160
V. G. DETHIER
were being carried, were apolar. The basis of these conclusions was that
an interchange without rotation of pieces of honey-trail caused disturbing effects at both ends of the path, but with rotation the disturbance
was greater and increased in proportion to the length of pathway exchanged. Similar experiments with segments of larva-trail produced no
comparable effects. By way of explanation Brun supposed that ants
leaving the nest lay down nest odor in decreasing concentration as they
go farther away from the nest and on the return journey lay down honey
odor which decreases in concentration as they approach the nest; but
that ants carrying larvae dispense equal amounts of larva odor all the
way and this overcomes the comparatively weak nest odor.
Chauvin (1948) and many other workers were unable to confirm
Bethe's and Brun's conclusions. Since the original idea was based largely
upon experiments in which sections of the track were reversed with the
consequent confusion of the ant, Chauvin suggested that the confusion
actually arose, not from a reversal of a polarized trail, but from the fact
that in reversing a section of the trail it was impossible to make the ends
meet. The ant thus loses the trail at the junction. Carthy was able to
show by developing the trails with lycopodium powder that this was
indeed the case. (In Eciton reversal is without serious inconvenience;
the confusion is only momentary and trailing is soon resumed.) Development of the trail also showed that it consisted of discrete streaks from
4 mm to 8 cm long, shaped as to resemble a drop of fluid extruded,
and drawn out in the direction of running. Thus, although the trail does
possess polarization, there is no indication that this has any meaning
for the ant. As Carthy has shown for A. fuliginosus, pieces of trail are
laid down while the ant is homing, and the balance of the distance is
covered by random wandering. With increasing familiarity additional
portions of the trail are completed, and finally entire trails are deposited,
especially while the ant is carrying larvae. The final result is a tangle
of trails so interwoven that polarization based upon the shape of the
deposits would be impossible. Furthermore, if an ant is allowed to
stumble onto the middle of a trail, it shows no indication of being able
to distinguish the homeward direction from the outgoing direction. In
addition, the trail is not followed exactly in all its intricacies. In maze
studies with Formica subsericea
and F. exsectoides
Schneirla (1929)
claimed a slight but unimportant (as far as orientation is concerned)
end-to-end polarization.
MacGregor (1948) investigated trail-laying and trail-following in
another species of ant, Myrmica ruginodis, and obtained some information which does not conform exactly with the foregoing facts. Having
first determined that a forager moves in a series of loops and that a
V. G. DETHIER
were being carried, were apolar. The basis of these conclusions was that
an interchange without rotation of pieces of honey-trail caused disturbing effects at both ends of the path, but with rotation the disturbance
was greater and increased in proportion to the length of pathway exchanged. Similar experiments with segments of larva-trail produced no
comparable effects. By way of explanation Brun supposed that ants
leaving the nest lay down nest odor in decreasing concentration as they
go farther away from the nest and on the return journey lay down honey
odor which decreases in concentration as they approach the nest; but
that ants carrying larvae dispense equal amounts of larva odor all the
way and this overcomes the comparatively weak nest odor.
Chauvin (1948) and many other workers were unable to confirm
Bethe's and Brun's conclusions. Since the original idea was based largely
upon experiments in which sections of the track were reversed with the
consequent confusion of the ant, Chauvin suggested that the confusion
actually arose, not from a reversal of a polarized trail, but from the fact
that in reversing a section of the trail it was impossible to make the ends
meet. The ant thus loses the trail at the junction. Carthy was able to
show by developing the trails with lycopodium powder that this was
indeed the case. (In Eciton reversal is without serious inconvenience;
the confusion is only momentary and trailing is soon resumed.) Development of the trail also showed that it consisted of discrete streaks from
4 mm to 8 cm long, shaped as to resemble a drop of fluid extruded,
and drawn out in the direction of running. Thus, although the trail does
possess polarization, there is no indication that this has any meaning
for the ant. As Carthy has shown for A. fuliginosus, pieces of trail are
laid down while the ant is homing, and the balance of the distance is
covered by random wandering. With increasing familiarity additional
portions of the trail are completed, and finally entire trails are deposited,
especially while the ant is carrying larvae. The final result is a tangle
of trails so interwoven that polarization based upon the shape of the
deposits would be impossible. Furthermore, if an ant is allowed to
stumble onto the middle of a trail, it shows no indication of being able
to distinguish the homeward direction from the outgoing direction. In
addition, the trail is not followed exactly in all its intricacies. In maze
studies with Formica subsericea
and F. exsectoides
Schneirla (1929)
claimed a slight but unimportant (as far as orientation is concerned)
end-to-end polarization.
MacGregor (1948) investigated trail-laying and trail-following in
another species of ant, Myrmica ruginodis, and obtained some information which does not conform exactly with the foregoing facts. Having
first determined that a forager moves in a series of loops and that a