203
Key words: Iris brandzae, 
subgenus Liminiris, series 
Spuriae, phytosociology of 
endangered species, ecology of 
rare plants, floristic composition, 
phytocoenological relevés. 
Ključne besede: Iris brandzae, 
podrod Liminiris, serija Spuriae, 
fitosociologija ogroženih vrst, 
ekologija redkih rastlin, floristična 
sestava, fitocenološki popisi.
Corresponding author:
Simona Dumitrița Chirilă
E-mail:  
simonachirilasc@yahoo.com
Received: 24. 10. 2023
Accepted: 8. 2. 2024
Wide habitat preference found in a rare, 
regional endemic species: Iris brandzae 
Prodán (Iridaceae Juss., subgenus 
Limniris, series Spuriae) in Romania
Abstract
Iris brandzae is one of the most threatened plant species in Romania. The current 
distribution of this species is very fragmented, and the population size has been re-
duced in the last 30 years. Information on the habitat preferences of the species has 
not been summarized yet. In this context, this study aimed to identify the habitat 
preferences of the species I. brandzae in Romania. The study was carried out in the 
NE (Moldova) and SE (Muntenia) regions of Romania. For the vegetation analysis, 
a total of 46 relevés were used. To classify the vegetation, we applied the hierarchi-
cal agglomerative clustering method, using the ß-flexible algorithm with ß = -0.25 
and the Bray-Curtis dissimilarity. The data were represented as mean percentage 
values, according to the the Braun-Blanquet scale . Relationships between floristic 
composition and environmental variables were analyzed with Detrended Corre-
spondence Analysis (DCA) and Canonical Correspondence Analysis (CCA). The 
vegetation analysis showed that I. brandzae grows in xerophilous, mesophilous and 
halophilous grasslands and ash-alpine alluvial forests. The species occurs in the 
communities of the Stipion lessingianae alliance, of the Festucetalia valesiacae order, 
mainly. The results of the CCA analysis indicate that the variation of the floristic 
composition of I. brandzae is influenced by the annual mean temperature (BIO1). 
Izvleček
Iris brandzae je ena najbolj ogroženih rastlinskih vrst v Romuniji. Trenutna 
razširjenost te vrste je zelo razdrobljena, velikost populacije pa se je v zadnjih 30 
letih zmanjšala. Habitatne preference vrste še niso bili raziskane, zato v tej raziskavi 
ugotavljamo, v katerih habitatih se vrsta I. brandzae pojavlja. Raziskavo smo nare-
dili v SV (Moldavija) in JV (Muntenia) regiji Romunije. Za analizo vegetacije smo 
uporabili skupaj 46 vegetacijskih popisov. Za klasifikacijo vegetacije smo uporabili 
metodo hierarhične klasifikacije z uporabo ß-fleksibilnega algoritma (ß = -0,25) 
in Bray-Curtisovim indeksom različnosti. Za mero abundance smo uporabili 
povprečne odstotne vrednosti v skladu s Braun-Blanquetovo skalo. Povezavo med 
floristično sestavo in okoljskimi spremenljivkami smo analizirali s korespondenčno 
analizo z odstranjenim trendom (DCA) in kanonično korespondenčno analizo 
(CCA). Analiza je pokazala, da I. brandzae uspeva na kserofilnih, mezofilnih in 
halofilnih traviščih ter jesenovih poplavnih gozdovih. Vrsta se pojavlja predvsem 
v združbah zveze Stipion lessingianae, reda Festucetalia valesiacae. Rezultati analize 
CCA kažejo, da na variabilnost floristične sestave I. brandzae vpliva povprečna 
letna temperatura (BIO1).
Simona Dumitrița Chirilă1, Kiril Vassilev2 & Alexandru Sabin Bădărău3
1 Danube Delta National Institute for Research and Development, Tulcea, Romania
2 Institute of Biodiversity and Ecosystem Research, Bulgarian Academy of Sciences, Sofia, Bulgaria
3 Faculty of Environmental Sciences and Engineering, Babeș-Bolyai University, Cluj-Napoca, Romania
DOI: 10.2478/hacq-2023-000923/2 • 2024, 203–212
23/2 • 2024, 203–212
204
Chirilă et al.
Wide habitat preference of Iris brandzae in Romania
Introduction
During the last three centuries, the habitats have under-
gone considerable changes as a result of anthropogenic 
activities and climate change, leading to a significant re-
duction of the geographical ranges of many plant species 
(O’Grady et al., 2004; Szczecińska et al., 2016). Thus, in 
the past, species such as Crambe tataria Sebeók (Chirilă 
et al., 2022), Pontechium maculatum (L.) Böhle & Hil-
ger and Iris aphylla subsp. hungarica Hegi had a more ex-
tensive distribution in Romania at the beginning of the 
20th century (Oprea, 2005; Dihoru & Negrean, 2009). 
However, these species are currently rare and endangered 
in the country (Mânzu et al., 2020; Chirilă et al., 2022). 
The steppe grasslands, where the previously mentioned 
species grow, are in danger due to the intensive growth of 
agriculture and the abandonment of traditional forms of 
land use (Hensen et al., 2005). Increased human activi-
ties have led to the expansion of pastures, which generally 
represent habitats that do not favor the development of 
these species (Bi et al., 2020). As a result, their popula-
tions in such habitats are numerically reduced. In this 
context, it is imperative to investigate the environmental 
and genetic factors influencing the performance of these 
species. Identifying whether habitat requirements are be-
tter indicators of rare and threatened species performan-
ce than population size and genetic diversity is critical. 
Conservation programs should be based on peripheral 
populations of the distribution range because peripheral 
populations have lower viability than central populations 
(Channell, 2004). 
Iris brandzae Prodán (Iridaceae Juss.) is one of the rare 
and endangered species in Romania. The area of the spe-
cies is restricted in Romania, the Republic of Moldova 
(Mânzu et al., 2020) and Ukraine (Derevenko, 2010; Vo-
lutsa, 2011). Although most populations were recorded 
in Romania (Oprea, 2005; Dihoru & Negrean, 2009), 
their number and size were significantly reduced in recent 
years (personal observations). In addition, the current ge-
ographical range of this species is very fragmented, being 
reported in different countries and regions, including Ro-
mania (Tutin et al., 1980; Dihoru & Negrean 2009), the 
Republic of Moldova (Tutin et al., 1980), Ukraine (Fe-
dorov & T⁀Svelev, 2001), Asia Minor and the Carpathi-
ans (Fedorov & T⁀Svelev, 2001). In Romania, the species 
was reported in Moldova and Muntenia regions (Oprea, 
2005). At Global and European level, the species has not 
yet been evaluated (EEA, 2023). In Romania, the status 
of the species varies, being considered vulnerable or rare 
(Oltean et al., 1994), vulnerable (Oprea, 2005), or with 
a low risk of extinction (Dihoru & Negrean, 2009). The 
range of the species is restricted due to factors such as 
overgrazing, land use change, irrigation, etc. (Dihoru & 
Negrean, 2009). 
Iris brandzae is a subendemic species (Oltean et al., 
1994) that was first described by Iuliu Prodan in 1935 
(British Iris Society, 1997). It was formed on the edge 
of the former Getic Gulf of the Sarmatian Sea. It later 
migrated to the plain after the retreat of the sea (Dihoru 
& Negrean, 2009). The species grows in hayfields, pas-
tures, and grasslands (Grințescu et al., 1966), more or less 
saturated places, grasslands or sparse forests (Fedorov & 
T⁀Svelev, 2001; Dihoru & Negrean, 2009), salt marshes, 
wet grasslands (Roger & Martyn, 1991; British Iris So-
ciety, 1997; Tutin et al., 1980). It is a mesophytic-mes-
ohygrophilous species that prefers well-drained, moist, 
humus-rich soils with a neutral pH, that grows on semi-
shaded slopes (Dihoru & Negrean, 2009).
Information available on the habitat preferences of the 
species I. brandzae in Romania is incomplete. In this con-
text, 46 phytocoenological relevés were analyzed: (1) to 
determine the floristic composition and phytogeographic 
characteristics of the identified groups; and (2) to deter-
mine the environmental variables that influence the vari-
ation of the floristic composition of the phytocoenoses of 
I. brandzae.
Materials and methods
Study species
Iris brandzae (Figure 1) is a perennial species with a height 
of 15 cm to 25 cm (Dihoru & Negrean, 2009). The stem 
is compressed, scabrous (Dihoru & Negrean, 2009), thin 
and striated, and the leaves of the fascicles are thin and 
narrow, in number of 3 (Grințescu et al., 1966). The fruit 
is short and cylindrical (Grințescu et al., 1966; Dihoru 
& Negrean, 2009). Flowering occurs from April to May 
(British Iris Society, 1997). The flowers are blue-violet or 
deep blue (Cassidy & Linnegar, 1987).
It belongs to subgenus Limniris (Tausch) Spach, the 
nominate section Limniris, series Spuriae (Diels) Law-
rence. This classification is sustained only by its very close 
resemblance to the other species in the series, especially 
Iris graminea L. and I. sintenisii Janka but the species was 
not included in any of the recent phylogenetic assessments 
of the subgenus or section (Wilson, 2004, 2006, 2009, 
2011; Rodionenko, 2009; Wheeler & Wilson, 2014; 
Dorofeeva & Zhurbenko, 2020; Aukhadieva et al., 2021). 
The series Spuriae consists of 12–17 species (Rodionenko, 
2007) which occur in various ecosystems from the west-
ern and central Palearctic forest-steppe, with the exception 
of Iris graminea and I. spuria L. which have their ranges 
extended across the forestry area of western and central 
23/2 • 2024, 203–212
205
Chirilă et al.
Wide habitat preference of Iris brandzae in Romania
Iris brandzae
Prodán 
200000 400000 600000 800000
500000
40
00
00
30
00
00
50
00
00
70
00
00
50
00
00
60
00
00
70
00
00
600000 700000
Văduva, 2008). The elevation of the analyzed areas varies 
from 20 m to 268 m. The annual precipitation was from 
384 to 569 mm, and the annual mean temperatures were 
from 9.1 °C to 10.8 °C. 
Vegetation sampling and 
classification
For the vegetation analysis, a total of 46 relevés (including 
469 taxa) were used: 16 of these relevés were taken from 
the Romanian Grassland Database (RGD; Vassilev et al., 
2018) made in the period 1964–2006, while the other 30 
relevés were made between April and June 2014–2021 
(personal data). The size of each personal record was 
100 m2, while the records retrieved from the Romanian 
BA
Figure 1: Iris brandzae Prodán: the habitat of the species (A), the flowering stage (B, C); Photos: S. D. Chirilă, May 2023). 
Slika 1: Iris brandzae Prodán: habitat vrste (A), faza cvetenja (B, C); Fotografije: S. D. Chirilă, maj 2023).
Figure 2: Map of the study area. (Map: S. D. Chirilă, October 2023;  
Raster: SRTM data, October 2023). 
Slika 2: Zemljevid preučevanega območja. (Zemljevid: S. D. Chirilă, 
oktober 2023; Raster: podatki SRTM, oktober 2023).
Europe (Meusel et al., 1965). Their ecology is very vari-
ate, from open xeric grasslands (I. pontica) up to closed 
canopy forests (I. graminea) and saline marshes (I. spu-
ria L., I. halophila Pall.) but most of them occur in typical 
forest-steppe environments (open woodlands and meadow 
steppe grasslands; Meusel et al., 1965). However all indi-
vidual species seems to have quite narrow, specific ecologi-
cal requirements (Rodionenko, 2007). This contrasts with 
our findings in concerning the ecology of Iris brandzae 
Prodan, which seems to cover a large variety of habitats. 
Study area
The study was carried out in localities in the northeastern 
regions of Romania – Moldova and Muntenia (Figure 2). 
The climate in this area is characterized by cold winters, 
hot summers, and frequent droughts (Chifu et al., 2006; 
●Relevés–Personal data  ●Relevés–Romanian Grassland Database
23/2 • 2024, 203–212
206
Chirilă et al.
Wide habitat preference of Iris brandzae in Romania
Grassland Database ranged in size from 20 m2 to 100 m2. 
To facilitate further data analysis, we imported the rel-
evés into JUICE 7.1 software (Tichý, 2002). Afterwards, 
we standardized the data set. This process involved: (i) 
the unification of taxonomy and nomenclature, and (ii) 
removal of taxa identified only to genus level. The final 
dataset included 46 relevés and 445 taxa.
To classify the vegetation, we applied the hierarchical 
agglomerative clustering method, using the ß-flexible al-
gorithm with ß = -0.25 and the Bray-Curtis dissimilarity, 
on the square-root transformed data. Quantitative data 
were represented as mean percentage values, according to 
the scale developed by the Braun-Blanquet (Braun-Blan-
quet, 1964) adapted by Borza & Boșcaiu (1965). Also, 
the dendrogram was made using the dis-similarity matrix 
and the flexible beta algorithm in the GINKGO program, 
included in the VegAna package (Bouxin, 2005). The op-
timal number of clusters was determined by the corrected 
Rand index (Rand, 1971), and the mean Silhouette index 
(Rousseeuw, 1987). After identifying the optimal number 
of clusters, we exported a synoptic table of plant com-
munities using the program JUICE version 7.1. The di-
agnostic species of each cluster were identified using the 
Phi coefficient in the JUICE program. Species with a phi 
coefficient value ≥0.5 were considered diagnostic. Later, 
they were validated by a permutation test (de Cáceres 
& Legendre, 2009) using the GINKGO software. The 
nomenclature and taxonomy of plant species follow the 
EURO+MED (2023), while the nomenclature of plant 
associations follow the Coldea et al. (2012, 2015). The 
cenotaxonomic affiliation of the plant associations fol-
lows Coldea et al. (2012, 2015). The nomenclature of the 
higher syntaxons units follows the Mucina et al. (2016). 
Habitat-level identification was performed using the EU-
NIS-ESy Expert System (Chytrý et al., 2020). 
Environmental variables
In the multivariate analysis, abiotic variables (elevation, 
aspect, slope, annual mean temperature, and annual pre-
cipitation) and edaphic variables (such as pH, P, and K 
content) were included. The elevation, aspect, and slope 
data were collected in the field, while information on an-
nual mean temperature and annual precipitation was ex-
tracted from the WorldClim database (Fick & Hijmans, 
2017). For personal data, soil samples were collected from 
each relevé. Afterwards, the chemical composition of the 
following elements was determined: phosphorus (P) and 
mobile potassium (K) according to ISO 11263 (1994); 
and soil pH according to SR EN ISO 10390 (2022). For 
the data obtained from the Romanian Grassland Database 
(RGD), the values for the chemical elements P, K, and soil 
pH (Ballabio et al., 2019) were extracted from the Euro-
pean Soil Database and Soil Properties (ESDAC).
Vegetation-environment 
relationship
To facilitate the interpretation of grouping and the ex-
ploration of floristic gradients, direct and indirect ordi-
nation analysis was performed. In this context, relation-
ships between floristic composition and environmental 
variables were analyzed with Detrended Corresponden-
ce Analysis using CANOCO version 5.1 (ter Braak & 
Šmilauer, 2018). Detrended Correspondence Analysis 
(DCA) resulted in a gradient length greater than 3, 
allowing an unimodal method to be applied. Thus, Ca-
nonical Correspondence Analysis (CCA) was used to ex-
plore which explanatory variables explained a relatively 
large and significant proportion of the variation in spe-
cies data. DCA analysis was performed to detect floristic 
gradients, and CCA (Canonical Correspondence Analy-
sis) analysis was applied to quantify the effect of each en-
vironmental variable on floristic composition, using the 
Monte Carlo permutation test (999 permutations). We 
used the variance inflation factor (VIF) in CANOCO 
to assess collinearity between independent variables. The 
variables with the VIF value < 5 were taken into account, 
they are not multicollinear (Table 1). The relevés distri-
bution map was made in QGIS version 3.28.3 (QGIS 
Development Team, 2022).
Table 1: VIF analysis between environmental variables.
Tabela 1: VIF analiza okoljskih spremenljivk.
Variables VIF
Elevation (m) 1.343
Bio1 – Annual temperature (°C) 2.253
Bio12 – Annual precipitation (mm) 1.257
pH 1.710
P – phosphorus (mg Kg-1) 1.318
K – potassium (mg Kg-1) 1.243
Results
Syntaxonomic scheme of the 
obtained clusters
Iris brandzae was identified in four main habitat types 
EUNIS (Chytrý et al., 2020): R1 Dry grasslands, R3 Sea-
sonally wet and wet grasslands, R6 Inland salt steppes, 
and T1 Broadleaved deciduous forests. In this context, 
46 relevés were classified into five vegetation classes, six 
orders, seven alliances, and nine plant associations.
23/2 • 2024, 203–212
207
Chirilă et al.
Wide habitat preference of Iris brandzae in Romania
Class: Festuco-Brometea Br.-Bl. et Tx. ex Soó 1947
Order: Festucetalia valesiacae Soó 1947
All.: Stipion lessingianae Soó 1947
Ass. Taraxaco serotinae-Festucetum valesiacae (Burduja 
et al. 1956, Răvăruţ et al. 1956) Sârbu et al. 1999
Ass. Jurineo arachnoideae-Stipetum lessingianae (Dobres-
cu 1974) Chifu et al. 2006
All.: Festucion valesiacae Klika 1931
Ass. Medicagini minimae-Festucetum valesiacae Wag-
ner 1941
Class: Festuco-Puccinellietea Soó ex Vicherek 1973
Order: Puccinellietalia Soó 1947
All.: Puccinellion limosae Soó 1933
Ass. Limonio gmelini-Artemisietum santonici (Soó 1927) 
Țopa 1939
Ass. Astero pannonici-Puccinellietum distantis Gehu 
et al. 1994
Class: Molinio-Arrhenatheretea Tx. 1937
Order: Molinietalia caeruleae Koch 1926
All.: Deschampsion cespitosae Horvatić 1930
Ass. Poo trivialis-Alopecuretum pratensis Regel 1925
Order: Potentillo-Polygonetalia avicularis Tx. 1947
All.: Potentillion anserinae Tx. 1947
Ass. Rorippo austriacae-Agropyretum repentis (Timár 
1947) R. Tx. 1950
Class: Carpino-Fagetea sylvaticae Jakucs ex Passarge 1968
Order: Fagetalia sylvaticae Pawłowski 1928
All.: Alnion incanae Pawłowski et al. 1928
Ass. Ulmeto campestris-Fraxinetum holotrichae Bor-
za ex Sanda 1970
Class: Salicornietea fruticosae Br.-Bl. et Tx. ex A. Bolòs y 
Vayreda et O. de Bolòs in A. Bolòs y Vayreda 1950
Order: Halimionetalia verruciferae Golub et al. 2001
All.: Artemisio santonicae-Puccinellion fominii She-
lyag-Sosonko et al. 1989
Ass. Nitrario schoberi-Artemisietum santonici Mitite-
lu 1982
Floristic composition and 
phytogeographic characteristics of 
the identified groups
The cluster analysis results are illustrated by a dendrogram 
and a synoptic table (Supplement E1). Based on the hier-
archical classification of the analyzed relevés, the vegeta-
tion was classified into seven distinct clusters within the 
data set (Figure 3). Cluster sizes varied, with cluster 6 rep-
resenting the lowest number of relevés (one relevé), while 
cluster 1 had the highest number (30 relevés). 
Table 2: Values obtained for diversity indices, chemical and bio-
climatic parameters. Values are means ± standard deviations (SD).
Tabela 2: Vrednosti diverzitenih indeksov, kemijskih in biokli-
matskih dejavnikov. Vrednosti so srednje vrednosti ± standardni 
odkloni (SD).
C
lu
ste
r
pH
El
ev
at
io
n
(m
)
Sl
op
e
(°) P
(m
g 
K
g-1
)
K
(m
g 
K
g-1
)
BI
O
1
(°C
)
BI
O
12
(m
m
)
1 7.2±0.6 149±52 12±8 50±33 337±65 9.5±0.2 556±11
2 6.7±0.4 96±28 4.3±1.2 30±2 390±40 9.6±0.1 505±105
3 6.2 51±16 25 246 10±0.4 539±29
4 6.6 117 27 341 9.6 547
5 6.8±0.1 90±97 30±14 28±6 437±6.4 9.9±0.7 532±32
6 7.3 175 4.4 42 327±27 9.5 563
7 6.7 268 8 27 360 9.9 537
8 5.8 97±5.5 21 204 10.8 518
Description of clusters
Cluster 1 – Taraxaco serotinae-Festucetum valesiacae 
association
The diagnostic species were Iris graminea (0.872, 0.001, 
***), Teucrium chamaedrys (0.861, 0.001, ***), Achil-
lea pannonica (0.816, 0.002, **), Stipa pennata (0.816, 
Figure 3: Dendrogram of relevés with I. brandzae in Romania.
Slika 3: Dendrogram vegetacijskih popisov z I. brandzae v Romuniji.
23/2 • 2024, 203–212
208
Chirilă et al.
Wide habitat preference of Iris brandzae in Romania
0.004, **), Nonea pulla (0.786, 0.003, **), Inula ger-
manica (0.762, 0.012, *), Adonis vernalis (0.755, 0.042, 
*), Schedonorus pratensis (0.732, 0.024, *), Polygala major 
(0.715, 0.016, *), Carex distans (0.693, 0.017, *), Leopol-
dia comosa (0.591, 0.035, *) and Festuca valesiaca (0.574, 
0.001, ***).
The community is heterogeneous and is characterized 
by numerous species characteristic of the Festuco-Brome-
tea class (Festuca valesiaca, Crambe tataria and Phlomis 
herba-venti subsp. pungens). This is the largest cluster, 
consists of 30 relevés (65%) in which 247 species were re-
corded. These communities occur at elevations from 20 
m to 244 m, in the Iași (66%) and Vaslui counties (27%). 
The sample area was 100 m2, with 21 to 101 species re-
corded, and a mean of 44 species. Vegetation coverage 
was from 70% to 100%. The western and southwestern 
aspects and small slopes, from 3.1° to 10° (53%) pre-
dominate. The soils vary from moderately acidic, from 
5.81 to 6.80 (33%), neutral, from 6.81 to 7.20 (17%) to 
slightly alkaline (50%). High concentrations of phosp-
horus (from 11 mg Kg-1 to 138 mg Kg-1) and potassium 
(from 240 mg Kg-1 to 603 mg Kg-1) were recorded. Also, 
the communities were identified in areas with annual 
mean temperatures from 9.1 °C to 10.2 °C, and annual 
precipitation from 521 mm to 569 mm. 
Cluster 2 – Limonio gmelini-Artemisietum santonici 
association
The diagnostic species are Artemisia santonicum subsp. 
santonicum (1.000, 0.002, **), Galatella villosa (0.993, 
0.001, ***), G. linosyris (0.986, 0.001, ***), Tripolium 
pannonicum (0.845, 0.014, *), Gypsophila muralis (0.816, 
0.013, *), Trifolium fragiferum subsp. bonannii (0.816, 
0.013, *), Veronica orchidea (0.813, 0.009, **), Aspara-
gus officinalis (0.795, 0.037, *), Bassia prostrata (0.753, 
0.016 *), Limonium gmelinii (0.749, 0.035, *), Agropyron 
cristatum subsp. pectinatum (0.747, 0.005 **), Atriplex lit-
toralis (0.716, 0.027, *), A. tatarica (0.716, 0.027, *) and 
Plantago schwarzenbergiana (0.716, 0.035, *).
This cluster consists of three relevés (7%), distribu-
ted in Iași County. These communities were recorded at 
elevations from 71 m to 126 m, on small slopes (from 
3.1° to 10°) with western and northwestern aspects. The 
annual mean temperature varied from 9.5 °C to 9.7 °C, 
and the annual precipitation varied from 384 mm to 566 
mm. The sample area was 25 m2, with the number of spe-
cies from 13 to 23, with a mean of 19 species. Vegetation 
coverage was from 77% to 95%. The soils are moderately 
acidic, with medium concentrations of phosphorus and 
very high concentrations of potassium. 
Cluster 3 – Poo trivialis-Alopecuretum pratensis 
and Rorippo austriacae-Agropyretum repentis 
associations.
The diagnostic species are Phragmites australis (0.885, 
0.001, ***), Alopecurus pratensis (0.816, 0.010, **), Bro-
mus commutatus (0.816, 0.010, **), Carex acutiformis 
(0.816, 0.010, **), C. melanostachya (0.816, 0.010, 
**), Rumex crispus (0.816, 0.010, **), Lotus corniculatus 
(0.769, 0.015, *) and Vicia villosa (0.741, 0.019, *).
This cluster consists of two relevés (4%), distributed in 
Vaslui and Iași counties, and is characterized by the pres-
ence of different species, such as Beckmannia eruciformis, 
Carex acutiformis, Centaurea jacea, Rumex crispus, Trifoli-
um hybridum, Agrostis stolonifera, etc. The number of spe-
cies varied from 45 to 48, with a mean of 47 species. Veg-
etation coverage was from 95% to 97%. The communities 
occur at elevation from 39 m to 62 m, where the annual 
precipitation were from 518 mm to 559 mm, and with an-
nual mean temperatures were from 9.7 °C to 10.3 °C. The 
soils are moderately acidic, with medium concentrations 
of phosphorus and high concentrations of potassium. 
Cluster 4 – Astero pannonici-Puccinellietum distantis 
association.
The diagnostic species is Puccinellia distans.
The cluster consists of two relevés (4%), distributed in 
Vaslui County. On the 30 m2 sample area, 22 species were 
identified, with a mean vegetation coverage of 82%. In-
stead, on the sample area of 20 m2, 20 species were identi-
fied, with a vegetation coverage from 77%. The commu-
nities were registered at elevation of 117 m, in areas with 
annual mean temperatures of 9.6 °C, and annual precipi-
tation of 547 mm. The soils are moderately acidic, rich in 
potassium, and poor in phosphorus. The most common 
species were Puccinellia distans.
Cluster 5 – Medicagini minimae-Festucetum 
valesiacae association
The diagnostic species are Anchusa ochroleuca (1.000, 
0.003, **), Anisantha tectorum (1.000, 0.003, **), Astra-
galus exscapus subsp. pubiflorus (1.000, 0.003, **), A. pon-
ticus (1.000, 0.003, **), Centaurea diffusa (1.000, 0.003, 
**), Cleistogenes serotina (1.000, 0.003, **), Hieracium vi-
rosum (1.000, 0.003, **), Xeranthemum annuum (1.000, 
0.003, **), Linum hirsutum (0.997, 0.002 **) and Bothri-
ochloa ischaemum (0.991, 0.001, ***).
This cluster consists of two relevés (4%), distributed in 
Vaslui County, were included. These communities were 
recorded at elevations from 21 m to 158 m. Annual pre-
cipitation varied from 509 mm to 554 mm, and annual 
mean temperatures varied from 9.5 °C to 10.4 °C. The 
mean number of species per 100 m2 was 69 (from 62 to 
23/2 • 2024, 203–212
209
Chirilă et al.
Wide habitat preference of Iris brandzae in Romania
75), and the vegetation coverage was from 93% to 100%. 
The soils are moderately acidic, with excessive concentra-
tions of potassium (437 mg Kg-1), and medium concen-
trations of phosphorus. 
Cluster 6 – Jurineo arachnoideae-Stipetum 
lessingianae association
The diagnostic species are Muscari racemosum (0.767, 0.001, 
***), Ferulago campestris (0.728, 0.015, *), Jurinea arach-
noidea (0.728, 0.025, *), Veronica arvensis (0.728, 0.027, *), 
Viola arvensis (0.728, 0.021, *), Camelina microcarpa (0.686, 
0.039, *) and Trinia kitaibelii (0.642, 0.034, *).
This cluster consists of one relevé (3%), distributed in 
Iași County. The communities occur in areas with annual 
precipitation was 563 mm, and annual mean tempera-
tures was 9.5 °C, at elevations 175 m. The aspects was 
southwest, and the slopes were small (from 3.1° to 10°). 
The mean number of species per 100 m2 was 55, and the 
vegetation coverage was from 98%. The soils are weakly 
alkaline, rich in phosphorus and potassium.
Cluster 7 – Nitrario schoberi-Artemisietum santonici 
association
The diagnostic species are Artemisia maritima (0.894, 
0.008, **), A. pontica (0.894, 0.006, **), Dianthus gutta-
tus (0.894, 0.006, **), Lepidium ruderale (0.775, 0.016, 
*), Nitraria schoberi (0.775, 0.017, *), Puccinellia distans 
subsp. limosa (0.775, 0.016, *), Spergularia marina (0.775, 
0.020, *), S. media (0.775, 0.017, *), Taraxacum besarabi-
cum (0.775, 0.020, *), Podospermum canum (0.665, 0.040, 
*) and Camphorosma annua (0.632, 0.048, *).
This cluster consists of three relevés (7%), distributed 
in Buzău County. The number of species was from 15 
to 22, with a mean of 19 species per 100 m2. Vegetation 
coverage varied from 74% to 82%. Communities of Ni-
trario schoberi occur at elevation of 268 m, on small slopes 
(from 3.1° to 10°) with southwest aspect. The soils are 
moderately acidic, poor in phosphorus and rich in potas-
sium. The annual mean temperature was 9.9 °C, and the 
annual precipitation was 537 mm.
Cluster 8 – Ulmeto campestris-Fraxinetum holotrichae 
association
The diagnostic species are Aegonychon purpurocaeruleum 
(1.000, 0.001, ***), Arctium lappa (1.000, 0.001, ***), 
Carex sylvatica (1.000, 0.001, ***), Colchicum autum-
nale (1.000, 0.001, ***), Ficaria verna (1.000, 0.001, 
***), Fraxinus angustifolia (1.000, 0.001, ***), F. pallisae 
(1.000, 0.001, ***), Ligustrum vulgare (1.000, 0.001, 
***), Lysimachia nummularia (1.000, 0., ***), Myosotis 
sylvatica (1.000, 0.001, ***), Oenanthe aquatica (1.000, 
0.001, ***), Poa trivialis (1.000, 0.001, ***), Pulmonaria 
officinalis (1.000, 0.001, ***), Quercus robur subsp. pe-
dunculiflora (1.000, 0.001, ***), Ulmus minor subsp. 
minor (1.000, 0.001, ***), Valeriana officinalis (1.000, 
0.001, ***) and Viola reichenbachiana (1.000, 0.001, ***).
This cluster consists of three relevés (7%), distributed 
in Buzău County. Communities with Fraxinus pallisae oc-
cur on flat terrain, with southwest aspect. The number of 
species was from 58 to 73, with a mean of 64 species per 
100 m2. Vegetation coverage was from 92% to 98%. An-
nual precipitation varied from 518 mm to 519 mm, and 
the annual mean temperature was 10.8 °C. The soils are 
moderately acidic, poor in potassium and phosphorus. 
The relationship between floristic 
composition and environmental 
variables
The main factors influencing the variation of the floristic 
composition are highlighted by indirect gradient analysis 
(Figure 4). Given that the length of the gradients of flo-
ristic similarity was 6.59 (Table 3), it was decided to use 
the unimodal method, i.e. Canonical Correspondence 
Analysis (CCA). 
Table 3: Summary of the DCA analysis.
Tabela 3: Povzetek rezultatov analize DCA.
Axis 1 Axis 2 Axis 3 Axis 4
Eigenvalues 0.8642 0.6683 0.3888 0.2819
Explained variation 
(cumulative) 11.66 20.68 25.93 29.74
Gradient length 6.59 4.5 2.52 2.6
Pseudo-canonical  
correlation (suppl.) 0.696 0.6758 0.4468 0.5444
cluster
   7
cluster 
2
cluster 
4
cluster 3
cluster
8
cluster
6
cluster
5
cluster
1
0 DCA1 7
0 
D
C
A2
 
5
Figure 4: DCA ordination diagram of the 46 vegetation relevés. 
Slika 4: DCA ordinacijski diagram 46 vegetacijskih popisov.
23/2 • 2024, 203–212
210
Chirilă et al.
Wide habitat preference of Iris brandzae in Romania
The DCA diagram highlights how various vegetation 
types are distributed on the first and second ordination 
axes. The main gradient of vegetation composition con-
cerning the first axis can be explained by positive cor-
relations with variables such as pH, phosphorus, and 
potassium. The second DCA axis was correlated with 
elevation and annual mean temperature (BIO1). This 
means that as elevation increases, the composition of the 
vegetation undergoes significant changes. Also, certain 
types of vegetation can be associated with more acidic or 
alkaline soils. Moreover, phosphorus and potassium con-
centrations can play an important role in the distribution 
of plant species (Figure 4).
BIO1 (annual mean temperature) and elevation are 
two key variables that were identified as having the great-
est impact. BIO1 explains 7.1% of the total variation and 
contributes 33.3% to the first ordination axis of the DCA 
plot. The elevation, which explains 5.2% of the total vari-
ation and contributes 24.7% to the first axis, is also sig-
nificant (Table 4).
Table 4: Results of the CCA ordination of the effect of 
variables on the floristic composition of communities with 
I. brandzae.
Tabela 4: Rezultati CCA ordinacije vpliva spremenljivk na 
floristično sestavo združb z vrsto I. brandzae.
Variables Contribution 
(%)
Explains 
(%)
pseudo- 
F
p- 
value
P 
(adj)
BIO1 33.3 7.1 3.3 0.001 0.006
Elevation 24.7 5.2 2.5 0.001 0.006
K – potassium 13.2 2.8 1.4 0.124 0.744
pH 13.9 2.9 1.4 0.055 0.33
BIO12 8.2 1.8 0.9 0.484 1
P – phosphorus 6.8 1.4 0.7 0.802 1
Discussion
The species I. brandzae occurs in Moldova and Muntenia 
regions (Mititelu et al., 1979–1980; Chifu et al., 2006, 
2014; Popescu, 2013; Sîrbu et al., 2019; Mânzu et al., 
2020). The results of our study, in which we identified 
eight distinct clusters associated with these communi-
ties, confirm the significant diversity present within these 
groups. All the other species in the section Limniris se-
ries Spuriae have quite limited ecological preferences and 
hence are rather specialized taxa. 
Iris spuria L. (with all the five subspecies), I. xanthospuria 
B. Matthew & T. Baytop, I. orientalis Mill., I. pseudonotha 
Galushko and I. halophila Pall. are always related to ha-
lophile wet grasslands, an intrazonal type of habitat from 
the nemoral, forest-steppe, and steppe zones. Iris kerneri-
ana Asch. & Sint. ex Baker, I. sintenisii Janka, I. ludwigii 
Maxim., I. notha M. Bieb. and I. pontica Zapal are present 
only in xeric grasslands, woodlands and thickets from the 
steppe and forest-steppe area (Fedchenko & Vvedenskii, 
1935; Mathew, 1984). The habitat of I. crocea Jacq. is not 
well known, the species being reported mostly from cul-
tivation and as a ruderal (Ali & Mathew, 2008). I. grami-
nea L. lives in open forests and mesic grasslands from the 
nemoral zone and the forest-steppe (Meusel et al., 1965). 
In contrast with all the mentioned species our study indi-
cates that I. brandzae Prodan is a species located solely in 
the forest-steppe, but in a much larger variety of different 
habitats, ranging from forests to mesic and xeric grasslands 
as the present study reveals for the first time. 
Many species characteristic only for the forest-steppe 
areas are also adapted to a wide variety of habitats inside 
their ranges (Erdős et al., 2018). One of the most thor-
oughly analyzed cases is Crambe tataria Sebeók (Chirilă, 
2024). The dimension of this phenomenon is not known 
nor are the eventual adaptative mechanisms behind it. 
The considerable ecological plasticity of I. brandzae is of a 
certain conservative importance. A species that is adapted 
to a large variety of habitats has better chances of survival 
than one that specializes in a single one. However, during 
our research we preliminary concluded that I. brandzae 
seems to be not so prone to resist diverse types of human 
impacts, like overgrazing associated with erosion and/or 
ruderalization – this being the theme of one of our future 
projects.
Conclusions
In Romania, the species I. brandzae is found in the Fes-
tuco-Brometea, Festuco-Puccinellietea, Carpino-Fagetea 
sylvaticae and Molinio-Arrhenatheretea vegetation class-
es. The species occurs most frequently in the Taraxaco 
serotinae-Festucetum valesiacae association. The floristic 
composition of the phytocoenoses with I. brandzae in the 
analyzed grasslands from Moldova is influenced by the 
annual mean temperature (BIO1). 
In Muntenia region, it occurs in halophilous grasslands 
and in alluvial ash-anine forests, with a low frequency 
and in small groups in various areas. The species I. bran-
dzae was identified in grasslands dominated by Nitraria 
schoberi, Puccinelia distans, Fraxinus angustifolia and F. 
pallisae. In Moldova region, the species can be found in 
halophilous, xerophilous and meso-xerophilous grass-
lands. In various localities in Iași and Vaslui counties, 
the species is rare or has a low frequency, in association 
with other dominant species such as Festuca valesiaca, 
Alopecurus pratensis, Artemisia santonicum, Bothriochloa 
ischaemum, Limonium gmelinii, Elytrigia repens or Puc-
cinellia distans. 
23/2 • 2024, 203–212
211
Chirilă et al.
Wide habitat preference of Iris brandzae in Romania
ORCID iDs
Simona Dumitrița Chirilă  https://orcid.org/0000-0003-
3397-1834
Kiril Vassilev  https://orcid.org/0000-0003-4376-5575
Alexandru Sabin Bădărău  https://orcid.org/0000-0001-
5113-2802
References
Ali, S.I., & Mathew, B. (2008). Iris in Flora of Pakistan, efloras, 
published on the Internet http://www.efloras.org [accessed 2024 
January 05]. Missouri Botanical Garden, St. Louis, MO & Harvard 
University Herbaria, Cambridge, MA.
Aukhadieva, E., Kalashnik, N., & Ishbirdin, A. (2021). Discussion 
of some taxonomy issues of species of the genus Iris L. based on 
biomorphological and karyological characteristics. In E3S Web of 
Conferences. EDP Sciences, 254, 06008.
Ballabio, C., Lugato, E., Fernández-Ugalde, O......, & Panagos, P. 
(2019). Mapping LUCAS topsoil chemical properties at European 
scale using Gaussian process regression. Geoderma, 355, 113912.
Bi, X., Li, B., Zhang, L., Nan, B., Zhang, X., & Yang, Z. (2020). Re-
sponse of grassland productivity to climate change and anthropogenic 
activities in arid regions of Central Asia. Peer J, 8, e9797.
Borza, A., & Boşcaiu, N. (1965). Introducere în studiul covorului vegetal 
(Introduction to the study of the vegetation carpet) [in Romanian]. 
Academia Republicii Populare Române.
Bouxin, G. (2005). Ginkgo, a multivariate analysis package. Journal of 
Vegetation Science, 16, 355–359.
Braun-Blanquet, J. (1964). Pflanzensoziologie. Grundzüge der 
Vegetationskunde. Springer.
British Iris Society (1997). A Guide to Species Irises: Their Identification 
and Cultivation. Cambridge University Press. p. 255.
Cassidy, G.E., & Linnegar, S. (1987). Growing Irises. Bromley: 
Christopher Helm. 68, 142.
Channell, R. (2004). The conservation value of peripheral populations: 
the supporting science. In Proceedings of the species at risk 2004 
pathways to recovery conference. Victoria, British Columbia, Canada, 
pp. 1–17.
Chifu, T., Irimia, I., & Zamfirescu, O. (2014). Festuco-Brometea. In: 
Chifu, T. (Ed.). Diversitatea fitosociologică a vegetației României. Vol. II. 
Vegetația erbacee antropizată. Tom 1. Vegetația pajiștilor (Festuco-Bro-
metea. In: Chifu T. The phytosociological diversity of Romania’s vege-
tation. Vol. II. Anthropogenic herbaceous vegetation. Part 1. Meadow 
vegetation) [in Romanian]. Institutul European, Iași, pp. 113–397.
Chifu, T., Mânzu, C., & Zamfirescu, O. (2006). Flora şi vegetaţia 
Moldovei (Flora and vegetation of Moldova) [in Romanian]. 
Universitatea Al. I. Cuza.
Chirilă, S. D., Cara, I.G., & Motrescu, I. (2022). Habitat preference 
of the endangered species Crambe tataria (Brassicaceae) from Romania. 
Tuexenia, 42, 275–296.
Chirilă, S.D. (2024). Analiza eco-cenotică și structura genetică a 
populațiilor de Crambe tataria Sebeók din România (Eco-cenotic 
analysis and genetic structure of Crambe tataria Sebeók populations in 
Romania) [in Romanian]. Presa Universitară Clujeană, pp. 272.
Chytrý, M., Tichý, L., Hennekens, S.M., & Schaminée, J.H.J. (2020). 
EUNIS Habitat Classification: expert system, characteristic species 
combinations and distribution maps of European habitats. Applied 
Vegetation Science, 23, 648–675.
Coldea, G., Indreica, A., & Oprea, A. (2015). Les associations 
végétales de Roumanie. 3-Les associations forestières et arbustives 
(Romanian plant associations. Forest and shrub associations) [in 
French]. Presa Universitară Clujeană, pp. 281.
Coldea, G., Oprea, A., Sârbu, I., Sîrbu, C., & Ștefan, N. (2012). Les 
associations végétales de Roumanie. Tome 2: Les associations anthropogénes 
(Romanian plant associations. Volume 2: Anthropogenic associations) 
[in French]. Presa Universitară Clujeană, pp. 482.
De Cáceres, M., & Legendre, P. (2009). Associations between species 
and groups of sites: indices and statistical inference. Ecology, 90, 
3566–3574.
Derevenko, T. (2010). Partes Horti botanici ChNU nuncupati. 
Yu Fedkovich in conservanda floristica diversitate Bukovina. In 
Conservation of plant diversity, pp. 310–315.
Dihoru, G., & Negrean, G. (2009). Cartea roșie a plantelor vasculare 
din România (Red book of vascular plants from Romania) [in Roma-
nian]. Academia Română, p. 290.
Dorofeeva, M.M., & Zhurbenko, P.M. (2020). Comparative study 
of ovule structure and development in some species of Iris subgenus 
Limniris (Iridaceae). Botanicheskij Zhurnal, 105(1), 15–31.
EEA (2023). Iris sintenisii subsp. brandzae (Prodàn) D.A.Webb & 
Chater. Data from: https://eunis.eea.europa.eu/species/186816. 
[accessed 2023 October 16].
Erdős, L., Ambarlı, D., Anenkhonov, O.A., Bátori, Z......, & Török, 
P. (2018). The edge of two worlds: A new review and synthesis on 
Eurasian forest-steppes. Applied Vegetation Science, 21, 345–362.
EURO+MED (2023). Euro+Med PlantBase – the information 
resource for Euro-Mediterranean plant diversity. Data from: http://
ww2.bgbm.org/EuroPlusMed [accessed 2023 October 16].
Fedchenko, B.A., & Vvedenskii, A.I. (1935). Iris in Komarov, V.L. 
(edit.) Flora of the USSR 4, Leningrad.
Fedorov, A., & Tsvelev, N.N. (2001). Flora of Russia 4. CRC Press.
Fick, S.E., & Hijmans, R.J. (2017). Worldclim 2: New 1-km spatial 
resolution climate surfaces for global land areas. International Journal of 
Climatology, 37, 14.
Grințescu, I., Nyarady, E.I., Paucă, A., Prodan, I., Șerbănescu, I., & 
Zahariadi, C. (1966). Iris brandzae, 500 p. In: Săvulescu, T. (1966). 
Flora Republicii Socialiste România. Academiei Republicii Socialiste 
România. p. 868.
Hensen, I., Oberprieler, C., & Wesche, K. (2005). Genetic structure, 
population size, and seed production of Pulsatilla vulgaris Mill. 
(Ranunculaceae) in Central Germany. Flora-Morphology, Distribution, 
Functional Ecology of Plants, 200(1), 3–14.
ISO 11263 (1994). Determination of phosphorus – Spectrometric 
determination of phosphorus soluble in sodium hydrogen carbonate 
solution. Data from: https://www.iso.org/standard/19241.html. 
[accessed 2023 October 11].
23/2 • 2024, 203–212
212
Chirilă et al.
Wide habitat preference of Iris brandzae in Romania
Mathew, B. (1984). Iris in Davis, P.H. (edit.) Flora of Turkey 8, 
Edinburgh University Press. 
Mânzu, C.C., Irimia, I., Cîșlariu, A.G., & Chinan, V.C. (2020). 
Chorological data for some rare plant species from ROSCI0222 
Sărăturile Jijia Inferioară-Prut and ROSPA0042 Eleșteele Jijiei 
și Miletinului (Iași County). Acta Horti Botanici Bucurestiensis, 
46, 35–54.
Meusel, H., Jäger, E., & Weinert, E. (1965). Vergleichende Chorologie 
der zentraleuropäischen Flora. Vol. 1. Fischer, Jena (Verbreitungskarte).
Mititelu, D., Ștefan, N., & Ciupercă, Gh. (1979–1980). Flora și 
vegetația rezervației „Pâclele” cu vulcani noroioși (Jud. Buzău) (Flora 
and vegetation of the reserve „Pâclele” with mud volcanoes) [in 
Romanian], pp. 99–120.
Mucina, L., Bültmann, H., Dierßen, K.........., & Tichý, L. (2016). 
Vegetation of Europe: hierarchical floristic classification system of 
vascular plant, bryophyte, lichen and algal communities. Applied 
Vegetation Science, 19, 3–264.
O’Grady, J.J., Reed, D.H......, & Frankham, R. (2004). What are the 
best correlates of predicted extinction risk? Biological Conservation, 
118, 513–520.
Oltean, M., Negrean, G., Popescu, A...., & Mihăilescu, S. (1994). 
Lista roşie a plantelor superioare din România (Red list of higher plants 
from Romania) [in Romanian]: Studii, sinteze, documentaţii de ecolo-
gie. 1. Academia Română, Institutul de Biologie București, pp. 1–52.
Oprea, A. (2005). Lista critică a plantelor vasculare din România 
(Critical list of vascular plants from Romania) [in Romanian]. Univ. 
“Alexandru Ioan Cuza”, Iași, pp. 668.
Popescu, I.E. (2013). Unicitatea patrimoniului natural din Rezervaţia 
de fâneţe seculare de la Valea lui David Iași (The uniqueness of the 
natural heritage of the Secular Hay Reserve from Valea lui David Iași) 
[in Romanian]. Mnemosyne, 4, 7–36.
QGIS Development Team (2022). QGIS version 3.28. Data from: 
https://qgis.org, [accessed 2023 October 14].
Rand, W.M. (1971). Objective criteria for the evaluation of clustering 
methods. Journal of the American Statistical Association, 66(336), 
846–850.
Rodionenko, G.I. (2005). On the independence of the genus Limniris 
(Iridaceae). Botanicheskij Zhurnal, 92(4), 547–554 [in Russian].
Rodionenko, G.I. (2009). A new system of the genus Iris (Iridaceae). 
Botanicheskij Zhurnal, 94(3), 423–435 [in Russian].
Roger, P. & Martyn, R. (1991). Perennials Vol. 1. Pan Books Ltd. p. 216.
Rousseeuw, P.J. (1987). Silhouettes: a Graphical Aid to the 
Interpretation and Validation of Cluster Analysis. Computational and 
Applied Mathematics, 20, 53–65.
Sîrbu, C., Oprea, A., & Peregrym, M. (2019). New data about 
Fritillaria meleagroides in Romania. Journal of Plant Development, 
26, 123–135.
SR EN ISO 10390 (2022). Soil, treated biowaste and sludge. 
Determination of pH. Data from: https://standards.iteh.ai/catalog/
standards/cen/35c1e689-fbc3-4bd2-a9e7-7162f61ae30b/en-
iso-10390-2022. [accessed 2023 October 14].
SRTM Data (2023). SRTM 90 m Digital Elevation Data. Data from: 
https://srtm.csi.cgiar.org/srtmdata/. [accessed 2023 October 14].
Szczecińska, M., Sramko, G., Wołosz, K., & Sawicki, J. (2016). 
Genetic diversity and population structure of the rare and endangered 
plant species Pulsatilla patens (L.) Mill in East Central Europe. PloS 
One, 11(3), e0151730.
ter Braak, C.J.F., & Šmilauer, P. (2018). Canoco reference manual 
and user’s guide: Software for ordination (version 5.10). Biometris, 
Wageningen University & Research.
Tichý, L. (2002). JUICE, software for vegetation classification. Journal 
of Vegetation Science, 13, 451–453.
Tutin, T.G., Heywood, V.H., Burges, N.A....., & Richardson, O.I.B.K. 
(1980). Flora Europaea. Volume 5: Alismataceae to Orchidaceae 
(Monocotyledones).
Vassilev, K., Ruprecht, E., Alexiu, V........., & Dengler, J. (2018). The 
Romanian Grassland Database (RGD): historical background, current 
status and future perspectives. Phytocoenologia, 48, 91–100.
Văduva, I. (2008). Clima României (Climate of Romania) [in 
Romanian]. Fundația România de Mâine.
Volutsa, O.D. (2011). Iris brandzae Prodán (Iridaceae) u flori 
Chernivec’koi’ oblasti. Aktual’ni problemy botaniky ta ekologii’. (In 
Ukrainian).
Wheeler, A.S., & Wilson, C.A. (2014). Exploring phylogenetic 
relationships within a broadly distributed northern hemisphere group 
of semi-aquatic Iris species (Iridaceae). Systematic Botany, 34(2), 
277–284.
Wilson, C.A. (2004). Phylogeny of Iris based on chloroplast matK 
gene and trnK intron sequence data. Molecular Phylogenetics and 
Evolution, 33, 402–412.
Wilson, C.A. (2006). Patterns of evolution in characters that define Iris 
subgenera and sections. Aliso, 22, 425–433.
Wilson, C.A. (2009). Phylogenetic relationships among the recognized 
series in Iris section Limniris. Systematic Botany, 34, 277–284.
Wilson, C.A. (2011). Subgeneric classification in Iris re-examined 
using chloroplastsequence data. Taxon, 60, 27–35.