Analysis of morphological and anatomical characteristics of Isoetes using Isoetes tennesseensis JESSICA M. BUDKE 1'3, R. JAMES HICKEY, AND KERRY D . HEAFNER 2 Budke, J. M., R. J. Hickey & K. D. Heafner (Miami University, Department of Botany, 316 Pearson Hall, Oxford, OH 45056, U.S.A.; email: jessica. budke@uconn.edu). Analysis of morphological and anatomical characteristics of Isoetes using Isoetes tennesseensis. Brittonia 57: 167-182. 2005.--The three known populations of Isoetes tennesseensis were examined to document and analyze their morphological and anatomical characters. Characters examined included velum coverage, lacunae, leaf form and size, sporangial wall cells, and ligule and labium morphology. Three types of morphological patterns were found: stable, variable, and dimorphic. The discovery of dimorphism between mega- and microsporophylls for velum coverage, velum length, sporangium size, and ligule shape is of particular interest. This study presents a model for future work designed to complement current species descriptions and it establishes a baseline for morphological comparisons with other Isoetes. Key words: phism. Isoetes, morphological characters, lycopod, pteridophyte, dimor- Introduction Pfeiffer (1922) published the first and only worldwide m o n o g r a p h of Isoetes L. She proposed a new sectional system based on megaspore surface morphology. The strengths o f this taxonomic system were its simplicity and utility; even herbarium specimens with little or no information regarding habitat could be readily classified to section. However, Pfeiffer stated that m e g a spore surface m o r p h o l o g y can vary within species or within individuals, thus other characters should be used to definitively s e p a r a t e s p e c i e s . D e s p i t e her w a r n i n g s , most l s o e t e s species described after 1922 have been distinguished primarily on this character. Hickey (1986a, 1986b, 1986c) 1 Current address: University of Connecticut, Department of Ecology & Evolutionary Biology, 75 North Eagleville Rd,, Unit 3043, Storrs, C T 062693043, U.S.A. 2 Current address: University o f Louisiana at M o n roe, M u s e u m of Natural History, Department of Biology, Monroe, L A 71209-0520, U.S.A. 3 Author for correspondence. strengthened Pfeiffer's cautions by showing that megaspore surface m o r p h o l o g y is subject to convergence as well as within-species, within-plant, and within-spore variation. In spite o f such variation, megaspores continue to be the dominant source of taxonomic characters, whereas other sources of morphological variation remain largely neglected. Only a few papers on North A m e r i c a n Isoetes have focused on comparative morphological variation. Two notable examples are the studies by Matthews and Murdy (1969) and Kott and Britton (1985). They concluded that most vegetative characters are environmentally plastic, developmentally dependent, or invariant. Specifically, Kott and Britton (1985) strongly supported species identification in Isoetes based primarily on spore characteristics; whereas Matthews and M u r d y (1969) concluded that characteristics examined in their study were continuously variable across populations. Therefore, these papers m a y have discouraged analyses of morphological characters Brittonia, 57(2), 2005, pp. 167-182. 9 2005, by The New York Botanical Garden Press, Bronx, NY 10458-5126 U.S.A. ISSUED: 30 June 2005 168 BRITTONIA such as velum coverage, sporangial pigmentation, leaf form and size, and ligule and labium morphology in Isoetes research. Consequently, authors rarely stress their taxonomic usefulness. In the North American literature, B o o m (1979) and R e e d (1965) stand apart in maintaining an emphasis on vegetative morphology and ecological data in distinguishing taxa. Their data, and those of Hickey (1986a), suggest that intensive analyses of species morphologies will yield stable taxonomic characters. There are several examples of Isoetes characterizations from outside North America which thoroughly integrate morphological characters. A good example is the work of Rauh and Falk (1959a, 1959b) on Stylites gemmifera Rauh and Falk (=L andicola (Amstutz) L. D. G6mez). In 159 pages, they fully describe the anatomy, morphology, development, and ecology of this Andean species. Hall (1971) demonstrated the utility of a number of novel anatomical features in delineating Isoetes of Ghana by documenting variability among characters. Additionally, Prada (1979) examined leaf anatomical characteristics o f lsoetes species from Spain including L lacustris and /. echinospora, species that are both present in North America. Well-documented, infraspecific variability for selected anatomical and spore characters has also been published for the African L melanotheca Alston (Pitot, 1959) and I. jaegeri Pitot (Pitot, 1968). Recently, Takamiya et al. (1997) utilized morphological characters to differentiate among both divergent and closely related species in Japan. Similarly, Hickey (1986a, 1986b, 1986c) and Small and Hickey (2001) have shown that morphology can be successfully used to differentiate taxa in Sou~h America. Despite these studies, neither the methodologies nor the morphological characters espoused have been used to circumscribe North American lsoetes species with much regularity. During the 20 ~h century the number of recognized species in lsoetes has increased from 64 (Pfeiffer, 1922) to an estimated 350 (Hickey et al., 2003). In North America the numbers have increased from 19 (Pfeiffer, 1922) to 23 (Taylor et al., 1993), with sev- [VOL. 57 eral more added subsequent to 1993. Typically, these new species are distinguished based on cytology and megaspore ornamentation. With species of different ploidy levels, cytology is highly informative. Unfortunately, cytological data are not available for most collections. In order to thoroughly embrace the variability within a taxon, the organism's entire morphology needs to be examined and documented. For Isoetes, this includes the microphylls, sporangia, corm, ligules, and labia. Research in our laboratory indicates that such morphological characters are of value in defining Isoetes taxa and can provide an important complement to published species descriptions. The goal of this study is to document and a n a l y z e m o r p h o l o g i c a l and a n a t o m i c a l c h a r a c t e r s using I s o e t e s t e n n e s s e e n s i s Luebke and Budke as a model against which other currently recognized Isoetes species can be compared, This recently described species is the only known Isoetes octaploid (2n = 88) in North America. Its unique chromosome number provided a definitive method of species confirmation for the populations used in our study. Methods During the summer of 2001, living material was collected from the three known, extant populations of Isoetes tennesseensis all of which are in the Hiwassee River, Polk County, Tennessee (Table I). Identities of specimens from all three populations were confirmed cytologically following procedures modified from Luebke and Budke (2003). Only confirmed octaploids were u s e d in this study. Morphological characters were measured using either a flexible ruler, a Wild M3C dissecting microscope, or an Olympus BHS compound microscope. For both microscopes an ocular micrometer was used. Specifically, leaf width was measured at the mid-subula length and ala width was measured at the widest point above the sporangium. Velum length was measured as the distance from the apex of the sporangium, down the middle, to the edge of the velum. The velum coverage of the sporangium was then calculated as a percentage of sporan- 2005] B U D K E ET AL.: A N A L Y S I S O F I S O E T E S 169 TABLE I POPULATIONS STUDIED OF Isoetes tennesseensis. ALL WERE COLLECTED FROM THE H1WASSEE RIVER, POLK CO., TENNESSEE, U.S.A Numbersof individuals Population collectedand examined 1 2 3 9 12 9 Voucher Data 24 June 2001, J. Budke, K. Heafner, E. Lickey, P. Cox, & J. Shaw 1 (MU) 13 July 2001, J. Budke, K. Heafner, E. Lickey, & K. Gustafson 8 (MIL, MU) 15 July 2001, J. Budke, K. Heafner, E. Lickey & K. Gustafson 17 (MIL, MU; type collection) gium length. Microspores were mounted onto glass slides in H o y e r ' s m e d i u m (Willey, 1971), and their lengths measured. The diameters of dry megaspores were m e a sured using the dissecting microscope. To examine sporangial wall cells, a portion o f the exposed adaxial wall was dissected and m o u n t e d in H o y e r ' s m e d i u m , and cell lengths were measured using the c o m p o u n d microscope. All specimens examined had sporangia that were fully mature. The standard by which we determined this was spore maturity and completeness of ornamentation. Specimens for histological analyses were fixed in FAA (formalin-aceto-alcohol), dehydrated in a graduated tertiary butyl alcohol series, and e m b e d d e d in paraffin (Johansen, 1940). Specimens were sectioned using a rotary m i c r o t o m e at a thickness of 12 ~m, fixed to the slide using M a y e r ' s adhesive (Johansen, 1940) and stained with a 0.2% solution of toluidine blue. All specimens were examined using the c o m p o u n d microscope. L e a f cross sections were m a d e f r o m the middle o f the leaf subula. For scanning electron microscopy (SEM), megaspores and microspores were air-dried, whereas leaf material was fixed in FAA, dehydrated in an ethanol series to 100%, and critical point dried. All samples were mounted onto S E M stubs using double-sided tape and sputter-coated with ca. 21 nm of gold-palladium. The SEM analyses were carried out using a Jeol T-200 (Tokyo, Japan) scanning electron microscope. Standard descriptive statistics, including mean, median, and standard deviation were calculated for each character using M I N I TAB | 13.32 (Minitab Inc., 2000). Dimorphic characters and variation a m o n g popu- lations were assessed with two-sample ttests using M I N I T A B . Results The corms of Isoetes tennesseensis were consistently bilobed, and in our collections had a m a x i m u m size of 23.0 m m in height, 21.0 m m in diam. C o r m s as small as 1.0 m m in height and 5.0 m m across were also collected. Dichotomous roots, typical for the genus, were produced along the circumbasal fossa. L e a f length in Isoetes tennesseensis was quite variable with mean leaf lengths from 78.2 m m to 124.1 m m across the populations. Standard deviations for this character across the three populations were consistent, but high in all cases (Table II). The mean leaf length for the species was 104.3 m m . Alae lengths showed similar interpopulational variation (Table II). Ala to leaf length ratios were calculated for all specimens examined (Table II). No scales or phyllopodia were o b s e r v e d in this species. In cross section, the mid-subulae of lsoetes tennesseensis were most c o m m o n l y half-terete with flattened adaxial and rounded abaxial sides, some were weakly trapezoidal with the adaxial and abaxial sides flattened and parallel, while a few were terete in outline (Fig. 1A; Table II). The adaxial surface was broader than the abaxial, an a s y m m e t r y paralleled internally by statistically significant differences (in all cases p < 0.001) in the widths and depths of the adaxial and abaxial lacunae (Table III). No statistically significant within-character differences were found across populations. The lacunae were separated f r o m each other by parenchymatous cell layers that were consistently two to three cell layers thick 170 BRITTONIA [ V O L . 57 TABLEII LEAF CHARACTERS MEASURED FOR lsoetes tennesseensis. DOPULATION NUMBERS REFER TO THOSE LISTED IN TABLE [ Character Pop. 1 Pop. 2 Leaf length, rnm Mean 120.1 78.2 SD IN] 21.02 [16] 23.02 [25] Range 82-148 42-118 Ala length, mrn Mean 30.7 17.5 SD IN] 6.25 [16] 6.96 [25] Range 18-39 6-34 Ala width, turn Mean 3.4 2.6 SD IN] 0.86 [16] 0.60 [25] Range 1.7-4.8 1.2-3.3 Ala/leaf length Mean 0.257 0.226 SD IN] 0.045 [16] 0.045 [25] Range 0.190-0.352 0.097-0.312 Subula (leaf) width, mm Mean 2.00 ] .47 SD IN] 0.592 [16] 0.314 [25] Range 1.2-3.8 1.0-2.2 Subula (leaf) W/D ratio (serial section data) Mean 1.28 1.31 SD IN] 0.265 [20] 0.230 [25] Range 0.91-1.88 0.80-1.84 a n d f r o m the o u t s i d e e n v i r o n m e n t b y the e p i d e r m i s as w e l l as three to f o u r cell l a y e r s of parenchyma. There was no evidence of e i t h e r s t o m a or p e r i p h e r a l f i b r o u s b u n d l e s . I n t e r n a l l y , the l e a v e s o f I s o e t e s t e n n e s s e e n sis h a d a c e n t r a l v e i n a l r e g i o n (Fig. 1A, B) w i t h e i t h e r o n e (64 s a m p l e s ) o r t w o ( t w o samples) small intrastelar canals. These can a l s w e r e e l l i p t i c in s e c t i o n a l v i e w (Table IV). S e c o n d a r y w a l l t h i c k e n i n g s w e r e n o t p r e s e n t w i t h i n the v e i n a l canals. T h e l a c u n a e o f the l e a v e s w e r e t r a v e r s e d b y d i a p h r a g m s o f stellate p a r e n c h y m a (Fig. 1 C - E ) . I n I s o e t e s t e n n e s s e e n s i s , t h e s e diap h r a g m s w e r e c o n s i s t e n t l y t w o to t h r e e cells in t h i c k n e s s . T h e c e l l s w e r e flattened in a p l a n e p a r a l l e l to the d i a p h r a g m a n d e a c h h a d five to s e v e n r a d i a l a r m s that a b u t s i m i l a r a r m s o f a d j a c e n t cells. I n l o n g i t u d i n a l v i e w , the m a i n b o d i e s o f the cells w e r e v e r t i c a l l y a l i g n e d a n d t i g h t l y appressed. Small hair-like protuberances cove r e d the cell surfaces. A t the j u n c t u r e s o f the r a d i a l arms, the p r o t u b e r a n c e s w e r e Pop. 3 Pop. 1-3 combined 124.1 25.21 [20] 91-159 104.3 31.79 [61] 42-159 30.6 8.08 [20] 18-45 25.2 9.62 [61] 6-45 3.1 0.82 [201 1.9-4.3 3.0 0.80 [61] 1.2-4.8 0.245 0.029 [20] 0.190-0.310 0.238 0.042 [61] 0.097-0.352 1.86 0.262 [20] 1.5-2.4 1.74 0.449 [61] 1.0-3.8 1.33 0.266 [21] 0.83-1.73 1.31 0.244 [66] 0.80-1.88 larger, c l a v a t e a n d f o r m e d p a i r e d w h o r l s , w i t h o n e w h o r l p e r cell (Fig. 1D, E). L i g u l e s (Fig. 2 A - K ) in I s o e t e s tenness e e n s i s d e g r a d e as l e a v e s a p p r o a c h full m a turity. A s a result, m a t u r e , intact l i g u l e s w e r e difficult to o b t a i n a n d c h a r a c t e r i z e . Most ligules associated with megasporophylls were depressed-ovate whereas those o f m i c r o s p o r o p h y l l s w e r e d e l t a t e (Table V). Ligule ratios between mega- and microsporophylls were significantly different (p = 0.001). In all c a s e s the l i g u l e s s h o w e d a s l i g h t l y a u r i c u l a t e base. The labia of Isoetes tennesseensis were g e n e r a l l y spatulate, o f t e n w i t h a f o r k e d or b i l a b i a t e a p e x (Fig. 2 A - K ) . L e n g t h s , w i d t h s a n d l e n g t h to w i d t h r a t i o s w e r e d e t e r m i n e d for the l a b i a (Table VI). N o s i g n i f i c a n t diff e r e n c e s in the l e n g t h s ( p = 0.344), w i d t h s ( p = 0.846), or ratios ( p = 0.724) w e r e f o u n d b e t w e e n l a b i a o f m e g a - and m i c r o sporophylls. The sporangia of Isoetes tennesseensis were basal and broadly elliptic. Noticeable 2005] B U D K E E T AL.: A N A L Y S [ S O F I S O E T E S 171 FIG. 1. M o r p h o l o g i c a l and anatomical features of lsoetes tennesseensis. A. Cross section of a microphyll, showing the difference in size between the abaxial and adaxial lacunae. B. Intrastelar canal. C. Cross-sectional view of the stellate d i a p h r a g m cells, triangle shaped pores, and protuberances. D. SEM photograph of the stellate diaphragm cells from a cross-sectional cut. E. SEM photograph of longitudinal section of a leaf showing the stacked stellate diaphragm cells. F. Median longitudinal section of microsporangia s h o w i n g the v e l u m where the labium is just starting to form. G. Median longitudinal section of microsporangia s h o w i n g the velum where the labium is fully developed. Specimen A, B and G (Budke et al. 17); C and F (Budke et aL 1); D and E (Budke et aL 8). Abbreviations: abl = abaxial lacunae; adl = adaxial lacunae; cb = cell body; ic = intrastelar canal; li = ligule; la = labium; micro = microsporangium; p = pores; pr = protuberances; ra = radial arm; s = sella; v velum. Scale bars in A, F a n d G = 1 mm; scale bar in B = 100 p,m; scale bar i n C = 200 Ixm; scale bar in D = 30 ~m; scale bar in E - 50 Ixm. 172 BRITTONIA [ V O L . 57 TABLE III LACUNAE CHARACTERS FROM LEAF CROSS SECTIONS OF ]soetes tennesseensis. POPULATION NUMBERS REFER TO THOSE LISTED 1N TABLE I Character Abaxial lacuna Mean SD [NI Range Abaxial lacuna Mean SD IN] Range Abaxial lacuna Mean SD IN] Range Adaxial lacuna Mean SD IN] Range Adaxial lacuna Mean SD [N] Range Adaxial lacuna Mean SD IN] Range Pop. 1 width, p,m 632 110.2 [401 450-850 depth, txm 591 125.0 [40] 350-900 W/D ratio 1.11 0.303 [40] 0.64-1.88 width, i~m 796 181.9 [40] 480-1130 depth, p~m 650 128.0 [401 400-900 W/D ratio 1.25 0.327 [40] 0.80-2.28 Pop. 2 Pop. 3 Pop. 1-3 combined 571 122.5 [52] 380-950 686 140.6 [42] 450-1000 625 133.l [134] 380-1000 509 94.8 [52] 300-730 606 92.4 [42] 430-750 564 112.4 [134] 300-900 1.15 0.297 [521 0.66-1.93 1.16 0.288 [42] 0.71-1.73 1.14 0.294 [1341 0.54-1.93 735 172.0 [52] 430-1180 890 231.5 [42] 480-1380 802 204.6 [134] 430-1380 591 125.3 [52] 350-940 64 l 117.3 [421 410-930 625 125.6 [134] 350-940 1.28 0.333 [52] 0.68-2.16 1.44 0.459 [421 0.58-2.51 1.32 0.382 [1341 0.58-2.51 differences exist between the absolute sizes of megasporangia and microsporangia. Megasporangia a v e r a g e d 5 . 2 • 4.1 m m w h e r e a s t h e m i c r o s p o r a n g i a a v e r a g e d 3.6 • 2.8 m m . T h e s p o r a n g i a d i f f e r e d s i g n i f i c a n t ly in b o t h l e n g t h ( p = 0 . 0 0 5 ) a n d w i d t h ( p < 0 . 0 0 1 ) , b u t n o t in l e n g t h to w i d t h r a t i o ( p = 0 . 2 6 8 ) . T h e s e data, a l o n g w i t h s e p a r a t e p o p u l a t i o n statistics, a r e p r o v i d e d in Table VII. Cross sections of sporangia s h o w e d t h a t t h e p l a c e n t a e w e r e flat to c o n v e x b a s e d o n t h e d e p t h to w i d t h ratio. TABLE I V ][NTRASTELAR CANAL CHARACTERS OF Isoetes te~nesseel~sis. POPULATION NUMBERS REFER TO THOSE LISTED IN TABLE I Character Pop. 1 Intrastelar canal width, Ixm Mean 38.3 SD [N] 7.67 [19] Range 22.5-55.0 Intrastelar canal depth, ~xm Mean 33.0 SD [N] 6.03 [191 Range 25.0-50.0 Intrastelar canal W/D ratio Mean 1.18 SD [N] 0.227 [19] Range 0.67-1.63 Pop. 2 Pop. 3 9 Pop. 1-3 combined 30.8 5.56 [26] 18.8-41.3 37.3 6.69 [21] 27.5-52.5 35.0 7.41 [66] 18.8-55.0 26.5 6.05 [261 13.8-37.5 29.3 5.09 [26] 20.0-42.0 29.3 6.27 [66] 13.8-50.0 1.22 0.313 [26] 0.70-1.86 1.3 0.391 [26] 0.71-2.21 1.24 0.321 [66] 0.67-2 ~21 2005] BUDKE ET AL.: ANALYSIS OF ISOETES 173 1 mm J O0 IJm FIG. 2. Leaf base characters of Isoetes tennesseensis. A - K . Variation in ligule and labium shape and size. Labia in the foreground, ligules in the background. Most ligules partially or extensively degraded; A, C, E, H. Have the least amount of ligule degradation. I, J. Note bifid labia. L - Q . Variation in sporangial cell wall shape and size; L-N. Wall cells of megasporangia. O - Q . Wall cells of microsporangia. All drawings to scale. Specimens A-D, E G, L, M, O and P (Budke et al. 17); H-J (Budke et al. l); K, N and Q (Budke et al. 8). 174 BRITTONIA [VOL. 57 TABLE V LIGULE CHARACTERSFOR Isoetes tennesseensis. POPULATION NUMBERS REFER TO THOSE LISTED IN TABLE I Character Pop. I Pop. 2 Pop. 3 Pop. 1-3 combined Ligule length, mm Mean SD [N] Range 1.46 0.549 [8] 0.57-2.30 0.88 0.503 [6[ 0.31-1.60 1.57 0.485 [8] 0.65-2.30 1.34 0.570 [22] 0.31-2.30 Ligule width, mm Mean SD [N] Range 1.61 0.439 [8] 0.80-2.10 1.05 0.210 [6] 0.81-1.30 1.85 0.463 [8] 1.10-2.50 1.55 0.504 [22] 0.80-2.50 Ligule L/W ratio Mean SD [N] Range 1.05 0.789 [81 0.44-2.88 0.82 0.379 [6] 0.31-1.23 1.88 0.296 [8] 0.41-1.27 0.92 0.530 [22] 0.31-2.88 Ligule length, m m - - m e g a s p o r o p h y l l s Mean 1.29 SD [N] 0.638 [4] Range 0.57-2.10 0.37 0.085 [2] 0.31-0.43 1.29 0.433 [4] 0.65-1.60 1.11 0.591 [10] 0.31-2.10 Ligule width, m m - - m e g a s p o r o h y l l s Mean 1.75 SD [N] 0.332 [4] Range 1.30-2.00 0.94 0.092 [2] 0.87-1.00 1.98 0.287 [4] 1.60-2.30 1.68 0.479 [10] 0.87-2.30 Ligule L/W ratio--megasporophylls Mean 0,73 SD [N] 0.353 [4] Range 0.44-1.24 0.40 0.130 [2] 0.31-0.49 0,64 0.167 [4] 0.41-0.80 0.63 0.262 [10[ 0.31-1.24 Ligule length, mm--microsporophylls Mean NA SD [N] NA Range NA 1.13 0.397 [4] 0.69-1.60 1.85 0.387 [4] 1.40-2.30 1.49 0.526 [8] 0.69-2.30 Ligule width, mm--microsporophylls Mean NA SD [N] NA Range NA 1.10 0.241 [4] 0.81-1.30 1.73 0.613 [4] 1.10-2.50 1.41 0.545 [8] 0.81-2.50 Ligule L/W ratio----microsporophylls Mean NA SD [N] NA Range NA 1.03 0.246 [4] 0.69-1.23 1.12 0.160 [4] 0.92-1,27 1.07 0.198 [8] 0.69-1.27 Sporangial wall cells are uniformly thin and. the cells generally lack pigment. Occasionally cells with dark cellular contents were noted. These were observed in only one out of nine sporangial mounts. Epidermal cells of megasporangia are elongate (Fig. 2 L - N ; Table VIII). Cells in a similar position on microsporangia are more isodiametric (Fig. 2 0 - Q ; Table VIII). Cell length-to-width ratios were found to be significantly different between mega- and microsporangia (p < 0.001). Velum coverage, measured as percent vertical coverage of the sporangium, ranged from 0 to 59% with significant differences between mega- and microsporophylls (p < 0.001). However, as indicated by the high standard deviations, the ranges for the two types of sporophylls overlap broadly (Table VII). Mean velum length also resulted in a significant difference between mega- and microsporophylls (p = 0.004). Megaspores of Isoetes tennesseensis were reticulate distally with occasional regions of open muri that approached a cristate condition. Proximally, megaspores 2005] B U D K E E T AL.: A N A L Y S I S O F I S O E T E S 175 TABLE VI LABIUM CHARACTERS FOR Isoetes lenrlessee?lsis. POPULATION NUMBERS REFER TO THOSE LISTED IN TABLE I Characler Pop. 1 Pop. 2 Pop. 3 Pop. 1-3 combined L a b i u m length, m m Mean SD [N] Range 1.13 0.473 [11 ] 0 . 3 6 - I. 90 1.49 0.467 [8] 0.74 - 2 . 3 0 1.83 0.345 [8] 1.40 - 2 . 4 0 1.44 0.514 [27] 0.36-2.40 L a b i u m width, m m Mean SD [N] Range 0.40 0.229 [11] 0.12-0.77 0.42 0.220 [8] 0.23-0.83 0.63 0.294 [8] 0.32-1.20 0.47 0.259 [27] 0.12-1.20 L a b i u m L / W ratio Mean SD IN[ Range 3.3 1.63 [11] 1.8-6.8 4.1 1.65 [8] 2.3-7.2 3.6 1.92 [8] 1.5-7.5 3.7 1.69 [27] 1.5-7.5 L a b i u m length, m m - - m e g a s p o r o p h y l l s Mean 1.27 SD [N] 0.565 [3] Range 0.81-1.90 1.16 0.314 [4] 0.74-1.40 1.92 0.354 [6] 1.40-2.40 1.54 0.517 [13] 0.74-2.40 L a b i u m width, m m - - m e g a s p o r o p h y l l s Mean 0.49 SD [N] 0.325 [3] Range 0.12-0.74 0.31 0.105 [4] 0.23-0.46 0.70 0.304 [6] 0.32-1.20 0.53 0.302 [13] 0.12-1.20 Labium L/W ratio--megasporophylls Mean 3.7 SD [N] 2.65 [3] Range 1.8-6.8 4.0 1.19 [4] 2.8-5.2 3.4 2.24 [6] 1.5-7.5 3.7 1.92 [13] 1.5-7.5 L a b i u m length, m m - - m i c r o s p o r o p h y l l s Mean NA SD [N] NA Range NA 1.87 0.404 [3] 1.50-2.30 1.55 0.071 [2] 1.50-1.60 1.74 0.336 [5] 1.50-2.30 L a b i u m width, m m - - m i c r o s p o r o p h y l l s Mean NA SD [N] NA Range NA 0.58 0.297 [3] 0.25~0.83 0.40 0.021 [2] 0.38~0.41 0.50 0.233 [5] 0.25-0.83 Labium L/W ratio--microsporophylls Mean NA SD [N] NA Range NA 4.1 2.70 [3] 2.3-7.2 3.9 0.39[2] 3.7-4.2 4.0 1.92 [5] 2.3-7.2 were echinate to echinate-cristate. Also, there was a narrow girdle in which the reticulations of the distal hemisphere gradually diminished to f o r m weakly developed papillae. Proximal and equatorial ridges were broad, bold, and generally devoid of o r n a m e n t a t i o n (e.g., L u e b k e & B u d k e , 2003). Megaspores ranged in diam. f r o m 625 Izm to 1025 ixm, with a mean value o f 801 Ixm. Microspores averaged 42.36 tzm in length and were consistently laevigate (Table VII). Discussion Based on our analyses of both morphological and anatomical characters of Isoetes tennesseensis, it was determined that the characters can be characterized as: stable, variable, or dimorphic. Stable characters showed relatively little variation, whereas variable characters were defined as those whose m o r p h o l o g y ranged widely f r o m the mean. Dimorphic characters were defined as statistically significant differences be- 176 BRITTONIA [VOL. TABLE VII SPORANGIUM, VELUM, AND SPORE CHARACTERSFOR Isoetes tennesseensis. POPULATION NUMBERS REFER TO THOSE LISTED IN TABLE I Pop. 2 Pop 3 Pop. 1-3 combined Sporangium length, m m Mean 3.9 SD [N] 1.16 [5] Range 2.5-5.2 Character Pop. 1 3.6 1.66 [17] 1.1-6.7 5.3 1.75 [16] 2.7-8.5 4.3 1.81 [38] 1.1-8.5 S p o r a n g i u m width, m m Mean 3.3 SD [N] 0.92 [5] Range 2.0-4.3 2.9 1.22 [171 1.2-5.4 4.0 1.00 [16] 2.4-5.7 3.4 1.20 [38] 1.2-5.7 Sporangium L / W ratio Mean 1.20 SD [NI 0.216 [51 Range 0.89-1.50 1.26 0.190 [17] 0.79-1.59 1.32 0.198 [16] 0.93-1.74 1.28 0.195 [38] 0.89-1.74 M e g a s p o r a n g m m length, m m Mean 3.9 SD [N] 1.16 [5] Range 2.5-5.2 5.2 1.60 [4] 3.4-6.7 5.9 1.75 [10] 3.6-8.5 5.2 1.71 [19] 2.5-8.5 M e g a s p o r a n g m m width, m m Mean 3.3 SD [N] 0.92 [5] Range 2.0-4.3 4.5 0.96 [4] 3.5-5.4 4.4 0.86 [10] 3.3-5.7 4.1 0.98 [19] 2.0-5.7 M e g a s p o r a n g m m L/W ratio Mean 1.12 SD [N] 0.216 [5] Range 0.89-1.50 1.13 0.132 [4] 0.97-1.29 1,32 0.204 [10] 1.00-1.74 1.25 0.200 [19] 0.89-1.74 M i c r o s p o r a n g m m length, m m Mean NA SD [N] NA Range NA 3.3 1.39 [13] 1.1-5.6 4.4 1.45 [6] 2.7-6.3 3.6 1.47 [19] 1.1-6.3 M i c r o s p o r a n g m m width, m m Mean NA SD [N] NA Range NA 2.5 0.82 [13] 1.2-4.1 3.3 0.92 [6] 2.4-4.6 2.8 0.92 [19] 1.2-4.6 M i c r o s p o r a n g l u m L / W ratio Mean NA SD [N] NA Range NA 1.28 0.232 [13] 0.79-1.59 1.32 0.205 [6] 0.93-1.55 1.32 0.187 [19] 0.93-1.59 % Velum c o v e r a g e - - m e g a s p o r o p h y l l s Mean 7.8 SD [N] 9.96 ]5] Range 0-24 22.3 10.2 [4] 12-36 14.0 8.9 [9] 4-24 14.1 10.3 [18] 0-36 % Velum c o v e r a g e - - m i c r o s p o r o p h y l l s Mean NA SD ]N] NA Range NA 39.5 10.2 [11] 26-59 28,9 13.2 [7] 13-50 35.4 12.3 [18] 13-59 Velum l e n g t h - - m e g a s p o r o p h y l l s , m m Mean 0.39 SD [N] 0.52 [5] Range 0.0-1.25 1.11 0.49 [41 0.61-1.54 0.83 0.63 [9] 0.21-1.87 0.77 0.60 [18] 0.0-1.87 Velum l e n g t h - - m i c r o s p o r o p h y l l s , m m Mean NA SD [N] NA Range NA 1.34 0.37 [11] 0.47 1.74 1.44 0.84 [7] 0.35-2.39 1.38 0.58 [18] 0.35-2.39 57 2005] B U D K E ET AL.: ANALYSIS OF ISOETES 177 TABLE VII CONTINUED Character Pop. 1 Megaspore diameter, p,m Mean 783 SD [N] 67.0 [20] Range 650-875 Microspore length, p,m Mean SD [N] Range NA NA NA Pop. 2 Pop. 3 Pop. ! 3 combined 799 95.7 [30] 625-1025 815 95.0 [30] 625-1025 801 89.9 [80] 625-1025 41.43 2.12 [100] 37.49-45.65 44.22 2.57 [50] 39.12-48.90 42.36 2.63 [150] 37.49-48.90 tween mega- and microsporophyll associated characters. STABLE CHARACTERS C o r m lobing has had a long history of use in Isoetes taxonomy. It was considered by early workers (Braun, 1864; Weber, 1922) to be a reliable and consistent character and even recent workers (Kott & Britton, 1985) have suggested that it is generally stable, although of reduced taxonomic utility in northeastern North America because of its interspecific uniformity. Nonetheless, the genus does show considerable variation and it is important to note the stability of this character in L tennesseensis. Lacunar wall thickness, uniform in Isoetes tennesseensis, is potentially correlated with habitat and the presence or absence of stoma. In species with stoma, walls vary in thickness from several cells thick down to just the epidermis in regions adjacent to the guard cells. In species lacking stoma, wall thickness appears to be consistent within species but m a y v a r y a m o n g taxa. Isoetes savatieri Franchet, for example, has robust lacunar walls, which add significantly to the rigidity of the leaf. Variation in lacunar wall thickness within species appears to be rare, but has been reported for L wormaldii Sim (Duthie, 1929). Hall (1971) and T a k a m i y a et al. (1997) showed that veinal (intrastelar) canal number varied a m o n g Isoetes species and can serve as a good taxonomic character. These canals have long been a source of interest to morphologists due to the repeated reports of isolated secondary wall thickenings located within the canals. These thickenings have been interpreted to be remnants of protoxylem tracheids and the canals have often been c o m p a r e d to p r o t o x y l e m lacunae of other taxa (e.g., Equisetum; Bierhorst, 1958, 1971). Recently, R o m e o et al. (2000) showed that the cells surrounding these lacunae are endodermal in nature and have a well-developed casparian strip, in accord with earlier observations of Scott and Hill (1900) and Williams (1943). In L tennesseensis and in species with multiple veinal canals, the canals are usually each bound by a distinct endodermal layer, which excludes the majority of tracheary elements of the leaf. One published exception appears to be L sinensis T. C. Palmer var. sinensis in which adjacent intrastelar canals share a short c o m m o n endodermal region. West and Takeda (1914) c o m m e n t e d on the cells surrounding these canals and suggested that they do not represent a true endodermis and coined the term pseudo-endodermis in light of the anomalous position and f o r m of the cells. Our data show that veinal canal number is stable within L tennesseensis, showing only rare variation in number (Fig. 1A, B; Table IV). The condition of this character in other North American species is unknown. Historically, little attention has been paid to the nature and d e v e l o p m e n t of the diaphragms of the lacunae, other than to note their presence. Hall (1971) described these translacunar diaphragms as being, " t w o or three cells in thickness, perforated by pores which are triangular, or circular, to elliptical." In Isoetes tennesseensis, the air spaces between the diaphragm cells are typically triangular in surface view (Fig. 1C, D). Ad- 178 BRITTONIA [VOL. 57 TABLE VIII SPORANGIAL WALL CELL CHARACTERSFOR Isoetes tennesseensis. POPULATION NUMBERS REFER TO THOSE LISTED IN TABLE I Character Pop. 1 Pop, 2 Pop. 3 Pop. 1-3 combined 138.8 47.25 [12] 65.0-240.0 117.5 26.67 [I2] 75.0-160.0 128.1 39.06 [24] 65.0-24,0 27.9 7.22 [12] 20.0-40.0 28.8 9.08 [12] 20,0-45.0 28.3 8.03 [24] 20.0-45.0 5.42 2.48 [12] 1.86-9.25 4.57 1.86 [24] 2.11-6.75 5.00 2.18 [24] 1.86-9.25 56.3 13.51 [12] 40.0-85.0 72.5 16.86 [12] 40.0-95.0 64.4 17.09 [24] 40.0-95.0 M i c r o s p o r a n g m m cell width, p,m Mean NA SD [Y] NA Range NA 35.4 6.89 [12] 25.0-45.0 28,3 9.13 [12] 15.0-45.0 31.9 8.70 [24] 15.0-45.0 M i c r o s p o r a n g i u m cell L / W ratio Mean NA SD IN] NA Range NA 1.63 0.42 [12] 1,00-2.40 2.91 1.47 [12] 1.38-6,33 2.27 1.25 [24] 1.00-6.33 M e g a s p o r a n g m m cell length, ~ m Mean NA NA SD IN] Range NA Megasporangmm cell width, ixm Mean NA NA SD [NI Range NA Megasporangmm cell L/W ratio Mean NA NA SD [NI Range NA Microsporangmrn cell length, Ixm Mean NA SD [N] NA Range NA ditionally, the diaphragms of L tennesseensis are c o m p o s e d of two to three layers of stellate p a r e n c h y m a cells. The cells of these layers are distinctly stacked one upon the other in vertical rows (Fig. 1E). This arr a n g e m e n t of diaphragm cells has not been noted in other species, perhaps because m o s t analyses of the diaphragms have concentrated on transverse rather than longitudinal views. The surface of the diaphragm ceils have been observed as having spines (protuberances) on the surface in African species (L abyssinica, and L nigritiana A. Br. ex Kuhn) by Hall (1971) and in Australian species (L coromandelina, L drummondii A. Br., and I. muelleri A. Br.) by Marsden (1979). lsoetes tennesseensis has numerous protuberances across the surfaces of the stellate cells of the diaphragms, but these are enlarged and m u c h more noticeable near the juncture o f adjacent arms (Fig. 1D). At that point the protuberances are arranged in two encircling whorls, one whorl per arm. Marsden (1979) also noted that these were well developed near the junctures of the steUate arms of the diaphragm cells in Australian species. Additionally these protuberances have been found in several Isoetes species of the Iberian Peninsula, including the circumboreal L lacustris (Prada, 1979; Prada & Rolleri, 2003). Hall (1971) used epidermal papillae, internal (lacunar) hairs, and subepidermal tubules to differentiate a m o n g Isoetes species in Ghana. Internal hairs not associated with the diaphragm cells but projecting into the lacunae have been noted in the African L abyssinica Chiovenda and I. melanotheca Alston (Pitot, 1959) and in the Indian L coromandelina L.f., L indica Pant and Srivastava, and L pantii G o s w a m i and A r y a (Bhambie, 1963). Epidermal papillae are k n o w n f r o m I. tenuifolia A. C. J e r m y (Hall, 1971), and these as well as epidermal striations of questionable similarity have been found in some Australasian species (Marsden, 1979). Epidermal papillae, internal 2005] B U D K E ET AL.: A N A L Y S I S OF ISOETES hairs, and subepidermal tubules are lacking in L tennesseensis. A sporangial character mentioned frequently in the Isoetes literature is pigmentation. Unfortunately, few authors have discriminated among the characters that are routinely subsumed under this general category. Specifically, there appear to be two potential sources o f sporangial coloration: cell wall pigmentation and lumen pigmentation. In the former, the walls of the exposed (not covered by vela) epidermal cells may be discolored by a non-lignin pigment (Hall, 1971). A second form of pigmentation appears to be due to discoloration within the lumen itself. Although this has not been definitely shown to be vacuolar, it would appear to be so. The limited spotting seen in L tennesseensis appears to be of this latter type. In addition to pigmentation, sporangial epidermal cells may show variation in cell wall thickness. The walls in L tennesseensis are uniformly thin. Previous descriptions o f s p o r a n g i a p i g m e n t a t i o n in American Isoetes have not been adequately discussed relative to pigment position or wall thickness to make comparisons with other species. Megaspore surface morphology is uniform among populations and among individuals within p o p u l a t i o n s . I n d i v i d u a l spores show variation in surface ornamentation between proximal and distal surfaces; a phenomenon that is c o m m o n in Isoetes. The species is identifiable on the basis of megaspore morphology depending on the experience of the worker. The consistently laevigate microspores are characteristic of most species with predominately reticulate megaspores. VARIABLE CHARACTERS The velum is of interest in this species due to its complex structure. In Isoetes tennesseensis, it is composed of velum proper, sella and, in some cases, labium tissue (Fig. 1E G). No mention o f similar variation appears in the literature, although the origin of the labium from an apparently proximal side of the sella in L nigritiana is of interest and deserves additional study. Complex vela of different forms have been reported. 179 Palmer (1932), for example, describes a complex velum in L leehleri Mett. var. anomala Palmer (=L karstenii A. Br. var. anomala (Palmer) Small & Hickey) composed of a thickened outer layer and an inner delicate layer. Palmer notes that these tw0 layers were "not always coincident." Examinations in our laboratory of the type material (Rimbach 171, US) o f this taxon have failed to find a true double velum. We suspect that what Palmer noted was a separation of the inner epidermis of the velum from the outer epidermis and associated internal parenchyma (see Fig. 1F). Because of textural differences in these layers, it is not surprising that they would have undergone d i f f e r e n t i a l s h r i n k i n g during s p e c i m e n preparation. In longitudinal sections of various species of Isoetes, especially those with extensive velum coverage, we have seen separation of these layers. Finally, Hall (1971) describes velum pleats on the surface of the nearly complete velum in L tenuifolia. These pleats appear to be verticallyelongate outgrowths of the velum, and although they do not appear to be artifacts of preservation, no indication is given about the frequency of their occurrence. Velum pleats were not observed in L tennesseensis. The morphology of the labium varies within Isoetes tennesseensis. Some leaves lack labia, some have just a short ridge, but most leaves have a spatulate labium, often with bifid apices. No dimorphism in labium morphology exists between the mega- and microsporophylls and the full extent of variation in this character can be found within a single plant. In general, North American Isoetes have small (depressed triangular) labia and hence L tennesseensis stands out among these species with its unique labium morphology (Fig. 2A-K). Only I. weberi Herter of southeastern Brazil shows a similar morphology (Hickey, 1985). The spatulate, bifid labium is quite extraordinary in Isoetes. The high degree of variation in this character exhibited by L tennesseensis may be related to a mixing of genomes in this high polyploid species (Luebke & Budke, 2003). Luebke and Budke (2003) described the leaves of Isoetes tennesseensis as terete in cross section. Our analyses, however, sug- 180 BRITTONIA gest that cross sections of the subulae can vary f r o m half-terete to weakly trapezoidal to terete, with the half-terete shape being the m o s t c o m m o n . This shape is dictated by the strong a s y m m e t r y of lacunae as seen in sectional view (Fig. 1A). The relative sizes of these adaxial and abaxial lacunae have been used by Takamiya et al. (1997) to differentiate between two varieties of L sinensis in Japan. L e a f cross-sectional shape has also been used as a key character for identifying and differentiating a n u m b e r of neotropical (Hickey, 1994; Small & Hickey, 2001) and Papuasian (Croft, 1980) species. Lacunae characterizations of other North A m e r i c a n species have not been previously published. However, ongoing w o r k in our laboratory suggests that lacunae and subula cross-sectional shape will be of diagnostic value. For example, species such as L butleri Engelm. have distinctly triquetrous leaves that can immediately be recognized in the field tactilely. In these species the abaxial lacunae are triangular in outline whereas the adaxial lacunae are quadrate. DIMORPHIC CHARACTERS Megasporangia and microsporangia of I s o e t e s t e n n e s s e e n s i s are dimorphic in size, yet maintain a consistent length to width ratio. To our knowledge, the only other indication of such d i m o r p h i s m is in L australis S. Williams, in which megasporangia are 0.75-1.25 X 0.5-0.75 m m whereas microsporangia are 1-1.5 X 0.5-0.6 m m (Williams, 1943). However, it is not known if these are significantly different. Kott and Britton (1985) showed considerable overlap in sporangium size for eight taxa in northeastern North America. High infraspecific variation and size overlap a m o n g species led them to conclude that sporangium size was not a useful taxonomic character. For similar reasons, very few I s o e t e s taxonom i s t s h a v e u s e d this c h a r a c t e r . A l s t o n (1959) is perhaps the last to use it with any real confidence. The presence of a statistically significant bimodality in sporangium size in a single species, however, suggests that sporangium size should not be i m m e diately ruled out as a potential character. If previous studies c o m p a r e d variously mixed [VOL. 57 samples o f mega- and microsporangia, then significant interspecific differences may have been overlooked. However, we recognize that sporangia in I s o e t e s are particularly difficult characters to utilize taxonomically due to ontogenetic and developmental variation. Our unpublished studies have shown that megaspore number and m e g a s p o r a n g i u m size in I. a n d i c o l a (Amstutz) L. D. G 6 m e z show a large amount of ontogenetic variation. Young plants produce as few as four spores per sporangium, and with age this number increases to eight, twelve, and higher. Sporangia increase in size accordingly. Size d i m o r p h i s m in sporangial wall cells was also noticed while examining cell shape. Wall cells of the megasporangia average 4.99X longer than wide, whereas the wall cells of the microsporangia are only 2.27• longer than wide. This within-plant variation is reminiscent of the interspecific variation documented by Hall (1971): sporangial wall cells of I s o e t e s tenuifolia are depicted as being 6 - 1 0 • longer than wide whereas those in I. a b y s s i n i c a are only 3 5X longer. Hall (1971) did not indicate f r o m which types of sporangia he obtained his data, nor if they were consistent across specimens and sporangia types. Critical analyses of cellular patterns o f the sporangium epidermis in other species m a y lead to insights about relationships and provide additional taxonomic characters. D i m o r p h i s m was also identified in v e l u m c o v e r a g e b e t w e e n m i c r o s p o r a n g i a and megasporangia. Initially, we ascribed the higher v e l u m coverage to incomplete enlargement of sporangia on inner microsporophylls; a strong case for not using percent v e l u m coverage of .sporangia for just such a reason was made by Kott and Britton (1985). U p o n closer examination, however, it was found that the variation in percent v e l u m coverage in I s o e t e s t e n n e s s e e n s i s reflects differences in v e l u m length. Our data s h o w that m i c r o s p o r a n g i a (with m e a n lengths of 3.62 m m and m e a n percent velum coverage of 35.4%) had a mean v e l u m length of 1.2 m m , whereas the older megasporangia (with a mean length of 5.20 m m and m e a n velum coverage of 14.1%) had a m e a n v e l u m length of 0.7 m m . Thus, not 2005] BUDKE ET AL.: ANALYSIS OF ISOETES only are there significant differences in velum coverage, but in overall velum length as well. Kott and Britton's (1985) argument was based on the premise that velum length was constant and that variation in percent cover was a reflection of differential realized growth of the sporangia. This is not the case in L tennesseensis. The ligule is an ephemeral structure that when young outpaces the growth of the young leaf (Smith, 1900), but with age, the ligule becomes dwarfed by the elongating leaf. In Isoetes tennesseensis shape differences in ligules were found between those of the micro- and megasporophylls. Ligules of the megasporophylls tended to be depressed-ovate whereas those of the microsporophylls were deltate. We suspect that these differences are artifacts associated with greater ligule degradation in the older megasporophyll leaves. The ligules of the microsporophylls are likely more representative of nearly entire ligules. Hence, this does not represent a true dimorphism. Ligule form in L tennesseensis needs further study in order to determine if this dimorphism is solely degradation related. D i m o r p h i s m in c h a r a c t e r s a s s o c i a t e d w i t h m e g a s p o r o p h y l l s and m i c r o s p o r o phylls has rarely been addressed in the Isoetes literature. To the best of our knowledge, other than spore differences, the only references to such p h e n o m e n a are those of Hickey (1985) with regard to sporangial pigmentation. In that w o r k it was noted that L gardneriana Mett. and L triangula U. Weber have black megasporangial walls and pale tan, diaphanous microsporangial walls. Additional observations on t h e nature of this sporangial pigmentation are wanting. Our observations of character dim o r p h i s m in sporangial size, velum coverage, and cell wall pattern, suggest that these characters need to be reevaluated on a case by case basis before they are discarded from taxonomic w o r k as suggested by s o m e r e s e a r c h e r s (e.g., M a t t h e w s & M u r d y , 1969). Conclusions A wide variety of morphological and anatomical characters have been examined in 181 our study. Unfortunately, we have not been able to address all aspects of the internal anatomy, specifically in association with the c o r m and roots. These two structures m a y provide additional sources o f characters and show unknown variation in Isoetes. Though previous taxonomic research on North American Isoetes has de-emphasized morphological characters, we have found that these characters are potentially significant. Other researchers have concluded that vegetative characters are environmentally plastic, developmentally dependent, or invariant (Matthews & Murdy, 1969; Kott & Britton, 1985). We concur that there is a sizeable amount of variability in some characters. However, we find that such variation can be quantified and can have biological and taxonomic importance. For example, even when absolute character measurements m a y not be discriminating, ratio data m a y show distinctive developmental patterns (Small & Hickey, 2001) or the degree of variation present m a y be statistically distinct a m o n g species (Hickey, 1978). Furthermore, the dimorphic patterns observed here suggest an additional source o f information that m a y have been overlooked in previous analyses. This study establishes a baseline for morphological comparisons with other Isoetes. We hope that these data will encourage researchers to examine the morphological and anatomical characters of Isoetes species not only in North America, but also worldwide. Such work will c o m p l e m e n t and expand our knowledge o f previously published species. This should result not only in a thorough evaluation of the potential taxonomic value of morphological characters, but also in a better understanding o f morphological evolution in Isoetes. Acknowledgments We would like to thank Pat Cox, Karl Gustafson, Ed Lickey, and Joey Shaw for collection assistance, Paul Russell for assistance with SEM, and Michael Barker for assistance with SEM and figure composition. We also acknowledge funding f r o m the Undergraduate S u m m e r Scholars Pro- 182 BRITTONIA gram at Miami University and the Willard S. Turrell Herbarium (MU). L i t e r a t u r e Cited Alston, A. H. G. 1959. The ferns and fern-allies o f west tropical Africa: supplement to flora of west tropical Africa. Ed. 2. Crown Agents for Oversea G o v e r n m e n t s and Administrations, London. Bhambie, S. 1963. Studies in pteridophytes. III. On the structure and development o f leaf and sporophyll of lsoetes coromandelina. Proc. Natn. Inst. Sci. India 29B: 169-190. Bierhorst, D. W . 1958. Vessels in Equisetum~ Amer. J. Bot, 45: 5 3 4 - 5 3 7 . 9 1971. Morphology of vascular plants. Macmillan, N e w York. Boom, B. M . 1979. Systematic studies o f the g e n u s lsoe-tes in the southern United States. M.S. thesis. The University of Tennessee, Knoxville. Braun, A. 1864. Les esp6ces d'Iso~tes de File Sardaigne. Ann. Sci. Nat. Bot. 5: 306-377. Croft, J. R. 1980. A taxonomic revision of lsoetes L. (Isoetaceae) in Papuasia. B l u m e a 26: 177-190. D u t h i e , A. V. 1929, The species o f lsoetes found in the union o f South Africa. Trans. Roy. Soc. S. Africa 17: 321-334. Hall, J. B. 1971. Observations on Isoetes in Ghana. Bot. J. Linn. Soc. 64: 117-139. I-Iiekey, R. J. 1978. Variability in the Lycopodium obscurum complex of North A m e r i c a and eastern Asia. M.S. thesis. Miami University, Oxford, Ohio. - - . 1985. Revisionary studies o f neotropical Isoetes. Ph.D. dissertation. T h e University of Connecticut, Storrs. 9 1986a. lso~tes megaspore surface morphology: nomenclature, variation, and systematic importance. Amer. Fern J. 76: 1-16. 9 1986b. The early evolutionary and morphological diversity of Iso~tes, with description of two neotropical species. Sys. Bot. 11: 3 0 9 - 3 2 1 . 9 1986c. On the identity o f lsoO'tes triquetra A. Braun. Taxon 35: 243-246. 9 1994. Isoetaceae. Pages 88-97. In: R. M. Tryon & R. G. Stolze, editors. Flora of Peru. Fieldiana Bot., n.s. 34: 1-123. , C. M a c l u f & W. C. T a y l o r . 2003. A Reevaluation o f Isoetes savatieri Franchet in Argentina and Chile. Amer. Fern. J. 93: 126-136. Johansen, D. A. 1940. Plant microtechnique. McGr.aw-Hill Book Co., Inc., N e w York. Kott, L. S. & D. M. Britton. 1985. Role of morphological characteristics of leaves and the sporangial region in the taxonomy of Isoetes in northeastern North America. Amer. Fern J. 75: 4 4 - 5 5 . L u e b k e , N. T. & J. M. B u d k e . 2003. Iso6tes tennesseensis (Is6etaceae), an octaploid quillwort from Tennessee. Amer. Fern J. 93: 184-190. Marsden, C. R. 1979. Morphology and t a x o n o m y of lsoetes in Australia, India, north-east and south-east Asia, C h i n a and Japan. Ph.D. dissertation. University of Adelaide, Adelaide. Matthews, J. F. & W . H. M u r d y . 1969. A study o f lsoetes c o m m o n to the granite outcrops of the [VOL. 57 southeastern Piedmont, United States. Bot. Gaz. 130: 5 3 - 6 1 . Minitab Inc. 2000. MINITAB Statistical Software Release 13.32 for Windows. Palmer, T. C. 1932. More about Lyoetes Lechleri Mett. Amer. Fern J. 22: 129-I32. Pfeiffer, N. E. 1922. Monograph of the Isoetaceae. Ann. Missouri Bot. Gard. 9: 7 9 - 2 3 2 . Pitot, A. 1959. Contribution ~t l'6tude des lsok'tes africains: Iso6tes melanotheca Alston. Bull. Inst. Fr. Afr. Noire, s6r. A 21: 9 0 0 - 9 2 0 . 91968. U n e nouvelle esp~ce africaine d'lso~tes: Isor jaegeri Pitot. Bull. Inst. fr. Afr. noire, s6r. A 30: 3 9 - 5 8 . Prada, C. 1979. Estudio de la anatomfa foliar de las especies espafiolas del g6nero lsoetes L. Lagascalia 9: 107-113. P r a d a , C. & C. H. Rolleri. 2003. Caracteres diagn6sticos foliates en t~ixones ib6ricos de lsoetes L. (Isoetaceae, Pteridophyta). Anales Jard. Bot. Madrid. 60: 371-386. R a u h , W . & H. Falk. 1959a. Stylites E. Amstutz, eine neue IsoOtaceae aus den h o c h a n d e n Perus. 1. Teil: Morphologie, Anatomic und Entwicklungsgeschichte der Vegetationsorgane. Sitzsungsber. Heidelberger Akad. Wiss, Math.-Naturwiss. K1. 1: 3-83. - & . 195919. Stylites E. Amstutz, eine neue Iso~taceae aus den h o c h a n d e n Perus. 2. Teil: Zur Anatomic des S t a m m e n mit besonderer Berficksichtigung der V e r d i c k u n g s p r o z e s s e . Sitzsungsber. Heidelberger Akad. Wiss. Math.-Naturwiss. K1. 2: 3-76. Reed, C. F. 1965. Isoetes in southeastern United States. Phytologia 12: 3 6 9 - 4 0 0 . Romeo, D., A. Trola, C. Burgarella & E. Bellini. 2000. Casparian strips in the leaf intrastelar canals o f Isoetes duriei Bory, a Mediterranean terrestrial species. Ann. Bot. 86: 1051-1054. Scott, D. H. & T. G. Hill. 1900. The structure of Isoetes hystrix. Ann. Bot. 14: 4 1 3 - 4 5 4 . S m a l l , R. L. & R. J. Hiekey. 2001. Systematics of the northern Andean Iso6tes karstenii complex. Amer. Fern J. 91: 4 1 - 6 9 . Smith, R. W . 1900. The structure and development of the sporophylls and sporangia o f lsoetes. Bot. Gaz. 29: 225-346. T a k a m i y a , M . , M. W a t a n a b e & K. O n o . 1997. Biosystematic studies on the genus lsoetes (Isoetaceae) in Japan. IV. Morphology and a n a t o m y of sporophytes, phytogeography and taxonomy. Acta Phytotax, et Geobot. 48:-89-122. T a y l o r , W . C., N. T. L u e b k e , D. M. B r i t t o n , R. J. Hickey & D. F. B r u n t o n . 1993. Iso~taceae Reichenbach, Quillwort family. Pages 6 4 - 7 5 . In: Flora of North America Editorial Committee, editors. Flora of North America. Vol. 2. Oxford University Press, New York. W e b e r , U. 1922. Zur anatomic u n d systematik der gattung Isoetes L. Nova Hedwigia 63: 2 1 9 - 2 6 2 . West, C. & H. T a k e d a . 1914. X. O n Isod'tesjaponica A. Br. Trans. Linn. Soc. Ser. 2, Bot. 8" 331-376. Willey, R. L. 1971. Microtechnique, a laboratory guide. Macmillian, New York. Williams, S. 1943. On lsoetes australis S. Williams, a n e w species from western Australia. Proc. Roy. Soc. Edinburgh. Sect B 1: 1-8.