Carl Smith


Territorial animals sometimes use conspicuous natural landmarks as boundaries to their territories. The utilization of territory-demarcating landmarks may have evolved to limit the costs of territorial defence, since the adoption of clearly defined boundaries by opponents in adjacent territories can reduce the overall rate of aggressive encounters, which can be energetically expensive or might result in injury. Here the role of artificial landmarks as boundaries was tested in territorial male rose bitterling (Rhodeus ocellatus), a fish with a resourcebased mating system. Pairs of size-matched territorial males were permitted to interact for short periods in an otherwise featureless aquarium with an obvious landmark at the shared boundary of their territory either present or absent. The presence of the territory-demarcating landmark significantly reduced both the frequency of territorial incursions by males into adjacent territories and the rate of territorial displays. Males showed individual differences in their propensity to enter the territory of a rival, irrespective of the presence of a territorydemarcating landmark. These results suggest that the cost of defence of a territory may be reduced by utilizing territory-demarcating landmarks, in accordance with the predictions of theoretical models.

Carl Smith and Martin Reichard

Sperm competition occurs when the spermatozoa of one male coincide with those of another to fertilise the same eggs. In some taxa males perform multiple ejaculations, which may function in sperm competition or in maintaining a baseline density of spermatozoa in the female reproductive tract to ensure fertilisation, a process that has been termed ‘topping up’. We investigated multiple ejaculations in the European bitterling (Rhodeus amarus), a freshwater fish that oviposits in freshwater mussels. We quantified spermatozoa in the mussel mantle cavity following ejaculation, and measured sperm motility parameters of males adopting different mating tactics. Following ejaculation spermatozoa density in the mussel increased linearly, peaked after 30 s, and then declined exponentially. Spermatozoa motility parameters did not differ between male mating tactics. We parameterised a model of sperm competition for R. amarus, which accurately predicted male fertilisation probability. We discuss these results in the context of multiple ejaculations and male mating tactics.

Martin Reichard, Carl Smith and Mara Casalini


Female density and resource availability are two key variables that shape mating systems. Theory predicts that reproductive skew will amplify with increased male density and decreasing availability of resources, though limited empirical evidence suggests that this may not always be the case. Here we tested mean crowding, defined as the number of males per unit of resource, and density per se, defined as the number of individuals present per unit area, to investigate their effect on the mating system of Rhodeus ocellatus, a fish with a promiscuous, resource-based mating system. Males were exposed to combinations of high and low levels of crowding and density, while the operational sex ratio was held constant. High levels of crowding significantly affected the proportion of mussel spawning sites defended by males and the proportion of mussels into which sperm was released. In contrast to theoretical predictions, neither density nor crowding influenced overall male aggressive behaviours. Density, but not crowding, had a significant effect on male courtship rate, which arose as a possible trade-off between intra-sexual competition and inter-sexual behaviour. We discuss the results in the context of mating system evolution.

Muna Agbali, Rowena Spence, Martin Reichard and Carl Smith

Research on sexual selection has tended to focus on indirect benefits of female mating decisions, and few attempts have been made to quantify the relative effect of direct and indirect selection simultaneously. Here we compared direct and indirect selection on female mating decisions in the rose bitterling (Rhodeus ocellatus), a fish with a resource-based mating system, using experimental treatments with equivalent consequences for female reproductive success. Direct selection was varied by manipulating the quality of sites available to females for oviposition, and indirect selection by presenting females with males of known genetic compatibility. Manipulating the strength of direct and indirect selection had specific, quantified consequences for embryo survival during incubation. There was a significant effect of both direct and indirect selection on female mating decisions, though direct selection accounted for more variance in female oviposition rate compared with indirect. No interaction between direct and indirect selection was detected. Although effects on female reproductive success were additive, selection for direct mate choice benefits appears to be stronger than for indirect benefits. A possible explanation for weaker selection on indirect benefits in the study species is because females are constrained in making mate choice decisions through alternative male mating tactics, which generate a sexual conflict.

Mirosław Przybylski, Carl Smith, Martin Reichard and Rowena Spence


We investigated reproductive rate in relation to oviposition site distribution and quality in the European bitterling, Rhodeus amarus, a freshwater fish that spawns on the gills of living unionid mussels. In a laboratory experiment male bitterling led females to groups of four mussels at a significantly higher rate than single mussels, irrespective of mussel species. Females spawned significantly more frequently on the gills of mussels in groups than on solitary mussels, and showed a preference for spawning on the gills of Unio pictorum in comparison with Anodonta anatina. In a field experiment the total number of eggs spawned on the gills of four mussels was significantly higher than that of single mussels, though the mean number of eggs per mussel was equivalent within species. There was a significant effect of species on the number of eggs spawned in mussels; U. pictorum and U. tumidus received more eggs than A. anatina and A. cygnea. We discuss these results in the context of mating system evolution.

Carl V. Taylor, Robert Powell, Martin L. Jolley, Donald D. Smith and John S. Parmerlee