Published August 1, 1969
Function of Insect Compound Eyes
Containing Crystalline Tracts
K J E L L B. D O V I N G and W I L L I A M H. M I L L E R
From the Section of Ophthalmology, Yale University School of Medicine, New Haven,
Connecticut 06510. Dr. D~ving's present address is Institute of Zoophysiology,University
of Oslo, Blindern, Oslo, Norway
T h e classical t h e o r y o f the function of the a r t h r o p o d c o m p o u n d eye was proposed b y J o h a n n e s Mfiller in 1826. Mfiller postulated t h a t e a c h o m m a t i d i u m
functions to m o n i t o r the intensity of the light f r o m the d i r e c t i o n t h a t it faces.
T h i s mosaic t h e o r y was generally a c c e p t e d for all c o m p o u n d eyes until 1891
w h e n S i g m u n d E x n e r p u b l i s h e d a m o n o g r a p h o n the physiology o f the c o m p o u n d eyes of crustaceans a n d insects. H e classified the c o m p o u n d eyes in
two groups a c c o r d i n g to their function: the apposition a n d the superposition
eyes. T h e superposition eyes are c h a r a c t e r i z e d b y c e r t a i n a n a t o m i c a l features,
the most i m p o r t a n t of w h i c h is t h a t the r h a b d o m e r e s are s e p a r a t e d f r o m t h e
crystalline cones b y considerable distances. T h e r h a b d o m e r e s a n d cones are,
however, c o n n e c t e d b y crystalline tracts. W h e n E x n e r e x a m i n e d these eyes to
25°
The Journal of General Physiology
Downloaded from on June 14, 2017
A~STRACT Image formation is studied in compound eyes of insects that contain crystalline tracts. In optical experiments the course of light is studied in
fresh scalps of dark-adapted eyes using point and extended sources. In the tract
region a point source gives a diffusely lighted area within which are punctate
spots about 10 times brighter. Because the position of these spots does not change
when the source is moved, and because their spacing agrees with estimates
based on the known scalp depth, we conclude that these spots represent light
radiating from the cut ends of tracts. An extended source gives a dim erect
image in the tract region that may come from the pattern of illumination radiating from the cut ends of the tracts. In electrophysiological experiments intracellular microelectrode recordings of responses to illumination are made from
single retinular cells of the skipper, Epargyreus clarus, an animal that lacks iris
pigment. Measurements of visual fields of single retinular cells by three methods give half-power beam widths of about 2 ° . Though not conclusive, these
experiments suggest that only the light contained in the tract is effective in
stimulating the retinular cell. This agrees with the theoretical study of Allen
(1968) and is inconsistent witll the superposition theory of Exner (1891) as
applied to certain moth and skipper eyes.
Published August 1, 1969
K. B. D¢vxNo AND W. H. MILLER InsectCompoundEyeswith CrystallineTracts
251
METHODS
Anatomy
The anatomy of the compound eyes is studied with the light microscope. The eyes
are fixed in 1% glutaraldehyde in phosphate buffer, pH 7.2, and embedded in
Maraglass. Thick sections of these eyes are examined using the light microscope in
order to measure the appropriate dimensions of the optical apparatus.
Optical Experiments
Completely dark-adapted living moths of the species Hyalophora cecropia, Antheraea
polyphemus, and Chaerocampaelpenor, and the skipper, Epargyreus clarus, are used. The
animals are decapitated and scalps of the eyes are made with a razor blade. The
scalps are transferred to a drop of 1% solution of glutaraldehyde in phosphate buffer
at pH 7.2 on a cover glass. The cover glass is mounted in a moist chamber with the
Downloaded from on June 14, 2017
study their function, he removed the rhabdomeres and the crystalline tracts
and substituted glycerine solution. In such cleaned out Lampyris eyes Exner
observed a superposition image at the level of the rhabdomeres. H e also observed a similar b u t poorer image in dark-adapted moth and crustacean eyes.
De Bruin and Crisp (1957) showed that the crystalline cones and tracts in
some crustacean eyes form optically continuous strands with a higher refractive index than the surrounding m e d i u m and they suggested that these structures function as wave guides.
T h e superposition theory was also questioned by N u n n e m a c h e r (1959,
1960) and Kuiper (1962). T h e y examined m a n y crustacean eyes and failed to
detect any superposition images. Kuiper did, however, confirm Exner's observations even to the finest details when he repeated the experiments on
Lampyris. N u n n e m a c h e r studied scalps of frozen eyes of Lampyris and saw a
superposition image, b u t it always fell behind the rhabdomeres. Recently,
Horridge (1968) has found that the tracts in the firefly, Photuris, act as light
pipes.
T h e previous studies have left a n u m b e r of unresolved questions about the
optical function of eyes with tracts. These include the problems of whether or
not the crystalline tract actually is a light pipe in dark-adapted eyes, how
m u c h light, if any, is outside the tracts, and what is the path of light that
stimulates the retinular cells. In an effort to resolve these problems we have
been working with Bernard and Allen (Allen and Bernard, 1967; Allen, 1968;
Miller, Bernard, and Allen, 1968) on theoretical, optical, anatomical, and
physiological studies of various tract eyes. In a theoretical and experimental
optical study of the tobacco hornworm moth compound eye Allen (1968)
found that an image is focused at the distal end of the tract and that the tract
is a light pipe. In the present investigation, optical and physiological experiments are reported that tend to support the light pipe theory.
Published August 1, 1969
252
THE
JOURNAL
OF
OENERAL
PHYSIOLOGY
• VOLUME
54
" I969
eye looking down and positioned in the center of a universal stage. The universal
stage is mounted on a Leitz compound microscope. The eye is illuminated with
collimated light from a microscope lamp with a 15 w incandescent bulb (running at
5 v and 2 amp) mounted under the preparation at a distance of 70 cm.
The lamp together with its filter holder and diaphragm is attached to a pivot arm
with its center of rotation at the preparation. A diagram of the experimental setup
is shown in Fig. 1. Most observations are made with a 550 nm interference filter and
a 2.0 log unit neutral density filter in the light b e a m Scalps are also examined
mounted on a front-surface mirror and under a dissecting microscope with the light
coming from the same direction as viewing.
Downloaded from on June 14, 2017
L
Fzoum~ 1. Schematic diagram of experimental
setup for optical experiments. Full explanation in
text.
El ectrophysi ology
All experiments reported here are on the silver spotted skipper Epargyreus darus. The
animal is decapitated and the head mounted with beeswax on a "jolter" (Tomita,
1965). A small hole is made in the cornea with a razor blade and then sealed with
silicone grease. Cell penetrations are made with glass microelectrodes filled with 2 N
KCI with a resistance between 100 and 200 Mohms. The potential changes in the
cells following illumination are measured with a precision electrometer (M4, W-P
Instruments, Hamden, Conn.) and displayed on an oscilloscope screen.
The light stimulus comes from a fiber optics bundle mounted on a perimeter at a
distance of 33 cm. The light source subtends an angle of about 0.1 o. A Xenon lamp
Published August 1, 1969
K. B. D~vINc AND W. H. MILLER Insect CompoundEyes with Crystalline Tracts
253
(Bausch and Lomb) serves as the light source. The light stimulus is controlled by a
series of neutral density filters (Kodak) and a shutter and is monitored with a silicon
photocell placed just under the eye.
The skippers are caught in the rural area around New Haven and used as soon as
possible. However, satisfactory recordings can he made from animals kept for 2-3
days at + 6 ° C . The animals are decapitated in dim light and prepared in red light.
As soon as a successful penetration is made, we determine the center of the visual
field and make recordings with the light in different positions, usually every 0.5 ° across
the entire field. The size of the visual field is determined in two ways. One is to measure the size of the response at the center of the visual field and find the angular
position off center that gives the same response at twice the intensity (half-power
point). The other is to find the light intensity that gives a threshold response (2 3
mv) at the center of the visual field and to increase the intensity by small steps. At
these increased intensities we determine the positions at which the increased light
gives the threshold response. The angular positions can be determined with an accuracy of -4-0.1 o
RESULTS
Exner's Firefly Experiment
T h e interior soft p a r t s of the firefly eye are carefully r e m o v e d w i t h a s h a r p e n e d
w o o d e n toothpick, so t h a t o n l y the c o r n e a w i t h its a t t a c h e d crystalline c o n e
remains. T h e hole is t h e n filled w i t h a 14 % a q u e o u s solution of glycerine,
n = 1.35. W e m o u n t the eye as s h o w n in Fig. 1 a n d d e s c r i b e d u n d e r M e t h o d s
a n d w i t h a 10 × o b j e c t i v e we o b s e r v e a n d p h o t o g r a p h the p a t t e r n of light at
different distances f r o m the c o r n e a as i n d i c a t e d in Fig. 2. O u r results c o n f i r m
E x n e r ' s a n d K u i p e r ' s observations. A s u p e r p o s i t i o n i m a g e of the light source
is f o r m e d at the level of the r h a b d o m e r e s as seen in Fig. 3 E. W h e n the focal
p l a n e is lifted a b o v e this level, the light rays f r o m the crystalline cones d i v e r g e
f o r m i n g a n h e x a g o n a l p a t t e r n (Fig. 3 F).
I f the p o i n t source s u b t e n d i n g an a n g l e of 0.6 ° is r e p l a c e d b y an e x t e n d e d
source w h i c h is a r u n i c R s u b t e n d i n g 4 ° lengthwise, we see small i n v e r t e d
i m a g e s 30 # distal to the tip of the crystalline cones (Fig. 4 A) a n d a n erect
s u p e r p o s i t i o n i m a g e at a d i s t a n c e of 340/a f r o m the front surface of the c o r n e a
(Fig. 4 g).
Downloaded from on June 14, 2017
FIOURE 2. Schematic diagram of section through firefly compound eye. Letters A-F
refer to level of plane of focus of microscope for pictures of Fig. 3.
Published August 1, 1969
254
THE
JOURNAL
OF
GENERAL
PHYSIOLOGY
- VOLUME
54
"
~969
FIGURE 4. Preparation same as Fig. 3 but with extended source. A, small inverted
images of runic R within the crystalline cones. B, erect superposition of image 340
from front surface of cornea.
Optical Experiments on Scalps of Fresh and Fixed Eyes
W e p r e p a r e scalps of the m o t h a n d skipper eyes as described u n d e r M e t h o d s .
U s i n g d i m light to p r e v e n t light a d a p t a t i o n , we study the light f r o m the p o i n t
source. P r o x i m a l to the crystalline cones there is a diffuse area of light t h a t
converges t o w a r d the back of the eye. T h i s diffuse spot of light is illustrated in
Fig. 5 B. T h e m i c r o s c o p e is focused on the p l a n e of section of the eye; the
Downloaded from on June 14, 2017
Fmum~ 3. Patterns of light from point source photographed in the planes indicated
by the arrows and letters on Fig. 2. Compound eye of the firefly, Photuris, prepared as
described in the text. Marker, 200 #.
Published August 1, 1969
K. B. D~WNC ANt) W. H. MILLER Insect CompoundEyes with Crystalline Tracts
255
d e p t h of the scalp is 630/~. T h e d e p t h of the scalp is measured by inverting the
p r e p a r a t i o n a n d focusing first on the corneal surface a n d then on the cover
slip. T h e d i a m e t e r of the diffuse spot seems to be larger in eyes with greater
radii of c u r v a t u r e a n d with larger sources of light.
In elpenor and polyphemus this diffuse area of light is smallest in front of the
r h a b d o m e s within the tract region. By careful m a n i p u l a t i o n of the universal
stage and by placing the light in different positions, small spots of light are
seen at the level of the cover slip, i.e., at the plane of sectioning the eye. T h e
FmURE 6. Scalp of compound eye of elpenor. A, illuminated with point source (see
tract at arrow). B, illuminated by extended source with superimposed runic R. Marker,
1 mm.
a l i g n m e n t of all parts of the optical p a t h is quite critical. T h e spots are approximately 1 log u n i t brighter t h a n the i m m e d i a t e surround (Fig. 5 A).
Occasionally by compromising, it is possible to illuminate two spots at once.
It is also possible to walk the spot from one place to another, either by m o v i n g
the light source or by rotating the universal stage.
Downloaded from on June 14, 2017
FICURE 5. Scalp of compound eye of elpenor. A, spot of light presumably radiating from
cut end of tract (arrow). B, same experiment as A but with 10 times the exposure length
to show diffuse halo. Marker, 1 mm.
Published August 1, 1969
TIlE JOURNAL
256
OF G E N E R A L
'FABLE
DIMENSIONS
OF
OF
THE
SOME TRACT-CONTAINING
FIELDS AS DETERMINED
Corneal
facet
diameter
Cone
length
PHYSIOLOGY
54
'
x969
I
OPTICAL
APPARATUS
EYES AND
BY O P T I C A L
Tract
length
• VOLUME
Rhabdom
length
#
THEIR VISUAL
METHODS
Radius of
curvature
Interommatidial
angle
/t
#
/z
Photuris
28
106
235
110
970
1.6
Antheraea polyphemus
Hyalophora cecropia
Manduca sexta
Epargyrus clarus
32
46
570
138
1170
1.6
31
29
100
100
600
900
160
200
1500
2000
1.2
0.8
27
100
400
200
1200
1.3
Visual field
tt
3.3
~2.0
4-5
Downloaded from on June 14, 2017
FIGURE 7. Scalp of elpenor after optical experiment. Phase microscopy. Note accordionlike deformation of tracts (arrow) and pigment position corresponding to dark-adapted
state. C, cornea. Marker, 25 /z.
W h e n the d e p t h of the scalp and the calculated i n t e r t r a c t distance at this
d e p t h are taken into consideration, we c o n c l u d e t h a t these spots of light represent the a p p e a r a n c e of the tracts viewed end on. T h e h e x a g o n a l p a t t e r n of the
tracts can be b r o u g h t out by c h a n g i n g the direction of the light.
Published August 1, 1969
K. B. DfVlNG AND W. H. MILLER Insect Compound Eyes with Crystalline Tracts
257
W h e n an extended source is substituted for the point source, we observe an
erect image as shown in Fig. 6. This image, however, is very dim and varies in
size according to the depth of the scalp; the shallower the scalp the bigger the
image. Also, this image can be seen only at the plane of section of the eye.
This indicates that the image is formed by light radiating from the ends of the
crystalline tracts. Another indication that the spots we observe come from the
tracts is that when the point source light position is changed, the spot position
remains the same; but it grows brighter or dimmer until it disappears entirely.
This experiment also allows one to estimate the total beam width for an
ommatidium. By inspection we find visual fields of the order of 2-5 o as shown
in Table I.
T h e appearance of one of the preparations after an optical experiment as
seen in a 1 # thick and unstained section viewed by phase microscopy is shown
in Fig. 7. This picture shows that the tracts tend to keep their normal relationships with the cones and surrounding material, although some of them tend
to contract in the m a n n e r of a released spring (arrow). Most important, this
section shows that the pigment has not moved and is in the fully dark-adapted
state. Although the histology is not done routinely, we always float the scalp
on a front-surface mirror after an experiment. W h e n the eye shows a full glow
(Fig. 8) viewed with light coming from the direction of viewing, we assume
that the distal iris pigment is still in the dark-adapted condition. It has been
shown that pigment migration towards the light-adapted position obliterates
the glow (HSglund, 1966).
E l ectr op hy si ol ogy
For our purposes the skipper compound eye is a desirable physiological preparation because although the eye has crystalline tracts surrounded by "iris"
cells, these iris cells do not contain pigment. This is illustrated by the phase
Downloaded from on June 14, 2017
FIGURE8. Appearance of Polvphemus scalp on front-surface mirror that is illuminated
from the direction of viewing. Presence of glow indicates that iris pigment is in the darkadapted position. Marker, 1 mm (from Allen, 1968).
Published August 1, 1969
258
THE
JOURNAL
OF
GENERAL
PHYSIOLOGY
•
VOLUME
54
"
I969
contrast light m i c r o g r a p h of Fig. 9. N o t e t h a t p i g m e n t is located b e t w e e n the
crystalline cones. T h i s p i g m e n t does not a p p e a r to m o v e w h e n the eye is
exposed to b r i g h t light and fixed in the light, b u t it is difficult to rule out the
possibility of r a p i d p i g m e n t m o v e m e n t s over short distances.
Downloaded from on June 14, 2017
FIGURE 9. Section of skipper eye as viewed using Zernike phase contrast system. Eye
was fixed in brightly lighted surroundings. No pigment is located in iris cells around
tracts. G, cornea; T, tract region; N, nuclei of retinular cells; R, rhabdoms. Note pigment between crystalline cones and between rhabdoms. Marker, 100 #.
T h e p e n e t r a t i o n of single cells in the skipper eyes is facilitated b y T o m i t a ' s
(1965) j o l t e r device. A t the b e g i n n i n g of the d o w n w a r d m o v e m e n t of the
electrode we f r e q u e n t l y p e n e t r a t e cells ( p r e s u m a b l y iris cells) t h a t give negative potentials of the o r d e r of 30 m y w h e n the eye is illuminated. T h e size of
the potential varies with the position of the light, b u t responses can be evoked
over large areas indicating t h a t these potentials are m e r e l y p i c k u p of the
r e c e p t o r potential in the p e n e t r a t e d cells.
Published August 1, 1969
K. B. DCVIN~ AND W. H. MILLER Insect CompoundEyes with Crystalline Tracts
259
G o o d p e n e t r a t i o n s of r e c e p t o r cells are i n d i c a t e d by a b r u p t negative
changes in DC potential of the o r d e r of 40 mv. W h e n stimulated with light,
these cells d o not show a n y sign of h y p e r p o l a r i z a t i o n , b u t give a d e p o l a r i z i n g
response w h e n light strikes the eye f r o m the right direction. W e usually search
for the visual field of single r e t i n u l a r cells using a 3.0 v n a k e d flashlight b u l b
t h a t is r u n at 1.5 v. T h e sharp d e p o l a r i z a t i o n we observe w h e n the light passes
the visual field is d r a m a t i c . T h e responses resulting f r o m w a v i n g the b a r e b u l b
FIGURE 10. Intracellular recording from skipper eye, presumably from a retinular cell
showing depolarizing responses caused by movement of bare flashlight bulb back and
forth across visual field of cell. Vertical marker, 5 mv; horizontal marker represents
angular movement of about 10°.
0.4
0.8
1.0
2.0
1.6
._.U'x--
mm.mmmp~ -v ~e ~mmr~mm*~s
J
m.4
I.I
8.(I
it. 8 ,
'
~
$.O
....j.
J
-
I
FIGURE 11. Intensity series recording of intracellular response. Top record, intracellular response. Bottom record, photocell. Numbers give relative intensities of illumination
in log units. Stimulus duration, 0.2 sec; vertical marker, 10 mv.
After this coarse d e t e r m i n a t i o n of the position of the visual field in space,
we position the fiber optics a p p r o x i m a t e l y a n d m o v e the light back and forth
carefully until we find the c e n t e r of the visual field.
In some instances we take pictures of the responses w i t h different light
intensities. A n e x a m p l e of an intensity series is shown in Fig. 11. W h e n the
light intensity increases a b o v e a certain level, an initial transient b e c o m e s evid e n t as shown at 2.4 log units intensity. T h e g r a p h of a n o t h e r intensity series
is shown in Fig. 12. T h e a m p l i t u d e s of the peak and the steady-state responses
Downloaded from on June 14, 2017
back a n d forth across an active visual field are shown in Fig. 10. A r o u g h
calculation based on the estimated speed of the bulb m o v e m e n t gives a b o u t 5 °
for the total b e a m w i d t h as so measured. A l t h o u g h this is a c r u d e measurem e n t , it is i m p o r t a n t because it is m a d e u n d e r circumstances t h a t almost
c e r t a i n l y rule out a n y effects of possible m i g r a t i o n of the p i g m e n t a r o u n d the
crystalline cones t h a t could result f r o m light a d a p t a t i o n .
Published August 1, 1969
260
THE
JOURNAL
OF
GENERAL
PHYSIOLOGY
• VOLUME
54
" I969
are plotted against the light intensity. For the higher intensities the discrepancy in amplitude between the peak and the steady state becomes evident,
and amplitude in the steady state reaches a m a x i m u m at around 3 log units
for this particular case.
A coarse indication of the receptive field of an individual retinular cell can
be obtained by stimulating the eye at different angles of incidence. Usually we
start about 5 ° off-center and expose the eye for every 0.5 ° across the entire
30
2O
o
O_
v~
~0
°o
I
LOG
2
INTENSITY
3
FmURE 12. Graph of initial transient (filled circles) and steady-state (open circles)
intracellular response amplitudes as a function of intensity of stimulus for a skipper
retinular cell.
-@
~8
.1
'
'
o I
s
I
II
s
li'
FIGURE 13. Intracellular responses from skipper eye as a function of point source
position. Stimulus duration, 0.2 sec; vertical marker, 20 mv.
field to 5 ° off-center on the other side. T h e design of the perimeter makes it
only possible to search the visual field in one direction, but since this direction
would vary for the different cells under investigation and no great differences
were found in visual fields, we assume that the visual fields are roughly symmetrical around the axes of the ommatidia. An example of the responses
evoked by stimulating at different angles with the same intensity of light is
shown in Fig. 13. In this case the initial peak is only evoked when the light is
close to the center of the field. Outside 4-3 ° no response can be observed. T h e
amplitude of the responses is plotted against the angle of incidence for
Downloaded from on June 14, 2017
hl
V~
Z
Published August 1, 1969
K. B. DCvmo Am) W. H. MxL~R
InsectCompoundEyes with Crystalline Tracts
26z
another cell in Fig. 14. T h e initial peak reaches about 21 mv in this case while
the m a x i m u m steady-state response is 12 my. Since the half-power points are
70 % of the m a x i m u m response, the visual field between half-power points is
2.1 o when using the initial peak as a criterion and 2.5 ° when using the amplitude of the steady state. Fig. 14 also demonstrates a frequently observed
phenomenon, namely that at 3-5 ° off-center the light stimulus produces a
25
2O
E
u')
Z
21c
t/3
W
n~
oo
POSITION
FIOURE 14. Intracellular response as a function of light position for initial transient
(filled circles) and steady state (open circles). Hyperpolarizing response is characteristic
of deteriorating preparation.
>~
>3
l-Z
I.IJ
(.92
0_ J
Fio~
15. Sensitivity of intraceUular response
as a function of light position.
°
o
POSITION
hyperpolarization of the m e m b r a n e potential. This is not observed in very
good penetrations with high and stable m e m b r a n e potentials, but it is seen
when the cells start to depolarize or in penetrations which do not show a good
m e m b r a n e potential. If large depolarizations occur, we observe only hyperpolarization responses that have the same general appearance as the responses
from the iris cells described above.
As mentioned above, threshold experiments give the most consistent results
for determining the visual fields of single retinular cells. W e also find this
Downloaded from on June 14, 2017
-bV
Published August 1, 1969
262
THE
JOURNAL
OF
GENERAL
PHYSIOLOGY
• VOLUME
54
"
I969
method easier than any other way of describing the visual field. A definite
advantage of this threshold measurement is that the particular cell under investigation is never strongly light-adapted. A threshold response of 9-3 mv is
evoked in the center of the field. T h e illumination is increased stepwise and the
position for a threshold response is determined at each step. In plotting the
values of light intensity which evoke a threshold response we use the expression
log sensitivity. This is the logarithm of the inverse of the light intensity actually
used. An example of such a graph is given in Fig. 15. F r o m such a graph the
half-sensitivity angle is easily determined and in this case is found to be 1.9 °.
For some units which could be held for a sufficient length of time both
threshold determinations and the amplitude of the response with different
angles of incidence for constant intensity are determined. An example of such
75
>I>
F~2
hJ
Z
hl
5o
I
u~
W
r~
(.9
O
.-I
25
I
o
FIGURE 16. Visual field of same skipper retinular
cell measured by the amplitude response (open
circles) and threshold methods (filled circles):.
POSITION
an experiment is shown in Fig. 16. T h e half-sensitivity angle as determined by
the threshold is 2.1 o, compared to 1.5 ° which is the half-power b e a m width
as determined by 70 % of the m a x i m u m response.
DISCUSSION
T h e optical experiments show that it is possible to obtain a superposition
image in the preparation used by Exner (1891) and Kuiper (1962) in which
both the rhabdomeres and the crystalline tracts are removed. A similar but
less well-defined picture of convergence at the retinal level is also obtained in
the freshly cut eyes of moths and skippers. Simultaneously one observes small
light spots at the calculated interommatidial distances corresponding to the
depth of the scalp. We conclude that these light spots represent the appearance
of the tracts as viewed end on. I n dark-adapted moths, we see an erect image
with the extended source; however, it is only seen at the level of the section
and increases in size in shallower scalps. We therefore conclude that the image
Downloaded from on June 14, 2017
I00
Published August 1, 1969
K. B. DevINe AND W. H. MILLER InsectCompoundEyes with Crystalline Tracts
263
Downloaded from on June 14, 2017
is formed from the conduction of light in the tracts and observed as the light
radiates from the tracts at the plane of the section.
In the original superposition theory formulated by Exner the eyes are postulated to form a superposition image only when dark-adapted. It is therefore
imperative to keep the eyes of the moths in the dark-adapted stage and great
care is taken to do this. A definite advantage of the skipper eye is that the
distal pigment is confined to the area around the cones; we have not observed
any changes in its position. T h e r e is no pigment migration as seen in the
moths because the cells surrounding the tracts are devoid of pigment. T h e
glow size and intensity of the skipper (Miller and Bernard, 1968) have not
been observed to change upon exposure for m a n y minutes to bright lights.
O u r optical experiments show that m u c h of the light is confined to the
crystalline tracts, but part of it is more or less diffusely transmitted outside
these structures. O u r electrophysiological experiments show, however, that the
light outside the tracts most probably does not enter the rhabdomeres and
excite the retinular cells. These findings are in agreement with the theoretical
study of light propagation in the m o t h eye conducted by Allen (1968). H e
measured the refractive index of all the optical apparatus in the tobacco hornw o r m moth, Manduca sexta, and on the basis of these experimental data calculated the light paths in the ommatidia. His results are in agreement with our
conclusions in that he predicts that for light incident on a facet within 3 °,
about 80 % of the light is transmitted in the crystalline tract. T h e half-power
b e a m width was found to be about 2 ° which coincides with the experimental
value found in the present physiological and optical study.
Electrophysiological experiments have been used to determine visual fields
in m a n y arthropods. In Calliphora Burkhardt et al. (1965) showed that the
visual field half-power b e a m width varies between 3.3 ° and 5.2 °. Vowles
(1966) showed that the half-power b e a m widths in Musca eyes vary from 2.5 °
to 3.0 ° in light-adapted and from 4.5 ° to 8.5 ° in dark-adapted eyes. T h e first
figures are horizontal field widths and the latter are vertical field widths.
Kirschfeld (1965) found the half-power b e a m width in Musca to be 7.7 °. In
Limulus the half-power b e a m widths have been determined by W a t e r m a n
(1954) and Kirschfeld and Reichardt (1964) and found to be 11.6 °.
I n Table II the visual field angles as observed with different methods in
different animals are tabulated. In all the apposition eyes the half-power b e a m
width (the angle at half the sensitivity) is m u c h larger than that found in the
skipper eye by electrophysiological means or in the moths and skipper by
optical experiments. It is of special interest that in the skipper no responses
whatsoever are observed electrophysiologically outside 2-4 ° off the center of
the visual field. This is in contrast to the findings in Musca and Calliphora by
Washizu et al. (1964), Kirschfeld (1965), and Vowles (1966) who observed
responses (though small) as far as 20-40 ° off axis. A probable explanation of
Published August 1, 1969
THE J O U R N A L OF G E N E R A L P H Y S I O L O G Y • VOLUME 54
264
" I969
these responses is that the screening pigment is transparent to red (Goldsmith,
1965). Since they used white light, some of the energy in the red part of the
spectrum m a y reach the retinular cell through adjacent ommatidia.
We determine the visual field of the single retinular cells of the skipper eyes
to be 2 . 1 ° 4 - 0.3 °, and because of the symmetry of the optical apparatus there
is no reason to believe that the visual field is different in different directions.
T h e visual field of the skipper and also of the m o t h eyes seems to be smaller
than that of any apposition eye studied so far. As pointed out in the Results,
TABLE
VISUAL FIELD
Animals
II
ANGLES FOR VARIOUS
Interommatidial angle
INSECT EYES
Visual field
Author
Optical
Locusta
Dixippus
Calliphora
H o r i z o n t a l 2.8 °
V e r t i c a l 1.4 °*
6.8 °
H o r i z o n t a l 7.2 °
V e r t i c a l 6.7 °
H o r i z o n t a l 2.4 °
V e r t i c a l 1.10 °*
7.5 °*
H o r i z o n t a l 4.0 °
V e r t i c a l 3.9 °
9.8 °
2.5 °*
7.6 °
K u i p e r (1962)
Autrum and
Wiedcmann
Autrum and
Wiedemann
Autrum and
Wiedemann
Autrum and
Wiedemann
(1962)
(1962)
(1962)
(1962)
Electrophysiological
Calliphora
Musca
Musca
H o r i z o n t a l 2.0-4.6 °
V e r t i c a l 2.6-3.4 °
H o r i z o n t a l 3.9 °
V e r t i c a l 2.3 °
H o r i z o n t a l 3.9 °
V e r t i c a l 2.3 °
Limulus
4-15 °
6.7 °
Epargyrus clarus
1.3 °
F r o n t a l 3.3 °
L a t e r a l 5.2 °
7.7o4_2.1o
Light-adapted,
2.5-3.0 °
Dark-adapted,
4.5-8.5 o
15 °
12 °
2.1°-4-0.3 °
B u r k h a r d t (1965)
K i r s c h f e l d (1965)
V o w l e s (1966)
W a t e r m a n (1954)
Kirschfeld and
K e i c h a r d t (1964)
Present study
* Q u o t e d b y A n t r u m a n d W i e d e m a n n f r o m v a r i o u s sources.
the interommatidial angle is also less for these eyes than for most apposition
eyes.
In Musca the ommatidia have separate rhabdomeres and these tend to be
differentially exposed to light from a point source (Kirschfeld, 1967). I n both
Musca and Calliphora the half-power b e a m width for a single retinular cell in
dark-adapted eyes is 7.8 °. T h e crystalline cones in fly eyes are 50 # long or
about one-third of the cone length in the skipper and in most moths. This
might account for part of the difference in visual fields. T h e Airy disc formed
by a point source would be expected to move a greater distance for increasing
Downloaded from on June 14, 2017
Apis
Apis
Published August 1, 1969
K. B. Derma
~a~u
W. H. MXLL~.R InsectCompoundEyes with Crystalline Tracts
265
Downloaded from on June 14, 2017
angles of incidence in the moth and skipper than in the fly with the result that
the image of the point source at the proximal end of the cone would only be
located on the tract for angles of incidence very close to on-axis.
Although the explanation given above is plausible, there are other possibilities that could explain the narrow visual fields we find in eyes containing
tracts. An ideal superposition image as defined by Exner should show such
narrow fields. In the absence of other information, the physiological experiments reported here cannot be taken as proof of image formation by tract
propagation. However, we know that the superposition image of Exner's experiment in the firefly is far from ideal. It is coarse and covers many ommatidia. In addition, the m e d i u m between the cones and the r h a b d o m is not
isotropic. When the theoretical and optical evidence (Allen, 1968; Miller et
al., 1968) and the optical evidence presented here are taken into consideration,
our physiological experiments are suggestive but not proof of tract propagation. Another unassessed factor which could contribute to the narrow fields in
the skipper is the presence of pigment that optically isolates the sensory parts
of the ommatidia.
T h e responses of the retinular ceils in the skipper eye are similar in form and
appearance to those observed in Musca and CaUiphora (Washizu et al., 1964;
Kirschfeld, 1965; Vowles, 1966; ~.nd Burkhardt, 1965). Although the rise time
of the initial peak in the skipper is fast, we do not observe any spike-like activity such as Naka and Eguchi (1962) found in the eye of the drone of the
honeybee. In many cells we find a negative response when the light is 2-4 °
outside the center of the visual field. These negative potentials of the retinular
ceils are only observed in ceils which show a low membrane potential or depolarization after penetration. T h e negative responses could be seen when
light hit the eye from 20 ° off-axis of the visual field. These negative responses
from the obviously damaged cells looked very m u c h like the responses from
the iris cells, and we believe that these potentials are extracellular pickup of
the activity from neighboring receptor cells. We do not know whether there
is any neural interaction between the ceils in different ommatidia, since we
did not specifically look for it in this study.
T h e evidence for tract propagation reported in this study makes it necessary
to reconsider the mechanism of dark adaptation in certain insect compound
eyes. In the dark-adapted position the distal screening pigment is situated
around the crystalline cones. During light adaptation the pigment migrates
toward the rhabdoms (Bernhard and Ottoson, 1960 a, b). During the adapting
process the sensitivity of the eye is decreased by 6 log units. T h e pigment
migration is responsible for about 3 decades of the total sensitivity decrease
(H6glund, 1966).
Kuiper (1962) suggests that the pigment has a high refractive index and its
presumed contact with the wall of the crystalline tract should frustrate total
internal reflection. Allen (1968) suggests that since in a dielectric wave guide
Published August 1, 1969
266
THE
JOURNAL
OF
GENERAL
PHYSIOLOGY
• VOLUME
54
"
1969
We thank Gary D. Bernard and John L. Allen for helpful contributions to this work.
This investigation was supported in part by a Public Health Service International Postdoctoral
Research Fellowship (1 F 05 YW 1209-01), United States Public Health Service research grant
NB 05730, and by the Connecticut Lions Eye Research Foundation, Inc.
Receivedfor publication 10 March 1969.
REFERENCES
ALLEN, J. L. 1968. The opdcal functioning of the superposidon eye of a nocturnal moth.
Thesis, Massachusetts Institute of Technology.
ALLEN, J. L., and G. D. BErnhARD. 1967. Superposidon optics, a new theory. Massachusetts
Institute of Technology, Research Laboratory of Electronics. Quart. Progr. Rep. No. 86.
113-122.
AUTRUM, H., and I. WmDEMANN. 1962. Versuche uber den Strahlengang im Insekten. Z.
Naturforsch. 17B-" 480.
BERNHARD, C. G., and D. Oa'rosoN. 1960 a. Comparative studies on dark adaptation in the
compound eyes of nocturnal and diurnal Lepidoptera. J. Gen. Physiol. 44:195.
BE~HARD, C. G., and D. Oa'TOSON. 1960 b. Studies on the relation between the pigment
migration and the sensitivity changes during dark adaptation in diurnal and nocturnal
Lepidoptera. J. Gen. Physiol. 44:205.
DE BRUIN, G. H. P., and D. T. CRISP. 1957. The influence of pigment migration on the vision
of higher erustacea, or. Exp. Biol. 34:447.
BURKrtARDT, D., I. Dr LA MOT'r, and G. SEITZ. 1965. Optical properties of the insect eye. Proceedings of the Symposium on Information Processing in Sight Sensory Systems, Pasadena,
California. 75.
EXNER, S. 1891. Die Physiologie der facettirten Augen yon Krebsen und Insecten. Franz
Deuticke, Vienna.
GOLDSM~WH,T. H. 1965. Do flies have a red receptor? or. Gen. Physiol. 49:265.
H/SGLUND, G. 1966. Pigment migration, light screening and receptor sensitivity in the compound eye of nocturnal Lepidoptera. Acta Physiol. Stand. 69(Suppl.):282.
Downloaded from on June 14, 2017
some of the light energy extends outside of the tract boundary, lossy pigment
surrounding the tract would absorb a portion of the energy outside the tract
and reduce the amount of energy reaching the rhabdomeres. Theoretical calculations have shown that this mechanism could account for the reduction in
sensitivity observed in the m o t h eye (Allen and Bernard, 1967).
O u r optical experiments on moth and skipper eyes indicate that light is
largely confined to the tracts and that these structures function as wave guides.
T h e total estimated visual field of tract propagation as indicated by the
measurement of moving the light is about 2-5 °. In our electrophysiological
experiments with the optical apparatus intact it could be ascertained that
light enters the retinular cells only from the direction that the o m m a t i d i u m
faces. T h e visual field is m u c h smaller than found for any other insects studied
so far. This therefore seems to be additional evidence for tract propagation.
O u r findings are in agreement with the theoretical calculations of light paths
based upon experimental studies of the refractive index of various parts of the
optical apparatus (Allen, 1968; Miller et al., 1968). These calculations, together with the results of the present experiments, are in conflict with the
superposition theory proposed by Exner (1891).
Published August 1, 1969
K. B. D~VXNO AND W. H. MILLER Insect Compound Eyes with Crystalline Tracts
267
Downloaded from on June 14, 2017
HORRmGE, G. A. 1968. Pigment movement and the crystalline threads of the firefly eye.
Nature (London). 218:778.
KIRSCHFELD, K. 1965. Das anatomische und das physiologische Sehfeld der Ommatidien im
Komplexauge von Musca. Kybernetik. 2:249.
KmSCHFELD, K. 1967. Die Projektion der optischen Umwelt auf das Raster der Rhabdomere
im Komplexauge yon musca. Exp. Brain Res. 3:248.
I~RSCHFELD, K., and W. REICHARDT. 1964. Die Verarbeitung station/irer optiseher Nachrichten im Komplexauge von Limulus. Kybernaik. 2:43.
KUXPER,J. W. 1962. The optics of the compound eye. Syrup. Soc. Exp. Biol. 16:58.
MILLER, W. H., and G. D. BERNARD. 1968. Skipper glow. Massachusetts Institute of Technology, Research Laboratory of Electronics. Quart. Progr. Rep. No. 88. 114-1 i9.
MILLER, "vV.H., G. D. BERNARD,and J. L. ALLEN. 1968. The optics of insect compound eyes.
Science. 162:760.
M/3LLER, J. 1826. Zur vergleichenden Physiologic des Gesichtsinnes des Menschen und der
Thiere. C. Cnobloch, Leipzig.
NAKA, K., and E. EGUCHI. 1962. Spike potentials recorded from the insect photoreceptor. J.
Gen. Physiol. 45:663.
NUNNEMACI-mR,R. F. 1959. The retinal image of arthropod eyes. Anat. Rec. 134:618.
NUNNEMACm~R, R. F. 1960. The slxucture and function of arthropod eyes. Proceedings 3rd
International Congress Photobiology, Copenhagen. pp. 428--429.
TOmTA, T. 1965. Eleetrophysiological study of the mechanisms subserving color coding in
the fish retina. Cold Spring Harbor Symp. Ouant. Biol. 30:559.
VOWLES, D. M. 1966. The receptive fields of cells in the retina of the housefly (Musca domestica). Proc. Roy. Soc. Set. B. Biol. So. 164:552.
WAsmzu, Y., D. BURKnAPa)T, and P. STECK. 1964. Visual field of single retinula ceils and
interommatidial inclination in the compound eye of the blowfly. Z. vergl. Physiol. 48:413.
WATRRMAN, T. H. 1954. Directional sensitivity of single ommatidia in the compound, eye of
Limulus. P~oc. Nat. Acad. Sci. U.S.A. 40:252.