Selective forces on the maintenance of outcrossing in an almost exclusively cleistogamous violet species

Abstract

Premise Cleistogamous species constitute interesting study systems to resolve the longstanding question of how outcrossing is maintained given that seed production is ensured through selfing. In this work, we investigate the selective forces that allow the persistence of producing self‐pollinated cleistogamous (CL) and chasmogamous (CH) flowers in Viola jaubertiana Marès & Vigin. Methods We monitored three populations at different elevation for two years, and studied the flowering phenology and the relative contribution of each flower morph to parental fitness. We tested whether allocation to CH and CL flowers differed across populations and if it covaried with herbivory and water stress conditions. We also performed hand‐pollination and bagging experiments in CH flowers to estimate inbreeding depression and heterosis. Results The CH flowers open in winter under unfavorable conditions for pollination, show high pollen limitation and no‐delayed selfing, and thus produce a low amount of seeds. Conversely, CL flowers appear in early spring, are physiologically cheaper to produce (i.e., dry weight is 3.4 times lower than that of CH flowers), and yield approximately 100 times more seeds than CH flowers. The CH flowers were favored under water stress and low herbivory. Crosses between populations showed up to 25% greater fitness than those within populations. Conclusions Despite the great pollen limitation in CH flowers, we suggest that the interaction among different environmental determinants and heterosis are probably sufficient forces to maintain chasmogamy in this long‐lived species, reducing deleterious fixed mutations in the selfed lines.

Climate change is associated with increased allocation to potential outcrossing in a common mixed mating species

“…Production of CH and CL flowers is largely sequential, with the main period of CH flowering occurring in the spring, and the main period of CL flowering in the summer and fall. However, similar to other violet species (Cortés‐Palomec and Ballard, 2006; Wang, 2013; Seguí et al, 2021), the transition between these two flower types is gradual, with overlapping production of CH and CL flowers occurring during this transition period. Occasional fall blooms of CH flowers can also occur (Austin, unpublished data), a phenomenon that occurs in other temperate spring ephemerals (e.g., Hopkins, 1937).…”

“…For example, in Viola (Violaceae)—a genus in which cleistogamy commonly occurs (Culley and Klooster, 2007)—CH flowers are often produced as temperatures rise in the spring, with CL flower production continuing as canopy closure reduces light availability (Culley, 2002). Furthermore, while interspecific variability exists in species with dimorphic cleistogamy (e.g., Campbell et al, 2016; Seguí et al, 2021), obligately selfing CL flowers are often produced in environments with low water availability and high temperature, suggesting that drought conditions are climatically stressful for many dimorphic cleistogamic species (e.g., Brown, 1952; Jones et al, 2013; Miranda and Vieira, 2016; Ansaldi et al, 2018).…”

Can heterosis and inbreeding depression explain the maintenance of outcrossing in a cleistogamous perennial?

“…This outbreeding depression was modest (–21%) for one cross, but strong (between –50 and –71%) for the other three crosses. There are no reported estimates of outbreeding depression in natural populations of a cleistogamous species, but to our knowledge only one published study has the relevant data (Seguí et al, 2021). In other species, outbreeding depression is most commonly observed in the F 2 and later generations, where negative epistatic interactions are thought to play a primary role (Lynch, 1991; Schierup and Christiansen, 1996; Edmands, 1999; Fenster and Galloway, 2000).…”

“…Weak inbreeding depression fits the expectation for small and/or highly selfing populations, where strongly deleterious alleles may have been purged from the population via some combination of selection or drift (Lande and Schemske, 1985;Glémin, 2003, but see Winn et al, 2011Waller, 2021;Toczydlowski and Waller, 2023). Outbreeding depression within populations has also been reported in the cleistogamous Viola jaubertiana (Seguí et al, 2021), and other highly selfing species (Paland and Schmid, 2003;Volis et al, 2011;Oakley and Winn, 2012;Gimond et al, 2013;Clo et al, 2021). Such outbreeding depression in within-population outcrosses could be due to a number of genetic mechanisms (Lande, 1985;Charlesworth, 1992;Schierup and Christiansen, 1996;Clo and Opedal, 2021), many of which are thought to be more likely in highly selfing species.…”

“…Despite the large number of published estimates of inbreeding depression, estimates of both inbreeding depression and heterosis in the same system are relatively uncommon (but see for example, Paland and Schmid, 2003; Weller et al, 2005; Oakley and Winn, 2012; Oakley et al, 2019; Stojanova et al, 2021). To our knowledge, such estimates are available for only one cleistogamous species (Seguí et al, 2021). Additionally, few studies in cleistogamous species estimate inbreeding depression independent of differences between flower types (reviewed in Oakley et al, 2007; but see Culley, 2000; Oakley and Winn, 2008; Munguía‐Rosas et al, 2013; Ansaldi et al, 2019).…”