To forecast how fast populations can adapt to climate change, it is essential to determine the evolutionary potential of different life-cycle stages under selection. In birds, timing of gonadal development and moult are primarily regulated by photoperiod, while laying date is highly phenotypically plastic to temperature. We tested whether geographic variation in phenology of these life-cycle events between populations of great tits (Parus major) has a genetic basis, indicating that contemporary genetic adaptation is possible. We carried out a common garden experiment in which we bred first- and second-generation pairs in captivity originating from eggs from Gotland (Sweden) and Hoge Veluwe (Netherlands), two populations that showed different temperature sensitivity of laying date in a recent meta-analysis. We recorded the phenology of egg-laying, moult and gonadal size in early spring. We found no significant differences in laying date between the populations, but they did differ in moult timing and testis size. This implies that under climate change the timing of gonadal development and moult, which are mainly regulated by photoperiod, will not respond to increased temperature but can respond by genetic adaptation in response to selection, while the opposite holds for laying date, perhaps indicating that plasticity is constraining genetic adaptation.

Setting up common garden experiment

Birds: In the spring of 2021, freshly laid, unincubated eggs were transported from eight populations to the NIOO: La Rouvière (France), Boshoek (Belgium), Wytham Woods (England), Gotland (Sweden) and four Dutch populations (Hoge Veluwe, Vlieland, Oosterhout and Liesbos). The eggs were then taken to a Dutch population (Bennekomse Bos, lat: 52,003; long: 5,708) where they were placed in foster nests for incubating and early parental care.  At 10 days post-hatching, these chicks were transported to the Netherlands Institute of Ecology and hand raised, following procedures as outlined in [1]. They were blood sampled and based on their genotypes (using five microsatellite regions Pma-TGAn33, PmaC25, PmaTAGAn71, PmaGAn27 and PmaD10 [2]) and that of the potential parents, they were assigned to a family following a standard protocol [3].

For a number of these populations there were severe problems leading to very low numbers of chicks (see Appendix 1 for more info on all eight populations). Hence, we only include offspring from Gotland (Sweden, lat: 57,063, long: 18,278) and Hoge Veluwe (Netherlands, lat: 52,041, long: 5,856) in our analysis. Note that due to the different timing of eggs laying the two populations the Gotland eggs were placed seven days later at the foster parent nests than the Hoge Veluwe eggs (7 May vs 30 April).

The following year (2022) we formed first generation (F1) breeding pairs (within populations) from these birds and kept them from January onwards in pairs in open aviaries at the Netherlands Institute of Ecology with ad libitum food (constant daily amount of food consisting of a mixture of minced beef, proteins and vitamins, sunflower seeds, fat, a mix of dried insects, a mixture of proteins, vitamins, minerals and trace elements (Ce´De´-mix), a surplus of calcium, water for drinking and bathing, nesting material and four nestboxes as nesting opportunities. The eggs produced where collected every morning and put in an egg turner (i.e. a device that gently rocks eggs during storage – a CocinaCo 154 Eggs Quail Turner Tray Container).

Eggs were, within five days of laying, taken to the Bennekomse Bos to foster parents. We put Gotland and Hoge Veluwe eggs together in foster broods (with a total clutch between 5 and 11 eggs, 9 on average) to ensure common conditions during incubation and early chick rearing. Note that as there was no difference in laying date between the populations for the F1 pairs, eggs were taken to their foster nests during the same period. At day 10 post-hatching the chicks were taken to the Netherlands Institute of Ecology, hand raised, blood sampled and assigned to a family.

In the following year (2023) 20 Gotland and 20 Hoge Veluwe second generation (F2) breeding pairs were set-up. As these F2 birds originate from eggs produced in a common garden setting, and thus any carry over effects of the location the eggs were produced are excluded, any differences between them will be genetic. In total 7 Hoge Veluwe and 8 Gotland F1 pairs and 20 Hoge Veluwe and 20 Gotland F2 pairs were used (number of pairs that produced a clutch was 7, 7, 20 and 17 respectively).

Aviaries: Breeding pairs were set up in 40 outdoor aviaries of 4 m x 2 m x 2 m (l x w x h) with on one side a mesh, allowing natural light and ambient temperatures. Despite being exposed to natural light, all aviaries are still darker than natural conditions. This causes the birds in aviaries to consistently lay later than wild birds if left without additional light. Thus, a fluorescent light tube provided additional light in the morning for all breeding pairs. In January and February lights were on from sunrise until midday (i.e. same as normal housing conditions) and from March onwards lights were on 2h and 15 min before sunrise until midday. This additional light was crucial for eggs to be laid while foster nests were still available to produce the F2 birds. The aviary building consisted of two rows of aviaries (20 West and 20 East facing) and to minimize the impact of any systematic variation in conditions between aviaries we kept the Gotland and Hoge Veluwe pairs in alternating aviaries.

Phenotyping

Laying date: Laying dates were recorded for both the Hoge Veluwe and Gotland F1 and F2 generations (see Appendix 1 for laying dates of F1 birds from other populations). Nest boxes were checked daily for nest building progress and new eggs, and the laying date was the day the first egg was laid. Eggs were replaced by plastic dummy eggs, and upon clutch completion females were allowed to incubate for four complete days after which nests and dummy eggs were removed on the fifth day. Frequently, pairs would initiate replacement clutches. Here, we only analyse the laying date of the first broods, i.e. the first clutch of the season.

Moult: Moult was scored once in both years at the end of the breeding season (F1 = 17^th of June in 2022 and F2 = 16^th of June in 2023). We inspected the right wing of each bird and gave 10 scores per bird, one for each primary feather (P1-P10), from 0 to 5 (0 old feather, 5 fully grown), reflecting how much it had grown [4]. Then, we obtained a moult score per individual by converting each of the 0-5 moult scores into an approximate proportion of feather grown [following 5], multiplied by the respective mass of that particular feather, and finally summed the values of all feathers. This resulted in a single value ranging between 0 and 1. The mass-corrected moult score serves as a proxy for moult timing because all birds were scored on the same day and feather mass increases fairly linearly throughout the season [5–7]. Thus, the larger the score, the furthest the bird is in its moult progress and consequently the earlier it started moulting.

Gonadal size: We measured the gonadal sizes of F2 birds on 22, 23, 26 & 27 Feb 2024. Birds, 19 Hoge Veluwe and 19 Gotland pairs, were kept in the same outdoor aviaries as in the F2 breeding season of 2023. The birds, alternating between pairs from Gotland and Hoge Veluwe, were put under isoflurane after which they were decapitated and the length (mm), width (mm) and fresh mass (mg) of their gonads measured. We calculated testis and ovary volume as V = 4/3.π.a².b, where a is width (mm)/2 and b is length (mm)/2.

Statistics

Laying date, moult score and gonad volume were analysed using generalized linear models using the lm function of lme4 [8] in R version 4.3.2 [9], with population (Gotland or Hoge Veluwe), generation (F1 or F2) and (for moult and gonads) sex (male or female), and their interactions, as explanatory variables. For testis the right testis was analysed (as one left testis was missing, reducing the sample size and as left and right testis volume were highly correlated (Pearson correlation: 0.96)). Following Schaper et al. [10], we use the ¹⁰log gonadal volume as the distribution of the logged values better follow a normal distribution.

References:

1.           Drent PJ, van Oers K, van Noordwijk AJ. 2003 Realized heritability of personalities in the great tit (Parus major). Proc. R. Soc. Lond. Ser. B-Biol. Sci. 270, 45–51.

2.           Saladin V, Bonfils D, Binz T, Richner H. 2003 Isolation and characterization of 16 microsatellite loci in the Great Tit Parus major. Mol. Ecol. Notes 3, 520–522.

3.           Greives TJ et al. 2015 Costs of sleeping in: circadian rhythms influence cuckoldry risk in a songbird. Funct. Ecol. 29, 1300–1307. (doi:10.1111/1365-2435.12440)

4.           Ginn HB, Melville DS. 1983 Moult in Birds. British Trust for Ornithology. See https://books.google.nl/books?id=Sd9FAAAAYAAJ.

5.           Underhill LG, Zucchini W. 1988 A model for avian primary moult. Ibis 130, 358–372. (doi:10.1111/j.1474-919X.1988.tb00993.x)

6.           Dawson A, Newton I. 2004 Use and validation of a molt score index corrected for primary feather mass. Auk 121, 372–379.

7.           Dawson A. 2003 A detailed analysis of primary feather moult in the Common Starling Sturnus vulgaris– new feather mass increases at a constant rate. Ibis 145, E69–E76. (doi:10.1046/j.1474-919X.2003.00161.x)

8.           Bates D, Machler M, Bolker BM, Walker SC. 2015 Fitting Linear Mixed-Effects Models Using lme4. J. Stat. Softw. 67, 1–48.

9.           R_Core_Team. 2021 R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing.

10.        Schaper SV, Gienapp P, Dawson A, Visser ME. 2013 Heritability of gonad size varies across season in a wild songbird. J. Evol. Biol. 26, 2739–2745. (doi:10.1111/jeb.12249)