Aquatic Mammals 1997, 23.3, 127-133 Description of selected behaviours of humpback dolphins Sousa chinensis Leszek Karczmarski1, Meredith Thornton2 and Victor G. Cockcroft3 1Centre for Dolphin Studies, Port Elizabeth Museum, P.O. Box 13147, Humewood 6013, South Africa, c/o Department of Zoology, University of Port Elizabeth, P.O. Box 1600, Port Elizabeth 6000, South Africa 2*Port Elizabeth Museum, P.O. Box 13147, Humewood 6013, South Africa 3Centre for Dolphin Studies, Port Elizabeth Museum, P.O. Box 13147, Humewood 6013, South Africa Introduction The intentional beaching of humpback dolphins when feeding, which was reported to occur in Despite its apparent wide distribution in the Indo- tidal channels of the Bazaruto Archipelago, Pacific, the humpback dolphin (Sousa chinensis) Mozambique (Peddemors & Thompson, 1994) was does not appear to be abundant anywhere and neither seen in Algoa Bay (present study) nor remains little known (Klinowska, 1991; Reeves & previously recorded in Eastern Cape waters. The Leatherwood, 1994). The natural history of hump- topography of the environment could be the reason. back dolphins has recently been studied in detail in The Eastern Cape region lacks extensive, shallow, the Algoa Bay region on the south Eastern Cape tidally influenced bays, lagoons or estuaries which coast of South Africa (Karczmarski, 1996). This seem to facilitate the intentional beaching of both long term study included many hours of field humpback dolphins off the Mozambican coast observations and provided an ideal opportunity to and bottlenose dolphins (Tursiops truncatus) collect data on dolphin behaviour (Karczmarski & from several sites world-wide (e.g. Hoese, 1971; Cockcroft, submitted). This paper describes several Silber & Fertl, 1995). The apparent inter-individual behaviours of humpback dolphins observed in the co-operation reported by Peddemors & Thompson Algoa Bay region throughout the three year period (1994) seems unusual when compared to the feeding between May 1991 and May 1994 and discusses behaviour of humpback dolphins in Algoa Bay in their probable ecological determinants. this study or Plettenberg Bay (Saayman & Tayler, 1979) and could be opportunistic and attributed to Description of selected behaviours habitat differences. Feeding Two foraging/feeding strategies were used by Respiratory and aerial behaviour humpback dolphins in Algoa Bay. Generally, feed- Humpback dolphins display a fairly stereotyped ing groups were widely dispersed, with individuals surfacing–breathing pattern, with the rostrum rising moving in various directions without an obvious steeply above the water before the forehead breaks pattern. All large groups of humpback dolphins the surface. When the blowhole is exposed to air, fed in this manner, though smaller groups (c6 most of the body remains submerged with only a dolphins) remained in fairly close, although varied, small part of the upper back and the anterior proximity to each other (roughly 1–20 m) and kept portion of the dorsal hump above the water. It moving up and down along several hundred meters seems possible that this pattern of surfacing, which of coastline approximately 200–300 m offshore. is similar to that of baiji (Lipotes vexillifer) from Consequently, with only these two feeding strat- China (Renjun et al., 1986), results in a relatively egies seen, the observed behaviour of humpback long period of blowhole exposure. Hui (1989) dolphins appears to be less diverse than that of suggested that dolphins may control the blowhole- some other dolphin species (e.g. Würsig, 1986; exposure time by controlling the emergence angle Belkovich et al., 1991). As seen from above the of porpoising leaps; increasing emergence angle water, cooperation between individuals, if any, increases exposure time with little modification of seems to be limited. forward speed. Consequently, the surfacing pattern of humpback dolphins may possibly benefit the *Present address: Sea Fisheries Research Institute, Private efficiency of ventilation before the next dive (Dral & Bag X2, Rogge Bay 8012, South Africa. Verwey, 1977), which for this species (26.312.7 s), 1997 EEAM 128 L. Karczmarski et al. seems to be slightly longer than average dive times and several respirations, the animals performed of some other coastal dolphins (Renjun et al., 1986; another ventral contact of similar length. These Shane, 1990; Lockyer & Morris, 1987), but similar apparent copulatory episodes were repeated several to that of baiji (Renjun et al., 1986). times. The pattern varied little between the 11 Although humpback dolphins in the Algoa Bay sightings when the apparent courtship and mating region displayed versatile aerial behaviours, includ- was observed. ing porpoising leaps, vertical leaps, side leaps, On three occasions, several adults (3–6) were somersaults and backward somersaults (see also involved in courting-like behaviour at the same time Saayman & Tayler, 1979), aerial behaviour was and remained together, in one group, for almost generally infrequent. The most common were 20 min. The animals swam at high speed in a circle ‘quasi-leaps’ (Hui, 1989), with the snout entering of 5–8 m in diameter. At least some of the dolphins the water while the middle of the body was clearly displayed the spiral-like movement described earlier above the water, but the tail not yet emerged. and varied aerial behaviour. After this spiral and Vertical leaps, side leaps and somersaults were circular movement the group swam rapidly to typical of behaviour categorized as ‘socializing & another spot where this activity was again initiated. playing’. This is similar for humpback dolphins in This sequence was repeated several times (Fig. 1d) several other areas along the east African coast and was followed by apparent copulation consisting (Saayman & Tayler, 1979). of several ‘copulatory episodes’. Despite the number of individuals involved, never more than a Courtship and mating single pair was seen ‘mating’ at one time. However, The apparent courtship and mating of humpback the remaining part of the ‘courting group’ always dolphins was seen several times during prolonged remained in close proximity to the ‘mating’ pair bouts of social behaviour and occurred within and it is unknown if the consecutive ‘copulatory groups of various sizes (4–16 animals including all episodes’ involved the same, or different individ- age classes) and under varied water clarity (2.0– uals. Similar behaviours occasionally occurred 7.2 m). The animals involved in courtship and among juveniles, but seldom lasted more than mating-like behaviour were most often temporarily 2–3 min, with brief (2–3 s) ‘copulatory episodes’. isolated from other members of the group and This apparent sexual activity of humpback dol- the number involved varied between two and six phins was sporadically accompanied/interrupted by dolphins. bouts of unusual behaviour. In one instance, a Behaviour appearing to be courtship was usually dolphin lay on its back, motionless at the surface, initiated by vigorous movement of two individuals exposing its ventral surface and pectoral fins above side by side, with consistent, helical interchanging the water and appeared to be resting. To respirate, of their relative positions (Fig. 1a). The animals the animal rolled about its vertical axis, breathing frequently exposed almost half of their body above and rolling again to the inverted position. Sometime the water and displayed prolonged body contact. later (3–5 min) the animal would roll onto its side This rapid movement was occasionally interrupted and join the activity of others. On another occasion by an aerial display by one of the individuals a dolphin hung vertically in the water, head down- (quasi-leaps, side leaps or vertical leaps). This ward, the tail and part of the caudal peduncle spiral-like movement lasted less than a minute protruding above the water. The dolphin then (precise timing was not obtained) and was followed slowly sank beneath the surface. by high speed movement to another location, There are several similarities between the court- mostly less than 100 m distant, where the behaviour ship and mating behaviour of humpback dolphins was repeated. The entire sequence was usually reported here and that described previously (par- repeated several times. ticularly Saayman & Tayler, 1979). The courtship It was subsequent to these energetic behaviours observed in Algoa Bay, however, was more com- that the dolphins apparently initiated true mating. plex than that which Saayman & Tayler (1979) This is best illustrated in the description of one termed ‘precopulatory activity’, or that which instance where an individual (identified as female) Zbinden et al. (1977) referred to as ‘apparent rolled onto its side and, inverted almost up-side- mating display’. The vigorous, simultaneous court- down, swam slowly beneath the surface. This ship of several animals seen in Algoa Bay resembles dolphin was then approached by its partner, both that observed in the Persian Gulf (Pilleri & Gihr, individuals interlocked ventrally (Fig. 1b) and 1973–74) and in the Indus delta region (Roberts remained together for about 25–30 s, swimming et al., 1983). The ‘vertical rising’ out of the water by slowly and rolling beneath the surface (Fig. 1c). two ventrally interlocked animals, however, was not Although neither erection nor intromission were seen in Algoa Bay. Generally, the apparent court- seen, this particular behaviour appeared to be copu- ship and mating behaviour of humpback dolphins lation. After