Karyomorphological features were observed in 13 samples of Aspidistra subrotata (Table 1). Of these, eight samples of five populations from Longzhou County and one sample from Nanning were tetraploid, while the other four diploid samples were from Mt. Daqing, Pingxiang County-level City, Mt. Poman, Baise City and Mt. Bavi, Hanoi.

Eight samples of five tetrapoloid populations of Aspidistra subrotata were all from Longzhou and were located not far away from each other (Map 1). Another single sample from Nanning was cultivated in Guangxi Medicine Botany Garden, which may be the clone plants of typical A. subrotata. All of them have a chromosome number of 2n = 76, with the karyotype formulated as 2n = 44 m + 8 sm + 24 st (Figure 1, 3). The karyotypes of the tetraploid samples are similar to one another. The first and largest pair was metacentric. The pair II was larger [larger than what?]and submetacentric. Chromosomes from III to VIII pair were larger [again, larger than what? Normally, these chromosome pairs would be expected to be progressively smaller] and submetacentric. The other chromosomes were smaller, of which pair IX was submetacentric and the others were metacentric (Figure 3). It was noted that the secondary constriction occurred in chromosome 12 of the sample NG (Figure 1c and 3c). The average lengths of chromosomes varied from 4.26 to 5.91 μm, while A₁ and A₂ were from 0.33 to 0.39 and 0.55 to 0.64 (Table 2).

Map 1.

Distribution of Aspidistra subrotata. Black circle represents the tetraploid population; gray circle represents the diploid population; empty triangle represents typical A. subrotata var. crassinervis; empty square represents typical A. subrotata var. angustifolia.

Figure 1.

Somatic Chromosome at mitotic metaphase in Aspidistra subrotata of 2n = 76. a Jilong population b 5th boundary marker population c Nonggang National Nature Reserve population, and the arrow shows the secondary constriction of the chromosome d 4th boundary marker population. Bar = 10 μm.

Four samples of three populations from Mt. Poman, Mt. Daqing, and Mt. Bavi of Aspidistra subrotata have a chromosome number of 2n = 38, uniformly formulated as 2n = 22 m + 4 sm + 12 st (Figures 2 and 4). The first and largest pair was metacentric. The pair II was larger and submedian centromeric. Chromosomes from III to VIII pair were larger [progressively smaller?] and subtelocentric. The other chromosomes were smaller, of which pair IX was submetacentric and the others were metacentric (Figure 4). No satellite was observed. The average lengths of chromosomes varied from 4.30 to 5.61 μm, while A₁ and A₂ were from 0.34 to 0.38 and 0.57 to 0.59 (Table 2).

Figure 2.

Somatic Chromosome at mitotic metaphase in Aspidistra subrotata of 2n = 38. a Mt. Poman population b, c Mt. Daqing population d Mt. Bavi population. Bar = 10 μm.

Figure 3.

Karyotype of Aspidistra subrotata of 2n = 76, formulated as 2n = 44 m + 8 sm + 24 st. a Jilong population b 5th boundary marker population c Nonggang National Nature Reserve population, and the empty ellipse shows the presence of the secondary constriction on the chromosome 12 d 4th boundary marker population. Bar = 10 μm.

Figure 4.

Karyotype of Aspidistra subrotata of 2n = 38, formulated as 2n = 22 m + 4 sm + 12 st. a Mt. Poman population b, c Mt. Daqing population d Mt. Bavi population. Bar = 10 μm.

Based on the observation in the field, flowers of Aspidistra subrotata were commonly found with the perigone lobes red–purple and the stigma white with more or less small red dots on the upper surface (Figure 5). However, some variation occurs in its leaf shape, from sublinear or narrowly elliptical to ovate–lanceolate, from without blotches to with paler, often white blotches, from without raised secondary veins to with raised secondary veins in either the tetraploid population of fourth boundary marker (Figure 6) or the diploid population of Mt. Daqing (Figure 7). There were tiny differences between tetraploids and diploids; the former may be distinguished from the latter by its rigid petiole as well as thick and deep green lamina as well as its ploidy level of chromosome.

Figure 5.

Flower morphology Aspidistra subrotata. a–c flowers of diploid plants from a Nonggang population b Nanning population c 4th boundary marker population d–f flowers of tetraploid plants from d, e Mt. Bavi population f Mt. Daqing population. Bar = 1 cm.

Figure 6.

Leaf morphology of tetraploid plants of Aspidistra subrotata. from 4th boundary marker population. a–c sublinear leaves with a. smooth face b blotches c blotches and raised secondary veins d–e narrowly lanceolate leaves with d smooth face e blotches f–h ovate–lanceolate leaves with f smooth face g blotches h blotches and raised secondary veins i plants with ovate–lanceolate leaves and ones with sublinear leaves grow together.

Figure 7.

Leaf morphology of diploid plants of Aspidistra subrotata from Mt. Daqing population. a–c leaves with smooth face. a. sublinear leaves b lanceolate leaves c ovate–lanceolate leaves d, e ovate–lanceolate leaves with inconspicuously raised secondary veins f lanceolate leaves with blotches and raised secondary veins.

The karyotypes of nine samples of six populations of Aspidistra subrotata are described here as tetraploid for the first time, with a chromosome number of 2n = 76 and the karyotype formulated as 2n = 44 m + 8 sm + 24 st. Among them, eight samples of five populations are all from Longzhou and are located not far away from each other (Table 1, Map 1). The secondary constriction is occasionally observed (Figure 1c and 3c). The other tetraploid sample is from the plants cultivated in Guangxi Medicine Botany Garden and is possibly the clone plants of typical A. subrotata. Unfortunately, the type specimen of A. subrotata have not been able to be checked until now. Our results may offer a hint that the plants cultivated in Guangxi Medicine Botany Garden were collected from the above county on the basis that the tetraploids so far have not been observed in any other place, and infer that the type of A. subrotata may be tetraploid from Longzhou. Another four samples of three populations from Mt. Poman, Mt. Daqing, and Mt. Bavi were diploid, with a chromosome number of 2n = 38 and the karyotype formulated as 2n = 22 m + 4 sm + 12 st. A difference from the previous report on the diploid states of A. subrotata is that no satellite was observed in the chromosomes of IX pair as sm in this study (Wang et al. 2000, Huang et al. (1997). Besides, the pair IX was reported as st in Huang et al. (1997), while one more pair of sm chromosomes occurs in Wang et al. (2000), compared with our results. The material of diploid specimens reported previously is all from cultivated plants and information about their wild population is unknown. Here, three localities are confirmed for the first time where diploid populations are distributed.

There is a long-standing debate on the ecological success of polyploids relative to diploids. Although some studies suggest that polyploids generally have larger ranges (Stebbins and Dawe 1987, Petit and Thompson 1999, Oberprieler et al. 2012), the present studies prefer to support that the correlation between ploidy and range or ecological attributes is inconsistent and appears to be taxon-specific (Martin and Husband 2009, Godsoe et al. 2013, Harbert et al. 2014). In our studies, it seems that the diploids occupy broader geographical and environmental niche spaces than the tetraploids (Map 1), which maybe offer an interesting example with which to explore the formation and the evolutionary dynamics of a new natural polyploidy complex from the limestone area of the tropical regions.

Our studies show that leaves of Aspidistra subrotata varied in leaf shape, color pattern, and venation in either the tetraploid population of the fourth boundary marker (Figure 6) or the diploid population of Mt. Daqing (Figure 7); the same case occurs in Phu Luang WS, north-eastern Thailand (Phonsena and De Wilde 2010), but unfortunately, the chromosome number of Thai material is unknown. As many polyploid plants are reported to be similar to their diploid parents and hence morphologically cryptic (Tate et al. 2005, Reis et al. 2014, Azizi et al. 2016), there is also tiny difference between diploids and tetraploids of A. subrotata. The tetraploid plants may be distinguished from the diploid plants by their rigid petioles as well as thick and deep green lamina. This type of leaf also occurs in the other three polyploid species in the genus Aspidistra, i.e. A. xilinensis, A. cruciformis and A. sutepensis, which seemly supports the notion that the polyploidy can exploit newly available habitats (Brochmann et al. 2004).

Although the leaf-shape of Aspidistra subrotata varies quantitatively between and within diploid or tetraploid population(s), no obvious discontinuity has been observed. It seems unreasonable to divide it into three varieties on the basis of leaf-shape. According to the independent distribution and external morphology in relation to the ploidal levels of chromosome of A. subrotata, two subspecies may be recognized; however, the taxonomic treatment of A. subrotata has not been properly dealt with until the types of all three varieties are checked and confirmed, with additional samples, geographical locations of collections, and molecular data analysis together. We hope this study will be helpful not only to better understand the origin and evolution of the species and the genus but also shed some light on the formation and the evolutionary dynamics of a new natural polyploidy complex in the limestone areas of the tropical regions.