ings, the attribution of these sounds to His- 

 triophoca is circumstantial since they are under- 

 water sounds from animals out of sight below the 

 surface. These sounds are unlike sounds attrib- 

 uted to any of the other animals known to inhabit 

 the area: gray whales (Asa-Dorian and Perkins 

 1967; Cummings et al. 1968; Fish et al. 1974), 

 walrus (Schevill et al. 1966; Ray and Watkins 

 1975), and the ringed seal and spotted seal 

 (Schevill et al. 1963; Stirling 1973; Ray pers. obs.). 

 The bearded seal, Erignathus barbatus, was seen 

 at times in low numbers during May 1967; some of 

 the recordings have a background that we recog- 

 nize as from Erignathus, but we eliminate it be- 

 cause: 1) the Histriophoca sounds are very differ- 

 ent from the Erignathus sounds heard at this 

 season (Ray et al. 1969); 2) in previous years when 

 only Erignathus was nearby, none of the His- 

 triophoca sounds was heard; 3) Histriophoca 

 sounds were heard in the presence of these seals 

 whether Erignathus were audible or not; and 4) 

 none of these sounds were heard unless His- 

 triophoca were observed in the area. 



The recordings were made in a variety of ice 

 conditions and ice is known to produce sounds 

 underwater (Schevill 1966; Watkins and Ray pers. 

 obs. ). The seal sounds did not vary with the ice and 

 did not match the kinds of sound we associate with 



ice. 



Underwater Sounds 

 Two types of underwater sounds were heard in 

 5- 



the presence of Histriophoca: a relatively intense 

 prolonged downward sweep in frequency and a 

 broadband puffing sound. These calls were heard 

 sporadically, with no obvious pattern to repeated 

 sounds nor to any answering calls. Nearby seals 

 could be heard at least once in 2 min and often 

 there were enough seals in audible range so that 

 when calling was most frequent we recorded 3 to 5 

 calls in 10 s. Since the seals were out of sight and 

 probably underwater during the recordings, we 

 could not correlate the sounds with behavior. 



The sweep sound (Figure 1) varied in frequency 

 from 7 to 0.1 kHz in downward sweeps of 2 to 5 kHz 

 each. Of the 120 sweep sounds measured, all but 

 one could be separated into three length categories 

 (Figure 2), each with somewhat different starting 

 and ending fundamental frequencies: 



Short sweeps, 1 s or less, sweeping from 2000-1750 Hz 



to 300 Hz. 

 Medium sweeps, 1.3 to 1.8 s, sweeping from 5300-2000 Hz 



to 100 Hz. 

 Long sweeps, 4 to 4.7 s, sweeping from 7100-3500 Hz 



to 2000 Hz. 



Short sweeps were common in the background 

 ambient sound, but only a few were heard from 

 nearby seals (16 measured). Midlength sweeps 

 were the ones most often heard from local seals (84 

 measured), and some of these began with a short 

 segment of sound at constant frequency for the 

 first 0.1 to 0.2 s before beginning the downward 

 frequency sweep (Figure 1). The long sweeps were 

 not particularly abundant but were conspicuous 



4- 



$ 3 



Seconds 



FIGURE 1. — The midlength sweep sound of Histriophoca often has a short portion of constant frequency before it begins to sweep 

 downward in frequency. Analyzing filter bandwidth was 45 Hz. Analyses of short and long sweeps (not figured separately) were 

 generally similar in character to the midlength sweep. 



451 



