The RA family has always been a fascinating if frustrating group. To this day I am not sure if certain animals I have given "R" names to are really members of the RA or XA families. I'm also not sure which clan area the RAs should be assigned to. The history of the RAs has not been clear and simple.

The first sighting of Remedios in October 1973; peering at us from behind a bush

I first met the RAs' matriarch on the western side of the Park on October 5, 1973. Harvey and I were with the KA family, which was walking casually past the vehicle while he photographed them, when a left-tusked female with a five-year-old calf appeared and began demonstrating at the car. She charged,but carefully kept a large bush between herself and the car, indicating that she was not being seriously aggressive. After a couple of mock charges she peered at us from over the bush looking a little ridiculous with her one, short, left tusk. Her calf was frightened by her mother's nervousness. We managed to get some poor photographs of both of them.

I saw this left-tusked female two more times that October: one time on her own at the edge of a known group and another time in a large aggregation of over 50. I only saw her twice in 1974 (this was still during the time I was in Amboseli only part-time), again once on her own near a small group and then with a large herd. Fortunately, she was easy to recognize, but no grouping or ranging pattern was emerging. Of the five times I had seen her, she was on the west side of the Park three times and on the east twice. Even though I did not know much about her I assigned her the group letter code "R" and named her Remedios. (It was at that time that I was reading Gabriel Garcia Marquez's One Hundred Years of Solitude and I liked the names in his book.)

In August 1975 just before I began the full-time study, I found Remedios with a new calf about two months old. From this time on I saw Remedios regularly. She appeared to associate with a variety of animals but finally there seemed to be some consistency in the associations. Her own older calf, now seven years old, and three other young females were usually with her. I named these females Renata, Rachel and Ruth and Remedios's older calf was eventually named Rebecca.

In Chapter 6, Ecological patterns of variability in demographic rates, Phyllis Lee, Keith Lindsay and Cynthia Moss investigate seasonal and annual variation in rates of conception and calf mortality using 30 years of demographic data for the Amboseli population.  

They show that conceptions are most frequent in February and May, and less frequent in October through December.  Conception rates closely follow food availability provided by two months grass growth in relation to the previous rainfall. Females who conceive for the first time are less closely tied into seasonal patterns of food availability.  While the Amboseli elephants' reproduction has a seasonal pattern, the year-round availability of swamp vegetation may buffer them against the severe condition loss and strict seasonality of reproduction seen elsewhere in Africa. 

Between years, however, annual rates of conception for available females (those not in the first 12 months of lactation or already pregnant) were strongly suppressed by droughts. During droughts, calves under the age of nutritional independence (24 mo) were likely to die. 

From Amboseli lifetable statistics the authors then examined early calf deaths in relation to the population size:  with more individuals in a population, the food available to all others is expected to be reduced in a food-limited ecosystem. During years when the elephant population concentrated within the central area of its range (due to poaching pressure outside), the rate of calf mortality was positively correlated with population size. 

However, after 1991, the effects of the ivory trade ban improved security outside the central well-protected area, and the elephants began to disperse more widely. With dispersal and access to additional, the effects of density on calf mortality were greatly reduced. 

The dynamic and unpredictable ('non-equilibrium') Amboseli ecosystem results in pulses of population growth and decline over the very long-term. Population models need to take ecosystem variance and human-induced mortality into account in order better to understand and predict the consequences of stochastic changes on the capacity for elephant populations to persist in a changing world. Understanding the factors that contribute to variability in demographic parameters and population growth or decline, both within and between populations, is of obvious importance to conservationists and managers across the African continent.