Pollen morphology and variability of native and alien, including invasive, species of the genus Spiraea L. (Rosaceae) in Poland.

The pollen morphology was studied in 25 taxa of the genus Spiraea L. The aims of this study were to describe the pollen morphology and variability and to determine whether the pollen features of alien, expansive or invasive Spiraea species differ from those of other taxa. The species of Spiraea were analysed for nine quantitative pollen traits as well as the following qualitative traits: the outline, shape and exine ornamentation. In total, 750 pollen grains were measured. Based on the pollen key exine ornamentation features, then individual Spiraea species were distinguished, while the other species formed groups of usually 2-3, up to 8 species. The most important pollen features included length, width and course of grooves and striae, presence or absence of perforations, as well as their number and diameter. The most variable taxa for all the nine biometric traits jointly were S. ×billardii, S. veitchii, S. nipponica and S. cana. The pollen of the invasive S. tomentosa differed from the other taxa studied, unlike the other invasive species (S. douglasii and S. japonica).

Introduction

The genus Spiraea L. belongs to the tribe Spiraeeae, the family Rosaceae Juss., to the order Rosales and to the rosid clade [1-4]. The taxonomy introduced by Rehder [5] divides the genus Spiraea into three sections: Chamaedryon Ser., Calospira K. Koch and Spiraria Ser. However, this was later developed by Yu and Kuan [6] into four sections and 10 series.

There are 80 to 100 species of this genus throughout the world, although the exact number of Spiraea species has not been specified [3, 7]. Spiraea is by far the largest and most widespread genus in the tribe Spiraeeae [3], which are native to the temperate Northern Hemisphere, with the greatest diversity in eastern Asia, mainly China, where approx 70 Spiraea species (47 endemic) are found [1, 7–9]. In Europe there are seven native species of Spiraea–S. cana, S. chamaedryfolia, S. crenata, S. decumbens, S. hypericifolia, S. media and S. salicifolia [10, 11].

Representatives of the genus Spiraea are very popular decorative plants in North America, Asia and Europe, including Poland. In the latter, many species of meadowsweet from Asia and North America are cultivated, while some of them, e.g. S. tomentosa, S. douglasii, and S. japonica L., have become naturalized and are invasive in some European countries [12-14], posing a threat to biodiversity in these areas [15].

According to Lambdon et al. [16] and the DAISIE [17], 21 alien Spiraea taxa (13 species and 8 hybrids) grow in Europe, of which 18 taxa were ‘aliens in Europe’ naturalized within the continent and 12 species were naturalized neophytes within the continent. A further five species were ‘aliens of European origin’ (five naturalized and three—naturalized neophytes). The third group–‘aliens to Europe’ consisted of 16 species (13 naturalized and nine—naturalized neophytes). Mirek et al. [18] reported the occurrence of two native species (S. salicifolia L., and S. media Schmidt) in Poland and the following eight alien species and hybrids: S. albiflora Zabel, S. chamaedryfolia L. em. Jacq., S. hypericifolia L., S. japonica L. f., S. ulmifolia Scop., S. ×vanhouttei (Briot) Zabel, S. ×arguata Zabel and S. ×bumalda Burv. In turn, Tokarska-Guzik et al. [15] listed six domestic, alien meadowsweet species and cultivars in Poland. These were locally established kenophytes—S. alba Du Roi (North America), S. chamaedryfolia L. em. Jacq. (southeastern Europe and southeastern Asia) and S. douglasii Hook. (western North America), a domestic, invasive species—S. tomentosa L. (from eastern North America), and two locally established cultivars—S. ×pseudosalicifolia Silverside (= S. salicifolia L. × S. douglasii Hook) and S. ×vanhouttei (Briot) Zabel. S. ×pseudosalicifolia Silverside and S. ×billardii Hérincq, analysed by our team, are often treated synonymously and traded as the same taxon. This is quite unfortunate because S. ×pseudosalicifolia Silverside is a hybrid of S. salicifolia L. × S. douglasii Hook, while S. ×billardii Hérincq is a hybrid of S. alba Du Roi × S. douglasii Hook [19].

Many palynologists believe that palynomorphological characteristics tend to be more useful to distinguish higher ranking taxa of the family Rosaceae (subfamily, tribe or genus) rather than those of lower rank (section, series or species) [20-41]. Especially exine sculpture was found to be an important feature to distinguish species of many genera from the family Rosaceae. The most important traits of exine sculpture were the number and width of striae and grooves, as well as the number, area and size of perforations [20–34, 41]. Many scientists have emphasized that for species delimitation within the family Rosaceae the length of polar axis (P), pollen shape (P/E ratio), operculum structure as well as the presence or lack of costae colpi are very important [20–26, 28, 29, 36, 37, 41, 42].

So far, the few studies on the pollen morphology of the genus Spiraea have been regional in character and covered at most a dozen species. Polyakova and Gataulina [38] described the pollen morphology of 12 Spiraea species from different regions of Siberia and the Far East. Hebda and Chinnappa [23] described some Spiraea species growing in Canada. In turn, Liu et al. [39] analyzed 18 Chinese Spiraea species, while Song et al. [40] described 17 Korean species. Wrońska-Pilarek et al. [32] examined pollen morphology and intraspecific variability of invasive S. tomentosa in Poland. In the PalDat database, Heigl [43] and Auer [44] briefly described S. chamaedryfolia and S. oblongifolia. All authors agree that the most important pollen features of the Spiraea species were connected with exine ornamentation, although some researchers also mentioned other features, such as the endoaperture diameter, equatorial and polar views, and length of the polar axis (P).

The presented study aimed to describe and analyze pollen morphology and intrageneric and interspecific variability of 25 Spiraea taxa—two species native to Poland and Europe and 23 alien species and hybrids—which were naturally distributed in Asia, Europe and North America, whence they came to Poland. This is the first palynological review of research on this genus in Europe. This study was also aimed at determining whether the pollen features of expansive and invasive Spiraea species differ from those of other taxa, and if so, whether and to what extent it may increase the effectiveness of the expansion of these species. It is believed that, due to the invasive nature of some of the studied species in Poland and Europe, any new data on its reproduction may be useful in the fight against the expansion of this species.

Material and methods

Palynological analysis

Inflorescences of the 25 Spiraea taxa under analysis (22 species, one variety and two hybrids) were collected from the herbarium belonging to the Botanical Garden of the Adam Mickiewicz University in Poznań (OB UAM) (Table 1). The plants grown there come from other botanical gardens or natural sites of individual Spiraea species (Table 1). All samples were taken from herbarium sheets prepared in 2019. A total of 24 samples came from shrubs not older than 40 years. The youngest was 12 years old, while the oldest was 39. Additionally, four samples came from shrubs that are 48, 55, 82 and 86 years old. These are plants (except for S. tomentosa L.), which are growing scattered over an area of about 22 ha. Their flowering takes place over a period of three months (from April to June/July).

Table 1

List of Spiraea taxa under analysis collected from the herbarium of the Botanical Garden of the Adam Mickiewicz University in Poznań and natural sites.

No.	Taxon name	The area of natural occurrence	Source and year of acquisition	BG AMU registration number	Herbarium sheet number
1	S. alba Du Roi	Canada, United States	Lack of data, 1995	7312	XI
2	S. alba var. latifolia (Aiton) Dippel	Canada, United States	Lack of data	0995	XXVIII
3	S. betulifolia Pall.	Japan, Russia	Experimentelle Botanische Garten, Georg-August-Universität Göttingen, 2004	7244	VI
4	S. cana Waldst. & Kit	Italy, Croatia, Bosna and Herzegovina, Serbia	Hesse Nursery, Weener on Ems, 1939	1206	XV
5	S. chamaedryfolia L.	Italy, Austria, Czech Republic, Slovakia, Bulgaria, Bosna and Herzegovina, Serbia, Croatia, Romania, Russia, Mongolia, China, Korea, Japan	The Kórnik Arboretum, 1983	2561	I
6	S. chinensis Maxim.	China	Lack of data, 2000	5014	XVI
7	S. dasyantha Bunge	China	The Botanical Garden of the University of Tartu, 1989	3333	XXVII
8	S. douglasii Hook.	Canada, United States	Lack of data, 2009	8586	X
9	S. elegans Pojark.	Russia, Mongolia, China	The Akureyri Botanical Garden, 1991	5083	XXIV
10	S. henryi Hemsl.	China	The Kórnik Arboretum, lack of data	0694	XX
11	S. hypericifolia L.	Bulgaria, Ukraine, Russia, Kazakhstan, Uzbekistan, Turkmenistan, Kyrgyzstan, Tajikistan, Mongolia, China	Lack of data, 1963	0907	XIII
12	S. japonica L. f.	Japan, Korea, China	Lack of data	6394	V
13	S. media Schmidt	Austria, Czech Republic, Slovakia, Hungary, Bulgaria, Romania, Croatia, Bosna and Herzegovina, Montenegro, Poland, Ukraine, Russia, Mongolia, China, Korea, Japan	Lack of data	0880	XVII
14	S. nipponica Maxim.	Japan	Lack of data, 1991	4175	XIX
15	S. pubescens Turcz.	Mongolia, Russia, China, Korea	Lack of data	6448	XXVI
16	S. salicifolia L.	Austria, Hungary, Romania, Bulgaria, Czech Republic, Poland, Russia, Mongolia, China, Korea, Japan	Arboretum Waasland Nieuwkerken-Waas, 1994	5088	IX
17	S. splendens É. N. Baumann ex K. Koch	Canada, United States	Lack of data	2547	XXIX
18	S. thunbergii Siebold ex Blume	China, Japan	Mlyňany Arboretum of the Slovak Academy of Sciences, 1984	2609	XII
19	S. trichocarpa Nakai	China, Korea	Royal Botanic Garden Edinburgh, 1987	3768	IV
20	S. uratensis Franch.	China	The Kórnik Arboretum,1956	0708	III
21	S. veitchii Hemsl.	China	Lack of data	0688	VIII
22	S. wilsonii Duthie	China	Hortus Botanicus Fominianus Universitatis Kioviensis, 2009	8408	XXII
23	S. ×billardii Hérincq.		The Kórnik Arboretum, 1985	2848	XXIII
24	S. ×cinerea Zabel ‘Grefsheim’		Lack of data, 1987	3010	II
25	S. tomentosa L.	United States, Canada	Poland, Bory Niemodlińskie, Tułowice, 2017, leg. B. Wiatrowska	PZNF	PZNF

PZNF–Herbarium of the Department of Forest Botany.

In this paper the taxonomic classification of the species from the genus Spiraea under analysis was adopted from Yü and Kuan [6], since the latest studies by Potter et al. [2, 3] and APG IV [4] do not provide systematic divisions within the studied genus. The taxa under examination represented all the four sections and eight (out of 10) series of the genus Spiraea [6]. Additionally, S. ×billardii Hérincq was included—a hybrid of two species (S. alba Du Roi × S. douglasii Hook) from the Spiraea section, and S. ×cinerea Zabel ’Grefsheim’, a hybrid of the species from the section Chamaedryon with a species from the section Glomerati Nakai.

The nomenclature of taxa was adopted from botanical studies characterizing flora of the corresponding regions of the world. For Europe it was Flora Europaea [10] and Illustrierte Flora von Mitteleuropa [11], for Central Asia and the Russian Far East—Flora of the U.S.S.R. [45], Deriewa i kustraniki SSSR [46] and Flora Sibiriae [], while for the Far East—Flora of Japan [] and Flora of China [49]. The names of hybrids were adopted from The European Garden Flora [50]. All plant names were compared and validated against the online databases [51, 52]. A list of the taxa analyzed together with their affiliation to particular sections and series is shown in Table 2.

Table 2

List of taxa analyzed with their affiliation to particular sections and series.

Section	Series	Taxa
Spiraea	-	S. salicifolia L., S. douglasii Hook., S. alba Du Roi,.S. alba var. latifolia (Aiton) Dippel; S. ×billardii Hérincq
Calospira K. Koch	-	S. betulifolia Pall., S. splendens É. N. Baumann ex K. Koch*
	Canescentes Yu	S. uratensis Franch., S. trichocarpa Nakai, S. canescens D. Don
	Henryanae Yu	S. veitchii Hemsl., S. henryi Hemsl., S. wilsonii Duthie
	Japonicae Yu	S. japonica L. f.
Chamaedryon Ser.	-	S. cana Waldst. & Kit, S. nipponica Maxim.
	Chamaedryfoliae Pojark.	S. chamaedryfolia L.
	Mediae Pojark.	S. media Schmidt
	Trilobatae Pojark.	S. chinensis Maxim., S. elegans Pojark., S. pubescens Turcz., S. dasyantha Bunge
Glomerati Nakai	Hypericifoliae Pojark.	S. hypericifolia L.
	Prunifoliae Yu	S. thunbergii Siebold ex Blume, S. ×cinerea Zabel ‘Grefsheim’

* species allocated to sections only—the section is not divided into series.

In accordance with the study by Wrońska-Pilarek et al. [53], each sample consisted of 30 randomly selected, mature and correctly formed pollen grains derived from a single individual (shrub). In total, 750 pollen grains were studied.

The pollen grains were prepared for light (LM) and scanning electron microscopy (SEM) using the standard acetolysis method described by Erdtman [54, 55]. An acetolysis mixture was prepared to consist of 9 parts acetic acid anhydride and one part concentrated sulfuric acid, while the acetolysis process lasted 2.5 minutes. The prepared material was divided into two parts: one part was immersed in an alcohol solution of glycerine (for LM) and the other in 96% ethyl alcohol (for SEM). Morphological observations were carried out using a light microscope (Biolar 2308, Nikon HFX-DX) and a scanning electron microscope (Jeol 7001TTLS). The pollen grains were measured in the equatorial view at a magnification of 640x.

Nine quantitative features of the pollen grains were analyzed, i.e., the length of the polar axis (P) and equatorial diameter (E), the length of the ectoaperture (Le), the thickness of the exine along the polar axis and equatorial diameter (Exp, Exe), as well as the P/E, Le/P, Exp/P and Exe/E ratios. Moreover, the following qualitative features were examined: the pollen outline and shape, and exine ornamentation (type, width and direction of grooves and striae, number and diameter of perforations).

The palynological terminology follows Punt et al. [56] and Halbritter et al. [57].

Statistical analysis

The normality of the distributions for the studied traits (length of polar axis—P, equatorial diameter—E, length of ectoaperture—Le, exine thickness along the polar axis—Exp, exine thickness along the equatorial diameter—Exe, as well as the P/E, Le/P, Exp/P and Exe/E ratios) was tested using Shapiro-Wilk’s normality test [58]. A multivariate analysis of variance (MANOVA) was performed based on the following model using a MANOVA procedure in GenStat 18: Y = XT+E, where: Y is (n×p)–the dimensional matrix of observations, n is the total number of observations, p is the number of traits (in this study p = 9), X is (n×k)–the dimensional matrix of design, k is the number of species (in this study k = 25), T is (k×p)–the dimensional matrix of unknown effects, and E is (n×p)–the dimensional matrix of residuals. Afterward, one-way analyses of variance (ANOVA) were performed to verify the null-hypothesis of a lack of a species effect in terms of the values of the nine observed traits, independently for each trait, based on the following model: y = ⎧+⎮+∑, where: y—jth observation of ith species, ⎧—the grand mean, ⎮—the effect of ith species and ∑—error observation. The minimal and maximal values of the traits as well as the arithmetic means and coefficients of variation (CV, in %) were calculated. Moreover, Fisher’s least significant differences (LSDs) were estimated at a significance level of α = 0.001. The relationships between the observed traits were estimated based on the species’ means using Pearson’s correlation coefficients. The results were also analyzed using multivariate methods. A canonical variate analysis was applied to present a multi-trait assessment of the similarity of the tested species in a lower number of dimensions with the least possible loss of information [59]. This made it possible to illustrate in the graphic form any variation in the species in terms of all the observed traits. The Mahalanobis distance was suggested as a measure of “polytrait” species similarity [60], the significance of which was verified using critical value D called “the least significant distance” [61]. Mahalanobis distances were calculated for the species. The differences between the analyzed species were verified by cluster analysis using the nearest neighbor method and Euclidean distances [62]. All the analyses were conducted using the GenStat 18 statistical software package.

Results

Morphological description of pollen

A description of pollen grain morphology of the Spiraea species under analysis is given below and illustrated in the SEM photographs (Figs 1–5). The morphological observations for the quantitative features are summarized in Tables 3–5 and S1 Table.

Fig 1

Pollen grains in polar view in S. douglasii, S. elegans, S. thunbergii, S. japonica, S. uratensis, A-E.

Fig 5

Striate exine ornamentation types according to Ueda and Tomita [63]; see Table 6.

A-I. A, S. media; B, S. pubescens; C, S. salicifolia; D, S. splendens; E, S. thunbergii; F, S. trichocarpa; G, S. uratensis; H, S. veitchii; I, S. tomentosa.

Table 3

Minimal, maximal and mean values as well as coefficients of variation (CV) for length of polar axis (P), equatorial diameter (E) and length of ectoaperture (Le) in studied taxa.

Trait	P				E				Le
Species	Mean		Min-Max	s.d.	Mean		Min-Max	s.d.	Mean		Min-Max	s.d.
S. alba	16.33	def	14–20	1.06	16.07	cd	14–20	1.11	15.27	c	12–20	1.70
S. alba var. latifolia	18.27	ab	16–20	1.02	17.87	b	16–20	1.04	17.53	ab	14–20	1.46
S. betulifolia	18.67	a	16–20	1.32	19.27	a	16–20	1.23	16.67	b	14–20	1.61
S. cana	16	defg	14–18	1.17	14.87	fghij	12–18	1.72	14.53	cde	12–18	1.57
S. chamaedryfolia	18.6	a	16–20	1.07	19.73	a	18–20	0.69	17.67	ab	16–20	1.40
S. chinensis	15.8	defg	14–18	1.10	15.6	cdef	14–18	0.97	14.93	cd	12–18	1.72
S. dasyantha	14.4	i	14–16	0.81	14.07	ij	12–16	0.64	13.8	def	12–16	1.10
S. douglasii	15.4	fgh	14–18	1.50	14.47	ghij	12–16	1.25	14.47	cdef	12–18	1.80
S. elegans	15.6	efg	14–16	0.81	15.73	cdef	14–16	0.69	15	cd	12–16	1.15
S. henryi	15.2	ghi	14–16	1.00	14.93	efghij	14–16	1.02	14.2	cdef	12–16	1.42
S. hypericifolia	18.27	ab	16–20	1.26	19.13	a	16–22	1.36	17.5	ab	16–20	1.55
S. japonica	15.6	efg	14–18	0.97	14.87	fghij	14–16	1.01	14.07	cdef	12–16	1.44
S. media	15.67	defg	14–18	1.18	15.8	cdef	14–18	1.10	14.67	cde	12–18	1.69
S. nipponica	16.6	cd	14–22	1.40	15.67	cdef	14–20	1.58	14.93	cd	12–22	1.80
S. pubescens	14.53	hi	12–16	1.04	14	j	12–16	1.05	14	cdef	12–16	1.39
S. salicifolia	19	a	18–20	1.02	19.33	a	16–20	1.09	18.27	a	16–20	1.26
S. splendens	16.47	de	14–20	1.25	15.2	defgh	14–18	1.13	15.2	c	14–18	1.13
S. thunbergii	15.4	fgh	14–18	1.07	15	efghi	12–18	1.26	13.6	ef	12–16	1.52
S. tomentosa	16.33	def	14–20	1.67	16.2	c	14–18	1.32	13.2	f	10–18	1.71
S. trichocarpa	15.8	defg	14–18	0.96	15.4	cdefg	14–16	0.93	14.6	cde	12–16	1.30
S. uratensis	17.53	bc	16–20	1.01	15.4	cdefg	14–16	0.93	17	ab	14–20	1.37
S. veitchii	16.53	de	14–20	1.28	15.87	cde	14–20	1.38	14.67	cde	12–20	2.12
S. wilsonii	15.73	defg	14–18	1.02	14.93	efghij	12–16	1.14	14.13	cdef	12–16	1.38
S. ×billardii	18.13	ab	14–20	1.38	16.27	c	14–20	1.64	17.13	ab	14–20	1.87
S. ×cinerea	15.67	defg	14–18	1.18	14.27	hij	12–16	1.02	14.67	cde	12–18	1.69
LSD0.001	0.989				0.992				1.317
ANOVA F	39.57				64.52				26.73
F value	<0.001				<0.001				<0.001

Table 5

Minimal, maximal and mean values as well as coefficients of variation (CV) for Le/P, Exp/P and Exe/E of analyzed species.

Trait	Le/P				Exp/P				Exe/E
Species	Mean		Min-Max	s.d.	Mean		Min-Max	s.d.	Mean		Min-Max	s.d.
S. alba	0.935	abcdef	0.75–1	0.083	0.06	def	0.013–0.100	0.018	0.061	cdefg	0.013–0.100	0.018
S. alba var. latifolia	0.961	abc	0.70–1	0.07	0.043	fg	0.011–0.089	0.019	0.045	fgh	0.011–0.089	0.022
S. betulifolia	0.894	ef	0.78–1	0.067	0.035	g	0.010–0.063	0.013	0.034	h	0.010–0.063	0.012
S. cana	0.908	abcdef	0.75–1	0.066	0.079	bc	0.033–0.114	0.022	0.085	ab	0.038–0.133	0.025
S. chamaedryfolia	0.95	abcde	0.80–1	0.06	0.049	efg	0.011–0.089	0.023	0.047	efgh	0.010–0.089	0.022
S. chinensis	0.945	abcdef	0.75–1	0.079	0.043	fg	0.013–0.063	0.012	0.044	gh	0.013–0.063	0.012
S. dasyantha	0.959	abcd	0.86–1	0.064	0.055	def	0.038–0.100	0.016	0.056	defg	0.043–0.114	0.017
S. douglasii	0.939	abcdef	0.86–1	0.067	0.056	def	0.033–0.114	0.019	0.059	defg	0.038–0.114	0.019
S. elegans	0.962	ab	0.86–1	0.059	0.068	bcd	0.038–0.114	0.02	0.067	bcd	0.038–0.100	0.02
S. henryi	0.935	abcdef	0.75–1	0.081	0.064	bcde	0.038–0.125	0.025	0.066	bcde	0.038–0.143	0.029
S. hypericifolia	0.959	abcd	0.80–1	0.064	0.048	efg	0.010–0.089	0.019	0.046	fgh	0.010–0.089	0.019
S. japonica	0.903	bcdef	0.75–1	0.091	0.065	bcde	0.033–0.114	0.019	0.068	bcd	0.038–0.114	0.019
S. media	0.935	abcdef	0.86–1	0.067	0.063	cde	0.033–0.100	0.018	0.063	cdefg	0.033–0.114	0.019
S. nipponica	0.899	def	0.78–1	0.062	0.079	bc	0.043–0.125	0.024	0.084	ab	0.043–0.125	0.026
S. pubescens	0.963	ab	0.86–1	0.062	0.058	def	0.013–0.100	0.019	0.061	cdefg	0.013–0.114	0.021
S. salicifolia	0.963	ab	0.80–1	0.062	0.036	g	0.010–0.089	0.015	0.034	h	0.010–0.089	0.015
S. splendens	0.926	abcdef	0.78–1	0.076	0.064	bcde	0.033–0.114	0.022	0.07	abcd	0.038–0.125	0.024
S. thunbergii	0.884	f	0.75–1	0.082	0.07	bcd	0.038–0.114	0.024	0.072	abcd	0.038–0.114	0.023
S. tomentosa	0.81	g	0.63–0.9	0.085	0.103	a	0.056–0.143	0.032	0.079	abc	0.038–0.143	0.03
S. trichocarpa	0.924	abcdef	0.86–1	0.063	0.069	bcd	0.033–0.100	0.018	0.07	abcd	0.038–0.114	0.019
S. uratensis	0.969	a	0.88–1	0.052	0.07	bcd	0.033–0.100	0.021	0.079	abc	0.038–0.114	0.022
S. veitchii	0.886	f	0.75–1	0.095	0.062	cde	0.033–0.114	0.022	0.064	cdef	0.038–0.114	0.022
S. wilsonii	0.9	cdef	0.75–1	0.083	0.058	def	0.033–0.114	0.023	0.061	cdefg	0.038–0.114	0.025
S. ×billardii	0.945	abcdef	0.78–1	0.07	0.047	efg	0.030–0.089	0.017	0.053	defgh	0.030–0.114	0.022
S. ×cinerea	0.935	abcdef	0.86–1	0.066	0.082	b	0.038–0.125	0.028	0.089	a	0.043–0.143	0.028
LSD0.001	0.061				0.018				0.019
ANOVA F	7.52				16.31				14.44
F value	<0.001				<0.001				<0.001

A-H. A, S. chinensis; B, S. ×cinerea ‘Grefsheim’; C, S. dasyantha; D, S. douglasii; E, S. elegans; F, S. henryi; G, S. japonica; H, S. alba var. latifolia.

The pollen grains of the studied species occur as radially symmetric, tricolporate, isopolar monads (Figs 1 and 2).

Fig 2

Pollen grains in equatorial view in S. cana, S. henryi, S. salicifolia, S. trichocarpa, S. veitchii, A-E.

According to Erdtman’s pollen size classification [54], all of the analyzed pollen grains were small (10.1–25 μm). The length of the polar axis (P) was 16.46 (12.00–22.00) μm. The smallest P (12.00 μm) were found in S. pubescens, while the largest ones (22.00 μm) in S. nipponica. The average value of this feature ranged from 14.00 to 19.73 μm (S1 Fig).

The mean length of the equatorial diameter (E) was 16.00 (12.00–22.00) μm. The smallest value of this feature (12.00 μm) was found in S. cana, S. douglasii, S. pubescens, S. thunbergii and S. ×cinerea ‘Grefsheim’, whereas the largest ones (22.00 μm) only in S. hypericifolia. The average value of this trait ranged between 13.20 and 18.27 μm (S1 Fig).

The outline in the polar view was mostly circular or more rarely elliptic, while in the equatorial view it was circular or elliptic (Figs 1 and 2).

The mean P/E ratio was 1.04, ranging from 0.78 in S. media to 1.43 in S. uratensis. The average value of this ratio was very similar and ranged from 0.81 to 0.97. The pollen shape was most frequently spheroidal (398 pollen grains– 53.1%) and prolate-spheroidal (215–28.7%), rarely oblate-spheroidal (61–8.1%), suboblate (37–4.9%), subprolate (37–4.9%), and very rarely prolate (2–0.3%).

The mean exine thickness was 0.98 μm (Exe)– 0.99 μm (Exp) (within the range of 0.20–2.00 μm). The relative exine thickness (Exp/P and Exe/E ratios) averaged 0.09 and 0.06 (Exp/P–from 0.01 to 0.90 and Exe/E–from 0.01 to 0.14), respectively.

Exine ornamentation was usually striate-perforate, rarely striate and in one species striate-reticulate (Figs 3–5). The striae and grooves usually ran parallel to the polar axis, but they also frequently formed loops. They were straight or forked and of varying length and width (Figs 3–5). Either the grooves were narrower than the striae, or the widths of the grooves and striae were similar and averaged from 0.1 to 0.3 μm. Circular or elliptic perforations of different diameters–small to large (from 0.1 to 1 μm)–were found at the bottom of the striae (Figs 3–5).

Fig 3

Striate exine ornamentation types according to Ueda and Tomita [63]; see Table 6.

A-H. A, S. nipponica; B, S. wilsonii; C, S. alba; D, S. betulifolia; E, S. ×billardii; F, S. cana; G, S. chamaedryfolia; H, S. hypericifolia.

Pollen of the individual Spiraea species under analysis was classified according to the striate exine ornamentation classification proposed by Ueda and Tomita [63] and Ueda [24] into five types (I-V) and six subtypes (I A,B, II A,B and III A,B). Ueda and Tomita [63] distinguished six types (I-VI) and six subtypes (I-III, each A and B). The above classification was applied, because it is based on the analysis of striate exine ornamentation in pollen of many species and genera from the family Rosaceae, and thus it best differentiates this type of ornamentation in the studied genus Spiraea. Type VI was not found in this study (Figs 3–5, Table 6). The greatest number of species (9) belonged to the IIA subtype, which was characterized by fairly distinct striae, narrow grooves and frequently numerous perforations of different diameters. Subtypes IA, IIIA, IIIB, IV and V were represented by two or three species, while type IB, by only one species (Figs 3–5, Table 6). In S. henryi only striate-reticulate exine ornamentation was found.

Table 6

Exine ornamentation types and subtypes of Spiraea species studied (according to the classification of Ueda and Tomita [63]).

Striate exine ornamentation type or subtype	Species
IA	S. dasyantha, S. hypericifolia, S. media
IB	S. cana
IIA	S. alba var. latifolia, S. chamaedryfolia, S. douglasii, S. pubescens, S. thunbergii, S. tomentosa, S. trichocarpa, S. veitchii, S. wilsonii
IIIA	S. alba, S. ×cinerea ’Grefsheim’, S. splendens
IIIB	S. betulifolia, S. ×billardii, S. elegans
IV	S. chinensis, S. japonica, S. uratensis
V	S. nipponica, S. salicifolia
striate-reticulate	S. henryi

Pollen grains usually possess three apertures–the colpi. These were distributed symmetrically, elongated, and narrowed toward the poles, with granular aperture membranes. The mean length of the colpi ranged between 10.00 and 22.00 μm, with an average of 15.27 μm. The length of the colpi typically constituted 93% of the polar axis length (P) and ranged from 63 to 100%. Their width was variable and usually the greatest in the equatorial region. The endoapertures were typically located in the middle of the colpi, less frequently asymmetrically, usually singly. They were circular or elliptic in outline with irregular margins.

Key to identification of the Spiraea species under analysis based on pollen features.

1 Exine ornamentation striate-reticulate…… …… …… …… …… …… …… …… …… …… …… …… .S. henryi

1*Exine ornamentation striate…………………………………………………………………2

2 Exine ornamentation striate without perforations…… …… …… …… …… …… …… …… …… …… ……3

2* Exine ornamentation striate with perforations…… …… …… …… …… …… …… …… …… …… …… …6

3 Exine ornamentation subtype IA (grooves distinct and of medium width, striae narrow; perforations few or absent, small)……………………………………………………………4

3* Exine ornamentation type IIIA (grooves distinct and wide, striae narrow; perforations few or absent, small)…….………………………………………………………………………5

4. Striae wavy……………………………………………………………………S. dasyantha

4* Striae straight……………………………………………………… …… …… … S. hypericifolia

5. Striae run across the P axis………………………………………………S. cinerea ’Grefsheim’

5* Striae extending along the P axis…………………………………….………….S. splendens

6 Exine ornamentation type I……….………………………………….………………………7

6* Other type of exine ornamentation………………………………………………………….8

7 Exine ornamentation subtype IA (grooves distinct and of medium width, striae narrow; perforations few or absent, small)………………………………………………………S. media

7* Exine ornamentation subtype IB (grooves distinct and of medium width, striae wide; perforations few or absent, small)…………………………………………………………S. cana

8 Exine ornamentation type II……….……………….………………………………………9

8* Other types of exine ornamentation………………….…………………………………….11

9 Exine ornamentation subtype IIA (grooves very distinct and of medium width, striae narrow; perforations numerous, with different diameters (mostly small)……….…………… …… …….10

10 Range of E narrow (14–18), range of P/E ratio large (0.778–1.286), exine thick (Exp—1.2733)……………………………………………………………………………S. tomentosa

10* Other features……………………………………….………………… .S. alba var. latifolia, S. chamaedryfolia, S. douglasii, S. pubescens, S. thunbergii, S. trichocarpa, S. veitchii, S. wilsonii

11 Exine ornamentation type III…….….……………………………………………….….12

11* Other type of exine ornamentation…….…………….………………………………….13

12 Exine ornamentation type IIIA (grooves distinct and wide, striae narrow; perforations few or absent, small)……………………………………….…………………………………S. alba

12* Exine ornamentation type IIIB (grooves distinct, flattened and wide, striae narrow; perforations numerous and mostly very large)………………………… …………….S. betulifolia, S. billardii, S. elegans

13 Exine ornamentation type IV (grooves distinct and of medium width, flat, striae very wide; perforations very numerous and most extremely large)……………………………S. chinensis, S. japonica, S. uratensis

13* Exine ornamentation type V (grooves indistinct, flat and blurred, striae wide; perforations numerous and usually large)…………………………………………S. nipponica, S. salicifolia

Intrageneric and interspecific variability of pollen grains

The results of the MANOVA performed indicated that all the samples were significantly different with regard to all of the nine quantitative traits (Wilk’s λ = 0.05895; F24,725 = 11.20; P<0.0001). The results of ANOVA indicated that the main effects of the species were significant for all the nine observed traits (Tables 3–5 and S1 Table). The mean values and standard deviations for the observed traits indicated high variability among the tested samples, for which significant differences were found in terms of all the analyzed morphological traits (Tables 3–5 and S1 Table) (Figs 6–9).

Fig 6

The density plot of P by Spiraea taxa.

Fig 9

The density plot of P/E by Spiraea taxa.

The interspecific variability of the Spiraea pollen grains was examined based on nine selected quantitative features. Statistical analysis of the studied features indicated high variability among the tested species. The most variable biometric traits were E and P (Table 3), while lower variability was observed in the Le/P ratio and Exe (Tables 4 and 5). A ranking of variability of the observed traits is as follows: E > P > Le > Exp/P > Exe/E > Exp > P/E > Exe > Le/P. The most variable species for all the nine traits jointly were S. ×billardii, S. veitchii, S. nipponica and S. cana, while those with lower variability were S. dasyantha, S. elegans, S. trichocarpa and S. chamaedryfolia.

Table 4

Minimal, maximal and mean values as well as coefficients of variation (CV) for exine thickness along the polar axis (Exp), exine thickness along the equatorial diameter (Exe) and P/E of analyzed species.

Trait	Exp				Exe				P/E
Species	Mean		Min-Max	s.d.	Mean		Min-Max	s.d.	Mean		Min-Max	s.d.
S. alba	0.98	defg	0.2–1.6	0.28	0.98	cde	0.2–1.6	0.28	1.019	defghi	0.88–1.14	0.066
S. alba var. latifolia	0.787	gh	0.2–1.6	0.34	0.793	ef	0.2–1.6	0.38	1.024	defghi	0.90–1.13	0.062
S. betulifolia	0.653	h	0.2–1	0.23	0.653	f	0.2–1	0.23	0.972	hij	0.80–1.25	0.086
S. cana	1.253	bcd	0.6–1.6	0.34	1.253	abc	0.6–1.6	0.34	1.089	abcd	0.88–1.33	0.134
S. chamaedryfolia	0.913	efgh	0.2–1.6	0.43	0.913	def	0.2–1.6	0.43	0.943	j	0.80–1.00	0.057
S. chinensis	0.68	h	0.2–1	0.19	0.68	f	0.2–1	0.19	1.014	efghij	0.88–1.14	0.061
S. dasyantha	0.793	fgh	0.6–1.6	0.25	0.793	ef	0.6–1.6	0.25	1.026	defghi	0.88–1.17	0.075
S. douglasii	0.853	efgh	0.6–1.6	0.28	0.853	def	0.6–1.6	0.28	1.067	bcdef	0.88–1.29	0.089
S. elegans	1.053	bcdefg	0.6–1.6	0.31	1.053	abcde	0.6–1.6	0.31	0.993	ghij	0.88–1.14	0.059
S. henryi	0.973	defg	0.6–2	0.4	0.973	cde	0.6–2	0.4	1.02	defghi	0.88–1.14	0.07
S. hypericifolia	0.873	efgh	0.2–1.6	0.36	0.873	def	0.2–1.6	0.36	0.958	ij	0.80–1.13	0.076
S. japonica	1.013	cdefg	0.6–1.6	0.28	1.013	bcde	0.6–1.6	0.28	1.052	bcdefg	0.88–1.14	0.077
S. media	0.987	cdefg	0.6–1.6	0.3	0.987	bcde	0.6–1.6	0.3	0.996	fghij	0.78–1.29	0.096
S. nipponica	1.313	b	0.6–2	0.42	1.313	a	0.6–2	0.42	1.066	bcdef	0.89–1.29	0.098
S. pubescens	0.847	efgh	0.2–1.6	0.29	0.847	def	0.2–1.6	0.29	1.041	cdefgh	1.00–1.33	0.082
S. salicifolia	0.673	h	0.2–1.6	0.28	0.66	f	0.2–1.6	0.27	0.986	ghij	0.90–1.13	0.074
S. splendens	1.053	bcdefg	0.6–2	0.36	1.053	abcde	0.6–2	0.36	1.086	abcde	0.88–1.14	0.075
S. thunbergii	1.087	bcde	0.6–1.6	0.38	1.087	abcd	0.6–1.6	0.38	1.03	cdefghi	0.88–1.17	0.076
S. tomentosa	1.667	a	02-sty	0.48	1.273	ab	0.6–2	0.49	1.016	efghi	0.78–1.29	0.138
S. trichocarpa	1.08	bcdef	0.6–1.6	0.29	1.08	abcde	0.6–1.6	0.29	1.028	cdefghi	0.88–1.14	0.066
S. uratensis	1.22	bcd	0.6–1.6	0.36	1.22	abc	0.6–1.6	0.36	1.142	a	1.00–1.43	0.085
S. veitchii	1.02	cdefg	0.6–1.6	0.37	1.02	bcde	0.6–1.6	0.37	1.046	cdefg	0.88–1.14	0.086
S. wilsonii	0.907	efgh	0.6–1.6	0.34	0.893	def	0.6–1.6	0.34	1.057	bcdefg	0.88–1.17	0.079
S. ×billardii	0.84	efgh	0.6–1.6	0.28	0.84	def	0.6–1.6	0.28	1.121	ab	1.00–1.29	0.095
S. ×cinerea	1.273	bc	0.6–2	0.42	1.273	ab	0.6–2	0.42	1.1	abc	1.00–1.29	0.077
LSD0.001	0.288				0.29				0.072
ANOVA F	13.97				9.7				10.13
F value	<0.001				<0.001				<0.001

The correlation analysis performed indicated statistically significant correlation coefficients for 25 out of 36 coefficients (Table 7, Fig 10). Twelve out of 25 significantly correlated pairs of traits were characterized by positive correlation coefficients. Negative correlation coefficients were observed between P and Exp/P (-0.486), P and Exe/P (-0.530), E and Exe (-0.376), E and P/E (-0.629), E and Exp/P (-0.524), E and Exe/E (-0.666), Le and Exp (-0.432), Le and Exe (-0.396), Le and Exp/P (-0.645), Le and Exe/E (-0.600), Exp and Le/P (-0.592), Exe and Le/P (-0.430), and Le/P and Exp/P (-0.586). In the case of eleven pairs of traits, no significant correlation was established (Table 7, Fig 10).

Table 7

Correlation coefficients between observed traits.

Trait	P	E	Le	Exp	Exe	P/E	Le/P	Exp/P	Exe/E
P	1
E	0.898***	1
Le	0.914***	0.815***	1
Exp	-0.245	-0.347	-0.444*	1
Exe	-0.285	-0.426*	-0.409*	0.946***	1
P/E	-0.241	-0.641***	-0.212	0.365	0.471*	1
Le/P	0.132	0.109	0.522**	-0.587**	-0.422*	-0.025	1
Exp/P	-0.481*	-0.542**	-0.644***	0.966***	0.922***	0.382	-0.579**	1
Exe/E	-0.527**	-0.683***	-0.596**	0.870***	0.948***	0.614**	-0.367	0.918***	1

* P<0.05

** P<0.01

*** P<0.001

Fig 10

Heatmap for Pearson’s correlation coefficients between observed traits for Spiraea taxa (r0.05 = 0.38; r0.01 = 0.49; r0.001 = 0.60).

In the presented clustering according to the neighbor method of Euclidean distances, all the examined Spiraea species were divided into three groups (Fig 11). The first group included eight species: S. alba var. latifolia, S. ×billardii, S. chamaedryfolia, S. hypericifolia, S. betulifolia, S. salicifolia and S. chinensis. The second one consisted of five species: S. henri, S. wilsonii, S. douglasii, S. dasyantha and S. pubescens. The third one comprised all the other species and was divided into two subgroups: A–S. alba, S. media, S. elegans, and B–all other species (Fig 11). The phylogenetic tree of distribution of pollen morphological characters was presented on the Fig 12. The obtained divisions were very similar to those presented in the dendrogram (Fig 11).

Fig 11

Clustering (Neighbor Joining Method) of Spiraea species based on all nine morphological traits.

Fig 12

The phylogenetic tree of distribution of pollen morphological characters.

Fig 13 shows variability of the quantitative traits in the 25 Spiraea taxa under investigation in terms of the first two canonical variables. In the graph coordinates of the point for particular species are values for the first and second canonical variables, respectively. The first two canonical variables accounted for 80.08% of the total multivariate variability between individual species. Significant positive linear relationships with the first canonical variable were found for Exe, P/E, Exp/P and Exe/E (Table 8), while the first canonical variable correlated negatively with P, E and Le (Table 8). The second canonical variable had a significant positive correlation with Exp, Exe, Exp/P and Exe/E, but a significant negative correlation with Le/P (Table 8).

Fig 13

Distribution of 28 Spiraea taxa in space of two first canonical variables.

Table 8

Correlation coefficients between the first two canonical variables and original traits.

Trait	First canonical variable	Second canonical variable
P	-0.9376***	0.067
E	-0.986***	0.054
Le	-0.869***	-0.271
Exp	0.381	0.817***
Exe	0.441*	0.612***
P/E	0.537**	0.022
Le/P	-0.160	-0.811***
Exp/P	0.581**	0.730***
Exe/E	0.680***	0.459*
Percentage of explained multivariate variability	66.06	15.75

* P<0.05

** P<0.01

*** P<0.001

The greatest variation in terms of all the traits, based on the Mahalanobis distances measured, was found for the species S. salicifolia and the invasive S. dasyantha (the Mahalanobis distance between them amounted to 6.138). The greatest similarity was found for the species S. japonica and S. wilsonii (0.432). The values of the Mahalanobis distances for all the pairs of treatments are presented in Table 9. Pollen of the invasive S. tomentosa was distinctive due to it exhibiting the greatest variation in terms of all the traits together, based on the Mahalanobis distances measured. Pollen of these species was characterized by a relatively narrow range of E values (14–18 μm), the largest range of the P/E ratio (0.78–1.29) and the highest number of pollen grains (20) with the thickest exine (2 μm) (Tables 3–5).

Table 9

Mahalanobis distances between analyzed Spiraea species were calculated based on nine quantitative traits.

Species	S. alba	S. alba var. latifolia	S. betulifolia	S. cana	S. chamaedryfolia	S. chinensis	S. dasyantha	S. douglasii	S. elegans	S. henryi	S. hypericifolia	S. japonica	S. media	S. nipponica	S. pubescens	S. salicifolia	S. splendens	S. thunbergii	S. tomentosa	S. trichocarpa	S. uratensis	S. veitchii	S. wilsonii	S. ×billardii	S. ×cinerea
	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17	18	19	20	21	22	23	24	25
1
2	2.36
3	4.19	2.59
4	1.82	3.44	4.80
5	4.12	2.63	1.87	4.70
6	1.12	2.86	4.67	2.51	4.69
7	2.26	4.36	6.02	2.77	6.10	1.68
8	1.48	3.51	5.28	1.84	5.31	1.27	1.26
9	0.84	2.96	4.70	1.97	4.63	1.28	1.78	1.36
10	1.25	3.47	5.14	1.84	5.19	1.24	1.27	0.97	0.93
11	3.48	2.01	1.86	4.17	0.70	4.04	5.46	4.67	4.00	4.56
12	1.21	3.34	4.98	1.36	5.06	1.54	1.76	0.86	1.20	0.86	4.45
13	0.60	2.83	4.54	1.75	4.42	1.09	1.92	1.26	0.64	0.93	3.79	1.00
14	1.33	3.06	4.64	1.16	4.46	2.25	2.84	1.98	1.67	1.77	3.92	1.39	1.46
15	2.12	4.21	5.86	2.44	5.93	1.69	0.44	1.03	1.65	1.10	5.30	1.54	1.79	2.64
16	4.22	2.12	1.62	4.91	1.81	4.67	6.14	5.32	4.76	5.22	1.66	5.11	4.59	4.75	5.95
17	0.97	2.84	4.57	1.38	4.64	1.59	2.15	1.22	1.22	1.28	4.02	0.92	1.20	1.15	1.92	4.68
18	1.44	3.58	5.19	1.48	5.17	1.81	2.01	1.26	1.44	1.13	4.58	0.61	1.12	1.37	1.85	5.32	1.38
19	4.09	5.04	5.81	3.88	5.66	4.70	5.05	4.45	4.22	4.41	5.33	4.01	4.08	3.67	4.93	6.11	4.06	3.85
20	0.77	3.02	4.78	1.48	4.75	1.38	1.88	1.12	0.75	0.78	4.13	0.61	0.60	1.17	1.70	4.84	0.85	0.82	3.99
21	1.82	2.74	4.82	1.91	4.54	2.47	3.21	2.21	2.21	2.41	3.95	2.12	2.16	1.59	2.97	4.55	1.48	2.37	4.38	1.93
22	0.84	2.66	4.32	1.59	4.28	1.59	2.55	1.59	1.44	1.58	3.69	1.10	1.04	1.02	2.41	4.46	1.01	1.15	3.75	0.96	1.83
23	1.12	3.20	4.84	1.51	4.93	1.27	1.73	0.76	1.29	0.92	4.31	0.43	0.98	1.51	1.55	4.97	0.93	0.82	4.04	0.77	2.12	1.01
24	2.02	1.58	3.68	2.70	3.84	2.51	3.72	2.67	2.61	2.95	3.22	2.58	2.50	2.48	3.52	3.41	1.99	2.92	4.66	2.45	1.75	2.07	2.45
25	2.00	3.76	5.37	0.95	5.37	2.52	2.43	1.64	1.90	1.62	4.79	1.28	1.89	1.58	2.06	5.37	1.43	1.51	4.22	1.45	1.97	1.97	1.55	2.88

Dα = 3.73

Discussion

Palynological studies on the species from the genus Spiraea L. are not numerous. The least amount of research has been carried out in Europe [32], perhaps due to the small number of native species (7), and also in North America [23]. Most research papers come from Asia, which is the world center for the distribution of this genus [38–40, 63]. According to previous studies, the diagnostic features of the Spiraea pollen grains were exine ornamentation (striae and groove length, width and direction, and perforation number and diameter) and P and the P/E ratio [23, 32, 38–40], as well as endoaperture diameter [38]. In the opinion of Liu et al. [39] as well as the authors of this study, P or the P/E ratio were of a lesser diagnostic value, because the values of both these traits were often quite similar in all the studied taxa. Neither this study nor studies carried out by other researchers confirmed the observation of Polyakova and Gataulina [38] that endoapertures of some Spiraea species can be wide (S. hypericifolia, S. salicifolia), while others narrow (S. alpine, S. crenata, S. schlothaurae).

For the genus Spiraea other palynologists distinguished a striate or rarely striate-granulate exine ornamentation [23, 38, 39, 63]. Only the striate exine ornamentation was reported by Wrońska-Pilarek et al. [32] in S. tomentosa, while in the study presented here striate were observed in only one species, where it was the striate-reticulate ornamentation. All authors agree that the most important features of Spiraea pollen were the striae and the groove course and dimensions, as well as the presence or absence of perforations. Polyakova and Gataulina [38] reported that in the meadowsweet species the striae were variable in length and width. They crossed and branched, for the most part meridionally and in different directions. In the opinion of the above-mentioned authors, S. salicifolia was characterized by the ‘four-grooved’ pollen grains. However, such a feature was not found in the study presented here. Liu et al. [39] claimed that in the Spiraea species exine ornamentation was ‘fringe and stripe reticular’. Song et al. [40] recognized that all Spiraeeae taxa have a striate exine ornamentation. Indeed, in their opinion four types of striate exine ornamentation could be recognized. The first three types could each be divided into two subtypes based on the diameter of the perforations, which may be small or large. The significant importance of the diameters and number of perforations in the diagnosis of Spiraea species was also emphasized by Wrońska-Pilarek et al. [32].

The pollen grains of Spiraea species were described most often as small-sized [32, 38–40, 43, 44, 64] and rarely as medium-sized [38, 40]. In the research presented here, all the studied pollen grains were small. Additionally, Polyakova and Gataulina [38] reported that the significant variability of pollen size was typical of S. media and S. alpine. However, these results (P of 17.8–21.5 μm) for S. media were not confirmed in this study–the P and E values were average, ranging between 14 and 18 μm (Table 3). The differences may be due to the fact that the cited authors examined more samples (5) of S. media from populations scattered from Siberia to the Far East. Palynologists have mentioned different pollen shape types in the meadowsweet species. These pollen grains ranged from prolate and subspheroidal, spherical, trilobate, dolicho-round, oblong and fusiform [39], through elongated spheroidal [38], and from oblate to prolate [40]. The shape of the pollen in S. tomentosa was most frequently spheroidal, rarely oblate-spheroidal or prolate-spheroidal and very rarely subprolate or prolate [32]. The results in this study indicate that pollen shape in the 25 studied taxa was most frequently spheroidal and prolate-spheroidal, much less often oblate-spheroidal, suboblate or subprolate, and very rarely prolate. The reason for these differences may be connected with the detailed description of pollen shape classes in the last two studies.

The authors, as well as the other cited researchers, did not find Ubisch bodies (orbicules) in the studied taxa. These results were confirmed by Song et al. [40], who noted that orbicule distribution patterns indicated that the absence of orbicules was a synapomorphic condition of the more derived clade, comprising the genera Pentactina, Petrophytum, Kelseya and Spiraea.

Of the three invasive species studied (S. douglasii, S. japonica and S. tomentosa), the first two were not distinguished by specific pollen features, while S. tomentosa was distinctive as showing the greatest variation in terms of all the traits jointly, based on the Mahalanobis distances measured. It was found that the pollen features of this species differed from the other studied taxa in terms of the relatively narrow range of E, the largest range of the P/E ratio and the highest number of pollen grains with the thickest exine. In earlier studies by Wrońska-Pilarek et al. [32] carried out on a sample of 900 pollen grains of S. tomentosa, similar data was obtained. Thus, it could be concluded that the invasive S. tomentosa described shows a large variability of pollen features and a thick exine, which may indicate a greater ability to adapt its pollen to changing environmental conditions, guaranteeing it greater efficiency in taking over new areas. However, in the authors’ opinion to confirm this thesis the pollen of other invasive plant species should be compared, not only pollen from the genus Spiraea.

For the first time the intrageneric and interspecific variability of Spiraea pollen grains were studied based on nine quantitative traits. Statistical analysis of the investigated features indicated high variability among the species. The most variable species for all the nine traits jointly were S. ×billardii, S. veitchii, S. nipponica and S. cana, while those with lower variability were S. dasyantha, S. elegans, S. trichocarpa and S. chamaedryfolia. The most variable biometric traits were the E and P, while lower variability was recorded in the Le/P ratio and Exe. The ranking of variability in the observed traits is as follows: E > P > Le > Exp/P > Exe/E > Exp > P/E > Exe > Le/P.

It needs to be stressed here that the above conclusions were drawn based on studies conducted applying the method of collecting flowers from single shrubs of individual taxa, an approach commonly used in palynology. Thus the recorded results may not illustrate the entire pollen variability in the investigated species. We are aware of this limitation in the sampling procedure; nevertheless, we are of an opinion that due to the definitely conservative character of pollen grains the obtained results are representative. This thesis is confirmed by the results of our numerous palynological studies, such as Wrońska-Pilarek et al. [31, 34, 65, 66] concerning pollen morphology and variability in dozens of species from many genera (e.g. Crataegus L., Rosa L., Rubus L., Salix L., Spiraea L.). They were performed using the same method and they were typically comparable to data cited in the above-mentioned papers given by other palynologists. A similar situation is found in the genus Spiraea.

The most useful pollen morphological features included exine ornamentation type or subtype determined based on traits of grooves and striae (length, width and course) and perforations (presence or absence, number, diameter). The other biometric features (pollen size and shape, exine thickness) turned out to be useless, as they were similar in all the studied taxa.

The presented results showed that the morphological traits of pollen grains from 25 Spiraea taxa made it possible to isolate then individual species, while the other species formed groups of typically 2–3, up to 8 species. The same groups included species both from different and the same sections and series, including related and unrelated ones. Therefore, the pollen features do not fully confirm the classical taxonomic division of the studied genus. Thus, the morphological features of the pollen can be used in the taxonomy of the genus Spiraea as auxiliary traits to describe a particular species and taxonomic relations within the examined genus.

The statistical analysis of the studied pollen traits indicated high variability among the tested species. The most variable biometric traits were E and P, while lower variability was observed for the Le/P ratio and Exe.

Pollen of the invasive S. tomentosa differed from the other studied taxa in terms of the narrow range of the E values, the largest range of the P/E ratio and the highest number of pollen grains with the thickest exine. Pollen of the other invasive species (S. douglasi and S. japonica) did not differ significantly from the other studied species.

Complete morphological observations of quantitative features.

(PDF)

Click here for additional data file.

LM micrographs of equatorial view of pollen showing colpori and exine ornamentation features (exine thickness).

A, S. alba; B, S. betulifolia; C, S. cana; D, S. chamaedryfolia; E, S. chinensis; F, S. dasyantha; G, S. douglasii; H, S. elegans; I, S. henryi; J, S. hypericifolia; K, S. alba var. latifolia; L, S. media; M, S. media; N, S. nipponica; O, S. pubescens; P, S. salicifolia; Q, S. splendens; R, S. thunbergii; S, S. trichocarpa; T, S. uratensis; U, S. veitchii; V, S. wilsonii; W, S. ×billardii; X, S. ×cinerea; Y, S. tomentosa, A-Y.

(PDF)

Click here for additional data file.