The fecundity-advantage-hypothesis (FAH) explains larger female size relative to male size as a correlated response to fecundity selection. We explored FAH by investigating geographic variation in female reproductive output and its relation to sexual size dimorphism (SSD) in Lacerta agilis, an oviparous lizard occupying a major part of temperate Eurasia. We analysed how sex-specific body size and SSD are associated with two putative indicators of fecundity selection intensity (clutch size and the slope of the clutch size-female size relationship), and with two climatic variables throughout the species range and across two widespread evolutionary lineages. Variation within the lineages provides no support for FAH. In contrast, the divergence between the lineages is in line with FAH: the lineage with consistently female-biased SSD (L. a. agilis) exhibits higher clutch size and steeper fecundity slope than the lineage with an inconsistent and variable SSD (L. a. exigua). L. a. agilis shows lower offspring size (egg mass, hatchling mass) and higher clutch mass relative to female mass than L. a. exigua, i.e. both possible ways to enhance offspring number are exerted. As the SSD difference is due to male size (smaller males in L. a. agilis), fecundity selection favouring larger females, together with viability selection for smaller size in both sexes, would explain the female-biased SSD and reproductive characteristics of L. a. agilis. The pattern of intraspecific life-history divergence in L.agilis is strikingly similar to that between oviparous and viviparous populations of a related species Zootoca vivipara. Evolutionary implications of this parallelism are discussed.
Roitberg_et_al_svl_data
Data for adult body size (snout-vent length, SVL)
Abbreviations of characters/columns (if not clear from their designations):
No, this column is for technical purposes of the author;
No_inv, inventary number (in the museum or in the working protocols of the primary researcher);
ssp, subspecies/clade: 1 L. a. agilis, 3 L. a. exigua;
reg3, code of the study sample as in Roitberg et al., 2015, Table 1;
sta, "stages": ad adult, gr gravid, m junger adult, sub subadult, j (=juv) juvenile;
NB: subadults and juveniles, as well as unsexed individuals, were excluded from all analyses in Roitberg et al. 2015;
sex: 1 males, 2 females, 3 unsexed;
Num , individual (study subject) number in the author's extended data base.
filter_adults: 1 individuals considered as "adults" in Roitberg et al. (2015), 2 other individuals.
Abbreviations of museums:
mkhar, Museum of Nature, Kharkiv National Karazin University;
mkiev, National Museum of Natural History Kyiv;
mtd, Senckenberg Natural History Collections Dresden;
zfmk, Zoological Research Museum A. Koenig Bonn;
zmb, Humboldt Natural History Museum Berlin;
zmh, Zoological Museum Hamburg;
zsm, Zoological State Collections Munich.
Comment 1. The codes “mkhar” and “mkiev” do not represent established abbreviations of the corresponding museums and used for simplicity.
Comment 2. This data table does not include Sample 3 (South Sweden) and Sample 9 (Russia, Ciscaucasia 1) because for these samples we used published mean values (respectively, Olsson, 1988 and Ushkalova, 1976).
Comment 3. SVL-values for Sample 4 (Poland) were obtained by extracting log(SVL)-values from published scatterplots (Borczyk & Paśko, 2011: Fig. 1, A,B) and subsequent re-calculating the raw values using the exponent function.
Comment 4. SVL-values for Sample 10 (Russia, Ciscaucasia 2) and Sample 11 (Russia, Ryazan Region) were obtained from published histograms (respectively, Lukina, 1966: Fig. 22 and Zharkova, 1973: Fig. 4).
Comment 5. SVL-values for the other samples are original data of the authors.
Roitberg_et_al_repro_data
Data for female reproductive output and maternal body size:
Abbreviations of characters/columns (if not clear from their designations):
No, this column is for technical purposes of the author;
No_inv, inventary number (in the museum or in the working protocols of the primary researcher);
ssp, subspecies/clade: 1 L. a. agilis, 3 L. a. exigua;
reg3, code of the study sample as in Roitberg et al., 2015, Table 1;
data: "real", usual data, which relate to real individuals; "sur", surrogate data which were generated for those samples where only mean values and standard deviations were available; these artificial samples were used in ANOVA/ANCOVA models (see Roitberg et al. 2015 for details and references);
svl, maternal snout-vent length;
cs_t, egg stage: 1 'enlarged follicles', 2 'oviductal eggs', 3 'corpora lutea', 4 'laid eggs';
cs, clutch size;
m2, post-oviposition female mass;
cm, clutch mass (the total mass of the freshly deposited eggs);
rcm, relative clutch mass (in %);
egg_m, mean egg mass per clutch;
hm, mean hatchling mass per clutch ;
Num, individual (study subject) number in the author's extended database.
ln_rcm, natural logarithm of rcm.
Abbreviations of museums:
mkhar, Museum of Nature, Kharkiv National Karazin University;
mkiev, National Museum of Natural History Kyiv;
mtd, Senckenberg Natural History Collections Dresden;
zfmk, Zoological Research Museum A. Koenig Bonn;
zmb, Humboldt Natural History Museum Berlin;
zmh, Zoological Museum Hamburg;
zsm, Zoological State Collections Munich.
Comment 1. The codes “mkhar” and “mkiev” do not represent established abbreviations of the corresponding museums and used for simplicity.
Comment 2. Sample 3 (South Sweden) should actually include some additional values which could not be extracted from the scatterplot (Olsson, 1993: Fig. 1) because the corresponding points were superimposed. The resulting bias is obviously minor, however: the published values for mean maternal SVL, mean clutch size, and the maternal SVL-clutch size correlation (Olsson, 1993; Olsson & Shine, 1997a) are close to those of Sample 3.
See Roitberg et al. (2015) for important comments to our study traits.
Roitberg_et_al_physical_characteristics_of_study samples
Data for geographic coordinates and climate
Abbreviations of characters/columns (if not clear from their designations):
ID, code of the study sample as in Roitberg et al. 2015, Table 1;
reg3 (as ID);
tmin1 – tmin12, monthly means of minimal temperatures;
tmax1 – tmax12, monthly means of maximal temperatures;
tmax1 – tmax12, monthly means of maximal temperatures;
prec1 – prec12, monthly sums of precipitation;
PC1-clim – PC2-clim, first two principal components of geographic variation for the above 36 climatic variables (see Roitberg et al. 2015 for explanations);
bio1 – bio19, the so called bioclimatic variables:
bio1, Annual Mean Temperature;
bio2, Mean Diurnal Range (Mean of monthly (max temp - min temp));
bio3, Isothermality (BIO2/BIO7) (* 100);
bio4, Temperature Seasonality (standard deviation *100);
bio5, Max Temperature of Warmest Month;
bio6, Min Temperature of Coldest Month;
bio7, Temperature Annual Range (BIO5-BIO6);
bio8, Mean Temperature of Wettest Quarter;
bio9, Mean Temperature of Driest Quarter;
bio10, Mean Temperature of Warmest Quarter;
bio11, Mean Temperature of Coldest Quarter;
bio12, Annual Precipitation;
bio13, Precipitation of Wettest Month;
bio14, Precipitation of Driest Month;
bio15, Precipitation Seasonality (Coefficient of Variation);
bio16, Precipitation of Wettest Quarter;
bio17, Precipitation of Driest Quarter;
bio18, Precipitation of Warmest Quarter;
bio19, Precipitation of Coldest Quarter.
Note that for most study samples, values for longitude, latitude, and climatic variables are average values for several sites from which the study animals were collected; sites which provided relatively small amount of data were not considered for computing these average values. Specifically, for Sample 16, most study animals come from surroundings of Nizhnyaya Neninka (Soltonskiy district); for Sample 17, most study animals were collected near Yailyu (Turochakskiy district). These localities were taken as reference sites for the corresponding samples.
Roitberg_et_al_geographic_distances
Geographic distances among study samples (in km) used in our Mantel tests:
See Fig. 1 in Roitberg et al. (2015) for geographic map and Table 1 for the phenotypic characteristics of the study samples.
The distances were calculated using the geographic coordinates. Note that for most study samples, values for the longitude and the latitude are average values for several sites from which the study animals were collected; sites which provided relatively small amount of data were not considered for computing these average values.
