Horning and Mellish: Predation on Eumetopias /ubatus by Sommosus pacificus in the Gulf of Alaska 
307 
salmon shark, and the Pacific sleeper shark. All 3 spe- 
cies have been documented feeding within Alaskan wa- 
ters and the Bering Sea (Compagno, 1984; Benz et ah, 
2004; Martin, 2005). However, lamnid sharks, including 
the salmon shark (Smith and Rhodes, 1983; Anderson 
and Goldman, 2001; Goldman et ah, 2004; Bernal et ah, 
2005) and the white shark (McCosker, 1987; Goldman, 
1997), have been shown to maintain body-core and 
stomach temperatures consistently 6-15°C above ambi- 
ent temperatures even in subarctic regions with SSTs 
below 5°C. Therefore, the Pacific sleeper shark emerges 
as the only potential predator of Steller sea lions in the 
North Pacific and Bering Sea region whose body-core 
temperatures likely remain at mid-water values. 
The postmortem ingestion by Pacific sleeper sharks 
of tags released through the attack of a different pred- 
ator is another possible interpretation. However, a de- 
layed secondary ingestion at the surface still would 
expose LHX tags to SSTs and ambient light, and the 
probability of a simultaneous encounter of a sea lion 
and 2 distinct predator species at depth would be very 
low, in part because it would expose Pacific sleeper 
sharks to predation risk (Ford et ah, 2011). 
The Pacific sleeper shark is found in the North Pa- 
cific and the Bering Sea. Pacific sleeper sharks have 
been documented in shallow coastal waters and even in 
estuarine waters in the arctic regions of the North Pa- 
cific and at oceanic depths (>2700 m) in temperate lati- 
tudes (Compagno, 1984; Ebert et ah, 1987; Hulbert et 
ah, 2006; Sigler et al., 2006; Yano et ah, 2007). Although 
the very large (up to 7 m in total length [TL] ) Greenland 
shark ( Somniosus microcephalus ) may have the north- 
ernmost range documented for species of this genus, the 
range of the Pacific sleeper shark extends into the Ber- 
ing Sea and past the Arctic Circle (Bright, 1959; Orlov, 
1999; Orlov and Moiseev, 1998, 1999; Benz et al., 2004; 
Hulbert et al., 2006; Courtney and Sigler, 2007). 
Pacific sleeper sharks have been documented to 
reach adult lengths up to 4.4 m TL (Compagno, 1984; 
Mecklenburg et al., 2002; Yano et al., 2007), but es- 
timates from historic and contemporary photographic 
and video recordings indicate that they may reach 
lengths in excess of 7 m TL (Hart, 1973; Compagno, 
1984). The estimated TL at sexual maturity is 3.7 m 
for females and 4.0 m for males (on the basis of the 
length of the few sexually mature individuals ever 
caught: Ebert et al., 1987; Courtney and Sigler, 2007; 
Yano et al., 2007). To date, only 5 mature females have 
been reported from bycatch 2 . Pacific sleeper sharks are 
slow growing and long-lived (Ebert et al., 1987), sup- 
porting the notion that mature adults should reach 
sizes well beyond 4-5 m TL. 
2 Tribuzio, C. A., K. Echave, C. Rodgveller, J. Heifetz, and K. 
J. Goldman. 2010. Assessment of the sharks in the Bering 
Sea and Aleutian Islands. In stock assessment and fishery 
evaluation report for the groundfish resources of the Ber- 
ing Sea/Aleutian Islands regions, p. 1451-1500. [Available 
from North Pacific Fishery Management Council, 605 West 
4th Ave., Suite 306, Anchorage, AK 99501.] 
Pacific sleeper sharks have a broad range of diet 
that includes benthic and mid-water crustaceans, ceph- 
alopods, salmonids, gadids, flatfish, and marine mam- 
mals (Bright, 1959; Compagno, 1984; Yang and Page, 
1998; Orlov, 1999; Orlov and Moiseev, 1998, 1999; Si- 
gler et al., 2006; Yano et al., 2007). Genetically iden- 
tified stomach contents have included cetaceans and 
phocids; however, a histological analysis of ingested tis- 
sue indicated scavenging rather than live captures in 
the case of all cetacean and some phocid samples, and 
the evidence was inconclusive for some freshly ingested 
phocid tissue (Sigler et al., 2006). 
Pacific sleeper sharks may appear to be improbable 
predators of fast-swimming and agile prey as large as 
juvenile Steller sea lions, especially given the adapta- 
tion of the functional morphology of their feeding ap- 
paratus toward suction feeding. Evidence shows that 
Pacific sleeper sharks have removed large circular sec- 
tions, cookie-cutter style, from whale carcasses (Bar- 
rett-Lennard et al., 2011). Such an injury, if inflicted 
on the lower abdominal area (between diaphragm and 
pelvic girdle) of large juvenile sea lions, could result in 
LHX tag ingestion and would be lethal if it occurred 
ante mortem. 
Several studies of shark predation on marine mam- 
mals (Arnold, 1972; Cockroft et al., 1989; Heithaus, 
2001; Bertilsson-Friedman, 2006) have indicated that 
predation attempts often involve approaches from be- 
low and the rear and may focus on abdominal areas. 
However, the attack behavior of the Pacific sleeper 
shark has not been studied. Pacific sleeper sharks do 
appear capable of preying on fast-swimming salmonids 
and scombrids (Ebert et al., 1987; Sigler et al., 2006). 
Larger and older Greenland sharks have been shown to 
prey on live pinnipeds (Fisk et al., 2002; Leclerc et al., 
2012), although such events may be ambush attacks 
on seals sleeping at the ocean surface (Watanabe et 
al., 2012). Southern sleeper sharks (Somniosus ant- 
arcticus) have been reported to attack live southern 
elephant seals ( Mirounga leonina) (Van Den Hoff and 
Morrice, 2008). 
Because of the potential habitat overlap, the role 
of the Pacific sleeper shark as a potential predator of 
Steller sea lions was investigated. A modeling exercise 
by Frid et al. (2009) revealed sea lion dive behavior as 
avoidance of predation risk from transient killer whales 
and Pacific sleeper sharks. The model output was sup- 
ported by telemetered data of sea lion dive behavior 
in Prince William Sound. Yet, no sea lion remains 
were identified in the stomachs of 198 Pacific sleeper 
sharks caught by long-lining near sea lion rookeries 
(Sigler et al., 2006). However, these 198 sharks were 
collected during 2 sampling periods in May and August 
and within 1-12 km of sea lion rookeries or haulouts 
(Sigler et al., 2006). 
Only 3 of the 15 predation events that were de- 
tected in our study occurred within the 4-month period 
of May-August (Table 1), and none of these 3 events 
occurred within 20 km of rookeries or haulouts. Fur- 
