Stable eusociality via maternal manipulation when resistance is costless

Abstract

In many eusocial species, queens use pheromones to influence offspring to express worker phenotypes. Although evidence suggests that queen pheromones are honest signals of the queen's reproductive health, here I show that queen's honest signalling can result from ancestral maternal manipulation. I develop a mathematical model to study the coevolution of maternal manipulation, offspring resistance to manipulation and maternal resource allocation. I assume that (i) maternal manipulation causes offspring to be workers against offspring's interests; (ii) offspring can resist at no direct cost, as is thought to be the case with pheromonal manipulation; and (iii) the mother chooses how much resource to allocate to fertility and maternal care. In the coevolution of these traits, I find that maternal care decreases, thereby increasing the benefit that offspring obtain from help, which in the long run eliminates selection for resistance. Consequently, ancestral maternal manipulation yields stable eusociality despite costless resistance. Additionally, ancestral manipulation in the long run becomes honest signalling that induces offspring to help. These results indicate that both eusociality and its commonly associated queen honest signalling can be likely to originate from ancestral manipulation.

Keywords: altruism; maternal effects; parent-offspring conflict; pheromone.

Figure 1

Tree description of the model. See text for details.

Figure 2

Stable eusociality via maternal manipulation with costless resistance. The plots show population-average values vs. generations. In the two top rows, offspring can be influenced by their mother to stay to help (shared control) (a-f). In the two bottom rows, offspring can stay without being influenced (offspring control) (g-l). In red shades, resistance to the maternal influence is favored to evolve (mother-offspring conflict). Because (a) resistance is initially ineffective, (b) the mother initially has some helpers that (c) allow her to reduce maternal care to the second brood, thereby (d) increasing the benefit that second-brood offspring receive from being helped which (a) eliminates selection for resistance. The genetic system is haplodiploid. Parameter values are in the Supporting Information 1 (SI1).

Figure 3

Stable eusociality via maternal manipulation can be obtained under smaller benefit-cost ratios than via offspring control despite costless resistance. The graphs show the outcome across values of the survival benefit for helped maternally provisioned offspring (b_p) vs. the survival cost for helping maternally provisioned offspring (c_p). In blue shade, eusociality is obtained with maternal control of offspring helping behavior (MC). In red shade, eusociality is obtained with either shared control (SC) or maternal control (MC). In green shade, eusociality is obtained with either offspring control (OC), shared control (SC), or maternal control (MC). When the cost for helping maternally provisioned siblings is maximal (here c_p = s₀ = 0.1), the initial workers are sterile. An evolutionary outcome is here considered eusociality if at the end of the process the two broods are present (n_i ≥ 1) and if there is at least one sterile helper in the first brood [n_p1p(1-q) ≥ 1; sterility occurs because in all panels c_n = s₀ = 0.1]. For the left column, the genetic system is diploid (a,c,e). For the right column, the genetic system is haplodiploid (b,d,f). In all panels, s₀ = 0.1. For the top row, s_max = 0.11 (a,b), the middle row s_max = 0.21 (c,d) and the bottom row s_max = 0.31 (e,f). Finally, b_n = b_pds_max/(ds_max – s₀) and d = 1. The remaining parameter values are in the SI1.