HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (6) Citing Articles via Google Scholar Google Scholar Articles by Cooper, R. L. Articles by Philbrick, C. T. Search for Related Content PubMed PubMed Citation Articles by Cooper, R. L. Articles by Philbrick, C. T. Agricola Articles by Cooper, R. L. Articles by Philbrick, C. T.

Comparative pollen morphology and ultrastructure of the Callitrichaceae¹

¹ Division of Science, Truman State University, Kirksville, Missouri 63501 USA; and ² Department of Biological and Environmental Sciences, Western Connecticut State University, Danbury, Connecticut 06810 USA

Received for publication June 5, 1998. Accepted for publication May 27, 1999.

ABSTRACT

The Callitrichaceae are an aquatic family of dicots that include the single, geographically cosmopolitan genus Callitriche. Callitriche contains 40–50 terrestrial, amphibious, and obligately submersed species, and it is the only known genus in the plant kingdom with co-occurring aerial and hydrophilous pollination syndromes. Pollen morphology and ultrastructure were described for 13 Callitriche species using scanning electron and transmission electron microscopy. Representative taxa of each growth form were examined; these included three terrestrial species (C. deflexa, C. peploides, and C. nuttallii), nine amphibious species (C. brutia, C. cophocarpa, C. cophocarpa-stagnalis hybrid, C. cribrosa, C. hamulata, C. heterophylla var. heterophylla, C. lusitanica, C. marginata, and C. trochlearis), and one obligately submersed species (C. truncata). Of the amphibious taxa, C. heterophylla var. heterophylla and C. trochlearis had internal geitonogamy, a type of internal self-fertilization. Pollen from all taxa was spheroidal, small, intectate, and lacked well-defined apertures. Taxa primarily differed with respect to exine thickness, surface ornamentation, and the presence or absence of aperture-like regions. The pollen of terrestrial species, as well as that of C. marginata, had well-developed exines with thick sculptured and basal layers. In general, amphibious taxa produced pollen with distinct, but thinner, exines than that of terrestrial taxa. Pollen of the amphibious taxa with internal geitonogamy had a thicker basal layer than species without internal geitonogamy, whereas the overall exine was reduced in C. hamulata and absent in C. brutia and C. lusitanica. Pollen of the obligately submersed C. truncata also lacked an exine. These palynological data were correlated with growth habits and related pollination biologies, as well as with phylogenetic interpretations of Callitrichaceae. Exine reduction or loss has evolved at least twice in the family, and it is associated with aneuploid reduction in chromosome number.

Key Words: aquatic plants • Callitrichaceae • Callitriche • hydrophily • morphology • pollen • pollination • ultrastructure

The Callitrichaceae (water starworts) are an aquatic family of dicots that include the single genus Callitriche. Callitriche contains 40–50 terrestrial, amphibious, and obligately submersed species (Philbrick and Jansen, 1991 ). Plants of all species of Callitriche are small, herbaceous, and limp-stemmed. Terrestrial species occur in seasonally wet areas, have prostrate stems that often grow in dense mats, and produce aerial flowers. Plants of amphibious species can grow either terrestrially, with submersed stems that reach to the water surface and produce floating rosettes of leaves, or as completely submersed stems that lack floating leaves. Several amphibious species lack the terrestrial growth form. Some amphibious species can produce flowers on both aerial and submersed stems, whereas others produce only aerial flowers. Obligately submersed species grow and flower entirely underwater (Philbrick and Osborn, 1994 ).

Species of Callitriche are monoecious and have reduced flowers that lack petals and sepals (Philbrick and Anderson, 1992 ). Pollination systems are diverse in Callitriche (Philbrick and Anderson, 1992 , and references therein), although the details of pollination remain unknown for most species. Both aerial (anemophilous) and aquatic (hydrophilous) pollination occur, and the type of pollination is often associated with growth habit. Anemophily is the most likely pollination system of both terrestrial species and terrestrial growth forms of amphibious species. Two types of water pollination have been reported in the genus: epihydrophily (pollination at the water surface) and hypohydrophily (pollination below the water surface). Epihydrophily has been reported for various amphibious species, although supporting evidence is usually anecdotal (see discussion in Philbrick and Anderson, 1992 ). Hypohydrophily occurs in the obligately submersed species, C. hermaphroditica L. (Philbrick, 1993 ) and C. truncata Gussone.

Geitonogamy (pollination among flowers on one plant) characterizes many species of all three growth habits. Furthermore, seven amphibious species have internal geitonogamy, a type of internal self-fertilization. Here, pollen grains germinate within indehisced anthers and the pollen tubes grow through vegetative tissues to nearby pistillate flowers where fertilization occurs (Philbrick, 1984 ; Philbrick and Anderson, 1992 ; Philbrick and Bernardello, 1992 ). Internal geitonogamy occurs in both aerial and submersed flowers of these species.

Uniformity in overall floral structure among species of Callitriche is interesting given the range of pollination systems that occurs in the genus. Many authors have discussed the high degree of floral diversity that occurs among closely related aerial- and underwater-flowering species in other angiosperm groups (Arber, 1920 ; Sculthorpe, 1967 ; Tomlinson, 1982 ; Dahlgren and Rasmussen, 1983 ; Philbrick, 1991 ). Perhaps the most dramatic difference between aerial-flowering and hydrophilous species pertains to the pollen. In particular, pollen of many hydrophilous taxa has an elongate shape (e.g., Pettitt, 1981 ; Cox, 1988 ) and reduced exine thickness (e.g., Osborn and Philbrick, 1994 ; Philbrick and Les, 1996 ). However, only limited data about pollen structure, especially ultrastructure, are known for Callitriche.

Most information about Callitriche pollen has been gained from studies with light microscopy (e.g., Erdtman, 1952 ; Moar, 1960 ; Díez, Talavera, and Garcia-Murillo, 1988 ) and scanning electron microscopy (SEM; Díez, Talavera, and Garcia-Murillo, 1988 ). Only three published studies used transmission electron microscopy (TEM) to describe pollen wall structure in the family (Martinsson, 1993 ; Osborn and Philbrick, 1994 ; Philbrick and Osborn, 1994 ). These investigations demonstrated that pollen of terrestrial and amphibious taxa of Callitriche had a well-defined exine, whereas that of the obligately submersed species C. hermaphroditica lacked an exine. However, C. hermaphroditica has two growth forms. Exineless pollen characterized the perennial form, whereas a rudimentary exine was present in pollen of the annual form (Osborn and Philbrick, 1994 ). A well-developed, two-layered intine is present in both growth forms.

Despite the observations of Martinsson 1993 , Osborn and Philbrick 1994 , and Philbrick and Osborn 1994 , these authors collectively examined the pollen from only nine taxa, including one terrestrial, seven amphibious, and one obligately submersed species. In the current study, pollen morphology and ultrastructure of 13 Callitriche species were investigated using SEM and TEM. Representative taxa of each growth form were examined, including three terrestrial species (C. deflexa A. Br. ex Hegelm., C. nuttallii Torrey, and C. peploides Nutt.), nine amphibious species (C. brutia Petagna, C. cophocarpa Sendter, C. cophocarpa-stagnalis hybrid, C. cribrosa Schotsman, C. hamulata Kütz., C. heterophylla Pursh var. heterophylla, C. lusitanica Schotsman, C. marginata Torrey, and C. trochlearis Fassett), and one obligately submersed species (C. truncata). Of the nine amphibious taxa, C. heterophylla var. heterophylla and C. trochlearis have internal geitonogamy.

The primary objectives of this paper were to describe pollen structure from the 13 species of Callitriche and to correlate new palynological data with growth habits and related pollination biologies of the taxa. Furthermore, these data, particularly information on exine thickness, were interpreted in a phylogenetic context.

MATERIALS AND METHODS

Plants of the 13 Callitriche species were collected from sites in Europe and North America (Table 1). Two growth forms of the amphibious C. hamulata were collected, those with aerial and those with submersed flowers. Voucher specimens were deposited at the Western Connecticut State University herbarium.

Three to five dehydrated anthers from different plants of each species were gradually infiltrated and embedded in Spurr epoxy resin. To identify appropriate specimens for ultrastructural studies (e.g., those with mature pollen), individual anthers were thick-sectioned on an ultramicrotome with glass or diamond knives. Thick sections (850 nm) were stained with either 10% toluidine blue O or Richardson's stain (50% methylene blue/50% Azure II) and examined on an Olympus BHS Compound Light Microscope.

To study pollen wall ultrastructure, anthers were thin-sectioned with glass or diamond knives. Thin sections (90–100 nm) were collected on 1 x 2 mm copper slot grids and dried onto formvar support films (Rowley and Moran, 1975 ). Grids were stained with 1% potassium permanganate (1 min), 1% uranyl acetate (15 min), and lead citrate (8 min; Venable and Coggeshall, 1965 ) and then examined and imaged using a JEOL JEM-100SX transmission electron microscope at 60–80 kV.

To study pollen surface morphology, dehydrated anthers were critical point dried. To isolate individual pollen grains, dried anthers were macerated using a syringe needle and mounted onto aluminum stubs with double-sided adhesive tape, which had been secured to the stubs with colloidal graphite. Stubs were sputter-coated with gold-palladium, and pollen was examined and imaged using a JEOL JSM-6100 scanning electron microscope at 5 kV.

RESULTS

Palynological data are presented with the individual taxa grouped according to growth habit. Callitriche marginata is grouped with the terrestrial taxa and C. lusitanica is grouped with the obligately submersed taxa, because of phylogenetic interpretations of these two amphibious species (see below; Philbrick and Jansen, 1991 ; Philbrick and Les, 1995 ). The palynological features for each species are described below; these, as well as additional characters, and statistical information are summarized in Table 2.

View larger version (179K): [in this window] [in a new window]   Figs. 1–12. Terrestrial taxa. Callitriche deflexa (Figs. 1–4). 1. Pollen grain showing spheroidal shape, microgemmate surface ornamentation, and three aperture-like regions. Bar = 2 µm. 2. Detail of pollen surface with pyramidal-shaped microgemmae in polygonal patterns. Bar = 1 µm. 3. Transverse section through pollen grain showing thin exinous basal layer in two aperture-like regions (arrows). Bar = 3 µm. 4. Transverse section of pollen wall showing sexinous microgemmae, thick basal layer, electron-dense line (arrow), and intine (I). Bar = 0.5 µm. Callitriche nuttallii (Figs. 5–8). 5. Pollen grain showing spheroidal shape and microgemmate surface ornamentation. Bar = 2 µm. 6. Detail of pollen surface showing pointed microgemmae in polygonal patterns. Bar = 1 µm. 7. Transverse section through pollen grain showing thin exinous basal layer and thicker intine in the three aperture-like regions (arrows). Bar = 3 µm. 8. Transverse section of pollen wall showing sexinous microgemmae, thick basal layer, electron-dense line (arrow), and intine (I). Bar = 0.5 µm. Callitriche peploides (Figs. 9–12). 9. Pollen grain showing spheroidal shape and microgemmate surface ornamentation with three aperture-like regions. Bar = 2 µm. 10. Detail of pollen surface showing pointed microgemmae in polygonal patterns. Bar = 1 µm. 11. Transverse section through pollen grain showing thin exinous basal layer and thicker intine in the three aperture-like regions (arrows). Bar = 2 µm. 12. Transverse section of pollen wall showing sexinous microgemmae, thick basal layer, electron-dense line (arrow), and thin intine (I). Bar = 0.5 µm

Callitriche peploides
Pollen spheroidal; 15.6 µm in diameter with three aperture-like areas (Fig. 9); surface ornamentation microgemmate with sculptural elements in polygonal patterns (Figs. 9–10); individual microgemmae pyramidal-shaped with pointed tips (Fig. 10); exine intectate and 1.11 µm thick; sexinous microgemmae with some lateral fusion near the bases (Figs. 11–12); nexine 0.37 µm thick and separated from intine by thin, electron-dense line (Fig. 12); nexine thinner and intine thicker in aperture-like regions (Fig. 11).

Callitriche marginata
Pollen spheroidal; 16.4 µm in diameter with three aperture-like areas (Fig. 13); surface ornamentation microgemmate with sculptural elements in polygonal patterns (Figs. 13–14); individual microgemmae mostly pyramidal-shaped and pointed (Fig. 14); exine intectate and 1.18 µm thick; sexinous microgemmae with lateral fusion near the bases (Figs. 15–16); nexine 0.27 µm thick and separated from intine by thin, electron-dense line (Fig. 16); some exine material extends into intine as globular protrusions (Fig. 16); nexine thinner in aperture-like regions (Fig. 15).

View larger version (55K): [in this window] [in a new window]   Figs. 13–16. Callitriche marginata. 13. Pollen grain showing spheroidal to subtriangular shape, microgemmate surface ornamentation, and three aperture-like regions. Bar = 3 µm. 14. Detail of pollen surface showing pyramidal-shaped microgemmae in polygonal patterns. Bar = 1 µm. 15. Transverse section through pollen grain showing thin exinous basal layer in the three aperture-like regions (arrows); note that the protoplasmic contents are partially plasmolyzed. Bar = 3 µm. 16. Transverse section of pollen wall showing sexinous microgemmae, thick basal layer, electron-dense line (arrow), and intine (I). Bar = 0.5 µm

View larger version (118K): [in this window] [in a new window]   Figs. 17–24. Amphibious taxa with internal geitonogamy. Callitriche heterophylla var. heterophylla (Figs. 17–20). 17. Pollen grain showing spheroidal shape and microgemmate surface ornamentation. Bar = 3 µm. 18. Detail of pollen surface showing fused microgemmae in polygonal to reticulate patterns. Bar = 1 µm. 19. Transverse section through pollen grain showing thin exinous basal layer and thicker intine in the three aperture-like regions (arrows). Bar = 3 µm. 20. Transverse section of pollen wall showing sexinous microgemmae and thick basal layer; note the globular exine protrusions (arrow) that penetrate the intine (I). Bar = 0.5 µm. Callitriche trochlearis (Figs. 21–24). 21. Pollen grain showing spheroidal shape and microgemmate surface ornamentation. Bar = 5 µm. 22. Detail of pollen surface showing round microgemmae covered by a thread-like material. Bar = 1 µm. 23. Transverse section through pollen grain showing thin exinous basal layer in two aperture-like regions (arrows). Protoplasmic contents and intine are not well-preserved in this grain due to initial fixation in formalin-acetic acid-ethanol rather than glutaraldehyde. Bar = 3 µm. 24. Transverse section of pollen wall showing sexinous microgemmae, thick basal layer, and intine (I). Bar = 0.5 µm

Amphibious taxa without internal geitonogamy
Taxa with distinct exines: Callitriche cophocarpa
Pollen spheroidal; 23.0 µm in diameter; inaperturate (Fig. 25); surface ornamentation microgemmate with sculptural elements in polygonal patterns (Figs. 25–26); individual microgemmae mostly round, some pyramidal-shaped (Figs. 25–26); exine intectate and 0.90 µm thick; many sexinous sculptural elements laterally fused (Figs. 27–28); nexine unevenly thick, averaging 0.19 µm, and slightly thicker under sculptural elements (Fig. 28); electron-dense line separating nexine from intine present, but less distinct than in other taxa (Fig. 28).

View larger version (177K): [in this window] [in a new window]   Figs. 25–36. Amphibious taxa without internal geitonogamy. Callitriche cophocarpa (Figs. 25–28). 25. Pollen grain showing spheroidal shape and microgemmate surface ornamentation. Bar = 3 µm. 26. Detail of pollen surface showing round microgemmae in polygonal patterns. Bar = 1 µm. 27. Transverse section through pollen grain showing uniformly thin basal layer. Bar = 5 µm. 28. Transverse section of pollen wall showing sexinous microgemmae, reduced basal layer, and intine (I). Bar = 0.5 µm. Callitriche cophocarpa-stagnalis hybrid (Figs. 29–32). 29. Pollen grain showing spheroidal shape and microgemmate surface ornamentation. Bar = 5 µm. 30. Detail of pollen surface showing round microgemmae in polygonal patterns. Bar = 1 µm. 31. Transverse section through pollen grain showing uniformly thin basal layer. Bar = 4 µm. 32. Transverse section of pollen wall showing sexinous microgemmae, reduced basal layer, and intine (I). Bar = 0.5 µm. Callitriche cribrosa33. Pollen grain showing spheroidal shape and microgemmate surface ornamentation. Bar = 2 µm. 34. Detail of pollen surface showing sculptural elements in polygonal patterns; note apical microechinae on individual rounded elements. Bar = 1 µm. 35. Transverse section through pollen grain showing uniformly thin basal layer. Bar = 3 µm. 36. Transverse section of pollen wall showing sexinous microgemmae, reduced basal layer, and thin intine (I). Bar = 0.5 µm

Callitriche cribrosa
Pollen spheroidal; 13.9 µm in diameter; inaperturate (Fig. 33); surface ornamentation microgemmate with sculptural elements in polygonal patterns; individual sculptural elements rounded with apical microechinae (Figs. 33–34); exine intectate and 0.80 µm thick; sexinous sculptural elements laterally fused; nexine unevenly thick, averaging 0.12 µm (Figs. 35–36); electron-dense line separating nexine from intine not distinct (Fig. 36).

Taxa with reduced or no exines—Callitriche hamulata
Pollen of both aerial and submersed flowers spheroidal and inaperturate (Fig. 37); aerial pollen 16.1 µm in diameter; submersed pollen significantly larger, 21.5 µm in diameter (Table 2); surface ornamentation of both pollen types consisting of small, granular microgemmae (Figs. 37–38); individual microgemmae distinct and with some lateral fusion; exine thin and intectate in both pollen types; exine significantly thicker in aerial pollen, 0.33 µm thick, and 0.19 µm thick in submersed pollen (Table 2); nexine uniformly thin in both pollen types (Figs. 39–40); nexine significantly thicker in aerial pollen, 0.14 µm thick, and 0.07 µm thick in submersed pollen (Table 2); intine well-developed, thick, and fibrillar in both pollen types (Figs. 39–40); intine 0.42 µm thick in aerial pollen and significantly thicker in submersed pollen, 0.67 µm thick (Table 2).

View larger version (123K): [in this window] [in a new window]   Figs. 37–44. Amphibious taxa without internal geitonogamy. Callitriche hamulata—submersed (Figs. 37–40). 37. Pollen grain showing spheroidal shape and granular surface ornamentation. Bar = 2 µm. 38. Detail of pollen surface showing granular microgemmae. Bar = 1 µm. 39. Transverse section through pollen grain showing reduced exine and thick intine. Bar = 3 µm. 40. Transverse section of pollen wall showing small, sexinous sculptural elements, thin basal layer, and thick intine (I). Bar = 0.5 µm. Callitriche brutia (Figs. 41–44). 41. Pollen grain showing spheroidal shape and slightly wrinkled surface. Bar = 5 µm. 42. Detail of pollen surface showing reduced surface ornamentation. Bar = 1 µm. 43. Transverse section through pollen grain showing lack of an electron-dense exine. Bar = 4 µm. 44. Transverse section of pollen wall showing fibrillar intine and lack of exine. Bar = 0.5 µm

Obligately submersed taxa (including Callitriche lusitanica)
Callitriche lusitanica
Pollen spheroidal; 17.3 µm in diameter; inaperturate (Fig. 45); surface ornamentation smooth to "wrinkled," due to slight desiccation of the grains during SEM preparation (Figs. 45–46); exine absent (Figs. 47–48); intine well-developed, unevenly thick, averaging 0.12 µm, and two-layered (Fig. 48); outer intine layer more densely fibrillar (Fig. 48).

View larger version (123K): [in this window] [in a new window]   Figs. 45–52. Obligately submersed taxa (including C. lusitanica). Callitriche lusitanica (Figs. 45–48). 45. Pollen grain showing spheroidal shape and wrinkled surface. Bar = 2 µm. 46. Detail of pollen surface showing wrinkled morphology. Bar = 1 µm. 47. Transverse section through pollen grain showing lack of an electron-dense exine. Bar = 3 µm. 48. Transverse section of pollen wall showing fibrillar intine and lack of exine. Bar = 0.5 µm. Callitriche truncata (Figs. 49–52). 49. Pollen grain showing spheroidal shape and smooth surface. Bar = 5 µm. 50. Detail of pollen surface showing lack of sculptural elements. Bar = 1 µm. 51. Transverse section through pollen grain showing lack of an electron-dense exine. Bar = 3 µm. 52. Transverse section of pollen wall showing unevenly thick, fibrillar intine and lack of exine. Bar = 0.5 µm

Discussion

Palynological characters
Pollen from all 13 Callitriche taxa in the present study was spheroidal, small, intectate, and lacked well-defined apertures. Taxa primarily differed with respect to surface ornamentation, exine thickness, and the presence or absence of aperture-like regions.

Although Callitriche pollen has been characterized as having true apertures (e.g., Erdtman, 1952 ; Moar, 1960 ; Díez, Talavera, and Garcia-Murillo, 1988 ), these have been described as either weakly defined or difficult to detect in transmitted light. In the present study, pollen from all 13 species lacked bona fide apertures, although aperture-like areas were identified in six species (C. deflexa, C. nuttallii, C. peploides, C. marginata, C. trochlearis, and C. heterophylla var. heterophylla). In addition, aperture-like areas have previously been described in four other species, including C. heterophylla var. bolanderi (Hegelm.) Fassett (Osborn and Philbrick, 1994 ), C. verna L. (= C. palustris), C. cophocarpa, and C. platycarpa Kütz. (Martinsson, 1993 ). Although Martinsson (1993) reported "aperturoid areas," or leptomata, in the pollen of C. cophocarpa using SEM and light microscopy, this species was determined to be inaperturate in the present study. Characterization of the apertural condition in Callitriche is difficult even with electron microscopy. Using SEM, aperture-like areas can be distinguished by irregular and less dense spacing of the sculptural elements; however, this character is not consistent. In our opinion, TEM is more revealing, as the aperture-like areas can be identified by thin regions in the basal layer (nexine).

Correlations with growth habit and pollination biology
Anemophily and "typical" geitonogamy occur in terrestrial species (Arber, 1920 ; Moar, 1960 ; Sculthorpe, 1967 ; Philbrick and Anderson, 1992 ). All three terrestrial species (C. deflexa, C. nuttallii, and C. peploides) had pollen grains with thick exines. The thick exine may play a role in preventing desiccation of the aerial-dispersed pollen in this growth habit. Furthermore, the pollen of terrestrial species had aperture-like areas, characterized by a reduction in the otherwise thickened basal layer. These thin exine regions may facilitate pollen tube germination. It is also noteworthy that the terrestrial taxa collectively had the smallest pollen diameters. This trend is correlated with the smallest overall anther size (Cooper et al., unpublished data) and the fewest number of pollen grains produced per anther (Philbrick and Anderson, 1992 ) in comparison with amphibious and obligately submersed species.

Pollen of the amphibious taxa generally had a thinner overall exine compared to that of the terrestrial taxa. Exine reduction in amphibious species was a result of both thinner sexine and nexine layers in most cases. Callitriche marginata, which does not produce flowers when plants are submersed, was the only exception. This species had the thickest exine of all amphibious taxa, and it is allied phylogenetically with the terrestrial taxa. Furthermore, surface ornamentation, patterning of sculptural elements, exine thickness and ultrastructure, and the presence of three aperture-like areas in the pollen of C. marginata are similar to that of the terrestrial taxa.

Two amphibious taxa in the present study had internal geitonogamy, C. heterophylla var. heterophylla and C. trochlearis, as did two other previously studied amphibious species: C. heterophylla var. bolanderi (Osborn and Philbrick, 1994 ; Philbrick and Osborn, 1994 ) and C. verna (= C. palustris; Martinsson, 1993 ). Although overall exine thickness was reduced in these taxa, the basal layer (nexine) was thicker than that of amphibious taxa without internal geitonogamy, and the pollen had aperture-like areas. Thinning of the exine in the aperture-like areas may facilitate pollen tube germination as in terrestrial taxa. In this case, however, pollen tube germination occurs precociously within indehiscent anthers (Philbrick and Anderson, 1992 ).

Taxa without internal geitonogamy, Callitriche cophocarpa, C. cophocarpa-stagnalis hybrid, and C. cribrosa, had a thinner exine overall, with the basal layer primarily being reduced. Pollen of these taxa was also inaperturate. Martinsson (1993) , however, reported aperture-like areas in C. cophocarpa and C. platycarpa. Although C. cophocarpa was determined to be inaperturate in the current study, it is noteworthy that these two species had slightly thicker basal layers than those of the other amphibious taxa without internal geitonogamy.

Amphibious taxa without internal geitonogamy can exhibit "typical" geitonogamy and/or anemophily if flowers are produced aerially; epihydrophily has also been reported when flowers are produced at the water level or when pollen is shed onto the water surface by aerial flowers (Schotsman, 1954, 1982 ; Sculthorpe, 1967 ; Philbrick and Anderson, 1992 ). The thinner exine and the absence of aperture-like regions in these taxa (except for C. platycarpa) may be correlated with epihydrophilous pollination. As such, the lack of aperture-like areas may decrease the likelihood of water uptake and thereby enable the pollen to be more buoyant for surface dispersal. Martinsson (1993) suggested that the "reticulate" surface ornamentation of C. cophocarpa, C. platycarpa, and C. stagnalis may enhance flotation for pollen transport at the water surface. Martinsson's (1993) hypothesis was based on a comparison with underwater-flowering Ruppia, in which air bubbles are released within submersed flowers, become trapped between the sculptural elements of the exine, and allow the pollen grains to float upward in the water column (Schwanitz, 1967 ; Pettitt and Jermy, 1975 ; Verhoeven, 1979 ). However, this hypothesis is not as relevant in Callitriche, because C. cophocarpa, C. platycarpa, and C. stagnalis produce flowers only in floating rosettes, but not when plants grow in submersed forms.

Callitriche hamulata, depending on the type of amphibious growth form, may produce aerial flowers exclusively, submersed flowers exclusively, or both on the same plant. When mature pollen from aerial and submersed flowers of C. hamulata was compared, many characters varied statistically (Table 2). Grain diameter and intine thickness were significantly greater in pollen of submersed flowers. By contrast, overall exine thickness, sexine thickness, and nexine thickness were significantly thicker in pollen of aerial flowers. As the exine is important for the prevention of water loss and maintenance of pollen viability, it might be expected that pollen from aerial flowers would have a thicker exine than that of submersed flowers. Nevertheless, exine thickness in aerial pollen of C. hamulata was extremely reduced when compared to other aerial-flowering Callitriche species. The smaller size of the aerial pollen of C. hamulata may minimize dehydration and thereby compensate for the overall exine reduction.

Two amphibious taxa lacked exines: C. brutia (the closest relative of C. hamulata) and C. lusitanica. These findings provide the first documentation of exine loss in the amphibious growth habit. However, pollen of C. brutia has previously been reported to have a reticulate surface ornamentation and an ellipsoidal shape (Díez, Talavera, and Garcia-Murillo, 1988 ). It is possible that Díez, Talavera, and Garcia-Murillo (1988) studied pollen from an aerial flower of C. brutia, whereas pollen from submersed flowers was examined in the current study.

New data on the obligately submersed C. truncata confirm the lack of an exine in this growth habit. Pollen of C. hermaphroditica has previously been shown to be exineless (Osborn and Philbrick, 1994 ; Philbrick and Osborn, 1994 ).

Evolution of hypohydrophily
Underwater pollination systems in angiosperms are derived from aerial floral systems (e.g., Faegri and van der Pijl, 1979 ; Philbrick and Les, 1996 ). Exine reduction or complete exine loss appear to be an essential component in the evolution of hydrophily (e.g., Philbrick, 1991 , and references therein). This has been documented in hydrophilous members of the Najadaceae (e.g., Pettitt and Jermy, 1975 ), Hydrocharitaceae (e.g., Pettitt, 1981 ), Cymodoceaceae (e.g., Pettitt, Ducker, and Knox, 1978 ; Pettitt, 1980, 1981 ), Zannichelliaceae (e.g., McConchie, Knox, and Ducker, 1982 ), and Ceratophyllaceae (e.g., Pettitt and Jermy, 1975 ; Les, 1988 ; Takahashi, 1995 ). The Callitrichaceae are unique in being the only known angiosperm family in which aerial, epihydrophilous, and hypohydrophilous pollination systems co-occur (Philbrick and Anderson, 1992 ). Therefore, the family provides the opportunity to evaluate questions concerning the evolution of underwater pollination.

The presence of an exine appears to be less critical when pollination is water-mediated. This is evident in Callitrichaceae, where exine structure and thickness are correlated with growth habit and pollination type. Pollen of the two obligately submersed taxa studied either lacked an exine (C. truncata and C. hermaphroditica—perennial form) or had a highly reduced exine (C. hermaphroditica—annual form); nevertheless, these plants pollinate successfully underwater. By contrast, pollen of aerial-flowering terrestrial species had well-defined exines that were considerably thicker than that of most amphibious taxa. Exine thickness was quite variable among amphibious species, especially with respect to the basal layer (nexine) in taxa with vs. without internal geitonogamy.

A similar correlation between pollen wall ultrastructure (i.e., exine reduction) and growth habit was evident within the single amphibious taxon C. hamulata. In C. hamulata, the statistically significant variation in exine thickness (and several other characters; Table 2) between fully mature pollen from aerial vs. submersed flowers may indicate that pollen in this species exhibits developmental plasticity. Such significant intraspecific variation in exine structure is not common. This ontogenetic hypothesis may also explain the variation in pollen of C. brutia observed in the present paper (exineless from submersed flowers) with that studied by Díez, Talavera, and Garcia-Murillo (1988 ; with a reticulate exine).

The palynological data indicate that underwater pollination in Callitriche may have had a recent origin. For example, the degree of exine loss is dissimilar between the two obligately submersed taxa studied. The pollen of C. truncata was completely exineless, and that of C. hermaphroditica was variable. Pollen of the perennial growth form of C. hermaphroditica lacked an exine, whereas pollen of the annual growth form had a rudimentary exine (Osborn and Philbrick, 1994 ; Philbrick and Osborn, 1994 ). Martinsson (1993) also identified extreme exine reduction in C. hermaphroditica, but not complete loss.

Furthermore, a shift from aerial pollination to hypo-hydrophily within the amphibious growth habit may be occurring within the C. hamulata/C. brutia clade. Although hydrophily has not been documented in this clade, these taxa exhibit exine reduction and exine loss (Fig. 53, clade H). Possible ontogenetic plasticity in the pollen wall of C. hamulata (i.e., from aerial vs. submersed flowers) provides further support for this hypothesis. Additionally, C. hamulata and C. brutia are unusual among amphibious Callitriche species in that they are able to produce seed in submersed flowers. However, seed set in these two taxa may not be related to hydrophily, but rather the result of agamospermy. Additional studies of the reproductive biology and systematics of these species are needed and may clarify these questions.

View larger version (30K): [in this window] [in a new window]   Fig. 53. Cladogram of 18 Callitriche species showing the distribution of pollen wall characters, particularly exine thickness and structure (solid black), as well as intine thickness (stippled). All pollen wall drawings have been reproduced at the same scale. See text for discussion of letter-indicated nodes and pollen wall correlations with chromosome number (2n; indicated in brackets following each species name). The cladogram has been modified from Philbrick and Les (1995) and is based on analyses of both rbcL and morphological characters. Pollen wall data for C. stagnalis, C. heterophylla var. bolanderi, and C. hermaphroditica are taken from Philbrick and Osborn (1994) , and those for C. platycarpa and C. verna (= C. palustris) are derived from Martinsson (1993)

Consideration of palynological characters relative to the proposed phylogeny provides support for several hypotheses concerning pollen structural evolution. Pollen of most Callitriche species has a well-developed exine. In contrast, exine reduction or absence occurs in five species: C. hermaphroditica, C. truncata, C. lusitanica, C. brutia, and C. hamulata. It is evident that exine reduction has evolved at least twice in Callitriche (Fig. 53, clades A and H). Exine reduction is associated with aneuploid reduction in chromosome number (except for C. cribrosa, see below). The three species in clade A have 2n = 6 or 8 and represent aneuploid reduction from 2n = 10. Callitriche brutia (2n = 28) and C. hamulata (2n = 38) seem to represent aneuploid reductions from polyploids (2n = 30 and 40, respectively), although alternative derivations of their chromosome numbers cannot be ruled out. Sexine reduction is not apparent in the aneuploid C. cribrosa (2n = 8), but the nexine (basal layer) is reduced. Exine reduction in clade A is also associated with hypohydrophilous pollination, although the submersed form of the amphibious C. lusitanica also lacks an exine. Possible relationships between pollen structure and pollination in this species remain to be elucidated.

Two apertural conditions occur in Callitriche: aperture-like and inaperturate. Aperture-like areas do not occur in species that either have reduced or absent exines; predictions about the evolution of this feature correspond to the evolution of exine reduction noted above. Inaperturate grains also characterize three species with well-developed exines: C. cophocarpa, C. cribrosa, and C. stagnalis. It is not possible to determine whether the inaperturate condition is ancestral in Callitriche given the current phylogenetic insight. However, pollen with true apertures is present in the probable outgroups (Scrophulariaceae and Hippuridaceae; e.g., Erdtman, 1952 ).

The present study is the first to incorporate data on pollen morphology and ultrastructure in a phylogenetic interpretation of the Callitrichaceae. Although these data do not fully answer all of the systematic queries associated with the family, the exine appears to be an excellent character to incorporate into cladistic analyses. Moreover, exine architecture correlates well with the various growth habits and associated pollination biologies found within this unique aquatic family.

FOOTNOTES

¹ The authors thank Karin Martinsson (Uppsala University; Sweden) and Alejandro Novelo (Universidad Nacional Autónoma de México) for assistance with field collection, as well as Bradford L. Day and Sarah S. Kreunen (both of Truman State University) for rendering line drawings of exine structure. This study was supported in part by Truman State University (Biology Alumni Research Stipend and Ronald E. McNair Program to RLC; Faculty Research Grant to JMO); Beta Beta Beta, National Biological Honor Society (Student Research Grant to RLC); the Council on Undergraduate Research (Summer Undergraduate Research Experience Fellowship sponsored by the National Science Foundation to RLC); and the National Science Foundation (DUE-ILI-9352323 to JMO; DEB-9496053 to CTP).

⁴ Current address: Department of Biological Sciences, University of Alberta, Edmonton, Alberta, Canada T6G 2E9.

⁵ Author for correspondence (e-mail: josborn{at}truman.edu) .

LITERATURE CITED

Arber, A. 1920 Water plants. University Press, London, UK.

Cox, P. A. 1988 Hydrophilous pollination. Annual Review of Ecology and Systematics 19: 261–280.

Dahlgren, R., and F. N. Rasmussen. 1983 Monocotyledon evolution: characters and phylogenetic estimation. In M. K. Hecht, B. Wallace, and G. T. Prance [eds.], Evolutionary biology, vol. 16, 255–395. Plenum, New York, New York, USA.

Díez, M. J., S. Talavera, and P. Garcia-Murillo. 1988 Contributions to the palynology of hydrophytic, non-hydrophilous angiosperms. 1. Studies with LM and SEM. Candollea 43: 147–158.

Erdtman, G. 1952 Pollen morphology and plant taxonomy: angiosperms. Almqvist and Wiksell, Stockholm, Sweden.

Faegri, K., and L. van der Pijl. 1979 The principles of pollination ecology, 3rd. rev. ed. Pergamon Press, Oxford, UK.

Les, D. H. 1988 The origin and affinities of the Ceratophyllaceae. Taxon 37: 326–345. [CrossRef][ISI]

Martinsson, K. 1993 The pollen of Swedish Callitriche (Callitrichaceae)—trends toward submergence. Grana 32: 198–209. [ISI]

McConchie, C. A., R. B. Knox, and S. C. Ducker. 1982 Ultrastructure and cytochemistry of the hydrophilous pollen of Lepilaena (Zannichelliaceae). Micron 12: 339–340. [ISI]

Moar, N. T. 1960 Studies in pollen morphology. 2. The pollen of New Zealand and two Australian species of Callitriche L. New Zealand Journal of Science 3: 415–421.

Osborn, J. M., and C. T. Philbrick. 1994 Comparative pollen structure and pollination biology in the Callitrichaceae. Acta Botanica Gallica 141: 257–266. [ISI]

Pettitt, J. M. 1980 Reproduction in seagrasses: nature of the pollen and receptive surface of the stigma in the Hydrocharitaceae. Annals of Botany 45: 609–622.

———. 1981 Reproduction in seagrasses: pollen development in Thalassia hemprichii, Halophila stipulacea, and Thalassodendron ciliatum. Annals of Botany 48: 609–622.

———, S. A. Ducker, and R. B. Knox. 1978 Amphibolis pollen has no exine. Helobiae Newsletter 2: 19–22.

———, and A. C. Jermy. 1975 Pollen in hydrophilous angiosperms. Micron 5: 377–405. [ISI]

Philbrick, C. T. 1984 Pollen tube growth within vegetative tissues of Callitriche (Callitrichaceae). American Journal of Botany 71: 882–886. [CrossRef][ISI]

———. 1991 Hydrophily: phylogenetic and evolutionary considerations. Rhodora 93: 36–50. [ISI]

———. 1993 Underwater cross-pollination in Callitriche hermaphroditica: evidence from random amplified polymorphic DNA markers. American Journal of Botany 80: 391–394. [CrossRef][ISI]

———, and G. J. Anderson. 1992 Pollination biology in the Callitrichaceae. Systematic Botany 17: 282–292. [CrossRef][ISI]

———, and L. M. Bernardello. 1992 Taxonomic and geographic distribution of internal geitonogamy in New World Callitriche (Callitrichaceae). American Journal of Botany 79: 887–890. [CrossRef][ISI]

———, and R. K. Jansen. 1991 Phylogenetic studies of North American Callitriche (Callitrichaceae) using chloroplast DNA restriction fragment analysis. Systematic Botany 16: 478–491. [CrossRef][ISI]

———, and D. H. Les. 1995 Systematics of North American and European Callitriche (Callitrichaceae). American Journal of Botany 82 (6, supplement): 156. (Abstract)

———, and ———. 1996 Evolution of aquatic angiosperm reproductive systems. BioScience 46: 813–826. [CrossRef][ISI]

———, and J. M. Osborn. 1994 Exine reduction in underwater flowering Callitriche (Callitrichaceae): implications for the evolution of hydrophily. Rhodora 96: 370–381. [ISI]

Rowley, J. C., and D. T. Moran. 1975 A simple procedure for mounting wrinkle-free sections on formvar-coated slot grids. Ultramicroscopy 1: 151–155. [CrossRef][Medline]

Schotsman, H. D. 1954 A taxonomic spectrum of the section Eucallitriche in the Netherlands. Acta Botanica Neerlandica 3: 313–384. [ISI]

———. 1982 Biologie florale des Callitriche: étude sur quelques espèces d'Espagne méridionale. Adansonia 4: 11–160.

Schwanitz, G. 1967 Untersuchungen zur postmeiotischen Mikrosporogenese. I. Morphogenese des Ruppia—pollens. Pollen et Spores 9: 9–48.

Sculthorpe, C. D. 1967 The biology of aquatic vascular plants. Edward Arnold, London, UK.

Takahashi, M. 1995 Development of structure-less pollen wall in Ceratophyllum demersum L. (Ceratophyllaceae). Journal of Plant Research 108: 205–208. [CrossRef][ISI]

Tomlinson, P. B. 1982 Helobiae (Alismatidae). In C. R. Metcalfe [ed.], Anatomy of the monocotyledons, vol. VII, 1–522. Clarendon Press, Oxford, UK.

Venable, J. H., and R. Coggeshall. 1965 A simplified lead citrate stain for use in electron microscopy. Journal of Cell Biology 25: 407.[Free Full Text]

Verhoeven, J. T. A. 1979 The ecology of Ruppia-dominated communities in western Europe I. Distribution of Ruppia representatives in relation to their autecology. Aquatic Botany 6: 197–268.

This article has been cited by other articles:

				V. Merckx, P. Schols, H. M.-v. d. Kamer, P. Maas, S. Huysmans, and E. Smets Phylogeny and evolution of Burmanniaceae (Dioscoreales) based on nuclear and mitochondrial data Am. J. Botany, November 1, 2006; 93(11): 1684 - 1698. [Abstract] [Full Text] [PDF]

				W. S. Judd and R. G. Olmstead A survey of tricolpate (eudicot) phylogenetic relationships Am. J. Botany, September 1, 2004; 91(10): 1627 - 1644. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (6) Citing Articles via Google Scholar Google Scholar Articles by Cooper, R. L. Articles by Philbrick, C. T. Search for Related Content PubMed PubMed Citation Articles by Cooper, R. L. Articles by Philbrick, C. T. Agricola Articles by Cooper, R. L. Articles by Philbrick, C. T.