ARTICLE IN PRESS
Flora 204 (2009) 146–156
www.elsevier.de/flora
Fruit anatomy of species of Solanum sect. Acanthophora (Solanaceae)
Franco E. Chiarini, Gloria E. Barboza
Instituto Multidisciplinario de Biologı´a Vegetal (CONICET-UNC), C.C. 495, 5000 Córdoba, Argentina
Received 11 October 2007; accepted 7 January 2008
Abstract
The fruits of 10 species of Solanum sect. Acanthophora were studied. Cross and/or longitudinal and/or tangential
microtome sections, stained mostly with astra blue/basic fuchsin, were made for microscopic examination. Three
different kinds of cells were found in the epidermis, immediately below which a hypodermis, consisting in any of four
types of structures, was always found. The mesocarp exhibits two histologically differentiated zones, an external one
(formed by normal or spongy parenchyma, according to the species), and an internal one, commonly juicy, and with
proliferations among the seeds. The diagnostic value of all these structures is assessed. Morpho-anatomical
information is used to define fruit types beyond the berry, traditionally described for Solanum, and the probable
dispersal syndrome related to them is discussed. Fruit similarities are slightly noticeable in several cases, while
differences may be the result of their sexual system – which affects specially the size – and their histology, which is
related to the dispersal syndrome. The comparison of our data with previous molecular phylogeny of the section
suggests that a significant morphological variation is not associated with significant DNA sequence changes.
r 2008 Elsevier GmbH. All rights reserved.
Keywords: Anatomy; Epidermis; Fruit; Phylogeny; Solanum; subgen. Leptostemonum
Introduction
Solanum subgen. Leptostemonum (Dunal) Bitter is a
large group (ca. 450 species) within Solanum, comprising
almost one-third of the genus (Bohs, 2005; Levin et al.,
2005; Weese and Bohs, 2007). The species of this
subgenus have a worldwide distribution, with the
greatest species richness in South America, Africa,
and Australia. Several of them are economically
important, for example S. melongena L. (aubergine
or brinjal eggplant), S. aethiopicum L. (scarlet eggplant),
S. quitoense Lam. (naranjilla or lulo) or S. sessiliflorum
Dunal (cocona). One of the main characteristics defining
Corresponding author. Tel./fax: +54 351 4331056.
E-mail address: chiarini@imbiv.unc.edu.ar (F.E. Chiarini).
0367-2530/$ - see front matter r 2008 Elsevier GmbH. All rights reserved.
doi:10.1016/j.flora.2008.01.010
the subgen. Leptostemonum is the presence of sharp
epidermal prickles on stems and leaves in all, except for
a few, taxa. Most members of the subgenus have stellate
hairs and attenuate anthers with small terminal pores.
Within subgen. Leptostemonum, sect. Acanthophora
Dunal is distinguishable and easily grouped due to the
existence of simple hairs on the upper leaf surface (Levin
et al., 2005; Nee, 1991; Whalen, 1984). This section is
monophyletic (Levin et al., 2005), and its diversity
center is located in eastern Brazil. It includes about
20 herbs and small shrubs mainly, adapted to disturbed
areas and secondary open forests. The importance of
sect. Acanthophora lies in the fact that some of its species
are considered invasive weeds. For instance, S. viarum
Dunal (‘‘tropical soda apple’’), is naturalized in USA,
Africa, and Asia (Bryson and Byrd, 1994; Welman,
ARTICLE IN PRESS
F.E. Chiarini, G.E. Barboza / Flora 204 (2009) 146–156
2003), while S. palinacanthum Dunal (‘‘pocote’’, ‘‘joá
bagudo’’); invades roadsides and crop fields in Argentina and Brazil (Auler Mentz and Oliveira, 2004; Nee,
1991). Conversely, other species of the section are
cultivated for their ornamental value (e.g. S. mammosum
L., ‘‘apple of Sodom’’, ‘‘cow’s udder’’ or ‘‘nipple fruit’’;
and S. capsicoides All., ‘‘cockroach berry’’).
Species of sect. Acanthophora exhibit great variation
in their fruit morphology. Some of them (S. mammosum,
S. palinacanthum) produce large fruits, reaching up to
5.5 cm in diameter and with a spongy mesocarp. Other
species produce smaller fruits with a thinner spongy
layer (S. aculeatissimum Jacq., S. viarum) while others
are said to be irregularly dehiscent (S. capsicoides,
S. platense Dieckmann) (Levin et al., 2005). Finally,
some species (e.g. S. atropurpureum Schrank) develop
small, red, juicy fruits. Despite this variability, no
anatomical studies have been performed in fruits
of Acanthophora species, except for S. mammosum
(Miller, 1969), so all the discussion about evolution of
fruit traits has been superficial. Moreover, the study of
fruit anatomy in the entire subgenus has been broadly
neglected. Thus, aiming to achieve a comprehensive
knowledge of the fruit anatomy within the species of
Leptostemonum, and to evaluate probable taxonomical
and evolutionary incidences, we started a carpological
analysis in members of different sections of the subgenus.
As a result, some groups, such as the Argentine
representatives of sect. Melongena (sensu Nee, 1999),
revealed their vast heterogeneity concerning fruit traits
(Chiarini and Barboza, 2007), supporting the paraphyly
of this section (Levin et al., 2006). We are now presenting
results in species of sect. Acanthophora, as a contribution
to clarify the fruit evolutionary pattern of this group, and
to understand the relationship between structure and
function.
Materials and methods
Ten wild species of Solanum sect. Acanthophora were
analysed. The following are the voucher data of the
studied material:
Solanum aculeatissimum Jacq. BRAZIL. Santa Catarina. Mun. Monte Castelo, 281380 1200 S, 501150 0600 W,
23-II-2006, Barboza et al. 1617 (CORD).
Solanum aenictum C. V. Morton. ARGENTINA.
Corrientes. Santo Tomé, 7-I-2002, Chiarini 536
(CORD).
Solanum atropurpureum Schrank. ARGENTINA.
Corrientes. Santo Tomé, 7-I-2002, Chiarini 531
(CORD); Goya, Arroyo Guazú, 291500 2600 S, 591
240 2400 W, 3-XII-2002, Barboza et al. 355 (CORD).
Solanum capsicoides All. ARGENTINA. Corrientes.
Ituzaingó, Isla Apipé Grande, 4-XII-2002, Barboza
147
et al. 394 (CORD); BRAZIL. Santa Catarina.
Trombudo Central, 24-XI-2003, Mentz et al. 274
(CORD, ICN); Garuva, 24-II-2006, Barboza et al.
1623 (CORD); São Paulo. From Salesópolis to
Paraibuna, SP 077, km 108.5, 25-II-2006, Barboza
et al. 1641 (CORD); Rio de Janeiro. Floresta de
Tijuca, 15-VII-2003, Barboza et al. s.n. (GUA 48406).
Solanum incarceratum Ruiz et Pav. BRAZIL. São
Paulo. Estrada Itú-Jundiaı́, 251150 3600 S, 471150 3400 W,
29-VI-03, Marcondes et al. s.n. (CORD 1028).
Solanum mammosum L. ECUADOR. Napo. XII1995, Hunziker s.n. (CORD 1024).
Solanum palinacanthum Dunal. ARGENTINA. Córdoba. Capital, 1-VI-2001, Chiarini 465 (CORD);
Tucumán. Famaillá, 1-IV-1977, Hunziker et al.
23081 (CORD); PARAGUAY. Caaguazú. 251220
4300 S, 561000 4200 W, 12-XII-2002, Barboza et al. 495
(CORD).
Solanum platense Dieckmann. ARGENTINA. Misiones. San Ignacio, ayo. Macaco, 7-XII-2002, Barboza et al. 441 (CORD).
Solanum tenuispinum Rusby. ARGENTINA. Salta.
Santa Victoria, Baritú, 2-X-2001, Barboza et al. 292
(CORD); Rosario de Lerma, 241580 1500 S, 651350
3700 W, 6-III-2002, Negritto et al. 293 (CORD);
Catamarca. Andalgalá, Rı́o Chacras, 271230 010 S,
651590 2900 W, 23-II-2003, Barboza et al. 629 (CORD).
Solanum viarum Dunal. ARGENTINA. Corrientes.
Santo Tomé, 6-I-2002, Chiarini 533, 537 and 538
(CORD); Ituzaingó, 14-V-2004, Barboza et al. 1006
(CORD); Misiones. Gral. Manuel Belgrano, 251420 2400 S,
54150 4800 W, 30-XI-2003 Barboza et al. 819 (CORD).
For microscopic examination, whole or cut up ripe
fruits were preserved in a formaldehyde–acetic acid–
ethanol mixture, then dehydrated in a 50–100% ethanol
series, and embedded in Paramats resin. Cross and/or
longitudinal and/or tangential microtome sections
10–12 mm thick were stained mostly with a 1% astra
blue solution in a 1% water/basic fuchsin solution in 501
ethanol. Astra blue stains cell wall polysaccharides such
as cellulose and pectins, while basic fuchsin shows
affinity for lignified, suberizied or cutinized walls, i.e.,
structures embedded in phenolic substances (Kraus
et al., 1998). Basic fuchsin also stains chloroplasts
and nucleic acids. In some cases, additional sections
were stained with a 0.05% cresyl blue solution in water
(Pérez and Tomassi, 2002).
The specimens were visualized using a Zeiss Axiophot
microscope. The images were captured with a digital
camera assembled to the microscope.
In order to enrich the discussion, data about the
sexual system of the species studied were drawn from
the literature, and missing information was found via
our observations in the field.
ARTICLE IN PRESS
148
F.E. Chiarini, G.E. Barboza / Flora 204 (2009) 146–156
Results
Most of the taxa have spherical fruits, with the
exception of S. tenuispinum, which has ovoid fruits, and
S. mammosum, which has an anomalous fruit, 3–5-lobed
at base.
According to Knapp (2002) the fruits of Solanum sect.
Acanthophora studied here are berries in a conventional
sense. In the Spjut’s (1994) system, and according to
our observations, fruits are either berries, i.e. simple
fruits with an indehiscent pericarp – containing many
seeds embedded in a solid fleshy mass, supported by an
epicarp that is less than 2 mm thick – or a carcerulus,
which is a fruit resembling the true berries, but with an
aerial space between the seeds and the pericarp when the
fruit is completely ripe, as is the case with peppers
(Capsicum spp.) (Fig. 1A–D, Table 1).
In general, fruits of the investigated species are small to
medium-sized, varying from 1.35 cm (S. incarceratum) to
4 cm (S. mammosum) in diameter. When mature, fruits
are of a single colour: red or orange-red (S. capsicoides
and S. atropurpureum), yellow (most of the species), or
greenish yellow (S. incarceratum and S. platense).
Seeds belong to two types: winged, i.e. a strongly
flattened seed, with the seed coat forming a prominent
wing, 0.8–2 mm wide (e.g. S. platense; S. capsicoides,
Fig. 2A; S. atropurpureum, Fig. 2B), or bulky, i.e. a
slightly flattened or non-flattened lenticular seed, whose
seed coat margin is not thickened at all (e.g. S. viarum,
Fig. 2C; S. mammosum, Fig. 2D).
The pericarp comprises three clearly distinguishable
zones: the exocarp, the mesocarp, and the endocarp.
S. aenictum (Fig. 3B). Ventilation clefts and stomata are
lacking in all cases.
The epidermis consists of different kinds of cells
according to the species, as follows:
(1) A unistrate layer of small, isodiametric cells, with
dense content and cellulosic walls, e.g. S. platense
(Fig. 4E) or S. aculeatissimum (Fig. 4D).
(2) A unistrate layer of relatively large, bottle-shaped
cells, here designed as lageniform cells, with a
noticeable nucleus, dense content and simple pits.
These cells can be weakly lignified, but not enough
so as to be considered true sclereids, e.g. S. aenictum
(Fig. 3B).
(3) A layer of brachysclereids (S. capsicoides, Fig. 4A
and B).
Immediately below the epidermis, a hypodermis is
differentiated, consisting in any of the four following
kinds of structures:
The cuticle is highly variable and usually thick
(especially in S. incarceratum, Fig. 4F), and it can be
smooth (e.g. S. atropurpureum Fig. 3D), undulate (e.g.
S. capsicoides Fig. 4A), or grooved (e. g. S. aculeatissimum, Fig. 4D). In all species, cuticular wedges are
present among the epidermal cells, these wedges being
notably deep in S. atropurpureum (Fig. 3D) and in
(a) Compressed parenchymatous cells, with dense content, arranged in 1 or 2 layers (e.g. S. incarceratum,
Fig. 4F).
(b) Rounded parenchymatous cells containing rhombohedric crystals of calcium oxalate, in a single
continuous layer or alternating with non-crystalliferous cells, below which some collenchymatous
layers are found (e.g. S. viarum, Fig. 3A).
(c) Sclerenchymatous cells, i.e., true brachysclereids,
naturally orange coloured, followed by 1–2, or
several true collenchymatous layers (e.g. S. capsicoides, Fig. 4A and B). Transitional forms between
sclereids and parenchyma cells develop along the
boundary between the exocarp and the mesocarp.
(d) Several layers of very thickened and lignified cell
walls, taller than wide, which could be considered
a sclerified collenchyma. In S. mammosum and
S. palinacanthum (Fig. 3E), the cells of these layers
adopt a strange ‘‘amoeboid’’ appearance due to their
irregular outline; a normal collenchyma is developed
underneath.
Fig. 1. Senescent fruits of the carcerulus type in Solanum sect.
Acanthophora species. A longitudinal cut shows the disposition
of the seeds: (A) S. aenictum, (B) S. viarum, (C) S. capsicoides,
and (D) S. platense. The bar represents 1 cm.
The epidermis and the hypodermis constitute a unit,
the exocarp, which generally has layers that gradually
decrease their degree of lignification from the outside to
the inside of the fruit.
Usually, when the fruit is immature, the cell layers
located below the epidermis (or below the crystalliferous
layer or layer of fibres, when present) have chloroplasts
and chromoplasts. In mature fruits, the chloroplasts
disappear and the cells become compressed. A collenchyma is always present, within which the number of
layers and the degree of lignification vary according to
the species.
Exocarp
Table 1.
Macroscopic and anatomical fruit features of the 10 species of Solanum sect. Acanthophora studied
Species
Fruit type
Sexual
system
Seeds type
Fruit
colour
Exocarp
Mesocarp
Hypodermis
External zone
Internal zone
Unistrate,
isodiametric cells
with dense content
(Fig. 4D)
Unistrate,
lageniform cells
(Fig. 3B)
12 layers of normal to
some sclerified
collenchyma (Fig. 4D)
20 layers of spongy
tissue over the veins and
20 underneath (Fig. 4C)
Tiny proliferations
among the seeds
(Fig. 4C)
12 layers of normal
collenchyma
6065 layers of spongy
tissue. Starch (in unripe
fruits)
Absent
Unistrate, short
lageniform and
trapezoidal cells
(Fig. 3D)
12 layers of
brachysclereids,
naturally orange
coloured (Fig. 4A
and B)
Unistrate,
isodiametric cells
with dense content
(Fig. 3D)
Unistrate, tall
lageniform cells
12 layer of normal
collenchyma (Fig. 3D)
A few layers of normal
to somewhat spongy
tissue
Tiny proliferations
Sclerenchyma or
sclerified collenchyma,
followed by normal
collenchyma (Fig. 4A)
20 layers of spongy
tissue over the veins and
20 underneath
Absent
12 layers of
compressed cells with
dense content (Fig. 3D)
5 layers of spongy tissue
over the veins and 5
underneath
Some tiny
proliferations
Special sclerified
collenchyma (amoeboid
appearance)
Absent
23 layers of special
sclerified collenchyma
(amoeboid appearance)
(Fig. 3E), followed by
normal collenchyma
12 collenchymatous
layers
40 layers of spongy
tissue over the veins and
60 underneath, with the
first layers dense, and
the following loose
Spongy tissue, 2030
layers over the veins
and 20 underneath (Fig.
3F)
S. aculeatissimum
C
A
Bulky
Yellow
Grooved
S. aenictum
C (Fig.
1A)
A
Bulky
Yellow
S. atropurpureum
B
H
Winged
(Fig. 2B)
Red
S. capsicoides
C (Fig.
1C)
A
Winged
(Fig. 2A)
Red
Grooved
with deep
cuticular
wedges
Smooth,
with deep
cuticular
wedges
Undulate
S. incarceratum
B
H
Winged
Yellow
greenish
Undulate
S. mammosum
C
A
Bulky
(Fig. 2D)
Yellow
Smooth,
with
wedges
S. palinacanthum
C
A
Bulky
Yellow
Grooved
Unistrate, tall
lageniform cells
(Fig. 3E)
S. platense
C (Fig.
1D)
A
Winged
Yellow
greenish
Grooved
S. tenuispinum
B
H
Winged
Yellow
Smooth
S. viarum
C (Fig.
1B)
A
Bulky
(Fig. 2C)
Yellow
Grooved
Unistrate,
isodiametric cells
with dense content
(Fig. 4E)
Unistrate,
lageniform cells
(Fig. 3C)
Unistrate,
isodiametric cells
with dense content
(Fig. 3A)
1 layer of living cells
containing a crystal,
followed by 23 layers
of laminar collenchyma
(Fig. 3A)
10 layers of spongy over
the veins and 20
underneath (Fig. 4E)
Absent
67 layers of spongy
tissue over the veins and
56 underneath
1820 layers of spongy
tissue over the veins and
20 underneath
Tiny proliferations
(Fig. 3C)
Absent
149
Abbreviations: A: andromonoecious species; B: berry; C: carcerulus; H: hermaphrodite species.
12 collenchymatous
layers (Fig. 3C)
Absent
ARTICLE IN PRESS
Epidermis
F.E. Chiarini, G.E. Barboza / Flora 204 (2009) 146–156
Cuticle
ARTICLE IN PRESS
150
F.E. Chiarini, G.E. Barboza / Flora 204 (2009) 146–156
Fig. 2. Seed types of Solanum sect. Acanthophora species: (A and B) winged seeds and (C and D) bulky seeds. (A) S. capsicoides, (B)
S. atropurpureum, (C) S. viarum, and (D) S. mammosum. The bar represents 1 cm. All pictures at the same scale.
Mesocarp
The number of layers of this tissue gives the pericarp
its thickness. The higher the number of mesocarp layers,
the thicker the pericarp. Fruits with a thick pericarp
contain usually more than 10 layers. The mesocarp
consists in two histologically differentiated zones: an
external one (immediately below the hypodermis,
evidenced with astra blue), and an internal one,
identified with basic fuchsin. The external zone may
adopt two forms, according to the kind of cell
arrangement:
(1) A spongy parenchymatic tissue, resembling the
albedo of the hesperidium, formed by big, very
vacuolated or almost empty, loosely connected cells,
with large intercellular spaces (e.g. S. palinacanthum,
Fig. 3F). The cells increase their size towards the
endocarp, and the cell walls get lose and undulated.
At maturity, the mesocarp is not in direct contact
with the seeds, therefore the fruit belongs to
the carcerulus type (Fig. 1A–D). The number
of parenchymatous layers varies according to the
species; for example, 40 layers over the veins and
60 beneath them are found in S. mammosum, while
only 10 layers over the veins and 8 underneath are
found in S. tenuispinum (Fig. 3C).
(2) A non-specialized or normal parenchyma, consisting
of regular, vacuolated, medium-sized cells with small
intercellular spaces (S. atropurpureum, Fig. 3D).
Regardless of the cellular arrangement of the mesocarp external zone, it is always followed by an internal
zone, which is commonly juicy, and develops proliferations among the seeds. The cells are large, with dense
content filled with grana, which disorganize and release
their own content to the locules and produce a mucilagelike substance that surrounds the seeds in the ripe fruit.
In several species this mucilaginous content turns black
at air contact, perhaps due to its phenolic or saponinic
nature.
The thickness of each zone varies notably according
to the species. For instance, in species with a thick
spongy external zone, the internal zone has few layers
or is altogether absent (S. palinacanthum, S. platense,
Fig. 4E) while in other species it is well developed
(S. atropurpureum, S. aculeatissimum, Fig. 4C).
ARTICLE IN PRESS
F.E. Chiarini, G.E. Barboza / Flora 204 (2009) 146–156
151
Fig. 3. Photomicrographs of fruit anatomy in Solanum sect. Acanthophora species, showing different structures: (A) S. viarum, (B) S.
aenictum, (C) S. tenuispinum, (D) S. atropurpureum, (E) S. palinacanthum, and (F) detail of the spongy mesocarp in S.
palinacanthum. The bars represent 50 mm in A–E, and 100 mm in F. Abbreviations: ac ¼ amoeboid cells; bs ¼ brachysclereids;
cc ¼ cells containing a crystal; co ¼ collenchyma; cw ¼ cuticular wedges; is ¼ isodiametric cells; lc ¼ lageniform cells;
sc ¼ sclerified collenchyma.
Stone cells or sclerosomes, widely present in many
sections of Solanum and in related genera, are absent
in sect. Acanthophora, at least in the species analysed
here.
Endocarp
Finally, no specific particularities were observed in the
endocarp. This layer, which is very difficult to observe
due to its delicate structure, is uniseriate and lacks
stomata in all cases.
Discussion
Trait assessment
Regarding the fruit itself, there is a slight similarity
among the different species, but several constraints
beyond phylogeny may be at play to result in this
manner. The external resemblance concerns size and
colour, while histological characters define each taxon.
For instance, the lageniform cells are present in some
species with large yellow fruits (e.g. S. aenictum and
S. palinacanthum), but not in other yellow-coloured ones
(e.g. S. aculeatissimum).
Although colour is a variable feature, yellow fruits
predominate. Colour has been attributed to the dispersal
syndrome (Van der Pijl, 1982), and brightly coloured
fruits would be more attractive to birds (Edmonds and
Chweya, 1997). This should be the case in red-fruited
species, but S. capsicoides or S. atropurpureum do not fit
this hypothesis very well. In fact, Solanum capsicoides
fruits have other features, such as their large size and
their spongy mesocarp, which make them unsuitable for
birds. In addition, both S. capsicoides and S. atropurpureum have winged seeds, presumably for wind
dispersion.
Cipollini et al. (2002) found a significant connection
among yellow colour, seed number, and mass. They also
described a pattern in which large, yellow fruits with a
low nutrient, toxic and highly dry matter content, differ
from small, red, and black fruits with a high nutrient,
non-toxic, and watery content. Nevertheless, colour
cannot be easily linked with any morpho-anatomical
structure in the species studied here, since fruits with the
same colour show different tissue types.
ARTICLE IN PRESS
152
F.E. Chiarini, G.E. Barboza / Flora 204 (2009) 146–156
Fig. 4. Photomicrographs of fruit anatomy in Solanum sect. Acanthophora species, showing different structures: (A) S. capsicoides in
cross-section, bar 50 mm; (B) S. capsicoides in tangential section, showing the sclereids, bar ¼ 50 mm; (C) pericarp of S.
aculeatissimum, bar ¼ 0.5 mm; (D) exocarp of S. aculeatissimum, bar ¼ 100 mm; (E) S. platense, bar ¼ 100 mm; and (F) S.
incarceratum, bar ¼ 100 mm. Abbreviations: bs ¼ brachysclereids; is ¼ isodiametric cells; p ¼ proliferations among the seeds;
sc ¼ sclerified collenchyma; st ¼ spongy tissue.
The cuticle has no diagnostic value, since it varies
from one species to another. This situation has already
been noticed in other ‘‘spiny Solanum’’ (Chiarini and
Barboza, 2007; Dottori and Cosa, 1999, 2003).
The epidermis has shown a significant diversification.
For instance, some species present isodiametric cells
with dense content, and others present lageniform cells
(which is apparently a distinctive feature of the section)
while sclereids are unique in S. capsicoides. These
different types of epidermal cells could be related to
fruit size and dehiscence type. Lageniform cells of large
fruits would be the result of a tendency for the epidermal
cells to become lignified and tall in response to external
constraints, while epidermal sclereids in S. capsicoides
could be related to the cracking of the fruit at the
senescent stage.
The presence of a hypodermis, mainly constituted by
collenchyma, is constant in the majority of the analysed
species. However, there is a specific variation in the
number of layers and the degree of lignification. The
collenchymatous hypodermis is common in fruits with
a thick outer skin, which is the case of many berries and
drupes, such as some species of Ribes, Berberis, and
Paris (Roth, 1977) and even berries of some members
of Solanaceae (Filippa and Bernardello, 1992; Valencia,
1985), and more precisely in Solanum (Dottori and
Cosa, 1999; Klemt, 1907; Roth, 1977). The function of
the hypodermal cells possibly comprises providing
mechanical support, or, in some cases, participating in
the dehiscence mechanism (Dyki et al., 1997; Klemt,
1907). The collenchyma is followed by a thick-walled
parenchyma, but it is difficult to draw a line between the
two tissue types, which is a well-known fact (Roth,
1977).
Some hypodermal cells, whose walls are impregnated
with lignin, resemble the outline of a true collenchyma.
Layers with such features are called here ‘‘sclerified
collenchyma’’. The amoeboid cells in the hypodermis of
S. palinacanthum and S. mammosum are another case
of a modified collenchyma. The sclerified collenchyma,
as well as the strong thickened walls of the amoeboid
cells, may be structures that make the fruits harder and
more resistant to deformation, and perhaps are a
defence against phytophagous insects. Indeed, fruit
features are usually interpreted in relation to vertebrate
dispersion and consumption, while insect and microbial
ARTICLE IN PRESS
F.E. Chiarini, G.E. Barboza / Flora 204 (2009) 146–156
attacks, which are all the more important, have become
neglected (Tewksbury, 2002).
Regarding the mesocarp, the presence of a spongy
parenchyma is very obvious in some species. This tissue,
characterized by large intercellular spaces and cells
that change their shape from rounded to elliptical,
elongate, and even stellate, was accurately described
in S. mammosum (Miller, 1969) with the name of
‘‘aerenchyma’’. Something similar occurs in the albedo
of the orange, where parenchyma cells develop arms in
different directions (Roth, 1977). We found that, apart
from S. mammosum, the spongy fruited Solanum
analysed here also present the so-called ‘‘aerenchyma’’.
The spongy tissue is not a very useful character to
distinguish sections, since it is present in several species
of sect. Acanthophora, but also in a species of section
Melongena (Chiarini and Barboza, 2007).
Regarding the pulp of the fleshy or juicy fruited
Solanum, both the placenta and, especially the pericarp,
contribute to form the pulp, a fact already reported for
Physalis peruviana (Valencia, 1985) and other Solaneae
(Filippa and Bernardello, 1992; Garcin, 1888; Murray,
1945). Instead, in Solanum lycopersicum (sub nom.
Lycopersicon esculentum) only the placentas are responsible for the formation of the pulp (Murray, 1945; Roth,
1977), which is very peculiar. The disorganization of the
inner mesocarp and the endocarp occurs in the same
way in the fleshy and juicy fruits analysed, as in
P. peruviana (Valencia, 1985).
The presence of a layer with a single prismatic crystal
in each cell, observed here in S. viarum, has been found
for several species of sect. Melongena (Chiarini and
Barboza, 2007; Dottori and Cosa, 2003). Deposition of
calcium oxalate may have evolved as a primary
mechanism for controlling the excess of calcium in
many plants, providing multiple benefits to different
organs (e.g. Franceschi and Horner, 1980; Sakai et al.,
1972; Thurston, 1976; Webb, 1999). Nevertheless, the
function of crystals in fruits remains unexplained to
date.
Nee (1991) has discovered that different kinds of seeds
can be found in species of the same section, probably in
response to environmental pressures. On their part,
Levin et al. (2005) discussed that, within sect. Acanthophora, species that share the same type of seed do not
form a natural group. In addition, we demonstrate that
one type of seed may not be related to any pericarp
characteristic, since winged seeds, for instance, are
present in species with spongy mesocarp (S. capsicoides,
S. platense) as well as in species with another type of
mesocarp (S. atropurpureum).
The most remarkable differences among fruits arise
from the sexual system (Table 1). In fact, the relationship between andromonoecy and fruit size detected by
Whalen (1984) and Whalen and Costich (1986) was also
found here, since most of the analysed species have few
153
hermaphrodite flowers per inflorescence (strong andromonoecy) and grow large fruits. Within these species,
size tends to increase through a proliferation of
parenchymatic cells in the mesocarp. The increased size
could be a way to accommodate more seeds per fruit
(Nee, 1986), which is accompanied by changes in
placentation patterns. In contrast, the weakly or nonandromonoecious species produce relatively smaller
fruits. Outstandingly, large fruits show the lengthened
lageniform cells, perhaps as a way to avoid deformation,
while small fruits (S. atropurpureum, S. tenuispinum and
S. incarceratum) present shorter epidermal cells.
Nevertheless, the relationship between fruit size or
sexual system and the different tissue types appears not
to be rigorously direct. For example, the spongy tissue is
mainly found in many-seeded, large fruits, in species
which are usually invasive herbs growing in ruderal
environments, in open, sunny areas of the Chaco region,
as S. viarum or S. aenictum (Matesevach, 2002; Whalen,
1984). Thus, the occurrence of these fruits is explainable
through ecological and reproductive reasons rather than
through phylogenetic relationships. However, the same
spongy tissue is also found in S. tenuispinum, which has
small fruits, and a different habit and habitat (this shrub
is typical of rainforests, the Yungas of northern
Argentina and southern Bolivia). Although the anatomical pattern of S. tenuispinum is similar to that of
species of drier regions, the spongy tissue is less thick.
The presence of such tissue in a non-andromonoecious
species could be a rudimentary feature shared with its
andromonoecious relatives.
Dispersion
Usually, fruits are classified into different dispersal
syndromes according to their morphological characters.
Van der Pijl’s (1982) criterion is usually followed, but
direct observation of the dispersion is seldom possible.
As a consequence, the fruits or seeds are assigned to a
dispersal syndrome on the basis of speculations, which
leads to puzzling discussions, as Levin et al. (2005)
pointed out. As mentioned before, there is an emphasis
on the interpretation of fruit traits as adaptations to the
interactions of vertebrates (Tewksbury, 2002), but our
results indicate that other points of view should be
regarded.
Most of the examined species produce a spongy fruit
that never dehisces without the action of external forces,
and whose placentas and seeds are not in contact at the
senescent stage, i.e. a carcerulus according to Spjut
(1994). For some species with such a kind of fruit
(S. viarum, S. aculeatissimum, S. mammosum), Cipollini
et al. (2002) suggested dispersion by large animals, like
mammals. Nee (1991) proposed a ‘‘shaker’’ mechanism
as the dispersal syndrome for the spongy fruits of two
ARTICLE IN PRESS
154
F.E. Chiarini, G.E. Barboza / Flora 204 (2009) 146–156
other species, S. capsicoides and S. platense, since he
observed a spontaneous rupture of the pericarp, which
we could not corroborate. Indeed, no special features
were detected in the fruit of S. platense, although
epidermal sclereids in S. capsicoides could be related to
the mentioned mechanism. The fact that the five
mentioned species have potentially poisonous fruits
(Cipollini and Levey, 1997), in addition to their spongy
structure, turns dispersion by vertebrates little plausible.
We suppose that all of these fruits with spongy tissue are
adapted to dispersion by drain water after a rainstorm,
the spongy tissue being an adaptation to flotation, as it
has been previously suggested (Bryson and Byrd, 1994;
Levin et al., 200; Nee, 1979, 1991).
The particularities of the fruit of S. atropurpureum
should be mentioned. This is a unilocular fruit with a
unique combination of red colour, a normal or not spongy
and slightly juicy mesocarp (in contrast to its relatives), and
winged seeds. Nee (1979, 1991) proposed epizoochory by
birds as a dispersal mechanism for this species and for
S. acerifolium Sendtn., a species with similar fruits, since
their seeds do not resist the passage through the digestive
system of birds. The intriguing combination of winged
seeds and a showy, juicy fruit, can be thus solved.
Section Acanthophora could have several dispersal
syndromes: hydrochory, epizoochory, endozoochory
and the ‘‘shaker’’ mechanism. Thus, it has been proven
that, despite the fact that their monophyly has already
been confirmed by molecular studies (Levin et al., 2005,
2006), there are diverging pathways in fruit evolution
within sect. Acanthophora.
Fruit traits and phylogeny
The similarities among fruits may not be synapomorphies, but they could be attributed to convergence in
reproductive strategies. Cipollini et al. (2002), in a study
that foregrounds phytochemical aspects, states that
there is no significant correlation among the fruit types
they distinguished and the phylogenetic lineages in
Solanum. For these authors, fruit typology may be the
consequence of physiological constraints, holding an
independent evolution of the different dispersal syndromes. Regarding morpho-anatomy, our data lead to
similar conclusions.
As mentioned before, an important factor influencing
fruit features in Solanum subgen. Leptostemonum could
be the andromonoecy. This sexual system is a convergent phenomenon, a homoplasic character that varies
secondarily (Whalen, 1984) and appears and disappears
within the clades or natural groups independently (Levin
et al., 2006; Whalen and Costich, 1986). The big, manyseeded fruits would then be a collateral effect of the
andromonoecy strategy, not a cause, since this sexual
system is independent of the fruit anatomy or the seed
type: there are andromonoecious species with big, dry or
juicy fruits, with winged or bulky seeds. Andromonoecious species tend to concentrate a high number of
ovules in few hermaphrodite flowers (Bertin, 1982;
Symon, 1979; Whalen and Costich, 1986); in consequence, they produce large fruits. Thus, the sexual
system may have played a very important role in the
Leptostemonum evolutionary lineages, since it affects not
only the reproductive biology of its species (reproductive
success, fitness, sexual allocation, etc.), but it also has an
influence on the morpho-anatomy of fruits and their
dispersal mechanism.
Phytochemical aspects are also very important in the
evolution of Solanum fruits. For instance, in Solanum
sect. Solanum, the concentration of steroidal glycoalkaloids decreases dramatically as the fruit matures (Carle,
1981; Cipollini et al., 2002), this being perhaps a
mechanism to deter seed predators while the fruit is
growing without deterring seed dispersers that feed on
mature fruits. Nevertheless, in subgenus Leptostemonum, the concentration of toxic compounds in ripe fruits
remains high (Cipollini and Levey, 1997). These authors
suggest that the compounds perform an antifungal
function, but that they also deter both predators and
dispersers. This may be the case of S. capsicoides,
S. viarum, and S. mammosum (Cipollini et al., 2002).
The results of our work suggest that phylogeny is not
the main factor influencing the different carpologic
characters found in subgen. Leptostemonum. We propose that, in many cases, a significant morphological
variation is not associated with significant DNA
sequence changes. Moreover, fruit traits seem to
respond quickly to selection constraints on the dispersal
syndromes. Molecular studies showed that some species,
like S. capsicoides and S. viarum, are closely related as
regards phylogeny (Levin et al., 2005), but differ notably
regarding fruit traits. We tend to think that similarities
and differences can be the result of the sexual system
and of ecological and physiological conditions.
Acknowledgements
The authors thank Consejo Nacional de Investicaciones
Cientı́ficas y Técnicas (CONICET, Argentina), Agencia
Córdoba Ciencia S. E. (Argentina), SECyT (UNC,
Argentina ), Coordenação de Aperfeiçoamento de Pessoal
de Nı́vel Superior (CAPES, Brazil) and Myndel Botanica
Foundation (Buenos Aires, Argentina) for financial
support.
References
Auler Mentz, L., Oliveira, P.L., 2004. Solanum (Solanaceae) na
região sul do Brasil. Pesqui. Bot. 54, 1–327.
ARTICLE IN PRESS
F.E. Chiarini, G.E. Barboza / Flora 204 (2009) 146–156
Bertin, R.I., 1982. The evolution and maintenance of
andromonoecy. Evol. Theory 6, 25–32.
Bohs, L., 2005. Major clades in Solanum based on ndhF
sequence data. In: Hollowell, V., Keating, R., Lewis, W.,
Croat, T. (Eds.), A Festschrift for William D’Arcy,
Monogr. Syst. Bot. Missouri Bot. Gard, vol. 104. Missouri
Botanical Garden Press, St. Louis, Missouri, pp. 27–50.
Bryson, C.T., Byrd, J.D., 1994. Solanum viarum (Solanaceae),
new to Mississippi. Sida 16, 382–385.
Carle, R., 1981. Investigations on the content of steroidal
alkaloids and sapogenins within Solanum sect. Solanum
( ¼ sect. Morella) (Solanaceae). Plant Syst. Evol. 138,
61–71.
Chiarini, F., Barboza, G., 2007. Anatomical study of different
fruit types in Argentine species of Solanum subgen.
Leptostemonum (Solanaceae). An. Jard. Bot. Madrid 164,
165–175.
Cipollini, M.L., Levey, D.J., 1997. Why are some fruits toxic?
Glycoalkaloids in Solanum and fruit choice by vertebrates.
Ecology 78, 782–798.
Cipollini, M.L., Bohs, L., Mink, K., Paulk, E., BöhningGaese, K., 2002. Secondary metabolites of ripe fleshy
fruits. Ecology and phylogeny in genus Solanum. In: Levey,
D.J., Silva, W.R., Galetti, M. (Eds.), Seed Dispersal
and Frugivory. Ecology, Evolution and Conservation.
CAB International Publishing, Wallingford, Oxforshire,
pp. 111–128.
Dottori, N., Cosa, M.T., 1999. Anatomı́a y ontogenia de fruto
y semilla en Solanum hieronymi (Solanaceae). Kurtziana 27,
293–302.
Dottori, N., Cosa, M.T., 2003. Desarrollo del fruto y semilla
en Solanum euacanthum (Solanaceae). Kurtziana 30, 17–25.
Dyki, B., Jankiewicz, L.S., Staniaszek, M., 1997. Anatomy and
surface micromorphology of Tomatillo fruit (Physalis
ixocarpa Brot.). Acta Soc. Bot. Pol. 66, 21–27.
Edmonds, J.M., Chweya, J.A., 1997. Black Nightshades.
Solanum nigrum L. and Related Species. IPGRI, Gatersleben, Germany.
Filippa, E.M., Bernardello, L.M., 1992. Estructura y desarrollo del fruto y semilla en especies de Athenaea, Aureliana
y Capsicum (Solaneae, Solanaceae). Darwiniana 31,
137–150.
Franceschi, V.R., Horner, H.T., 1980. Calcium oxalate
crystals in plants. Bot. Rev. 46, 361–428.
Garcin, M.A.G., 1888. Sur le fruit des Solanées. J. Bot. 2,
108–115.
Klemt, F., 1907. Über den Bau und die Entwicklung einiger
Solanaceenfrüchte. Inaugural-Diss., Berlin.
Knapp, S., 2002. Tobacco to tomatoes, a phylogenetic
perspective on fruit diversity in the Solanaceae. J. Exp.
Bot. 53, 2001–2022.
Kraus, J.E., de Sousa, H.C., Rezende, M.H., Castro, N.M.,
Vecchi, C., Luque, R., 1998. Astra blue and basic fuchsin
double staining of plant material. Biotech. Histochem. 73,
236–243.
Levin, R.A., Watson, K., Bohs, L., 2005. A four-gene study of
evolutionary relationships in Solanum section Acanthophora. Am. J. Bot. 92, 603–612.
Levin, R.A., Myers, N.R., Bohs, L., 2006. Phylogenetics
relationships among the ‘‘Spiny Solanums’’ (Solanum
155
subgenus Leptostemonum, Solanaceae). Am. J. Bot. 93,
157–169.
Matesevach, A.M., 2002. Solanum, Subgen. Leptostemonum.
Flora Fanerogámica Argentina. Fascı́culo, vol. 79. CONICET, Córdoba.
Miller, R.H., 1969. A morphological study of Solanum
mammosum and its mammiform fruit. Bot. Gaz. 130,
230–237.
Murray, M.A., 1945. Carpellary and placental structure in the
Solanaceae. Bot. Gaz. 107, 243–260.
Nee, M., 1979. Patterns in biogeography in Solanum, section
Acanthophora. In: Hawkes, J.G., Lester, R.N., Skelding,
A.J. (Eds.), The Biology and Taxonomy of the Solanaceae,
Linnean Society Symposium Series, vol. 7. The Linnean
Society of London, London, pp. 569–580.
Nee, M., 1986. Placentation patterns in the Solanaceae. In:
D’Arcy, W.G. (Ed.), Solanaceae, Biology and Systematics.
Columbia University Press, New York, pp. 169–175.
Nee, M., 1991. Synopsis of Solanum section Acanthophora, a
group of interest for glycoalkaloids. In: Hawkes, J.G.,
Lester, R., Nee, M., Estrada, N. (Eds.), Solanaceae III,
Taxonomy, Chemistry, Evolution. Royal Botanic Gardens,
Kew, pp. 257–266.
Nee, M., 1999. Synopsis of Solanum in the new world. In: Nee,
M., Symon, D.E., Lester, R.N., Jessop, J.P. (Eds.),
Solanaceae IV. Advances in Biology and Utilization. Royal
Botanic Gardens, Kew, pp. 285–333.
Pérez, A., Tomassi, V.H., 2002. Tinción con azul brillante de
cresilo en secciones vegetales con parafina. Bol. Soc.
Argent. Bot. 37, 211–215.
Roth, I., 1977. Fruit of Angiosperms. In: Linsbauer, K. (Ed.),
Encyclopaedia of Plant Anatomy, vol. 10. Borntraeger,
Berlin, pp. 1–675.
Sakai, W.S., Hanson, M., Jones, R.C., 1972. Raphides with
barbs and grooves in Xanthosoma sagittifolium (Araceae).
Science 178, 314–315.
Spjut, R.W., 1994. A systematic treatment of fruit types. Mem.
New York Bot. Gard. 70, 1–82.
Symon, D.E., 1979. Sex forms in Solanum (Solanaceae)
and the role of pollen collecting insects in Solanum,
section Acanthophora. In: Hawkes, J.G., Lester, R.N.,
Skelding, A.J. (Eds.), The Biology and Taxonomy
of the Solanaceae, Linnean Society Symposium Series,
vol. 7. The Linnean Society of London, London,
pp. 385–395.
Tewksbury, J., 2002. Fruits, frugivores and the evolutionary
arms race. New Phytol. 156, 137–139.
Thurston, E.L., 1976. Morphology, fine structure and
ontogeny of the stinging emergence of Tragia
ramosa and T. saxicola (Euphorbiaceae). Am. J. Bot. 63,
710–718.
Valencia, M.L.C. de, 1985. Anatomı́a del fruto de la
uchuva, Physalis peruviana L. Acta Biol. Colomb. 1,
63–89.
Van der Pijl, L., 1982. Principles of Dispersal in Higher Plants,
third ed. Springer, New York, Heidelberg, Berlin.
Webb, M.A., 1999. Cell-mediated crystallization of calcium
oxalate in plants. Plant Cell 11, 751–761.
Weese, T., Bohs, L., 2007. A three-gene phylogeny of the genus
Solanum (Solanaceae). Syst. Bot. 32, 445–463.
ARTICLE IN PRESS
156
F.E. Chiarini, G.E. Barboza / Flora 204 (2009) 146–156
Welman, W.G., 2003. The genus Solanum (Solanaceae) in
Southern Africa: subgenus Leptostemonum, the introduced
sections Acanthophora and Torva. Bothalia 33, 1–8.
Whalen, M.D., 1984. Conspectus of species groups in
Solanum subgenus Leptostemonum. Gentes Herbarum 12,
179–282.
Whalen, M.D., Costich, D.E., 1986. Andromonoecy in
Solanum. In: D’Arcy, W.G. (Ed.), Solanaceae, Biology
and Systematics. Columbia University Press, New York,
pp. 284–302.