Temperate zone ants overwinter using a variety of mechanisms. The genus Formica overwinters entirely as adults. In at least one species it has been demonstrated that winter workers are more corpulent than summer foragers, storing resources in their bodies and mobilizing them for early brood production in spring. Here we examine overwintering by the common western thatch ant, F. obscuripes. Excavation of a winter nest revealed only workers, distributed in multiple chambers in a roughly spherical region from 0.5 to 1.05 m deep. Worker size, as measured by head width, was weakly bimodal, with fewer workers in the small vs. large size class. We measured dry weights of workers from the winter nest and workers collected the previous summer from the surfaces of multiple nests in the vicinity, including our excavated nest. Controlling for size, there was no evidence of bimodality in winter worker weight, and winter workers were 59.7% heavier than summer foragers. These results suggest that F. obscuripes workers are at their maximum corpulence going into their first winter, expend their stored fat during spring, and mostly die before overwintering a second time. It remains uncertain whether workers can regain corpulence.

Overwintering, seasonal adaptation, worker corpulence

Insects encounter annual variation in abiotic conditions that influence important life events both physically and physiologically (Chown and Nicolson 2004). In the case of insects that live in the temperate zone, one of these significant abiotic conditions is seasonal changes and the need to successfully overwinter by retreating from the cold to avoid cold exposure (Sinclair 2014; Sunday et al. 2014). Winter months for insects have proven to be a time of food shortage (Leather et al. 1995) and high mortality for some species (Leather 1993). One physiological mechanism used to overcome these challenges while overwintering is the storage and use of lipids (Lease and Wolf 2010).

Ants (family Formicidae) have long-lived adults that in the temperate zone require mechanisms to overwinter (Kipyatkov 2001). Temperate zone ants, like many other insects, overwinter below ground to avoid low winter temperatures (e.g., Kondoh 1968a; Włodarczyk 2020) . It is an effective strategy, but this can also delay spring emergence (Włodarczyk 2020).

Another mechanism that can aid overwintering ants is lipid storage. Seasonal lipid fluctuation has been studied in Formica japonica, Solenopsis invicta, Pogonomyrmex barbatus, and P. badius (Kondoh 1968b; Tschinkel 1993, 1998; Roeder et al. 2023). Pogonomyrmex badius has a large pulse of early-spring sexuals that cannot be produced solely from current foraging intake. The metabolic and labor resources needed for sexual production are stored in the bodies of the young workers, whose fat content reaches the annual minimum after producing sexuals. After these sexuals have flown, the colony once again switches to producing workers and storing excess foraging intake as worker fat for over-wintering, and for producing the next year’s sexuals (Tschinkel 1998). Pogonomyrmex barbatus was shown to have patterns of dry weight similar to those of P. badius, with a low in the summer and high in the winter months (Roeder et al. 2023). Formica japonica has heavier workers in the winter and a bimodality of corpulent and lean workers (Kondoh 1968b). Formica japonica workers emerge as adults with an intermediate corpulence level before gaining more fat prior to overwintering (Kondoh 1969).

In the genus Formica, egg production is not continuous throughout the year. Egg production stops in mid to late summer so that when winter arrives, they have no larvae, pupae, or eggs to care for, and colonies will only contain worker ants and one or more queens (Eidmann 1943; Hasegawa and Yamaguchi 1994). They overwinter underground and then immediately in spring, following snow melt, eggs are produced once again. Workers begin to rear a cohort of sexual brood before any significant foraging commences (Lopatina 2018). During spring there is presumably a transition from use of stored food resources to new food obtained from foraging, with a premium on acquiring food as early as possible. For example, in Formica polyctena (wood ants), workers not only fight for territory in the spring but also capture and kill opposing colony workers that they feed to the queen and sexual larvae, accelerating the production of sexuals (Mabelis 1984).

Formica obscuripes , the western thatch ant, is a successful ant species present in a variety of habitats of the western United States, at elevations from 1500–2500 m (Wheeler 1913; Cole 1932). They use twigs and dry plant material to create dome-shaped thatch mounds. Workers are very active on the mound surface from early spring to late summer, but then disappear below ground during winter. Winter conditions of F. obscuripes’ nests have not been investigated. We investigated the natural history of overwintering F. obscuripes by excavating a nest in winter and examining the distribution and abundance of workers in chambers. We obtained head widths and dry weights of workers to examine whether workers showed the bimodality observed in F. japonica by Kondoh (1968b). We also obtained sizes and weights of workers on the surfaces of multiple mounds during the summer prior to excavation. We predicted that winter workers would be heavier than the summer foragers due to the additional stored fat for overwintering.

Across the three days of the winter nest excavation there was variation in surface ground temperature from –1 to –13 °C. The temperature of the chambers varied from –2 to 5 °C.

During excavation we encountered 27 chambers. The chambers contained workers only, with no brood and no evidence of any stored food. Three individual workers were encountered in a tunnel at 30 cm depth; these were excluded from any further analysis. The first chamber encountered was 45 cm deep. Chambers varied greatly in shape and size. Some chambers were pancake-shaped, some were more cylindrical, while others were irregular and appeared opportunistically formed along rocks or roots. Chambers were found at 105 cm depth, but tunnels continued deeper, suggesting more of the colony remained at greater depth. No queen was found, suggesting the queen might have remained in a deeper chamber. In a previous excavation of a different nest we found a dealate queen in the deepest chamber of the nest, at 150 cm depth. No larvae or pupae were found. The chambers were scattered in a general spherical volume. In total, 4,417 workers were collected. The least populated chamber contained 13 workers and the most populated chamber contained 579 workers. The average chamber population was 170 ants. The distribution of workers with respect to depth shows the greatest density at an intermediate depth, 70–80 cm (Fig. 2).

The over-wintering colony workers had a head width that trended towards bimodality but the dip test was not significant (p = 0.0525) (Fig. 3). There was no evidence of spatial segregation of workers by HW or log(DW) in the colony. There was no relationship of worker size or dry weight to depth (HW: r² = 0.0004, p = 0.64; log(DW): r² = 0.001, p = 0.43). There was also no relationship of worker dry weight to chamber (ANOVA log(DW): F(25, 486) = 1.521, p = 0.05213). There was however a slight relationship between worker size to chamber (ANOVA HW: F(25, 486) = 1.642, p = 0.02692, r² = 0.03). This relationship does not appear strong though as the lowest median size of a chamber was 1.595 mm with the largest being 1.929 mm. The residuals of the HW × log(DW) regression showed no signs of bimodality and had a roughly normal distribution.

Workers in the over-wintering colony were heavier than the summer-collected workers (Fig. 4) (ANCOVA: F(1, 728) = 1407, p < 0.001). The average dry weight of the summer and winter workers was 3.3 mg and 5.27 mg, respectively. The winter workers were 59.7% heavier than the summer foragers.

Figure 2. 

Distribution of Formica obscuripes workers with respect to depth in an over-wintering colony. The distribution may be truncated, with an unexcavated portion of the colony below 110 cm.

Figure 3. 

Distribution of Formica obscuripes worker head width in an over-wintering colony. Workers show a trend toward bimodality.

Figure 4. 

Relationship of dry weight to head width of Formica obscuripes workers, comparing summer foragers to workers in an over-wintering colony.

In our winter excavation no external food sources, eggs, larvae, or pupae were found in the chambers. A similar lack of stored food or brood was found in an exploratory excavation of another nest in 2020. This confirms the observation in other Formica species that they over-winter without brood or any externally stored food resources. Based on our excavated colony, F. obscuripes appears to share many similarities with F. japonica. Formica obscuripes and F. japonica had a similar distribution of workers in the nest, with a greater proportion at intermediate depths (Kondoh 1968a). Although not as obvious as the HW bimodality of F. japonica, we also saw trends of HW bimodality in our F. obscuripes nest.

We found that the dry weight of winter workers was much greater than the dry weight of summer foragers. Several mechanisms could explain this. If worker corpulence can fluctuate, increasing or decreasing rapidly, the difference could be due to temperature dependence of metabolism (Shik et al. 2019). Summer foragers may have less lipid content as their social roles require more exposure to higher temperatures and more energy consumption as they forage outside of their thatch mounds. Temperature has been shown to be a possible factor in ant lipid content fluctuation for Pogonomyrmex barbatus (Roeder et al. 2023).

Alternatively, new workers may eclose with maximum corpulence, expend it in the spring and summer, and then never regain it. In the typical progression of temporal polyethism, workers may emerge in their first summer, engage entirely in brood care or nest maintenance until they overwinter, divest their lipid stores in spring, and then become external foragers. Our summer workers from nest surfaces were probably a biased sample of the entire nest population, consisting largely of these workers near the end of their lives. Kondoh (1969) observed two separate classes of summer workers in F. japonica, one consisting of newly emerged workers that stayed in the nest and were more corpulent, the other consisting of foragers which were leaner. We did not observe any bimodality of corpulence in the winter nest, which suggests that few of the foraging workers survive to overwinter a second time. Longevity of foraging workers is known to be very short (Porter and Jorgensen 1981; Schmid-Hempel and Schmid-Hempel 1984; Gordon and Hölldobler 1987; Conway 1996). If there are older foragers in the overwintering population, it is likely that they represent a very small fraction of the overall population. It remains unknown whether lean workers can ever regain corpulence.

A better understanding of how corpulence changes and functions through the seasons in ants is important in this time of rapid climate change and warming. Corpulence dynamics may be a major contributor to ant colony phenology and mortality as summers become warmer and winters shorter.

We would like to thank all those who provided feedback and reviewed this manuscript. We want to thank the University of Utah’s Office of Undergraduate Research for their Undergraduate Research Opportunity Program for funding this work.