close

Вход

Забыли?

вход по аккаунту

?

Comparison of genetic and anthropological interpretations of population isolates in Aguacatenango Chiapas Mexico.

код для вставкиСкачать
Comparison of Genetic and Anthropological
Interpretations of Population Isolates in
Aguacatenango, Chiapas, Mexico
ROBERT P. ERICKSON,I SARA NERLOVE,2 WILLIAM
AND A. KIMBALL ROMNEY3
Departments of Anthropology and Medicine, Stanford University,
Stanford, California 94305
P. CREGER
ABSTRACT
Intensive demographic and genetic studies were made of Aguacatenango,
a partial isolate population in Chiapas, Mexico. The census showed higher rates of
birth than mortality, with a stable population maintained by emigration. Complete
enumeration showed many discrepancies between the ideal and actual mating
patterns. Though ideally a n endogamous town, many marriages were to outsiders.
Also many occurred between the two allegedly endogamous barrios of the town. Blood
group determinations on a third of the population provided a second description of
the isolate. This description suggested that although the population is nearly a
homogeneous genetic unit, the elder residents of one bani0 show genetic differences
from those of the other. That the barrios might have indeed been genetic isolates
was further suggested by a better fit to Hardy-Weinberg equilibria for each barrio
separately than for the isolate as a whole. These blood group data were compared to
blood group data on one other isolate in the region revealing marked gene frequency
differences. It is suggested that similar studies are needed o n many populations if
the effects of particular population structures on gene drift and selection are to be
elucidated.
Interpretation of paputation isolates
There has been increasing interest in
the simultaneous study of polymorphic
physiological characters and demographic
parameters in isolate populations (Neel et
al., '64; McKusick et al., '64; Arends et al.,
'67). Assessment of the contribution of
gene drift and natural selection to genotype frequency changes has been sought
in studies of small, primitive communities
which have high childhood death rates.
Such communities with historical records
of some depth have provided data on migration distances, consanguinity and the
"founder" effect for tests of population
genetic model (Cavalli-Sforza et al., '64;
Roberts, '68). Cultures in different parts
of the world have widely divergent customs
in regard to the choice and number of
mates, the stability of a particular mating,
the sex and number of offspring desired,
etc. The effects of these aspects of a population's mating structure (which are not
fully expressed in the co-efficient of inbreeding and differential fertility of population) on changes in allele frequency are
AM. J. PHYS.ANTHROP.,32: 105-120.
being analyzed (Neel et al., '64; Gajdusek,
'64; Yasuda and Morton, '67).
The present study was made in Aguacatenango, Chiapas, Mexico, a small, primitive Tzeltal-speaking Mayan community
with a high childhood death rate, for which
there are some historical records available.
One aim of the study was to examine this
semi-isolate Mayan Amerindian population
in order to provide further information on
the variables involved in allele frequency
changes. Another aim was to compare the
allele frequencies of this semi-isolate to
those of another nearby in order to contribute to the studies of divergence among
related groups (Neel and Salzano, '64). In
the course of the examination of allele frequencies, the focus of the study became the
reconciliation of some discrepancies between the anthropological and genetic descriptions, both of which emphasize the
importance of migration and thus of "open"
population models.
1
Present address:
National Institute of Arthritis
and Metabolic Diseases, Bethesda, Maryland.
2 Present address:
University of California, Irvine,
School of Social Sciences.
3 See footnote 2.
105
106
ERICKSON,
NERLOVE, CREGER AND ROMNEY
Historical background of Tzotzil and
Tzeltal-speaking groups
At present Tzotzil and Tzeltal, two
closely related Mayan languages, number
about 160,000 speakers who are situated
in the highlands of Chiapas. Archeological and linguistic evidence suggests that
about 1200 years ago a Mayan group migrated from the mountainous heights that
now form the northern rim of Guatemala
to the Chiapas Highlands, crossing some
50 miles of low plateau that intervenes between the two highland areas. The archeological evidence indicates that as of about
a millenium ago, the highland region became more densely inhabited and characterized by uniformity both in settlement
pattern and ceramic inventory ( Adams and
McQuown, ’59). These changes can be regarded as the consequence of an influx
of population with cultural similarities
(particularly in ceramics j from the zone
of Classic Mayan civilization in the lowlands to the east. The immigrating Mayans
spread over the highlands and differentiation of their language into “dialects” began. According to lexicostatistical indices
(Swadesh, ’59), Tzotzil and Tzeltal separated about 1,200 years ago. The Tzotzil
and Tzeltal Indians did not remain completely free of foreign influences after
settling in the highlands. During the ninth
and tenth centuries, a Nahuatl-speaking
group from central Mexico penetrated into
the area as far as the western edge of the
Chiapas Highlands; and during the early
sixteenth century, two of the northern
Tzotzil groups were conquered by Montezuma I1 (Calnek. ’59).
In 1524, Luis Marin began the conquest
of the Tzotzil and Tzeltal Indians. His
victories were facilitated by mutual animosity among subgroups (clan groups
with ceremonial centers replete with ball
courts and small pyramids) of Indians.
The Zinacantecos helped him defeat the
Chamulans and Huistecos (these 3 Tzotzil
groups are still present in Highland Chiapas). Marin did not attempt to stay in
the Chiapas Highlands, but he returned in
1528 to found Ciudad Real, now Ciudad
de las Casas. Thus began the colonial
period, the dominant characteristics of
which seem to have been established dur-
ing the fourth and fifth decades of the
sixteenth century.
Ethnohistorical evidence shows that
Aguacatenango has had a continuous population almost since the time of the Spanish conquest. The first census report found
for Aguacatenango is for the year 1611
AD. when the population was 720 individuals (Calnek, ’61). Later, a second town,
Quesaltepec, was added to Aguacatenango
during one of the directed population
movements in the seventeenth century.
Quesaltepec probably originated independently but first appears on the census roles
as a parcialidade (land possession) of
Aguacatenango.
Aguacatenango: A Tzeltal community
Aguacatenango, a community of Tzeltal
speakers situated in the highlands of Chiapas, Mexico, constitutes the partial isolate
of Mayan Amerindians of this study. In the
summer of 1963, its Indian population was
1405 (fig. 1). It is a nuclear town whose
residents live in an area of several acres
in which there is space for a limited cultivation of fruits, vegetables and some
maize, which is the staple crop. Most of
the farming, however, is done in the surrounding country to which the men commute in order to cultivate their milpa or
maize fields.
A few families of Zadinos (people who
consider themselves to be non-Indian and
a part of Mexican national culture) live in
Aguacatenango. They run small shops selling a variety of goods. One ladino is the
“engineer” for the electric and water supplies and a power corn-grinder. Four of
the Indian homes serve also as cantinas
(bars) which have electric record players
with public address systems over which the
purchase of trago (native distilled fermented sugar cane juice) is encouraged.
Most households have cold running water
which comes through a faucet in the yard,
a few have one electric light, and some
have houses with tile roofs which are considered the height of achievement. There
is a three-year primary school which
teaches a moderate percentage of the children, mostly males, enough Spanish so
that they can get along outside of the village, while a rare individual will learn
the rudiments of reading and writing. The
Isthmus
'. -._.
-'..
__
.'..
'... ..
-._
1
.
1
.
.
-..
\..'
I
Fig. 1
Location of Aguacatenango.
Mexican National Indian Institute (I.N.I. )
has provided an Indian high school i n
nearby Ciudad de las Casas, the highland
commercial center. A few Indians from
various villages are thus accommodated
for further education and some of the
graduates return to the villages as teachers
or nurses. I.N.I. has also built a clinic in
Aguacatenango which has such a nurse
and is visited weekly by a doctor.
Ethnographic data provide ideal descriptions of mating behavior in Aguacatenango
108
ERICKSON, NERLOVE, CREGER A N D ROMNEY
which include lineage exogamy, and town
and barrio endogamy. The kinship structure consists of a localized group of patrilineal kinsmen which constitute at least
the nucleus of a lineage. This nucleus
functions as a unit within which the individual seeks advice in time of indecision
and protection in time of trouble (Metzger,
’56). These “loose-lineages’’ involve some
three generations of patrilocal families.
Four generations might be involved when
a linking relative is recently deceased and
still remembered. Members of a patrilineal
lineage share the same family name of
Spanish origin, e.g., Mendez, but one such
name may be shared by several lineages in
Aguacatenango.
Exogamy within the patrilocal looselineage is required, but genealogies are seldom remembered past three generations so
many third-cousin marriages can occur.
The inhabitants of Aguacatenango are
aware of a few exceptions to the ideal of
patrilineal exogamy : there are known occurrences of marriages between patrilineal
first cousins. Limiting the duration of a
given loose-lineage are human memory and
strong male sibling rivalry which tends to
speed the separation of the lineage into
several lineages. Termination of a lineage
occurs for a lack of male offspring to continue it.
Aguacatenango is divided into two barrios (territorial divisions) one on each
side of the main central plaza in which
the Catholic church is situated. The plaza
forms part of the barrio boundary and the
remainder of it runs between bordering
household plots in an unmarked, but well
recognized, manner (fig. 2). The origin
of the division of Aguacatenango into two
barrios is unknown although it may be a
remnant of old clan organizations, or it
may be the results of the incorporation of
the parcialidade, Quesaltepec, into the
town. It is marital, not political or religious
interaction, which is limited between the
two barrios. The normative descriptions of
behavior indicate barrio endogamy although, again as with the ideal of lineage
exogamy, exceptions are known.
Following the Mexican Revohtion, the
town acquired lands about four miles away
on a lower plateau. Some families moved
to be near their milpa on this plateau,
forming two “colonies” there. These colonies are gradually developing their own
rules of endogamy although they often immigrate back to Aguacatenango. The colonists still account for the major portion of
immigration to Aguacatenango as well as
emigration from it. One colony is named
El Puerto; the other remains unnamed.
SAiMPLE AND METHODS
Anthropological field work in Aguacatenango began in 1956 and has been continued by several different anthropologists
to the time of the present study. Demographic data for this study were collected
with the aid of interpreters in the summer
of 1963 during a period of seven weeks of
intensive field work. The genealogical data
were gathered during the course of a complete census of the 253 households then
comprising the town of Aguacatenango.
The simultaneous gatherings of genealogies and census and the availability of
genealogical data from previous studies
provided maximal opportunity for crosschecking these data. For each household,
questions asked included women’s fertility
histories (including number of still births),
birth place of each member of the families
of orientation and procreation, estimated
date, and age at death of all deceased relatives, number and order of spouses, duration and outcome of marriages, and the
present location of siblings or children who
have left the village. Estimated ages of
elderly and deceased persons, as well as
numbers of still births and infant deaths
are unavoidably crude. Birth records were
available for over a third of the population
and were used whenever possible for establishing ages. Death certificates were seldom issued.
Initial blood group studies were performed on 250 samples of blood drawn on
several weekends during the summer of
1962 (Erickson et al., ’63). These samples
were tested with suitable antisera for A,
B, M, N, S, C, D, E, c, e, Kell, Cellano,
Duffy and Diego antigens. Known blood
cells (Pano-cells, Knickerbocker Co., New
York, N. Y.) and duplicate samples (single
blind) were used to control the groupings
studied. These controls showed many inaccuracies in the Rh system.
I N T E R P R E T A T I O N S O F P O P U L A T I O N ISOLATES
TO Outlier SE
4
~o Outlier P
I\
4
Fig. 2
Map of Aguacatenango, Chiapas, Mexico.
109
110
ERICKSON, NERLOVE,
CREGER AND ROMNEY
Demography
The results of the preliminary blood
group study led to a n investigation of a
The complete census revealed that the
larger sample of the population for the an- barrio division is not the only spatial divitigens which were found to vary signifi- sion with social implications of the popucantly in the isolate. Many biochemical lation of Aguacatenango (fig. 2). There
traits, such as red blood cell proteins are six outliers, four of which are named.
(Lisker et al., '66) or serum proteins (11, 111, IV, and V numbered for purposes
(Lisker et al., '67; Melartin et al., '67) are of reference). Each of the outliers has its
not useful markers for this population be- own cross, which is a n indication of some
cause of insufficient variability in geno- autonomy.
type frequencies. Blood samples were colOutlier I, which is adjacent to Barrio 2,
lected from 460 individuals on two consec- is comprised of eight households of Tzeltalutive days during the summer of 1964. speakers. All members of these households
They were taken from individuals who were were born outside or had parents who were
willing to give blood in exchange for five born outside Aguacatenango. Outliers 11,
pesos (4OC U. S . ) . The Mexican national 111, and IV form a semicircle around the
election day was selected for the first day periphery of Barrio 1. Two additional outof blood collection as everyone gathered in liers are across the road from Aguacatethe town for the balloting and ensuing fes- nango (Outliers V and VI). Outlier V is
tival. The town elders (officials), with comprised of seven households of Aguacawhom the anthropologists had maintained tenangueros who moved from a group of
good rapport through the years, were the households on the periphery of Barrio 1.
first to donate. About 20 more older men, Outlier VI is comprised of 12 households
closely associated with the elders, followed made up of people from Huistan, a Tzotzilthem, but all other mature males were ex- speaking town several miles from Aguacatenango.
tremely reluctant to donate. Instead, they
In summary, the named Outliers 11, 111.
sent their wives and children, who thus IV, and V have more intimate connections
made up three-quarters of the sample. both spatially and socially, with the two
Young males were particularly eager to do- barrios of Aguacatenango. Outliers I and
nate blood as they could keep the money VI have no blood or affinal ties with either
for themselves to spend at the festival.
of the two barrios.
The clotted samples were packed in ice,
Three households cannot properly be
shipped, and arrived at Stanford Univer- considered part of either outliers or barrios
sity five days later. These cells were tested by virtue of both birth place and interacfor A, B, M, N, S, C, D, E, c, and e anti- tions of the residents. There is one family
gens. Of the sample of 460 donors, 400 of Tzotzil Indians from the town of
were definitely identified on the geneal- Huistan near Outlier 111, and there are two
ogies and their blood samples were used families cf Tzeltal Indians of unknown
for computations of phenotype frequen- origin near the northeastern periphery of
cies. Of the 60 donors dropped from the Barrio 2.
The household and population distribusample, blood group data from 50 of these
tion
across barrios and outliers in Aguacawere not used because the individuals were
not from Aguacatenango proper, but from tenango is presented in table 1. Population
pyramids of age and sex composition for
El Puerto, one of the two colonies nearby. Barrio 1, Barrio 2, and Outliers 11-V are
The remaining ten blood samples were dis- presented in figure 3 . Fifty-seven per cent
carded because they were duplicates or of all the males are 20 years of age or
there had been a mix-up in sample num- younger. The birth rate is 67 per 1,000.
bers. Blood group data from each of the A five-year period was selected for investi400 subjects studied were punched on key- gation of mortality rates among nonsort cards along with the donor's age, emigrant Aguacatenangueros ( 1300 indibarrio number and the name of his patri- viduals). The total number of deaths in
each age interval of five years was divided
lineal lineage.
111
INTERPRETATlONS O F POPULATION ISOLATES
TABLE 1
Distribution of the households and population of Aguacutenango ucross burrio and outliers
No. of
households
Barrio I
Barrio I1
Outliers 11-V
Outlier I
Outlier V I
Misc.
No. of males
No. of females
212
255
164
16
218
75
95
60
8
12
Totals
3
8
276
175
24
29
7
253
676
729
v)
51-60
a
a
41-50
0
w
”
31-40
4
21-301
b l r t n - 10
L
zoo
160 120 80
40 o
40
NUMBER
DO
120 160 200 240 280
Fig. 3 Age and sex distribution of barrios I
and I1 and outliers 11-V (for 1300 individuals,
631 males and 669 females - read 233 males
under 10 rather than 205).
by five to give a n average estimate for
deaths occurring in one year. A separate
interval was used for one year and under.
The deaths occurring in this interval include still births. The mortality in Barrio
1, Barrio 2, and Outliers 11-V is presented
in table 2.
Fertility was tabulated on the basis of
completed families for the population of
Barrio 1, Barrio 2, and Outliers 11-V (table
3 ) . The sample upon which the tabulation
is based includes all women who were 50
years and over during the summer of 1963.
None of these women had given birth
within the last five pears. There were 53
such women, one of whom was dropped
from the sample because of inadequate
information. Twenty-six of the 52 women
in the sample were currently living with
their husbands. The number of children
included on a woman’s fertility history
includes still-borns and deceased as well
as all living children. The particular father
of the child and the child’s present residence were not considered.
21
No. of individuals
430
53 1
339
40
50
15
1,405
Population structure of
Barrios 1 and 2
The genetic contributions from the outliers, from the other barrio, and from outside villages to Barrios 1 and 2, respectively, were not considered. These genetic
contributions were, at best, difficult to
calculate. The time depth which i t is feasible to include in such calculation is quite
limited. Data for the calculation were available for 961 individuals. The contributions
from the outliers were computed on the
basis of the location in which subjects’
parents reside. The contribution from one
barrio to another was computed from all
cross-barrio marriages. Those offspring
from cross-barrio marriages who live in
Barrio 1 are considered contributions of
Barrio 2 to Barrio 1. Similarly, those offspring from cross-barrio marriages who reside in Barrio 2 are considered contributions of Barrio 1 to Barrio 2. The genetic
contributions from outside the town have
been tabulated on the basis of outsiders
married to Aguacatenangueros with offspring living in Aguacatenango. These genetic contributions to each barrio from
these various sources are shown in table 4.
The chi-squares show significant differences between the barrios in the various
rates of genetic contributions.
The general picture of genetic contributions indicates that Barrio 2 is more isolated than Barrio 1. The proportion of the
population of the outliers originating in
either of the two barrios could not be ascertained. As noted, the connections of
Outlier V are with Barrio 1. It seems that
the populations of Outliers 11, 111, and IV
also derive from Barrio 1. Note, however,
that a genetic contribution from groups
which have derived from the group which
112
ERICKSON, NERLOVE, CREGER AND ROMNEY
TABLE 2
Mortality in Aguacatenango froin interview data
Deaths
Deaths
FeMales
males
5 year
1 and under
23
+
15
4
2
3
1
0
1
1
3
2
2
2
0
27
15
7
0
2
2
0
1
1
1
0
2
1
0
Totals
59
59
2-5
6-10
11-15
16-29
21-25
26-30
31-35
36-40
4 1-50
51-60
61-70
71-80
80
Totals
1 year
4.60
5.40
3.00
3.00
0.80
1.40
0.00
0.40
0.40
0.60
0.20
0.40
0.00
0.00
0.20
0.20
0.20
0.20
0.60
0.20
0.40
0.00
0.40
0.40
0.40
0.20
0.00
0.00
Population at
risk as of
Summer 1963
10.00
6.00
2.20
0.40
1.00
0.60
0.00
0.40
0.40
0.80
0.40
0.80
0.60
0.00
Estimated
death rate
per 1,000
169.9
29.0
10.4
2.9
7.3
6.3
0.0
5.6
6.3
6.1
12.9
5.3
5.0
0.0
59
207
211
138
137
96
92
72
64
131
31
15
12
7
1,272
TABLE 3
Fertility in Aguacatenango f r o m interview data
Frequency distribution
of no. of husbands
(average no. of husbands: 1.64)
Number of husbands
a woman has had
0
1
2
3
4
5
Number of women
over 50 years
having had
that number
of husbands
0
30
13
7
1
1
Frequency distribution of
no. of children
(average no. of children: 6.94)
Number of
children
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
52
is the recipient of their contribution is
characterized by an element of circularity.
In contrast, Barrio 2 receives a greater
genetic contribution from non-Aguacatenangueros. It is clear that people are mov-
Number of women
over 50 years
having had
that number
of children
1
3
1
5
4
2
3
14
6
3
3
4
1
0
0
0
0
0
1
0
1
52
ing out of the barrios as the size of the
population has remained relatively stable
for the last seven years (Metzger, '56 for
1956 census; Mexican government for
1960 census; present study for 1963 cen-
113
INTERPRETATIONS O F POPULATION ISOLATES
TABLE 4
Genetic contributions to barrio I and 11 from outliers, other barrio, and non-Aguacatenangueros
as tabulated from interview data
Number of individuals now living in
Birthplace of
Barrio I (pop. 430)
Barrio11 (pop.531)
14
7
54 (41 ’)
9
4
0
6
Outlier I1
Outlier I11
Outlier IV
Outlier V
5
x2
=72.2
< 0.01 (0.001at least)
Total from Outliers
84
15
p
Barrio I
Barrio I1
-
21
xa=6.86
-
p
13
28
p
64
12.1%
xe=75.2
131
30.5%
Total
Total %
1 Indicates
34
< 0.01
x2~2.83
Non-Aguacatenangueros
and offspring
p
< 0.05
< 0.01 (0.001 at least)
the contribution of the Giron family which is discussed in the text below.
sus) though the birth rate is 67 per 1,000
and the adjusted rate is 21 per 1,000. A
comparison of the population pyramid to
the age-related death rate suggest that the
narrowing of the population pyramid from
ages 20-50 is largely due to emigration.
A particularly well-documented case of
the immigration, multiplication and spread
of a kindred of outsiders in Aguacatenango
is that of the Girons who came from San
Rafael and account for a large part of the
outside contributions to Barrio 1 (table 4).
The distribution by generation and location of the 72 members of the Giron family
who are in Aguacatenango is presented in
TABLE 5
Distribution of girons i n Aguacatenango by
generation and location (total 7 2 )
Generation
G-1 (oldest)
6-2
Barrio
1
IV
v
2
0
1
0
2
0
0
2
2
4
4
2
6
0
0
0
7
12
0
8
1
Total
15
22
20
1
Males
Females
11
4
Total
15
0
0
0
0
1
1
Males
Females
0
1
Total
Males
Females
Total
6-3
6-4(youngest)
-::? -:;?
Males
Females
table 5. Not all Girons in Aguacatenango
are descendants of the three siblings of the
oldest generation (G-1) who live in Aguacatenango. Some Girons living in Aguacatenango are descended from the daughter
of another sibling of that generation who
has never lived in Aguacatenango.
Blood group typing
It was not feasible to double-test the
blood samples, and controls detected certain laboratory errors. Control typings for
anti4 (direct agglutination) revealed false
negatives. Four out of the 460 CDEce typings showed “impossible” phenotypes although there had been no trouble in typing
controls. The inaccuracy in the N and S
antigen typings affects MNS comparisons,
between the data of this study and that of
other studies, but does not affect the intragroup comparisons that have been made
because the members of families and of the
various subgroups that have been defined
came to donate blood in random order.
Therefore, errors in typing should be random with each group.
Considering the degree of mating across
barrios, it seems likely that the population
of Aguacatenango would be genetically
homogeneous, However, the chi-square
analysis of blood group allele frequencies
between the two barrios in the preliminary
data suggested that they should be considered as partial genetic isolates (Erickson
et al., ’ 6 3 ) .
114
ERICKSON,
NERLOVE, CREGER AND ROMNEY
The present blood group study does not
substantiate at statistically significant
levels that the two barrios are partial genetic isolates. Nonetheless, there is an interesting trend of genetic difference when
one compares the barrios and the various
sub-samples, i.e., corrected (described below), over 35 years, and under 15 years,
within them for Rh and MNSs frequencies.
Table 6 shows the composite phenotype
frequencies of blood groups for Barrios
1 and 2, while table 7 presents their derived genotype frequencies. Chi-squares for
differences between the barrios and between the various sub-samples examined
within the barrios are presented in table 8.
The biggest differences between barrios are
seen in the sub-sample of individuals 35
years and older from each barrio. When
only individuals of 15 years or younger are
considered, the barrios seem almost identical in both blood group systems. To remove
the bias in the data caused by disproportionate representation of various families,
a corrected sub-sample for each barrio
was constructed. The two sub-samples
could not contain more than two individuals from each patrilineal lineage, nor,
where possible, could they contain more
than two members of the same nuclear
family. Using these two sub-samples to
compare barrios for Rh and MNSs frequencies, it is shown in tables 6 and 8 that
the phenotypes are not affected by correcting for family representation.
Calculations were made to see if phenotype frequencies fit Hardy-Weinberg equilibria. The results of these calculations are
presented in table 9. It can be seen that
each barrio approaches Hardy-Weinberg
equilibrium for Rh but the total population does not. Through Barrio 2 approaches equilibrium, neither Barrio 1 nor
the total population of Aguacatenango is
in equilibrium as determined by the calculations for the MNSs data. The MNSs data
are suspect because of trouble with the
anti-N typings. Nonetheless, the errors in
MNSs typing should equally affect the calculations for each barrio. The MNSs data,
as well as the Rh data, however, fit the
Hardy-Weinberg equilibria better for each
barrio separately.
Kell Cellano, Duf€y, Diego and Gm and
Inv allotype frequencies are presented in
table 10.
The Duffy and Diego phenotype frequencies can be utilized along with the
total Aguacatenango Rh frequencies to
compare this partial isolate to the one other
partial isolate that has been sampled in
the Tzotzil-Tzeltal area: Chalchihuitan, a
TABLE 6
Composite o f blood group phenotype frequencies f m barrios 1 and 2 i n Aguacatenango
Phenotype
0
A
ccddee
ccDee
ccDEE
ccDEe
CCDee
CcDee
CCDEe
CcDEE
CcDEe
NNs
NNS
NMS
MMS
MNs
MNS
Total
Barrio
1
Barrio
2
243
4
0
1
34
21
69
20
150
1
1
1
14
14
49
17
3
3
12
84
23
10
100
42
39
32
247
6
452
20
6
44
24
33
24
151
Barrio
Barrio
Corrected 1 Corrected 1
35 years
Barrio 1
35 years
Barrio 2
15 years
Barrio 3
15 years
Barrio 2
57
2
0
0
7
117
1
16
3
37
0
0
0
2
4
16
6
78
0
0
1
11
7
22
0
0
1
8
5
2
6
37
a4
2
0
0
11
10
22
10
1
4
27
11
2
23
20
15
15
56
0
0
0
6
5
21
4
3
4
13
7
1
19
9
12
a
3
22
3
2
23
14
7
10
86
56
59
a
9
7
a
0
0
16
7
35
8
3
7
39 2
15
3
50
13
21
16
118
a
3
3
222
12
2
27
6
22
9
78
Corrected for family sampling by taking no more than two individuals from each identified patrilineal lineage and, when possible, not from the same nuclear family within that lineage.
2 Three Rh mistypes in Barrio 1 and 1 in Barrio 2 deleted.
3 MNSs mistype i n Barrio 1 deleted.
1
INTERPRETATIONS O F POPULATION ISOLATES
115
Tzotzil town studied by Matson and Swanson ('63). Table 11 presents these data.
Wide differences are observed.
DISCUSSION
The demographic data show a broadbased population pyramid which narrows
quickly because of high infancy and child
mortality. Most of the remaining population survives through middle age. The
mortality figures do not account for all of
the further narrowing of the pyramid. Part
of this narrowing is the result of emigration. The population data show an excess
of females, which probably relates both to
the "social" deaths of males (e.g., murder
for witchcraft) and to the usual rate of
differential viability. Fertility is noted to
be variable with a mode of seven children,
with a range of none to 20. The age of 15
years is used as the initial age of reproduction. The data in figure 3 and tables 2 and
3 can then be used to calculate indices of
selection intensity following Neel and
Chagnon ('68). I,, the portion of the selection intensity due to mortality (of females)
prior to the age of reproduction equals
0.767; while If, the portion due to fertility
differences among women reaching the age
of reproduction, equals 2.682. These indices are considerably higher than those of
the "primitive" Xavante (Neel et al., '64).
We have been unable to calculate the
breeding size of the population accurately,
although only a few mature individuals
have had no offspring. Although the genealogies show that there are a fair number of third-cousin marriages, we have
also been unable to calculate a coefficient
of inbreeding. If one could state the coefficient of inbreeding as well as the differential fertility accurately, the desirable next
step would be to analyze their respective
contributions to the evolution of the present gene frequencies in the population.
There are numerous barriers to such an
analysis.
First, it is necessary to know what the
selection coefficients are in this population
for certain blood group phenotypes or
genes closely linked to them. In order to
know these selection coefficients, knowledge of the phenotypes of both parents and
children in a number of families is needed.
Such knowledge would enable the detec-
116
ERICKSON, "ERLOVE, CREGER AND ROMNEY
TABLE 8
Chi-square values and probabilities
Between barrios
Comparison
Rh'
Chi-square
Barrio I to Barrio I1
Corrected Barrio I to
Probability
0.5
X s Z =5.23
0.20
~ 5 %
0.4
xs3= 1.96
0.75
x f =7.98
0.08
x 2 =6.94
0.15
xs2=1.75
0.89
x4z =3.94
0.40
=5.23
corrected Barrio I1
1
Chi-square
xsZ~ 4 . 9 6
Older than 35, Barrio
I to Barrio 11
Younger than 15, Barrio
I to Barrio I1
MNSs
Probability
Rare Rh phenotype pooled as in table 9.
TABLE 9
Closeness o f fit t o Hardy-Weinberg equilibria i n Aguacatenango barrios 1 and 2 considered
both separately and together
Aguacatenango
Observed
Expected
Barrio 1
Observed
Expected
0
3
6
45
49
17
14
14
0.168
5.438
3.088 xS=7.053
51.151 Dr=5
44.031 p
0.05
18.485
14.200
10.497
1.940
24
84
69
20
34
21
1
0.273
8.158
6.149 x2=12.946
93.188 Dr=5
60.831 p == 0.025
22.138
34.561
16.687
2.014
64.118
59.918
124.990 x2=31.278
109.865 D r = 2
13.966 p
0.001
24.143
42
32
100
39
10
23
40.302
25.922
87.772 x2=24.105
62.929 D,=2
7.795 p
0.001
11.279
66
56
MNSs
Observed
0
0.441
6
13.593
18
9.237 x2= 19.718
129 144.898Dr=5
118 104.684 p
0.01
40.430
37
48.340
49
27.474
35
3.904
2
0
3
12
<
Rh
Barrio 2
Expected
'g
16
43
<
1
24
44
33
6
20
-<
>
23.926
24.132
37.745 x2= 7.350
45.527 Dr=2
5.942 p
0.025
13.728
<
TABLE 10
Additional blood group and serum allotype data f o r Aguacatenango
System
Phenotype
KellCellano
K(+ )k( - )
+ +
K( )k(
)
K(-)k(-)
no.
%
0
0
0
100
0
132
Genotype
R
12
Allele
Frequency
+
0
1.0
*
Duffy
Fy(a+ 1
Fy(a- 1
81
99
44.96
55.04
FY"
not F p
0.258 0.051
0.7420.051
Diego
Di(a+)
Di(a-)
9
124
6.8
93.2
Dia
not Dia
0.0352 0.023
0.965 f 0.023
Gm Allotypes
Gm 1 , 2 , 1 3
Gm 1,13
Gm 1 , 2
Gm 1
2
3
74
77
1.28
1.92
47.44
49.36
Gml
Gm1.2
Gm'Ja
0.7000
0.2839
0.0161
Inv Allo-
Inv(l+a+)
types
Inv( 1+ a - )
85
3
49
62.04
2.19
35.77
Inv(1-a-)
117
INTERPRETATIONS O F POPULATION ISOLATES
TABLE 11
Comparison of blood group phenotypes between two semi-isolates: Chachihuitan and Aguacatenango
Expected
Observed
Phenotype
CCDee
CcDEE
CcDEe
CcDEe
ccDEE
ccDEe
ccDee, ccdee, CCDEe
2:-
Chalchinango
hitan
Chalchi-
tz:z
hitan
nango
15
0
36
3
22
2
2
118
18
129
37
48
35
9
22.45
3.04
27.35
6.75
11.81
6.24
1.86
110.55
14.96
137.15
33.25
58.19
30.76
90.14
80
394
79.50
475.00
76
4
81
99
52.35
27.65
80
180
80.00
2.472
3.040
2.335
2.083
8.782
2.881
0.011
0.502
0.618
0.434
0.423
1.784
0.584
0.001
~e’=26.060 p
117.70
62.30
10.62
24.73
xie=67.02
15
65
9
124
9.02
10.98
14.96
118.04
80
133
20.00
133.m
3.95
0.50
< 0.01
10.15
21.52
p
< 0.001
2.37
0.30
tion of statistically significant shifts from behavior and differential fertility on allele
the theoretically expected (based on the frequency changes, our primary involvehypothesis of no selection) segregation ment became that of defining the variables
ratios (Morton, ’59). Selection coefficients under study. Verbal statements about
might be taken from Morton’s data on a breeding behavior recorded in the genealNortheastern Brazilian population which ogies incorporate significant error and culshowed selection only in the ABO system turally defined social entities may be
(Morton et al., ’66). It is not a safe as- widely discrepant with the actual populasumption, however, that the same selec- tion entities. Such a discrepancy has been
tion forces are at work in the two dissim- noted previously with respect to marriage
ilar environments. It is only after one patterns (Kunstadter et al., ’ 6 3 ) .
knows what portion of allele frequency difThe problem of “illegitimacies” is a genferences might be due to selection that the eral one, which serves particularly to comeffects of total population size, breeding
size and coefficient of inbreeding on ge- plicate studies of differential fertility. For
netic drift can be considered. The contribu- example, the Girons, who have made such
tion of the population’s mating system to a disproportionately large apparent contrigenetic drift is similarly unamenable to bution to the population, may have made a
still larger contribution by way of “illegiticalculation.
Secondly, the mathematics for analysis macies.” For 17 families there are blood
of population genetics are predicted on samples for both parents and a total of 37
sampling a fairly large population, not on of their offspring. Children with blood typ
complete enumeration of a small popula- ings which are multiply incompatible with
tion. The difficulty of trying to analyze a those of their parents included one whose
small population has been noted (Sutter, blood is incompatible with that of the
’63; Arends et al., ’67) but the solution mother. It is usually assumed that the
mother indicated is a biological parent, but
does not appear to be forthcoming.
Although the goal of this study was to this rule can be in error if step-parentage
provide data for comparative studies of or adoption are not indicated by informthe effects of parameters such as mating ants (Buettner-Janusch et al., ’64).
118
ERICKSON, NERLOVE, CREGER AND ROMNEY
If an Aguacatenaguero is asked if there
are any “outsiders” who live in town, he
would indicate ladinos living in the town
and the inhabitants of Outliers I and VI.
He considers the town to be endogamous.
He also considers each barrio to be endogamous but close questioning would elicit
the fact that he knows of exceptions to the
endogamy rule. It is only when a complete
census is taken that the disparity between
real and ideal social entities is discovered.
It is important to consider, however, how
much of that disparity has to do with the
exogamous marriages necessitated by imbalances in the spouse pool. Even geographically, the apparent isolate can be
shown to consist of micro-units. One-third
of Aguacatenango’s population lives outside of the barrios in outliers. The barrios
show a high degree of intermarriage and
outsiders are not only present, but their
offspring are a considerable proportion of
the population.
The suggestion, then, of genetic isolation between the barrios found in the preliminary study is not only interesting but
also somewhat puzzling. In the present
study, the differences do not reach signscance at usually accepted levels of probability for chi-square tests, but there is a
clear trend of increasing differences with
increasing age of the individuals. Such a
trend suggests that perhaps there had once
been greater isolation between the barrios
and that it is now breaking down. Such
age-related genotype differences are not
found in populations which are not isolated. For instance, studies of several
serum protein factors in West Greenland
Eskimos show no regional variations, even
for individuals over 55 years of age (Persson, ‘68). The evidence of better fits to the
Hardy-Weinberg criterion for genetic equilibria in each barrio separately is also suggestive of barrio isolation. If the possibility
is considered that each barrio might have
a separate isolate, then it is surprising to
find for each barrio a fit to Hardy-Weinberg
equilibria which is premised on a large
population for panmixis (Sutter, ’63). Yet
good fits to Hardy-Weinberg equilibria in
much smaller and more highly inbred populations are being found (Nee1 et al., ’64;
Arends et al., ’67). It will not be possible
to understand the genetic structure of
small populations until further genetic
studies of small populations are made.
There are three hypotheses that would
explain gene frequency differences between the two barrios: 1. differential immigration, 2. separate origin or 3. differential selection. In distinguishing between
these hypotheses, it is important to decide
whether the gene exchange between barrios observed in 1963 is a recent innovation. If the inter-barrio exchange is of long
standing, large initial allele frequency differences would be expected or the mechanisms maintaining allele frequency differences need to be very strong. It seems
highly unlikely that the population living
in each barrio is exposed to a selective environment distinct enough to maintain
allele frequency differences. The forces of
modern communication and transportation, however, which can serve to contribute to a breakdown of old customs of endogamy conceivably could have greatly increased the degree of genetic contribution
to Aguacatenango from the outside. Along
with a breakdown in town endogamy
could be a breakdown in barrio endogamy
and thus account for a recent onset of
higher gene exchanges between barrios.
If one barrio has always received more
outsiders or each barrio has received outsiders from a different area, differential
genetic contribution to the two barrios
could explain blood group allele frequency
differences. A t present, Barrio 1 has a significantly higher rate of barrio exogamy
(mostly involving outliers) while Barrio 2
has a greater number (not significant) of
outsiders. Further blood group studies of
“outsiders” and individuals who have
changed barrios would allow us to determine whether their movements are important for the possible genotype differences that we find between the two barrios.
More data on phenotypes of complete families such as the Girons are needed for
better assessment of the effects of differential fertility on modal genotypes in this
village. In Japan, Yanase (’65) has found
decreased fertility for immigrants rather
than the increased fertility we find in this
one case. Perhaps a combination of differential immigration and differential fertility (with or without selection) would determine barrio allele frequency differences.
INTERPRETATIONS OF POPULATION ISOLATES
It has been mentioned that the origin
of the barrios is unknown. If it represents
an old clan distinction, it is probably that
the clans had similar allele frequencies (by
reason of common ancestry, exposure to
nearby identical selective environments
and a continued small degree of inter-clan
gene exchange. Note that this argument
represents the sort of hypothesis that our
methods could help substantiate.) It is
also possible that the barrios might have
originated when Quesaltepec was added to
Aguacatenango during the directed population movements in the seventeenth century. If so, two barrios, with rules of barrio
endogamy, might have been set up to maintain isolation. It is interesting to note that
paricialidades had some political autonomy
(Calnek, ’61) which provides more reason
to believe that the combined populations
would not immediately merge.
The comparison of modal genotype frequencies between Aguacatenango and
Chalchihuitan, a northern Tzotzil semiisolate, show significant differences of
varying degree for different blood group
systems. The differences found suggest
that outsiders, but from within the
Tzotzil-Tzeltal area, would contribute significant blood group heterogeneity to the
population. Blood group data from many
other semi-isolates in this region are
needed before a meaningful discussion of
gene clines and their relaticnship to local
historical events and geographical location
is possble. The other blood group genotype
and allotype frequencies presented in table
9 may be helpful in this regard. Blood
group data (Matson and Swanson, ’59, ’61,
’63) and allotype data (Steinberg et al.,
’61) on samples of Tzotziles and Tzeltales
are available as a preliminary indication
of the genotype variation expected.
ACKNOWLEDGMENTS
The work was supported by a grant from
the Stanford University School of Medicine, Dean’s Office Fund, by the Pamela
Weigl Memorial Fund for Research in Diseases of the Blood, Stanford University, by
a U.S.P.H.S. grant, No. CH 00050-04, to
Dr. A. Kimball Romney, and by two Russell Sage Foundation grants to Robert P.
Erickson. Antisera for Kell, Cellano, and
Duffy antigens were kindly donated by
119
Ortho-Pharmaceuticals, Raritan, New Jersey. The antisera for determination of
Diego antigen was due to the courtesy of
Dr. Miguel Layrisse, Caracas, Venezuela.
The allotype determinations were kindly
performed by C. Ropartz. We are indebted
to Robert Armstrong, John Hotchkiss, and
Vicente Gomez for help with the field work.
Jo Anne Dalziel, Joan Harris, and David
Clark are to be thanked for excellent laboratory work. Paul Comer and Peter Workman have aided us with calculations.
Duane Metzger and Howard Cann have
provided valuable discussion.
LITERATURE CITED
Adams, R. M., and N. A. McQuown 1959 Prehistory, and post-conquest developments. Report of the University of Chicago, “Man-inNature” Project in Chiapas, Mexico, mimeographed.
Arends, T., G. Brewer, N. Chagnon, M. L. Gallango, €I. Gershowitz, M. Layrisse, J. Neel, D.
Shreffler, R. Tashian and L. Weitkamp 1967
Intratribal genetic differentiation among the
Yanomama Indians of Southern Venezuela.
Proc. Natl. Acad. Sci. U. S.A., 57: 1252.
Buettner-Hanusch, J., J. R. Bove and N. Young
1964 Genetic traits and problems of biological
parenthood in two Peruvian Indian tribes. Am.
J. Phys. Anthrop., 22: 149-154.
Calnek, E. E. 1959 Ethnohistorical notes. Report of the University of Chicago “Man-inNature” Project in Chiapas, Mexico, mimeographed.
1961 Distribution and location of the
Tzotzil and Tzeltal pueblos of the highlands of
Chiapas from earliest times to the present.
University of Chicago, Anthropology Department, mimeographed.
Cavalli-Zforza, L. L., I. Barrai and A. W. F. Edwards 1964 Analysis of Human Evolution
Under Random Genetic Drift. Cold Spring Harbor Symp. Quant. Biol., XXM: 9-20.
Erickson, R. P., D. Clark, W. P. Creger and A. K.
Romney 1963 The comparison of genetic
and anthropological interpretations of breeding
patterns and breeding isolates. Abstracts, 62nd
Annual Meeting of the American Anthropological Association: 16.
Gajdusek, D. C. 1964 Factors governing the
genetics of primitive human populations. Cold
Spring Harbor Symp. Quant. Biol., XXIX: 121136.
Kunstadter, P., R. Buhler, F. F. Stephan and C.
F. Westoff
1963 Demographic variability
and preferential marriage patterns. Am. J.
Phys. Anthrop., 21: 511-520.
Lisker, R., G.Zarate and A. Loria 1966 Studies
on several genetic hematological traits of Mexicans IX. Abnormal hemoglobins and erythrocytic Glucose-6-phosphate dehydrogenase deficiency in several Indian tribes. Blood, 27: 824.
120
ERICKSON, NERLOVE, CREGER AND ROMNEY
Lisker, R., G. Zarate and E. Rodriguez 1967
Studies on several genetic hematological traits
of the Mexican population XIV. Serum polymorphisms in several Indian tribes. Am. J.
Phys. Anthrop., 27: 27-32.
McKusick, V. A., J. A. Hostetter, J. 0. Egeland
and R. Eldridge 1964 The distribution of
certain genes in the old order Amish. Cold
Spring Harbor Symp. Quant. Biol., XXIX: 99114.
Matson, A., and J. Swanson 1959 Distribution
of hereditary blood antigens among the Maya
and non-Maya Indians in Mexico and Guatemala. Am. J. Phys. Anthrop., 17: 49-74.
1961 Distribution of hereditary blood
antigens among American Indians in middle
America: Lacondon and other Maya. Amer.
Anthrop., 63: 1292.
1963 Distribution of hereditary blood
antigens among Indians in Middle America 11:
Tzotzil and other Maya. Am. J. Phys. Anthrop.,
21: 1-14.
Melartin, L., B. S. Blemberg and R. Lisker 1967
Albumin Mexico, a new variant of serum albumin. Nature, 215: 1288.
Metzger, D. 1956 The social organization of
Aguacatenango, Chiapas: a preliminary statement. Unpublished.
Morton. N. E. 1962 Segregation and linkage.
In: Methodology and Human Genetics. W. J.
Burdett, ed. Holden-Day, San Francisco, 17.
Morton, N. E., H. Krieger and M. P. Mi 1966
Natural selection on polymorphisms in Northeastern Brazil. Amer. J. Hum. Genet., IS: 153.
Neel, J. V., and N. A. Chagnon 1968 The Demography of Two Tribes of Primitive Relatively
Unacculurated American Indians. Proc. Natl.
Acad. Sci., 59: 680.
Neel, J V., and F. M. Salzano 1964 A prospectus for genetic studies of the American Indian.
Cold Spring Harbor Symp. Quant. Biol., XXIX:
85-98.
Neel, J. V., E. M. Salzano, P. C. Junqueira- F.
Keiter and D. Maybury-Lewis 1964 Studies
on the Xavante Indians of the Brazilian Mato
Grosso. Am. J. Hum. Genet., 16: 52-140.
Persson, I. 1968 The distribution of serum
types i n West Greenland Eskimos. Acta Genet.,
Basel, 18: 261.
Roberts, D. F. 1968 Genetic effects of population size reduction. Nature, 220: 1084.
Steinberg, A. G., M. S. Cordova and L. Ruben
1967 Studies o n several genetic hematologic
traits of Mexicans XV. The Gm allotypes of
some Indian tribes. Amer. J. Hum. Genet., 19:
747-756.
Sutter, J. 1963 The relationship between hum a n population genetics and demography. In:
The Genetics of Migrant and Isolate Populations, E. Goldschmidt, ed. Williams and Wilkins, New York, 160.
Swadesh, M. 1959 Linguistics as an instrument of prehistory. Southwestern J. of Anthrop., 15: 20-35.
Yanase, T. 1965 Degree of endogamy and
effective migration rate in isolated populations.
Proc. Japan Acad., 41: 169.
Yasuda, N., and N. E. Morton 1967 Studies on
human population structure. Proc. I11 Intern.
Congr. Human Genet., Johns Hopkins Press,
Baltimore, 249.
Документ
Категория
Без категории
Просмотров
1
Размер файла
1 185 Кб
Теги
anthropological, aguacatenango, population, chiapas, isolated, mexico, interpretation, genetics, comparison
1/--страниц
Пожаловаться на содержимое документа