In this study we reported for the first time the chromosome number and the karyotype of Chrysolaena cristobaliana. Besides, the karyotype of Chrysolaena sceptrum was recorded for the first time, as well as a new cytotype.

The base chromosome number x=10 is considered characteristic of Chrysolaena and clearly distinguishes this genus from the remaining American groups of the tribe. This number also has been found in the two species here analyzed, which is consistent with previous studies carried out in others Chrysolaena species (Dematteis 1997a, 1997b, 1998, 2002, 2009, Dematteis et al. 2007, Angulo and Dematteis 2009a, 2009b, Via do Pico and Dematteis 2012a, 2012b). The only cytological record available for Chrysolaena sceptrum was reported by Dematteis (2002) for a population from Paraguay, which showed an octoploid cytotype with 2n=8x=80. In our study we report a new cytotype (tetraploid, 2n=4x=40) and the first karyotype analysis of the species. Chrysolaena sceptrum has been considered by some authors as a synonym of Chrysolaena cognata (Jones 1981, Robinson 1988), or a variety of this species (Cabrera 1944). However, there are many differences between these two entities, such as the shape and width of the leaves, and the number of florets per head, among the most conspicuous features (Dematteis 2009). Dematteis (2009) refers to the number of chromosomes as another distinguishing characteristic between these two species. Chrysolaena cognata presents 2n=20, 40, and 60, and Chrysolaena sceptrum presents octoploid cytotype (2n=80). Nevertheless, further counts reported also the octoploid cytotype (2n=80) for Chrysolaena cognata (Via do Pico and Dematteis 2012b), and in this study we report the tetraploid cytotype (2n=40) for Chrysolaena sceptrum. For this reason, the number of chromosomes already would not be a useful character to distinguish the two species each other. Although Chrysolaena sceptrum has been cited for Argentina (Cabrera 1944), it has not been found to date in that country. The specimens cited by Cabrera (1944) clearly belong to Chrysolaena cognata, which is widely distributed in Argentina. The geographic distribution of Chrysolaena sceptrum is exclusive of Mato Grosso and Mato Grosso do Sul in Brazil and east of Paraguay. Morphological differences listed above and geographical distribution would be the most conspicuous features to distinguish between Chrysolaena cognata and Chrysolaena sceptrum.

Chrysolaena cristobaliana has never been cytologically analyzed. This study reports as novelty the base chromosome number (x=10), the chromosome number (2n=7x=70, heptaploid) and the karyotype of the species. Chrysolaena cristobaliana and Chrysolaena sceptrum are closely related species. They are distributed in the same geographic region, and even populations of both entities can be found living in the same area. The main morphological features that differentiate these two species are the branch of the stem and the leaf shape. Chrysolaena cristobaliana presents densely branched and elliptical, lanceolate to oblanceolate leaves, whilst Chrysolaena sceptrum has single stems and narrowly lanceolate to linear leaves. The chromosome counts realized here show that these two entities can also be distinguished by the chromosome number.

Chrysolaena cristobaliana is also closely related to Chrysolaena cognata, one of the most widely distributed species of the genus. However, both species differ in the leaf shape, the pubescence type, the florets number and the geographical distribution. The results of this study added the chromosome number as a feature to distinguish these two closely related species.

Jones (1995) estimated that accessory chromosomes occur in about 10–15% of flowering plant species. Their distribution among angiosperm families is quite heterogeneous. They have been described in more than 150 species of Asteraceae and Poaceae. Generally, these chromosomes are heterochromatic, small, and very variable in number between individuals (Guerra 1988). The irregular distribution of B chromosomes among Angiosperm families suggests that species in certain groups are more likely to bear them than species in other families (Levin et al. 2005). Numerous cases of accessory chromosomes are known in the tribe Vernonieae (Angulo and Dematteis 2009b, Dematteis 1998, Galiano and Hunziker 1987, Oliveira et al. 2007a, Angulo and Dematteis 2012). The occurrence of accessory chromosomes in Chrysolaena cristobaliana and Chrysolaena sceptrum are new records for the genus. In both species, Bs presented a metacentric morphology and similar size. Previous studies reported accessory chromosomes in Chrysolaena flexuosa, Chrysolaena cognata, Chrysolaena propinqua, and Chrysolaena verbascifolia (Via do Pico and Dematteis 2012b). In most of this species, Bs are present in a low frequency (1-4), as reported here in Chrysolaena sceptrum. However, in Chrysolaena verbascifolia it has been observed between 0 and 7 B chromosomes per cell, as well as in Chrysolaena cristobaliana. This variation was observed between individuals of the same population and even within the same individual. Another case in which it has been observed a high frequency of B chromosomes is Lepidaploa canescens (Kunth) H. Robinson, 1990, (sub nom=Vernonia geminata). In this species it has been observed from 0 to 6 accessory chromosomes amongst the cells of a given individual (Oliveira et al. 2007a). The B chromosomes of Chrysolaena cristobaliana and Chrysolaena sceptrum varied in cells of the same plant, which suggests non disjunction in the mitotic anaphase. Variation in chromosome number in the same plant is a rule used to discriminate B chromosomes from the normal chromosome complement (Jones and Rees 1982). Apparently, there is no difference in frequency between diploid and polyploids species (Jones and Rees 1982, Palestis et al. 2004, Trivers et al. 2004), but, there is a trend suggesting that Bs have a higher frequency in species with a large genome size (Trivers et al. 2004). So far, in the analyzed species of Chrysolaena, there is no differences in frequency between diploids and polyploids, since in Chrysolaena verbascifolia (2n=20) and Chrysolaena cristobaliana (2n=70) it has been observed the same frequency of Bs.

In species of plants and animals, that carries B chromosomes, those individuals from a given population with and without Bs, cannot generally be phenotypically distinguished from each other. However, in some species, there are instances in which B chromosomes change certain morphological characteristics (Jones and Rees 1982, Jones and Houben 2003) or cause some selectively advantageous effects (Teoh and Jones 1978, Jones and Rees 1982). Apparently, the B chromosomes found in Chrysolaena species have no effect on the phenotype or development of the individuals.

Chrysolaena cristobaliana and Chrysolaena sceptrum were never been karyotypically characterized. Karyotype analysis is essential for the cytogenetic characterization of species and to examine the variation between its individuals and/or populations. The comparison of karyotypes of different species also allows the taxonomic and evolutionary analysis of a taxon, such as a genus. Besides, many times, differences in karyotype asymmetry can indicate how these chromosomes have diversified in size and morphology within a group (Guerra 1988). The karyotypes of the taxa here studied are formed by metacentric and submetacentric chromosomes, and are quite symmetrical due the predominance of metacentric chromosomes. This is a common feature in species of the tribe Vernonieae (Ruas et al. 1991, Dematteis 1997a, 1997b, 1998, Dematteis and Fernández 1998, 2000, Oliveira et al. 2007a, 2007b, Angulo and Dematteis 2009b, Via do Pico and Dematteis 2012b). Chrysolaena sceptrum presents the most symmetric karyotype, which is reflected in the highest average centromeric index and the lowest intrachromosomal asymmetry coefficient (A₁). Chrysolaena cristobaliana shows the highest interchromosomal asymmetry coefficient (A₂), which reflects the amplitude of its chromosome size. The asymmetry indexes, A₁ and A₂, calculated show similar values to other species of Chrysolaena previously analyzed (Angulo and Dematteis 2009b, Via do Pico and Dematteis 2012b). According to Stebbins (1971), both species fit into the symmetry category 1B, since the relation between the longest and shortest chromosome was between 2:1 - 4:1. This author suggested that classes 1B and 1C are absent in higher plants and only occur in animals (particularly in reptiles), which have karyotypes characterized by great differences in chromosome size, but with predominantly median or submedian centromeres (Stebbins, 1971). However, several subsequent works demonstrated that the class 1B is present among plants, e.g. in the genera Helianthus Linnaeus, 1753, Crotalaria Linnaeus, 1753, and Onobrychis Miller, 1754 (Gupta and Gupta 1978; Kulshreshtha and Gupta 1981; Almada et al. 2006; Hesamzadeh Hejazi and Mahdi Ziaei Nasab 2010).

Chrysolaena cristobaliana presents the longest karyotype (TKL), which is correlated with its ploidy level (2n=7x=70). Although the species differed in their chromosome number and total karyotype length, they had similar chromosomal morphology and asymmetry indices. Therefore, karyotype data do not seem to be of great use for group taxonomy, since the chromosomes are small and the karyotype differences cannot be detected rapidly by the analysis of one or a few cells of each species, but only by comparing average measurements. According to Ruas et al. (1991), Dematteis (1996, 1998), Dematteis and Fernández (1998, 2000) and Oliveira et al. (2007a, b), despite the occurrence of variation in chromosome number among species of Vernonia sensu lato, the karyotypes with conventional techniques does not discriminate well the species already studied, due to small variation in chromosome size and centromeric position.

The species of Chrysolaena exhibit abundant polyploidy. The majority of species studied so far, include diploid and polyploid populations. There are some exceptions of species with a single known cytotype, such as the diploid Chrysolaena verbascifolia (Dematteis 1997, Via do Pico and Dematteis 2012b). The review of chromosomal studies reveals that the tetraploid cytotype is the most common. From nine species in which chromosome number is known, seven had populations with tetraploid cytotypes (Dematteis 1997a, 1997b, 1998, 2002, 2009, Dematteis et al. 2007, Angulo and Dematteis 2009b, Via do Pico and Dematteis 2012a, b). Ploidy differences are not restricted to comparisons between species, but also occur frequently within species (Miller 1978, Burton and Husband 1999, Weiss et al. 2003). Available data on Chrysolaena and the results of this study show that the genus is cytologically complex, and polyploidy is an important mechanism in the differentiation and adaptation of species. The species, as well as, different populations of same species present numerous ploidy levels, even odd cytotypes. Some species would be polyploid series and would present more than one cytotype. Besides, there may be a species complex, with entities morphologically related but with different ploidy levels. In this group of species it could be occurring continuous processes of polyploidization and hybridization, which would lead to the different cytotypes observed. According to Jones (1979) and Ruas et al. (1991), the Vernonieae of the New World, in contrast to those of the Old World, show marked diversity in chromosome number and a high ratio of polyploid species. Polyploids may combine to give rise to a complex of polyploid species, which promotes morphological and ecological changes that hinder the taxonomic treatments (Galiano and Hunziker 1987, Dematteis 2002).

Moreover, the chromosome number found in Chrysolaena cirstobaliana is the second report of impair ploidy level for the genus Chrysolaena. In Chrysolaena cognata a mixed population with pentaploid (2n=5x=50) and hexaploid specimens (2n=6x=60) was found in Misiones, Argentina (Dematteis 2002). However, the population of Chrysolaena cristobaliana is not mixed, and all individuals analyzed showed heptaploid ploidy level (2n=7x=70). Within the Vernonieae, another case of odd polyploidy was reported in Lessingianthus macrocephalus (Less.) H. Robinson, 1988, which presented 2n=11x=176, one of the higher ploidy levels found within the Asteraceae (Angulo and Dematteis 2012). The high ploidy level with odd chromosome complement, suggests that an irregular meiosis behavior could lead to sterility. Generally, in plants of the Asteraceae family, with these features, is very common the apomictic reproduction. Apomixis is usually defined as a natural process that allows clonal reproduction through seeds, avoiding meiosis and fertilization, and resulting in offspring that are genetically identical to the maternal plant (Nogler 1984). Apomixis was recorded in 2.9 % of the genera of the Asteraceae family and is very common, e.g., in the tribe Eupatorieae (Farco et al. 2012). Considering this background, the population of Chrysolaena cristobaliana analyzed in this study probably present apomictic reproduction. However, other studies should be conducted to test this issue.