Limits...
A test of male infanticide as a reproductive tactic in a cichlid fish

View Article: PubMed Central - PubMed

ABSTRACT

Infanticide and offspring cannibalism are taxonomically widespread phenomena. In some group-living species, a new dominant individual taking over a group can benefit from infanticide if doing so induces potential mates to become reproductively available sooner. Despite widespread observations of infanticide (i.e. egg cannibalism) among fishes, no study has investigated whether egg cannibalism occurs in fishes as a result of group takeovers, or how this type of cannibalism might be adaptive. Using the cooperatively breeding cichlid, Neolamprologus pulcher, we tested whether new unrelated males entering the dominant position in a social group were more likely to cannibalize eggs, and whether such cannibalism would shorten the interval until the female's next spawning. Females spawned again sooner if their broods were removed than if they were cared for. Egg cannibalism occurred frequently after a group takeover event, and was rarer if the original male remained with the group. While dominant breeder females were initially highly aggressive towards newcomer males that took over the group, the degree of resistance depended on relative body size differences between the new pair and, ultimately, female aggression did not prevent egg cannibalism. Egg cannibalism, however, did not shorten the duration until subsequent spawning, or increase fecundity during subsequent breeding in our laboratory setting. Our results show that infanticide as mediated through group takeovers is a taxonomically widespread behaviour.

No MeSH data available.


Proportion of brood surviving at (a) 1 h and (b) 24 h post-manipulation. Kernel density plots are also presented to better visualize the shape of the skewed data in each boxplot. ** indicates p < 0.01.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC5383833&req=5

RSOS160891F3: Proportion of brood surviving at (a) 1 h and (b) 24 h post-manipulation. Kernel density plots are also presented to better visualize the shape of the skewed data in each boxplot. ** indicates p < 0.01.

Mentions: Over the 1 h trials, egg cannibalism was directly observed in both control and takeover groups. Over the first hour of the trials, group takeovers resulted in the cannibalism of 177 eggs (in total), while controls resulted in the cannibalism of 58 eggs (in total). Cannibalism was directly observed in seven of the 25 takeover trials (all committed by the takeover male) and in five of the 14 control groups (four of these were committed by the control breeder male, and once by the control breeder female). After 1 h, there were no differences between the control and takeover groups in terms of the proportions of the broods surviving (GLMqb, t = −0.84, N = 39, p = 0.41; figure 3a), though larger broods were more likely to have a higher proportion of offspring surviving after 1 h (GLMqb, t = 2.05, N = 39, p = 0.048). After 24 h, however, far fewer offspring remained in the takeover groups than the control groups (GLMqb, t = −2.77, N = 39, p = 0.009; figure 3b). Female resistance was not related to the proportion of offspring surviving at either time point (both p > 0.2).Figure 3.


A test of male infanticide as a reproductive tactic in a cichlid fish
Proportion of brood surviving at (a) 1 h and (b) 24 h post-manipulation. Kernel density plots are also presented to better visualize the shape of the skewed data in each boxplot. ** indicates p < 0.01.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC5383833&req=5

RSOS160891F3: Proportion of brood surviving at (a) 1 h and (b) 24 h post-manipulation. Kernel density plots are also presented to better visualize the shape of the skewed data in each boxplot. ** indicates p < 0.01.
Mentions: Over the 1 h trials, egg cannibalism was directly observed in both control and takeover groups. Over the first hour of the trials, group takeovers resulted in the cannibalism of 177 eggs (in total), while controls resulted in the cannibalism of 58 eggs (in total). Cannibalism was directly observed in seven of the 25 takeover trials (all committed by the takeover male) and in five of the 14 control groups (four of these were committed by the control breeder male, and once by the control breeder female). After 1 h, there were no differences between the control and takeover groups in terms of the proportions of the broods surviving (GLMqb, t = −0.84, N = 39, p = 0.41; figure 3a), though larger broods were more likely to have a higher proportion of offspring surviving after 1 h (GLMqb, t = 2.05, N = 39, p = 0.048). After 24 h, however, far fewer offspring remained in the takeover groups than the control groups (GLMqb, t = −2.77, N = 39, p = 0.009; figure 3b). Female resistance was not related to the proportion of offspring surviving at either time point (both p > 0.2).Figure 3.

View Article: PubMed Central - PubMed

ABSTRACT

Infanticide and offspring cannibalism are taxonomically widespread phenomena. In some group-living species, a new dominant individual taking over a group can benefit from infanticide if doing so induces potential mates to become reproductively available sooner. Despite widespread observations of infanticide (i.e. egg cannibalism) among fishes, no study has investigated whether egg cannibalism occurs in fishes as a result of group takeovers, or how this type of cannibalism might be adaptive. Using the cooperatively breeding cichlid, Neolamprologus pulcher, we tested whether new unrelated males entering the dominant position in a social group were more likely to cannibalize eggs, and whether such cannibalism would shorten the interval until the female's next spawning. Females spawned again sooner if their broods were removed than if they were cared for. Egg cannibalism occurred frequently after a group takeover event, and was rarer if the original male remained with the group. While dominant breeder females were initially highly aggressive towards newcomer males that took over the group, the degree of resistance depended on relative body size differences between the new pair and, ultimately, female aggression did not prevent egg cannibalism. Egg cannibalism, however, did not shorten the duration until subsequent spawning, or increase fecundity during subsequent breeding in our laboratory setting. Our results show that infanticide as mediated through group takeovers is a taxonomically widespread behaviour.

No MeSH data available.