Falling canopy ants glide home

Steve Yanoviak tosses ants from very high places: tropical forest canopy trees. In the 10 February, 2005 issue of the journal, Nature, Yanoviak, ant biologist, Mike Kaspari, and biomechanics expert, Robert Dudley, publish an amazing observation: canopy ant workers (Cephalotes atratus L) jettisoned from branches 30 m above the ground, glide backwards to the trunk of the same tree with incredible accuracy. This is the first published account of directed gliding in wingless insects.

Eighty-five percent of falling C. atratus workers glide back to their home tree. Marked ants often came right back to the branch where they started within ten minutes of falling or being dropped off! “I first noticed directed descent behavior on BCI [the Smithsonian’s Barro Colorado Island field station in Panama] in 1998 while working on a canopy ant project with Mike Kaspari. Some spiny C. atratus workers got stuck in my hand while I was sitting in a tree crown. When I brushed them off, they appeared to glide rather than fall haphazardly,” Yanoviak recalls.

“Early on, when Steve was dropping ants from the radio tower on BCI I got really excited because I could see their very clear ’J’ trajectory.” explains Kaspari, zoology professor at the University of Oklahoma and Research Associate at the Smithsonian. Kaspari introduced Steve to Robert Dudley, physiologist at the University of California, Berkeley and also a Smithsonian Research Associate.

Yanoviak’s caught the ants’ initial vertical drop, a quick swivel to orient the hind legs in the direction of the “trunk” and a steep, directed glide and landing on the vertical surface on video he recorded in Peru. Dudley’s high-speed video enabled the authors to quantify the velocity and angle of the glide trajectory as Yanoviak dropped ants from the balcony of the lab on Barro Colorado against a backdrop of white bedsheet.

Additional experiments at Yanoviak’s field sites in Costa Rica and near Iquitos, in Peru, where he currently works collecting mosquitos for the University of Texas Medical Branch and the University of Florida Medical Entomology Laboratory, showed that the ants visually orient toward tree trunks. In addition, small workers don’t fall as far as their larger counterparts before making contact with the tree. Yanoviak: “We still don’t understand exactly what mechanisms the ants use to change direction and to maintain a steady glide path through the air.”

Yanoviak and Kaspari also asked whether all canopy ants glide. “I was the guy that stayed on the ground,” Kaspari recounts, “while Steve dropped Paraponera [bullet ants that pack a nasty sting] down at me.” Canopy ants in two groups: the Cephalotini and the Pseudomyrmecinae glide; arboreal Ponerines and Dolichoderines did not. “The tropical forest canopy, home of half of the rainforest’s animal biodiversity, is a risky place. Canopy creatures live on the edge of a very long fall. For an ant, a 30 m fall to the forest floor is akin to me falling 3.5 miles. An ant falling to the forest floor enters a dark world of mold and decomposition, of predators and scavengers, where the return trip is through a convoluted jungle of dead, accumulated leaves. Gliding is definitely the way to go, and we won’t be surprised if we find more examples of this behavior among wingless canopy insects,” Kaspari concludes.