A Mink, J.N., J.R. Singhurst, and W.C. Holmes. 2011. Dryopteris celsa (Dryopteridaceae) and E.J. Palmer 29404: Solution of a Texas mystery. Phytoneuron 2011-11: 1-8. DRYOPTERIS CELSA (DRYOPTERIDACEAE) AND PALMER E.J. 29404: SOLUTION OF A TEXAS MYSTERY Mink Jeffrey N. Department of Biology Baylor University Waco. Texas 76798-7388 U.S.A. [email protected] R Jason Singhurst Wildlife Diversity Program Texas Parks and Wildlife Dq>aitment 4200 Smith School Road Austin, Texas 78704 U.S. Walter Holmes C. Department of Biology Baylor University Waco, Texas 76798-7388 U.S.A. ABSTRACT A study of the specimens of EJ. Palmer 29404 from Bowie County, Texas, validates the occurrence of Dryopteris celsa in Texas and discloses that the specimens are mixed in various combinations with Osmunda cinnamomea (Osmundaceae). KEY WORDS: Dryopteris. Dryopteridaceae, Osmunda cinnamomea. Osmundaceae, Texas, ELI. Palmer. Dryopteris taxonomy has been problematic in Texas, although the genus represented by is only five species (Mink et al. 2010). This has been caused by several factors, one being the difficulty in delineating the genus Dryopteris from Thelypteris. For example, Gould (1962) included eight Of species of Dryopteris in his checklist of Texas plants. these eight, only D. feiix-mas (L.) Schott, is known The now presently to occur in the state. other species are placed in Thelypteris or its segregates. Also complicating the systematics have been an underestimation of geographic distributions (e.g. Dryopteris celsa by Small 1938) as a result of mis identification (as discussed here) and the occurrence of hybridization within the genus. The North American species of Dryopteris are widely distributed throughout northern temperate areas and freely hybridize (Wagner 1971; Walker 1962). These hybrid relationships are well-established (e.g., fertile and sterile crosses) and acknowledged (Wagner 1943), but this hybridization intensifies the difficulty and complexity of species identification in the field (Wagner 8c & Musselman 1979; Lellinger 1985; Montgomery Wagner 1993; Peck 2000). Despite fair understanding of hybrid relationships, identification of hybrid Dryopteris has confused definitive confirmation of distinctive morphological characters and subsequent nomenclature of the genus. Although no species of Dryopteris were described from Texas vouchers, parent taxa names were imported from areas where hybridization was common. This hybridization relevant to the Texas is fern flora because names of various fertile tetraploids (e.g., D. celsa and D. crislata) were reflected in the identification of historic Texas collections by reference to more eastern and northern taxa D. (i.e., cristata). Fertile tetraploids ultimately have become recognized as part of the modern Texas flora (e.g., D. celsa and D. ludoviciana), even though parental taxa do not occur in the state. Inclusion of Dryopteris cristata (L.) A. Gray in Texas and controversy regarding the presence & or absence of D. celsa (W. Palmer) Knowlton, W. Palmer. Pollard in Texas primarily associated is MO, UMO), with interpretation E.J. Palmer 29404 (GH, which was collected along "Margins of sandy bog, near Texarkana, Bowie County, Oct. 27, 1925." The specimen was determined to be D. cristata (in part) by Correll (1956). This determination was included in the floristic account by Correll and Johnston (1970. 1972) and Correll and Correll (1972). Complicating the identification Two has been the mixed nature of these specimens, which the subject of this paper. of the is specimens distributed under this number are now known to be a mixture of Osmunda cinnamomea L. andD. celsa. — MO For this study, four herbarium sheets of Palmer 29404 were located two at and one UMO MO UMO GH each at and GH. The and specimens were received on loan, while the specimen was studied from a high resolution digital image. The original labels of all four sheets identified the Osmunda cinnamomea. ferns as UMO The sheet (Fig. 1) contains one frond, identifiable as sterile Osmunda cinnamomea by the cinnamon colored woolly hairs in the axils of the pinnae. As written on the sheet, the specimen was from the herbarium of E.J. Palmer and was a gift to the University of Missouri. There are no annotations on this sheet. Since this was in the collector's personal herbarium, it is presumed to be the "best" sheet and suggests that O. cinnamomea was the object of the collection. This was confirmed by reference to Palmer's field notebook, where he indicated that 29404 O. cinnamomea is may (R.C. Kennedy, pers. comm.). This specimen be involved with circumstances surrounding the & exclusion of Dryopteris celsa from Texas in the Flora of North America (Montgomery Wagner 1993). MO MO Two sheets of Palmer 29404 are housed at (Figs. 2, 3). 925450 has a sterile Osmunda cinnamomea frond, identified by the cinnamon colored hairs at the bases of the pinnae and MO more or less entire margins of the pinnae. 2139996 consists of a sterile frond of Dryopteris celsa, which is distinguished from Osmunda by the brownish to tan triangular scales at the base of the stipe and pinnules, serrated margins of the pinnae, and presence of foveolae near the margins of the mm upper surface of the leaf. The vein in the foveola is swollen into a clavellate shape of ca. 0.2 long and ends before reaching the margin. Several lines of evidence indicate mat these two specimens once constituted a single sheet: conventional practice would prohibit two sheets of the MO same number (or specimen) sent to the same place in exchanges; the label of 2139996 (D. celsa) is photocopied, as determined by the "toner" nature of the print, abundance of toner specking over the surface, and lack of impact type impressions found on the labels of the other sheets distributed under cm this number; the label is smaller in size (10.7 x 5.9 vs. 10.8 x 7 cm) than the labels of other specimens with this number; and the higher accession number of the Dryopteris sheet, which is suspicious. That these two specimens were originally mounted on a single sheet confirmed by is Correll's (1956) statement: "A frond of this species [Dryopteris cristata, a misidentification of D. celsa] is mounted with a frond of Osmunda cinnamomea on sheets in the Herbarium of the Missouri Botanical Garden and the Gray Herbarium. Although an error in mounting the specimens on the herbarium sheets might have occurred, the fact that two sheets in two different herbaria, labeled "Osmunda cinnamomea" each have a frond of this species and of cinnamomea would not appear to substantiate this assumption. It appears more likely that the mixture of fronds of the two species was obtained in the field and was distributed accordingly" 151-153). (pp. This also lends support to the fronds being collected at the same time and place Bowie Co., (i.e., Texas). The two specimens (prior to removal to separate sheets) were annotated by Correll using an "ANNOTATION offset printed annotation label. Across the top was printed LABEL." Under this was inked in blue: "Dryopteris cristata (L.) Gray." Immediately below this was inked in "Osmunda "DETERMINED BY cinnamomea L." Further below the Osmunda annotation was offset-printed DONOVAN S. CORRELL," followed by a handwritten "1951" at the right margin. Upon separation of the fronds, the annotation label was removed from the original sheet and split lengthwise between the two binomials, as determined by the matching proportions of the raw cut edges. To the Dryopteris portion (upper half) of the label was written in: "del Donovan S. Correll 1951, "but not in (MO the hand of Correll. This was affixed to a new herbarium sheet 2139996) with the photocopied label and Dryopteris frond. Two additional annotations are now on the sheet, one by T.H. Peck 1985, the other by R. Cranfill 1981, both identifying the specimen as D. celsa. The Osmunda cinnamomea specimen was also removed from the original sheet and remounted on a new herbarium sheet, along with the original label. The elliptical Missouri Botanical Garden stamp with the accession number (MO 925450) was cut from the original sheet and glued to this new sheet. The original label and accession number were included with the Osmunda sheet, possibly because made the identification it may correct and have been presumed to be what Palmer intended to collect. In addition to this information on the herbarium sheet the bottom portion of the cut annotation label identifying the A specimen as O. cinnamomea was included. small part of the lower left corner is missing from this now-Osmunda annotation label but otherwise visibly unaltered after being sectioned. Since the is sheet (MO 925450) not annotated by anyone other than Correll. who saw the two fronds mounted is on the same sheet, it is assumed that no one else lias critically studied the specimen, possibly because there no reason to assume there more than one Palmer 29404 specimen in the same herbarium. is is Mention must be made that the annotation of these two sheets was not done by Correll, but by an unknown person using CorrelFs original annotation label, but in a doubly altered form (cut and amended). It appears reasonable to assume that the original mixed specimens were mounted in the mid to late 1920s, shortly after its collection in 1925. It is likely that the specimens were split and remounted in the early 1970s (a time when photocopying was developed enough to achieve acceptable reproduction). According to John Pruski (pers. comm.) of the Mssouri Botanical Garden, MO MO He 2300000 was mounted in 1976, making the separation date above proximate. also said splits mixed sheets when they are noticed, and then the policy is to write on each sheet what was formerly mixed with Unfortunately, such notation was not done in this case. it. GH The sheet (Fig. 4) has sterile fronds of Dryopteris celsa (left) and Osmunda cinnamomea (right) on it. The Osmunda determination is confirmed by Correll as "! DSC" in his hand. No date is given. Near the Dryopteris frond penciled 'Dryopteris cristata (L.) A. Gray" followed below with is DSC "! 1947." Comparison of handwriting shows that none of this appears to be in the hand of Correll. Generally, the exclamation mark before a name (or initials) means that the person considers the determination as correct, in this case, the Osmunda frond. Lack of an exclamation mark means a new or initial determination is ma—de and the new name written in. Nonetheless, the identification of the Dryopteris frond incorrect the specimen D. celsa, not D. cristata. is is MO GH Correll (1956) mentioned the mixed nature of the sheet and the original unseparated sheet and used in his citation of Dryopteris cristata in Texas as "Palmer 29404 (in part)." The mixed nature of the sheets was also explained in the discussion following the specimen citation (see above). This was also mentioned in Correll and Johnson (1970, 1972) and Correll and Correll (1972), as "Palmer 29404, " but in these circumstances no further explaination was afforded as in Correll p.p., UMO (1956). Since Correll (1956) never examined the sheet, he didn't know the full extent of the MO mixture. Nor could he have been aware of the later separation of the fronds on the sheet. Later MO workers should have been alerted to the alterations in the now two specimens through a literature MO review of the references mentioned in this paragraph, or, if examining both specimens, the unusual appearance of the sheets. Other than for the current authors, there no evidence that anyone is else has examined both sheets. The above mentioned mis identification by Correll, unequally mixed sheets, and later separation of specimens have combined to create a perplexing situation that could only be untangled by examination of all the specimens involved. Other than the mis identification, only Correll (1956) correctly assessed the problem. For other botanists, the outcomes depended upon which specimen(s) UMO GH was/were studied. In summary, the specimen Osmunda cinnamomea, the specimen is is MO mixed (Dryopteris ce/sa and O. cinnamomea)., while the originally mixed sheet has been split [MO [MO into two specimens, one 925450] O. cinnamomea and the other 2139996] D. celsa. is is Finally, this does eliminate the confusion associated with the reports ofD. celsa in Texas. ACKNOWLEDGEMENTS MO UMO MO, UMO, The authors thank the curators of and for providing specimens and GH PH TEX-LL and for digital images and permission to include them in the publication. and assisted through searches for additional specimens of Palmer 29404. John Pruski (MO) furnished MO information on the policy of splitting mixed specimens at and approximate dates of when the (UMO) accession numbers of specimens were mounted. Robin C. Kennedy provided information MO from E.J. Palmer's field log. Digital photography and computer preparation of images from and UMO were taken and processed by Darrell S. Vodopich (Biology Department, Baylor University). LITERATURE CITED Correll, D.S. 1956. Ferns and Fern Allies of Texas. Texas Research Foundation, Renner. Correll, D.S. and H.B. Correll. 1972. Aquatic and Wetland Plants of the Southwestern United States. Stanford Univ. Press; Stanford, California. Correll D.S. and M.C. Johnston. 1970. Manual of the Vascular Plants of Texas. Texas Research Foundation, Renner. Correll, D.S. and M.C. Johnston. 1972. Manual of the vascular plants of Texas: Additions and I. corrections. Amer. Midi. Naturalist 88: 490-496. A - Gould, F.W. 1962. Texas Plants Checklist and Ecological Summary. MP-585. Texas Agricultural Experimental Station, College Station. A & & Lellinger, D.B. 1985. Field Manual of the Ferns Fern- Allies of the United States Canada. Smithsonian Inst. Press, Washington, D.C. New Mink J.N., J.R. Singhurst, and W.C. Holmes. 2010. Dryopteris marginalis (Dryopteridaceae): to Texas. Phytoneuron 53 1-6. : Montgomery, J.D. and W.H. Wagner. 1993. Dryopteris. Pp. 280-288 in Flora of North America New Editorial Committee, Flora of North America, Vol. 2. Oxford University Press, York. Peck, J.H. 2000. Floristic status of log ferns (Dryopteris) in Arkansas. Arkansas Acad. Sci. 54: J. 88-102. Small, J.K. 1938. Ferns of the Southeastern United States. The Science Press, Lancaster, Pennsylvania. Wagner, W.H., 1943. Hybridization by remote control. Amer. Fern 33: 71-73. Jr. J. W.H, Wagner, Jr. 1971. Evolution of Dryopteris in relation to the Appalachians. Pp. 147-192 in P.C. Holt (ed.), The Distributional History of the Biota of the Southern Appalachians. Part 2, Flora. Virginia Polytechnic Inst, and State Univ., Res. Div. Monogr. 2, Blacksburg. Wagner, W.H. Jr. and L.J. Musselman. 1979. Log ferns (Dryopteris celsa) and their relatives in the , Dismal Swamp. Pp. 127-139 in P.W. Kirk Jr. (ed.), The Great Dismal Swamp. Univ. Press of Virginia, Charlottesville. Walker, S. 1962. Further studies in the genus Dryopteris: The origin of D. clintoniana, D. celsa and Amer. 497-503. related taxa. Bot. 49: J. Mink, Singhurst, and Holmes: Dryopteris celsa'm Texas 5 UMO) Figure Osmunda cinnamomea (EJ. Palmer 29404, 1. link, Singhurst, and Holmes: Dryopteris celsa'm Texas 5 Mink, Singhurst, and Holmes: Dryopterisi Figure Dryopyeris celsa (EJ. Palmer 29404, MO). Separated from Osn 3. MO original sheet (see text). Mink, Singhurst, and Holmes: Dryopten GH Figure 4. sheet ofE.J. Palmer 29404. composed of Dryopteris celsa (left) arid Osmunda * (right).