A revision of gypsophila, bolanthus, ankyropetalum and phryna1

TL;DR: The only previous revision of Gypsophila is that published by Williams (1889), which includes 76 species i.e. about 3/5th of the number recognized here as discussed by the authors.

Abstract: This work was undertaken at the suggestion of Dr. J. Lanjouw and Dr. F. P. Jonker. The citation heading this paragraph indicates that the group of Caryophyllaceae with which it deals, presents unusual taxonomic difficulties. At first, it was intended to restrict the revision to the genus Gypsophila. However, in the course of the work it was realized that the small genera Bolanthus, Ankyropetalum and Phryna could not be left out of account as they had been regarded by some authors as subdivisions of Gypsophila and by others as near relatives of this genus. For this reason a complete revision of these genera too was included. The only previous revision of Gypsophila is that published by Williams (1889). His study, largely based on data derived from the literature, includes 76 species i.e. about 3/5th of the number recognized here. His views on the generic limits were strongly influenced by those expressed by BENTHAM in BENTHAM and HOOKER’s Genera Plantarum 1 (1862). Later authors did not follow him in this respect, and generally preferred BOISSIER’s delimitation (1867), so e.g. PAX and HOFFMANN in the 2nd Edition of ENGLER und PRANTL, Planzenfam. (1934). PAX had already accepted this delimitation in the first edition (1889).

Summary

Limits of the genus and revision of Literature

• The name Gypsophila was mentioned for the first time in Linne's Diss.
• Apparently, Linne did not consider Haller's work, Stirp. Helv. 1: 381 (1742), where Tunica was regarded as a separate genus.
• In a cross section of the stem one sees the epidermis, a thin cortex, a ring of sclerified tissue which differs in thickness from one species to the other, then a ring of vascular tissue, and in the centre the pith.
• When in the middle of the internode a leaf trace leaves the vascular ring, the ring is interrupted over a certain distance, while the trace deviates towards the periphery of the stem.
• In form the leaves vary from ovate to subulate, lanceolate leaves being most cojmmon.

He agreed with

• The authors third conclusion is fully in accordance with the idea of the presence of a genetic spiral.
• When the authors compare two opposite leaves, they find that they agree in their asymmetry and in the direction in which their tops deviate from the median plane.
• The planes of symmetry of the central flower of the dichasium and those of the two central flowers of the side branches do not coincide, as indicated in Eichler's figure, but their transverse section forms a triangle.
• Then the central mass assumes an oblong shape, and begins to form the carpels; meanwhile at the base of the epipetalous stamens the petals make their appearance (Fig. 2 ).

Aestivation of Sepals

• During the taxonomic study of this genus, I noticed that some calyces were dextrorsely quincuncial, whereas others were sinistrorsely quincuncial.
• This difference was noticed in the same plant and even in the same dichasium.
• Branchlets from living plants with their buds were fixed in alcohol and observed afterwards under a binocular microscope.
• Attention was focussed on two important points, viz.

The Corolla

• The corolla is in the genus Gypsophila the least variable part of the flower.
• The petals are generally 5 in number (in some aberrant flowers there may be 6 or 4), free and alternating with the sepals.
• They vary from cuneiform to linear and are mostly retuse at the apex.
• In a few species they are as long as the calyx, but mostly they are longer and spreading.
• Most Gypsophila species have white flowers, others are more or less pink, and a very few have white flowers with purple veins.

Gypsophila flowers have

• No smell, that is recognizable to us, but at the base of the petals they possess nectar glands which attract insects by which pollination may be effected.
• Every petal has three nerves that meet at the top.
• These three nerves spring from a single strand that departs in the disc from the stele at the same level as the strands which enter the sepals or at a slightly higher one.
• No: garden number; d: dextrorsely; s; sintrorsely; quin: quincuncial; dich: dichasium; fl: flower.

St. Hilaire

• Celakovsky assumed that the axis may be involved in the development of the gynoecium, but regarded the ovules as born on appendages.
• Sinnott and Bailey (1914) and Eames (1931) introduced a new theory about the nature of the primitive carpel.
• 3) The development of the outer part of the ovary wall and that of the dissepiment take place in similar directions, the dissepiment developing more quickly on the sides, while the ovary wall proper develops more quickly in the middle, so that no disjunction between the dissepiment and the ovary wall is to be recognized.

Capsule and Seeds

• The ripe fruit opens into two times as many valves as there are styles.
• The opening is not impeded by the presence of the calyx, because the latter is always shorter than the capsule.
• The seeds have mostly the shape of a compressed snail.
• In some species they are more or less similar to Saponaria seeds, which means that they are more or less globose with a distinct hilum.
• Those of the section Heterochroae are more or less similar to those of Acanthophyllum in being slightly oblong.

Cytology

• In recent years, cytological data have contributed greatly to the solution of difficult taxonomic problems.
• Whether at the level of species, genera or higher groups, cytological research has proved to be of fundamental importance to the delimitation and characterization of the taxa.
• To the present author it seems urgently necessary to combine orthodox taxonomic methods with cytological ones.
• In others they have proved to be a key to a better understanding of the relation between various taxonomic groups.

Pollen Morphology

• In recent years pollen-morphological characters have often been used to furnish additional data for taxonomic valuation.
• In many cases they proved to be of importance for the delimitation of families, genera or species; their value varies, however, from one group to the other, as is the case with many other morphological characters.
• The author tested the usefulness of pollen-morphological characters for the present revision.
• Preparations were made of pollen grains of 8 species of Gypsophila, of some species of Saponaria (16-35.5); average diameter of annulus plus pore 5.5 fi (2.6-9) ; average thickness of the exine 2.7 ju.
• The pollen-grains of Saponaria, Tunica, Ankyropetalum and Bolanthus exhibit the same general pattern; the figures in the above description apply, however, only to the 8 species of Gypsophila of which the pollen was investigated by him.

Geography

• Christ, is the only species that inhabits the N.E. corner of Africa .
• This part of the area, which may rightly be called the main variation centre of the genus Gypsophila, includes the Black Sea region, Turkey, The Caucasus, north Iraq and north Iran.
• Few other variation centres can be mentioned here.
• In the Balcan Peninsula there is one with five species representing two different sections belonging to two different subgenera.
• Its species occupy small areas which do not overlap.

chelerioides),

• And the monotypic genus Phryna, formerly included in Gypsophila, also occur there.
• All these facts point to the conclusion that the genus Gypsophila originated in its present main variation centre, and most likely in that part of the area which covers the Caucasus, the Transcaucasian region and East Turkey.

Ecology

• It inhabits for a great part the Irano-Turanian phytogeographic region which was described and delimited by Eig, A. (1931).
• This region is characterized by a very low amount of precipitation restricted to the winter and spring months, and by a long dry season.
• The summer is very hot and the winter very cold, with a temperature contrast between day and night species were compiled from indications on herbarium labels.
• They are admittedly incomplete and often of a too general nature.
• It would be worth-while to collect more information about the habitats of all these taxa since this may shed some more light on their affinities ad delimitation.

Uses

• Its small flowers, which vary in colour from pink to white and which have no smell, are not sufficiently attractive to justify its introduction in their gardens.
• Two Gypsophila species are, nevertheless used in some countries as ornamental plants, viz.
• Some species of the subgenus Pseudosaponaria have pretty small flowers, and there is a chance that they too may be introduced in their gardens.
• G. arrostii and G. acutifolia are cultivated for the production of this substance.
• In Iran and Turkmenia, the roots of G. bicolor are used instead of soap; this use depends on the presence of saponin.

Material

• The present work is based on the study of about 7000 herbarium sheets.
• In order to save space not all herbarium sheets studied by the writer are cited.

WU

• Botanisches Institut und Botanischer Garten der Universitat Wien, also known as Wien, Austria.
• The author wishes to express his gratitude to the directors and curators of the above-mentioned Herbariums who made the material entrusted to their care available to him for this study.
• Furthermore, he should like to thank the staff members of the Berlin-Dahlem Museum and of the Kew Herbarium for their kind hospitality during the author's visit to these institutions.

Bot. de

• This species is capable of propagating easily by means of cuttings which form roots directly.
• Occasionally cushions without any roots may produce new leaves and flowers when they are put in a suitable damp place (SiSkin in Komar., FI.
• In Turkmenia, where woody plants are rare, the cushions of G. aretioides are used for fuel, also known as Economical use.

spergulifolia:

• Stems few, with scattered glandular hairs above; leaves not densely imbricate at the base of the stem, always 2 on one node.
• Type: West Albania: Mt. Puka near Alessia, Grisebach s.n. Holotype (GOET), isotype (K).
• Mr. Schwarz in 1938 discovered this new species in his own collection, but his work on this collection was interrupted by the war and the new species could not be described, also known as Note.
• From Gypsophila spergulifolia it differs by having shorter, flat leaves, lower stems, and being quite glabrous.

Sept.

• A mountain plant, growing especially along streams, at 500-2500 m alt.
• Type: Iran: Kiih-e-Alwand, Auchcr-Eloy 542. Jun., fr. Jul. Type: Iran: Between Siiltanabad and Kashan near Shek-ab, Th. Strauss s.n. Isotypes (E, JE), holotype (B destroyed).
• This species was placed by Bornmiiller in the section Pseudo-acanthophyllum Boiss.
• In serpentino-lapidosis alpinis, ca. 2700 m alt.

23675 (O. Polunin). Holotype (E), isotype (K).

• Though the general habit is clearly different, the inflorescence and the flower morphology have much in common.
• Tim geographical distribution points also to affinity between these two species.
• Yet, G. graminifolia can easily be recognised by having non-spinose, long leaves, by the long pedicel, and by the number of ovules.

Tourn.

• G. iberica Barkoudah spec. nov. Gypsophila struthium auct.
• At the same time, the similarity and overlapping geographic distribution of the two species points strongly to a common origin.

Carpathian Mountains in

• Romania and E. Hungary; Rodopi Planina in Bulgaria.
• This species differs from Gypsophila olympica Boiss.
• By its flat, long, obtuse leaves with broadened base, its often distinct pedicel, its glabrous calyx and its longer petals.

Geographic distribution:

• This species differs from Gypsophila fastigiata L. by its broader leaves, much looser inflorescence, acute calyx lobes, broadly clawed petals, fewer ovules and obtusely tuberculate seeds.
• Only known from Wilayet Bitlis in Turkey.

(FI).

• This species differs from the two last species, G. acutifolia Fisch. and G. stevenii Fisch., by its narrow leaves, long pedicel, rotundate petals, and fewer ovules.
• Unfortunately, the available material of these three species was scanty and not very well preserved.
• It is not impossible that a thorough study of these three species would show them to be only subspecies of one more widely construed species.

I. BARKOUDAH

• Type: Transcaucasia: Nakhichevan, Grossheim 9 (BM, K).
• On sandy hills and subdesertic sandy soil.
• Non Freyn et Sint; Bornmiiller, Fedde Repert.
• This species differs from G. melampoda Bien, by its perennial habit, narrower acute leaves, sinuate petals, and 12 ovules.

On sandy soil.

• This species is geographically far isolated from other Gypsophila species.
• In Pinus brutia forests with maquis vegetation of Cistus creticus, Myrtus communis, and Erica arborea.
• Type: Turkey: Wilayet Elazig , along the Euphrates, Aucher 549.

Diagnosis and delimitation from other genera

• Perennial, puberulent to hirsute, grey-green to velvety herbs with thin short stems and small leaves; inflorescence contracted; bracts foliaceous; pedicel mostly shorter than the calyx.
• Bolanthus hirsutus (Labil.) Barkoudah, also known as Type species.
• 1) The calyx is pentagonal, turbinate at base, narrow, without any trace of anastomosing veins, with thick ribs alternating with broad hyaline intervals; no Saponaria species has a calyx with this combination of characters.
• 2) The petals of Bolanthus, though patent like those of Saponaria, are not provided with a corona; the claw is very minutely winged, and the neck between the limb and the claw is mostly papillose; across the limb mostly a characteristic purple stripe is present.
• 3) The Bolanthus ovules, especially the upper ones, have long funicles, and this is not so in Saponaria.

Eig

• (1931) considers a great part of Greece, South Turkey and the coastal mountains of Syria, Lebanon and Palestine as part of the east-mediterranean phytogeographic subregion.
• They grow mostly in crevices of calcareous rock and on stony slopes.

General Morphology

• The thick woody caudex, the several stems which are renewed every year, and the small leaves and flowers are very similar in all its species.
• The perennial habit and the strong root may be regarded as adaptations to their special habitat.
• It is noteworthy that the calyx teeth develop before the connate part of the calyx; this is just as in Gypsophila.
• The petals mostly show a spreading limb and a long linear claw separated from the limb by a contraction.
• Cross-fertilization can easily be realized in this way, because the long-tongued insects which visit these flowers in search of the nectar secreted by the disc, bring with them pollen grains from other flowers, and these pollen grains are easily given off to the various parts of the stigmatic surface with which the proboscis comes into contact.

Relation with other genera

• This genus may be placed between Gypsophila and Saponaria.
• This affinity with Gypsophila can be recognized especially in two subgenera of the latter, viz.
• It has to be kept in mind that this does not mean that such Gypsophila species will ever change into Bolanthus species, but that the genus Gypsophila, which comes nearest to the primitive type of the Silenoideae, reveals in some of its species part of the evolutionary tendencies which may have given rise to such genera as Bolanthus.
• This same supposition can be applied also to the similarity between Bolanthus and Acanthophyllum.
• Stem with scattered glandular hairs; flowers terminal and axillary; pedical as long as calyx; ovules 20 7.

var. hirsutus

• Stems 8-20 cm high, glandular-pubescent with some longer hairs; inflorescences dense, terminal and axillary; pedicel shorter than the calyx; calyx 4-5 mm long, puberulent with longer hairs on the ribs.
• Stems 10-15 cm high, puberulent without longer hairs; inflorescences lax; pedicel as long as or slightly longer than the calyx; calyx 4-5 mm long, puberulent without longer hairs on the ribs.

Full text

A
revision
of
Gypsophila,
Bolanthus,
Ankyropetalum
and
Phryna
(With
Arabic
summary)
Y.I.
Barkoudah
(.Botanical
Museum
and
Herbarium,
[received
March
6th,
1962)
Utrecht)
1
2
CONTENTS
Introduction
2
GYPSOPHILA
4
A.
General
Part
4
1.
Limits
of
the
genus
and
revision
of
literature
4
2.
General
morphology
9
The
root;
the
stem;
the
leaf;
on
the
phyllotaxy
of
Gypsophila
;
the
inflorescence;
remarks
on
the
dichasial
cymes
found
in
Gypsophila;
ontogeny
of
the
flower
in
the
genus
Gypsophila;
the
calyx;
aestiva-
tion
of
sepals;
the
corolla;
aestivation
of
the
corolla;
the
androecium;
the
ovary;
placentation;
capsule
and
seeds.
3.
Cytology
26
4.
Pollen
morphology
28
5.
Geography
29
6.
Ecology
31
7.
Uses
32
B.
Taxonomic
Part
33
1.
Material
33
2.
Diagnosis
and
key
to
the
genera
of
the
Diantheae
34
3
Diagnosis
and
subdivisions
of
the
genus
35
4.
Key
to
the
species
4g
5.
Species
descriptions
55
BOLANTHUS
157
1.
Diagnosis
and
delimitation
from
other
genera
159
2.
Geography
160
3.
General
morphology
jgg
4.
Relation
with
other
genera
Ig2
5.
Key
to
the
species
jg2
6.
Species
descriptions
jgg
ANKYROPETALUM
170
1.
Delimitation
of
the
genus
and
discussion
of
the
literature
. . . .
171
2.
Key
to
the
species
172
3.
Species
descriptions
173
PHRYNA
175
1.
Diagnosis
176
*)
The
author
was
enabled
to
carry
out
this
work
by
a
scholarship
received
from
Damascus
University
and
by
a
special
grant
covering
part
of
the
publishing
costs.
2
)
Future
address:
Botanical
Department,
Damascus
University,
Damascus,
Syria.

2
Y.
I.
BARKOUDAH
2.
Taxonomic
discussion
177
Index
of
names
198
Index
of
exsiccatae
191
References
187
Excluded
species
180
Species
of
uncertain
status
and
names
of
uncertain
application
. .
.
178
INTRODUCTION
Die
Gattung
ist
schwierig
zu
umgrenzen,
da
sie
sehr
nahe
Beziehungen
zeigt
zu
Saponaria
und
Acanthophyllum.
Pax
und
Hoffmann
With
regard
to
the
delimitation
of
the
genus
as
well
as
to
its
sub-
divisions
the
present
author
agrees
largely
with
Williams
(1889),
and
less
so
with
Boissier
(1867).
The
sections
Pseudoacanthophyllum
Boiss.
and
Jordania
Boiss.,
included
by
Boissier
in
Gypsophila,
are
transferred
here
to
Acanthophyllum.
The
section
Bolanthus
Boiss.,
included
in
Gypsophila
by
Boissier
and
in
Saponaria
by
Williams,
is
treated
in
the
present
work
as
a
separate
genus;
the
same
is
done
with
Ankyropetalum,
which
Williams
included
in
Gypsophila.
The
sub-
section
Vittatae
Williams
(of
the
section
Macrorrhizaea
Boiss.,
which
is
regarded
in
this
work
as
a
subgenus
of
Gypsophila
),
is
transferred
here
to
Saponaria.
Gypsophila
montana
Balf.
is
transferred
to
Sapo-
naria;
G.
alsinoides
Bunge
to
Arenaria.
In
Williams’
revision
(1889)
76
Gypsophila
species
were
recog-
nized;
of
these
8
are
excluded
in
the
present
paper.
Of
those
species
which
were
described
afterwards,
31
are
recognized
here;
with
the
19
species
described
in
this
work
this
makes
50;
the
inclusion
of
is
that
published
by
Williams
(1889).
His
study,
largely
based
on
data
derived
from
the
literature,
includes
76
species
i.e.
about
3/5th
of
the
number
recognized
here.
His
views
on
the
generic
limits
were
strongly
influenced
by
those
expressed
by
BENTHAM
in
BENTHAM
and
HOOKER’s
Genera
Plantarum
1
(1862).
Later
authors
did
not
follow
him
in
this
respect,
and
generally
preferred
BOISSIER’s
delimitation
(1867),
so
e.g.
PAX
and
HOFFMANN
in
the
2nd
Edition
of
ENGLER
und
PRANTL,
Planzenfam.
(1934).
PAX
had
already
accepted
this
delimitation
in
the
first
edition
(1889).
Gypsophila
The
only
previous
revision
of
and
by
others
as
near
relatives
of
this
genus.
For
this
reason
a
complete
revision
of
these
genera
too
was
included.
Gypsophila
could
not
be
left
out
of
account
as
they
had
been
regarded
by
some
authors
as
subdivisions
of
Phryna
and
Ankyropetalum
Bolanthus,
However,
in
the
course
of
the
work
it
was
realized
that
the
small
genera
Gypsophila.
with
which
it
deals,
presents
unusual
taxonomic
difficulties.
At
first,
it
was
intended
to
restrict
the
revision
to
the
genus
Caryophyllaceae
This
work
was
undertaken
at
the
suggestion
of
Dr.
J.
Lanjouw
and
Dr.
F.
P.
Jonker.
The
citation
heading
this
paragraph
indicates
that
the
group
of

3
GYPSOPHILA,
BOLANTHUS,
ANKYROPETALUM
AND
PHRYNA
these
50
species
made
it
necessary
to
distinguish
2
new
sections,
3
new
subsections
and
2
new
series.
In
the
present
revision
therefore
126
species
of
Gypsophila
are
recognized.
Of
these
19
are
new;
besides,
1
new
subspecies
and
2
new
varieties
are
described.
At
the
end
follows
a
list
of
names
of
uncertain
application.
Out
of
the
very
many
species
of
this
group
of
genera,
only
nine,
all
of
them
native
to
Syria,
were
known
to
the
author
in
the
field.
This
deficiency
suggested
the
idea
that
it
would
be
worth-while
to
grow
as
many
species
as
possible
in
the
garden.
Mr.
E.
A,
Mennega,
to
whom
the
author
is
much
indebted,
succeeded
in
securing
seed
samples
of
21
different
species
from
other
botanical
gardens.
Not
all
the
seed
samples
which
were
received
were
correctly
named;
in
fact,
misidentification
of
Gypsophila
species
appeared
to
be
more
frequent
in
botanical
gardens
than
in
herbariums.
Though
the
climatic
conditions
in
the
Netherlands
are
not
favourable
for
growing
Gypsophila
species,
the
gardener
of
Cantonspark
(Baarn)
succeeded
in
bringing
some
of
these
species
to
flower.
The
section
dealing
with
general
morphology
is
based
largely
on
material
collected
from
these
living
plants.
The
author
used
this
material
also
for
studying
the
cytology
of
these
species.
However,
this
part
of
the
work
did
not
yield
many
results,
in
the
first
place
because
these
plants
were
mostly
well-known
horticultural
or
else
European
species
which
had
already
been
studied
cytologically,
and
also
because
they
represented
only
2
sections
of
the
genus
Gypsophila
and
because
no
critical
species
were
among
them.
The
author
intends
to
continue
in
the
future
his
investigations
in
this
direction.
It
is
hoped
that
the
present
revision
will
help
to
clear
up
the
taxonomic
position
of
these
genera.
The
author
realizes
that
some
of
the
problems
concerning
the
relation
of
these
genera
with
other
members
of
the
tribe
Diantheae
still
remain
to
be
solved.
Cyto-
taxonomic
studies
would
presumably
contribute
valuable
data.
This
applies
especially
to
the
genera
Acanthophyllum,
Bolanthus,
Ankyro-
petalum
and
Phryna
and
to
those
Gypsophila
species
which
seem
to
approach
these
genera.
Comparative
ecological
research
too
might
help
to
understand
the
differentiation
of
the
species.
A
study
of
this
kind
would
be
particularly
interesting
where
closely
related
species
are
found
to
inhabit
adjacent
geographic
areas.
Acknowledgement.
The
author
wishes
to
express
his
deep
gratitude
to
Prof.
Dr.
J.
Lanjouw,
under
whose
supervision
and
in
whose
institute
this
work
was
carried
out;
to
Prof.
Dr.
C.
E. B.
Bremekamp
and
to
Dr.
K.
U.
Kramer
for
their
constructive
criticism
of
the
manuscript;
to
Dr.
Elisabeth
Mennega
who
corresponded
with
other
herbariums
in
order
to
obtain
the
material
on
loan
and
who
assisted
him
in
his
search
for
type
specimens;
to
Dr.
W.
Punt
for
his
assistance
in
the
pollen
morphological
part.

4
Y.
I.
BARKOUDAH
GYPSOPHILA
A.
GENERAL
PART
1.
Limits
of
the
genus
and
revision
of
Literature
The
name
Gypsophila
was
mentioned
for
the
first
time
in
Linne’s
Diss.
Chen.:
41
(1751).
There,
Linne
described
7
species
of
Gypso-
phila,
out
of
which
one
was
afterwards
removed
to
Arenaria
and
another
one
to
Tunica.
Later,
Linne
described
2
other
species,
one
of
which
was
also
a
Tunica.
That
Linne failed
to
recognize
the
difference
between
Tunica
and
Gypsophila
was
due
to
the
purely
numerical
basis
of
his
system.
Apparently,
Linne
did
not
consider
Haller’s
work,
Stirp.
Helv.
1:
381
(1742),
where
Tunica
was
regarded
as
a
separate
genus.
Scopoli
(1772)
revised
Linne’s
work,
and
made
the
bound-
aries
between
Gypsophila
and
Tunica
sufficiently
clear.
Bentham
in
Bentham
and
Hooker
(1862)
made
the
difference
quite
pronounced
by
referring
the
two
genera
to
two
different
subtribes
of
the
tribe
Sileneae;
in
the
latter
he
distinguished
three
subtribes,
viz.
Diantheae
(T
unica).
Drypideae
and
Sileneae
(Gypsophila),
according
to
the
seed
and
embryo
shape.
The
mistake
made
by
Desfontaine
(1800),
Sibthorp
and
Smith
(1806),
Seringe
in
DC.
Prod.
(1824),
Bunge
in
Ledebour,
FI.
Ross.
(1842),
Grisebach
(1843),
Dufour
(1860)
and
Janka
(1867),
by
neglecting
the
differences
between
Tunica
and
Gypsophila,
has
very
seldom
been
repeated
in
modern
literature.
The
peltate
seed
with
facial
hilum,
the
straight
embryo
placed
in
the
middle
of
the
seed,
viz.
with
the
endosperm
all
around,
i.e.
also
on
the
dorsal
side
of
the
seed,
are
characters
which
do
not
occur
in
any
Gypsophila
species.
Moreover,
the
grass-like
habit
and
the
presence
of
stigmatic
papilae
all
along
the
styles
are
typical
Tunica
characters
which
are
very
rarely
met
in
Gypsophila.
It
is
true
that
Gypsophila
caricifolia
Boiss.
and
Gypsophila
graminifolia
Bark,
are
in
their
habit
more
or
less
similar
to
Tunica,
but
the
flowers,
fruits
and
seeds
of
these
two
species
are
of
the
kind
which
is
typical
for
Gypsophila.
This
suggests
that
these
two
species
developed
in
their
vegetative
organs
on
lines
parallel
to
those
in
which
the
vegetative
parts
of
Tunica
were
differentiated;
this
how-
ever,
does
not
mean
that
there
exists
a
near
relation
between
these
two
genera.
The
difference
between
Gypsophila
and
Saponaria
too
was
not
clear
to
Linne,
for
he
put
Gypsophila
porrigens
(L.)
Boiss.
in
Saponaria;
at
an
earlier
date
he
had
referred
it
to
Silene.
Lamarck,
FI.
Fr.
(1778)
did
not
accept
Gypsophila
as
a
separate
genus,
and
inserted
its
species
in
Saponaria.
A.
Braun
(1843)
says
that
though
it
is
easy
to
separate
Saponaria
from
the
other
genera
of
the
Lychnideae,
it
is
very
difficult
to
distinguish
it
from
Gypsophila.
Bentham
(1862)
says
that
Gypso-
phila
and
Saponaria
are
so
intimately
blended
with
each
other
that
no
distinct
boundary
line
can
be
drawn
between
them,
and
some
of
the
small-flowered
species
may
according
to
him
with
almost
equal
right
be
referred
to
the
one
as
to
the
other.
Boissier
(1867)
draws

5
GYPSOPHILA,
BOLANTHUS,
ANKYROPETALUM
AND
PHRYNA
attention
to
the
points
of
resemblance
found
between
the
sections
Macrorrhizaea,
Bolanthus
and
Hagenia
of
Gypsophila
on
the
one
hand
and
the
genus
Saponaria
on
the
other,
but
as
he
regards
the
differ-
ences
in
the
shape
of
seed
and
embryo
as
more
important
than
the
points
in
which
they
resemble the
genus
Saponaria,
he
inserts
these
sections
under
Gypsophila,
though
at
an
earlier
date
most
of
the
species
belonging
to
them
had
been
referred
by
him
to
Saponaria.
Simmler
(1910)
says
that
the
possibility
of
finding
a
well-defined
boundary
line
between
Gypsophila
and
Saponaria
is
greatly
hamper-
ed
by
the
presence
of
countless
intermediate
forms,
though
in
general
these
genera
are
quite
different.
Neumayer
(1921)
unites
Gypsophila
with
Saponaria
under
the
name
of
the
latter.
Pax
and
Hoffmann
(1934)
say
“Die
Gattung
ist
schwierig
zu
umgrenzen,
da
sie
sehr
nahe
Beziehungen
zeigt
zu
Saponaria
und
Acanthophyllum”.
They
follow
in
the
main
Boissier’s
delimitation
though
accepting
the
small
changes
which
had
been
introduced
by
Williams
in
the
latter.
Siskin
(1936)
too
accepts
Boissier’s
delimitation
of
Gypsophila
from
Saponaria.
After
studying
the
species
Saponaria
ocymoides
L.,
S.
cerastioides
Fisch.,
S.
orientalis
L.,
S.
calabria
Guss.,
S.
bellidifolia
Sm.
on
the
one
hand,
and
Gypsophila
picta
Boiss.,
G.
floribunda
(Kar.
et
Kir)
Turcz.
and
G.
kermanensis
(Bornm.)
Stroh
on
the
other,
the
present
author
arrived
at
the
conclusion
that
the
character
“Unguem
petalorum
non
lamellatura,
seminis
radicula
elongata”
which
were
Boissier’s
main
arguments
in
separating
Gypsophila
from
Saponaria
can
not
be
re-
garded
as
absolutely
reliable
differences
between
these
two
genera.
The
embryo
of
Saponaria
orientalis
and
that
of
S.
cerastioides
have
a
radicle
which
is
easily
recognizable,
though
not
markedly
elongate,
but
in
the
embryo
of
Gypsophila picta
the
radicle
is
certainly
not
more
distinctly
elongate
than
in
these
two
species.
In
Gypsophila
floribunda
and
G.
kermanensis
the
elongation
of
the
radicle
is
more
pronounced
because
the
seed
is
oblong,
a
character
that
is
related
to
the
length
of
the
funicle
and
to
the
position
of
the
seed.
The
“unguis
lamellatus
vel
non-lamellatus,”
i.e.
the
presence
or
absence
of
wings
along
the
claw
of
the
petal
is
not
decisive
either.
Saponaria
cerastioides
has
a
bare
claw,
and
S.
orientalis
has
a
very
narrow
claw
with
hardly
noticeable
wings.
In
Gypsophila
picta
the
claw
is
winged
at
the
base
only,
in
the
other
Gypsophila
species,
however,
it
is
indeed
not
winged
at
all.
Let
us
now
consider
some
other
characters
of
these
three
Gypsophila
species.
All
of
them
appear
to
have
foliaceous
bracts,
a
calyx
which
is
better
described
as
cylindrical-tubulose
than
as
campanulate-turbinate,
and
which
is
provided
with
teeth
not
exceeding
the
length
of
the
connate
part
and
with
green
bands
separated
from
each
other
by
very
narrow
hyaline
stripes,
the
latter
containing
hardly
any
calcium
oxalate
druses;
they
are
further
provided
with
a
well-developed
androphore,
stamens
with
capillary
filaments,
a
narrowly
ovoid
ovary,
instead
of
a
terminal
stigma,
a
stigmatic
surface
extending
all
along
the
style,
and
an
ovoid
capsule
opening
with
small
teeth,
not
a
deeply
4-valved
one.
All
these
features
are
more
or
less
typical
Saponaria