Skip to main content
NCBI home page
Annals of Botany logoLink to Annals of Botany
. 2017 Jan 8;119(6):957–964. doi: 10.1093/aob/mcw261

Small genomes and large seeds: chromosome numbers, genome size and seed mass in diploid Aesculus species (Sapindaceae)

Anna Krahulcová 1, Pavel Trávníček 1, František Krahulec 1,*, Marcel Rejmánek 2
PMCID: PMC5604552  PMID: 28065925

Abstract

Background and Aims Aesculus L. (horse chestnut, buckeye) is a genus of 12–19 extant woody species native to the temperate Northern Hemisphere. This genus is known for unusually large seeds among angiosperms. While chromosome counts are available for many Aesculus species, only one has had its genome size measured. The aim of this study is to provide more genome size data and analyse the relationship between genome size and seed mass in this genus.

Methods Chromosome numbers in root tip cuttings were confirmed for four species and reported for the first time for three additional species. Flow cytometric measurements of 2C nuclear DNA values were conducted on eight species, and mean seed mass values were estimated for the same taxa.

Key Results The same chromosome number, 2n = 40, was determined in all investigated taxa. Original measurements of 2C values for seven Aesculus species (eight taxa), added to just one reliable datum for A. hippocastanum, confirmed the notion that the genome size in this genus with relatively large seeds is surprisingly low, ranging from 0·955 pg 2C–1 in A. parviflora to 1·275 pg 2C–1 in A. glabra var. glabra.

Conclusions The chromosome number of 2n = 40 seems to be conclusively the universal 2n number for non-hybrid species in this genus. Aesculus genome sizes are relatively small, not only within its own family, Sapindaceae, but also within woody angiosperms. The genome sizes seem to be distinct and non-overlapping among the four major Aesculus clades. These results provide an extra support for the most recent reconstruction of Aesculus phylogeny. The correlation between the 2C values and seed masses in examined Aesculus species is slightly negative and not significant. However, when the four major clades are treated separately, there is consistent positive association between larger genome size and larger seed mass within individual lineages.

Keywords: Aesculus, chromosome number, genome size, phylogeny, seed mass

INTRODUCTION

The genus Aesculus L. comprises woody plants, all native to the temperate Northern Hemisphere (Harris et al., 2009). Aesculus is known to be widespread in Tertiary forests that linked Europe, Asia and North America, and contemporary species are considered to be relics of these ancestral populations (Hardin, 1960; Danoghue and Smith, 2004). The number of recognized species ranges from 12 to 19, depending on acceptance of seven eastern Asian species, all putative constituents of section Calothyrsus (Xiang et al., 1998; Turland and Xia, 2005; Harris et al., 2009). Traditional assignment of species to five sections was exclusively based on morphology (Hardin, 1957a, b, 1960). Later, proposed Aesculus phylogenies were based either on both molecular and morphological data (Xiang et al., 1998) or exclusively on morphology (Forest et al., 2001). Recently, a robust phylogeny of Aesculus constructed by Harris et al. (2009) on the basis of extensive data on morphology, DNA sequences and fossils revealed four major clades, namely (1) an Asian clade, i.e. sect. Calothyrsus excluding A. californica (Spach) Nutt.; (2) sect. Macrothyrsus + A. californica; (3) sect. Pavia + sect. Parryana; and (4) sect. Aesculus (Table 1).

Table 1.

Four major clades in Aesculus recognized by Harris et al. (2009) and sections described by Hardin (1957a, b, 1960)

Clada Taxa involved
1 Asian clade (sect. Calothyrsus excl. A. californica): A. assamica, A. chinensis, A. indica*, A. polyneura, A. tsiangii, A. wangii, A. wilsonii
2 Sect. Macrothyrsus + A. californica: A. californica*, A. parviflora*
3 Sect. Pavia + sect. Parryana: A. flava, A. glabra*, A. glabra var. arguta*, A. hankensii, A. pavia*, A. parryi*, A. sylvatica
4 Sect. Aesculus: A. hippocastanum*, A. turbinata*
Open in a new tab
*

Taxa included in this study.

The chromosome number of 2n = 40, published so far for nine non-hybrid Aesculus species, corresponds to the diploid level (Fedorov, 1969; IPCN Chromosome Reports Database, 2015; Table 2). The triploid (2n = 60) and tetraploid (2n = 80) chromosome numbers (Fedorov, 1969) refer exclusively to taxa which arose from interspecific hybridization, namely to A. ×carnea, A. ×plantierensis and A. ×rubicunda. Chromosomes in Aesculus are small sized, as inferred from depicted metaphase chromosomes in A. hippocastanum L. (Pogan et al., 1980); the chromosome length in this species ranges approximately from 1 μm to 2 μm. Nevertheless, certain differences in chromosome size among species may exist, as, for example, between reportedly larger chromosomes of A. pavia L. compared with chromosomes of A. hippocastanum (Skovsted, 1929). Accordingly, differently sized parental chromosomes of A. pavia and A. hippocastanum were allegedly also detectable in their hybrid A. ×carnea (Skovsted, 1929); however, this finding was later questioned (Upcott, 1936). So far, the nuclear genome size has only been reported for A. hippocastanum (Bennett et al., 1982; Hanson et al., 2002 in Bennett and Leitch, 2005; Table 3). Using Feulgen densitometry, the first analysis of nuclear DNA content measured 0·250 pg 2C–1 in this species (Bennett et al., 1982). Such an extremely low C-value would assign A. hippocastanum to angiosperms with the smallest genomes, which are represented, for example, by Arabidopsis thaliana (Bennett and Leitch, 2005). Later on, using flow cytometry, about a five times higher value of nuclear DNA content was determined in A. hippocastanum, namely, 1·200 pg 2C–1 (Hanson et al., 2002 in Bennett and Leitch, 2005). Evidently, this value, although still indicating a rather small nuclear DNA amount in Aesculus, is considered to be more likely correct than that published previously (Bennett and Leitch, 2012).

Table 2.

Origins, sources and chromosome numbers of ten Aesculus taxa

Taxon Clade Area of origin Source Chromosome number* Counted
Reported
A. californica (Spach) Nutt. 2 California A n = 20a 2n = 40
A. glabra Willd. var. glabra 3 Midwestern USA B n = 20b 2n = 40
A. glabra Willd. var. arguta (Buckley) B.L. Rob. 3 Texas to Kansas C – 2n = 40
A. hippocastanum L. 4 Balkan Peninsula B 2n = 40c,d,e –
A. indica (Camb.) Hook I 1 Nordwest Himalaya D – 2n = 40
A. indica (Camb.) Hook II 1 Nordwest Himalaya E – 2n = 40
A. parryi A. Gray 3 Baja California F – 2n = 40
A. parviflora Walter 2 Southeast USA B n = 20b 2n = 40
A. pavia L. 3 Southeast USA B n = 20b 2n = 40
A. turbinata Blume 4 Japan C – 2n = 40
Open in a new tab

Sources: A, collected in Stebbins Cold Canyon, California; B, Průhonice Park, Czech Republic; C, Cusanelli Seed Company, Beachwood, NJ; D, University of California, Davis Arboretum; E, Quarry Hill Botanical Garden, CA; and F, collected 12 km south of El Rosario, Baja California.

*

Reported chromosome numbers in the following references:

Table 3.

Genome size and seed mass of ten Aesculus taxa

Taxon Clade 2C nuclear DNA content (pg ± s.e.)
 n Mean seed mass (g ± s.e. or range)
Reported Measured
A. californica (Spach) Nutt. 2 – 1·023 ± 0·004 5 39·7 ± 2·6*
A. glabra Willd. var. glabra 3 – 1·275 ± 0·004 5 7·8 (6·7–9·4)†
A. glabra Willd. var. arguta (Buckley) B.L. Rob. 3 – 1·273±0·010 3 11·4 (9·6–14·1)†
A. hippocastanum L. 4 1·20§ 1·122 ± 0·010 3 15·6 (13·3–19·6)†,‡
A. indica (Camb.) Hook I 1 – 1·063 ± 0·001 4 16·7 ± 0·8*
A. indica (Camb.) Hook II 1 – 1·084 ± 0·004 4 59·8 ± 3·9*
A. parryi A. Gray 3 – 1·202 ± 0·004 3 7·2 ± 0·4*
A. parviflora Walter 2 – 0·955 ± 0·006 3 3·8 ± 0·6*
A. pavia L. 3 – 1·253 ± 0·006 5 7·5 ± 0·3*
A. turbinata Blume 4 – 1·092 1 12·4 ± 0·6*
Open in a new tab

n indicates the number of individuals analysed.

*

Mean mass of 30 seeds.

Aesculus consists of perennial trees or shrubs reproducing by rather robust seeds (horse chestnuts) that are commonly eaten and dispersed by mammals. In A. hippocastanum, seed dormancy is released by storage for several months in moist and cold conditions. The seeds are recalcitrant, which means they are short lived and sensitive to desiccation (Obroucheva and Lityagina, 2007). Seed recalcitrance was also found in A. parviflora Walter, and most other Aesculus species are expected to possess this attribute (Seed Information Database, 2008). Two starchy cotyledons constitute the main mass of the seed (Beger, 1924; Hardin, 1957b; Obroucheva and Lityagina, 2007). What makes the genus Aesculus unique is that several of its species possess the highest seed masses not only in the family Sapindaceae and order Sapindales (besides Xylocarpus granatum, Meliaceae), but also among most of the angiosperm orders with the exception of Arecales, Ericales, Fabales and Laurales (Linkies et al., 2010).

In general, seed mass is an essential factor influencing seed dispersal. Because the nucleotypic effects of the amount of DNA on cell size and volume are additive in multicellular structures and organs, a causal relationship between nuclear genome size and seed mass is expected (Bennett, 1987). A positive correlation between both traits was identified within species and across species from the same genus and family (Bennett, 1987; Dąbrowska, 1992; Grotkopp et al., 2004; Knight et al., 2005; Knight and Beaulieu, 2008; Kim et al., 2015). However, no significant correlation between genome size and seed mass was found in 92 species of the genus Acacia, subgenus Phillodineae (Gallagher et al., 2011), and there is no significant correlation between genome size and seed size in diploid species of palms in the tribe Cocoseae and genus Syagrus (Henderson et al., 1995; Gunn et al., 2015). A broad-scale evaluation of data gathered for large numbers of species across genera, families and orders showed that species with a small 2C DNA content have a wide range of seed masses, while a large 2C DNA content is associated more with large seeds (Beaulieu et al., 2007). Considering the seed dispersal pattern, the highest values of genome size and seed mass were found in those plants whose seeds are dispersed via endozoochory (Bai et al., 2013).

In this study, we sought to find out whether the potential variation in genome size among Aesculus species is related to variation in their seed mass. Determining the chromosome numbers, we verified the ploidy level in eight available Aesculus taxa. A visual karyotyping should detect the potential gross interspecific differences in total metaphase chromosome sizes that have formerly been described in Aesculus (Skovsted, 1929). Such differences, if verified, may be associated with variation in the nuclear DNA content.

MATERIALS AND METHODS

Plants representing nine Aesculus taxa, namely eight species and one additional variety of A. glabra, were used in this study (Tables 1 and 2). Except A. hippocastanum, where material from three mature trees was used, seedlings obtained from seeds provided material for assessment of chromosome number and nuclear DNA content. The seeds, originating from natural populations, Průhonice park (Czech Republic), botanical gardens or seed companies (Table 2), were obtained in autumn in 2013 and 2014. Seeds were kept for several weeks in a refrigerator at 4 °C, after which they were planted separately in pots with moist garden soil and left in a cellar (temperature range approx. 8 − 12 °C). As soon as the seeds started to germinate (predominantly from January to March), the pots were transferred to daylight at room temperature. Those remaining seeds that did not germinate in a cellar by March were transferred to an unheated greenhouse where some of them eventually started to germinate. Young plants about 40 cm tall were repotted and cultivated in an open bed in the experimental garden at Průhonice. The final number of available individuals was limited by a low germinability of seeds and poor survival of young seedlings (especially in A. turbinata). Voucher specimens of all taxa are deposited in the Herbarium of the Institute of Botany, Průhonice, Czech Republic (PRA).

Chromosomes were counted for one to two plants for each of the taxa studied; A. hippocastanum was not examined, because its chromosome number has been confirmed several times in the literature (Table 2). Root tip cuttings of cultivated seedlings were used for chromosome preparations. The material was pre-treated with either a saturated solution of α-bromonaphthalene or 0·002 m 8-hydroxyquinoline for 3·5 h at room temperature, fixed overnight in a cold mixture of ethanol and acetic acid (3:1) and then stored in 70 % ethanol. After that, the root tips were hydrolysed in 1 n HCl at 60 °C for 10 min and rinsed in water, and the cut off meristematic tissue was squashed in a drop of lacto-propionic orceine (Dyer, 1963). The chromosomes were counted in at least five metaphases per plant, using both standard and phase contrast illumination (Olympus BX-51 microscope). Nuclear genome sizes were determined via flow cytometry using a Partec CyFlow instrument (Partec GmbH, Görlitz, Germany) equipped with a 532 nm solid state laser as the source of excitation light. The flow cytometric measurements were made for one to five plant individuals per taxon (Table 3). Samples were prepared by the two-step protocol described by Doležel et al. (2007). Solanum pseudocapsicum (1C = 1·295 pg; Temsch et al., 2010) was selected as an appropriate reference standard for analyses, and propidium iodide was used as the fluorescent dye. Young intact leaf tissue of an Aesculus sample and a proportional piece of the internal reference standard were processed together in each analysis (for details of the procedure, see Suda et al. 2010). At least three flow cytometric analyses performed on different days were conducted to avoid day to day fluctuations and used for calculation of the genome size values for included taxa. Weighing of 30 freshly mature seeds was performed for most of the species (Table 3), while the published data on seed mass for A. glabra, A. glabra var. arguta and A. hippocastanum were taken from Bonner et al. (2008). Statistical analyses were performed in R version 3.2.3 (R Core Team, 2015).

RESULTS

The same chromosome number, 2n = 40, was determined in all investigated taxa (Table 2). Chromosome numbers for A. indica, A. parryi and A. turbinata are reported here for the first time. Chromosomes in all species are small, approx. 1 − 2 μm long (Fig. 1). Comparing the metaphase chromosome sets among taxa examined, we did not find any conspicuous differences in chromosome size.

Fig. 1.

Fig. 1.

Open in a new tab

Representative example of metaphase chromosomes from root tip meristem of Aesculus (2n = 40 in all four species). (A) A. indica II; (B) A. parryi; (C) A. pavia; (D) A. turbinata.

The genome size was measured for the first time for A. californica, A. glabra var. glabra, A. glabra var. arguta, A. indica, A. parviflora, A. parryi, A. pavia and A. turbinata (Table 3). The nuclear DNA content in Aesculus ranges from 0·955 pg 2C–1 in A. parviflora to a 2C value of 1·275 pg in A. glabra var. glabra (Table 3), i.e. genome sizes differ nearly 30 % between species with the smallest and largest genomes within the genus. The pattern of variation in genome size and a post-hoc Tukey HSD test suggest a statistically distinct 2C value in most of the species analysed. Nevertheless, neither the two varieties of A. glabra, nor the three taxa with medium genome size (A. indica II, A. turbinata and A. hippocastanum), are statistically distinguishable from each other. The genome sizes seem to be distinct and non-overlapping among the four major Aesculus clades (Table 3; Fig. 2A, B). These distinct ranges of 2C values point to evolutionary genome size conservatism within clades. The 2C values increase from clade 2 (A. parviflora + A. californica), through clade 1 (A. indica) and clade 4 (A. turbinata + A. hippocastanum), to clade 3 (A. parryi, A. pavia, A. glabra and A. glabra var. arguta) (Fig. 2A, B). Mean seed mass values for seven examined Aesculus taxa ranged from 3·8 to 59·8 g (Table 3). Overall, the correlation between the 2C values and seed masses across all examined species (ten taxa) is slightly negative, but not significant (P = 0·35). However, when the four major clades are treated separately, there is consistent positive association between larger genome size and larger seed mass within individual clades (Fig. 2A). The sample size is too small (even after including an intraspecific relation in A. indica) to perform a meaningful sign test.

Fig. 2.

Fig. 2.

Open in a new tab

(A) The relationship between seed mass and 2C DNA content in selected species of the four major lineages of Aesculus. 1 (from left to right): A. indica I–A. indica II; 2: A. parviflora–A. californica; 3: A. parryi–A. pavia–A. glabra var. arguta–A. glabra var. glabra; 4: A. turbinata–A. hippocastanum. (B) Phylogram of Aesculus based on analysis of DNA and morphology (modified, adapted from Harris et al., 2009, p. 17, fig. 4A) supplemented by genome size data (pg DNA 2C–1, column in the middle). Numbering of lines (right), that delimit the four major lineages of Aesculus, corresponds to (A).

DISCUSSION

The chromosome number of 2n = 40, confirmed here for four Aesculus species previously reported in the literature, known for five other species in the literature and reported for the first time here for three additional species (four taxa) (Table 2), seems to be conclusively the universal 2n number for all non-hybrid species in this genus. Interestingly, 2n = 40 is a unique number in Sapindaceae and very rare in the order Sapindales (Lannea and Mangifera in Anacardiaceae may be the only exceptions) (Fedorov, 1969).

Original measurements of 2C values for seven Aesculus species (eight taxa), added to the single reliable datum for A. hippocastanum, confirmed the notion that the genome size in this genus with relatively large seeds is surprisingly low (0·955–1·275 pg 2C–1, compared with mean and median 2C values of angiosperms which are 11·8 and 5·0 pg, respectively; Suda et al., 2015). According to the classification of DNA amounts among embryophytes by Leitch et al. (2005), the genome sizes in Aesculus species belong to the category of ‘very small’ (≤2·8 pg 2C–1) genomes. Mean seed mass values for seven Aesculus species were in agreement with published data, insofar as data for studied species were available (Bonner et al., 2008; Seed Information Database, 2008). The only notable exception was A. parviflora (mean seed mass 3·8 g; Table 3), for which the published data were higher by 2–6 g (table 5·6 in Chanon, 2005). Two extremely different seed mass values for accessions I and II of A. indica may represent two ‘unofficial’ varieties lakut handun (small seeded) and budh handun (large seeded) found in the Kashmir valley (Rafiq et al., 2015).

Small genome sizes in Aesculus are not unusual among angiosperm woody species with large seeds. The reported 2C value for Mangifera indica is 1·8 pg (Bennett and Leitch, 2005). On the other hand, the 2C value for Persea americana, which has seed mass values comparable with some large-seeded Aesculus species, is somewhat larger (4·6 pg; Bennett and Leitch, 2012). Taking only temperate woody angiosperms into account, where mean and median 2C values are 2·24 and 1·6 pg, respectively (Ohri, 2005), Aesculus genome size is still rather small. Also, it is small within its own family, Sapindaceae (mean 2C value = 2·71 ± 0·37 pg, median = 2·02 pg, n = 62; Bennett and Leitch, 2012; Coulleri et al., 2014).

A positive association between larger genome size and larger seed mass within individual clades (all slopes in Fig. 2A are positive) is consistent with other studies that found a positive correlation between genome size and seed mass (Bennett, 1987; Dąbrowska, 1992; Grotkopp et al., 2004; Knight et al., 2005; Knight and Beaulieu, 2008; Kim et al., 2015). Slight differences in genome sizes between two A. indica accessions (I and II) are also positively associated with differences in their seed mass values (Table 3; Figs 2A and 3). Even if this is only one comparison, it is consistent with positive intraspecific correlations between genome size and seed mass in some other species (Caceres et al., 1998; Chung et al., 1998; Benor et al., 2011; Aliyu, 2014). The results presented here support the generalization made by Beaulieu et al. (2007): ‘Large seed masses have evolved in species with both small and large genomes, but large genome species rarely have small seed sizes … genome size may set a minimum seed mass, that increases with increasing genome size, but the maximum seed mass for any given genome size may be determined by other factors.’ The range of seed masses in A. indica (Figs 2A and 3) is an excellent illustration of this last point. Large seeds in Aesculus may have several adaptive advantages evolved over relatively short evolutionary times (Leishman and Westoby, 1994; Mendoza and Dirzo, 2009).

Fig. 3.

Fig. 3.

Open in a new tab

Two fruit/seed morphotypes, I (UCD Arboretum) - upper row and II (Quarry Hill Botanical Garden) - lower row, of Aesculus indica.

Based on species used in this study, there is a clear division among the four major Aesculus clades in 2C DNA content (Fig. 2A, B). Clade 3 (sections Pavia and Parryana), with the largest genomes in the genus, seems to be isolated from other clades. Clade 2 (A. californica + A. parviflora) exhibits the lowest 2C values, and the two remaining major clades (Asian clade and sect. Aesculus) are situated in the middle. These results provide an extra support for the most recent reconstruction of Aesculus phylogeny (Harris et al., 2009; Fig. 2B). In spite of the fact that our sample size is too small for any statistical evaluations, a positive trend of an increase of seed mass and 2C values in individual lineages (Fig. 2A) is in agreement with many generalizations that have been proposed. Unfortunately, the exact mechanism of this relationship is still not known (Beaulieu et al., 2007). Finally, invasiveness (spread in areas of introduction) of seed plants is often associated with small genome sizes, similar to those we are finding in Aesculus (Kubešová et al., 2010; Pandit et al., 2014). Several Aesculus species have been introduced as ornamentals many times into many areas where they had not been native. However, there is not one Aesculus species among >750 invasive tree and shrub species currently known globally (Rejmánek and Richardson, 2013). This may be in agreement with the fact that small genome sizes are more reliable predictors of invasiveness in some gymnosperms and herbaceous angiosperms than in woody angiosperms (Grotkopp et al., 2004; Chen et al., 2010). Surprisingly, studies of Aesculus seed dispersal are rather rare (Thompson and Thompson, 1980; Hoshizaki et al., 1999; Irie and Tsuyuzaki, 2001) and focused mostly on the Japanese species A. turbinata. Lack of efficient dispersal mechanisms for relatively large seeds is probably the major reason for limited spread of introduced Aesculus species. Therefore, Aesculus species seem to be relatively safe, inconsequential introductions. Nevertheless, dispersal by vertebrates and by water remains a possibility in some areas.

Including both the data published earlier and those presented here, the uniform chromosome number of 2n = 40 is currently known for 13 extant recently recognized Aesculus species. Consequently, at least 70 % of all Aesculus species are diploid, considering the maximum total number of 19 non-hybrid species recognized at present. In spite of invariable chromosome number detected within the genus, the nuclear DNA content varies among the eight examined species, ranging up to a 30 % difference between the species with the smallest and largest genome. The 2C values seem to be distinct and non-overlapping among the major Aesculus clades, supporting the most recent reconstruction of Aesculus phylogeny (Harris et al., 2009). Irrespective of detected variation, the genome sizes in Aesculus species rank among very small genomes that are, however, not exceptional among woody angiosperms. Although Aesculus is in general a large-seeded genus, there is a substantial variation in seed mass among species. The correlation between 2C values and seed masses in examined Aesculus species is slightly negative and not significant. If the four major clades are treated separately, a consistent positive association between larger genome size and larger seed mass is suggested within individual lineages. Although the small genome sizes may suggest that the introduced Aesculus species should be invasive, the relatively large seeds are probably the main reason for the lack of efficient dispersal mechanisms.

ACKNOWLEDGEMENTS

We thank Daniel Rejmánek for obtaining A. parryi seeds, and Jiří Burda for obtaining seeds of three Aesculus species from Průhonice park. James Doyle and Eva Grotkopp are acknowledged for valuable comments on the first draft of this article. This work was supported by the Czech Academy of Sciences (long-term research & development project No. RVO 67985939 to A.K., P.T. and F.K.) and by the University of California Agricultural Experiment Station (to M.R.).

LITERATURE CITED

  1. Aliyu OM. 2014. Analysis of absolute nuclear DNA content reveals a small genome and intra-specific variation in Cashew (Anacardium occidentale L.), Anacardiaceae. Silvae Genetica 63: 285−293. [Google Scholar]
  2. Bai C-k, Cao B, Li G-s.. 2013. Correlations of plant seed dispersal pattern with genome size and 1000-seed mass. Chinese Journal of Ecology 32: 832−837. [Google Scholar]
  3. Beaulieu JM, Moles AT, Leitch IJ, Bennett MD, Dickie JB, Knight CA.. 2007. Correlated evolution of genome size and seed mass. New Phytologist 173: 422−437. [DOI] [PubMed] [Google Scholar]
  4. Beger H. 1924. 77. Fam. Hippocastanaceae. Rosskastaniengewächse In: Hegi G, ed. Illustrierte Flora von Mitteleuropa, V/1. München: J.F. Lehmanns Verlag, 296–309. [Google Scholar]
  5. Bennett MD, Smith JB, Heslop-Harrison JS.. 1982. Nuclear DNA amounts in angiosperms. Proceedings of the Royal Society B: Biological Sciences 216: 179−199. [DOI] [PubMed] [Google Scholar]
  6. Bennett MD. 1987. Variation in genomic form in plants and its ecological implications. New Phytologist 106 (Suppl.): 177−200. [Google Scholar]
  7. Bennett MD, Leitch IJ.. 2005. Nuclear DNA amounts in angiosperms – progress, problems and prospects. Annals of Botany 95: 45−90. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Bennett MD, Leitch IJ.. 2012. Plant DNA C-values database (release 6·0, December 2012) http://data.kew.org/cvalues/ (last accessed 8 December 2016).
  9. Benor S, Fuchs J, Blattner FR.. 2011. Genome size variation in Corchorus olitorius (Malvaceae s.l.) and its correlation with elevation and phenotypic traits. Genome 54: 575−585. [DOI] [PubMed] [Google Scholar]
  10. Bonner FT, Karrfalt RP, Nisley RG, eds. 2008. The woody plant seed manual. USDA Forest Service, Agriculture Handbook 727. [Google Scholar]
  11. Caceres ME, De Pace C, Scarascia Mugnozza GT, Kotsonis P, Caccarelli M, Cionini PG.. 1998. Genome size variations within Dasypyrum villosum: correlations with chromosomal traits, environmental factors and plant phenotypic characteristics and behavior in reproduction. Theoretical and Applied Genetics 96: 559−567. [Google Scholar]
  12. Chanon AM. 2005. Studies on the reproductive capacity of Aesculus parviflora and Aesculus pavia: opportunities for their improvement through interspecific hybridization. PhD Thesis, Ohio State University, USA.
  13. Chen G-Q, Guo S-L, Yin L-P.. 2010. Applying DNA C-values to evaluate invasiveness of angiosperms: validity and limitation. Biological Invasions 12: 1335−1348. [Google Scholar]
  14. Chung J, Lee J-H, Arumuganathan K, Graef GL, Specht JE.. 1998. Relationships between nuclear DNA content and seed and leaf size in soybean. Theoretical and Applied Genetics 96: 1064−1068. [Google Scholar]
  15. Coulleri JP, Urdampilleta JD, Ferrucci MS.. 2014. Genome size evolution in Sapindaceae at subfamily level: a case study of independence in relation to karyological and palynological traits. Botanical Journal of the Linnean Society 174: 589−600. [Google Scholar]
  16. Dąbrowska J. 1992. Chromosome number and DNA content in taxa of Achillea L. in relation to the distribution of the genus. Prace Botaniczne 49: 1−83. [Google Scholar]
  17. Danoghue MJ, Smith SA.. 2004. Patterns in the assembly of temperate forests around the Northern Hemisphere. Philosophical Transactions of the Royal Society B: Biologial Sciences 359: 1633−1644. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Daws MI, Lydall E, Chmielarz P, Leprice O, et al. 2004. Developmental heat sum influences recalcitrant seed traits in Aesculus hippocastanum across Europe. New Phytologist 162: 157−166. [Google Scholar]
  19. Dobeš C, Vitek E.. 2000. Documented chromosome number checklist of Austrian vascular plants. Wien: Verlag des Naturhistorischens Museums. [Google Scholar]
  20. Doležel J, Greilhuber J, Suda J.. 2007. Estimation of DNA content in plants using flow cytometry. Nature Protocols 2: 2233−2244. [DOI] [PubMed] [Google Scholar]
  21. Dyer AF. 1963. The use of lacto-propionic orcein in rapid squash methods for chromosome preparations. Stain Technology 38: 85−90. [Google Scholar]
  22. Fedorov AA, ed. 1969. Chromosome numbers of flowering plants. Leningrad: Nauka. [Google Scholar]
  23. Forest F, Drouin JN, Charest R, Brouilett L, Bruneau A.. 2001. A morphological phylogenetic analysis of Aesculus L. and Billia Peyrr. (Sapindaceae). Canadian Journal of Botany 79: 154−169. [Google Scholar]
  24. Gallagher RV, Leishman MR, Miller JT. et al. 2011. Invasiveness in introduced Australian acacias: the role of species traits and genome size. Diversity and Distributions 17: 884−897. [Google Scholar]
  25. Grotkopp E, Rejmánek M, Sanderson MJ, Rost TL.. 2004. Evolution of genome size in pines (Pinus) and its life-history correlates: supertree analyses. Evolution 58: 1705−1729. [DOI] [PubMed] [Google Scholar]
  26. Gunn BF, Baudouin L, Beulé T, et al. 2015. Ploidy and domestication are associated with genome size variation in palms. American Journal of Botany 102: 1625−1633. [DOI] [PubMed] [Google Scholar]
  27. Hanson L, Leitch IJ, Bennett MD.. 2002. Unpublished data from the Jodrell Laboratory, Royal Botanic Garden Kew. In: Bennett MD, Leitch IJ. 2012. Plant DNA C-values database (release 6·0, December 2012) http://data.kew.org/cvalues/CvalServlet?querytype=2. (last accessed 8 December 2016).
  28. Hardin JW. 1957a. A revision of the American Hippocastanaceae. Brittonia 9: 145−171. [Google Scholar]
  29. Hardin JW. 1957b. A revision of the American Hippocastanaceae − II. Brittonia 9: 173−195. [Google Scholar]
  30. Hardin JW. 1960. Studies in the Hippocastanaceae. V. Species of the Old World. Brittonia 12: 26−38. [Google Scholar]
  31. Harris AJ, Xiang Q-Y, Thomas DT.. 2009. Phylogeny, origin and biogeographic history of Aesculus L. (Sapindales) − an update from combined analysis of DNA sequences, morphology and fossils. Taxon 58: 108−126. [Google Scholar]
  32. Henderson A, Galeano G, Bernal R.. 1995. Palms of the Americas. Princeton, NJ: Princeton University Press. [Google Scholar]
  33. Hoshizaki K, Suzuki W, Nakashizula T.. 1999. Evaluation of secondary dispersal in a large-seeded tree Aesculus turbinata: a test of directed dispersal. Plant Ecology 144: 167−176. [Google Scholar]
  34. Irie K, Tsuyuzaki S.. 2001. Dispersal timing, palatability and caching of acorns of Aesculus turbinata. Plant Biosystems 145: 798−801. [Google Scholar]
  35. IPCN Chromosome Reports Database, Missouri Botanical Garden. 2015. http://www.tropicos.org/Project/IPCN (last accessed 8 December 2016).
  36. Kim S, Han M, Rayburn AL.. 2015. Genome size and seed mass analyses in Cicer arietinum (Chickpea) and wild Cicer species. HortScience 50: 1751−1756. [Google Scholar]
  37. Knight CA, Beaulieu JM.. 2008. Genome size scaling through phenotype space. Annals of Botany 101: 759−766. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Knight CA, Molinari NA, Petrov DA.. 2005. The large genome constrain hypothesis: evolution, ecology and phenotype. Annals of Botany 95: 177−190. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Kubešová M, Moravcová L, Suda J, Jarošík V, Pyšek P.. 2010. Naturalized plants have smaller genomes than their non-invading relatives: a flow cytometric analysis of the Czech alien flora. Preslia 82: 81−96. [Google Scholar]
  40. Leishman MR, Westoby M.. 1994. The role of large seeds in shaded conditions: experimental evidence. Functional Ecology 8: 205−214. [Google Scholar]
  41. Leitch IJ, Soltis DE, Soltis PS, Bennett MD.. 2005. Evolution of DNA amounts across land plants (Embryophyta). Annals of Botany 95: 207−217. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Linkies A, Graeber K, Knight C, Leubner-Metzger G.. 2010. The evolution of seeds. New Phytologist 186: 817−831. [DOI] [PubMed] [Google Scholar]
  43. Mendoza E, Dirzo R.. 2009. Seed tolerance to predation: evidence from the toxic seeds of the buckeye tree (Aesculus californica; Sapindaceae). American Journal of Botany 96: 1255−1261. [DOI] [PubMed] [Google Scholar]
  44. Měsíček J, Javůrková-Jarolímová V.. 1992. List of chromosome numbers of the Czech vascular plants. Praha: Academia. [Google Scholar]
  45. Obroucheva NV, Lityagina SV.. 2007. Dormancy release and germination in recalcitrant Aesculus hippocastanum seeds. Dendrobiology 57: 27−33. [Google Scholar]
  46. Ohri D. 2005. Climate and growth form: the consequences for genome size in plants. Plant Biology 7: 449−458. [DOI] [PubMed] [Google Scholar]
  47. Ordnuff R, Lloyd R.. 1965. Documented chromosome numbers of plants. Madroño 18: 122−126. [Google Scholar]
  48. Pandit MK, White SM, Pocock MJO.. 2014. The contrasting effects of genome size, chromosome number and ploidy level on plant invasiveness: a global analysis. New Phytologist 203: 697−703. [DOI] [PubMed] [Google Scholar]
  49. Pogan E, Wcislo H, Jankun A.. 1980. Further studies in chromosome numbers in Polish Angiosperms. Part XIII. Acta Biologica Cracoviensia, Series Botanica 22: 37−69. [Google Scholar]
  50. R Core Team. 2015. R: A language and environment for statistical computing R Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org/ (last accessed 18 August 2016).
  51. Rafiq SI, Jan K, Singh S, Saxena DC.. 2015. Physicochemical, pasting, rheological, thermal and morphological properties of horse chestnut starch. Journal of Food Science and Technology 52: 5651−5660. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Rejmánek M, Richardson DM.. 2013. Trees and shrubs as invasive alien species – 2013 update of the global database. Diversity and Distributions 19: 1093−1094. [Google Scholar]
  53. Seed Information Database. 2008. Royal Botanic Gardens, Kew (release 7·1, May 2008). http://data.kew.org/sid/SidServlet?Clade=&Order=&Family = &APG=off&Genus=Aesculus&Species=&StorBehav=0 (last accessed 18 August 2016).
  54. Skovsted A. 1929. Cytological observations of the genus Aesculus L. with some observations on Aesculus carnea Willd., a tetraploid species arisen by hybridization. Hereditas12: 64−70. [Google Scholar]
  55. Suda J, Meyerson LA, Leitch IJ, Pyšek P.. 2015. The hidden side of plant invasions: the role of genome size. New Phytologist 205: 994−1007. [DOI] [PubMed] [Google Scholar]
  56. Suda J, Trávníček P, Mandák B, Berchová-Bímová K.. 2010. Genome size as a marker for identifying the invasive alien taxa in Fallopia section Reynoutria. Preslia 82: 97−106. [Google Scholar]
  57. Temsch EM, Greilhuber J, Krisai R.. 2010. Genome size in liverworts. Preslia 82: 63−80. [Google Scholar]
  58. Thompson DC, Thompson PS.. 1980. Food habits and caching behavior of urban grey squirrels. Canadian Journal of Zoology 58: 701−710. [Google Scholar]
  59. Turland N, Xia N.. 2005. A new combination in Chinese Aesculus (Hippocastanaceae). Novon 15: 488−489. [Google Scholar]
  60. Upcott M. 1936. The parents and progeny of Aesculus carnea. Journal of Genetics 33: 135−149. [Google Scholar]
  61. Xiang Q-Y, Crawford DJ, Wolfe AD, Tang Y-C, DePamphilis CV.. 1998. Origin and biogeography of Aesculus L. (Hippocastanaceae): a molecular phylogenetic perspective. Evolution 52: 988−997. [DOI] [PubMed] [Google Scholar]

Articles from Annals of Botany are provided here courtesy of Oxford University Press

RESOURCES