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

First published online November 4, 2005
Journal of Experimental Biology 208, 4223-4230 (2005)
Published by The Company of Biologists 2005
doi: 10.1242/jeb.01905

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 HighWire Citing Articles via Google Scholar Google Scholar Articles by Wolf, H. Articles by Wehner, R. Search for Related Content PubMed PubMed Citation Articles by Wolf, H. Articles by Wehner, R.

Desert ants compensate for navigation uncertainty

Harald Wolf^1,* and Rüdiger Wehner²

¹ Department of Neurobiology, University of Ulm, D-89069 Ulm, Germany
² Institute of Zoology, University of Zurich, CH-8057 Zurich, Switzerland

View larger version (18K): [in a new window]   Fig. 1. `Goal expansion' and `error compensation' strategies. Feeders (F) were surrounded by target circles painted on the desert floor, as shown for three nest-feeder distances D. The target circles were used for recording the ants' downwind approach distances d. The ants' approach towards the feeder may be governed by two alternate strategies: expectation according to the `error compensation strategy' (solid line) and expectations according to the `goal expansion strategy' (dashed lines; for details see text). Downwind angle is indicated for comparison with d; at 20 m nest-feeder distance, a schematic distribution of approach distances is illustrated, typical of, although narrower than, those recorded in the present experiments (compare Fig. 4).

View larger version (17K): [in a new window]   Fig. 2. Relationship between downwind approach distance d and wind speed. (A) The distribution of wind speeds in the desert near Maharès (bin width 0.5 m s-1; recordings (n=1961) taken between ca. 08:00 h and 16:00 h, between 19 August and 5 September 2002, and 22 June and 3 July 2004). (B,C) The relationship between wind speed (bin width 0.1 m s-1) and downwind approach distance d. In B, data from all experiments are pooled; in C only data recorded for the 5 mnest-feeder distance are shown. (D) The pooled data from all experiments, as in B, but the individual values were corrected for the dependency of d on nest-feeder distance D, effectively eliminating any bias introduced by this dependency into the relationship between wind speed and d. Values are means ± S.D. (D), +1 S.D. (B,C).

View larger version (15K): [in a new window]   Fig. 3. Relationship between downwind approach distance d, and wind direction. (A) The distribution of wind directions in the desert near Maharès (bin width 15°, east is at 90°; recording times as in Fig. 2A; wind directions above 180° were not observed during the present experiments). (B) The relationship between wind direction and d; values are means ± 1 S.D. (C) Four approaches of an individually marked ant to a feeding site are superimposed (actual feeder was just 30 mm in diameter; its location is indicated by a 67 cm circle centred on the feeding site). As indicated by arrows, wind directions were slightly different from one approach to the next. The ant's final approaches were always against the wind. The broken circle centred on the feeder has a radius of 1.75 m.

View larger version (13K): [in a new window]   Fig. 4. Distribution of downwind approach distances d, for different nest-feeder distances D. The histograms illustrate the distributions recorded at the nest-feeder distances given above the bin peaks (bin widths 0.5 m). Different histograms are distinguished by different shadings. For the different nest-feeder distances D, the numbers of recordings (n) and ant individuals (N) were as follows: D=5 m, n=747, N>21; D=10 m, n=420, N>24; D=20 m, n=668, N>29; D=40 m, n=127, N>8; D=50 m, n=41, N>11; D=60 m, n=165, N>10; D=75 m, n=6, N=3. Half widths of the distributions correspond to angles of 8.8° for D=5 m, 5.8° for D=10 m, 3.8° for D=20 m, 5.2° for D=40 m, 2.0° for D=50 m and 2.4° for D=60 m.

View larger version (16K): [in a new window]   Fig. 5. (A) Relationship between downwind approach distance d and nest-feeder distance D. Same data set as in Fig. 4. Dotted line indicates the best-fit regression (d=0.4D+0.56), thin lines mark 95% confidence intervals. Measurements for each individual were pooled before calculating means, S.D. and regression line. (B) Relationship between the scatter of the downwind approach distance d and the nest-feeder distance D. Same data set as in A and Fig. 4. The dotted line indicates a trend line (scatterd=0.008D+0.443, values in m; a regression line was not constructed since the data points were not normally distributed, because the scatter was calculated as absolute values of the difference between d and the mean of d). 25% and 75% percentiles are given. The values for 40 m nest-feeder distance showed unusually high scatter, as a result of the experimental conditions; see text.

View larger version (19K): [in a new window]   Fig. 6. Schematic illustration of the significance of the ants' `error compensation strategy'. The goal (feeder) is surrounded by an uncertainty range (grey area). This uncertainty range has a directional (angular) and a distance (linear) component. For details and rationale see text.