Vol. 71, No. 11/November 1981/J. Opt. Soc. Am.
Pokorny et al.
1327
Color matching in autosomal dominant tritan defect
Joel Pokorny and Vivianne C.Smith
Eye Research Laboratories, The University of Chicago, 939 East 57th Street, Chicago, Illinois 60637
Loe N. Went
Department of Human Genetics, Leiden University, Wassenaarseweg 72, 2405 Leiden, The Netherlands
Received April 1, 1981
We evaluated color matching in 39 observers with an autosomal dominant tritan defect. Eleven tritans were dichromats with a 10 field, and only two were dichromats with an 80 field. Twenty-one of the tritan observers had
normal blue-green equations with an 80 field, indicating that autosomal dominant tritans have short-wavelengthsensitive cones. Some of the tritan observers showed a shifted blue-green equation, which was ascribed to rod activity.
Observers with defective color vision confuse colors that to
the color-normal observer may appear quite different. Tritan
observers are characterized by color confusions, such as green
with blue or yellow with violet. Kbnig' described the tritan
defect as characteristic of the normal observer when a small
field of view is used. Later, he2 described similar color-vision
defects accompanying eye disease. Kbnig apparently did not
consider that he had seen the tritan defect as a congenital color
defect in persons without ophthalmoscopic abnormality.
Waardenburg, 3 however, thought that at least one of Kbnig's
patients was a congenital tritanope. Similar to protanopia
and deuteranopia, tritanopia is usually considered a loss
system, 4 affecting the short-wavelength-sensitive (SWS)
cones.
Tritan defects were reported 5 -7 rarely and without a hereditary basis until Wright 8 used a magazine article (Picture
Post) to attract the attention of possible tritan observers. As
a result, Wright 8 confirmed tritanopia in 17 persons and
measured 1.20 color-matching characteristics for 7 of them.
Kalmus 9 demonstrated an autosomal dominant inheritance
in his study of the families of 22 tritans.
The mechanism of tritanomaly, a trichromatic form of tritan defect, is puzzling. Tritanomaly is classified as an alteration system, 4 implying that one or more of the cone visual
photopigments differ in their absorption spectra from that
occurring in normal trichromats. Since tritans have normal
Rayleigh matches (unless there is concomitant congenital or
acquired red-green color defect), their middle- and longwavelength-sensitive (MWS and LWS) photopigments are
normal. Tritanomaly is expected when photoreceptors other
than normal SWS cones are active in blue-green matches.
Parallel to the use of the Rayleigh equation for detection of
protanomalous and deuteranomalous trichromacy, tritanomaly should be detected by a match of violet and green primaries to a blue-green test field; tritanomals should use an
abnormally high proportion of blue primary in their matches.
Blue-green equations are subject to large interobserver
variation among color-normal observers; an equation for tritanomaly would, as Wright 8 cautioned, be sensitive to interobserver variation in density of macular pigmentation.
0030-3941/81/111327-08$00.50
The use of blue-green equations for recognition of tritanomaly
is therefore not so certain as the use of the Rayleigh equation
for recognition of congenital X-linked red-green color-vision
defects.
Engelking' 0 described the original published case of tritanomaly in his student Hartung. Hartung" found the defect
in two of his maternal uncles, the more severely affected being
described as almost tritanopic."11" 2 Although initial data'0' 2
suggested X-chromosomal-linked inheritance, subsequent
2 4
-' stressed phenotypic variation with autosomal
pedigrees 9 "1
dominant inheritance in tritanomaly.' 3 -'8 Cole et al. 15 suggested that there was no evidence that tritanomaly was
transmitted as a separate genetic entity, but they could not
distinguish tritanomaly from incomplete tritanopia in their
pedigree.
We had the opportunity to study many individuals with an
autosomal dominant tritan defect in The Netherlands by
using a Moreland anomaloscope. We were interested in two
issues. First, we wished to learn if the size of the field of view
affected color-matching performance. One of the Wright 8
tritanopes showed dichromacy19' 20 for a 1.750 parafoveal field.
However, larger field sizes were not presented. We thought
it possible that phenotypic variation might be different for
10 and 80 fields. Second, we wished to evaluate all affected
family members for tritanomaly. For this purpose we used
the blue-green equation specifically designed for detection
of tritanomaly.2 1-2 4
Screening and Criteria for Selection of Tritans
In contrast with tests to screen for red-green defects, there
are relatively few tests to screen tritan defects. The tritan
tests are unvalidated, having been used on rather few tritan
observers. A recently developed test that is a sensitive indi-
cator of tritan defects is the TNO tritan test,2 5 ' 26 in which the
observer is required to detect a 0.80, 456-nm rectangular
stinulus that flickers at 0.5 Hz on a 140 6000-cd-m- 2 yellow-adapting field. A slide with four neutral filters is interposed before the blue test field. The filter densities are calculated so that after chromatic adaptation of 1-2 min to the
© 1981 Optical Society of America
1328
Pokorny et al.
J. Opt. Soc. Am./Vol. 71, No. 11/November 1981
Table 1. Summary of Color Matching for Autosomal Dominant Tritan Observers
Family
Observer
Pedigree 1
II-1
11-5
11-6
11-8
11-9
11-10
II-15
Information
siblings
Dichromatic
Coefficient"
Red-Green
Defect
20
40
Moreland Equationb
8°
Age
Sex
75
70
68
66
64
63
55
F
M
M
M
F
M
M
X
-
X
X
X
X
X
X
-
47
39
38
M
F
M
23
21
20
F
F
F
33
31
20
M
M
M
10
80
1°
S
S
S
S
S
N?
N
-
S
S
S
-
S
-
S?
X
-
S
X(RE)
-
LE:S;RE:N?
U
N
LE:N?RE:N?
N?
S
F
F
N
N
N
F
F
F
N
W
N
LE:S;RE:W
W
S
LE:N;RE:N
N
N
F
N
F
III-1
III-5
III-6
children of II-1
III-18
III-19
111-20
children of II-5
III-29
III-30
III-34
children of II-10
III-36
III-38
III-40
children of II-13
(deceased)
32
28
23
F
M
F
X(OU)
X
X
III-43
children of II-15
24
F
V1
V
IV-2
IV-16
child of III-1
child of 111-5
11
M
F
X
U
-
-
-
S
9
U
U
U
S
N
U
siblings
47
46
35
32
F
M
M
M
V
F
F
F
W
W
W
W
N
child of II-1
11
W
N
siblings
F
F
F
N
W
W
F
Pedigree 2
II-1
11-2
II-3
II-4
III-1
Pedigree 3
II-2
II-4
II-5
11-6
III-9
111-11
III-12
children of 11-4
child of 11-5
Pedigree 4
III-17
IV-15
IV-20
IV-22
Pedigree 5
11-1
a
b
children of III-17
U
U
U
U
V
X
X
V
V/
X
X
DA
DA
DA
x
X
X
VI
x
V
D
VA
V
M
X
X
54
49
47
45
M
F
F
M
x
22
17
11
M
F
M
55
DA
X
X
DA
V/
X
V
X
x
X
X
-
W
-
F
-
F
N
N
N
X
F
N
'I
X
F
F
S
N
W
N
V/
VI
F
W
x
x
X
X
X
F
X
29
25
20
F
F
F
X
x
39
F
X
X
v/, accepts; X, refuses;-, not tested; U, untestable.
N, normal; S, shift; W, wide; F, full; U, untestable.
yellow field, detection of the test field should be mediated by
SWS cones for slide positions 1 and 2 and by SWS, MWS, or
LWS cones for slide positions 3 and 4. The observer with
normal color vision or an X-linked red-green defect should
detect the test field for all four slide positions. The observer
with a functional reduction in SWS sensitivity may detect the
test field only at slide position 3 or 4.
It was anticipated that elderly observers or observers with
acquired color-vision defects4 would also fail the TNO tritan
test. We therefore adopted the following as minimal criteria
for diagnosis of autosomal dominant tritan defects:
(1)
(2)
The individual should fail the TNO tritan test.
Positive family history. One parent should also fail
Pokorny et al.
the TNO tritan test, or family study should reveal an autosomal dominant mode of inheritance.
(3) There should be no indication of optic atrophy or other
ophthalmologic disease.
We tested 39 observers with an autosomal dominant tritan
defect who met these criteria.
Observers
The tritan observers came from five families (Table 1).
Pedigree l
This was a large pedigree in which we tested 22 tritan members. The index case (III-19) was found in a screening of 400
students25 by using the TNO tritan test. Subsequently, other
family members representing three generations were tested.
We evaluated by anomaloscope 22 of the 26 who failed the
TNO tritan test as well as 7 other family members.
Pedigree 2
The index case of this family was found in a screening at a
medical instrument convention by using the TNO tritan
test.26 A tritan defect was subsequently established in three
siblings and a nephew of the index case, all of whom we
tested.
Pedigree 3
This family was described as family B by van de Merendonk
and Went. 27 The index case (111-10), a deuteranomalous
trichromat, was found in a school screening for red-green
defects by using pseudoisochromatic plates as screening devices. 27 An autosomal dominant tritan defect was subsequently discovered in eight family members, of whom we
tested seven.
Pedigree 4
This family was described by Went et al.,28 who identified
eleven tritan members, of whom we have tested four.
Pedigree 5
The index case of Hungarian origin was found in the same
screening as Pedigree 1.25 Subsequently, the mother was
examined and found to have a tritan defect.
Methods
Vol. 71, No. 11/November 1981/J. Opt. Soc. Am.
1329
that of the Nagel anomaloscope. 4 31 The spectral test field
was variable in luminance; the primary mixture field was
variable from instrument unit 0 at primary 1 to instrument
unit 424 at primary 2. The field was freely viewed; each observer used his preferred eye. The observers adapted to the
ambient field illumination. Observers were cautioned not to
stare continuously at the field but rather to use a glance
technique in which they looked away every 10-12 sec. This
technique was used to minimize the changes in color appearance that occur with continuous viewing of blue-green test
fields.
Experiment 1: Evaluation of Dichromacy in 39 Observers
with an Autosomal Dominant Tritan Defect
We evaluated two equations. First, we checked each observer's Rayleigh equation to establish whether red-green
vision was normal or showed evidence of X-chromosomelinked red-green color defect. We then presented a dichromatic match to evaluate the presence of tritanopia.
Procedure
1. The Rayleigh Match. The test-field wavelength was 589
nm; the mixture field primaries were 545 and 670 nm. The
field size was 2°; the field luminance was approximately 5 cd
m_2 . For those observers over 16 years of age for whom preliminary screening indicated normal red-green vision, a
self-adjustment technique was used. The observer controlled
the test-field luminance knob and the primary ratio control.
The observer who asked to adjust the test-field luminance and
primary ratio to achieve a color match. The majority of observers were able to do this without difficulty. For younger
observers and for those for whom we suspected a red-green
color defect, the primary ratio adjustments were made by the
experimenter; the observer reported whether the circular field
appeared to be of uniform color. 4'3'
2. Dichromatic Coefficients. The test-field wavelength was
589 nm; the mixture-field primaries as used by Wright were
650 and 480 nm. The mixture-field luminancb varied with
the primary ratio. At the characteristic tritanopic match, the
field luminance was 15 cd m- 2 . The field size was 10, 20, 40,
or 8°. A self-adjustment technique was used. For younger
observers, matches characteristic of tritanopes were presented.
A 10 or 20 field was presented initially; if a 10 match could not
be made the procedure was discontinued. If 10 and 20
matches were accepted, the procedure was repeated for the
40 and 80 fields.
Equipment
A Moreland universal anomaloscope 29 was used. Modification of this interference-filter anomaloscope has been described. 30 The instrument permits a circular bipartite field
of variable extent to be used, with the left half-field giving the
test field and the right half-field giving the matching field.
We used stops yielding field sizes between 10 and 8° visual
angle. Three-cavity interference filters (Ditric) were used for
this study. Wavelengths of peak transmission and half-height
bandwidth were measured in the anomaloscope with a calibrated laboratory-constructed spectroradiometer. Halfheight bandwidths ranged from 7 to 14 nm.
Procedure
The majority of observers were tested in their homes. The
anomaloscope was used in a manner essentially identical with
Results
The majority of the tritan observers had normal red-green
color vision. X-chromosomal-linked defects occurred in three
pedigrees. Protanomaly and deuteranomaly occurred in
Pedigree 1, protanopia and deuteranomaly in Pedigree 2, and
deuteranomaly in Pedigree 3. Five of the tritan observers,
two in Pedigree 1 (I1-29, III-34), two in Pedigree 2 (II-2, II-4),
and one in Pedigree 3 (III-9), showed a combination of a tritan
defect with deuteranomaly. Segregation of X-chromosomal
and autosomal dominant genes was observed in Pedigrees 1
and 3.
We tested 37 tritans (20 in Pedigree 1 and all in Pedigrees
2-5): 25 refused the dichromatic match with a 10 field, 11
accepted the 10 match, and one who was not tested with the
10 field accepted the 20 field match (Table 1). Fewer tritans
1330
Pokorny et at.
J. Opt. Soc. Am.NVol. 71, No. 11/November 1981
accepted the dichromatic match as the field size increased, and
only two observers showed tritanopia for an 8° field. Among
the five deuteranomalous tritans, two accepted a 1° dichromatic match and none accepted the 8° dichromatic match.
Our data indicate that dichromacy is not typical in autosomal dominant tritan defect. The majority of the tritans
were trichromats even with a 1° field. In our test families we
note that observers with congenital deuteranomaly do not
have a more severe tritan defect than observers with normal
red-green vision.
Experiment 2: The Moreland Equation
The Moreland equation is a blue-green equation optimized
by Moreland and Kerr2 1' 22 to minimize the effect of interobserver macular pigment variation. It is similar to the equation
previously derived by Speranskaya. 2 3 The equation is affected by the age of the observer, but norms are available.24
We used 430 nm for the blue primary rather than the 440 nm
used by Moreland and Kerr 21l22 because pilot data indicated
that color-normal observers showed less match-midpoint
variation when 10 and 80 matches were compared. Since 430
nm nearly coincides with the tritan confusion wavelength for
500 nm, we hoped to obtain good correlation between fullrange Moreland equations and acceptance of dichromatic
coefficients.
Procedure
We used primaries of 500 and 430 nm in the mixture field and
480 nm with a small amount of 580-nm desaturation in the test
field. 2 1'22 The mixture-field luminance varied with the
mixture ratio and was 5 cd m-2 at the normal match for a
young observer. A fixed mixture ratio was presented, and the
observer was instructed to adjust the luminance of the test
field and report if an exact color match occurred. After each
setting, the mixture ratio was reset and the full matching range
for the equation was established. For the majority of observers the desaturant was fixed. 21'22 If an observer refused
the normal match, we varied the desaturant to check that the
normal blue-green ratio was refused at all desaturant levels.
Field sizes of 10, 20, 40, and 80 were used; in some observers,
all four were evaluted; in others, only the 10 and 83 fields were
evaluated. We tested the preferred eye; if time (and patience)
permitted we checked the fellow eye.
Normal data were obtained from 10 observers who worked
in the laboratories. All were under 45 years of age; six had
normal color vision; four had congenital X-linked deutan
defects. The Moreland equation does not distinguish between normal and deutan observers. We also noted that rod
intrusion does not affect 83 Moreland matches. All observers
could make acceptable 80 matches without reporting annoying
desaturant effects. The 80 matches were if anything easier
to make than the 10 matches. In addition, we tested seven
members of Pedigree 1 who passed the TNO tritan test to
verify the Moreland equation in the field-testing situation.
These observers included one deuteranomalous trichromat.
All gave normal Moreland equations.
Results
We classified the data according to four groups (Table 1): (1)
Normal: A normal match showed normal midmatching position for age and a narrow matching range (fewer than 20
instrument units). (2) Shifted: A shifted match showed a
600-
I-0.
LL
0
A
0.6
N
N
b. 3
0 0
-LJ
N
U
0
0.2
0.4
0.6
0.8
tO0
MIXTURE RATIO
Fig. 1. Comparison of 10 (open symbols) and 80 (filled symbols) color
matches for the Moreland equation (480 nm + 580 nm - 500 nm + 430
nm). The flux in the 480-nm + 580-nm test field is shown as a function of the proportion of 430-nm primary. The solid line shows 10
brightness matches made by a 10 tritanope; the dashed line shows 80
matches made by a complete achromat. Observer symbols: inverted
triangles, 1,III-1; squares, 1, III-18; triangles, 1, 1II-40; diamonds, 1,
IV-2; circles, 4, IV-22.
midmatching position and range that did not fall within the
normal position for age. Matching ranges of shifted matches
were usually narrow. In no case did variation of the desaturant change the match location. (3) Wide: A wide match
included the normal match and showed a matching range that
was 6-18 times wider than normal (i.e., 120-350 instrument
units). (4) Full: a full matching range including either primary (425 instrument units) or an extremely wide matching
range excluding only one or another primary (over 400 instrument units).
With a 11 field, 19 observers showed a full range match, 2
showed a widened matching range, and 11 showed a shifted
match. We could not establish a match for two observers
(Pedigree 1, II-8 and III-6). One observer (Pedigree 1, III-5)
had a shifted match with one eye while the other eye matched
approximately normally; yet another observer (Pedigree 1,
III-36) showed a shifted match in one eye and a wide match
in the other. For the 80 field, twenty-four observers gave a
normal match; eight gave a wide match; one gave a full match;
five gave shifted matches. One (Pedigree 1, IV-15) was impossible to test. For the 1° field, a shifted match was seen in
five persons over 64 years of age and in six observers (excluding III-5) ranging in age between 9 and 47 years. For the
83 field, the shifted match occurred only in the five observers
aged over 64 years of age. Five younger observers who had
a shifted match for the 10 field had a normal match for the 80
field.
The brightness matches for the various ratios in the
matching range were plotted for the five tritans with shifted
10 and normal 80 matches (Fig. 1).
The figure shows 83
matches in filled symbols and 10 matches in open symbols.
The arrows show average midmatching points and flux settings of our 10 control observers. The heavy solid line shows
the 10 brightness matches of one of the 10 tritanopes (Pedigree
1, III-20). This line intersects the normal matches and agrees
with the 83 matches of the five tritans.
Pokorny et al.
Vol. 71, No. 11/November 1981/J. Opt. Soc. Am.
The 10 shifted matches require a high proportion of 430-nm
primary and are made with low 480-nm flux. The matches
do lie on the 10 dichromat's match, i.e., the 1° dichromat is a
reduction system of both 10 and 80 tritan matches. A shifted
match may be indicative either of an alteration system, 4 which
could occur if a photopigment other than SWS cones mediated
the match, or of an absorption system, which could occur if
an abnormal spectrally selective filter attenuated light before
the retina. The acceptance of the shifted 10 matches by I'
dichromats suggests an alteration rather than an absorption
system. In a study of incomplete achromatopsia, 2 we found
an incomplete achromat who had a similarly shifted 80
Moreland equation. This suggested that rods might be the
receptor system mediating the shifted 10 matches. The
dashed line in Fig. 1 shows 8° matches made by a complete
achromat3 2 The 10-shifted matches occur near the intersection of the achromatic and tritanopic matches. It is,
therefore, possible that the matches are rod mediated. In
order to investigate this possibility, we decided to test other
wavelengths.
I; 111-18
0
-0.41
-0.8I
1331
-I
- L-
---
-1.2
-1.6
1
I; M-40
-J
0
I
-l.2
o -1.6
Experiment 3: Extended Match Series for the Moreland
Equation
According to von Kries, 33 an alteration system can be differentiated from an absorption system when a series of test
wavelengths is used. In an absorption system the normal and
abnormal matches remain in constant proportion, i.e., the
amount of shift is independent of test wavelength. 4 The algebraic basis is shown by Alpern et al.34 In contrast, in an
alteration system the normal and abnormal matches change
in proportion, i.e., the amount of shift varies with test wavelength.
Procedure
Following the example of the extended Rayleigh series, we
used the same primaries, 500 and 430 nm, and desaturant, 580
nm, as for the Moreland equation. We used test wavelengths
440, 450, 460, 470, and 490 nm. For each of these test wavelengths, two of the authors (JP and VS) made a large-field (80)
match, varying the amount of desaturant for a best color
match. The desaturant was then fixed at that level. A 10
tritanope (Pedigree 1, 111-20) made 10 brightness matches, and
a complete achromat made 80 brightness matches for the five
test wavelengths. Two of the tritans with shifted 10 matches
(Pedigree 1, III-18 and Pedigree 4, IV-22) performed the full
series of 10 and 80 matches. A third tritan (Pedigree 1, III-40)
was available only for a limited time and was tested at 460,470,
and 490 nm.
The procedure was as before. The test wavelength with
fixed desaturant was chosen. For various ratios of primary
ratio, the tritans adjusted the flux of the test field and reported
whenr a color match was obtained.
Results
The logarithm of the midmatching point and range was calculated; the data for a test wavelength of 480 nm were included
in the calculation. The logarithmic differences between 80
and 10 as a function of test wavelength (Fig. 2) are shown. In
this plot, an absorption system should give data that show no
dependence on wavelength. Despite rather large error
bounds, data for the three observers show a systematic decrease as test wavelength increases. The solid line is an es-
0n
4; IV-22
0
-J
-0.4
-0.8
-1.2
-1.6
I
44 0
I
I
460
I
I
480
TEST WAVELENGTH (nm)
Fig. 2. The results of the extended-wavelength series (X + 580 nm
500 nm + 430 nm). The difference between logarithms of the
blue-green ratio (B/G) for 10 and 80 matches is plotted as a function
of test wavelength. The solid line is theoretical (see text).
-
timation of the differences expected if rods were active in the
10 matches and SWS cones were active in the 80 matches. For
the analysis, the CIE 1964 100 observer Tx was taken as representing the SWS cone spectral sensitivity, and the CIE 1951
Vx' represented the rod spectral sensitivity. These data are
consistent with an alteration system.
Using the same data format as for Fig. 1, the mixture ratios
and brightness settings were plotted (Fig. 3). The normal
matches and tritan matches fall on the tritanope's brightness
matches. The 80 matches of the tritans fall near the 80
matches of the normal observers. The 10 matches are shifted
toward the 430-nm primary and occur near the complete
achromat's brightness matches. The two independent
analyses of the data are consistent with the hypothesis that
the shifted 10 matches are mediated by activity of a rhodopsin
photopigment with peak sensitivity near 490-500 nm and that
80 matches are mediated by SWS cones.
Discussion
All the tritans had an abnormal Moreland equation with a 10
field. The 10 Moreland match therefore agreed with the TNO
1332
Pokorny et at.
J. Opt. Soc. Am.NVol. 71, No. 11/November 1981
I
gm
J 90(
N
I
I
X
I
I
470 nm
09O0
-J
LI
U-
IL
I-
440
w
I
nm
en
e)
LI
I.-
i- 6 0 0
Li.
0
0
-+N
X
x
-J
IL.
W
w30C
30 0
-J
-J
Ix
w
0
g
0.2
04
0.6
0
0.8
0i
'Di
0.6
0.8
I
MIXTURE RATIO
(d)
MIXTURE RATIO
(a)
.J
I\
'C
-J
9 0 0
45o nm
IU)
60C
0
'C
U.
X
-J
LL
w300
-J
ii: I
I
D
I
0.2
Ip
I
0.4
I
0.6
I
I
0.8
I
t.
1.0
MIXTURE RATIO
(b)
MIXTURE RATIO
(c)
tritan test in establishing functional SWS cone abnormality.
The color matches showed interfamilial and intrafamilial
variation, just as has been reported previously for screening
and discrimination tests. The dichromatic coefficients and
Moreland equations showed some variability: not all dichromats had full-range Moreland matches; conversely, many
tritans with full-range 1° Moreland matches were not 10 dichromats. There was also some indication of intereye variability. Observer III-36 gave a shifted 10 Moreland match
in one eye but a wide match in the other. Unfortunately, time
constraints and general difficulties of field testing made it
impossible for us to obtain full data on both eyes from all observers. Observer II1-5 showed a shifted match in one eye
and a normal match in the other eye. There was a mild vi-
MIXTURE RATIO
(0)
Fig. 3. Comparison of 80 (filled symbols) and
10 (open symbols) matches for five test wavelengths on the extended-wavelength series (X
+ 580 nm =_500 nx + 430 m). The flux in the
test field is plotted as a function of the proportion of 430-nm primary. The solid line
shows 1° brightness matches of a I1 txitanope;
the dashed line shows 8° brightness matches
of a complete achromat. Test wavelengths
were 440 nm [panel (a)], 450 nm [panel (b)], 460
nm [panel (c)], 470 nm [panel (d)], and 490 nm
[panel (e)]. Observer symbols: squares, 1,
III-18; triangles, 1, II-40; circles, 4, IV-22.
sual-acuity deficit in the color-normnal eye, but ophthalmoscopic examination was not revealing. We thought that eccentric fixation in that eye might explain the normal color
match since the majority of tritan observers made normal
matches with 80 fields; we have no other explanation of the
result. The data from the 22 observers of Pedigree 1 showed
some differences from the data of the 17 observers from
Pedigrees 2-5. Some of these differences may be ascribed to
age. Six of the seven Generation II members in Pedigree I
were 19-20 years older than any of the tritans in Pedigrees 2-5.
If these Generation II members were omitted, the differences
between Pedigree 1 and Pedigrees 2-5 for the 80 test field
would be reduced. Nevertheless, we found only four I dichromats of 16 Generation III-IV tritans in Pedigree 1, compared with eight 10 dichromats of 17 tritans in Pedigrees 2-5,
Pokorny et al.
whereas we noted a shifted 10 Moreland match in six young
members of Pedigree 1 and in only one member of Pedigrees
2-5.
Finally, we did find evidence of an alteration system 4 in
some of the observers. We suggested that the shifted matches
reflect rod activity. We have previously noted that rods may
be active in color matching of color-defective observers. 3 4-3 6
Of note in this study was rod activity in a 10 colorimetric field.
We suppose that certain observers have small or nonexistent
rod-free zones or are able to use stray light in making 10 discriminations. Provided that rods are active, their signals may
give information to those observers whose color vision has
been compromised by hereditary or acquired defects. Among
tritan observers over 64 years of age, a shifted blue-green
equation was seen even with an 8° field. The shift was larger
than would be expected on the basis of age-related changes
in ocular media transmission21 ' 22 and may reflect a variation
of a normal retinal aging phenomenon. We hypothesize that
if SWS function diminishes, rods may become an active color
receptor. Some clinical Moreland equation data may be interpreted in this way: Moreland et al.38 noted a shifted
blue-green equation in one eye of a 55-year-old patient with
maculopathy (visiual acuity, 0.7) similar to the shifted bluegreen matches that we observed in young and old tritan observers. Further, Moreland 39 has reported shifted small-field
matches in some diabetic patients.
We found that most tritans made a completely normal
blue-green color match using an 80 field. Some continue to
show chromatic discrimination loss, but only 2 of 39 tritans
showed 8° dichromacy. We infer that the normal 80 Moreland equations indicate that the majority of tritans have SWS
cones in their retinas. Our studies do not indicate the cause
of the tritan defect. The tritan defect might reflect an abnormal number or distribution pattern of SWS cones or a
neural defect.
ACKNOWLEDGMENTS
Noor van Heel, Lida Spruyt, and Inge van Leeuwen assisted
in the field studies and in color screening of the tritan families.
This research was performed during 1979-1980 in the Department of Human Genetics, Leiden University, and was
presented at the Optical Society of America Annual Meeting,
October 1980. This work was supported in part by USPH
NIH NEI research grants EY 00901 (Pokorny) and EY 01876
(Smith) and by a grant from ZWO, the Dutch Organization
for the Advancement of Pure Research (Pokorny).
REFERENCES
1. A. Konig, "Uber den menschlischen Sehpurpur und seine Bedeutung fur das Sehen," Sb. Akad. Wiss. Berlin, 577-593
(1894).
2. A.K6nig, "Ueber Blaublindheit," Sb. Akad. Wiss. Berlin, 718-731
(1897).
3. P. J. Waardenburg, Genetics and Ophthalmology. Vol. II,
NeuroOphthalmologicalPart. Chap. 28: Dyschromatopsia.
(Royal VanGorcum, Assen, The Netherlands, 1963).
4. J. Pokorny, V.C.Smith, G.Verriest, and A.J. L. G.Pinckers, eds.,
Congenitaland Acquired Color Vision Defects (Grune &Strat-
ton, Inc., New York, 1979).
5. M. Levy, "Ueber einen Fall von angeborener beiderseitiger Tritanopie (Blaublindheit)," Albrecht von Graefes Arch. Klin. Exp.
Ophthalmol. 62, 464-480 (1905).
Vol. 71, No. 11/November 1981/J. Opt. Soc. Am.
1333
6. D. Farnsworth, "Description of a subject congenitally deficient
in violet-yellow vision," J. Opt. Soc. Am. 33, 350A (1943).
7. F. P. Fischer, M. A. Bouman, and J. T. Doesschate, "A case of
tritanopy," Doc. Ophthalmol. 5-6, 73-87 (1951).
8. W. D. Wright, "The characteristics of tritanopia," J. Opt. Soc.
Am. 42,509-521 (1952).
9. H. Kalmus, "The familial distribution of congenital tritanopia
with some remarks on some similar conditions," Ann. Hum.
Genet. 20,39-56 (1955).
10. E. Engelking, "Die Tritanomalie, ein bisher unbekannter Typus
anomaler Tricbromasie," Albrecht von Graefes Arch. Klin. Exp.
Oplithalmol. 116, 196-244 (1925).
11. H. Hartung, "Ueber drei familiare Falle von Tritanomalie," Klin.
Monatsbl. Augenheilkd. 76, 229-240 (1926).
12. E. Engelking-and H. Hartung, "Spektraluntersuchungen fiber
die Minimalfeldhelligkeiten des Tritanomalen und seine Unterschiedsempfindlichkeit fur Aenderunger der Helligkeit," Arch.
Ophthalmol. 118, 211-220 (1927).
13. R. A. Crone, "Combined forms of congenital colour defects. A
pedigree with atypical colour blindness," Br. J. Ophthalmol. 40,
462-472 (1956).
14. G. H. Henry, B. L. Cole, and J. Nathan, "The inheritance of cogenital tritanopia with the report of an extensive pedigree," Ann.
Hum. Genet. 27, 219-231 (1964).
15. B. L. Cole, G. H. Henry, and J. Nathan, "Phenotypical variations
of tritanopia," Vision Res. 6, 301-313 (1966).
16. 1.Schmidt, "Ueber zwei Falle von angeborener Blaugelbsehstorung," Klin. Monatsbl. Augenheilkd. 109,635-652 (1942).
17. I. Schmidt, "On congenital tritanomaly," Vision Res. 10,717-743
(1970).
18. D. P. Smith, B. L. Cole, and A. Isaacs, "Congenital tritanopia
without neuroretinal disease," Invest. Ophthalmol. 12,608-617
(1973).
19. E. A. Hough and K. H. Ruddock, "The parafoveal visual response
of a tritanope and an interpretation of the VA' sensitivity functions
of mesopic vision," Vision Res. 9, 935-946 (1969).
20. K. H. Ruddock, "Parafoveal colour vision responses of four dichromats," Vision Res. 11, 143-156 (1971).
21. J. D. Moreland and J. Kerr, "Optimization of stimuli for tritanomaloscopy," Mod. Probl. Ophthalmol. 19, 162-166 (1978).
22. J. D. Moreland and J. Kerr, "Optimization of a Rayleigh-type
equation for the detection of tritanomaly," Vision Res. 19,
1369-1375 (1979).
23. N. I. Speranskaya, "Tritanomaly," Dokl. Akad. Nauk. SSSR 130,
1362-1365 (1960).
24. J. D. Moreland, "Temporal variations in anomalscope equations,"
Mod. Probl. Ophthalmol. 19, 167-172 (1978).
25. L. van Heel, L. N. Went, and D. van Norren, "Frequency of tritan
disturbances in a population study," in Colour Vision Deficiencies V, G. Verriest, ed. (Hilger, Bristol, U.K., 1980) pp. 256260.
26. D. van Norren and L. N. Went, "New test for the detection of
tritan defects evaluated in two surveys," Vision Res. (to be published).
27. S. van de Merendonk and L. N. Went, "Two cases of inherited
deutan and tritan disturbances in the same person, and a study
of their families," in Colour Vision Deficiencies V, G. Verriest,
ed. (Hilger, Bristol, U.K., 1980), Chap. 5, pp. 268-272.
28. L. N. Went, H. Volker-Dieben, and E. C. De Vries-De Mol,
"Colour vision, ophthalmological and linkage studies in a pedigree
with a tritan defect," Mod. Probl. Ophthalmol. 13, 272-276
(1974).
29. J. D. Moreland and W. B. Young, "A new anomaloscope
employing interference filters," Mod. Probl. Ophthalmol. 13,
47-55 (1974).
30. J. Pokorny and V. C. Smith, "Effect of field size on red-green color
mixture equations," J. Opt. Soc. Am. 66, 705-708 (1976).
31. A. Linksz, An Essay on Color Vision and Clinical Color- Vision
Tests (Grune & Stratton, New York, 1964).
32. J. Pokorny, V. C. Smith, A. J. L. G. Pinckers, and M. Cozijnsen,
"Classification of complete and incomplete autosomal recessive
achromatopsia," Albrecht von Graefes Arch. Ophthalmol. Klin.
Exp. (to be published).
33. J. von Kries, "Normal and anomalous color systems," in Treatise
on Physiological Optics, H. L. F. von Helmholtz, ed., 3rd ed.,
1334
J. Opt. Soc. Am./Vol. 71, No. 11/November 1981
translated by J. P. C. Southall (Optical Society of America,
Rochester, N.Y., 1924), Vol. II, pp. 395-425.
34. M. Alpern, B. Bastian, E. N. Pugh, Jr., and W. Gras, "Altered
ocular pigments, photostable and labile," Mod. Probl. Ophthalmol. 17, 273-291 (1976).
35. V. C. Smith and J. Pokorny, "Large-field trichromacy in protanopes and deuteranopes," J. Opt. Soc. Am. 67, 213-220 (1977).
36. J. Pokorny and V. C. Smith, "A variant of red-green color defect,"
Vision Res. 21, 211-217 (1981).
Pokorny et al.
37. J. Pokorny and V. C. Smith, "New observations concerning
red-green color defects," Color Res. Appl. (to be published).
38. J. D. Moreland, M. Maione, F. Carta, and L. Scoccianti, "Acquired
'tritan' deficiencies in macular pathology," Mod. Probl. Ophthalmol. 19, 270-275 (1978).
39. J. D. Moreland, "Tritanomaloscopy of diabetics undergoing
photocoagulation therapy," in Colour Vision Deficiencies V, G.
Verriest, ed. (Hilger, Bristol, U.K., 1980), pp. 299-305.