QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

First published online August 17, 2006
Journal of Experimental Biology 209, 3301-3308 (2006)
Published by The Company of Biologists 2006
doi: 10.1242/jeb.02370

This Article Summary Full Text Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Ronacher, B. Articles by Wehner, R. Search for Related Content PubMed PubMed Citation Articles by Ronacher, B. Articles by Wehner, R.

Integrating two-dimensional paths: do desert ants process distance information in the absence of celestial compass cues?

B. Ronacher^1,*, E. Westwig^1, and R. Wehner²

¹ Department of Biology, Humboldt University, Invalidenstr. 43, D 10099 Berlin, Germany
² Department of Zoology, University of Zürich, Winterthurerstrasse 190, CH 8057 Zürich, Switzerland

View larger version (22K): [in a new window]   Fig. 1. `Open' controls, with sky view throughout the whole maze. (A,B) `Z'-shaped maze, with eastward (A) and westward (B) orientation of the middle leg. (C) Straight control (schematic top views of all channels). (A) Expected direction 315° (filled arrowhead), mean vectors, at 1 m, 2 m and 3 m radii: 319.9°/319.8°/324.2° (arrows); vector lengths 0.91/0.96/0.98; N=24, 23, 20. Mean vector not significantly different from expectation at 1 m and 2 m; at 3 m significantly different (P<0.01). (B) Expected direction 45°, mean vectors, 42.1°/47.9°/47.2°; vector lengths 0.98/0.97/0.98; N=23, 23, 23. Mean vector not significantly different from expectation at all radii. (C) Expected direction 0° (north), mean vectors, 3.1°/2.1°/6.8°; vector lengths 0.93/0.96/0.97; N=25, 20, 11. Mean vector not significantly different from expectation at all radii. Note that some data points obscure others with identical azimuth.

View larger version (24K): [in a new window]   Fig. 2. Maze with covered middle segment, same orientations and dimensions as in Fig. 1A,B. The Perspex cover excluding UV radiation extended by 15 cm into the N-S-channel segments on both sides. (A) Eastward orientation. Mean vectors, at 1 m, 2 m and 3 m radii: 345.8°/350.3°/350.4°; vector lengths 0.89/0.90/0.94; N=23, 23, 16. Mean vector not significantly different from 0° (open arrowhead) at all radii, but highly significantly different from 315° (filled arrowhead). (B) Westward orientation. Mean vectors 12.9°/15.0°/16.3°; vector length 0.91/0.92/0.95; N=22, 22, 18. Mean vector not significantly different from 0° at 1 m, but significantly different from 0° at 2 m and 3 m. P<0.05. At all radii highly significantly different from 45°. Mean vectors of A and B significantly different (Watson-Williams test with Bonferroni correction, P<0.01).

View larger version (24K): [in a new window]   Fig. 3. Maze with covered middle segment and open bends. The Perspex did not cover the bends, allowing a view of the sky up to 15 cm into the middle segment on both sides. (A) Eastward orientation. Conventions as in Fig. 2: mean vectors 347.2°/351.1°/344.0°; at all radii not significantly different from 355° (open arrowhead), but highly significantly different from 315° (P<0.01); vector lengths 0.86/0.94/0.97; N=25, 25, 15. Note that in this maze a -5° or +5° deviation from a purely northern course is now expected, even if the ants had ignored the covered middle segment, since an uncovered middle segment of 30 cm has to be taken into account for the calculation of the expected home vector. (B) 26.4°/27.1°/36.8°; 0.95/0.96/0.98; N=21, 21, 12. At all radii mean vectors significantly different both from 5° (open arrowhead) and from 45° (P<0.01), except at 3 m radius (not significantly different from 45°). Mean vectors of A and B significantly different (Watson-Williams test with Bonferroni correction, P<0.01 to 0.001).

View larger version (26K): [in a new window]   Fig. 4. Z-shaped maze in a different orientation (in 2005, first and third leg pointing to the east, middle to the south) (see Materials and methods). (A) Open control. Expected orientation 315° (filled arrowhead), mean vectors: 319.1°/316.8°/317.1°/312.0 (at 1 to 4 m radii); not significantly different from expected 315° at all radii; v.l. 0.97/0.98/0.98/0.98; N=22, 22, 22, 13. (B) Maze with covered middle segment. Mean vectors: 276.1°/273.8°/272.2°/270.2°; v.l. 0.95/0.96/0.97/0.97; N=25, 25, 25, 18. Mean vectors not significantly different from expected 270° (open arrowhead) at all radii.

View larger version (29K): [in a new window]   Fig. 5. Homing directions after training in open control maze, in relation to time of day. Expected values 45° and 315°. After mirroring the data of eastbound training they were not significantly different from those after westbound training (see Fig. 1A,B), and therefore the former data were mirrored for this plot. No larger deviations from 45° at some times of the day, nor a significant overall trend was visible, indicating that directional errors induced by the restricted sky view within the channels had been cancelled out due to the perpendicular orientation of segments in the `open' controls.