Clonality and ploidy levels are positively associated with plant invasiveness. However, there is still no consensus on whether polyploidization can promote the invasion of alien plants by enhancing clonality. Our recent long-term community succession study found that the more vigorous clone of introduced polyploid Solidago canadensis succeeded into mono-dominant community, which seems to be a positive correlationship between polyploidization and clonal reproduction. However, how polyploidization improves the clonal reproduction of S. canadensis remains unknown. Here, we compared clonal growth ability among diploids and polyploids of S. canadensis from native and introduced ranges in a common garden. Results showed that the rhizomes of S. canadensis originated from axillary buds of dense nodes at the basal stem of seedling and then produced into clonal ramets. Diploids had denser nodes and more buds, developed more rhizomes per unit mass and produced more clonal propagules at the early growth stage compared with polyploids. However, the number of juvenile and secondary rhizomes, as well as the diameter and length of rhizomes in polyploid populations was significant higher than those of diploids, and those clonal traits in introduced polyploids were significant higher than in native polyploids. Moreover, a phalanx growth form was observed in native and introduced diploid populations, which allocated about 3% and 5% of the total biomass to rhizomes, respectively, resulting in short and weak rhizomes. However, native and introduced polyploids allocated about 35% and 40%, respectively, of the total biomass to rhizomes, resulting in long and strong rhizomes, which were guerrilla growth forms. This study firstly shows that polyploidization enhanced the effective clonal reproduction of S. canadensis through pre-adaptation and rapid post-adaptation evolution, and consequently contributed to its successful invasion.

Diploid (2n = 2x = 18), tetraploid (2n = 4x = 36), and hexaploid (2n = 6x = 54) seeds of S. canadensis were collected from the native region (US and Canada) and invasive region (Europe and China). The detailed descriptions of material collection and identification had been presented by Cheng et al. (2021). These seeds were stored in refrigerator at 4℃, before we used these seeds, they had been stored in the refrigerator for about five ~ eight years in the Weed Research Laboratory, Nanjing Agricultural University, China. The ‘NA2x’, ‘NA4x’, ‘NA6x’, ‘IN2x’, ‘IN4x’ and ‘IN6x’ were used to represent the native diploid, native tetraploid, native hexaploid, introduced diploid, introduced tetraploid, and introduced hexaploid, respectively. Each geo-cytotype was comprised of 4 populations with 20 individual plants (see detailed information of different populations in supplementary material Appendix S1: Table S1).

Morphogenesis of rhizome of S. canadensis

In order to determine the origin of rhizome of S. canadensis, seeds of S. canadensis of different geo-cytotypes (NA2x, NA4x, NA6x, IN2x, IN4x, and IN6x) were sown in pots (11-cm-wide, 8-cm-deep, 1 L) with a 1:1 mixture of sterilized compost and vermiculite. From cotyledon stage to the stage of rhizome formation, 20 seedlings (4 populations × 5 individuals) from each geo-cytotypes population were chosen at each stage that two new leaves outgrew, respectively. Plants were pulled out of the pots and washed the soil carefully to ensure that the roots and rhizomes are intact and undamaged. The originations of rhizomes were observed with a stereomicroscope (SZX7, OLYMPUS, Japan). The selected typical seedlings were recorded with a camera.

Difference of basal nodes and the number of axillary buds of different geo-cytotypes of S. canadensis

After determining that the rhizome of S. canadensis originated from axillary buds, we counted the number of axillary buds at the basal stem of S. canadensis to study the potential of rhizome production in different geo-cytotype populations. The basal stem here refers to the stem starting between nodes 1 and 10 (node 1 = cotyledonary node).

Comparison of rhizomes and clonal ramets of different geo-cytotypes of S. canadensis in pot common garden experiment and field test

Seeds of six geo-cytotypes (4 populations with 20 individuals were selected from each geo-cytotype) were sowed in small germination pots respectively, which were filled with the same soil substrate as described above and grown for one month. Then, uniform plants in each population were selected and transplanted into 1 L pots (10-cm-wide, 9-cm-deep) with the same soil substrate. Each of the pots contained one individual and was placed in a greenhouse. Pots were randomly arranged on tables and changed their positions weekly. They were watered every two ~ four days and manually removed the other weeds regularly. Three months later, they were transplanted into 6.6 L pots (24-cm-wide, 21-cm-deep) with a 1:2 mixture of sterilized compost and soil and each of the plots contained 1 seedling. The plots were placed in the outdoor of Pailou Teaching and Research Station of Nanjing Agricultural University (32°02′N, 118°37′E), Nanjing, China. This region is characterized as being a subtropical monsoon climate, with an average annual rainfall of 1090mm and the mean annual temperature is 15°C (lowest in Feb at 2.7°C and highest in July at 28.1°C). They were watered, weeded, and sprayed with pesticides regularly. After these plans grew for six months in the outdoor (October 2018), the number of clonal ramets and rhizomes developed from axillary buds at the basal stem of each genet was counted by removing the topsoil around the basal stem of genet to expose stem above the cotyledon node without damaging the roots, and the height (cm) and diameter (mm) of each ortet were measured with a straight-edge and a vernier caliper, respectively. After these measurements, the ramets were cut off in winter except the basal stem (5~6 cm) and belowground parts and some soils were added to these pots. In the next two years, we only watered, weeded, sprayed with pesticides regularly and cut off the aboveground part in winter. Three years later, in September 2021, the number, diameter (mm), height/length (cm), and biomass (g) of ramets and rhizomes of different geo-cytotypes were measured by digging out these plants from pots. Because the rhizomes are produced not only by the axillary buds on the basal stem of mother ramet and daughter ramet but also by the axillary buds on the rhizome, we defined rhizome that produced from mother ramet as ‘old rhizomes’, rhizome that produced from daughter ramet as ‘juvenile ramet’ and rhizome that produced from old rhizome as ‘secondary rhizome’ in this study.

To determine what is the difference of clonal reproduction among the six geo-cytotypes in a field environment, we used the field test of Cheng et al. (2020) to analyze the clonality of different cytotypes of S. canadensis. This study was also conducted at the Pailou Teaching and Research Station of Nanjing Agricultural University from July 2011 to November 2015. The field experiment was established without artificial watering and fertilization during the whole period of experiment. Plots were randomly designed with four replicates for the six geo-cytotypes. S. canadensis were planted in a plot (3 × 2 m) including 12 plants and named the plot with the corresponding geo-cytotype (see supplementary material Appendix S1: Figure S1). Local weeds were also transplanted in every plot in order to create a scene where S. canadensis compete with local weeds (detailed information had been shown in paper (Cheng et al. 2020)). In November 2011 to November 2015, before the S. canadensis seeds matured, the aboveground of S. canadensis and four local weeds were cut off every year during the winter, maintaining the growth of belowground root and rhizome. The spacer length between mother ramet and daughter ramet (cm) of each geo-cytotype with 48 mother ramets (4 plots × 12 individuals), in the first year were measured. From 2012 to 2015, the total number of ramets in each plot was counted in the spring and the autumn respectively. At the same time, the average clonal ramets produced by each ortet, the total number of ramets in each plot, plant height (cm) and fresh biomass (g) of aboveground of different geo-cytotypes in the autumn of the first year were measured.