bs_bs_banner
Botanical Journal of the Linnean Society, 2013, 172, 555–571. With 3 figures
Comparison of pollen grain morphological features of
selected species of the genus Crataegus (Rosaceae) and
their spontaneous hybrids
DOROTA WROŃSKA-PILAREK1*, JAN BOCIANOWSKI2 and
ANDRZEJ M. JAGODZIŃSKI3,4
1
Department of Forestry Natural Foundations, Poznań University of Life Sciences, Wojska Polskiego
71 d, 60-625 Poznań, Poland
2
Department of Mathematical and Statistical Methods, Poznań University of Life Sciences, Wojska
Polskiego 28, 60-637 Poznań, Poland
3
Institute of Dendrology, Polish Academy of Sciences, Parkowa 5, 62-035 Kórnik, Poland
4
Department of Game Management and Forest Protection, Poznań University of Life Sciences, Wojska
Polskiego 71 c, 60-625 Poznań, Poland
Received 26 March 2012; revised 31 August 2012; accepted for publication 15 January 2013
The aim of our study was to verify, on the basis of statistical analyses of nine quantitative morphological features of
pollen grains, the hypothesis that pollen grains of three parental species of Crataegus (C. laevigata, C. monogyna,
C. rhipidophylla) differed from the pollen of three spontaneous hybrids of these species (C. ¥ macrocarpa, C. ¥ media,
C. ¥ subsphaericea). Contrast analysis revealed that a majority of the pollen features of hybrid species were
characterized by significantly higher mean values than those of parental species. Analysis of pollen shape classes
indicated that the parental species clearly differed from each other in contrast with hybrids, which were characterized
by a similar proportion of pollen in individual pollen shape classes. Statistical analyses showed that the pollen grains
of two parental species, C. laevigata and C. monogyna, were most similar to one another. Pollen grains of typical
C. rhipidophylla were similar to the pollen of hybrids and the mean values of almost all studied pollen features [P,
E, Exp, Exp/P, Le, d, d/E (PAI)] of C. rhipidophylla var. rhipidophylla were intermediate between those of
C. monogyna and C. rhipidophylla var. lindmanii. This corroborates Zieliński’s conjecture that C. rhipidophylla is
probably an old, conserved hybrid between C. monogyna and C. calycina (= C. lindmanii = C. rhipidophylla var.
lindmanii). According to the analysis of canonical variables, C. ¥ macrocarpa and C. ¥ media pollen grains were most
similar. C. ¥ subsphaericea and C. rhipidophylla var. rhipidophylla and C. rhipidophylla var. lindmanii formed
another pair (group), and C. monogyna and C. laevigata constituted separate ‘single species groups’. © 2013 The
Linnean Society of London, Botanical Journal of the Linnean Society, 2013, 172, 555–571.
ADDITIONAL KEYWORDS: interspecific variability – pollen morphology – pollen variability – statistical
analysis.
INTRODUCTION
The genus Crataegus L. (hawthorns) comprises
approximately 150–1200 species (depending on the
species concept employed), distributed mainly in the
Northern Hemisphere. Among these are 50–100 Old
World Crataegus spp. occurring in Europe, northern
Africa and western Asia (Christensen, 1992).
*Corresponding author. E-mail: pilarekd@up.poznan.pl
Crataegus is a well-defined genus, traditionally
included in the tribe Crataegeae, subfamily Maloideae, of Rosaceae (Phipps, 1983, 1988; Robertson
et al., 1991; Christensen, 1992; Christensen & Janjic,
2006). According to Campbell et al. (2007), Crataegus
is included in the subtribe Pyrinae (formerly subfamily Maloideae). Christensen & Zieliński (2008)
included Crataegus in the tribe Pyreae. The intrageneric classification of maloid genera has been studied
by Phipps (1983), Campbell & Dickinson (1990) and
Phipps et al. (1991).
© 2013 The Linnean Society of London, Botanical Journal of the Linnean Society, 2013, 172, 555–571
555
556
D. WROŃSKA-PILAREK ET AL.
The taxonomy of Crataegus is problematic owing to
biological and historical factors (Dönmez, 2008). As in
other large maloid genera, frequent interspecific
hybridization, extensive polyploidy and apomixes all
have the potential to blur the boundaries between
species (Byatt, Ferguson & Murray, 1977; Dönmez,
2004; Talent & Dickinson, 2005; Dickinson, Lo &
Talent, 2007; Ennos et al., 2012). Christensen (1992)
also emphasized that Crataegus has a complicated
taxonomic history, because numerous species have
been described by several authors and the circumscriptions of the species have varied widely. In his
studies, Christensen (1992, 1997) presented a synthesis of hawthorns in Europe and proposed a decrease
in the numbers of recognized taxa and a new nomenclature. Within the boundaries of Crataegus, Phipps
et al. (1990) and Phipps, O’Kennon & Lance (2003)
recognized 40 series in about 15 readily distinguishable sections, although there was considerable variation in each of these groups. Christensen (1992)
distinguished four series and three subseries in
section Crataegus.
During the course of our study, the pollen morphology of three Crataegus spp. [C. laevigata (Poir.) DC.,
C. monogyna Jacq., C. rhipidophylla Gand.] and three
spontaneous hybrids of these species (C. ¥ macrocarpa
Hegetschw., C. ¥ media Bechst., C. ¥ subsphaericea
Gand.) were investigated (Table 1). According to
Christensen’s (1992) classification of the genus, the
species analysed belong to the large section Crataegus,
series Crataegus, subseries Erianthae (Pojarkova)
Christensen (C. laevigata) and subseries Crataegus
(all remaining taxa; Table 1).
It is evident from the palynological studies of Crataegus carried out so far that their pollen grains are
very variable. Pollen grains of the described species
are in radially symmetrical monads. Two pollen
types are most frequent (tricolpate and tricolporate),
but trisyncolporate, tetracolporate, pericolporate and
inaperturate pollen types also occur. The length of the
polar axis (P) varies greatly, ranging from 15 to 65 mm;
similarly, the equatorial diameter (E) can range from
15 to 47 mm. Pollen grain sizes are mostly medium,
rarely small or large. The pollen shape (P/E ratio)
varies from suboblate, through spheroidal to prolate.
Many different exine sculpture types can be found
in Crataegus, including striate-perforate, regulateperforate, rugulate-striate-perforate and microreticulate (Reitsma, 1966; Byatt, 1976; Kuprianowa &
Alyoshina, 1978; Eide, 1981; Fedoronchuk & Savitskii,
1985; Dickinson & Phipps, 1986; Xin, Zhang & Wang,
1986; Hebda, Chinnappa & Smith, 1988; González
Romano & Candau, 1989; Hebda & Chinnappa, 1990;
Moore, Webb & Collinson, 1991; Christensen, 1992;
Hebda & Chinnappa, 1994; Zhou, Wie & Wu, 2000;
Dönmez, 2008).
Palynological studies on intraspecific hybrids have
generally focused on artificial hybrids and have rarely
been concerned with spontaneous hybrids found in
nature. Such investigations have been based on the
analysis of selected, single quantitative or qualitative
features of pollen grains (usually comprising pollen
size and exine sculpture) of individual interspecific
hybrids. The results obtained are not consistent.
Some authors have maintained that the pollen sizes
of interspecific, artificial hybrids are significantly
larger than those of their parents and that this is
associated with ploidy (Nair, Katiar & Srivastava,
1977; Srivastava, Pal & Nair, 1977; Rizaeva,
Akhmedova & Abdullaev, 1985; Hossain, Inden &
Asahira, 1990; Van der Walt & Littlejohn, 1996;
Franssen et al., 2001; Rhee et al., 2005), whereas
others have claimed that hybrids (usually polyploids)
can have pollen sizes similar or smaller than those of
their parents (Olsson, 1974; Srivastava, 1978; Rao,
Kumar & Nair, 1979; Ohashi, Hoshi & Iketani, 1991;
Chaturvedi, Ram & Pal, 1993; Delaporte, Conran &
Sedgley, 2001; Franssen et al., 2001; Jiang et al.,
2001; Lu et al., 2002; Dönmez, 2008; Karlsdóttir et al.,
2008).
The objective of our study was to verify, based on
the statistical analyses of nine quantitative features
of 900 pollen grains, whether and to what degree
pollen grains of the three parental species studied
(C. laevigata, C. monogyna, C. rhipidophylla) differ
from the pollen grains of three spontaneous hybrids
(C. ¥ macrocarpa, C. ¥ media, C. ¥ subsphaericea).
MATERIAL AND METHODS
The taxonomic classification of Crataegus and the
studied taxa was adopted from Christensen (1992),
with the exception of C. ¥ kyrtostyla Fingerh., considered to be a hybrid between C. monogyna and C. rhipidophylla. As explained by Christensen & Zieliński
(2008), in the past, C. monogyna ¥ rhipidophylla has
often been referred to as C. ¥ kyrtostyla Fingerh.
(1829) [see, for example, Franco (1968) and Christensen (1992)], but recently an original specimen of
C. kyrtostyla was located at Bonn and, although the
specimen is in a fairly poor condition, it is evident that
Lippert’s (1978) contention that C. kyrtostyla is synonymous with C. monogyna is correct (Christensen,
1997). As a consequence, C. subsphaericea Gand.
(1871) is now the oldest legitimate name for the hybrid
C. monogyna ¥ rhipidophylla (Christensen, 1997).
Therefore, following Christensen & Zieliński (2008),
we adopted the name C. ¥ subsphaericea for this
taxon.
The palynological terminology follows Punt et al.
(2007) and Hesse et al. (2009). The study was conducted on three parental species of Crataegus and
© 2013 The Linnean Society of London, Botanical Journal of the Linnean Society, 2013, 172, 555–571
No.
Series
Subseries
Species/taxon
Locality
Position
Collector, herbarium
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Crataegus
Crataegus
Crataegus
Crataegus
Crataegus
Crataegus
Crataegus
Crataegus
Crataegus
Crataegus
Crataegus
Crataegus
Crataegus
Crataegus
Crataegus
Crataegus
Crataegus
Crataegus
Erianthae
Erianthae
Erianthae
Erianthae
Erianthae
Crataegus
Crataegus
Crataegus
Crataegus
Crataegus
Crataegus
Crataegus
Crataegus
Crataegus
Crataegus
Crataegus
Crataegus
Crataegus
Crataegus laevigata (Poir.) DC.
Crataegus laevigata (Poir.) DC.
Crataegus laevigata (Poir.) DC.
Crataegus laevigata (Poir.) DC.
Crataegus laevigata (Poir.) DC.
Crataegus monogyna Jacq.
Crataegus monogyna Jacq.
Crataegus monogyna Jacq.
Crataegus monogyna Jacq.
Crataegus monogyna Jacq.
Crataegus rhipidophylla Gand.
Crataegus rhipidophylla Gand.
Crataegus rhipidophylla Gand.
Crataegus rhipidophylla Gand.
Crataegus rhipidophylla Gand.
Crataegus ¥ macrocarpa Hegetschw.
Crataegus ¥ macrocarpa Hegetschw.
Crataegus ¥ macrocarpa Hegetschw.
51°55′6.239″N, 17°2′7.548″E
50°17′49.344″N, 16°39′7.848″E
51°18′26.424″N, 17°3′31.14″E
51°28′18.911″N, 16°54′30.383″E
50°24′2.304″N, 18°3′49.463″E
49°28′51.744″N, 22°42′3.671″E
52°14′53.951″N, 17°5′17.951″E
53°9′44.136″N, 22°48′13.355″E
49°32′48.983″N, 21°50′58.092″E
52°52′34.103″N, 18°41′36.707″E
49°28′52.968″N, 20°1′53.508″E
49°44′30.551″N, 19°35′43.403″E
53°18′0.36″N, 17°26′51.144″E
50°17′45.312″N, 16°52′25.86″E
50°17′49.344″N, 16°39′7.848″E
52°16′53.328″N, 17°1′34.067″E
50°26′47.471″N, 16°15′54.431″E
53°29′26.303″N, 16°19′7.536″E
Kaczmarek, KOR
Kosiński, KOR
Gostyńska, KOR
Gostyńska, KOR
Browicz, KOR
Jakuszewski, KOR
Gostyńska, KOR
Niedziałek, KOR
Gostyńska, KOR
Gostyńska, KOR
Browicz, Zieliński, KOR
Browicz, Zieliński, KOR
Boratyński, KOR
Kosiński, KOR
Kosiński, KOR
Gostyńska, KOR
Browicz, Gostyńska, KOR
Gostyńska, KOR
19
20
21
22
23
24
25
26
27
28
29
30
Crataegus
Crataegus
Crataegus
Crataegus
Crataegus
Crataegus
Crataegus
Crataegus
Crataegus
Crataegus
Crataegus
Crataegus
Crataegus
Crataegus
Crataegus
Crataegus
Crataegus
Crataegus
Crataegus
Crataegus
Crataegus
Crataegus
Crataegus
Crataegus
Crataegus ¥ macrocarpa Hegetschw.
Crataegus ¥ macrocarpa Hegetschw.
Crataegus ¥ media Bechst.
Crataegus ¥ media Bechst.
Crataegus ¥ media Bechst.
Crataegus ¥ media Bechst.
Crataegus ¥ media Bechst.
Crataegus ¥ subsphaericea Gand.
Crataegus ¥ subsphaericea Gand.
Crataegus ¥ subsphaericea Gand.
Crataegus ¥ subsphaericea Gand.
Crataegus ¥ subsphaericea Gand.
Ostrowo, Wielkopolskie Province
Bystrzyca Kłodzka, Dolnośla˛skie Province
Kocie Góry, Dolnośla˛skie Province
Żmigród, Dolnośla˛skie Province
Mechnica, Opolskie Province
Krościenko, Podkarpackie Province
Kórnik, Wielkopolskie Province
Leśniki, Podlaskie Province
Rymanowa Zdrój, Podkarpackie Province
Aleksandrów, Kujawsko-Pomorskie Province
Nowy Targ, Małopolskie Province
Sucha Beskidzka, Małopolskie Province
Czarmuń, Kujawsko-Pomorskie Province
Stronie Śla˛skie, Dolnośla˛skie Province
Bystrzyca Kłodzka, Dolnośla˛skie Province
Borówiec, Wielkopolskie Province
Kudowa, Dolnośla˛skie Province
Kamienna Góra, Zachodniopomorskie
Province
Dubiecko, Podkarpackie Province
Życzanów, Małopolskie Province
Pre˛gowo, Pomorskie Province
Kulin, Kujawsko-Pomorskie Province
Ciechocinek, Kujawsko-Pomorskie Province
Milicz, Dolnośla˛skie Province
Chapsko, Kujawsko-Pomorskie Province
Sobieszyn, Lubelskie Province
Grzegorzowice, Świe˛tokrzyskie Province
Smogorzów, Opolskie Province
Kulin, Kujawsko-Pomorskie Province
Wrotkowo, Warmińsko-Mazurskie Province
49°49′32.808″N, 22°23′24.467″E
49°30′2.988″N, 20°40′37.991″E
54°15′17.855″N, 18°28′46.019″E
52°40′12.215″N, 19°7′35.471″E
52°52′51.671″N, 18°47′38.292″E
51°31′51.815″N, 17°16′32.412″E
52°37′37.559″N, 17°56′4.848″E
51°35′36.815″N, 22°9′44.424″E
50°52′19.991″N, 21°9′6.011″E
51°9′23.615″N, 17°45′33.911″E
52°40′12.215″N, 19°7′35.471″E
54°14′43.223″N, 22°20′49.379″E
Browicz, Gostyńska, KOR
Gostyńska, KOR
Gostyńska, KOR
Gostyńska, KOR
Bojarczuk, KOR
Gostyńska, KOR
Gostyńska, KOR
Boratyński, KOR
Gostyńska, KOR
Gostyńska, KOR
Gostyńska, KOR
Naklicki, KOR
POLLEN FEATURES OF SELECTED CRATAEGUS TAXA
© 2013 The Linnean Society of London, Botanical Journal of the Linnean Society, 2013, 172, 555–571
Table 1. List of localities of Crataegus taxa studied. KOR – Herbarium of the Institute of Dendrology, Polish Academy of Sciences, Kórnik, Poland
557
558
D. WROŃSKA-PILAREK ET AL.
three spontaneous hybrids of these species. The list of
the taxa analysed is shown in Table 1.
Pollen samples were collected in the Herbarium of
the Institute of Dendrology of the Polish Academy of
Sciences in Kórnik (52°14′12″N, 17°05′55″E) – KOR
(Poland). Several randomly selected flowers were collected from each individual shrub. The plant material
was collected by specialists, experienced dendrologists
from the Institute of Dendrology of the Polish
Academy of Sciences in Kórnik. We have selected
herbarium specimens representing different taxa,
and the specimens were re-reviewed by Professor
Jerzy Zieliński, a leading dendrologist specializing in
Rosaceae, including hawthorns (Zieliński, 1977, 1982;
Christensen & Zieliński, 2008).
Each shrub was represented by 30 pollen grains.
Plant material was derived from 30 natural sites
located in Poland, so that each taxon had five localities (Table 1).
All samples were acetolysed according to the
method described by Wrońska-Pilarek (1998, 2011)
and Wrońska-Pilarek & Jagodziński (2011). The
acetolysing mixture was made up of nine parts of
acetic acid anhydride and one part of concentrated
sulphuric acid, and the process of acetolysis lasted
2.5 min. Observations were carried out with both
a light microscope (Biolar 2308, Nikon HFX-DX)
and a scanning electron microscope (SEM) (Hitachi
S-3000N).
A pollen sample consisted of 30 pollen grains from
each shrub. In total, 900 pollen grains were measured. Only mature, correctly formed pollen grains
were studied. However, malformed pollen grains were
also noticed in the samples, and their percentage was
determined by examining 400 pollen grains (four randomly selected samples of 100 pollen grains) for each
taxon.
Nine quantitative features were analysed, i.e. the
length of the polar axis (P), equatorial diameter (E),
length of the ectocolpi (Le), thickness of the exine
along the polar axis (Exp), distance between the
apices of two colpi (d) and four ratios: P/E, Le/P, Exp/P
and d/E. Two qualitative features were also analysed:
pollen outline and shape.
The normality of P, E, P/E, Exp, Exp/P, Le, Le/P, d
and d/E data distributions was tested using Shapiro–
Wilk’s normality test (Shapiro & Wilk, 1965). A oneway analysis of variance (ANOVA) was carried out to
determine the effects of species on P, E, P/E, Exp,
Exp/P, Le, Le/P, d and d/E. The minimal and maximal
values of the characteristics and the arithmetic
means and coefficients of variation were calculated.
When critical differences were noted, multiple comparisons were carried out using least-significant
difference (LSD) for each trait; based on this, homogeneous groups (not significantly different from each
other) were determined for the traits analysed. The
relationships between P, E, P/E, Exp, Exp/P, Le, Le/P,
d and d/E were estimated using Pearson’s correlation
coefficients (Sokal & Rohlf, 1995). The Mahalanobis
distance (Mahalanobis, 1936) between the species,
determined using P, E, P/E, Exp, Exp/P, Le, Le/P, d
and d/E, can be treated as the phenotypic distance
between the species. Graphical distribution of the
species and hybrids on a plane, including all the
features jointly, was possible through the application
of canonical variable analysis. All data analysis was
performed using the statistical package GenStat v.
10.1 (GenStat, 2007).
RESULTS
GENERAL
POLLEN MORPHOLOGICAL DESCRIPTION
A description of the pollen grain morphology of Crataegus spp. studied is given below and illustrated
with SEM photographs (Figs 1–15). Pollen grains
occur as monads. The pollen grains of Crataegus
spp. were tri-zonocolporate (Figs 1–8). All analysed
pollen grains were of medium size (25.1–50 mm;
Table 2) according to Erdtman’s (1952) pollen size
classification.
The average length of the polar axis (P) for the
three parental Crataegus spp. was 36.50 mm (range,
28–46 mm) and for the three hybrids 39.07 mm (range,
30–50 mm). Parental species usually had smaller
pollen than hybrids. The exception was C. rhipidophylla, the pollen grains of which were relatively
large and equalled some of the hybrid pollen grains.
Among parental species, the smallest P features were
found for the pollen of C. laevigata (28 mm) and the
largest for C. rhipidophylla (46 mm). However, in
hybrids, the smallest P was found in C. ¥ macrocarpa
(30 mm) and the largest in C. ¥ subsphaericea (50 mm;
Table 2).
The mean length of the equatorial diameter (E) for
the parental Crataegus spp. was 33.20 mm (range,
24–46 mm) and for the three hybrids 36.04 mm (range,
26–46 mm). In parental species, the shortest equatorial
diameter (E) occurred in the pollen of C. monogyna
(24.0 mm), and the longest was in C. rhipidophylla
(46 mm). In hybrids, the shortest equatorial diameter
(E) was found in C. ¥ macrocarpa (26 mm) and the
longest in C. ¥ subsphaericea (46 mm; Table 2).
In all taxa studied, the outline in polar view
was mostly elliptic or circular, more rarely triangular
with obtuse apices, whereas, in equatorial view,
the outline was mostly elliptic, rarely circular
(Figs 1–5).
The mean P/E ratio for parental Crataegus spp. was
1.11, ranging from 0.81 in C. rhipidophylla to 1.67 in
C. monogyna, and for hybrids was 1.09, ranging from
© 2013 The Linnean Society of London, Botanical Journal of the Linnean Society, 2013, 172, 555–571
POLLEN FEATURES OF SELECTED CRATAEGUS TAXA
Figures 1–15. See caption on next page.
© 2013 The Linnean Society of London, Botanical Journal of the Linnean Society, 2013, 172, 555–571
559
560
D. WROŃSKA-PILAREK ET AL.
Figures 1–15. Polar and equatorial views, apertures and exine sculpture in scanning electron microscopy. Figures 1–3.
Crataegus laevigata, C. rhipidophylla, C. monogyna: pollen grains in equatorial views, colpi and exine sculpture.
Figures 4, 5. Crataegus rhipidophylla, C. ¥ media: pollen in polar views, three colpi and exine sculpture. Figure 6.
Crataegus laevigata: two pollen grains in equatorial and polar views, colpi and exine sculpture. Figure 7.
Crataegus ¥ macrocarpa: long colpus and exine sculpture. Figure 8. Crataegus ¥ macrocarpa: colporus and exine sculpture. Figure 9. Crataegus ¥ media: seven pollen grains in polar and equatorial views, colpori and exine sculpture.
Figures 10–15. Crataegus laevigata, C. monogyna, C. rhipidophylla, C. ¥ macrocarpa, C. ¥ media, C. ¥ subsphaericea:
striate exine sculpture with striae, grooves and perforations.
0.86 in C. ¥ subsphaericea to 1.77 in C. ¥ macrocarpa
(Table 2).
Pollen grains of the parental Crataegus spp. differed distinctly with respect to their pollen shape
(Table 3). Crataegus monogyna had the largest
number of elongated pollen grains from such shape
classes as prolate-spheroidal (50.7%) to subprolate
(33.3%). The proportion of the remaining classes was
low, ranging from 2.0 to 7.3%, and oblate pollen
grains were not found in this species. Pollen grains
of C. laevigata were primarily spheroidal (49.3%)
and oblate-spheroidal (20.7%). In addition, this
species was distinguished by having a relatively high
proportion of oblate pollen (12.7%), a moderate frequency of prolate-spheroidal pollen (16%) and a low
proportion of pollen grains from the subprolate class
(1.3%). No prolate pollen grains occurred in C. laevigata. Crataegus rhipidophylla was characterized
by the most uniform distribution of pollen in individual shape classes. Its pollen grains belonged
mainly to the following three shape classes: spheroidal (30%), prolate-spheroidal (26%) and subprolate
(24.7%). The remaining classes were less frequent
(1.3–10.7%; Table 3).
Pollen grains of the three hybrids, in contrast with
parental species, were characterized by similar
shapes. In all taxa, prolate-spheroidal pollen grains
were most numerous (44.7–48.7%) and grains from
the spheroidal and subprolate classes were also frequent, ranging from 20.7 to 28% and 19.3 to 26%,
respectively. Oblate-spheroidal and prolate pollen
grains occurred rarely (5.3–6.7%) and very rarely
(1.3–2%), respectively. No oblate pollen grains were
found in any of the analysed species (Table 3). In
hybrids, prolate-spheroidal pollen grains were most
numerous (46.4%), pollen grains from subprolate
(22.9%) and spheroidal classes were frequent and
similar, whereas oblate-spheroidal (5.8%) and prolate
(1.6%) pollen grains were the least numerous. No
oblate pollen grains were found (Table 3).
The mean exine thicknesses were 0.69 mm (0.2–
2.0 mm) (parental species) and 0.82 mm (0.4–2.0 mm)
(hybrids; Table 2). The exine was thinnest in C. monogyna (0.2 mm) and thickest in C. laevigata and
C. rhipidophylla (2.0 mm). In hybrids, the exine thickness was slightly less variable, ranging from 0.4 mm
in C. ¥ media and C. ¥ subsphaericea to 2.0 mm in
C. ¥ macrocarpa.
The relative thickness of the exine (Exp/P ratio) for
parental species and hybrids was similar, and averaged 0.019 (0.009–0.040) and 0.021 (0.008–0.063;
Table 2), respectively. In all taxa studied, the exine
sculpture was striate with or without perforations.
Striae were long and more or less parallel to the colpus
(Figs 1, 7, 12) or short or medium length and interwoven, forming bends (Figs 2, 6, 10). Circular or elliptic
perforations with different, usually small, diameters
were found at the bottom of the grooves (Figs 9, 11, 15).
Pollen grains usually possessed three apertures
(colpori; Figs 4, 5). Colpi were arranged meridionally,
regularly, more or less evenly spaced, fusiform in
outline and were long: mean lengths of 32.45 mm
(24.00–46.00 mm) (parental species) and 35.82 mm
(24.00–48.00 mm) (hybrids) (Figs 1–3; Table 2). On
average, the length of the colpi constituted 89% of the
polar axis length in parental species and 92% in
hybrids. On average, species had slightly shorter colpi
than hybrids. Their width was variable and usually
greatest in the equatorial region.
The polar area index (PAI) averaged 0.20 (0.05–
0.40) in parental species and 0.15 (0.05–0.35) in
hybrids. The lowest mean values of this index in
parental species were recorded in C. rhipidophylla
(0.16) and the highest in C. monogyna (0.24); in
hybrids, mean values of PAI were less variable and
smaller, ranging from 0.15 in C. ¥ macrocarpa to 0.16
in C. ¥ subsphaericea (Table 2).
The percentage of deformed pollen grains in the
samples (400 grains per taxon) was variable and
ranged from 20 to 75% (Fig. 16). The results are
surprising because, on average, the share of deformed
pollen grains from parental species was higher (48%)
than in the hybrids (30%). The highest frequency of
deformed pollen was found in samples of C. rhipidophylla (75%) and the lowest in C. ¥ macrocarpa (20%).
In parental species, the lowest percentage of deformed
pollen grains occurred in C. monogyna (25%), an
intermediate percentage in C. laevigata (45%) and the
highest percentage in C. rhipidophylla (75%). In
hybrids, the percentages of deformed pollen grains
were 20% in C. ¥ macrocarpa, 25% in C. ¥
subsphaericea and 45% in C. ¥ media.
© 2013 The Linnean Society of London, Botanical Journal of the Linnean Society, 2013, 172, 555–571
POLLEN FEATURES OF SELECTED CRATAEGUS TAXA
Figures 1–15. Continued
© 2013 The Linnean Society of London, Botanical Journal of the Linnean Society, 2013, 172, 555–571
561
562
D. WROŃSKA-PILAREK ET AL.
Figures 1–15. Continued
© 2013 The Linnean Society of London, Botanical Journal of the Linnean Society, 2013, 172, 555–571
P (mm)
Taxon
C. laevigata
C. monogyna
C. rhipidophylla
C. ¥ macrocarpa
C. ¥ media
C. ¥ subsphaericea
LSD0.001
ANOVA P > F
Min–Max
28–40
30–44
32–46
30–46
32–46
30–50
E (mm)
Mean
CV (%)
33.73d
37.09c
38.67b
38.17bc
38.69b
40.35a
1.19
< 0.001
7.16
6.48
7.48
10.07
6.8
10.27
Exp (mm)
Taxon
C. laevigata
C. monogyna
C. rhipidophylla
C. ¥ macrocarpa
C. ¥ media
C. ¥ subsphaericea
LSD0.001
ANOVA P > F
Min–Max
0.4–2
0.2–1.6
0.4–2
0.6–2
0.4–2
0.4–2
Min–Max
26–38
24–40
26–46
26–44
28–44
30–46
P/E
Mean
CV (%)
Min–Max
Mean
CV (%)
31.35c
32.09c
36.15ab
35.55b
35.41b
37.17a
1.36
< 0.001
7.34
9.37
13.26
11.14
8.55
10.13
0.875–1.385
0.938–1.667
0.810–1.462
0.9–1.769
0.889–1.5
0.857–1.412
1.081b
1.166a
1.086b
1.080b
1.099b
1.089b
0.044
< 0.001
9.07
11.46
13.33
10.03
9.54
8.8
Mean
CV (%)
30.43d
31.95c
34.96b
34.85b
35.84ab
36.77a
1.42
< 0.001
8.31
10.68
10.6
12.28
10.33
11.8
Exp/P
Le (mm)
Mean
CV (%)
Min–Max
Mean
CV (%)
0.716c
0.492d
0.857ab
0.888a
0.833abc
0.739bc
0.137
< 0.001
53.69
51.3
47.77
40.46
45.22
47.04
0.01–0.059
0.005–0.04
0.009–0.059
0.013–0.063
0.009–0.063
0.008–0.044
0.021ab
0.013c
0.022ab
0.024a
0.022ab
0.018b
0.004
< 0.001
55.02
51.53
46.8
42.78
47.67
47.45
Le/P
d (mm)
Min–Max
Mean
C. laevigata
C. monogyna
C. rhipidophylla
C. ¥ macrocarpa
C. ¥ media
C. ¥ subsphaericea
LSD0.001
ANOVA P > F
0.8125–1
0.6–1
0.75–1
0.6842–1
0.75–1
0.56–1
0.902a
0.862b
0.904a
0.912a
0.925a
0.912a
0.023
< 0.001
CV (%)
4.48
9.39
6.36
5.77
6.48
6.05
Min–Max
2.0–12
4.0–12
2.0–12
2.0–12
2.0–10
2.0–10
24–38
24–40
28–46
26–44
24–44
28–48
d/E (PAI)
Mean
CV (%)
Min–Max
Mean
CV (%)
6.367b
7.647a
5.707bc
5.280c
5.287c
5.787c
0.725
< 0.001
29.81
23.87
34.9
39.51
33.33
31.69
0.063–0.353
0.105–0.4
0.045–0.333
0.05–0.3529
0.05–0.2632
0.05–0.2941
0.203b
0.239a
0.161c
0.148c
0.150c
0.157c
0.021
< 0.001
28.21
23.95
37.77
37.16
33.49
32.77
563
Taxon
Min–Max
POLLEN FEATURES OF SELECTED CRATAEGUS TAXA
© 2013 The Linnean Society of London, Botanical Journal of the Linnean Society, 2013, 172, 555–571
Table 2. Range (minimum–maximum, Min–Max), mean values and coefficient of variation (CV) of pollen features studied. One-way analyses of variance
(ANOVAs) were performed separately for each of the traits. Same letters indicate a lack of statistically significant difference between the analysed taxa according
to the least-significant difference (LSD) test (P < 0.001)
564
D. WROŃSKA-PILAREK ET AL.
Table 3. Percentage of pollen grains distributed among shape classes (P/E ratio) according to Erdtman’s (1952)
classification: oblate-spheroidal (0.89–0.99); spheroidal (1.00); prolate-spheroidal (1.01–1.14); subprolate (1.15–1.33);
prolate (1.34–2.00); perprolate (> 2.01)
Pollen shape class
Taxon
Oblate
Oblate-spheroidal
Spheroidal
Prolate-spheroidal
Subprolate
Prolate
C. laevigata
C. monogyna
C. rhipidophylla
C. ¥ macrocarpa
C. ¥ media
C. ¥ subsphaericea
12.7
20.7
2.0
10.7
6.7
5.3
5.3
49.3
6.7
30.0
28.0
20.7
21.3
16.0
50.7
26.0
44.7
46.0
48.7
1.3
33.3
24.7
19.3
26.0
23.3
7.3
7.3
1.3
2.0
1.3
1.3
C. xsubsphaericea
C. xmedia
C. xmacrocarpa
C. rhipidophylla
C. laevigata
C. monogyna
0
10
20
30
40
50
60
70
80
Deformed pollen grains (%)
Figure 16. Percentage of deformed pollen grains.
INTERSPECIFIC
VARIABILITY
Statistically significant differences were found among
the taxa examined for all nine pollen grain features
analysed (P < 0.001; Table 2). The results of ANOVAs
are summarized in Table 2.
Most of the observed quantitative features were
significantly correlated (Table 4). However, feature d
was not correlated with P, E, P/E, Exp or Exp/P, Exp
was not correlated with P, and Le/P was not correlated with Exp or Exp/P. Fourteen of 28 significantly
correlated pairs of features were characterized by
positive correlation coefficients (Table 4).
The analysis of phenotypic distance (Mahalanobis
distance) between the species revealed that pollen
grains of C. laevigata and C. monogyna were most
similar with respect to all the features examined
treated jointly (Mahalanobis distance 1.68), whereas
they differed significantly from the pollen grains of
C. rhipidophylla and from those of the hybrids
(Mahalanobis distances 2.07 and 2.04, respectively)
(Table 5). The three hybrids differed slightly from
one another and from C. rhipidophylla, and differed
significantly from the pollen of C. laevigata and
C. monogyna (Table 4). The diagram of the first two
canonical variables (Fig. 17) was used to divide the
taxa examined into four groups. The first group
comprised C. ¥ macrocarpa and C. ¥ media. The
second included C. ¥ subsphaericea and C. rhipidophylla. The other two groups included just one
species, C. monogyna or C. laevigata (Fig. 17). The
first two canonical variables accounted for 91.92% of
the total multivariate variability among species and
hybrids.
In the contrast analysis between parental species
and their hybrids (Table 6), for the majority of the
features analysed, hybrids were characterized by a
significantly higher mean of the given feature in
comparison with the mean of the parental form (i.e.
negative contrast value). In the case of P, E and Le
features, all hybrids were characterized by a significantly higher mean in comparison with the parental
forms (Table 6). For Exp, Exp/P and Le/P, two of
the three hybrids exhibited a significantly higher mean
of these features in comparison with the parental
forms. In the case of Exp and Exp/P, these included
C. ¥ macrocarpa (C. laevigata ¥ C. rhipidophylla) and
C. ¥ media (C. monogyna ¥ C. laevigata) and, for
Le/P, C. ¥ subsphaericea (C. monogyna ¥ C. rhipidophylla) and C. ¥ media. Considerably fewer hybrid/
feature combinations were found to have a lower mean
of the given feature in comparison with the mean of its
parental forms. This happened in all hybrids in the
case of d and d/E and in C. ¥ subsphaericea and
C. ¥ media for the P/E feature. Hybrids did not differ
significantly from the mean of the parental species in
only the following four cases: in C. ¥ macrocarpa for
P/E and Le/P features and in C. ¥ subsphaericea for
Exp and Exp/P features (Table 6).
© 2013 The Linnean Society of London, Botanical Journal of the Linnean Society, 2013, 172, 555–571
POLLEN FEATURES OF SELECTED CRATAEGUS TAXA
565
Table 4. Correlation matrix for the observed features
Feature
P
E
P/E
Exp
Exp/P
Le
Le/P
d
d/E
P
E
P/E
Exp
Exp/P
Le
Le/P
d
d/E
1
0.575***
0.284***
0.054
-0.146***
0.833***
0.090**
-0.020
-0.191***
1
-0.609***
0.257***
0.139***
0.534***
0.144***
-0.029
-0.343***
1
-0.232***
-0.287***
0.170***
-0.089**
0.008
0.216***
1
0.975***
0.079*
0.062
-0.059
-0.138***
1
-0.091**
0.038
-0.05
-0.094**
1
0.624***
-0.165***
-0.314***
1
-0.277***
-0.304***
1
0.941***
1
*P < 0.05, **P < 0.01, ***P < 0.001.
Table 5. Phenotypic distance between the taxa studied, calculated as the Mahalanobis distance based on P, E, P/E, Exp,
Exp/P, Le, Le/P, d and d/E
Taxon
C. laevigata
C. monogyna
C. rhipidophylla
C. ¥ macrocarpa
C. ¥ media
C. ¥
subsphaericea
C. laevigata
C. monogyna
C. rhipidophylla
C. ¥ macrocarpa
C. ¥ media
C. ¥ subsphaericea
0
1.68
2.07*
1.91
1.96
2.38*
0
2.04*
2.29*
2.26*
2.24*
0
0.77
0.85
0.96
0
0.36
0.92
0
0.81
0
*P < 0.05.
DISCUSSION
Statistical analyses of the morphological variability of
the pollen grains of the taxa of Rosaceae have rarely
been undertaken. Extensive research on the range of
pollen grain variation of several species of Rosa L., at
the inter- and intraspecific levels, and on flower variation, at the individual and population levels, were
conducted by Wrońska-Pilarek & Jagodziński (2009,
2012). There are a few other published studies in
which coefficients of variation of selected pollen grain
morphological traits have been analysed for the taxa
of Rubus L. and Spiraea L. (Naruhashi & Takano,
1980; Polyakova & Gataulina, 2008; Wrońska-Pilarek,
Jagodziński & Maliński, 2012).
In the palynological literature, it is difficult to find
information associated with comparative studies on
species (parentals) and their spontaneous hybrids.
Researchers have focused on pollen grain morphological features of selected species (parentals) and their
artificial hybrids. Usually, only a few basic quantitative features of pollen grains (pollen size, P/E ratio,
exine sculpture elements, less frequently the number
of apertures) or qualitative features (exine sculpture
type, operculum shape) have been analysed.
The results of the investigations carried out so far
are equivocal. Some researchers maintain that taxa
(hybrids) of higher chromosome number (frequently
polyploids) are habitually characterized by larger
pollen size than diploid species (parentals); others are
of the opposite opinion. It is no wonder that controversy exists because, as explained by Grant (1971),
polyploidy is correlated with many factors (climate,
latitude, elevation, type of habitat, hybridity, life
form, breeding system, cell and chromosome size and
structure, sex chromosome mechanism or genotype).
Rizaeva et al. (1985) found that, in nine species of
Gossypium L. (Malvaceae) and in various hybrids
involving these species, pollen size was associated
with ploidy. Tetraploid, pentaploid and hexaploid
hybrids exceeded their parentals in pollen diameter
(E), which, in the F1 diploid hybrids, was intermediate between that of the parental species. The greatest variation in pollen size was found in the
pentaploid hybrid (Gossypium barbadense L.
‘S6037’ ¥ [G. harknessii
Brandegee ¥ G. trilobum
(DC.) Kearney] ¥ ‘S6037’). The same conclusions were
also reached by Sørensen (1989) in his comparative
studies of the pollen morphology of Pachyrhizus
erosus (L.) Urb., P. tuberosus Spreng. and P. ahipa
© 2013 The Linnean Society of London, Botanical Journal of the Linnean Society, 2013, 172, 555–571
566
D. WROŃSKA-PILAREK ET AL.
Figure 17. Distribution of six Crataegus taxa as indicated by first two canonical variables.
Table 6. Results of contrast analysis between parental species and their hybrids
Contrast
Feature
C. ¥ media
C. monogyna ¥ C. laevigata
C. ¥ macrocarpa
C. laevigata ¥ C. rhipidophylla
C. ¥ subsphaericea
C. monogyna ¥ C. rhipidophylla
P
E
P/E
Exp
Exp/P
Le
Le/P
d
d/E
-3.28***
-3.69***
0.024*
-0.229***
-0.00435***
-4.65***
-0.043***
1.72***
0.071***
-1.97***
-1.80***
0.003
-0.101**
-0.00170
-2.16***
-0.009
0.76***
0.034***
-2.47***
-3.05***
0.036**
-0.064
-0.00067
-3.32***
-0.028***
0.89***
0.043***
*P < 0.05, **P < 0.01, ***P < 0.001.
Parodi and their hybrids (Fabaceae). The thesis was
further corroborated by Hossain et al. (1990), who
maintained that pollen grains of amphidiploids,
interspecific hybrids between Brassica oleracea L.
var. capitata L. ‘Yoshin’ and B. rapa L. [as B. campestris L.] var. pekinensis (Lour.) Kitam. ‘Kaga’ and
B. rapa var. chinensis (L.) Hanelt ‘Honsaiai’ (Brassicaceae), were significantly longer and wider than
those of their diploid parentals, presumably because
of the phenotypic expression of the hybrid genomes
and ploidy effects. Van der Walt & Littlejohn (1996)
recorded significant differences in pollen size of
Protea L. (Proteaceae) between clones, interspecific
hybrids and species. Rhee et al. (2005) reported that,
in interspecific hybrid Lilium (L.) (Liliaceae) after
in vitro chromosome doubling, in hybrid FA96-18, the
pollen length was 67.5 mm in haploids and 107.3 mm
in diploids.
The researchers quoted below reached different
conclusions. Olsson (1974) investigated pollen grain
features from parental plants of Linaria vulgaris
Mill. and L. repens Mill. (Plantaginaceae) and controlled crosses of F1 and F2 generations. In this case, a
high percentage of grains with a reduced equatorial
diameter (E) was typical of the F1 individuals. The
median values of equatorial diameter were smaller
© 2013 The Linnean Society of London, Botanical Journal of the Linnean Society, 2013, 172, 555–571
POLLEN FEATURES OF SELECTED CRATAEGUS TAXA
than the smallest values of the equatorial diameter
for the parental species by 3–4 mm, i.e. by c. 20%.
Similar results were found for the length of the polar
axis (P) in parental species and hybrids. According to
Srivastava (1978), pollen grains of three cultivars of
Cajanus cajan (L.) Huth and their hybrids (Fabaceae)
are characterized by similar length values of the polar
axis (P). The morphological analysis indicated dominance of the male parent in the hybrid pollen grains.
Investigations conducted by Delaporte et al. (2001) on
interspecific hybrids of Eucalyptus L’Hér (Myrtaceae),
which focused on numerous morphological traits,
including operculum length and width, failed to
provide unambiguous results. The operculum lengths
of hybrids from the F1 generation and from the openpollinated second generation were similar and, on
average, were 10.9 and 10.1 mm, whereas, in the four
Eucalyptus spp. analysed, they were slightly higher
and ranged from 13.1 to 13.3 mm. The operculum
width is a feature which is much more variable than
the length, and thus measurement results were more
variable. The operculum width was found to be
similar in hybrids from the F1 generation and from
the open-pollinated second generation (10.9 and
10.3 mm), whereas, in individual parental species, the
measurements of this feature differed widely (2.1–
14.5 mm). Franssen et al. (2001) examined pollen morphological variation among species of Amaranthus L.
(Amaranthaceae) and interspecific hybrids from the
point of view of differences between the monoecious
and dioecious Amaranthus spp. According to these
researchers, pollen grains of the dioecious species
usually had a greater number of apertures on the
visible surface. However, pollen diameters (E) did not
differ between the monoecious and dioecious plants.
Pollen of the hybrids was similar in size to that of the
maternal parent, but the number of apertures was
intermediate between those of the parental species.
This indicates that pollen characteristics may be
controlled by the female, contrary to the claim of
Srivastava (1978), and that hybrids may be more
prevalent than originally thought. Jiang et al. (2001)
ascertained that the pollen sizes of F1 hybrids did not
exceed those of parental taxa (Gastrodia elata Blume
forma glauca S. Chow and G. elata forma elata;
Orchidaceae). These results indicated that pollen
grains from the F1 hybrids expressed general or transitional morphology of the pollen grains of the parental taxa. Lu et al. (2002) observed the shape, size,
germinal aperture and exine sculpture of the pollen
grains of a new hybrid Eriobotrya japonica (Thunb.)
Lindl. ‘Zaozhong 6’ and parents ‘Moriowase’ and ‘Jiefangzhong’ (Rosaceae). The analysed pollen features
of the three cultivars differed in terms of size and
exine sculpture, with those of ‘Zaozhong 6’ being transitional [e.g. the pollen shapes (P/E ratio) of ‘Morio-
567
wase’, ‘Jiefangzhong’ and the hybrid (‘Zaozhong 6’)
were 1.346, 1.268 and 1.193, respectively, and the
average lengths of the polar axis (P) were 20.71, 24.33
and 23.83 mm, respectively]. Dönmez (2008), examining the pollen morphology of Turkish Crataegus taxa,
did not observe a significant increase in pollen size in
polyploid species.
The only article dealing with natural (spontaneous)
hybrids was published by Karlsdóttir et al. (2008),
who examined the pollen size and shape of 22
natural, triploid Betula L. hybrids and parental
species (Betulaceae). The mean diameter of the pollen
grains from the triploid hybrids was not statistically
significantly different from that of B. nana L. pollen,
but was significantly smaller than the mean value of
B. pubescens Ehrh. pollen.
The results obtained in our investigations were
also ambiguous. A strong argument in favour of the
thesis that the spontaneous hybrids studied were
characterized by greater pollen grain diameter in
comparison with parental species results from contrast analysis (Table 5). For most of the analysed
pollen features, hybrids were characterized by significantly higher values than the mean of the parental species. Moreover, differences between parental
species and hybrids were also confirmed by the
shape analysis of pollen grains of the taxa examined
(Table 3). Pollen grains of the parental species
clearly differed from one another with respect to this
feature, in contrast with hybrids, which were characterized by similar proportions of pollen grains in
individual shape classes, but it should be emphasized that, in C. rhipidophylla, the proportion of
pollen grains in individual pollen shape classes was
similar to that in hybrids.
The remaining statistical analyses examining the
relationships among individual taxa showed that the
pollen grains of two parental species, C. laevigata and
C. monogyna, were most similar to each other and
differed significantly, not only with respect to the
pollen grains of the three spontaneous hybrids,
but also with regard to the pollen of the third
parental species, C. rhipidophylla (Table 5). However,
based on the results of the canonical variable analysis, C. ¥ macrocarpa (C. laevigata ¥ C. rhipidophylla)
and C. ¥ media (C. monogyna ¥ C. laevigata) were
the most similar pollen grains. A second pair was
formed by C. ¥ subsphaericea (C. monogyna ¥ C.
rhipidophylla) and C. rhipidophylla, whereas C. monogyna and C. laevigata formed separate, ‘single
species groups’ (Fig. 17). The effectiveness of the
multi-feature analysis of the Crataegus spp. and
hybrids was corroborated by the fact that the first two
canonical variables accounted for 91.92% of the
overall variability. The usefulness of the method of
canonical variables is further corroborated by its
© 2013 The Linnean Society of London, Botanical Journal of the Linnean Society, 2013, 172, 555–571
568
D. WROŃSKA-PILAREK ET AL.
widespread application (Rybiński et al., 2009, 2011;
Seidler-Łożykowska & Bocianowski, 2012).
Pollen grains of C. rhipidophylla are larger than
those of C. laevigata and C. monogyna, and are more
similar to the pollen of the hybrids examined.
Although, in classical taxonomy, C. rhipidophylla is
considered as a ‘good species’, in other words it is
characterized by distinct features which distinguish it
from the remaining Crataegus spp., Zieliński (1977),
based on morphology (features of leaves, stipules,
fruit and sepals), classified C. curvisepala Lindm.
(= C. rhipidophylla) as a species occupying an intermediate position between C. monogyna and C. calycina
Peterm. (= C. lindmanii Hrabětová). Zieliński (1977)
maintained that it is probably an old, conserved hybrid
between C. monogyna and C. calycina [= C. rhipidophylla var. lindmanii (Hrabětová) P.A.Schmidt].
This thesis was also supported by the analysis of
Christensen (1992), which, based on morphological
features, found that 30% of Crataegus spp. were hybrid
in origin. Therefore, we decided to determine whether
the pollen morphology supports Zieliński’s (1977)
theory. Even though our study is predominantly based
at the species level, we compared the pollen morphology of C. rhipidophylla at the varietal level. We collected, in addition, three samples of C. rhipidophylla
var. lindmanii, each of 30 pollen grains (Góra Zelejowa, Świe˛tokrzyskie Province, 50°49′5.988″N,
20°27′33.012″E; Ojcowski Park Narodowy, Małopolskie Province, 50°12′42.44″N, 19°49′49.33″E; Dopikujdowa,
Małopolskie
Province,
49°31′6.959″N,
19°47′18.599″E), and compared these for all the palynological features obtained for C. rhipidophylla var.
rhipidophylla. The samples were also re-reviewed by
Professor Jerzy Zieliński. There were statistically significant differences (P < 0.0001) between C. rhipidophylla var. rhipidophylla and C. rhipidophylla var.
lindmanii in P (mean values: 38.67 vs. 42.02 mm,
respectively), E (36.15 vs. 39.67 mm, respectively), Exp
(0.857 vs. 1.066 mm, respectively), Exp/P (0.024 vs.
0.053, respectively) and Le (34.96 vs. 37.51, respectively). For the remaining pollen morphological features, there were no statistically significant differences
between the varieties studied. This results corroborate
Zieliński’s (1977) conjecture, because the mean values
of almost all the pollen features studied [P, E, Exp,
Exp/P, Le, d, d/E (PAI)] for C. rhipidophylla var.
rhipidophylla are intermediate between those of
C. monogyna (the lowest values of the pollen grains
studied) and C. rhipidophylla var. lindmanii
(= C. calycina = C. lindmanii) (the highest values of
the pollen grains studied) (Table 2).
Moreover, Christensen’s (1992, 1997) classification
confirmed the complex character of this taxon
because, according to that researcher, the present
C. rhipidophylla embraces two species distinguished
earlier, namely C. curvisepala and C. lindmanii
(= C. calycina). Considerable variation of C. rhipidophylla was also confirmed by Christensen (1992), who
divided this species into two varieties: C. rhipidophylla var. rhipidophylla, with fruits crowned by
spreading or reflexed sepals, and C. rhipidophylla
var. lindmanii (Hrabětová-Uhrová) Christensen, with
fruits crowned by erect or suberect sepals.
According to this study, the characteristics of the
pollen grains of the Crataegus taxa studied are considered to be auxiliary features in the taxonomy,
because they do not allow the diagnosis of individual
species and only allow the distinction of groups of
species (Fig. 17). Similar results were obtained by
Reitsma (1966), Byatt (1976), Byatt et al. (1977), Eide
(1981), Hebda et al. (1988) and Hebda & Chinnappa
(1994). These results support Christensen’s (1992)
hypothesis that the interpretation and practical use
of pollen structure in Crataegus and other genera of
Maloideae is severely limited.
According to Muniyamma & Phipps (1979), in Crataegus, the occurrence of malformed pollen grains is
related to extensive hybridization. Formation of the
deformed pollen grains may be dependent on disturbed meiosis, noted in some of the sexual hybrids of
Crataegus (Byatt et al., 1977). In the taxa studied, the
deformation of pollen grains includes mainly the fracture of the aperture area and the loss of turgor
pressure and change in pollen shape (such pollen
grains are flat, not spherical or ellipsoidal) (Figs 5, 6,
9). This is mainly associated with acetolysis, in which
pollen grains are exposed to concentrated chemicals
and high temperature. Nevertheless, this process is
necessary to obtain comparable measurement results
of pollen grain morphology with published data from
other palynologists, who, in most cases, measured
acetolysed pollen grains (Reitsma, 1966; Byatt, 1976;
Eide, 1981; Hebda et al., 1988; González Romano &
Candau, 1989; Hebda & Chinnappa, 1990, 1994;
Moore et al., 1991; Christensen, 1992; Zhou et al.,
2000; Dönmez, 2008). Fedoronchuk & Savitskii (1985)
found that a high proportion of deformed and sterile
pollen grains was characteristic for species of probable hybrid origin and also of wide-ranging species,
especially C. monogyna (80–85% abnormal grains).
Our studies have not fully confirmed this hypothesis.
A similar result was found only for C. rhipidophylla
var. rhipidophylla samples (75% deformed grains); for
the remaining taxa, the proportion of abnormal pollen
grains was much lower and ranged from 20 to 45%
(Fig. 16). Completely different results from those published by Fedoronchuk & Savitskii (1985) were
obtained for C. monogyna, in which we found only
25% malformed pollen. The results are surprising and
show that, on average, the percentage of deformed
pollen grains in parental species is higher (46%) than
© 2013 The Linnean Society of London, Botanical Journal of the Linnean Society, 2013, 172, 555–571
POLLEN FEATURES OF SELECTED CRATAEGUS TAXA
in hybrids (30%) (Fig. 16). However, this interpretation is not straightforward. In the case of parental
species, the high mean value of malformed pollen
grains is affected by C. rhipidophylla var. rhipidophylla samples. Taking into account C. monogyna,
C. laevigata, C. rhipidophylla var. lindmanii and the
three hybrids, the share of deformed pollen grains is
similar (20–45%). On this basis, one might lean
towards the idea that the parental and hybrid taxa
examined do not differ significantly in terms of the
percentage of deformed pollen grains.
ACKNOWLEDGEMENTS
We kindly thank Professor Jerzy Zieliński (Institute
of Dendrology, Polish Academy of Sciences, Kórnik,
Poland) for the identification of plant material and Dr
Lee E. Frelich (University of Minnesota, St Paul, MN,
USA) for linguistic support and valuable comments
on an early draft of the manuscript. We would like to
thank the reviewers for their detailed and valuable
comments on the manuscript.
REFERENCES
Byatt JI. 1976. Pollen morphology of some European species
of Crataegus L. and of Mespilus germanica L. (Rosaceae).
Pollen et Spores 18: 335–349.
Byatt JI, Ferguson IK, Murray BG. 1977. Intergeneric
hybrids between Crataegus L. and Mespilus L.: a fresh look
at an old problem. Botanical Journal of the Linnean Society
74: 329–343.
Campbell CS, Dickinson TA. 1990. Apomixis, pattern of
morphological variation, and species concepts in subfam.
Maloideae (Rosaceae). Systematic Botany 15: 124–135.
Campbell CS, Evans RC, Morgan DR, Dickinson TA,
Arsenault MP. 2007. Phylogeny of subtribe Pyrinae (formerly the Maloideae, Rosaceae): limited resolution of a
complex evolutionary history. Plant Systematics and Evolution 266: 119–145.
Chaturvedi M, Ram T, Pal M. 1993. Pollen morphology in
Chorisia species and their hybrid. Phytomorphology 43:
25–28.
Christensen KI. 1992. Revision of Crataegus sect. Crataegus
and nothosect. Crataeguineae (Rosaceae–Maloideae) in the
old world. Ann Arbor, MI: The American Society of Plant
Taxonomists.
Christensen KI. 1997. Typification of Crataegus kyrtostyla
Fingerh. In: Wisskirchen R, ed. Notulae and Floram Germanicam I. Feddes Repertorium 108: 1–109.
Christensen KI, Janjic N. 2006. Taxonomic notes on European taxa of Crataegus (Rosaceae). Nordic Journal of
Botany 24: 143–147.
Christensen KI, Zieliński J. 2008. Notes on the genus
Crataegus (Rosaceae–Pyreae) in southern Europe, the
Crimea and western Asia. Nordic Journal of Botany 26:
344–360.
569
Delaporte K, Conran J, Sedgley M. 2001. Morphological
analysis to identify the pollen parental of an ornamental
interspecific hybrid Eucalyptus. Scientia Horticulturae 89:
55–72.
Dickinson TA, Lo E, Talent N. 2007. Polyploidy, reproductive biology, and Rosaceae: understanding evolution and
making classifications. Plant Systematics and Evolution
266: 59–78.
Dickinson TA, Phipps JB. 1986. Studies in Crataegus
(Rosaceae, Maloideae), XIV. The breeding system of Crataegus crus–galli sensu lato in Ontario. American Journal of
Botany 73: 116–130.
Dönmez AA. 2004. The genus Crataegus L. (Rosaceae) with
special reference to hybridization and biodiversity in
Turkey. Turkish Journal of Botany 28: 29–37.
Dönmez EO. 2008. Pollen morphology in Turkish Crataegus
(Rosaceae). Botanica Helvetica 118: 59–70.
Eide F. 1981. Key for northwest European Rosaceae pollen.
Grana 20: 101–118.
Ennos RA, Whitlock R, Fay MF, Jones B, Neaves LE,
Payne R, Taylor I, De Vere N, Hollingsworth PM. 2012.
Process-Based Species Action Plans: an approach to conserve contemporary evolutionary processes that sustain
diversity in taxonomically complex groups. Botanical
Journal of the Linnean Society 168: 194–203.
Erdtman G. 1952. Pollen morphology and plant taxonomy.
Angiosperms. An introduction to palynology 1. Stockholm:
Almquist & Wiksell.
Fedoronchuk NM, Savitskii VD. 1985. Palynomorphological study of the Ukrainian species of the genus Crataegus
(Rosaceae). Botanicheskii Zhurnal 70: 1190–1196 [in
Russian].
Franco J. 1968. Crataegus calycina Peterm. and C. curvisepala Lindman. In: Heywood VH, ed. Flora Europaea
Notulae Systematicae ad Floram Europaeam spectantes
7. Feddes Repertorium 79: 1–68. doi:10.1002/fedr.
19680790102.
Franssen AS, Skinner DZ, Al–Khatib K, Horak MJ. 2001.
Pollen morphological differences in Amaranthus species and
interspecific hybrids. Weed Science 49: 732–737.
GenStat. 2007. GenStat release 10 reference manual. Rothamsted, Hertfordshire: Lawes Agricultural Trust.
González Romano ML, Candau PA. 1989. Contribution to
palynological studies in the Rosaceae. Acta Botanica
Malacitana 14: 105–116.
Grant V. 1971. Plant speciation. New York, London: Columbia University Press.
Hebda RJ, Chinnappa CC. 1990. Studies on pollen morphology of Rosaceae in Canada. Review of Palaeobotany and
Palynology 64: 103–108.
Hebda RJ, Chinnappa CC. 1994. Studies on pollen morphology of Rosaceae. Acta Botanica Gallica 141: 183–193.
Hebda RJ, Chinnappa CC, Smith BM. 1988. Pollen morphology of the Rosaceae of western Canada I. Agrimonia to
Crataegus. Grana 27: 93–113.
Hesse M, Halbritter H, Zetter R, Weber M, Buchner R,
Frosch–Radivo A, Ulrich S. 2009. Pollen terminology. An
illustrated handbook. Vienna: Springer.
© 2013 The Linnean Society of London, Botanical Journal of the Linnean Society, 2013, 172, 555–571
570
D. WROŃSKA-PILAREK ET AL.
Hossain MM, Inden H, Asahira T. 1990. Pollen morphology
of interspecific hybrids of Brassica oleracea and B. campestris. HortScience 25: 109–111.
Jiang L, Wan SY, Wang SB, Yu CJ. 2001. Scanning electron
microscopy (SEM) observation of pollen morphology in Gastrodia elata Bl. f. glauca S. Chow and G. elata Bl. f. elata
and their hybrids. Journal of Huazhong Agricultural
University 20: 89–91.
Karlsdóttir
L,
Hallsdóttir
M,
Thórsson
AT,
Anamthawat–Jónsson K. 2008. Characteristics of pollen
from natural triploid Betula hybrids. Grana 47: 52–59.
Kuprianowa LA, Alyoshina LA. 1978. Pollen dicotyledonarum florae partis Europeae URSS Lamiaceae–
Zygophyllaceae. Leningrad: Nauka.
Lippert W. 1978. Zur Gliederung und Verbreitung der
Gattung Crataegus in Bayern. Die Veröffentlichungen der
Bayerischen Botanischen Gesellschaft 49: 165–198.
Lu XM, Chen JY, Zhang LM, Zheng SQ, Yu D, Liao RY.
2002. Observation and comparison on pollen morphology of
a new hybrid loquat variety ‘Zaozhong 6’ and its parentals.
Acta Horticulturae Sinica 29: 271–273.
Mahalanobis PC. 1936. On the generalized distance in statistics. Proceedings of the National Institute of Science of
India 12: 49–55.
Moore PD, Webb JA, Collinson ME. 1991. Pollen analysis.
London: Blackwell Scientific Publications.
Muniyamma M, Phipps JB. 1979. Cytological proof of
apomixis in Crataegus (Rosaceae). American Journal of
Botany 66: 149–155.
Nair PKK, Katiar K, Srivastava GS. 1977. Pollen morphology of an Erythrina hybrid and their parentals. Current
Science 46: 824–825.
Naruhashi N, Takano H. 1980. Size variation of pollen
grains in some Rubus species. Journal of Phytogeography
and Taxonomy 28: 27–32.
Ohashi H, Hoshi H, Iketani H. 1991. Taxonomy and pollen
morphology of hybrids between Sorbus and Micromeles in
the genus Sorbus (Rosaceae subfamily Maloideae). Journal
of Japanese Botany 66: 110–124.
Olsson U. 1974. A biometric study of the pollen morphology
of Linaria vulgaris (L.) Miller and L. repens (L.) Miller
(Scrophulariaceae) and their hybrid progeny in F1 and F2
generations. Grana 14: 92–99.
Phipps JB. 1983. Crataegus – a nomenclator for sectional
and serial names. Taxon 32: 598–604.
Phipps JB. 1988. Crataegus (Maloideae, Rosaceae) of the
southeastern United States, I. Introduction and series Aestivales. Journal of the Arnold Arboretum 69: 401–431.
Phipps JB, O’Kennon RJ, Lance RW. 2003. Hawthorns
and medlars. Cambridge: Royal Horticultural Society.
Phipps JB, Robertson KR, Rohrer JR, Smith PG. 1991.
Origin and evolution of subfam. Maloideae (Rosaceae). Systematic Botany 16: 303–332.
Phipps JB, Robertson KR, Smith PG, Rohrer JR. 1990. A
checklist of the subfamily Maloideae (Rosaceae). Canadian
Journal of Botany 68: 2209–2269.
Polyakova TA, Gataulina GN. 2008. Morphology and variability of pollen of the genus Spiraea L. (Rosaceae) in
Siberia and the Far East. Contemporary Problems of
Ecology 1: 420–424.
Punt W, Hoen PP, Blackmore S, Nilsson S, Le Thomas A.
2007. Glossary of pollen and spore terminology. Review of
Palaeobotany and Palynology 1431: 1–81.
Rao PM, Kumar CR, Nair PKK. 1979. Pollen morphology of
Cotyledon grandiflora and Echeveria sp. and their hybrid.
Current Science 48: 909–910.
Reitsma TJ. 1966. Pollen morphology of some European
Rosaceae. Acta Botanica Neerlandica 15: 290–379.
Rhee HK, Cho HR, Kim KJ, Kim KS. 2005. Comparison of
pollen morphology in interspecific hybrid lilies after in
vitro chromosome doubling. Acta Horticulturae 673: 639–
643.
Rizaeva SM, Akhmedova MZ, Abdullaev AA. 1985.
Pollen viability and pollen grain morphology in interspecific
cotton
hybrids
differing
in
origin
and
ploidy.
Sel’skokhozyaistvennaya Biologiya 9: 63–66 [in Russian].
Robertson KR, Phipps JB, Rohrer JR, Smith PG. 1991. A
synopsis of genera in Maloideae (Rosaceae). Systematic
Botany 16: 376–394.
Rybiński W, Szot B, Bocianowski J, Rusinek R. 2011.
Geometric properties of grasspea seeds and their mechanical loads. International Agrophysics 25: 271–280.
Rybiński W, Szot B, Rusinek R, Bocianowski J. 2009.
Estimation of geometric and mechanical properties of seeds
of Polish cultivars and lines representing selected species of
pulse crops. International Agrophysics 23: 257–267.
Seidler-Łożykowska K, Bocianowski J. 2012. Evaluation
of variability of morphological traits of selected caraway
(Carum carvi L.) genotypes. Industrial Crops and Products
35: 140–145.
Shapiro SS, Wilk MB. 1965. An analysis of variance test for
normality (complete samples). Biometrika 52: 591–611.
Sokal RR, Rohlf FJ. 1995. Biometry: the principles and
practice of statistics in biological research. New York:
Freeman WH and Co.
Sørensen M. 1989. Pollen morphology of species and interspecific hybrids in Pachyrhizus Rich. ex DC. (Fabaceae:
Phaseoleae). Review of Palaeobotany and Palynology 61:
319–339.
Srivastava V. 1978. Pollen morphology of Cajanus cajan (L.)
Willd. (Leguminosae) cultivars and their hybrids. Grana 17:
107–109.
Srivastava V, Pal M, Nair PKK. 1977. A study of the pollen
grains of Amaranthus spinosus Linné and A. dubius Mart
ex Thellung and their hybrids. Review of Palaeobotany and
Palynology 23: 287–291.
Talent N, Dickinson TA. 2005. Polyploidy in Crataegus and
Mespilus (Rosaceae Maloideae): evolutionary inferences
from flow cytometry of nuclear DNA amounts. Canadian
Journal of Botany 83: 1268–1304.
Van der Walt ID, Littlejohn GM. 1996. Pollen morphology,
male hybrid fertility and pollen tube pathways in Protea.
South African Journal of Botany 62: 236–246.
Wrońska-Pilarek D. 1998. Pollen morphology of the Polish
species of the genus Ribes L. Acta Societatis Botanicorum
Poloniae 67: 275–285.
© 2013 The Linnean Society of London, Botanical Journal of the Linnean Society, 2013, 172, 555–571
POLLEN FEATURES OF SELECTED CRATAEGUS TAXA
Wrońska-Pilarek D. 2011. Pollen morphology of Polish
native species of the Rosa genus (Rosaceae) and its relation
to systematics. Acta Societatis Botanicorum Poloniae 80:
221–232.
Wrońska-Pilarek D, Jagodziński AM. 2009. Pollen morphological variability of Polish native species of Rosa L.
(Rosaceae). Dendrobiology 62: 71–82.
Wrońska-Pilarek D, Jagodziński AM. 2011. Systematic
importance of pollen morphological features of selected
species from the genus Rosa (Rosaceae). Plant Systematics
and Evolution 295: 55–72.
Wrońska-Pilarek D, Jagodziński AM. 2012. Intra- and
inter-individual variability of selected quantitative features
of pollen grain morphology based on the example of Rosa
canina L. (Rosaceae). Dendrobiology 67: 25–39.
571
Wrońska-Pilarek D, Jagodziński AM, Maliński T. 2012.
Morphological studies of pollen grains of the Polish endemic
species of the genus Rubus (Rosaceae). Biologia 67: 87–
96.
Xin X, Zhang YM, Wang J. 1986. Observations on the pollen
morphology of major Crataegus species in China. Scientia
Agricultura Sinica 3: 94–95 [in Chinese].
Zhou LH, Wie ZX, Wu ZY. 2000. Pollen morphology of
Maloideae of China (Rosaceae). Acta Botanica Yunnanica
22: 47–52 [in Chinese].
Zieliński J. 1977. Crataegus curvisepala Lind. and C. microphylla C. Koch in Bulgaria. Arboretum Kórnickie 22: 29–38.
Zieliński J. 1982. Crataegus heldreichii Boiss. in Bulgaria.
Fragmenta Floristica et Geobotanica Polonica 28: 625–627.
© 2013 The Linnean Society of London, Botanical Journal of the Linnean Society, 2013, 172, 555–571