Classification of antennal olfactory receptors of the cockroach, Periplaneta americana L


Fujimura, K.; Yokohari, F.; Tateda, H.

Zoological Science 8(2): 243-255

1991


Response spectra and characteristic stimulus intensity-response curves were investigated in the antennal receptor system of P. americana. Responses to 50 arbitrarily selected chemicals were recorded from receptor cells of all the various types of olfactory sensilla, identified by external structure. Most receptor cells (87%) examined could be classified into 1 of 8 groups and 1 pheromone-sensitive cell, on the basis of similarities in response spectra of each receptor cell. There was some overlap in the response spectra of the different groups. Stimulus-response curves of single receptor cells to various components were not always in parallel, thereby suggesting that the receptor may have multiple sites. Receptor cells classified physiologically into the same group are occasionally seen in 2 types of sensilla with different stimulus conducting systems. It is suggested that the receptor membrane may contribute mainly to the discriminatory properties of the sensory organ, while the stimulus conducting system plays an auxillary role.

ZOOLOGICAL
SCIENCE
8:
243-255
(1991)
1991
Zoological
Society
of
Japan
Classification
of
Antennal
Olfactory
Receptors
of
the
Cockroach,
Periplaneta
americana
L.
K.
FUJIMURA',
F.
YOKOHARI
2
and
H.
TATEDA
3
'Department
of
Biology,
Faculty
of
Science,
Kyushu
University,
Fukuoka
812,
and
Department
of
Physiology,
2
Department
of
Biology,
Faculty
of
Science,
Kyushu
University,
Fukuoka
812
and
Biological
Laboratory,
Faculty
of
Science,
Fukuoka
University,
Fukuoka
814-01
and
3
Department
of
Biology,
Faculty
of
Science,
Kyushu
University,
Fukuoka
812;
Japan
ABSTRACT—Response
spectra
and
characteristic
stimulus
intensity-response
curves
were
investigated
in
the
antennal
olfactory
receptor
system
of
the
cockroach,
Periplaneta
americana.
1.
Responses
to
50
arbitrarily
selected
chemicals
were
recorded
from
receptor
cells
of
all
the
various
types
of
olfactory
sensilla,
identified
by
external
structure.
2.
Most
receptor
cells
(87%)
examined
could
be
classified
into
one
of
eight
groups
and.
one
phermone-sensitive
cell,
on
the
basis
of
similarities
in
response
spectra
of
each
receptor
cell.
3.
There
was
some
overlap
in
the
response
spectra
of
the
different
groups.
4.
Stimulus-response
curves
of
single
receptor
cells
to
various
components
were
not
always
in
parallel,
thereby
suggesting
that
the
receptor
may
have
multiple
sites.
5.
Receptor
cells
classified
physiologically
into
the
same
group
are
occasionally
seen
in
two
types
of
senilla,
that
in
senilla
with
different
stimulus
conducting
systems.
We
suggest
that
the
receptor
membrane
may
contribute
mainly
to
the
discrimina-
tory
properties
of
the
sensory
organ
while
the
stimulus
conducting
system
plays
an
auxiliary
role.
INTRODUCTION
The
antennal
olfactory
receptor
system
in
sever-
al
species
of
insects
has
been
investigated
both
physiologically
and
morphologically
[1-5].
In
some
species,
some
functional
groups
of
olfactory
receptor
cells
were
classified
and
were
correlated
to
morphologically
identifiable
types
of
sensilla
containing
these
cells
by
Sass
[6-8].
In
Periplaneta
americana,
olfactory
receptor
cells
that
respond
to
odors
of
single
pure
substances
can
be
classified
into
seven
response
spectra
[6].
Utilizing
a
similar
stimulus
procedure
Selzer
found
some
new
re-
sponse
groups
on
the
antennal
receptors
of
Peri-
planeta
americana
[9,
10].
In
their
experiments,
however,
the
stimulus
intensity
was
controlled
by
means
of
the
concentration
of
the
stimulus
source
Accepted
November
8,
1990
Received
September
8,
1990
Present
address:
Department
of
Physiology,
Nagasa-
ki
University
School
of
Medicine,
Nagasaki
852,
Japan
2
Present
address:
Biological
Laboratory,
Faculty
of
Science,
Fukuoka
University,
Fukuoka
814-01,
Japan
liquid
and
not
that
of
the
stimulus
air
itself.
Hence
their
classification
of
the
receptor
cells
was
based
on
the
responses
to
different
intensities
of
stimuli.
We
attempted
to
determine
the
response
spectra
on
the
basis
of
responses
to
a
constant
intensity
of
stimulus
air,
i.e.
uniform
partial
vapor
pressure
of
the
substance
in
air.
To
classify
functional
charac-
teristics
of
the
antennal
olfactory
system,
it
is
important
to
determine
the
stimulus
intensity-
response
relations
of
each
receptor
group
[6,
7,
9,
10].
The
antennal
olfactory
system
of
the
Periplaneta
comprises
several
morphological
types
of
olfactory
sensilla
[1,
11,
12].
To
elucidate
the
function
of
the
antennal
olfactory
system,
we
examined
receptor
cells
in
all
identifiable
types
of
olfactory
sensilla.
Some
types
of
sensilla
have
already
been
classified
[5,
6, 9,
10,
13].
We
report
here
the
response
spectra
of
different
antennal
olfactory
receptors
housed
in
all
types
of
sensilla.
The
spectra
are
based
on
responses
to
a
constant
intensity
of
stimuli
and
on
the
stimulus
response
relations
of
each
group
of
receptor
cells
244
K.
FUJIMURA,
F.
YOKOHARI
AND
H.
TATEDA
to
several
substance,
including
components
of
the
female
sex
pheromone.
The
response
spectra
observed
were
then
analyzed
in
an
effort
to
corre-
late
cell
groups
to
specific
sensillar
types.
MATERIALS
AND
METHODS
Materials
Adult
male
and
female
cockroaches,
Periplaneta
americana
L.
from
the
laboratory
colony
were
used.
These
cockroaches
were
reared
at
23-27
°
C,
and
50-60%
relative
humidity.
Water
and
rat
pellet
food
were
available
ad
libitum.
Stimulants
Fifty
kinds
of
commercially
available
chemicals
were
arbitrarily
selected
as
a
stimulus
source;
included
were
normal
alcohols
and
derivatives,
terpene
or
aromatic
compounds
(Table
1).
The
purity
of
some
of
the
chemicals
which
elicited
large
responses
from
receptor
cells
was
determined
by
gas
chromatography.
For
studies
on
response
spectra,
the
stimulus
intensity
was
controlled
at
0.1
mmHg
partial
vapor
pressure
at
25
°
C
by
diluting
the
test
chemical
with
ethanol.
The
mixing
ratio
of
a
given
test
chemical
to
ethanol
was
calculated
by
the
Antoine
or
Claperyon-Clausius
equation
using
their
saturated
vapor
pressure
values
[14,
15],
to
derive
desired
intensities
of
stimuli.
Chemicals
with
saturated
vapor
pressures
under
0.1
mmHg
at
25
°
C
or
which
dissolved
poorly
in
ethanol
were
not
diluted.
To
examine
the
relationship
between
stimulus
intensities
and
response,
the
chemicals
were
diluted
stepwise
with
ethanol.
We
used
ethanol
as
the
solvent
for
the
following
reasons:
1)
Ethanol
elicited
little
or
no
responses
from
all
the
olfactory
receptor
cells
studied
by
Sass
[6]
as
well
as
seen
in
our
preliminary
experiments.
2)
Most
chemicals
used
in
this
study
dissolve
in
ethanol
at
25
°
C.
3)
The
stimulus
intensity
can
be
calculated
using
the
above
equations
when
the
solvent
is
a
pure
substance
such
as
ethanol.
Crude
sex
pheromone
was
prepared
by
the
following
methods.
Virgin
females
were
segre-
gated
from
the
males
before
the
imaginal
molt.
The
collected
feces
(200
mg
in
dry
weight)
of
the
virgin
female
were
allowed
to
steep
in
distilled
water
(100
m1).
The
supernatant
(20
ml)
was
ex-
tracted
three
times
with
n-hexane
(5
ml)
at
5°C.
The
combined
hexane
solution
(15
ml)
was
conde-
nsed
under
a
vacuum
at
45
°
C
to
give
about
0.5
ml
hexane
extract.
The
amount
of
pheromone
was
estimated
by
the
behavioral
test
of
Rust
[16]
and
Tobin
et
al.
[17].
About
50
pl
of
the
hexane
extract
was
almost
equivalent
to
10
-4
-10
-5
pg
peripla-
none
B.
Male
fecal
extracts
were
prepared
using
the
same
procedure
and
served
as
the
control.
Stimulation
The
antenna
was
first
exposed
to
an
odorless
dried
air
stream
from
a
glass
nozzle.
This
air
was
obtained
by
passing
it
successively
through
active
carbon
and
silica
gel.
Stimuli
were
delivered
by
an
electromotor-driven
syringe,
outlet
(4
mm
in
dia-
meter)
of
which
was
about
10
mm
from
the
anten-
na.
The
flow
velocity
was
180
cm
/sec
at
the
outlet.
The
syringe
contained
a
piece
of
filter
paper
(4
cm
2
)
containing
odorant
solution
(50
pl).
With
respect
to
pheromone
stimulation,
the
prepared
hexane
extract
(50
pl)
was
absorbed
by
the
filter
paper
and
after
the
hexane
had
evaporated,
the
filter
paper
was
put
into
the
syringe.
The
stimulus
duration
and
the
stimulus
interval
were
0.6
sec
and,
at
least,
3
min,
respectively.
The
odorant
was
ventilated
from
the
system
following
each
stimulus
period.
Recording
The
recording
procedure
was
similar
to
that
used
by
Yokohari
and
Tateda
[18].
The
animal
was
immobilized
by
coiling
with
ice
and
their
limbs
were
attached
to
a
holder
using
paraffin.
The
antenna
was
also
immobilized
at
2-3
mm
intervals
on
the
holder.
The
active
electrode
was
an
etched
tungsten
wire
(0.5
mm
in
diameter)
with
a
tip
diameter
of
less
than
1
pm.
The
electrode
tip
was
placed
into
the
basal
cavity
of
a
sensillum.
The
indifferent
electrode,
a
platinum
wire
(0.3
mm
111
diameter),
was
inserted
into
a
distal
cut
end
of
the
antenna.
Electrical
events
were
recorded
using
a
standard
method.
The
response
magnitude
was
represented
by
the
impulse
number
at
0.05-0.45
sec
after
the
stimulus
onset
minus
the
impulse
number
at
0.05-
0.45
sec
before
the
onset.
Olfactory
Recepotors
of
Periplaneta
245
Identification
of
sensillum
At
termination
of
the
recording,
the
location
of
the
sensillum
was
marked
for
identification
by
removing
some
of
bristles,
and
the
distribution
of
surrounding
sensilla
was
sketched.
The
piece
of
antenna
with
the
marked
sensillum
was
dehy-
drated
through
a
graded
acetone
series,
dried
in
air
and
the
dried
antenna
was
coated
with
gold
in
an
ion
coater.
Observations
were
carried
out
using
HITACHI
S-430
scanning
electron
microscope
(SEM).
Morphological
classification
of
sensilla
We
classified
the
olfactory
sensilla
on
the
anten-
na
into
the
following
types,
according
to
the
exter-
nal
structure
observed
by
SEM.
1)
Type
S
sensillum
has
a
smooth-surface
and
a
blunt-tip,
and
curves
gradually
towards
the
tip
of
the
flagellum.
This
type
of
sensillum
was
further
classifiable
into
subtypes,
on
the
basis
of
its
length,
for
both
sexes;
Type
S-I
is
8-12
pm
long
in
both
sexes,
and
type
S-II
is
18-22
pm
long
in
males
and
13-16
pm
in
females.
Type
S
sensillum
belongs
to
a
single
walled
WP-sensillum
described
by
Altner
and
Prillinger
[2].
2)
Type
G
sensillum
has
a
grooved-surface
and
a
terminal
pore
and
the
length
is
ca.
7
pm.
It
is
straight
or
bends
slightly
towards
the
tip
of
the
flagellum.
This
type
was
also
subclassifiable
into
two
types,
on
the
basis
of
the
number
of
longitu-
dinal
grooves
on
its
side
wall;
type
G-I
has
18-26
grooves
and
type
G-II
has
24-32
grooves,
in
both
sexes.
As
the
number
of
grooves
overlaped
be-
tween
G-I
and
G-II
types,
an
further
criterion
was
needed
for
subclassification
in
the
case
of
the
sensilla
with
24-26
grooves.
For
practical
pur-
poses
we
subclassified
the
sensilla
which
housed
a
Cold
receptor
cell
into
G-II
and
the
others
into
G-I
among
the
sensilla
having
24-26
grooves.
The
G
sensillum
belongs
to
the
double-walled
WP-
sensillum
in
the
category
of
Altner
and
Prillinger
[2].
3)
Type
T
sensillum
is
a
long,
thin
sensillum
(20-
40
pm
in
length)
with
a
sharp-tip.
The
distal
two-thirds
of
the
hair
are
thinner
than
the
basal
o
ne-third
and
bend
sharply
towards
the
tip
of
the
flagellum.
The
T
type
belongs
to
the
single-walled
WP-sensillum
described
by
Altner
and
Prillinger
[2].
RESULTS
Physiological
classification
of
receptor
cells
Responses
were
recorded
from
about
250
olfac-
tory
receptor
cells,
68
of
which
are
included
in
Table
1.
The
response
to
the
chemicals
which
dissolved
poorly
in
ethanol
or
whose
saturated
vapor
pressure
was
under
0.1
mmHg
at
25
°
C
was
extrapolated,
assuming
that
the
response
magni-
tudes
increase
along
a
standard
response
curve
(see
2nd
section).
The
response
magnitudes
were
classified
into
four
grades;
100-85%,
84-50%,
49-
20%
and
less
than
20%
of
the
maximum
(see
Table
1).
To
classify
the
receptor
cells,
we
statistically
examined
the
response
magnitudes.
Similarities
of
the
response
spectra
were
evaluated
by
calculating
the
correlation
coefficient
r
for
spectra
of
each
possible
pair
of
all
68
receptor
cells.
First,
all
responses
of
each
receptor
to
chemicals
examined
were
standardized
to
the
response
to
the
most
effective
chemical
for
the
receptor,
though
the
response
magnitudes
are
shown
in
4
graded
groups
in
Table
1.
Secondly
rs
of
the
spectra
of
all
pairs
(2278
pairs)
were
calculated.
If
the
r-value
was
over
0.5,
the
paired
receptor
cells
were
considered
to
correlate.
In
this
fashion
the
similarity
of
the
receptor
cells
was
evaluated
to
correlate
or
not
to
do
so.
The
combinations
of
cells
evaluated
to
correlate
with
a
given
cell
differed
from
cell
to
cell
and
there
were
few
cells
with
the
same
combina-
tion
of
correlate
cells.
For
this
reason,
the
similar-
ity
index
i
was
introduced.
The
similarity
index
i
is
defined
as
N
I
/N
2
,
where
N
1
is
number
of
correlate
cells
common
to
both
compared
cells
and
N
2
is
number
of
correlate
cells
to
at
least
one
of
them.
We
arbitrarily
selected
the
i
value
greater
than
0.5
as
an
indication
of
similarity
(see
Discussion).
This
means
that
the
pair
of
cells
which
shares
at
least
half
the
number
of
correlate
cells
of
each
cell
were
interpreted
to
be
similar.
Consequently,
we
iden-
tified
eight
groups
(I-VIII)
of
receptor
cells
having
the
same
or
almost
the
same
combinations
of
similar
cells
(Table
2)
and
several
solitary
cells.
246
K.
FUJIMURA,
F.
YOKOHARI
AND
H.
TATEDA
TABLE
1.
Response
spectra
of
antennal
olfactory
receptor
cells
of
Periplanew
americana
Cell
Number
Stimulus
Substances
02
05
03
04
09
11
16
13
20
26
A8
2
4
07
04
06
10
08
15
14
19
Al
21
A2
2
n-propanol
0
0
0
0 0
0
0 0
0
0
0 0
3
n-butanol
++++4µ
1+}4}4}++
0 0 0 0
+
0
0
0
0 0
++
0
4
n-pentanol
4
44+ 411-
4
41+
+f
414
1+
44-
4+
+1A+
+
1+ 1+
++0
+
:
4
4
g
5
n-hexanol
+I-
+
-FF
+1+
+F
*
44-+FF
41F
th
1+F
-I+
14F
+I-
44-14-
+
+1-
+
-4-
,
0
+
6
n-heptanol
0
+
0
0
0 0 0
+
+F
4+
44+
+I+
th
*
4+
*
+ +
++
1
.
i,L
7
n-octanol
2)
0
0
0
0
0 0
0
0 0
0+F+144+-114
-H-
°
+
-
f
t
-
0
41-
8
n-nonanol
1),
2)
0
0
0
0
0
414
-141-
4
}
1
0
0
1+
+
9
n-decanol
1),
2)
0
0
0
+I-
44+
41-
0 0
0
th-
0
0
+
.
10
n-propanal
0
0
0 0
0 0 0
0
0
0 0
0 0
0 0
0
0
.
11
n-butanal
0
0
0
0 0 0 0 0
+
+
0 0
0 0 0 0
0
12
n-pentanal
0
0
0
0 0 0 0
+11-•41--11-
0
000
0
0
0
0
(
0
0
1
)
1
0
13
n-hexanal
.
0
0
0 0 0 0
+HE
1++F+00+000
14
n-heptanal
0 0
0
0 0 0 0 0 0
0
+
1+
+I-
+
0 0
0 0 0
0
++
15
n-octanal
000
0
0
0 0 0 0 0
0-H-1+-1+0
0 0
0
44-+
0
16
formic
acid
0
0 0
0 0
0
0
+
0
0 0 0 0
0
0
0
17
acetic
acid
0
+
0
0 0
•+
0
•+•
0
00
0 0 0
0
0
18
propanoic
acid
0
0
0
0 0
0
0 0
+
0
0
0
0
0 0 0
0
0 0
19
n-butylic
acid
•+
0
0
0
0
0 0 0
0-1+++F000
0 0
0
0
20
n-valeric
acid
0
0
0
0
0
0+
0
4
--
+I-
-H-
0 0
0 0 0
0
21
n-caproic
acid
1),
2)
0 0
0
0
0
0
0+
0
0—
4-1-
I*
0 0
0+
0
,
0
22
enanthic
acid
1),
2)
0
0
0 0
0 0
0
...
....
.
0
0
4H-0
(.
23
isopropyl
acetate
000
0 0 0
0
0
0
0
0
0
0
0
0
+
0
0 0
0
)
0
0
24
n-butyl
acetate
0
0 0
0
+
0
+I-
+
+
0 0
0
+
0
'
0
:
.
+
+
44-
27
ethyl
amyl
acetate
0
0
0
+
0 0
0
0
0
0 0
+
+
0 0
+
+
+
25
n-amyl
acetate
0
+
41-
++0
0
0
0
0
+
+
1
i
+
+
0 0 0
+
0
+
+
00
0
0
26
ethyl
n-caproate
0
0 0
0 0 0
0
0+ 0+
0
th-
441-
H-
0
+
+
28
glycerol
1),
2)
0
0
+
0 0
0
0
+
0 0
0
0
29
d-limonene
0
0
0
0 0 0 0 0
000
4+•
0 0 0 0
4+
0
0+
0*
0
30
l-limonene
31
d-carvone
0
0
0
0 0 0
0 0
0 0
0
0
0
++
-H
+++
0
4+
-
0
+
+
1)
0
0 0
0 0 0 0 0 0 0
0
0
+
0
0+
0
+ +
+
4+
0
414
0
32
1-carvone
1)
000000000000+0000+++++
0
*
0
33
menthone
1)
+
0
+F
-H-
+
+F
0
0
0
0
1-F
0
+
0 0
1+ 1+
-H-
-FF
*
A+
+
+I-
+
34
fenchone
0
0 0
0
0
0 0
0
0 0
0
0
444+
4i
-F1
:
-
1+
+
*
0
0
.
++
:
35
fl-ionone
1),
2)
0
0
0
0 0 0
0 0
0 0
0
++
36
terpineol
1)
0
0
0
0 0
0 0
+
+
0
+
0
+
0
0
*
*
*
*
-14-
i+
+
4+
H
-
0
39
santalol
1),
2)
0
+
0 0 0 0
0
+
+
0
0
0
4+•
+I-
++
-
-
*
7
'4,
4+
:
-
-
f
o
++0
4:
38
citronellol
37
geraniol
1),
2)
0
0
0
0 0 0 0
+
0 0
0
-F1-
+F
4+ 4+
+
1+
1),
2)
0
0
0 0
0 0 0 0 0 0
0
0 0
+
0
0
40
citral
1),
2)
0 0
+
0
0
0
0
* +
+
+ +
ii-
,+,
+
0
-,-,
0
41
cineol
0
0
0 0
0 0
0
0 0
0
+
0
+
0
+
*
43
benzyl
alcohol
1),
2)
1+
-I+1+
+
*
+I-
+
1+
0
+
0
*
0 0
411
+
0
0
°
0
-
4-
0
- H -
+
+
0
00
+0
42
bornyl
acetate
1)
0
0
0 0
0 0 0 0
+
0
0
0
0 0
0
44
benzyl
aldehyde
0
0
0
0 0 0 0
0
0
0
0
0
+
0
0
+
+
+F
++
--
(1
'
0
45
benzyl
formate
1)
+I+
0
+I-
-H-
++
0
0+
0+
0
0
0
0
0
0
0
0
(I
0
0
46
a-phenethyl
alcohol
1)
0
0
0
+
0
0
+
-FF
-H-
+
0
+
0
0
+4
-I+
1+
-FF
+
+
0
+
47
/9-phenethyl
alcohol
1)
0+
0
1+++
0
+
0 0
00+•
0+14
4+++1-+F+
0 0
+
48
phenyl
acetate
0
0 0
0
0
0
+ +
+
.
.
* *
*4*
4+
0
+
,,
'
49
p-anis
aldehyde
1),
2)
0 0 0 0 0
0 0 0
0
++
-H-
-H-
++
0
0
0#
+F
+
H
"
50
cinnamic
aldhyde
1),
2)
0 0 0 0 0
0 0
0
0
1+
-H-
+
+
0
0 0
+F
-FF
-H
-
Cell
Groups
1
I
i
1 I
I
I
ethanol
0 0
0
0
0
0 0 0 0 0 0
0
0 0 0 0 0 0 0
0
0
(()
0.
II
III
IV
The
stimulus
intensity
was
set
at
0.1
mmHg
partial
vapor
pressure
at
25
°
C
for
all
stimuli,
except
the
chemicab
marked
2).
The
responses
of
each
cell
were
standardized
to
its
response
to
the
most
effective
stimulus.
flf'l
100-85%
of
the
maximum
response;
+I-,
84-50%;
+,
49-20%;
0,
less
than
20%.
Dots
indicate
substances
nc.
tested.
Boiling
points
of
chemicals
marked
1)
are
higher
than
200
°
C.
Chemicals
marked
2)
were
used
without,
0
0 0
0
0
)
TABI
F
1.
Continued
Olfactory
Recepotors
of
Periplaneta
247
37
A7
34
32
28
30
33
25
35
31
27
29
42
Cell
Number
A9
45
38
39
40
49
47
51
53
50
41
43
44
A6
52
48
67
54
64
66
58
55
60
57
59
B1
65
70
71
68
0
0
0
00000000000
0
00000 000
0
000000
0
00000000
0
0000
.
0
000
0000
0
0000 000
0
000000
0
000
0
00
.
0
0
0000
0000
0
0000 000
0
000000
0
0
0
0
00
0
0
0 0
000000000
0
0000 000
0
000000
0
0
0
0
00
00
.
0 0 0
00+000000
0
0000 000
0
000000
0
0
0
0
00
00
1
+
0 0
0000000000+
0
0000 000
0
000000
0
000
0
00
00
+•
0000000000+
+
00+0
000
0
000000
0
000
0
00
00
0 0
0
00
0004+
+
0*
000
00
0
000
0
00
00
0
00
000004
-H-
0
04
000
00
0
000
0
00
0
. .
o
000
00
0
0
0000 000
0
000000
0
0
000000
0
000
00
0
0
0000
000
0
0000
0
0
00000000
0
0000
0000+0
0000
000
0
000000
0
0+000000
000
0
0000+0000+
0
0+00 000
0
0000
0
0
0+000000
0
000
0
000
000+0+
+
00+00 000
0
0000
0
0
00000000
0
00+0
o
0000000+0+
+
0
.
4+40
+0+0000000
0
00000000
0
004++
0
000
000
0
+0**4
+
0
4
*
•4+
4
+1+
00000000
0
0
0
000
000
1+
00000 000
0
0
000
4+
+0000000
00
0
000
00000+ 0.000 0+0
+
0
000
0+4+00+0
0
0000
0
0
000
000
00 00000 000
0
00
000
4+
410-4+++0++
0
0000
0
000
000000* 000+0
000
0
000000
0
44
4
-
414444
+
4
44-4+440
0
0
0
000
0000
++4++4++
+++
04+0000
0
0
—I-R*10+4+0f*
0
0
0
0 0
4
*
*
+I+
444
40
0
0
0
0
Hi
+4+ 431-
4+[
ff
11+
0
0
0
0 0
0
0
0
0
0
00000 000
0
000000
0
0000
00
0
0000
0 0
0
0
0 0
0
0
0+
0
0
00000
0
000
0
0
0000
40+00000000
0+0++
000
0
+00000
0
0000
00
0
0000
+000000
000
0
000+
000
0
000000
0
0000
00
0
0000
0000000
00
0
000
00
0
0
0000
0
00000
00
0
0000
00
00
O•
0
0000 000
0
0
00
0
0
0
0
00
00
0
00
•+++00
0
4+
000
00
0
0
00
0
0
0
000
00
0
0000
**•+
0
0
0
0+
0
+
0
0
00
0
0
0000
0
0
000
00
0
0000
+
+ +
0 0
0 0
+
0
+
+44-4
++40
if
+000
0
+
000000
0 0
0+
0+1+i++000++0
*
0+4
+
+
0
+
0000
0
0
000000
0 0
00
+
+
+I-
i+
+
+
i+
i+
*
-4
0
4
0
0
0
i+
4000
0
+
000000
0 0
00
0
0
0 0
0
0 0
0
+
+00
000 0400000
0
+
00000
0 0
00
0
0
+
0
0
0
0
+• 0404400+
*• 000000
0
0
00
*
+I+
+
I+
*
+I+
* *
+If
*
0
+
00000 000
0
000000
0
00000000
0
0000
0
+0
0***++4+14+
*14+4+44+
+000000000
0
00000000
0
0000
0 0
0
0
0
+
0
0
0
4+
4+
+
i+
0
0
-I+
*
+
000
0
000000
0
00
00000
0
00+0
4+000000
00+-
+0
0
+1+0400
00 00000
0
-H-
i+
0 0
0
0+
0*
0
00000000
0
0
0
i+
*
+
0 0
0 0
0
+
+1-
+
0000
0
0
00
00
0
00000000
0
00
0
0
++1+++00
0+0
0
0044
+04
0
00 00
0
+0
0
0
00
0
00+4
0
000000004+1+
+
00+00 000
0
000000
0
00000000
0
000
+00•
0000+0+
+00++0
000
0
000000
0
00000000
0
0000
0000000
000
+
itf
4 4
++*
+40000
0
00
0
00
0
0000
000000000-H-0
0
000+0 000
0
000000
0
00000000
0
000
0000000000+
4++++4++
000
+
000000
0
00000000
0
000
000+•
00
•+++
0
000+
+00
00
00
0
0000
00
0
0000
0
000
000000*
+F
4+
4+
+I+
44f
4-F
000
000040-H-i+
+FE 4+F
*
4+
44
*+
*
µ
0
000000
0
0
000000
0
00000000
00000000
0
0
000
0
00
VI
VII
VIII
dilution
because
they
dissolve
poorly
in
ethanol
or
their
saturated
vapor
pressure
was
less
than
0.1
mmHg.
The
Table
2).
responses
to
these
substances
were
determined
by
(see
The
cells
were
classified
into
8
extrapolation.
groups
Olfactory
Recepotors
of
Periplaneta
249
TABLE
2.
Continued
2,
3,
Cell
No.
Sensillar
Type
07,
06,
05,
01
04
s-I
10,
09,
08,
11,
S-I
15,
13
16,
14,
12,
S-I
9,
20,
Al,
21,
S-I
22,
A7,
37,
25,
S-11
34,
35
32,
31,
28,
27,
S-II
30,
29,
33
A9,
50,
45,
41,
G-I
38,
43,
39,
44,
4
0.
A6,
49,
47
52,
51,
48
60,
64,
57,
66,
G-II
5
9,
58,
B1,
65,
71.
70,
68
T
A8.
A2,
24, 23,
S-II
42,
67,
53,
55,
or
54
G-I
The
stimulus-response
relationships
were
ex-
amined
for
each
group.
The
threshold
of
the
receptor
cell
to
a
given
substance
was
defined
by
the
stimulus
intensity
which
elicited
half
the
max-
'mum
response
to
this
substance.
Group
I
Cells
in
Group
I
were
excited
by
an
n-alcohol
series
having
4
to
6
carbons,
especially,
pentanol.
They
were
also
strongly
excited
by
benzyl
alcohol,
benzyl
formate
and
menthone.
The
maximum
response
was
seen
with
pentanol,
with
a
magnitude
of
59
±
11
(SD)
impulses/0.4
sec,
and
a
threshold
of
4.5
x
10
-3
mmHg.
The
thresholds
to
butanol,
hexanol
and
benzyl
alcohol,
and
benzyl
formate
and
menthone
were
10
-2
and
10
-1
mmHg,
respectively
(Table
2).
Group
II
Cells
in
Group
II
were
excited
by
n-alcohols
from
C5
to
C7,
and
especially
hexanol.
They
were
also
excited
strongly
by
a-phenethyl
and
benzyl
alcohols.
The
maximum
response
was
59±
19
impulses/0.4
sec.
Thresholds
were
4.2
x
10
-3
mmHg
for
hexanol,
10
-2
mmHg
for
pentanol
and
heptanol,
and
10
-1
mmHg
for
hexanol
and
a-phenethyl
alcohol
and
benzyl
alcohol
(Table
2).
Group
///
Cells
in
Group
III
excited
by
various
substances
as
shown
in
Table
2.
Most
belonged
to
normal
chain
groups.
Among
them
octanol
and
nonanol
elicited
strong
responses;
the
maximum
response
was
60±10
impulses/0.4
sec,
and
the
threshold
was
7.5X
10
-5
mmHg
for
octa-
nol,
5.8x
10
-5
mmHg
for
nonanol.
The
second
most
effective
compounds
we're
heptanol
and
hex-
anol,
the
thresholds
being
10
-3
mmHg
for
both
compounds.
Group
IV
Cells
in
Group
IV
were
excited
by
various
substances
which
belong
to
the
terpenes
and
aromatic
compound
groups,
n-alcohols
and
esters
(Table
2).
Thresholds
exceeded
10
-2
mmHg
for
all
the
compounds
examined.
None
of
the
compounds
yielded
a
staturated
resposne
even
at
the
maximum
intensity
of
stimulation
(about
10
-1
mmHg).
The
response
was
about
55
impulses
/0.4
sec
at
the
maximum
intensity
of
sti-
mulation.
Group
V
Cells
in
Group
V
were
almost
exclusively
excited
by
terpene.
Thresholds
ex-
ceeded
10
-2
mmHg
for
all
the
compounds
ex-
amined
(Table
2).
The
largest
response
was
36
±
7
impulses/0.4
sec.
Figure
la
and
b
show
the
stimu-
lus-response
curves
characteristic
for
cells
in
Group
V.
The
response
curves
reached
plateau
at
about
0.1
mmHg
of
the
most
effective
compounds,
cineol
and
menthone
in
some
cells
(Fig.
la).
The
plateau
was
not
observed
in
the
response
curves
of
?50
K.
FUJIMURA,
F.
YOKOHARI
AND
H.
TATEDA
41
a
3
14
41
20
-
10
-
36
0
0
-6
-4
-
2
-
1
0
-6
-5
-4
-3
-
log
vapor
pressure
(mmHg)
FIG.
1.
Stimulus-response
curves
of
two
cells
(a.
Cell
A7;
b.
Cell
22
in
Tables
1
and
2)
in
Group
V.
The
curve
almost
in
parallel,
except
for
the
curve
for
pentanol
(4).
The
number
of
each
curve
refers
to
the
sti
substance,
as
enumerated
in
Table
1.
30
-
b
s
are
muls
33
0.5
ti
a
-6
-5
-4
-3
-2
-1
0
log
vapor
pressure
(mmHg)
FIG.
2.
Relationships
between
partial
vapor
pressures
of
terpineol
and
the
responses
of
the
cells
in
Group
VI
(II,
11
cells
of
males;
0,
7
cells
of
females).
Responses
relative
to
the
maximum
response
are
plotted
against
partial
vapor
pressure
of
terpineol
on
a
log
scale.
Threshold
is
below
10
-4
mmHg
and
the
response
is
saturated
at
10
-2
mmHg.
other
cells
(Fig.
lb).
The
curves
were
almost
parallel
with
the
exception
of
the
one
for
pen
tanol
Group
VI
Cells
in
Group
VI
are
characte-
rized
by
a
marked
large
response
to
terpineol;
the
sigmoidal
stimulus-response
relationship
is
shown
in
Figure
2.
The
threshold
was
below
10
-5
mmHg
and
the
maximum
response,
45
±
11
impulses/0.4
sec,
was
obtained
at
about
10
-2
mmHg
(Table
2).
These
cells
were
also
excited
by
menthone,
gera-
niol
and
citronellol.
Thresholds
were
10
-2
mmHg
for
citronellol.
The
stimulus-response
curves
for
different
compounds,
except
for
menthone
and
bornyl
acetate,
are
nearly
in
parallel
to
the
curve
for
terpineol
within
the
range
tested,
whereas
the
curves
for
menthone
and
bornyl
acetate
did
not
appear
to
be
in
parallel
to
the
curve
for
pentanol
(Fig.
3a,
b).
Group
VII
Cells
in
Group
VII
are
characte
-
rized
by
a
strong
excitation
to
undiluted
santalol
at
a
vapor
pressure
of
2
x
10
-5
mmHg.
These
cells
are
excited
also
by
cinnamic
aldehyde,
citral,
ena
0-
I
e
I
s
111P
.•
ell
a
40-
a
40-
b
36
33
O
33
41
20-
4)
-
20
a.
42
41
4
14
0
0
-1
0
-6
-5
-4
-3
-2
-1
0
log
vapor
pressure
(mmHg)
FIG.
3.
Stimulus-response
curves
of
two
cells
in
Group
VI
for
various
compounds.
The
stimulus-response
cure
almost
in
parallel
to
the
curve
for
terpineol
(36)
except
for
menthone
(33)
and
bornyl
acetate
(42).
Olfactory
Recepotors
of
Periplaneta
251
33
40
SO
0
0
13
36
20
0
•0
-
-3
-2
-1
-6
-5
-4
-
log
vapor
pressure
(mmHg)
4.
Stimulus-response
curves
of
the
cells
in
Group
VII.
a.
Relative
responses
of
the
three
compounds
to
undiluted
santalol.
Each
point
is
the
average
response
of
six
cells.
Vertical
bars
indicate
standard
deviations.
b.
Spike
responses
of
cell
A9
are
plotted
against
partial
vapor
pressures
of
various
compounds.
The
cell
in
Group
VII
is
characterized
by
a
strong
excitation
to
sanatalol.
The
curves
of
various
compounds
are
almost
in
parallel.
13
21
20
c'
40
19a.
20
0
. .
.
0
20
-5
-4
-3
-2
-1
0
-5
-4
-3
-2
-1
log
vapor
pressure
(mmHg)
FIG
5.
Stimulus-response
curves
of
cells
in
Group
VIII.
a.
Relative
responses
of
various
compounds
to
undiluted
enanthic
acid.
Each
point
is
the
average
response
of
nine
cells
which
responded
exclusively
to
n-fatty
acids.
Vertical
bars
indicate
standard
deviations.
b.
Stimulus-response
curves
of
cell
70,
which
responded
also
to
some
compounds
other
than
n-fatty
acids.
The
spike
response
is
plotted.
Most
cells
in
Group
VIII
respond
only
to
n-fatty
acid,
whereas
some
respond
not
only
to
n-fatty
acids
but
also
to
octanol
(15)
and
bornyl
acetate
(42).
15
42
a
b
60
a)
thic
acid,
and
some
other
terpene
and
aromatic
compounds.
Figure
4a
shows
the
stimulus-
response
relationships
to
santalol,
cinamic
aldehy-
de
and
citral;
the
responses
were
in
paralled
and
the
thresholds
were
2
x
10
-5
mmHg
for
santalol,
10-3
mmHg
for
cinnamic
aldehyde,
and
10
-2
mmHg
for
citral
and
enanthic
acid
(Table
2).
The
stimulus-response
curves
to
less
effective
com-
pounds
were
also
observed
to
be
in
parallel
to
the
Curve
for
santalol
(Fig.
4a).
Some
cells
responded
Only
to
santalol
and
citral,
threshold
being
10
-2
mmHg
(Table
2).
Group
VIII
Cells
in
Group
VIII
are
char-
acterized
by
large
responses
to
n-fatty
acids
with
Over
four
carbons.
Most
cells
responded
only
to
a
-
fatty
acids,
whereas
other
cells
responded
also
to
octanol
and
bornyl
acetate
(Fig.
5a,
b).
Figure
5a
shows
the
stimulus-response
relations
of
cells
re-
sponding
only
to
fatty
acids;
the
responses
were
almost
in
parallel
to
one
another.
The
maximum
response,
which
was
elicited
by
enanthic
acid,
was
60±13
impulses/0.4
sec.
The
thresholds
were
1.4
X
10
-4
mmHg
to
enanthic
acid,
and
1.3
x
10
-3
mmHg
to
caproic
acid.
Some
cells
were
strongly
excited
by
caproic
acid
and
enanthic
acid,
but
weakly
by
valeric
acid
(Fig.
5b).
Pheromone
receptor
Cells
sensitive
to
the
crude
extract
of
female
feces
were
found
in
the
type
S-II
sensillum
of
the
male
(Fig.
6)
(not
given
in
Tables
2
and
3).
The
receptor
cells
responded
to
50
pi
hexane
extract
of
the
female
feces
with
20±
3.6
impulses/0.4
sec
but
not
to
a
100-fold
diluted
a
80
b
U
39
cu
40
.14
252
K.
FUJIMURA,
F.
YOKOHARI
AND
H.
TATEDA
terpineol
menthone
i44,4
I
4*,
extract
(female)
extract
(male)
menthone
extract
(female)
P2
mV
100msec
Fia.
6.
Extracellular
recordings
of
responses
of
olfactory
receptor
cells.
These
recordings
from
a
type
S-I1
sensillum
of
a
male
reveal
three
different
types
of
cells
present
in
the
sensillum.
A
cell
in
Group
VI
(largest
spikes)
responds
to
terpineol
at
4.5
x
10
-4
mmHg
with
39
spikes/0.4
sec
(a),
and
to
menthone
at
10
-4
mmHg
with
19
spikes/0.4
sec
(c),
which
is
lost
by
a
preceding
application
of
terpineol
(a).
A
cell
in
Group
V
(smallest
spikes)
is
excited
by
the
application
of
menthone
(a,
c).
A
cell
(middle
spikes)
is
excited
by
a
crude
extract
of
female
feces
of
50
pl
(b,
c),
but
not
by
the
extract
of
male
feces
(b).
stimulus.
These
cells
were
not
excited
by
male
extracts
nor
by
the
some
of
the
compounds
ex-
amined.
These
same
cells
were
absent
in
the
females.
Solitary
cells
Several
receptor
cells
listed
in
Table
1
do
not
belong
to
any
group,
in
our
clas-
sification
shown
in
Table
2.
Relationships
between
cell
groups
and
sensillar
types
All
sensilla
from
which
responses
were
recorded
were
observed
with
SEM,
and
all
groups
of
recep-
tor
cells
correlated
with
the
sensillar
types.
As
shown
in
Table
2,
the
cells
in
Groups
I,
H,
III,
and
IV
were
found
exclusively
in
the
S-I
sensilla
along
with
one
cell
which
usually
did
not
respond
to
any
stimuli
examined.
The
cells
of
Group
V
occurred
in
the
type
S-II
sensillum
together
with
cells
of
Group
VI,
solitary
cells
or
functionally
non-
identified
cells,
in
both
sexes.
The
following
combinations
were
identified;
No.
22
cell
(V)
and
No.
31
cell
(VI),
No.
24
(solitary)
and
No.
32
(VI),
and
No.
A8
(solitary)
and
No.
33
(VI)
in
Table
1.
Moreover,
cells
in
Group
VI
were
sometimes
found
along
with
female
sex
pheromone
sensitive
cells
in
the
male
(Fig.
6).
The
cells
in
Group
VII
occurred
in
type
G-I
sensilla
along
with
one
or
two
functionally
non-identified
cells,
in
both
sexes.
Cells
which
belonged
to
Group
VIII
were
found
in
two
morphological
types.
Cells
responding
exclu-
sively
to
fatty
acids
in
Group
VIII
were
found
in
the
type
G-II
sensillum
along
with
a
cold
receptor
cell
and
one
or
two
non-identified
cells.
Cells
in
this
group
responding
also
to
fatty
acids
and
to
octanol
were
found
in
the
type
T
sensillum
along
with
a
cell
which
did
not
respond
to
any
stimulus
examined.
DISCUSSION
Solvent,
solution
and
stimulus
intensity
We
used
ethanol
to
dilute
the
stimulus
subst-
ances
and
most
of
the
chemicals
we
used
dissolved
well
in
ethanol
(cf.
[6-8,
19,
20]).
Although
ethanol
did
elicit
a
slight
degree
of
excitation
from
a
few
receptor
cells
in
the
S-I
sensilla,
the
magni-
tude
were
negligible.
The
stimulus
intensity
was
determined
from
the
partial
vapor
pressure
of
the
compound
calculated
from
the
equation
of
Antoine
or
Claperyon-
Clausius.
Therefore,
the
values
of
the
stimulus
intensities
were
accurate
for
the
compounds
whose
boiling
points
were
low,
but
were
inaccurate
by
about
one
log
unit
for
compounds
whose
boiling
points
are
over
200
°
C.
Such
differences
are
re-
sponsible
for
response
magnitude
differences
of
at
most
50%
of
the
maximum
response,
in
some
cases.
However,
this
does
not
seriously
affect
the
reliability
of
our
classification
of
the
receptor
cells:
Olfactory
Recepotors
of
Periplaneta
253
As
the
stimulus
intensity
of
given
chemicals
was
set
in
the
same
value
throughout
the
experiment,
it
did
not
affect
the
similarity
index
i,
even
if
the
value
differed
from
that
of
other
chemicals.
However,
there
is
the
possibility
that
the
abso-
lute
threshold
for
the
compounds
with
high
boiling
points
may
be
over-
or
underestimated
within
the
range
of
one
log
unit
(Table
2).
Thresholds
to
chemicals
which
did
not
dissolve
in
ethanol
and
whose
saturated
vapor
pressures
were
under
0.1
mmHg
were
obtained
by
extrapolation.
Hence
the
thresholds
might
be
over-
or
underestimated
when
the
response
curves
to
these
chemicals
did
not
parallel
the
standard
curve.
The
error
depends
on
the
difference
between
the
slopes.
The
most
highly
effective
compounds
were
analyzed
for
purity,
using
gas
chromatography.
Among
them,
santalol
was
shown
to
have
two
high
peakds
and
one
low
one.
The
two
high
peakds
were
due
to
a-santalol
and
/3-santalol,
but
the
third
could
not
be
determined.
Thus,
there
is
the
possibility
that
the
receptors
might
respond
to
the
third
one.
Nevertheless,
we
used
this
compound
because
it
elicited
large
and
stable
responses
exclu-
sively
from
receptors
in
Group
VII.
Cell
classification
and
stimulus-response
rela-
tionships
We
introduced
statistical
methods
to
classify
the
physiological
responses
of
receptor
cells.
Al-
though
the
boundary
of
the
similarity
index
i
was
arbitrarily
set
at
0.5,
59
cells
(87%)
of
the
68
listed
in
Table
1
could
be
included
in
one
of
8
groups,
each
of
which
included
at
least
four
receptor
cells.
Thus,
our
procedure
appears
to
be
appropriate
for
analysis
of
this
population
size.
Most
of
the
receptor
cells
could
be
classified
into
8
groups.
However,
nine
cells
could
not
be
clas-
sified
and
were
excluded
from
these
groups.
This
means
that
all
antennal
olfactory
receptors
of
Periplaneta
do
not
always
fit
this
classification
scheme.
Boeckh
and
Ernst
[4]
stated
in
a
review
of
the
work
done
in
their
laboratory
that
the
antennal
receptor
cells
of
the
same
animal
could
be
clas-
sified
into
25
groups.
The
difference
between
their
L
b
result
and
ours
seems
to
originate
from
the
differ-
ence
s
in
the
classification
methods
and
in
the
s
timulus
substances
used.
The
responses
spectra
for
the
receptor
cells
in
the
same
group
were
not
always
identical,
perhaps
because
of
the
the
properties
of
the
receptor
cells.
This
phenomenon
may
be
explained
in
at
least
two
ways,
as
follows:
1)
Multiple
kinds
of
receptor
cells
are
included
in
single
groups.
Each
receptor
cell
has
a
single
receptive
site
different
from
other
receptor
cells.
Differences
between
receptor
cells
are
too
small
to
be
detected
using
our
procedure
of
classification.
2)
A
single
kind
of
receptor
cell
is
included
in
a
single
group,
but
the
receptor
cell
has
multiple
kinds
of
receptive
sites.
The
term
site
is
in
reference
to
a
locus
or
group
of
loci
holding
in
common
the
same
ion
channel.
Similar
results
were
reported
by
Selzer
[10].
Most
stimulus
response
curves
were,
roughly
speaking,
parallel
to
one
another
in
the
single
receptor
cells
and
a
few
were
not
in
parallel
(Figs.
lb,
3b).
This
means
that
the
receptor
cell
has
several
kinds
of
receptive
sites.
Thus,
groups
of
our
classification
may
conform
to
the
following
interpretation:
The
receptor
cells
in
each
group
have
one
common
(predominant)
receptive
site
and
the
other
sites
are
inferior.
The
predominant
site
has
a
low
specificity
and
binds
many
kinds
of
compounds.
The
inferior
sites
vary
among
the
receptors
in
one
group.
These
sites
have
a
high
specificity
so
that
they
bind
only
a
few
kinds
of
compounds.
According
to
this
interpretation,
it
is
the
predominant
site
that
determines
into
which
group
the
receptor
is
to
be
classified.
We
examined
the
response
to
sex-pheromone
extract
in
order
to
physiologically
classify
the
receptor
cells.
The
sex-pheromone
sensitive
cells
responded
only
to
sex-pheromone
extract,
among
the
chemicals
examined.
Though
santalol
has
been
described
as
a
component
of
the
sex-pheromone
mimics
of
Periplaneta
americana
[21]
this
receptor
did
not
respond
to
this
substance.
Accroding
to
Sass
[22],
there
are
two
types
of
pheromone
sensi-
tive
cells
in
the
same
sensillum;
one
is
excited
by
pheriplanone
A
and
the
other
by
pheriplanone
B.
However,
only
one
type
of
cell
was
excited
by
the
sex-pheromone
extract
in
our
experiment.
The
difference
may
relate
to
the
procedure
of
prepara-
tion
of
the
sex-pheromone
extract.
We
examined
the
response
to
some
substances
listed
in
Table
1
which
elicited
large
respones
of
the
receptors
in
254
K.
FUJIMURA,
F.
YOKOHARI
AND
H.
TATEDA
Group
I
to
VIII,
but
the
sex-pheromone
sensitive
cell
was
not
excited
by
any
of
them.
Though
we
did
not
confirm
if
these
pheromone
sensitive
cells
belonged
to
one
of
the
our
8
groups
because
we
did
not
systematically
examine
the
responses
of
the
cells
to
all
the
chemicals
listed
in
Table
1,
the
response
spectra
of
this
receptor
do
not
seem
to
correlate
with
those
of
any
cell
in
Group
Ito
VIII
and
this
receptor
appears
to
be
highly
specific
only
to
the
sex-pheromone.
Relationships
between
cell
groups
and
sensillar
types
According
to
Altner
et
al.
[5],
cells
sensitive
to
n-alcohols
are
located
in
single-walled
sensilla
with
pore
tubules
(corresponding
to
S-I
and
S-II
sensilla
in
our
classification);
the
cells
sensitive
to
n-acids
are
in
the
double-walled
sensilla
containing
secre-
tion
material
in
spoke
canals
(corresponding
to
G-I
and
G-II).
However,
the
cells
sensitive
to
n-
alcohols
(group
I-III)
were
found
only
in
type
S-I
sensilla,
but
the
cells
sensitive
to
n-acids
(Group
VIII)
were
found
in
type
G-II
and
T
sensilla
(Table
2).
Type
G-II
sensilla
belong
to
the
double-walled
group
of
sensilla
and
type
T
sensilla
to
single-
walled
sensilla
in
the
category
of
Altner
et
al.
[5].
On
the
other
hand,
the
cells
in
Group
III
were
excited
considerably
by
a
series
of
n-fatty
acids
but
housed
in
type
S-I
sensilla,
i.e.,
single
walled
sensilla.
Similar
results
were
reported
on
the
cyclohexylamine
cell
observed
by
Selzer
[10].
Therefore,
it
may
be
difficult
to
conclude
that
cells
sensitive
to
alcohol
are
housed
in
single-walled
sensilla
and
those
cells
sensitive
to
n-fatty
acid
locate
in
double-walled
sensilla.
Furthermore,
cells
whose
responses
were
recorded
simultaneous-
ly
from
more
than
two
receptor
cells
show
diffe-
rent
response
spectra.
Cells
in
Group
V
are
often
present
with
cells
in
Group
VI
in
the
same
sensil-
lum.
Consequently,
the
stimulus
conducting
sys-
tem
may
roughly
control
the
access
of
substances
to
the
receptor
membranes,
and
the
discriminatory
properties
of
the
sensory
organs
may
be
a
function
of
the
receptor
membrane.
Speculation
related
to
discrimination
of
odorous
substances
There
are
numerous
odorous
substances
in
the
habitat
of
the
animal
and
possession
of
a
receptor
cell
highly
specific
to
each
of
all
the
substances
is
unlikely.
On
the
other
hand,
it
might
not
be
important
for
the
animal
to
exactly
differentiate
odorous
substances,
except
for
pheromones
but
it
might
be
necessary
to
discriminate
favorable
from
unfavorable
odors.
Most
receptors
of
Periplaneta
americana
respond
to
various
odors.
However,
the
sex-pheromone
must
be
clearly
differentiate
both
from
non-specific
odors
and
from
the
sex-
pheromones
of
the
other
species
for
purposes
of
mating.
The
animal
may
well
have
developed
a
very
highly
specific
receptor
to
its
own
sex-
pheromones.
ACKNOWLEDGMENTS
We
express
our
gratitude
to
Dr.
D.
R.
Stokes
(Emory
University,
USA)
and
Dr.
M.
Ohara
for
correcting
the
English.
They
also
are
greatly
indebted
to
Dr.
Y.
Toh
for
his
encouragement.
This
work
was
supported
in
part
by
grants
from
the
Ministry
of
Education,
Science
and
Culture,
Japan
(No.
574294
to
F.
Y.
etc.).
REFERENCES
1
Schaller,
D.
(1978)
Antennal
sensory
system
of
Periplaneta
americana
L.
Cell
Tissue
Res.,
191:
121-
139.
2
Altner,
H.
and
Prillinger,
L.
(1980)
Ultrastructure
of
invertebrate
chemo-,
thermo,
and
hygroreceptors
and
its
functional
significance.
Int.
Rev.
Cytol.,
67:
69-139.
3
Steinbrecht,
R.
A.
(1984)
Arthropoda:
Chemo-,
thermo,
and
hygroreceptors.
In
"Biology
in
the
Integument
I".
Ed.
by
A.
G.
Bereiter-Hahn,
K.
S.
Matoltsy,
and
K.
S.
Richards,
Springer
Verlag,
Berlin/Heidelberg/New
York,
pp.
523-553.
4
Boeckh,
J.
and
Ernst,
K.
D.
(1987)
Contribution
of
single
unit
analysis
in
insects
to
an
understanding
of
olfactory
function.
J.
Comp.
Physiol.,
A
161:
549-
565.
5
Altner,
H.,
Sass,
H.
and
Altner,
I.
(1977)
Rela
-
tionships
between
structure
and
function
of
antenna'
chemo-,
hygro-,
and
thermoreceptive
sensilla
in
Periplaneta
americana.
Cell
Tissue
Res.,
176:
389-
405.
6
Sass,
H.
(1976)
Zur
nervosen
Codierung
von
Geruchsreien
bei
Periplaneta
americana.
J.
Comp•
Physiol.,
107:
49-65.
7
Sass,
H.
(1978)
Olfactory
receptors
on
the
antenna
of
Periplaneta:
Response
constellations
that
encode
food
odors.
J.
Comp.
Physiol.,
128:
227-233.
Olfactory
Recepotors
of
Periplaneta
255
Sass,
H.
(1980)
Physiological
and
morphological
identification
of
antennal
olfactory
receptors
of
in-
sects.
In
"Joint
Congr
Chemoreception
ECRO
IV-
ISOT
VIII".
Ed.
by
van
der
Starre,
RL
Press,
London
Washington
DC,
pp.
194.
Selzer,
R.
(1981)
The
processing
of
a
complex
food
odor
by
antenna!
olfactory
receptors
of
Periplaneta
americana.
J.
Comp.
Physiol.,
144:
509-519.
Selzer,
R.
(1984)
On
the
specificities
of
antenna!
olfactory
receptor
cells
of
Periplaneta
americana.
Chem.
Senses,
8:
375-395.
Schafer,
R.
and
Sanchez,
T.
V.
(1976)
The
nature
and
development
of
sex
attractant
specificity
in
cockroaches
of
the
genus
Periplaneta.
I.
Sexual
dimorphism
in
the
distribution
of
anternnal
sense
organ
in
five
species.
J.
Morph.,
149:
139-158.
Toh,
Y.
(1977)
Fine
structure
of
antennal
sense
organs
of
the
male
cockroach
Periplaneta
americana.
J.
Ultrastruct.
Res.,
60:
373-394.
Hartmann,
N.
(1987)
Function
and
development
of
sensory
cells
in
'attractant
sensilla':
physiological
differences
between
larvae
and
adults
of
Periplaneta
americana
L.
Chem.
Senses,
12:
210.
Jordan,
T.
E.
(1954)
Vapor
Pressure
of
Organic
Compounds.
Interscience
Publisher,
Inc.,
New
York.
Weast,
R.
C.
(1980)
CRC
Handbook
of
Chemistry
and
Physics.
61st
Ed.
CRC
Press
Inc.,
Boca
Raton,
Florida,
USA.
16
Rust,
M.
K.
(1976)
Quantitative
analysis
of
male
responses
released
by
female
sex
pheromone
in
Periplaneta
americana.
Anim.
Behavior,
24:
681-
685.
17
Tobin,
T.
R.,
Seelinger,
G.
and
Bell,
W.
J.
(1981)
Behavioral
responses
of
male
Periplaneta
americana
to
periplanone
B,
a
synthetic
component
of
the
female
sex
pheromone.
J.
Chem.
Ecol.,
7:
969-979.
18
Yokohari,
F.
and
Tateda,
H.
(1976)
Moist
and
dry
hygroreceptors
for
relative
humidity
of
the
cock-
roach
Periplaneta
americana
L.
J.
Comp.
Physiol.,
106:
137-152.
19
Kafka,
W.
A.
(1970)
Molekulare
Wechselwir-
kungen
bei
der
Erregung
einzelner
Riechzellen.
Z.
vgl.
Physiol.,
70:
105-143.
20
Vareschi,
E.
(1971)
Duftunterscheidung
bei
der
Honigbiene
Einzelzell-Albeitungen
and
Verhalten-
sreaktionen.
Z.
vgl.
Physiol.,
75:
143-173.
21
Bower,
W.
S.
and
Bodenstein,
W.
G.
(1971)
Sex
pheromone
mimics
of
the
American
cockroach.
Nature,
232:
259-261.
22
Sass,
H.
(1983)
Production,
release
and
effective-
ness
of
two
female
sex
pheromone
components
of
Periplaneta
americana.
J.
Comp.
Physiol.,
152:
309-
317.
IMP