Middleborns disadvantaged? Testing birth-order effects on fitness in pre-industrial Finns

Abstract

Parental investment is a limited resource for which offspring compete in order to increase their own survival and reproductive success. However, parents might be selected to influence the outcome of sibling competition through differential investment. While evidence for this is widespread in egg-laying species, whether or not this may also be the case in viviparous species is more difficult to determine. We use pre-industrial Finns as our model system and an equal investment model as our null hypothesis, which predicts that (all else being equal) middleborns should be disadvantaged through competition. We found no overall evidence to suggest that middleborns in a family are disadvantaged in terms of their survival, age at first reproduction or lifetime reproductive success. However, when considering birth-order only among same-sexed siblings, first-, middle- and lastborn sons significantly differed in the number of offspring they were able to rear to adulthood, although there was no similar effect among females. Middleborn sons appeared to produce significantly less offspring than first- or lastborn sons, but they did not significantly differ from lastborn sons in the number of offspring reared to adulthood. Our results thus show that taking sex differences into account is important when modelling birth-order effects. We found clear evidence of firstborn sons being advantaged over other sons in the family, and over firstborn daughters. Therefore, our results suggest that parents invest differentially in their offspring in order to both preferentially favour particular offspring or reduce offspring inequalities arising from sibling competition.

Figure 1. Alternative predictions regarding fitness effects of birth-order.

(a) Resource allocation according to the equity heuristic as a function of birth rank in families with one to four children. Spheres in the upper part of the figure represent the proportion of resources obtained for a child in each of several birth ranks, in sibship sizes ranging from one to four, and across seven periods of growth (‘years’). The bars in the lower part show the cumulative resource distribution for children in an only-child, a two-child, a three-child and a four-child family. Whereas an equity motive produces a fair distribution at any given point in time, it yields a cumulative distribution of investments that is unequal. Adapted with permission from Hertwig, Davis & Sulloway (2002). (b) Fitness as a function of birth-order category (first-, middle- and last-borns): three example alternative predictions to that posited by Hertwig et al. (2002). Left panel: if parental experience is important for offspring success, or if parental investment increases as expected residual reproductive value decreases, all else being equal, middleborns and lastborns should be at an advantage. Middle panel: if parents favour firstborns (from which fitness benefits are likely to be greatest), a higher fitness is expected for firstborns compared with other categories; if parents favour middleborns which are likely to be at a competitive disadvantage, similar success is expected irrespective of birth-order category. Right panel: if resources diminish over time, lastborns are expected to be at a disadvantage.

Figure 2. Probability of surviving to adulthood (age 15).

(a) As a function of overall birth-order category: the probability of surviving to adulthood was marginally significantly influenced by whether an individual was a firstborn, middleborn or lastborn among all offspring (χ² ₂ = 5.76, p = 0.056). (b) As a function of intra-sex birth-order category (among daughters): the probability of surviving to adulthood was significantly influenced by whether a daughter was a firstborn, middleborn or lastborn (χ² ₂ = 6.48, p = 0.04). Figures show predicted values of the minimal model and error bars represent standard errors of the means.

Figure 3. Reproductive success.

(a) Lifetime reproductive success (LRS: number of children raised to age 15) as a function of intra-sex birth-order category (among sons): firstborn sons had the highest LRS, as compared to their younger brothers (χ² ₂ = 5.26, p = 0.07). No such effect was found among sisters (χ² ₂ = 1.60, p = 0.45, not shown in the figure). (b) Lifetime fecundity (number of children produced) as a function of intra-sex birth-order category (among sons): firstborn sons had the highest fecundity, as compared to their younger brothers (χ² ₂ = 12.91, p = 0.002). No such effect was found among sisters (χ² ₂ = 1.33, p = 0.5, not shown in the figure). Figures show predicted values of the minimal model, error bars represent standard errors of the means and sample sizes are indicated at the bottom of the bars.

Figure 4. Firstborn sons start reproducing earlier, probably due to their greater access to resources.

(a) Age at first reproduction as a function of intra-sex birth-order category: middleborn sons began reproduction later than firstborn sons (F_2,560 = 6.54, p = 0.002). No such effect was found among sisters (F_2,564 = 0.78, p = 0.5, not shown in the figure). (b) Proportion of men who owned a land among first-, middle- and lastborn sons: firstborn sons were twice more likely to become landowners (χ² ₂ = 25.71, p<0.0001).