Cuticle ultrastructure of Pseudofrenelopsis gansuensis: Further taxonomical implications for Cheirolepidiaceae

Cretaceous Research 71 (2017) 24e39

Contents lists available at ScienceDirect

Cretaceous Research journal homepage: www.elsevier.com/locate/CretRes

Cuticle ultrastructure of Pseudofrenelopsis gansuensis: Further taxonomical implications for Cheirolepidiaceae €tan Guignard a, b, Xiao-Ju Yang c, *, Yong-Dong Wang c Gae Universit e Lyon 1, F-69622, Lyon, France CNRS, UMR 5023 LEHNA, 7-9 rue rapha€ el Dubois, 69622, Villeurbanne cedex, France c Nanjing Institute of Geology and Palaeontology, Chinese Academy of Sciences, 39 East Beijing Road, Nanjing, 210008, China a

b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 15 August 2016 Received in revised form 10 November 2016 Accepted in revised form 14 November 2016 Available online 19 November 2016

A study of the cheirolepidiaceous conifer Pseudofrenelopsis gansuensis from the Lower Cretaceous of Wangqing Jilin Province in China was conducted in detail using scanning and transmission electron microscopy techniques. In total, nine ultrastructural features were recognized for the cuticle of this fossil plant, which are helpful in the distinguishing between cuticles of ordinary epidermal cells, subsidiary cells, guard cells and hypodermal cells of the stomatal apparatus. A three dimensional reconstruction of the cuticle ultrastructure was obtained. Pseudofrenelopsis gansuensis is the second species of this genus for which the cuticle ultrastructure has been statistically examined with 30 measurements and estimated confidence interval values. The close comparison of the cuticle ultrastructure characters, including statistical data, among Cheirolepidiaceae and other fossil conifers provides potential evidence of the taxonomic significance of this genus: ten characters are potentially valuable for specific separation, eleven parameters for generic separation and three parameters seem also to be useful for Family determination. The differences in the chemical composition according to preliminary statistical element analyses of the cuticles based on three ratios in two species of Pseudofrenelopsis, P. dalatzensis and P. gansuensis, should also be examined in future studies. © 2016 Elsevier Ltd. All rights reserved.

Keywords: Pseudofrenelopsis Cheirolepidiceae Cuticle ultrastructure Taxonomy Palaeoecology EDS analyses

1. Introduction The application of transmission electron microscopy (TEM) in the study of Cheirolepidiaceae (Archangelsky and Taylor, 1986; Archangelsky and Del Fueyo, 1989; Guignard et al., 1998; Villar de Seoane, 1998; Axsmith et al., 2004; Del Fueyo et al., 2008; Yang et al., 2009; Mairot et al., 2014) has achieved fruitful results since the eve of the 21st century. Recently, generation of detailed statistical data and a reconstruction of the cuticle ultrastructure of cheirolepidiaceous conifers have been provided (Yang et al., 2009; Mairot et al., 2014). Pseudofrenelopsis dalatzensis and Suturovagina intermedia are a few of the taxa that have been studied in detail. There are significant differences between Suturovagina and Pseudofrenelopsis in four cuticle ultrastructural characters, and the proportions of different layers also markedly differed (Mairot et al., 2014). The proportions of the cuticle proper and cuticular layer in

* Corresponding author. Fax: þ86 25 83357026. E-mail addresses: [email protected] (G. Guignard), [email protected] (X.-J. Yang). http://dx.doi.org/10.1016/j.cretres.2016.11.006 0195-6671/© 2016 Elsevier Ltd. All rights reserved.

the epidermal cell cuticular membranes and wavy and polylamellate and granular layers in the cuticle proper differed among cheirolepidiaceous conifers, but few statistical data are available. The present study describes the cuticle ultrastructure of another cheirolepidiaceous conifer, Pseudofrenelopsis gansuensis, to provide further anatomical details and elucidate the potential taxonomic and palaeoecological significance of the ultrastructural features observed using transmission electron microscopy. In addition, trial element analyses were also applied to examine the cuticles of two species of Pseudofrenelopsis for similar purposes. 2. Material and methods Samples were collected from Luozigou, Wangqing County, Jilin Province, northeastern China (43
530 4800 N, 130
030 2600 E, Fig. 1). The plant-bearing strata in Luozigou Basin belong to the Lower Cretaceous Dalazi Formation (Oishi, 1941; Zhou et al., 1980; Bureau of Geology and Mineral Resources of Jilin Province, 1997). The type section of the Dalazi Formation is well developed in Dalazi village, Zhixin Town, Longjing County in Yanji Basin, approximately 150 km to the southwest of Luozigou Basin in eastern Jilin Province, China

G. Guignard et al. / Cretaceous Research 71 (2017) 24e39

Fig. 1. Sketch map showing the fossil locality in Wangqing, Jilin Province, China (indicated by ).

(Li et al., 2016). Until recently, several cheirolepidiaceous conifers have been reported in Yanji Basin (Chow and Tsao, 1977; Zhang et al., 1980; Zhang, 1986; Zhou, 1995), but only one conifer species has been reported in Luozigou (Yang and Deng, 2007). After several field trips to Luozigou, Yang XJ collected abundant shoots of cheirolepidiaceous conifers belonging to Pseudofrenelopsis, and some of these conifers were branched. Further examination showed that the cuticle is not similar to Pseudofrenelopsis dalatzensis (Chow et Tsao) Cao ex Zhou (Zhou, 1995), but resembled Pseudofrenelopsis gansuensis based on a fragmentary leafy shoot from the Cretaceous in Jiuquan Basin, Gansu Province, Northwest China (Deng et al., 2005). Interestingly, Pseudofrenelopsis dalatzensis occurs rather frequently in the Dalazi Formation of Yanji Basin, while P. gansuensis is the most common plant from Luozigou Basin. Although the two cherolepidiaceaeous conifers from the two localities are generally similar in gross morphology, the cuticles of these plants are markedly different (Yang and Deng, 2007). In the present study, the cuticle ultrastructure of P. gansuensis was examined using SEM and TEM. Pieces of cuticles were removed from the specimens, treated with hydrofluoric acid (HF) for 18 h, and subsequently macerated in Schulze's solution (nitric acid and potassium chlorate). The time for oxidation depended upon the degree of coalification of the compressed specimen, typically approximately 5e8 h. For the next step (see also Kerp, 1990), after the cuticles became yellow and translucent, the samples were rinsed with water and treated with dilute ammonia (5%) for a few seconds to half a minute, followed by thorough rinsing with water. The samples for scanning electron microscopy (SEM) were mounted on stubs using double-sided adhesive tape, coated with gold, observed and photographed using a Hitachi S-4300 SEM at the SEM Laboratory of the Swedish Museum of Natural History, Stockholm, Sweden. The samples for transmission electron microscopy (TEM) were prepared according to Lugardon (1971), a technique that is also used for fossil pollen and spores and living plant cuticles (conifers and angiosperms; Bartiromo et al., 2012, 2013). A total of 12 cuticle samples were obtained from the specimens (2 Epon resin blocks were obtained from slightly treated material PB21041 and 10 blocks were obtained from treated material PB21045). Among these blocks, 220 ultrathin 60e70 nm sections (70 sections from specimen PB21041 and 150 sections from specimen PB21045) were obtained and collected on uncoated 300 Mesh copper grids (200

25

transversal sections, i.e., perpendicular to the leaf length; 20 longitudinal sections, i.e., parallel to the leaf length). Ultrathin sections were selected, observed and photographed using a Philips CM 120 at 80 kV at the Centre de Technologie des Microstructures (CTm) of Lyon-1 University, France; and a few sections were examined using a Hitachi H-7650 at 80 kV at the Life Sciences Laboratory Center of Nanjing Agricultural University in China. To obtain information on the approximate stomatal and cuticle structures, 100 transversal and 10 longitudinal 1-mm sections were mounted on 22 glass slides and subsequently examined. These sections were stained according to Richardson et al. (1960). Consistent with Richardson et al. (1960), the samples were stained with methylene blue, Azur II solution (½e½), prior to obtaining micrographs under a Zeiss axioscope 2 light microscope using Zen 1.0.1.0 software. EDS analysis was performed on TEM images using SIRIUS SD ENSOTECH and IDFIX software, with an acceleration voltage of 120 kV, 1e3 spot sizes, and 60e120 s processing time at a constant time of 4 mseconds. For each of the two Pseudofrenelopsis species compared in the present study, 10 copper 300 Mesh uncoated grids were used, devoid of uranyl acetate and lead citrate staining. Among the available elements, Cu and Al were eliminated from the results as components of the grid, Os was eliminated as a component of the embedding technique, Si was eliminated as a component of the oils used in the TEM, and C and O were also eliminated as major components of the EPON embedding resin. Element analyses of the matrix containing fossil plants were undertaken using an energy dispersive X-ray system attached to LEO-1530VP in Nanjing. Specimens (PB21041e21045), SEM stubs and negatives were housed at Nanjing Institute of Geology and Palaeontology, Chinese Academy of Sciences in China. TEM material and negatives were stored at Lyon-1 University, Villeurbanne, France. The following terms and abbreviations are used in the text, tables, figures and appendices: OEC ¼ ordinary epidermal cell cuticle; SC and GC ¼ subsidiary and guard cell, respectively, of the cuticle of the stomatal apparatus; CM ¼ cuticular membrane (CP þ CL); CP ¼ cuticle proper (A ¼ A1 þ A2); A1 ¼ outer polylamellate layer of the cuticle proper; A2 ¼ inner mainly granular layer; CL ¼ cuticular layer (B); B1 ¼ outer fibrilous layer; B2 ¼ inner most granular layer; OL ¼ opaque lamellae of the polylamellate layer (A1); and TL ¼ translucent lamellae of the polylamellate layer (A1). 3. Results 3.1. Gross morphology and cuticle structure There are numerous vegetative shoots in the collection. Most of these structures are unbranched shoots, up to 110 mm long and 3e8 mm wide (Fig. 2). The internodes are typically 7e14 mm long. The leaves are adpressed on the axis, loosely arranged in a simple spiral. The free area is triangular, approximately 0.5e1 mm long with an acute apex and decurrent base that encircles the entire internode. The outer surface bares distinct fine, parallel, longitudinal ridges and grooves that converge towards the leaf apex. The abaxial cuticle of the leaf is approximately 8e28 mm thick, with well-defined longitudinal stomatal and non-stomatal files (approximately 7e9 in stomatal files per mm) that are clearly visible on both outer and inner surfaces (Fig. 3, A, B). The outer surface is normally smooth without hairs or papillae (Fig. 3, A). The epidermal cells have special periclinal walls that irregularly split to form fissures along the anticlinal wall on the outer surface, except at the cell corner (Fig. 3, A). In the stomatal files, epidermal cells are more or less square in outline or isodiametrically polygonal, approximately 12e30 mm  14e25 mm in size; in the non-stomatal

26

G. Guignard et al. / Cretaceous Research 71 (2017) 24e39

Fig. 2. Pseudofrenelopsis gansuensis Deng, Yang et Lu. Gross morphology. Specimen PB21041 (A), PB21043 (B) and PB21045 (C).

files, epidermal cells are arranged in one or two rows, square or somewhat rectangular, approximately 12e35 mm  12e18 mm in size (Fig. 3, BeD). The stomata are typically arranged in uniseriate rows or occasionally in short biseriate rows (Fig. 3, C, D, F), 11e12 in number per mm in file. In stomatal files, the stomata are deeply sunken, separated by 1e4 ordinary epidermal cells or occasionally contiguous, but never share subsidiary cells (Fig. 3, C, G). Encircling cells and Florin rings are absent. The stomata are elliptical, approximately 60e80 mm  60e90 mm in diameter, actinocytic and longitudinally or obliquely oriented (Fig. 3, D, E, G). Each stoma bears 5e8 subsidiary cells (typically 6e7) with no papillae. The guard cells are only partly cutinized. Cutinized hypodermis was well developed in some shoots. The hypodermal cells are elongate or rectangular, covering the non-stomatal files in the inner surface of the cuticle (Fig. 3, F, H). 3.2. Cuticle ultrastructure Four types of abaxial cuticles from the internode, that is, those of the ordinary epidermal cells, subsidiary cells, guard cells and hypodermal cells, were observed in detail (Figs. 4e6). Two types (ordinary epidermal and subsidiary cells) comprised the cuticle proper A (CP ¼ A1 þ A2) and cuticular layer B (CL ¼ B1 þ B2), while

the cuticle of the guard cell comprises the cuticle proper (A) and that of the hypodermal cell contains only the B layer. Polylamellate layer A1 is primarily lamellate and rarely non-lamellate, typically wavy and not straight. The data described below are the mean values calculated based on 30 measurements and the percentage of each component of the cuticle is also shown (Table 1, Appendix A). The ordinary epidermal cell cuticle (Figs. 4, CeD and 5) comprises 2e8 (mean on 30 measurements: 4.47, standard deviation 2.05) opaque (4.63 nm) and translucent (6.60 nm) lamellae (Fig. 5, DeL), approximately 1% (0.097 mm) of the total cuticle thickness. The granular layer A2 is 1.01 mm in thickness, approximately 10.6% of the total cuticle. The cuticular layer B comprises a fibrilous layer B1 and a granular layer B2 (Fig. 5, A), and approximately 88.4% of the total cuticle; layer B1 is 7.56 mm (79.7%) and layer B2 is 0.83 mm (8.7%) (Fig. 5, MeU). The subsidiary cell cuticle (Fig. 6, AeT) is 9.60 mm in thickness. The polylamellate layer A1 is 0.141 mm thick and approximately 1.5% of the total cuticle, while the granular layer A2 is 0.96 mm in thickness and 10% of the total cuticle. Layer A1 comprises 2e7 opaque (5.87 nm thick) and translucent (6.83 nm thick) lamellae. Layer A1 is typically wavy, particularly at the boundary between subsidiary and guard cell cuticles (Fig. 6, EeG), scarcely straight,

G. Guignard et al. / Cretaceous Research 71 (2017) 24e39

27

Fig. 3. Pseudofrenelopsis gansuensis. Scanning electron micrographs of internode abaxial cuticle. Specimens: PB21043 (A), PB21045 (BeH), slightly treated. A. Outer surface, note fissures usually present on the periclinal walls along anticlinal walls. B. Inner surface with distinct longitudinal arrangement of stomatal- and non-stomatal files. C. Inner surface showing stomata arranged closely in stomatal files. D. Inner surface view showing stomata with deep pits and epidermal cells with thick anticlinal walls. E. Inner surface view showing a stoma with 7 subsidiary cells. Guard cell cuticles are partly preserved in the middle of the stomatal apparatus. F. Inner surface view showing well developed hypodermal cuticles (arrows) in rows, which nearly consist of a whole cuticle layer. G. Inner surface view showing hypodermal cells (arrows) developed well along the non-stomatal files. H. Inner surface view showing hypodermal cells (arrows) developed well along the stomatal- and non-stomatal files.

present in all parts of the cuticle (Fig. 6, I, N, R), and also in the outer chamber parts (Fig. 6, JeK, PeQ). The guard cell cuticle (Fig. 6, UeX) is the thinnest cuticle (1.88 mm) among the four type, and comprised of layer A and was devoid of layer B. The polylamellate layer A1 is 0.093 mm thick and approximately 4.8% of the total cuticle. This layer is also wavy in some parts (Fig. 6, V), but more typically straight than epidermal and subsidiary cell cuticles (Fig. 6, WeX). The lamellae are 1e5 nm

thick (mean of 30 measurements: 2.53 nm, standard deviation 1.33). The thickness of the opaque lamellae is 6.81 nm, while that of the translucent lamellae is 8.79 nm. The granular layer A2 is 1.79 mm thick and approximately 95.2% of the total cuticle. Only a portion of the hypodermal cell cuticle (Fig. 5, SeU) was examined, comprising the granular layer, approximately 3.12 mm in thickness. Although the cuticle is similar in ultrastructure to the external A2 granular layer, its location far from the outer portion of

28

G. Guignard et al. / Cretaceous Research 71 (2017) 24e39

Fig. 4. Light micrographs of glass slide 1 mm sections of the cuticle. OP ¼ outer part, IP ¼ inner part, EC ¼ epidermal cell cuticle, CR ¼ cell remnants, CL ¼ cell location, HC ¼ hypodermal cell cuticle, SC ¼ subsidiary cell cuticle, GC ¼ guard cell cuticle, OC ¼ outer chamber of the stomatal apparatus, IC ¼ inner chamber of the stomatal apparatus. A. Scheme of the two types of material observed for these sections. B. General view of the two sides of a flattened axis with cuticle (arrows) in the outermost part and cell remnants in the middle part. Slightly treated. C. Cuticle of epidermal cell, very accidented in its outer part in this case. Slightly treated. D. Rare area with remnants of hypodermal cuticle and epidermis cell locations. Treated. EeK. Various sections showing different aspects of stomatal apparatuses, more or less filled with granular extracuticular material due to the slight treatment. Note the subsidiary cell cuticle sometimes very erected over the epidermal cell cuticle zone. L. A very clear section with two subsidiary cell cuticles and the two guard cell cuticles delimiting the boarder between the outer and inner chambers. Treated.

the cuticle belongs to the B2 granular layer. This cuticle ultrastructure is rarely observed in fossil plants, as the hypodermis is typically absent and/or easily damaged. Because few data have been obtained thus far, the thickness measured of hypodermis cuticle is not used in the statistics and comparisons, although these observations are precious. 3.3. EDS Although a degree of error was introduced when the cuticle was analysed using the EDS system (Fig. 7), reflecting resin embedding, the elemental measurement results were added with the eventual portion of the resin within the cuticle. To provide only comparable

values of the cuticles, among the potential ratios, only three elements (N/S, N/Cl and N/K) with homogeneous values for the resin (with insignificant differences from the ManneWhitney test for five measurements, Appendices E.1 and F; also indicated as “no” in Appendix G.3) were selected here for the ordinary epidermal cell cuticle, although they showed similar errors to the elements in the resin (Appendices E, F, G.2 and 3). Moreover, in the two resins of each taxon, the majority of these elements have a much lower presence than in the cuticle. In addition to the confidence interval values, these cuticle values were also determined to be significantly different or not using the ManneWhitney test. Within each of the two taxa cuticles (Appendix G.3), N/S distinguished three groups of layers, N/K distinguished two groups of layers, and N/Cl was almost

G. Guignard et al. / Cretaceous Research 71 (2017) 24e39

29

Fig. 5. Ultrastructural transmission electron micrographs of epidermal and hypodermal cell cuticles. OP ¼ outer part, IP ¼ inner part, EC ¼ epidermal cell cuticle, CR ¼ cell remnants, HC ¼ hypodermal cell cuticle, SC ¼ subsidiary cell cuticle, GC ¼ guard cell cuticle, AW ¼ anticlinal wall of the epidermal cell cuticle, SC ¼ subsidiary cell cuticle, GC ¼ guard cell cuticle, OC ¼ outer chamber of the stomatal apparatus, IC ¼ inner chamber of the stomatal apparatus, EXM ¼ extra-cuticular material. The cuticle is made with polylamellate and wavy A1 and A2 layers (¼cuticle proper), fibrilous B1 and granular B2 (¼cuticular layer) layers. Except micrographs D and Q-U which are longitudinal ultrathin sections from treated material, all others are transversal sections from slightly treated material. AeR. Epidermal cell cuticles. A. General view. B. In this general view, A1 polylamellate layer is going deeply in the A2 subjacent granular layer. C refers to micrograph C of the same figure. C. Detail of B, with A1 layer and its polylamellae visible in some parts. D. Rare case of A1 unlamellate layer. E. Another case not often observed, with A1 numerous and straight polylamellae covering the A2 granular layer. FeI. Usual cases with wavy polylamellae of the A1 layer above A2 granular layer. J. Very unusual observation of the A1 polylamellate layer in a wide opened space of the outermost part of the cuticle. K refers to micrograph K of the same figure. K. Detail of micrograph J with numerous polylamellae of the A1 layer visible in some parts. L. Outer and middle parts of cuticle with cuticle proper (A1 þ A2) and B1 fibrilous layer (part of cuticular layer). MeO. Different aspects of B1 fibrilous layer. P. Middle and inner parts of cuticle with A2 (part of cuticle proper), fibrilous B1 and granular B2 layer (¼cuticular layer). QeR. Inner parts of cuticles with B1 fibrilous and B2 granular layers. SeU. Hypodermal cell cuticles. SeT. Different aspects of hypodermal cuticles, below epidermis cell locations and more or less entirely preserved. (U refers to micrograph U of the same figure). U. Detail of hypodermal cuticle of micrograph T, made with B2 granular layer.

30

G. Guignard et al. / Cretaceous Research 71 (2017) 24e39

G. Guignard et al. / Cretaceous Research 71 (2017) 24e39 Table 1 Statistical values, based on 30 measurements of the thickness for the four kinds of cell cuticles. Ordinary epidermal cell cuticle

Hypothetical hypodermal cell cuticle

Mean MineMax

%

st-d

Mean

MineMax

%

st-d

CM

9.49

4.68e14.95

100

3.61

3.12

1.81e6.13

100

1.18

CP (A) A1 OL/nm TL/nm A2

1.10 0.097 4.63 6.60 1.01

0.45e2.28 11.6 0.48 0.015e0.348 1 0.09 1e10 2.86 2e11 3.10 0.18e2.26 10.6 0.52

Absent Absent Absent Absent Absent

CL(B) B1 B2

8.39 7.56 0.83

3.85e13.60 3.5e12.98 0.02e4.19

3.12 1.81e6.13 Absent 3.12 1.81e6.13

100

1.18

100

1.18

88.4 3.58 79.7 3.44 8.7 0.96

Stomatal apparatus Subsidiary cell cuticle

Guard cell cuticle

CM

9.60

3.62e17.42

100

3.70

CP (A) A1 OL/nm TL/nm A2

1.10 0.141 5.87 6.83 0.96

0.34e2.14 11.5 0.57 1.88 0.026e0.277 1.5 0.072 0.093 4e9 1.72 6.81 4e13 8.28 8.79 0.20e2.11 10 0.61 1.79

CL (B) B1 B2

8.50 7.67 0.82

2.77e15.56 1.95e13.55 0.18e2.97

88.5 3.44 79.9 2.96 8.6 0.65

1.88

0.80e4.38

100

0.80e4.38 100 0.049e0.132 4.8 3.5e15 2.5e17 0.72e4.27 95.2

0.91 0.91 0.024 3.59 4.86 0.91

Absent Absent Absent

Note that Minemax ¼ minimum and maximum values of thickness observed. % ¼ percentage of each detailed part of the cuticle. st-d ¼ standard deviation. The cuticular membrane CM is made up with cuticle proper CP (¼A ¼ A1 þ A2 layers) and cuticular layer CL (¼B ¼ B1 þ B2 layers). A1 layer cuticle is composed of alternate lamellae, opaque OL and translucent TL. Except for the very thin OL and TL measured in nm, all other measurements are in mm.

homogeneous among the four layers. In addition, N/S had the largest variation between the two taxa (between 0.15 and 1.81 in mean), while N/K was between 0.12 and 1.41 and N/Cl was between 0.09 and 0.68. Combining these two data sets, each layer of each taxon has its own characteristics. 4. Discussion and comparisons 4.1. Potential taxonomical implications 4.1.1. Ultrastructures of four types of cuticles of Pseudofrenelopsis gansuensis Apart from the cuticles of ordinary epidermal cells and stomatal apparatuses, the hypodermis cuticle was examined using TEM for the first time in Cheirolepidiaceae. This structure comprises only the granular zone (B2) and is devoid of cuticle proper A and B1 of the fibrilous layer. This rare observation is supported by a

31

comparison with the extant plant cuticle. In ultrastructure, this cuticular structure was comparable with the hypodermis Phorandendron flavescens (Pursh) Nutt. var. macrophyllum growing on Juglans hindsii (Jeps.) Jeps (cf. Calvin, 1970). Also according to Calvin (1970), the cuticle specimen shown in his fig. 27, which is similar to Fig. 5, SeU herein, is a “region where sub-epidermal cells have a cuticular layer”. Jeffree (2006) reported an “internal cuticle” that extends deeply to the surface of the “palisade and spongy mesophyll cells”. The figures (e.g., fig. 35) provided by Calvin (1970) illustrating the presence of cuticle in the inner part of the leaf are unfortunately too low in magnification to show the ultrastructural details, but this structure is clearly granular and belongs to the B2 granular part of the B layer. The three other main types of cuticles (ordinary epidermal cells OEC, subsidiary and guard cell cuticles SC and GC) were compared based on the confidence interval value (CI) of nine characters (Appendices B and G.1). First, homogeneity was revealed as the statistical value of two similar characters (opaque and translucent lamellae of the A1 layer) among the three cuticles, which are slightly variable in thickness. Among six characters (total CM, A, A2, B, B1, B2), OEC and SC were similar to each other and significantly different from the guard cell cuticles. However, the A1 layer is homogeneous (in confidential interval) between OEC and GC, but different from that of the subsidiary cell cuticle. The results are summarized in the key of Table 2 and demonstrated in the threedimensional reconstruction (Fig. 8). The key is based on the presence or absence of cuticular layer B, according to the thickness of the A1 layer that clearly separates OEC-SC from GC. The proportions and measurements of the other layers of the cuticle were enhanced, and the precise measurement of the A1 layer was used to distinguish OEC and SC. The resemblance of OEC and SC of P. gansuensis could be correlated with the non-specific function of SC, as these layers are both quite similar and flat (no papillae), with a majority of insignificant proportions of layers; in most other taxa these cuticles are different. The GC is quite different, as it is a layer thin, as also observed in other Cheirolepidiaceae. In Pseudofrenelopsis dalatzensis, this layer is on average 1.77 mm thick, 5.63 mm for SC and 12.27 and 9.41 mm for OEC (Yang et al., 2009); in Hirmeriella muensteri, the GC was 0.8 mm thick, 8 mm for SC and 11.5 mm for OEC (Guignard et al., 1998). These findings are also consistent with the results of several previous detailed ultrastructural studies on the cuticles of other fossil groups compared with Cheirolepidiaceae, where the stomatal apparatus could be studied. In the genus Dichopteris (Pteridospermales), GC is 3.6 mm, 14.1 mm for SC and 14.7e17.8 mm venard et al., 2005). In pteridosperms, the genus for OEC (The Pachypteris has a GC of 3.99 mm, a SC of 5.30 mm and an OEC of 13.68 to 12.26 mm (Guignard et al., 2004); in the fossil species of Ginkgoites (G. ticoensis, Ginkgoales), the cuticle is 0.46 mm thick for GC, 1.47 mm thick for SC and 1.02 mm thick for OEC (Del Fueyo et al.,

Fig. 6. Ultrastructural transmission electron micrographs of stomatal apparatus cuticle, details of subsidiary and guard cell cuticles. OP ¼ outer part, IP ¼ inner part, EC ¼ epidermal cell cuticle, CR ¼ cell remnants, HC ¼ hypodermal cell cuticle, SC ¼ subsidiary cell cuticle, GC ¼ guard cell cuticle, AW ¼ anticlinal wall of the epidermal cell cuticle, EXM ¼ extracuticular material. The cuticle is made with polylamellate and wavy A1 and A2 layers (¼cuticle proper), fibrilous B1 and granular B2 (¼cuticular layer) layers. All micrographs are provided from transversal ultrathin sections of slightly treated material. AeC. General views. A. Section of a whole stomatal apparatus. B and C refer to micrographs B and C of this figure. BeC. Details of the two parts of A. Note two different aspects of guard cell cuticles, on both sides of the stomatal apparatus. G refers to micrographs G on the same figure. DeT. Details of subsidiary cell cuticles. D. Whole cuticle of subsidiary cell, A1 and A2, B1 and B2 layers. EeG. Details of several boundaries between subsidiary and guard cell cuticles, G being an enlargement of B. Note the wavy A1 polylamellate layer. H. Part of stomatal apparatus with subsidiary cell cuticle with seemingly detached and rounded part at the top of the outer chamber (compare with 15 and outer views of the SEM figure). IeK refer to micrographs IeK on the same figure. I. Detail of the outermost part of the subsidiary cell of H, with wavy A1 layer (arrow) made with polylamellae hardly visible in this case. JeK. Details of H, wavy A1 layer (arrow) along the outer chamber of the stomatal apparatus, with polylamellae more or less visible. L. Middle and lower parts of subsidiary cell cuticle. M. One side of the outer chamber of a stomatal apparatus, with wavy A1 layer, unlamellate in this case. N. Outer most part of the cuticle of a subsidiary cell, with wavy A1 polylamellate layer. O. Part of stomatal apparatus with seemingly detached and elongated part at the top of the outer chamber (compare with H and outer views of the SEM figure). P and Q refer to micrographs P and Q of the same figure. PeQ. Details of the two sides of the outer chamber of O, with wavy A1 layer, non-lamellate in this case. R. Outer most part of the cuticle of a subsidiary cell, with wavy A1 polylamellate layer covered with extracuticular material. SeT. Middle and lower parts of subsidiary cell cuticle. UeX. Details of guard cell cuticles. UeV. Entire cuticle with hardly visible A1 layer (arrows), wavy in some parts. WeX. Enlargements of outermost parts, with clear polylamellate layer rather straight.

32

G. Guignard et al. / Cretaceous Research 71 (2017) 24e39

Fig. 7. Examples of EDS analysis.

Table 2 Key to the identification of the three types of cuticles observed in P. gansuensis. General features

Proportions of layers

Measurements

*Total thickness: 9.49e9.60 mm / A1: 0.097 mm ordinary epidermal cell cuticle *A: 1.10 mm / A1: 0.14 mm subsidiary cell cuticle *A2: 0.96e1.01 mm A1 mainly lamellate. straight or *B: 8.39e8.50 mm wavy: 0.097e0.141 mm *B1: 7.56e7.67 mm *B2: 0.82e0.83 mm *Total thickness: 1.88 mm / Guard cell cuticle Absence of cuticular layer CL ¼ B *100% cuticle proper *A: 1.88 mm (CP ¼ A ¼ A1 þ A2): 4.8% A1. 95.2%A2 *A1: 0.093 mm A1 mainly lamellate. straight or *A2: 1.79 mm wavy: 0.05e0.10 mm qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi Note that only the distinctions of measurements with the confidence interval (¼ x± var n  1:96 giving 95% a risk) of the significant ultrastructural cuticle characters are noted with their mean values. Insignificant characters, i.e. opaque and translucent lamellae, are not indicated in this key. This table has to be read from the left to the right side. Presence of cuticular layer CL ¼ B

*11.5e11.6% cuticle proper CP ¼ A ¼ A1 þ A2 10e10.6% A2 *88.4e88.5% cuticular layer CL ¼ B ¼ B1 þ B2 79.7e79.9% B1. 8.6e8.7% B2

2013); in G. skottsbergii (Guignard et al., 2016), the GC is 0.47 mm and 2.89 and 2.07 mm for SC and OEC, respectively. The thickness of the guard cell cuticle is correlated with the function of closing or opening the stoma, as a thin cuticle is much more malleable than a thick cuticle. However, for Ginkgo, the guard cell cuticle is the thickest, and all other cuticles are nearly equal in thickness, with an average of 1.76 mm for GC, 1.49 mm for SC and 1.58 and 1.10 mm for OEC (Guignard and Zhou, 2005). In some living plants, such as Agave Americana, Plantago major and Ardisia crenata, the A1 polylamellate layer is correlated with impermeability (Fischer and Bayer, 1972; Wattendorff and Holloway, 1980, 1982; Wattendorff,

1992; Jeffree, 2006). The differences of A1 between OEC, SC and GC in various fossil taxa might imply different impermeable functions (Yang et al., 2009). In the present fossil materia, the thickness in proportion to the A1 polylamellate layer (4.8% versus 1 and 1.5% for OEC and SC) likely increased the impermeability of this type of cell. 4.1.2. Ultrastructure and taxonomy In Pseudofrenelopsis, as in other cheirolepidiaceous plants, the majority of specific characters are microscopic (Hill et al., 2012). For example, P. dalatzensis differs from P. gansuensis in the absence of

G. Guignard et al. / Cretaceous Research 71 (2017) 24e39

33

Fig. 8. Three-dimensional reconstruction of the cuticle of P. gansuensis, and comparisons with P. dalatzensis. Statistical percentages are indicated for each layer.

papillae and the Florin ring in the stomatal apparatus and presence of fissures between ordinary cells, in addition to the less developed hypodermal cells (Yang and Deng, 2007). Based on studies of P. dalatzensis (Yang et al., 2009) and Suturovagina intermedia (Mairot et al., 2014), the present study represents the third statistical analysis of the cuticle ultrastructure conducted in Cheirolepidiaceae based on 30 measurements of different types of cuticles. Although there is little data compared with the number of taxa in Cheirolepidiaceae among the potential 27 features of the two taxa of Pseudofrenelopsis (Table 3, Appendix C) (9 features  3 types of cuticles; typically 21, as 6 characters are not present in the SC or GC of both taxa), 10 of the features are potentially significant at a specific level, as their confidence interval values can be useful to distinguish P. dalatzensis from P. gansuensis. For specific separation, the cuticle characters of the A, A2, B, B1 and B2 layers are most significant. The characters of guard cells are not important (for only 2 out of 6 characters). Although cuticle proper A is often considered in cuticle studies, as it is the most informative external layer, cuticular layer B, which closer to the cell wall and occasionally more or less merged with it, is nevertheless more significant than cuticle proper A. The total thickness of the cuticular membrane CM and A1 layer and the thickness of the translucent lamellae are much less useful. The results are interesting,

providing the first evidence of the potential value of the cuticle ultrastructure in the specific identification of Pseudofrenelopsis, consistent with the findings of Axsmith et al. (2004) and Axsmith (2006), who showed through gross morphology and SEM cuticular characters, including cuticle details, that Pseudofrenelopsis material represented ‘different biological species’. Eleven cuticle ultrastructures were homogeneous in the genus Pseudofrenelopsis and could be potentially important for generic separation. In the three types of cells, the cuticle ultrastructures of GC are more significant than those of OEC and SC. Among the different layers, cuticle proper A is more markedly homogeneous in structure at the generic level than cuticular layer B. This number is reduced to 5 generic features compared with the genus Suturovagina (Table 4), as GC features could not be compared because GC is missing in this extremely thick cuticle. At the level of Family Cheirolepidiaceae (Table 4, Appendix D), the cuticle ultrastructures have been studied in detail for only one other genus, Suturovagina, which might be useful in comparison (Table 4, Appendix D). However, among the available comparable features among the three taxa (8  2 type of cells OEC and GC ¼ 16), three features are homogenous among Pseudofrenelopsis and Suturovagina (based on the present study and data from Yang et al.,

34

G. Guignard et al. / Cretaceous Research 71 (2017) 24e39

Table 3 Comparisons of the ultrastructural cuticle characters among two species within the genus Pseudofrenelopsis.

ordinary epidermal cell cuticle

stomatal apparatus cuticle subsidiary cell cuticle

guard cell cuticle

P.

P.

P.

P.

P.

P.

gansuensis

dalatzensis

gansuensis

dalatzensis

gansuensis

dalatzensis

9.60 μm

5.63 μm

1.42-1.88 μm

1.10 μm

5.63 μm

1.77-1.88 μm

CM cuticular

A+B

9.41-9.49 μm

membrane CP cuticle proper

A = A1 +

1.10 μm

8.45 μm

A2 CP cuticle proper

A1 N/S ratio

0.097-0.10 μm 0.17

N/K ratio CP cuticle

0.046 μm

0.093 μm

1.81 unmeasured character

0.12-0.42 0.14-1.17

N/Cl ratio

0.093-0.141 μm

1.01 μm

8.35 μm

N/K ratio

0.17 0.13 0.12

1.53 0.68 1.41

cuticular

B = B1 +

8.39 μm

0.96 μm

8.50 μm

absent

layer

B2

character absent

7.56 μm

0.50 μm

7.67 μm

absent

character absent

proper

A2 N/S ratio N/Cl ratio

0.96 μm

5.54 μm

unmeasured character

unmeasured character

CL

CL cuticular

B1

layer

unmeasured character

0.44-0.56

N/S ratio N/Cl ratio

0.09

0.48

N/K ratio

1.25

0.47

B2

0.83 μm

0.45 μm

N/S ratio

0.60

0.15

unmeasured character

CL cuticular

1.72-1.79 μm

0.82 μm

absent

unmeasured character character absent

layer

0.11-0.13

unmeasured character

unmeasured character

3.72-4.63 nm

5.87-6.32 nm

5.64-6.81 nm

6.16 -6.60 nm

6.83-8.48 nm

N/Cl ratio N/K ratio CP cuticle

OL/nm

proper TL/nm

1.19

0.11

8.79 nm

5.65 nm

qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi Note that the figures in blue colour are potential specific characters, indicating the mean using the confidence interval (¼ x± var n  1:96 giving 95 % a risk); the figures in grey colour are potential generic characters, indicating the minimum and maximum mean, not significantly different between the two species. See Appendix A and the P. dalatzensis data of Yang et al. (2009) for the values, Appendices C, D and F for detailed comparisons. OL ¼ opaque lamellae, TL ¼ translucent lamellae of the A1 polylamellate layer.

2009; Mairot et al., 2014), and could be considered, of course very cautiously, as characteristic of this Family. They are: 1/the cuticle membrane CM comprising A1 þ A2 þ B1 þ B2 layers, 2/the presence of a wavy polylamellate A1 layer, 3/the thickness of the translucent lamellae of the A1 layer. Moreover it is interesting to notice that although the cuticle ultrastructural data of another genus Hirmeriella (Guignard et al., 1998) is not included in this table, as these features are not sufficiently statistically detailed, this genus also bears these three common features. The present observation emphasizes the potential taxonomic importance of

these cuticle ultrastructural features for Cheirolepidiaceae, which are described in previous studies (Guignard et al., 1998; Yang et al., 2009; Mairot et al., 2014). The second ultrastructural feature (wavy polylamellate A1 layer) is particularly characteristic. This feature was also observed in some other genera of potentially cheirolepidiaceous affinity (Yang et al., 2009), such as Tarphyderma (Archangelsky and Taylor, 1986, their figs. 25 and 28) and Glenrosa (Zhou et al., 2000, their pl. 3 and fig. 9). It is not clearly observed in some genera that are referred to this family from the South Hemisphere, such as Tomaxellia biforme (Villar de

G. Guignard et al. / Cretaceous Research 71 (2017) 24e39

35

Table 4 Potential taxonomical level of significance of ultrastructural cuticle characters within Cheirolepidiaceae.

ordinary epidermal cell cuticle

stomatal apparatus cuticle

P= Pseudofrenelopsis S = Suturovagina

subsidiary cell cuticle P. gansuensis

P. dalatzensis

CM (cuticular

general

membrane)

description

CM (cuticular membrane)

A (+ B) thickness

CP (cuticle

A = A1 + A2

proper)

thickness

CP (cuticle

A1

proper)

thickness

CP (cuticle proper)

(cuticle

A2 thickness

CP (cuticle proper) CP (cuticle proper)

generic character 9.41-9.49 μm 100%

specific character 1.10 μm 11.6%

specific character 8.45 μm 89.8%

generic character 0.097-0.10 μm 1-1.1%

specific character 1.01 μm 10.6%

specific character 8.35 μm 88.7%

OL opaque lamellae

P. dalatzensis

S. intermedia

generic or specific character ? 114.6 μm 100%

specific character 9.60 μm 100%

specific character 5.63 μm 100%

generic or specific character ? 117.4 μm 100%

generic or specific character ? 73.9 μm 64.5%

specific character 1.10 μm 11.5%

specific character 5.63 μm 100%

generic or specific character ? 117.4 μm 100%

generic or specific character ? 0.26 μm 0.22%

generic character 0.093-0.141 μm 1-1.6%

generic or specific character ? 0.25 μm 0.21%

family character wavy layer

A1

proper)

P. gansuensis

family character wavy A1 + A2 + B1 + B2

description

CP

S. intermedia

generic character 3.72-4.63 nm

thickness/nm

generic or specific character ? 73.6 μm 64.3% generic or specific character ? 6.54 nm

specific character 0.96 μm 10%

specific character 5.54 μm 98.4%

generic character 5.87-6.32 nm

generic or specific character ? 117.1 μm 99.8% generic or specific character ? 8.27 nm

TL

family character 5.73 6.32 nm

translucent lamellae thickness/nm

Note. Comparable characters using confidence interval data (¼ x±

qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi var  1:96 giving 95 % a risk) between the genera Pseudofrenelopsis and Suturovagina are provided, and also n

general descriptions for two characters. Characters are indicated with the mean of thickness values, or minimum and maximum values, and percentages of thickness if relevant. CL (cuticular layer ¼ B) being only present in some cases and though not comparable, is not represented in this table. Blue colour represents potential specific characters, significantly different between the two specific P. gansuensis and P. dalatzensis (data from Yang et al. (2009)). For comparisons with the generic Suturovagina (data from Mairot et al. (2014)), as just one species was studied it is indicated with “generic or specific character?” as the level of significance cannot be checked more precisely. For this latter taxon guard cell features are not indicated as they were not present in the extremely thick material observed. Grey colour represents potential generic character, not significantly different between the two Pseudofrenelopsis specific (and though merged together) but significantly different between the two genera Pseudofrenelopsis and Suturovagina. Orange colour represents potential family characters, not significantly different between the two genera Pseudofrenelopsis and Suturovagina and though merged together. See Appendix A, the data of Yang et al. (2009) and Mairot et al. (2014) for the values, Appendix D for the detailed comparisons.

Seoane, 1998), however as high magnifications and statistics are not available for this taxon this is still a question, for instance her photo 4 (Plate II) seems to show a quite thick A1 layer which is not straight at all. Moreover, in other taxa than Cheirolepidiaceae, where the A1 layer is present and clearly visible, it is never wavy, but always

straight, such as in Sphenobaiera of Ginkgoales (Wang et al., 2005), or is even divided in two straight sub-layers, A1 upper A1U and A1 lower A1L, as in Ginkgo (Guignard and Zhou, 2005; Del Fueyo et al., 2006; Del Fueyo et al., 2013; Guignard et al., 2016) and Pachypteris (Guignard et al., 2004).

36

G. Guignard et al. / Cretaceous Research 71 (2017) 24e39

The third ultrastructural feature (thickness of the translucent lamellae of the A1 layer) is likely also a potential distinguished feature of the Cheirolepidiaceae, however, this feature is with less adequate, as only a few cuticles have been observed. The mean thickness of the translucent A1 layer was consistently higher in Pachypteris gradinarui Popa (Guignard et al., 2004, their table 1) than that of all of the Cheirolepidiaceae cuticles examined measured thus far. In Ginkgoales, among the numerous types of cuticles observed, these lamellae are thinner in most cases, such as in Ginkgoites ticoensis (Del Fueyo et al., 2013, their table 1), except for that of OEC and Ginkgoites skottsbergii (Guignard et al., 2016, their table 1). This finding was also largely consistent for the fossil Ginkgo yimaensis (Guignard and Zhou, 2005, their table 4) and male and female trees of living Ginkgo biloba (Guignard and Zhou, 2005, their tables 2 and 3), except for that of GC of the male ginkgo tree. It has also to be added that to our knowledge, among comparable data where international description (Holloway, 1982; Archangelsky and Taylor, 1986) of the cuticle layers can be used, no other fossil families and Orders are identical to the Cheirolepidiaceae in these three cuticle ultrastructural features, enhancing the characteristics just discussed. The cuticular membrane of another Coniferales (genus Mirovia of the Miroviaceae, Nosova et al., 2016), is heterogeneous among the 3 species observed (with A1, divided in A1U and A1 L or not; A2, divided in A2 U and A2 L or not; B1). In the other only Coniferales so far known (living Pinus of the Pinaceae, Bartiromo et al., 2012), the cuticle is very different as just made with B1 layer. Czekanowskiales have a cuticle proper A (¼A2) þ a cuticular layer B differentiated in two sublayers (¼B1 þ B2) (Zhou and Guignard, 1998). Some of the Ginkgoales have a cuticle proper A (A1U þ A1L) and a cuticular layer B (B1) (Wang et al., 2005; Guignard and Zhou, 2005; Del Fueyo et al., 2006; Del Fueyo et al., 2013). The cuticular membranes of Komlopteris and Dichopteris of the Pteridospermales comprise only venard et al., 2005), granular layer A2 (Guignard et al., 2001; The while those of Ruflorinia of the Caytoniales comprise A2 and B (B1 and B2) for the adaxial cuticle and B (B1 and B2) for the abaxial cuticle (Carrizo et al., 2014). In the Bennettitales, three types of CM have been observed: outer layer lamellate, inner layer alveolate; outer layer alveolate or reticulate, inner layer lamellate-reticulate; and the outer layer reticulate, inner layer lamellate (Villar de Seoane, 1999, 2001, 2003). Cycadalean cuticles vary in ultrastructure, but are easily distinguishable (Artabe et al., 1991; Archangelsky et al., 1995; Passalia, Del Fueyo and Archangelsky, 2010). In another genus (Pachypteris (Corystospermaceae, Guignard et al., 2004), of which the cuticles have been examined using a statistical approach, the upper cuticle comprises A (¼A1 upper þ A1 lower þ A2) and B (¼B1) and the lower cuticle comprises A (¼A1 þ A2) and B (¼B1). 4.1.3. EDS analyses The presence and function of the molecules in cuticles have primarily been examined in a few living taxa (Fernandez and Eichert, 2009; Dominguez et al., 2011; Fernandez and Brown, 2013). Although some elements have been increasingly examined, particularly sulphur, with respect to the environment and stomatal density for whole leaves (Haworth and McElwain, 2008a,b; Retallack, 2009; Haworth et al., 2012; Elliott-Kingston et al., 2014), there are still few cuticle analyses of the elements. However, Bartiromo et al. (2012, 2013) showed that although sulphur is present in volcanic fumaroles, this compound is completely absent from the cuticles of Pinus (gymnosperm) and Erica (angiosperm), as examined using EDS with TEM sections, although in the case of fossil Pseudofrenelopsis, the presence of sulphur in the two taxa could indicate a different environment or a different metabolism, as this element is evocated for living plants

(Pautler et al., 2013; citing; Droux, 2004). The nitrogen concentration has been examined in orchids (Paphiopedilum and Cypripedium) for adaptive significance allied with cuticle thickness (Guan et al., 2011) and in Laurus and Ceratonia (Grammatikopoulos et al., 1998), associated with the water permeability of the cuticle of ivy (Hedera helix; Santrucek et al., 2004). Potassium plays a role in plant cuticle development (genera Agave and Clivia, Wattendorff and Holloway, 1984), and its penetration through the cuticle has been examined in various € nherr, 2002; Elshatshat et al., 2007), particuangiosperms (Scho € nherr and Luber, 2001). larly in Pyrus and Citrus (Scho For fossil plants, the presence and function of elements, such as N, S, Cl and K, has been difficult to envisage; however, statistically significant differences can be interpreted at least fruitfully as taxonomical features. Among the 12 features of the OEC cuticle (3 ratios  4 layers), at the species level for the two species Pseudofrenelopsis gansuensis and P. dalatzensis, between the four layers (A1, A2, B1, B2), EDS analysis reveals 8 new elements features (¼significant differences between the two taxa using the ManneWhitney test) with N/S, N/Cl, and N/K ratios (Table 3; Appendices E.3, E.4, F). Although the number of measurements is not so high (5 measurements for each value, due to the high technique and time required for each measurement), the standard deviation is almost quite low, though these values are quite homogeneous and trustable (Appendix E.2). Comparing the four layers (A1, A2, B1, and B2) in each taxon (Pseudofrenelopsis gansuensis and P. dalatzensis, Appendix G.3), N/S is the best for distinction, followed by N/K, and N/Cl. Comparing the same layer between two taxa, the three ratios enable distinction (Table 3); however, N/S and N/K enable better distinction (3 out of 4 layers) than N/Cl (2 out of 4). At the generic level, (¼insignificant differences between the two taxa using the ManneWhitney test), 4 ratios are relevant. Concerning the potential influence of the sediments on the chemical composition of fossil cuticles, within each taxon, the cuticle is not homogeneous, showing significant differences between the 4 layers (Appendix G.3), enhancing the precise details of each layer and a rather low influence of sediments, consistent with the general variation and heterogeneity in the chemical composition of each layer previously studied in living plants (Jeffree, 2006; Stark and Tian, 2006; Pollard et al., 2008; Dominguez et al., 2011), corresponding to the adequate quality of the preservation of the fossil material. These potential taxonomical elemental ratios should be added to the recent Gingkoites study (Gingkoaceae; Guignard et al., 2016), where two fossil taxa show several ratios (N/Cl, K/S, N/Ca, S/Ca, and K/Ca) that are potentially useful at the specific or generic level, thereby increasing the knowledge of fossil taxonomy in future studies. Taxonomic features are examined in fossilization and might consider elemental studies (Zodrow and Mastalerz, 2009), but with much more complex biomolecules (D'Angelo and Zodrow, 2015, 2016). Using aromatic compounds and aliphatic residues for paleosols of Canadian arctic material (Simpson et al., 2003; Simpson et al., 2007), Kelleher and Simpson (2006) and Pautler et al. (2013) revealed a primarily angiosperm quaternary fossil vegetation, with a majority of dicotyledons and a minority of monocotyledons. More precisely, Al might be used as one of the elements for the distinction between genera or species (gymnosperms in D'Angelo et al., 2010); seed ferns, conifers and cycad-related fossils (D'Angelo and Zodrow, 2011; D'Angelo et al., 2011); pteridophyll cuticles in Zodrow and Mastalerz (2007); seed ferns (Zodrow et al., 2010; D'Angelo et al., 2012); and also for N and Cl (tree ferns in Stoyko et al., 2013). Although limited in taxa number (2), the present elemental Pseudofrenelopsis study reveals a potential interest for taxonomy, with both genus and species features.

G. Guignard et al. / Cretaceous Research 71 (2017) 24e39

Concerning the potential influence of the rock sediments on the chemical composition of fossil cuticles, matrix-containing Pseudofrenelopsis gansuensis and P. dalatzensis were analysed using an energy dispersive X-ray (EDX) system attached to a LEO1530VP. The results showed a markedly different composition between these species: the absence of a matrix for N and Cl for Pseudofrenelopsis gansuensis and the absence of N, S and Cl for P. dalatzensis. Notably, within each taxon, each cuticle shows significant differences between the 4 layers (Appendix G.3), enhancing the precise details of each layer and no influence of the rock sediments. 4.2. Palaeoecological considerations Most Cheirolepidiaceae bear xeromorphic characters and are generally considered to be halophytes and thermophilous (Alvin, venard et al., 2000; Watson, 1988; Zhou, 1982; Du et al., 2014; The 1983). The genus Pseudofrenelopsis and other cheirolepidiaceous conifers from different areas worldwide exhibit an increased ecological amplitude: the cheirolepidiaceous conifers of the English Wealden flora likely lived in river margin swamps (Oldham, 1976), and the Pseudofrenelopsis-inhabited soils with variable water availability (Alvin, 1982, 1983); Pseudofrenelopsis parceramosa from North America, Europe and China might have adapted to different salinity environments or even non-saline habitats (Upchurch and Doyle, 1981; Zhou, 1995); additionally, cheirolepidiaceous conifers Otwayia from high-latitudes of southeastern Australia might adapt to humid conditions (Tosolini et al., 2015). Pseudofrenelopsis gansuensis also exhibits a number of xeromorphic characters: the leaves are adpressed on the axis, loosely arranged in a simple spiral, and end in a reduced free triangular apex. The thick cuticles and the presence of cutinized hypodermis are additional morphological characters that prevent excessive evaporation from the leaves. The leaf cuticle ultrastructure of Pseudofrenelopsis gansuensis is also of significance. A thick peripheral, curved and wavy polylamellate zone (A1 layer) is present in the cuticle proper of Pseudofrenelopsis gansuensis. Until recently, all cheirolepidiaceous conifers examined in detail showed this common cuticle ultrastructural characters: Hirmeriella muensteri from the Liassic of Franconia, Germany (Guignard et al., 1998), Pseudofrenelopsis dalatzensis from the Lower Cretaceous of Longjing, China (Yang et al., 2009) and Suturovagina intermedia from the Lower Cretaceous of Nanjing, China (Mairot et al., 2014). In addition, the polylamellate zone (A1 layer) in the cuticle proper is present in Tarphyderma and Glenrosa, both potentially belonging to the Cheirolepidiaceae (Archangelsky and Taylor, 1986; Srinivasan, 1992; Zhou et al., 2000). 5. Conclusions The Early Cretaceous cheirolepidiaceous conifer Pseudofrenelopsis gansuensis from Wangqing Jilin Province in China was conducted in detail using light microscope, scanning and transmission electron microscopy techniques, which improve our understanding of ultrastructure characters of Cheirolepidiaceae. The ultrastructure characters of hypodermis cuticle was firstly observed in Cheirolepidiaceae, which comprises only the granular zone (B2) and is devoid of cuticle proper A and B1 of the fibrilous layer. Statistical measurements and the use of confidence interval, comparisons with the few other cheirolepidiacean taxa providing statistics and also other Orders and Families, allowed cautiously to distinguish with their thicknesses 10 potential specific characters (5 for OEC: A cuticle proper and A2 layer, B cuticular layer, B1 and B2 layers; 3 for SC: total cuticular membrane CM, A and A2; 2 for GC: A1 and the translucent lamellae thickness TL of the A1 layer), 11

37

potential generic characters (4 for OEC: cuticular membrane CM, A1, opaque OL and translucent lamellae of the A1 layer; 3 for SC: A1, opaque and translucent lamellae of the A1 layer; 4 for GC: CM, A and A2, opaque OL lamellae of the A1 layer), 3 potential Family characters (presence of A1 polylamellate layer, A2 granular layer, fibrilous B1 and granular B2 layers; presence “waves” in A1 layer; thickness of translucent TL lamellae of the A1 layer). The cuticle ultrastructures of all species studied of Cheirolepidiaceae bear three common features: the entire cuticle comprises the A1 þ A2 þ B1 þ B2 layers, the presence of a wavy polylamellate A1 layer and the thickness of the translucent lamellae of the A1 layer. Three elemental analysis ratios (N/S, N/K, N/Cl) of Pseudofrenelopsis dalatzensis and P. gansuensis shows that the elements of the sediments had no influence on the cuticle. Although the number of taxa is still too less to have a general view of the cuticle ultrastructure interest, these studies are clearly more and more precise and promising. Acknowledgements The study was supported by the National Natural Science Foundation of China (grant NFSC 41472011), and the National Basic Research Program of China (grant 2012CB822003) as well as a State Key Project for Research and Development from Ministry of Science and Technology (2016YFC0600406), Strategy Pilot Program of the Chinese Academy of Sciences (XDB18030502). We wish to thank the Editor for his fruitful comments, also Dr Mihai Popa and an unknown reviewer. Many thanks also to the technical staff of the University Lyon 1 «centre technologique des microstructures CTm”, atrice Burdin and Xavier Jaurand; Sophie Cros for especially Be preparing most of technical work. We also thank Prof. E.M. Friis for providing all facilities during one of the three authors, Dr. X-J Yang studied in Swedish Museum of Natural History, and to Prof. Z-Y Zhou for helpful criticism and improvement of the study. References Alvin, K.L., 1982. Cheirolepidiaceae: biology, structure and palaeoecology. Review of Palaeobotany and Palynology 37, 71e98. http://dx.doi.org/10.1016/00346667(82)90038-0. Alvin, K.L., 1983. Reconstruction of a Lower Cretaceous conifer. Botanical Journal of the Linnean Society 86, 169e176. http://dx.doi.org/10.1111/j.1095-8339.1983.tb00724.x. Archangelsky, S., Del Fueyo, G., 1989. Squamastrobus gen. n., a fertile podocarp from the Early Cretaceous of Patagonia, Argentina. Review of Palaeobotany and Palynology 59, 109e126. http://dx.doi.org/10.1016/0034-6667(89)90010-9. Archangelsky, S., Taylor, T.N., 1986. Ultrastructural studies of fossil plant cuticles. II. Tarphyderma gen. n., A Cretaceous conifer from Argentina. American Journal of Botany 73, 1577e1587. Stable URL: http://www.jstor.org/stable/2443925. Page Count: 11. Archangelsky, A., Andreis, R., Archangelsky, S., Artabe, A., 1995. Cuticular characters adapted to volcanic stress in a new Cretaceous cycad leaf from Patagonia, Argentina. Considerations on the stratigraphy and depositional history of the  Formation. Review of Palaeobotany and Palynology 89, 213e233. SSDI Baquero 0034 6667 (95) 00011 9. Artabe, A.E., Zamuner, A.B., Archangelsky, S., 1991. Estudios cuticulares en cycado  siles. El ge nero Kurtziana Frenguelli 1942. Ameghiniana 28, 365e374. psidas fo http://www.academia.edu/1151697_ISSN_0002-7014. Axsmith, B.J., 2006. The vegetative structure of a Lower Cretaceous conifer from Arkansas: further implications for morphospecies concepts in the Cheirolepidiaceae. Cretaceous Research 27, 309e317. http://dx.doi.org/10.1016/ j.cretres.2005.07.001. Axsmith, B.J., Krings, M., Waselkov, K., 2004. Conifer pollen cones from the Cretaceous of Arkansas: implications for diversity and reproduction in the Cheirolepidiaceae. Journal of Paleontology 78, 402e409. http://dx.doi.org/10.1666/ 0022-3360(2004)078%3C0402:CPCFTC%3E2.0.CO;2. Bartiromo, A., Guignard, G., Barone-Lumaga, M.R., Barattolo, F., Chiodini, G., Avino, R., Guerriero, G., Barale, G., 2012. Influence of volcanic gases on the epidermis of Pinus halepensis Mill. in Campi Flegrei, Southern Italy: a possible tool detecting volcanism in present and past floras. Journal of Volcanology and Geothermal Research 233e234, 1e17. http://dx.doi.org/10.1016/j.jvolgeores.2012.04.002. Bartiromo, A., Guignard, G., Barone-Lumaga, M.R., Barattolo, F., Chiodini, G., Avino, R., Guerriero, G., Barale, G., 2013. The cuticle micromorphology of in situ Erica arborea L. exposed to long-term volcanic gases in Phlegrean Fields,

38

G. Guignard et al. / Cretaceous Research 71 (2017) 24e39

Campania, Italy. Environmental and Experimental Botany 87, 197e206. http:// dx.doi.org/10.1016/j.envexpbot.2012.10.006. Bureau of Geology and Mineral Resources of Jilin Province, 1997. Stratigraphy (Lithostratic) of Jilin Province. China university of Geosciences Press, Wuhan, 324pp. Calvin, C.L., 1970. Anatomy of the Aerial Epidermis of the Mistletoe, Phoradendron flavescens. Botanical Gazette 131, 62e74. http://dx.doi.org/10.1086/336513. Carrizo, M.A., Del Fueyo, G.M., Medina, F., 2014. Foliar cuticle of Ruflorinia orlandoi nov. sp. (Pteridospermophyta) from the Lower Cretaceous of Patagonia. Geobios 47, 87e99. http://dx.doi.org/10.1016/j.geobios.2014.01.003. Chow, T.Y., Tsao, C.Y., 1977. On eight new species of conifers from the Cretaceous of East China with reference to their taxonomic position and phylogenetic relationship. Acta Sinica 16, 165e181 (in Chinese with English abstract). D'Angelo, J.A., Zodrow, E.L., 2011. Chemometric study of functional groups in different layers of Trigonocarpus grandis ovules (Pennsylvanian seed fern, Canada). Organic Geochemistry 42, 1039e1054. http://dx.doi.org/10.1016/ j.orggeochem.2011.06.022. D'Angelo, J.A., Zodrow, E.L., 2015. Chemometric study of structural groups in €tte, Canada): chemotaxomedullosalean foliage (Carboniferous, fossil Lagersta nomic implications. International Journal of Coal Geology 138, 42e54. http:// dx.doi.org/10.1016/j.coal.2014.12.003. D'Angelo, J.A., Zodrow, E.L., 2016. 3D chemical map and a theoretical life model for Neuropteris ovata var. simonii (index fossil, Asturian, Late Pennsylvanian, Canada). international Journal of Coal Geology 153, 12e27. http://dx.doi.org/ 10.1016/j.coal.2015.11.007. D'Angelo, J.A., Zodrow, E.L., Camargo, A., 2010. Chemometric study of functional groups in Pennsylvanian gymnosperm plant organs (Sydney Coalfield, Canada): implications for chemotaxonomy and assessment of kerogen formation. Organic Geochemistry 12, 1312e1325. http://dx.doi.org/10.1016/j.orggeochem.2010.09.010. D'Angelo, J.A., Escudero, L.B., Volkheimer, W., Zodrow, E.L., 2011. Chemometric analysis of functional groups in fossil remains of the Dicroidium flora (Cacheuta, Mendoza, Argentina): Implications for kerogen formation. International Journal of Coal Geology 2, 97e111. http://dx.doi.org/10.1016/j.coal.2011.05.005. D'Angelo, J.A., Zodrow, E.L., Mastalerz, M., 2012. Compression map, functional groups and fossilization: a chemometric approach (Pennsylvanian neuropteroid foliage, Canada). International Journal of Coal Geology 90, 149e155. http:// dx.doi.org/10.1016/j.coal.2011.11.009. Del Fueyo, G., Villar de Seoane, L., Archangelsky, S., Guignard, G., 2006. Estudios cuticulares de Ginkgoites Seward del Creta'cico Inferior de Patagonia. Revista del Museo Argentino de Ciencias Naturales, NS 8, 143e149. Available from: macn. secyt.gov.ar. Del Fueyo, G., Archangelsky, S., Llorens, M., Cuneo, R., 2008. Coniferous ovulate cones from the Lower Cretaceous of Santa Cruz Province, Argentina 4. International Journal of Plant Sciences 169, 799e813. http://dx.doi.org/10.1086/ 533608. Del Fueyo, G., Guignard, G., Villar De Seoane, L., Archangelsky, S., 2013. Leaf cuticle anatomy and the ultrastructure of Ginkgoites ticoensis Archang. from the Aptian of Patagonia. International Journal of Plant Sciences 406e424. http://dx.doi.org/ 10.1086/668221. Deng, S.H., Yang, X.J., Lu, Y.Z., 2005. Pseudofrenelopsis (Cheirolepidiaceae) from the Lower Cretaceous of Jiuquan, Gansu, northwestern China. Acta Palaeontologica Sinica 44, 505e516. Dominguez, E., Heredia-Guerrero, J.A., Heredia, A., 2011. The biophysical design of plant cuticles: an overview. New Phytologist 189, 938e949. http://dx.doi.org/ 10.1111/j.1469-8137.2010.03553.x. Droux, M., 2004. Sulfur assimilation and the role of sulfur in plant metabolism: a survey. Photosynthesis Research 79, 331e348. http://dx.doi.org/10.1023/B: PRES.0000017196.95499.11. Du, B., Zhang, M., Dai, S., Sun, B., 2014. Discovery of Pseudofrenelopsis from the Lower Cretaceous of Liupanshan Basin and its paleoclimatic significance. Cretaceous Research 48, 193e204. http://dx.doi.org/10.1016/j.cretres.2013.12.003. Elliott-Kingston, C., Haworth, M., McElwain, J.C., 2014. Damage structures in leaf epidermis and cuticle as an indicator of elevated atmospheric sulphur dioxide in early Mesozoic floras. Review of Palaeobotany and Palynology 208, 25e42. http://dx.doi.org/10.1016/j.geobios.2014.01.003. €nherr, J., 2007. Some cesium and potassium salts Elshatshat, S., Schreiber, L., Scho increase the water permeability of astomatous isolated plant cuticles. Journal of Plant Nutrition and Soil Science 170, 59e64. http://dx.doi.org/10.1002/ jpln.200625042. Fernandez, V., Brown, P.H., 2013. From plant surface to plant metabolism: the uncertain fate of foliar-applied nutrients. Frontiers in Plant Science 1e5. http:// dx.doi.org/10.3389/fpls.2013.00289. Fernandez, V., Eichert, T., 2009. Uptake of hydrophilic solutes through plant leaves: current state of knowledge and perspectives of foliar fertilization. Critical Reviews in Plant Sciences 28, 36e68. http://dx.doi.org/10.1080/07352680902743069. Fischer, D.A., Bayer, D.E., 1972. Thin section of plant cuticles, demonstrating channels and wax platelets. Canadian Journal of Botany 50, 1509e1511. http:// dx.doi.org/10.1139/b72-186. Grammatikopoulos, G., Kyparissis, A., Drilias, P., Petropoulou, Y., Manetas, Y., 1998. Effects of UV-B radiation on cuticle thickness and nutritional value of leaves in two mediterranean evergreen sclerophylls. Journal of Plant Physiology 153, 506e512. http://serials.unibo.it/cgi-ser/start/en/spogli/df-s.tcl? prog_art¼5992564&language¼ENGLISH&view¼articoli.

Guignard, G., Zhou, Z.Y., 2005. Comparative studies of leaf cuticle ultrastructure between living and the oldest known fossil ginkgos in China. International Journal of Plant Sciences 166, 145e156. http://dx.doi.org/10.1086/425667. venard, F., Van Konijnenburg-Van Cittert, J.H.A., 1998. Cuticle ultraGuignard, G., The structure of the cheirolepidiaceous conifer Hirmeriella muensteri (Schenk) Jung. Review of Palaeobotany and Palynology 104, 115e141. PII: S0034-6667(98)00052-9. Guignard, G., Boka, K., Barbacka, M., 2001. Sun and shade leaves? Cuticle ultrastructure of Jurassic Komlopteris nordenskioeldii (Nathorst) Barbacka. Review of Palaeobotany and Palynology 114, 191e208. PII: S0034-6667(00)00078-6. Guignard, G., Popa, M., Barale, G., 2004. Ultrastructure of Early Jurassic fossil plant cuticles: Pachypteris gradinarui Popa. Tissue and Cell 36, 263e273. http:// dx.doi.org/10.1016/j.tice.2004.04.002. Guignard, G., Del Fueyo, G.M., Villar de Seoane, L., Carrizo, M.A., Lafuente Díaz, M.A., 2016. Insights into the leaf cuticle fine structure of Ginkgoites skottsbergii lund. from the Albian of Patagonia and its relationship within Ginkgoaceae. Review of Palaeobotany and Palynology 232, 22e39. http://dx.doi.org/10.1016/ j.revpalbo.2016.05.002. Guan, Z.J., Zhang, S.B., Guan, K.Y., Li, S.Y., Hu, H., 2011. Leaf anatomical structures of Paphiopedilum and Cypripedium and their adaptive significance. Journal of Plant Research 124, 289e298. http://dx.doi.org/10.1007/s10265-010-0372-z. Haworth, M., McElwain, J., 2008a. Hot, dry, wet, cold or toxic? Revisiting the ecological significance of leaf and cuticular micromorphology. Palaeogeography, Palaeoclimatology, Palaeoecology 262, 79e90. http://dx.doi.org/10.1016/ j.palaeo.2008.02.009. Haworth, M., McElwain, J., 2008b. Reply to the comment on “Hot, dry, wet, cold or toxic? Revisiting the ecological significance of leaf cuticular micromorphology” by M. Haworth and J.C. McElwain. Palaeogeography, Palaeoclimatology, Palaeoecology 262 (2008) 79e90. Palaeogeography, Palaeoclimatology, Palaeoecology 273, 209e211. http://dx.doi.org/10.1016/ j.palaeo.2008.11.008. Haworth, M., Elliott-Kingston, C., Gallagher, A., Fitzgerald, A., McElwain, J.C., 2012. Sulphur dioxide fumigation effects on stomatal density and index of nonresistant plants: implications for the stomatal palaeo-[CO2] proxy method. Review of Palaeobotany and Palynology 182, 44e54. http://dx.doi.org/10.1016/ j.revpalbo.2012.06.006. Hill, C.R., Yang, X.J., Zhou, Z.Y., Doyle, J.C., 2012. Exceptionally preserved conifer twigs of Pseudofrenelopsis from the marine Lower Cretaceous of Yorkshire, England. Acta Palaeontologica Sinica 51, 395e410. Holloway, P.J., 1982. Structure and histochemistry of plant cuticular membranes: an overview. In: Cutler, D.F., Alvin, K.L., Price, C.E. (Eds.), The Plant Cuticle. Linnean Society, London, pp. 1e32. Jeffree, C.E., 2006. The fine structure of the plant cuticle. In: Riederer, M., Müller, C. (Eds.), Biology of the Plant Cuticle. Blackwell, UK, pp. 10e125. Kelleher, B.P., Simpson, A.J., 2006. Humic substances in soils: are they really chemically distinct? Environmental Science & Technology 40, 4605e4611. http://dx.doi.org/10.1021/es0608085. Kerp, J.H.F., 1990. The study of fossil gymnosperms by means of cuticle analysis. Palaios 5, 548e569. Li, G., Ohta, T., Batten, D.J., Sakai, T., Kozai, T., 2016. Morphology and phylogenetic origin of the spinicaudatan Neodiestheriafrom the Lower Cretaceous Dalazi Formation, Yanji Basin, north-eastern China. Cretaceous Research 62, 183e193.  la connaissance de la morphoge  ne se et de la Lugardon, B., 1971. Contribution a es isospore es. Unpublished Thesis. structure des parois sporales chez les Filicine Toulouse University, France (in French). Mairot, C., Guignard, G., Yang, X.J., Zhou, Z.Y., 2014. Cuticle micro-and ultrastructure of Suturovagina Chow et Tsao (Cheirolepidiaceae): taxonomic and palaeoecological implications. Review of Palaeobotany and Palynology 205, 9e21. http://dx.doi.org/10.1016/j.revpalbo.2014.02.011. Nosova, N., Yakovlevaa, O., Ivanovaa, A., Kiritchkovab, A., 2016. First data on the wna. ultrastructure of the leaf cuticle of a Mesozoic conifer, Mirovia Reymano Review of Palaeobotany and Palynology 233, 115e124. http://dx.doi.org/ 10.1016/j.revpalbo.2016.07.008 0034-6667/©2016. Oishi, S., 1941. Note on some Mesozoic plants from Lo-Tzu-Kou, province Chientao, “Manchoukuo”. Journal of Faculty of Science of Hokkaido Imperial University Series 4, 167e176. Oldham, T.C.B., 1976. Flora of the Wealden plant-debris beds of England. Palaeontology 19, 437e502. Passalia, M.G., Del Fueyo, G., Archangelsky, S., 2010. An Early Cretaceous zamiaceous cycad of South West Gondwana: Restrepophyllum nov. gen. from Patagonia, Argentina. Review of Palaeobotany and Palynology 161, 137e150. http:// dx.doi.org/10.1016/j.revpalbo.2010.04.001. Pautler, B.G., Sanborn, P.T., Simpson, A.J., Simpson, M.J., 2013. Molecular characterization of organic matter in Canadian Arctic paleosols for paleoecological applications. Organic geochemistry 63, 122e138. http://dx.doi.org/10.1016/ j.orggeochem.2013.08.009. Pollard, M., Beisson, F., Li, Y., Ohlrogge, J.B., 2008. Building lipid barriers: biosynthesis of cutin and suberin. Trends in Plant Science 13, 236e246. http:// dx.doi.org/10.1016/j.tplants.2008.03.003. Retallack, G.J., 2009. Comment on “Hot, dry, wet, cold or toxic? Revisiting the ecological significance of leaf cuticular micromorphology” by M. Haworth and J.C. McElwain. Palaeogeography, Palaeoclimatology, Palaeoecology 262 (2008) 79e90. Palaeogeography, Palaeoclimatology, Palaeoecology 273, 207e208. http://dx.doi.org/10.1016/j.palaeo.2008.08.019.

G. Guignard et al. / Cretaceous Research 71 (2017) 24e39 Richardson, K.C., Jarrett, L., Finke, E.H., 1960. Embedding in epoxy resins for ultrastaining sectioning in electron microscopy. Stain Technology 35e313. http:// dx.doi.org/10.3109/10520296009114754. Santrucek, J., Simanova, E., Karbulkova, J., Simkova, M., Schreiber, L., 2004. A new technique for measurement of water permeability of stomatous cuticular membranes isolated from Hedera helix leaves. Journal of Experimental Botany 401, 1411e1422. http://dx.doi.org/10.1093/jxb/erh150. €nherr, J., 2002. Foliar nutrition using inorganic salts: Laws of cuticular peneScho tration. Proceedings of the International Symposium on Foliar Nutrition of perennial Fruit Plants. Acta Horticulturae 594, 77e84. €nherr, J., Luber, M., 2001. Cuticular penetration of potassium salts: Effects of Scho humidity, anions, and temperature. Plant and Soil 236, 117e122. http:// dx.doi.org/10.1023/A:1011976727078. Simpson, A.J., Kingery, W.L., Hatcher, P.G., 2003. The identification of plant derived structures in humic materials using three-dimensional NMR spectroscopy. Environmental Science & Technology 37, 337e342. http://dx.doi.org/10.1021/ es025956j. Simpson, A.J., Simpson, M.J., Smith, E., Kelleher, B.P., 2007. Microbially derived inputs to soil organic matter: are current estimates too low? Environmental Science & Technology 41, 8070e8076. http://dx.doi.org/10.1021/es071217x. Srinivasan, V., 1992. Two new species of the conifer Glenrosa from the Lower Cretaceous of North America. Review of Palaeobotany and Palynology 72, 245e255. http://dx.doi.org/10.1016/0034-6667(92)90029-G. Stark, R.E., Tian, S., 2006. The cutin biopolymer matrix. In: Riederer, M., Müller, C. (Eds.), Biology of the Plant Cuticle. Blackwell Publishing, Oxford, pp. 126e144. Stoyko, S.S., Rudyk, B.W., Mar, A., Zodrow, E.L., D'Angelo, J.A., 2013. Powder X-ray diffraction and X-ray photoelectron spectroscopy of cutin from a 300 Ma tree fern (Alethopteris pseudograndinioides, Canada). International Journal of Coal Geology 106, 35e38. http://dx.doi.org/10.1016/j.coal.2013.01.006. venard, F., Guignard, G., Gomez, B., Zhou, Z.Y., 2000. Genre Glenrosa, exemple de The rophytes ou accommodation physiologique? Monografias de la plantes xe Academia de Ciencias, fisicas Químicas y Naturales de Zaragoza, Spain 16, 45e51. venard, F., Barale, G., Guignard, G., Daviero-Gomez, V., Gomez, B., Philippe, M., The Labert, N., 2005. Reappraisal of the ill-defined Liassic pteridosperm Dichopteris using an ultrastructural approach. Botanical Journal of the Linnean Society London 149, 313e332. http://dx.doi.org/10.1111/j.1095-8339.2005.00439.x. Tosolini, A.M.P., McLoughlin, S., Wagstaff, B.E., Cantrill, D.J., Gallagher, S.J., 2015. Cheirolepidiacean foliage and pollen from Cretaceous high-latitudes of southeastern Australia. Gondwana Research 27, 960e977. http://dx.doi.org/10.1016/ j.gr.2013.11.008. Upchurch, G., Doyle, J., 1981. 1. Paleoecology of the conifers Frenelopsis and Pseudofrenelopsis (Cheirolepidiaceae) from the Cretaceous Potomac Group of Maryland and Virginia. In: Romans, R.C. (Ed.), Geobotany 2. Plenum Press, New York, pp. 167e202. Villar de Seoane, L., 1998. Comparative study of extant and fossil conifer leaves from the Baquero Formation (Lower Cretaceous), Santa Cruz Province, Argentina. Review of Palaeobotany and Palynology 99, 247e263. PII 0034-6667(97)000511-1. Villar de Seoane, L., 1999. Otozamites ornatus sp. nov., a new bennettitalean leaf species from Patagonia, Argentina. Cretaceous Research 20, 499e506. http:// dx.doi.org/10.1006/cres.1999.0158. Villar de Seoane, L., 2001. Cuticular study of Bennettitales from the Springhill Formation, Lower Cretaceous of Patagonia, Argentina. Cretaceous Research 22, 461e479. http://dx.doi.org/10.1006/cres.2001.0266. Villar de Seoane, L., 2003. Cuticle ultrastructure of the Bennettitales from the  Formation (Early Aptian), Santa Cruz Province, Argentina. Anfiteatro de Tico Review of Palaeobotany and Palynology 127, 59e76. PII: S0034-6667(97)000511. venard, F., Dilcher, D., Barale, G., Mosbrügger, V., Yang, X., Wang, Y., Guignard, G., The Mei, S., 2005. Cuticular anatomy of Sphenobaiera huangii (Ginkgoales) from the Lower Jurassic of Hubei, China. American Journal of Botany 92, 709e721. http:// dx.doi.org/10.3732/ajb.92.4.709.

39

Watson, J., 1988. The Cheirolepidiaceae. In: Beck, C.B. (Ed.), Origin and Evolution of Gymnosperms. Columbia Universiy Press, New York, USA, pp. 382e447. Wattendorff, J., 1992. Cryoultrasections of Agave Americana cuticles extracted for cuticular lipids: time-dependant penetration of KMnO4 and changes of the lamellar structure. Annals of the Sixth Cell wall meeting Nijmegen Netherlands, 68. Wattendorff, J., Holloway, P.J., 1980. Studies on the ultrastructure and histochemistry of plant cuticles: the cuticular membrane of Agave americana L. in situ. Annals of Botany 46, 13e28. http://aob.oxfordjournals.org/. Wattendorff, J., Holloway, P.J., 1982. Studies on the ultrastructure and histochemistry of plant cuticles: isolated cuticular membrane preparations of Agave americana L. and the effects of various extraction procedures. Annals of Botany 49, 769e804. http://aob.oxfordjournals.org/. Wattendorff, J., Holloway, P.J., 1984. Periclinal penetration of potassium permanganate into mature cuticular membranes of Agave and Clivia leaves: new implications for plant cuticle development. Planta 161, 1e11. http://dx.doi.org/ 10.1007/BF00951453. Yang, X.J., Deng, S.H., 2007. Discovery of Pseudofrenelopsis gansuensis from the Lower Cretaceous of Wangqing, Jilin Province, and its Significance in Correlation of Cretaceous Red Beds in China. Acta Geologica Sinica 81, 905e911. venard, F., Wang, Y.D., Barale, G., 2009. Leaf cuticle ulYang, X.J., Guignard, G., The trastructure of Pseudofrenelopsis dalatzensis (Chow et Tsao) Cao ex Zhou (Cheirolepidiaceae) from the Lower Cretaceous Dalazi Formation of Jilin, China. Review of Palaeobotany and Palynology 153, 8e18. http://dx.doi.org/10.1016/ j.revpalbo.2008.06.002. Zhang, C.B., 1986. The middle-late Early Cretaceous strata in Yanji Basin, Jilin Province. Journal of the Changchun College of Geology 2, 15e28 (in Chinese with English abstract). Zhang, W., Zhang, Z.C., Zheng, S.L., 1980. Spermatophyta. In: Shenyang Institute of Geology and Mineral Resources, Chinese Academy of Geological Sciences (Ed), Palaeontological Atlas of Northeast China, Mesozoic and Cenozoic. Geological Publishing House, Beijing, pp. 263e307 (in Chinese). Zhou, Z.Y., 1983. A heterophyllous cheirolepidiaceous conifer from the Cretaceous of East China. Palaeontology 26, 789e811. Zhou, Z.Y., 1995. On some Cretaceous pseudofrenelopsids with a brief review of cheirolepidiaceous conifers in China. Review of Palaeobotany and Palynology 84, 419e438. http://dx.doi.org/10.1016/0034-6667(94)00078-X. Zhou, Z.Y., Guignard, G., 1998. Leaf cuticle ultrastructure of two czekanowskialeans from the Middle Jurassic Yima Formation of Henan, China. Review of Palaeobotany and Palynology 102, 179e187. PII : S0034-6667(97)00063-8. Zhou, Z.Y., Chen, P.J., Li, B.X., Li, W.B., Wen, S.X., Zhang, L.J., Ye, M.N., Liu, Z.S., Li, Z.P., Yang, X.L., 1980. Younger Mesozoic non-marine deposits of the Yanbian area, eastern Jilin. Bulletin of the Nanjing Institute of Geology and Palaeontology, Academia Sinica 1, 1e21 (in Chinese). venard, F., Barale, G., Guignard, G., 2000. A new xeromorphic conifer Zhou, Z.Y., The from the Cretaceous of East China. Palaeontology 43, 561e572. http:// dx.doi.org/10.1111/j.0031-0239.2000.00140. Zodrow, E.L., Mastalerz, M., 2007. Functional groups in a single pteridosperm species: variability and circumscription (Pennsylvanian, Nova Scotia, Canada). International Journal of Coal Geology 70, 313e324. http://dx.doi.org/10.1016/ j.coal.2006.06.010. Zodrow, E.L., Mastalerz, M., 2009. A proposed origin for fossilised Pennsylvanian plant cuticles by pyrite oxidation (Sydney Coalfield, Nova scotia, Canada). Bulletin of Geosciences 84, 227e240. http://dx.doi.org/10.3140/bull.geosc .1094. Zodrow, E.L., D'Angelo, J.A., Mastalerz, M., Cleal, C.J., Keefe, D., 2010. Phytochemistry of the fossilized-cuticle frond Macroneuropteris macrophylla (Pennsylvanian seed fern, Canada). International Journal of Coal Geology 84, 71e82. http:// dx.doi.org/10.1016/j.coal.2010.07.008.

Appendices. Supplementary data Supplementary data related to this article can be found at http://dx.doi.org/10. 1016/j.cretres.2016.11.006.