CC BY
25
FULL COMMUNICATIONS
PLANT SCIENCE
Structural organization of epidermis in leaves of extant species of Gnetum L. (Gnetales) and Middle Jurassic Bennettitales
Ianina Bogdanova and Lyudmila Kartseva
Laboratory of Paleobotany, Komarov Botanical Institute, Russian Academy of Sciences, ul. Professora Popova, 2, Saint Petersburg, 197022, Russian Federation
Address correspondence and requests for materials to Ianina Bogdanova, [email protected]
Abstract
Citation: Bogdanova, I. and Kartseva, L. 2023. Structural organization of epidermis in leaves of extant species of Gnetum L. (Gnetales) and Middle Jurassic Bennettitales. Bio. Comm. 68(2): 105-121. https://doi. org/10.21638/spbu03.2023.205
Authors' information: Ianina Bogdanova, MSc, Curator of Collections, orcid.org/0000-0002-0378-0773; Lyudmila Kartseva, Chief Engineer, orcid.org/0000-0002-8183-3033
Manuscript Editor: Anton Nizhnikov, Department of Genetics and Biotechnology, Faculty of Biology, Saint Petersburg State University, Saint Petersburg, Russia
Received: March 22, 2022;
Revised: September 30, 2022;
Accepted: March 16, 2023.
Copyright: © 2023 Bogdanova and Kartseva. This is an open-access article distributed under the terms of the License Agreement with Saint Petersburg State University, which permits to the authors unrestricted distribution, and self-archiving free of charge.
Funding: The reported study was funded by RFBR according to research project No. 18-34-00912 and partly was carried out on the equipment of the Center of Collective Research "Cellular and Molecular Technologies of Studying Plants and Fungi" BIN RAS (St. Petersburg).
Ethics statement: This paper does not contain any studies involving human participants or animals performed by any of the authors.
Competing interests: The authors have declared that no competing interests exist.
An attempt to assess the identification and evaluation of correlations between the epidermal characters of leaves in Bennettitales and the comparison of the results with the structural organization of leaves of extant plants has been carried out. Epidermis of four species of Gnetum L. and four species of Jurassic Bennettitales (Nilssoniopteris Nathorst and Ptilophyllum Morris) was studied. The coefficients of variation for epidermal characters Gnetum range from 6.4 to 24.0%, Ptilophyllum — 15.7-63.5% and Nilssoniopteris — 18.0-39.9%. In both groups of plants, the sinuosity of the tangential cell walls of the epidermal cells in the upper and lower epidermis is a stable character (Cv < 18.0 %). In the Ptilophyllum and Gnetum, the length of stomata demonstrates a low level of variability (Cv < 16.8%). A significant range of the coefficients of variation in both Gnetum and Bennettitales show the number of epidermal cells per 1 mm2 of the upper and lower epidermis (17.5 % < Cv < 31.9 %), the area of the epidermal cells in the upper and lower epidermis (21.2 % < Cv < 63.5 %) and the number of stomata per 1 mm2 of epidermis (29.3 % < Cv < 39.9 %). Similarities in the correlation structure of the epidermal characters are revealed in correlations between sinuosity of the tangential cell walls of the epidermis, the number of stomata per 1 mm2 and their size; the length of stomata and the number of epidermal cells; stomatal index and the number of epidermal cells of the epidermis. In Gnetum, the number of differentiated stomata correlates with the number of aborted stomata. In N. angustifolia and P. caucasicum, the number of stomata correlates with the number of papillae per 1 mm2. Similarities in correlational structure of epidermis in Nilssoniopteris, Ptilophyllum, and Gnetum could be ecological adaptations or ontogenetic characters, such as the development of stomata that has been described by other researchers. Keywords: paleobotany, plant fossils, Bennettittales, Gnetum, leaf, structural organization
Introduction
Gnetales and Bennettitales are two taxa of gymnosperms whose evolution occurred in parallel (Crane, 1996; Donoghue and Doyle, 2000; Friis et al., 2007). Both taxa appeared at approximately the same time, at the beginning of the Me-sozoic era: Gnetales — at the Permian-Triassic boundary (Wang, 2004), Bennettitales — at Late Triassic (Krassilov and Bugdaeva, 1988). They developed under similar ecological conditions (Crane, 1996). In this regard, both groups revealed some common structural characters. For example, bisexual strobili (Carmichael and Friedman, 1996; Friedman, 1998; Donoghue and Doyle, 2000; Kiritchkova and Nosova, 2012): The male cones of extant Gnetales have nonfunctional ovules, which are considered by some authors as traces of the past bisexuality in ancestral plants (Haycraft and Carmichael, 2001). Another example is the similarity in the structure of epidermis and stomata. Thomas and Bancroft (Thomas and Ban-
croft, 1913) described the epidermis of Bennettitales and found that the stomata are paracytic. This stomatal type is not found in cycads, to which the Bennettitales were originally assigned, and among other gymnosperms it is known only in Gnetum and Welwitschia (Florin, 1931, 1933; Nautiyal, Singh, and Pant, 1976) and Podozamites (Shi et al., 2017), but it is also widely distributed among flowering plants (Donoghue and Doyle, 2000).
Florin (Florin, 1931, 1933) described the development of stomata in Bennettitales: The mother cell of the stomata in the protoderm divides unequally twice, leading to the formation of subsidiary cells, and as a result of the final equal division, guard cells of the stomata are formed. Thus, the origin of subsidiary and guard cells is the same — from one meristemoid. However, epidermal cells at the poles of the stomata develop from neighboring protodermal cells — they have a different origin than subsidiary and guard cells. This shows the similarity in the development of the stomatal apparatus in leaves of Bennettitales and Gnetum species (Rudall and Bateman, 2019).
Indian researchers (Nautiyal, Singh, and Pant, 1976), who studied in detail the histogenesis of epidermis in Gnetum, found that the paracytic stomata has a dual origin: guard cells and subsidiary cells are formed from one meristemoid — mesogene origin, and neighboring epidermal cells at the poles of stoma arise from other cells of protoderm — perigene origin (Rudall and Bateman, 2019). According to the classification of D. D. Pant (Pant, 1965), this is a variant of meso-perigenic stomata.
Angiosperms have a wide range of ways of origin of paracytic stomata, depending on whether the cells surrounding the stomata develop from the same mother cell as the guard cells, or from other protodermal cells, or also have a dual origin (Pant, 1965).
Stomatal traits have played a significant role in inferring seed-plant phylogeny, and the presence of paracytic stomata in Bennettitales represents a synapomorphy linking them with angiosperms in most analyses. Rudall and Bateman (Rudall and Bateman, 2019) conclude that the stomata of Gnetum represent by far the most likely analogue for stomatal development in Bennettitales.
The leaf epidermis of Bennettitales is well-investigated (Florin, 1931, 1933; Doludenko and Svanidze, 1969; Kiritchkova and Nosova, 2012). However, studies comparing quantitative characters and revealing the structural organization of epidermis of extinct plants using multivariate statistics methods have not been carried out. We performed an attempt to adapt a method of principal component analysis for the leaves of fossil plants, which could be useful in traditional studies of taxonomy and phylogeny and become an alternative to molecular taxonomy. Due to this analysis, it is possible to compare the structures of extinct and extant plants.
The purpose of our study was to estimate the common variability and correlations between the structural characters of the epidermis in extinct species of Bennettitales (genera Nilssoniopteris Nathorst and Ptilophyllum Morris) and extant species of Gnetum, and to evaluate common characters and differences or similarities in the structural organization of leaves.
Fossilized leaves of Bennettitales differ from each other, ranging from a few centimeters to more than a meter in length. The leaves are predominantly unipin-nate. Rarely, the leaves are simple. The species are described according to the epidermal characters of leaves, in particular, the structure of the stomatal apparatus, the shape of the cells, the presence of papillae and trichomes (Rudall and Bateman, 2019). Two genera of Jurassic Bennettitales were the objects of our study: Nilssoniop-teris with a simple leaf blade and Ptilophyllum with a leaf blade dissected into segments.
The cuticles, on which imprints of cells of epidermis are present, are well-preserved in fossils. This makes it possible not only to describe the structure accurately, but also to evaluate the quantitative characters, and to try to identify, at least partially, some characters of the structural organization of leaves in Bennettitales. For example, to assess the correlations between the structural characters of leaf epidermis. The results obtained for epidermis of extinct plants were compared with similar data for extant Gnetum via cuticular analysis (Kerp, 1990).
In recent decades, many fossils of Gnetales have been found (Krassilov and Bugdaeva, 1988; Rydin and Friis, 2010; Friis et al., 2007), but there are no species described on epidermal characters (Crane, 1996). However, the epidermis of reproductive structures and shoots has been described (Rydin and Friis, 2010). According to molecular data, the Gnetum arose approximately in the Eocene or Middle Miocene. Of all the extant Gne-tales, only Gnetum is not represented in the fossil record (Won and Renner, 2006), possibly due to the fact that their leaves are similar to the leaves of flowering plants, and it is difficult to distinguish them in fossils by morphological and anatomical characters (Crane, 1996). Only Protognetum from Middle Jurassic is known, which combines the vegetative characters of Ephedra with reproductive characters of Gnetum (Yang, Xie, and Ferguson, 2017). Gambaryan and Kuznetsov (Gambaryan and Kuznetsov, 2021), creating a model of a hypothetical ancestor of flowering plants, suggest that the hypothetical plant of Gnetales from the Jurassic could resemble in appearance the modern tree species Gnetum gnemon L.
Unlike other modern genera (leaves of Ephedra L. are reduced, and Welwitschia have only two leaves, which grow constantly due to intercalary meristem), species of Gnetum have simple leaves, which are ideal model objects to compare with the leaves of Bennettitales.
Materials and methods
Plant material. Fossils of Bennettitales were taken from collection No. 1073, Laboratory of Paleobotany, BIN RAS. The samples were collected by G. V. Delle (1980) in Georgia, river Barula, Tsesi village. Dating: Middle Jurassic (Callovian) (Doludenko and Svanidze, 1969).
The cuticle of 73 specimens belonging to 2 genera was studied. Ianina O. Bogdanova made the determination.
Nilssoniopteris Nathorst — 2 species:
N. angustifolia Doludenko, 9 samples: BIN 1497, 1498, 1521, 1536, 1538, 1540, 1541, 1560, 1564;
N. longifolia Doludenko, 16 samples: BIN 1509, 1510, 1519, 1520, 1525, 1526, 1527, 1528, 1530, 1532, 1537, 1550, 1556, 1557, 1562, 1563.
Ptilophyllum Morris — 2 species:
P. caucasicum Doludenko et Svanidze, 20 samples: BIN 1573, 1586, 1593, 1595, 1597, 1598, 1599, 1608, 1613, 1619, 1620, 1621, 1623, 1639, 1642, 1649, 1650, 1651, 1656, 1683;
P. okribense f. ratchense Doludenko et Svanidze, 28 samples: BIN 1568, 1572, 1581, 1584, 1589, 1592, 1600, 1601, 1602, 1604, 1606, 1607, 1615, 1617, 1622, 1631, 1632, 1635, 1637, 1638, 1646, 1647, 1648, 1657, 1658, 1672, 1679, 1680.
Gnetum L.: upper and lower epidermis from 76 mature leaves belonging to 4 species was studied.
G. gnemon L. — 24 leaves: 21 leaves were collected in greenhouse No. 20 of the Botanical Garden BIN RAS, 3 leaves were taken from herbarium LE, collected in Vietnam: LE 01068309, LE 01068330.
G. gracilipes C. Y. Cheng — 16 leaves. The samples were collected in Yunnan Province, China (N 21.63, E 101.44, 2018).
G. latifolium Blume — 11 leaves. The material was taken from the herbarium of the BIN RAS (LE), collected in Yunnan Province and Hainan Island (China), Vietnam: LE 01068250, LE 01068251, LE 01068253, LE 01068254, LE 01068256, LE 01068257, LE 01068259, LE 01068260, LE 01068262, LE 01068263, LE 01068301.
G. montanum Markgr. — 25 leaves: 21 leaves were collected in greenhouse No. 20 of the Botanical Garden BIN RAS and 4 leaves were taken in LE herbarium, collected in Laos and Vietnam: LE 01068323, LE 01068326, LE 01068327.
Material selection. Gnetum: fragments of epidermis from the middle part of the leaf blade, located between the main vein and the leaf margin, were studied. Living material was fixed in 70° ethanol. Bennet-titales: the fossil material was taken from the middle part of a whole leaf blade (Nilssoniopteris), or from the middle part of a leaf segment or a whole segment (Ptilophyllum).
Herbarized material. Herbarized leaf fragments were placed in a mixture of glycerol, distilled water and ethanol (70°) in a ratio of 1 : 1 : 1 in a thermostat (temperature 50 °C). The material was washed in distilled water and used for further research.
Maceration of epidermis. Fragments of the leaves of Bennettitales were kept in hydrofluoric acid (HF). Fragments of the leaves of gnetums and Bennettitales were placed in a solution of KClO3 and HNO3 (Yang, Xie, and Ferguson, 1976). Then they were washed with distilled water and placed in a KOH solution. After that, the fragments were again washed with distilled water. Using a dissecting needle, the upper and lower epidermis of gnetums and the cuticle of Bennettitales were separated. The separated epidermis and cuticle were placed in 70° etha-nol. Subsequently, fragments of the epidermis of gnetums were stained with safranin. Fragments of the epidermis and the fossil cuticle were poured into a glycerol-gelatin.
Light and scanning electron microscopy. Viewing and photographing of the slides were carried out using a Leica EZ4 binoculars, Leica DM500 microscopes, Leica DM1000 microscope and Leica EC3 digital camera (Germany). The obtained data were processed in the ImageJ. Photographing of the fossilized leaves of Bennettitales was carried out using a Canon EOS 770 camera.
For scanning electron microscopy, the leaf fragments were dehydrated in a series of ethanol of increasing concentrations (20°, 50°, 70°, 80°, 90°, 96°, 100°). Then, they were dipped into mixtures of acetone and ethanol (100°), acetone and isoamyl acetate and iso-amyl acetate only. Dehydrated specimens underwent critical point drying using liquid carbon dioxide (CO2). Dried objects were applied to a stub and sprayed with ions of gold. We used a scanning electron microscope JSM-6390LA (Japan) for the examination of the prepared specimens.
Description of the epidermis. The cuticular analysis was used for description of fossils (Kerp, 1990). A feature of the inner surface of the cuticle of fossil plants is that the surface of the epidermis is imprinted on it: a sculpture of epidermal cells, stomata and trichomae. Accordingly, it becomes possible to analyze the structure of epidermis of extinct plants and compare it with the epidermis of extant plants according to key characters.
For description of qualitative characters of the epidermis, the classification of Vasiliev (Vasiliev, 1988) was used; for description of shapes, sinuosity of tangential walls and projections of the epidermal cells the classification of Zakharevich (Zakharevich, 1954) was used.
Statistical methods. The obtained data were processed in the Statistica 8.0. The coefficient of variation
(Cv) was calculated to determine the general variability of the characters. The assessment of the level of common variability of characters was carried out in accordance with the classification of Mamaev (Mamaev, 1973).
The coefficient of sinuosity of the tangential cell walls of the epidermal cells in upper epidermis was calculated using the formula (Vasiliev, 1988):
Supp
Sin_upp — sinuosity of epidermal cells in lower epidermis; Per_upp — perimeter of epidermal cells in upper epidermis, ^m; Supp — area of epidermal cells in upper epidermis, ^m2. The sinuosity of epidermal cells in lower epidermis was calculated in the same way.
Correlations between characters were determined using principal component analysis (Kendall and Stewart, 1976). For statistical methods, 13 characters that describe the structure of epidermis were selected. They are listed in tables 1 and 2.
Table 1. Statistical characteristics of the characters of the leaf structure for Gnetum species
Characters G. gnemon G. gracilipes
M a mx Xmin Xmax M a mx Xmin Xmax
Ncell_upp 2875 361 587 2166 3425 1537 375 384 1120 2567
Supp 319 49 65 249 409 732 163 183 380 961
Per_upp 99 10 20 79 125 144 21 36 92 171
Sin_upp 1.39 0.15 0.28 1.08 1.66 1.34 0.07 0.33 1.19 1.42
NcelMow 2401 370 490 1609 3040 1414 480 475 972 1856
Slow 495 121 101 367 814 873 247 218 432 1273
Per_low 119 19 24 92 153 160 26 40 103 196
Sin_low 1.34 0.14 0.27 1.05 1.56 1.37 0.06 0.34 1.25 1.50
Nstom 296 84 60 188 543 205 68 51 135 438
Ndstom 202 52 41 127 295 182 48 46 122 336
Lstom 25 2 5 23 30 29 1 7 27 32
Kstom 7.12 1.65 1.45 4.88 11.76 7.50 0.41 1.88 6.55 8.00
G. latifolium G. montanum
Ncell_upp 1231 335 371 978 2179 906 188 181 424 1312
Supp 819 146 247 465 1061 1112 325 222 718 2393
Per_upp 120 24 36 77 162 176 36 35 117 299
Sin_upp 1.13 0.08 0.34 1.01 1.24 1.32 0.12 0.26 1.06 1.53
Ncell_low 1101 73 332 990 1250 970 145 194 521 1187
Slow 1061 171 320 714 1293 1240 335 248 809 2457
Per_low 146 27 44 90 192 193 30 39 148 285
Sin_low 1.21 0.09 0.36 1.07 1.33 1.39 0.11 0.28 1.22 1.65
Nstom 168 43 51 121 285 162 29 31 98 232
Ndstom 151 36 46 118 248 137 27 27 72 213
Lstom 32 2 10 28 36 30 2 6 27 33
Kstom 10.66 1.38 3.21 9.09 14.18 10.08 1.04 2.02 7.88 12.56
Notes: M — mean; mx — standard error of the mean; o — standard deviation; Xmin — minimum; Xmax — maximum.
Characters: Ncell_upp — number of epidermal cells per 1 mm2 of upper epidermis; Supp — area of epidermal cells in upper epidermis, |_im2; Per_upp — perimeter of epidermal cells in upper epidermis, |_im; Sin_upp — sinuosity of epidermal cells in lower epidermis; NcelMow — number of epidermal cells per 1 mm2 of lower epidermis; Slow — area of epidermal cells in the lower epidermis, |_im2; Per_low — perimeter of epidermal cells in lower epidermis, |_im; Sin_low — sinuosity of epidermal cells in lower epidermis; Nstom — number of stomata per 1 mm2 of epidermis; Ndstom — number of differentiated stomata per 1 mm2 of epidermis; Lstom — length of stomata, |_im; Kstom — stomatal index, %.
Results and discussion
Morphological and anatomical characteristics of the
mature leaves of Gnetum species. The studied leaves are simple, divided into blade and petiole, small, with a blade area of 11.5-18 cm2 (G. gnemon, G. monta-
num) or medium size, with a blade area of 40-152 cm2 (G. gnemon, G. gracilipes, G. latifolium, G. montanum) (Fig. 1A, E; Fig. 2A, E).
Leaves are hypostomatic. In the upper epidermis above large veins of G. gnemon, G. gracilipes, G. monta-num occasional stomata are noted.
Fig. 1. Epidermal structure of Gnetum leaves. G. gnemon: A — leaf; B — fragment of upper epidermis; C — fragment of lower epidermis with differentiated (s) and aborted (as) stomata; D — fragment of lower epidermis with stomata. G.gracilipes: E — leaf; F — fragment of upper epidermis; G — fragment of lower epidermis with differentiated and aborted stomata.
Scale bar: A, E — 1 cm; B, D, E, G — 50 pm; C — 10 pm.
Fig. 2. Epidermal structure of Gnetum leaves. G. latifolium: A — leaf; B — fragment of upper epidermis; C — fragment of lower epidermis with differentiated (s) and aborted (as) stomata. G. montanum: D — leaf; E — fragment of upper epidermis; F — stoma in lower epidermis; G — fragment of lower epidermis with stomata.
Scale bar: A, D — 1 cm; B, C, E — 50 jm; F — 10 jm; G — 30 jm.
The upper epidermis (Fig. 1B, F; Fig. 2B, E; Table 1) consists of very small-sized cells (~3400 cells per 1 mm2) in G.gnemon; small-sized cells (1600-2500 cells per 1 mm2) in G.gnemon and G. latifolium; medium-sized cells (900-1600 cells per 1 mm2) in G. gracilipes, G. lati-
folium and G. montanum; large-sized cells (400-900 cells per 1 mm2) in G. gracilipes and G. montanum.
The shape of the epidermal cells is diverse. The projection of the cell area above the mesophyll is square, rectangular or polygonal; above small and large veins the
projection is rectangular or square. The outlines are rectilinear, wavy, sinuous or sinuous-wavy. The coefficient of sinuosity of the tangential cell walls ranges from 1.006 to 1.66. Minimal values (< 1.1) are in the epidermis of G. gnemon, G. latifolium, and G. montanum; maximal values (1.66) are in the epidermis of G. gnemon.
The lower epidermis (Fig. 1C, D, F, G; Fig. 2C, D, F, G) consists of small-sized cells (2400-3040 cells per 1 mm2) in G.gnemon; medium-sized cells (1350-1800 cells per 1 mm2) in G. gnemon, G. gracilipes and G. latifolium; large-sized cells (1000-1250 cells per 1 mm2) in G. gracili-pes, G. latifolium and G. montanum; very large-sized cells (~520 cells per mm2) in G. montanum.
The shape of the epidermal cells of the lower epidermis is also diverse. The projection of their area above the mesophyll is elongated, polygonal or rectangular; above small and large veins it is rectangular. The outlines are wavy, large-wavy, rectilinear, sinuous, sinuous-wavy. The coefficient of sinuosity of tangential cell walls ranges from 1.05 to 1.65. Minimal values (< 1.1) are in the epidermis of G. gnemon and G. latifolium; maximal values (1.65) are in G. montanum.
There are very few stomata on the surface of epidermis (~70 stomata per 1 mm2) in G. montanum; few stomata (110-120 stomata per 1 mm2) in G. gracilipes, G. latifolium and G. montanum; average number of stomata (210-230 stomata per mm2) in G. gnemon, G.gra-cilipes, G. latifolium and G. montanum; many stomata (275-330 stomata per 1 mm2) in G. gnemon, G. gracilipes and G. latifolium. The stomata are arranged irregularly. In the areas of the epidermis above large veins, occasional stomata are noted.
The stomatal Index is very low (4.9 %) in the epidermis of G. gnemon; low (6.5-11.8 %) in G. gnemon, G. gracilipes, G. latifolium and G. montanum; medium (12.6-14.2 %) in G. latifolium and G. montanum.
The predominant type of stomata is paracytic (Fig. 1C, D, G; Fig. 2C, F, G). There were noted latero-cytic, anomocytic, encyclocytic, hemiparacytic types, as well as transitional variants of stomata (for example, transitional to encyclocytic type in G. gnemon, G. latifo-lium and G. montanum; transitional to paracytic type in G. latifolium; transitional to laterocytic and hemipara-cytic types in G. montanum). Aborted stomata are noted in the epidermis of G. gnemon, G. gracilipes, and G. latifolium (Fig. 1C, G; Fig. 2C). In all the studied specimens on the surface of the lower epidermis, cork warts were found. They distributedabove the mesophyll and along large veins.
Morphological and anatomical characteristics of the leaves of Bennettitales
Nilssoniopteris. N. angustifolia. The leaves are simple, linear, gradually narrowed at the apex and base (Fig. 3A). The apex is pointed. The petiole is thin. The
veins are thin, simple, and dichotomous near the origin of the shaft. The length of the leaves is more than 100 mm; the width is 5.5-23 mm in the middle part.
The upper epidermis (Fig. 3B, Table 2) undivided into costal and intercostal fields, consists of large-sized cells (530-900 cells per 1 mm2), or medium-sized cells (900-1200 cells per 1 mm2). There are more or less distinct longitudinal rows of cells. The projection of cell area is square or elongated; outlines are sinuous-wavy. The coefficient of sinuosity of the tangential cell walls ranges from 1.2 to 2.1. Trichomae are absent.
The lower epidermis (Fig. 3C-G) consists of medium-sized cells (1350-1450 cells per 1 mm2), or large-sized cells (610-1350 cells per 1 mm2). The epidermis is divided into costal and stomatal fields (Fig. 3E). The projection of the area of the epidermal cells in the stomatal fields is round or elongated; outlines are sinuous-wavy or low-sinuous. In costal fields, cells are arranged in rows. Their projection is rectangular or square; outlines are sinuous-wavy, wide-wavy, low-sinuous. The coefficient of sinuosity of the tangential cell walls ranges from 1.2 to 1.7. Most of the epidermal cells in the lower epidermis, both in stomatal and costal fields, have one central convex-shaped papilla; their imprints are well preserved on the cuticle (Fig. 3C, D, F). The number of papillae varies from 300 to 745 per 1 mm2. Trichomae are absent.
Leaves are hypostomatic. There are from one to three stomata per width of the stomatal field (Fig. 3E, G). Their apertures are oriented at different angles within the field. Occasional stomata can occur in costal fields and above large veins. The type of stomata is paracytic (Fig. 3F, G). The stomatal index is low (9.5-11 %), medium (11-16 %) or large (16-17.0 %). There are very few (~70 per 1 mm2), few (100-160 per 1 mm2), average (160-250 per 1 mm2), and many stomata (~260 per 1 mm2) on the epidermal surface. The tangential cell walls of subsidiary cells are even and thick.
N. longifolia. The leaves are simple, linear, narrow, gradually and very slightly narrowed at the apex and base (Fig. 4A). The apex is rounded. The petiole is thin. The veins are thin. The length of the leaves is more than 190 mm; the width is 9.1-24.7 mm in the middle part.
The upper epidermis (Fig. 4B, Table 2) consists of medium-sized cells (900-1300 cells per 1 mm2), or large-sized cells (440-900 cells per 1 mm2). The projection of the area of the epidermal cells is rectangular or square, above the veins it is elongated. The outlines are sinuous-wavy, wide-wavy. The coefficient of sinuosity of the tangential cell walls ranges from 1.1 to 1.9. Tricho-mae are absent.
The lower epidermis (Fig. 4C-G) consists of large-sized cells (510-600 cells per 1 mm2) or very large-sized cells (600-1100 cells per 1 mm2). The epidermis is divided into costal and stomatal fields (Fig. 4C). The projection of the area of the epidermal cells in the stomatal
Fig. 3. Leaves of N. angustifolia. A — fragment of leaf blade, spec. BIN 1560; B — cuticle fragment with imprint of upper epidermis, inner view, spec. BIN 1538; C — cuticle fragment of lower epidermis with papillae (p), external view, spec. BIN 1538; D — abaxial side of leaf, a cuticle fragment with papillae, spec. BIN 1538; E — cuticle fragment with imprint of lower epidermis with stomata (s), inner view, spec. BIN 1564; G — abaxial side of leaf, cuticle fragment with stomata, spec. BIN 1536; 7 — stoma, inner view, spec. BIN 1560. Scale bar: A — 1 cm; B — 20 pm; C, D, E — 50 pm; F — 100 pm; G — 10 pm.
fields is square or rounded; in costal fields, on the sides, it is square and elongated in the central part. The outlines are sinuous-wavy, wide-wavy. The coefficient sinuosity of the tangential cell walls ranges from 1.1 to 2.4.
The leaves are hypostomatic (Fig. 4C-G). Per width of the stomatal field there are three-four stomata
(Fig. 4C). Their apertures are oriented at different angles within the field. Occasional stomata can be found in costal fields or above large veins. On the epidermal surface there are very few (59-100 stomata per 1 mm2), few (100-160 stomata per 1 mm2), or average number of sto-mata (160-250 stomata per 1 mm2) (Fig. 4C). The type
Fig. 4. Leaves of N. longifolia. A — fragment of leaf blade, spec. BIN 1562; B — cuticle fragment with imprint of upper epidermis, inner view, spec. BIN 1525; C — abaxial side of leaf, cuticle fragment with stomatal strips, spec. BIN 1528; D — abaxial side of leaf, a cuticle fragment with stomata (s) and trichome base (h), spec. BIN 1527; E — trichome base, external view, spec. BIN 1530; F — cuticle fragment with imprint of lower epidermis with stomata, inner view, spec. BIN 1509; G — stoma, inner view, spec. BIN 1520. Scale bar: A — 1 cm; B, E — 10 pm; C — 200 pm; D — 50 pm; F, G — 20 pm.
of stomata is paracytic (Fig. 4D, F, G). The stomatal index is low (6.8-11 %) or medium (11-15.1 %). The tangential cell walls of subsidiary cells are even or curved.
Rarely, in the stomatal fields, there are bases of tri-chomae (Fig. 4D, E): large and round. Their area is about 1000-1100 ^m2.
Ptilophyllum. P. caucasicum. The leaves are pinnate, narrowly or broadly lanceolate, dissected into segments (Fig. 5A). The area of the segments ranges from 13.6 to 140 mm2. Segments are alternate, widely spaced or closely pressed to each other. The shape varies from wide and short with a rounded top to narrow and long with a pointed apex.
Table 2. Statistical characteristics of the characters of the leaf structure for Nilssoniopteris and Ptilophyllum
Characters N. angustifolia N. longifolia
M a mx Xmin Xmax M a mx Xmin Xmax
Ncell_upp 795 251 281 538 1234 797 255 266 444 1263
Supp 1466 421 518 952 2177 1373 524 458 648 2231
Per_upp 244 76 86 156 388 228 75 76 141 338
Sin_upp 1.58 0.30 0.56 1.25 2.08 1.54 0.29 0.51 1.08 1.93
NcelMow 967 272 342 618 1453 759 214 253 516 1136
Slow 1326 210 469 1065 1632 1639 467 546 1030 2285
Per_low 215 40 76 156 271 270 94 90 156 450
Sin_low 1.47 0.21 0.52 1.16 1.72 1.65 0.38 0.55 1.10 2.45
Npap 524 133 185 303 744 0 0 0 0 0
Nstom 166 70 59 73 261 140 54 40 59 222
Lstom 37 8 9 32 39 31 10 10 30 48
Kstom 12.45 2.95 4.40 9.54 16.96 11.83 2.94 3.94 6.77 15.12
P. caucasicum P. okribensef. ratchense
Ncell_upp 926 268 178 574 1773 803 113 207 535 953
Supp 2023 1234 389 607 4668 2284 1510 590 558 4830
Per_upp 341 140 66 152 646 380 138 98 199 635
Sin_upp 1.94 0.34 0.37 1.19 2.61 2.09 0.29 0.54 1.62 2.69
NcelMow 1050 232 202 655 1629 938 168 242 561 1254
Slow 1870 771 360 301 3702 1837 1352 474 482 4392
Per_low 301 100 58 114 531 314 147 81 129 613
Sin_low 1.76 0.34 0.34 1.30 2.49 1.90 0.29 0.49 1.34 2.41
Npap 567 137 146 285 744 0 0 0 0 0
Nstom 111 38 21 56 179 156 60 40 59 255
Lstom 39 7 7 29 48 37 5 9 31 46
Kstom 8.86 3.05 1.70 4.35 13.98 12.11 2.79 3.13 7.07 16.88
Notes: M — mean; mx — standard error of the mean; o — standard deviation; Xmin — minimum; Xmax — maximum.
Characters: Npap — number of papillae per 1 mm2 of epidermis. Abbreviations of the characters of epidermal structure are the same as in the Table 1.
The upper epidermis (Fig. 5B, Table 2) consists of small-sized cells (570-900 cells per 1 mm2), medium-sized cells (900-1600 cells per 1 mm2), or large-sized cells (1600-1770 cells per 1 mm2). The projection of the area of the epidermal cells is square or rectangular, above the large veins it is rectangular. Outlines are sinuous-wavy, wide-wavy, sinuous. The coefficient of sinuosity of the tangential cell walls ranges from 1.2 to 2.6. Trichomae are absent.
The lower epidermis (Fig. 5C-F) consists of medium-sized cells (1350-1600 cells per 1 mm2), or large-sized cells (655-1350 cells per 1 mm2). The epidermis is divided into costal and stomatal fields (Fig. 5E). The projection of the
area of the epidermal cells in the stomatal fields is rounded or square; in costal fields it is square, rectangular. Outlines are sinuous-wavy, wide-wavy, sinuous. The coefficient of sinuosity of the tangential cell walls ranges from 1.3 to 2.5. Most of the epidermal cells in the lower epidermis, both in costal and stomatal fields, have one central convex-shaped papilla. Their imprints are clearly visible on the upper side of the cuticle. On the lower side of the cuticle, they have narrow recesses-cavities in the central parts of the epidermal cells (Fig. 5C, D, E). The number of papillae varies from 285 to 745 per 1 mm2. Trichomae are absent.
Leaves are hypostomatic. The stomata are arranged in two or three per width of stomatal field (Fig. 5E), their
Fig. 5. Leaves of P. caucasicum. A — a fragment of leaf blade with segments, spec. BIN 1600; B — a cuticle fragment with imprint of upper epidermis, inner view, spec. BIN 1568; C — a cuticle fragment with imprint of lower epidermis (s) and papillae (p), external view, spec. BIN 1584; D — a stoma, external view, spec. BIN 1584; E — an abaxial side of leaf, a cuticle fragment with stomatal strips, spec. BIN 1635; F — a stoma and epidermal cells with imprints of papillae (on), inner view, spec. BIN 1607.
Scale bar: A — 1 cm; B, D — 20 pm; C — 50 pm; E — 200 pm.
apertures are oriented across the veins. Quite often, sto-mata are found in costal fields, less often they can be found above large veins. On the epidermal surface, there are very few (56-100 stomata per 1 mm2), few (100-
160 stomata per 1 mm2), or average number of stomata (160-180 stomata per 1 mm2). The type of stomata is paracytic (Fig. 5F). The stomatal index is very low (4.36 %), low (6-11 %) or medium (11-14.0 %). The stomata
Fig. 6. Leaves of P. okribense. A — fragment of leaf blade with segments, spec. BIN 1573; B — cuticle fragment with imprint of upper epidermis, inner view, spec. BIN 1639; C — abaxial side of leaf, a cuticle fragment with stomatal strips; spec. BIN 1619; D — stoma, external view, spec. BIN 1573; E — abaxial side of leaf, a cuticle fragment with stomata (s), spec. BIN 1608; F — stoma, inner view, spec. BIN 1683. Scale bar: A — 1 cm; B — 20 pm; C, E — 100 pm; D, F — 50 pm.
are submerged; the subsidiary cells hang over the guard cells, forming an H-shaped cavity (Fig. 5D).
P. okribense f. ratchense. The leaves are linear, dissected into segments (Fig. 6A). The segments are opposite or slightly offset from each other, narrow, long with
a pointed apex. The area of the segments ranges from 8.4 to 270 mm2.
The upper epidermis (Fig. 6B, Table 2) consists of medium-sized cells (900-950 cells per 1 mm2), or large-sized cells (535-900 cells per 1 mm2). The projection of
the area of the epidermal cells is square or rectangular; outlines are sinuous-wavy, wide-wavy, sinuous. The coefficient of sinuosity of the tangential cell walls ranges from 1.6 to 2.7. Trichomae are absent.
The lower epidermis (Fig. 6C-F) consists of large-sized cells (600-1253 cells per 1 mm2), very large-sized cell (560-600 cells per 1 mm2). The epidermis is divided into costal and stomatal fields (Fig. 6C). The projection of the area of the epidermal cells is flattened, square or rectangular, above the large veins it is rectangular. Outlines are sinuous-wavy and sinuous in costal fields, wide-wavy and sinuous-wavy in stomatal fields. The coefficient of sinuosity of the tangential cell walls ranges from 1.3 to 2.4. Trichomae are absent.
Leaves are hypostomatic. The stomata are located in indirect two-row stripes, their apertures are oriented across the veins (Fig. 6C). Stomatal fields may merge with each other. Stomata are often located in costal fields, less often — above large veins. On the epidermal surface, there are very few (59-100 stomata per 1 mm2), few (100-160 stomata per 1 mm2), or an average number of stomata (160-250 stomata per 1 mm2). The type of sto-mata is paracytic (Fig. 6E, F). The stomatal index is low (7.1-11 %), medium (11-16 %) or large (16-16.9 %). The stomata are submerged; the subsidiary cells hang over the guard cells, forming an H-shaped cavity (Fig. 6D).
Table 3. Factor structure of epidermal characters in gnetums and Bennettitales
Characters Factor 1 Factor 2 Factor 3
Ncell_upp 785 -231 478
Supp -831 -256 151
Per_upp -889 -271 232
Sin_upp -733 -269 349
NceNJow 765 -317 527
Slow -840 -089 064
Per_low -896 -169 184
Sin_low -740 -244 280
Npap -373 -617 -270
Nstom 752 -565 138
Ndstom 643 -669 -167
Lstom -600 221 133
Kstom -277 -444 -784
FD, % 52.5 14.3 12.1
Factors 1-3 — factor loadings on Factor 1 and Factor 2. FD — factor dispersion, %. Zero and the dot in front of the values of the characters are omitted. Characters with |r| > 0.5 are marked in bold. Abbreviations of the characters of epidermal structure are the same as in Table 1.
Correlations of the characters in epidermis of
gnetums and Bennettitales. According to the results of PCA, three factors were revealed (Table 1). Their total factorial variance (XFD) is 78.9 %.
Factor 1 (|r| > 0.5; FD = 52.5 %) combined 11 main characters of epidermis: area of tangential cell walls of epidermal cells in upper and lower epidermis, their perimeter and sinuosity of the tangential cell walls, number of epidermal cells per 1 mm2 of upper and lower epidermis, number of differentiated and aborted stomata per 1 mm2 of epidermis, length of stomata. Character-indicator: perimeter of epidermal cells in lower epidermis (r = -0.896). The larger are epidermal cells, the larger are their area and perimeter, the lower is the sinuosity of the tangential cell walls, the lower is the number of stomata per 1 mm2 of epidermis, but stomata are larger and most of them are functional. The lower are the epidermal cells, the lower are their area and perimeter, the higher the sinuosity of their tangential cell walls, the more stomata are per 1 mm2, and the lower their size.
Factor 2 (|r| > 0.5; FD = 14.3 %) included three characters of the epidermis: number of papillae per 1 mm2, total number of stomata per 1 mm2 and number of differentiated stomata per 1 mm2. Character-indicator: number of differentiated stomata per 1 mm2 of epidermis (r = -0.669). This factor shows differences in epidermal structure among Bennettitales, which have central papillae on the surface of the epidermal cells (N. angus-
tifolia and P. caucasicum), Bennettitales without papillae (N. longifolia and P. okribense f. ratchense) and gnetums. For Bennettitales with papillae: the more stomata are per 1 mm2 of epidermis, the more papillae are in the epidermis. The fewer the number of stomata per 1 mm2 of epidermis, the fewer number of papillae are in the epidermis. For gnetums: the higher the total number of stomata per 1 mm2 of epidermis, the more of them are differentiated. The lower the total number of stomata per 1 mm2 of epidermis, the fewer of them are differentiated.
Factor 3 (|r| > 0.5; FD = 12.1 %) includes two characters: stomatal index and number of epidermal cells per 1 mm2 of lower epidermis. Character-indicator: stomatal index (r = -0.784). This factor describes the ratio of the specific proportion of stomata and the size of epidermal cells. The larger are epidermal cells, the lower is the stomatal index. The smaller are epidermal cells, the higher is the stomatal index.
In the diagram of factor space (Fig. 7), the specimens of the investigated leaves diverge along the factor 1. Gne-tums form a fairly dense cluster, in which individual species stand out well. On the other hand, Bennettitales form a large cluster that occupies a significant part of the factor space. It is rather problematic to single out individual species or genera within this cluster. Nevertheless, the second factor more or less distinguishes two clusters, including species with central papillae on the epidermal cells (N. an-
•
♦ L
* • a O O a>°
• o o □ oo o
• • • • * ♦ ♦
• 4
•
-3
-1 0 Factor 1
o Gnetum gnemon
□ Gnetum gracilipes o Gnetum latifolium
A Gnetum montanum
♦ Nilssoniopteris angustifolia
♦ Nilssoniopteris iongifoiia
• Ptilopylium caucasicum
• Ptilophylium okribense f. ratchense
Fig. 7. Distribution of leaf samples of gnetums and Bennettitales in the scatterplot of Factor 1 and Factor 2.
Gnetum species are marked with white outline icons; Bennettitales with papillae in epidermal cells are marked by gray icons; Bennettitales species without papillae are marked by black icons.
gustifolia and P. caucasicum) and species without papillae (N. longifolia and P. okribense f. ratchense).
As already shown in previous articles, the epidermis of gnetums is rather similar. Differences between species were revealed in the ratio of the structural elements of epidermis, as well as in the number of aborted stomata and their ratio with the number of differentiated stomata (Pau-tov, Pagoda, and Krylova, 2012; Pautov and Pagoda, 2015). Aborted stomata are found in the epidermis of many species of Gnetum. Using G.gnemon as model species, it has been shown that their origin occurs during the second phase of leaf development during the period of mass differentiation of paracytic stomata (Pautov and Pagoda, 2013, 2014). The ratio of differentiated and aborted stomata is reflected in the correlation structure of a leaf (Pautov, Pagoda, and Krylova, 2012). Presumably, the presence of aborted stomata may be a kind of mutation developed in the course of evolution for effective regulation of transpiration.
Epidermis of Bennettitales is very diverse even within the same species. Nevertheless, despite strong differences in the epidermal structure, a number of
similarities were revealed in the correlation structure of epidermis of gnetums and Bennettitales. Firstly, the similarity in the ratio of the elements that form the epidermis (Factor 1, Table 3): number of epidermal cells and number of stomata per 1 mm2 of epidermis. The revealed correlations in this case are the basic characteristics of the epidermis and are also inherent in flowering plants (Pautov, 2009). Secondly, the similarity of correlations between the size of the leaf (its large-sized and small-sized cells) and the degree of sinuosity of tangential cell walls of the epidermal cells (Factor 1, Table 3). The shape of the epidermal cells plays a significant role in the leaf morphogenesis of seed plants (Pautov, 2009; Pautov and Vasilieva, 2010). This character can, in particular, be involved in maintaining the flat shape of the leaf blade, and affect the density of stomata. Thirdly, in gnetums and Bennettitales, a negative correlation was found between the size of stomata and the size of epidermal cells (Factor 1, Table 3). This can be explained as an ontogenetic character of leaf development (the more cells have divided, the more small-sized they are, and
epidermis contains many small stomata; the fewer cells have divided, the more large-sized they are, and epidermis contains a low number of large stomata), as well as an increase in the efficiency of transpiration regulation.
Character variability in the epidermis of gnetums and Bennettitales. The coefficient of variation of the investigated characters in epidermis of Gnetum ranges from 6.4 to 24.0 %. Most of them have medium and high levels of common variability. A very low level (Cv < 7.5 %) is for the length of stomata. Low level (8-12 %) is for sinuosity of epidermal cells in upper and lower epidermis. Medium level (13-20 %) is for stomatal index, the perimeter of epidermal cells in upper and lower epidermis and the number of epidermal cells per 1 mm2 of lower epidermis. Increased level (21-30 %) is for the area of tangential cell walls of epidermal cells in upper and lower epidermis, the number of epidermal cells per 1 mm2 of upper epidermis and the number of stomata per 1 mm2 of epidermis.
The coefficient of variation of the investigated characters in the epidermis of Nilssoniopteris ranges from 18.0 to 39.9 %. Most of them demonstrate increased and high levels of common variability. The medium level of variability is for sinuosity of epidermal cells in upper and
lower epidermis. Increased level is for stomatal index, the perimeter of epidermal cells in lower epidermis, the number of epidermal cells per 1 mm2 of lower epidermis. High level (31-40 %) is for the length of stomata, the area of tangential cell walls of epidermal cells in upper epidermis, perimeter of epidermal cells in upper epidermis, the number of epidermal cells per 1 mm2 of upper epidermis, the number of stomata per 1 mm2 of epidermis.
The coefficient of variation of the investigated characters in the epidermis of Ptilophyllum ranges from 15.7 to 63.5 %. Most of them have increased and high levels of common variability. The average level of variability is for sinuosity of epidermal cells in upper and lower epidermis, the length of stomata, the number of epidermal cells per 1 mm2 of lower epidermis Increased level is for the number of epidermal cells per 1 mm2 of upper epidermis, stomatal index. High level is for the number of stomata per 1 mm2 of epidermis, the perimeter of epidermal cells in upper and lower epidermis. Very high level (Cv > 40 %) is for the area of tangential cell walls of epidermal cells in the upper and lower epidermis.
In the diagram in Fig. 8, the values of the coefficients of variation are shown. The characters of epider-
Fig. 8. The common variability of the leaf structure characters in gnetums and Bennettitales.
The vertical axis indicates the coefficient of variation (Cv, %). Abbreviations of the characters of epidermal structure are the same as in Table 1.
mal structure in Bennettitales are generally higher than those of gnetums. The common variability of characters in Nilssoniopteris and Ptilophyllum remains at increased and high levels, while in gnetums they have medium and increased levels. Nevertheless, in general, the values of the coefficients of variation in characters of Bennettitales and gnetums are comparable and, despite a clear scatter in values, they demonstrate a number of similarities.
Firstly, in both gnetums and Bennettitales, the sinuosity of the tangential cell walls of the epidermal cells, compared with the variability in other characters, has the lowest level of variability (Cv < 18.0 %), i. e., this is a fairly stable character. This is consistent with taxonomic concepts. The sinuosity of the tangential cell walls in epidermal cells is successfully used as a diagnostic character for the determination of Bennettitales species (Do-ludenko and Svanidze, 1969; Kiritchkova and Nosova, 2012; Rudall and Bateman, 2019). Secondly, the length of stomata is also a stable character. It is relatively invariable in species of Ptilophyllum (Cv = 16.8 %) and slightly variable in gnetums (Cv = 6.4 %). The size of the stomata also has a taxonomic significance for Ptilophyllum (Do-ludenko and Svanidze, 1969). Thirdly, there is considerable variation in the values of the coefficients of variations for the number of epidermal cells per 1 mm2 of upper and lower epidermis (17.5 % < Cv < 31.9 %), the area of tangential cell walls of epidermal cells in upper and lower epidermis (21.2 % < Cv < 63.5 %), the number of stomata per 1 mm2 of epidermis (29.3 % < Cv < 39.9 %). These characters determine the size of epidermal cells and, consequently, the leaf size, which is consistent with the results of PCA (Factor 1, Table 3). Nevertheless, despite significant difference in the level of common variability of the structural elements of epidermis in Bennettitales, the ratio in the number of stomata and the number of epidermal cells — the stomatal index (13.0 % < Cv < 28.8 %) — remain more or less stable not only within the species, but within the genera as well.
Conclusion
The investigated structural organization of the epidermis of the leaves of Bennettitales is comparable and similar in many aspects to the leaves of extant gnetums. Given that both groups are considered as possible ancestors or closer relatives of flowering plants (Crane, 1996; Friis et al., 2007; Sokoloff and Timonin, 2007; Gambaryan and Kuznetsov, 2021), it will be further possible to compare the obtained results with those on primitive flowering plants. Perhaps it could be possible to reveal some general directions of leaf evolution based on the transformation of quantitative characters.
Leaves of Bennettitales and especially their stomatal characters have become central to their description, identification and systematization. Although these characters
are also open to considerable interpretation and debate. Rudall and Bateman (Rudall and Bateman, 2019) identify close similarities between paracytic stomata of Bennettitales and relict species Gnetum. They indicate that the pair of lateral subsidiary cells of Bennettitales, as in Gnetum, are both mesogene cells. We found significant similarities in the variability and correlation structure of the epidermal structure of the extinct Bennettitales Nilssoniopteris and Ptilophyllum and extant gnetums.
An important question is whether the similarities in leaf morphology between Bennettitales, Gnetum and flowering are related to their shared ancestry or indicate shared ecological adaptations.
Rudall and Bateman (Rudall and Bateman, 2019) concluded that stomata of Gnetum represent by far the most likely analogue for the stomatal development in Bennettitales. The similarities in correlational structure of Bennettitales and gnetums could also represent an ecophysiological marker for adaptation or developmental features because many correlations in structural organization are basic for seed plants, especially the correlation between the size of the leaf and the degree of sinuosity of tangential cell walls of the epidermal cells, which indicate the shape of the epidermal cells. Correlation between the size of stomata and size of epidermal cells could be explained both as ontogenetic character of leaf development and as regulation of transpiration.
The present study provides the first successful attempt to evaluate the structural organization of the epidermis in leaves of Bennettitales and compare it with the leaves of extant plants using a principal component analysis. Together, these features could infer the development to this diverse and relatively derived lineage that co-existed with the earliest recognizable angio-sperms, and help to use these characters in phylogeny reconstructions. However, the data obtained only on the cuticle of fossil plants are not enough to get a complete picture.
Acknowledgments
We express our gratitude to Marina A. Romanova of the Department of Botany, Saint Petersburg State University for editing the article and Anna N. Melnikova of the Department of Botany, Saint Petersburg State University for consultations on methodogy and light microscopy tools; Dmitry V. Gromyko of the Laboratory of Paleobotany for consultations on paleo-botanical methods, Svetlana S.Popova for collecting plant material from China and Natalia V. Nosova for providing the opportunity to work with a collection of fossil plants.
References
Carmichael, J. S. and Friedman, W. E. 1996. Double fertilization in Gnetum gnemon (Gnetaceae): its bearing on the evolution of sexual reproduction within the Gnetales and the anthophyte clade. American Journal of Botany 83(6):767-780. https://doi.org/10.1002/j.1537-2197.1996.tb12766.x
Crane, P. R. 1996. The fossil history of the Gnetales. International Journal of Plant Sciences 157(6):S50-S57. https:// doi.org/10.1086/297403 Doludenko, M.P. and Svanidze, Ts.I. 1969. Late Jurassic flora of Georgia; 116 pp. in Trudy geologicheskogo instituta. Vyp. 178. (In Russian) Donoghue, M.J. and Doyle, J.A. 2000. Seed plant phylog-eny: demise of the anthophyte hypothesis? Current Biology 10:106-109. https://doi.org/10.1016/S0960-9822(00)00304-3 Florin, R. 1931. Untersuchungen zur Stammesgeschichte der Coniferales und Cordaitales. Kungl. Svenska Vetenskap-sakademiens handlingar 10(1):1-588. Florin, R. 1933. Studien über die Cycadales des Mesozoikums... Erorterungen über die Spaltoffnungsapparate der Bennettitales. Stockholm: Almqvist och Wiksells bok-tryck 12(5):1-134. Friedman, W.E. 1998. The evolution of double fertilization and endosperm: an "historical" perspective. Sex Plant Reproduction 11:6-16. https://doi.org/10.1007/ s004970050114 Friis, E.M., Crane, P.R., Pedersen, K. R., Bengtson, S., Donoghue, P.S.G., Grimm, G.W., and Stampanoni, M. 2007. Phase-contrast X-ray microphotography links Cretaceous seeds with Gnetales and Bennettitales. Nature 450(22):549-552. https://doi.org/10.1038/nature06278 Gambaryan, P.P. and Kuznetsov, A.N. 2021. Gnetum and Nymphaeaceans as models for a scenario of the origin of morphotype of flowering plants. Biology Bulletin Reviews 1 1(3):237-253. https://doi.org/10.1134/ S2079086421030038 (In Russian) Haycraft, C.J. and Carmichael, J. S. 2001. Development of sterile ovules on bisexual cones of Gnetum gnemon (Gneta-ceae). American Journal of Botany 88:1326-1330. https:// doi.org/10.2307/3558344 Kendall, M.G. and Stewart, A. 1976. Multivariate statistical analysis and time series. 736 pp. Nauka Publ. Moscow. (In Russian)
Kerp, H. 1990. The study of fossil gymnosperms by means of cuticular analysis. Palaios 5:548-569. https://doi. org/10.2307/3514861 Kiritchkova, A. I. and Nosova, N.V. B. 2012. Jurassic continental deposits of the Middle-Caspian Basin. Part 2: Facies, taphonomy, interregional correlations, flora (Pinophyta: Pteridospermae, Cycadales, Bennettitales, Ginkgoales, Czekanowskiales, Coniferales). 338 pp. VNIGRI Publ. St. Petersburg. (In Russian) Krassilov, V.A. and Bugdaeva, E.V. 1988. Gnetalean plants from the Jurassic of Ust-Balej, East Siberia. Review of Palaeobotany and Palynology 53:359-374. https://doi. org/10.1016/0034-6667(88)90040-1 Mamaev, S.A. 1973. Forms of intraspeciflc variability of woody plants. 284 pp. Nauka Publ. Moscow. (In Russian) Nautiyal, D.D., Singh, S., and Pant, D.D. 1976. Epidermal structure and ontogeny of stomata in Gnetum gnemon, G. montanum and G. ula. Phytomorphology 26(3):282-296. Pant, D. D. 1965. On the ontogeny of stomata and other homologous structures. Plant Science Series 1:1-24. Pautov, A.A. 2009. Phylomorphogenesis patterns of the plant vegetative organs. 219 pp. St. Petersburg University Press. St. Petersburg. (In Russian)
Pautov, A. A., Pagoda, I. O., and Krylova, Ye. G. 2012. The leaf structure of Gnetum gnemon (Gnetaceae). Botanicheskii zhurnal 97(12):1497-1505. https://doi.org/10.1134/ S1234567812120038 (In Russian) Pautov, A.A. and Pagoda, I.O. 2013. The distribution of aborted stomata in the leaf blade epidermis of Gnetum gnemon (Gnetaceae). Vestnik Leningradskogo universiteta. Ser. 3. Biologia 3(4):26-33. (In Russian) Pautov, A.A. and Pagoda, I.O. 2014. Growth and development of leaf in Gnetum gnemon (Gnetaceae). Botanicheskii zhurnal 9:1010-1015. https://doi.org/10.1134/ S1234567814090055 (In Russian) Pautov, A.A. and Pagoda, I.O. 2015. Structural diversity of the leaf epidermis in the genus Gnetum (Gnetaceae). Botanicheskii zhurnal 100(2):171-177. https://doi. org/10.1134/S0006813615020064 (In Russian) Pautov, A.A. and Vasilieva, V.A. 2010. Role of the epidermis basic cells shape in leaf morphogenesis of Hamameli-daceae representatives. Botanicheskii zhurnal 95(3):338-347. (In Russian) Rudall, P.J. and Bateman, R.M. 2019. Leaf surface development and the plant fossil record: stomatal patterning in Bennettitales. Biological Reviews 94(3):1179-1194. https://doi.org/10.1111/brv.12497 Rydin, C., Friis, E. M. 2010. A new Early Cretaceous relative of Gnetales: Siphonospermum simplex gen. et sp. nov. from the Yixian Formation of Northeast China. BMC Evolutionary Biology 10(1):183. https://doi.org/10.1186/1471-2148-10-183
Shi, G., Herrera, F., Herendeen, P. S., Leslie, A. B., Ichinnorov, N., Takahashi, M., and Crane, P.R. 2017. Leaves of Podoza-mites and Pseudotorellia from the Early Cretaceous of Mongolia: stomatal patternsand implications for relationships. Journal of Systematic Palaeontology 16(2):111-137. https://doi.org/10.1080/14772019.2016.1274343 Sokoloff, D.D. and Timonin, A.C. 2007. Morphological and molecular data on the origin of angiosperms: On a way to a synthesis. Journal of General Biology 68:83-97. Thomas, H.H. and Bancroft, N. 1913. On the cuticles of some recent and fossil Cycadean fronds. Transactions of the Linnean Society 8:155-204. https://doi. org/10.1111/j.1095-8339.1913.tb00284.x Vasiliev, B. R. 1988. Leaf structure of woody plants of various climatic zones. 208 pp. Leningrad University Press. Leningrad. (In Russian) Wang, Z.-Q. 2004. A new Permian gnetalean cone as fossil evidence for supporting current molecular phylogeny. Annals of Botany 94:281-288. https://doi.org/10.1093/aob/ mch138
Won, H. and Renner, S.S. 2006. Dating dispersal and radiation in the gymnosperm Gnetum (Gnetales) — clock calibration when outgroup relationships are uncertain. Systematic Biology 55(4):610-622. https://doi. org/10.1080/10635150600812619 Yang, Y., Xie, L., and Ferguson, D.K. 2017. Protognetaceae: a new gnetoid macrofossil family from the Jurassic of northeastern China. Perspectives in Plant Ecology, Evolution and Systematics 28:67-77. https://doi.org/10.1016/j. ppees.2017.08.001 Zakharevich, S. F. 1954. About the method of description an epidermis of the leaf. Vestnik Leningradskogo universiteta. Ser. 3. Biologia 4:65-75. (In Russian)
