Mechanisms of breeding aggregations in fish

Breeding aggregations are ubiquitous in the animal kingdom, yet experiments are lacking to explain the mechanisms that cause animals to cluster their breeding territories. Interest in animal aggregations is currently undergoing a renaissance with a growing awareness that multiple types of aggregations such as colonies, fish shoals and foraging flocks of birds may be formed by common mechanisms. The emerging concept of public information has the potential to serve as a common thread unifying various types of aggregations. This proposal follows up on an existing FWF project and aims to perform novel experiments that address two hypotheses explaining breeding aggregations: (1) the classic, unresolved hypothesis that high density breeding provides safety from predators and (2) the newer hidden lek hypothesis that sexual selection contributes to breeding aggregations. We examine the taxonomic generality of mechanisms of aggregation by testing hypotheses on fish that have been primarily developed from birds. We combine tightly controlled behavioral experiments in aquariums with complementary field experiments in Lake Tanganyika, including DNA analyses of parentage and other genetic parameters. We have chosen two fish species whose breeding territories are potentially aggregated by different components of mate choice, one of which is also a model species for testing predation avoidance. In our current project we have shown experimentally in three-spined sticklebacks Gasterosteus aculeatus that females prefer males with territories near a more attractive male, supporting the hidden lek hypothesis. We will build on this result by determining whether males increase their mating success by aggregating their breeding territories around more attractive males. Our principal study species is Neolamprologus caudopunctatus, a monogamous cichlid that breeds in dense colonies in which we propose experiments testing two potential mechanisms of colony formation: 1. Breeders choose to settle near neighbors to increase their chances of pairing with a better mate in the future, and 2. Breeders extract public information from predator warnings that neighbors signal to their offspring. The study of these two contrasting model systems will permit us to evaluate the generality of mechanisms that form various kinds of breeding aggregations.

The question of why many animals breed in crowded colonies has vexed ecologists for decades. Colonies are aggregations of breeding sites used only to raise offspring. The most venerable hypothesis, arising in the 1930s, is that colonial animals benefit from high density breeding by finding safety from predators in dense groups. Although this idea has been examined throughout the history of the field, surprisingly, there have been no studies that have experimentally manipulated the formation of colonies. One reason for this gap is that the predominant study animals birds are not conducive to such experiments. We side-stepped this constraint by setting up fish colonies in our institutes giant 16,000 liter ring-shaped aquarium. We introduced 16 males and 16 females of our study species, Neolamprologus caudopunctatus, into the tank and allowed them to pair up, build nests and lay eggs. Over a 2-year period, we observed the formation of 14 colonies, with alternative colonies having predators or no predators present in the tank. We found that breeders aggregated their nests at significantly higher densities when predators were present, making this the first study to experimentally demonstrate that predation can drive colony formation.We combined such aquarium experiments with field research in which our team studied N. caudopunctatus in their native Lake Tanganyika in southern Africa. By mapping the nests and obtaining DNA samples from families in a colony of 126 nests, we obtained substantial new knowledge about the behavior of this colonial species. After performing DNA analyses in our institutes genetics lab to determine the relatedness of individual fish to each other, we discovered that there were no extra-pair fertilizations. This means, that in contrast to most socially monogamous bird species, adulterous matings apparently do not occur wherein the female of the pair produces offspring that were fathered by an outside male. Instead, our DNA analyses revealed the common occurrence of apparent adoption. We found that most broods were a mixture of natal fry that were produced by both parents and foreign fry that were unrelated to both parents. For the first time in a study of adoption, we were able to genetically match a number (12) of foreign fry to their actual parents, and in several cases, the distances between the birth and foster nests were so far (maximum >40 m) as to make it virtually inconceivable that tiny juveniles could have traversed them without being predated by numerous predators surrounding the colony. In human terms, it would be like a toddler crossing a busy city on its own without mishap. Further analyses led to evidence that these fry were carried in the mouths of their parents (which occurs in this species) and deposited into other nests, where adoptive parents apparently allowed the unrelated fry to mix with their own. Such unselfish behavior appears to be a Darwinian paradox (i.e. organisms are expected to maximize their own reproductive success, and would obtain zero evolutionary fitness by raising unrelated young). However, there is an adaptive explanation for adoption, namely that ones own offspring are less likely to be predated in a larger group. Thus our adoption study adds to the evidence of our colony formation experiment that predation may be a driving force of coloniality, to which animals evolve multiple behavioral adaptations.