The phenotype of a plant individual can be influenced by the environmental condition experienced by its parent, a phenomenon called parental effects or trans-generational effects (Agrawal, 2002; Mousseau et al., 2009; González et al., 2016; Waterman and Sultan, 2021). Parental effects can influence offspring not only via seeds produced through sexual reproduction (Bossdorf et al., 2010; Mondoni et al., 2014; Baker et al., 2019; Veselá et al., 2021) but also via ramets (asexual individuals) produced through clonal growth or vegetative reproduction (Latzel and Klimešov, 2010; Dong et al., 2018a; González et al., 2018; Portela et al., 2020). Both sexual and clonal parental effects have important ecological and evolutionary implications as they can potentially influence plant fitness, intraspecific and interspecific interactions, population dynamics, and community structure (Badyaev and Uller, 2009; Droste et al., 2010; Geng et al., 2016; Herman and Sultan, 2016; Pérez-Ramos et al., 2019).

Increasing evidence shows that parental effects may be adaptive, enhancing the fitness of the offspring when established in an environment similar to their parents (Herman et al., 2012; Rasmann et al., 2012; Latzel et al., 2014; Dong et al., 2017; Baker et al., 2018). For instance, when offspring of Polygonum persicaria grow in shade, offspring produced by shaded parents perform better than offspring produced by parents under sunlight (Baker et al., 2018). Adaptive parental effects have been reported in several plant species in response to a variety of biotic and abiotic factors (Lacey and Herr, 2000; Whittle et al., 2009; Dong et al., 2017; Waterman and Sultan, 2021). As a result, parental effects are recognized as an important source of phenotypic variation that may have an essential role in local adaptations (Herman and Sultan, 2011; Holeski et al., 2012; Dong et al., 2018a).

Many studies have shown that parental effects caused by differences in the quality of resources provisioned to offspring may be more likely to be adaptive because they can persist through the life cycle (Roach and Wulff, 1987; Herman and Sultan, 2011; Germain et al., 2013; Zas et al., 2013). For example, larger seedlings of Pinus pinaster can come from larger seeds that are produced in favorable parental environments (Zas et al., 2013). Compared to sexual propagules (e.g., seeds), vegetative propagules (e.g., ramets) of clonal plants are larger in size and mass, and thus their potential for resource provisioning may be relatively high (Dong et al., 2019). For instance, González et al. (2017) found that parent ramets of Trifolium repens grown in better conditions produced larger/heavier vegetative propagules that enabled offspring to grow better. Dong et al. (2018a, 2019) and Portela et al. (2020) showed that clonal fragments of Alternanthera philoxeroides produced in favorable parental environments benefited the subsequent growth of clonal offspring. These studies suggest that resource provisioning is one of the most important mechanisms underlying clonal parental effects (González et al., 2017; Dong et al., 2018a, 2019; Portela et al., 2020).

A high proportion of aquatic species are clonal and capable of rapid vegetative reproduction (Sosnová et al., 2010; Wang et al., 2016; Zhang et al., 2019; Adomako et al., 2021). A clone often consists of a large network of interconnected ramets belonging to different vegetative generations (Dong, 2011). However, previous studies testing clonal parental effects on offspring performance have involved clonal offspring of only a single vegetative generation (Dong et al., 2017, 2019; González et al., 2017; Portela et al., 2020), and have not considered the potential differences among offspring ramets of different vegetative generations. As the size and biomass of offspring ramets commonly become smaller with increasing vegetative generation (Wang et al., 2014), their ability of resource provisioning may become weaker. Therefore, the magnitude of clonal parental effects on offspring performance may differ when offspring of different vegetative generations are considered.

To test how clonal parental effects influence the offspring growth of different vegetative generations, with a focus on provisioning as a possible mechanism, we conducted a greenhouse experiment on an aquatic clonal plant Pistia stratiotes. We grew a single ramet of P. stratiotes under low or high nutrients, and its offspring ramets of three different generations, i.e., primary, secondary, and tertiary offspring ramets were also subjected to low or high nutrients. Specifically, we tested three hypotheses: (1) clonal parental effects would impact offspring growth of all three vegetative generations of P. stratiotes; (2) for all three vegetative generations, offspring produced by the parent ramet under high nutrients would perform better than offspring produced by the parent ramet under low nutrients, because providing high nutrients to the parent ramet may allow them to produce high-quality clonal offspring; and (3) the magnitude of clonal parental effects on offspring performance would become smaller with increasing the vegetative generation, i.e., parental effects would be the highest on the growth of the primary ramets, the lowest on that of the tertiary ramets, and in between on that of the secondary ramets.

As expected, both in the first and the second phase, current high nutrients increased the production of biomass and the number of new offspring ramets of P. stratiotes, agreeing with many previous findings on other aquatic plants (Zhao et al., 2006; Jampeetong and Brix, 2009; Zhang et al., 2020). Besides the direct positive effect of high nutrients on the growth of offspring ramets, the benefit was also from physiological integration from their parent ramets growing under the favorable conditions of high nutrient availability. Physiological integration can allow acropetal transport of resources from well-established ramets to developing offspring. Benefits from physiological integration have been repetitively reported by various previous studies (e.g., Slade and Hutchings, 1987; Saitoh et al., 2002; Roiloa and Retuerto, 2006; Dong et al., 2015; Elgersma et al., 2015; Portela et al., 2021; Wang et al., 2021). In our study, it is logical to anticipate that the benefit gained by offspring connected to parents with high nutrient availability was due to the support received from their parents growing under favorable conditions.

High nutrients increased ramet number of different offspring generations but had different effects on ramet weight (i.e., biomass per ramet). Previous studies found that plants produced lighter and likely smaller ramets in response to high-density environments and produced heavier and, likely, also bigger ramets in response to low-density environments (Wang et al., 2014; Adomako et al., 2021). Therefore, the relationship between ramet biomass and ramet number was caused by the different responses to plant density (Wang et al., 2014). In our study, high nutrients increased the biomass of the F1 offspring ramets but had little effect on the biomass of the F2 and the F3 ramets. This is mainly because plants did not occupy the whole surface of containers when producing the F1 ramets, so high nutrients increased both ramet biomass and number under the low-density environment. However, the high nutrients increased the number of the F2 and the F3 ramets, resulting in increased plant density and intensity of intraspecific competition, which may limit the growth of the F2 and the F3 ramets. Thus, ramet biomass of the F2 and the F3 ramets was not significantly greater under high nutrients than in low nutrients.

The interaction of parental nutrient level and offspring nutrient level significantly affected the growth of different offspring generations. Compared to low nutrients experienced by the parent ramet, parental high nutrients inhibited subsequent growth of offspring in high nutrients for all three offspring generations but increased growth of the F3 offspring ramets in low nutrients. These results support the first hypothesis that parental effects could impact offspring growth of different vegetative generations of clonal plants. Similarly, previous studies have shown that parental environment effects can persist across vegetative generations in clonal plants (e.g., González et al., 2017; Dong et al., 2018b; Portela et al., 2020; Zhang et al., 2021). Our results suggest that paternal effects are fast in clonal plants as the vegetative reproduction is rapid.

The selected offspring ramets from the first phase that grew under high nutrients were relatively bigger and had nearly two times greater initial mass than did offspring taken from the first phase grew under low nutrients. However, such a sizeable advantage of ramets inhibited subsequent growth of offspring in high nutrients for all three vegetative generations. Contrary to the second prediction, the benefit obtained by offspring generations (F1, F2, and F3) growing in high nutrients was significantly higher when their parent ramets grew in low nutrients. A plausible explanation for this unexpected result could be based on biomass partitioning responses. As predicted by the optimal partitioning theory (Bloom et al., 1985; Hilbert, 1990; Gleeson and Tilman, 1992), an increase in the proportional biomass allocated to roots would be expected in parents growing under low nutrient conditions. Furthermore, trans-generational plasticity predicts that the plastic response experienced by the parental generation could be transferred to their offspring generations, as a mechanism to facilitate offspring establishment (Latzel et al., 2014; Dong et al., 2017, 2018a). Therefore, it is reasonable to predict that parent ramets of P. stratiotes growing under low nutrients would have increased the root production, and this plastic response would have been transmitted to the subsequent offspring generations (F1, F2, and F3). In this situation, offspring ramets having a higher proportion of roots and growing under high nutrients would be able to make better use of them and achieve maximum nutrient acquisition efficiency, which translates into the highest growth. Unfortunately, leaf, stem, and root of P. stratiotes were harvested together, and biomass allocation ratios were not available in our study. Therefore, to truly test this plausible explanation, studies should consider not only parental effects in terms of biomass production but also in terms of biomass allocation ratio.

Our finding was opposite to previous studies showing that the high-quality offspring produced in favorable parental environments benefited subsequent growth of offspring (Zas et al., 2013; Dong et al., 2018a, 2019; Portela et al., 2020), as stated by the “silver-spoon” effect, where parent plant growing in favorable conditions can provide more resources to their offspring (Roach and Wulff, 1987). In addition, our results also do not fit with another potential benefit of parental effects, which states that offspring generations could gain an advantage when establishing in the same or similar conditions to those experienced by their parents (Dong et al., 2017, 2018a).

The “silver-spoon” effect was detected when the F3 offspring ramets of P. stratiotes were grown under low nutrients but was absent when the F1 and F2 offspring ramets were grown under low nutrients. Also, the magnitude of parental nutrient effects on the subsequent growth of the F1, F2, and F3 offspring ramets in high nutrients were similar (Figures 3–5). These results do not support the third hypothesis that the magnitude of parental effects would decrease with increasing vegetative offspring generations. This is very likely because the difference in the size advantage of offspring of different generations cannot transmit to their growth benefits, as discussed above.

We conclude that nutrient-induced clonal parental effects can influence offspring growth of different vegetative generations, suggesting that clonal parental effects can transmit fast. However, heavier and, likely, bigger ramets produced under high nutrients in the parental generation could not increase the subsequent growth of offspring. Thus, parental provisioning in favorable conditions may not always increase the growth of their offspring, partly depending on root allocation but not ramet biomass.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

L-MZ and F-HY designed the experiment. L-MZ, J-FZ, W-HY, C-YQ, and D-HW performed the experiment and collected data. L-MZ and J-FZ analyzed the data. L-MZ wrote the first draft of the manuscript. F-HY and SR contributed substantially to the revisions. All authors contributed to the article and approved the submitted version.

The research was supported by the National Natural Science Foundation of China (Grants 32071527 and 32101261).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.