Elephant Seals on the Farallones, Population Structure of an Incipient Breeding Colony

ELEPHANT SEALS ON THE FARALLONES, POPULATION STRUCTURE OF AN INCIPIENT BREEDING COLONY

BURNEY J. LE BOEUF, DAVID G. AINLEY, AND T. JAMES LEWIS Downloaded from https://academic.oup.com/jmammal/article/55/2/370/842919 by guest on 25 September 2021

ABSTRACT.-The colonization of Southeast Farallon Island, California, by north ern elephant seals, Mirounga angustirostris, began in 1959 with the appearance of a single animal. By 1971, over 100 juveniles were counted during the fall and spring haulout. Breeding began on the island in 1972. The age of juveniles and their pattern of island use for the two years preceding the birth was similar to that of Ana Nuevo Island, a nearby rookery, The majority of seals onshore during the fall and spring peaks ,",,-ere juveniles less than two years old. Tagged individuals sighted on Southeast Farallon Island were born on rookeries to the south: Aiia Nuevo Island, San Miguel Island, and San Nicolas Island. Males were observed more frequently than females. Utilization of the hauling~out place by juveniles was not determined primarily by availability of space. During the first breeding season, adult females arrived on the island before the first male was observed.

The purpose of this paper is to describe some aspects of breeding colony formation in northern elephant seals, M irounga angustirostris. Specifically, we describe events and the individuals involved in initiating breeding on a California island, and the pattern of island use prior to the commencement of breeding as well as the age, sex, and origin of the immigrants. The popu~ hUon structure of an incipient breeding colony of pinnipeds has not been studied previously. On 20 January 1972, a northern elephant seal was born on Southeast Faral~ Ion Island, one of a group of islands west of San Francisco, California. This is the first elephant seal birth recorded here and illustrates that this species is continuing to expand its breeding range northward since its near~extennination during the last century. It has been speculated that elephant seals once bred here over 100 years ago, a reasonable speculation since the island lies within the fonner breeding range of the species (Radford et aZ., 1965). Northern elephant seals were once found from Cabo San Lazaro in Baja California, Mexico, to Point Reyes peninsula north of San Francisco, a range of approximately 1621 kilometers (km) (Scammon, 1874). Intensive exploita~ tion by man during the early part of the 19th century reduced the species to near extinction by 1869 (Scammon, 1874). From 1884 to the middle 1930's the species bred only on Isla de Guadalupe, an isolated island 243 km west of central Baja California; in 1892 the total population was estimated at less than 100 animals (Bartholomew and Hubbs, 1960), vVith protection pro vided by the Mexican government in 1922, joined later by the United States government, the elephant seal began a remarkable comeback which continues to the present day. According to Rice et al. (1965), initial observations of elephant seals on islands within their former range occurred as follows: Islas San Benito (1918), 370 May 1974 LE BOEUF ET AL.-POPULATION OF ELEPHANT SEALS 371

San Miguel Island (1925), Islas Los Coronados (1948), Santa Barbara Island (1948), San Nicolas Island (1949), Ario Nuevo Island (1955), Anacapa Island (1958), Isla Cedros (1965), and Santa Rosa Island (1965). At present, ele phant seals breed on all the above islands except Anacapa Island and Santa Rosa Island. In nearly all the above localities, the year of first breeding is Downloaded from https://academic.oup.com/jmammal/article/55/2/370/842919 by guest on 25 September 2021 unknown but probably fonowed initial reoccupation by several years. This interval is known for Ano Nuevo Island and Southeast Farallon Island. At Ana Nuevo Island, the first birth occurred in 1961, six years after the first animal was sighted (Radford et al., 1965). The first birth on the Farallones occurred 13 years after Thoresen sighted a single elephant seal there in July 1959 (Thoresen, 1959). Because Southeast Farrallon Island has been under biological surveillance for the past few years and several thousand elephant seals from various rook eries to the south have been tagged during this period (Le Boeuf, unpublished data), we can describe the seasonal distribution of animals on this island with respect to sex and age for the two years before the first birth occurred. Iden tification of marked animals enables us to specify the birthplace of some of the animals, the length and regularity of their visits, and other locations where they have been observed. We will then compare these data to similar data from a breeding colony, Ano Nuevo Island, located 89 km south of the Farallones. METHODS Censuses.-Since November 1969, researchers associated with the Point Reyes Bird Ob servatory have made daily censuses of elephant seals on Southeast Farallon Island and periodic visits to other islands in the Farallon group. Southeast Farallon Island, covering 95 acres and located 49 km west of the San Francisco Golden Gate Bridge, is situated north of all other northern elephant seal rookeries (Fig. 1). Researchers at the University of California at Santa Cruz took censuses on Aiio Nuevo Island daily during the breeding season and weekly during the rest of the year. During each census at Afio Nuevo Island the identity of tagged individuals was re corded and the sex and age category of each animal was recorded when possible. The following categories were distinguished; pups (less than four months old and observed from January through April), breeding or adult females (3--4 years and older), subadult males (5 to 8 years old), and adult males (9 years and older). With the exception of pups, the differential size of the animals in the above categories indicated sex. The sex of a pup was detemlined by the presence or absence of a penile opening approximately 12. centi meter (cm) posterior to the umbilicus. Only the juvenile category (approximately 1 to 4 years old) presented an identification problem. The close morphological resemblance between juveniles of both sexes and breeding females made quick and reliable distinctions in the field difficult, particularly between late March to early May when both categories ap· peared grouped together at Ano Nuevo Island. Furthennore, because sex differences among them are minimal, juveniles could only be sexed when lying on their sides or backs. Tagging.-Le Boeuf and his associates began tagging elephant seals in 1968. Since that time over 8,000 Mirounga from seven different rookeries have been marked individually, the majority of these being newly weaned pups about 30 days old. In 1968, tags were affixed to the fleshy trailing edge of the hindflipper, a practice which resulted in a high frequency of tag losses. This technique was discontinued in subsequent seasons and since then tags have been placed in the webbing of the hindflippers where they are less likely Vol. 55, No.2 372 JOURNAL OF MAM).·fALOGY I / ( Downloaded from https://academic.oup.com/jmammal/article/55/2/370/842919 by guest on 25 September 2021 FARALLON IS .• SAN \ FRANc/SCO \ At10 NUEVO I.. \ \ \ \ CALIFORNIA\\

SAN MIGUEL I. • ~ .. \ LOS SAN NICOLAS I.• ANGELES

I. DE GUADALUPE '

IS. SAN BENITO •. I. CEDROS I ~.... 'C-

Flc. I.-The Farallon Islands in relation to other northern elephant seal rookeries. May 1974 LE BOEUF ET AL.-POPULATION OF ELEPHANT SEALS 373

TABLE I.-Chronicle of er;ents leading up to birth, presumed copulation, alld departure of prillcipals in 1972.

Male on ANI and FAR Cowls) and pup on FAR Date Activity Date Activity Downloaded from https://academic.oup.com/jmammal/article/55/2/370/842919 by guest on 25 September 2021 28 Dec. 1971 Immigrant subadult sighted, tagged (GI159), and marked* LPB on ANI 2:8 Dec. to 4 Feb. 1972 present on ANI 17 Jan. 1972 2 cows arrive 20 Jan. I cow (C1) gives birth to pup (PI) 2.0 Jan. other cow departs 5 to 8 Feb. absent from ANI 6to 9 Feb. present on FAR near C1 9 to 10 Feb. present on ANI l3 to 16, 18 to 20, and 22 Feb. present on FAR 12 Feb. CI departs iO to 21, and 27 Feb. to 3 March present on ANI 21 April last date PI observed on FAR

* All males on ANI during the breeding SeaSOn are given names which are bleached into their pelage (Le Boeuf, 1971).

to be detached. Reference to the tagged animals in this paper means aninlab that are one to five years old. Two types of cattle ear tags have been employed. The first is 1 by 4% cm and is made of monel. One or both of the outer sides is embossed with a number and the inner sur face contains the inscription: "NOTIFY U CALIF, SANTA CRUZ USA." TIlese tags wear well but are difficult to read except at close quarters. The other more frequently used tag is larger, 2 by 4% cm made of colored plastic embossed with large black numbers. The inner surface bears the same inscription as the monel tag. These plastic tags must be re placed after about three years because of wear but they offer the advantage of being easy to read from a distance of several meters without disturbing the animals. Elephant seals from different rookeries have been marked with different colored tags as follows: Isla de Guadalupe (blue), Islas San Benito and Isla Cedros (white), San Miguel Island (yellow), San Nicolas Island (red), Ano Nuevo Island (green), and Southeast Farallon Island (pink). Throughout the rest of this paper the following abbreviations \vill be used: Southeast Farallon Island (FAR), Ano Nuevo Island (ANI), San Miguel Island (Si\U), and San Nicolas Island (SNI).

RESULTS Birth and Subsequent Breeding Table 1 summarizes important events associated with the birth of the single pup on 20 January 1972 on FAR. It should be noted that the adult females were present before the arrival of the subadult male. Because neither female 374 JOURNAL OF MAM:tvrALOGY Vol. 55, No.2

was tagged their birthplace and age are unknown. Parturition occurred four days after arrival and the female nursed her pup for 24 days. Thus, the fe male was on shore continuously for 28 days. The temporal patterning of the birth, nursing period, and departure is typical of females of the species and is within the range of variation observed at ANI where, on the average, marked Downloaded from https://academic.oup.com/jmammal/article/55/2/370/842919 by guest on 25 September 2021 females gave birth seven days after arrival and nursed their pups for 27 days (Le Boeuf, 1972). Although we did not witness copulation, it is most likely that it occurred since the only male on the island, LPB was present and remained near the female on days 18 to 21 following the birth (his name, bleached into his pelage on both sides of his body, facilitated identification). He may have been in the vicinity on days 22 to 24 as well. Female elephant seals are usually in estrus at the end of the lactation period, the last three to five days before they leave the rookery (Le Boeuf et al., 1972). Although the male was not fully grown, he was sexually mature and capable of mating. The male in attendence, LPB, was identified as an immigrant to ANI since he did not bear a tag when he first arrived there. Virtually all pups born on ANI have been tagged each breeding season since 1962, the year after breed ing began on this island. T. C. Poulter and R. Jennings, Stanford Research Institute, tagged pups from 1962 to 1967 (Annual reports to the Division of Beaches and Parks, State of California) and Le Boeuf has done so from 1968 to the present. By comparison with known age animals, we estimate that LPB was six years old, a subadult male. Normally, males of this age class have a very low social rank in competition with adult males and do not take part in breeding (Le Boeuf and Peterson, 1969). LPB spent only a few days on the breeding beaches and was never observed copulating at ANI, although the beaches were observed throughout each day during the breeding season. He spent the majority of his time at ANI resting in nonbreeding areas with other males of similar age. LPB was seen on both ANI and FAR on the same day on two occasions, 9 and 20 February. On 20 February, he was recorded on ANI at 0800 hours and recorded on FAR shortly after the census was taken at 1320 hours. If the censuses were made just before departure and just after arrival, this male's travel speed was approximately 17 km per hour. Breeding occurred again during the 1973 season. In early December fiO"ht- , 0 ing was observed among four subadult males. Two cows arrived on 14 and 15 January and both of them gave birth between 17 and 19 January in the same location as the birth of the previous year. LPB arrived on FAR on 18 January after having been marked on ANIon 23 December 1972 and was ob served there off and on until as late as 14 January 1973. LPB, now approxi ~ately seven years old, appeared considerably more mature than in the pre VIOUS year. His neck shield was starting to develop and his nose starting to dangle, both signs of increasing maturity. He immediately dominated the May 1974 LE BOEUF ET AL.- POPULATJON OF ELEPHANT SEALS 375 ~ SUCAR-LOAF I S LA~ D / ~ Colden Ca t e l r 1d g~ 4'i k .

n SIII>RMAN ' 5 \<,'" Downloaded from https://academic.oup.com/jmammal/article/55/2/370/842919 by guest on 25 September 2021 IIAY ~b \ " P ~c vailin8 S .. ~ Us

P O I~T

MAINT OP aAY

~ " " . OC,

FIG. 2.- Southeast Farallon Island ( after Bowman, 1961). Elephant seals haulout at Low Arch, East Landing, North Landing, and Sea Lion Cove. The largest aggregations occur at Low Arch where as many as 100 animals were observed. other four younger males in residence. Two of these males bore tags (U C 1604 and UC 1614 ) identifying them as five year aIds who were born on ANI in 1968. During his stay, LPB alternated more or less daily between spending time with either of the two cows who were separated by approximately 50 meters (m ). Similarly, the number two male, UC 1614, spent time with the female vacated by LPB. On 7 February, UC 1614 was seen copulating with one of the cows, 21 days after she had given hirth. At the time, LPB was near the other female. LPB remained on FAR until 16 February (after the females had departed) and he returned to ANI for a day on 18 February and on 1 March. It is almost certain that LPB copulated with both fem ales but our observations were too infrequent to confirm this. Seventeen pups were born in 1974 but only five of them survived to weaning age (28 days) . Six breeding females bore tags indicating that they had been born on Aila Nuevo Island or San Miguel Island. LPB was the alpha male and he monopol ized breeding. Vol. 55, No.2 376 JOURNAL OF MAMMALOGY

140 ,I • 1973 120 • 1972

• 1971 • Downloaded from https://academic.oup.com/jmammal/article/55/2/370/842919 by guest on 25 September 2021 o 1970

[l 1969 • 100 • ' >- • . () 80 Z • W ::0 0 w 60 • 0:: • "-

20 .' JAN FEB MAR APR MAY JUN JUL MONTH FIG. 3.-The distribution of elephant seals on Southeast Farallon from Kovember 1969 to April 1973 (except for the period December 1969 through March 1970).

Census Data Elephant seals were observed at four different locations on FAR (Fig. 2). The largest concentrations (over 100 animals) were seen at Low Arch, a sandy beach measuring 15 by 25 ill, and all births occurred in this area or on the sand flat just above it. Groups of seals were also seen on the 10 by 15 m cobblestone beach near North Landing particularly in 1971, but in 1972 this beach was occupied by California sea lions, Zalophus califol'nianus. A few elephant seals were sighted occasionally on the cobblestone beach at East Landing and on the sandy beach at Sea Lion Cove. No elephant seals were seen on other islands in the Farallon group during visits on 30 October 1971 and 1 May 1972, times when peak numbers of seals were present on FAR. It is unlikely that seals ever hauled ant on these other islands since the shores of these islands are steep~sided and have nO beaches. The distribution of elephant seals censused on FAR from November 1969 to April 1973 is shown in Fig. 3. Several points are worth noting. L Only a few animals were observed during the breeding season, which May 1974 LE BOEUF ET AL.-POPULATION OF ELEPHANT SEALS 377

1200 1100 ... 1972 1000 .1971

900 Downloaded from https://academic.oup.com/jmammal/article/55/2/370/842919 by guest on 25 September 2021

800 >- '-' 700 w 600 ::J'" wCi 500 Ct: I.L 400

300

200

100

JAN FEB MAR APR MAy JUN JUL AUG SEp OCT NOV DEC MONTH FIG. 4.-The distribution of elephant seals On Ano Nuevo Island in 1971 and 1972.

extends from December to rnid~March (Le Boeuf, 1972) before breeding COm~ menced in 1972. 2. Another low census period occurred during July and August, the time when adult males and large subadult males haul out to molt at established rookeries, for example, ANI (Orr and Poulter, 1965). Sexually mature males molt at the same locations where they breed (Le Boeuf, unpublished obser vations). Because no breeding took place on FAR before the winter of 1972 and there were no breeding males in residence in previous years, the island was virtually vacant during the months of July and August. 3. Two peak periods are evident, one in the spring (April and May) and the other in the autumn (October and November). The spring peak was higher than the autumn peak in 1970 but the situation was reversed in 1971 and 1972. Information about the individuals seen during these two periods will be pre sented later. 4. The seal population increased over the period of observation. The last census taken on 5 May 1973 showed 189 animals present. 5. A comparison of the seasonal distribution of animals on FAR with that of the ANI breeding colony reveals similarities and differences (Fig. 4). The most obvious difference lies in the distribution during the breeding season. Otherwise, there was a spring and autumn peak on both islands. The highest census on ANI was recorded during the spring, a result found in previolls years Vol. 55, No. 32 378 JOURNAL OF MAMMALOGY

TABLE 2._Proportion and percentage of tagged animals sighted on Southeast Faral/on from September 1968 to June 1972 to number tagged.

Adult Subadult and Tagging location Pups and weaners Juveniles females adult males Totals Downloaded from https://academic.oup.com/jmammal/article/55/2/370/842919 by guest on 25 September 2021 Ano Nuevo Island plastic 42/750:::: 5.6 0/10:::: 0 0/49:::: 0 1/215:::: .5 43/1024 :::: 4.2 monel* 14/330:::: 4.2 O~O 0/5~ 0 1/59:::: 1.7 15/394:::: 3.8 San Miguel Island plastic 70/2044 :::: 3.4 0/73 =0 0/10 = 0 0/44 =0 70/2180 :::: 3.2 monel* 13/1099 :::: 1.2 0/11:::: 0 0/7:::: 0 0/28:::: 0 13/1145:::: 1.1 San Nicolas Island plastic 13/575:::: 2.3 0/6 =0 0/5 =0 0/15:::: 0 13/601:::: 2.2 monel'" 1/145:::: 0.7 0/5 =0 0/6 ~ 0 0/15 ~ 0 1/171:::: 0.6

~ Some animals were marked with both a plastic and II monel tag. Sightjngs c.f these animals arc recorded as pla~tic returns only.

on ANI going back to 1963 (Orr and Poulter, 1965; Le Boeuf, unpublished observations). In two of three years, the highest census on FAR was during the autumn. Additionally, there were animals in age categories on ANI during each peak which were not found on FAR. Virtually all animals seen on FAR were less than five years old. In addition to similarly aged animals being seen on ANI, older animals were also present at various times in the year. Subadult males were present from early November through the end of March. They reappeared to molt in early June and remained until about the end of August. Adult males were present from early December until mid· March and molted during July and August, Adult females hauled out to breed in mid-December and de parted in mid-:March, Some females returned to ANI beginning in late March and remained through the month of April. They were not seen at other times of the year. Further comparisons between FAR and ANI will be restricted to tagged animals,

Tagging Data One hundred and fifty-five tagged elephant seals were sighted on FAR dur ing the interval 1 September 1968 to 10 June 1972. These rehlrns yield infor mation regarding the sex, age, origin, length of stay and movements of the "Fal'allon seals." Island of origin.-Seals that were born and tagged on three rookeries were Sighted on FAR (Table 2). These islands and the distances between them and the Farallones to the north are ANI (89 km), SMI (497 km), and SNI (617 km). Although 2009 elephant seals were tagged on Isla de Guadalupe and 398 seals were tagged on Isla Cedros and Islas San Benito, none of these animals was ever sighted on FAR. These islands lie over 1000 km south of the Farallones. May 1974 LE BOEUF ET AL.-POPULATION OF ELEPHANT SEALS 379

Of the tagged animals sighted on FAR, 54 per cent were marked as pups on SMI, the largest of the three populations involved, 37 per cent of the recov eries were marked as pups at ANI, and the remaining nine per cent were born on SNL If the number of animals sighted is considered in relation to the num ber tagged at each site, the return is related to proximity to the Farallones. Downloaded from https://academic.oup.com/jmammal/article/55/2/370/842919 by guest on 25 September 2021 The closest island, ANI, gave the highest return rate and the most distant is land, SNI, exhibited the lowest return rate. Islands further south yielded no returns. For the three rookeries listed in Table 2, it is dear that the young of the year dispersed northward and utilized FAR as a hauling out place. Adults of both sexes and juveniles did not show this pattern until after breed ing began in 1972. Plastic tags yielded a higher return rate than monel tags (3.3 vs. 1.7 per cent) but the relative return rate of the two types is somewhat confounded by the fact that some animals were marked with both tags. These double tagged individuals were usually observed because of the more visible plastic tag. Hence, only the plastic tag was recorded as a return in Table 2. The greater visibility of the plastic tags made them superior to the monel tags dur ing the short period covered in this study. Age.-Because most elephant seal pups are born during the interval 1 Jan uary to 10 February, age was estimated by assuming that all pups were born on 20 January, the peak and median day of the 40-day pupping period. The age estimate is thus correct to within plus or minus 20 days. Tagging was done in February or March. The age category best represented during the months surrounding the au tumn peak (September through December) was young born the previous January. Out of 152 tagged animals observed during this period, 71 per cent were in their first year that is 8 to 11 months old, 23 per cent were in their second year, four per cent were in their third year, and two per cent were in their fourth year. None of the untagged animals observed at this time prior to 1973 appeared to be older than 4 years of age. Young of the year did not arrive on FAR early enough for the spring peak (April and May). This is in line with the previous finding that most pups remain at their birthplace until the end of April (Le Boeuf, 1972). The majority of the animals in resi dence during the spring peak were 15 to 16 months old (73 per cent); the remainder were 27 to 28 months old (20 per cent) and 39 to 40 months old (six per cent). Only 10 tagged animals were sighted during the months of January, February, March, June, July, and August. These were simply devi ants arriving earlier or later than the majority of animals. Thus, it is clear that the animals observed on FAR before breeding began were juveniles less than five years of age. Length of stay.-Estimating the average number of days that young seals were observed on FAR is hazardous for several reasons. First of all, failure to observe an individual did not necessarily mean he was not in the area. Be- 380 JOURNAL OF MAMMALOGY Vol. 55, No.2

T 3 Interisland movements of seals sighted on Southeast Faralloll The two short ver_ ti:~;Cl~n;~a'h yea' >ep""'" tloe ,p,ing and autumn peak in cen"""S ~'~a;,0Z::,wg:l:f ;;dc is used: "= time of birth; F = Farallon Is.; A = Aiia Nuevo Is;

Birthplace Tag number 1969 1970 1971 1972 Downloaded from https://academic.oup.com/jmammal/article/55/2/370/842919 by guest on 25 September 2021 San Nicolas Island 383 , F A 463 • A F 825 , F A , Ana Nuevo Island 95 F A 192 • F A A 223 • F F A 255 • , F F A 419 , F A F A 444 , F A 517 , A F 566 F A A 581 • A F 866 , A F 954 • F A 1060 , F F A 1061 , A F San Miguel Island 278 " A F 309 • A F 343 • F A A 445 ,• F A A 487 A F 511 , F A 518 • S 'jA F 724 • A F A 783 , AA A F F F 817 • , F S 833 , F A 848 F F A 927 • F 1271 • F A A 2127 A F 2141 F A 2149 F A F A cause these animals normally group together in tight pods, tags were often hidden from view. Some seals were in the water and could not be identified when the censuses were taken. Lastly, censuses were missed on occasion. Because of these problems, we will not attempt a detailed quantitative anal ysis. An accurate account of the pattem and length of stay of seals on a haul- May 1974 LE BOEUF ET AL.-POPULATION OF ELEPHANT SEALS 381

TABLE 4.-Sex ratio of elephant seals tagged as pups and sighted on Southeast Farallan.

Birthplace Males Females Sex ratio Ana Nuevo Island 28 18 1.56: 1 2.78*

San Miguel Island 36 29 1.24: 1 1.61 Downloaded from https://academic.oup.com/jmammal/article/55/2/370/842919 by guest on 25 September 2021 San Nicolas Island 7 6 1.17:1 Overall 71 53 1.34:1 3.16*

*P<.05 ing out place would require a separate study utilizing more obvious marks on individuals than the tags we employed. The most frequent visits to FAR, 55 per cent of those recorded, were of short duration lasting one to five days. Less than 10 per cent of the animals were seen in the area for more than a month. Nearly all visits of five days or more were made during either the spring or autumn. Long visits, when the animals were thought to be in the area for 20 days or more, were somewhat more frequent in the spring, when most juveniles molt, than in the autumn. The following profiles further illustrate the typical visit patterns of these young seals: Green tag 223 (male born on ANI in January 1969; sighted on FAR on 12 April, 4 and 14 to 17 May, 20 to 21, 23, and 25 September, 5 and 21 October 1970; sighted on ANI in April 1971); Yellow 1335 (male born on SMI in January 1970; sighted on FAR on 12, 16, 18 to 23, and 25 November 1970, and 11 to 12, and 26 to 28 October 1971); Red 276 (female born on SNI in January 1969; sighted on FAR on 4 to 6 February 15 to 16,25, and 27 May, 12 and 15 to 18 July, 13 September, 6 and 9 to 10 October, 14 to 15 December 1971). Sightings at other locations.-Thirty-three of the 155 tagged animals sighted on FAR were also observed elsewhere The approximate time and location of these multiple sightings is shown in Table 3. Evidently there is considerable back and forth movement of juveniles between ANI and FAR. Indeed, fifteen of seventeen pups tagged on SMI were observed on both ANI and FAR. Some animals were seen on both islands during the same peak period. Only one ani· mal, Yellow 817, was known to move back and forth between SMI and FAR. Sex differences.-Despite an equal sex ratio at birth and at the time of tag ging (Le Boeuf et al., 1972), more males than females were sighted on FAR (Table 4). The difference is significant only for recoveries from ANI. There were no sex differences in the spring haulout when all recoveries were grouped; however, there were significantly more males than females pres· ent during the fall period (1.25:1, z =2.07, P < .05, N = 81). All z values in this paper are corrected for continuity (Hardyck and Petrinovich, 1969). This was due to the sex difference in animals from SMI (1.47: 1, Z = 2.29, P < .05, N = 42). Recoveries from the other two islands during the fall period re· vealed no significant sex difference. Sex differences with age were evaluated by recording the sex of animah 382 JOURNAL OF MAMMALOCY Vol. 55, No.2 present each month of the year. The sex ratio of seals was as follows: first year, 1.25:1 (N ~ 90); second year, L1.13 (N ~ 98); third year, 2.38:1 (N ~ 27); fourth year, 2.0:1 (N = 12). The differences were significant for the first (z ~ 2.00, P < .05) and third year (z ~ 4.05, P < .05). The sample size was too small in the fourth year to permit a statistical test. Downloaded from https://academic.oup.com/jmammal/article/55/2/370/842919 by guest on 25 September 2021

Comparisom with Afia Nuevo Island Tag 8ightings.-Tagged seals sighted on ANI originated from the same two southern rookeries as the animals seen on FAR. Two hundred and fifty-eight seals sighted on ANI were born on 8MI and SNI. The majority were from SMI (87.6 per cent). Except for three juveniles, all recovered animals were tagged as pups; no seal marked as an adult on a southern rookery (Table 2) was ever sighted on ANI. Furthermore, no sighting was made of animals tagged on other Baja California islands. Most of the recovered animals (228 as opposed to 30) bore plastic rather than monel tags. The tag return rate from SMI and SNI for both types of tags was better on ANI than on FAR as follows: plastic tags, 8.2 as compared with 3.0 per cent; monel tags, 2.3 as compared with 1.1 per cent, This differ ence was probably due to the pl'esence of more observers on ANI and more frequent monitoring. Age.-The distribution of tagged animals four years of age or less sighted on ANI was similar to that seen on FAR during each month of the year. For example as on FAR, most of the seals present during the fall months were in their first year (82 per cent of 302 animals observed), Eighty per cent of the animals observed in April and May were in their second year. Length of stay.-The length of time spent in the ANI area by seals tagged on 8MI and 8NI was of the same order as visits to FAR. Sex differences.-The only sex difference on FAR was that more males than females from ANI were sighted (Table 4). There were no sex differences in ANI-tagged and ANI-recovered animals (1:1.05, N ~ 119). The only sex dif ference evident on ANI was for animals tagged at SNI where the females re covered outnumbered the males (1: 1.60, z = 2.15, P < .05, N = 26). However, this finding is suspect because the number of seals recovered was small and we know of no reason why 8NI-tagged females should differ from the 8MI tagged females who were recovered no more frequently than males (1.04:1, N ~ 151). The distribution of sexes during the spring and autumn peaks was similar on ANI and FAR. On both islands, there were no sex differences during the spring haulout with respect to island of origin or when all islands were grouped, On both recovery islands, SMI-tagged males outnumbel'ed females during the autnmn peale (ANI ~ 1.50,1, z ~ 3.33, P < .05, N ~ 75). Also during the au tumn period, ANI-tagged, ANI-recovered females were more often sighted May 1974 LE BOEUF ET AL.-POPULATION OF ELEPHANT SEALS 383 than males (1: 1.35, z = 2.04, P < .05, N = 54). All other sex comparisons were not significant. Males in their first and third years outnumbered females of similaT age on FAR. The only sex difference with age found on ANI was with animals in their HI'St year; males from SMI were more frequent than females (1.24:1, Downloaded from https://academic.oup.com/jmammal/article/55/2/370/842919 by guest on 25 September 2021 z = 2.06, P < .05, N = 101) and females from SNI were more frequent than males (1,1.73, z ~ 2.71, P < .05, N ~ 30). An overall comparison On ANI, like the one which was done on FAR, resulted in no sex differences with age for animals that were 5 years of age or less.

DISCUSSION Because pinniped breeding colonies are established during a brief period in time, the initial process has been rarely observed or studied. With possibly a few exceptions (Radford et al., 1965; O'Gorman, 1961), new rookeries are not noticed until breeding has taken place for a few years (for example, Peter~ Son et al., 1968). Only then are studies of the development of the colony ini~ tiated. The present study is an exception and its value lies in demonstrating the pattern of island use by elephant seals prior to the initiation of breeding. Our data show that the annual distribution of juvenile elephant seals was no different on a hauling-out place than on a rookery. The major difference in the composition of animals on FAR and ANI was that older animals were present in winter and summer on the rookery and absent at this time on the hauling-out place until breeding began there in the winter of 1972. There was a dramatic increase in the number of animals on FAR during the 1973 breed~ ing season. Utilization of the hauling-out place by juveniles was not determined pri marily by availability of space. Juveniles did not haul out on FAR in winter and summer before breeding began even though the beaches were empty. More important factors detennining the rhythm of island use would seem to be molting and the availability of food in the vicinity. Most juveniles molt during the spring Or autumn haulout (Le Boeuf, unpublished observation). The precipitous rise and fall of the census curve in Figs. 3 and 4 correlates with fluctuations in the salinity and temperature of water and upwellings along the central California coast (Bolin and Abbott, 1963), fluctuations which reflect changes in phytoplankton (which peaks in June), and possibly elephant seal prey. Although there has -been much speculation concerning whether males or females arrive first when a new pinniped colony is formed there are little data bearing on this point. O'Gorman (1961) counted one adult female, her pup, and seven adult males in the new colony of South Shetland fur seals, Arto cephalus tropicalis (Gray), at Cape Shirreff in the South Shetlands. Peterson et al. (1968) counted 60 females, 40 pups, and only one male in the new colony 384 JOURNAL OF MA~n,.fALOGY Vol. 55, No.2

of Alaska fur seals, Callorhinus ursinus, on San ~11iguel Island, California. In both studies the sex of the first animals to arrive was not determined with cer tainty. On the contrary, our data on elephant seals show unequivocably that two adult females arrived 21 days before the first male was observed.

It is interesting that the only males in residence during the first two breed Downloaded from https://academic.oup.com/jmammal/article/55/2/370/842919 by guest on 25 September 2021 seasons on FAR were young subadults. A similar situation was found at San Nicolas Island around the time when it was first colonized in 1949 and Barth olomew (1952) remarked on the absence of adult males here and in other newly formed colonies. Males less than seven years old are rarely successful at mating on established rookeries (Le Boeuf, 1974).

ACKNOWLEDGMENTS We thank M. Bonnell, R. Gantt, and R. Whiting for field assistance, M. Whitt, D. Robin son, and Prof. C. Hubbs for transportation, the Bureau of Sports Fisheries and Wildlife for their cooperation, and J. Smail for comments on the manuscript and for raising funds to maintain the FaralJon Biological Station. This work was l:iupported in part by National Sci ence Foundation Grant GB-16321 to B. J. Le Boellf. This is contribution No. 48 of the Point Reyes Bird Observatory.

LITERATURE CITED BARTHOLO)o.lEW, G. A. 1952. Reproductive and social behavior of the northern elephant seal. Univ. California Publ. Zoo1., 47:369-472. BARTHOLOMEW, G. A., AND C. L. HUBBS. 1960. Population growth and seasonal move ments of the northern elephant seal, Mirounga angllstirostris. Mammalia, 24: 313-324. BOLIN, R. L., AND D. P. ABB01~. 1963. Studies on the marine climate and phytoplankton of the central coastal area of California, 1954-1960. California Coop. Ocean. Fish. Invest. Rept. 9:23--45. BOW~{AN, R. I. 1961. Latel:ipring observations on birds of South Farallon Island, Cal ifornia. Condor, 63 :410-416. HARDYCK, C. D., AND L. F. PETBINOVICH. 1969. Introduction to statistics for the be havioral sciences. \,y. B. Saunders Co., Philadelphia, Pennsylvania, 302 pp. LE BOEUF, B. J. 1971. The aggression of the breeding bulls. Natural History, 80:82-94. ---. 1972. Sexual behavior in the northern elephant seal Miro!lnga angustirostris. Behaviour, 41:1-26. ---. 1973. Male~male competition and reproductive success in elephant seals. Amer. Zool.,14:161-174. LE BOEUF, B. J., A:>:10 R. S. PETEHSO:>:1. 1969. Social status and mating activity in elephant seals. Science, 163:91-93. LE BOEUF, B. ]., R. J. WHITING, A:>:10 R. F. GANTT. 1972. Perinatal behavior of northern , elephant seal females and their young. Behaviour, 43:121-156. o Gom'fAN, F. A. 1961. Fur seals breeding in the Falkland Islands Dependencies Nature 192:914-916. . , , ORR, R. T., AN~ T .. C. POULTER. 1965. The pinniped population of Ana Nuevo Island, CahforllIa. Proc. California Acad. ScL, 32:377-404. PETEHSON, R. S., B. J. LE BOEUF, AND R. L. DELo,\'G. 1968. Fur seals from the Bering Sea breeding in California. Nature, 219:899-901. RADFORD, K. W., R. T. ORR, A}/D C. L. HUDBS. 1965. Reestablishment of the northern elephant seal (Mirounga angustirostris) off central California. Pmc. California Acad. Sci., 31:601-612. May 1974 LE BOEUF ET AL.-POPULATION OF ELEPHANT SEALS 385

RICE, D. W., K. W. KENYON, AND D. B. LLUeH. 1965. Pinniped population at Islas Guadalupe, San Benito, and Cedros, Baja California, in 1965. Trans. San Diego Soc. Nat. Hist., 14:73-84, SCAL'\.IMON, C. M. 1874. The marine mammals of the Northwestern coast of North Amer ica. John H. Camlany and Co., San Francisco, 319 pp. THORESEN, A. C. 1959. A biological evaluation of the Farallon Islands. Report to the Downloaded from https://academic.oup.com/jmammal/article/55/2/370/842919 by guest on 25 September 2021 Division of Wildlife, Fish and Wildlife Service, Pacific Region, Portland, Oregon, 28 pp. Crown College, University of California, Santa Cruz, 95064 and Point Reyes Bird Ob seroatory, Mesa Road, Bolinas, California 94924. Submitted 5 August 1972. Accepted 4 July 1973.