than wide, the mass of sperm had almost a spherical shape. The average size of the sperm mass was 148 urn in length by 138 urn in width. At 6.5 months the gametophytes began to form archegonia (Fig. 14). Gametophytes with 1-3 archegonia were on average 1.0 mm long and 0.7 mm wide. The archegonia had short necks with usually 2-3 tiers of neck cells exposed above the gametophyte surface (Fig. 15). Their average length from base of egg to tip of the neck was 160 um (Fig. 16). Once the gametophytes formed archegonia at 6.5 months they were sexually mature. Their gametangia are functional under these cultural conditions because sporophytes developed in cultures with moisture on the surface of the nutrient medium (Fig. 17). Functional fertile spikes did not form on these young sporophytes, however in some cases an abortive fertile spike was associated with the first leaf (Fig. 17, arrow). DISCUSSION AND CONCLUSIONS The average germination times for green pteridophyte spores and non-green fern spores are 1.5 and 9.5 days respectively (Lloyd and Klekowski, 1970). Spores of the Ophioglossaceae were not studied by Lloyd and Klekowski (1970). The average germination times of 54 and 37 days for spores of Ophioglossum (excluding O. crotalophoroides) and Botrychium respectively (Table 1) support the generalization that spore germination is slow for this s opercular cell of an a AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 3 Cm.riopllVtr with two voung archegonia 500 urn. 15. Archegonia, bar = 250 um. gametophyte with an an mynniimi (arrow) arrow indicates abortive fertile spike, bar - (arrows) and s-.mkr•n antln •ridi a (arrowheads), bar = thn mgh apical region o 16. Longitudi and antheridia 17. Young sporophyte = 100 u 5 mm. family. The germination of spores of O. crotalophoroides in 8 days is the second fastest germination reported for the Ophioglossaceae (Table 1). Compared with the average germination times for this family, the spores of O. crotalophoroides germinate rapidly. Spore germination in this species is also faster than the average germination time (9.5 days) for other non-green fern spores (Lloyd and Klekowski, 1970). The pattern of cell divisions in the early development of the gametophytes of O. crotalophoroides is basically the same as reported for Botrychium and other species of Ophioglossum (Whittier, 1981). There was nothing unusual about WHITTIER: GAMETOPHYTE MATURATION IN OPHIOGLOSSUM CROTALOPHOROIDES TABLE Ophioglossum crotalophoroides1 engelmannir moluccanunr pendulum2 pusillum' 1. 143 Days to spore germination in the Ophioglossaceae. 8 71 Present study Whittier. umpubl. 3 36 90 Campbell, 1907 Campbell, 1907 Whittier, unpubl. the first 4-5 divisions after germination. The formation of mucilage on the exposed proximal cell of O. crotalophoroides appears normal for this family. It has been found previouly in species of Botrychium (Melan and Whittier, 1989). The production of mucilage on the rhizoids of O. crotalophoroides is typical for the Ophioglossaceae. Mucilage has been found on the rhizoids of Botrychium species examined from axenic culture (Whittier and Peterson, 1984). It has not been reported for other species of Ophioglossum because they did not develop rhizoids under culture conditions. The gametangia that developed were similar to those found on gametophytes of O. crotalophoroides from soil. The antheridia were almost completely sunken with a bistratose jacket and, as reported by Mesler (1976), a single opercular cell. Archegonia on gametophytes from culture had short exposed necks that are similar to those on gametophytes from soil (Mesler, 1976). The gametangia on these gametophytes were normal for Ophioglossum (Pant et al, 1984). A major difference between the gametophytes of O. crotalophoroides from soil and culture is the absence of a mycorrhizal fungus in the cultured gametophytes. This is typical for normally mycorrhizal gametophytes growing in axenic culture. The sugar in the nutrient medium replaces the need for the mycorrhizal fungus as a carbon source. Whether the fungus under natural conditions supplies additional organic materials to the gametophyte is unknown at this time. Gametophyte lengths at day 100 from sowing and at the times of early antheridia and archegonia formation provided a chance to determine average growth rates. The growing time was computed as the time from sowing minus the time to germination. Using these calculations the average growth in length 144 AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 3 (2003) per day for gametophytes of O. crotalophoroides was 3.3 pm for the first 3 months after germination and 5.5 um for 4.2 and 6.2 months after germination. These rates were faster than the 2.5 um per day for gametophytes of Botrychium virginianum and B. dissectum forma obliquum for 4 months of growth after germination in culture (Whittier unpubl.). The average length and width of gametophytes of O. crotalophoroides with 1-3 antheridia was 0.7 mm by 0.5 mm. The 10 smallest gametophytes from soil with only antheridia averaged 1.0 mm long by 0.7 mm wide for B. dissectum and 1.0 mm long by 0.8 mm wide for B. virginianum (Foster, 1964). The average sizes for Botrychium gametophytes with antheridia from soil are presented because they were smaller than gametophytes of these species from culture with 1-3 antheridia (Whittier unpubl.). The average size of the 10 smallest gametophytes of B. dissectum with archegonia from soil was 1.6 mm long by 1.2 mm wide and that of the 6 smallest gametophytes of B. virginianum with archegonia was 1.9 mm by 1.4 mm (Foster, 1964). These average sizes were again smaller than those for gametophytes of these species bearing 2-3 archegonia from culture (Whittier, unpubl.). The average sizes of the Botrychium gametophytes with archegonia are larger than those of O. crotalophoroides bearing 1-3 archegonia which averaged 1.0 mm long and 0.7 mm wide. These comparisons show that the gametophytes of O. crotalophoroides from culture develop gametangia at smaller sizes than either Botrychium species. The time to sexual maturity for O. crotalophoroides at 6.5 months from sowing the spores is much faster than the 22 months reported for B. dissectum in culture (Whittier, 1972). Besides maturing faster than B. dissectum, these gametophytes mature much quicker than the perennial gametophytes studied by Bruchmann (1904), Campbell (1911), and Pant et al. (1984). The only gametophytes of the Ophioglossaceae that may mature as fast are possibly the annual gametophytes of O. moluccanum (Campbell, 1907). The rapid maturation of the gametophytes of O. crotalophoroides in axenic culture helps to confirm the report of rapid reproduction in this species by Mesler (1976). The accelerated maturation of these gametophytes is promoted by each of the following factors. Quick spore germination initiated gametophyte development sooner. Rapid growth produced larger gametophytes in a shorter time. The formation of antheridia and archegonia on smaller gametophytes reduced the amount of growth necessary to attain maturity. Collectively, these conditions—rapid germination, rapid growth, and reduced amount of gametophyte tissue necessary for gametangia formation—bring about the accelerated sexual maturity of these gametophytes. iank R. Dale Thomas at Northeast Louisiana I nhersih tor assistance in obtaining the spores crotalophoroides. The spores were supplied or were from plants at sites located by him. I also c R. L. Peterson for use of his laboratory facilities at the University of Guelph (Canada) where WHITTIER: GAMETOPHYTE MATURATION IN OPHIOGLOSSUM CROTALOPHOROIDES 145 LITERATURE CITED B. 1963. Le gametophyte des Ophioglossacees. Considerations biologiques. Bull. Soc. Linn. Normandie 4:81-97. H. 1904. Ueber das Prothallium und die Keimpflanze von Ophioglossum vulgatum L. Bot. Zeitung, 2. Abt. 62:227-247. CAMPBELL, D. H. 1907. Studies on the Ophioglossaceae. Ann. Jardin Bot. Buitenzorg 6:138-194. CAMPBELL, D. H. 1911. The Eusporangiatae. Publ. Carnegie Inst. Wash. No. 140. FOSTER, D. B. 1964. The gametophytes and embryogeny of five species of Botrychium. Ph.D. Thesis, Cornell University, Ithaca, New York. JOHANSEN, D. A. 1940. Plant Microtechnique. McGraw-Hill, New York. LLOYD, R. M. and E. J. KLEKOWSKI JR. 1970. Spore germination and viability in Pteridophyta: Evolutionary significance of chlorophyllous spores. Biotropica 2:129-137. MELAN, M. E. and D. P. WHITTIER. 1989. Characterization of mucilage on the proximal cells of young gametophytes of Botrychium dissectum (Ophioglossaceae). Amer. J. Bot. 76:1006-1014. BOULLARD, BRUCHMANN, Walt. Amer. J. Bot. 63:443^148. M. R., R. D. THOMAS and J. G. BRUCE. 1975. Mature gametophytes and young sporophytes of Ophioglossum nudicaule. Phytomorphology 25:156-166. NAYAR, B. K. and S. KAUR. 1971. Gametophytes of homosporous ferns. Bot. Rev. 37:295-396. Nozu, Y. 1961. The gametophyte of Helminthostachys zeylanica and Ophioglossum vulgatum. Phytomorphology 11:199-206. PANT, D. D., D. D. NAUTIYAL and D. R. MISRA. 1984. Gametophytes of Ophioglossaceae. Phyta Mon. MESLER, V. 1989. Developmental Biology of Fern Gametophytes, Cambridge University Press, Cambridge, U.K. D. E. G., B. H. FLETCHER and H. E. STREET. 1959. Studies on the growth of excised roots. VII. RAGHAVAN, SHEAT, Phytologist 58:128-154. D. P. 1964. The effect of sucrose; on apogamy in Cvrlomium falcatum. Amer. Fern WHITTIER, D. P. 1972. Gametophytes of Botrychium dissectum as grown in sterile culture. Bot. Gaz. 133:336-339. D. P. 1981. Spore germinaton and young gametophyte development of Botrychium and Ophioglossum in axenic culture. Amer. Fern J. 71:13-19. WHITTIER, D. P. and R. L. PETERSON. 1984. The gametophyte of Botrychium lunarioides and its mucilage coated rhizoids. Canad. J. Bot. 62:2854-2860. WHITTIER, D. P. and T. A. STEEVES. 1960. The induction of apogamy in the bracken fern. Canad. J. Bot. 38:925-930. WHITTIER, D. P. and R. D. THOMAS. 1993. Gametophytes and sporophytes of Botrychium axenic culture. Internat. J. Plant Sci. 154:68-74. WHITTIER, WHITTIER, Nomenclatural and Taxonomic Notes on the Pteridophytes of Costa Rica, Panama, and Colombia, III DAVID B. LELLINGER Botany Section, National Museum of Na1 Washington, DC 20560-016 The purpose of this paper and those that preceded it (Lellinger 1977a, 1977b, 1985) is to publish lectotypes, new combinations, and new species of pteridophytes that for the most part will be included in the forthcoming volume of my Ferns and Fern-allies of Costa Rica, Panama, and the Choco. All Morton photos were seen at US. Aspidium macrophyllum var. pittieri Christ in Dur. & Pitt., Bull Soc. Roy. Bot. Belgique 35, Mem. 208. 1896.—LECTOTYPE: Tsaki, Talamanca, Pcia. Limon, Costa Rica, ca. 200 m, Tonduz 9483 (US!; isolectotypes BR, CR!), designated here. The other syntypes are: Rio Yurquin [Zhorkin], Pcia. Limon, 50 m, Pittier 8523 (BR) and Puerto Viejo, Pcia. Heredia, Costa Rica, Biolley 6924 (BR; isosyntype CR!). The name is a synonym of Tectaria incisa Cav. Asplenium trianae Mett. in Tr. & Planch., Ann. Sci. Nat. Bot. V, 2:233. 1864.— LECTOTYPE: "Prov. de Barbacoas, via de Tuquerres," Depto. Narino, Colombia, 900 m, Triana in 1853 (BM-Morton photo 7049; isolectotype B), chosen here. The other syntype is: Ingara, Depto. Choco, Colombia, 340 m, Triana (B). the basionym of Diplazium trianae (Mett. in Tr. & Blechnum I'herminieri subsp. lehmannii (Hieron.) Lellinger, comb. nov. Blechnum lehmannii Hieron., Bot. Jahrb. Syst. 34:473. 1904.—TYPE: Rio Timbiqui, Depto. Cauca, Colombia, 100-500 m, Lehmann 8928 (B— Morton photo 10024; isotypes F!, K, US!). This subspecies has sterile laminae that taper gradually and uniformly toward the base of the lamina. It has a cordilleran distribution from Costa Rica to Bolivia and Brazil. In contrast, Blechnum I'herminieri (Bory ex Kunze) Mett. subsp. I'herminieri has sterile laminae that are abruptly tapered above the base LELLINGER: NOMENCLATURAL AND TAXONOMIC NOTES Blechnum loxense var. stenophyllum (Klotzsch) Lellinger, comb. nov. Lomaria stenophylla Klotzsch, Linnaea 20:346. 1 1847.—TYPE: Peru, Dombey (B-Morton photo 10092: isotype P-Morton photo 4399). Lomaria squamulosa Desv., Mem. Soc. Linn. Paris 6:290. 1827.—TYPE: Peru, Dombey (B-Morton photo 10092; isotypes P-Morton photo 4399, US!). This variety differs from the typical variety in having bicolorous stipe and rachis scales. It occurs from Colombia to Bolivia, whereas the typical variety ranges to Costa Rica and Venezuela. The epithet stenophylla has been more used than squamulosa, and so I have chosen to use the former at the varietal level. Diplazium ribae (Pacheco & R. C. Moran), Lellinger, comb. nov. Callipteris ribae Pacheco & R. C. Moran, Brittonia 51:375, f. 21. 1999.—TYPE: El Llano-Carti Road 17.4 km from the Interamerican Highway, Com. S. Bias, Panama, 350 m, deNevers, Herrera 6- Gonzalez 3924 (MO; isotype UC!). In my opinion, subdivision of the genus Diplazium sensu lato would best be delayed until more information about species that may be related to, but are not included in Callipteris, is at hand. A few other tropical American species and many Old World species of Diplazium sensu lato have some of the characters of Callipteris, and it is important that these species be dealt with in detail. Hypolepis rubiginosopilosula Lellinger, sp. nov. Rhizoma repens, 2(4?) mm in diam., leviter brunneopilosum. Stipites 30-45 cm longi spinosi ad basin rufobrunnei, distaliter pallide aurantiacobrunnei. Rachides sparse spinosae flavovirides, distaliter catenatopilosulae. Laminae lanceatae vel oblongo-lanceatae, 3-pinnato-pinnatifidae, usque ad 100 cm longae ca. 50 cm latae, ad basin pinnis lanceatis aequilateralibus, distaliter pinnis oblongis; costulis stramineis leviter catenatopilosulis, pilis articulatis leviter rufobrunneis; pinnulis secundariis vel segmentibus 4-7 mm latis oblongis ad apicem rotundatis chartaceis abaxialiter leviter glandulosis; venulis complanatis fuscis, hydathodis elongatis; soris ad apicem venarum in lobis demissis; indusiis 0.1(0.3) mm latis erosis sparse ciliatis. TYPE: Vicinity of El General, Pcia. S. Jose, Costa Rica, 1160 m, Skutch 2975 (US-2 sheets). PARATOPES: COSTA RICA: Cartago: Muheco, 1500-1500 m, Standley & Torres 51203 (US); Heredia: Parque Nal. Braulio Carrillo betw the R. Peje and the headwaters of the R. Sardinal, Atlantic slope of V. Barba, 1200-1300 m, Grayum 7820 (CR, MO, US); S. Jose: La Palma, 1459 m, Tonduz 12529 (US). FERN JOURNAL: VOLUME 93 NUMBER 3 (2003) PANAMA: Chriqui: Holcomb's Trail, 10 mi above Boquete, 1625-1650 m, Killip 5235 (US). This species, which occurs in the Cordillera Central of Costa Rica and the Cordillera de Talamanca of Costa Rica and Panama at 1100-1700(2100) m elevation, has generally been called H. rigescens (Kunze) T. Moore. The type of that species is from Est. Bahia, Brazil; it is known to me only from Morton photo 16280 of an isotype in Firenze (FI). Based on frond outline and on location and elevation of the type, I believe H. rigescens to be the earliest name for a species probably confined to the lowland Brazilian coastal forest that has usually been called H. mitis Kunze ex Kuhn. [Hypolepis stolonifera Fee is another synonym). Therefore, I have provided a new epithet for the Central American-Andean material, which differs from the Brazilian material in being equally pilosulous on both surfaces and in having erose, obviously ciliate indusia, rather than in being glabrous adaxially and in having erose, eciliate indusia. Apparently H. rubiginosopilosula is most closely related to H. viscosa (Karst.) Mett. in Tr. & Planch. The principal differences are that the latter has pilose, rather than sparsely pilosulous axes (both have catenate hairs) and lacks spines on the stipes, rachises, and costae. In Costa Rica, H. viscosa grows at 2100-2600 m elevation, entirely above the elevational range of H. rubiginosopilosula. The foregoing differences and differences in range [H. viscosa is known from Costa Rica to Venezuela and Ecuador) make it unlikely that specimens of H. rubiginosopilosula are merely variants of H. Lastreopsis squamifera (C. Chr.) Lellinger, comb. nov. Dryopteris exculta var. squamifera C. Chr., Kongel. Danske Vidensk. Selsk. Skr., Naturvidensk. Afd. VIII, 6:96. 1920. SYNTYPES: Navarro, Pcia. Cartago, Werckle 16741 (P), 16753 (P), and 16764 (P). This species is distinct from L. exculta in its pinnae, which are twice as far apart as those of L. exculta, and in its narrowly lanceate, brown, subclathrate scales, which are unlike the linear, blackish, clathrate scales of L. exculta. Lomariopsis salicifolia (Kunze) Lellinger, comb. nov. Lomaria salicifolia Kunze, Linnaea 9:58. 1834.—TYPE: Yurimaguas, Depto. Loreto, Peru, Poeppig in Dec 1830 (LZ destroyed). Lomariopsis fendleri D. C. Eaton, Mem. Amer. Acad. Arts N.S., 8:195. I860.—TYPE: Venezuela, Fendler 335 (YU; isotypes K fragm NY!, MO!). Despite a careful search by Dr. Bruno Wallnofer, no isotype of Poeppig's specimen was found at W, which has the first set of Poeppig's collections. According to my notes, which may be in error, NY apparently had a fragment of the isotype, but Dr. Robbin Moran could not find at the present time. Kunze compared his species to what is now called Lomariopsis sorbifolia (L.) Fee, a well known Antillean species. Although the characters he used to LELLINGER: NOMENCLATURAL AND TAXONOMIC NOTES 149 distinguish L. salicifolia are mostly those of the genus, that makes it more certain that Kunze had a Lomariopsis at hand. The only other possibilities are Blechnum, which does not routinely climb up tree trunks, and Stenochlaena, which does look very much like Lomariopsis. The latter is a strictly Old World genus, and Poeppig never collected in the Old World. Of the five species of Lomariopsis attributed to Peru by Moran (2000, p. 59), only L. fendleri D. C. Eaton has the frond and pinna dimensions and lamina shape approaching those of L. sorbifolia. The other Peruvian species are much larger plants whose laminae do not taper gradually at the base. Therefore, it is certain that Kunze's name is correctly applied to this species. Nephrodium sodiroi Baker, J. Bot. Brit. For. 15:16. 1877.—LECTOTYPE: The type specimen, "Andes of Ecuador," Sodiro (K-Tryon photo US!), is a mixed collection. According to Tryon and Stolze (1991), the type consists of a sterile frond of Bolbitis nicotianifolia (Swartz) Alston, a rhizome and stipe of an unidentifiable species of Lomariopsidaceae, and a fertile, Tectaria pinna. I here designate as lectotype the sterile frond of Bolbitis nicotianifolia, which is the basis for all of Sodiro's description, except for the soriation and indusia, which are taken from the Tectaria fragment. The name Nephrodium sodiroi thus becomes a synonym of Bolbitis nicotianifolia. The name Tectaria chimborazensis (C. Chr.) C. Chr., which has an adequate type specimen, applies to the Tectaria fragment and other material of this species that had been called Tectaria sodiroi (Baker) Maxon. Polypodium chirripoense Lellinger, sp. nov. Plantae epiphyticae. Rhizoma late repens 3-4 mm diam., phyllopodiis 4-12 mm longis, 0.3-2.5 cm distantibus, nigrum paleaceum, paleis lanceolatis peltatis appressis integris ca. 3-5 mm longis 1 mm latis ad marginem apicemque stramineis ad centrum atrobrunneis, marginibus apicibusque deciduis irregulariter erosis, pagina rhizomatis demum detecta. Stipites rachides laminaeque pilosulae vel pilosae, pilis 0.1-0.5 mm longis catenatis 3-6-cellulis, subhyalinis. Stipites (5)10-25 cm longi (0.8)1-2 mm lati exalati distaliter sulcati, ad basin atrobrunnei vel atrocastanei ad apicem brunnei, pilosi glabrescentes. Laminae anguste lanceolatae 13-32(45) cm longae (5)7-10 cm latae papyraceae ad basin obtusae pinnatae (pinnis basalaribus reductis 1.5-2.5(4) cm longis) ad apicem acutae vel acuminatae pinnatisectae vel pinnatifidae, pinnis segmentisque integris vel crenatis ad basin truncatis ad apicem attenuatis non falcatis leviter pilosulis (rachidibus pilosis), venulis (l)2(3)-furcatis; soris rotundis 0.75-1.5 mm diam. leviter supramedialibus 1-seriebus, sporangiis sparse pilosulis. TYPE: 1 km NW of Villa Mills on Interamerican Highway, behind the hotel La Georgina, Pcia. Cartago, Costa Rica, 2900 m elevation, Lellinger 853 (US; isotype CR). 150 AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 3 (2003) PARATYPES: Costa Rica: Cartago: Vicinity of Millsville, Pan-American Highway ca. 3 km above Nivel, 3000-3300 m, Holm & litis 604 (US; isoparatype GH). Cartago-San Jose: Upper slopes, western ridge of Cerros Cuerici, 3160 m, Davidse 24696 (UC!; isoparatype MO); Cerros Cuerici, near the summit, 3200-3394 m, Davidse 24783 (UC!; isoparatype MO). San Jose: Southwest slopes of Cerro Chirripo, along trail from Canaan to summit, near the cavern, 9800-10300 ft, Lellinger & Evans 105 (US!; TENN!). Limon: Atlantic side of Cerro Chirripo, 10400-11000 ft, Lellinger Er Evans 164 (TENN!); Atlantic side of the Kamuk Massif, E of the main peak, 3000-3300 m, Davidse & Herrera 29327 (MO!). This species is closely related to Polypodium ursipes Moritz ex C. Chr., with which it has been confused. It differs from that species in having dark brown, pilose rather than mostly grayish, densely pilosuluous rachises, round rather than elongate sori, and generally thinner rhizomes. Both species have basally tapering rather than truncate laminae and have similar rhizome scales. Polypodium chirripoense appears to be restricted to the central portion of the Cordillera de Talamanca and to grow at higher elevations (2900-3394 m) than does P. ursipes. It is mostly terrestrial, on fallen logs, or epipetric, but has been recorded as growing epiphytically, usually in mossy oak forests. Pteridium caudatum subsp. arachnoideum (Kaulf.) Lellinger, comb. nov. Pteris arachnoidea Kaulf., Enum. Fil. 190. 1824.—Type: Brazil, Chamisso (LE-Tryon photo GH). Because of differences in ploidy, totally or largely distinct ranges, and consistent differences in morphology, the specimens of this genus fall into at least two species, in the New World certainly into P. aquilinum (L.) Kuhn and P. caudatum (L.) Maxon. The major taxa within each of these species are, for similar reasons, logically treated as subspecies, although some of them may eventually prove to be independent species, based on cytological or other evidence. Tectaria xmichleriana (D. C. Eaton) Lellinger, comb. nov. Lindsaea michleriana D. C. Eaton, Mem. Amer. Acad., N.S. 8:213. 1860.— Type: Colombia, Depto. Choco, Near the falls of the Rio Truando, Schott 8 (YU photo and fragm US!; isotype NY!). This is the hybrid of T. incisa Cav. X T. panamensis (Hook.) Try on & A. Tryon. The latter species was formerly known as Dictyoxiphium panamense LITERATURE CITED 77a. Nomenclatural Notes on Some Ferns of Costa Rica, Panama, i LELLINGER: NOMENCLATURAL AND TAXONOMIC NOTES 151 D. B. 1977b. Nomenclatural and taxonomic notes on the Pteridophytes of Costa Rica, Panama, and Colombia, I. Proc. Biol. Soc. Wash. 89:703-732. LELLINGER, D. B. 1985. Nomenclatural and taxonomic notes on the Pteridophytes of Costa Rica, Panama, and Colombia, II. Proc. Biol. Soc. Wash. 98:366-390. MORAN, R. C. 2000. Monograph of the neotropical species of Lomahopsis (Lomariopsidaceae). Brittonia 52:55-111. TRYON, R. M. and R. G. STOLZE. 1991. Pteridophyta of Peru. Part IV. 17. Dryopteridaceae. Fieldiana, LELLINGER, SHORTER NOTES New Records for the Pteridoflora of Chiapas, Mexico.—In order to write the inventory of Pteridophytes of the Biosphere Reserve of "El Triunfo" and of "La Sepultura" and of other areas North of the state of Chiapas, an intensive plant collection was made. As a result, two new species ferns registered for Chiapas adding to the total number reported by Smith (Fl. Chiapas 2:1-370, 1981] and Breedlove {Listado Flonsticc de Mexico, IV, Flora de Chiapas, Instituto de Biologia-UNAM, 1986). These records should be added to the 693 species registered by Riba and Perez-Farrera (Amer. Fern. J. 90:104-111, 2000), to make a total of 695 species. This number is higher than the number of species registered for Oaxaca by Mickel and Beitel (Mem. New York Bot. Gard. 46:1-568. 1988) giving Chiapas the richest fern flora in Mexico. Elaphoglossum ipshookense Mickel {M.A. Perez-Farrera 435, Herbarium of the Escuela de Biologia UNICACH; UAMIZ) was collected in the municipality of Jiquipilas, Cerro Hojas Moradas, 6 Km W of the town "Los Alpes", Sierra Madre of Chiapas, "La Sepultura Biosphere Reserve, in mesophilous mountain forest, 1800 mis (16° 20' 30" N; 93° 42' 30" W)." This species is closely related to E. tectum (H. & B. ex Willd.) Moore, but differs from it in having a small blade and peltate scales on the petiole, rachis and upper surface of the sterile blade. This species was, until recently, only know from one collection [Mickel 4748, NY) from the Zempoaltepetl Hill, Mixe district, Oaxaca (Mem. N.Y. Bot. Gard. 46:1-568. 1988). Anemia guatemalensis Maxon [M.A. Perez-Farrera 1452, Herbarium of the "Escuela de Biologia" UNICACH; UAMIZ) was collected in Altamirano municipality, on the margins of the Tzaconeja river, 8 Km W of Altamirano in the physiographic region of the Eastern mountains in a Quercus forest, 1210 mis (16° 42' 10" N; 91° 59' 35" W). This species is very similar to A. karwinskyana [C. Presl.) Prantl., but differs from it in having a 2 pinnatepinnatifid blade and ovate to elongate-ovate segments. This species is distributed in southern Mexico and Central America south to Costa Rica. These new records are confined to the physiographic region of the Sierra Madre of Chiapas. This area is important as a Mesoamerican corridor for the distribution of pteridoflora. The first author thanks The Nature Conservancy, The Mac Arthur Foundation and SIBEJ-CONACYT, through the project 98SIBEJ-06-018, financial support of the project "Floristic Inventory of the "El Triunfo" Biosphere Reserve. We also thank Jesus de la Cruz Rodriguez, Oscar Farrera Sarmiento, Francisco Hernandez Najarro, Emerit Melendez Lopez and Tomas Acero for their help in the fieldwork and processing of plants.—MIGUEL ANGEL PEREZ FARRERA, Escuela de Biologia, UNICACH, A.P. 782, Tuxtla Gutierrez, Chiapas, 29000, Mexico, BLANCA PEREZ-GARCIA, SHORTER NOTES 153 UAM-Iztapalapa Ap. Postal 55-535, Mexico, D. F. 09340, RAMON RIBA. UAMIztapalapa Ap. Postal 55-535, Mexico, D. F. 09340 and MARIA E. LOPEZ-MOLINA, Escuela de Biologia, UNICACH, A.P. 782, Tuxtla Gutierrez, Chiapas, 29000, Mexico. Corrections and Additional Information on Ferns from the Semi-Arid Region of Brazil.—The publication by Ambrosio and de Melo (Amer. Fern J. 91(4): 227-228. 2001) of three new records from the semi-arid region in northeastern Brazil requires clarification. The purported new records involve Acrostichum danaeifolium Langsd. & Fisch., Thelypteris interrupta (Willd.) Iwatsuki, and Marsilea quadrifolia L. The taxonomic conclusions by Ambrosio and de Melo were based on a comparision of their findings with a list published by Barros et al. (Biol. Bras. 1: 143-159. 1989a). Although the paper by Barros et al. (1989a, op. cit) presented an interesting list of species for the "Caatinga" in Pernambuco State ("Caatinga" is a local name to referring to semi-arid vegetation), it is only a preliminary account of the pteridophytes found in this region, and is by no means a complete statement of our knowledge of the ferns from this area. According to Proctor (Ferns of Jamaica: 591. 1985), M. quadrifolia is native to southern Europe, Asia, and Japan, and is naturalized in North America. Johnson, in a revision of Marsilea for the New World (Syst. Bot. Monogr. 11:187. 1986), showed its distribution in North America and also presented interesting comments on accidental dispersal of M. quadrifolia by man, birds, and water in United States. Johnson did not mention this species for Brazil. Kuhn (in Martius, Flora Brasiliensis v. 2, part 1: 650-652, tab. 80, fig. 1-5. 1881) cited two species of Marsilea for semi-arid regions in Brazil: M. polycarpa Hook. & Grev. and M. deflexa A. Braun. Johnson also cited the same two species and presented a distribution map showing M. polycarpa in the Petrolina region (Pernambuco State). The material cited by both Kuhn and Johnson [Martius s.n., M) was collected during the historic travels of Martius through Brazil, in the state of Bahia, near Juazeiro. It is well known that the Martius expediton visited several Brazilian semi-arid regions including those in northern Minas Gerais, Bahia (city of Juazeiro), Pernambuco (city of Registro do Juazeiro: oldest name for Petrolina), and Piaui (city of Oeiras) states. Juazeiro is located south of the city of Petrolina and between the two cities is the Sao Francisco River. Barros et al. (Acta Bot. Brasil. 2(1-2): 47-84. 1989b) also recorded M. quadrifolia from "Sertao do Araripe", another semi-arid zone in the state of Pernambuco. No information about these historical occurrences or literature was included in the note by Ambrosio and de Melo (2001, op. cit.). I conclude that M. quadrifolia is a misidentification and thus not a new record for the area. Most likely, the material from Petrolina collected by Ambrosio (Ambrosio 52, TSAH) is one of the species previously cited by Kuhn and Johnson for that region in Brazil. Marsilea polycarpa can be distinguished from M. deflexa by its numerous, small (less than 3 mm long), terete sporocarps borne on the proximal 2/3 of the stipes 154 AMER FERN JOURNAL: VOLUME 93 NUMBER 3 (2003) (vs. 1-4 sporocarps 4-6 mm long, angled in cross section, with several lateral ribs, and on proximal % of the stipes). Thelypteris interrupta was previously cited for this same region by Baker (in Martius, Flora Brasiliensis v. 2, part 1: 486-487, t. 30, fig. 13. 1870) and by Andrade-Lima (Anais XX Congr. Nac. Bot.: 33-39. 1969) as Nephrodium unitum R. Br., and by Barros et al. (1989b, op. cit.) as Thelypteris totta (Thunb.) Schelpe. This species is recognized by its long-creeping rhizomes, proximal pinnae the longest or nearly so, basal veins from adjacent segments united at an obtuse angle below the sinus with an excurrent vein to the sinus, and laminae chartaceous to subcoriaceous, 1-pinnate-pinnatifid, abaxially with sessile reddish glands. Acrostichum danaeifolium was previously cited by Baker (in Martius, 1870, op. cit.) as common and widespread in Brazil, but its occurrence in Pterolina could be, in fact, a new record. I am greatful to Dr. Alan R. Smith for constructive comments on the manuscript.—JEFFERSON PRADO, Instituto de Botanica, Segao de Briologia e Pteridologia, C. P. 4005, 01061-970 Sao Paulo - SP, Brazil. Diellia mannii (D. C. Eaton) Robins. (Aspleniaceae) Rediscovered in Hawai'i.—Diellia mannii (D. C. Eaton) Robins, is a rare endemic species of the island of Kauai. It was first collected by H. Mann and W. T. Brigham as Microlepia mannii D. C. Eaton (Mann, Enumeration of Hawaiian plants. Proc. Amer. Acad. Arts and Sci. 7, 1867) sometime between 1864 and 1865. Last known collections were probably made by V. Knudsen during the period 1871-1886. About 24 collected specimens of D. mannii are deposited in different herbaria around the world. Some of those may originate from the same individuals (Wagner, Univ. Calif. Publ. Bot. 26:1-167, 1952). Although these collections provide little information about exact sites and habitats, all the specimens probably were collected in Western Kauai in the general area of Halemanu, in dry or mesic forests on the steep slopes of gulches, at an altitude of 500-1000 m (Wagner, Wagner & Flynn, Contr. Univ. Michigan Herb. 20: 241-260, 1995). Diellia mannii has probably always been a rare and very local fern species. Already in 1902, Diels (Polypodiaceae, pp. 139-339 in Engler & Prantl Die naturlichen Pflanzenfamilien Bd.l (Abt.4), Verlag von Wilhelm Engelmann, Leipzig) referred to it as a rarity of Kauai. The note of A. S. Knudsen from 1914 (Wagner et al., 1995) included mention of D. mannii as a very rare fern that has almost disappeared from the Halemanu in Koke'e Mountains. The status of the species has been assessed as probably extinct (Fosberg & Herbst, Allertonia 1: 1-72. 1975; Wagner, Wagner, Palmer & Hobdy, Contr. Univ. Michigan Herb. 22: 135-187. 1999), not seen after 1900 (Wagner et al. 1999; U.S. Fish & Wildlife Service species List. 2000), but considered to be a species of concern as "further field research may reveal that D. mannii still exists somewhere in (Koke'e Resource Conservation Program) during forest weeding work in Halemanu, Koke'e State Park. The only known individual of Diellia mannii is growing on a steep (ca 40°45°) northwest-facing slope just above a gulch bottom at an altitude of 1050 m. The natural community was at one time most likely dominated by an AcaciaMetrosideros montane mesic forest. Currently, the original vegetation has been degraded and the area is dominated by Corynocarpus laevigatus J. R. Forster & G. Forster. A few native trees {Acacia koa A. Gray, Metrosideros polymorpha Gaud., Hedyotis terminalis (Hook. & Arnott) W. L. Wagner & Herbst, Nestegis sandwicensis (A. Gray) Degener, I. Degener & L. Johnson and Coprosma waimea Wawra) are also present but of these, only A. koa has seedlings. Canopy coverage is ca 75%. The understory is sparse, with a coverage of ca 15%, consisting mainly of ferns and some grasses [Panicum nephelophilum Gaud.). Diellia mannii grows in the middle part of the slope where Asplenium macraei Hook. & Grev. is the most frequent pteridophyte species. Less commonly native Athyrium microphyllum (Sm.) Alston, Doodia kunthiana Gaudich., Dryopteris glabra (Brack.) Kuntze and Microlepia strigosa (Thunb.) 156 AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 3 (2003) C. Presl, and the naturalized Blechnum glandulosum Link and Christella parasitica (L.) H. Lev. were also found. The soil is silty with decomposing basalt, dry to moderately moist and sparsely covered with leaf litter. Suitable habitat conditions for D. mannii cover an area of ca 100-200 m2. In June of 2002, the Diellia mannii plant had five slightly arching, finely dissected fronds of 20-36 cm in length. Of these, two were older and senescent, three were younger and one was still uncurling. Like other species of Diellia, it had persistent stipe bases. Stipes were 2-3 mm in diameter and densely covered with tan-brown clathrate scales. Pinnae of erect young fronds were nearly perpendicular to the rachis. According to Wagner's 1952 description basal pinnules should be somewhat shorter than median pinnules. On this individual the basal pinnules were larger than the median pinnules and the pinnae were more elongate triangular than lanceolate, as per the original description of Mann (1867) and specimens described by Hillebrand as D. knudsenii var. oc (Hillebr.) Diels and D. knudsenii var. p Hillebr. (Hillebrand, Flora of the Hawaiian Islands. A description of their phanerogams and vascular cryptogams. London, New York, Heidelberg, 1888; Diels, 1902). The only individual of Diellia mannii in Halemanu is healthy and fertile. Unfortunately no other individuals at any life stage have been found in the area, despite a thorough search. Whereas the principal associate species, the highly variable and finely dissected A. macraei, is present in all life-stages. Asplenium macraei becomes fertile at quite an early age—young and small individuals having fronds with linear sori. Juvenile individuals of D. mannii have never been found. On the basis of previous research (Aguraiuja, CBM:s Skriftserie 3:7-24. 2000), it is hypothesized that juvenile D. mannii has much longer fronds than young fertile A. macraei and so it should be relatively easy to differentiate between the individuals of these two species in their early life stages. The main threats to Diellia mannii include trampling of the forest understory and possible herbivory by introduced feral deer and pigs; spatial competition with non-native species such as Blechnum glandulosum, Christella parasitica, Rubus argutus Link and Erharta stipoides Labill., which possess the ability to spread rapidly and effectively cover large areas in the forest understory; catastrophic extinction through environmental events; and reduced reproductive vigour as the result of limited numbers of existing individuals. Considering the highly endangered status of D. mannii the surrounding area should be fenced. Efforts for the monitoring and propagation of this fern should be supported. This study was financed by the Estonian Science Foundation (grant No. 4468 to M. Zobel). We thank Koke'e Natural History Museum and Koke'e Resource Conservation Program for their kind support.—R. AGURAIUJA, Institute of Botany and Ecology, University of Tartu; Tallinn Botanic Garden, Kloostrimetsa tee 52, Tallinn 11913, Estonia and K. R. WOOD, National Tropical Botanical Garden, Department of Conservation, 3530 Papalina Road, Kalaheo, Kaua'i, Hawaii 96741. Kaempferol and Quercetin 3-0-(2",3'-di-0-p-coumaroyl)-glucosides from Pteris vittata.—Previous work on the flavonoids of Pteris vittata L. has led to the identification of luteolinidin 5-O-glucoside by Harborne (Phytochemistry 5:589-600, 1966); in addition acid hydrolysis of extracts of this fern has led to the identification of kaempferol, quercetin, leucocyanidin and leucodelphinidin by Voirin (Ph. D. Thesis, University of Lyon, p. 151, 1970). More recently 3-C-(6'"acetyl-(3-cellobiosyl)-apigenin (Amer. Fern J. 89:217-220, 1999) and 6-C-f3-cellobiosyl-isoscutellarein-8-methyl ether together with quercetin 3-0glucuronide and rutin (Amer. Fern J. 90:42-47, 2000) have been identified by Imperato and Telesca. In addition three kaempferol glycosides (3-O-glucoside, 3-O-glucuronide and 3-0- (X", X"-di-protocatechuoyl)-glucuronide) together with quercetin 3-0-(X", X"-di-protocatechuoyl)-glucuronide (Amer. Fern J. 90:141-144, 2000) and two di-C-glycosylflavones (3,8-di-C-arabinosylluteolin and 6-C-arabinosyl-8-C-glucosylluteolin) (Amer. Fern J. 92:244-246, 2002) have been found by Imperato. For the present paper, three flavonoids (I—III) have been isolated from aerial parts of Pteris vittata L. collected in the Botanic Garden of the University of Naples. The fern was identified by Dr. R. Nazzaro (University of Naples); a voucher specimen [149.001.001.01] has been deposited in the Herbarium Neapolitanum (NAP) of the University of Naples. Flavonoids (I—III) have been isolated by preparative paper chromatography in BAW (ji-butanol:acetic acid:water, 4:1:5, upper phase), 15% HOAc (acetic acid) and BEW (n-butanol:ethanol:water, 4:1:2.2) from an ethanolic extract of aerial parts of Pteris vittata. Further purification was carried out by Sephadex LH-20 column chromatography eluting with methanol. Rf values on Whatman No 1 paper (0.74 in BAW; 0.43 in 15% HOAc; 0.42 in CHCl3-HOAc-H20, 30:15:2 (CAW)) and ultraviolet spectral analysis with the customary shift reagents Xmax (nm) (MeOH) 312, 263; +A1C13 394, 302, 275; +A1C13/HC1 393, 301, 274; +NaOMe 382, 274; +NaOAc 380, 273 suggested that compound (I) may be a flavonoid with free hydroxyl groups at positions 5, 7 and 4'. According to Harborne and Williams (pp. 376-441, in J. B. Harborne, T. J. Mabry and H. Mabry, eds., The Flavonoids, Chapman and Hall, London, 1975) the ultraviolet spectrum of flavonoid (I) suggested that this compound may be acylated with a cinnamic acid since the cinnamic acid absorption is superimposed on the flavonoid spectrum. Total acid hydrolysis (2N HCl; 1 hr at 100°C) gave kaempferol and D-glucose whereas alkaline hydrolysis (2N NaOH; 2 hr in a sealed tube at room temperature) gave kaempferol 3-Oglucoside (astragalin) and p-coumaric acid. Electrospray mass spectrum showed a pseudomolecular ion at m/z 763 [M+H+Na]+ and an ion at m/z 1503 [M 2+H+Na]+ (dimer); hence two p-coumaroyl groups are linked to kaempferol 3-O-glucoside. Treatment of I with acetone in the presence of dry CuS04 gave a monoisopropylidene derivative; methylation (methyl iodide in the presence of Ag20 in dimethylformamide in the dark with stirring; 18 hr at room temperature) of the isopropylidene derivative gave a permethyl ether which showed [M+H]+ at m/z 879 in the El-mass spectrum; hence hydroxyl groups at positions 4" and 6" AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 3 (2003) ~o- CH = CH C O of D-glucose are free according to Woo et al. (Phytochemistry 18:353-355, 1979). The above data show that flavonoid (I) is kaempferol 3-0-(2",3"-di-0-pcoumaroyl)-glucose (Fig. 1) which is a new natural product. The position of acyl groups were confirmed by 13C NMR spectrum (Table 1). C-6" and C-4" resonate at 8 60.8 and 8 68.1 respectively. These data show that hydroxyl groups at C-4" and C-6" of flavonoid (I) are free since the corresponding carbon atoms of astragalin resonate at 5 61 and 5 70.1, respectively as described in a review of Markham and Chary (pp. 19-51, in J. B. Harborne and T. J. Mabry eds., The Flavonoids: Advances in Research, Chapman and Hall, London, New York, 1982); the upfield shift of C-4" of flavonoid (I) is due to acylation at C-3". The chemical shift (5 76.4) of C-5" of I was similar to that of the corresponding carbon of astragalin (5 76.5); this observation confirms that hydroxyl groups at C-4" and C-6" are free. C-2" of flavonoid (I) resonated at 8 73.3 (C-2" of astragalin resonates at 5 74.2) since the downfield shift due to acylation is absent and there is an upfield shift due to acylation at C-3"; when a p-coumaroyl group is at C-2 of D-glucose, the downfield shift is often not observed (Markham and Chary, 1982 ) as in this case. C-3" resonated at 8 77.5 because there is a downfield shift due to acylation at C-3" and an upfield shift due to acylation at C-2"; the corresponding carbon of astragalin shows a signal at 8 77.2. C-l" of flavonoid (I) resonated at 8 99.1 showing an upfield shift due to acylation at C-2" since the corresponding carbon of astragalin resonates at 8 101.4. The structure of flavonoid (I) was i Flavonoid I Flavonoid I (sugar moiety) 3.20 3.40 3.56 4.80 5.38 5.52 6.05 6.22 6.37 (1H, (1H (2H (1H (1H (1H (1H (1H (2H m , H-4") m, H-5") 6.67 6.84 7.31 7.52 7.95 (4H (2H (4H (2H (2H br d, J=8.8, H-6'", H-8'" H- "", H-8"") d, J=9, H-3', H-5') "", H-9"") br d, J=8.8, H-5'", H-9'" br d, J=16, H-3'", H-3"") d>J=9,H-2',H-6') m, H-3") d, J=8, H-l") m, H-2") J=2. H-6) d. 1=2, H-8) br d, J=16 , H-2'" and H Flavonoid II 3.21 (1H, m, H-4") 3.42 (1H m, H-5") 3.57 (2H br d, 1=11.3, H-6") m, H-3") 5.40 (111 d, J=8, H-l") m, H-2") 1'507 (1H d, J=2, H-6) 6.24 (1H d, J=2, H-8) 6.38 (2H br d, J=16, H-2'", H-2"") 6.65 (4H 6.85 (1H d, 1=8.5, H-5') 7.31 (4H br d, J=8.8, H-5'", H-9'", 7.45 (2H br d, J=16, H-3'", H-3"") 7.55 (1H d, J=2, H-2') 7.68 (1H dd. 1=2, 1=8.5, H-6') ; C-6" C-4" C-2" C-5" C-3" C-l" 73.3 76.4 77.5 (sugar moiety) 60.8 73.3 76.5 C-6" C-4" C-2" C-5" C-3" 99.3 "", H-9"") confirmed by aH NMR spectrum (Table 1); acylation at positions 2" and 3" was confirmed by the presence of two oxymethine protons (8 5.52 (H-2") and 6 4.80 (H-3")) which showed a marked downfield shift as described in a review of Markham and Geiger (pp. 441-473 in, J. B. Harbome ed., The Flavonoids: Advances in Research since 1986, Chapman and Hall, London, 1994). Rf values on Whatman No 1 paper (0.70 in BAW; 0.39 in 15% HOAc; 0.38 in CAW) and ultraviolet spectral analysis with the customary shift reagents >.max (nm) (MeOH) 313, 262; +AlCl3 433, 329 (sh), 279; +A1C13/HC1 401, 365 (sh), 272; +NaOMe 405, 321 (sh), 277; +NaOAc 382, 271 suggested that compound (II) may be a flavonoid with free hydroxyl groups at positions 5, 7, 3' and 4'. In addition the ultraviolet spectrum of compound (II) was similar to that of flavonoid (I) suggesting that compound (II) may be a flavonoid acylated with a cinnamic acid. Total acid hydrolysis (2N HCl; 1 hr at 100°C) gave quercetin 160 AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 3 (2003) and D-glucose whereas alkaline hydrolysis gave p-coumaric acid and quercetin 3-O-glucoside. Electrospray mass spectrum showed a pseudomolecular ion at m/z 779 [M+H+Na]+ and an ion at m/z 1535 [M 2+H+Na]+ (dimer); these data show that two p-coumaroyl groups are linked to quercetin 3-O-glucoside. Treatment of flavonoid (II) with acetone in the presence of dry CuS04 gave a mono-isopropylidene derivative; methylation (with the method used for flavonoid (I)) gave a permethyl ether which showed [M+H]+ at m/z 909 in the El-mass spectrum. This result shows that hydroxyl groups at positions 4" and 6" of D-glucose are free according to Woo et al. (Phytochemistry 18:353-355, 1979). 1H- and 13C NMR spectra of flavonoid (II) were quite similar to those of flavonoid (I) and were in agreement with these observations (Table 1). The above data show that flavonoid (II) is quercetin 3-0-(2", 3"-di-0-p-coumaroyl)glucose (Fig. 1), a new natural product. The ultraviolet spectrum of compound (III) was similar to those of (I) and (II) suggesting that III may be a flavonoid acylated with a cinnamic acid. Total acid hydrolysis of III gave kaempferol and D-glucose whereas alkaline hydrolysis gave kaempferol 3-O-glucoside (astragalin), p-coumaric acid and ferulic acid. Since ultraviolet spectrum of III in the presence of usual shift reagents showed the presence of free hydroxyl groups at positions 5, 7 and 4', hydroxycinnamic acids are linked to D-glucose. Electrospray mass spectrum showed a pseudomolecular ion at m/z 793 [M+H+Na]+; hence one p-coumaroyl group and one feruloyl group are linked to D-glucose. The above data show that flavonoid (III) is kaempferol 3-0-(X"-0-p-coumaroyl-X"-0-feruloyl)-glucose. The presence of flavonoids (I—III) in Pteris vittata L. represents the first occurrence of diacylated flavonoid glycosides in Pteridophyta. Flavonoid glycosides with only one hydroxycinnamoyl group have previously been isolated from the fern genera Adiantum, Asplenium, Davalha, Pteridium, Brainea and Cheilanthes as described in a review of Markham (pp. 427-468, in J. B. Harborne ed., The Flavonoids: Advances in Research since 1980. Chapman and Hall, London, New York, 1988) and in a review of Imperato (pp. 39-75, in R. Uma ed., Current Topics in Phytochemistry, Vol. 3. Research Trends, Trivandrum, 2000). Recently it has been suggested that the Pteridaceae may be considered advanced from a phylogenetic point of view since flavone O-glycosides and Oglycosyl-C-glycosylflavones have been found in this family (Imperato, 2000). The presence of flavonoids (I—III) in Pteris vittata L. confirms the above suggestion since acylation of flavonol 3-O-glycosides may be considered an advanced biochemical character according to Markham (1980). The author thanks Murst (Rome) for financial support. Mass spectral data were provided by SESMA (Naples).—FILIPPO IMPERATO, Dipartimento di Chimica, Universita della Basilicata, 1-85100 Potenza, Italy. SHORTER NOTES 162 AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 3 (2003) high in the canopy of dry tropical forests in the eastern foothills of the Andes in Peru and Bolivia. It is most common in rain-protected valleys, at 200-400 meters elevation. In Peru little of these forests remain. There are, however, two protected regions in Peru where Platycerium andinum is native. One is the vast Parque Nacional Cordillera Azul in the Departments of San Martin, Huanuco, Loreto and Ucayali, which began as a reserve 7 September 2000. The other is El Quinillal reserve in the Department of San Martin, created 9 June 2001 due to an effort started by Roy Vail. In the herbarium of the University of San Marcos in Lima, Peru, there are eight vouchers with complete herbarium label data. Seven of these are from the valley of the central Huallaga River in the Department of San Martin, which indicates to us that Platycerium andinum was common there when the original forests were still present. Three new localities are reported here, one each from the Departments of Junin, Loreto and Puno. The Department of Loreto report of Platycerium andinum is very well documented in photographs, even though no herbarium material was deposited (pp 127 in Alverson, W. S., L. O. Rodriguez, and D. K. Moskowits (eds), Peru: Biabo Cordillera Azul, Rapid biological Inventories Report 2. The Field Museum, Chicago, 2001). During a study between 23 August and 14 September 2000, Platycerium andinum was found near the Pauya Campamento Orilla del Rio study site, (07°36'17.0-22.5"S. 75°56'26.3-28.0 W, ca. 360 m] in lowland forest on an alluvial fan terrace near the shore of the Rio Pauya in semidecidous forests. This location is in the first watershed east of the central Huallaga River. The Department of Junin report began in 1984 when an insect fancier, Mr. Clinton Callegari, reported to R. Fernandez that there were Platycerium in the area of Puerto Ocopa in the Chanchamayo valley. A field trip was made by R. Fernandez to verify the report. Much forest was being destroyed during construction of a new road from Boca Satipo to Puerto Ocopa, 85 kms N. E. of Satipo. Platycerium andinum was observed on tall trees of an unidentified species of Rubiaceae locally called "mohena." This is the other Platycerium andinum specimen in the herbarium of the University of San Marcos. (1 November 1984 USM R. Fernandez and C. Callegari 683) Its identity has been confirmed by B. Leon and R. G. Stoltze. The Department of Puno report is from M. Percy Nuez who located a single cluster of Platycerium andinum in a logged isolated pocket of dry tropical forest in Sandia Punto. Vouchers from this collection are in the herbarium of the Field Museum of Natural History, Chicago, and the University of San Marcos in Lima, Peru (9 August 2001 USM M. Percy Nuez et al 30273). This collection is unique because its altitude, 1,100 meters, is more than twice that previously reported for Platycerium andinum. Far more exploration is needed to locate other isolated pockets of dry tropical forest, and to determine their importance to the distribution of Platycerium andinum. It is possible that there are very narrow transition bands of dry tropical forest in which Platycerium andinum also occurs, or that the fern is adaptable enough for specimens to occasionally be found outside the dry tropical forest. Either would account for the following three reports: the discovery of an isolated specimen near Pucallpa in the Department of Ucayali by plant dealer Lee Moore in 1962 (pictured pp 1143 in Graph, A. B., Exoticia International (Series 4) Library Edition, Volume 1, Roehrs Company, East Rutherford NJ, 1982; described later in, Moore, L., The Discovery of Platycerium andinum, LAWS Journal Vol. 23, No. 2: 30-31, 37 Feb. 1996); the report to Fernandez by an Austrian herpetologist of a Platycerium in the Panguana area of the Llullapichis River near Puerto Inca in the Department of Ucayali; and Hennipman and Roos finding a Platycerium andinum herbarium specimen with a "rather illegible—label with the locality 'Ecuador'," (pp 84 in Hennipman E., Roos, N. C, A monograph of the fern genus Platycerium, North-Holland Publishing Company Amsterdam, 1982). More exploration needs to be done in northern Peru for small transition bands of dry tropical We consider it very possible that Platycerium andinum will be found in southern Peru between the Tambopata Candamo Reserve and the boarder with Bolivia, since this area is not far from the dry tropical forest of the Machiriapo River valley in Bolivia, a location where Platycerium andinum was located (A. Gentry, R. Foster, in A Biological Assessment of the Alto Madidi Region and adjacent areas of Northwest Bolivia May 18-June 15, 1990, Rapid Assessment Program, Conservation International, December 1991).—RICARDO FERNANDEZ, Museo de Historia Natural, Apdo. 14-0434, Lima 14, Peru and ROY VAIL, 200 Ridge, Mena, Arkansas, 71953 USA. A Modern Multilingual Glossary for Taxonomic Pteridology, by David B. Lellinger. 2002. Pteridologia 3:5-263. Published by the American Fern Society. Hardcover [ISBN 0-933500-02-5]. 263 pp. $28.00. Every field of study requires its own metric: a standard that can be employed to establish precision and insure accurate communication. Lellinger's glossary is that standard for systematic pteridology. The first sentence in the Introduction states that "Accurate communication is the essence of plant taxonomy." Without doubt, accuracy and its alter ego, conciseness, are the reasons scientific terminology is so extensive. In taxonomy, single words have evolved to depict precise, narrowly specific morphological conditions. Thus, a relatively short string of nouns and modifiers can provide a summation of a species hypothesis as well as define predicted boundaries with sister taxa. Unlike species, however, terms have no type specimens and in their absence the application of terms is likely to vary across a discipline as much as common names do across a continent. One need only look at any recent general biology text to see the degeneration of terminology. Examine, for example, the application of the word carpel in the more widely used biology or botany texts and it is clear that there is no common concept behind this widely used term. It is used variously for the entire gynoecium, for a pistil, or for an evolutionary and structural component of a compound pistil. This inappropriate diversity of usage is enhanced by the absence of well distributed, recent morphological glossaries. All too often the conceptual underpinnings of terms are lost to the everyday user. The "Glossary" contains an Introduction, a chapter on consulted references, 13 chapters of terminology, and four separate indices. As is true for all sections of the book, the short, explanatory Introduction is reproduced in English, French, Portuguese, and Spanish. The multilingual approach is unique and thus provides a single international source for fern characterizations. The main body of the work is divided into the following sections: Figure, Order and Division, Position, Growth, Substance, Surface, Gametophytes, Sporophytes, Anatomy, Cytology, Ecology and Distribution, Evolutionary Relationships, and Nomenclature. As should be expected in a work of such magnitude there are some regrettable omissions. Three specific examples that I have noted are the absence of aneuploid, dysploid, and epitype. I also would like to have seen a reproduction of the chart of terminology of simple symmetrical plane shapes published by the Systematics Association Committee for Descriptive Biological Terminology (Taxon 11:245-247, and reproduced in W. T. Steam. 1983. Botanical Latin, 3rd ed. David & Charles Publ., Great Britain), although because both are mandatory 'at-hand books', I do not lack for its absence. In the short time that I have had this glossary, I have used it at least once or twice every week. Already it is becoming a bit dog-eared from use. Thankfully I have three copies at had—one in my office, a second in my lab, and a third in our herbarium library. Lellinger's book is a must for all professional, and many avocational, pteridologists.—R. JAMES HICKEY, Botany Department, Miami University, Oxford, OH 45056. American Fern Journal 93(3):166-168 (: Index to Distribution Maps of Pteridophytes in Asia, 2nd Edition, by Toshiyuki Nakaike. Supplement No. 1 to the Journal of the Fernist Club, Tokyo. Vol. 3 (2002). 151 pages. Paper-bound (ISSN 0287-3257), USA $15.00 including postage. 8.25 by 11.75 inches. (In Japanese and English). Place orders to T. Nakaike, Natural History Museum & Institute, Chiba, 955-2, Aoba-cho, Chuo-ku, Chiba City 260-8682, Japan. Knowing the distribution of organisms is of basic importance in biological sciences. The study of ecology, evolution, biogeography, conservation, and many other disciplines are dependant upon knowing where organisms are distributed. For the scientist and naturalist, the publication, Index to Distribution of Maps of Pteridophytes in Asia is a welcome addition to the resource literature. Because of the broad application of this publication and because it may be used by readers of English, I would like to draw it to the attention of Western botanists. This book contains the literature sources that show distribution maps of Asian pteridophytes. The first few pages (pp. I-VII, in Japanese) point out the importance of distribution maps to biology and give the history of the index. The first edition (Nakaiki, 1998) was privately published and commemorated the completion of the monumental 8-volume work entitled Illustrations of Pteridophytes of Japan (Kurata & Nakaike, eds., Vol. 1-8. Pp. 5333. University of Tokyo Press, Tokyo, Japan, with the cooperation of the Japan Fernist Club [In Japanese with Latin names]). Among the information for each pteridophyte entry in these volumes is a map showing its distribution in Japan. In the 18 years it took to complete these volumes, much information needed to be added and updated. Updating the pteridophyte distribution on maps lead to the first edition of the Index to Distribution Maps of Pteridophytes in Asia. Further expansion and updating developed into this second edition (Nakaike, T. 1998. Index to Distribution Maps of Pteridophytes in Asia. Private press edition, Tokyo. 99 pp. [In Japanese and English, Latin names]). The pages in the next section (pp. VII & VIII, in Japanese) explains how to use the index, and delineates the terms and symbols used. The symbols are self-explanatory in Japanese or English . All ranks of Asiatic taxa from families to cultivars and nothospecies having maps are listed in the index. Taxa that extend beyond the Asiatic area are also included in the index if they are mapped. The following section (pp. 1-8, in English) is entitled Literature Cited. The 148 literature citations give the author, date of the publication, title of the paper, volume, number, page or publisher. The date of the latest literature citation is for 2001. The body of the index follows immediately (pp. 9-124). The names of the genera and of the species used are based on those given in the first edition HOSHIZAKI: REVIEW 167 of the Index to Distribution Maps of Pteridophytes in Asia (Nakaike, 1998), Illustrations of Pteridophytes of Japan, Vol. 8, pp. 467-473 (Kurata & Nakaike, 1997), and The New Flora of Japan Pteridophytes Revised and Enlarged (Nakaike, T. 1992. Shibbundo Co., Ltd. Publishers, Tokyo. 868 pp. [In Japanese with Latin names]). Synonyms are listed in the index and cross-referenced to the accepted name; names without authors (nomina nuda) but with maps are also included in the index. All entries are arranged alphabetically. After each taxon entry the region or country of the distribution map is given followed by the name of the author, the date of publication and the page. By noting the author of the maps and the bibliographic citation, the complete reference may be located in the Literature Cited section. A separate index (pp. 125-151) lists the Japanese names of the ferns in Japanese script. These names are cross-references to the scientific names in English. Every other page of the body of the index, whether in English or Japanese has a black and white drawing of a fern occupying slightly less than one-quarter of the page. Where the index is in English, captions to the picture are in English. The fern illustrated corresponds to a fern listed on the same page. In the Japanese index, the captions are in Japanese. Some of these handsome line drawings are credited to older publications but most are from recent or as yet to be published Japanese work. This book admirably fulfills its foremost function, and that is to help the researcher find distribution maps for Asian ferns. The nomenclature is updated and generic names are similar to those in Western usage. In any case, since common synonyms are listed, unfamiliar generic names are not a problem. The absence of author citations to the scientific names may be confusing for a few species. The inclusion of hybrids, varieties and cultivars in the listing is helpful for these categories are often omitted in other botanical indexes. Another use for this index is that it can serve as a checklist of all the known Japanese ferns. Names of Asian ferns and their literature sources are difficult to locate in many Western botanical libraries and most Western botanists are not familiar with Asian fern literature, so this index may be used as a reference source for a variety of purposes. The extensive listing of updated Asian fern names, though not complete, makes it a handy reference to rapidly check spelling and to locate other studies on Asian ferns through the literature cited. Particularly well represented are fern distribution maps of China (Guangzhou Province in particular), Thailand, Nepal (Katmandu) and monographs that have maps of Asiatic species. Other areas having an abundance of maps are Korea and Taiwan. Less frequently cited as having fern maps are Circumpolar areas, Malesia, Russia, India, Burma, and Vietnam. A few listings appear for the Mideast and Turkey. The paucity of maps does not necessarily mean that the fern distributions are not known, but rather it may be because maps are lacking. Considerable care was put into the editing of this publication and typographical errors are very rare, no small task when English is not your native language. The author is to be congratulated for undertaking such a laborious task to give fern workers such a helpful resource book. It will make the AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 3 (2003) research process that much easier in many disciplines and will be a handy reference to use for Asian ferns. That the working part of the book is in English and is so reasonably priced will make this publication well worth a place on the reference shelf of fern researchers in a variety of disciplines. I wish to express my gratitude to Takeko Hayashi for the translations from Japanese and to Kenneth A. Wilson for his editorial help.—BARBARA JOE HOSHIZAKI, 557 N. Westmoreland Ave., Los Angeles, CA 90004-2210. Note.—The Illustrations of Pteridophytes of Japan i Kurata. S. & T. Nakaike (eds.), 1979. Illustrations of Pteridophytes of Japan, vol. 1. 628 ] 1981. Illustrations of Pteridophytes of Jap; 1983. Illustrations of Pteridophytes of Japan, 1985. Illustrations of Pteridophytes of Japan, 1987. Illustrations of Pteridophytes of Japan, 1990. Illustrations of Pteridophytes of Japan, 1994. Illustrations of Pteridophytes of J 1997. Illustrations of Pteridophytes of J University of Tokyo Press, Tokyo, Japan. (In Japanese wi 3 1753 00312 7229 INFORMATION FOR AUTHORS Authors are encouraged to submit ma ligation in the American Fern Journal. Acceptance of papers for publication depends on merit as judged by two or more referees. Authors are encouraged to contribute toward publishing costs; however, the payment or non-payment of page charges will affect neither the acceptability of manuscripts nor the date of publication. Authors should adhere to the following guidelines; manuscripts not so prepared may be returned for revision prior to review. Submit manuscripts in triplicate (xerox copies acceptable), including review copies of illustrations and originals of illustrations. After review, submission of final versions of manuscripts on diskette (in PC- or Mac-compatible formats) is strongly encouraged. Use standard 8V2 by 11 inch paper of good quality, not "erasable" paper. Double space manuscripts throughout, including title, authors' names and addresses, short, informative abj and keys), literature cited, tables (separate from text), paragraphs separate from figures). Arrange parts of manuscript in order just given. Include author's name and page number in upper right corner of every sheet. Provide margins of at least 25 mm all around on typed pages. Do not submit right-justified copy, avoid footnotes, and do not break words at ends of lines. Make table headings and figure captions self-explanatory. Use S.I. (metric) units for all measures (e.g., distance, elevation, weight) unless quoted or cited from another source (e.g., specimen citations). For nomenclatural matter (i.e., synonymy and typification), use one paragraph per basionym (see Regnum Veg. 58:39-40. 1968). Abbreviate titles of serial publications according to Botanico-Periodicum-Huntianum (Lawrence et al., 1968, Hunt Botanical Library, Pittsburgh) and its supplement (1991). References cited only as part of nomenclatural matter are not included in literature cited. For shorter notes and reviews, omit the abstract and put all references parenthetically in text. Use Index Herbariorum (Regnum Veg. 120:1-693. 1990) for designations of hert page width with caption on the same page. Provide margins of a illustrations, design originals for reprodu amount. In composite blocks, abut edges of adjacent photographs. Avoid sequence and numbering of figures (and of tables) with order of citation in text. Explain scales and symbols in figures themselves, not in captions. Include a scale and reference to latitude and longitude in each map. Proofs and reprint order forms are sent to authors by the printer. Authors should send corrected proofs to the editor and reprint orders to the printer. Authors will be assessed charges for extensive alterations made after type has been set. For other matter of form or style, consult recent issues of American Fern Journal and The Chicago Manual of Style, 14th ed. (1993, Univ. Chicago Press, Chicago). Occasionally, departure from these guidelines may be justified. Authors are encouraged to consult the editor for assistance with any aspect of manuscript preparation. Papers longer than 32 printed pages may be sent to the Editor of Pteridologia (Memoir Editor, see cover 2). PTERIDOLOGIA ISSUES IN PRINT noir series of the American rtystichum in Western North Send your order with a check or money orde Inc.. 7t U.S. National Herbarium MRC-166, Smiths< DC 20560. AMERICAN FERN JOURNAL ON MICROFICHE Volumes 1-61 of the American Fern Journal are available as archival quality, silver positive microfiches. Single volumes or the entire run may be purchased. The fiches are easily read with 10X or greater magnification (using a dissecting microscope and transmitted illumination or A hche reader). Silver negative microfiches of vols. 1-50 are also available. The price is $4.00 per volume or $244.00 per set of 61 volumes, postpaid. Send your inquiry or order with a check or money order to: American Fern Society. Inc., cc Dr. James D. Montgomery, Ecology III. Inc., R.D. 1, Box 1795, Berwick, PA 18603. VISIT THE AMERICAN FERN SOCIETY'S WORLD WIDE WEB HOMEPAGE: http ://w ww.amerfernsoc.org/ AMERICAN FERN JOURNAL Volume 93 Number4 October-December 2003 QUARTERLY JOURNAL OF THE AMERICAN FERN SOCIETY Six New Species of Tree Ferns from the Andes Marcus Lehnert 169 Isoetes tennesseensis (Isoetaceae), an Octoploid Quillwort from Tennessee 184 Asplenium ofeliae (Aspleniaceae), a New Species from Luzon, Philippines A. Edward Salgado 191 Lycopodiella xgilmanii (Lycopodiaceae), a New Hybrid Bog Clubmoss from Northeastern North America Arthur Haines 196 The Common Staghorn Fern, Platycerium bifurcation, ilume 93 (2003) The American Fern Society Council for 2003 HRISTOPHER H. HAUFLER, Dept. of Ecology and Evolutionary Biology, Univen W. CARL TAYLOR, 800 W. Wells St., Milwaukee Public K I 53233-1478. JAMES D. CAPONETTI, Dept. of Botany, University of Tennessee, Knoxville, TN 37916-1110. GEORGE YATSKIEVYCH, Missouri Botanical Garden, P. O. Box 299, St. Louis, MO 63166-0299 JAMES D. MONTGOMERY, Ecology III. 804 Salem Blvd., Berwick, PA 18603-9801. EDITOR Botany Department,Miami University. Oxford, OH 45056, ph. (513) 529-6000, e-mail: hickeyrj@muohio.edu R. JAMES HICKEY GERALD J. GASTONY ASSOCIATE EDITORS Dept. of Biology, Indiana University, Bloomington, IN 47405-6801 Lawrence, KS 66045-2106 New York Botanical Garden, Bronx, NY 10458-5126 ROBBIN C. MORAN The ••American Fern Journal" (ISSN 0002 8444) is an illustrated quarterly devoted to the general stud> of terns. It is owned b\ ; . The American Fern Society % Missouri Botanical Garden, P. O. Box 299, St. Louis, MO 63166-0299. Periodicals postage paid at Subscriptions f and Mexico; $45.00 to elsewhere in the world (-$2.00 &25.00 + ind Mo »; 1 ruber up S300.00 +• $140.00 S7.(K) mailinsi surcharge be\ond I S.A Canad POSTMASTER: Send address changes to Box 299. St. Louis, MO 63166-0299. AMERICAN FERN JOURNAL, Missouri Botanical (Sard en, P. O. FIDDLEHEAD FORUM The editor of •'• Bty welcomes contributions from mem non-members, including miscellaneous notes, offers to exchange or purchase materials, personalia. ;ews of non-technical books on ferns. SPORE EXCHANGE Ms. Denia Mandt, 12616 Ibbetson Ave., Downey, CA 91 1 Spores exchanged GIFTS AND BEQUESTS services to members and to other- . otei t sted r gifts are always welcomed and are ta MISSOURI BOTANICAL ican Fern Journal 93(4):169-183 (2003) NOV 1 7 2003 GARDEN LIBRARY Six New Species of Tree Ferns from the Andes MARCUS LEHNERT brecht-von-Haller-Institut fur Pflanzenwissenschaften, Abteilung Systematische Botanik, Universitat Gtittingen, Untere Karspule 2, 37073 Gottingen - Germany ) are described and Bolivia; Alsophila ; n Colombia to The Family Cyatheaceae shows a pantropical distribution with a clear preference for the moist inner tropics. About 200 species are known from the Neotropics. Unlike Guatemala (Stolze, 1976), Peru (Tryon and Stolze, 1989) or Venezuela (Smith, 1995) no taxonomic treatment of the Cyatheaceae exists yet for Bolivia. In the scope of my master's thesis I conducted a revision of the Bolivian tree ferns (Cyatheaceae and Dicksoniaceae) that involved field studies of most species (Lehnert, 2002). Among the 34 recognized species, six are new to science and are described here. Traveling through Ecuador and Peru in 2002,1 had the chance to study many additional specimens of the new species in the herbaria of Quito (QCA, QCNE), Trujillo (HUT), and Lima (USM), and have been able to study the habit of some species for the first time. A full treatment of the Bolivian tree ferns is in preparation, but several taxonomic and systematic problems, especially among the Cyathea caracasanadelgadii alliance, remain to be resolved. The generic concept for the Cyatheaceae used here follows Lellinger (1987). The main literature consulted included Conant (1983), Gastony (1973), Moran (1991, 1995), Stolze (1984), Tryon (1971, 1972, 1976), and Windisch (1977, 1978). Alsophila minervae M. Lehnert, sp. nov. TYPE.—Bolivia, Dept. La Paz, Prov. Nor Yungas. 2 km de Chuspipata hacia Coroico, 16°22'S 67°94'W, 2900 m, 14 Septiembre 1997, M. Kessler 11900 (holotype: UC; isotypes: GOET, LPB). Fig. 1 A, B. Alsophila indusio globoso, foliis bipinnato-pinnatfidis, sectione apicali gradatim reducta, pinnis apicem versus alatis. Trunk to 3 m tall and 15 cm in diameter, with squaminate spines, without old petiole bases. Fronds to 220 cm long; petiole to 70 cm long, verrucate and with squaminate spines, petiole scales to 7 X 1.4 mm, with dark brown center and broad whitish margins, one apical seta. Petiole scurf consisting of light brown squamellae. Lamina ca. 100 cm wide, bipinnate-pinnatifid to -pinnatisect, its apical section gradually contracted (Fig. IB). Pinnae and pinnules sessile; distal portions of pinnae slightly green alate. Margins crenulate to serrulate (Fig. 1A). AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 4 . Kessler 1190 (UC). C, D. Alsophila mostellaria M. Lehru 1451 (UC). D. Sterile pinnule, abaxially, M. Kessler 11451 Costae and costules densely covered with brown hairs on both sides, hairs to 0.5 mm long, additional scurf of brown squamellae and few larger scales abaxially, especially in the junctures of rachis and costae. Leaf axes reddish- to orangebrown. Veins sparsely covered with brown trichomidia, 0.2 mm long; stellate hairs rarely present; with many flattish and bullate squamellae abaxially, LEHNERT: SIX NEW SPECIES OF TREE FERNS 171 whitish to brown with fringed margins (Fig. 1A). Fertile veins regularly forked except for the basal ones (Fig. 1A). Sori inframedial to costal; indusium cyatheoid to subsphaeopteroid, with irregular dehiscence (Fig. 1A). Paraphyses shorter than sporangia. Spores not examined. PARATYPES.—Peru. Dept. Pasco. Prov. Oxapampa. trail to summit of Cordillera Yanachaga via Rio San Daniel, 10°23'S 78°27'W, 2500 m, 17 July 1984, D.N. Smith, A. & H. Boetger 7817 (USM). Bolivia. Dept. La Paz. Prov. Nor Yungas. Carretera Chuspipata - Yolosa, entre Chuspipata y Sacramento Central, 16°17'S 68°48'W, 2700 m, 10 Noviembre 2002, M. Lehnert 495 (GOET, LPB, UC). Dept. Cochabamba. Prov. Carrasco. 115 km antigua carretera entre Cochabamba y Villa Tunari, 17°08'S 64°38'W, 2500 m, 05 Julio 1996, M. Kessler 7015 (GOET, LPB, UC). Alsophila minervae is named after the Roman goddess of wisdom. Alsophila minervae is sympatric with A. erinacea (Karst.) Conant but clearly prefers higher elevations; they are best distinguished by the normally broader petiole scales, well developed indument and distally green alate pinnae of A. minervae. The gradually contracted lamina apex (Fig. IB) clearly separates this new species from all other Andean Alsophila species with squaminate spines, which typically have an abruptly reduced apex. Alsophila minervae grows in the understory of wet montane forest at 25002900 m and ranges from central Peru south to central Bolivia. Alsophila mostellaria M. Lehnert, sp. nov. TYPE.—Peru, Dept. Amazonas, Prov. Bongara, Road Pedro Ruiz - Florida, wet quebrada along road side, 05°51.7'S 77°58.4'W, 2200 m, 05 August 2002, Lehnert 243 (holotype: USM; isotypes: GOET, UC). Fig. 1 C, D. Alsophila basi petioli pinnis subaphlebiatis instructa, petiolis spinulis brevibus squamiformibus obtectis, lamina bipinnato-pinnatifida apicem versus abrupte reducta; soris costae approximatis, indusio sphaeropteroideo usque meniscoideo fatiscente. Trunk to 6 m high and 9-10 cm in diameter, nearly black, with squaminate spines, completely covered by old spiny petiole bases, appearing sulcate. Fronds to 250 cm long and 110 cm wide. Petioles to 70 cm long, stramineous to orange-brown,proximally nearly black, with short squaminate spines, scurf sparse or absent. Petiole scales long-ovate, 1.6 X 8 mm, discordantly bicolorous with blackish brown center and brown margins, one apical seta. Aphlebioid pinnae in 1-2 pairs basally on the petiole, ca. 10 cm long (Fig. lC). Rachis stramineous abaxially, brown to black adaxially, with few squaminate spines basally, sparsely hairy adaxially, becoming denser towards the apex, without hairs abaxially; ephemeral scurf sometimes present. Lamina bipinnatepinnatifid, apical section abruptly reduced, pinnae and pinnules subsessile to short-stalked, margins crenulate or weakly serrate (Fig. ID). Pinnules to 75 mm long and 19 mm wide. Veins sparsely covered with brown trichomes, 0.6 mm long; no stellate trichomes present. Fertile veins forked. Sori subcostal to 172 AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 4 (2003) costal; indusium sphaeropteroid in young, fresh material, abraded by the sporangia to a meniscoid shape when mature or dried. Paraphyses as long as the sporangia or shorter. Spores not examined. PARATYPES.—Peru. Dept Cajamarca. Prov. Santa Cruz. Distrito Catache, upper Rio Zana valley, ca. 5 km above Monte Seco, near base camp clearing, ca. 1800 m, 02-04 May 1987, M. O. Dillon et al. 4883 (HUT). Dept. Amazonas. Prov. Bongara. Shillac, N by trail from Pedro Ruiz, 05°49'S 78°01'W, 2300 m, D.N. Smith 8t S. Vasquez-S. 4879 (MO, USM); Road Pedro Ruiz - Florida, wet quebrada along road side, 05°51.7'S 77°58.4'W, 2200 m, 05 August 2002, M. Lehnert 241 Er 242 (GOET, UC, USM). Dept. Pasco. Prov. Oxapampa. Road La Merced - Oxapampa, ca. 23 km from Oxapampa, 10°44.4'S 75°21.2'W, 1500 m, 27 August 2002, M. Lehnert 321 Er 322 (GOET, UC, USM); En propriedad del Sr. Espinoza, sigiuendo el riachuelo, (ca. 10°44'S ca. 75°21'W, ca. 1800 m), 21 Abril 1988, B. Leon Er K. Young 1743 (USM). Bolivia. Dept. La Paz, Prov. Caranavi. Serrania Bella Vista, 41 km de Caranavi hacia Sapecho, 15°41'S 67°30'W, 1450 m; 25 Agosto 1997, M. Kessler 11451 (GOET, LPB, UC). This new species is named after the comedy Mostellaria, also known as the "ghost comedy", by the Roman dramaturgian T. Maccius Plautus. Alsophila mostellaria also has been a "ghost" because for so long it was known only from a single sterile specimen. Alsophila mostellaria is unique among Andean tree ferns in having aphlebioid pinnae (Fig. IC). It differs from the similar aphlebiate species A. setosa Kaulf. and A. capensis (L.f.) J.Sm. in its sphaeropteroid to meniscoid indusium; the other two have hemitelioid indusia. Sterile specimens of A. capensis are easily separated by the higher dissection of the aphlebiae (= real aphlebiae), gradually reduced lamina apex, and the absence of squaminate spines on the petiole. Alsophila setosa and A. mostellaria share aphlebioid pinnae (= aphlebiae of coarse dissection), an abruptly reduced lamina apex, and long squaminate spines. Both species appear to be close allies, and there are no striking differences in sterile material: A. setosa may have some lateral seta on the petiole scales and generally has a deeper dissection of the pinnules. These features have not been observed in A. mostellaria. The most similar non-aphlebiate species regarding indument and lamina dissection is Alsophila incana (Karst.) Conant, which in Bolivia occurs only as far south as the Bolivian - Tucuman region. In Peru, however, both species are sympatric in Prov. Oxapampma, Dept. Pasco. Sterile specimens from this region cannot be told apart if the petiole is missing. In the field, both species are easily distinguished by the trunk which is free of petiole bases and light brown as a result of a thick cover of scales in A. incana, whereas A. mostellaria is nearly black and covered with old petiole bases. This species occurs from northern Peru to central Bolivia in wet forests and quebradas at an altitude of 1450-2300 m. Cyathea zongoensis M. Lehnert, sp. nov. TYPE.—Bolivia, Dept. La Paz, Prov. Murillo; Rio Zongo Valley, 22.5 km below dam at Lago Zongo, 16°09'S LEHNERT: SIX NEW SPECIES OF TREE FERNS 173 68°07'W, 3000 m. Cloud forest, low at 4-8 m tall. Abundant epiphytes, especially liverworts, 09 October 1982, J.C. Solomon 8429 (holotype: UC; isotype: MO). Fig. 2. Cyathea exindusiata acaulescens simili C. frigidae (Karst.) Domin et C villosae Willd., a C. frigida paraphysibus longioribus, a C. villosa squamis petioli bicoloribus latioribusque differt. Rhizome creeping or ascending, covered with old petiole bases, ca. 5 cm in diameter, apex hidden. Petioles 60-80 cm long, blackish brown, muricate to tuberculate, scurf absent. Petiole scales to 12 mm long and 5.75 mm wide, long-ovate, pointed, weakly contorted, discordantly bicolorous, brown with yellowish brown margins or yellowish brown with white margins, the center with darker areas (Fig. 2D). Rachis smooth, with dark brown hair adaxially, no hairs abaxially, just a few brown scales. Lamina 50-90 cm long and 70 cm wide, bipinnate-pinnatisect, coriaceous; apex gradually reduced (Fig. 2B). Segment margins deeply crenate to entire, revolute (Fig. 2A). Pinnae sessile to stalked, alternate; pinnules to 16 mm wide and 71 mm long, sessile or very short stalked, deeply pinnatisect, basally truncate, apically obtuse to acute (Fig. 2A). Costa smooth, with many white to light brown hairs adaxially, fewer hairs and some brown scales abaxially. Costules smooth, with many contorted white hairs, additional scales abaxially. Veins on both surfaces with long white hairs, abaxially more so than adaxially (Fig. 2C). Flattish brown scales as well as white bullate ones abaxially. Fertile veins mostly forked (Fig. 2C). Sori inframedial to medial, frequently situated above the furcation; indusium absent. Paraphyses longer than the sporangia, translucent white, contorted over the sorus, easily abraded (Fig. 2C). Spores not examined. Cyathea zongoensis is named after the type locality, the Zongo valley near La Paz. This collection was originally determined as Trichipteris frigida (Karst.) Tryon (= Cyathea frigida (Karst.) Domin) by Barrington. However, C. frigida has short paraphyses and fringed scales on the abaxial costules that cover the sori (Karsten 1860, Barrington 1978). In contrast, C. zongoensis lacks such scales and the sori are hidden under a veil of paraphyses (Fig 2C). These characters match C. villosa Willd. which also occurs in Bolivia but has uniformly reddishbrown, heavily contorted petiole scales {vs. bicolorous, weakly contorted ones (Fig. 2D) in C zongoensis) and grows in different habitats, namely open sunny woods and pastures at 700-1800 m elevation. As recently discovered, young fertile plants of Cyathea brevistipes R.C. Moran show the same habit as C. zongoensis (Fig. 2B), this species can be distinguished by its sphaeropteroid indusium, and more scales and less hair on the lamina than in C. zongoensis. Known only from one collection from the Rio Zongo valley in the Prov. Murillo, Dept. La Paz, in humid timberline scrub at 3000 m. The Zongo valley has been visited by numerous botanists, so the lack of collections suggest that this species is genuinely rare. The probability that it is a hybrid should not be excluded; but it seems unlikely as no potential parents grow nearby. The similar, exindusiate Cyathea frigida (Karst.) Domin has not . \L: VOLUME 93 NUMBER 4 FIG. 2. Cyathea zongoensis M. Lehnert. A. Middle pinnule of upper pinna, adaxiallv, J.C. Solomon 8429 (UC). B. Habit, J.C. Solomon 8429 (UC). C. Sori; upper sorus with sporangia removed to show the long paraphyses; upper fertile vein unforked, lower one forked, J.C. Solomon 8429 (UC). D. Petiole scale, J.C. Solomon 8429 (UC). LEHNERT: SIX NEW SPECIES OF TREE FERNS 175 been reported from Bolivia or the adjacent Peruvian Depts. Puno and Cuzco so far. The equally similar, indusiate C. brevistipes R. C. Moran grows at 3000 m near Cotapata some 30 km from the Zongo valley, but hybrids between indusiate and exindusiate Cyathea species normally have some remnants of an indusium (Tryon 1976). The highest reaching exindusiate Cyathea species in Bolivia, C. conjugata (Hook.) Domin, grows in the Zongo valley only below 2500 m. A hybrid between this and any other Cyathea species must be suspected to be a much stouter plant than C. zongoensis. Cyathea xenoxyla M. Lehnert, sp. nov. TYPE.—Bolivia, Dept. Cochabamba, Prov. Chapare, Entre Villa Tunari y Cochabamba, cerca del puente "Rio Carmen May", 17°10'S 65°44'W, 1950 m, 01 Septiembre 2000, M. Lehnert 049 (holotype: GOET; isotypes: LPB, UC). Fig. 3. Cyathea trunco nudo, non duro, spinulis deficientibus, cicatricibusque foliorum notato. Petioli laevigati spisse indumento isabellino e squamellis minimis luteis brunneisque mixtis obtecti; lamina bipinnato-pinnatifida, glabra vel in pagina inferiore leviter squamulis luteolis castaneisque provisa; indusium cyathiforme margine fragili. Trunk to 3 m high, 5-10 cm in diameter; soft, inclined or ascending, smooth, lacking spines or old petiole bases, the apex not hidden between petiole bases of green fronds (Fig. 3A). Indument of small squamellae mainly near the apex; when wet easily abraded and giving the trunk a slimy feel. Frond scars oval, conspicuous. Adventitious buds occur regularly (Fig. 3A). Petiole to 110 cm long, without scales when fully grown, with some long corticinate spines (Fig. 3A) and an indument like that of the trunk. Petiole bases with some large pneumathodes of reddish color. Young croziers initially appearing dark castaneous, due to the black scales with light brown margins, later appearing light brown as a result of the visibility of scurf between the scales as the croziers expand (Fig. 3A). Rachis at least basally muricate, slightly hairy adaxially, glabrous abaxially. Lamina 70-120 cm wide and to 120 cm long, mostly bipinnate-pinnatifid to tripinnate, apical section gradually reduced. Pinnae short-stalked, alternate; pinnules sessile, to 22 mm wide and 105 mm long, segment margins crenate to serrate, basally also double serrate (Fig. 3B, C). Costae/costules with swollen junctures (Fig. 3C), hairy adaxially, hairs light brown to brown, glabrous abaxially, scales pale brown, few and scattered. Veins slightly hairy on both sides of the lamina; hairs brown adaxially, light brown to white abaxially. Leaf axes and veins bearing scales abaxially, white bullate ones as well as light brown flattish ones, sometimes with weakly fringed margins (Fig. 3D). Rachis and leaf axes orange-brown to stramineous. Fertile veins forked. Sori subcostal (Fig. 3B); indusium discoid to subsphaeropteroid, easily abraded and then appearing hemitelioid (Fig. 3B). Paraphyses shorter than the sporangia. Spores without perispore, exospore smooth, finely porate. PARATYPES.—Colombia. Prov. Putumayo. Cerro Portachuelo, camino Sibunday a Pepino, 2300 m, 27 Agosto 1965, D.D. Soejarto 1567 (USM). Ecuador. Prov. Pichincha. Canton Quito, Rio Gualajito Reserve, 10 km W of Chiriboga, km 59 I •y ~~v" LEHNERT: SIX t IE FERNS 177 of old road Quito - Santo Domingo, 00°14'S 78°48'W, 1900 m, 08 July 1991, A. 8r L. Fay 3278 (QCA); Canton Quito, Rio Gualajito Reserve, 10 km W of Chiriboga, km 59 of old road Quito - Santo Domingo, 00°14'S 78°48'W, 1900 m, 10 July 1991, A. & L. Fay 3356 (QCNE); Estacion Cientifica Rio Gualajito, in quebrada Las Palmeras, along road Chiriboga - El Triunfo, 00°14'S 78°47'W, 1800-1900 m, 10 June 1990, B. OUgaard 98019 (AAU, QCNE). Prov. Pastaza. Road N of Mangayacu, km 1.8 (W of Mera), 01°26'S 78°07'W, 1400 m, 13 November 1994, B. Qllgaard & H. Navarrete, 105650 (AAU, QCNE). Prov. Zamora - Chinchipe. New road Loja - Zamora, 13 km E of the pass, 04°00'S 79°02'W, 2000 m, 14 February 1991, B.C. Moran & C. Bohrbach 5384 (MO, QCNE); Ca. 4 km E of Paquisaca, 03°55'S 78°35'W, 1250 m, 06 February 1989, B. OUgaard, J.E. Madsen, L. EUemann 8r B.J. Eriksen 90438 (AAU, QCNE); Road Loja - Zamora, ca. 13 km E of the pass, just before junction with old road, 03°58'S 79°05'W, 2030 m, 08 March 1989, B. QUgaard, J.E. Madsen 8r L. EUemann 90890 (AAU, QCNE). Peru. Dept. Amazonas. Prov. Condorcanqui. Cordillera del Condor, Puerto de la Vigilancia Alfonso Ugarte (PV 3), cabeceras del Rio Comainas, tributario al oeste del Rio Cenepa; 03°55.0'S 78°25.4'W, 1000-1300 m, 20 Julio 1994, H. Beltran & B. Foster 1083 (USM). Prov. Chachapoyas, Road Chachapoyas - Mendoza, 52 km from Chachapoyas, ca. 10 km behind Molinopampa, 06° 14,2'S 77°35,9'W, 2400 m, 04 August 2002, M. Lehnert 229 (GOET, UC, USM). Dept. San Martin. Prov. Rioja. Road Moyobamba - Pedro Riuz, km 395, trail into forest, 03 August 2002, M. Lehnert 216 (GOET, UC, USM). Dept. Ucayali, Prov. Coronel Portell. Dobson (?), 14 Agosto 1946, B. Ferreyra s.n. (USM). Dept. Pasco, Prov. Oxapampa, Trail to summit of Cordillera Yanachaga via Rio San Daniel, 10°23'S 75°27'W, 2500 m, 17 July 1984, D.N. Smith, H. & A. Boetger 7846 (USM). Dept. Cuzco. Prov. La Convencion. Distrito Echarate, Llactahuaman, N del Rio Apurimac, NE del Pueblo Libre, S de la Cordillera de Vilcabamba, 12°51'55.5"S 73°30'40"W, 1650 m, 14 Julio 1998, /. Baldeon, W. Nanray &- B. De la Boca 3077 (USM). Bolivia. Dept. Cochabamba. Prov. Carrasco. A 3 km aproxidamente desde del campamento Locotal, en direccion NO, a lo largo de la antigua senda de Kara Huasi a Pojo, 17°46'12"S 64°45'62"W, 2200 m, /. Jimenez 340 (GOET, LPB, UC). Prov. Chapare. 115 km antigua carretera Cochabamba - Villa Tunari, 17°08'S 65°38'W, 2350 m, Bosque siempreverde, cerrado, virgen, 05 Julio 1996, M. Kessler 7007 (GOET, LPB, UC); 130 km antigua carretera Cochabamba - Villa Tunari, 17°07'S 65°36'W, 2000 m, Bosque siempreverde, cerrado, virgen, 13 Julio 1996, M. Kessler 7220 (GOET, LPB, UC). This species is named for its peculiar soft trunk (Greek ^svoq = strange, TO QXov = wood). The spiny petiole with its dense scurf and lack of scales (Fig. 3A) makes it possible to identify this species even when lamina samples are poor Cyathea xenoxyla M. Lehnert. A. Trunk apex with croziers and adventitious buc 1 (GOET). C. 1 r 7220 (UC). 178 AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 4 (2003) {R. Ferreyra s.n., USM). This character combination is not present in species with similar lamina dissection like C. amazonica R.C. Moran and C. multiflora J.Sm., or in C. pallescens (Sod.) Domin which has similar scales on the abaxial lamina surface. The naked fleshy trunk (Fig. 3A) is the best field characteristic of C. xenoxyla. This distinctive species is perhaps related to Cyathea mucilagina R.C. Moran from Costa Rica and Peru, which also grows in very moist woods and seems to have a similar habit, but that species lacks indusia, has persistent petiole scales, and has abaxially winged costules. This is a locally common tree fern in the undergrowth of mature humid montane forests at 1000-2500 m elevation; it evidently prefers moist to swampy soils. Among the new Cyathea species described here, this is the only one with a fairly wide range, reaching from central Bolivia to southern Colombia and possibly farther north. Cyathea arnecornelii M. Lehnert, sp. nov. TYPE.—Bolivia. Dept. La Paz, Prov. Nor Yungas; Chuspipata a Yolosa, km 7; 16°17'S 67°48'W, 2700 m, Bosque secondario, 01 Agosto 2000, M. Lehnert 003 (holotype: GOET; isotypes: LPB, UC). Fig. 4. Cyathea trunco duro, cicatricibus foliorum notato, apice dense squamulis brunneis cinereo-marginatis obtecto. Juncturae costarum cum rhachidi costulisque aerophoros ferentes. Differt a C. caracasana (Klotzsch) Domin indusio hemitelioideo, a C. multiflora J.Sm. indusio squamulis in receptaculo insertis obtecto. Trunk to 3-4 m high, 7-12 cm in diameter; smooth, no spines or old petiole bases, apex not hidden between the petiole bases of the green fronds, stem scales mostly deciduous, present only at the apex; apex often broader than the trunk due to densely arranged croziers (Fig. 4A), these covered with deciduous scales; outward on the young croziers scales dark brown to almost black with broad grey margin, inwards more and more the color of the margins prevailing, ending in uniformly gray scales (Fig. 4B). Petiole 50 cm long, smooth to slightly verrucate, without indument (Fig. 4A, B) except for some persistent crozier scales. Rachis smooth, hairy adaxially, glabrous abaxially. Lamina 150 cm long and 50-80 cm wide, coriaceous, bipinnate-pinnatifid to tripinnate, the apical section gradually reduced. Pinnae stalked, alternate; pinnules sessile, to 21 mm wide and 65 mm long. Segments obtuse to rounded, margins crenate to serrate (Fig. 4C). Costae/costules smooth, normally with prominent aerophores at their bases (Fig. 4C); moderately to densely covered with light brown hairs adaxially; with many trichomidia and squamules abaxially, but only costules with few trichomes. Veins glabrous or with occasional white hairs on both sides, squamules white to light brown, flattish and bullate; no hairs between the veins. Fertile veins forked. Sori subcostal; indusium hemitelioid, small and ascending, normally covered by scales inserted at the receptacle (Fig. 4C). Paraphyses shorter than the sporangia. Spores not examined. PARATYPES.—Bolivia. Dept. La Paz. Prov. Nor Yungas. 2 km de Chuspipata hacia Coroico, 16°22'S 67°49'W, 2900 m, Bosque secundario; 14 Septiembre LEHNERT: SIX NEW SPECIES OF TREE FERNS i M. Lehnert. A. Trunk, photo M. Lehnert 0 3 (GOET). C. eptacle; hemitelioid indusium; arrow indicating s 180 AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 4 (2003) 1997, M. Kessler 11905 (GOET, LPB, UC); 5 km de Chuspipata hacia Coroico, 16°23'S 67°48'W, 2800 m, Bosque secundario; 19 Septiembre 1997, M. Kessler 12092 (GOET, LPB, UC). I name this species in memory of the biology student Arne Cornelius from Gottingen University, Germany, who died under tragic circumstances while conducting field work in Borneo. The smooth, scale-free petiole (Fig. 4A, B) as well as the aerophores (Fig. 4C) are the most significant features of sterile material of this species; similar species like C. caracasana (Klotzsch) Domin or C. delgadii Sternb., which sometimes have a swollen juncture of costae and costules, have different scale coloration and rarely truly inermous petioles. In the field, the broad trunk apex is most remarkable. Fertile material of C. arnecornelii is easily recognized by the combination of hemitelioid indusium and additional scales on the receptacle (Fig. 4C). Known from humid montane forest, disturbed secondary forests, and even along roads, at 2700-2900 m elevation, only near Chuspipata (Dept. La Paz, Prov. Nor Yungas, Bolivia). Cyathea carolihenrici M. Lehnert, sp. nov. TYPE.—Bolivia. Dept. La Paz, Prov. Nor Yungas, Cotapata Santa Barbara, 16°18'S 67°52'W, 3150 m, Bosque nublado, 06 Agosto 2000, M. Lehnert 011 (holotype: GOET; isotypes: LPB, UC). Fig. 5. Cyathea indusio globoso, trunco squamulis fusco-brunneis obtecto; lamina bipinnato-pinnatifida usque tripinnata, glabra vel inferiore squamulis castaneis minutis fimbriatisque vestita. Differt a Cyathea caracasana (Klotzsch) Domin colore squamularum, a C. pallescenti (Sod.) Domin lamina in pagina superiori glabra. Trunk to 7 m tall and 10-15 cm in diameter; small plants (160 cm tall, M. Lehnert 011) with persistent old spiny petiole bases, apex hidden among petiole bases of the green fronds; trunk of larger plants unknown. Petiole 100 cm long, verrucate to aculeate; petiole scales discordantly bicolorous, variable, either golden brown with dark central stripe or brown with broad white margin (Fig. 5C). Petiole scurf consisting of brown trichomidia and squamellae. Rachis smooth or with scattered small corticinate spines, hairs absent, scurf of minute fringed brown squamellae. Lamina ca. 150 cm long and 130-140 cm wide, bipinnate-pinnatifid to tripinnate (Fig. 5A), apical section gradually contracted, coriaceous. Pinnae long stalked; pinnules short to long stalked, to 40 mm wide and 130 mm long (Fig. 5A). Segments rounded, margins revolute, slightly crenulate to entire (Fig. 5B). Costae smooth to muricate, costules smooth; costae and costules densely covered with brown hairs adaxially, with fewer or no hairs and additional scurf of minute brown squamellae abaxially (Fig. 5A, D). Leaf axes dark brown, in strong contrast to the lamina. Veins sparsely covered with brown hairs adaxially, few hairs and many squamellae like those on the costulae abaxially (Fig. 5D). Fertile veins forked (Fig. 5A, D). Sori subcostal; indusium subsphaeropteroid, dark brown and persistent, LEHNERT: SIX NEW SPECIES OF TREE FERNS 5. Cyathea carolihenrici M. Lehnert. A. Fertile pinnule abaxiallv, M. Lehnert 011 (GOET). B. I (GOET). C Petiole scale, M. Lehnert Oil (GOET). D. Sori; typical quamules on costule and mid vein of segment; arrow indicating squamule arising from indusium, A. Lehnert Oil (GOET). FIG. 182 AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 4 (2003) without umbo, sometimes with small scale arising from the indusium (Fig. 5D). Paraphyses as long as the sporangia or a bit longer. Spores with verrucate exospore and baculate perispore. PARATYPE.—Bolivia. Dept. La Paz. Prov. Nor Yungas. Trocha al Valle de Coscapa, Parque Nacional de Cotapata, 16°12'S 67°53'W, 3000 m, Bosque siempreverde, virgen de 15 m de altura, 12 Septiembre 1997, M. Kessler 11875 (GOET, LPB, UC). I name this species after my grandfather Karl-Heinz Hass. Cyathea carolihenrici is a typical elfin forest tree fern; the coriaceous lamina and the sphaeropteroid indusium (Fig. 5D) are features shared by many other species in this habitat. It is most similar to C. pallescens (Sod.) Domin from which it differs in its glabrous upper lamina surface (Fig. 5B); both share a firm indusium and discordantly bicolorous petiole scales (Fig. 5C). Other similar species with a persistent indusium have a stronger contrast in the petiole scale colors than C. carolihenrici: C. boliviano R.M. Tryon has a white, broad petiole scale margin with occasional dark cell groups, C. straminea Karst. has almost entirely white scales. Cyathea caracasana (Klotzsch) Domin has an ephemeral indusium and concordantly bicolorous petiole scales. Cyathea carolihenrici is unique in its indument of minute, castaneous scales on the abaxial leaf axes (Fig. 5D). Cyathea carolihenrici grows at 3000-3150 m in very humid elfin forests near Unduavi and the nearby Cotapata National Park in the Prov. Nor Yungas, Dept. La Paz. Many thanks to: Michael Kessler and Alan R. Smith for their M. Schwerdtfeger for the Greek translations; G. Wagenitz and I Latin diagnoses; A. N. Schmidt-Lebuhn for editing the figures; I of godfathers for the new species, and following herbaria for n me: AAU, HUT, LPB, MO, QCA, QCNE, UC, US, USM (acron LITERATURE CITED D. S. 1978. A Revision of Trichipteris (Cyatheaceae). Contr. Gray Herb. 208:3-93. D. S. 1983. A Revision of the Genus Alsophila (Cyatheaceae) in the Americas. J. An Arbor. 64:333-382. GASTONY, G. J. 1973. A Revision of the Fern Genus Nephelea. Contr. Gray Herb. 203:83-148. HOLMGREN, P. K., N. H. HOLMCREN and L. C. BARNETT. 1990. Index Herbariorum. Part 1: The Herh of the World. New York Botanical Garden, Bronx, NY. KARSTEN, H. 1860. Flora Columbiae, Band 1. F. Diimmler, Berlin. LJ:H\I:KT. M. 2002 Revision der Baumfarne Boliviens (Familien Cyatheaceae und Dicksoniac* Diplomarbeit (Master's Thesis), Albrecht-von-Haller-Institut fur Pflanzenwissenschal Universitat Gottingen, Germany. LELLINGER, D. B. 1987. The Disposition of Trichopteris (Cyatheaceae). Amer. Fern J. 77:90-94. MORAN, R. C. 1991. Eight New Species of Tree Ferns (Cyathea, Cyatheaceae) from the Amer BARRINCTON, CONANT, LEHNERT: SIX NEW SPECIES OF TREE FERNS 183 R. C. 1995. Five new species and two new combinations of ferns (Polypodiopsida) from Ecuador. Nordic J. Bot. 15:49-58. SMITH, A. R. 1995. Cyatheaceae. In J. E. Steyermark, B.K. Hoist & P.E. Berry (eds.). The Flora of the Venezuelan Guayana, Vol. II. Missouri Botanical Garden Press, St. Louis, MO. STOLZE, R. G. 1976. Fern and fern allies of Guatemala. Part 1. Ophioglossaceae through Cyatheaceae. Fieldiana Botany 6:1-522. STOLZE, R. G. 1984. Two New Tree Ferns from Panama. Amer. Fern J. 74:101-104. TRYON, R. M. 1971. The American Tree Ferns Allied to Sphaeropteris horrida. Rhodora 73:1-19. TRYON, R. M. 1972. Taxonomic Ferns Notes, VI. New Species of American Cyatheaceae. Rhodora MORAN, R. M. 1976. A Revision of the Genus Cyathea. Contr. Gray Herb. 206:19-98. R. M. and R. G. STOLZE. 1989. Pteridophyta of Peru. I. Fieldiana Bot., N.S. 20:110-139. P. G. 1977. Synopsis of the Genus Sphaeropteris with a revision of the Neotropical exindusiate species. Bot. Tahrb. Syst. 92:176-198. WINDISCH, P. G. 1978. The Systematics of the group of Sphaeropteris hirsuta (Cyatheaeceae). In The Botany of the Guayana Highland. Memoirs of the New York Botanical Garden 29:2-22. TRYON, TRYON, WINDISCH, Isoetes tennesseensis (Isoetaceae), an Octoploid Quillwort from Tennessee NEIL T. LUEBKE Department of Botany, Milwaukee Public Museum, Milwaukee, WI 53233 JESSICA M. BUDKE Department of Botany, Miami University, Oxford, OH 45056 Hiwassee River in Tennessee. Past collections of this species have crospora [=1. lacustris). Isoetes tennesseensis differs from /. lacust ispore and microspore morphology and distribution. Speculation o In July 1978, Eugene Wofford and Michael Dennis collected quillworts from the Little Tennessee River at Jones Ferry, Tomatlo Ford, the southwestern end of Davis Island, and from the Hiwassee River approximately 1.1 miles southeast of the bridge on highway 411 in Tennessee. These collections, as well as subsequent ones from the Hiwassee River, have been identified as Isoetes macrospora Dur. (Dennis et al., 1979; Boom, 1979; Taylor et al, 1993). The population of /. lacustris L. (= /. macrospora) in eastern Tennessee is roughly 450 miles from the nearest known outlying population at Passage Creek in northern Virginia (Svenson and Griscom, 1935). Both of these populations are disjunct from the more northern /. lacustris (Taylor et al., 1993). Dennis et al. (1979) hypothesized that these outlying populations of /. macrospora could be the result of either long-range dispersal by waterfowl from northern populations or relics of a previously wider distribution. Except for the difference in geography, J. macrospora and /. lacustris are indistinguishable from each other. Chromosome number, as well as leaf and spore morphology is the same. Therefore, Isoetes macrospora has recently been placed in synonymy with the European /. lacustris (Taylor et al., 1993). In North America, /. lacustris ranges from Greenland and Newfoundland west to Saskatchewan. It typically occurs in cool, oligotrophic lakes, ponds, and streams. Isoetes lacustris is distinguished by its dark green, rigid leaves and large megaspores that range from 550 to 750 urn in diameter (Taylor et al., 1993). Megaspores typically have a cristate to reticulate ornamentation (Fig. 1 A-C) and a densely papillate girdle below the equatorial ridge (Fig. 1 C). Kott and Britton (1980), Taylor and Luebke (1988), and Britton and Goltz (1991) have reported chromosome counts of In = 110 for J. lacustris (Fig. 2 A). Recent studies of plants from the Tennessee populations have shown that past identifications of these plants as J. lacustris are incorrect. These populations represent an undescribed species. In this paper we present our evidence from morphological and cytological studies and describe and name this new taxon. LUEBKE & Bl Dl ^SEENSIS 185 MATERIALS AND METHODS Mature megaspores and microspores were taken from live plants and herbarium specimens. Photomicrographs of spores were obtained with a Hitachi S-570 scanning electron microscope. Measurements of megaspore diameters and microspore lengths were made using Olympus SZX12 and Nikon Microphot-FX microscopes outfitted with ocular micrometers. A minimum of 20 megaspores and 20 microspores were measured from fertile specimens. Megaspores were measured dry while microspores were placed in a drop of water on a slide and covered with a coverslip before being measured. Procedures used for obtaining chromosome counts follow Jong (1997) with some modifications. Plants of /. tennesseensis were floated in deionized water in a growth chamber under a cycle of 12 hours of light, 12 hours of darkness and a constant 18 °C until new roots had formed. New roots approximately 6 mm long were harvested in late morning and pretreated in a saturated solution of paradichlorobenzene (PDB) in the dark at room temperature for four hours. Roots were then fixed in Farmer's Solution (3:1 96% ethyl alcohol: glacial acetic acid), left at room temperature for one hour and then stored in the freezer. For staining, roots were hydrolyzed in IN HCL for ten minutes at 60°C, soaked in three different changes of 96% ethyl alcohol for fifteen minutes each, blotted, placed in Whitman's hematoxylin stain for approximately one hour, and then destained in glacial acetic acid for five minutes. Megaspores of /. tennesseensis have bold, broad tri-radiate and equatorial ridges and an obscure to slightly papillate girdle below the equatorial ridge. Ornamentation on the proximal half may be sparse to dense and varies from cristate to rugate (Fig. 1 E-G). The muri on the distal face are bold with even crests, forming a broken to somewhat regular pattern with areolae of various shapes. Megaspore size ranges from 616-946 um in diameter with a mean of 753 um in diameter (N = 40; SD = 71.76). These megaspores differ from those of /. lacustris in both size and ornamentation. Megaspores of J. lacustris are slightly smaller ranging in size from 550-750 um in diameter (Taylor et al., 1993). Kott and Britton (1983) report a mean diameter of 640 um. In addition, /. lacustris megaspores have a narrow tri-radiate and equatorial ridge, with a densely papillate to occasionally smooth girdle. The ornamentation pattern on the proximal hemisphere is broken, short cristate whereas on the distal face it varies from cristate to nearly reticulate with narrow muri and uneven crests (Fig. 1 A-C). Microspores of /. tennesseensis have a laevigate surface and range in size from 33-40 um long with a mean length of 36 um (N = 40; SD = 2.10) (Fig. 1 H). The microspores in /. lacustris (Fig. 1 D) are larger in size ranging from 37-50 um long with a mean of 43 um long (N = 20; SD = 3.25) and have papillose ornamentation (Kott and Britton, 1983; Taylor et al., 1993). Chromosome counts from the squashed root tips of eight plants showed that AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 4 LUEBKE & BUDKE: ISOETES TENNESSEENSIS I. tennesseensis is an octoploid, 2n = 88 characteristic of /. lacustris. This is the fin reported for North America. Based on our examination of recent and past collections from the Little Tennessee and Hiwassee Rivers we describe the following new species: Isoetes tennesseensis N. T. Luebke & J. M. Budke, sp. nov. TYPE.—U.S.A. Tennessee: Polk Co., Hiwassee River, ca. 1 mile downstream of the crossing of Tellico-Reliance Road, 15 July 2001, /. Budke, K. Heafner, E. Lickey and K. Gustafson 17 (holotype: MIL; isotype: MU). Figs. 1 E-H, 2 B, C. Planta aquatica. Caudex bilobatus. Folia 15-35, atro-olivacea, usque ad 11 cm alta, rigida; subula recta usque recurvata apicem versus; alae basim versus, pallida brunneae. Ligula anguste elongata usque triangulata. Labium spathulatum. Velum tegens sporangium <20%. Sporangium basale, ovale, cum maculis brunneis. Megasporae albidae, 616-946 urn diametro, cristatoreticulatae, cum cristis triradiatis et crista aequatoria lata. Microsporae pallide canae in massa, 33-40 urn longae, laevigatae. Chromosomatum numerus 2n = 88. Plant aquatic. Rootstock 2-lobed. Leaves 15-35, dark olive-green, up to 11 cm tall, rigid. Subula straight to recurved toward tip, terete in cross-section, ca. 1.5 mm wide at mid length (Fig. 2 C). Alae on either side of the base of the microphyll up to 4.5 cm tall, pale brown. Ligule narrowly elongate to triangular. Labium spathulate. Velum covering < 20% of sporangium. Sporangium basal, oval, 4-6 mm long and 1-1.5 mm wide, lightly brownstreaked. Megaspore white, 616-946 um in diameter, x = 753 urn; cristate to rugate proximally, the ornamentation sparse to dense; cristate to reticulate distally, with tall, thick-walled muri of even height; proximal surface with bold tri-radiate ridges; equatorial ridge with an obscure to slightly papillate girdle below. Microspores light gray in mass, 33-40 um long, x = 36 um; laevigate. Chromosome number In — 88. PARATYPES— U.S.A. Tennessee: Monroe Co., Little Tennessee River: Jones Ferry, B.E. Wofford et al. 78-133 (TENN); Tomatlo Ford, B.E. Wofford and W. Dennis, 78-134 (TENN); southwest end of Davis Island, B.E. Wofford and W. M. Dennis, 78-135 (TENN); gravel bars several miles upstream from hwy 411 bridge, B. Boom 318 (TENN); upstream side of Davis Island near Mile 15, W.M. FlG. 1. SEM photomicrographs of Isoetes lacustris an Taylor 4902 (MIL): A: proximal view of megaspore; B. i megaspore. D. Taylor 5010 (MIL): Microspore. E-H. holotype): E. proximal view of megaspore; F. dista i. microspore. ! (MIL). B. Somati 3 (MIL). C. Plants ice, Ten) Dennis et al. (TENN). Polk Co., Hiwassee River: shallow shoals at intersection of Hwy. 30 and State Road 2518, B.E. Wofford and A.M. Evans 78-168 (TENN); along Hwy 30 ca. 0.6 mi NW of bridge at Reliance, W.C. Taylor 5189 (MIL); at Tenn 315 and 30, Reliance and scattered 1.6 mi downstream, K.D. Heafner et al. 00-042 (MIL, MU); ca. 0.25 mi upstream of crossing of Tellico-Reliance Road, /. Budke et al. 8 (MIL, MU). Distribution.—At present, Isoetes tennesseensis is known from southeastern Tennessee, in the Little Tennessee River in Monroe County and the Hiwassee River in Polk County. Specimens oil. tennesseensis have not been found in the Little Tennessee River since the construction of a dam and the permanent raising of the water level. However, it is likely that plants could persist in some areas of this river where conditions are suitable. Isoetes tenneseensis grows in the cool waters of the Hiwassee River (Fig. 2 D). An upstream dam results in water levels rising and falling on a regular basis. On average the water is two meters deep but can vary across the river. Plants of /. tennesseensis are constantly submerged and appear to be obligate LUEBKE & BUDKE: ISOETES TENNESSEENSIS 189 aquatic as evidenced by their lack of stomata. River substrate varies, including cobble, sand, and crevice-ridden shale. Plants were found growing wedged in the sand-filled crevices of the shale or partially buried in sandy cobble. To date, Isoetes tennesseensis has only been found in a few locations along the Hiwassee River. Searches for the plant farther upstream from the known locations and in other river systems have not revealed other populations. It is unknown whether /. tennesseensis still occurs in the Little Tennessee. Further field investigation is necessary. Until more is known about the species' distribution, it is suggested that the known populations be afforded protection. It does not appear that the population at Passage Creek, Virginia is this new species. Rebecca Bray has counted the chromosomes from these plants and reports that they are 2n = 110 (Personal Communication). Examination of specimens from this population also reveals that they differ from J. tennesseensis in leaf and spore morphology, but are similar to /. lacustris. Megaspores range in size from 580-705 (im and fall within the range for I. lacustris. Kott and Britton (1983) found that spore size can be correlated with ploidy level in Isoetes. This does not seem to hold with this species since megaspore size of the octoploid, /. tennesseensis (x = 753 (im) is larger than that for the decaploid, J. lacustris (x = 640 \im). However, this correlation between ploidy level and spore size is reflected in the microspore size where those of I. tennesseensis are smaller (x = 36 urn) in comparison to those of/, lacustris (x = 43 |im). Isoetes tennesseensis is the only octoploid quillwort reported for North America and only the third worldwide. The others are /. pseudojaponica M. Takamiya, Mitsu. Watan. & K. Ono which occurs in Japan (Takamiya, 1999; Troia, 2001) and I. andina Hook, from South America (Taylor et al., 2002). Preliminary studies of comparisons of nuclear ribosomal ITS nucleotide sequences suggest a possible origin of /. tennesseensis. The comparison indicates that /. tennesseensis is similar to J. engelmannii A. Braun and J. valida (Engelm.) Clute and shares several ITS nucleotide sites and indels with each. Isoetes engelmannii and I. valida, both diploids (2/7 = 22), and their allotetraploid {2n = 44), /. appalachiana D. F. Brunton & D. M. Britton (Napier et al., 2002) are sympatric within the area of /. tennesseensis. Further comparison of six cloned ITS genomic sequences showed all six were similar to I. engelmannii. From these preliminary studies a pedigree is proposed for /. tennesseensis that suggests it is the result of the backcrossing of /. engelmannii with /. appalachiana to form a sterile triploid (2n = 33) which doubled its chromosomes to form a fertile hexaploid (2n = 66). The result of I engelmannii backcrossing with this hexaploid would produce a sterile tetraploid that with the doubling of its chromosomes would produce a fertile octoploid. Further molecular investigations of /. tennesseensis may reveal more information about the origin of this new species. AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 4 (2003) ) Pat Cox, Jim Hickey, Dan Brunton, i Peterson and Mary Ann Polasek for their 5 and suggestions which strengthened LITERATURE CITED B. M. 1979. Systematic studies of the genus Isoetes in the southeastern United States. Thesis. University of Tennessee, Knoxville. D. M. and J. P. GOLTZ. 1991. Isoetes prototypus, a new diploid species from eastern Canada. Can. J. Bot. 69:277-281. DENNIS, W. M., A. M. EVANS and B. E. WOFFORD. 1979. Disjunct populations of Isoetes macrospora in southeastern Tennessee. Amer. Fern J. 69:97-99. JONG, K. 1997. Laboratory Manual of Plant Cytological Techniques. Royal Botanic Garden Edinburgh. Korr, L. S. and D. M. BRITTON. 1980. Chromosome numbers for Isoetes in northeastern North America. Can. J. Bot. 58:980-984. KOTT, L. S. and D. M. BRITTON. 1983. Spore morphology and taxonomy of Isoetes in northeastern North America. Can. J. Bot. 61:3140-3163. NAPIER, N. S., S. B. HOOT and W. C. TAYLOR. 2002. Unraveling a tangled web: hybrid and allopolyploid origins of Isoetes. Botany 2002 Abstracts, 142 Abstr. TAKAMIYA, M. 1999. Natural history and taxonomy of Isoetes (Isoetaceae). Acta Phytotax. Geobot. BOOM, BRITTON, SVENSON, H. K. and L. GRISCOM. 1935. Isoetes macrospora in the Shenandoah Valley. Amer. Fern J, W. C. and N. T. LUEBKE. 1988. Isoetes xhickeyi: a naturally occurring hybrid between /. echinospora and /. macrospora. Amer. Fern J. 78:6-13. TAYLOR, W. C, N. T. LUEBKE, D. M. BRITTON, R. J. HICKEY and D. F. BRUNTON. 1993. Isoetaceae, pp. 6475. In: FNA Editorial Committee, eds., Flora of North America, North of Mexico, Volume 2. Oxford University Press, New York. TAYLOR, W. C, N. T. LUEBKE and S. B. HOOT. J Isoetes based on morphology, chroi TAYLOR, TROIA, A. 2001. The genus Isoetes L. (Lycophyta, Isoetaceae): synthesis of karyological data.1 AMERICAN FERN JOURNAL QUARTERLY JOURNAL OF THE AMERICAN FERN SOCIETY Editor R. James Hickey Botany Department, Miami University, Oxford, OH 45056 hickeyrj@muohio.edu Associate Editors Gerald J. Gastony, Department of Biology, Indiana University, Bloomington, IN 47405-6801 Christopher H. Haufler, Department of Ecology and Evolutionary Biology, University of Kansas, Lawrence, KS 66045-2106 James H. Peck, Department of Biology, University of Arkansas-Little Rock, Little Rock, AR 72204 R., and K. R. Wooi> />;, Rediscovered in Hawai'i - AGUARAIUJA, M. S. (see W. D. BARKER, M. S. (see R. J. BARKER, BARKER, I) ( Eaton) Robins. (Asplcniaccac) HICKEY) M. S. and W. D. HAUK. An Evaluation of Scepteridium dissectum (Ophioglossaceae) with ISSK \lnKi-. lmpl^> to-- !•> • s , ., Swematics I. C. L. (see A. B. BARROS, J. M. (see N. T. BUDKE, . HAUK) MARCON) LUEBKE) S., M. SRIVASTAVA and R. SRIVASTAVA. Contribution to the Gametophyte Morphology of the Fern Genus Lomagramma J. Sm. in India CHANDRA, P. G. (see H. W. DAVISON, M., (see S. L. DOOLEY, KELLER) NONDORF) M. A. P., B. PEREZ-GARCIA, R. Pteridoflora of Chiapas, Mexico FARRERA, FERNANDEZ, R., AND R. M. (see A. B. GUERRA, VAIL. RIBA and M. E. LOPEZ-MOLINA. New Records New Records for t Baker in Peru MARCON) A. Lycopodu a i I ycopodiaceae), a New Hybrid Clubmoss from Northeastern North America HAINES, HAUFLER, C. H. (see H. W. HAUK, W. D. (see M. S. HAUK, W. D. and M. S. Ohio HICKEY. KELLER) BARKER) BARM > angustisegmentum in R. J. Review: The Cycads HICKEY, R. J. Review: A Modern Multilingual Glossary for Taxonomic Pteridology HICKEY, R. J., M. S. BARKER and M. Argentina and Vicinity HICKEY, R. J., C. MACLUF and W. C. TAYLOR. A Rt e Franchet in Argentina and Chile HOSHIZAKI, PONCE. An Adiantopsis Hybrid from Northeastern 1 t n ot / t B. J. Review: Index to Distribution Maps of Pteridophytes in Asia, 2nd Edition F. Kaempferol and Quercitin 3-0-(2", 3"-di-0-/j-coumaroyl)-glucosides from Pteris vittata IMPERATO, ITAM, K. (see U. KELLER, YUSUF) H. W., P. G. DAVISON, C. H. HAUFLER and D. B. LESMEISTER. Polypodium appalachianum: An Unusual Tree Canopy Epiphyte in the Great Smoky Mountains res in Natural Soils M. Six New Species of Tree Ferns from the Andes IT, 3ER, D. B. Nomendjtui.il and hixonomtc \oiv> mi the i'kauloplutos of Costa Panama, and Colombia, III B. and A. R. SMITH. M. E. (see M. A. P. LOPEZ-MOLINA, N. T. and J. M. from Tennessee LUEBKE, C. (see R. J. MACLUF, RK New Species and New Combinations of Grammitidaceae FARRERA) Isoetes tennesseensis (Isoetaceae), an Octoploid Quillwort BUDKE. HICKEY) A. B., I. C. L. BARROS and M. GUERRA. A Karyotype Comparison Between Two Closely Related Species ol Acrostichum MARCON, K, P. ROJAS and M. PALACIOS-RIOS. Moth 1 fern Acrostichum danaeifolium as Host for MEHLTRETER, MORAIS, P. O. (see A. SALINO) MORAN, R. C. (see L. PACHECHO) S. L., M. DOOLEY, M. PALMIERI and L. J. SWATZELL. The Effects of pH. Temperature, Light Intensity, Light Quality, and Moisture Levels on Spore Germina- NONDORF, L. and R. C. MORAN. Lectotypification of Several Names Currently Placed in Diplazium (Woodsiaceae) PACHECHO, M. (see K. PALACIOS-RIOS, PALMIERI, M. (see S. L. MEHLTRETER) NONDORF) R. W. The Common Staghorn I ern Plat in Southern Florida PEMBERTON, PEREZ-GARCIA, B. (see M. A. P. M. (see R. J. PONCE, < / htm '> t'nrcatum, Naturalizes FARRERA) , HICKEY) PRADO, J. New Species in Adiantum from Brazil, PRADO, J. Corrections and Additional Information on Ferns from the Semi-Arid Region c Brazil M. Soil Spore Bank of Ferns in a Gallery Forest of the Ecological Station c Panga, Uberlandia, MG, Brazil RANAL, RIBA, R. (see M. A. P. ROJAS, P. (see K. SALGADO, SALINO, SMITH, A. E. FARRERA) MEHLTRETER) \ liaceae), a New Species from Luzon, Philippin A. and P. O. (Tectariaceae) A. R. (see B. MORAIS. LE6N) New Comb teal American Ctenit SRIVASTAVA. M. (see S. CHANDRA) SRIVASTAVA, R. (see S. CHANDRA) TAYLOR, VAIL, W. C. (see R. J. R. (see R. . HICKEY) FERNANDEZ) WHITTIER, D. P. Rapid Gametophyte Maturation in Ophioghs WHITTIER, D. P. The Gametophyte of Diphasiastrwn sitchensi WILSON, K. A.. Rc\ ic« : //i/n'<//',-".\ l-'cms and Fern Allies .. . IL K. ITAM, F. ABDULLAH, I. Gleichenia (Gleicheniaceae) , I. (see U. ZAINAL and A. SUKARI. Leaf l-'knoiumU i YUSUF) Volume 93, Number 1, January-March, pages 1^8, issued 12 May 2003 Volume 93, Number 2, April-June, pages 49-96, issued 30 May 2003 Volume 93, Number 3, July-September, pages 97-168, issued 17 Septembt Volume 93, Number 4, October-December, pages 169-212, issued 10 Nov. i Fern Journal 93(4):1 Asplenium ofeliae (Aspleniaceae), a New Species from Luzon, Philippines The Philippine archipelago lies entirely within the tropics and belongs to the phytogeographic region known as Malesia. The archipelago consists of about 7,107 islands, islets and reefs scattered over 1,295,000 Km2 of the western Pacific Ocean (Tan & Rojo, 1989). The Philippine fern flora is rich and well known, although only one comprehensive fern flora has ever been published (Copeland, 1958-1960). Thirty-one families, 151 genera, and 958 species were reported in the last published checklist (Salgado, 1990). Since that publication, new species, new records for the country, and other record changes have been published (Barcelona et al. 1996; Salgado, 1996; Hovenkamp, 1998; Nooteboom, 1998; Barcelona and Price, 1999). By the time a new fern flora can be prepared, the final number of fern species will probably approach 1,000. The genus Asplenium is represented in the Philippines by at least 43 species (Salgado, 1990). While studying Philippine Asplenium in greater depth since the publication of the checklist, it became obvious that this number of species is too low. Some species are in reality groups of species, and others have been erroneously reduced to synonymy. Several names have been traditionally used in the Philippines and other parts of Asia to designate these species groups (see Tardieu-Blot and Ching, 1936; Holttum, 1955). In revising the Philippine species of Asplenium sect. Hymenasplenium, I found specimens in K, L, PRC and US that had been identified as Asplenium unilaterale Lam., but actually represented a new species. The type of Asplenium unilaterale was collected by P. Commerson in Mauritius. It is a common, widespread species reported from Africa to Polynesia (Christensen, 1943; Copeland, 1960; Burrows, 1990). This variable species is commonly found in humid ground, among rocks, and on ravine embankments. In the Philippines it grows from about 150 to 2500 m. Asplenium unilaterale is recognized by its dorsiventral, creeping rhizome, pinnate frond, oblong lamina, the basiscopic side of the pinnae with a very narrow lamina Y3 to Vi the length of the pinna the margin gradually expanding then tapering toward the apex, the pinna apex acute or narrowly rounded, the acroscopic pinna margin dentate or crenate (Fig. 1. C), and the oblong sori oblique to the costa occupying the base or center of the veins I JOURNAL: VOLUME 93 NUMBER 4 FIG. 1. A. Asplenium ofeliae: Abaxial side of pinnae showing the acroscopic margin with deep and shallow sinuses, and the straight, basiscopic margin with the subapical tooth [Merrill Phil. Pits. 700, US). B. Asplenium ofeliae: Lobe bearing two teeth, each tooth has a veinlet ending in a notch Pits. 700, US). C. Asplenium unilateral: Abaxial side of laminae showing a dentate acroscopic margin and the expanded lamina below the costa [Ramos &• Edano BS37952, US). D. Asplenium unilaterale: Abaxial side of pinna showing the irregularly crenate acroscopic margin [C. A. Wenzel 547, US). T = subapical tooth; D = deep sinus; S = shallow sinus; N = notch. (Fig. 1. D). Iwatsuki (1975) grouped A. unilaterale and its allied species A. excisum C. Presl, A. subnormale Copel., A. filipes Copel. (syn. A. unilaterale var. udum C. B. Clarke), and A. cheilosorum Kunze in section Hymenasplenium, which is characterized by their dorsiventral, long-creeping rhizome, phyllopodia or swollen stipe bases, a characteristic anatomy of the meristeles and a unique chromosome number (x = 39) within Asplenium (Lovis, 1973; Mitui et al., 1989). SALGADO: ASPLENIUM OFEUAE 193 Asplenium ofeliae Salgado, sp. nov.—TYPE: Philippines: Benguet, Luzon, May 1911, Merrill Phil. Pits. 700 (holotype US!; isotype PRC!). Asplenio unilaterali Lam. affine. Stipites atropurpurei hirsuti, pilis stramineis coarctatis; laminae oblongae; pinnae subsessiles dimidiatae, margine basiscopica distaliter dente subapicala munita, margine acroscopica lobata propter sinus profundos et denticulata propter incisuras non profundas, incisuris et sinibus alternantibus, venis gracilibus non prominentibus ad basin Rhizomes short-creeping, ca. 3 mm in diameter, with small phyllopodia, densely covered with stramineous hairs, scales few, black, clathrate, entire. Fronds alternating on the dorsal side of the rhizome, ca. 0.5 cm distant, (11)15-20(22) cm long and (1.9)2.2-3.1(3.4) cm wide, pinnate; stipes (4)59(10) cm long, terete, atropurpureous, polished, profusely hairy near the base, the hairs long, yellow, multiseriate, becoming shorter distally, often forming a mat on the surface of the stipe; laminae longer than the stipes, (9)10-14(16) cm long, oblong, acuminate, thin, truncate at the base; rachises shallowly grooved, marginate to the lower pinnae with a chlorophyllous, narrow wing, glabrescent or hairy, hairs stramineous and multiseriate; pinna pairs 16-25, subopposite to alternating, the basal pinna pair as long as the median pairs, median pinnae (1.0)2-3(3.4) cm long, 0.5-0.7 cm wide, sessile or short stalked with a decurrent narrow wing on the acroscopic side of the stalk, oblong, with a broadly rounded, dentate apex, the acroscopic pinna base at a right angle to the costa or broadly cuneate, the basiscopic margin almost completely excised, less than 1 mm wide for half or more the length of pinna, straight, ending in a horizontal, subapical tooth (Fig. 1. A), acroscopic pinna margin with lobes formed by deep incisions between the secondary veins, ]/3 to Vi to the costa, forming marginal teeth, teeth rounded, with an apical notch, apex pinnatifid with a thin wing along the rachis; veins free, visible, thin, costa straight for about 2A of the length of the pinnae then turning towards the acroscopic margin, acroscopic secondary veins separated by the deep marginal incisions, not forked or dividing only once, each vein or venule produced at the fork extending into a rounded marginal tooth and reaching the apical notch, two or three basiscopic veins present, the first basiscopic vein paralleling the margin and ending in the subapical tooth; sori 3-5 mm long, distal on the pinnae, mostly in an oblique row on the acroscopic side of the costa and close to it, never reaching the base of the teeth, 0-2 sori on the basiscopic side of the costa and usually parallel to the margin; indusia thin, yellowish or brown, entire. : ofeliae in honor of Ofelia Brana-Salgado, my PARATYPES.—PHILIPPINES. LUZON: Benguet: Monte Tonglon(= Mt. Santo Tomas), 2250 m, northern Luzon, Mar. 1897, Loher 1245 (US!); Haight's Place, Jan 22-28, 1909, Topping 1132 (US!); Pauai, Jan 23-28,1909, a second Topping 194 AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 4 (2003) 1132 (US!); Mt. Santo Tomas, Feb 2,3, 1909, Topping 1188 (US!). Ifugao: Mt. Data, Sept. 1921, Ramos & Edano BS40257 (US!, K!); Mt. Data, May 3, 1946, Alcasid PNH 1748 (L!). Asplenium ofeliae has been collected very few times since the end of the nineteenth century. It is distinguished from A. unilaterale and its allied species by hairy stipes and rachises, oblong pinnae with a broadly rounded, toothed apices, by the straight basiscopic pinna margin, ending in a horizontal, subapical tooth (Fig. 1. A), by its incised upper margin with alternating deep incisions separating the secondary veins, and shallow incision separating venules and the rounded teeth with a marginal notch at the tip (Fig. 1. B). Asplenium ofeliae is a terrestrial fern found between 1200-2300 m in the central highlands of northern Luzon, Philippines. These mountains receive heavy rainfall during the monsoon and typhoon season from May to November. There is a period of drought from about December to April. In the Philippines, mountain summits above 1500 m are naturally covered with mossy or cloud forests often shrouded in clouds and mist for several hours every day. Humidity is normally high at these elevations. The herbarium specimens from which the species has been described do not include ecological notes or descriptions of the locations where they were collected. Asplenium ofeliae may be saxicolous like two of its close relatives, A. unilaterale and A. subnormale, but its habitat has not been established with certainty. LITERATURE CITED BARCELONA, ). F., B. F. HERNAEZ and M. G. PRICE. 1996. Philippine Schizaea. Asia Life Sciences 5: J. F., and M. G. PRICE. 1999. New Philippine Goniophlebium (Polypodiaceae: Pteridophyta). Fern Gaz. 15:261-264. BURROWS, J. E. 1990. Southern African Ferns and Fern Allies. Fourways, South Africa: Frandsen BARCELONA, C. 1943. A Revision of the Pteridophyta of Samoa. Bernice P. Bishop Mus. Bull. 177. Honolulu: Bernice P. Bishop Museum. E. B. 1958-1960. Fern Flora of the Philippines, vols. I-III. Natl. Inst. Sci. Tech., Bureau of Printing, Manila. R. E. 1955. Ferns. A Revised Flora of Malaya, vol. II. Ferns of Malaya. Government Printing Office, Singapore. HOVENKAMP, P. H. 1998. Polypodiaceae. Flora Malesiana, Series II, v. 3. IWATSUKI, K. 1975. Taxonomic studies of Pteridophyta. X. Acta Phytotax. Geobot. 27:39-54. Lovis, J. D. 1973. A biosystematic approach to phylogenetic problems and its application to the Aspleniaceae. In A. C. Jermy, J. A. Crabbe and B. A. Thomas, eds., The Phylogeny and Classification of the Ferns. J. Linn. Soc, Bot., Suppl. 1, v. 67, p.p. 211-228. MITUI, K., N. MURAKAMI and K. IWATSUKI. 1989. Chromosomes and systematics of Asplenium sect. Hymenasplenium. (Aspleniaceae). Amer. J. Bot. 76:1689-1697. CHRJSTENSEN, COPELAND, HOLTTUM, SALGADO: ASPLENIUM OFEUAE 195 H. P. 1998. Davalliaceae. Flora Malesiana, Series II, v. 3. A. E. 1990. A checklist of Philippine ferns. Philipp. J. Sci. 119:107-148. SALGADO, A. E. 1996. Taxonomic and Nomenclatural Notes on Philippine Ferns. I. The Identity of Asplenium anisodontum C. Presl (Aspleniaceae). Edinburgh J. Bot. 53:271-274. TAN, B. C. and J. P. Rojo. 1989. The Philippines. In D. G. Campbell and D. Hammond, eds., Floristic Inventory of Tropical Countries. The New York Botanical Garden, New York. TARDIEU-BLOT, M. L. and R. C. CHING. 1936. Revision des especes confondues avec VAsplenium laserpitiifolium Lam., avec description d'especes nouvelles asiatiques de ce groupe. Notul. NOOTEBOOM, SALGADO, Lycopodiella Xgilmanii (Lycopodiaceae), a New Hybrid Bog Clubmoss from Northeastern North America ABSTRACT.—Lycopt [escribed as a new hybrid from It is the result of L. appressa X L. inundata as inferred from morphology provided for Lycopodiella in northeastern North America that includes r Lycopodiella sensu Holub is a distinctive, small genus of wetland clubmosses. It differs from all other genera of lycopods in possession of largely deciduous shoots that overwinter as unions and subpeltate sporophylls with a narrow, elongate, leaf-like apical portion. Lycopodiella is further characterized by hemisaprophytic gametophytes, subglobose sporangia, superficial horizontal shoots that normally produce unbranched upright shoots terminated by a single strobilus, and a base chromosome number of x = 78 (Bruce, 1975; 011gaard, 1987; Wagner and Beitel, 1992). Despite the fact there are only six known species of Lycopodiella in North America (Wagner and Beitel, 1993), the genus is complex. Factors such as cryptic and environmentally influenced morphology, extensive hybridization, and ploidy-level differences contribute to an often bewildering array of morphologies seen in regional collections. Bruce (1975) critically examined Lycopodiella in the southeast and Great Lakes regions of North America. Of great importance is that he documented the existence of diploid and tetraploid taxa. Further, he showed that two types of hybrids existed - those with well formed spores produced by species of similar ploidy level and those with malformed spores produced by species of different ploidy level. Though he also examined northeastern material for his study, only a few paragraphs were devoted to discussion of taxonomic problems in New England and maritime Canada. This paper describes a new hybrid that has caused substantial confusion in the literature and in herbarium collections. Lycopodiella appressa (Chapman) Cranfill is one of the most distinctive species of bog clubmoss in North America. Oddly, it is also one of the more misunderstood taxa. Fernald (1950), for example, interpreted L. inundata (L.) Holub (using the name Lycopodium inundatum L.) as passing freely into L. appressa (using the name Lycopodium inundatum var. bigelovii Tuckerman). This statement is based on failure to recognize hybrid individuals, which obscure the morphological gap between L. appressa and L. inundata. These hybrids, noted from northeastern North America by Bruce (1975) and Gillespie (1962), have largely gone unnoticed in regional collections. Also, failure to geographic cline in certain morphological characters HAINES: LYCOPODIELLA xGILMANII 197 may have contributed to the problem. Northern Lycopodiella specimens are shorter, have thinner shoots, and produce fewer upright shoots compared with southern specimens. Following the arguments of Wagner (1968), a binomial name is here provided for L. appressa X L. inundata in order to call attention to this hybrid and its contribution to the taxonomic difficulties faced by students of the genus. Lycopodiella Xgilmanii A. Haines, hybr. nov.—TYPE. USA: Connecticut., Tolland County: low, open, wet areas in abandoned borrow pit, Koller Wildlife Management Area, growing with Lycopodiella appressa, Scirpus cyperinus, Muhlenbergia uniflora, Alnus incana ssp. rugosa, and Rhynchospora capitellata, at ca. 122 m elevation, Tolland, 23 Oct 2001, Haines and Mehrhoffs.n. (holotype: GH). Figs 1 and 2. Caulis horizontalis 0.9-1.5 mm latus, prostratus, folia 3.8-6 X 0.5-0.8 mm, dentibus marginalibus utrinque 0-3(-4). Caulis erectus 1 vel 2, 8-18.5 cm altus. Strobili 28-75 X (6-)7-12 mm, sporophyllis (4.6-)5-6.4(-7.l) X 0.5-0.75 mm, ascentibus, dentibus marginalibus utrinque 0-2. Hybrid of Lycopodiella appressa and L. inundata. Horizontal stem prostrate, 7-21 cm long, 0.9-1.5 mm in diameter exclusive of the leaves. Leaves of the horizontal stem 3.8-6 mm long, with 0-3(-4) minute teeth per margin, leaves on the distal portion of stem with relatively more teeth. Upright shoot 1 or 2 per horizontal stem segment, 8-18.5 cm tall, the leaves with entire margins or those in the basal portion of shoot minutely toothed. Strobili (22-)24-75 mm long, (6-)7-12 mm wide, representing (20-)28-45 percent of the upright shoot height. Sporophylls with 0-2 slender teeth per margin, ascending (loosely appressed), (4.6-)5-6.4(-7.1) mm long, 0.5-0.75 mm wide. Spores mostly 4853 um, varying from ca. 5-90 percent malformed. PARATYPES.—CANADA. Nova Scotia: Yarmouth County. Peaty and sandy margin of Salmon (Greenville) Lake, 25 Aug 1921, Fernald and Long 23077 (GH); Sandy and cobbly beach of Cedar Lake, 6 Oct 1920, Fernald and Linder 19567 (GH). UNITED STATES. Connecticut. Fairfield County: Large colony in moist mossy area, coastal field, with others, 27 Sep 1940, Eames 12049 (CONN, NEBC). New Haven County: In moist sandy place, Milford, 26 Sep 1908, Blewitt 1183 (NEBC); Wet sandy soil by R.R. E of Towautic Sta., 9 Sep 1917, Harger 6992 (NEBC). Tolland County: Koller Wildlife Area, borrow pit, Tolland, 10 Aug 1991, Mehrhoff 14914 (CONN, NEBC); Wet area in ruts of abandoned road, on hillside mined for gravel, E side of Route 32, 0.6 km southbound from 1-84 overpass, Willington, 23 Oct 2001, Haines s.n. (NEBC). Maine. Cumberland County: Seasonally wet floor of abandoned quarry north of Pleasant Street near Freeport town line, growing with Lycopodiella inundata, Muhlenbergia uniflora, and Rhyncbospora capitellata, Brunswick, 7 Sep 2002, Haines s.n. (MAINE); Quaking bog, Cumberland, 6 Sep 1902, Chamberlain s.n. AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 4 HAINES: LYCOPODIELLA xGILMANII 199 (NEBC); Bog, The Meadow, Cumberland, 27 July 1903, Chamberlain 484 (NEBC); Quaking bog, The Meadow, Cumberland, 12 Sep 1903, Chamberlain s.n. (MAINE); Sphagnum bog, Cumberland, 18 Aug 1900, Chamberlain s.n (BRU); Open, wet areas in a sandy depression, the surface covered by a thin layer of organic soil and/or Sphagnum, at 39 meters elevation, with Muhlenbergia uniflora, Rubus hispidus, Drosera intermedia, and Viola lanceolata, growing in close proximity to L. inundata, Falmouth, 2 Sep 2001, Haines s.n. (MAINE, NEBC). Hancock County: Aunt Betty Pond Road, Bar Harbor, 29 Aug 1908, Band s.n. (MAINE). Massachusetts. Barnstable County: Damp sandy and peaty border of Israel Pond, Barnstable, 31 Jul 1913, Fernald 8381 (GH). Bristol County: Open sandy swamp, North Easton, Easton, 2 Aug 1905, Forbes s.n. (CONN, NASC). Dukes County: McKinley Road bog, Marthas Vineyard, 23 Sep 1913, Bicknell 11592 (NEBC); Cranberry Bog, Chillmark, 21 Sep 1916, Seymour 1015 (GH). Hampden County: Wet sphagnous flat by gravel pit N of Winchell Road, Granville, 28 Jul 1989, Sorrie and Lovejoy 4803 (NEBC). Hampshire County: Sandy, low area on Plain Road, Hatfield, 30 Aug 1976, Ahles 82399 (CONN). Middlesex County: Round Pond, Tewksbury, 9 Sep 1901, Pease 111 (NEBC); Sphagnum bog, border of Round Pound, Tewksbury, 18 Sep 1909, Fernald s.n. (CONN). Norfolk County: Narrow open fen bordering small pond behind Haemetics building, with Lycopodiella appressa, Drosera rotundifolia, Juncus canadensis, and Eleocharis tuberculosa, "peatland morphotype", Braintree, 18 Sep 2001, Haines and Lubin s.n. (GH); Low sand margin of Ponkapog Pond, among sedges, Canton, 1 Aug 1908, Ware 652 (SCHN); Purgatory Swamp, Dedham, Faxon s.n. (GH); Wellesley, 22 Sep 1909, Wight s.n. (SCHN). New Hampshire. Carroll County: Sandy strand of Ossipee Lake, Ossipee, 2 Sep 1936, Weatherby 6874 (NEBC); S shore of Ossipee Lake among the sedge mat, Center Ossipee, Ossipee, 31 Aug 1975, Hellquist 11010 (NASC). Chesire County: Shore of Pond, Jaffrey, 22 Sep 1894, Deane s.n. (SPR). Strafford County: Open floor of abandoned borrow pit, growing with Rhynchospora capitellata, Muhlenbergia uniflora, Viola lanceolata, Schizachyrium scoparium, Alnus incana, Lycopodiella appressa, and L. inundata, ca. 54 m elev., Lee, 10 Oct 2002, Haines, Lubin, and Abair s.n. (GH) New York. Hamilton County: Shore of East Stoner Lake, 18 Aug 1934, Muenscher and Clausen 4113 (GH). New Jersey. Borough County: Closter, Austin s.n. (GH). Rhode Island. Providence County: Wet fields, 27 Aug 1892, Providence, Collins s.n. (GH). Washington County: Damp sands near Grace Point, Block Island, New Shoreham, Fernald, Long, and Torrey 8387 (NEBC). Vermont. Windsor County: View Pond, Woodstock, 31 Aug 1921, Kittredge 3a (NEBC); Edge of View Pond, South Woodstock, Woodstock, 31 Aug 1921, Kittredge B807 (NEBC). Kics. 1-3. Lycopodiella Xgiln lanii and Lycopodiella appressa. 1 . Lycopodiella XgHm mil, specimorphotype with tall strobili (relati ve to total upright sh oot height) ascending sporophylls. 2 . Lycopodiella Xgilmanii sporophylls, note the slender teeth near base. 3. L. appressa sporophylls, note that when teeth are presenl :, they are short and 1 200 AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 4 (2003) The epithet has been chosen to honor Arthur Gilman of Vermont, a careful student of free-sporing tracheophytes. His expertise and tireless responses to inquiries has greatly assisted my studies of lycopods. Lycopodiella Xgilmanii does demonstrate some variation in morphology. Most collections of L. Xgilmanii show relatively tall strobili comprising more than 30% of the total upright shoot height, a character state contributed by L. inundata (Figure 1). This form is found on saturated soils with high sand content, such as abandoned borrow pits and coastal outwash plain pond shores. In contrast, when L. Xgilmanii is found in hydric organic soils with extensive bryophyte cover, such as acid fens and lake-border fens, the strobilus is relatively short. This strobilus reduction in the "peatland morph" is paralleled in northeastern L. inundata and has been noted in Michigan for other species of bog clubmosses (Robert Preston, University of Michigan, pers. comm.j. Lycopodiella Xgilmanii usually has ascending sporophylls at maturity. Rarely, however, collections have loosely appressed sporophylls until very late in the season when they spread further from the axis. The latter form has been seen from northeastern Connecticut and appears to merely represent dwarfed individuals with short upright shoots. Lycopodiella Xgilmanii also appears to have two forms based on spore morphology - those with abortive spores and those with well formed spores. This suggests one of the parents may occur in two ploidy levels (likely L. appressa; see discussion under that species in Haines 2003). All of the variations of L. Xgilmanii are united by similarities in morphology of horizontal shoots, upright shoots, leaves, and sporophylls, spore size, and in geography (i.e., they occur within the region of sympatry of L. appressa and L. inundata). The holotype of L. Xgilmanii has a high fraction of aborted spores. Despite previous confusion, Lycopodiella Xgilmanii is readily separated from Lycopodiella appressa by examination of sporophylls and horizontal stems. Lycopodiella Xgilmanii has sporophylls commonly exceeding 5 mm long with 0-2 slender teeth per margin (Figure 2) and horizontal shoots, excluding the leaves, 0.9-1.5 mm thick. Lycopodiella appressa, on the other hand, has sporophylls usually shorter than 5 mm long with entire margins or infrequently with a short, broad tooth on one or both margins (Fig. 3; rarely the teeth prolonged and slender) and horizontal shoots 1.2-3.5 mm thick. Further, most collections of L. Xgilmanii have ascending sporophylls at maturity, rather than the appressed sporophylls of L. appressa. In northeastern herbaria, L. Xgilmanii is most often labeled as Lycopodium inundatum var. bigelovii Tuckerman. Examination of the type specimen [Tuckerman s.n., GH!) shows this name to be a synonym of L. appressa, in contradiction to the statements of Gillespie (1962), who believed the name applied to hybrids involving L. appressa and L. inundata. Lycopodiella Xgilmanii is close in morphology to L. Xcopelandii (Eiger) Cranfill (= L. alopecuroides X L. appressa), which also has long, ascending sporophylls at maturity. Lycopodiella Xcopelandii is, however, a more robust plant with somewhat arching stems and more densely imbricate leaves (see key; couplet 4). Lycopodiella Xgilmanii is responsible for reports of HAINES: LYCOPODIELLA xGILMANII 201 L. margueritae in New England (Bruce, 1975 - as L."appressed inundata"; Angelo and Boufford 1986; and several unpublished media), a tetraploid species of the Great Lakes region (Bruce et al., 1991). Though the plants are similar in overall outline, sporophyll orientation, etc., L. margueritae is a larger plant with thicker horizontal shoots (mostly 1.3-2.2 mm thick), wider horizontal shoot leaves (0.8-1.2 mm wide), and larger spores (mostly 58-65 urn; Bruce, 1975 and Bruce et al, 1991). Lycopodiella Xgilmanii is probably also responsible for reports of L. margueritae from Pennsylvania (Roads and Block, 2000), but I have not yet seen specimens to confirm this. KEY TO LYCOPODIELLA OF NEW ENGLAND la. Sporophylls tightly to loosely appressed at maturity (i.e., late August through September), the bases spreading less than 15 degrees from the strobilus axis; strobilus 3-7 mm wide inclusive of the sporophylls 2b. Sporophylls (4.6-)5-6.4(-7.1) mm long, at least some with 1 or more slender, marginal teeth 0.3-0.6 mm long; horizontal shoots 0.9-1.5 mm thick exclusive of the leaves, usually producing 1 or 2 upright shoots per segmen long on one or both margins (rarely the teeth prolonged); horizontal shoots 1.2-3.5 mm thick exclusive of the leaves, usually producing 2-6 upright shoots per segment; well formed spores mostly 50-55 um in diameter L. oppress* lb. Sporophylls at maturity ascending to horizontally spreading, the bases spreading 30-90 ' e tips inwardly curved); strobilus 6-20 mm re margins; horizontal shoots verv slender. vith 1 upright shoot L. inundatt Kt least some of the leaves of the horizontal stems with 1 or more slender, marginal eeth; horizontal stems thicker, 0.9-3.1 mm in diameter, 7-36 cm long, each shoot legment with 1-5 upright shoots bases spreading from the axis 30-50 degrees; Horizontal shoots 0.9-1.5 mm thick, flat to the ground, commonly rooting 1.5-6.0 cm distal to the proximal-most upright shoot, usually producing 1 or 2 upright shoots per segment; many sporophylls and leaves with 1 or more (common forms of) L. Xgilmani e bases spreading from the axis 70-90 degrees; strobili 10-20 mm wide inclusive of the sporophylls 6a. Strobilus representing 6-38 percent of the total upright shoot height; horizontal stems strongly arching, frequently more than 3 cm above the substrate, commonly rooting 7.5-36 cm distal to the proximal-most upright shoot; leaves Strobilus representing 34-55 percent of the total upright sis horizontal stems somewhat arching, usually less than 2.5 cm above the / rooting 7.0-13.5 cm distal to proximal-most upright shoot; s with 0-3 teeth per margin L. X robust : VOLUME 93 NUMBER A atefully thanked for his contribution to this study. Da\ ert Preston donated information through discussion and helpful co: iked for comments on the manuscript. C. John Burk, Lisa Haines, Do n Reznicek, Dorothy Spaulding, and Emily Wood are also thanked LITERATURE CITED ANGELO, R. and D. E. BOUFFORD. 1986. Atlas of the Flora of New England: Pteridophytes an J. G. 1975. Systematics and Morphology of Subgenus Lepidotis of the Genus Lycopodiu, (Lycopodiaceae). Ph.D. Thesis, University of Michigan, Ann Arbor, MI. BRUCE, J. G., W. H. WAGNER, JR. and J. M. BEITEL. 1991. Two New Species of Bog Clubmosse Lycopodiella (Lycopodiaceae), from Southwestern Michigan. Michigan Bot. 30:3-10. th FERNALD, M. L. 1950. Gray's Manual of Botany, 8 edition. Van Nostrand Reinhold Company, Ne York, NY. GILLESPIE. J. P. 1962. A Theory of Relationships in the Lycopodium inundatum Complex. Ame Fern J. 52:19-26. HAINES, A. 2003. The Families Huperziaceae and Lycopodiaceae of New England. V.F. Thomas Cc Bar Harbor, ME. 0LLGAARD, B. 1987. A Revised Classification of the Lycopodiaceae s. lat. Opera Bot. 92:153-178. ROADS, A. F. and T. A. BLOCK. 2000. The Plants of Pennsylvania. University of Pennsylvania Pres Philadelphia, PA. WAGNER, W. H., JR. 1968. Hybridization, Taxonomy, and Evolution. Pp. 113-138 in V.H. Heywooi BRUCE, W. H., JR. and J. M. BEITEL. 1992. Generic Classification of Modern North America Lycopodiaceae. Ann. Missouri Bot. Gard. 79:676-686. W. H., JR. and J. M. BEITEL. 1993. Lycopodiaceae. Pp. 18-37 in Flora of Nor1 America Editorial Committee, eds., Flora of North America, vol 2. Oxford I imcrsitv Pres New York, NY. WAGNER, WAGNER, SHORTER NOTES The Common Staghorn Fern, Platycerium bifurcatum, Naturalizes in Southern Florida.—Platycerium bifurcatum (Cav.) C. Chr. is a popular ornamental staghorn fern that is widely cultivated in the tropics and subtropics and under protection in cooler climates. Native to Australia, New Guinea and Indonesia (Jones, D. L. 1987. Encyclopedia of Ferns. Timber Press, Portland, Oregon.), it thrives out-of-doors in southern Florida, where large cultivated plants suspended by chains hung from residential trees or houses are a frequent sight. These ferns are exceptionally abundant in some areas. For instance, at least 19 large cultivated P. bifurcatum plants grow in 7 of the 11 yards on one street in Ft. Lauderdale, Broward County, southeastern Florida. Early in 2001, young sporophytes of a staghorn were observed growing on a live oak tree [Quercus virginiana Mill.) in a residential neighborhood in Ft. Lauderdale. By January 2002, one of these plants had grown fertile fronds, enabling it to be identified as P. bifurcatum. The same tree bore an estimated 25 younger, non-spore bearing plants. Two large P. bifurcatum plants hang from trees across the street, within 50 meters of the colonized tree. About one half mile away, two P. bifurcatum plants "volunteered" on a live oak growing next to a yard with many large spore-producing P. bifurcatum plants. These colonizations appear to be a local phenomenon related to the close proximity of fertile plants. Later in January 2002, I found P. bifurcatum growing in a native live oak forest at Tree Tops County Park in southwestern Broward County. During a three hour survey of the park, a total of 19 plants were located on 11 large live oaks. Three plants had fertile leaves, and two of these were large clumps of plants more than one meter across with numerous basal and foliage fronds, and remaining 16 sporophytes were of various sizes but all had at least one foliage frond. Four of the 11 trees had more than one plant, all of which were on separate branches. Most of the plants were on the upper or lateral sides of the larger branches 4.5 to 9 m off the ground. These 11 host trees were scattered within a forest stand about 600 meters in length. Tree Tops Park is approximately 11 km west of the Ft. Lauderdale residential oaks with the colonizing P. bifurcatum plants. In January 2002, I also surveyed the mixed hardwood forest at the Broward County Flamingo Environmentally Sensitive Lands Site, about eight km. west of Tree Tops Park. A single medium-sized plant of P. bifurcatum with multiple basal and foliage fronds was found during the two-hour search of the site. No fertile leaves were apparent on this plant, which was growing in a live oak about five meters above the ground. An unpublished list of the plants at Tree Tops Park and the adjacent Pine Island Ridge Preserve, compiled by P. Howell in 1995, included P. bifurcatum (P. Howell pers. com.). The plant was a single young sporophyte found 204 AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 4 (2003) growing on an oak the forest in 1994 (P. Howell, pers. com.), which suggests that the fern was naturalizing in the park by that date. The age of the colonizing staghorn plants may be judged by their size. Under optimal conditions, it can take P. bifurcation up to a year to grow from a spore to a young sporophyte to initiate foliage fronds, and another 3-4 years to produce fertile fronds (B.J. Hoshizaki, pers. com.) Multiple basal fronds can be produced after about two years. This suggests that 3 of the 19 plants at Tree Tops Park are about a year old, 5 are 2 years or older, 8 are between 2 and 5 years old, and three are probably more than 5 years old. Because the growing conditions in Broward County are probably suboptimal due to cool and dry winter weather, the plants are probably older than they appear. Both Tree Tops Park and the Flamingo Preserve have residential areas within one km of the park which could be spore sources. The source of spores for the Ft. Lauderdale residential oak colonization is likely nearby cultivated plants that occur within 100 meters of the tree. It is not known whether Platycerium gametophytes are able to self fertilize (B. J. Hoshizaki, pers.com.). The ability to self fertilize would make naturalization easier because only one spore would be needed to establish a plant and population. Self fertilization seems desirable in epiphytic ferns growing on tall trees in dense forests. Ferns that are long distance dispersers are more likely to self fertilize (Peck, J., C. Peck and D. Farrar. 1990. Amer. J. Bot 80:126-126.). The two climbing ferns invasive in Florida, Lygodium japonicum (Thunb. ex Murray) Sw. and L. microphyllum (Cav.) R. Br., can self fertilize (Lott, M. S., J. C. Volin, R. W. Pemberton and D. F. Austin. 2003. Amer. J. Bot. 90:1144-1152.). In Australia, P. bifurcatum occurs in tropical and subtropical Queensland, and extends into temperate New South Wales (Jones, D. L. 1987. Encyclopedia of Ferns. Timber Press, Portland, Oregon.). The fern has survived -9°C on Mount Boss in New South Wales and it occurs at 240-450 m near Sydney (Graf, A. B. 1992. Tropica, Color Cylopedia of Exotic Plants, 4th Edition. Roehrs Co., East Rutherford, New Jersey.). Sydney is located about ca. 34 degrees south latitude, whereas Broward County, Florida lies at ca. 26 degrees north. A commercial nursery of P. bifurcatum in West Palm Beach County, just north of Broward County, has survived many freezing temperatures during its 40 years of operation (D. Rowett, pers. com.). The nearby weather station at Loxahatchee recorded low temperature between -3 and -4°C for eight years between 1961 and 1990 (Southeast Regional Climate Center, 2002. sercc® cirrus.dnr.state.se.us). Older staghorn plants may be able to tolerate freezes because their rhizomes are insulated by the masses of base fronds and sometimes have the ability to produce new base and foliage fronds if the old ones are killed. Florida's dry season can kill young plants, but larger plants are resistant to drought (Dave Rowett, pers. com.). These factors suggest that plants, should persist in southern Florida and based on low-temperate tolerance, P. bifurcatum should be able to extend it distribution northward. If P. bifurcatum plants become very dense on trees, they could displace native epiphytes. In the oak forests presently colonized, most of the branches, including those with P. bifurcatum are covered with resurrection fern [Pleopeltis polypodioides Humb. & Bonpl. ex Willd.), and five species of bromeliads [Tillandsia balbisiana Schult. & Schult.f., T. fasiculata Sw., T. recurvata (L.) L., T setaceae Sw., T. usneoides (L.) L., T. utriculata L.) are common. Two of these bromeliads, T. fasciulata and T. utriculata, are classified as endangered by the State of Florida because of the attack of an exotic weevil which specifically feeds on bromeliads (Coile, N.C. 2000. Notes on Florida's endangered and threatened plants. Florida Division of Plant Industry, Bureau of Entomology, Nematology and Plant Pathology-Botany Section Contribution No. 38, 3rd edition, p.122.). If P. bifurcatum becomes abundant in other preserves, which are rich in rare endangered epiphytic orchids and bromeliads, it could become more serious threat. Its presence in Tree Tops and Flamingo represents another exotic species in natural areas already plagued with abundant introduced species. It is a more obviously nonnative component of the forests, than are the exotic figs [Ficus spp.) and shoebutton ardisia [Ardisia elliptica Thunb.), which have native counterparts. Given the incipient naturalization, despite an apparent long history of cultivation, and its modest abundance, it seems unlikely that P. bifurcatum will approach the severity of other invasive ferns in Florida. Examples of such include Lygodium microphyllum (Cav.) R. Br. (Pemberton, R. W. and A. Ferriter. 1998. Amer. Fern J. 88:165-175.), L. japonicum (Thunb.] Sw., Nephroplepis cordifolia (L.) C. Presl., N. multiflora (Roxb.) F.M. Jarrett exC.V. Morton, and Tectaria incisa Cav.. All of these are Category 1 invasive exotics (Austin et al, http://www.fleppc.org/99list.htm). Platycerium bifurcatum has probably had a long history of cultivation in southern Florida. The 1887 sales catalogue of the Royal Palm Nursery, Oneca, Manatee Co., lists P. alcicorne (Willem.) Tardieu. This species may have actually been P. bifurcatum, a similar species (Hoshizaki, B. J. and R. C. Moran. 2001. Fern Grower's Manuel. Timber Press, Portland, OR.). Platycerium bifurcatum tolerates Florida's subtropical climate better than P. alcicorne, native of eastern Africa and Madagascar (Hoshizaki and Moran, 2001). While P. bifurcatum may have naturalized previously, it did not persist. The plant's many horticultural forms (Hoshizaki and Moran, 2001,) and tropical to warm temperate distribution (Jones, 1987,) suggests considerable genetic variation. With increased population and residential gardening, it is likely that there are many more genotypes of P. bifurcatum plants present today and this may account for the current naturalization. Platycerium bifurcatum has naturalized in Hawaii, where it was documented to occur on three islands in 1991 (Wilson, K. A. 1996. Pacific Sci. 50:127-141.). With the naturalization of P. bifurcatum in Florida, the number of exotic ferns and fern allies in the state is now 34 (Wunderlin, R. P. 1998. Guide to the Vascular Plants of Florida. University Press of Florida, Gainesville). Wunderlin lists 32 species as introduced, to which Salvinia minima Baker can be added because of the recent recognition of the plant's exotic status (Jacono, C. C, T. D. Davern and T. D. Center. 2001. Castanea 66:214-226.). These 34 represent about one-third of Florida's fern species, the same proportion of naturalized seed plants in the state. Thus there seems to be no difference in the 206 AMERICAN FERN JOURNAL: VOLUME 93 NUMBER 4 (2003) ability of ferns and seed plants to naturalize in Florida. In Hawaii, however, where about half of the flora is comprised of naturalized species, only 19% of the ferns are naturalized (Wilson, K. A. 1996. Pacific Sci. 50:127-141; 2002. Amer. Fern J. 92:179-183), suggesting that ferns are less likely than seed plants to naturalize on those islands. Both the proportion of ferns that are naturalized and the severity of associated problems are greater in Florida than in Hawaii. Barbara Joe Hoshizaki, Patricia Howell, Broward County Florida Parks and Recreation, Dave Rowett Just Stags Nursery, Palm Beach Co. Florida, provided helpful information. Robbin Moran, The New York Botanical Garden, and Barbara Joe Hoshizaki kindly reviewed and improved the manuscript.—ROBERT W. PEMBERTON, Invasive Plant Research Laboratory, USDA-Agricultural Research Service, 3205 College Ave., Ft. Lauderdale, FL 33314 high quality of ar tides in the Am< mean Fern Journal and i Fern Society and iks to the following revie d patience in the year 2003. DAVID LELLINGER JERRY MCCLURE TOM RANKER MICK RICHARDSON PAUL RUSSEL JAMES PECK GERALD GASTONY VALERIE PENCE JAMES P. MICHAEL VINCENT PITTS L. CHRISTOPHER HAUFLER THERESA KERRY HEAFNER MICHAEL PRICE PITTS-SINGER GEORGE YATSKIEVYCH { Karyotype Comparison Between Two Closely Related Species of Acrostichum, 116 i Re-evaluation of Isoetes savatieri Franchet in Acrospermum, 86 Acrostichum, 116-125; aureum, 5 Hum, 49-55, 116-125, 153, M. S. (see W. D. HAUK) M. S. (see R. J. HICKEY) M. S. and W. D. HAUK. An Evaluati Scepteridium dissectum (Ophiogli Markers: Implica for Scepteridium Systematics, 1 BARROS, I. C. L. (see A. B. MARCON) 29, 103, 149; brasiiiense, 103, 10; glandulosum, 1! 46 (comb. BARKER, BARKER, BARKER, 147 . \Vu.)i). (D.C. Eaton) Robins. (Aspleniaceae) Redis- . dissectum. decompositun turn, 1-16; disi turn var. tenuifolium, 2: dissectum var. typicum, 2; gallicomontanum, 143; jenma- - - INDEX TO VOLUME 93 Ceratopteris richardii, 123 Cereus giganteus, 56 S., M. SRIVASTAVA and R. SRIYASTAYA. Contribution to the Gametophyte Morphology of the Fern Genus Lomagmmma J. Sm. In India, 25 , 20-24; CHANDRA, Chusquea scandens, i Cochlidium, 8 Corrections and Additional Information Ferns from the Semi-Arid Regioi Egenolfia, 26; vivipara, 30 Elaphoglossum, 26, 29, 30; cuspidatum, ipshookense, 152; tectum, 152 Enterosora, 81, 84; asplenioides, 84 Fagus grandifolia, 93 FARRERA, M. A. P., B. PEREZ-GARCIA. R. RIP.A E. LOPEZ-MOLINA. New Records for Pteridoflora of Chiapas, Mexico, 152 FIFWMJI.'. R.. AND R. VAIL. New Records >,m Spores in Natural Soils, 70 16; blotiana, 45; birta, 45; intermedia, 46; longissima, 45; microphylla, 45, 46 Grnmmitis. 81: anfractuosa, 86; jamesonioides, P. G. (see H. W. KELI Dennstaedtia punctilobula, II Dicranopteris, 44; pectinata, ' Dicranum montanum, 39; via DAVISON, IDC. Eaton) Robins. (Aspleniaceae) Rediscovered in Hawai'i, 154 Diphasiastrum complanatum, 22; digitatum, Haplohymenium triste, 39 Harpaleieunea ovata subsp. integra, 39 HAUFLER, C. H. (see H. W. KELLER) HAUK, W. D. (see M. S. BARKER) HAUK, W. D. and M. S. BARKER. Botrychium lanceolatum subsp. angustisegmentum in FERN JOURNAL: VOLUME 93 NUMBER 4 (2003) Hawai'i's Ferns and Fern Allies, Review, 95 Hedyosmum, 83 Hedyotis terminalis, 155 in the Great Smoky Mountains National Park, 36 Ko, W. Germination of Fern Spores in Natural Helminthostachys zeylanica, 137 HICKEY, R. J. Review: A Modern Multilingual Glossary for Taxonomic Pteridology, 164 HICKEY, R. J. Review: The Cycads, 47 Lastreopsis, 32; exculta, 148; squamifera, 146, 148 {COmb. nov.) Leaf Flavonoids in the Genus Gleichenia Glei- R. J., M. A. BARKER and M. PONCE. An Adiantopsis Hybrid from Northeastern Argentina and Vicinity, 42 B. J. Review: Index to Distribution Maps of Pteridophytes in Asia. 2nd Edition, 166 Hyobanche, 14 Hypolepis mitis, 148; rigescens, 148; rubiginosopilosula, 146, 147 [spec, nov.), 148; stolonifera, 148; viscosa, 148 HICKEY, HOSHIZAKI, Isoetes, 135; alcalopbila, 134; andina, 189; appalachiana, 189; canadensis, 135; chilensis [nom. nud.}, 133; cbubutiana, 126, 1 N C it! PYaCed in Diplazium (Woodsiaceae), 90 M. Six New Species of Tree Ferr from me Andes, 169 Lejeunea lamacerina subsp. gemminata, 3' rutna 39; „Iirinia 39 LELLINGER, D. B. Nomenclatural and Taxonom Notes on the Pteridophytes of Costa Ric Panama, and Colombia, III, 146 Lellingeria, 81 LE6N, B. and A. R. SMITH. New Species an LEHNERT, Leucoaon brachypus, 39 Lindera benzoin, 93 Liriodendron tulipifera, [ Lomagramm 135; bieronymii, 127; luetzelburgii, 134; neyeri [nom. nud.], 130, 189; riparia, 135; saccharata, 135; mrafj'eri, 126-129, 131-135; sforJcii, 134; fennesseensis, 184, 185,187-188 [spec, nov), nud.}, 130, 133; valida, 189 Isoetes tennesseensis (Isoetaceae), an Octoploid Quillwort from Tennessee, 184 Isoetites, 135 ITAM, K. (see U. YUSUF) . „ . ,. „„ „_ Jamesomella autumnalis, 37, 39 Quercitin 3-0-[2\ 3"-di-0-p157 H. W., P. G. DAVISON, C. H. HAUFLER and D. B. LESMEISTER. Polypodium appalachianum: An Unusual Tree Canopy Epiphyte KELLER, Lomariopsidaceae, 149 Lomariopsis, 25, 148, 149; fendleri, 148, 149; teracea, 30; salicifolia, 146,148 (comii. •v.); sorbifolia, 148, 149 LOPEZ-MOLINA, E. (see M. A. P. FARRERA) LUEBKE, N. T. and J. M. BUDKE. Isoetes tenneswort fr °m Tennessee, 184 Lycopodiella, 196,197; alopecuroides, 200, 201; alopecuroides X appressa, 200; appressa, 196 > 199-201; appressa X inundata, 197; Xcopelandii, 200, 201; xgilmanii, 196, 197-199 [bybr. nov.), 199-201; inundata, 200, 201; margueritae, 201; ^ , t • ' Xrobusta, 201 Lycopodiella Xgilmanii (Lycopodiaceae), a New Hybrid Bog Clubmoss from Northeastern North America, 196 Lycopodium, 22; inundatum, 196; inundatum var. bigelovii, 196, 200 Lygodium japonicum, 204, 205; microphyllum, 204, 205; venustum, 107-110 INDEX TO VOLUME 93 C. (see R. J. HICKEY) Macrothelypteris torresiana, 103, 108-110 MARCON, A. B., I. C. L. BARROS and M. GUERRA. A Karyotype Comparison Between Two Closely Related Species of Acrostichum, MACLUF, Marsilea, 153; deflexa, 153; polycarpa, 153; Megalavtrum, 32 MFHIIRFI-LR K P. ROJAS and M. PALACIOS-RIOS. Moth Larvae-damaged Giant Leather-fern L. and R. C. MORAN. Lectotypificatic Several Names Currently Placed i Diplazium (Woodsiaceae), 90 PACHECHO, of _ Secondary Colonization by Ants, Melpomene, 81, 86; anfractuosa, 86; Pr, ;,'-::: Metrosideros, 155; polymorpha, 155 Metzeeria sp 39 Miconia, 86 " R. W. The Common Staghoi Platycerium bifurcatum. Natural PEMBERTON, Microlepia mannii, 154; strigosa, 15 todium, 81 R. C. (see L. PACHECHO) -damaged Giant Leather-i i danaeifolium a ary Colonization by Ants, 49 Myrmelachista mexicana, 52 Myrsine, 86 [AN, Nephrodium sodiroi, 149 (lectotype); muftjzn, 154 Nephrolepis cordifoliu. 205: exaltata. 70-74; multiflora, 205 American m Baker in Penstemon, 14 PEREZ-GARCIA, B. (see N Pheidole sp., Phlebodium a var - calomelanos, 1U7-11U Platycerii ..ndinum, 160163 ; bifurcatum, 203-205 Pleopeltis polypodioides, 205 Polypodium, 50; an/racfuosum, 86; appalacbiammii 36-41; chirripoense, 146,149 (spec. noy.), 150; rum> 87, 88; Polypodium appalachianum: An Unus ual Tree Smoky Mountains National Park, 36 PONCE, M. (see R. J. HICKEY) obconica, 39 L, M. Soil Spore bank of Ferns in a Gallery Forest of the Ecological Station of Panga. Uberlandia, AMERICAN FERN JOURNAL: VOLUME 9 Rapid Gametophyte Maturation in Ophiogh sum crotalophoroides, 137 Review: A Modern Multilingual Glossary j Taxonomic Pteridology, 164 Review: Index to Distribution Maps of. A. Si KARI. Leaf Flavonoids in the Genus Gleichema (Gleicheniaceae), 44 Iff II 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 prepared may be returned for revision prior to review. Submit manuscripts in triplicate (xerox copies acceptable), including review copies o! illustrations and originals o\~ . After review, submission of final versions of manuscripts on diskette or Mac-compatible formats) is strongly encouraged. Use standard 8Vi by i paper of good quality, not "erasable" paper. Double space manuscripts including title, authors* names and addresses, short, informative abstract, text (including heads and keys), literature cited, tables (separate from text), and figure captions (grouped as consecutive paragraphs separate from figures). Arrange parts oi manuscripl in order just given. Include author's name and page number in upper right corner of every sheet. Provide margins of at least 25 mm all around on typed pages. Do not submit right-justified copy, avoid footnotes, and do not break words at ends of lines. Make table headings and figure captions self-explanatory. Use S.I. (metric) units for all measures (e.g., distance, elevation, weight) unless quoted or cited from another source (e.g., specimen citations). For nomenclatural matter (i.e., synonymy and typification), use one paragraph per basionym (see Regnum Veg. 58:39-40. 1968). Abbreviate titles of serial publications according to Bot Lawrence et al.. 1968, Hunt Botanical Library, Pittsburgh) and its supplement (1991). References cited notes and reviews, omit the abstract and put all references parenthetically in text. Use Index Herbariorum (Regnum Veg. 120:1-693. 1990) for designations of herIllustrations should be proportioned to fit page width with caption on the same page. Provide margins of at least 25 mm on all illustrations. For continuous-tone illustrations, design originals for reproduction without reduction or by uniform amount. In composite blocks, abut edges of adjacent photographs. Avoid combining continuous-tone and line-copj in single illustrations or blocks. Coordinate sequence and numbering of figures (and of tables) with order of citation in text. Explain scales and symbols in figures themselves, not in captions. Include a scale and reference to latitude and longitude in each map. Proofs and reprint order forms are sent to authors by the printer. Authors should send corrected proofs to the editor and reprint orders to the printer. Authors will be assessed charges for extensive alterations made after type has been set. For other matter of form or style, consult recent issues of American Fern Journal and The Chicago Manual of Style, 14th ed. (1993, Univ. Chicago Press, Chicago). Occasionally, departure from these guidelines may be justified. Authors are encouraged to consult the editor for assistance with any aspect of manuscripl preparation. Papers longer than 32 printed pages may be sent to the Editor of Pteridologia PTERIDOLOGIA ISSUES IN PRINT Western North 2A. Lellinger, David B. 1989. The Ferns and Fern-allies of Costa Rica, Panama, and the Choco (Part 1: Psilotaceae through Dicksoniaceae). 364 pp. $32.00 postpaid. Send your order with a check or money order tc Inc.. % U.S. National Herbarium MRC-166, Smithsonk DC 20560. AMERICAN FERN JOURNAL ON MICROFICHE Volumes 1-61 of the American Fern Journal are available as archival quality, silver positive microfiches. Single volumes or the entire run may be purchased. The fiches are easily read with 10X or greater magnification (using a dissecting microscope and transmitted illumination or a fiche reader). Silver negative microfiches of vols. 1-50 are also available. The price is $4.00 per volume or $244.00 per set of 61 volumes, postpaid. Send your inquiry or order with a check or money order to: American Fern Society, Inc., % Dr. James D. Montgomery, Ecology III, Inc., R.D. 1, Box 1795, Berwick. PA 18603. VISIT THE AMERICAN FERN SOCIETY'S WORLD WIDE WEB HOMEPAGE: http://www.amerfernsoc.org/