230
2.5 mm lateral to the mid-line, for the anterior group at 3.5 and 2.5 mm and
for the posterior group at 4.5 and 3.5 mm, respectively. The lesion damage
lay largely anterior and posterior to the plane of the anterior commissure.
After operation the exposed surface was covered by a cap of dental cement
and the skin sealed with Histacryl tissue glue (Braun, Melsungen). Four days
were allowed for recovery before testing recommenced. After behavioural
testing all animals were perfused with saline and 10% formalin solution. The
brains were serially sectioned (6 μm) and stained in cresyl violet for the re-
cording of tissue damage (Fig. 1).
Histology
Histological examination showed that six birds belonged to the anterior
and six to the posterior Iesioned group. The lesions extend approximately 2
mm anterior or posterior to the plane of the anterior commissure. The bilat-
eral extent of the lesion at five evenly-spaced intervals for all birds in each
group is shown by the shaded area in Fig. 1; the maximum extent is shown
by the hatched area.
The anterior lesion removed bilaterally the dorsal midline hyperstriatum
accessorium and dorsal hippocampus; about 75% of the hyperstriatum inter-
calatum and 50% of the hyperstriatum dorsale (lateral part intact) were also
damaged. The dorsomedial midline of the anterior hyperstriatum ventrale
and the midline border of the neostriatum were damaged in four of six birds
in the plane of greatest lesion damage.
The posterior lesion removed the dorsal and midline hippocampus bilater-
ally. The midline quadrant of the area parahippocampalis was damaged in
anterior sections, but this became progressively more spared posteriorly.
Damage to the dorsomedial and midline hyperstriatum ventrale anteriorly
and the part of the neostriatum posteriorly that contains the anteromedial
Field L was also sustained.
Data from two birds with lesions extending anterior and posterior to the
anterior commissure showed characteristics of both groups during testing
and were not included in the analysis. Three sham controls were exposed to
the total operative procedure. The dura was cut, but no aspiration probe was
applied.
Training
The birds were trained once a day to approach a dish to feed after hearing
a food-trill and to stay away at all other times. Each session consisted of 20
repeated trills played from a Uher 4400 portable tape recorder at intervals of
1, 2, 3 or 4 min in a pseudo-random order. Each train-trill (Fig. 2) was the
same example of a natural call given by a feeding Guinea fowl that no test
animal had met. When a bird approached the food dish it was rewarded with
20—30 maize grains. Prior testing had shown single grains to be ineffective.