Eviction Behaviour in the Kalahari Meerkats

Evictions are common in a substantial number of social species, yet they have seldom been studied in detail, and we know relatively little about the causes and consequences of eviction. While previous research has largely focused on the profiles of subordinate females (evictees) that make them most likely to be evicted, in this dissertation I examine eviction behaviour in detail from the dominant females’ perspective (evictor) using the Kalahari meerkats (Suricata suricatta). I focus on how eviction behaviour varies with current environmental, social, and individual conditions and determine whether eviction behaviour is personality-driven or context-dependent (Chapter Three). I find no evidence of eviction behaviour being personality-driven as eviction behaviour is not consistent within individuals over time. However, it appears that evictions are strongly associated with the social environment, especially with the profiles of subordinate females. The dominants evict more and higher proportion of females when there are more subordinate females present and when the age difference between the subordinate and the dominant is small. The dominant females also start evicting earlier in their pregnancy if pregnant subordinates are present in the group. In Chapter Four, I demonstrate a correlation between faecal androgen metabolites and eviction behaviour. I find that androgens increase significantly during pregnancy in dominant females and the androgens levels are consistent within individuals over multiple pregnancies. Androgens levels appear to be associated with the offset of eviction however not the number of evictions. This suggests a threshold relationship between androgens and evictions. I also find no relationship between androgen levels experienced in utero and offspring weight, or survival. However, all individuals that have successfully become dominant later in life, have all experienced relatively low levels of androgens in utero. Though this relationship is interesting, the reason for this is still unknown. Finally, I examine the fitness consequences of eviction on the dominant female. Litters born to dominant females that have evicted more subordinates during their pregnancies have higher chance of survival to 1 week as well as to emergence from their natal burrow. This result supports the hypothesis that the dominants evict subordinates in order to lower the risk of infanticide by the subordinates. Lastly, using a structural equation model, I show the relationship between the profile of dominant females, eviction behaviour, and litter emergence supporting my hypothesis of evictions indirectly increasing litter survival through the eviction of pregnant subordinate females.