Short periods of extreme heat during development may continue harming bumblebees long after they emerge as adults. A controlled study of Bombus terrestris found that cyclic thermal stress during pupation reduced successful adult emergence by as much as 30% and doubled subsequent adult mortality risk at the highest temperature tested.
The findings, published in Communications Biology on 25 September 2026, show that developmental heat exposure can create delayed costs that are not captured simply by counting how many insects survive an initial heat event. The results may therefore matter for attempts to understand how increasingly frequent heat extremes affect pollinators.
Heat exposure during a vulnerable life stage
Bumblebees are important pollinators in natural ecosystems and agriculture, but they are also relatively cold-adapted insects. Previous research has shown that elevated temperatures can affect colony growth, behaviour and reproduction. Less is known about whether thermal stress experienced during metamorphosis leaves lasting damage that becomes visible only after adulthood begins.
Researchers Maximilian M. Mandlinger and Christoph Kurze used a controlled in vitro system to isolate the direct effects of temperature during the pupal stage. Late-stage larvae from commercial B. terrestris colonies were reared under controlled conditions and, at pupation, individuals were randomly assigned to one of three temperature regimes.
The control group remained at a constant 34°C. Two thermal-stress groups experienced a repeating 12-hour cycle between 34°C and either 37°C or 38°C. The cycling continued until adult emergence, which typically occurred after eight to nine days for workers and nine to ten days for males. Individuals from five or six colonies contributed to each experimental treatment, allowing the analyses to account for variation between colonies.
A further group of bees developed naturally in their natal colonies and was used as an additional comparison for the adult survival experiment.
Adult emergence fell as thermal stress intensified
The probability of successfully reaching adulthood changed significantly with thermal treatment. Relative to pupae kept continuously at 34°C, the 37°C cyclic treatment produced a 15% lower adult emergence rate. The 38°C cyclic treatment produced a 30% lower emergence rate.
The overall effect of cyclic thermal stress on pupal survival was highly significant. The difference between the 37°C and 38°C treatments was also statistically significant, indicating a progressively larger developmental cost as the daytime thermal maximum increased.
Heavier pupae were more likely to emerge successfully, suggesting that developmental resources may influence an individual’s ability to tolerate stressful conditions. Temperature treatment itself did not significantly change the duration of pupation. Instead, developmental duration differed mainly by sex.
Heat exposure continued to affect bees after emergence
The most consequential result appeared after the surviving bees had already become adults. Thermal stress during pupation significantly reduced adult survival in both workers and males.
In the Cox proportional hazards analysis, adults that had experienced the cyclic 38°C treatment as pupae had twice the mortality risk of adults reared at a constant 34°C. The hazard ratio was 2.00, with a 95% confidence interval from 1.61 to 2.48 and p below 0.0001.
The 37°C cyclic treatment was associated with a smaller increase in adult mortality risk. Its hazard ratio was 1.39, with a 95% confidence interval from 1.024 to 1.89 and p = 0.035.
Across the adult survival analysis, the researchers followed 370 individuals: 113 colony-reared bees, 101 reared at 34°C, 80 exposed to the cyclic 37°C treatment and 76 exposed to the cyclic 38°C treatment.
Importantly, adult mortality risk did not significantly differ between bees reared in vitro at the 34°C control temperature and bees that developed naturally in their natal colonies. The hazard ratio for the colony-reared comparison was 0.95, with a 95% confidence interval from 0.70 to 1.30. This result supports the use of the controlled rearing system for isolating the thermal treatment effect.
Wing deformities became more common at 38°C
The researchers also examined adult morphology. Pupae exposed to the cyclic 38°C treatment developed deformed wings 3.6 times more frequently than those kept at 34°C and 4.5 times more frequently than naturally colony-reared bees. Both comparisons were statistically significant.
There was no significant overall treatment effect on antennal deformities. However, when wing and antennal deformities were considered together, adults displaying these abnormalities had substantially poorer survival. Morphological deformation was associated with a hazard ratio of 2.38 for adult mortality, with a 95% confidence interval from 1.94 to 2.92.
This makes visible developmental abnormalities potentially useful indicators of more extensive physiological damage. For a flying pollinator, malformed wings also have an obvious functional consequence because severe deformation can prevent normal flight, foraging and reproduction.
The damage was not explained by depleted energy reserves
The researchers tested several additional traits to understand why heat-exposed bees died sooner. They found no clear evidence that the deferred mortality could be explained by differences in pupal weight loss, adult emergence weight, head width or relative lipid content at the end of life.
The absence of a significant difference in relative lipid reserves suggests that the survival penalty was not simply the result of heat-exposed pupae emerging with depleted energy stores. The authors instead argue that lasting structural, physiological or genetic damage may contribute to the reduced resilience observed in adulthood.
Why delayed effects matter for pollinator research
Studies of environmental stress frequently focus on immediate mortality. These results show why that approach can underestimate biological consequences. A bee may survive a developmental heat event and emerge successfully, yet still carry damage that substantially shortens its adult life.
For social insects, shorter worker lifespans could affect colony functioning even when an initial heatwave does not produce catastrophic mortality. Workers contribute to nursing, nest maintenance, thermoregulation and foraging, while males depend on adult survival and functional flight for reproduction.
The study also identifies the pupal stage as a potentially important developmental bottleneck. Metamorphosis involves extensive tissue reorganisation, and exposure to stressful temperatures during this period may have consequences that are not fully reversible once development is complete.
Important limitations
The experiment was deliberately controlled, which strengthens causal interpretation of the temperature treatments but limits direct translation to natural colonies. The cyclic regimes alternated predictably between 34°C and fixed daytime maxima of 37°C or 38°C. Real heatwaves vary in duration, timing, humidity and intensity, while colonies can alter nest ventilation and other thermoregulatory behaviours.
The work also focused on one species, Bombus terrestris. Different bumblebee species have different thermal tolerances, nesting habits and geographic distributions, so the numerical effects should not be assumed to apply uniformly to all pollinators.
Adult bees in the longevity assay were housed individually under controlled conditions. This allowed survival to be measured consistently but did not reproduce the social environment, workload, pathogen exposure and foraging demands experienced by bees in functioning colonies.
Finally, the experiment demonstrates that developmental heat caused later survival costs under these conditions, but it does not establish which molecular or physiological mechanism produced the deferred mortality.
What the findings suggest
The study provides experimental evidence that the effects of thermal extremes can cross life stages. At the highest cyclic temperature tested, fewer pupae became adults, surviving adults faced twice the subsequent mortality risk, and wing deformities became markedly more common.
These carry-over effects could make the ecological impact of extreme heat larger than estimates based only on immediate deaths. Incorporating delayed developmental consequences into models of insect responses to climate stress may therefore improve assessments of pollinator vulnerability.
Source Information
Study: Cyclic thermal stress during pupation triggers irreversible carry-over effects in a key pollinator
Authors: Maximilian M. Mandlinger and Christoph Kurze
Journal: Communications Biology
Published: 25 September 2026
Article: Volume 9, Article 1228
DOI: 10.1038/s42003-026-10940-3
