Abstract
Background and Aims
Botanical literature is filled with studies which tried to demonstrate that leaves of many gymnosperms have an endodermis with Casparian bands or a sheath of sclerified cells around leaf veins. Direct photographic evidence of an endodermis with Casparian bands (strips) is lacking for the leaves of many gymnosperms. Our goal was to confirm with direct evidence an endodermis with Casparian bands or a vein sheath in leaves of representative gymnosperms via histochemical staining and epifluorescence microscopy, while extending previous work by examining understudied petioles and leaf bases.
Methods
We sectioned leaves fresh with razor blades and viewed them unstained or stained, usually with berberine hemisulfate, counterstained with gentian violet, and phloroglucinol HCl. Bright-field, epifluorescence and/or laser confocal microscopies on a Zeiss LSM700 were used for viewing and imaging.
Key Results
Most members of Pinaceae of Pinophyta (13 species of genera Cedrus, Pseudotsuga, Larix, Picea and Pinus) had endodermis with distinct Casparian bands in needles. We identified endodermis in the distal petiolar regions of nine of these same species. Sclerified vein sheaths or partial vein sheaths were observed in 10 species of 41 studied among Cycadophyta (Cycas), Ginkgophyta (Ginkgo) and the Pinophyta Podocarpaceae (Dacrycarpus, Dacrydium), Cupressaceae (Metasequoia), Taxaceae (Amentotaxus) and Pinaceae (Keteleeria, Abies, Tsuga).
Conclusions
Endodermis with Casparian bands is only characteristic of most species of Pinaceae; vein sheaths are found in three genera of Pinaceae and six genera of the other families of gymnosperms, but most gymnospermous leaves lack endodermis and vein sheath, particularly in petiolar and leaf base regions. The presence of endodermis with Casparian bands could have contributed to the adaptation of the Pinaceae to extreme environments; members of genera such as Picea and Pinus are the typical treeline species in many mountain ranges across the world.
Keywords: Endodermis, Casparian bands or strips, conifers, gymnosperms, needles, phylloclades, vein sheath
INTRODUCTION
The leaves of gymnosperms have been variously described as linear with abaxially/adaxially oriented veins, broad with petioles, needles with petioles, cataphylls or scale leaves, cladodes with obliquely oriented veins, or phylloclades with perpendicularly orientated veins and stem-like concentrically arranged vascular bundles (Napp-Zinn, 1966; Bierhorst, 1971; Gifford and Foster, 1989; Dörken and Stützel, 2011; Kaplan, 2022). Within these leaves, structures surrounding the vascular tissues, or veins, have been described and designated in many different ways, but usually structures termed vascular bundle sheaths (vein sheaths) or endodermis have been assigned to them; both of these are historically known.
From Chamberlain (1935) to Napp-Zinn (1966), Esau (1977), Gifford and Foster (1989), Biswas and Johri (1997) and Kaplan (2022), texts and major books have long extolled the characteristics of gymnosperm leaves, including the occurrence of endodermis, especially in leaves of pines and related Pinaceae. In his review less than 30 years ago, however, Lersten (1997) noted that there was no real proof of Casparian bands in an endodermis in gymnosperm leaves, and earlier, in his major tome on gymnosperm leaves, Napp-Zinn (1966, citing mostly much earlier researchers) did not present any photographic evidence of leaf endodermis or of sclerified vein or vascular bundle sheath (or ‘Leitbündelscheiden’, pp. 156–157, 226–227). More recently, the Dörken group has studied many of the families, genera and species of gymnosperms (e.g. Dörken et al., 2010, 2019, 2021, 2023a, b; Dörken and Stützel, 2011, 2012; Dörken, 2012, 2013a, b, 2014; Dörken and Parsons, 2016; Dörken and Lepetit, 2018; Dörken and Nimsch, 2018, 2021), but their techniques did not usually allow direct Casparian band evidence.
Endodermis and vein sheaths are different anatomical structures, although both form the innermost layer of root and shoot cortex or leaf mesophyll, and will be illustrated initially in the first figure of the Results section. The endodermis in vascular plants is a specific structure that has been most characterized in roots. In its simplest form, as described and discussed variously by Peterson and colleagues (Seago et al., 1999; Enstone et al., 2003; Ma and Peterson, 2003), Schreiber and Franke (2011) and Soukup and Tylová (2018), it comprises a ring or layer of cells of the innermost cortex with mostly lignified Casparian bands impregnated in its radial and transverse cell walls: a State I endodermis has Casparian bands only. Endodermis can also form additional chemical structures; a State II endodermis has Casparian bands with associated suberin lamellae within the limits of the primary cell wall and outside the protoplast. Further, endodermis can form a State III endodermis with secondarily lignified wall lining the inside of the suberin lamellae, often with its thickness greatest on the inner tangential wall in root endodermis (and outer tangential wall of an innermost layer of an exodermis, which may form the outermost layer of cortex). An endodermis in stems has also been well documented (Seago, 2020), as has endodermis of some gymnosperm roots (e.g. Wilcox, 1968 for Pinus resinosa; Bonacorsi and Seago, 2016 for Ginkgo biloba; Song et al., 2019 for Cunninghamia lanceolata; Brundrett and Tedersoo, 2020 for various gymnosperms with mycorrhizae).
Importantly, as noted by Enstone et al. (2003), Geldner (2013), Soukup and Tylová (2018) and others, the endodermis regulates the movement of water, dissolved minerals and carbohydrates into and out of the vascular cylinder of roots; further, the endodermis filters and arrests unwanted harmful molecules from being translocated in the xylem strand up to the shoot while at the same time allowing necessary inorganic molecules essential for plant growth. While water and dissolved minerals can pass apoplastically from cell wall to cell wall, symplastically or even transcellularly through the protoplasts of cortex cells, all movement of water and minerals passes through the endodermis only symplastically or transcellularly due to the presence of Casparian bands in the radial/transverse walls of the endodermis. It is known that in the needles of conifers, endodermis also plays an important ecological role in the survival of pines in extreme cold habitats; for example, Stegner et al. (2023) suggested that the lignification of the endodermis blocks the penetration of ice from the frozen vascular tissue into the mesophyll, allowing needles to remain active during extended periods of extreme cold temperatures. This, in turn, may be associated with oxygen and carbon dioxide pathways between vein and mesophyll (Roden et al., 2015; Sage and Khoshravesh, 2016) as well as movement of nutrients like nitrogen (Gao et al., 2023; Valderrama-Martin et al., 2024) and carbohydrate (Liesche et al., 2011) and structural support (e.g. Leegood, 2007; Meicenheimer et al., 2008).
An endodermis with Casparian bands has been well demonstrated in leaf blades or needles in relatively few species, with the most striking examples being in Pinus nigra and P. resinosa of the Pinaceae by Meicenheimer et al. (2008), where P. nigra had prominent Casparian bands and P. resinosa had prominent outer tangential, secondarily lignified walls in its endodermis. Other authors (e.g. Campbell, 1972; Delian and Săvulescu, 2022) have presented photographs of P. nigra clearly showing endodermis, sometimes with well-demonstrated Casparian bands. Often published illustrations are specimens in which Casparian bands do not show, but the cells have a characteristic endodermal shape and pattern, as in Esau (1977) for Larix and Pinus and Gifford and Foster (1989) for Pinus. Using a variety of techniques, Soar (1922) had drawings of leaf endodermis in Pinus, Picea, Abies, Larix, Pseudotsuga, Cedrus and Tsuga. Others (e.g. Hu and Wang, 1984; Eguchi et al., 2004; Bercu et al., 2010; Ghimire et al., 2015a) demonstrated images with endodermis patterns in a variety of Pinaceae, but Casparian bands were not visible. Vein sheath was labelled in Abies (Dörken, 2014; Dörken and Lepetit, 2018); these studies were not suggestive of an endodermis with Casparian bands.
Evidence seems to show that endodermis in gymnosperm leaves may be a more recently derived trait and of interest evolutionarily. Fossil cycads, for example, do not have endodermis in cataphylls or scale leaves (Hermsen et al., 2006), and they may be among the oldest of gymnosperms (e.g. Taylor et al., 2009; Crisp and Cook, 2011; Lu et al., 2014; Lubna et al., 2021). Fossils from basal gymnosperms like Podocarpaceae (Andruchow-Colombo et al., 2023) do not show any evidence of endodermis in leaves. On the other hand, fossil pines, like Pinus hokkaidoensis (Stockey and Ueda, 1986) and P. arnoldii (Klymiuk et al., 2011), are recent in evolutionary appearance and have been shown to have a distinct endodermis (Taylor et al., 2009; Gernandt et al., 2016).
A vein sheath or its modifications consists of cortex sclerenchyma cells surrounding the vascular bundles that are sclerified or lignified without Casparian bands. The occurrence and nature of vein sheaths in gymnosperm leaves have been little addressed in major tomes (e.g. Bierhorst, 1971; Gifford and Foster, 1989; Kaplan, 2022), but they have been variously described as a ‘layer of lignified cells’ in Ginkgo by Dörken (2013b, p. 90), a ‘parenchyma sheath’ or ‘unstratified endodermis with thin-walled large cells’ by Cargo et al. (2000, p. 38), or a zone of ‘scattered thin-walled lignified fibers’ in Cycas by Griffith et al. (2014, p. 241). Vein sheaths are clearly irregular in pattern and structure, but without Casparian bands. In addition to facilitating the movement of water, nutrients, photosynthates and gases between vascular tissues and mesophyll, the vein sheath reinforces and supports the vascular tissues and even the inner mesophyll (Esau, 1977; Leegood, 2007), potentially serving a similar function as endodermis.
Among other gymnospermous families, evidence of endodermis with Casparian bands or vein sheath is spotty and sometimes lacking. Sethi et al. (2020) reported a sclerenchyma sheath in Cycas. Cargo et al. (2000, p. 38) stated ‘an unistratified endodermis with thin-walled large cells’ in Ginkgo biloba, but it was not so described by Duarte et al. (2013); Dörken’s (2013a) bright-field micrographs demonstrated a vein sheath in its lamina. Liesche et al. (2011, p. 184) reported a ‘suberized bundle sheath’ and ‘endodermis-like bundle sheath’ for the same lamina image in Ginkgo, whereas Klimko et al. (2015, p. 185) noted a ‘parenchymatous sheath’ around its veins. Bundle sheaths appear to be present in some members of the Podocarpaceae (Andruchow-Colombo et al., 2024) and not others (Lee, 1952; Griffith, 1957). Endodermis is claimed but not demonstrated in leaves of Cupressaceae by Yousif and Shahbaz (2023); presumed endodermis was also noted in Juniperus (Bercu et al., 2010; Güvenc et al., 2011). In their broad survey of leaf characteristics in conifers (7 families, 37 genera, 111 species), Yao and Hu (1982) reported the presence or absence of endodermis or vein sheath, but their drawings did not demonstrate endodermis with Casparian bands or vein sheaths. The scale leaves or cataphylls reported by many (e.g. Soar, 1922; Sacher, 1955; Kaplan, 2022) have reduced vascular tissue and no apparent endodermis.
Thus, the evidence for endodermis with Casparian bands in gymnosperm leaves is incomplete and uncertain. The situation in petioles or leaf bases has been far less documented, even though Napp-Zinn (1966), for example, devoted much to leaf origin – without a single depiction of a mature leaf with petiole. Lee (1952, p. 395) illustrated by drawing a petiole and its leaf vascular bundle without an ‘endodermal sheath’ in the leaf of Dacrydium taxoides of the Podocarpaceae. An illustration by Cross (1940) showed a vein sheath in the petiolar or proximal region of a Cunninghamia lanceolata leaf. Doak (1935) had only drawings of the fascicle bases in pine and leaf bases in Araucariaceae. Soar’s (1922) drawings of the origin of needles in fascicles of Pinus sylvestris showed no endodermis in lower or proximal levels of the developing needles and an incomplete endodermis in Picea excelsa and in Abies pectinata. Needles of Pinus mugo and P. monophylla were photographed by Bierhorst (1971) with endodermis around the two vascular bundles of a needle; his photograph of a P. mugo fascicle without endodermis is one of the few examples of photographic evidence of endodermis presence or absence in petiolar regions within a fascicle of gymnosperm leaves. Bright-field images of Picea mariana needles by Way and Sage (2008) show a typical endodermis but without manifest Casparian bands. Hu and Yao (1981) and Hu and Wang (1984) reported an endodermis in the needle and stem of Cathaya argyrophylla; an endodermis is apparent in its leaves without visible Casparian bands and in a petiole in a ring around the vein, but endodermis or vein sheath was not mentioned in its petiole. Duarte et al. (2013) had micrographs of petioles in Ginkgo biloba without noting vein sheath or endodermis, but Dörken (2013a) clearly illustrated one vein sheath around the two vascular bundles in its petioles. Yet Klimko et al. (2015) labelled an endodermis in Ginkgo petioles. The dearth of photographic evidence and conflicting reports indicate that more investigations are merited for petioles in gymnosperms.
Given the paucity of photographs adequately demonstrating an endodermis with Casparian bands or vein sheaths in gymnosperm leaves, we examined leaves (the Dörken group; Napp-Zinn, 1966; Kaplan, 2022) and other gymnospermous leafy appendages like cladodes (Dörken and Stützel, 2011), phylloclades (Dörken et al., 2021; Kaplan, 2022), cataphylls (Kaplan, 2022) and their associated leaf petioles or petiolar regions (or leaf bases where no petiole was apparent), which have been rarely included in analyses of leaves. We demonstrate the presence or absence of endodermis with Casparian bands along radial (and sometimes transverse) cell walls, mostly by epifluorescence of magnified images after selective staining and microscopy. We concentrate on the endodermis and the xylem of the vascular tissue because xylem is much easier to demonstrate in relation to Casparian bands than is phloem tissue. For many species, we use higher-magnification close-ups of veins to show more clearly the presence or absence of endodermis Casparian bands. We will note other leaf features, like transfusion tissues, which may be appropriate to anatomical characterizations and functions in veins. It is sometimes very difficult to determine a vein sheath with lignified cell walls but no Casparian bands. These observations should help researchers studying the associations among plant structure and function in various ecological/environmental conditions. We have at least one image for all species studied so that we can demonstrate the situations in many species and genera, but we emphasize, with more detail, select species among 11 families, 30 genera and 41 species.
MATERIALS AND METHODS
For two centuries, Oswego, NY, USA, has been a small but major port on Lake Ontario, easternmost of the North American Great Lakes that flow into the St Lawrence River and then the Atlantic Ocean. During that time, ships have brought seeds or plants into the Oswego area from many places around the world; for example, on the SUNY Oswego campus there are non-native species of gymnosperms, including dawn redwood or Metasequoia glyptostroboides, which is believed to have been planted on campus in the 1950s, not many years after it was reported extant in China, and Chamaecyparis pisifera, located within 50 m of Lake Ontario shoreline. In the city of Oswego, well-established trees of Ginkgo biloba, Cryptomeria japonica, and Picea abies occur within 200 m of Lake Ontario and the Port of Oswego, and in the surrounding towns of Minetto, Oswego and Scriba many other non-native as well as native gymnosperms are found.
With that background, we were long interested in these gymnosperms. Leaves were obtained from plants growing in the city of Oswego, NY, USA, the towns of Minetto, Oswego and Scriba, NY, USA, the campus and greenhouse of SUNY at Oswego, and a few botanical gardens and growers (examples are named in the Acknowledgements section). Multiple leaves of deciduous species like Ginkgo biloba and Larix occidentalis were sampled during early to late summers, June into October. Evergreen species like Chamaecyparis pisifera and Pinus strobus were sampled during their first year from early summer into winter, and we examined leaves in summer and winter (we note, however, that summer to winter first-year leaves were similar in structure). Leaves and stems were fresh-sectioned with Wilkinson sword razor blades (split in half to make single blades; sometimes the halves were held together to section through a leaf, a technique taught to J.L.S. by Chaodong Yang). Specimens were then run through stain procedures or mounted directly in dH20 or 80 % glycerine.
Following methods in Seago et al. (1999), Soukup and Tylová (2018) and Seago (2020), staining procedures of freshly cut specimens included berberine hemisulfate for 1–2 h to reveal leaf cell walls under epifluorescence, berberine hemisulfate counterstained briefly in 1 % gentian violet or in 0.5 % toluidine blue O for 30–60 s to reveal lignins in endodermis Casparian bands and xylem under epifluorescence, fluorol yellow for 1 h to reveal suberin under epifluorescence, phloroglucinol HCl for 5–10 min to stain lignin in cell walls in bright-field, Sudan red 7B or Sudan IV for 1 h to reveal suberin in bright-field, and toluidine blue O for 30–60 s as a metachromatic stain to colour different wall chemicals of various tissues in bright-field. These procedures allowed us to visualize endodermis with Casparian bands, vein sheaths, xylem tracheids and transfusion tissues (e.g. Hu and Yao, 1981). We do not include images of all of the staining methods.
We used a Zeiss LSM700 with a camera at SUNY Oswego for the vast majority of images under epifluorescence at 509 nm, but a few at 419 nm. Many Zeiss LSM700 images were taken in bright-field and in laser confocal microscopy at 488 nm, and an Amscope bright-field scope with a Samsung Galaxy A42 camera was used for some images.
RESULTS
After first presenting illustrations of specimens with endodermis, vein sheaths or no layers conspicuously surrounding veins (Fig. 1), the remainder of the Results will begin with the Cycadophyta, Ginkgophyta and Gnetophyta – gymnosperms not usually classified with the Pinophyta; these will then be followed by the Pinophyta from Araucariaceae through Pinaceae. These descriptions will include references to specific, appropriate literature.
Fig. 1.
(A–C) Pinus strobus. (A) Transverse section of fresh, unstained needle, ring of endodermis without conspicuous Casparian bands, xylem of vein. Scale bar = 175 μm. (B) transverse section of needle stained in Bef (see abbreviations list at the end of the figure legend) and viewed in epifluorescence. Casparian bands easily visible, outer tangential cell walls of endodermis reveal some light lignin staining at arrow. Scale bar = 175 μm. (C) Close-up of BGVef-stained section with radially oriented Casparian bands of endodermis around xylem and some transfusion tissue. Scale bar = 60 μm. (D–F) Ginkgo biloba. (D) Transverse section of small section of laminar leaf with two large veins stained in Bef. Vein sheaths at s are around each vein, with xylem labelled. Scale bar = 400 μm. (E) Close-up of one such large vein with vein sheath, stained in berberine gentian violet and viewed in epifluorescence. Scale bar = 100 μm. (F) Small vein stained in BGVef viewed under epifluorescence does not reveal an endodermis with Casparian bands or vein sheath. Scale bar = 65 μm. (G, H) Sequoia sempervirens. (G) Transverse section of needle stained in Bef to show xylem of vein and transfusion tissue on either side of xylem, viewed in epifluorescence to show lack of endodermis and vein sheath. Scale bar = 350 μm. (H) Close-up of a vein stained in BGVef under epifluorescence to illustrate the lack of endodermis. Scale bar = 70 μm. Abbreviations in legends: bf, bright-field; ef, epifluorescence at 509 or 419 nm; lc, laser confocal at 488 nm; B, berberine hemisulfate; BGV, berberine/gentian violet; Pg, phloroglucinol 20 % HCl; Sr, Sudan red 7B; unst, unstained. Labels in photographs: c, concentric vascular bundles (stem-like); e, endodermis; s, vein sheath; t, transfusion tissue; v, vein; x, xylem.
Structures surrounding leaf veins
Figure 1A is a transverse section of a fresh-cut, unstained bright-field image of a white pine (Pinus strobus) needle in which there is a clear ring of cells, i.e. endodermis, surrounding the central vein but no easily visible Casparian bands. Epifluorescence images reveal endodermis around a vein in Fig. 1B, C with the radially oriented Casparian bands showing clearly around the xylem tracheids and vein transfusion tissue. As seen in Fig. 1D, E, there is a vein sheath or ring or zone of lignified sclerenchyma cells around large veins in a small section of a broad maiden hair/ginkgo (Ginkgo biloba) leaf. Staining for Casparian bands in a small vein reveals that veins and vein sheath cells do not have Casparian bands (Fig. 1F). An example of a flattened gymnospermous needle in a coastal redwood (Sequoia sempervirens) without endodermis or vein sheath is shown in epifluorescence in Fig. 1G, H, where there are xylem tracheids and transfusion tissue visible in a vein, but no endodermal Casparian bands or sclerified cells.
Cycadophyta, Cycadophyceae
Cycas revoluta
The compound leaf of Cycas revoluta has leaflets (Fig. 2A) and a petiolar region (Fig. 2B) with a singular vein of little xylem surrounded by a vein sheath (Fig. 2B) but no endodermis. This is similar to the drawings in Lamb (1923) and to the photographs in Jeffrey (1917), Gifford and Foster (1989) and Stevenson et al. (1996). Griffith et al. (2014, p. 244) demonstrated a thick, U-shaped ‘limiting layer of vein’ with phloroglucinol stain, but no endodermis. The related Zamia leaves have many veins with only vein sheaths (Beck, 2010). Vovides et al. (2020) and Glos et al. (2022) show vein sheath very clearly in standard microscopic preparations in another cycad, Ceratozamia, and in Zamia, respectively, as well as Coiro et al. (2020) in various microscopies for many Zamiaceae.
Fig. 2.
Specimens of Cycadophyta, Ginkgophyta, Gnetophyta and Pinophyta (Araucariaceae, Podocarpaceae, Sciadopityaceae). (A, B) Cycas revoluta. (A) Leaf section, vein with vein sheath, Pgbf (see abbreviations list in legend of Fig. 1). Scale bar = 300 μm. (B) Petiole, vein sheath, Bef. Scale bar = 150 μm. (C, D) Ginkgo biloba petiole (see also lamina in Fig. 1D–F). (C) Distal lamina, vein sheaths, Bef. Scale bar = 300 μm. (F) Proximal petiole near stem, one vein with two oblique halves, vein sheath, Pgbf. Scale bar = 300 μm. (E, F) Gnetum gnemon. Leaf blade, three main veins and lateral veins, no endodermis, no vein sheath, Pgbf. Scale bar = 400 μm. (F) Petiole, veins, two labelled, no endodermis, no vein sheath, Bef. Scale bar = 250 μm. (G) Ephedra. Scale leaves against stem, xylem cells are not revealed, Pgbf. Scale bar = 250 μm. (H, I) Araucaria araucana. (H) Leaf blade section with multiple veins and no endodermis or vein sheath, Bef. Scale bar = 500 μm. (I) Petiolar region at base of leaf, veins without endodermis or vein sheaths, Bef. Scale bar = 150 μm. (J–L) Phyllocladus alpinus. (J) Section of mid-phylloclade, c on concentric vascular bundles, v’s denote positions of some veins, Bef. Scale bar = 900 μm. (K) Phylloclade section of main concentric vascular bundles (two in number) and two veins labelled, no endodermis, Bef. Scale bar = 120 μm. (L) Concentric vascular bundles at phylloclade base, BGVef, no endodermis. Scale bar = 300 μm. (M) Dacrycarpus dacrydioides. Needle-like cataphyll, no endodermis, vein sheath cells, Bef. Scale bar = 100 μm. (N, O) Dacrydium elatum. (N) Cataphyll, vein sheath cells, Bef O. Scale bar = 100 μm. (O) Close-up of vein, no endodermis, vein sheath cells, BGVef. Scale bar = 60 μm. (P–R) Podocarpus macrophyllus. (P) Needle, vein with transfusion tissue, no endodermis, no vein sheath, Bef. Scale bar = 375 μm. (Q) Close-up of vein, no endodermis, BGVef. Scale bar = 90 μm. (R) Petiole, no endodermis or vein sheath, Bef. Scale bar = 270 μm. (S–U) Sciadopitys verticillata. (S) Transverse section of cladode, two obliquely oriented veins, Bef. Scale bar = 750 μm. (T) Close-up of one vein, xylem, no endodermis. Scale bar = 90 μm. (U) Scale leaf, no endodermis or vein sheath, little to no xylem. Scale bar = 150 μm.
Ginkgophyta, Ginkgophyceae
Ginkgo biloba
The deciduous broad leaf lamina (Fig. 1D, E) and the petiole (Fig. 2C, D) of G. biloba long shoots and spur shoots (Gifford and Foster, 1989) are characterized by adaxial/abaxial vein orientation in the lamina with a vein sheath (Fig. 1D, E). There is no endodermis with Casparian bands around the leaf lamina or petiole veins as shown by berberine gentian violet staining in a small peripheral lamina vein (Fig. 1F). Along most of their length, petioles also have a vein sheath, extending even between the two veins of the petiole (Fig. 2C). Nearer petiole bases, xylem may be united and surrounded by a less distinct lignified vein sheath (Fig. 2D); there is no endodermis. This corroborates images by Dörken (2013b) and by Leigh et al. (2011) but contradicts the claims of an endodermis by Cargo et al. (2000) and Klimko et al. (2015, for petiole), neither of whom used techniques to demonstrate Casparian bands. Our Ginkgo leaves contrast with roots of G. biloba, which have been well documented (Bonacorsi and Seago, 2016) to have an endodermis and exodermis with Casparian bands and suberin lamellae, and even its first stem internode has such exodermis, which was revealed using similar techniques to the present study (from the same batch of seeds used to produce one of the plants in the present study).
Gnetophyta, Gnetaceae
Gnetum gnemon
The broad leaf lamina (Fig. 2E) and narrower petiole (Fig. 2F) of G. gnemon are multi-veined with no endodermis (Fig. 2F) or vein sheath around the veins (Fig. 2E). Tomlinson and Fisher (2005) noted no endodermis but an ‘indistinct’ vein sheath in Gnetum, while Ajuru (2018) noted an endodermis, which does not appear evident.
Ephedraceae
Ephedra tweedieana
Appressed to the stem, scale leaves (Fig. 2G) of E. tweedieana have no endodermis or vein sheath; these can also lack xylem in very small-scale leaves. Dörken (2012) showed no endodermis or vein sheath.
Pinophyta, Araucariaceae
Araucaria araucana
The relatively wide leaf of A. araucana has several veins but no endodermis or vein sheath (Fig. 2H). Leaf tips reduce to one vein and leaf bases have a few veins with no endodermis (Fig. 2I). Thomson’s (1913) early photographs do not reveal a possible endodermis. Mastroberti and Mairath (2003) label endodermis in their images, but there is no clear ring of cells nor Casparian bands. The bright-field images of A. araucana by Rivera et al. (2022) or of Wollemia by Burrows and Bullock (1999) and Vasheka et al. (2019) do not mention or show endodermis or vein sheath. Multi-veined leaf and petiole images of Agathis do not reveal endodermis or vein sheath in Arbicheva and Pautov (2018).
Podocarpaceae
Members of the Podocarpaceae have not been shown by most to have an endodermis around their vasculature structures in phylloclades, cataphylls or needles (Lee, 1952; Griffith, 1957; Kausik, 1975; Tomlinson et al., 1989; Dörken and Parsons, 2016; Dörken et al., 2021; Andruchow-Colombo et al., 2024). The species that we examined have no endodermis around their leaf veins or concentric vascular bundles in phylloclades or leaves.
Phyllocladus alpinus
A phylloclade (Dörken et al., 2021) is illustrated at mid-phylloclade level in Fig. 2J. Viewed in berberine (Fig. 2J) and berberine gentian violet (Fig. 2K), some of the vascular tissues are stem-like and arranged in a concentric fashion, hence concentric vascular bundles, with very few xylem elements and no endodermis (Fig. 2K); there may be one stem-like stele (concentric vascular bundles) or a few stem-like vascular bundles. Lateral veins generally have a few xylem elements without endodermis and are oriented perpendicular to the epidermis, not adaxially/abaxially (Fig. 2K). At the phylloclade base (Fig. 2L) there is a ring of vascular bundles, i.e. a stem stele, sometimes with sclerified cells and no endodermis. All veins and concentric vascular bundles are without vein sheath or endodermis (Dörken et al., 2021).
Dacrycarpus dacrydioides
For the needle-like ends of the cataphylls and their bases, there is no endodermis, but vein sheath cells occur (Fig. 2M). Dörken and Parsons (2016) presented no evidence of endodermis or vein sheath in their standard histochemical preparations.
Dacrydium elatum
In D. elatum, needle-like ends of the cataphylls (Fig. 2N) do not have an endodermis (Fig. 2O), but there are vein sheath cells in the cataphylls and their bases (Fig. 2N). In one of the few studies to look at petioles, Lee (1952, p. 396) similarly noted that there was no ‘endodermal sheath’ in the leaf or petiole. Dörken and Parsons (2016) also showed no evidence of endodermis or a vein sheath in Dacrydium cupressinum; however, there was some evidence of a vein sheath in the bright-field micrograph of a leaf in Sahni and Mitra (1927).
Podocarpus macrophyllus
Like Griffith (1957) or Kausik (1975), under berberine (Fig. 2P) and berberine gentian violet (Fig. 2Q) staining our specimens of P. macrophyllus had no endodermis or vein sheath in the needle or petiole (Fig. 2R). We did observe transfusion tissue under berberine but not under berberine gentian violet staining (Fig. 2P; see Griffith, 1957).
Sciadopityaceae
Sciadopitys verticillata
The shoot structures on our plants were cladodes (Dörken and Stützel, 2011) in which two well-separated veins have an oblique orientation seen under berberine epifluorescence (Fig. 2S); there is no endodermis (Fig. 2T, berberine gentian violet) even at their base and no obvious vein sheath. Scale leaves have a very small vein with little to no xylem and no endodermis or vein sheath of lignified cells (Fig. 2U). The Dörken and Stützel (2011) bright-field images show that there is no evident ring of endodermis, but there is a vein sheath (or vein sheet, per Dörken and Stützel, 2011). The image of a fossil Sciadopitys (Stockey et al., 2005) shows two veins but no endodermal or sheath region.
Cupressaceae
The leaves of members of the Cupressaceae have such a seeming variety of morphologies from phylloclades, needles and flattened laminae to cataphylls (Gifford and Foster, 1989; Kaplan, 2022) that we were not surprised to find that many researchers reported a variety of situations for their vascular tissues. For example, Cross (1940) showed no evidence of endodermis but did report a vein sheath in Cryptomeria, whereas Ivănescu et al. (2007) and Yousif and Shahbaz (2023) noted but did not demonstrate an endodermis in various species not covered herein; in addition, some of the vascular tissues photographed and referred to were stem vascular bundles, not leaf or phylloclade tissues. Fossil Cupressaceae with phylloclade-like morphologies have a wider central vein or probable concentric vascular bundles much like typical, extant phylloclade-bearing plants (Stockey et al., 2005; Dörken et al., 2023a, b; the current study for Phyllocladus of the Podocarpaceae and Fokienia of the Cupressaceae).
Cunninghamia lanceolata
Neither needles (Fig. 3A) nor petioles (Fig. 3B) have endodermis or a distinctive, lignified vein sheath in C. lanceolata. Cross (1940) showed no endodermis but noted a vein sheath without good evidence (even referring to a wrong photograph).
Fig. 3.
Cupressaceae and Taxaceae. (A, B) Cunninghamia lanceolata. (A) Needle, no endodermis or vein sheath, BGVef (see abbreviations list in legend of Fig. 1). Scale bar = 500 μm. (B) Petiole, no endodermis or vein sheath, Bef. Scale bar = 200 μm. (C) Athrotaxis cupressoides. Scale leaf, no endodermis or vein sheath. Scale bar = 125 μm. (D, E) Metasequoia glyptostroboides. (D) Needle transverse section, vein sheath, Bef. Scale bar = 400 μm. (E) Petiole vein with vein sheath, xylem. Scale bar = 100 μm. (F, G) Sequoiadendron giganteum. Needle transverse section, no obvious endodermis or vein sheath, BGVef. Scale bar = 215 μm. (G) Close-up of vein, xylem and transfusion tissue, BGVef. Scale bar = 70 μm. (H, I) Cryptomeria japonica. (H) Needle section, BGVef. Scale bar = 200 μm. (I) Vein, xylem, transfusion tissue, no endodermis or vein sheath, BGVef. Scale bar = 150 μm. (J) Thuja occidentalis cataphyll, no endodermis or vein sheath, BGVef. Scale bar = 150 μm. (K) Chamaecyparis pisifera. Vein of cataphyll against stem, Bef. Scale bar = 200 μm. (L–O) Fokienia hodginsii. (L) Mid-phylloclade section, concentric vascular bundles at c, two lateral veins at v, BGVef. Scale bar = 500 μm. (M) Concentrix vascular bundles in centre of phylloclade, no endodermis. Scale bar = 100 μm. (N) One of two veins of phylloclade, no endodermis, BGVef. Scale bar = 60 μm. (O) Phylloclade base, two veins nearer reorientated concentric vascular bundles, no endodermis, BGVef. Scale bar = 300 μm. (P) Juniperus chinensis cataphyll, no endodermis or vein sheath, BGVef. Scale bar = 200 μm. (Q) Juniperus horizontalis cataphyll, no endodermis or vein sheath, xylem and transfusion tissue, Bef. Scale bar = 200 μm. (R, S) Cephalotaxus harringtonia. (R) Needle, no endodermis or vein sheath around vein, BGVef. Scale bar = 450 μm. (S) Petiole, vein without endodermis or vein sheath, BGVef. Scale bar = 150 μm. (T, U) Amentotaxus formosana. (T) Broad leaf with one vein, xylem, transfusion tissue and partial vein sheath, Bef. Scale bar = 700 μm. (U) Petiole, vein with partial vein sheath, BGVef. Scale bar = 300 μm. (V–X) Taxus cuspidata. (V) Needle, xylem, transfusion tissue, no apparent endodermis or sheath, Bef. Scale bar = 400 μm. (W) Close-up of xylem, no endodermis or vein sheath, BGVef. Scale bar = 75 μm. (X) Petiole, vein, no endodermis or vein sheath, Bef. Scale bar = 150 μm.
Athrotaxis cupressoides
The tightly positioned, scale-like cataphylls of A. cupressoides lack an endodermis and vein sheath (Fig. 3C). Only a faint vein sheath was reported by Napp-Zinn (1966, noting a 19th-entury report).
Metasequoia glyptostroboides
The needle (Fig. 3D) and petiolar end of the needle (Fig. 3E) in M. glyptostroboides have no endodermis, but they do have a vein sheath in needles and petiolar regions. Sterling (1949, p. 468) stated that ‘No distinct endodermal layer is present’. Standard preparations and micrographs in Ng and Smith (2020) reveal a possible vein sheath, but no endodermis, including in fossil Metasequoia. In Metasequoia roots, however, endodermis and exodermis have been clearly demonstrated (Yang et al., 2019).
Sequoiadendron giganteum
We found no endodermis or vein sheath in the needle-like end of the cataphylls (Fig. 3F, G, with transfusion tissue) or at the bases of cataphylls of S. giganteum. Chin and Sillett (2016) showed no endodermis, but they did demonstrate fibres around the vein of a cataphyll.
Sequoia sempervirens
There is no endodermis or vein sheath in needles (Fig. 1G) or petioles (Fig. 1H) of S. sempervirens. After Napp-Zinn (1966, noting far earlier researchers) noted only faint vein sheath in Sequoia, photographs in Oldham et al. (2010) and Chin et al. (2022) did not reveal any evidence of endodermis, although a vein sheath or fibres may have been present.
Cryptomeria japonica
Our sections of C. japonica did not reveal an endodermis or vein sheath with lignified cell walls in either the needle (Fig. 3H) or the petiolar base, both with xylem and transfusion tissues (3I). Cross (1940, p. 577) stated that an ‘incipient endodermal sheath may be present’. Azuma et al. (2015) labelled a vein sheath in needle leaves of C. japonica, but they did not reveal sclerification.
Thuja occidentalis
Where cataphylls arise from stems and partially surround the stem (Fig. 3J), they have no endodermis or vein sheath. Dörken (2013a) also did not find endodermis or vein sheath in Thuja plicata.
Chamaecyparis pisifera
Chamaecyparis has cataphylls that terminate in a needle-like tip; neither their needle-like portions (Fig. 3K) nor bases have an endodermis or vein sheath, although there is transfusion tissue in the vein (Shiraki et al., 2016). Ivănescu et al. (2007) noted, without demonstrating, that its leaves had endodermis.
Fokienia hodginsii
The photosynthetic organs of F. hodginsii are phylloclades (Fig. 3L). The central vascular tissue comprises stem-like concentric vascular bundles (Fig. 3L, M; also see Dörken et al., 2021), and the two lateral vascular strands are leaf veins (Fig. 3L, N), each with a single bundle of xylem; these are arranged with xylem lateral to phloem rather than in an adaxial/abaxial pattern without endodermis or vein sheath (Fig. 3N). Near phylloclade bases, the veins have their perpendicular-to-epidermis orientation, but the stem vasculature is opposite that found in the phylloclade (Fig. 3O); phylloclades lack endodermal Casparian bands or vein sheaths (Fig. 3L–O). The phylloclade pattern was shown long ago by Saxton (1930) for Fokienia, with a variation of the position of the two lateral veins in Libocedrus. While Fokienia has been found to be very closely related to Chamaecyparis and Thuja of the Cupressaceae (Lu et al., 2014; Zang et al., 2019), its leaves are clearly more similar structurally to Phyllocladus of the Podocarpaceae (Zang et al., 2019; Dörken et al., 2021); phylloclades were termed branchlets by Saxton (1930), and Dörken et al. (2021) clarified their nature and structures. These phylloclades in Fokienia are not like the leaves or cataphylls of Chamaecyparis or Thuja. Yousif and Shahbaz (2023) had bright-field micrographs of cupressaceous plants with stem-like vasculature in the leaves of Platycladus; they also stated that an endodermis was present in these genera, including around secretory canals, but an endodermis was not revealed in their micrographs.
Juniperus
We did not find an endodermis or vein sheath in the needle-like or scale leaves and petiolar regions of J. chinensis (Fig. 3P) or J. horizontalis (Fig. 3Q). Güvenc et al. (2011) labelled endodermis in standard preparations of Juniperus species, but there was no evidence of Casparian bands.
Cephalotaxaceae
Cephalotaxus harringtonia
Neither the needle (Fig. 3R) nor the petiole (Fig. 3S) has an endodermis or vein sheath in C. harringtonia. Similarly, Biswas and Johri (1997) stated that the lone vein does not have a vein sheath in C. fortunei, nor did Hu (1984) reveal endodermis in Cephalotaxus, although transfusion tissue was present. However, the autofluorescence photograph in Elpe et al. (2017) has an apparent vein sheath.
Taxaceae
Ghimire et al. (2014, p. 379) found an endodermis ‘scarcely evident’ or ‘parenchymatous’ in the Taxaceae, yet they labelled an endodermis in Taxus. Photographs in Elpe et al. (2017) reveal a possible vein sheath in Taxus, Pseudotaxus and Torreya.
Amentotaxus formosana
The leaf lamina has no endodermis, but there are vein sheath cells in the leaf lamina (Fig. 3T) and petiole (Fig. 3U); the extension of sheath fibres/sclerenchyma throughout the lamina from the single vein ending is striking (Fig. 3T). For A. formosana, Chin and Lin (2022, p. 4) also identified a vein sheath around the singular vein, as well as a vascular cambium in the vein and the extension of transfusion cells with ‘elongated sclerenchyma’ throughout the mesophyll (see also Ghimire et al., 2014). Elpe et al. (2017) labelled some cells as vascular sclereids in Amentotaxus argotaenia and Torreya californica, but did not demonstrate endodermis or vein sheath.
Taxus cuspidata
In Taxus, the needle (Fig. 3V, W) and petiole (Fig. 3X) have no endodermis or vein sheath (Roden et al., 2015), but transfusion tissue is readily apparent (Fig. 3V). Gostin and Predan (2022) found no endodermis in bright-field images, but Ghimire et al. (2014) labelled an endodermis in bright-field sections, but no Casparian bands were visible. Elpe et al. (2017) showed a possible vein sheath in T. baccata and Pseudotaxus chienii in fluorescence microscopy.
Pinaceae
Dörken and Stützel (2012) and Dörken et al. (2010) have demonstrated with standard preparation techniques and bright-field microscopy an endodermis in many members of the pines. Interestingly, Du et al. (2020, p. 1) categorized the differences between the ‘flattened leaves’ of Abies, Larix, Pseudolarix and Pseudotsuga and the ‘needlelike leaves’ of Cedrus, Picea and Pinus in regard to photosynthetic mesophyll capacity and adaxial/abaxial tissue polarity, but endodermis presence does not fully follow their categories; our findings below will show that only certain species of Pinus have oblique vein orientations. A bright-field image of Cathaya leaf shows an apparent endodermis (Hu and Wang, 1984); Pant and Basu (1977) show Casparian bands only in drawings. Fossilized needles of Pinus foisyi (Miller, 1992) demonstrate endodermis.
Cedrus atlantica
Cedrus atlantica has a clearly demonstrable endodermis with Casparian bands in its needles (Fig. 4A, B), but in petioles there are no Casparian bands or vein sheaths (Fig. 4C); veins are oriented adaxially/abaxially. In bright-field images of needles, Fahn (1990) clearly demonstrated endodermal Casparian bands in C. deodara, and Bercu et al. (2010) and Bakkali and Amraoul (2022) illustrated endodermis with Casparian bands in C. atlantica.
Fig. 4.
Pinaceae. (A–C) Cedrus atlantica. (A) Needle, endodermis vein, Bef (see abbreviations list in legend of Fig. 1). Scale bar = 400 μm. (B) Vein with Casparian bands in endodermis, BGVef. Scale bar = 40 μm. (C) Petiole, no endodermis, Bef. Scale bar = 125 μm. (D–F) Keteleeria davidiana. (D) Needle transverse section, vein, no endodermis, vein sheath, Bef. Scale bar = 400 μm. (E) Needle vein, xylem, no endodermis, BGVef. Scale bar = 100 μm. (F) Petiole vein with vein sheath, Bef. Scale bar = 200 μm. (G) Close-up of vein, vein sheath, Bef. Scale bar = 60 μm. (H–J) Abies balsamea. (H) Section of needle, no endodermis, vein sheath, Bef. Scale bar = 200 μm. (I) Close-up of vein, vein sheath with arrow on radial cell wall, Bef. Scale bar = 60 μm. (J) Close-up of petiole vein, vein sheath, xylem, no endodermis, Bef. Scale bar = 60 μm. (K–M) Abies concolor. (K) Two-veined needle, vein sheath, Bef. Scale bar = 225 μm. (L) Close-up of two veins, no endodermis, some vein sheath cells visible. Scale bar = 75 μm. (M) Petiole, two veins with vein sheath, Bef. Scale bar = 125 μm. (N–Q) Tsuga canadensis. (N) Needle, vein sheath around vein, unstef. Scale bar = 350 μm. (O) Close-up of vein xylem without endodermis, BGVef. Scale bar = 60 μm. (P) Petiole, two veins, no endodermis, vein sheath, unstef. Scale bar = 100 μm. (Q) Petiole, two veins, lignified cells of vein sheath, Pgbf. Scale bar = 60 μm. (R–T) Pseudotsuga menziesii. (R) Needle, vein surrounded by endodermis, Bef. Scale bar = 200 μm. (S) Close-up of vein with endodermis Casparian bands, BGVef. Scale bar = 30 μm. (T) Petiole, xylem of vein, endodermis with Casparian bands and thickened outer tangential walls, Bef. Scale bar = 100 μm.
Keteleeria davidiana
In our specimens of K. davidiana, there is a vein sheath in needles (Fig. 4D) and petioles (Fig. 4F), but no demonstrable endodermis with Casparian bands in needles (Fig. 4E) or petioles (Fig. 4F, G). Pant and Basu (1977) showed only drawings of the ring of endodermis in Keteleeria. Pant and Basu (1977) noted that there was an endodermis with Casparian bands, suberin and lignified secondary walls, but these were not apparent in their photographs of the closely related Cathaya, only in drawings with a ring of endodermis. Hu and Wang (1984) noted Casparian bands without demonstrating them in Cathaya needles, but not in petioles; a ring of cells in the position of endodermis was visible in both leaf parts. Han (1984) showed only an incomplete ring of endodermis without identifiable Casparian bands in Keteleeria, but no such ring of endodermis was in Cathaya.
Abies
Around the two veins of needles, A. balsamea has a vein sheath with some lignified cell walls (Fig. 4H) even on radial walls in the position of Casparian bands (Fig. 4I, arrow), but no distinct endodermis. In petioles, there is a vein sheath without endodermis (Fig. 4J). Abies concolor has a vein sheath around the two veined needles without endodermis (Fig. 4K) and in petioles (Fig. 4M); Casparian bands via berberine gentian violet epifluorescence are not revealed in either needles (Fig. 4L) or petioles. Dörken (2014) and Dörken and Lepetit (2018) illustrated a vein sheath in the related A. alba, although others (e.g. Bercu et al., 2010; Ghimire et al., 2015a) illustrated endodermis, but Casparian bands were not evident. Endoh et al. (2012) had a ring of endodermis under bright-field, but Casparian bands were not visible in cryo-SEM images.
Tsuga canadensis needles (Fig. 4N) and petioles (Fig. 4P, Q) have a vein sheath without endodermis (Fig. 4O). There is usually just one vein in needles, although it may appear somewhat divided, whereas petioles have greater division between two units of the xylem. Soar (1922) showed lignified and suberized endodermal cells in a drawing, and Gambles and Dengler (1974) clearly demonstrated a partial vein sheath in T. canadensis.
Pseudotsuga menziesii
Needles (Fig. 4R, S) and petioles (Fig. 4T) have distinct endodermis Casparian bands (Fig. 4S); they also have lignified outer tangential secondary walls without suberin lamellae (Fig. 4T). Soar (1922), Pădure et al. (2008), Chin and Sillett (2019) and Apple et al. (2002) also demonstrated the Casparian bands in this species.
Larix
In L. laricina, endodermis with Casparian bands is present in needles (Fig. 5A, B) and leaf bases or petioles (Fig. 5C) with some evidence of lignified tangential walls in needles in berberine staining (Fig. 5A, low magnification). In L. kaempferi (Fig. 5D, E) and L. occidentalis (Fig. 5F, G) endodermis is present in needles and leaf bases (petioles); in L. kaempferi outer tangential endodermal cell walls were thickened under low magnification in berberine staining (Fig. 5D) and in L. occidentalis at higher magnification under phloroglucinol staining (Fig. 5G). Endodermis in Larix has been shown by Eguchi et al. (2004) and Bercu and Popoviciu (2013) without demonstrable Casparian bands.
Fig. 5.
Pinaceae. (A–C) Larix laricina. (A) Needle, endodermis around vein, Bef (see abbreviations list in legend of Fig. 1). Scale bar = 100 μm. (B) Close-up of vein with endodermis Casparian bands, BGVef. Scale bar = 60 μm. (C) Close-up of needle base, vein with Casparian bands of endodermis, BGVef. Scale bar = 50 μm. (D, E) Larix kaempferi. (D) Needle, endodermis around vein, Bef. Scale bar = 150 μm. (E) Base of needle with endodermis Casparian bands, BGVef. Scale bar = 175 μm. (F, G) Larix occidentalis. (F) Needle, endodermis with Casparian bands around vein, BGVef. Scale bar = 130 μm. (G) Needle base, endodermis with Casparian bands and outer tangential walls of endodermal cells stained for lignin, Pgbf. Scale bar = 50 μm. (H–K) Picea abies. (H) Needle, one vein, endodermis, Bef. Scale bar = 150 μm. (I) Needle vein, endodermis Casparian bands. Scale bar = 75 μm. (J) Distal petiole with endodermis, Bef. Scale bar = 120 μm. (K) Distal petiole to show Casparian bands of endodermis, Bef. Scale bar = 70 μm. (L–N) Picea glauca. (L) Needle, endodermis around vein, BGVlc. Scale bar = 150 μm. (M) Needle vein, endodermis Casparian bands, BGVef. Scale bar = 35 μm. (N) Petiole, no endodermis, BGVef. Scale bar = 1000 μm. (O–Q) Picea pungens. (O) Needle, endodermis, BGVef. Scale bar = 220 μm. (P) Needle vein, endodermis Casparian bands, BGVef. Scale bar = 45 μm. (Q) Petiole, no endodermis, Bef. Scale bar = 100 μm.
Picea abies, P. glauca and P. pungens In the three species of Picea studied here, P. abies has endodermis in the needles (Fig. 5H, I) and distal petioles (Fig. 5J, K), but not in proximal petioles; the outer tangential walls fluoresce lightly in berberine (Fig. 5H). Picea glauca has distinctive endodermis in needles (Fig. 5L, M), but no endodermis or vein sheath in petioles (Fig. 5N), and P. pungens also has endodermis with Casparian bands in needles (Fig. 5O, P), but no endodermis or vein sheath in petioles (Fig. 5Q). Endodermis was demonstrated in P. abies by Soukupová et al. (2000), in P. glauca by Hacke et al. (2015) and in P. pungens by Soukupová et al. (2001) and Qin et al. (2014). Endodermis was also shown in needles of Picea species by Ghimire et al. (2015a) without evident Casparian bands.
Pinus
Pinus strobus (Fig. 1A–C) has single vein needles with a distinctive endodermis with Casparian bands (Figs 1B, C and 6A) and some lignification on tangential cell walls (Fig. 6A) but no suberin lamellae (Fig. 6B); the xylem of the vein is often in two slightly separated units (Fig. 1C). In the needle bases within the middle-lower fascicle or petiolar region, there is no endodermis or vein sheath (Fig. 6C); there is endodermis in the most distal petiolar regions (not shown). Pinus mugo needles have an endodermis with Casparian bands around the two veins (Fig. 6D); the fascicle base has two petiolar needles without endodermis or vein sheath around the veins (not shown). In P. sylvestris, the two-veined needles have a distinctive endodermis with lignified Casparian bands (Fig. 6E, F) but no suberin lamellae (Fig. 6G); the outer tangent wall is thickened but stained unevenly for lignin (Fig. 6F). Casparian bands are found in distal fascicled needles (Fig. 6H, I) but not in proximal fascicled needles (Fig. 6J), indicating that endodermis has developed before needles become separate in fascicles. Scale leaves have no endodermis and often show no xylem (Fig. 6K).
Fig. 6.
(A–C) Pinus. strobus (see also Fig. 1A–C). (A) Needle, part of vein, endodermis with slightly lignified outer walls (arrow), Pgbf (see abbreviations list in legend of Fig. 1). Scale bar = 30 μm. (B) Part of vein, endodermis with no evidence of suberin lamellae, SR7Bbf. Scale bar = 35 μm. (C) Fascicle five-needle bases, no endodermis Casparian bands, Bef. Scale bar = 300 μm. (D) Pinus mugo. Needle, endodermis Casparian bands, Bef. Scale bar = 150 μm. (E–K) Pinus sylvestris. (E) Needle, two distinct veins, endodermis ring Casparian bands, Bef, Scale bar = 350 μm. (F) Needle, part of vein, endodermis with slightly lignified outer tangential walls (arrow), Pgbf. Scale bar = 35 μm. (G) Part of vein, endodermis with no evidence of suberin lamellae, SR7Bbf. Scale bar = 30 μm. (H) Distal fascicle of two needles, endodermis, Bef. Scale bar = 150 μm. (I) Close-up of part of endodermis with Casparian bands from Fig. 5V, BGVef. Scale bar = 50 μm. (J) Proximal fascicle needles with two veins each, no endodermis. Scale bar = 150 μm. (K) Scale leaf, vein with no endodermis and no visible xylem. Scale bar = 70 μm. (L–Q) Pinus nigra. (L) Transverse section of needle with two veins, endodermis with distinct Casparian bands, Bef. Scale bar = 150 μm. (M) Needle, part of vein, endodermis with Casparian bands with slightly thickened outer walls, Pgbf. Scale bar = 35 μm. (N) Part of vein, endodermis with no evidence of suberin lamellae, SR7Bbf. Scale bar = 30 μm. (O) Distal fascicle of two needles, endodermis, Bef. Scale bar = 150 μm. (P) Close-up from Fig. 5O to show Casparian bands in distal fascicle needle, BGVef. Scale bar = 50 μm. (Q) Proximal fascicle with no endodermis, BGVef. Scale bar = 150 μm. (R–Y) Pinus resinosa. (R) Needle transverse section, endodermis with thickened outer tangential cell walls, Bef. Scale bar = 150 μm. (S) Needle, part of vein, endodermis with slightly lignified outer walls, Pgbf. Scale bar = 35 μm. (T) Part of vein, endodermis with no evidence of suberin lamellae, SR7Bbf. Scale bar = 30 μm. (U) Distal fascicle of two needles, endodermis, Bef. Scale bar = 150 μm. (V) Close-up of part of endodermis from Fig. 5V, BGVef. Scale bar = 35 μm. (W) Proximal fascicle needles with two veins each, no endodermis. Scale bar = 150 μm. (X) Scale leaf, no endodermis, vein with no visible xylem. Scale bar = 70 μm.
In P. nigra leaves, needles have endodermis with Casparian bands around the two veins (Fig. 6L); Casparian bands stain for lignin (Fig. 6M, on either side of arrow or e), but outer tangential cell walls do not stain for lignin via phloroglucinol HCl in bright-field or berberine gentian violet in epifluorescence, although they appear thicker (Fig. 6M at arrow). There is no evidence of suberin lamellae (Fig. 6N). In unstained specimens viewed in epifluorescence (Fig. 6O, difficult to see) distal and partially separated needles within a fascicle develop an endodermis with Casparian bands (Fig. 6P). In proximal, unseparated needles within the fascicle there is no endodermis in the two needles (Fig. 6Q). Pinus resinosa has mature needles with two veins surrounded by an endodermis (Fig. 6R) with Casparian bands (Fig. 6R, S) and distinctly secondarily thickened outer tangential cell walls of lignin (Fig. 6R, S); they do not have suberin lamellae (Fig. 6T). Distal, slightly separated needles within a fascicle have Casparian bands in endodermis around each two-veined needle (Fig. 6U, V), whereas proximal or petiolar needles in fascicles do not have endodermis (Fig. 6W). Scale leaves have little to no xylem and no endodermis (Fig. 6X). In three species of Pinus with two veins within an endodermis we find varying obliquely oriented veins (Fig. 6E, L, R; cf. Du et al., 2020).
Chamberlain (1935), Keng and Little (1961), the Dörken group (e.g. Dörken and Stützel, 2012) and Kaplan (2022), have had illustrations of pine needles with an endodermis. Kuusk (2024) has recently shown bright-field images of needles of P. nigra that seem to have more lignin in their endodermal cell walls after 2 years than in their first-year needles after phloroglucinol HCl treatment. Endodermis in P. sylvestris showed clearly in images via laser confocal images in Lin et al. (2001) and in bright-field by Dörken and Stützel (2012). No cataphyll endodermis was revealed in bright-field images in P. lambertiana by Sacher (1955). Carde (1978) questioned the occurrence of Casparian bands in P. pinaster needles, but Gao et al. (2023) clearly demonstrated Casparian bands. Like Meicenheimer et al. (2008) for P. resinosa, Donaldson and Williams (2018) demonstrated unequivocally that there were secondarily lignified outer tangential cell walls in P. radiata, but their fluorescence photographs also showed clearly Casparian bands without suberin lamellae in the endodermis using techniques like ours. The bright-field images of various Abies, Picea (Ghimire et al., 2015a) and Pinus species (e.g. P. sylvestris, P. parviflora, P. densiflora, P. taeda; Ghimire et al., 2015b) also showed thickened outer tangential endodermal cell walls. Miller (1992) also demonstrated clearly thickened outer tangential cell walls in the fossilized endodermis of P. foisyi. In P. nigra, Meicenheimer et al. (2008) had bright-field and fluorescent sections of P. nigra with no secondary cell walls, but Lux et al. (2017) and many others (e.g. Ghimire et al., 2015a, 2015b; Donaldson and Williams, 2018 (P. radiata)) had bright-field photographs with distinct outer tangential cell walls that were thickened and lignified. Soar (1922) illustrated (but did not present photographs of) the lack of endodermis in the proximal regions of the fascicled base of pines and its presence developing prior to emergence of the needles between the scale leaves of fascicles. Few other researchers have determined the presence or absence of endodermis or vein sheaths in petioles or scale leaves in pines.
DISCUSSION
As shown in Table 1, an endodermis with Casparian bands is found around leaf veins only in the Pinophyta family Pinaceae, where it is found in most members (Cedrus atlantica; Pseudotsuga menziesii; Larix laricina, L. kaempferi and L. occidentalis; Picea abies, P. glauca and P. pungens; Pinus mugo, P. nigra and P. resinosa with lignified secondary outer tangential walls, P. strobus and P. sylvestris). Petioles, in contrast, almost always lack endodermis; the notable exceptions with Casparian bands in petiolar regions are Pseudotsuga menziesii, Larix laricina, Picea abies, and distal petiolar regions within fascicles of Pinus nigra, P. resinosa, P. strobus and P. sylvestris. Within the 11 genera of Pinaceae and the 8 we studied, only needles of Keteleeria, Abies and Tsuga lack distinctive Casparian bands in an endodermis-like ring around veins, and only these same three Pinaceae genera have vein sheaths. Both Keteleeria and Abies are interesting because their needles and petioles showed indications of cell wall modifications that might represent incipient endodermis. Thus, in detail, with our study and with other examples in the literature, we have resolved the issues of Napp-Zinn (1966) and Lersten (1997) that endodermis with Casparian bands around veins had not been positively identified in needles of many members of the Pinaceae.
Table 1.
The presence of endodermis and vein sheath in representative gymnosperms.
| Phylum and family | Species | Endodermis with Casparian bands | Vein sheath | ||
|---|---|---|---|---|---|
| Leaf | Petiole | Leaf | Petiole | ||
| Cycadophyta, Cycadaceae | Cycas revoluta | ✓ | ✓ | ||
| Ginkgophyta, Ginkgoaceae | Ginkgo biloba | ✓ | ✓ | ||
| Pinophyta, Araucariaceae | Araucaria araucana | ||||
| Pinophyta, Podocarpaceae | Phyllocladus alpinus | ||||
| Dacrycarpus dacrydioides | * | * | |||
| Dacrydium elatum | * | * | |||
| Podocarpus macrophyllus | |||||
| Pinophyta, Sciadopityaceae | Sciadopitys verticillata | ||||
| Pinophyta, Cupressaceae | Cunninghamia lanceolata | ||||
| Athrotaxis cupressoides | |||||
| Metasequoia glyptostroboides | ✓ | ✓ | |||
| Sequoia sempervirens | |||||
| Sequoiadendron giganteum | |||||
| Cryptomeria japonica | |||||
| Thuja occidentalis | |||||
| Chamaecyparis pisifera | |||||
| Fokienia hodginsii | |||||
| Juniperus horizontalis | |||||
| Juniperus chinensis | |||||
| Pinophyta, Cephalotaxaceae | Cephalotaxus harringtonia | ||||
| Pinophyta, Taxaceae | Amentotaxus formosana | * | * | ||
| Taxus cuspidata | |||||
| Pinophyta, Pinaceae | Cedrus atlantica | ✓ | |||
| Keteleeria davidiana | ✓ | ✓ | |||
| Abies balsamea | ✓ | ✓ | |||
| Abies concolor | ✓ | ✓ | |||
| Tsuga canadensis | ✓ | ✓ | |||
| Pseudotsuga menziesii | ✓ | ✓ | |||
| Larix occidentalis | ✓ | ✓ | |||
| Larix kaempferi | ✓ | ✓ | |||
| Larix laricina | ✓ | ✓ | |||
| Picea abies | ✓ | ✓ | |||
| Picea pungens | ✓ | ||||
| Picea glauca | ✓ | ||||
| Pinus nigra | ✓ | ✓ | |||
| Pinus resinosa | ✓ | ✓ | |||
| Pinus strobus | ✓ | ✓ | |||
| Pinus sylvestris | ✓ | ✓ | |||
| Pinus mugo | ✓ | ||||
| Gnetophyta, Gnetaceae | Gnetum gnemon | ||||
| Gnetophyta, Ephedraceae | Ephedra tweedieana | ||||
Taxa are listed according to Yang et al. (2022).
*Partial vein sheath.
In contrast to endodermis, vein sheaths are prominent in the most early evolved gymnosperms, namely Cycadophyta and Ginkgophyta, but they are present in relatively few members of the Pinophyta and are absent in the Gnetophyta. Viewed in current phylogenies (Gernandt et al., 2016; Yang et al., 2022), the distribution of this trait (Table 1) suggests a loss in several lineages or several independent origins. A few gymnosperms in the Pinophyta families of Podocarpaceae (Dacrycarpus dacrydioides and Dacrydium elatum), Cupressaceae (Metasequoia glyptostroboides) and Taxaceae (Amentotaxus formosana) have vein sheaths or sclerified cells in positions of possible vein sheaths. The derived lineage Gnetophyta (Gnetum gnemon and Ephedra tweedieana) has no endodermis or vein sheath in leaves.
If we consider the fossil record (Taylor et al., 2009), for example, Robison (1977) had micrographs of fossil Pinus fascicles to show that an endodermis, without visible Casparian bands, was present in distal needles bases, but not proximal bases. He also claimed that there was only a thick-walled sheath in an earlier fossil, Prepinus (Robison, 1974). Then, Stockey and Ueda (1986) produced light micrographs of five-needled fossil Pinus hokkaidoensis that clearly reveal a ring of endodermal cells without visible Casparian bands around the vein. Taylor et al. (2009, in a photograph by R. A. Stockey) had a fossil five-needle pine, P. hokkaidoensis, like P. strobus, with a quite distinct endodermis ring around the vein in each needle with distinctly thickened radial and outer tangential walls. The absence of an endodermis with Casparian bands in the Abietoideae of the Pinaceae fossil records supports the idea that it is a more recently evolved structure and that it may have evolved multiple times. It would help to have more anatomical images of fossil members of the Abietoideae and Pinoideae of the Pinaceae to track the evolution of endodermis with Casparian bands. The fossil record of other families like the Cupressaceae seems particularly difficult for obtaining a clear picture of internal structure with vein sheaths or no vein-encircling structures (e.g. Taylor et al., 2009; Herrera et al., 2017), complicating the task of tracking the evolution of vein sheath across gymnosperms.
Soar (1922) was among the first to discuss any physiological and ecological roles of Pinaceae leaves in enabling them to withstand temperatures in alpine and subalpine regions. She described the leaves as xerophytic in nature, with defined endodermis, placing them in contrast with angiosperm leaves where there is no endodermis. The development and structure of endodermis with Casparian bands in needles/leaves of the members of Pinaceae could have contributed to their ability to evolve in, colonize and radiate into extreme environments, especially in cold, high-elevation environments where other gymnosperms are not found. Indeed, members of Pinaceae such as Picea and Pinus are the typical treeline species in many mountain ranges around the world (e.g. Kaku, 1971; Skärby et al., 1998; Miller, 1999; Wu et al., 2005; Taylor et al., 2009; Roden et al., 2009, 2015; Leslie et al., 2018; Stegner et al., 2023) where blowing ice and rime ice, consequent desiccation and extreme low temperatures limit other species (e.g. Kaku, 1971; Skärby et al., 1998; Wu et al., 2005; Roden et al., 2009, 2015; Taylor et al., 2009; Stegner et al., 2023). As Wu et al. (2005) have importantly shown, endodermis with Casparian bands in needles of Pinus bungeana is more permeable than endodermis with Casparian bands in roots; a salient reason could be the lack of suberin lamellae as we have shown for other species of Pinus. In Picea mariana, Way and Sage (2008) showed temperatures had a pronounced effect upon vein and leaf size, but it was not apparent in their photographs if there were any effects upon the obvious endodermis which we have shown in other Picea species. Endoh et al. (2012) showed in a species of Abies that changes in needles were structurally visible during freezing events. Freeze tolerance gained through an endodermis with Casparian bands is a logical driver of its evolution and persistence in Pinaceae.
The lack of a clear, distinct endodermis in Keteleeria and Abies of Abietoideae could relate to their subtropical habitat and origin (Zhang et al., 2023) – and a possible lack of selection pressure for endodermis, potentially explaining why this anatomical feature is absent in these lineages but present in Cedrus of Abietoideae and members of Pinoideae. The possible incipient endodermis seen in these genera could suggest an intermediate evolutionary step between vein sheaths and endodermis. Selection pressure for the development of endodermis is understandable for species like Abies balsamea, which is a treeline species subject to extreme cold temperatures.
This relationship between leaf structures and the plant environments of the gymnosperm has also been studied with regard to levels of CO2 (Sage and Khoshravesh, 2016), as well as nutrients and other compounds. In terms of the control and facilitation of movements of nutrients, phosphates and gases through needle veins, endodermis and mesophyll, Gao et al. (2023) have demonstrated the importance of endodermis with Casparian bands (see also Uchida et al., 2012). Other researchers have shown how photosynthetic and respiratory processes with CO2 and O2 movements could be salient in considering how the endodermis functions in conifer leaves (e.g. Sutinen et al., 2000; Eguchi et al., 2004; Roden et al., 2015; Sage and Khoshravesh, 2016). These interactions and relationships between leaf structural and environmental effects need to be examined far more thoroughly. Indeed, this goes beyond normal leaf functions to the effects of air pollution on these gymnosperm leaves in which there is endodermis – very evident in bright-field images of Pinus nigra by Delian and Săvulescu (2022).
As noted in the Results, members of the Pinaceae generally have a State I endodermis with Casparian bands only, i.e. there are no suberin lamellae in the endodermis cells, as per Seago et al. (1999), Enstone et al. (2003), Soukup and Tylová (2018) and Yang et al. (2019). The occurrence of a secondarily lignified outer tangential wall might seem to make these cells more like an exodermis, except that they are lacking suberin lamellae. However, a typical exodermis has Casparian bands and suberin produced concomitantly in plant roots (Pazourek and Votrubová, 1997; Seago et al., 1999; Seago, 2002; Enstone et al., 2003; Schreiber and Franke, 2011; Bonacorsi and Seago, 2016; Soukup and Tylová, 2018) or in plant stems (Seago and Fernando, 2013; Bonacorsi and Seago, 2016). Where produced, secondary cell wall lignification between suberin lamellae and protoplasm in these cells is normally produced in the innermost tangential cell wall of endodermis and the outermost cell wall in exodermis (e.g. Meyer et al., 2009), as found in gymnosperm roots (e.g. Bonacorsi and Seago, 2016; Brundrett and Tedersoo, 2020).
Therefore, the development and structure of endodermis in conifer leaves are both similar to and different from conifer roots. Conifer leaf endodermis is typically in State I, Casparian bands only. Where lignified, secondary cell wall thickening occurs, as in Pinus resinosa needles and Pseudotsuga petioles, it occurs only in the outer tangential cell wall and without suberin lamella formation. However, our findings also show that some outer tangential cell wall thickening in needles seen under epifluorescence occurs in these outer tangential walls in Cedrus, Metasequoia, Larix (deciduous species all), Picea, Pseudotsuga and most Pinus species, but these do not stain for lignin by the phloroglucinol test as they do in Pinus resinosa (Meicenheimer et al., 2008). However, other Pinus species, e.g. P. radiata (Donaldson and Williams, 2018) and P. laricio (Coulter and Chamberlain, 1901; Popham, 1966; Clegg and Cox, 1978), have what is like an innermost State III exodermis, except that there are no suberin lamellae and no U-shaped lignified walls. Interestingly, Griffith et al. (2014) showed a U-shaped wall in Cycas hainanensis leaf via phloroglucinol staining, but no suberin lamellae were demonstrated.
Age of needles must be very important to these differences (Soar, 1922; Cross, 1940; Ewers and Schmid, 1981; Sutinen et al., 2000; Wu et al., 2005; Sack and Scoffoni, 2013; Vasheka et al., 2019; Bozkurt et al., 2023; Huusk, 2024). In her recent thesis Huusk (2024) had bright-field images of phloroglucinol-stained Pinus nigra, P. halapensis and P. pinea in which needles older than first year in P. nigra appeared to reveal more heavily lignified cell walls under low magnification; yet, needles may remain on some pines like P. sylvestris for 3–4 years (Pensa and Jalkanen, 1999) and many Picea species retain needles for years. There would seem to be major physiological and ecological ramifications for leaves and the plants if leaf endodermis develops more lignified cell walls, even if only in outer tangential walls, over time as leaves survive and remain active, especially through seasons of warmth and then freezing cold, etc. Lastly, larger questions are: do Pinaceae represent earlier or later evolved gymnosperms (cf. Taylor et al., 2009; Lu et al., 2014; Gernandt et al., 2016; Yang et al., 2022) and thus when did endodermis evolve in gymnosperms and Pinaceae?
ACKNOWLEDGEMENTS
We are extremely grateful to the following persons for aiding this project by encouragement, providing access to plants, providing background information or helping to obtain some supplies: Marilyn A. Seago, Melodie Seago, Zenab Mohamed, Samouel Beguin, Yulia Artemenko, Ashley Mears, Beverly Jones, Cathy Van Patten, Chaodong Yang, Aleš Soukup, Edita Tylová, Olga Votrubová, Stan Kollar at Kollar Nursery, Silas Zoeller at Far Reaches Botanical Conservancy, Ontario Orchards, Wilson Bros Gardens, Juniper Level Botanic Garden at Plant Delights Nursery, Conifer Kingdom, Woodies Garden Goods and the AoB reviewers and editors.
Contributor Information
James L Seago, Department of Biological Sciences, SUNY Oswego, Oswego, NY 13126, USA.
Kamal I Mohamed, Department of Biological Sciences, SUNY Oswego, Oswego, NY 13126, USA; Rice Creek Field Station, SUNY Oswego, Oswego, NY 13126, USA.
Kristen R Haynes, Rice Creek Field Station, SUNY Oswego, Oswego, NY 13126, USA.
FUNDING
We received no funding except that we had the use of the facilities and Zeiss LSM700 microscope at SUNY Oswego.
AUTHOR CONTRIBUTIONS
The study was conceived and begun by the senior author, but all three authors participated in the preparation and examination of specimens, as well as the microscopy and photography. The senior author began and wrote the early version of the manuscript, but all three authors contributed to the writing.
REFERENCES
- Ajuru MG. 2018. Morpho-anatomical studies of Gnetum africanum Welw and Gentum buchholzianum Engl. (Gnetaceae Lindley) in Nigeria. Journal of Applied Life Sciences International 18: 1–8. [Google Scholar]
- Andruchow-Colombo A, Escapa IH, Aagesen L, Matsunaga KKS. 2023. In search of lost time: tracing the fossil diversity of Podocarpaceae through the ages. Botanical Journal of the Linnean Society 203: 315–336. [Google Scholar]
- Andruchow-Colombo A, Pozner R, Abarzúa AM, Escapa IH. 2024. Leaf and megasporangiate cone anatomy of two Podocarpaceae endemic to western Patagonia: Saxegothaea conspicua and Lepidothamnus fonkii. International Journal of Plant Sciences 185: 159–176. [Google Scholar]
- Apple M, Tiekotter K, Snow M, et al. 2002. Needle anatomy changes with increasing tree age in Douglas-fir. Tree Physiology 22: 129–136. [DOI] [PubMed] [Google Scholar]
- Arbicheva AL, Pautov AA. 2018. Leaf periderm supports longevity and functionality of crown leaves in Agathis species (Araucariaceae). Brazilian Journal of Botany 41: 155–165. [Google Scholar]
- Azuma W, Ishii HR, Kuroda K, Kuroda K. 2015. Function and structure of leaves contributing to increasing water storage with height in the tallest Cryptomeria japonica trees of Japan. Trees 30: 141–152. [Google Scholar]
- Bakkali NE, Amraoul MB. 2022. Morphological and anatomical characterization of ecotype needles of Cedrus atlantica in Morocco. International Journal of Forestry Research 2022: 1–11. [Google Scholar]
- Beck CB. 2010. An introduction to plant structure and development. Cambridge: Cambridge University Press. [Google Scholar]
- Bercu R, Broasca L, Popoviciu R. 2010. Comparative anatomical study of some Gymnospermae species leaves. Botanica Serbia 34: 21–28. [Google Scholar]
- Bercu R, Popoviciu DR. 2013. Anatomical comparative study of Larix decidua Mill. and Picea abies (L.) Karsten (Pinaceae) leaf. Annals of the Romanian Society for Cell Biology 18: 172–177. [Google Scholar]
- Bierhorst DW. 1971. Morphology of vascular plants. New York: Macmillan. [Google Scholar]
- Biswas C, Johri BM. 1997. The gymnosperms. Heidelberg: Springer. [Google Scholar]
- Bonacorsi NK, Seago JL Jr. 2016. Root development and structure in seedlings of Ginkgo biloba. American Journal of Botany 103: 355–363. [DOI] [PubMed] [Google Scholar]
- Bozkurt AE, Coşkunçelebri K, Terzioğlu S. 2023. Variation in needle anatomy of Scots pine (Pinus sylvestris L.) populations according to habitat and altitudinal zones in Türkiye. Šumarski List 147: 215–225. [Google Scholar]
- Brundrett MC, Tedersoo L. 2020. Resolving the mycorrhizal status of important northern hemisphere trees. Plant and Soil 454: 3–34. [Google Scholar]
- Burrows GE, Bullock S. 1999. Leaf anatomy of Wollemi pine (Wollemia nobilis, Araucariaceae). Australian Journal of Botany 47: 795–806. [Google Scholar]
- Campbell R. 1972. Electron microscopy of the development of needles of Pinus nigra var. maritima. Annals of Botany 36: 711–720. [Google Scholar]
- Carde J-P. 1978. Ultrastructural studies of Pinus pinaster needles: the endodermis. American Journal of Botany 65: 1041–1054. [Google Scholar]
- Cargo JMC, Gattuso S, Gattuso M. 2000. Morphoanatomical and micrographic characters of Ginkgo biloba L. (Ginkgoaceae) leaf. Acta Farmacéutica Bonaerense 19: 35–39. [Google Scholar]
- Chamberlain CJ. 1935. Gymnosperms: structure and evolution. Chicago: University of Chicago Press. [Google Scholar]
- Chin ARO, Guzmán-Delgado P, Sillett SC, et al. 2022. Shoot dimorphism enables Sequoia sempervirens to separate requirements for foliar water uptake and photosynthesis. American Journal of Botany 109: 564–579. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Chin S-T, Lin H-Y. 2022. A note on the structural peculiarities of the leaves of Amentotaxus formosana Li in southern Taiwan. Collection Research 34: 1–9. [Google Scholar]
- Chin ARO, Sillett SC. 2016. Phenotypic plasticity of leaves enhances water-stress tolerance and promotes hydraulic conductivity in a tall conifer. American Journal of Botany 103: 796–807. [DOI] [PubMed] [Google Scholar]
- Chin ARO, Sillett SC. 2019. Within-crown plasticity in leaf traits among the tall conifers. American Journal of Botany 106: 174–186. [DOI] [PubMed] [Google Scholar]
- Clegg CJ, Cox G. 1978. Anatomy and activities of plants. London: John Murray. [Google Scholar]
- Coiro M, Jelmini N, Neuenschwander H, et al. 2020. Evolutionary signal of leaflet anatomy in the Zamiaceae. International Journal of Plant Sciences 181: 697–715. [Google Scholar]
- Coulter JM, Chamberlain CJ. 1901. Morphology of spermatophytes, Vol. 1. New York: D. Appleton. [Google Scholar]
- Crisp MD, Cook LG. 2011. Cenozoic extinctions account for the low diversity of extant gymnosperms compared with angiosperms. New Phytologist 192: 997–1009. [DOI] [PubMed] [Google Scholar]
- Cross GL. 1940. Some histogenetic features of the shoot of Cryptomeria japonica. American Journal of Botany 28: 573–582. [Google Scholar]
- Delian E, Săvulescu E. 2022. Anatomical and physiological changes in needles of Pinus nigra J.F. Arnold reveal urban traffic air pollution driven effects. Science Papers, Series B, Horticulture 46: 674–684. [Google Scholar]
- Doak CC. 1935. Evolution of foliar types, dwarf shoots, and cone scales of Pinus: with remarks concerning similar structures in related forms. Illinois Biological Monographs 13: 7–106. [Google Scholar]
- Donaldson L, Williams N. 2018. Imaging and spectroscopy of natural fluorophores in pine needles. Plants 7: 10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Dörken VM. 2012. Leaf-morphology and leaf-anatomy in Ephedra altissima Desf. (Ephedraceae, Gnetales) and their evolutionary relevance. Feddes Repertorium 123: 243–255. [Google Scholar]
- Dörken VM. 2013a. Leaf dimorphism in Thuja plicata and Platycladus orientalis (thujoid Cupressaceae s. str., Coniferales): the changes in morphology and anatomy from juvenile needle leaves to mature scale leaves. Plant Systematics and Evolution 299: 1991–2001. [Google Scholar]
- Dörken VM. 2013b. Morphology, anatomy and vasculature in leaves of Ginkgo biloba L. (Ginkgoaceae, Ginkgoales) under functional and evolutionary aspects. Feddes Repertorium 124: 80–97. [Google Scholar]
- Dörken VM. 2014. Morphology and anatomy of needle-leaves and cladode-like structures in Abies firma Siebold & Zucc. (Pinaceae, Coniferales) and their evolutionary relevance. Feddes Repertorium 125: 61–71. [Google Scholar]
- Dörken VM, Hill RS, Jordan GJ, Parsons RF. 2021. Evolutionary and ecological significance of photosynthetic organs in Phyllocladus (Podocarpaceae). Botanical Journal of the Linnean Society 196: 343–363. [Google Scholar]
- Dörken VM, Ladd PG, Parsons RF. 2019. The foliar characters in Callitris (Callitroideae, Cupressaceae s. str.) and their evolutionary and ecological significance. Feddes Repertorium 130: 247–271. [Google Scholar]
- Dörken VM, Ladd PG, Parsons RF. 2023a. Leaf dimorphism in Cupressus nootkatensis D. Don and Cupressus vietnamensis (Farjon & T.H. Nguyên) Silba (Cupressaceae) and its ecological and evolutionary significance. Trees 37: 1267–1279. [Google Scholar]
- Dörken VM, Lepetit B. 2018. Morpho-anatomical and physiological differences between sun and shade leaves in Abies alba Mill. (Pinaceae, Coniferales): a combined approach. Plant, Cell & Environment 41: 1683–1697. [DOI] [PubMed] [Google Scholar]
- Dörken VM, Nimsch H. 2018. A monograph of leaf characters in the genus Abies (Abietoideae, Pinaceae). Oberwinter: Kessel. [Google Scholar]
- Dörken VM, Nimsch H. 2021. Spiessstannen (Cunninghamia, Cupressaceae s.l.): eine Gattungsübersicht. Schweizerische Beiträge zur Dendrologie 71–72: 55–66. [Google Scholar]
- Dörken VM, Parsons RF. 2016. Morpho-anatomical studies on the change in the foliage of two imbricate-leaved New Zealand podocarps: Dacrycarpus dacrydioides and Dacrydium cupressinum. Plant Systematics and Evolution 302: 41–54. [Google Scholar]
- Dörken VM, Rushforth K, Hudson T, Nimsch H. 2023b. Morphology and anatomy of leaves and vegetative shots in Calocedrus Kurz (Cupressoideae, Cupressaceae s. str.) in light of species identification. Feddes Repetorium 134: 271–288. [Google Scholar]
- Dörken VM, Stephan G, Stützel T. 2010. Morphology and anatomy of anomalous short shoots in Pinus (Pinaceae) and their evolutionary meaning. Feddes Repetorium 121: 133–150. [Google Scholar]
- Dörken VM, Stützel T. 2011. Morphology and anatomy of anomalous cladodes in Sciadopitys verticillata Siebold & Zucc. (Sciadopityaceae). Trees 25: 199–213. [Google Scholar]
- Dörken VM, Stützel T. 2012. Morphology, anatomy and vasculature of leaves in Pinus (Pinaceae) and its evolutionary meaning. Flora 207: 57–62. [Google Scholar]
- Du H, Ran J-H, Feng Y-Y, Wang X-Q. 2020. The flattened and needlelike leaves of the pine family (Pinaceae) share a conserved genetic network for adaxial-abaxial polarity but have diverged for photosynthetic adaptation. BMC Evolutionary Biology 20: 131. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Duarte MR, Souza DC, Costa RE. 2013. Comparative microscopic characters of Ginkgo biloba L. from South America and Asia. Latin American Journal of Pharmacy 32: 1118–1123. [Google Scholar]
- Eguchi N, Fukatsu E, Funada R, et al. 2004. Changes in morphology, anatomy, and photosynthetic capacity of needles of Japanese larch (Larix kaempferi) seedlings grown in high CO2 concentrations. Photosynthetica 42: 173–178. [Google Scholar]
- Elpe C, Knopf P, Stützel T, Schulz C. 2017. Diversity and evolution of leaf anatomical characters in Taxaceae s.l.—fluorescence microscopy reveals new delimiting characters. Journal of Plant Research 131: 125–141. [DOI] [PubMed] [Google Scholar]
- Endoh K, Fujikawa S, Arakawa K. 2012. Freezing behavior of cells in evergreen needles of fir (Abies sachalinensis). Cryobiology and Cryotechnology 58: 125–134. [Google Scholar]
- Enstone DE, Peterson CA, Ma F. 2003. Root endodermis and exodermis: structure, function, and responses to the environment. J Plant Growth Reg 21: 335–351. [Google Scholar]
- Esau K. 1977. Plant anatomy. New York: John Wiley & Sons. [Google Scholar]
- Ewers FW, Schmid R. 1981. Longevity of needle fascicles of Pinus longaeva (bristlecone pine) and other North American pines. Oecologia 51: 107–115. [DOI] [PubMed] [Google Scholar]
- Fahn A. 1990. Plant anatomy. New York: Pergamon Press. [Google Scholar]
- Gambles RL, Dengler NG. 1974. The leaf anatomy of hemlock, Tsuga canadensis. Canadian Journal of Botany 52: 1049–1056. [Google Scholar]
- Gao C, Marker SJV, Gundlach C, Poulsen HF, Bohr T, Schulz A. 2023. Tracing the opposing assimilate and nutrient flows in live conifer needles. Journal of Experimental Botany 74: 6677–6691. [DOI] [PubMed] [Google Scholar]
- Geldner N. 2013. The endodermis. Annual Review of Plant Biology 64: 531–558. [DOI] [PubMed] [Google Scholar]
- Gernandt DS, Holman G, Campbell C, et al. 2016. Phylogenetics of extant and fossil Pinaceae: methods for increasing topological stability. Botany 94: 863–884. [Google Scholar]
- Ghimire B, Lee C, Heo K. 2014. Leaf anatomy and its implications for phylogenetic relationships in Taxaceae s.l. Journal of Plant Research 127: 373–388. [DOI] [PubMed] [Google Scholar]
- Ghimire B, Lee C, Yang J, Heo K. 2015a. Comparative leaf anatomy of some species of Abies and Picea (Pinaceae). Acta Botanica Brasilica 29: 346–353. [Google Scholar]
- Ghimire B, Lee C, Yang J, Heo K. 2015b. Comparative leaf anatomy of native and cultivated Pinus (Pinaceae) in Korea: implications for the subgeneric classification. Plant Systematics and Evolution 301: 531–540. [Google Scholar]
- Gifford EM, Foster AS. 1989. Morphology and evolution of vascular plants. New York: W. H. Freeman. [Google Scholar]
- Glos RAE, Salzman S, Calonje M, et al. 2022. Leaflet anatomical diversity in Zamia (Cycadales: Zamiaceae) shows little correlation with phylogeny and climate. Botanical Review 88: 437–452. [Google Scholar]
- Gostin IN, Predan GMI. 2022. Structure and micromorphology of the leaf in some Taxus species. Analele Universităţii din Oradea, Fascicula Biologie XXIX: 221–227. [Google Scholar]
- Griffith MM. 1957. Foliar ontogeny in Podocarpus macrophyllus, with special reference to transfusion tissue. American Journal of Botany 44: 705–715. [Google Scholar]
- Griffith MP, Magellan TM, Tomlinson PB. 2014. Variation in leaflet structure in Cycas (Cycadales: Cycadaceae): does anatomy follow phylogeny and geography? International Journal of Plant Sciences 175: 241–255. [Google Scholar]
- Güvenc A, Hürkul MM, Erdem A. 2011. The leaf anatomy of naturally disturbed Juniperus L. (Cupressaceae) species in Turkey. Turkish Journal of Botany 35: 251–260. [Google Scholar]
- Hacke UG, Lachenbruch B, Pittermann J, Mayr S, Domec J-C, Schulte PJ. 2015. The hydraulic architecture of conifers. In: Hacke U. ed. Functional and ecological xylem anatomy. Switzerland: Springer, 39–75. [Google Scholar]
- Han W. 1984. A scanning electron microscopic observations of the leaves in some conifers. Acta Botanica Sinica 26: 276–380. [Google Scholar]
- Hermsen EJ, Taylor TN, Taylor EL, Stevenson DM. 2006. Cataphylls of the Middle Triassic cycad Antarcticycas schopfii and new insights into cycad evolution. American Journal of Botany 93: 724–738. [DOI] [PubMed] [Google Scholar]
- Herrera F, Shi G, Knopf P, et al. 2017. Cupressaceae conifers from the Early Cretaceous of Mongolia. International Journal of Plant Sciences 178: 19–41. [Google Scholar]
- Hu Y-S. 1984. Comparative anatomy of the leaves of Cephalotaxus (Cephalotaxaceae). Acta Phytotaxonomica Sinica 22: 289–296. [Google Scholar]
- Hu YS, Wang FH. 1984. Anatomical studies of Cathaya (Pinaceae). American Journal of Botany 71: 727–735. [Google Scholar]
- Hu Y-S, Yao B-J. 1981. Transfusion tissue in gymnosperm leaves. Botanical Journal of the Linnean Society 83: 263–272. [Google Scholar]
- Huusk V. 2024.Role of leaf anatomy in photosynthesis modifications during plant ontogeny and upon environmental acclimation. PhD Thesis, Estonian University of Life Sciences, Tartu, Estonia.
- Ivănescu L, Toma C, Rugină R. 2007. Histo-anatomical researches regarding some species of Cupressaceae. Analele ştiinţifice Univ AI I Cuza Iaşi 53: 34–39. [Google Scholar]
- Jeffrey EC. 1917. The anatomy of woody plants. Chicago: University of Chicago Press. [Google Scholar]
- Kaku S. 1971. A possible role of the endodermis as a barrier for ice propagation in the freezing of pine needles. Plant & Cell Physiology 12: 941–948. [Google Scholar]
- Kaplan DR. 2022. Kaplan’s principles of plant morphology. Boca Raton: CRC Press. [Google Scholar]
- Kausik SB. 1975. The leaf structure in Podocarpus brevifolius (Stapf.) Foxw. Proceedings of the Indian Academy of Sciences 81B: 197–206. [Google Scholar]
- Keng H, Little EL Jr. 1961. Needle characteristics of hybrid pines. Silvae Genetica 10: 131–146. [Google Scholar]
- Klimko M, Korszun S, Bykowska J. 2015. Comparative morphology and anatomy of the leaves of Ginkgo biloba L. cultivars. Acta Scientiarum Polonorum Hortorum Cultus 14: 169–189. [Google Scholar]
- Klymiuk AA, Stockey RA, Rothwell GW. 2011. The first organismal concept for an extinct species of Pinaceae: Pinus arnoldii Miller. International Journal of Plant Sciences 172: 294–313. [Google Scholar]
- Kuusk V. 2024. Role of leaf anatomy in photosynthetic modifications during plant ontogeny and upon environmental acclimation. PhD Thesis, Estonian University of Life Sciences, Estonia.
- Lamb SMA. 1923. Leaflets of Cycadaceae. Botanical Gazette 76: 185–202. [Google Scholar]
- Lee CJ. 1952. The anatomy and ontogeny of the leaf of Dacrydium taxoides. American Journal of Botany 39: 393–398. [Google Scholar]
- Leegood RC. 2007. Roles of the bundle sheath cells in leaves of C3 plants. Journal of Experimental Botany 59: 1663–1673. [DOI] [PubMed] [Google Scholar]
- Leigh A, Zwieniecki MA, Rockwell FE, Noyce CK, Nicotra AB, Holbrook NM. 2011. Structural and hydraulic correlates of heterophylly in Ginkgo biloba. New Phytologist 189: 459–470. [DOI] [PubMed] [Google Scholar]
- Lersten NR. 1997. Occurrence of endodermis with a Casparian strip in stem and leaf. Botanical Review 63: 265–272. [Google Scholar]
- Leslie AB, Beulieu J, Holman G, et al. 2018. An overview of extant conifer evolution from the perspective of the fossil record. American Journal of Botany 105: 1531–1544. [DOI] [PubMed] [Google Scholar]
- Liesche J, Martens HJ, Schulz A. 2011. Symplasmic transport and phloem loading in gymnosperm leaves. Protoplasma 248: 181–190. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lin J, Jach ME, Ceulemans R. 2001. Stomatal density and needle anatomy of Scots pine (Pinus sylvestris) are affected by elevated CO2. New Phytologist 150: 665–674. [Google Scholar]
- Lu Y, Ran J-H, Guo D-M, Yang Z-Y, Wang X-Q. 2014. Phylogeny and divergence times of gymnosperms inferred from single-copy nuclear genes. PLoS One 9: e107679. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lubna SA, Khan AL, Jan R, Khan A, Kim K-M, Lee I-J. 2021. The dynamic history of gymnosperm plastomes: insights from structural characterization, comparative analysis phylogenomics, and time divergence. Plant Genome 14: e20130. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lux A, Baláž M, Kummerová M, et al. 2017. Visual guide to plant anatomy. Prague: Academia. [Google Scholar]
- Ma F, Peterson CA. 2003. Current insights into the development, structure, and chemistry of the endodermis and exodermis of roots. Canadian Journal of Botany 81: 405–421. [Google Scholar]
- Mastroberti AA, Mairath JA. 2003. Leaf anatomy of Araucaria angustifolia (Bertol.) Kuntze (Araucariaceae). Revista Brasileira De Botanica 26: 343–353. [Google Scholar]
- Meicenheimer RD, Coffin DW, Chapman EM. 2008. Anatomical basis for biophysical differences between Pinus nigra and P. resinosa (Pinaceae) leaves. American Journal of Botany 95: 1191–1198. [DOI] [PubMed] [Google Scholar]
- Meyer CJ, Seago JL, Peterson CA. 2009. Environmental effects on the maturation of the endodermis and multiseriate exodermis of Iris germanica roots. Annals of Botany 103: 687–702. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Miller CN Jr 1992. Silicified Pinus remains from the Miocene of Washington. American Journal of Botany 79: 754–760. [Google Scholar]
- Miller CN Jr 1999. Implications of fossil conifers for the phylogenetic relationships of living families. Botanical Review 65: 239–277. [Google Scholar]
- Napp-Zinn K. 1966. Anatomie des Blattes. I. Blattanatomie der Gymnospermen. Handbuch der Pflanzenanatomie, Band VIII, Teil 1. Berlin: Borntraeger. [Google Scholar]
- Ng M, Smith SY. 2020. Evaluating stasis in Metasequoia (Cupressaceae): testing the relationship between leaf traits and climate. International Journal of Plant Sciences 181: 157–174. [Google Scholar]
- Oldham AR, Sillett SC, Tomsecu AMF, Koch GW. 2010. The hydrostatic gradient, not light availability, drives height-related variation in Sequoia sempervirens (Cupressaceae) leaf anatomy. American Journal of Botany 97: 1087–1097. [DOI] [PubMed] [Google Scholar]
- Pădure IM, Bădulescu L, Dediu T, Burzo I. 2008. Morpho-anatomical and phytochemical researches regarding Pseudotsuga menziesii (Mirbel) Franco (Pinaceae). Analele ştiintifice ale Universităţii Al. I. Cuza Inşi LIV: 33–39. [Google Scholar]
- Pant DD, Basu N. 1977. A comparative study of the leaves of Cathaya argyrophylla Chung & Kuang and three species of Keteleeria Carrière. Botanical Journal of the Linnean Society: Linnean Society of London 75: 271–282. [Google Scholar]
- Pazourek J, Votrubová O. 1997. Atlas of plant anatomy. Prague: Peres. [Google Scholar]
- Pensa M, Jalkanen R. 1999. Needle chronologies on Pinus sylvestris in northern Estonia and southern Finland. Silva Fennica 33: 171–177. [Google Scholar]
- Popham RA. 1966. Laboratory manual for plant anatomy. St Louis, MO, USA: Mosby. [Google Scholar]
- Qin X, Sun N, Ma L, Chang Y, Mu L. 2014. Anatomical and physiological responses of Colorado blue spruce to vehicle exhausts. Environmental Science and Pollution Research 21: 11094–11098. [DOI] [PubMed] [Google Scholar]
- Rivera BK, Sáez PL, Cavieres LA, et al. 2022. Anatomical and biochemical evolutionary ancient traits of Araucaria araucana (Molina) K. Koch and their effects on carbon assimilation. Tree Physiology 42: 1957–1974. [DOI] [PubMed] [Google Scholar]
- Robison CR. 1974. A new species of Prepinus from the late Cretaceous of New Jersey. American Journal of Botany 61: 18. [Google Scholar]
- Robison CR. 1977. Pinus triphylla and Pinus quinquefolia from the Upper Cretaceous of Massachusetts. American Journal of Botany 64: 726–732. [Google Scholar]
- Roden JS, Canny MJ, Huang CX, Ball MC. 2009. Frost tolerance and ice formation in Pinus radiata needles: ice management by the endodermis and transfusion tissues. Functional Plant Biology 36: 180–189. [DOI] [PubMed] [Google Scholar]
- Roden JS, Kahmen A, Buchmann N, Siegwolf R. 2015. The enigma of effective path length for 18O enrichment in leaf water of conifers. Plant, Cell & Environment 38: 2551–2565. [DOI] [PubMed] [Google Scholar]
- Sacher JA. 1955. Cataphyll ontogeny in Pinus lambertiana. American Journal of Botany 42: 82–91. [Google Scholar]
- Sack L, Scoffoni C. 2013. Leaf venation: structure, function, development, evolution, ecology and applications in the past, present and future. New Phytologist 198: 983–1000. [DOI] [PubMed] [Google Scholar]
- Sage RF, Khoshravesh R. 2016. Passive CO2 concentration in higher plants. Current Opinion in Plant Biology 31: 58–65. [DOI] [PubMed] [Google Scholar]
- Sahni B, Mitra AK. 1927. Notes on the anatomy of some New Zealand species of Dacrydium. Annals of Botany 41: 75–90. [Google Scholar]
- Saxton WT. 1930. Notes on conifers: IV. Some points in the leaf anatomy of Fokienia hodginsii Henry and Thomas and Libocedrus macrolepis B. and H. Annals of Botany 44: 167–171. [Google Scholar]
- Schreiber L, Franke RB. 2011. Endodermis and exodermis in roots. eLS. Chichester: John Wiley. 10.1002/9780470015902.a0002086.pub2. [DOI] [Google Scholar]
- Seago JL Jr. 2002. The root cortex of the Nymphaeaceae, Cabombaceae, and Nelumbonaceae. Journal of the Torrey Botanical Society 129: 1–9. [Google Scholar]
- Seago JL Jr. 2020. Revisiting the occurrence and evidence of endodermis in angiosperm shoots. Flora 273: 151709. [Google Scholar]
- Seago JL Jr, Fernando DD. 2013. Anatomical aspects of angiosperm root evolution. Annals of Botany 112: 223–238. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Seago JL Jr, Peterson CA, Enstone DE, Scholey CA. 1999. Development of the endodermis and hypodermis of Typha glauca Godr. and Typha angustifolia L. roots. Canadian Journal of Botany 77: 122–134. [Google Scholar]
- Sethi P, Gupta H, Ellakiya K, Kumar B. 2020. Histology of a xerophytic leaflet: Cycas circinalis and Cycas revoluta (Cycadaceae). Chemistry & Material Sciences Research Journal 2: 6–14. [Google Scholar]
- Shiraki A, Azuma W, Kuroda K, Ishii HR. 2016. Physiological and morphological acclimation to height in cupressoid leaves of 100-year-old Chamaecyparis obtuse. Tree Physiology 37: 1327–1336. [DOI] [PubMed] [Google Scholar]
- Skärby L, Ro-Poulsen H, Wellburn FM, Shepphard LJ. 1998. Impacts of ozone on forests: a European perspective. New Phytologist 139: 109–122. [Google Scholar]
- Soar I. 1922. The structure and function of the endodermis in the leaves of the Abietineae. New Phytologist 21: 269–292. [Google Scholar]
- Song C, Shen W, Du L, Wen J, Lin J, Li R. 2019. Development and chemical characterization of Casparian strips in the roots of Chinese fir (Cunninghamia lanceolata). Trees 33: 827–836. [Google Scholar]
- Soukup A, Tylová E. 2018. Apoplastic barriers: their structure and function from a historical perspective. In: Sahi VP, Baluška F. eds. Concepts in cell biology – history and evolution, Plant Cell Monographs, Vol. 23. Springer. 155–183. [Google Scholar]
- Soukupová J, Cvikrová M, Albrechtová J, Rock BN, Eder J. 2000. Histochemical and biochemical approaches to the study of phenolic compounds and peroxidases in needles in needles of Norway spruce (Picea abies). New Phytologist 146: 403–414. [Google Scholar]
- Soukupová J, Rock BN, Albrechtová J. 2001. Comparative study of two spruce species in a polluted mountainous region. New Phytologist 150: 133–145. [Google Scholar]
- Stegner M, Buchner O, Gesslbauer M, et al. 2023. Frozen mountain pine needles: the endodermis discriminates between the ice-containing central tissue and the ice-free fully functional mesophyll. Physiologia Plantarum 175: e13865. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Sterling C. 1949. Some features in the morphology of Metasequoia. American Journal of Botany 36: 461–471. [Google Scholar]
- Stevenson DW, Norstog KJ, Molsen DV. 1996. Midribs of cycad pinnae. Brittonia 48: 67–74. [Google Scholar]
- Stockey RA, Kvaček J, Hill RS, Rothwell GW, Kvaček Z. 2005. The fossil record of Cupressaceae s. lat. In: Farjon A. ed. A monograph of Cupressaceae and Sciadopitys. Kew: Royal Botanic Gardens, 54–68. [Google Scholar]
- Stockey RA, Ueda Y. 1986. Permineralized pinaceous leaves from the Upper Cretaceous of Hokkaido. American Journal of Botany 73: 1157–1162. [Google Scholar]
- Sutinen M-L, Repo T, Sutinen S, Lasarov H, Alvila L, Pakkanen TT. 2000. Physiological changes in Pinus sylvestris needles during early spring under sub-Arctic conditions. Forest Ecology and Management 135: 217–228. [Google Scholar]
- Taylor TN, Taylor EL, Krings M. 2009. Paleobotany: the biology and evolution of fossil plants. Burlington: Elsevier. [Google Scholar]
- Thomson RB. 1913. On the comparative anatomy and affinities of the Araucariaceae. Philosophical Transactions of the Royal Society B 204: 1–50. [Google Scholar]
- Tomlinson PB, Fisher JB. 2005. Development of non-lignified fibers in leaves of Gnetum gnemon (Gnetales). American Journal of Botany 92: 383–389. [DOI] [PubMed] [Google Scholar]
- Tomlinson PB, Takaso T, Rattenbury JA. 1989. Developmental shoot morphology in Phyllocladus (Podocarpaceae). Botanical Journal of the Linnean Society 99: 223–248. [Google Scholar]
- Uchida N, Lee JS, Horst RJ, et al. 2012. Regulation of inflorescence architecture by intertissue layer ligand-receptor communication between endodermis and phloem. Proceedings of the National Academy of Sciences of the United States of America 109: 6337–6342. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Valderrama-Martin JM, Ortigosa F, Canton FR, Avila C, Cañas RA, Cánovas FM. 2024. Emerging insights into nitrogen assimilation in gymnosperms. Trees 38: 273–286. [Google Scholar]
- Vasheka O, Spoletini A, Catoni R, Pepe M, Gratani L. 2019. Age-related changes in anatomical and morphological leaf traits of Wollemia nobilis. Modern Phytomorphology 13: 47–53. [Google Scholar]
- Vovides AP, Pérez-Farrera MA, Gutiérrez-Ortega JS, et al. 2020. Cycad anatomy: old techniques solving new problems. Cycads 5: 33–44. [Google Scholar]
- Way DA, Sage RF. 2008. Elevated growth temperatures reduce the carbon gain of black spruce [Picea mariana (Mill.) B.S.P.]. Global Change Biology 14: 624–636. [Google Scholar]
- Wilcox HE. 1968. Morphological studies of the root of red pine, Pinus resinosa. I. Growth characteristics and patterns of branching. American Journal of Botany 55: 247–254. [Google Scholar]
- Wu X, Lin J, Lin Q, Wang J, Schreiber L. 2005. Casparian strips in needles are more solute permeable than endodermal transport barriers in roots of Pinus bungeana. Plant and Cell Physiology 46: 1799–1808. [DOI] [PubMed] [Google Scholar]
- Yang Y, Ferguson DK, Liu B, et al. 2022. Recent advances on phylogenomics of gymnosperms and a new classification. Plant Diversity 44: 340–350. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Yang C, Zhang X, Wang T, et al. 2019. Phenotypic plasticity in the structure of fine adventitious Metasequoia glyptostroboides roots allows adaptation to aquatic and terrestrial environments. Plants 8: 501–512. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Yao B-J, Hu Y-S. 1982. Comparative anatomy of conifer leaves. Acta Phytotaxonomica Sinica 20: 275–294. [Google Scholar]
- Yousif I, Shahbaz SE. 2023. Comparative anatomical study of leaves in five Cupressaceae Gray. Taxa in Kurdistan region – Iraq. Science Journal of University of Zakho 11: 22–29. [Google Scholar]
- Zang M, Su Q, Weng Y, et al. 2019. Complete chloroplast genome of Fokienia hodginsii (Dunn) Henry et Thomas: insights into repeat regions variation and phylogenetic relationships in Cupressaceae. Forests 10: 528. [Google Scholar]
- Zhang Q, Shen X, Jiang X, Fan T, Liang X, Yan W. 2023. MaxEnt modeling for predicting suitable habitat for endangered tree Keteleeria davidiana (Pinaceae) in China. Forests 14: 302–315. [Google Scholar]






