The Galápagos Islands have a climate characterized by two distinct seasons shaped by oceanic currents and trade winds. The hot season (January–May) brings elevated sea and air temperatures, along with variable rainfall, while the cold season (July–December) features lower temperatures and persistent cloud cover, which delivers moisture primarily to the highlands. In contrast, the lowland zones remain dry throughout most of the year (Trueman and D’Ozouville 2010; Watson et al. 2010; Dueñas et al. 2021). The in situ study site is located within this dry lowland zone.

Research was conducted at two main locations, both situated at approximately 15 m a.s.l.:

Historically, a second population was recorded near Caleta Tagus, along the Bolívar Channel (0.294°S, 91.356°W; 40 m a.s.l.), in an area characterized by volcanic substrates similar to those at Playa Tortuga Negra (Naumann and Geist 2000). However, this population had not been observed in over 60 years and was considered locally extinct. In 2024, restoration efforts by the GV2050 Program successfully reintroduced individuals to this site, marking the beginning of a secondary population (Naumann and Geist 2000; Jaramillo Díaz et al. 2018, 2024) (Fig. 1).

Morphology

Among all described species of Galvezia, the subspecies leucantha stands out for its unique floral morphology. This subspecies is characterized by slender glabrescent pedicels considerably bent at the apex; a calyx composed of five ovate puberulent sepals; bilabiate waxy-white tubular corollas (10–12 mm long) covered externally with nonglandular trichomes; light brown subglobose capsules (3–4.5 mm long); and dark brown seeds with several cristate-echinate ridges (up to 0.5 mm long) (Fig. 2).

Phylogenetic studies place Galvezia as a monophyletic genus within Antirrhineae, closely related to Gambelia and Schweinfurthia (Ogutcen and Vamosi 2016) and to the monospecific genus Pseudorontium (Guzmán et al. 2017). It comprises four species adapted to arid environments from coastal Peru and Ecuador to the Galápagos Islands. Among them, G. leucantha is the only Galápagos endemic, with distinct floral and vegetative features compared to its mainland relatives (Wiggins 1968; Elisens 1989, 1992; Guzmán et al. 2017).

The floral morphology of G. leucantha subsp. leucantha is closely tied to its evolutionary history and pollination strategies. As previously documented by Guzmán et al. (2017), it has shifted from bird to insect pollination, with Leptotes parrhasioides, Nylanderia sp., and Xylocopa darwinii acting as pollinators. Although Wiggins and Porter (1971) illustrated the general morphology of this subspecies in “Flora of the Galápagos Islands”, the Lankester Composite Dissection Plate (LCDP) presented here provides the first standardized and detailed visual record.

This study confirmed key morphological differences among subspecies. For example, G. leucantha subsp. leucantha displays waxy-white corollas and glabrous to sparsely pubescent surfaces, clearly contrasting with the pigmented purple, pubescent corollas of subsp. porphyrantha and subsp. pubescens. The use of the LCDP enabled high-resolution documentation that supports taxonomic clarification and enhances morphological visualization of this critically endangered taxon. This approach has also proven effective in other Galápagos endemics (Calle-Loor and Jaramillo Díaz 2024).

Despite ongoing conservation efforts, all subspecies maintain small, declining populations, underscoring their vulnerability and increasing risk of extinction (Elisens 1992; Jaramillo Díaz and Tye 2018).

The data suggest that seed viability in G. leucantha is primarily influenced by the reproductive conditions at the time of seed production, particularly the number of reproductive individuals and pollination frequency. Seeds collected in 2016 showed low viability, which may be attributed to limited floral availability and insufficient population density to sustain effective pollination. These conditions likely reduced opportunities for cross-pollination, resulting in the production of physiologically poor-quality seeds, as reported in other rare or fragmented species (Ashman et al. 2004; Knight et al. 2005; Castilla et al. 2016).

In contrast, seeds collected in 2020 and 2023 showed improved viability, coinciding with the increase in the number of established, flowering individuals due to in situ restoration actions. Higher plant density and floral resource availability may have promoted pollinator activity and reproductive success, resulting in better-quality seeds. Similar patterns have been observed in other conservation programs where population reinforcement has enhanced seed viability (Betz et al. 2013; Stewart and Byrne 2025).

These results highlight that restoration-driven increases in population size and reproductive activity are likely the primary drivers of improved seed viability, emphasizing the importance of in situ management for enhancing reproductive success in critically endangered plant species.

The results showed no significant differences in seed viability between NaClO and distilled water treatments. However, this result should be interpreted with caution due to the low statistical power associated with the small sample size (n = 3 paired replicates). Therefore, the current data are insufficient to draw conclusions about the effect of NaClO on seed viability. Further studies with broader concentration ranges and larger sample sizes are recommended. In this case, this was not feasible due to the limited availability of seeds, as the species remains critically endangered.

Although both treatments were evaluated, the results suggest that initial seed quality shaped by pollination and fecundation conditions plays a more decisive role in viability than treatment type. Seeds produced under favorable reproductive conditions responded better, whereas those from low-density contexts exhibited high rates of non-viability regardless of treatment.

These findings have relevant implications for ex situ conservation and ecological restoration efforts targeting other species at high conservation risk. Prioritizing seed collection during years with high flowering and sufficient population coverage is recommended, along with the establishment of regular viability monitoring protocols to inform seed use and replacement in seed banks. The observed improvement in seed viability following population reinforcement suggests that restoration efforts in G. leucantha are effectively enhancing its reproductive functionality, a critical factor for long-term conservation success.

The germination trials revealed that G. leucantha germinated most successfully on filter paper. However, most germinated seedlings from this treatment failed to survive post-germination, likely due to poor anchorage and nutrient access. This outcome is consistent with challenges observed in other species with small, dust-like seeds, such as orchids. In orchid propagation, small seed size and limited reserves demand rapid establishment and external nutrient uptake (Hall et al. 1982; Deconninck and Gerakis 2021).

These findings suggest that although filter paper is valuable for assessing germination capacity, substrates such as germination mix with sand are more suitable for full seedling development. Future studies should test modifications such as nutrient supplementation (e.g., Hoagland’s solution), controlled light exposure, and even in vitro cultivation to improve seedling survival.

Additionally, NaClO pretreatment significantly enhanced germination, likely due to its combined effect as a disinfectant and scarifying agent, as documented in other small-seeded species (Bae et al. 2013). Optimizing NaClO concentration and exposure time—possibly using precise methods such as the syringe technique (Ponert et al. 2012)—could further improve outcomes.

A two-step approach is recommended: filter paper or in vitro media for early germination monitoring, followed by transfer to supportive substrates for survival and growth.

The controlled propagation experiment provided insights into the early growth dynamics of G. leucantha. Under laboratory conditions, seedlings exhibited a biphasic growth pattern, followed by accelerated elongation, likely associated with branch formation. This biphasic growth pattern is typical of stress-tolerant pioneer species that first invest in root or structural anchorage before allocating resources to aerial growth. However, by May 2024, only seven individuals remained in the laboratory, as the others had already been transplanted to field sites, indicating that seedlings reaching sufficient size were progressively incorporated into restoration efforts rather than reflecting mortality under controlled conditions.

The growth evaluation included only 23 plants, which limited the robustness of the analysis and only allowed a visual description based on Fig. 4; therefore, the results should be interpreted with caution. The graph shows that variability increased over time, possibly reflecting differences in seed vigor, microhabitat effects within nursery trays, or physiological plasticity. Final cumulative stem lengths averaged ~70 cm, but this metric may underestimate biomass accumulation; future studies should incorporate complementary indicators, such as leaf area index or shoot dry weight, for a more robust assessment of growth performance and should increase the sample size.

As a species adapted to colonize volcanic scoria and eroded lava substrates, G. leucantha exhibits traits characteristic of xerophytic pioneer plants, including tolerance to nutrient-poor and drought-prone environments. In other volcanic ecosystems, similar species have been shown to establish successfully on substrates with low organic matter and limited water retention, contributing to early successional processes by stabilizing soil and facilitating nutrient cycling (Fridriksson 1987). Given this ecological role, early seedling development of G. leucantha may be particularly sensitive to the physical and chemical properties of the germination medium. Additionally, volcanic sands, such as those from the Darwin Volcano, may supply essential minerals (Nusbaum et al. 1991). However, the specific mineral content of the sand currently used in the propagation trials remains uncharacterized, as no local laboratory performs soil analyses.

Altogether, these findings emphasize the importance of understanding how volcanic sands and other low-nutrient substrates affect the early growth of pioneer species. Optimizing these abiotic conditions, alongside controlled nursery practices, will be critical for enhancing seedling performance and success in restoration efforts.

The phenological cycle of G. leucantha shows a seasonal flowering peak between October and January, which likely corresponds with climatic cues and moisture availability in the dry lowlands of Isabela. This pattern is consistent with phenological behaviors in other endangered island flora. For instance, species in the Pitcairn and Bonin Islands exhibit short, climate-sensitive reproductive windows, often disrupted by habitat loss or invasive species (Waldren et al. 1995; Tetsuo et al. 2008). Island endemics generally show high sensitivity to environmental fluctuations, making phenology a key indicator of ecological stability (Caujapé-Castells et al. 2010). However, it is also possible that ephemeral, less abundant flowering events may occur but are not detected through the sampling scheme. Considering this, and the logistical and cost constraints of monitoring, increased sampling effort, i.e., more frequent sampling dates, over a few seasons is recommended to determine whether additional short phenological peaks exist. Long-term monitoring of flowering trends in G. leucantha is therefore essential to improve knowledge of the species and to detect potential climatic shifts, reproductive limitations, or pollination failures that could compromise regeneration. This knowledge is also critical for timing seed collection and restoration actions in both in situ and ex situ contexts.

The results indicate that, for G. leucantha, a clear size threshold for successful establishment was identified, significantly increasing survival during the early months following outplanting. This threshold likely reflects a minimum structural size necessary to reduce transplant shock, withstand abiotic stress, and possible resist herbivory in the wild. Taller individuals may benefit from more developed root systems, greater resource storage capacity, and increased resistance to desiccation or mechanical damage, factors commonly linked to transplant success in arid island ecosystems.

Similar patterns have been documented in other endemic or rare species. For instance, in a restoration program of Scalesia affinis in the Galápagos, survival rates improved significantly when water-saving technologies were used, but success also depended on the initial robustness of transplants (Plunkett et al. 2023). In the critically endangered Apium bermejoi from Menorca, individuals transplanted at advanced developmental stages showed over 50% survival after 3 months, with successful fruiting in more than 80% of the survivors (Rita and Cursach 2013).

Size-dependent success has also been observed in other restoration systems. For example, a study on Eunicella cavolini, an endemic coral species, found that fragments larger than 10 cm had up to 90% survival, while smaller ones suffered much higher mortality (Topçu et al. 2023).

These examples suggest that plant or fragment size is a widely applicable predictor of post-transplant performance, especially in restoration programs that reintroduce individuals into harsh or degraded environments. In the case of G. leucantha, the findings support the implementation of a minimum height threshold (~60 cm) for outplanting, ensuring that individuals have reached a stage of structural maturity conducive to survival. Establishing such developmental thresholds can improve the efficiency of conservation interventions by reducing early mortality and maximizing resource investment. Additionally, given the short monitoring period following this cohort outplanting, it was not possible to detect potential competition among G. leucantha plants. Although the literature recommends larger areas (≥ 50 m²) for restoration techniques using plant clusters or nuclei and greater distance between them (≥ 50 m) (Zahawi et al. 2013; Corbin et al. 2016; Holl et al. 2020), the plot size was logistically constrained because of the terrain, i.e., slope, and the cost of fencing material. However, no other species were abundant in the plots, which suggests that plot size may not have a strong effect on negative interactions. On the other hand, plants started flowering shortly after repatriation, which could facilitate pollination by increasing detectability by potential pollinators and, subsequently, increase the likelihood of seed release on site.

Finally, although the evaluation period in this study was short, i.e., 2.5 months, the results already indicated that waterboxx was the most effective treatment for maintaining initial survival (100%), whereas biochar performed the worst (~16%). This may inform recommendations for future repatriation efforts. Further assessment of all treatments after the major dry season is necessary to determine the overall effects of these tools.