American Journal of Plant Sciences, 2011, 2, 1-14
doi:10.4236/ajps.2011.21001 Published Online March 2011 (http://www.SciRP.org/journal/ajps)
1
Numerical Taxonomic Study of Some Tribes of
Gramineae from Egypt
Ahmed Osman1, Mohammed Zaki2, Sohar Hamed1, Nagwa Hussein1
1South Valley University, Qena, Egypt; 2Cairo University, Giza, Egypt.
Received October 10th, 2010; revised November 26th, 2010; accepted December 9th, 2010.
ABSTRACT
A systematic study of eleven tribes of Gramineae surveyed 34 characters including fruit morphology, fru it anatomy and
palynology. The results were conducted to some numerical analysis aspects. On the basis of UPGMA (Unpaired Group
Method of Average) clustering and PCA (Principal component analysis), the results show congruence between the
UPGMA clustering and PCA method, in suggesting two major clad s/groups and five subclads.
Keywords: Gr amineae, Numerical Taxonomy, UPGMA, Cladistic Tree
1. Introduction
Poaceae (grasses) is one of the most species-rich flower-
ing plant families and includes many economically im-
portant crops. Parallel evolution of such features as the
annual habit, C4 photosynthesis and several highly char-
acteristic reproductive structures has facilitated a series
of major radiations within Poaceae, culminating in the
existing global distribution of about 10000 species and
700 genera [1,2]. A phylogeny of Poaceae was recently
established using a combination of multiple data sets
from both molecules and morphology [3], enabling im-
proved understanding of relationships between basal and
derived grasses.
Poaceae tribes and genera are subject to different stud-
ies in order to understand the phylogenetic relationships
between taxa. Many attempts have been made to address
phylogenetic relationships of Chloridoideae; synonym
Eragrostoideae that comprises approximately 146 genera
and 1360 species, whose adoption of efficient C4 photo-
synthesis had led to its successful proliferation in the
tropics and subtropics [1]; mainly based on the basis of
morphological and molecular data [4,5], but general
agreement is still lacking. The grass tribe Triticeae in-
cludes some of the world’s most important cereals and a
significant number of important forage grasses [6]. Due
to its renownedly complicated evolutionary history and
its economic importance there has been an increasing
interest in producing molecular phylogenies for the
Triticeae. Attempts to unravel the relationships in the
group have been based on many different types of data
e.g. isozymes [7], restriction site data [8,9] and sequence
data from a number of different coding and/or
non-coding regions, viz.5S RNA [10]. Among the mod-
ern tools for plant taxonomy, reference [11] stated that
increasing use has taken place of computers for data stor-
age and analysis during the past twenty years. Data de-
rived from all tools of taxonomical investigations has to
be analyzed mathematically and cladistic trees have to be
drawn. Despite of the criticism of using cladistic analysis
in taxonomy, cladistic methods have become a most
useful technical tool for clarifying intrafamilial relation-
ships. Moreover; the advantages of using more rigorous
techniques to elaborate natural classifications or evolu-
tionary diagrams instead of those that have been used
traditionally in botany have been well presented by [12].
A phylogenetic analysis of Triticeae was performed by
means of numerical methods due to [13]. Five methods,
each based on extreme assumptions of parameters so
interpreted under [14] evolutionary model, were used.
The most parsimonious tree obtained served as a base for
subsequent elaboration of the final tree, taking into con-
sideration genetic information primarily, and for the
erection of the proposed phylogenetic classification of
Triticeae. A key is provided for identification of the
groupings in the tribe. The proposed classification is
discussed in the light of previous classifications, even
though none of them were phylogenetic in the sense of
Hennig. Reference [15] have introduced a cladistic ana-
lysis, primarily based on morphological data from 40

Numerical Taxonomic Study of Some Tribes of Gramineae from Egypt 2
taxa representing the 24 genera of the Triticeae. They
used Bremer support as a measure of branch support. The
trees based on morphology and on molecular data are
largely incongruent. Also; [16] and [17]; in their study
showed the relationships of graminid/restiid of poales in
a cladistic tree. This report aims to apply numerical tax-
onomy; UPGMA and Cluster Analysis; to reveal better
the relationships between genera within a tribe and tribes
within the family based on the data collected from the
previous investigations for caryopses morphology and
anatomy and pollen grains morphology.
2. Materials and Methods
2.1. Plant Material
The study dealt with 34 species belonging to 25 genera
of 11 tribes of Gramineae; Andropogoneae, Aristideae,
Arundineae, Aveneae, Brachypodieae, Bromeae, Er-
agrostideae, Paniceae, Poeae, Stipeae and Triticeae. The
study based on herbarial specimens dried and kept in the
QNA Herbarium (in South Valley University, Qena,
Botany Department) and some species received on loan
from CAI Herbarium for the palynological study (Table
1). In the following analysis, species and genera consti-
tuted the OTUs (Operational Taxonomic Units). In order
to broadly sample variation, the OUTs consist of a num-
ber of collections from different localities in Egypt, illus-
trated in Table 2.
2.2. Characters Observations
Table 3 shows the characters and character states scored
for fruit anatomy, fruit morphology and pollen morphol-
ogy, averaged for each OUT. A total of 37 characters
were measured, comprising 22 qualitative and 15 quanti-
tative characters. For recording the total characters; a
main using of different microscopic techniques; light,
scanning electron and stereomicroscope were used for
investigating different samples and recording data col-
lected. Table 4 shows the data matrix used for analysis
of taxa studied. For some of the OTUs, some characters’
observations were lacked and these omissions were
coded as missing data (−0.999).
2.3. Data Analysis
Two types of analysis were performed with STATIS-
TICA version 5.0 computer software. Firstly, the total
data coded were analyzed by the Unpired Group Method
of Average (UPGMA) clustering. Construction of the
tree illustrating the relationships between the studied
species was performed using Arthimetic Average (UP-
GMA) proposed by [18]. Secondly, factor analysis and
factor loadings were applied to determine the major and
specific characters that aid in separation using the same
program. A principal component analysis (PCA) was
also performed according to [19].
3. Results
Figure 1 shows the UPGMA cladistic tree comprising all
OTUs in the present study. The tree separated into two
major clads at 100 dissimilarity distance. The first major
clad at 53 dissimilarity distance, comprised only two
species of the total number; Panicum turgidum and
Arundo donax; while the second major clad at 93 dis-
similarity comprised the rest 32 species.
The second major clad separated into two branches,
the first branch includes five subclads: 1) A subclad at 86
dissimilarity distance with five species; Lamarckia aurea,
Oryzopsis miliacae, Polypogon monspeliensis, Eragrostis
cilinensis and Stipagrostis ciliata. 2) A subclad at 84 dis-
similarity distance with Aegilops kotshyi. 3) A subclad at
80 dissimilarity distance comprises six species; Aegilops
ventricosa, Hordeum murinum ssp. leporinum, Lolium
perenne, Bromus scoparius, Brachypodium distachyum
and Avena fatua. 4) A subclad at 74 dissimilarity dis-
tance includes nine species; Stipa capensis, Dactylis
glomerata, Stipa lagascae, Bromus rubens, Echinocoloa
colona, Coelachryum bervifolium, Schismus arabicus,
Stipa parviflora and Aristida funiculata and 5) A subclad
at 54 dissimilarity distance with nine species, Poa annua,
Polypogon maritimus, Eragrostis minor, Phalaris minor,
Avena barbata, Aristida mutabilis, Cenchrus ciliaris,
Leptochloa fusca and Aristida adscensionis. The second
branch of the second major clad comprises two species;
Dactylochtenium aegyptium and Sorghum variegatum; at
66 dissimilarity distance.
Factor analysis using Principal Component Analysis
(PCA) showed that the most intrinsic characters en-
hanced separation of the total OTUs are fruit shape, color
type and fruit surface sculpture of the morphological
characters, of the fruit anatomical characters; section
outline shape, hull cells type, aleurone cells shape and
orientation, scutellum shape and thickness and en-
dosperm thickness are intrinsic characters for separation.
Meanwhile, all the pollen characters are good data for
separating of taxa; pollen class, shape, size, surface
sculpture, annulus thickness, pore diameter, pollen wall
thickness, sexine and nexine thickness. The characters of
separation are of high factor loadings ≥ (± 0.7) Table 5.
These represented by a percentage of the total variation
as 24.01% from the three factors extracted as; factor 1 is
responsible for 16.54% of the variation, factor 2 is re-
sponsible for 4.34% of the variation and factor 3 is re-
sponsible for the minimum value of the total variation; is
3.13%. The plot of 34 OTUs on the first two factors ex-
tracted in the PCA method is shown in (Figure 2). Plot
of factor 1/2 shows two groups. 1) Group of 6. Arundo

Numerical Taxonomic Study of Some Tribes of Gramineae from Egypt 3
Table 1. List for the investigated taxa with their geographical region.
No. Species Herb. Collecting region Year Collector
1 Sorghum varigatum QNA N 2004 A.K. OSMAN
2 Aristida adscensionis QNA GE 2004 A.K. OSMAN
3 Aristida funiculate QNA GE 2004 A.K. OSMAN
4 Aristida mutabilis QNA GE 2004 A.K. OSMAN
5 Stipagrostis ciliata QNA R 2004 A.K. OSMAN
6 Arundo donax QNA N 2009 N.R.A. HUSSEIN
7 Schismus arabicus QNA M 2006 A.K. OSMAN
8 Aven a b a r b a t a QNA N 2005 A.K. OSMAN
9 Avena fatua QNA N 2005 A.K. OSMAN
10 Phalaris minor QNA M 2006 A.K. OSMAN
11 Polypogon maritimus QNA N 2006 A.K. OSMAN
12 Polypogon monspeliensis QNA N 2006,07 A.K. OSMAN
13 Brachypodium distachym QNA M 2006 A.K. OSMAN
14 Bromus rubens QNA M 2005 A.K. OSMAN
15 Bromus scoparius QNA M 2006 A.K. OSMAN
16 Coelachyrum bervifolim QNA GE 2004 A.K. OSMAN
17 Dactylochtenium aegyptium QNA N 2006 A.K. OSMAN
18 Eragrostis cilianensis QNA S 2005 A.K. OSMAN
19 Eragrostis minor QNA M 2005 A.K. OSMAN
20 Leptochloa fusca QNA S 2005 A.K. OSMAN
21 Cenchrus ciliaris QNA GE 2004 A.K. OSMAN
22 Echinochloa colona QNA N 2006 A.K. OSMAN
23 Panicum turgidum QNA N 2005 A.K. OSMAN
24 Dactylis glomerata QNA M 2006 A.K. OSMAN
25 Lamarckia aurea QNA M 2006 A.K. OSMAN
26 Lolium perenne QNA N 2009 N.R.A. HUSSEIN
27 Poa annua QNA N 2005 A.K. OSMAN
28 Oryzopsis miliacea CAI M 2006 A.K. OSMAN
29 Stipa capensis QNA M 2006 A.K. OSMAN
30 Stipa lagascae QNA M 2006 A.K. OSMAN
31 Stipa parviflora QNA M 2006 A.K. OSMAN
32 Aegilops kotshyi QNA M 2006 A.K. OSMAN
33 Aegilops ventricosa CAI M 2006 A.K. OSMAN
34 Hordium murinum Subsp. Leporinum CAI S 2005 A.K. OSMAN
QNA = Qena Faculty of Science Herbarium (QNA a proposed Agronym); M = Mediterranean region; N = Nile region; R = Red sea coastal region; S = Sinai;
GE= Gabel Elba.

Numerical Taxonomic Study of Some Tribes of Gramineae from Egypt 4
Table 2. List for the names of total OTUs studied and their corresponding numbers and tribe names.
OTUs no. Species name Tribes OTUs no. Species name Tribes
Sp.1 Sorghum varigatum Andro Sp.18 Eragrostis cilianensis Eragro
Sp.2 Aristida adscensionis Arist Sp.19 Eragrostis minor Eragro
Sp.3 Aristida funiculate Arist Sp.20 Leptochloa fusca Eragro
Sp.4 Aristida mutabilis Arist Sp.21 Cenchrus ciliaris Panic
Sp.5 Stipagrostis ciliate Arist Sp.22 Echinochloa colona Panic
Sp.6 Arundo donax Arund Sp.23 Panicum turgidum Panic
Sp.7 Schismus arabicus Arund Sp.24 Dactylis glomerata Poeae
Sp.8 Avena barbata Aven Sp.25 Lamarckia aurea Poeae
Sp.9 Avena fatua Aven Sp.26 Lolium perenne Poeae
Sp.10 Phalaris minor Aven Sp.27 Poa annua Poeae
Sp.11 Polypogon maritimus Aven Sp.28 Oryzopsis miliacea Stipeae
Sp.12 Polypogon monspeliensis Aven Sp.29 Stipa capensis Stipeae
Sp.13 Brachypodium distachym Brach Sp.30 Stipa lagascae Stipeae
Sp.14 Bromus rubens Brom Sp.31 Stipa parviflora Stipeae
Sp.15 Bromus scoparius Brom Sp.32 Aegilops kotshyi Triti
Sp.16 Coelachyrum bervifolim Eragro Sp.33 Aegilops ventricosa Triti
Sp.17 Dactylochtenium aegyptium Eragro Sp.34 Hordium murinum Subsp. Leporinum Triti
Andro= Andropogoneae, Arist= Aristideae, Arund= Arundineae, Aven= Aveneae, Brach = Brachypodieae, Brom= Bromeae, Eragro= Eragrostideae, Panic=
Paniceae,Triti= Triticeae.
Figure 1. Cladogram of 34 species studied by UPGMA method.

Numerical Taxonomic Study of Some Tribes of Gramineae from Egypt 5
Table 3. Characters and character states used in the analysis of Gramineae tribes.
Characters Character states Code
Fruits Morphological characters
1. Fruit shape
Elliptic
Rectangular
Cordate
Oblong
Linear
Narrow cordate
Cordate with hollow part
Circular
Oval
Oval with acute protrusion
(tall) oblong with tapered ends
Elliptic with tapered ends (acute elliptic)
1
2
3
4
5
6
7
8
9
10
11
12
2. Fruit coloring Onecoloured (uniformly coloured)
Bicoloured
1
2
3. Color type
Light brown
Dark brown with light yellowish sheath
Brown
Beage
Dark red
Light beage
Light beage and light green strips
Brown with pale beage sheath
Beage with light violet sheath
Brown and light orange
Beage and brown protrusions
Shiny beage and light brown small spots
Dark beage
Beage and brown black spots
Light brown and dark brown ends
Light brown with light green sheath
Brown with light green sheath
Gradient beage with light green and brown spot at grain top
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Trichomes on grains
4. Trichomes type Simple hairs 1
5. Trichome presence
Absent (glabrous)
Present
Present on grain sheath
1
2
3
6. Hair length
Short
Long
With different lengths
1
2
3
7. Hair coloring Shiny transparent (colorless)
Colored
1
2
8. Position o f attaching
Around all grain surface
At grain edges
Around all surface, condensed at the top
At the top
At the base
Few around all surface, condensed at the base
On the side margins of the grain sheath
1
2
3
4
5
6
7
9. Fruit surface sculpture
Reticulate with straight cell wall
Striate
Scabrate
1
2
3

Numerical Taxonomic Study of Some Tribes of Gramineae from Egypt 6
Reticulate with undulate granulate cell wall
Compound reticulate with foveolate
Reticulate with undulate cell wall
Striate to scabrate
Scalariform
Striate at the intermediate and reticulate with straight cell wall next to hilum
Compound reticulate with tubrculate
Smooth
Rugose
Reticulate with straight to undulate cell wall
Ribbed pattern cell wall
Compound reticulate with granulate
Scaly surface
4
5
6
7
8
9
10
11
12
13
14
15
16
Fruits Morph metrical characters
10. Fruit weight ( mg)
=(0.03-2.12)
=(2.12-4.21)
=(4.21-6.3)
=(6.3-8.39)
=(8.39-10.48)
=(10.48-12.57)
=(12.57-14.66)
1
2
3
4
5
6
7
Fruit size
11. Wide (µm)
=(29-112)
=(112-195)
=(195-279)
1
2
3
12. Length (µm)
=(1.9-52.9)
=(52.9-103.9)
=(103.9-154.9)
=(154.9-205.9)
=(205.9-256.9)
=(256.9-307.94)
=(307.94-358.9)
1
2
3
4
5
6
7
Fruit anatomy
13. Section outline shape
Circular
Circular to cordate
Circular to oval
Semi-circular
Oval
Oval to cordate
Cordate
Oval to rectangular with curved corners
Oval to rectangular
C- shaped
Circular to triangular with obtused corners
Triangular with obtused corners
Triangular
Rectangular with obtused corners
Rectangular with obtused corners to oval
Rectangular with acuted corners
Rectangular
Polygonal
Linear with folded ends
Circled linear
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
14. Hull cells type
Parenchyma
Epithelial
Not observed
1
2
3

Numerical Taxonomic Study of Some Tribes of Gramineae from Egypt 7
15. Aleurone cells shape
Rectangular
Rectangular and quadrate
Quadrate
Rectangular and polygonal
Not observed
1
2
3
4
5
16. Aleurone cells oreintation
Horizontal
Vertical
Horizontal and vertical
Not observed
1
2
3
4
17. Scutellum cells shape
Strip of cells
Elliptic mass of cells
Rectangular to cordate mass of cells
Quinqangular mass of cells
Oval mass and strip of cells
Triangular mass and strip of cells
Not observed
1
2
3
4
5
6
7
18. Endosperm differentiation Differentiated
Differentiated and Undifferentiated
1
2
19. Type of endosper m Starchy
Starchy and fluidy
1
2
20. Section wide (µm)
=(5.33-31.66)
=(31.66-57.99)
=(57.99-84.32)
=(84.32-110.65)
=110.65-136.98)
=(136.98-163.31)
=(163.31-189.64)
1
2
3
4
5
6
7
21. Section length (µm)
=(33.33-71.94)
=(71.94-110.55)
=(110.55-149.16)
=(149.16-187.77)
=(187.77-226.38)
=(226.38-264.99)
=(264.99-303.6)
1
2
3
4
5
6
7
22. Hull cells thickness (µm)
Not ovserved
=(9.39-14.79)
=(14.79-20.19)
=(20.19-25.59)
=(25.59-30.99)
=(30.99-36.39)
0
1
2
3
4
5
23. Seed coat thickness (µm)
=(2.43-8.63)
=(8.63-14.83)
=(14.83-21.03)
=(21.03-27.23)
=(27.23-33.43)
=(33.43-39.63)
1
2
3
4
5
6
24. Aleurone layer thickness
(µm)
Not observed
=(1.92-9.42)
=(9.42-16.92)
=(19.92-24.42)
=(24.42-31.92)
=(31.92-39.42)
=(39.42-46.92)
0
1
2
3
4
5
6
25. Scutellum thickness ( µm)
Not observed
=(5.54-85.94)
=(85.94-166.43)
=(166.34-246.74)
0
1
2
3

Numerical Taxonomic Study of Some Tribes of Gramineae from Egypt 8
=(246.74-327.14)
=(327.14-407.54)
=(407.54-487.94)
Thicknesses in the range codes 1&4
Thicknesses in the range codes 1&3
4
5
6
7
8
26. Endosperm thickness (µm)
=(21.52-125.02)
=(125.02-228.52)
=(228.52-332.02)
=(332.02-435.52)
=(435.52-539.02)
=(539.02-642.52)
Two different thicknesses in range of codes 1&2
1
2
3
4
5
6
7
Pollen grains Morphological characters
27. Pollen class Monoporate
Diporate
1
2
28. Annulus Pollen annulate 1
29. OperculumPollen operculate1
30. Pollen size Pollen small
Pollen medium
1
2
31. Pollen shape
Pollen Oblate-spheroidal
Pollen Suboblate
Pollen Prolate-spheroidal
Pollen spheroidal
1
2
3
4
32. Surface sculpturing
Areolate
Granulate
Scabrate
verrucate
1
2
3
4
Pollen grains Morph metrical characters
33. Annulus thickness (µm)
=(1.16-1.52)
=(1.52-1.88)
=(1.88-2.24)
=(2.24-2.6)
=(2.6-2.96)
=(2.96-3.32)
1
2
3
4
5
6
34. Pore diameter (µm)
=(1.57-2.17)
=(2.17-2.77)
=(2.77-3.37)
=(3.37-3.97)
=(3.97-4.57)
1
2
3
4
5
35. Pollen wall thickness (µm)
=(0.698-0.848)
=(0.848-0.998)
=(0.998-1.148)
=(1.148-1.298)
=(1.298-1.448)
1
2
3
4
5
36. Sexine thickness (µm)
=(0.434-0.454)
=(0.454-0.474)
=(0.474-0.494)
=(0.494-0.514)
=(0.514-0.534)
=(0.534-0.554)
1
2
3
4
5
6
37. Nexine thickness (µm)
=(0.314-0.354)
=(0.354-0.394)
=(0.394-0.434)
=(0.434-0.474)
=(0.474-0.514)
=(0.514-0.554)
1
2
3
4
5
6

Numerical Taxonomic Study of Some Tribes of Gramineae from Egypt 9
Table 4. Data matrix used in the numerical analysis for some Gramineae tribes.
Sp.1 Sp.2 Sp.3 Sp.4Sp.5 Sp.6 Sp.7Sp.8Sp.9Sp.10Sp.11Sp.12Sp.13 Sp.14 Sp.15Sp.16
1 3 5 5 5 5 −0.999 9 4 4 3 3 4 5 11 5 7
2 2 1 1 1 1 −0.999 2 1 1 1 1 1 1 1 1 1
3 18 9 8 3 1 −0.999 12 6 4 3 1 16 5 5 5 5
4 0 1 1 1 1 −0.999 0 1 1 1 0 0 1 1 1 0
5 0 1 1 1 1 −0.999 0 1 2 1 0 0 1 1 1 0
6 0 2 2 2 2 −0.999 0 2 2 1 0 0 1 1 1 0
7 0 5 5 5 5 −0.999 0 3 1 1 0 0 2 1 7 0
8 2 7 6 2 14 −0.999 11 2 2 3 7 12 2 15 2 6
9 1 1 1 1 1 −0.999 1 1 4 1 1 1 2 2 1 1
10 3 3 1 1 1 −0.999 1 2 5 3 1 1 2 3 1 2
11 1 3 1 1 1 −0.999 1 1 4 1 1 1 1 1 2 1
12 16 5 20 4 1 −0.999 12 6 17 5 1 2 10 15 10 10
13 1 1 3 2 2 −0.999 3 3 3 3 3 1 3 3 3 1
14 1 1 5 1 1 −0.999 1 1 1 4 2 1 1 2 1 1
15 1 1 4 3 1 −0.999 1 2 2 2 2 2 1 2 2 1
16 2 2 7 2 1 −0.999 1 4 1 1 1 7 7 7 7 1
17 1 1 1 1 1 −0.999 1 1 1 1 1 1 2 1 1 1
18 1 1 2 1 1 −0.999 1 1 1 1 1 1 1 1 1 1
19 2 1 2 2 2 −0.999 2 4 4 2 2 1 2 2 1 2
20 3 1 1 1 1 −0.999 1 3 5 2 1 1 3 2 2 2
21 1 1 0 1 1 −0.999 0 0 0 0 0 2 0 0 0 3
22 2 1 1 1 1 −0.999 2 1 4 2 1 1 2 1 2 2
23 1 1 0 1 1 −0.999 1 5 5 3 2 2 1 3 3 1
24 2 2 0 2 1 −0.999 1 5 1 1 1 0 0 0 0 1
25 1 1 1 2 4 −0.999 3 5 6 4 4 2 7 2 1 1
26 1 −0.999 −0.999 −0.9991 1 −0.999 −0.9991 −0.999 −0.999 −0.9992 −0.999 2 −0.999
27 2 −0.999 −0.999 −0.9991 2 −0.999 −0.9992 −0.999 −0.999 −0.9992 −0.999 2 −0.999
28 2 −0.999 −0.999 −0.9994 1 −0.999 −0.9991 −0.999 −0.999 −0.9994 −0.999 4 −0.999
29 3 −0.999 −0.999 −0.9992 1 −0.999 −0.9993 −0.999 −0.999 −0.9992 −0.999 2 −0.999
30 2 −0.999 −0.999 −0.9993 2 −0.999 −0.9993 −0.999 −0.999 −0.9991 −0.999 4 −0.999
31 1 −0.999 −0.999 −0.9991 2 −0.999 −0.9994 −0.999 −0.999 −0.9992 −0.999 2 −0.999
32 5 −0.999 −0.999 −0.9991 1 −0.999 −0.9991 −0.999 −0.999 −0.9991 −0.999 2 −0.999
33 2 −0.999 −0.999 −0.9991 2 −0.999 −0.9991 −0.999 −0.999 −0.9992 −0.999 2 −0.999
34 3 −0.999 −0.999 −0.9994 3 −0.999 −0.9993 −0.999 −0.999 −0.9992 −0.999 3 −0.999

Numerical Taxonomic Study of Some Tribes of Gramineae from Egypt 10
Sp.17 Sp. 18 Sp. 19 Sp. 20Sp. 21 Sp. 22 Sp. 23Sp. 24Sp. 25Sp. 26Sp. 27Sp. 28Sp. 29Sp. 30 Sp. 31 Sp. 32 Sp. 33Sp. 34
1 9 8 1 1 3 3 6 10 12 2 1 9 4 5 5 4 2 4
2 2 1 1 2 2 2 1 1 2 1 1 1 1 1 1 1 1 1
3 11 3 1 10 14 7 4 3 15 1 3 13 1 3 4 17 1 4
4 0 0 0 0 0 0 0 0 1 0 1 1 2 1 1 1 1 0
5 0 0 0 0 0 0 0 0 2 0 1 1 3 1 1 1 1 0
6 0 0 0 0 0 0 0 0 2 0 2 2 2 2 1 2 2 0
7 0 0 0 0 0 0 0 0 5 0 4 4 6 1 5 4 4 0
8 10 16 1 1 6 5 1 6 6 2 8 12 9 13 2 2 2 4
9 1 1 1 1 1 1 1 1 1 2 1 1 1 1 2 3 5 2
10 1 1 1 1 2 3 5 1 2 3 1 2 1 1 2 4 5 2
11 1 1 1 1 1 1 1 1 1 1 1 1 1 1 5 1 1 1
12 18 1 2 1 5 13 −0.99911 −0.9998 3 1 1414 19 7 9 8
13 3 3 3 3 3 1 −0.9993 −0.9993 3 3 3 3 3 3 3 3
14 1 1 5 1 1 1 −0.9995 −0.9991 3 3 1 2 5 1 2 1
15 1 1 4 1 1 1 −0.9994 −0.9992 1 1 1 1 4 1 2 2
16 6 5 1 1 1 1 −0.9997 -0.9997 7 7 1 7 7 3 7 7
17 1 1 1 1 1 1 −0.9991 −0.9991 1 1 1 1 1 1 1 1
18 1 1 1 1 1 1 −0.9991 −0.9991 1 1 1 1 1 1 1 1
19 2 2 2 1 2 2 −0.9993 −0.9993 2 2 1 2 1 6 5 6
20 1 1 1 1 3 1 −0.9991 −0.9993 1 1 1 1 2 5 6 4
21 0 0 0 0 0 5 −0.9990 −0.9990 0 0 0 0 0 0 0 0
22 2 1 2 1 1 1 −0.9991 −0.9994 6 6 1 2 2 3 6 5
23 1 1 0 1 1 2 −0.9990 −0.9995 2 2 1 1 0 5 4 6
24 7 8 1 1 1 1 −0.9990 −0.9990 0 0 1 0 0 6 0 0
25 2 3 4 3 4 4 −0.9994 −0.9993 3 3 5 3 1 5 6 3
26 2 −0.999 −0.999 −0.999 −0.999 −0.999 1 −0.999 −0.9991 −0.9991 −0.999−0.999 −0.999 −0.999 2 2
27 2 −0.999 −0.999 −0.999 −0.999 −0.999 2 −0.999 −0.9992 −0.9992 −0.999−0.999 −0.999 −0.999 2 2
28 3 −0.999 −0.999 −0.999 −0.999 −0.999 1 −0.999 −0.9994 −0.9991 −0.999−0.999 −0.999 −0.999 4 1
29 2 −0.999 −0.999 −0.999 −0.999 −0.999 1 −0.999 −0.9994 −0.9992 −0.999−0.999 −0.999 −0.999 1 2
30 2 −0.999 −0.999 −0.999 −0.999 −0.999 4 −0.999 −0.9993 −0.9991 −0.999−0.999 −0.999 −0.999 5 3
31 1 −0.999 −0.999 −0.999 −0.999 −0.999 1 −0.999 −0.9994 −0.9992 −0.999−0.999 −0.999 −0.999 3 2
32 3 −0.999 −0.999 −0.999 −0.999 −0.999 2 −0.999 −0.9993 −0.9992 −0.999−0.999 −0.999 −0.999 4 1
33 2 −0.999 −0.999 −0.999 −0.999 −0.999 2 −0.999 −0.9992 −0.9992 −0.999−0.999 −0.999 −0.999 2 2
34 3 −0.999 −0.999 −0.999 −0.999 −0.999 1 −0.999 −0.9993 −0.9993 −0.999−0.999 −0.999 −0.999 3 2

Numerical Taxonomic Study of Some Tribes of Gramineae from Egypt 11
Figure 2. Scatter-plot of 34 studied taxa plotted against the first factor by the second factor.
donax and 23. Panicum turgidum. 2) Group including the
rest 32 species. There are some characters; character of
trichomes type, annulus of the pollen and also the oper-
culum of the pollen grain (Table 3); which not been fit-
ted into the data matrix because they are of only one code,
so they were excluded in the analysis because they had
no variation in the matrix.
4. Discussion
In the present study a large number of grains macro- and
micro-morphological, anatomical and pollen grains char-
acters were scored and numerical methods (UPGMA and
PCA) were applied to study the relationship among
eleven Poaceae tribes and estimate the level of variation
within and among these tribes. UPGMA gives insight
into degree of similarity among the studied species and
whether they form groups // clusters and gives an indica-
tion of the level of variation within and between tribes.
PCA reflects which characters are important on the axes,
and indicates the significant characters based on the
highest factor score (Table 5). Therefore it becomes
clear which characters cause the separation between
groups and can be useful to distinguish taxa. Pollen
grains showed the most powerful significant characters,
whereas all characters have been recorded are of high
factor scores. Generally, our results show congruence
between the UPGMA clustering and PCA analysis in
suggesting two main groups and five subgroups which
included the distribution of eleven tribes studied.
Our UPGMA results show that the tribe Andropo-
goneae is separated in one branch of the cladistic tree, the
tribe Aristidieae is separated in three branches of the tree
through three different subclads // subgroups. The tribe
Arundineae with two species is separated in two
branches of different clads in the tree. The tribe Aveneae
is separated in five branches of the tree while the tribe
Brachypodieae is separated in one branch and the
Bromeae is separated in two branches within two differ-
ent subclads. The Eragrostideae separated in four
branches and the Paniceae separated in three branches.
The Stipeae separated in four branches in only two clads
while the Triticeae separated in three branches in also
two clads. All the mentioned species, tribes, major clades
and subclads are arranged as the following:
The first main clads // groups of two species 6. Arundo
donax (Tribe: Arundineae) and 23. Panicum turgidum
(Tribe: Paniceae). While the second main group includes
a large variety of taxa from different tribes; 32 species of
tribes: Andropogoneae, Aristideae, Arundineae, Aveneae,
Brachypodieae, Bromeae, Eragrostideae, Paniceae, Sti-
peae and Triticeae. These tribes are separated through
five distinct subgroups: 1) Sub-clad of species 25. La-
marckia aurea, 28. Oryzopsis miliacea, 12. Polypogon
monspeliensis, 18. Eragrostis cilianensis and 5. Stipa-
grostis ciliata belonging to Poeae, Stipeae, Aveneae, Er-
agrostideae, and Aristideae. 2) Sub-clad only of species

Numerical Taxonomic Study of Some Tribes of Gramineae from Egypt 12
Table 5. Factor loadings showed the most intrinsic characters enhanced separations of the studied species.
Factor Loadings
Rotation: Un-rotated
Extraction: Principal components
Characters
Factor 1 Factor 2 Factor 3
1. Fruit shape 1.578348 0.871665 −1.24653
2. Coloring mode −0.23281 0.691848 0.173174
3. Color type 2.27393 2.051546 −3.32845
4. Trichome presence −0.54117 0.571291 0.467989
5. Hair length −0.50498 0.545862 0.444604
6. Hair coloring −0.36187 0.617713 0.489408
7. Position of attaching 0.194314 0.461728 0.273855
8. Fruit surface sculpture 2.058189 2.095991 −0.0891
9. Fruit weight (mg) −0.15868 −0.01955 0.347091
10. Fruit wide (mm) 0.125955 0.069094 −0.25377
11. Fruit length (mm) −0.18849 0.39183 0.305645
12. Section outline shape 3.173544 −3.89842 −0.03563
13. Hull cells type 0.443799 0.150657 1.117598
14. Aleurone cells shape 0.12138 0.430593 1.567286
15. Aleurone cells orientation 0.052685 0.232461 1.360735
16. Scutellum shape 1.010615 −1.21997 0.271807
17. Endosperm differentiation −0.33176 0.452823 0.673013
18. Endosperm ty pe −0.33784 0.490211 0.698834
19. Section wide (µm) 0.296755 −0.18378 0.934019
20. Section length (µm) 0.097103 −0.48171 0.569057
21. Hull cells thickness (µm) −0.60307 0.690619 0.471997
22. Seed coat thickness (µm) 0.18211 −0.36414 0.73636
23. Aleurone layer thickness (µm) 0.130375 −0.39784 0.654374
24. Scutellum cells thickness (µm) −0.18054 1.09338 0.69615
25. Endosper m thickness (µm) 0.764397 −0.05966 1.460169
26. Pollen class −1.07075 −0.2744 −0.64054
27. Pollen siz e −1.05331 −0.44013 −1.01514
28. Pollen shap e −0.92265 −0.79847 −0.89728
29. Pollen surface sculpture −0.98552 −0.54043 −0.84148
30. Annulus thickness (µm) −0.96595 −0.83463 −1.30983
31. Pore diameter (µm) −1.02089 −0.78764 −0.8329
32. Pollen wall thickness (µm) −1.00316 −0.51699 −0.97543
33. Sexine thickness (µm) −1.06557 −0.39242 −1.01249
34. Nexine thickness (µm) −0.97452 −0.69914 −1.23458
Percentage per PCA 16.54 4.34 3.13
Percentage for total variation for the three factors extracted 24.01 %
*PCA: Principal Component Analysis

Numerical Taxonomic Study of Some Tribes of Gramineae from Egypt 13
32. Aegilops kotshyi belongs to Triticeae. 3) Sub-clad of
33. Aegilops ventricosa, 34. Hordium murinum Subsp.
Leporinum, 26. Lolium perenne, 15. Bromus scoparius,
13. Brachypodium distachym and 9. Avena fatua be-
longing to Triticeae, Poeae, Bromeae, Brachypodieae and
Aveneae. 4) Sub-clad of species 29. Stipa capensis, 24.
Dactylis glomerata, 30. Stipa lagascae, 14. Bromus
rubens, 22. Echinochloa colona, 16. Coelachyrum bervi-
folium, 7. Schismus arabicus, 31. Stipa parviflora and 3.
Aristida funiculata belonging to Stipeae, Poeae, Bromeae,
Paniceae, Eragrostideae, Arundineae and Aristideae. 5)
Sub-clad of species 27. Poa annua, 11. Polypogon mari-
timus, 19. Eragrostis minor, 10. Phalaris minor, 8. Avena
barbata, 4. Aristida mutabilis, 21. Cenchrus ciliaris, 20.
Leptochloa fusca and Aristida adscensionis within tribes:
Poeae, Aveneae, Eragrostideae, Paniceae and Aristideae.
Several various monophyletic species which regarded
as sister-groups are distinct within five subclads men-
tioned. Firstly, in the tree (Figure 1) Paniceae and Arun-
dineae are a two-species sister-group to the rest whole
cluster of the tree, on the other hand, Eragrostideae and
Andropogoneae are another two-species sister-group
within the second branch of the second major clad in the
tree. Moreover, other species within Paniceae, Er-
agrostideae and Arundineae are separated through some
different subclads; thus Andropogoneae alone can be
conspicuously differentiated from other tribes by means
of its characteristic features for the fruit morphology,
fruit anatomy and pollen grains morphology. Secondly,
different tribes consume sister-grouping within each of
the five subclads distinguished. Tribe Poeae conform a
monophyletic sister-group in subclad (1) in a cluster of
Stipeae, Aveneae, Eragrostideae and Aristideae and in
subclad (5) in a cluster of Stipeae, Eragrostideae, Aris-
tideae, Bromeae, Paniceae and Arundineae. Therefore,
Poeae is preferably separated from these tribes depending
on its own marked pollen grains characters.
Moreover, the Triticeae shows a distinct variation that
can aid the comparison of the relationships between
Triticeae, Bromeae and Brachypodieae revealed by [16],
where they suggested that the Brachypodieae is the sister
group of the Triticeae while the Bromeae is the sister
group of the Brachypodieae plus the Triticeae. Brachy-
podium is the sister-group of a clad including both Bro-
mus and the Triticeae. While, [5] illustrated that the rela-
tionships between Bromus and the Triticeae is unre-
solved, so there is a possibility that the Triticeae is a
non-monophyletic group. Meanwhile, in our results, the
Triticeae is a monophyletic sister-group to the neighbor-
ing clad of Triticeae, Poeae, Bromeae, Brachypodieae
and Aveneae (sub-clad 3). This clad which can be sepa-
rated conspicuously through the Aveneae which is a
monophyletic branch through Avena fatua, in addition to
the separation of Poeae among tribes of subclads 1 and 5.
Thus the Triticeae, Bromeae and Brachypodieae are
closely related as confirmed by their palynological simi-
larity, in addition to the compatibility of the fruit mor-
phological (Table 5) that enhanced the understanding of
the degree of similarity between taxa of these tribes. The
Stipeae is a sister-group of the sub-clad (4), with exclud-
ing tribes Poeae, Bromeae, Paniceae, Arundineae and
Eragrostideae from this sub-clad, thus the Stipeae is
separated from the Aristideae and also the similarity de-
gree between them can be conducted to the characters of
the fruit morphology and pollen grain morphology illus-
trated in Table 5. Therefore, the applied methods of
UPGMA and PCA can be used to study the variation
within the tribe and the tribes in the family to determine
the relationships between genera and tribes. Our results
revealed there is a much separation between tribes An-
dropogoneae, Arundineae, Aristideae, Stipeae, Poeae and
Eragrostideae. However, tribes Triticeae, Bromeae and
Brachypodieae showed much closer relationships. In
addition to the consideration of those tribes Aveneae,
Eragrostideae and Stipeae are the most heterogeneous
tribes because the taxa of these tribes found to be inter-
spersed with taxa from tribes Poeae, Paniceae and Aris-
tideae.
REFERENCES
[1] W. D. Clayton and S. A. Renvoize, “Grasses of the
World,” Genera Graminum, Her Majesty’s Stationary Of-
fice, London, 1986.
[2] H. P. Linder and P. J. Rudall, “The Evolutionary History
of Poales,” Annual Reviews in Ecology and Systematics,
Vol. 36, No. 1, 2005, pp. 107-124.
doi:10.1146/annurev.ecolsys.36.102403.135635
[3] Grass Phylogeny Working Group, “Phylogeny and Sub-
familial Classification of the Grasses (Poaceae),” Annals
of the Missouri Botanical Garden, Vol. 88, No. 3, 2001,
pp. 373-457. doi:10.2307/3298585
[4] K. W. Hilu and K. Wright, “Systematics of Gramineae: A
Cluster Analysis Study,” Taxon, Vol. 31, No. 1, 1982, pp.
9-36. doi:10.2307/1220585
[5] K. W. Hilu and L. A. Alice, “A Phylogeny of Chlori-
doideae (Poaceae) Based on matK Sequences,” System-
atic Botany, Vol. 26, No. 2, 2001, pp. 386-405.
[6] O. Seberg and S. Frederiksen, “A Phylogenetic Analysis
of the Monogenomic Triticeae (Poaceae) Based on Mor-
phology,” Botanical Journal of the Linnean Society, Vol.
136, No. 1, 2001, pp. 75-97.
doi:10.1111/j.1095-8339.2001.tb00557.x
[7] C. L. McIntyre, “Variation in Isozyme Loci in Triticeae,”
Plant Systematics and Evolution, Vol. 160, No. 1-2, 1988,
pp. 123-142. doi:10.1007/BF00936714
[8] J. V. Monte, C. L. McIntyre and J. P. Gustafson, “Analy-
sis of Phylogenetic Relationships Using RFLPs,” Theo-

Numerical Taxonomic Study of Some Tribes of Gramineae from Egypt 14
retical and Applied Genetics, Vol. 86, No. 5, 1993, pp.
649-655. doi:10.1007/BF00838722
[9] R. J. Mason-Gamer, E. A. Kellogg, “Chloroplast DNA
Analysis of the Monogenomic Triticeae: Phylogenetic
Implications and Genome-Specific Markers,” In: J. J. Jau-
har, Ed., Methods of genome analysis in plants. Boca
Raton, CRC Press, Florida, 1996, pp. 301-325.
[10] E. A. Kellogg and R. Apple, “Intraspecific and Inter-
specific Variation in 5S RNA Genes are Decoupled in
Diploid Wheat Relatives,” Genetics, Vol. 140, No. 1,
1995, pp. 325-343.
[11] W. K. Taia, “Modern Trends in Plant Taxonomy,” Asian
Journal of Plant Sciences, Vol. 4, No. 2, pp. 2005, 184-
202.
[12] L. R. Parenti, “A Phylogenetic Analysis of the Land
Plants,” Biological Journal of the Linnean Society, Vol.
13, No. 3, 1980, pp. 225-242.
doi:10.1111/j.1095-8312.1980.tb00084.x
[13] B. R. Baum, “A Phylogenetic Analysis of the Tribe
Triticeae (Poaceae) Based on Morphological Characters
of the Genera,” Canadian Journal of Botany, Vol. 61, No.
2, 1983, pp. 518-535.
[14] J. Felsenstein, “Alternative Methods of Phylogenetic
Inference and Their Interrelationship,” Systematic Zool-
ogy, Vol. 28, No. 1, 1979, pp. 49-62.
doi:10.2307/2412998
[15] K. Bremer, “Gondwanan Evolution of the Grass Alliance
of Families (Poales),” Evolution, Vol. 56, No. 7, 2002, pp.
1374-1387.
[16] R. R. Sokal and C. D. Michener, “A Statistical Method
for Evaluating Systematic Relationships,” University of
Kansas Scientific Bulletin, Vol. 28, 1958, pp. 1409-1438.
[17] A. K. Osman, “Numerical Taxonomic Study of Some
Tribes of Compositae (Subfamily Asteroideae) from
Egypt,” Pakistan Journal of Botany, Vol. 43, No. 1, 2010,
pp. 171-180.
[18] O. Seberg, S. Frederiksen, C. Baden and I. Linde-Laursen,
“Peridictyon, a New Genus from the Balkan Peninsula,
and Its Relationship with Festucopsis (Poaceae),” Will-
denowia, Vol. 21, No. 1-2, 1991, 87-104.
[19] S. Frederiksen and O. Seberg, “Phylogenetic Analysis of
the Triticeae (Poeae),” Hereditas, Vol. 116, No. 1-2, 1992,
pp. 15-19. doi:10.1111/j.1601-5223.1992.tb00198.x