The anthocerote chloroplast: A review

Abstract

CONTENTS Summary 169 III. Variations in chioroplast (plastid) ultrastructure 180 I. Introduction 170 IV. Chloroplast division 185 II. The chloroplast of the mature gametophyte 170 V. Associations with other organelles 185 (a) General considerations 172 VI. Evolutionary considerations 187 (b) The pyrenoid 176 VII. Taxonomic uses and conspectus 188 (c) Thylakoids 178 Acknowledgements 189 (d) Starch 178 References 189 (e) Chioroplast number, size and shape SUMMARY This review covers previous data, together with new information from our laboratories, on the subject of the anthocerote chioroplast. Unlike all other archegoniates, most species of anthocerote have pyrenoids in their chloroplasts. The pyrenoid is the site of accumulation of the first enzyme in the C3 photosynthetic cycle, ribulose bisphosphate carboxylase/oxygenase. Unlike most algae, the hornwort pyrenoid is composed of distinct subunits, numbering up to several hundred. Pyrenoid morphology is quite variable among the genera in shape, fine structure, and distribution of inclusions. Another unique feature of the anthocerote chloroplast is the presence of thylakoids that connect adjacent granal stacks at right angles to the long axis of the granum (so-called channel thylakoids), resulting in a 'spongy' arrangement of the thylakoid system. The granal stacks of anthocerotes are like the 'pseudograna' of green algae because they lack the highly-curved end membranes typical of all other embryophytes. The channel thylakoids are enriched in photosystem (PS) I and the grana are enriched in PS II. The chloroplast envelope is a double membrane structure with regions of appression, much like that of other green plants. The apical cell of the gametophyte contains chloroplasts similar to the mature chloroplasts of the thallus, although certain gametophytic tissues may contain underdeveloped plastids. Chloroplasts in cells around Nostoc colonies and in cells invaded by mycorrhizal fungi have thylakoids mainly in pairs, and small or absent pyrenoids. A number of similarly reduced plastids are noted in the placental cells at the sporophybe/gametophyte junction and in developing spores. The greatest reduction is observed in spermatid cell plastids, which at maturity consist of only a small starch grain surrounded by the envelope. Chromoplast-like organelles are found in the cells of the antheridial jacket in some genera; these contain numerous osmiophilic globules that are probably pigment aggregations. Colourless bead-like plastids occur in the rhizoids; these seem to develop by fragmentation of the single chioroplast in the rhizoid initials, concomitant with the loss of chlorophyll. Chloroplast division is a tightly controlled process and, in uniplastidic species, always occurs just before nuclear division, with the participation of a unique system of chioroplast-associated microtubules. The number of chloroplasts per cell is quite variable in some genera, although most species have but a single chloroplast in each cell of the gametophyte. Chioroplast shape is also variable from ellipsoidal, dumbbell-shaped, to irregular. These data indicate that the anthocerote chloroplast is unique among the embrvophvtes and are in line with the notion of an isolated position in the plant kingdom. Certain features of chioroplast morphology appear to be typical of certain genera and might prove useful in taxonomic decisions at the generic level.

Full text

New Phytol. (1992), 120, 169 190
The anthocerote chloroplast: a review
By KEVIN C. VAUGHN1, ROBERTO LIGRONE2,
HEATHER A. OWEN3, JIRO HASEGAWA4, ELLA 0. CAMPBELL5,
KAREN S. RENZAGLIA6 and JULIAN MONGE-NAJERA7
1. Southern Weed Science Laboratory, USDA/ARS, P.O. Box 350, Stoneville, MS 38776, USA
2. Dipartimento di Biologia vegetate, via Foria 223, 1-80139 Napoti, Italy
3. Zoology Department, Miami University, Oxford, OH 45056, USA
4. Laboratory of Applied Biology, Faculty of Agriculture, Kyoto University, Kyoto 606, Japan
5. Botany and Zoology Department, Massey University, Palmerston North, New Zealand
6. BiologyDepartment, E. Tennessee State University, Johnson City, TN 37614, USA
7. Biologia Tropical, Universidad de Costa Rica, Ciudad Universitaria ‘Rodgigo Facio’, Costa Rica
(Received 19 March 1991; accepted 6 October 1991)
CONTENTS
Summary 169 III. Variations in chioroplast (plastid) ultrastructure 180
I. Introduction 170 IV. Chloroplast division 185
II. The chloroplast of the mature gametophyte 170 V. Associations with other organelles 185
(a) General considerations 172 VI. Evolutionary considerations 187
(b) The pyrenoid 176 VII. Taxonomic uses and conspectus 188
(c) Thylakoids 178 Acknowledgements 189
(d) Starch 178 References 189
(e) Chioroplast number, size and shape
SUMMARY
This review covers previous data, together with new information from our laboratories, on the subject of the anthocerote chioroplast.
Unlike all other archegoniates, most species of anthocerote have pyrenoids in their chloroplasts. The pyrenoid is the site of
accumulation of the first enzyme in the C3 photosynthetic cycle, ribulose bisphosphate carboxylase/oxygenase. Unlike most algae,
the hornwort pyrenoid is composed of distinct subunits, numbering up to several hundred. Pyrenoid morphology is quite variable
among the genera in shape, fine structure, and distribution of inclusions. Another unique feature of the anthocerote chloroplast is the
presence of thylakoids that connect adjacent granal stacks at right angles to the long axis of the granum (so-called channel
thylakoids), resulting in a ‘spongy’ arrangement of the thylakoid system. The granal stacks of anthocerotes are like the
‘pseudograna’ of green algae because they lack the highly-curved end membranes typical of all other embryophytes. The channel
thylakoids are enriched in photosystem (PS) I and the grana are enriched in PS II. The chloroplast envelope is a double membrane
structure with regions of appression, much like that of other green plants. The apical cell of the gametophyte contains chloroplasts
similar to the mature chloroplasts of the thallus, although certain gametophytic tissues may contain underdeveloped plastids.
Chloroplasts in cells around Nostoc colonies and in cells invaded by mycorrhizal fungi have thylakoids mainly in pairs, and small or
absent pyrenoids. A number of similarly reduced plastids are noted in the placental cells at the sporophybe/gametophyte junction
and in developing spores. The greatest reduction is observed in spermatid cell plastids, which at maturity consist of only a small
starch grain surrounded by the envelope. Chromoplast-like organelles are found in the cells of the antheridial jacket in some genera;
these contain numerous osmiophilic globules that are probably pigment aggregations. Colourless bead-like plastids occur in the
rhizoids; these seem to develop by fragmentation of the single chioroplast in the rhizoid initials, concomitant with the loss of
chlorophyll. Chloroplast division is a tightly controlled process and, in uniplastidic species, always occurs just before nuclear
division, with the participation of a unique system of chioroplast-associated microtubules. The number of chloroplasts per cell is
quite variable in some genera, although most species have but a single chloroplast in each cell of the gametophyte. Chioroplast
shape is also variable from ellipsoidal, dumbbell-shaped, to irregular. These data indicate that the anthocerote chloroplast is unique
among the embrvophvtes and are in line with the notion of an isolated position in the plant kingdom. Certain features of chioroplast
morphology appear to be typical of certain genera and might prove useful in taxonomic decisions at the generic level.
Key words: Anthocerote, Anthocerotophyta, chloroplast, Bryophyta, electron microscopy, pyrenoid.
I. INTRODUCTION
The anthocerotes are a small group of land plants characterized by having non-meristematic uniplastidic cells and pyrenoids in the
chloroplasts. Although this makes the anthocerotes a very well defined group, in an isolated position within the embryophvtes, their
taxonomy is extremely controversial, because of the lack of clear-cut diagnostic features.
Several hundred species have been described so far, mostly from tropical and subtropical areas, but actually the group is likely to
comprise no more than a hundred species, with many of them reported in the literature under synonyms (Schuster, 1984; Hasegawa,
1988). Six genera are commonly recognized, i.e. Anthoceros, Phaeoceros, Notothylas, Dendroceros, Folioceros and Megaceros
(Schofield, 1984). Folioceros is considered to be a subgenus of Anthoceros by Hasegawa (1988), whereas Häsel de Menendez
(1988) recognizes two additional genera, .Leiosporoceros [treated as a subgenus of Phaeoceros by Hasegawa (1988)) and
Sphaerosporoceros (considered under Anthoceros by other taxonomists).

Inter- and infrageneric differences in plastid number and morphology have been known for a long time (Campbell, 1907;
McAllister, 1914,1917; Kaja, 1954), but this topic has so far been addressed at the ultrastructural level in only a limited range of
species (Burr, 1970; Valentine, Campbell & Hopcroft, 1986).
This paper reports a study of the chloroplast in a number of anthocerote species (see Table 1), including representatives of
Folioceros and Sphaerosporoceros, genera that have not yet been investigated. Variations in plastid structure in different tissues of
both the gametophyte and sporophyte are also reported. The data are integrated with information from the literature on hornwort
plastids, to evaluate the evolutionary and taxonomical significance of plastids in the anthocerotes.
II. THE CHLOROPLAST OF THE MATURE GAMETOPHYTE
(a) General considerations
nthocerote chloroplasts contain all of the elements used in defining a chloroplast (Fig. 1 A). The chloroplast is bound by
a double-membraned envelope that varies in the closeness of the two membranes from well separated to completely appressed. The
hylakoids are distributed throughout the interior of the chloroplast and are surrounded by stroma. In the case of species with
pyrenoids, these areas are more electron-opaque than the surrounding stroma. In many species, plastoglobuli, spherical aggregations
of lipid, are found throughout the stroma, often clustered in small groups.
uccessful fixation and embedding of anthocerote chloroplasts are not easy tasks. Because of this, previous analyses of
these chloroplasts are incomplete or inaccurate. Vaughn et at. (1990a) report that a fixation in 6 0 (v/v) glutaraldehyde in 005 ii
PIPES (pH 74), washing in cacodylate buffer, and post- fixation in 200 (v/v) 0s04 in cacodylate buffer are satisfactory for
gametophytic tissue in many species. Addition of caffeine (for some Megaceros spp.) and tannic acid (for some Notothytas spp.) are
useful in providing optimal resolution, by binding phenolics (caffeine) or enhancing membrane binding of osmium (tannic acid).
Sporophytic tissue is fixed well with this same protocol and generally does not require the additives for successful fixation. All
bryophytes are difficult to embed in epoxy resins so a slow infiltration scheme (exchanges over several days), with agitation or
rotation of the sample in 10000 resin, is most effective for successful embedding (Duckett, Renzaglia & Smith, 1988). Spurr’s or
other low viscosity resins are the most often-utilized. Of the non-epoxy resins, L.R. White resin is the most commonly used for
immunocytochemistry. Lowicryl resin, although of very low viscosity, apparently does not penetrate the tissue easily so that a very
long infiltration protocol (4—6 d at 0—4 °C) is required for successful embedding.
Table 1. List of anthocerote species that have been investigated ultrastructurally
Species Ref.*
Anthoceros
punctatus L. 1, 2, 6, 9
laminiferus Steph. 1
formosae Steph. 1
lametlatus Steph. 1
fusiformis Aust. 1
ecklonii Steph. 8
Dendroceros
granulatus IVIitt. 1, 3
tubercularis Hatt. 1, 4
japonicus Steph. 1
ualidus Steph. 1, 3
javanicus Nees 1
cavernosus Haseg. 1
Folioceros
fuciformis Bharadw. 1
appendicutatus Haseg. 1
Megaceros
longispirus Steph. 1, 3
leptohvmenius Steph. 1, 3, ?5
flageltaris Steph. 1, 3, 5
sp. from North Carolina, USA 1, 6
giganteus Steph. 3, 5
vincentianus Camp. 1, 5
denticulatus Steph. 1, 3
Notothytas
orbicularis Sull. 1, 5, 6
temperata Haseg. 1
breutetii Gottsche 1
Phaeoceros
himalayensis Prosk. 5
laevis (and carolinianus) Prosk. 1
coriaceus Steph. 1
microsporus Hässel 1
miyakeanus Hatt. 7
Sphaerosporoceros
adscendens Hässel 1
*References: 1 This report and/or Vaughn et al. (1990 a); 2 Chauhan & Schraudholf (1986); 3 Valentine et at. (1986); 4 Ligrone &
Renzaglia (1990); 5 Burr (1970); 6 Duckett & Renzaglia (1988); 7 Toyama (1974); 8 Wilsenach (1963); 9 Ajiri & Ueda (1986).

Figure 1. (A) Electron micrograph of a chloroplast of Sphaerosporoceros adscendens, a rather ‘typical’ anthocerote chloroplast.
Subunits of the multiple pyrenoid (P) are more electron-opaque than the surrounding stroma. Channel thylakoids connect grana
stacks together end-to-end. Numerous plastoglobuli (arrows) are electron-opaque spheres found in the stroma. S, starch; bar, 50
sum. (B) Electron micrograph of the pyrenoid region of Phaeoceros laevis. The pyrenoid subunits (P) are more electron-opaque than
the surrounding stroma. The pyrenoid is dissected by stroma as well as thylakoids organized into both grana (g) and single (stroma)
lamellae. Bar, 05 sum. Inset: Nomarski differential interference micrograph of an Anthoceros punctatus chloroplast with a
prominent multiple pyrenoid (P), x 200.
(b) The pyrenoid
McAllister (1914, 1927), using light microscopy, was the first to describe the pyrenoid of Phaeoceros laevis as a multiple
pyrenoid; that is, a pyrenoid consisting of up to 200 small disc-like structures aggregated in the centre of the chloroplast. This makes
it distinct from the ‘unit’ pyrenoids of most algae (e.g. McKay & Gibbs, 1989). Anthocerote pyrenoids are especially striking using
Nomarski differential interference microscopy so that the pyrenoid subunits, or pyrenosomes are seen in positive relief (Fig. 1 B
inset).
Electron microscopic investigations (Fig. 1 B) of these structures in P. laevis reveal that the pyrenoid region is traversed
by thylakoids, generally both stacked and unstacked, and also by areas of nonpyrenoid stroma. The pyrenoid subunits are
approximately 05 ,um wide and on the average about 15—20 pm long in median longisections through the plastid. In none of the
published studies have serial sections been made through the pyrenoid regions to determine if the units are discrete or are one highly
dissected structure, intercalated by thylakoids and non-pyrenoid stroma. Generally, the pyrenoid occupies the central area of the
chloroplast and is surrounded by starch granules, especially in older tissue.
here are numerous variations in this pyrenoid structure. In pyrenoids of Notothylas and Fotioceros, pyrenoglobuli occur
at the periphery of the pyrenoid subunits (Fig. 2). Although the composition of these pyrenoglobuli is unknown, it is probable that
they are lipid-containing, like the plastoglobuli of higher plants (Steinmüller & Tevini, 1985). Chloroplasts of Sphaerosporoceros
and Anthoceros have clusters of plastoglobuli in the stroma in the region of the pyrenoid but not within the pyrenoid itself (Fig. 1
A). In Dendroceros, the pyrenoid matrix characteristically exhibits a pattern of electron-opaque globules with finer threadlike
structures radiating from their surface. These globules are dispersed throughout the pyrenoid and rather evenly spaced, as though a
repeating substructure exists in the pyrenoid (Fig. 3). The staining characteristics and shapes of these opaque globules is similar to
the pyrenoglobuli found in Fotioceros and Notothylas, indicating that the bodies in Dendroceros pyrenoids are also lipids. The
pyrenoid subunits of most species are disc-shaped. Those of Notothytas and Dendroceros are quite irregular in shape, although the
final shape of the pyrenoid aggregation in Dendroceros is similar to that noted in the other genera. Notothylas species have orbicular
pyrenoids, similar in overall shape to the pyrenoids found in green algae such as Coleochaete. The degree of dissection of the

pyrenoid by non-pyrenoid stroma is also quite variable. For example, in Notothylas orbicularis and N. breutetii, the pyrenoid
subunits are nearly appressed, separated only by stroma thylakoids and a relatively small area of stroma, whereas in other species
(including N. temperata) the subunits are widely separated by non-pyrenoid stroma and both stromal granal lamellae are present in
the pyrenoid area.
Not all of the anthocerote species have pyrenoids. All of the Megaceros species investigated by Valentine et at. (1986)
and by us lack a pyrenoid; the electron opacity of the stroma is even throughout. Burr (1970) found that in some Megaceros species,
particular areas of the stroma had more starch grains or granules, like the starch accumulation around the pyrenoid in pyrenoid-
containing species. Similar starch distributions were not found by either Valentine et at. (1986) or us (Fig. 4) when these same
species were examined.
Anthocerosfusiformis has chloroplasts in which the starch-containing layers form a halo around a central area devoid of
starch (Fig. 5A). No pyrenoid is visible in this area, although the arrangement of starch granules is very much like that in species
with typical pyrenoid. This starch distribution is unlikely to be an artifact of fixation or growth condition as it was observed in
specimens of this species grown under different growth and light conditions. A similar starch distribution is found in P. coriaceus
(Fig. 5B). Moreover, observation of apical and subapical cells of both of these species with ‘pseudopyrenoids’ reveals a similar
starch distribution to that found in the mature regions. Light microscopic stains for protein reveal an even staining throughout the
plastid both in those species with anomalous starch distribution and the Megaceros species studied by Burr (1970), whereas
pyrenoids are quite distinct from the remainder of the stroma using these same a site of accumulation of ribulose-1 ,5-bisphosphate
stains. carboxylase/oxygenase (Rubisco) (e.g. Mc Kay & In algae, the biochemical and immunocyto- Gibbs, 1989). Vaughn et at.
(1990 a) have recently chemical evidence clearly shows that the pyrenoid is examined the chloroplasts of a number of hornwort
species and found that all of the Rubisco is concentrated in the pvrenoids, with only background labelling elsewhere in the non-
pyrenoid stroma (Fig. 6A). Pvrenoid inclusions, such as pvrenoglobuli, are not labelled. These data indicate that the hornwort
pyrenoid is homologous with the algal pyrenoids and that the Rubisco in the pyrenoid is active, because all of it is present in the
pyrenoid. Chloroplasts without pyrenoids, including those with starch-free areas of stroma in the centre of the chloroplast, have an
even distribution of Rubisco throughout the stroma (Fig. 6B; Vaughn et a!., 1990a).
Figure 2. Electron micrographs of the pyrenoid region of (A) Notothylas orbicularis, (B) N. temperata, and (C) Folioceros
fuciformis. In N. orbicularis, the pyrenoid subunits are spaced closely together with only single lamellae separating the subunits.
Pvrenoglobuli border the subunits (arrows). N. temperata pyrenoids are more dissected by non-pyrenoid stroma (s) and lamellae
occasionally occur in the grana stacks as well. The Folioceros pvrenoid has numerous pvrenoglobuli (arrows) as well as well-
developed grana stacks that separate the subunits. Bars: (A, B) 05 /m; (C) 1 0μm.

Figure 3. Dendroceros chloroplasts. (A) Low magnification electron micrograph of D. tubercularis plastid revealing the
irregular chioroplast outline and the centrally localized pyrenoid (P) with irregularly shaped subunits. Bar, 50 1um. (B) Details of
the pyrenoid region of D. validus, revealing the repeating pattern of electron-opaque pyrenoid inclusions surrounded by a spoke-
like array of electron opaque lines. Bar, 05 pm. Inset is a Nomarski differential interference micrograph of the wing of the thallus of
D. tubercularis. Note the irregular plastid shape in this profile, x 200.

Figure 4. Megaceros chioroplasts. (A) M. leptohymenius. (B) M. flagellaris. No pyrenoids are found in these chloroplasts and the
starch is randomly distributed. Bar, 50 pm. Nu, nucleus; s, starch. Inset is a bright-field light micrograph of an upper epidermal cell
of M. flagellaris. Two chloroplasts are present, x 200.

Figure 5. Chloroplasts with unusual starch distributions. (A) Anthocerosfusiformis with starch (s) concentration near the plastid
envelope, leaving a starch-free zone (*) in the centre of the chloroplast. (B) The Phaeoceros coriaceus chloroplast has a
concentration of starch (a) near the envelope and, in addition, has an irregular morphology. Bars, 20 /im. Inset: Nomarski
differential contrast micrograph of P. coriaceus chloroplasts showing the dumbbell-shaped chloroplast often found in this species,
x 400.
(c) Thylakoids
he presence of a pyrenoid is not the only distinguishing characteristic of the anthocerote chloroplast, as the thylakoid
architecture is also unique. The granal stacks are connected to each other through the thylakoids orientated perpendicular tothe long
axis of the granum (Figs 1, 6C). This gives - a ‘spongy’ appearance to the thylakoid system. The intergranal thylakoids were termed
‘channel thylakoids’ by Burr (1970) in that they dissect the stroma into channels. The granal stacks of the anthocerote chloroplast
are also atypical in that they lack the highly-curved grana end membranes present in the chloroplasts of other embryophytes (Fig. 6,
inset). Manton (1962) referred to anthocerote grana as ‘pseudograna’ to differentiate them from the true grana found in higher
plants. Channel thylakoids and grana without end membranes are found in all of the species observed to date, indicating that these
features may be diagnostic of hornworts (Crandall-Stotler, 1986). The channel thylakoids were misinterpreted in early electron
microscopic studies (Menke, 1961; Manton, 1962). These workers assumed that the stromal space between two adjacent channel
thylakoids was a swollen thylakoid lumen, i.e. the two outer membranes of adjacent channel thylakoids were assumed to bound the
lumen of the thylakoid. The advent of improved fixation techniques which preserved more of the stromal matrix reconciled this
misconception (Wilsenach, 1963). The immunocytochemical localization of Rubisco in this space in Megaceros (Vaughn et at.,
1990a) is further proof that it is stroma, not a swollen lumen.

ittle is known of the molecular organization of the photosynthetic apparatus in the anthocerotes. So-called ‘green’ gels, in
which the thylakoid proteins are partially denatured, reveal similar chlorophyll proteins to those found in higher plants [the P700
chlorophyll a protein, the chlorophyll a/b light-harvesting complex, and the CP47 and CP49 polypeptides of photosystem (PS) II
(Vaughn, unpublished)]. Cytochemical localization of PS I activity indicates that the channel thylakoids and the outer thylakoids of
the grana stack are enriched in PS I (Fig. 7A).
he reduction of the osmiophilic tetrazolium thiocarbamyl nitroblue tetrazolium, which accepts electrons from PS II
(Vaughn, 1987), occurs mainly in the partition region of the grana stacks. Thus, the distribution of photosystems in the grana stacks
is similar to that in higher plants and most algae, the channel thylakoids being the equivalent of the stroma lamellae of higher plant,
connecting grana and enriched in PS I.
he only other thylakoid protein investigated in the anthocerotes is polyphenol oxidase. Phaeoceros, Notothylas, and
Anthoceros spp. have polyphenol oxidase activity as measured by spectrophotometric enzyme activity, gel activity stain, and
cytochemistry (Sherman, Vaughn & Duke, 1991). Polyphenol oxidase is found in some charalean algae and most higher plants
(Sherman et at., 1991), but the function of the protein is not known (Vaughn, Lax & Duke, 1988).
Figure 6. Immunocytochemical localization of Rubisco in (A) Sphaerosporoceros adscendens and (B) Megaceros
flagellaris. In the pyrenoid-containing S. adscendens, the immunogold labelling is concentrated over the pyrenoid whereas the
labelling is found throughout the stroma in M. flagellaris; g, granum. (C) Electron micrograph of thylakoid membranes of
Anthoceros punctatus showing the characteristic channel thylakoids (arrows) that connect tops and bottoms of the grana (G).
Inset: electron micrograph of thylakoid membranes of the higher plant Vicia faba showing the highly-curved grana end membranes
(absent in anthocerotes) and stroma lamellae (arrows) connecting the grana stacks in the long axis parallel to the stack. Bars; (A, C,
D) O5,im; (B) 1O,um.

(d) Starch
In most of the pyrenoid-containing anthocerotes, a layer rich in starch granules surrounds the pyrenoid (Fig. 1 A). Mature
chloroplasts are often so full of starch, that the pyrenoid is difficult to discern by light microscopy. Valentine et at. (1986) found no
starch in two species of Dendroceros and only small starch grains were observed in sporophyte plastids and adjacent gametophytic
plastids of D. tubercutaris (Ligrone & Renzaglia, 1990). More abundant starch is found in sporophyte chlorenchyma in a number of
Dendroceros species. Small starch grains are observed around the pyrenoids in chloroplasts of gametophytes of D. tubercularis, D.
japonicus, and D. javanicus, but these are relatively rare, regardless of the culture of the plants or the part of the thallus examined.
Valentine et at. (1986) speculated that the unusual lipid bodies found in the cytoplasm may represent an alternative form of storage
product.
(e) Chtoroptast number, size and shape
lthough the uniplastidic condition of gametophyte cells is to be considered one of the most distinctive features of the
anthocerotes, some exceptions are known. Notably, in the genus Megaceros, the apical cell and young epidermal cells have a single
chloroplast, but mature epidermal cells generally contain 2—4 chloroplasts. The parenchyma cells are generally at least biplastidic,
but occasionally they may contain up to 14 distinct chloroplasts (Burr, 1970; Campbell, 1982). Exceptions to the uniplastidic
condition have also been reported in other genera, e.g. Phaeoceros coriaceus, Phaeoceros hallii, Fotioceros fuciformis, Anthoceros
fusiformis, Anthoceros formosae (Bartlett, 1928; Bharadwaj, 1958; Campbell, 1982; Hasegawa, 1984). Gametophyte cells around
the archegonia may have two chloroplasts in species with uniplastidic cells (Schuster, 1984).
As to the chloroplast number in sporophytic cells, an incorrect idea that the hornworts generally have two chloroplasts in
sporophytic cells, may have been accepted without any substantiating observations. Indeed, we can see such descriptions in some
textbooks (Lotsy, 1909; Campbell, 1918; Smith, 1955; Parihar, 1962). Chloroplast number in sporophytic cells varies in each
species, however. For example, Anthoceros husnotii, Anthoceros crispus, Anthoceros hawaiensis, Dendroceros tubercularis,
Folioceros appendiculatus, Folioceros funciformis, Phaeoceros laevis, and Phaeoceros miyakeanus have a single chloroplast in
sporophytic cells (Scherrer, 1915; Bartlett, 1928; Bharadwaj, 1958; Hasegawa & Wada, unpublished), where such species as
Anthoceros formosae, Anthoceros fusiformis, Anthoceros punctatus, and Anthoceros subbrevis contain two chloroplasts in their
sporophytic cells (Bartlett, 1928; Bharadwaj, 1958; Hasegawa & Wada, unpublished). Moreover, Bartlett (1928) reported the four
chloroplast condition in sporophytic cells of Phaeoceros haliji and Phaeoceros pearsonii. In Megaceros spp. where gametophyte
cells contain more than two chloroplasts, the chioroplast number in sporophytic cells could be as many as twelve (Bartlett, 1928).
The shape of the anthocerote chloroplast is quite variable. In the genera with multilayered thalli, notably Phaeoceros, Anthoceros
and Folioceros, the upper epidermal cells contain the most well- developed chloroplasts. These are of giant sizes (over 70 pm long)
and typically consist of a large central portion, containing the pyrenoid and starch, and a relatively thin peripheral area (Fig. 1 A).
The peripheral area is highly pleomorphic and is responsible for the great variability in chloroplast shape. This is particularly
evident in the chloroplasts of Dendroceros (Fig. 3, and inset). In Dendroceros and Megaceros, the chloroplasts are often very
closely associated with the nucleus, that generally lies in a constriction of the dumbbell-like chloroplast (Fig. 7B). Some of the most
unusual plastid shapes are found in Phaeoceros coriaceus, whose chloroplasts may be dumbbell-shaped, divided into two halves
except by a fine thread, or oddly divided at angles (Fig. 5 B, and inset). Similar unusual plastid shapes have been observed in Ant
hoceros formosae (Bharadwaj, 1958; Hasegawa, 1984).
Some of the variability in chloroplast structure may be due to light-dependent movements. For example, Burr (1968)
found that several Megaceros species had very photoactive chloroplasts that were laminate under low light intensities but contracted
into elliptical structures under high light intensities. The chloroplast morphology seems to be modulated by culture conditions as
well (Valentine, 1984). The pleomorphism of the peripheral portions of chloroplasts in such genera as Phaeoceros, Anthoceros and
Folioceros may be related to the absence of starch and pyrenoid in these areas. Whereas the stroma and thvlakoids may respond
readily, the starch and pyrenoid, being less malleable, will resist movements. Campbell (1971) speculated that photo- activity may
be one reason why Megaceros chloroplasts have lost pyrenoids.

Figure 7. For legend see opposite.
Figure 7. Cytochemical detection of photosystem I activity in Phaeoceros laeuis. Reaction is present on the channel thvlakoids
and the tops and bottom of the grana stacks (arrows). (B) Chloroplast of the Megaceros species from North Carolina in which the
nucleus (N) is associated with a constriction in the centre of the chloroplast. (C) Chloroplast (C) in a cell adjacent to a Nostoc (N)
colony of Notothylas orbicularis are much less- developed than those found in other portions of the thallus. Thylakoids are mainly
in doublets and pyrenoids are absent. Bars: (A) 05 /im; (B) 5() /im; (C) 2() pm.

III. VARIATIONS IN CHLOROPLAST (PLASTID) U L T R A STRUCTURE
In the genera with a multilayered thallus, the chloroplasts of the lower epidermal cells are smaller and much less
pleomorphic than in the upper epidermal cells (Burr, 1968; Valentine, 1984).
The single apical cell of the thallus has chloroplasts with a well developed thylakoid system (Duckett & Renzaglia,
1988). This appears to be a general feature of lower embryophytes, including mosses, hepatics and most pteridophytes, and is in
stark comparison to the underdeveloped proplastids of young apical cells of higher plants (Wellburn, 1987). Plastids with a reduced
thylakoid system are, however, of frequent occurrence in gametophyte and sporophyte cells of anthocerotes.
In all of the genera we and others have examined, the thallus cavities harbouring Nostoc colonies are outlined by a layer
of cells that contain small pleomorphic plastids with few thylakoids, no starch, and pyrenoids reduced or absent (Fig. 7C, Duckett et
at., 1977; Honegger, 1980). Similarly, the plastids in cells of Phaeoceros infected by mycorrhizal fungi are less developed than the
chloroplasts of uninfected parenchyma cells and lack starch and a pyrenoid (Ligrone, 1988). Most likely, these changes in plastid
structure reflect physiological interactions with the symbionts (Duckett et at., 1977; Ligrone, 1988).
Several species of anthocerotes form tubers during conditions that are unfavourable for growth. Ligrone & Lopes (1989)
examined the ultrastructural changes that occur during the formation of tubers in Phaeoceros laevis. Chloroplast structure is
remarkably well preserved even at final stages of tuber formation, with extensive depositions of starch, perhaps as a reserve
substance for regrowth when conditions become more favourable.
Rhizoids may develop from any epidermal cell (Renzaglia, 1978). As the rhizoid differentiates, the chloroplast of the
epidermal cell fragments into beadlike structures that spread throughout the cytoplasm. Whether these are truly multiple plastids or a
single plastid connected by thin connections is not known. Concomitantly, the chloroplast dedifferentiates, losing both the pyrenoid
and the thylakoids. Starch is still present. No green colour or bright red fluorescence, indicative of chlorophyll, is found in the
mature rhizoid of most species, indicating that the plastids are leucoplasts or amyloplasts at maturity or are lost entirely. Some
chlorophyll is observed in rhizoids of Dendroceros species, especially in those close to the apex.
When angiosperms are exposed to prolonged periods of darkness, etioplasts develop. Ligrone & Fioretto (1987) kept
growing sporophytes ofF. laevis in darkness for prolonged periods and monitored the ultrastructural effects as well as changes in the
chlorophyll content over time. Unlike angiosperms, no prolamellar bodies are noted in the developing hornwort sporophytes, even
after 60 days in darkness. The chloroplasts of dark-grown sporophytes have bigger granal stacks and less numerous channel
thylakoids than in the light-grown controls. Paracrystalline membrane arrays, faintly reminiscent of prolamellar bodies, are found
occasionally in developing chloroplasts but disappear during development.
One of the greatest changes to the anthocerote chloroplast occurs during the development of the male gametes. The
plastids in sperm cells of all of the anthocerotes examined are much smaller than the chioroplasts of the thallus (Renzaglia &
Carothers, 1986). At the mid-spermatid stage, the chloroplast is already very reduced, containing a small tear-shaped starch grain
and a few vesicle-like thylakoids (Fig. 8A). At maturity, other than a tiny bit of stroma and the starch grain, there is no substructure
to the chioroplast (Fig. 8 B). The mature spermatozoids of Notothylas and Phaeoceros lack plastid DNA (Renzaglia & Duckett,
1989). Antheridial chambers are obvious on the thallus surfaces as small patches of a bright orange colour in some of the genera
(e.g. Phaeoceros, Notothytas). The orange colour is due to a layer of chromoplast-containing cells, the antheridial jacket, which
surrounds the spermatids (Duckett, 1975). The jacket cell plastids are similar to chromoplasts found in higher plants in that they
have fewer thylakoids than the chloroplasts of the adjoining gametophytic tissue and an abundance of plastoglobuli, which
presumably contains the carotenoid pigments (Fig. 8C). In other genera (e.g. Megaceros), the antheridial jacket cells contain
chioroplasts rather than chromoplasts.
There are no previous reports on the ultrastructure of the female gametangia, although there have been several light
microscopic studies (e.g. Renzaglia, 1978). Ultrastructural investigations of archegonia of Phaeoceros laevis (Renzaglia,
unpublished) reveal the presence of long, flat plastids with a rudimentary thylakoid system in the egg cells (Fig. 9A). The plastids of
sterile archegonial cells, such as the neck and ventral canal cells, are similar in structure but also contain small pyrenoids and a
slightly more elaborate thylakoid system than the egg cell. The presence of long, small-diametered well-developed chloroplasts in
the egg cells, coupled with the absence of chloroplast DNA and the small size of the plastids in the male gametes, indicate the
maternal inheritance of chloroplast DNA in the anthocerotes, as in most higher plants (Vaughn et at., 1980), although dissimilar to
the paternal inheritance of plastids in gymnosperms (Sears, 1980).
There is much greater variation in plastid structure in the sporophyte than in the gametophyte tissue. At the
sporophyte/gametophyte junction in Phaeoceros (Gambardella, Ligrone & Castaldo, 1981; Gambardella & Ligrone, 1987),
Anthoceros (Chauhan & Schraudolf, 1986) and Fotioceros (Vaughn & Hasegawa, unpublished), the gametophyte transfer cells
contain pleomorphic plastids with a reduced thylakoid system and are generally devoid of starch (Fig. bA). The adjoining haustorial
cells of the sporophyte exhibit large plastids with a rudimentary inner membrane system, a scarcely recognizable pyrenoid, abundant
membranous envelope-associated structures similar to peripheral reticulum, and large starch grains of spheroidal shape (Fig. 10B).
In Dendroceros tubercutaris (Ligrone & Renzaglia, 1990), the gametophytic transfer cells have well-developed
chloroplasts, whereas the haustorial cells have large plastids lacking thylakoids, pyrenoid starch, and containing numerous swollen
vesicles. The opposite situation is found in Notothylas, where the haustorial cells have well-developed chloroplasts with abundant
starch and a distinct pyrenoid, whereas the gametophytic transfer cells have plastids with a reduced thylakoid system and no
recognizable pyrenoid (Ligrone, Renzaglia & Duckett, unpublished). These variations in plastid structure may possibly reflect
differences in the physiological relationships between the two generations (Ligrone & Renzaglia, 1990). The plastids from cells of
the basal meristem have a similar morphology: long and flat plastids with somewhat swollen-appearing thylakoids, generally in
doublets or small grana stacks, and a distinct pyrenoid, in species with pyrenoids. Chioroplasts of both the sporophyte wall and the
columella cells are similar to those of the gametophyte, except that the thylakoid system is less elaborate. At early developmental
stages, spores (or spore mother cells) have relatively well-developed chloroplasts that contain both thylakoids and pyrenoids (Fig.
bC), but later the plastids are converted to amyloplasts in most of the hornwort species (Fig. 1OD, see also Ajiri & Ueda, 1986).
Similar plastid morphologies are noted in elater mother cells. The spores of Dendroceros are unique amongst anthocerotes in that
they become multicellular before release from the capsule and appear green because of the development of chloroplasts (Renzaglia,
1978). Chloroplasts in multicellular Dendroceros spores are similar to those found in the later development of the gametophytic
tissue.
Although the plastids of the sporophyte basal meristem are recognizable as chloroplasts, there is a gradient of
development from the basal meristem to the more mature chloroplasts found in the columella and chlorenchyma cells. Wilsenach
(1963) utilized this gradient of development to follow chloroplast ontogeny in Anthoceros ecktonii. His data indicate that thylakoids
may develop from ingrowths of the inner chioroplast envelope, the stacks form from folding of lamellae, and channel thylakoids
from fusion of lamellae between stacks. Ligrone & Fioretto (1987) were able to repeat these observations using more modern
fixation protocols on Phaeoceros laevis. In higher plants, there has been some doubt whether the inner-envelope derived vesicles in

fact contribute to thylakoid development (Vaughn & Duke, 1987). In anthocerotes, the connections between welldeveloped
thylakoids and envelope vesicles makes the role of these vesicles in thylakoid development more tenable.
Figure 8. Ultrastructural characteristics of plastids in antheridia. (A) Plastid (P) from a mid-stage spermatid of
Megacerosflagellaris. Only a few membranes (arrows) and a starch grain (s) are noted. (B) Mature spermatid of Folioceros
fuciformis with plastid reduced to a single envelope-covered starch grain. N, nucleus; f, flagella. (C) Chromoplast from the
antheridial jacket cell of Folioceros fuciformis. Although some thylakoids remain in the plastid, there are large numbers of
plastoglobuli (arrows), probably the site of carotenoid accumulation. Bars; (A, B) O2pm; (C) 1Opm.

Figure 9. Archegonial plastids of Phaeoceros laevis are relatively long and flat structures (arrow) with a reduced membrane system.
These are much more developed than the plastids found in the spermatids (Fig. 8B), however. Ventral canal cells (VC) have similar
plastids to those in the archegonia. Stages of chloroplast division in the sporophyte meristem of Folioceros fuciformis. (B)
Association of the nucleus (N) with a constriction in the chloroplast. Note the accumulation of electron-opaque material on the
envelope at the site of this constriction (arrows). (C) In addition to the constriction ring, a band of filaments (arrow) is also noted in
the chloroplast. Bars: (A) 20 pm; (B, C) 05 pm.

Figure 10. Ultrastructure of plastids from the placenta and sporophyte. (A) Transfer cell plastid of Phaeoceros laevis with a
multiple pyrenoid (p) but with a greatly reduced thylakoid system. (B) Plastid from a haustorial cell of the sporophyte of Notothylas
breutelii with very few internal membranes and a vacuole-like inclusion (*). (C) Plastid from a young spore of Folioceros
fuciformis, still with some thylakoids. The plastid encircles the nucleus (N). (D) Plastid from a nearly mature spore of Notothylas
orbicularis with much less electron-opaque stroma, only remnants of thylakoids (arrows) and a pyrenoid (p). Bars: (A, B) 05 /Lm;
(C, D) 20 ,um.

IV. CHLOROPLAST DIVISION
Because many anthocerote species have uniplastidic cells, it is not surprising that the division of the single plastid would
be tightly controlled in order to prevent the formation of apoplastidic and biplastidic cells by unequal cytokinesis. Lander (1935)
described the division process from light microscopic studies of Z\Totothylas orbicularis. At early stages, the chloroplast is closely
associated with the nucleus and is perpendicular to the future cell plate. The chloroplast then elongates and forms a central
constriction with the two halves lying along opposite cell walls. Nuclear division follows and the formation of a cell plate eventually
completes the chloroplast division.
Lander’s light microscopic study is a surprisingly complete description, whose details have recently been substantiated
and expanded with the use of immunofluorescence and electron microscopy. At the end of the chloroplast constriction, a ring of
electron-opaque material accumulates (Fig. 9B). This is the first description of a constriction ring in the anthocerotes, although
similar rings have been observed in chloroplast division in many plants (Oross & Possingham, 1989; Tewinkel & Volkman, 1987;
Kuroiwa, 1989). It is assumed that all of these constriction rings are actin. In addition, anthocerote chloroplasts contain a bundle of
fibrillar material similar in structure to actin microfilament that is oriented in the direction of division in the plastid as well (Fig.
9C). McCurdy & Williamson (1987) detected a protein that was recognized by monoclonal anti-actin in purified pea chioroplasts,
but similar experiments on Phaeoceros plastids were negative (McCurdy, personal communication). Brown & Lemmon (1985,
1988) showed that the preprophase band of microtubules transects the plane of chloroplast division. Other microtubules, forming an
axial microtubule system, are oriented parallel to the long axis of the plastid (Brown & Lemmon, 1989) and appear to have a role in
plastid division. During mitosis, the nearly divided chloro plas is positioned with a mass concentrated at each pole connected by a
constricted region. Spindle microtubules are organized at the tips of the chloroplast envelope, indicating that this area is a
microtubule organizing centre (Fig. 12A). There fore successful chloroplast division is a requisite for nuclear division, ensuring that
no apoplastidic cells are formed, as the chloroplast serves as a spindle microtubule organizing centre. As cytokinesis is completed
by the formation of a cell plate, the plastid division is completed. The formation of the cell plate in hornworts is produced by a
phragmoplast, like other embryophytes and some green algae (Stewart & Mattox, 1975) rather than a phycoplast like the remaining
green algae (Stewart & Mattox, 1975). Peroxisomes, or perhaps a highly branched peroxi some occur at the cell plate, as is typical
of both advanced algae and higher plants (Owen, 1983). In both the light microscopic studies of Lander (1935) and the electron
microscopic and immunofluorescent studies of Brown & Lemmon (1985, 1988) and Owen (1983), dividing cells were observed in
the basal meristem of the sporophyte. In the apical cell of the gametophyte, a similar pattern of chloroplast division is noted,
although the chloroplasts are more developed than those at the sporophyte meristem (e.g. Fig. 11). Investigations of meiotic cells
also reveal the involvement of plastids in meiotic divisions as spindle microtubule organizing centres (Brown & Lemmon, 1990).
No examples of plastid division by methods other than median constriction (Vaughn et al., 1990b) have been observed in
anthocerotes.
In the genus Megaceros, most of the thallus cells are multiplastidic, with the exception of the apical cell and some
epidermal cells. Division of the chioroplast appears similar to that in uniplastidic cells, however. One of the chloroplasts associates
with the nucleus and the plane of the subsequent division is set (Burr, 1968). This ensures at least one plastid in each new cell after
cytokinesis even though all of the other chioroplasts may end up in a single cell. Thus, the mechanism of a coordinated nuclear and
chloroplast division appears to be ubiquitous in the anthocerotes and is perhaps of taxonomic importance in defining
Anthocerotophyta.
Like most bryophytes, the anthocerote gametophytes have a remarkable ability to regenerate whole plants from parts of
the old. Burr (1969) investigated the changes in chloroplast number and division pattern during the regeneration process in
Megaceros flagellaris. Surprisingly, she found that, during formation of new apical cells from the multiplastidic cells of the thallus,
the chloroplast number was reduced to one. This was achieved through cell divisions without plastid division, walling off of plastids
without the nucleus, and chloroplast fusion.
V. ASSOCIATIONS WITH OTHER ORGANELLES
Besides the associations with the nucleus and microtubules (above), several other associations of anthocerote
chloroplasts with other organelles are worthy of note.
In higher plants, peroxisomes are often associated with the chloroplast envelope. It is believed that this close relationship
is due to the presence of the important photorespiratory enzyme, glycollate oxidase, in the peroxisome (Vaughn, 1990). Similarly,
peroxisomes are often found associated with the anthocerote chloroplast envelope and have been detected both in species that have
and species that lack pyrenoids. Peroxisomes are also present during chioroplast and nuclear division, generally at the plastid
isthmus (Owen, 1983; Brown & Lemmon, 1985).
In developing plastids of the sporophyte, arrays of both smooth and rough endoplasmic reticulum (ER) are associated
with the chloroplast envelope. This plastid-associated ER was first noted by Ligrone & Fioretto (1987) in Phaeoceros laevis and has
been found in all of the developing sporophytes of anthocerotes examined by us as well. Ligrone & Fioretto (1987) speculated that it
may have a role in plastid development, possibly for the transport of lipids and/or proteins into the chloroplast.

Figure 11. Early stages of plastid division (A) and telophase nuclear division (B) figures in the apical cell of a Folioceros sp. from
Japan. N, nucleus; M, mucilage; cp, cell plate; C, chloroplast; bar, 20 1um.

Figure 12. (A) Metaphase figure in which the termini of the microtubules is found on the chioroplast envelope (*) which appears to
be a microtubule organizing centre. Arrows mark microtubules; C, chromosome. (B) Chioroplast of Coleochaete scutata contain a
pyrenoid in which only single thylakoids cross the pyrenoid area (P). N, nucleus; bars, 2Opm.
VI. EVOLUTIONARY CONSIDERATIONS
Recently, there has been much discussion of the possibility of Coleochaete representing the extant green algal genus
most like the progenitor of higher plants (Delwiche et al., 1989). Unfortunately, from the present viewpoint, most of the
investigations of these algae have centred on aspects of the biology of these plants other than the chioroplast morphology and
development. Coleochaete chloroplasts (Fig. 12) share many characteristics with the anthocerotes, especially Notothylas. The granal
stacks do not have end membranes in Coleochaete, but rather are pseudograna as is found in the anthocerotes. In addition, the
thylakoids are arranged in channels, a situation which, among embryophytes, is noted only in the anthocerotes. The pyrenoid of
Coleochaete is globular and is transected into subunits by single thylakoids. Most of these pyrenoids have very little non-pyrenoid
stroma intercalating the subunits of the pyrenoid. This pyrenoid structure is very similar to that found in Notothylas orbicularis or N.
breutelii: a globular structure dissected only by stroma lamellae. Although many bryologists consider that the genus Notothylas is
advanced (i.e. its small sporophyte enclosed within the gametophyte has evolved by reduction from a more elaborate sporophyte
like that found in the other anthocerote genera), the opposite situation could be true. In fact, using cladistics, Mishler & Churchill
(1985) concluded that the most parsimonious cladogram is one in which Notothylas is separated from the anthocerotes, and set in a
closer affiliation with the hepatics and charalean algae than the other anthocerotes. However, it must be noted that similarities in

sperm ultrastructure (Renzaglia & Duckett, 1989) and placental organization (Ligrone et al., unpublished) are indicative of a close
relationship between Notothylas and other anthocerotes.
Duckett & Renzaglia (1988) examined micro- graphs from published work of a number of charalean algae on the line to
higher plants, as postulated by Stewart & Mattox (1975). These authors also observe that the Coleochaete chloroplast is structurally
similar to those of anthocerote species, but they note a particularly close similarity to the chloroplasts of Dendroceros and
Megaceros (rather than Notothylas). Although the plastids in Coleochaete are similar in shape to those observed in Megaceros, our
investigations reveal that the pyrenoid of Coleochaete is very similar to those found in Notothylas. The Dendroceros pyrenoid has
highly distinctive inclusions and is quite different in shape from the Coleochaete pyrenoid.
Both Kaja (1954) and Burr (1970) presented possible evolutionary schemes for the anthocerotes based upon chloroplast
structure at the light or electron microscopic level, respectively. In Kaja’s (1954) scheme, the Notothylas chioroplast is considered
the most algal-like, with various other pyrenoid configurations as intermediate, and the pyrenoid-less (interpreted by Kaja as many
dispersed pyrenoids) Megaceros fiagellaris as the most advanced. Burr (1970) examined mainly species of Megaceros and found
variations in chioroplast number and size. However, her analysis of chloroplast ultrastructure in these species appears flawed
because the areas she described as ‘pyrenoids’ appear to be nothing more than starch-free zones of stroma. Ultrastructural studies of
Valentine et al. (1986) and us have revealed no pyrenoids in some of the same species referred to by Burr (1970) as having a
pyrenoid actually lack a pyrenoid. In our studies, a series of variations in pyrenoid ultrastructure may be constructed similar to
Kaja’s scheme. This scheme draws an evolutionary sequence which is difficult to reconcile with the many other morphological
characteristics that indicate that the sporophvte of Notothylas is derived, and not ancestral (Renzaglia, 1978). In addition,
ultrastructural investigations are still restricted to a small range of species and clearly there is a large variation in plastid
ultrastructure, including apparently pyrenoidless species in the genus Notothylas (Udar & Singh, 1981). Thus, despite the number of
anthocerote species examined for chloroplast structure, drawing any kind of phylogenetic scheme even for this one character may be
premature.
The anthocerote chloroplast is clearly unique among the archegoniates and these observations support the view that the
anthocerotes are an isolated taxonomic group. Crandall-Stotler (1980, 1986) and Schuster (1984) argue on morphological criteria
that the anthocerotes should be placed in their own division, the Anthocerotophyta. The presence of uniplastidic cells in non-
meristematic cells of the gametophyte, multiple pyrenoids, pseudograna, and channel thylakoids are all chloroplast characters found
nowhere else in the embryophytes and certainly not at all in the hepatics, a group with which the anthocerotes traditionally have
been linked.
VII. TAXONOMIC USES AND CONSPECTUS
Pyrenoid structure is extremely variable and may be an aid in the taxonomy of this group. For example, the Notothylas
spp. have rather rounded pyrenoids and irregularly shaped subunits. In Dendroceros, the pyrenoid subunits are of irregular shape,
with numerous electron-opaque inclusions (pyrenoglobuli) surrounded by electron-opaque threads. The Folioceros pyrenoids also
have pyrenoglobuli, but only along the periphery of the subunits, and the subunits are similar in shape to that found in Anthoceros,
Sphaerosporoceros, and Phaeoceros. Although Folioceros has been disputed as a good genus by some (Schuster, 1987; Hasegawa,
1988), the presence of a concentration of pyrenoglobuli at the periphery of the pyrenoid subunits is a potential taxonomic marker for
this taxon. In Megaceros, no pyrenoids were detected in any of the species examined. As additional taxa are examined, these
generalities or generic parameters in chloroplast structure may not prove to be absolute.
Chloroplast structure and number were utilized in a recent taxonomic revision by Hasegawa (1988). Megaceros
giganteus (formerly Dendroceros giganteus) has a conspicuous costa and unistratose wings, an anatomical organization that closely
recalls that found in Dendroceros species. However, mature gametophyte cells of M. giganteus are multiplastidic (Hasegawa, 1988)
and the chloroplasts lack a pyrenoid (Valentine et al., 1986). Thus, the grouping of this species in Megaceros reflects the greater
similarity of the chloroplasts of this species with Megaceros than with Dendroceros.
This review has centred on morphological features but this is in large part because there are almost no
physiological/biochemical studies on the anthocerote chloroplast. Anthocerotes are not found abundantly in the field (except in
unique localities such as the cool, wet areas of New Zealand) and, even under optimal conditions, material at the same physiological
state in sufficient quantity to perform physiological/biochemical studies is difficult to obtain. The thylakoids of the hornworts
represent a unique system of membranes and it would be desirable to learn how the various pigment—protein complexes are
arranged in the membranes. Immuno cytochemica techniques, already used to localize Rubisco in the pyrenoid (Vaughn et at., 1990
a), may be useful in determining the distribution of the complexes in the photosynthetic apparatus. Similarly, it may be helpful to
determine if other carbon- fixation enzymes besides Rubisco are present in the pyrenoids of the anthocerote. Thus,
immunocytochemistry offers a solution to the problem of limited tissue of a given physiological state and should be invaluable in
determining many other facets of the anthocerote chloroplast.
ACKNOWLEDGEMENTS
Thanks are extended to Ms Lynn Libous-Bailey and to Ms Ruth H. Jones for their excellent technical assistance during
the course of these experiments. Drs Barbara Crandall-Stotler, David Glennie, Barbara Polly, Charles Bryson, and Roy Brown
provided samples for some of the studies described herein. Drs R. J. Smeda, T. D. Sherman, W. Pettigrew, and K. G. Wilson
provided helpful comments on the manuscript. Dr Gabriella Häsel de Menendez provided identification of some Central American
species.
Mention of a trademark or product does not constitute endorsement of the product by the US Department of Agriculture
and does not imply its approval to the exclusion of other products that may also be suitable.

REFERENCES
AJIRI, T. & UEDA, R. (1986). Electron microscope observations on the sporogenesis in the hornwort, Anthoceros punctatus L.
Journal of the Hattori Botanical Laboratory 40, 1—26.
BARTLETT, E. M. (1928). A comparative study of the development of the sporophyte in the Anthocerotaceae, with special
reference to the genus Anthoceros. Annals of Botany 42, 409—430.
BHARADWAJ. D. C. (1958). Studies in Indian Anthocerotaceae. II. The morphology of Anthoceros cf. gemmulosus (Hattori)
Pande. Journal of the Indian Botanical Society 37, 7 5—92.
BROWN, R. C. & LEMMON, B. E. (1985). Preprophasic establishment of division polarity in monoplastidic mitosis of hornworts.
Protoplasma 124, 175—183.
BROWN, R. C. & LEMMON, B. E. (1988). Preprophasic microtubule systems in the development of the mitotic spindle in
hornworts (Bryophyta). Protoplasma 143, 11 21.
BROWN, R. C. & LEMMON, B. E. (1989). Morphogenetic plastid migration and spindle dynamics in simple land plants.
Proceedings of the 47th Annual Meeting of the Electron Microscopy Society of America 764—765.
BROWN, R. C. & LEMMON, B. E. (1990). The quadripolar microtubule system and meiotic spindle ontogeny in hornworts
(Bryophyta: Anthocerotae). American Journal of Botany 77, 1482 1490.
BURR, F. A. (1968). Chloroplast structure and division in Megaceros species. Ph.D. Dissertation, University of California,
Berkeley.
BURR, F. A. (1969). Reduction in chioroplast number during gametophyte regeneration in Megaceros flagellaris. The Bryologist
72, 200—209.
BURR, F. A. (1970). Phylogenetic transitions in the chioroplasts of the Anthocerotales. I. The number and ultrastructure of the
mature plastid. American Journal of Botany 57, 97 110.
CAMPBELL, D. H. (1918). The Structure and Development of the Mosses and Ferns. 3rd edn, pp. 1—708. New York.
CAMPBELL, E. H. (1907). Studies on some Javanese Anthocerotaceae. I. Annals of Botany 21, 467—486.
CAMPBELL, E. 0. (1971). Problems in the origin and classification of bryophytes with particular reference to liverworts. New
Zealand Journal of Botany 9, 6 78—688.
CAMPBELL, E. 0. (1982). Notes on some Anthocerotae of New Zealand (2). Tuatara 25, 65 70.
CHAUHAN, E. & SCHRAUDOLF, H. (1986). Ultrastructural studies on the placental region in Anthoceros punctatus L. Beitrage
zur Biologie der Pfianzen 61, 357—372.
CRANDALL-STOTLER, B. (1980). Morphogenetic designs and a theory of bryophyte origins and divergence. Bioscience 30,
580—584.
CRANDALL-STOTLER, B. (1986). Musci, hepatics, and anthocerotes — An essay on analogues. In: New Manual of Bryology, vol.
II (Ed. by R. M. Schuster), pp. 1093—1129. Hattori Botanical Laboratory, Nichinan, Japan.
DELWICHE, C. F., GRAHAM, L. E. & THOMPSON, N. (1989). Lignin-like compounds and sporopollenin in Coleochaete, an
algal model for land plant ancestry. Science 245, 399—401.
DUCKETT, J. G. (1975). An ultrastructural study of the differentiation of antheridial plastids in Anthoceros laevis L. Cytobiologie
10, 432—438.
DUCKETT, J. G., PRASAD, A. K. S. K., DAvIEs, D. A. & WALKER, S. (1977). A cytological analysis of the Nostoc—bryophyte
relationship. New Phytologist 79, 349—362.
DUCKETT, J. G. & RENZAGLIA, K. S. (1988). Ultrastructure and development of plastids in the bryophytes. Advances in
Bryology 3, 33—93.
DUCKETT, J. G., RENZAGLIA, K. S. & SMITH, E. A. (1988). Preparative techniques for transmission electron microscopy of
bryophytes. In: Methods in Bryology (Ed. by J. M. Glime), vol. 1, pp. 181 192. Hattori Botanical Laboratory, Nichinan, Japan.
ENDERLIN, C. S. & MEEKS, J. C. (1983). Pure culture reconstruction of the Ant hoceros—Nostoc symbiotic association. Planta
158, 157—165.
GAMBAREIELLA, R. & LIGRONE, R. (1987). The development of the placenta in the anthocerote Phaeoceros laevis (L.) Prosk.
Planta 172, 439—447.
GAMBARDELLA, R., LIGRONE, R. & GASTALDO, R. (1981). Ultra-structure of the sporophyte foti in Phaeoceros. Crytogamie,
Bryologie et Lichenologie 2, 23—45.
HASEGAWA, J. (1984). Taxonomical studies on Asian Anthocerotae. IV. A revision of the genera Anthoceros, Phaeoceros and
Folioceros in Japan. Journal of the Hattori Botanical Laboratory 57, 242 272.
HASEGAWA, J. (1988). A proposal for a new system of the Anthocerote, with a revision of the genera. Journal of the Hattori
Botanical Laboratory 64, 87—95.
HASEL 05 MENNDEZ, G. G. (1988). A proposal for a new classification of the genera within Anthocerotophyta. Journal of the
Hattori Botanical Laboratory 64, 71—86.
HONEGGER, R. (1980). Zytologie der Blaualgen-HornmoosSymbiose bei Anthoceros laevis aus Island. Flora 170, 290—302.
KAJA, H. (1954). Untersuchungen uber die Chromatophoren und Pyrenoide der Anthocerotales. Protoplasma 44, 136—153.
KUROIWA, T. (1989). The nuclei of cellular organelles and the formation of daughter organelles by the ‘plastid-dividing ring’. The
Botanical Magazine (Tokyo) 102, 291—329.
LANDER, C. A. (1935). The relation of the plastid to nuclear division in Anthoceros laevis. American Journal of Botany 22,
42—5 1.
LIGRONE, R. (1988). Ultrastructure of a fungal endophyte in Phaeoceros laevis (L.) Prosk. (Anthocerotophyta). Botanical Gazette
149, 92—100.
LIGRONE, R. & FIORETTO, A. (1987). Chloroplast development in light- and dark-green sporophytes of Phaeoceros laevis (L.)
Prosk. (Anthocerotophyta). New Phytologist 105, 301—308.
LIGRONE, R. & LoPES. C. (1989). Ultrastructure, development and cytochemistry of storage cells in the ‘tubers’ of Phaeoceros
laevis. Prosk. (Anthocerotophyra). New Phytologist 112, 3 17—325.
LIGRONE, R. & RENZAGLIA, K. S. (1990). The sporophyte— gametophyte junction in the hornwort, Dendroceros tubercularis
Hatt. (Anthocerotophyta). New Phytologist 114, 497—505.
LOTSY, J. P. (1909). Vortrage über botanische Stammegeschichte, vol. 2, pp. 1—902. Jena.
MANTON, I. (1962). Observations on plastid development in the meristem of Anthoceros. Journal of Experimental Botany 13, 325
333.
MCALLISTER, F. (1914). The pyrenoid of Anthoceros. American Journal of Botany 1, 79 95.
MCALLISTER, F. (1927). The pyrenoids of Anthoceros and Notothylas with especial reference to their presence in spore mother
cells. American Journal of Botany 14, 246—257.

MCCURDY, D. W. & WILLIAMSON, R. E. (1987). An actin-related protein inside pea chloroplasts. Journal of Cell Science 87,
449—456.
MCKAY, R. M. L. & GIBBS, S. P. (1989). Immunocytochemical localization of ribulose 1,5-bisphosphate carboxylase/ oxygenase
in light-limited and light-saturated cells of Chlorella pyrenoidosa. Protoplasma 129, 31—37.
MENKE, W. (1961). Uber die chioroplasten von Anthoceros punctatus. S. Mitteilung zur Entwicklungsgeschichte der Plastiden.
Zeitschrift für Naturforschung 16, 334—336.
MISHLER, B. D. & CHURCHILL, S. P. (1985). Transition to a land flora: phylogenetic relationships of the green algae and
bryophytes. Cladistics 1, 305 328.
OROSS, J. W. & POSSINGHAM, J. V. (1989). Ultrastructural features of the constricted region of dividing plastids. Protoplasma
150, 13 1—1 38.
OWEN, H. A. (1983). Cell structure and division in Phaeoceros (Anthocerotae) and its comparison with Charophycean algae. MS.
Thesis, Miami University, Oxford, OH.
PARIHAR, N. S. (1962). An Introduction to Embryophyta. I. Bryophyta, 4th edn, pp. 1—377. Allahabad.
RENZAGLIA, K. S. (1978). A comparative morphology and developmental anatomy of the Anthocerotophyta. Journal of the
Hattori Botanical Laboratory 44, 31—90.
RENZAGLIA, K. S. & CAROTHERS, Z. B. (1986). Ultrastructural studies of spermatogenesis in the Anthocerotales 4. The
blepharoplast and mid-stage spermatid of Notothylas. Journal of the Hattori Botanical Laboratory 60, 97—104.
RENZAGLIA, K. S. & DUCKETT, J. G. (1989). Ultrastructural studies of spermatogenesis in Anthocerotophyta. V. Nuclear
metamorphosis and the posterior mitochondrion of Notothylas orbicularis and Phaeoceros laevis. Protoplasma 151, 137—
150.
SCHERRER, A. (1915). Untersuchungen uber Bau und Vermehrung der Chromatophoren und das Vorkommen von Chondriosomen
bei Anthoceros. Flora 107, 1 56.
SCHOFIELD, W. B. (1984). Introduction to Bryology. Macmillan Publishing Co., New York.
SCHUSTER, R. M. (1984). Morphology, phylogeny, and classific atio of the Anthocerotae. In: New Manual of Bryology, vol. II
(Ed. by R. M. Schuster), pp. 1071—1092. Hattori Botanical Laboratory, Nichinan, Japan.
SCHUSTER, R. M. (1987). Preliminary studies on Anthocerotae. Phytologia 63, 193—201.
SEARS, B. B. (1980). Elimination of plastids during spermatogenesis and fertilization in the plant kingdom. Plasmid 4, 233 255.
SHERMAN, T. D., VAUGHN, K. C. & DUKE, 5. 0. (1991). A limited survey of the phylogenetic distribution of polyphenol
oxidase. Phytochemistry 30, 2499—2506.
SMITH, G. M. (1955). Cryptogamic Botany, vol. 2, 2nd edn, pp. 1—399. New York.
STEINMULLER, D. & TEvINI. M. (1985). Composition of plastoglobuli. I. Isolation and purification from chloroplasts and
chromoplasts. Planta 163, 201—207.
STEWART, K. D. & MATTOX, K. R. (1975). Comparative cytology, evolution and classification of the green algae with some
considerations of the origin of the organisms with chlorophylls a and b. Botanical Review 41, 104—135.
TEWINKEL, M. & VOLKMAN, D. (1987). Observations on dividing plastids in the protonema of the moss Funaria hygrometrica
Sibth. Arrangement of microtubules and filaments. Planta 172, 309 320.
TOYAMA, 5. (1974). Electron microscope studies on the morphogenesis of plastids. IX. The ph> logeny of photosynthetic
apparatus with special reference to lamellar structure. Japanese Journal of Botany 20, 413-438.
UDAR, R. & SINGH, D. K. (1981). Notothylas khasiana Udar et. Singh sp. nov. from Schillong, India. Journal of the Indian
Botanical Society 69, 112—117.
VALENTINE, L. J. (1984). A comparison of selected New Zealand species of Megaceros and Dendroceros (A nthocerotophyta),
with special reference to chioroplast gross morphology and ultrastructure. Honours Thesis, Massey University, Palmerston
North, New Zealand.
VALENTINE, L. J., CAMPBELL, E. 0. & HOPCROFT, D. H. (1986). A study of chloroplast structure in three Megaceros species
and three Dendroceros species (Anthocerotae) indigenous to New Zealand. New Zealand Journal of Botany 24, 1—8.
VAUGHN, K. C. (1987). Two immunological approaches to the detection of ribulose 1,5-bisphosphate carhoxylase in guard cell
chloroplasts. Plant Physiology 84, 188—196.
VAUGHN, K. C. (1990). Subperoxisomal localization of glycolate oxidase. Histochemistry 91, 99—105.
VAUGHN, K. C., DEBONTE, L. R., WILSON, K. G. & SCHAEFFER, G. W. (1980). Organelle alteration as a mechanism for
maternal inheritance. Science 208, 196—198.
VAUGHN, K. C. & DUKE, S. 0. (1987). Transport of proteins into the chloroplasts. In: Models in Plant Physiology, vol. 1 (Ed. by
D. W. Newman & K. G. Wilson), pp. 69—73. CRC Press, Boca Raton, FL.
VAUGHN, K. C., CAMPBELL, E. 0., HASEGAWA, J., OWEN, H. A. & RENZAGLIA, K. S. (1990 a). The pyrenoid is the site of
ribulose I ,5-bisphosphate carboxylase/oxygenase accumulation in the hornwort (Bryophyta: Anthocerotae) chloroplast.
Protoplasma 156, 117—129.
VAUGHN, K. C., HICKOK, L. G., WARNE, T. R. & FARROW, A. C. (1990b). Structural analysis and inheritance of a clumped
chloroplast mutant in the fern Ceratopteris. Journal of Heredity 81, 146—151.
VAUGHN, K. C., LAX, A. R. & DUKE, S. 0. (1988). Polyphenol oxidase: The chioroplast oxidase with no established function.
Physiologia Plantarum 72, 659 665.
WELLRURN, A. R. (1987). Plastids. International Review of C3- tology 17, 149 210.
WILSENACH, R. (1963). Differentiation of the chioroplast in Anthoceros. Journal of Cell Biology 18, 41 9—428.