Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 53
Chapter V
TII E COMPARABILITY OF IRRADIATION
SUBCLASSES
~ 1
of the radiation exposure categories of their
fathers and mothers, as these categories were
defined in the preceding chapter. The crux of
our problem is a comparison of the characteris-
tics of the infants comprising these subclasses.
As has been emphasized several times already,
all of the indicators of possible genetic damage
utilized in this investigation may also be in-
fluenced by a variety of other factors. It will
be the purpose of this chapter to undertake a
detailed comparison of the parents of the in-
fants comprising these subclasses with respect
to certain possible differences which might in-
fluence the outcome of pregnancy. Of the many
possible differences which could be explored,
we shall restrict ourselves to those which in our
opinion are most pertinent to the problem at
hand and for which we have more nearly satis-
factory data.
In the comparisons which will be presented,
usually no allowance has been made for the fact
that because of repeated pregnancies the same
couple may be represented more than once
among the parentage of a subclass. The tests of
significance that follow assume no duplicate
registration and thus yield underestimates of
the errors of differences. It might be pointed
out that, in the main, we should underestimate
variances in the parental population more than
in the offspring population.
5.1 Cor7sa'Jg~inity. It is generally recog-
nized that the offspring of consanguineous
marriages more often exhibit the consequences
of genetic homozygosity than do the offspring
of non-consanguineous unions. To the extent
that there is a recessive, incompletely recessive,
or semi-dominant genetic component in the
cow
THE infants examined in Hiroshima and Naga- etiology of congenital defect, stillbirth, or neo-
saki during the course of this study may be natal death, such homozygosity might be ex-
annortioned among 25 subclasses on the basis pected to alter the frequency of these events
sufficiently to obscure real irradiation differ-
ences or to create spurious ones. The occurrence
within the populations under study of a non-
random distribution of consanguineous mar-
riages thus might introduce a source of bias into
the findings. This, then, was the theoretical
consideration which led to the inclusion of an
item in the Genetics Short Form regarding
parental consanguinity. The problem seemed of
special importance because of the relatively high
frequency of consanguineous unions in Japan
(Neel, Kodani, Brewer, and Anderson, 1949~.
In Tables 5.1 and 5.2 the present data are
examined as regards a relationship between ex-
posure subclass and the frequency of consan-
guinity.~ Although all known degrees of con-
sanguinity were recorded by the clerks, in the
analysis attention has been restricted to mar-
riages of first cousins, first cousins once re-
moved, and second cousins. A few uncle-niece
unions, and some involving more remote de-
grees of consanguinity than second cousin
marriages, have been recorded during the study;
all such unions have been excluded from these
considerations because of the uncertainty regard-
ing the exhaustiveness of the ascertainment.
There is significantly more consanguineous
marriage in Nagasaki than in Hiroshima, this
presumably a reflection of social and cultural
differences between the two localities. Moreover,
within the cities there is heterogeneity among
exposure subclasses. In general this consists in
less consanguineous marriage among the "ex-
posed" (categories 2, 3, 4, 5) than among the
The differing forms utilized in the presentation
of the results of the analysis of the various tables
will be evident following the reading of Chapter VI.
53
OCR for page 54
54 Genetic Ejects of Atomic Bombs Chapter V
TABLE 5.1 CONSANGUINITY ( FIRST COUSIN, FIRST COUSIN ONCE
REMOVED, SECOND COUSIN ~ BY CITY AND PARENTAL EXPOSURE
Hiroshima
Fathers n
1 2 3 4-5 Total
Jrn18,7231,61164842221,404
1 r1,2429742231,404
p.0663.0602.0648.0545.0656
~ n5,7211,9934162648,394
24 r3111172716471
~ p.0544.0587.0649.0606.0561
v, In2,3204515451613,477
3g r9723164140
up.0418.0510.0294.0248.0403
In1,2082171161271,668
4-54 r7796698
l P.0637.0415.0517.0472.0588
~ n 27,972 4,2721,725974 34,943
Totals r 1,727 2469149 2,113
tp .0617 .0576.0528.0503 .0605
Nagasaki
Fathers
A
1 234-5 Total
cS
In16,3382,42025814919,165
14 r1,5412111381,773
up.0943.0872.0504.0537.0925
f n10,1414,48330118815,113
2\ r68329322101,008
up.0674.0654.0731.0532.0667
~ n823298109391,269
3g r69147494
up.0838.0470.0642.1026.0741
~ n5961293530790
=54 r3310 245
up.0554.0775 .0667.0570
~ n 27,898 7,330 703 406 36,337
Total$ r 2,326 528 42 24 2,920
l P .0834 .0720 .0597 .0591 .0804
a In this and subsequent tables, the term "Fathers ' will be used as an
abbreviation for Father's Exposure Category, and the term "Mothers"
similarly.
OCR for page 55
OCR for page 56
OCR for page 57
OCR for page 58
OCR for page 59
OCR for page 60
OCR for page 61
OCR for page 62
OCR for page 63
OCR for page 64
OCR for page 65
OCR for page 66
OCR for page 67
OCR for page 68
OCR for page 69
OCR for page 70
OCR for page 71
Representative terms from entire chapter:
unrelated parents
r6~ cam of ~ i~6r7~I ~55
unexposed (category I i, aRbough the tendency
is statlstlcally sl~nlEcant only filth respect to
the mothers. fibers is also ~ significant clty-
m~ber interaction. Category 1 contains ~ hub
proportion of persons Hobo did not legaDy re-
~de in thy of the No Titles ~ Me time of
the bo~lngs. It may be surmised that ~ s~-
stantla1 pt~orbon of these persons bee rural
antecedents; the freguen~ of consangulnl~
Kahn to be blabber in rural communizes (NeeI
et at., 1949~.
hem par~ta1 We ~ an] pad~,3 and ~ of
the lndlcators of possible genetic dodge herein
studied is attested to ~ ~ volumloous literature
Be. Clocco, 1938; Yerusbal~, 1938; Pen-
ros~ 1939; Mushy, 1947; Lanitman, 1948;
Record and Eon, 1949; Suthedand, 1949;
Cagey 1950; Lee and ~^e~ 19 go;
Worcester, Stevenson, and Rice, 1950; H~-
naue~ 1951; Earn an] Penrose, 1951; Novel,
195 3; Salber and Bradsha~, 195 3; Con
and Gordon, 1953; Eyed, 1954). Our own
TABLE ~2 CHI~^RB ANALYSIS OF THE DEFOE NCY OF CONSANGOINEOOS ABRIDGES BY Cal AND
PARENTAL EXPOSURE
Source DE X ~P
Tota1 51 26I~53 <(.OOI
Interactions, first order
Chi ~- 16.913
CE ~L610
B4F 9 6317
Chin Endue
City
Bloth~\ exposure 1
B{otbe~ exposure 2
Blotber~ exposure
brothers exposure
Blothert
Hlroshlma (~)
I
56 Genetic Effects of Atomic Bombs Chapter V
TABLE 5.3 MEAN MATERNAL AGE BY CITY AND PARENTAL EXPOSURE
(Based on unrelated parents orally)
Hiroshima
Fathers
, ~
1 2 3 4 5
v'
; -
~:
o
Total
rtin17,4811,51460613626320,000
lx27.046927.325027.516527.485328.270027.1013
2;n5,4101,8763891081407,923
six26.836630.131129.295628.092628.750027.7883
r n2,22342852981763,337
31x26.952329.140230.998130.012329.039527.9961
4 in38888452419564
mix27.786128.363629.666731.166729.684228.2340
in7431206526521,006
5lx-27.125229.366729.338528.538531.480827.7972
Total in26,2454,0261,63437555032,830
l--27.008728.909129.198928.514728.850927.3988
Nagasaki
Fathers
, ~
1 2 3 4 5 Total
~ c
o
X
fin14,7972,2092454110017,392
(x28.542729.142129.526528.487830.280028.6425
din9,4584,1902795112714,105
l-x27.608730.847730.311830.117629.897628.6540
in75428410212231,175
Six-27.657829.841531.529431.583330.652228.6204
4 in105351443161
(x29.933330.485734.642932.000034.666730.6025
i n4588421120584
5)x26.744528.476229.190513.000029.400027.1490
1 T t lrn25,5726,80266110927333,417
~ 0 a Ax28.144730.220730.264829.578030.117228.6299
The Comparability of Irradiation; Subclasses 57
through 5.8. It will be noted that the compari-
son is restricted to unrelated parents. In the fol-
lowing chapter we shall advance reasons for
excluding children born to related parents from
the analysis. Because of the potential great im-
portance of age-parity differences in this analy
2, 3, 4, or 5) than when non-exposed (cate-
gory 1) . This appears to be true with reference
to fathers' but not mothers' exposure in Naga-
saki. An obvious explanation of the findings
with regard to the relation between paternal ex-
posure and mean maternal age is the fact that
TABLE 5.4 ANALYSIS OF VARIANCE: MOTHER'S AGE BY CITY AND PARENTAL EXPOSURE
(Unrelated parents)
(In this and subsequent tables, one asterisk will be used to denote a value significant at the 5 per cent level,
and two asterisks, a value significant at the 1 per cent level.)
Sums of squares
Sourceof deviationsDFMean squareF
Main effects
City (C) 21,983.321551 1 21,983.3216 830.86**
Father (F) 43,401.030323 4 10,850.2576 410.08**
Mother (M) 947.687594 4 236.9219 8.95**
Interactions
First order
CF 74.432938 4 18.6082 1.42
CM 3,107.667017 4 776.9168 29.36**
M F 18,682.813301 16 1,167.6758 44.13**
Higher orders 871.870114 16 54.4919 2.06**
Between classes 94,513.849388 49 1,928.8541 72.90**
Within classes 1,751,484.087605 66,197 26.4587
Total ; . 1,845,997.936993 66,246
TABLE 5.5 THE DISTRIBUTION OF MEAN SQuAREs FOR MATERNAL AGE BY CITY, SEX OF INFANT, AND
PARENTAL EXPOSURE ~ Unrelated parents )
(The sample numbers are the same as in Table 5.3. The form of the analysis will be discussed in Chapter VI.)
v,
o
Hiroshima
Fathers
A
Sex 1 2 3 4-5
1 i: 22.160 27.336 29.653 25.529
l: 21.628 24.205 29.271 23.310
2~3
'it
3{d
5{:
21.999 28.964
22.327 29.351
24.216 29.592
23.381 29.429
26.112 28.495
24.226 36.713
33.690 28.247
40.519 34.122
32.618 25.009
31.245 36.640
47.317 35.253
27.148 36.738
Nagasaki
- -
~
o
Sex 1
(d' 25.376
1~: 25.822
2l,:
rep
3~:S:
4-5{:
Fathers
28.335 37.926
26.361 38.289
30.266 49.061
29.384 39.629
27.956 28.292
27.720 48.456
2 3
36.213 40.472
33.257 32.491
43.489
42.377
Xc: = 121.021** DF = 15 Xc: = 222.236** DF = 15
XS? = 144.251~* DF= 15 X:- = 192.154** DF= 15
sis, it has seemed wise to anticipate this decision
with respect to this tabulation.
Not only do the cities differ but there is also
significant heterogeneity within cities among
exposure subclasses as regards both mean ma-
ternal age and mean parity. In Hiroshima there
is a clear tendency for maternal age to be greater
when the mother or father is exposed (category
4-5
38.920
32.455
31.397
40.381
42.831 18.423
37.398 31.729
32.515 28.743
39.000 55.269
the younger males tended to be in military
service while the older males remained at home
(and so were exposed), and mother's age is
correlated with father's. Heterogeneity between
the subclasses is also indicated by the fact that
most of the interaction terms are also significant.
No less variable than the mean maternal ages
and parities are the mean squares (variance esti
58 Genetic Ejects of Atomic Bombs Chapter
TABLE 5.6 MEAN PARITY BY Corm AND PARENTAL EXPOSURE
(Based on unrelated parents only)
Hiroshima
Fathers
1 2 3 4 5 Total
v,
sit
o
By
J n17,4811,51460613626320,000
1 ~2.42282.41082.37622.36762.46772.4207
2 r n5,4101,8763891081407,923
~ x2.42223.61513.23392.79632.94292.7588
3 ~ n2,22342852981763,337
~ a;2.43683.28973.91873.91363.31582.8370
4 r n38888452419564
x2.60313.22733.48893.87503.36842.8511
r n7431206526521,006
5 six2.40783.53333.06153.42313.51922.6680
T talon 26,245 4,026 1,634 375 550 32,830
° l ~2.4261 3.1167 3.1377 2.9947 2.8364 2.5596
Nagasaki
Fathers
1 2 3 4 5
: -
o
Total
1 r n14,7972,2092454110017,392
~ x2.90083.09463.15512.48783.15002.9295
2 J n9,4584,1902795112714,105
six2.72943.97543.68103.29413.64573.1287
r n75428410212231,175
3 six2.85943.72544.00005.00003.69573.2060
4 r n105351443161
x3.47623.80005.35714.75006.33333.7950
J n4588421120584
six2.62013.16673.61901.00003.75002.7705
T t ltn 25,572 6,802 661 109 273 33,417
0 a ~2~8335 3.6680 3.5688 3.2110 3.5055 3.0247
The Comparability of Irradiation Subclasses 59
mates) for these two variables.4 From Tames
5.5 and 5.8 we note a significant heterogeneity
among exposure classes, within each sex and
city, in the variance estimates. It can also be
shown that the variance estimates differ signifi-
cantly between cities. In general, the variance
in post-war Japan, those persons closer in having
suffered in material ways to a greater extent than
those farther out. Economic status appears to
be related to some of the indicators of radiation
effect here under consideration, notably still-
birth frequency, birthweight, and neonatal
TABLE 5.7 ANALYSIS OF VARIANCE: PARITY BY CITY AND PARENTAL EXPOSURE
(Unrelated parents)
Sums of squares
Sourceof deviationsDFMean squareF
Main effects
city (C) 1,663.729494 1 1,663.7295 531.52**
Father (F) 6,053.779421 4 1,513.4449 483.51**
Mother (M) 852.862735 4 213.2157 68.12**
Interactions
First order
CF 181.504471 4 45.3761 14.50* *
CM 62.510339 4 15.6276 4.99**
MF 2,637.923744 16 164.8702 52.67* *
Higher orders 77.508327 16 4.8443 1.55
Between classes 13,593.325165 49 2,774.1480 886.27**
Within classes 207,206.531342 66,197 3.1302
Total 220,799.856507
66,246
TABLE 5.8 THE DISTRIBUTION OF MEAN SQuAREs FOR PARITY BY CITY, SEX OF INFANT, AND
PARENTAL EXPOSURE ~ Unrelated parents ~
(The sample numbers are the same as in Table 5.6.)
. .
Hlroshlma
Fathers
Sex
1 2
3
4-5
1J'd 2.255 2.156 2.360 2.067
:: 2.076 2.311 1.691 2.217
2 ;: 0.242 3.438 4.023 2.286
:: 2.232 3.840 3.659 2.729
3 2.361 3.610 4.460 4.286
:: 2.134 3.274 3.785 4.251
4_5;c; 2.239 4.171 4.999 5.328
i: 2.206 4.526 3.884 3.441
Nagasaki
, ~
Fathers
, ^- ~
Sex 1 2 3 4-5
1 J'3 2 867
~ ~ 2.899
2<(s?
3{?
4-5~:
3.022 4.933
2.795 5.183
3.059 4.849
3.624 4.647
3.258 2.967
3.434 4.878
4.111 4.717
4.036 3.297
4.652
4.284
4.306
3.350
4.434
4.529
X~ = 281.717*4 DF = 15 Xcp = 337.847* ~DF = 15
X~?2-324.417** DF= 15 X; =397.544*4 DF= 15
estimates increase with increasing conjoint death (e.g., Ebbs et
parental exposure. 1944;
5.3 Economic stains. On a priori grounds et al.,
it seemed quite possible that there might be a
correlation between distance from the hypocenter
at the time of the explosion and economic status
4 The effect of heterogeneity of the variances on
the tests given in Tables 5.5 and 5.7 will be discussed
in Section 6.6.
3.266
2.328
3.984
3.992
4.474
5.862
6.495
5.359
al., 1942a, b; Balfour,
Antonov, 1947; Smith, 1947; Burke
1949; Dean, 1950; Nixon, 1950). It
therefore seemed advisable to make some at-
tempt to evaluate this possibility.
The economic status of the home into which
the child was born was estimated for all preg-
nancies terminating abnormally, as well as every
pregnancy whose registration number ended in
60 Genetic Effects of Atomic Bombs Chapter V
TABLE 5.9 ECONOMIC STATUS BY CITY AND PARENTAL EXPOSURE
(The figure given for economic status is proportion of homes falling
into the "poor" and "very poor" categories.)
Hiroshima
Fathers
4-5Total
f~ n1,77014365462,024
1jr14446 154
P.0814.0280.0923 .0761
~ n5691714728815
2: r49244279
P.0861.1404.0851.0714.0969
n218265619319
~s 35 r1456126
~p.0642.1923.1071.0526.0815
~ n103211011145
4-54 r911112
lP.0874.0476.~ 000.0909.0828
J n2,6603611781043,303
Total~ r21634174271
I~P.0812.0942.0955.0385.OS20
Nagasaki
Fathers
_~
1234-5
Jn1,45922727141,727
1- r226338 267
l n.1549.1454.2963 .1546
Jn89438733141,328
12> r1257451205
~p.1398.1912.1515.0714.1544
11 n6129106106
3j r752216
IP.1148.1724.2000.3333.1509
~ n56124375
4-5 J r921-12
IP.1607.1667.2500-.1600
1
~ n2,47065574373,236
Total: r367114163500
:~p.1486.1740.2162.0811.1545
The Com parability 0
zero, in connection with the use of the Genetics
Long Form (Sees. 2.2 and 2.3~. This evalua-
tion was made according to a very rough scoring
system admitting of five economic classifica-
tions: very poor, poor, average, well-to-do, and
rich. The evaluation was carried out by the
Japanese physician at the time of the home visit.
The physicians were instructed that it was ex-
pected that most homes would be graded as
average, i.e., that existing rather than pre-bomb-
ing standards should serve as the yardstick of
comparison. Inasmuch as in the course of time
a physician employed in the program found
himself in all sections of the city, it was felt
that this crude system of rating would pick up
any marked differences between exposure cate-
gories.
The findings are given in Tables 5.9 and
5.10. The data are presented in simplified form,
in terms of proportion of all homes graded as
"poor" or "very poor." In view of the subjective
nature of the ratings, it is difficult to attach sig-
nificance to the apparent difference between the
two cities. There is no apparent relation between
economic status and mother's exposure category,
nor is there a relation between father's category
and economic status. At first glance this observa-
tion would seem to be counter to common sense,
since it would seem obvious that persons close
to the hypocenter would have suffered greater
material losses than those more distant, losses
which would reflect themselves in economic
level. However, the effect of the atomic bomb
was but one of a number of sources of severe
economic dislocation in post-war Hiroshima
and Nagasaki. The obvious conclusion from the
findings summarized in this section would be
that these other causes of economic readjustment
tended to complicate and even to nullify the
effects of the bombs.
5.4 Frequency of positive serological te it f or
syphilis. In view of the well-known relation-
ship between syphilis and stillbirth and neonatal
death, a serological test for syphilis was carried
out in the ABCC laboratories on each woman
whose registration number terminated in "0,"
as well as for any woman whose pregnancy
terminated abnormally. In about 3 per cent of
the terminations no test was performed, usually
because of the disinclination of the mother to
submit to venipuncture. The tests used were the
cardiolipin microflocculation and the Kline.
Tables 5.11 and 5.12 present the findings.
f Irradiation Subclasses 61
There is a significant difference between cities,
but no indication of heterogeneity between
parental exposure classes.
5.5 Frequency of induced abortions Ovid of
dilatation Arid curettage of the Sterns fD and C7.
During the post-war years, in an effort to
control birth rates, the Japanese government
relaxed the indications for legal "therapeutic"
abortion. Exact figures on the frequency with
which abortions were performed are impossible
to obtain for a variety of reasons. However, be-
cause of the possibility that the late complica-
tions of uterine infection following an induced
TABLE 5.10 CHI-SQUARE ANALYSIS OF THE
DISTRIBUTION OF ECONOMIC STATUSES
BY CITY AND PARENTAL EXPOSURE
(Unrelated parents)
DF
Total 31
Interactions, first order
CM
CF
ME
Main effects
Cities ( C )
Fathers (Fir
Hiroshima 3
Nagasaki 3
Sum ..........
Mothers (M) ~
Hiroshima .......
Nagasaki ........
a Adjusted for cities.
X
120.378
3 2.268
3 0.365
9 17.683
1
S2.529
3.782
6.262
6 10.044
3 3.354
3 0.028
6
3.382
p
< 0.001
.50-.70
.90-.95
.02-.05
< 0.001
.20-.30
.05-.10
.10-.20
.30-.50
> .99
.70-.80
abortion might influence pregnancy termination,
an attempt was made to obtain some informa-
tion on this point through the use of the Genet-
ics Short Form. Each registrant was questioned
concerning the occurrence of abortions. The
abortions which were reported were divided
into two categories, depending on occurrence
before or after the date of the atomic bombing,
and further subdivided as to whether they were
spontaneous or induced. The findings concern-
ing induced abortions are shown in Tables 5.13
and 5.14. The figures refer to the proportion of
women reporting one or more induced abor-
tions. Each mother is scored only once, on the
basis of the first registered pregnancy. Induced
abortions are reported with a significantly
greater frequency in Hiroshima than in Naga-
saki and, in both cities, appear to be more
62 Genetic EJects of Atomic Bombs Chapter V
TABLE 5.11 FREQUENCY OF POSITIVE SEROLOGY BY PARENTAL EXPOSURE,
CITY, AND GROUP
(Unrelated "zero" parents only)
Hiroshima
2 {r
|:n
33 r
~tP
rn
,
4-5 ~ r
UP
En1,610131
14 r613
I P.0379.0229
\
Fathers
~_ . -
123 4-5
41
145
10
.0690
2551
11
.0400.0196
18
197
5
0254
84
.0595
43
4
.0930
9
.1111
Total
1,842
71
.0488.0385
26723
28
.0387
291
8
.0275
122
7
.0574
2
.0769
118
.0556
11
1
.0909
En2,400319163962,978
Total: r8314116114
tp.0346.0439.0675.0625.0383
Nagasaki
Fathers
-
-
1234-5Total
~ n1,40121426141,655
14 r70114-85
lip.0500.0514.1538 .0514
~ n85336332111,259
2g r42232 67
l p.0492.0634.0625-.0532
in6225106103
34 r11 - 11
p.1774 -.1068
Cn
4-5: r
UP
55 12 4 1 72
.0545
.0417
Jn2,37161472 323,089
Total r126346 166
l P.0531.0554.0833 .0537
The Comparability of Irradiation Subclasses
common following the bombing than before.
However, there is no apparent, consistent rela-
tion to exposure history. Three of the interac-
tion terms are significant.
In view of the possibility that it might be
easier to obtain a history of a "therapeutic" D
and C than of an induced abortion, a question
concerning D and C was included in the Ge-
netics Long Form. The Endings in those termi-
nations where the registration number ended in
"O" are shown in Tames 5.15 and 5.16. The
figures refer to the proportion of women report-
ing one or more D and C's, with no reference
TABLE 5.12 CHI-SQUARE ANALYSIS OF THE
FREQUENCY OF POSITIVE SEROLOGIES BY
CITY AND PARENTAL EXPOSURE
(Unrelated parents)
Source DF X 2
Total 31 60.330
Interactions, first order
CM ...............
CF ...........
ME ...........
Main effects
Cities (C) ........
Fathers (F) a
Hiroshima .......
Nagasaki ........
3
.... 3
9
. . ~
.. 1
3
3
Sum 6
Mothers (M) a
Hiroshima
Nagasaki
Sum
a Adjusted for cities.
3 2.136
3 6.099
. 6 8.235
63
nancies in Nagasaki than in Hiroshima. Fur-
thermore, particularly with respect to mothers'
exposure in Nagasaki, there is apparent a tend-
ency for the more heavily irradiated mothers
to have borne more children.
5.7 Porrer~tal coopero~tiorz. If for any reason
one group of parents was more cooperative than
another in permitting an examination of their
p
6.645
3.522
2.876
.001-.01
6.469 .05-.10
3.106 .30-.50
9.575
newborn child, herein lies a source of bias.
Fortunately, parental cooperation was excellent
throughout the course of this study. Refusals to
permit the ABCC physician to examine a child
were infrequent, so infrequent that no analysis
of the phenomenon has been carried out. It
should in this connection be pointed out that
such refusals could influence only the data on
malformation, since sex, birthweight, and the
occurrence of stillbirth were reported by the
midwife, and since, further, it would be im
°°~~ pi practical to attempt to conceal a neonatal death.
In Japan as in the United States, the birth of
a malformed child tends to stigmatize the par
ents in the lay mind. In a country where mar
riages are still often arranged to a large extent
by the families concerned, and where the koseki
(census register) is freely consulted when a
marriage is under consideration, there are at
work social factors encouraging the concealment
.~0-.20 of congenital abnormality. In this connection,
the possibility has to be recognized of subtle
influences tending to make one of the groups
under study more cooperative than another
group. More specifically, one had to recognize
the possibility that some of the "exposed"
group, learning from the newspapers that con-
genital abnormalities were a possible aftermath
of exposure to an atomic bomb, would cooperate
more completely than the controls, and thus
introduce a bias.
The chief opportunity for concealing major
congenital defect lay in children who were still-
born or died shortly following birth, and whose
bodies were disposed of without being seen by
an ABCC physician. As mentioned earlier (Sec.
2.5), an attempt was made to obtain autopsies
on as many children who were stillborn or died
during the neonatal period as possible. In 1950
and 1951, when the autopsy program was in
full stride in Hiroshima, autopsies were obtained
on approximately 50 per cent of all children
who were stillborn or died during the neonatal
period. Another 15 per cent of all children who
were stillborn or died during the neonatal
.50-.70
.~0-.20
.20-.30
as to whether the event preceded or followed
the bombings. There is a striking difference be-
tween the cities, similar in direction to that re-
ported with respect to induced abortions, but
much greater in magnitude. However, no differ-
ence with respect to exposure groups is apparent,
nor are any of the interaction terms significant.
5.6 The frequency of repeat registrations.
Depending upon the number of pregnancies
which she experienced during the period cov-
ered by this study, a woman residing in Hiro-
shima or Nagasaki could bear one, two, three,
four or even more infants whose examination
fell within the scope of the program. Table
5.17 summarizes the mean number of infants
per mother in relation to the exposure categories
of mother and father, by city. There is a tend-
ency towards a higher average number of preg
66r Ger~etic Effects of Atomic Bombs Chapter V
TABLE 5.13 FREQUENCY OF MOTHERS REPORTING ONE OR MORE INDUCED
ABORTIONS BY PARENTAL EXPOSURE, CITY, AND TIME
(Unrelated parents)
Hiroshima before AUg. 6, 1945
, ^ ~
Fathers
-
Total
rrn4,738310115795,242
14 r2071644231
(p.0437.0516.0348.0506.0441
pn1,037902137782,154
2g r42286177
(p.0405.0310.0438.0128.0357
rn43716528183966
3gr1936129
tp.0435.0182.0214.0120.0300
v,
o
-
Jn1867362 66387
4-5> r274 13
I~P.0108.0959.0645 ~.0336
l
n6,3981,450 5953068,749
Total] r27054 206350
(p.0422.0372 .0336.0196.0400
Hiroshima after Aug. 6, 1945
Fathers
1 2 3 4-5 Tc~tal
~n 10,599 799 426 223 12,047
I1 ) r 581 50 20 19 670
p .0548 .0696 .0469 .0852 .0556
rn 2,860 402 128 85 3,475
2g r 187 18 9 4 218
p .0654 .0448 .0703 .0471 .0627
n 1,185 127 99 42 1,453
J 3~` r 77 8 7 - 92
tp .0650 .0630 .0707 .0633
rn 599 51 39 19 708
4-5 g r 45 4 7 56
(p .0751 .0784 .1795 .0791
__
~n 15,243 1,379 692 369 17,683
Total: r 890 80 43 23 1,036
I
The Comparability of Irradiations Subclauses
TABLE 5.13 Continued
Nagasaki before Aug. 9, 194S
, ~
Fathers
1 2 3 4-5
-
Total
rrn5,61670665346,421
1j r12117 138
(p.0215.0241- .0215
En1,9182,322128804,448
2/ r27273158
(p.0141.0116.0234.0125.0130
En1711205829378
34 r 1 12
lip .0083 .0345.0053
En87482110166
4-5gr --11
(p -.1000.0060
On7,7923,19627215311,413
Total] r1484533199
tp.0190.0141.0110.0196.0174
Nagasaki after Aug. 9, 1945
Fathers
^
-
~n 7,953
1: r
IMP
v,
: -
a~
o
1 2
182
.0229
4,692
98
.0209
Jn
2~ r
UP
En
33 r
UP
rn
4-5 ~ r
UP
361
.0111
291
5
.0172
1,073
23
.0214
1,028
20
.0195
120
4
.0333
56
1
.0179
Total
129 859,240
2 207
.0155 .0224
90 555,865
3 -121
.0333 -.0206
25
13 8
509
8
.0157
368
6
.0163
En13,2972,277257 15115,982
Totals r289485 -342
(p.0217.0211.0195 -.0214
65
66 Genetic Efects of Atomic Bombs Chapter V
TABLE 5.14 CHI-SQUARE ANALYSIS OF THE FRE
QUENCY OF MOTHERS REPORTING ONE OR MORE
INDUCED ABORTIONS BY PARENTAL EXPO
SURE, CITY, AND TIME (Unrelated parents ~
Source DF X 2 p
Total 63 574.384 <.001
Interactions, first order
CT .....
TM ....
TF ....
CM ....
CF ....
MF . . .
. . ..
1 3.057
3 16.265
3.725
3
3
9
.05 -.10
.00 i-.01
.20 -.30
0.087 .01 -.02
2.616 .30 -.50
30.308 <.001
Main effects
Cities (C)
Before bombing . . . 1 95.230 < .001
After bombing 1 270.261 <.OO1
Sum 2 365.491 <.001
Time (T)
Hiroshima (H) 1 40.677 <.OO1
Nagasaki (N) 1 5.401 .02 -.05
Sum 2 46.078 < .001
Mothers (M)
(H) Before bomb
ing 3 6.191 .10-.20
After bomb
ing 3 9.001 .02 -.05
(N) Before bomb
ing 3 15.700 .001-.01
After bomb
ing 3 0.459 .90 -.95
Fathers (F)
H ) Before bomb
ing 3 5.039 .10-.20
After bomb
ing 3 3.127 .30-.50
(N) Before bomb
ing 3 3.898 .20-.30
After bomb
~ng
Sum
.80 -.90
... 3
... 12 12.915 30.-.50
0.851
The Comparability of Irradiation Subclasses
TABLE 5.15 FREQUENCY OF DILATATION AND CURETTAGE BY PARENTAL
EXPOSURE AND CITY: ZERO TERMINATIONS
Hiroshima
us
Fathers
~- ~
1 2 3 4-5Total
(n1,77014365462,024
1\ r26619105300
up.1503.1329.1538.1087.1482
rn5691714728815
24 r751656102
up.1318.0936.1064.2143.1252
rn218265619319
3/ r34111147
up.1560.0385.1964.0526.1473
En103211011145
4-S] r1742124
up.1650.1905.2000.0909.1655
fn2,660 3611781043,303
Totals r392 402813473
l p.1474 .1108.1573.1250.1432
Nagasaki
Fathers
1 2 3 4-5 Total
1<,
2 <,
con
3:
UP
1 n1,45922727141,727
I r2531 29
p.0171.0132.0370 .0168
in89438733141,328
r12121126
p.0134.0310.0303.0714.0196
1 n6129106106
;r-11 2
up .0345.1000-.0189
En56124375
~53 r
En2,47065574373,236
Total] r37163157
up.0150.0244.0405.0270.0176
67
68 Genetic Efects of Atomic Bombs Chapter NO
TABLE 5.16 CHI-SQUARE ANALYSIS OF THE FRE
QUENCY OF DILATATION AND CURETTAGE BY
PARENTAL EXPOSURE AND CITY: ZERO
TERMINATIONS (Unrelated parents)
Source DF X 2
31 372.060
Total
Interactions, first order
CM 3 3.099
CF 3 7.426
MF 9 10.799
Main effects
Cities (C)
1
346.140
< .001
.30-.50
.05-.10
.20-.30
< .001
Mothers (M) a
Hiroshima 3 3.212 .30-.50
Nagasaki 3 1.718 .50-.70
Sum ......
Fathers (F) a
Hiroshima . . .
Nagasaki ....
4.930 .50-.70
3
.20-.30
.1~.20
Sum 6 9.219 .10-.20
a Adjusted for cities.
TABLE 5.17 MEAN NUMBER OF REGISTERED PREGNANCIES PER MOTHER
BY PARENTAL EXPOSURE AND CITY
Hiroshima
Fathers
1 2
us
at
Total
:n13,5771,155490301 15,523
l x1.2881.3111.2371.326 1.288
2: n3,9641,472308189 5,933
) ~1.3651.2741.2631.312 1.335
~ n1,653318420121 2,512
Six1.3451.3461.2601.298 1.328
4En8211568299 1,158
-5 lx 1.378 1.333 1.341 1.2221.356
r n 20,015 3,101 1,300 71025,126
Totall 1.311 1.298 1.257 1.3031.307
Nagasaki
Fathers
1 2 3 4-5
Total
1 ~ n10,5811,5821669712,426
Ax1.3981.3961.4761.4541.400
2 in6,2462,9711881229,527
lx1.5141.4101.4841.4591.481
r n4761967520767
3: x1.6321.4491.3601.7501.532
4 5 r n345812421471
lx-1.6321.4691.4581.3331.582
r n17,6484,83045326023,191
Total:.1.4491.4081.4591.4691.441
The Comparability of Irradiation Subclasses
period were seen by a physician in the employ
of the ABCC but did not come to autopsy. It
follows that opportunities for the concealment
of defect existed with respect to only 35 per cent
of all stillbirths and neonatal deaths. In 1952
and 1953 the efficiency of the autopsy program
increased, to the point where some 60-70 per
cent of the possible material was being autop-
sied, with a corresponding decrease in onoor-
tunities for concealment. Tables 13.1 through
13.4 analyze the material coming to autopsy
with reference to its randomness. It can be
shown that there is no detectable bias as regards
the exposure history of the parents. Since this
autopsy material includes the majority of the
stillbirths and neonatal deaths, it seems unlikely
that there exists the possibility of a serious bias
as regards parental exposure among the still-
births and neonatal deaths who were not seen.
If, for instance, unirradiated mothers of still-
born (malformed) children were especially
prone to dispose of these infants without their
coming to the attention of the ABCC, then,
since approximately 10 per cent of stillborn
children are malformed, this should depress the
representation of unexposed mothers in the
parents of the autopsy material; such a dispro-
portion was not observed. On the other hand,
the data are of course not sufficiently extensive
to exclude small biases in this direction. In this
connection, however, it must be borne in mind
that all of the infants falling in this 25 per cent
were seen by a Japanese midwife or obstetrician,
which provides a partial safeguard against the
concealment of defect, even though there were
occasionally encountered striking shortcomings
in midwife reporting.
5.S Late seq~elae of exposure to the bombs.
-Reference has already been made (Sec. 4.1)
to the occurrence among the parents of the in-
fants under study of certain late sequelae of
exposure to the atomic bombs. The best docu-
mented of these late effects are cataracts (Cogan,
Martin, and Kimura, 1949; Sinsky, 195S) and
leukemia (Folley, gorges, and Yamawaki,
1952; Lange, Moloney, and Yamawaki, 1954;
Moloney and Lange, 1954; Moloney and Kas-
tenbaum, 1955 ~ . Refractory anemia may also
be a delayed manifestation of radiation injury
(Lange, Wright, Tomonaga, Kurasaki, Mat-
suoke, and Matsunaga, 1955 ~ . These events
occur with a frequency which cannot be ignored
in a study of this type. Thus, the findings with
respect to leukemia are given in Table 5.18. To
69
date, the over-all frequency of this disease in
individuals who at one time displayed one or
more of the three radiation symptoms, epilation,
petechiae, or gingivitis, is 0.5 per cent.
While the occurrence of cataracts would not
be expected to influence pregnancy outcome,
there is little doubt that a disease with the
systemic manifestations of leukemia has pro-
found effects. Whether in addition to leukemia
there are other serious sequelae which have so
far escaped detection cannot be said. Fillmore
(1952) was not able to detect any significant
sequelae in a general medical examination of
78 persons who had received sufficient radiation
at the time of the bombings to develop, later,
radiation cataracts. On the other hand, Lorenz
and his collaborators (1954) have demonstrated
a shortened life span in a variety of animals
exposed to chronic irradiation, associated with
the development of such conditions as pancyto-
penia and lymphoid, pulmonary, ovarian, and
mammary tumors. Whether there are compar-
able effects in man is not yet known.
Whereas techniques exist for coping with
the age-parity and consanguinity differences be-
tween the groups of parents with whom we are
concerned, it is much more difficult in any plan
of analysis to make allowance for the somatic
effects of irradiation which might influence the
outcome of a pregnancy. These somatic effects
should be exerted largely through the mother.
Thus, "effects" apparently consequent upon
maternal radiation which are not confirmed by
a corresponding analysis with regard to pa-
ternal radiation history must be viewed with
reservations.
5.9 The changing proportion of control arid
irradiated from year to year. Table 2.1 pre-
sented a summary of the number and propor-
tion of registrations with at least one parent
falling into categories 4 or 5 for the years 1948
through 1953. There was apparent a recent de-
crease in both the absolute and the relative
representation of the more heavily irradiated.
If there were any marked tendency for the level
of diagnostic accuracy or parental cooperation
with respect to congenital malformation to
change during the course of the study, or if be-
cause of post-war improvements in medical and
economic levels the stillbirth or the neonatal
death rate fell appreciably or the birthweight
increased, then here again are factors capable
of introducing a source of spurious conclusions.
Tables 5.19 and 5.20 summarize our annual
70 Generic Effects of Atomic Bombs Chapter V
TABLE 5.18 INCIDENCE OF LEUKEMIA IN THE HIROSHIMA SuRV~VORs OF THE
ATOMIC BoMs~NG As RELATED TO DISTANCE FROM THE HYPOCENTER
AND THE PRESENCE OF SEVERE RADIATION COMPLAINTS
(After Moloney and Kastenbaum, 1955)
Distance from Population a
hypocenter , ^_
(meters) SRC b NRC c Total
0- 999 .......................... 750 450 1,200
1,000-1,499 .......................... 2,250 8,250 10,500
1,500-1,999 .......................... 1,750 16,950 18,700
2,000-2,499 .......................... 950 16,250 17,200
2,500 and over 850 49,650 50,500
Total 6,550 91,550 98,100
Cases of leukemia Incidence
~, ~
SRC NRC Total SRC NRC Total
14 1 15 1: 53 1: 450 1: 80
15 9 24 1:150 1: 917 1: 438
3 2 5 1:583 1: 8,475 1: 3,740
1 1 2 1:950 1: 16,250 1: 8,600
0 4 4 1: 12,412 1: 12,625
33 17 50 1:198 1: 5,385 1: 1,962
a Population estimated and rounded off to the nearest 50 persons. These
population figures were based on the Commission's 1949 radiation census and
the Japanese national census (1950~. Numbers of survivors with severe radia-
tion complaints were estimated from an analysis of the pregnancy registration
data.
b SRC: severe radiation complaints (heavily irradiated).
c NRC: no radiation complaints (lightly irradiated).
TABLE 5.19 THE FREQUENCY OF MALFORMATIONS BY YEAR AMONG
THE OFFSPRING OF PARENTS NEITHER OF WHOM WAS
EXPOSED TO THE ATOMIC BOMBS
Hiroshima
Total births
Malformations
Percentage . .
....... 1,756 4,005
14 34
0.80 0.85
DF= 4
X a = 2.496
1948 1949 1950 1951 1952 Total
3,602 3,324
34 36
0.94 1.08
0.70 > P > 0.50
3,084 15,771
35 153
1.13 0.97
Nagasaki
A_ ~
1948-49 1950 1951 1952 Total
Total births 3,9343,2433,1893,12313,489
Malformations 31364532144
Percentage 0.791.111.411.021.07
X2=6.584 DF=3
0.10 > P > 0.05
The Comparability of Irradiation Subclasses
TABLE 5.20 THE FREQUENCY OF STILLBIRTHS BY YEAR AMONG THE
OFFSPRING OF PARENTS NEITHER OF WHOM WAS
EXPOSED TO THE ATOMIC BOMBS
Hiroshima
Be. .
1948 1949 1950 1951 1952 Total
Total births 1,742 3,971 3,564 3,287 3,037 15,601
Stillbirths 51 87 64 58 61 321
Percentage 2.93 2.19 1.80 1.76 2.01 2.06
x2=9.544 DF_4 0.05 > P> 0.02
Total births ..........
Stillbirths . . .
Percentage ............
% 2= 4.372 DF= 3
figures for the frequency of gross malformations
and stillbirths for the first five years of the study
among the offspring of parents neither of whom
was exposed to the atomic bombs. The £re-
quency of malformations and stillbirths appears
to remain sufficiently constant from year to
year, that time trends should not complicate the
analysis of these two indicators. On the other
hand, we shall in a subsequent section (Sec.
10.4) present evidence that time trends may be
of importance in the birthweight analysis.
5.10 The backgroz~r~d of group 1 individuals.
-There remains one final problem for dis-
cussion. There are important differences in back-
ground between the parents in radiation cate-
gory 1 and those in categories 2 through 5. The
latter have all been urban dwellers since 1945
or before. The former parents, although in part
composed of residents of Hiroshima and Naga-
saki who were away at the time of the bombings,
and also of urbanites who have come to Hiro-
shima and Nagasaki from the other cities of
Japan, in addition include a high proportion
of emigrants from rural areas and repatriates
from Manchuria, Korea, and Formosa. In addi-
tion, among the men there are many with ex-
tensive overseas military service. Although there
are well-known urban-rural differences with
respect to the indicators of possible genetic
effect here considered, it is by no means clear
to what extent these are apparent and to what
71
Nagasaki
1948-49 1950 1951 1952 Total
3,200
61
1.91
3,134 3,084 13,320
49 67 240
1.56 2.17 1.80
0.30 ~ P > 0.20
extent real, and if the latter, whether they are
socio-economic or biological in origin. Finally,
if biological, there are no data as to how long
they persist after urbanization. In the face of
this situation, one can only conclude that the
significance of any finding based solely on a
difference between category 1 and categories 2
through 5 collectively must be viewed with
reservations.
5.~! Summary. Of the differences between
the individuals in the various exposure subcate-
gories which have come to light in the forego-
ing analysis, some appear to be inconsequential,
of such a nature that they can be ignored with
safety. Others of the differences appear to repre-
sent sources of potentially significant bias. These
latter differences fall into a natural dichotomy,
depending on whether or not allowance can be
made for them in an analysis. Thus, one way or
another, differences in the frequency of con-
sanguinity, age, and parity can be circumvented.
This is not so either with respect to the occur-
rence of late sequelae of the bombing among the
more heavily irradiated, the possible implica-
tions of the progressive decline during the six
years covered by this study in the proportion of
category 4's and 5's among the registrants, or
the differences in background among persons
falling into category 1 as opposed to categories
2 through 5. In the following chapter, the steps
taken to meet this situation will be described.
