Cooperative Feeding Behavior of Crocodilians 1

In 1774, while traveling on the St. Johns River in Florida, U.S.A., in the vicinity of Lake Dexter, William Bartram observed that "... alligators were in such incredible numbers and so close together from shore to shore that it would have been easy to have walked across their heads had the animals been harmless" (Van Doren 1928: 118). That observation has been cited by later authors (Reese 1915: 8-9; King 1969: 33, 1972: 15-18; Florida Game and Fresh Water Fish Commission 1989) as evidence of the primordial numbers of American alligators, Alligator mississippiensis, prior to their decimation by hide hunters in the 1900's. Though Bartram clearly encountered numerous large alligators, many more than can be seen at the same locality today, his observations at Lake Dexter relate not to abundance but to a specific behavior, cooperative fishing. Compare Yamashita's (1991) photograph above of the behavior in Caiman yacare with Bartram's 1774 detailed description of the alligators arrayed across the mouth of Lake Dexter where it flows north into the St. Johns river:

"... I soon accounted for the prodigious assemblage of crocodiles [American alligators] at this place, which exceeded every thing of the kind I had ever heard of.

"How shall I express myself so as to convey an adequate idea of it to the reader, and at the same time avoid raising suspicions of my veracity? Should I say, that the river (in this place) from shore to shore, and perhaps near half a mile above and below me, appeared to be one solid bank of fish, of various kinds, pushing through this narrow pass of the St. Juan's [St. Johns River] into the little lake, on their return down the river, and that the alligators were in such incredible numbers, and so close together from shore to shore, that it would have been easy to have walked across on their heads, had the animals been harmless? What expressions can sufficiently declare the shocking scene that for some minutes continued, whilst this mighty army of fish were forcing the pass? During this attempt, thousands, I may say hundreds of thousands, of them were caught and swallowed by the devouring alligators. I have seen an alligator take up out of the water several great fish at a time, and just squeeze them betwixt his jaws, while the tails of the great trout [largemouth black bass] flapped about his eyes and lips, ere he swallowed them ... This scene continued at intervals during the night, as the fish came to the pass. After this sight, shocking and tremendous as it was, I found myself somewhat easier and more reconciled to my situation; being convinced that their extraordinary assemblage here was owing to the annual feast of fish; and that they were so well employed in their own element, that I had little occasion to fear their paying me a visit." [van Doren 1928: 118-119].

Bartram wrote that description of social fishing in Lake Dexter alligators long after he returned to Philadelphia. It appears in his 'TRAVELS ... ' which was published in 1791. Bartram documented his observations in daily journals which have been lost. However, in 1773-1774, shortly after he observed the cooperative feeding behavior, Bartram used his journals to compose two reports to John Fothergill in England. In those reports, Bartram described the same incident somewhat differently:

"It is scarcely credible what an immence number of Fish these monsters destroy, especially at these passes, the River being here[,] as I observed before[,] very Narow. The Trout [largemouth black bass] who pass here in their way to & from the numerous lakes & endless Lagoons & Marshes towards the head of this Vast River, where they go to spawn. The Alegator post themselves forming a line across whe[re] we see them opening their voracious Jaws into which the fish are intrap't. They heave their heads and upper part of their body upright[,] opening their throats to swallow them, & I have seen them with two or three great Trout in their mouth at a time[,] choping them up[,] the fishes tail hanging out." [Bartram 1943: 152].

Bartram described yet another incident in which fish were devoured by a group of alligators in the sinkhole that drains the eastern end of the Alachua Savanna (= Paynes Prairie):

"In and about the Great Sink, are to be seen incredible numbers of crocodiles [alligators] ... they are so abundant, that, if permitted by them, I could walk over any part of the bason and the river upon their heads, which slowly float and turn about like knotty chunks or logs of wood, except when they plunge or shoot forward to beat off their associates, pressing too close to each other, or taking up fish, which continually crowd in upon them from the river and creeks draining from the savanna, especially the great trout [largemouth black bass], mudfish, catfish, and various species of bream ...."[van Doren 1928: 178].

Water pouring into the sinkhole from the surrounding savanna is the ideal situation in which alligators might line-up across the creek mouths to capture fishes that ride the currents into the sink. However, alligators plunging forth to beat off others that got too close suggests that these alligators were not feeding cooperatively, but simply opportunistically catching whatever fish they could find in the sink while fending off competitors.

However, apart from Bartram's lucid account of his 1774 observation, this behavior in alligators has been reported by few other authors. Like many later authors, Neill (1971: 25) questioned the accuracy of Bartram's observations with the following comment, "As this episode took place by night, the scientist must ask by what light it was observed in such exciting detail. The report must have been constructed in large part from the imaginative interpretation of night sounds." However, a second instance of cooperative fishing in alligators was witnessed by Allen Chesser in 1890 in Buzzard Roost Lake in Okefenokee Swamp, Georgia, and reported by Francis Harper (1926: 417-418, see also Harper 1930: 57):

"Now I want ter tell yer erbout a sight er Alligators I seed one time ... That occurred at the Buzzard Roost Lake ... Hyere the lake [at this point Chesser scratched a diagram on the ground], an' hyere's a little run [a channel] goes out erbout thirty yards broad. An' right hyere at the en' is a little round lake. These Alligators, I suppose, they must 'a' driv all the fish out er this big lake, an' down this road [the outlet]. It 'us in between daylight and sunrise. I heerd the racket before I got there ... The Alligators cared nothin' fer us. There must 'a' ben three hundred uv 'em. They'd ketch fish that long [indicating about a foot and a half]. Ef they'd ketch a perch, yer'd hear 'im flutterin' in their mouth—thrr, jest like a-that. An' the funny part, there'd be a Gator sometimes that high [indicating about a yard] out of er the water—an' ernother un on ter 'is tail. He'd think it 'us a fish.

"When he'd ketch a fish, jest stick 'is haid up thataway, an' ernother un tryin' ter get it away frum 'im. They'd pay no attention ter us. We stayed there till the sun wuz erbout an hour high.

"We fell ter shootin', an' it wuz either fourteen er sixteen we killed before they took any notice a-tall. An' when they did take a notion ter get away, there wuz a sight ter look at—when they commenced smellin' the blood. They started down that road [the outlet]. They wuz that thick, I could 'a' walked down that road on Gator haids."

Earlier, John James Audubon (1827) reported alligators feeding together in Louisiana, but it is not clear from his description whether it was an instance of cooperative fishing (as suggested by the alligators lying close together in the inflowing currents) or simply an instance in which alligators opportunistically take advantage of fish that have been trapped in lakes and ponds as the water recedes during the summer (as suggested by the dying fish and falling water level):

"When alligators are fishing, the flapping of their tails about the water may be heard at half a mile ... There, at a sight, hundreds of alligators are seen dispersed over the lake, their head, and the upper part of the body, floating like a log ... It is then that you see and hear the alligator at his work ... You see them lying close together. The fish that are already dying by thousands, through insufferable heat and stench of the water, and the wounds of the different winged enemies [herons, egrets, ibis] constantly in pursuit of them, resort to the Alligator's Hole to receive refreshment, with a hope of finding security also, and follow down the little currents flowing through the connecting sluices; but no! for, as the water recedes in the lake, they are here confined. The alligators thrash them and devour them ... You plainly see tails of the alligators moving to and fro, splashing, and now and then, when missing a fish, throwing it up in the air."

Surprisingly, despite the great body of research that has focused on alligators in the last 40 years, cooperative fishing in this species appears not to have been reported by later authors. It has been documented in a number of other crocodilians. Pooley and Gans (1976) describe cooperative feeding in the Nile crocodile, Crocodylus niloticus:

"Another example of cooperation may be seen in the early spring, when rivers rise and the water flows into channels leading to pan, or natural depressions, along the river. Subadult crocodiles often form a semicircle where a channel enters a pan, facing the inrushing water and snapping up the fish that emerge from the river. Each crocodile stays in place and there is no fighting over prey. Any shift in position, of course, would leave a gap in the crocodiles' ranks through which the fish could escape, so that what might be a momentary advantage for one crocodile would be a net loss for the group."

The behavior in Nile crocodiles was later described in greater detail by Pooley (1989: 88-90):

"In Lake St. Lucia, Natal, South Africa, there are annual migrations of shoalfish out of and into the lake from the Indian Ocean either to spawn or to feed. Species include kob, spotted grunter but more importantly, striped mullet (Mugil cephalus). The annual movement of the mullet shoals is fairly constant, and between mid-April and mid-May each year large numbers of crocodiles move down from northern and open stretches of the lake in response to the fish shoaling; others move up from river systems to the south.

"They congregate in an area known as the Narrows, a channel less than 500 meters (550 yards) in width. Numbers peak during May but decrease rapidly thereafter. Examples of cooperative feeding can be observed with several crocodiles spreading out in a semicircular or line formation, which blocks the passage of the fish. Each crocodile maintains its place in line and snaps at approaching fish. There is no fighting over prey; shifting position and leaving a gap in the ranks would lessen the chances of successful prey capture.

"In other Zululand rivers similar behavior may be seen in summer when rivers flood and water spills into channels leading to natural pans. The crocodiles form a barrier where a channel enters the pan, facing the inrushing water and snapping up river fishes such as bream (genus Talapia) and catfish."

Schaller and Crawshaw (1982) found similar behavior in Caiman crocodilus in Brazil:

"Adjoining ponds sometimes had connecting channels, and ... Usually only one or two caiman occupied a particular passage, but once, when heavy rain created a wide riffle between ponds, 7-15 caiman fished in it all day ... Pooley and Gans (1976) noted that subadult African crocodiles 'often form a semicircle where a channel enters a pan, facing the inrushing water and snapping up the fish that emerge from the river.' The authors considered this an example of cooperative hunting. We observed similar behavior when caiman gathered at road culverts ... Such behavior is a form of proto-cooperation ... rather than cooperation in the sense of a joint action in the performance of a defined task; any help fishing crocodilians gave each other seemed inadvertent."

Thorbjarnarson (1991) found it in Caiman crocodilus in Venezuela:

"In shallow, moving water caiman would use two techniques for catching prey. Both of these feeding behaviors were seen principally at night when it was difficult to make extensive observations without disturbing the animals. The first technique was to orient the body parallel to the flow of water and capture prey by making rapid sideswipes. This behavior was observed several times, including once when seven caiman were oriented in two rows facing into the current (31 December 1984), with their mouths slightly opened and their heads elevated in the water."

Yamashita (1991) photographed this behavior in Caiman yacare and summarized it as:

"Many social interactions occur in shallow open ponds during the dry season when very large concentrations of this species congregate in the limited habitat. Among these are social fishing behavior shown in the photograph."

While American alligators are plentiful throughout their range, the relative scarcity of easily seen, multitudinous, dense populations has contributed to the lack of reports of cooperative feeding in this species in recent years. As reported above, Schaller and Crawshaw (1982) observed cooperative feeding in Caiman crocodilus gathered at road culverts. However, a few crocodilians lined up facing into the narrow current pouring out of a culvert lacks the visual impact of large numbers of them arrayed side-by-side in rows across a broad channel as described in most reports of cooperative feeding. As a consequence, cooperative feeding around culverts has tended to go unnoticed and unreported. Small numbers of American alligators have been observed lined up and cooperatively feeding in the flow from culverts in north Florida (Kent Vliet, personal observation in Paynes Prairie State Preserve) and in south Florida (Wayne King, personal observation in Everglades National Park). It is a sight familiar to tourists, but one that biologists have tended to overlook.

Gans (1989) asked the rhetorical question, "The few observations on food acquisition by crocodilians are mostly serendipitous ... Do the few reports on cooperation represent accidental events or part of a more general phenomenon?" Cooperative feeding is quite different from the non-cooperative, feeding frenzy that is seen when groups of wild crocodilians devour fish trapped in drying pools or captive ones are fed at a single location in a farm or exhibit enclosure. In such free-for-all situations, the animals fight for food, injuries may occur, and smaller individuals often are excluded until the dominant animals are sated and move away. By contrast, all the reports of cooperative fishing demonstrate that:

It occurs where concentrations of fish are moving with the current flowing through a channel into or out of a body of water;
the crocodilians face into the current and arrange themselves side-by-side in a row across the flow of water;
the feeding crocodilians may be sufficiently close together to give the impression that a person could walk across the channel on their heads or backs;
when one crocodilian moves out of position, another takes it place;
while fighting over prey may occur it is infrequent; and
the intensity of feeding is so great that initially the crocodilians may pay little attention to humans and other disturbances.

Though reports of cooperative feeding are scarce, the shared ecological and behavioral components of the above cited, independently recorded, observations on American alligators, Nile crocodiles, spectacled and Yacare caiman confirm that cooperative feeding in crocodilians is not an accidental event.