Three—
History and Present Status of Southern Elephant Seal Populations

Richard M. Laws

ABSTRACT. The total world population of southern elephant seals in 1990 was estimated at 664,000. Of the three (or four) main stocks, South Georgia and Peninsula Valdes account for 60% of the total; Isles Kerguelen, 28%; and Macquarie Island, 12%. The species was hunted for its oil in the eighteenth and nineteenth centuries and subsequently recovered under protection. The only large-scale industry this century was a government-licensed one at South Georgia, from 1909 to 1964, which was restricted to adult males. The history and rational basis for this industry is described here, including some effects of sealing, population decline from 1931 to 1951, and subsequent recovery to sustained yield level under a research-based management plan. Pup production was estimated, population models were drawn up, and the size of the South Georgia population in 1951 was calculated. A comprehensive survey in the 1985 breeding season indicated the same annual pup production as for 1951 and a population of 357,000 in 1990. Population models indicate that some 75% of the adult male population is at sea during the breeding season, which has implications for aquatic mating of virgin females and nonpregnant mature females.

A literature review of other stocks examines various estimates of pup production by site and years between 1949 and 1990 and converts these by a raising factor to estimates of total population sizes by years. During this period, different stocks of southern elephant seals have increased (Península Valdes, Argentina, by 144% since 1975); have probably remained stable after rapid recovery from exploitation (South Georgia, 15% due to postexploitation recovery of the male stock); or have decreased dramatically at annual rates varying between 2.1% and 8% at different places and periods. Total percentage decreases since 1949 are estimated at 50% for Heard Island, 84% for Marion Island, 57% for Macquarie Island, 96% for Campbell Island, and 93% for Signy Island.

This chapter sets the scene for the more detailed discussions presented by those studying key colonies. Those discussions deal with the factors responsible for changes in population number and advance hypotheses to account for the precipitous decline of certain populations.

Fig. 3.1
Breeding distribution of southern elephant seals in 1990.  Closed circles:  known
breeding colonies (area proportional to the estimated population sizes except for
colonies of less than 5,000).  Open circles:  probable small breeding
populations. Fuller data are presented in table 3.1.

In terms of numbers, the southern elephant seal, M. leonina , is one of the more abundant seal species in the world. There are three main stocks, defined (Laws 1960) as the South Georgia stock, the Kerguelen stock, and the Macquarie Island stock (fig. 3.1). The first is the largest numerically and includes elephant seal breeding colonies in the Scotia arc (South Georgia, South Orkney Islands, South Shetland Islands, South Sandwich Islands) and Gough and Bouvet islands, together with South America and

the Falkland Islands. Although more recent information indicates movements between South American and Falkland Islands elephant seal colonies (Laws, unpubl.), up to now no movements have been reported between them and the remaining colonies; so there may be four stocks. The second stock includes Kerguelen and Heard islands, Marion and Prince Edward islands, and Iles Crozets. Single births have been recorded at Amsterdam and St. Paul islands and South Africa (Laws 1960). Finally, the Macquarie Island Stock includes Macquarie Island, Campbell Island, Auckland Islands, and Antipodes Islands, and one birth has been reported from Tasmania (Laws 1960). A very small number of births have also been reported from the Antarctic continent (Murray 1981). The centers of the three main stocks are roughly 90° and 107° longitude apart, except in the Pacific Ocean sector (163°), which is devoid of suitable islands in the right latitudes and supports no breeding colonies. Genetic studies by N. J. Gales, M. Adams, and H. R. Burton (1989) reported on elephant seal blood samples from Heard and Macquarie islands that were examined for protein variation. They concluded that the two populations "may have diverged genetically, and a very limited gene flow exists between the islands, a finding consistent with limited information from mark-recapture studies" (57). It is likely that a similar division exists between them and the South Georgia stock.

The sealers of the eighteenth and nineteenth centuries came first to hunt the more valuable fur seals for their pelts but did not ignore the larger species, which was hunted for its oil. The right whalers in southern waters also took elephant seals. As fur seals declined, the sealers concentrated more on elephant seals (Bonner 1982; Headland 1989). In his chronological list of Antarctic expeditions, R. Headland (1989) analyzed 930 voyages to the Antarctic between 1786 and 1928, the majority by U.S. (46.4%) and British (19.1%) ships, and showed the frequency of sealing voyages by year. There were progressively lower peaks, separated by periods of partial recovery, about 1820, 1842, 1855, 1875, and 1906. The early voyages primarily took fur seals, which were hunted almost to extinction; the objective of the later voyages was primarily the elephant seal. It is not known how many elephant seals were taken during this period, but it was probably in excess of one million of both sexes, assuming that the original populations totaled at least 600,000 to 750,000, as in this century (Laws 1960; McCann 1985). The sealing industry had virtually ended by 1909, and protection was subsequently conferred on most stocks, which recovered from the overhunting.

The only large-scale elephant sealing during this century was the government-licensed industry at South Georgia between 1909 and 1964 (see Laws 1953a , 1960; Bonner 1958; and Headland 1984), although small numbers were taken in other places, particularly Kerguelen, where 12,000 bulls were taken between 1958 and 1964 (van Aarde 1980). Over the course of sixty-five years, some 260,000 elephant seals, predominantly adult males, were taken at South Georgia (Headland 1984). The industry came to an

end in 1964 as a result of the collapse of the whaling industry at South Georgia; the sealing industry was on a firm sustainable yield basis. As will be shown, the size of the South Georgia elephant seal population in 1951, when government-controlled sealing was at its height, was similar to the size of the population in 1985, some twenty years after sealing had ended. However, the populations in the Kerguelen and Macquarie islands stocks declined steadily over forty years from 1950, even in the absence of sealing. It seems that the elephant seal population in Argentina has been increasing in recent years.

Below I describe the recent history of twentieth-century sealing at South Georgia, because of its relevance to estimating abundance and trends, and what is known about population sizes this century in the three (or four) main stocks, which are considered separately. Estimates, updated to 1990, are given for the total world population of the species.

South Georgia Stock

The Sealing Industry, 1910–1964

The biological basis underpinning rational exploitation of this species is its highly polygynous land-breeding behavior. The sexes are nearly equal in numbers at birth, and a single adult male can serve a large number of cows. In fact, the overwhelming majority of matings are carried out by a very small minority of breeding bulls; according to M. N. Bester and I. S. Wilkinson (this volume), "dominant bulls controlling the harems achieved over 98% of all matings." A large surplus of males can therefore be taken from the beaches without affecting recruitment to the population. Polygyny has resulted in the socially mature male being on average about eight times the weight of the average breeding female (Laws 1984). As the sealing industry sought oil from the blubber, it was most convenient to take the bulls in the earlier part of the breeding season in the spring, when the oil yield is greatest. Because the larger males haul out first on the beaches, the sealers automatically selected the largest animals. Oil yields from autumn sealing are lower, and this practice was discontinued in the 1950s.

To prevent recurrence of the former indiscriminate slaughter, the Falklands Islands government introduced the Seal Fishery Ordinance in 1899 (improved by further ordinances in 1904, 1909, and 1921), with the intention of placing the industry on a rational basis. The coastline of South Georgia was divided into four sealing divisions (and four reserves where no seals were taken). The divisions were generally worked in rotation, each year one of them being closed to sealing. Licenses to take a stipulated number of adult male elephant seals were issued annually to the sealing company. The first license was issued in 1909 to the oldest established whaling company at South Georgia, which held a monopoly until the 1961–1962

season. There was no sealing in 1962–1963, and licenses were issued to a Japanese company for the last two seasons, 1963–1964 and 1964–1965. With the cessation of autumn sealing in the 1950s, sealing began early in September and ended about mid-November. The sealing methods are described in detail by me (Laws 1953a ) and by Bonner (1958). Although attempts were made to utilize carcasses, this was found to be uneconomical and the sole product of the industry continued to be blubber oil.

For many years, the annual quota under license was fixed at 6,000. But in 1948, it was raised to 7,500; in 1949 and 1950, to 9,000; and in 1951, was set at 8,000. Until 1952, the quota was divided equally among the three divisions worked each year, although seal abundances in the four divisions have probably always been unequal. The sealing regulations were altered in 1952 as a result of recommendations made by me (Laws 1960).

Effects of Sealing

Previously, I have presented a variety of evidence that indicated that from the late 1930s onward there were adverse changes in the population, despite the conservation measures in force (Laws 1960). Most significant, it was shown that the average catch of seals per catcher's day's work (CDW) in October had progressively declined by 28% from 1931 to 1951. Over these two decades, there had also been a decline in the oil yield per seal and a lengthening of the season, trends that had begun well before the higher catches in 1948–1951. Consideration of the sex ratios on the breeding beaches, harem sizes (on exploited beaches and in a reserve), and the low average age of males on the breeding beaches supported the indications that the demonstrated decline in the catch per unit effort represented a real decrease in the size of the adult male component of the population.

In Laws 1960, I reported counts of seals on beaches in Divisions I–III and one count in Division IV. These counts were corrected for date and for pup mortality, based on my research, and I concluded that approximately 20,000 pups were born in Division I and 15,000 in Division II. Supplementary information was obtained from the sealers which indicated that seals were much more abundant in Divisions III and IV, and I assumed that 32,500 pups were born in each of these divisions, making an estimated total of approximately 100,000 pups that survived to weaning. Assessing pup mortality from birth to weaning at 2% gave a figure of about 102,000 pups at birth.

I constructed population models to assess the size of the total population in each division, based on my demographic research, on the size of the commercial catches in previous years, and on the age composition of the commercial kill of males in 1951, based on growth layers in the teeth (Laws 1952, 1953b ). Some assumptions were made, and predictions about certain future statistics were advanced, the confirmation of which later appeared to support the assumptions. The results of these calculations were estimates of

the size of the population in each division and of the annual accessions to each age class (see Laws 1960 for details). Other information on breeding behavior suggested that the stock of breeding bulls in each division should be kept equal to one-twelfth of the stock of adult cows, and on this basis surplus accessions of bulls for all divisions combined were calculated at 5,842, close to the average annual catch. But the catches in Divisions I and II for this period were, respectively, about 24% and 65% higher than the estimated surpluses for those divisions, whereas the catches in Divisions III and IV were well below.

Therefore, a new system was introduced in 1952 to control sealing. Most important, divisional quotas were set for the five years 1952–1956, proportional to the estimated size of the available surplus in each division. A safeguard was introduced to monitor the age composition of the catch, by requiring canine teeth to be collected from a 5% sample of the catch which would subsequently be examined to determine age structure of the catch. A minimum length regulation was introduced.

Favorable results were immediately apparent, with a rise in the catch per CDW of 26% over the first 4-year period. In general, sealing operations were conducted earlier (midseason date two weeks earlier) and the average oil yield increased substantially. About 300 teeth were collected annually between 1952 and 1964 for age determination, and W. N. Bonner made control collections from a total of over 400 seals caught. The mean age of the catch sample in 1951 was 6.64 ± 0.34 (SE) years; in 1952, it was 7.20 ± 0.14; and it increased to level off at about 7.71 ± 0.19, 7.70 ± 0.18, and 7.69 ± 0.15 years in 1961, 1963, and 1964, respectively (Laws 1979). This leveling off confirmed that the catch size was sustainable; it corresponds to the estimated average age on attaining social maturity, and the management strategy was to monitor the age distribution of the catch and to adjust annual quotas, as necessary, so as to keep the average age of the catch at 7½ to 8 years. The success of this policy gave confidence in the general accuracy of the estimates of population abundance and life table parameters (see below).

Population Size in 1951

In formulating my recommendations, I had constructed provisional life tables for the two sexes. These life tables adjusted for 102,000 births annually indicated an average midyear population of about 310,000 elephant seals at South Georgia in 1951, that is, about 3.1 times the number of births (see Laws 1960 for details). I reviewed available information on other breeding colonies of the overall South Georgia stock and concluded that its total size was a little more than 315,100.

McCann (1985) revised the life tables, assuming higher female fecundity (88% vs. 82.5%). He found no change in the number of pups born in Cumberland Bay, no evidence for an increase or decrease in the cow

population, and no significant change in the age at first pupping since 1964. He concluded that "a suitable factor by which to convert pup counts to total population of all animals older than pups of the year at the end of the breeding season is 3.5" (14). McCann (1980) also found no difference in body size in 1976–1977 compared to 1948–1951. He reported a three- to fourfold increase since exploitation ended in numbers of bulls ashore, an increase in the mean age of harem bulls, and increased average harem size; he also found that bulls hauled out earlier in the season and stayed ashore longer. There was little change in the number of breeding cows or their age structure. (Comparisons were with results in my papers, based on fieldwork in 1948–1951, listed in Laws 1960.) McCann (1985) concluded that the South Georgia population in 1980 was 350,000 (actually, it should have been 357,000 if calculated from births × 3.5).

McCann's estimate was derived from the 1951 estimate of pups born and his life tables for an unexploited population. In 1951, however, at the height of the sealing period, even assuming a similar pup production, the total population should have been somewhat lower, because the numbers of adult males were reduced in the exploited population. Comparison of Laws (1960) and McCann (1985) shows that the female life tables are really quite similar (perhaps not surprising as McCann adopted some important parameters from me, especially first- and second-year survival); the male life tables differ only in the higher first- and second-year mortality assumed by me and in the reduced survival of older males in the exploited population.

I have therefore recalculated the exploited male life table, by raising the first-year survival to the female level, as McCann did; this gives almost identical survival rates for the two models up to 6 years. Above 6 years the selective removal of older males reduced the adult male population, so that in the exploited 1951 population model there was virtually no survival beyond age 12, compared with age 20 in McCann's (1985) life table. I have calculated the number of adult males estimated by the two models, assuming 54,000 male pups at birth in both exploitation and postexploitation models; McCann reported a sex ratio of 53% males at birth. For the 1951 model there were 17,280 males aged 7 to 12 years (none were older), and for 1985 there were an estimated 36,000 aged 7 to 20 years (the upper limit). The difference is nearly 19,000, suggesting that the 1951 total South Georgia population size was actually about 338,000 (357,000 – 19,000). Sealing ended in 1964, so the recovery in adult male numbers should have been virtually complete by 1970 as full cohorts successively acceded to the adult male population.

Population Size in 1985

The elephant seal population of South Georgia was surveyed comprehensively during the 1985 breeding season (McCann and Rothery 1988). There were 87,711 females and 10,260 adult males counted on shore, and the

female counts were corrected using a model of haul-out distribution over time to adjust for the date of count. The annual pup production was estimated to be about 102,000, the same as the estimate in Laws (1960). I actually counted 26,260 live pups in Divisions I and II and gave an estimate for pup production, corrected for date of count, of 31,075; I rounded up this estimate to 35,000, to allow for the beaches I did not visit. This can be compared with T. S. McCann and P. Rothery's (1988) corrected total of 34,665 in these two divisions, which is very close. McCann and Rothery obtained corrected pup counts of 42,892 and 24,676 for Divisions III and IV, respectively, and suggested that as the numbers in Divisions I and II were approximately the same in 1985 as in 1951, the same should be true of the totals in Divisions III and IV. Thus, there is a good basis for assuming that the number of pups was indeed about 102,000 in both periods, and applying the factor 3.5 gives a population other than pups of 357,000 in 1985. (A possible but small source of error lies in the adoption of my 1960 estimate of pup mortality (2%); e.g., Condy [1978] estimated preweaning mortality at 6%.)

Thus, the total pup production at South Georgia was very similar in two estimates 34 years apart. This suggests that the South Georgia population may have been almost stable since 1951, apart from an increase in adult males after sealing ended in 1964. However, McCann and Rothery (1988) show that substantial annual fluctuations occurred in the number of cows ashore on certain beaches, and their data suggest a possible decline in the 1960s. It seems likely therefore that over the last 40 years, the South Georgia population size has fluctuated around 350,000.

The 1985 count and model presented above indicate that of some 36,000 adult males (aged 7 to 20 years) estimated to be in the South Georgia population, only some 10,000 were counted on land at any one time during the breeding season, and about 26,000 were therefore in the sea, since they have not been seen elsewhere on land. This is relevant to hypotheses about the mating of the virgin females; they are very rarely seen on the breeding beaches in the breeding season, yet the 1985 model (McCann 1985) indicates that some 21,000 virgin females (equivalent to about a sixth of all mature females) mate for the first time at age 3. I observed aquatic matings and believed that newly mature females mate for the first time not on land but in the sea, probably during the usual breeding season of the adult cows (Laws 1956). (With 88% fecundity, the model also indicates that 12% of all adult females—an estimated 14,000—would be nonpregnant in the pupping season and probably also would mate aquatically). Behavior during the aquatic phase of the life cycle may have implications for the causes of the decline in other southern elephant seal populations (SCAR 1991; Bester and Wilkinson, this volume; Hindell, Slip, and Burton, this volume; Fedak et al., this volume).

Other Populations in the South Georgia Stock

Patagonia and the Falkland Islands

I estimated less than 1,000 pups born in the Falkland Islands, representing a population of less than 3,500 (Laws 1960); there is no recent information. For Patagonia, at Península Valdes, I. S. Carrara (1952) estimated a population of approximately 1,000 in 1951; J. A. Scolaro (1976) estimated 3,933 pups born in 1975, implying a population of 13,800 (3.5 × births); D. F. Vergani, M. N. Lewis, and Z. B. Stranganelli (1987) estimated 6,737 pups born there in 1982 (pop. 23,579), increasing at 5.1% annually between 1975 and 1982. C. Campagna and M. Lewis (1992) give the results of eight surveys from 1969 to 1990, corrected for date and area surveyed, concluding that 7,455 pups were born in 1982 (pop. 25,092) and 9,636 pups were born in 1990 (pop. 33,726), having increased at 3.2% annually since 1982.

South Orkney Islands

I estimated the size of the breeding population I studied at Signy Island to be 300 in 1948–1950 (Laws 1960). Subsequently, this has decreased to less than 20 (unpublished British Antarctic Survey records).

King George Island

Vergani (pers. comm.) reported 708 pups born in 1980, declining to 560 by 1990, which represents total populations of approximately 2,500 and 2,000.

Nelson Island

Vergani, Lewis, and Stranganelli (1987) reported 106 pups born in 1985, representing a population of approximately 370.

Avian Island

One birth has been reported from Avian Island (68° S.) (unpublished British Antarctic Survey record).

Gough Island

Bester (1980; pers. comm.) reported 32 pups born in 1977 and 28 in 1990, suggesting a stable population of about 105.

South Sandwich Islands and Bouvet Island

The species probably breeds in small numbers on these islands, but there is no information.

These populations represent a total of about 39,720 in 1990, about 10% of the total South Georgia stock in 1990, which is estimated at about 397,000 compared with McCann's (1985) estimate of some 369,000 based on counts spread over earlier years (table 3.1). The difference is due to a real increase in the Península Valdes population and the upward revision of the earlier South Georgia estimate to allow for the recovery of the adult male stock as discussed above.

The Iles Kerguelen Stock

Iles Kerguelen:
Courbet Peninsula

Here there are difficulties in making comparisons from year to year because of uncertainties about the proportion of the population censused. Almost all breed on the Courbet Peninsula between Point Molloy and Cap Noir, here called the study area. R. J. van Aarde (1980) summarized previous counts of breeding cows plus weaned pups (roughly equal to pups born) in the study area in 1958, 1960, and 1970 and presented his own figures for 1977.

These were, respectively, 41,970, 40,050, 55,252, and 42,420. However, the first two are underestimates because they were uncorrected and so are not comparable. On the basis of the estimates for 1970 and 1977, van Aarde calculated that there had been a decline of 4.6% annually over this period (and, raising by 3.5 times, this corresponds to a decline in total population from 193,382 to 148,470). Bester (1982) counted 36,900 pups in the study area in 1979, representing a population of 129,150; a continued decline of 4.6% from 1977 would have resulted in 38,607 pups (total 135,124). C. Guinet, P. Jouventin, and H. Weimerskirch (1992) obtained a figure of 41,000 pups in 1989 and suggest that the population has stabilized; this would indicate a population of about 143,500 in 1990.

Heard Island

R. Carrick and S. E. Ingham (1962) gave an estimate of 23,000 pups for 1949, representing a total population of 80,500. By 1985, Burton (1986) estimated pup production at 13,000, representing a population of 45,500, which had therefore declined by 2.4% annually, averaged over the period from 1949 to 1985. Continuing decline at this rate would result in a population of 40,355 in 1990, a total decline of 50%.

Marion Island

In 1973 and 1976, total numbers in various age classes were counted in the period November 13–19, when combined pup and weaner counts were at a maximum (Condy 1978); no adjustments to these were therefore necessary except for preweaning mortality (5.99%). A mean of 1,049 pups and weaners were counted, which adjusted for mortality gave an estimated 1,115 births annually in this period. Raised by 3.5, this indicates a total population of 3,903 in 1976. R. W. Rand (1962) had counted the pups, weaners, and dead pups on the island in the second half of November 1951 and obtained a figure of 3,662, corresponding to a total population of 12,817. (Condy calculated a 69.5% decline between 1951 and 1976.) P. R. Condy (1977) analyzed three comparable counts from 1951, 1965, and 1974 and estimated the rate of decrease of the Marion Island breeding population over this period at 4.8% a year.

To study the trend between 1973 and 1982 in more detail, J. D. Skinner and R. J. van Aarde (1983) analyzed results of censuses from a subset of the breeding beaches, representing about a third of the annual pup production, for the period 1973–1982 (excepting 1978); they are not directly comparable to Condy's (1977) figures. This indicated an exponential decrease in pup production over the period 1973–1982 at a mean annual rate of 8%.

The most recent data come from Bester and Wilkinson (this volume) and Wilkinson and Bester (in press). The first relate to a main study area representing about 10% of the coastline, censused from 1974 to 1989 fol-

lowing the methods of Condy (1978). In 1976 and 1986–1989, total pup counts of the whole island were made, which confirmed that numbers in the main study area provide a representative index of the total population. They conclude that the overall decline between 1974 and 1989 was 4.8% annually, but in 1983–1989 it had slowed to 1.9%. Wilkinson and Bester (in press) estimated the total pup production for the island in 1989 at 585; a further 1.9% decline would reduce this to 574 in 1990. Raised by 3.5, this gives an estimated total population of 2,009 in 1990, a decline of 83.7% since 1951 and of 48.5% since 1976. In percentage terms, this population has been most adversely affected of all except Signy Island and Campbell Island (see below).

Prince Edward Island

Condy (1978) also counted 386 pups and weaners on the east coast of Prince Edward Island in 1977; applying Condy's (1978) correction factor for preweaning mortality gives an estimate of 411, raised to a total population of 1,439. If this population also declined on average at 4.8% annually between 1974 and 1989 and 1.9% in 1989–1990, its 1990 population would be 782.

Iles Crozet

A. Barrat and J. L. Mougin (1978) reported about 3,000 pup births in 1976, corresponding to a total population of 10,500, having declined by 5.8% annually between 1966 and 1976 (SCAR 1991). Guinet, Jouventin, and Weimerskirch (1992) reported 612 pups in 1989, representing by inference a total population of 2,142, having declined at 5.7% annually between 1980 and 1989. Extrapolating a further year at this rate of decline implies a population of 2,019 in 1990 (SCAR 1991).

The populations comprising the Iles de Kerguelen stock represent a total of about 189,000 in 1990 (table 3.1), about 28% of the total world population, having declined dramatically over the last 40 years, one population by over 80%. This is in marked contrast to the South Georgia stock, which has increased at South Georgia since 1951 by 20,000 (15.3%) as a result of the postexploitation recovery of the male component and at Península Valdes by 20,000 (144%) since 1975, due to natural increase.

The Macquarie Island Stock

Macquarie Island

M. A. Hindell and H. R. Burton (1987) have recently described the past and recent status of the elephant seal at Macquarie Island. This is one of the most continuously studied populations since observations began in

1949. Until the early 1980s, only adult females and males were counted in the isthmus study area where most of the seals haul out to breed (Carrick and Ingham 1960; Carrick et al. 1962). However, Hindell and Burton (1988) showed that the haul-out pattern of breeding females is so predictable that a census on any day within two weeks of the haul-out maximum can be used to estimate the maximum number accurately. In the 1985 breeding season, daily counts of adult females, weaned pups, dead pups, and breeding males were made on the isthmus study area, and a complete island census was carried out on a single day, a week before the peak number; an estimate of the peak number of females was derived from the haul-out curve modeled from the isthmus data.

Hindell and Burton (1987) estimated the maximum number of females ashore on the isthmus for 21 years between 1949 and 1985. They showed that there was a significant correlation between years and maximum number of females, with an average annual rate of decrease of 2.1%. There was a similar percentage decrease in the numbers of males ashore.

The estimated total island pup production in 1959 was 44,501, compared with 24,713 in 1985. Applying the 3.5 raising factor (McCann 1985) gives a total population of 156,000 in 1959 and 86,500 in 1985, that is, 44.6% less than in 1959. Hindell and Burton (1987) concluded that the population was probably declining at least from 1950, and they estimated that the whole island population in 1949 was about 183,000. Extrapolating from 1985 to 1990, assuming a continued decline at 2.1% per annum, gives a total population of 77,791 in 1990, which represents an overall decline of 57.5% since 1949.

Campbell Island

There is a small population of elephant seals at Campbell Island. J. H. Sorensen (1950) gave estimates from counts of the numbers of pups born in 1941, 1942, 1945, and 1947, increasing from 75 to 100 in 1941 to 191 in 1947. Taking these figures at face value, this suggests an average of about 140 pups born in the 1940s and, applying the raising factor used by McCann (1985), a total population of about 500 at this time. The only recent information appears to be from R. H. Taylor and G. A. Taylor (1989), who give a pup production of only 5, representing a total population numbered in tens of animals, let us say 20. This would represent a decline of 96% since the 1940s.

Antipodes Island

Taylor and Taylor (1989) also give an estimate of 113 for the pup production at Antipodes Island in 1978. There is no information on the trend in this population, but if there has been no change, applying the raising factor gives a population of about 400 in 1990.

Thus the Macquarie Island stock, which is essentially confined to Macquarie Island for breeding, had an estimated total size of about 78,000 in 1990 (table 3.1), representing only 12% of the total world population, having declined dramatically from about 183,000 in 1949, a decrease of 57% over 42 years. This is again in marked contrast to the South Georgia stock and more like the Iles Kerguelen stock.

Conclusions

As the work summarized here indicates, during the period from 1949 to the present, different populations of southern elephant seals have increased (Península Valdes), probably remained stable after rapid recovery from the exploitation that ended in 1964 (South Georgia), or decreased dramatically (Iles Kerguelen, Marion Island, Macquarie Island, Signy Island, Campbell Island).

The population sizes are derived from estimates of pups born, raised by the factor of 3.5 to obtain estimates of the size of populations, excluding pups. Different raising factors should be applied to increasing, stable, or declining populations, and in the absence of adequate demographic knowledge of the different populations, the estimates of annual pup production are more accurate than population abundance estimates. However, it is desirable to have population abundance estimates for other purposes (such as estimating population biomass and food consumption; Boyd, Arnbom, and Fedak, this volume) and since a common factor is applied, the derived population trends are the same as for pup production. Subject to this reservation, table 3.1 shows an estimated total population of the southern elephant seal in 1990 of about 664,000, which compares with an estimated total population of 758,000 derived from data obtained at different times during the period 1949–1980 (McCann 1985), an overall decline of some 12.3%. The decline is not universal, however, and 60% of the total world population is in the South Georgia (and South American) stock, which is stable or increasing.

If we look at the populations that have decreased, currently about 40% of the total world population, they have decreased on average by about 31%, but some have decreased to a much greater extent; for example, the major stock at Macquarie Island has declined by an estimated 57% since 1949. Information about the possible causes of the declines is presented elsewhere in this volume (chaps. 4, 5, and 6). A workshop meeting convened by the Scientific Committee on Antarctic Research (SCAR 1991) also addressed the possible causes of past and present population trends but did not reach any firm conclusion. It seems that the causes of the declines may differ among populations; however, the causes remain speculative and essentially unknown. In any case, the contrast between the population

trends in the northern elephant seal, M. angustirostris (Stewart et al., this volume), and the southern elephant seal is striking and is an urgent topic for further research.

There is no evidence to suggest that the population declines are the result of factors influencing the seals while ashore, indicating that they result from factors operating while the animals are at sea (SCAR 1991). In this respect, the evidence presented here indicating that a large proportion of adult males and newly mature and fully adult females remain in the sea during the breeding season may be relevant. For successful mating, it seems unlikely that these components of the population are widely dispersed at this time. They may be in the general vicinity of the breeding islands, or relatively concentrated in other parts of the species' foraging range. Tracking studies using satellite geolocation equipment should eventually provide an answer to this question.

Much of this chapter describes the government-controlled industry at South Georgia between 1909 and 1964. This is a topic that is not covered elsewhere in this volume and has been included here because of the light it throws on the population dynamics of the species—and because this stock constitutes some 60% of the total population of the species. Despite the removal of some 260,000 bulls during a 65-year period, the South Georgia population appears to have remained fairly stable or to have fluctuated only slightly, a confirmation that the industry was on a firm sustainable yield basis after 1952. This elephant seal stock in fact provides a classic example of rational utilization of a natural renewable resource.

References

Barrat, A., and J. L. Mougin. 1978. L'elephant de mer Mirounga leonina de L'île de la Possession, Archipel Crozet (46°25¢ S, 51°45¢ E). Mammalia 42: 143–174.

Bester, M. N. 1980. The southern elephant seal Mirounga leonina at Gough Island. South African Journal of Zoology 15: 235–239.

———. 1982. An analysis of the southern elephant seal Mirounga leonina breeding population at Kerguelen. South African Journal of Antarctic Research 12: 11–16.

Bonner, W. N. 1958. Exploitation and conservation of seals in South Georgia. Oryx 4: 373–380.

———. 1982. Seals and Man . Seattle: University of Washington Press.

Burton, H. 1986. A substantial decline in numbers of the southern elephant seal at Heard Island. Tasmanian Naturalist 86: 4–8.

Campagna, C., B. J. Le Boeuf, M. Lewis, and C. Bisioli. 1992. Equal investment in male and female offspring in southern elephant seals. Journal of Zoology, London 226: 551–561.

Campagna, C., and M. Lewis. 1992. Growth and distribution of a southern elephant seal colony. Marine Mammal Science 8: 387–396.

Carrara, I. S. 1952. Lobos marinos, pinguinos y guaneras de las costas del littoral

maritimo e islas adyacentes de la Republica Argentina. Universidad Nacional La Plata, Facultad de Sciencias Veterinarias, Catedra de Higiene e Industrias, Publ. Espec.

Carrick, R., S. E. Csordas, S. E. Ingham, and K. Keith. 1962. Studies on the southern elephant seal, Mirounga leonina (L.). III. The annual cycle in relation to age and sex. CSIRO Wildlife Research 7: 119–160.

Carrick, R., and S. E. Ingham. 1960. Ecological studies on the southern elephant seal, Mirounga leonina (L.), at Macquarie Island and Heard Island. Mammalia 24: 325–342.

———. 1962. Studies on the southern elephant seal, Mirounga leonina (L.). V. Population dynamics and utilization. CSIRO Wildlife Research 7: 198–206.

Condy, P. R. 1977. Annual cycle of the southern elephant seal Mirounga leonina (Linn.) at Marion Island. South African Journal of Zoology 14: 95–102.

———. 1978. The distribution and abundance of southern elephant seals Mirounga leonina (Linn.) on the Prince Edward Islands. South African Journal of Antarctic Research 8: 42–48.

Gales, N. J., M. Adams, and H. R. Burton. 1989. Genetic relatedness of two populations of the southern elephant seal, Mirounga leonina. Marine Mammal Science 5: 57–67.

Guinet, C., P. Jouventin, and H. Weimerskirch. 1992. Population changes and movements of southern elephant seals on Crozet and Kerguelen archipelagos in the last decades. Polar Biology 12: 349–356.

Headland, R. 1984. The Island of South Georgia . Cambridge: Cambridge University Press.

———. 1989. Chronological List of Antarctic Expeditions and Related Historical Events . Cambridge: Cambridge University Press.

Hindell, M. A. 1991. Some life-history parameters of a declining population of southern elephant seals, Mirounga leonina. Journal of Animal Ecology 60: 119–134.

Hindell, M. A., and H. R. Burton. 1987. Past and present status of the southern elephant seal (Mirounga leonina ) at Macquarie Island. Journal of Zoology, London 213: 365–380.

———. 1988. The history of the elephant sealing industry at Macquarie Island and an estimate of the pre-sealing numbers. Papers and Proceedings of the Royal Society of Tasmania 122: 159–176.

Laws, R. M. 1952. A new method of age determination for mammals. Nature 169: 172.

———. 1953a . The elephant seal industry at South Georgia. Polar Record 6: 746–754.

———. 1953b . A new method of age determination for mammals with special reference to the elephant seal, Mirounga leonina Linn. Falkland Islands Dependencies Survey, Scientific Reports 2: 1–11.

———. 1956. The elephant seal (Mirounga leonina Linn.). II. General, social, and reproductive behaviour. Falkland Islands Dependencies Survey, Scientific Reports 13: 1–88.

———. 1960. The southern elephant seal (Mirounga leonina Linn.) at South Georgia. Norsk Hvalfangst-Tidende 49: 446–476, 520–542.

———. 1979. Monitoring whale and seal populations. In Monitoring the Marine Environment , ed. D. Nichols, 115–140. New York: Praeger.

———. 1984. Seals. In Antarctic Ecology , vol. 2, ed. R. M. Laws, 621–715. London: Academic Press.

McCann, T. S. 1980. Population structure and social organization of southern elephant seals, Mirounga leonina (L.). Biological Journal of the Linnaean Society 14: 133–150.

———. 1985. Size, status and demography of southern elephant seal (Mirounga leonina ) populations. In Sea Mammals of South Latitudes: Proceedings of a Symposium of the 52d ANZAAS Congress in Sydney—May 1982 , ed. J. K. Ling and M. M. Bryden, 1–17. Northfield: South Australian Museum.

McCann, T. S., and P. Rothery. 1988. Population size and status of the southern elephant seal (Mirounga leonina ) at South Georgia 1951–1985. Polar Biology 8: 305–309.

Murray, M. D. 1981. The breeding of the southern elephant seal, Mirounga leonina L., on the Antarctic continent. Polar Record 20: 370–371.

Rand, R. W. 1962. Elephant seals on Marion Island. African Wildlife 16: 191–198.

SCAR. 1988. Report of a meeting held at Hobart, Tasmania, August 23–25, 1988. Biomass Report Series , no. 59: 1–61.

———. 1991. Report of the workshop on southern elephant seals, Monterey, California, May 22–23, 1991.

Scolaro, J. A. 1976. Censo de elefantes marinos (Mirounga leonina L.) en el territorio continental Argentino. Informes Técnicos 1.4.2, Centro Nacional Patagonico, Puerto Madryn, R. Argentina, 1–12.

Skinner, J. D., and R. J. van Aarde. 1983. Observations on the trend of the breeding population of southern elephant seals, Mirounga leonina , at Marion Island. Journal of Applied Ecology 20: 707–712.

Sorensen, J. H. 1950. Elephant seals of Campbell Island. Cape Expedition Series 6: 1–31.

Taylor, R. H., and G. A. Taylor. 1989. Re-assessment of the status of southern elephant seals (Mirounga leonina ) in New Zealand. New Zealand Journal of Marine and Freshwater Research 23: 201–213.

van Aarde, R. J. 1980. Fluctuations in the population of southern elephant seals Mirounga leonina at Kerguelen Island. South African Journal of Zoology 15: 99–106.

Vergani, D. F., M. N. Lewis, and Z. B. Stranganelli. 1987. Observation on haul-out patterns and trends of the breeding populations of southern elephant seals at Península Valdes (Patagonia) and Stranger Point (King George Island). SC-CCAMLR-VI/BG/36 , Hobart, Tasmania, 1–9.

Vergani, D. F., and Z. B. Stanganelli. 1990. Fluctuations in breeding populations of elephant seals Mirounga leonina at Stranger Point, King George Island 1980–88. In Antarctic Ecosystems , ed. K. R. Kerry and G. N. Hempel, 241–245. Berlin: Springer Verlag.

Wilkinson, I. S., and M. N. Bester. 1990. Duration of postweaning fast and local dispersion in the southern elephant seal, Mirounga leonina , at Marion Island. Journal of Zoology, London 222: 591–600.

———. In press. Population parameters of the southern elephant seal population at Marion Island. Antarctic Science .