Abstract

Background and aims – Tendril-climbing vines are widespread within the genus Passiflora, but in P. subg. Astrophea, the basal clade, different growth forms are observed, varying from trees to woody vines, in contrast to the herbaceous climbers of the other subgenera. Considering the shift from woody forms to specialized climbers observed in most derived lineages, this study aims to test the existence of structural characteristics that differentiate stems from P. subg. Astrophea from stems in the other, more derived subgenera. Therefore, we describe the anatomy of the stem of four of the five recognised subgenera of Passiflora, to identify and compare the structural diversity across the genus. Only Tetrapathea, a small subgenus confined to Oceania, was excluded from this study.

Material and methods – Basal branches segments of 12 Passiflora species representing four subgenera (P. subg. Astrophea, P. subg. Decaloba, P. subg. Deidamioides, P. subg. Passiflora) were examined using hand sections and methacrylate microtomy with light microscopy. Samples were stained for qualitative and quantitative analysis. Comparisons between quantitative traits (vessel element diameter and length, percentage of cross-sectional area occupied by secondary xylem and sclerenchyma) among subgenera were performed using Kruskal-Wallis tests followed by Mann-Whitney U post hoc tests.

Key results – Stem cross-sections showed geometric diversity, varying from circular to deeply winged forms. The characters that distinguish P. subg. Astrophea from the other species are the periderm (present/absent), the percentage of woodiness (16–20/>32), and the percentage of sclerenchyma (3/4–7/>13). Periderm with lignified phellem layers was present in all species of P. subg. Astrophea, while absent in the other subgenera. Species of P. subg. Astrophea exhibited more developed vascular cylinders, with substantially higher proportions of secondary xylem (up to ~46% of the stem cross-section) and larger amount of sclerenchyma (up to ~20%) than species of the other subgenera. Overall P. subg. Passiflora had the largest mean vessel diameters, while the species of P. subg. Astrophea which were either lianas or scandent shrubs had the smallest diameters.

Conclusion – The transition from woody forms to specialized climbers in Passiflora is characterized primarily by a reduction of lignified supporting tissues (wood and sclerenchyma) combined to a delay in periderm onset demonstrating a trend towards a lighter stem construction in the tendril-climbers.

Keywords

branches, lianas, mechanical tissues, periderm, wood

Introduction

Passiflora L. (Passifloraceae) is the largest genus of lianas in the Neotropics (Acevedo-Rodríguez et al. 2015; Feuillet and Acevedo-Rodríguez 2020), comprising about 600 species, subdivided into five (Feuillet and MacDougal 2003; Krosnick et al. 2009, 2013) or six subgenera (Mezzonato-Pires et al. 2025). Among them, the monophyly of P. subg. Astrophea (DC.) Mast., P. subg. Decaloba (DC.) Rchb., P. subg. Passiflora L., and P. subg. Tetrapathea (DC.) P.S.Green is supported by molecular phylogenetic analyses (Muschner et al. 2003, 2012; Krosnick et al. 2013).

Passiflora subg. Astrophea comprises about 60 species distributed in South and Central America (Ulmer and MacDougal 2004), with the plains of northern South America being the centre of diversity for the subgenus (Ulmer and MacDougal 2004). Brazil is the country with the highest number of recorded species (31 species), followed by Colombia (20 species), predominantly occurring in the Amazon rainforest (Mezzonato-Pires et al. 2018).

Several molecular phylogenies (Muschner et al. 2003, 2012; Krosnick et al. 2009, 2013) support the basal placement of P. subg. Astrophea in Passiflora, which presents the most atypical morphological traits within the genus, together with a distinct chromosome number (n = 12), which is considered the base number for Passiflora (Hansen et al. 2006). Moreover, P. subg. Astrophea is the most differentiated group within Passiflora, with a notable diversity of growth forms, ranging from shrubs and small trees to lianas, distinguishing it from other Passiflora subgenera in which tendril-climbers are prevalent (Feuillet and MacDougal 2003; Muschner et al. 2012; Mezzonato-Pires et al. 2017). Some authors correlate the basal position of this clade with the diversity of growth forms and distinctive morphology in P. subg. Astrophea (Feuillet and MacDougal 2003; Muschner et al. 2003, 2012; Krosnick et al. 2013).

In addition, species of P. subg. Astrophea have the following diagnostic characters: entire leaf blades (most Passiflora species have deeply lobed leaves), with two small glands at the distal portion of the petiole or at the base of leaf blade base; tendrils absent or modified into spines; small bracts subtending the flower or inflorescence; solitary flowers, in pairs or in inflorescences, white with a yellow corona or pink to purplish, or red to orange hypanthium; erect and membranaceous operculum; unilocular berry-type fruits (Feuillet and MacDougal 2007; Mezzonato-Pires et al. 2018, 2020).

According to phylogenetic analyses, P. subg. Astrophea diverged much earlier from the other lineages, what would explain the more distinct characteristics observed in this subgenus when compared to the others (Muschner et al. 2012). In several phylogenetic analyses, the sister groups of Passiflora (Adenia Forssk., Paropsia Noronha ex Thouars) indicate that the ancestral state would be represented by erect woody plants, with shrubby or tree-like species (Muschner et al. 2003, 2012; Hansen et al. 2006; Hearn 2009; Krosnick et al. 2013), which indicates that this is an ancestral condition in Passiflora, present in P. subg. Astrophea.

Recent taxonomic and morphological studies have contributed to expanding the knowledge about P. subg. Astrophea, which is still an understudied group within the genus. Mezzonato-Pires et al. (2015, 2017) conducted palynological and seed analysis within the subgenus, contributing to the identification of species with delimitation problems as well as to the revision of the subgenus. Mezzonato-Pires et al. (2020) also published a taxonomic review of P. subg. Astrophea in Brazil, in which 31 species and 2 varieties were recognized and included in the sections P. sect. Botryastrophea (Harms) Killip, P. sect. Capreolata J.M.MacDougal & Feuillet, P. sect. Leptopoda Killip ex Feuillet & Cremers, and P. sect. Pseudoastrophea (Harms) Killip.

Some studies in the literature address the morphology and leaf anatomy of Passiflora, such as Farias et al. (2016), who studied the leaf anatomy of some species belonging to P. subg. Decaloba; Pérez and d’Eeckenbrugge (2017), who performed a morphological characterization of some species of the genus Passiflora; Chitwood and Otoni (2017) analysed the morphometric characteristics of the leaves of different species of the genus Passiflora in order to identify the different patterns of leaf vascularization; Mezzonato-Pires et al. (2015, 2017) conducted studies on the importance of pollen and seed morphology in the taxonomy of P. subg. Astrophea.

Despite advances in morphological, palynological, and phylogenetic studies, the anatomy of stems in Passiflora remains relatively unexplored in a comparative and comprehensive manner. Available studies are sporadic and limited to a few species, which makes it difficult to access evolutionary patterns of growth forms in the genus, especially in P. subg. Astrophea. The following studies are mainly descriptive, searching for diagnose features: Beraldo and Kato (2010) conducted a brief characterization of the morphoanatomy of the leaf and stem of P. edulis Sims; Zerpa and Gómez (2014) compared the characteristics of the stems of three Passiflora species (P. edulis, P. quadrangularis L., and P. ligularis Juss.); Chinniah and Thiagarajan (2015) studied the stem and leaf of P. incarnata L.; Pereira-Sühsner et al. (2016) who characterized the stem and leaves of P. caerulea L.; Rajput and Baijnath (2016) analysed the stem anatomy of five Passiflora species aiming to characterize the anatomical variation between species. More recently, Pereira-Sühsner (2023) conducted an extensive study of species from Paraguay in the subgenera P. subg. Decaloba and P. subg. Passiflora, covering genetic resources, anatomy, morphology, and cytogenetics, contributing to the taxonomy, genetic improvement programs, and conservation of the genus Passiflora.

Considering the shift from woody forms to herbaceous vines observed in most derived lineages of Passiflora, we present a descriptive anatomical comparison of the stems of 12 species of four out of five recognised subgenera, with the exception of P. subg. Tetrapathea, a small group restricted to Oceania. This study aims to test the existence of structural characteristics to differentiate stems of P. subg. Astrophea from stems in the other, more derived subgenera, P. subg. Passiflora, P. subg. Decaloba, and P. subg. Deidamioides.

Material and methods

Sample collection

Entire branches of five individuals belonging four species of P. subg. Passiflora were collected in parks and green areas near the University of Paraná and Botanical Garden (Curitiba, Paraná, Southern Brazil). These areas were selected according to the abundance of individuals. The species were identified with the aid of identification keys, and confirmed by a specialist.

The other 19 samples were obtained from herbarium material (voucher barcodes available in Table 1), representing one species of P. subg. Decaloba, one species of P. subg. Deidamioides, and six species of P. subg. Astrophea; for this last subgenus (Astrophea), a pre-selection was carried out to represent woody forms (wood vines scandent/shrub).

Table 1.

Download as

CSV

XLSX

Species studied and their respective herbarium barcode, distribution by state, and growth form. All vouchers were deposited in the Herbarium of the University Federal of Paraná, Department of Botany (UPCB).

Subgenus	Species	UPCB barcode	Distribution in Brazil (Flora e Funga do Brasil 2023)	Growth form
Astrophea	P. cerradensis	0021789	MG, GO, DF, MT	Liana
		0021793
	P. haematostigma	0034448	SC, PR, SP, RJ, MG	Liana
		0034450
	P. longiracemosa	0022129	AM, PA	Scandent shrub, liana
		0034286
	P. mansoi	0021551	Centre-Western region TO, BA, MA	Liana, shrub
		0022154
		0034291
	P. pentagona	0024276	RJ, ES	Scandent shrub, liana
		0024278
		0024275
	P. rhamnifolia	0024523	Southeastern Region, BA	Liana
		0024522
Decaloba	P. capsularis	0022386	South and Southeastern Regions, BA, PI, MS, GO, AM	Vine
		0021693
Deidamioides	P. contracta	0024293	BA, PE, AL, ES	Vine
		0024269
		0024287
Passiflora	P. actinia	0064822	RJ, SP, PR, RS, SC	Vine
	P. alata	0064976	Brazil	Vine
		0065207
	P. caerulea	0065499	South Region, SP, RJ, MG	Vine
	P. edulis	0065300	Brazil	Vine

Among the species of P. subg. Astrophea sampled, three are exclusive lianas (P. cerradensis Sacco, P. haematostigma Mart. ex Mast., and P. rhamnifolia Mast.), while the other three are either lianas or scandent shrubs (P. longiracemosa Ducke, P. mansoi Mast., and P. pentagona Mast.); all the other species studied are tendril-climbers (Table 1).

The branches selected from herbarium samples or collected in the field were approximately similar in length, ranging between 42 and 21 cm in length, and all samples were taken at the most distal part, ensuring that all had a fully grown vascular cambium. Thus, we assured that all samples were collected at the most developed stage for a given size allowing a reliable comparison among the species. By focusing on branches at approximately the same length, we controlled for variables that might influence growth, ensuring that the observed differences were attributable to species-specific traits.

Samples were taken to the Laboratory of Plant Anatomy and Biomechanics, Department of Botany, UFPR (Federal University of Paraná) for anatomical study. Vouchers of the collected species were prepared and deposited in the Herbarium of the University Federal of Paraná, Department of Botany (UPCB) (Table 1).

Preparation of samples for anatomical study

One sample from each individual of each species was fixed in 70% ethanol for subsequent anatomical sectioning. Hand-free cross sections were obtained with a razor blade, approximately 20–30 µm thick. The samples obtained in the herbarium were stored in 50% ethanol for herbarium reversal, which was performed by placing the samples in boiling water (100°C) for 5 minutes, leaving them in the water until they cooled to room temperature so that they could be sectioned. Once the cross sections were obtained, they were stained with Safranin and Astra Blue (1:9) and mounted on semi-permanent slides with glycerine gelatine (Kraus and Arduin 1997).

In addition to the hand-cut sections, samples of all species were also embedded in methacrylate resin for microtomy sections, using a Leica historesin kit. Initially, the samples stored in 50% ethanol were dehydrated in solutions of ethanol of increasing concentrations (70% until 95%) and then subjected to periodic vacuums, during short periods, in a desiccator connected to a vacuum pump for 7 days in a 95% ethanol solution. Subsequently, the samples were transferred to a pre-infiltration solution (activated resin and 95% ethanol, 1:1), for 15 days and stored in a refrigerator. During this period, the samples were subjected to vacuum at least twice a day for 5 minutes. After that, samples were transferred to the activated resin mix for the infiltration (pure resin), and kept during 7 days, subjected to vacuum at least twice a day for 5 minutes and also stored in the refrigerator. After this period, the samples were moved to a solution of resin and polymeriser and placed in silicone moulds. The samples were left to dry at room temperature for 3 days and then stored for subsequent microtomy. Microtomy was performed in a rotary microtome Leica RM 2125 RT (Leica Biosystems, Germany) at 10–15 µm thick. The sections were then stained using Toluidine blue and mounted on semi-permanent slides with glycerinated gelatine (Kraus and Arduin 1997).

Samples of all species were prepared for maceration. The samples were cut into small fragments, and the material was placed in a mixture of hydrogen peroxide 30% and glacial acetic acid (1:1) (Franklin 1945, modified), heated at 60°C for approximately 24 h. Next, the mixture was drained and the material washed with distilled water. Then the material was stained with 1% safranin in 50% ethanol and mounted in slides with a 50% glycerine mount (Kraus and Arduin 1997). The slides were finished with a 50% glycerine mount (Kraus and Arduin 1997).

All histological slides were photographed at CTAF - UFPR (Centre for Advanced Fluorescence Technologies), using an OLYMPUS BX51 microscope for subsequent qualitative and quantitative anatomical description.

Quantitative analyses

Diameter and length of vessel elements of the species studied were measured (20 vessels per specimen), as well as the percentage of cross-sectional area occupied by xylem and lignified tissues (fibres or sclerenchyma), using Optimas software (v.6.5.172, Media Cybernetics, Inc., Rockville, MD, USA). Basic statistics (mean and standard deviation) and comparative analyses using non-parametric Kruskal-Wallis tests followed by Mann-Whitney post hoc tests were performed using PAST 3.0 (Hammer et al. 2001; Suppl. materials 1–3).

Results

Stem anatomical descriptions of the subgenera (Figs 1, 2, 3, 4, 5, 6; Table 2)

Passiflora subg. Astrophea

The cross-sectional outline of the stems is generally elliptical-circular (Fig. 1A–F). In P. cerradensis and P. rhamnifolia, it is regular (Fig. 1A, F), but for the other species, P. haematostigma, P. longiracemosa, P. mansoi, and P. pentagona, they are very wavy, with prominent tips in P. haematostigma and P. pentagona (Fig. 1B, E).

Figure 1.

Overview of cross-section stems of species of Passiflora subg. Astrophea: P. cerradensis (A), P. haematostigma (B), P. longiracemosa (C), P. mansoi (D), P. pentagona (E), and P. rhamnifolia (F). Note: generally elliptical-circular shape; in P. haematostigma (B), P. longiracemosa (C), P. mansoi (D), and P. pentagona (E), they are quite wavy, to varying degrees.

Figure 2.

Detail of the general organisation of stems in cross section of species of P. subg. Astrophea. Overall, the species have the same organisation: note the presence of trichomes; uniseriate epidermis covered by cuticle; in most species, a periderm is being formed underneath the epidermis, composed of several layers of sclerified phellem cell-walls, with the epidermis still visible above; below, layers of collenchyma can be observed; in the phloem there are phenolic contents. A. P. longiracemosa: note the thick cuticle and the well-developed layers of sclerified phellem; epidermis still visible, with long trichomes (arrows). B. P. mansoi: thin cuticle covering epidermis; collenchyma located just below; phenolic contents absent in the phloem; note phellogen below lignified layers of phellem (arrow). C. P. rhamnifolia: thick cuticle covering the epidermis; short trichomes (arrow); well-developed layers of collenchyma; note phenolic contents in the phloem. D. P. haematostigma: thick cuticle covering uniseriate epidermis; long trichomes (arrow); phenolic contents in the cortex and phloem; note phellogen below lignified layers of phellem. E. P. pentagona: note thick cuticle covering epidermis; well-developed phellem, composed of several layers of cells with lignified cell-walls. F. P. cerradensis: thick cuticle covering epidermis; collenchyma formed by layers of cells with thickened walls; phenolic contents in cortex and phloem; well-developed perivascular fibre bundles. Abbreviations: Co - collenchyma; Ct - cuticle; Fb - perivascular fibre bundles; Ph - phloem; Sc- sclerechyma; Xl - xylem.

Figure 3.

A–C. Xylem in cross section and macerated (D) of species of Passiflora subg. Astrophea. A. P. cerradensis: predominantly solitary vessels; rays 1–2 cells wide. B. P. longiracemosa: predominantly solitary vessels; scanty paratracheal parenchyma (arrow); rays 1–5 cells wide. C. P. pentagona: vessels solitary and multiples of two; apotracheal parenchyma in short lines (arrow); rays 1–3 cells wide. D. P. cerradensis: dissociated vessel element with simple perforation plate (arrow) and fibre.

Figure 4.

Detail of the general organisation of stems in cross section of Passiflora capsularis (A–C), P. subg. Decaloba and P. contracta (D–F), P. subg. Deidamioides. A. Conspicuous wings, long trichomes (arrows) and well-developed secondary xylem in front of the wings. B. Detail of one wing, showing the fibres forming a continuum (arrow). C. Detail of the secondary xylem located in front of a wings: vessels solitary and multiple of two; phloem with phenolic contents. D. Circular outline of the stem, with sharp undulations in P. contracta. E. Detail of the epidermis, showing cuticle, phenolic contents in the cortex and phloem, and perivascular fibres with thick cell-walls. F. Detail of the secondary xylem, showing predominantly solitary vessels and lignified uniseriate rays. Abbreviations: Ct - cuticle; Fb - perivascular fibre bundles; Ph - phloem; Xl - Xylem.

Figure 5.

Stems cross-sections of species of Passiflora subg. Passiflora. A. P. actinia, elliptical-circular shape. B. P. alata, quadrangular shape with wings at each angle. C. P. caerulea, circular-pentagonal shape. D. P. edulis, pentagonal shape.

Figure 6.

Detail of the general organisation of stems in cross section and dissociated xylem of species of P. subgen. Passiflora. A. P. actinia: note the single-layered epidermis covered by thin cuticle; layers of collenchyma below; bundles of perivascular fibres with thick lignified cell-walls; vascular cylinder in early secondary growth. B. P. alata: thin cuticle covering epidermis, less developed bundles of perivascular fibres; druses in the cortex and phloem. C. P. caerulea: same tissue organisation as in the previous species; more developed vascular cylinder. D. P. edulis: note the larger vessels than in the previous species in the secondary xylem. E. P. alata: more developed vascular cylinder than in other species; predominantly uniseriate rays. F. P. actinia: dissociated vessel element with simple perforation plate (arrow) and bordered pits. Abbreviations: Co - collenchyma; Ct - cuticle; Fb - perivascular fibre bundles; Ph - phloem; Xl - xylem.

Table 2.

Download as

CSV

XLSX

Qualitative and quantitative traits of the species studied (absent: ab; present: +) for Passiflora subg. Astrophea (CE - P. cerradensis, HA - P. haematostigma, LO - P. longiracemosa, MA - P. mansoi, PE - P. pentagona, RH - P. rhamnifolia), P. subg. Decaloba (CP - P. capsularis), P. subg. Deidamioides (CO - P. contracta), and P. subg. Passiflora (AC - P. actinia, AL - P. alata, CA - P. caerulea, ED - P. edulis).

Astrophea

Decaloba

Deidamioides

Passiflora

CE

HA

LO

MA

PE

RH

CP

CO

AC

AL

CA

ED

Stem cross section

elliptical-circular

elliptical-circular, wavy

elliptical-circular

3-winged

elliptical-circular

quadrangular

pentagonal

Epidermis

Cuticle

thick

thin

thick

thin

thick

very thin

Trichomes

ab

long

short

long

short

ab

Peridermis

Layers lignified phellem

0–3

7–8

10–12

9–10

3–4

0–5(–6)

ab

Cortex

Collenchyma layers

2–3

4–5

2–3

3–4

5–6

1–2

2–3

1–2

3–4

Druses

ab

+

ab

+

ab

+

Phenolic content

ab

+

ab

+

Phloem

Phenolic content

ab

+

ab

+

Secondary xylem

Vessel grouping

1 (2–3)

1 (–2)

1 (2–3)

solitary

1 (2–3)

solitary

1 (2–3)

Ray width

1–2 cells

1–4 cells

1–7 cells

4–5 cells

1–5 cells

1–4 cells

uniseriate

Starch in rays

ab

+

Pith

Starch

ab

+

ab

Druses

ab

+

ab

+

ab

+

ab

+

Lignification

+

ab

+

ab

+

Hollow

+

ab

+

ab

+

Dermal tissue

All species have a uniseriate epidermis, composed of cells with a rounded outline, covered by a thin to thick cuticle (Fig. 2A), except for P. mansoi, which has a thin cuticle (Fig. 2B). In all species, except P. cerradensis, uniseriate trichomes were observed (Fig. 2C–E); P. haematostigma and P. longiracemosa have longer trichomes than the other species (Fig. 2A, D). The stomata were present at the same level as the epidermal cells, which means, the ordinary epidermal cells are aligned with the guard-cells of the stomata, on the same level in the cross-section. All species, except one individual of P. rhamnifolia (Fig. 2C) and P. cerradensis (Fig. 2F), have a periderm already formed underneath the epidermis, composed of a stratified phellem formed by layers of cells with thick lignified cell-walls (Fig. 2A, B, D, E). Passiflora haematostigma has 7 to 8 layers; P. longiracemosa, 10 to 12 layers; P. mansoi, 9 to 10 layers; P. pentagona, 3 to 4 layers; and P. rhamnifolia, 5 to 6 layers. Although most species and individuals have an already formed periderm, the epidermis and cuticle are still visible.

Fundamental tissue

Below the periderm or epidermis, two to six layers of collenchyma are observed; in P. cerradensis, and seven to eight layers in P. rhamnifolia. In both species, the collenchyma is located just below the epidermis, as the periderm is absent (Fig. 2C, F). The presence of phenolic compounds was verified in the cortical parenchyma of P. cerradensis, P. haematostigma, P. longiracemosa, and P. pentagona (Fig. 2A, D, F); druses are present in the cortex of P. haematostigma and P. rhamnifolia. Perivascular fibre bundles are distributed throughout the entire circumference of the cortex in all species, composed of cells with thick, secondary lignified cell-walls (Fig. 2A–F).

Vascular system

In all species, phenolic contents are abundant in the phloem (Fig. 2A–F). Secondary xylem: the species have predominantly solitary vessels (Figs 2E–F, 3A–B), rarely multiples of two (Figs 2E, 3A–B); in P. cerradensis, P. haematostigma, and P. pentagona, rare multiples of three occurred; vessel elements have simple perforation plates in all species (Fig. 3B–C); vessel elements have alternate bordered pits (Fig. 3C); fibres include distinct areolate pits (Fig. 3C); sparse paratracheal parenchyma are present in all species (Figs 2E–F, 3A–B); apotracheal parenchyma in short lines were also observed in P. pentagona (Fig. 3C). In all species, parenchyma rays contain starch (Figs 2F, 3A). In P. cerradensis, the parenchyma rays are uni- to biseriate in cross-section (Fig. 3A); in P. rhamnifolia (Fig. 2C), P. longiracemosa (Fig. 3B) and P. pentagona (Fig. 3C), rays are uni- to multiseriate in cross-section, one to four cells wide.

The pith of all species is parenchymatic (Figs 1A–F, 3C). The species P. cerradensis, P. haematostigma, and P. longiracemosa have phenolic compounds in the pith. Passiflora cerradensis has a fistulous region in the pith. Druses were observed in the pith of all species.

Passiflora subg. Decaloba

The cross-section of the stems of P. capsularis L. is triangular, with three pronounced wings (Fig. 4A).

Dermal tissue

The epidermis is uniseriate, composed of rounded cells with a thin cuticle (Fig. 4B); stomata are located at the same level as the epidermal cells; very long uniseriate trichomes were present (Fig. 4B).

Fundamental tissue

In the cortex, a layer of collenchyma is found below the epidermis (Fig. 4B); followed by two to three layers of parenchyma, containing druses and phenolic compounds; bundles of perivascular fibres formed by cells with very thick secondary walls occur around the entire circumference of the vascular cylinder (Fig. 4A–B); each of the three wings is filled with fibres forming a continuum along the wings (Fig. 4A–B).

Vascular system

The secondary phloem is collapsed, but the presence of phenolic contents can be noted (Fig. 4C). Secondary xylem: opposite to the wings, the secondary xylem is more developed, composed predominantly of solitary vessels, and also multiples of two to three (Fig. 4C); vessel elements have simple perforation plates and alternate bordered pits; scanty paratracheal parenchyma present among axial elements; fibres have bordered pits. The rays are lignified and generally uniseriate in cross section (Fig. 4C).

The parenchymatic pith has a fistulous central region and the presence of phenolic contents (Fig. 4A).

Passiflora subg. Deidamioides

The cross-section of the stem is elliptical-circular, with sharp undulations (Fig. 4D).

Dermal tissue

The epidermis is uniseriate, formed by rectangular rounded cells, covered by a thick cuticle (Fig. 4E), with few short, scattered uniseriate trichomes. The stomata are at the same level as the epidermal cells.

Fundamental tissue

Just below the epidermis, one to two layers of collenchyma are present, followed by layers of parenchyma, with a large number of phenolic substances (Fig. 4E); there are bundles of well-developed perivascular fibres with thickened cell-walls, distributed around the vascular cylinder (Fig. 4E).

Vascular system

The secondary phloem is collapsed, with the presence of many cells containing phenolic substances. In the secondary xylem, the vessels are predominantly solitary, rarely multiples of two (Fig. 4F); vessel elements have simple perforation plates and alternated bordered pits; sparse paratracheal parenchyma is present among axial elements; fibres have bordered pits. The rays are lignified and predominantly uniseriate (Fig. 4F).

The pith is parenchymatic, with phenolic contents (Fig. 4D).

Passiflora subg. Passiflora

The cross-sectional outline of the stems varies among the species: P. actinia Hook. is elliptical-circular, irregular; P. alata Curtis is quadrangular, with pronounced edges (“wings”); P. caerulea is pentagonal (Fig. 5A–D); P. edulis tends to be elliptical-circular, slightly pentagonal.

Dermal tissue

All species have a single-layered epidermis, composed of rounded cells covered by a thin cuticle, and stomata are at the same level as the epidermal cells (Fig. 6A–D).

Fundamental tissue

The cortex has 2 to 4 layers of collenchyma in P. actinia (Fig. 6A); 3 to 4 layers in P. alata and P. caerulea (Fig. 6B–C); 4 to 5 layers in P. edulis (Fig. 6D), located below the epidermis, followed by layers of parenchyma in sequence. All species have bundles of perivascular fibres with very thick cell-walls around the vascular cylinder (Fig. 6A–D). The pronounced edges of P. alata also contain each a bundle of fibres (Fig. 5B). In P. caerulea, the bundles of perivascular fibres are more prominent at the edges (larger area) (Fig. 5C). Druses were found in the cortical parenchyma in all species.

Vascular system

The phloem has large sieve tube elements, predominantly solitary or in multiples of two, with one to three companion cells; sieve tube elements have simple and oblique sieve plates. The secondary xylem is in the early stages of differentiation. The vessel elements are predominantly solitary (Fig. 6A–E), with simple perforation plates and alternate bordered pits (Fig. 6C); sparse paratracheal parenchyma is present among axial elements, and fibres have bordered pits. Parenchymatic rays are uniseriate (Fig. 6B, C, E).

The pith is parenchymatic. All species frequently contain druses and starch in the pith. Fistulous portions are observed in the pith (Fig. 5D).

Quantitative analysis

Diameter and length of vessel elements (Fig. 7; Table 3; Suppl. materials 1, 2)

Passiflora subg. Astrophea

The mean diameter varies from 38.66 μm (21.06–95.63 μm) in P. longiracemosa to 58.98 μm (38.02–90.99 μm) in P. rhamnifolia; P. haematostigma and P. rhamnifolia, which are predominantly lianas, have the largest mean diameters, respectively 53.01 μm (30.29–99.30 μm) and 58.98 μm (38.02–90.99 μm), significant different from the other species (p < 0.05); P. longiracemosa, P. mansoi, and P. pentagona, the ones that can be either lianas or scandent shrubs, have similar mean values for vessel diameter, respectively 42.68 μm (26.42–88.03 μm) and 43.03 μm (22.82–95.37 μm), significantly different from the others (p < 0.05) (Fig. 7A; Table 3; Suppl. material 1). The values of P. pentagona and P. mansoi overlap with those of P. caerulea, which is a liana. The mean length of vessel elements varies from its lowest value in P. haematostigma, with 323.30 μm (157.00–496.10 μm) up to the largest in P. cerradensis, with 543.70 μm (348.80–697.70 μm); there are significant differences detected for P. cerradensis, which differs from the other species with the highest mean value for this variable (Fig. 7B; Table 3).

Figure 7.

Boxplots of vessel diameter (µm) (A) and vessel element length (µm) (B) for all species studied. (P. subg. Astrophea: CE - P. cerradensis; HA - P. haematostigma; LO - P. longiracemosa; MA - P. mansoi; PE - P. pentagona; RH - P. rhamnifolia; P. subg. Decaloba - Dec: CP - P. capsularis; P. subg. Deidamioides - Dei: CO - P. contracta; P. subg. Passiflora: AC - P. actinia; AL - P. alata; CA - P. caerulea; ED - P. edulis).

Table 3.

Download as

CSV

XLSX

Measurements of tangential diameter (DTV) and length of vessel elements (CEV); percentage of cross sectional area occupied by sclerenchyma and fibres - %ES and xylem - %XI) (mean, standard deviation, minimum and maximum values) for all species studied (Passiflora subg. Astrophea: CE - P. cerradensis; HA - P. haematostigma; LO - P. longiracemosa; MA - P. mansoi; PE - P. pentagona; RH - P. rhamnifolia; P. subg. Decaloba: CP - P. capsularis; P. subg. Deidamioides: CO - P. contracta; P. subg. Passiflora: AC - P. actinia; AL - P. alata; CA - P. caerulea; ED - P. edulis).

Astrophea

Decaloba

Deidamioides

Passiflora

CE

HA

LO

MA

PE

RH

CP

CO

AC

AL

CA

ED

DTV (µm)

42.68 ± 16.38 (26.42–88.03)

53.01 ± 18.42 (30.29–99.40)

38.66 ± 18.94 (21.06–95.63)

39.58 ± 11.80 (27.43–94.10)

43.03 ± 18.81 (22.82–95.37)

58.98 ± 17.29 (38.02–90.99)

46.46 ± 15.71 (29.30–98.12)

40.66 ± 23.17 (20.85–123.40)

61.60 ± 23.63 (33.58–106.54)

51.56 ± 29.99 (27.33–155.13)

51.22 ± 20.94 (32.01–110.86)

50.48 ± 20.14 (32.84–150.99)

CEV (µm)

543.70 ± 107.51 (348.82–697.71)

323.31 ± 129.92 (157.01–496.11)

340.61 ± 147.63 (137.82–582.11)

427.21 ± 113.21 (219.93–735.32)

412.02 ± 94.71 (234.13–590.21)

424.3 ± 65.1 (329.71–582.14)

358.31 ± 87.43 (201.01–501.83)

347.93 ± 73.34 (239.01–460.92)

274.81 ± 77.92 (178.43–482.91)

428.11 ± 102.83 (265.41–614.62)

271.42 ± 55.62 (190.51–373.42)

289.31 ± 74.62 (189.81–396.41)

%XI

43 ± 0.2 (42.9–43.5)

27 ± 0.8 (26.4–28.3)

39 ± 0.1 (38.5–38.9)

40 ± 0.1 (3.5–40.1)

46 ± 0.1 (37.0–45.9)

36 ± 0.1 (35.9–46.4)

32 ± 0.4 (31.8–32.9)

17 ± 1.6 (14.7–18.7)

23 ± 14.4 (12.0–48.5)

17 ± 7.5 (11.9–30.0)

21 ± 4.0 (16.7–25.7)

18 ± 8.5 (12.5–32.2)

%ES

5 ± 0.4 (4.8–5.4)

14 ± 2.5 (12.1–15.7)

12 ± 0.1 (11.6–13.9)

20 ± 0.1 (19.6–22.8)

13 ± 0.1 (13.2–15.4)

14 ± 0.1 (13.6–15.6)

5 ± 0.1 (4.6–4.7)

7 ± 0.1 (6.6–6.8)

2 ± 1.0 (1.3–3.8)

2 ± 0.8 (0.8–2.4)

4 ± 0.6 (3.0–4.6)

3 ± 1.0 (1.8–4.0)

Passiflora subg. Passiflora

The mean vessel diameter for the subgenus shows that P. actinia has the largest values, with 61.60 μm (33.58–106.54 μm), similar to P. caerulea (51.22 μm; 32.01–110.86 μm), differing significantly from P. alata and P. edulis (p < 0.05), with mean values of 51.56 μm (27.33–155.13 μm) and 50.48 μm (32.84–50.99 μm), respectively; P. caerulea is similar to all species; P. alata and P. edulis have similar mean values for vessel diameter (Fig. 7A; Table 3; Suppl. material 1).The mean value for the vessel elements length varies from its largest in P. alata, with 428.09 μm (265.39–614.65 μm) to the smallest of 271.40 μm (190.49–373.43 μm) in P. caerulea; the statistical analysis shows significant differences only to P. alata with regard to the other species studied (Fig. 7B; Table 3; Suppl. material 2).

Considering the values in general, P. subg. Passiflora present the highest mean values for vessel diameter. However, there is some overlap between some of the species (P. rhamnifolia and P. haematostigma (lianas) – P. subg. Astrophea has similar mean values to P. actinia (tendril climber) – P. subg. Passiflora) with the largest mean vessel diameters (Fig. 7A; Table 3; Suppl. material 1). Passiflora capsularis (P. subg. Decaloba) has mean values similar to those of the climbing species of both subgenera Astrophea and Passiflora, while P. contracta Vitta (P. subg. Deidamioides) differs from most of the species, with mean values similar to the species of Astrophea which are either lianas or shrubs (smallest values). With regard to the lengths of the vessel elements, although there is less variation between species, there is an overlap in values between the species. P. cerradensis (P. subg. Astrophea) has the highest mean value for this variable (Fig. 7B), differing from all other species (p > 0.05). Passiflora actinia (P. subg. Passiflora), P. haematostigma, and P. longiracemosa (P. subg. Astrophea) have the lowest mean values for this variable (Fig. 7B; Suppl. material 2).

Percentage of cross-sectional area occupied by xylem and sclerenchyma (Fig. 8; Table 3; Suppl. material 3)

Passiflora subg. Astrophea

The percentage of xylem area varies from 27% (26.4–28.3) in P. haematostigma up to 46% (37.0–45.9) in P. pentagona. The values for the other species are, in ascending order, as follows: P. rhamnifolia 36% (37.0–45.9), P. longiracemosa 39% (38.5–38.9), P. mansoi 40% (38.5–40.1), and P. cerradensis 43% (42.9–43.5). As for the percentage of total sclerenchyma area, this varies from 5% (4.8–5.4) in P. cerradensis up to 20% (19.6–22.8) in P. mansoi. The remaining values in ascending order are as follows: P. longiracemosa 12% (11.6–13.9), P. pentagona 13% (13.2–15.4), P. rhamnifolia 14% (13.6–15.6), and P. haematostigma 14% (12.2–15.7) (Fig. 8; Suppl. material 3).

Figure 8.

Boxplots of the percentages of cross-sectional area (% CSA) of sclerenchyma (A) and xylem (B) for all four subgenera. Different letters indicate significant differences (p < 0.05).

Passiflora subg. Decaloba and P. subg. Deidamioides

The percentage of xylem area ranges from 32% (31.8–32.9) to 17% (14.7–18.7) for P. capsularis and P. contracta, respectively. Regarding the percentage of total sclerenchyma area, we have 4.6% (4.6–4.7) and 6.7% (6.6–6.8) for P. capsularis and P. contracta, respectively. Given the low n, there are no statistically significant differences detected between these two taxa, although they differed from P. subg. Astrophea with regard to the percentage of xylem area for P. subg. Deidamioides, and the percentage of sclerenchyma area for P. subg. Decaloba; both differ from P. subg. Passiflora for percentage of sclerenchyma area, but not for percentage of xylem area (Fig. 8; Suppl. material 3).

Passiflora subg. Passiflora

The percentage of xylem area ranges from 17% (11.9–30.0) to 18% (12.5–32.2) for P. alata and P. edulis, respectively, up to 21% (16.7–25.7) in P. caerulea and 23% (12.0–48.5) in P. actinia. In relation to the percentage of total sclerenchyma area, P. alata has 2% (0.8–2.4) and P. caerulea 4% (3.0–4.6); the other species, P. actinia 2% (1.3–3.8) and P. edulis 3% (1.8–4.0). Comparing the four subgenera of Passiflora, there are significant differences between P. subg. Passiflora and P. subg. Astrophea regarding the percentage of xylem area (p < 0.01) and also regarding the percentage of sclerenchyma area (p < 0.01). No significant difference is found between the other subgenera (Fig. 8; Suppl. material 3).

Discussion

Passifloraceae is a Neotropical family in which the climbing habit predominates, and Passiflora is among the ten genera richest in climbing species (Sperotto et al. 2023). Here, we discuss adaptive aspects of the woody habit, predominant in P. subg. Astrophea, and of the herbaceous climbing habit, common in the other Passiflora subgenera, identifying the modifications that took place in this woody to herbaceous transition within the genus.

The secondary xylem of vines and lianas frequently shows numerous structural modifications making the stem more flexible and voluble, assisting anchorage to the support and growth toward the canopy while avoiding severe structural damage during this process caused by the twisting of the stem associated with the climbing process (Ewers et al. 1991; Rowe and Speck 2005; Angyalossy et al. 2015; Isnard and Feild 2015; Cunha Neto 2023). Among these modifications typical of climbers are large-diameter vessels combined with narrow-diameter vessels (called vessel dimorphism), an increase in axial and radial parenchyma concomitant to a reduction in fibres, and vascular variations associated with non-cylindrical stems geometries (Angyalossy et al. 2015; Leme et al. 2021; Cunha Neto 2023). These traits compose the “lianescent vascular syndrome”, associated with higher hydraulic conductivity and greater stem flexibility at the cost of reduced investment in supporting tissues (Angyalossy et al. 2015). In this study, we observed some of these traits, such as different stem geometries, large and narrow vessels, and a high percentage of parenchyma in the stem, mainly in the form of a well-developed pith, although vascular variants were not detected in the branches studied.

Cross-sections of stems in the group studied show a wide diversity of geometries, ranging from circular to deeply winged, in P. subg. Decaloba and P. subg. Passiflora, while species of P. subg. Astrophea and P. subg. Deidamioides have only an elliptic-circular outline. Pereira-Sühsner (2023) also observed a variety of geometries in her study of the same subgenera (P. subg. Passiflora and P. subg. Decaloba), and it is worth noting that she analysed younger stems cross sections, with very little secondary growth, meaning that these different geometries are prior to the arise of secondary growth. In self-supporting trees and shrubs, the circular cross-section of stems develops from the typical activity of a vascular cambium that produces secondary phloem outward and secondary xylem inward, in the same proportion in all directions (Carlquist 1991; Spicer and Groover 2010). In climbing plants, the deeply modified geometries of the stem in cross-section arise, in many cases, from vascular variants, either from the primary or secondary vascular system (Isnard and Silk 2009; Luna-Márquez et al. 2021; Cunha Neto 2023; Kaçamak et al. 2025), or at least from some differences in cambial activity which produces unequal deposition of xylem and phloem (Rajput and Baijnath 2016) or even by irregular division of cells outside the vascular system, causing profound changes in the cross-sectional geometry of the stem (Soffiatti and Rowe 2020). In any case, in climbers, these geometrical changes will contribute to stem flexibility and survival from injuries, optimising the growth of the species throughout the climbing process (Carlquist 1991; Ewers et al. 1991; Speck and Rowe 2003; Isnard and Silk 2009; Soffiatti and Rowe 2020). In the Passiflora species studied, these geometries appeared before the emergence of any cambial variants and could be a product of the primary vascular system configuration or different patterns of cortical cell divisions, influenced by mechanical constraints (Bassel et al. 2014).

Yang et al. (2025) also reported different geometries and cambial variants in adult stems of several species of Passiflora but notably, the presence of cambial variants was not always associated with non-circular geometries and not all species had cambial variants (Yang et al. 2025). Rajput and Baijnath (2016) also recorded some differences in adult stem geometries, such as lobbed stem for P. foetida L., and a cambial variant type “shallow phloem wedges” in adult stems of P. edulis and P. vesicaria L. (all P. subg. Passiflora). The authors attributed both the differentiated geometry and the cambial variation observed to the unequal activity of vascular cambium, which produces the vascular components of phloem and xylem in an irregular manner. In our study, cambial variants were not detected, either in woody or herbaceous plants. The literature shows that it can vary in Passifloraceae and it does not occur in all species (Ayensu and Stern 1964; Rajput and Baijnath 2016; Pereira-Sühsner 2023; Yang et al. 2025). Besides, as noted by Ayensu and Stern (1964), for some species of Passiflora, the cambial variants only appear at very old stages of development, which are not usually included in herbarium collections, which mainly consist of young branches.

Considering diameters of vessel elements, the Passiflora species studied presented relatively small diameters when compared to those reported for other Passiflora (Ayensu and Stern 1964; Rajput and Baijnath 2016; Yang et al. 2025), although not too different from the values obtained for P. coccinea Aubl. (Ewers et al. 1990). These differences observed in relation to the information in the literature are somewhat anticipated, as we used samples collected from branches, which typically have smaller vessel diameters than those of the main stems, as vessel diameter scales with stem size and tapers from base to top (Anfodillo et al. 2006; Olson and Rosell 2013; Olson et al. 2014). It is worth noting that, from species of P. subg. Astrophea analysed, the ones growing as scandent shrubs or lianas had the smallest mean vessel diameters (P. longiracemosa, P. mansoi, and P. pentagona; Fig. 7A; Table 3), while species growing only as climbers had larger diameters, especially those of P. subg. Passiflora, and also P. rhamnifolia and P. haematostigma, both lianas from P. subg. Astrophea having also among the largest vessel diameters. Because lianas possess extremely long stems but often smaller stem cross-sectional area than self-supporting species, increased conduits diameter can compensate for smaller transverse stem area (Ewers et al. 1990; Leme et al. 2021; Kaçamak et al. 2025). The other two species representing P. subg. Decaloba, characterised by smaller vines, and P. subg. Deidamioides, which can be more woody, exhibited values that that were close to those of P. subg. Passiflora and P. subg. Astrophea, respectively. Functionally, lianas and vines have large-diameter vessels (Ewers and Fisher 1989; Carlquist 2013; Rosell and Olson 2014; van der Sande et al. 2019), a pattern that becomes clearer in comparative studies within the same genus where transitions between lianescent and erect self-supporting habits occur (Leme et al. 2021). By contrast, vessel element lengths recorded here do not differ from those reported for other Passiflora species (Rajput and Baijnath 2016; Yang et al. 2025).

The subgenera differed in the amount of mechanical (lignified) tissues, reflected both in the quantity of secondary xylem and in the abundance of fibres or sclerenchyma, allied to a thick and well-developed peridermis. Passiflora subg. Astrophea stands out by having a more developed vascular cylinder and a larger amount of mechanical tissue than the species belonging to P. subg. Decaloba, P. subg. Deidamioides, and P. subg. Passiflora. Studies comparing taxa with lianescent and erect habits report increased percentages of cross-sectional area occupied by lignified tissues in the erect taxa, which affords them greater support (Leme et al. 2021). Woodiness is ancestral in angiosperms (Doyle 2012), and changes towards herbaceousness occurred multiple times across different lineages (Klimeš et al. 2022; Luo et al. 2023). Herbaceous forms are considered phylogenetically derived, although many shifts back to woodiness is also observed (Carlquist 2013; Lens et al. 2013; Zizka et al. 2022). In Passiflora, the evolutionary trend towards specialized lighter architectures is evident, and the basal lineage, represented by P. subg. Astrophea, exhibits remarkable woody robustness, characterised by a more developed secondary xylem cylinder and an early onset of a thick periderm.

Conclusion

Our findings demonstrate that the woody Passiflora subg. Astrophea is characterized by well-developed vascular cylinders, the occurrence of great amount of lignified tissue (sclerenchyma) besides a well-developed peridermis. These characteristics confer on these species a more robust and rigid structure than that observed in herbaceous vines of Passiflora. The underlying anatomical traits that accompanies the shift from woody to herbaceous are the reduction in mechanical (lignified) tissues (wood and sclerenchyma) and delayed periderm formation in more recent lineages, which may have facilitated the widespread emergence of lightweight herbaceous climbers. The robustness of P. subg. Astrophea contrasts with the persistent epidermis and a more parenchymatic structure present in the tendril-climbers clades. Future studies with larger samples from all subgenera, including adult stems, will certainly contribute to more in-depth discussions about growth forms diversification patterns and inherent anatomical changes in Passiflora.

Acknowledgements

I would like to thank the Coordination for the Improvement of Higher Education Personnel (CAPES) for financial support provided through the Demanda Social Program Scholarship (CAPES-DS).

References

Acevedo-Rodríguez P, et al. (2015) Lianas and Climbing Plants of the Neotropics. Smithsonian National Museum of Natural History. https://naturalhistory.si.edu/research/botany/research/lianas-and-climbing-plants-neotropics [accessed 01.09.2025]

Anfodillo T, Carraro V, Carrer M, Fior C, Rossi S (2006) Convergent tapering of xylem conduits in different woody species. New Phytologist 169: 279–290. https://doi.org/10.1111/j.1469-8137.2005.01587.x

Angyalossy V, Pace MR, Lima AC (2015) Liana anatomy: a broad perspective on structural evolution of the vascular system. In: Schnitzer SA, Bongers F, Burnham RJ, Putz FE (Eds) Ecology of Lianas. John Wiley and Sons, Chichester, 253–287. https://doi.org/10.1002/9781118392409.ch19

Ayensu ES, Stern WL (1964) Systematic anatomy and ontogeny of the stem in Passifloraceae. Contributions from the United States National Herbarium 34: 45–73.

Bassel GW, Stamm P, Mosca G, Barbier de Reuille P, Gibbs DJ, Winter R, Janka A, Holdsworth MJ, Smith RS (2014) Mechanical constraints imposed by 3D cellular geometry and arrangement modulate growth patterns in the Arabidopsis embryo. PNAS 111(23): 8685–8690. https://doi.org/10.1073/pnas.1404616111

Beraldo J, Kato ETM (2010) Morfoanatomia de folhas e caules de Passiflora edulis Sims (Passifloraceae). Revista Brasileira de Farmacognosia 20(2): 233–239. https://doi.org/10.1590/S0102-695X2010000200016

Carlquist S (1991) Anatomy of vine and liana stems: a review and synthesis. In: Putz FE, Mooney HA (Eds) The Biology of Vines. Cambridge University Press, Cambridge, 53–71. https://doi.org/10.1017/CBO9780511897658.004

Carlquist S (2013) More woodiness/less woodiness: evolutionary avenues, ontogenetic mechanisms. International Journal of Plant Sciences 174(7): 964–991. https://doi.org/10.1086/670400

Chinniah V, Thiagarajan VRK (2015) Anatomical investigation on the leaves and stem of Passiflora incarnata (Passifloraceae). International Journal of Research in Ayurveda and Pharmacy 6(4): 537–543. https://doi.org/10.7897/2277-4343.064101

Chitwood DH, Otoni WC (2017) Morphometric analysis of Passiflora leaves: the relationship between landmarks of the vasculature and elliptical Fourier descriptors of the blade. GigaScience 6(1): giw008. https://doi.org/10.1093/gigascience/giw008

Cunha Neto IL (2023) Vascular variants in seed plants — a developmental perspective. AoB Plants 15: plad036. https://doi.org/10.1093/aobpla/plad036

Doyle JA (2012) Molecular and fossil evidence on the origin of angiosperms. Annual Review of Earth and Planetary Sciences 40: 301–326. https://doi.org/10.1146/annurev-earth-042711-105313

Ewers FW, Fisher JB (1989) Variation in vessel length and diameter in stems of six tropical and subtropical lianas. American Journal of Botany 76(10): 1452–1459. https://doi.org/10.1002/j.1537-2197.1989.tb15126.x

Ewers FW, Fisher JB, Chiu S-T (1990) A survey of vessel dimensions in stems of tropical lianas and other growth forms. Oecologia 84: 544–552. https://doi.org/10.1007/BF00328172

Ewers FW, Fisher FB, Fichtner K (1991) Water flux and xylem structure in vines. In: Putz FE, Mooney HA (Eds) The Biology of Vines. Cambridge University Press, Cambridge, 119–152. https://doi.org/10.1017/CBO9780511897658.007

Farias V, Maranho LT, Mushner VC, Soffiatti P (2016) Leaf anatomy of Passiflora subg. Decaloba (Passifloraceae): taxonomic implications. Rodriguésia 67(1): 29–43. https://doi.org/10.1590/2175-7860201667103

Feuillet C, MacDougal JM (2003) A new infrageneric classification of Passiflora L. Passiflora: The Journal and Newsletter of the Passiflora Society International 13(2): 34–38.

Feuillet C, MacDougal JM (2007) Passifloraceae. Passifloraceae Juss. ex Roussel, Fl. Calvados, ed. 2:334 (1806), ‘Passifloreae’, nom. cons.. In: Kubitzki K (Ed.) Families and Genera of Vascular Plants. Volume 9. Springer-Verlag, Berlin, 270–280. https://doi.org/10.1007/978-3-540-32219-1_35

Feuillet C, Acevedo-Rodríguez P (2020) Passifloraceae. In: Acevedo-Rodríguez P, et al. (Eds) Lianas and Climbing Plants of the Neotropics. Smithsonian National Museum of Natural History. https://naturalhistory.si.edu/sites/default/files/media/file/passifloraceae-final-small.pdf [accessed 27.09.2023]

Flora e Funga do Brasil (2023) Jardim Botânico do Rio de Janeiro. https://floradobrasil.jbrj.gov.br/consulta/#CondicaoTaxonCP [accessed 27.09.2023]

Franklin GL (1945) Preparation of thin sections of synthetic resins and wood-resin composites, and a new macerating method for wood. Nature 155: 51. https://doi.org/10.1038/155051A0

Hansen AK, Gilbert LE, Simpson BB, Downie SR, Cervi AC, Jansen RK (2006) Phylogenetic relationships and chromosome number evolution in Passiflora. Systematic Botany 31: 138–150. https://doi.org/10.1600/036364406775971769

Hammer Ø, Harper DAT, Ryan PD (2001) PAST: Paleontological Statistics Software Package for Education and Data Analysis. Palaeontologia Electronica 4(1): art. 4.

Hearn D (2009) Descriptive anatomy and evolutionary patterns of anatomical diversification in Adenia (Passifloraceae). Aliso 27: 13–38. https://doi.org/10.5642/aliso.20092701.03

Isnard S, Silk WK (2009) Moving with climbing plants from Charles Darwin’s time into the 21st century. American Journal of Botany 96: 1205–1221. https://doi.org/10.3732/ajb.0900045

Isnard S, Feild TS (2015) The evolution of angiosperm lianescence: a perspective from xylem structure-function. In: Schnitzer SA, Bongers F, Burnham RJ, Putz FE (Eds) Ecology of Lianas. John Wiley and Sons, Chichester, 221–238. https://doi.org/10.1002/9781118392409.ch17

Kaçamak B, Réjou-Méchain M, Grolleau M, Loumeto J-J, Loubota Panzou GJ, Rowe N (2025) Linking anatomical diversity with functional life strategies in a liana community of central Africa. New Phytologist 248: 1225–1244. https://doi.org/10.1111/nph.70413

Klimeš A, Šímová I, Zizka A, Antonelli A, Herben T (2022) The ecological drivers of growth form evolution in flowering plants. Journal of Ecology 110: 1525–1536. https://doi.org/10.1111/1365-2745.13888

Kraus JE, Arduin M (1997) Manual Básico de Métodos em Morfologia Vegetal. EDUR, Seropédica, Rio de Janeiro.

Krosnick SE, Ford AJ, Freudenstein JV (2009) Taxonomic revision of Passiflora subgenus Tetrapathea including the monotypic genera Hollrungia and Tetrapathea (Passifloraceae), and a new species of Passiflora. Systematic Botany 34(2): 375–385. https://doi.org/10.1600/036364409788606343

Krosnick SE, Porter-Utley KE, MacDougal JM, Jørgensen PM, McDade LA (2013) New insights into the evolution of Passiflora subgenus Decaloba (Passifloraceae): phylogenetic relationships and morphological synapomorphies. Systematic Botany 38(3): 692–713. https://doi.org/10.1600/036364413X670359

Leme CLD, Pace MR, Angyalossy V (2021) The “Lianescent Vascular Syndrome” statistically supported in a comparative study of trees and lianas of Fabaceae subfamily Papilionoideae. Botanical Journal of the Linnean Society 197(1): 25–34. https://doi.org/10.1093/botlinnean/boab015

Lens F, Davin N, Smets E, del Arco MJ (2013) Insular woodiness on the Canary Islands: a remarkable case of convergent evolution. International Journal of Plant Sciences 174(7): 998–1013. https://doi.org/10.1086/670259

Luna-Márquez L, Sharber WV, Whitlock BA, Pace, MR (2021) Ontogeny, anatomical structure and function of lobed stems in the evolution of the climbing growth form in Malvaceae (Byttneria Loefl.). Annals of Botany 128(7): 859–874. https://doi.org/10.1093/aob/mcab105

Luo A, Xu X, Liu Y, Li Y, Su X, Li Y, Lyu T, Dimitrov D, Larjavaara M, Peng S, , Chen Y, Wang Q, Zimmermann NE, Pellissier L, Schmid B, Wang Z (2023) Spatio-temporal patterns in the woodiness of flowering plants. Global Ecology and Biogeography 32(3): 384–396. https://doi.org/10.1111/geb.13627

Mezzonato-Pires AC, Milward-de-Azevedo MA, Mendonça CBF, Gonçalves-Esteves VP (2015) Pollen morphology and detailed sexine of Passiflora L. subgenus Astrophea (Passifloraceae). Plant Systematics and Evolution 301: 2189–2202. https://doi.org/10.1007/s00606-015-1223-0

Mezzonato-Pires AC, Mendonça CBF, Milward-de-Azevedo MA, Gonçalves-Esteves VP (2017) The systematic value of pollen morphology of Passiflora subg. Astrophea (Passifloraceae). Phytotaxa 298(1): 1–19. https://doi.org/10.11646/phytotaxa.298.1.1

Mezzonato-Pires AC, Milward-de-Azevedo MA, Mendonça CBF, Gonçalves-Esteves VP (2018) Taxonomy, palynology and distribution notes of seven new species of Passiflora L. (Passifloraceae s.s.) newly recorded from Brazil. PhytoKeys 95: 1–14. https://doi.org/10.3897/phytokeys.95.22342

Mezzonato-Pires AC, Milward-de-Azevedo M, Mendonça C, Mendonça F, Gonçalves-Esteves V (2020) A taxonomic revision of Passiflora subg. Astrophea (Passifloraceae sensu stricto) in Brazil. Phytotaxa 473: 1–60. https://doi.org/10.11646/phytotaxa.473.1.1

Mezzonato-Pires AC, Bernacci LC, Teixeira HGSF, Cruz PF, Pacheco G, Mendonça CBF, Gonçalves-Esteves V (2025) Palynological characterization and taxonomical delimitation of the subgenera within Passiflora L. (Passifloraceae: Malpighiales), the largest genus of lianas in the Neotropics European Journal of Taxonomy 981: 239–264. https://doi.org/10.5852/ejt.2025.981.2829

Muschner VC, Lorenz AP, Cervi AC, Bonatto SL, Souza-Chies TT, Salzano FM, Freitas LB (2003) A first molecular phylogenetic analysis of Passiflora (Passifloraceae). American Journal of Botany 90: 1229–1238. https://doi.org/10.3732/ajb.90.8.1229

Muschner VC, Zamberlan PM, Bonatto SL, Freitas LB (2012) Phylogeny, biogeography and divergence times in Passiflora (Passifloraceae). Genetics and Molecular Biology 35(4 Suppl.): 1036–1043. https://doi.org/10.1590/S1415-47572012000600019

Olson ME, Rosell JA (2013) Vessel diameter–stem diameter scaling across woody angiosperms and the ecological causes of xylem vessel diameter variation. New Phytologist 197: 1204–1213. https://doi.org/10.1111/nph.12097

Olson ME, Anfodillo T, Rosell JA, Petit G, Crivellaro A, Isnard S, León- Gómez C, Alvarado-Cárdenas LO, Castorena M (2014) Universal hydraulics of the flowering plants: vessel diameter scales with stem length across angiosperm lineages, habits and climates Ecology Letters 17: 988–997. https://doi.org/10.1111/ele.12302

Pereira-Sühsner CD (2023) Caracterización de recursos genéticos del género Passiflora L. (Passifloraceae) de Paraguay. PhD Thesis, Universidad Nacional de Misiones, Argentina. https://hdl.handle.net/20.500.12219/5570 [accessed 25.03.2026]

Pereira-Sühsner CD, Ferrucci MS, Honfi AI (2016) Morfo-anatomía foliar y caulinar de Passiflora caerulea L. (Passifloraceae). Steviana 8(2): 75–83. https://doi.org/10.56152/StevianaFacenV8N2A1_2016

Pérez JO, d’Eeckenbrugge GC (2017) Morphological characterization in the genus Passiflora L.: an approach to understanding its complex variability. Plant Systematics and Evolution 303(4): 531–558. https://doi.org/10.1007/s00606-017-1390-2

Rajput KS, Baijnath H (2016) Stem anatomy of some species of Passiflora (Passifloraceae). IAWA Journal 37(3): 431–443. https://doi.org/10.1163/22941932-20160145

Rosell JA, Olson ME (2014) Do lianas really have wide vessels? Vessel diameter–stem length scaling in non-self-supporting plants. Perspectives in Plant Ecology, Evolution and Systematics 16(6): 288–295. https://doi.org/10.1016/j.ppees.2014.08.001

Rowe N, Speck T (2005) Plant growth forms: an ecological and evolutionary perspective. New Phytologist 166: 61–72. https://doi.org/10.1111/j.1469-8137.2004.01309.x

Speck T, Rowe NP (2003) Modelling primary and secondary growth processes in plants: a summary of the methodology and new data from an early lignophyte. Philosophical Transactions of the Royal Society B: Biological Sciences 358(1437): 1473–1485. https://doi.org/10.1098/rstb.2003.1347

Sperotto P, Roque N, Acevedo-Rodríguez P, Vasconcelos T (2023) Climbing mechanisms and the diversification of neotropical climbing plants across time and space. New Phytologist 240(4): 1561–1573. https://doi.org/10.1111/nph.19093

Spicer R, Groover A (2010) Evolution of development of vascular cambia and secondary growth. New Phytologist 186(3): 577–592. https://doi.org/10.1111/j.1469-8137.2010.03236.x

Soffiatti P, Rowe NP (2020) Mechanical innovations of a climbing cactus: functional insights for a new generation of growing robots. Frontiers in Robotics and AI 7: 64. https://doi.org/10.3389/frobt.2020.00064

Ulmer T, MacDougal JM (2004) Passiflora: Passionflowers of the World. Timber Press, Portland.

van der Sande MT, Poorter L, Schnitzer SA, Engelbrecht BMJ, Markesteijn L (2019) The hydraulic efficiency–safety trade-off differs between lianas and trees. Ecology 100(5): e02666. https://doi.org/10.1002/ecy.2666

Yang S-Z, Chen P-H, Chen J-J (2025) Structure and development of the secondary xylem in ten species of Passiflora from Taiwan. Taiwania 70(2): 252–261. https://doi.org/10.6165/tai.2025.70.252

Zerpa ZAM, Gómez OAV (2014) Morphoanatomy of the stem of three species of the genus Passiflora L. (Passifloraceae). Agrobiología 26: 114–120.

Zizka A, Onstein RE, Rozzi R, Weigelt P, Kreft H, Steinbauer MJ, Bruelheide H, Lens F (2022) The evolution of insular woodiness. PNAS 119(37): e2208629119. https://doi.org/10.1073/pnas.2208629119

Supplementary materials

Supplementary material 1

Comparative non parametric Kruskall-Wallis test and post hoc Mann-Whitney test of vessel tangential diameters for all studied species (Passiflora subg. Astrophea: CE - P. cerradensis; HA - P. haematostigma; LO - P. longiracemosa; MA - P. mansoi; PE - P. pentagona; RH - P. rhamnifolia; P. subg. Decaloba: CP - P. capsularis; P. subg. Deidamioides: CO - P. contracta; P. subg. Passiflora: AC - P. actinia; AL - P. alata; CA - P. caerulea; ED - P. edulis).

Download file (1.15 kb)

Supplementary material 2

Comparative non parametric Kruskall-Wallis test and post hoc Mann-Whitney test of vessel length for all studied species (Passiflora subg. Astrophea: CE - P. cerradensis; HA - P. haematostigma; LO - P. longiracemosa; MA - P. mansoi; PE - P. pentagona; RH - P. rhamnifolia; P. subg. Decaloba: CP - P. capsularis; P. subg. Deidamioides: CO - P. contracta; P. subg. Passiflora: AC - P. actinia; AL - P. alata; CA - P. caerulea; ED - P. edulis).

Download file (1.09 kb)

Supplementary material 3

Comparative non parametric Kruskall-Wallis test and post hoc Mann-Whitney test for the percentage of cross-sectional area of sclerenchyma and xylem for the four subgenera (Astrophea, Decaloba, Deidamioides, and Passiflora).

Download file (408.00 bytes)