201
q 2000 The Society for the Study of Evolution. All rights reserved.
Evolution, 54(1), 2000, pp. 201?209
LATITUDINAL VARIATION FOR TWO ENZYME LOCI AND AN INVERSION
POLYMORPHISM IN DROSOPHILA MELANOGASTER FROM
CENTRAL AND SOUTH AMERICA
JAN VAN ?T LAND,1,2 WILLEM F. VAN PUTTEN,1,3 HAROLDO VILLARROEL,4 ALBERT KAMPING,1 AND
WILKE VAN DELDEN1,5
1Population Genetics, University of Groningen, P.O. Box 14, 9750 AA, Haren, The Netherlands
4Universidad de Playa Ancha, Valpara??so, Chile
Abstract. Many organisms show latitudinal variation for various genetically determined traits. Such clines may involve
neutral variation and originate from historical events or their maintenance may be explained by selection. For Drosophila
melanogaster, latitudinal variation for allozymes, inversions, and quantitative traits has been found on several con-
tinents. We sampled D. melanogaster populations in Panama and along a transect of 40 latitudinal degrees on the west
coast of South America. Negative correlations with latitude were found for AdhS and aGpdhF allele frequencies and
for the frequency of the cosmopolitan inversion In(2L)t in AdhS aGpdhF chromosomes. A positive correlation existed
between wing length and latitude. Significant correlations were found between these traits and climatic variables like
temperature and rainfall. The observed clines show considerable resemblance to those found on other continents.
Gametic disequilibrium between AdhS and aGpdhF occurred predominantly at higher latitudes and was caused by the
presence of In(2L)t. The reasons for the clinal distributions are discussed and it is argued that selection is the most
likely explanation. However, the exact nature of the selective force and the interactions of allozymes with each other
and with In(2L)t are complex and not fully understood. In tropical regions In(2L)t-containing genotypes have higher
fitness than ST/ST and Adh and aGpdh hitchhike with the inversion, but there is also evidence for balancing selection
at the Adh locus.
Key words. Alcohol dehydrogenase, a glycerophosphate dehydrogenase, Drosophila melanogaster, gametic disequi-
librium, inversions, latitudinal clines, polymorphism.
Received January 8, 1999. Accepted August 9, 1999.
Many species show an ecogeographical distribution with
several traits varying with longitude or latitude. Such clines
can be attributed either to adaptation to different environ-
ments, historical events, or to random processes like genetic
drift, gene flow, or isolation by distance (Endler 1977). Ex-
tensive research has been carried out with respect to the ex-
istence of latitudinal clines for Drosophila melanogaster. The
cline for allele frequencies of the alcohol dehydrogenase lo-
cus (Adh) is perhaps one of the best-studied examples, and
this cline (with an increasing frequency of the AdhS allele
toward the equator) has been observed on several continents,
with an emphasis on North America and Australasia (Berger
1971; Johnson and Schaffer 1973; Vigue and Johnson 1973;
Voelker et al. 1977; Oakeshott et al. 1982; Capy et al. 1986;
Anderson et al. 1987; Berry and Kreitman 1993; Bubli and
Imasheva 1997). A similar clinal pattern has been found in
populations in Europe and Africa (Capy et al. 1986; David
et al. 1989; Be?nassi and Veuille 1995; Veuille et al. 1998).
The general view is that the worldwide cline for Adh is
the result of selection (Oakeshott et al. 1982; Anderson et
al. 1987; Berry and Kreitman 1993), with temperature or
temperature-related factors being the main selective agents
(Malpica and Vassallo 1980; Van Delden 1982; McKenzie
et al. 1994). The ADHS enzyme has a higher in vitro stability
at higher temperatures (Vigue and Johnson 1973). However,
flies homozygous for AdhS showed a higher in vivo mortality
2 Present address: Organization for the Advancement of Tropical
Research, NWO, Den Haag, The Netherlands.
3 Present address: Netherlands Institute for Ecological Research,
Heteren, The Netherlands.
5 Corresponding author. E-mail: w.van.delden@biol.rug.nl.
at high (i.e., 358C) temperatures (Van Delden and Kamping
1980). In addition, the AdhFF genotype is better adapted to
the presence of alcohol in the medium, which is presumed
to be available in larger quantities in fruits in temperate re-
gions (Van Delden et al. 1978; Van Delden 1982 and ref-
erences therein).
However, it is not yet clear whether selection acts directly
on the Adh gene or whether this cline simply reflects selection
at linked loci or chromosomal regions (see Berry and Kreit-
man 1993). Two important candidates for this type of selec-
tion are the a-glycerophosphate dehydrogenase (sGpdh) gene
and the cosmopolitan inversion In(2L)t. The aGpdh locus is
related functionally to the Adh locus through the NADH:
NAD1 ratio (Cavener 1983). In(2L)t includes the aGpdh lo-
cus, one of its breakpoints is located close to the Adh locus,
and is nearly always found in combination with AdhS and
aGpdhF (Voelker et al. 1978). Clinal patterns have been ob-
served for In(2L)t frequency, varying from 0% to 5% in tem-
perate populations to 70% toward the equator (Mettler et al.
1977; Inoue and Watanabe 1980; Yamaguchi et al. 1980;
Knibb et al. 1981; Knibb 1982, 1983; Inoue et al. 1984;
Anderson et al. 1987; Das and Singh 1991; Singh and Das
1992; Inoue and Igarashi 1994; Veuille et al. 1998). For
aGpdh, a clinal pattern was also observed: the frequency of
the F allele increases towards the equator (Johnson and Schaf-
fer 1973; Oakeshott et al. 1982, 1984; Knibb 1983; Gibson
et al. 1991; Parkash and Shamina 1994).
As for Adh, the latitudinal cline for aGpdh is thought to
be related to temperature. The aGpdhF allele is more resistant
to higher temperatures (Alahiotis et al. 1977; Voelker et al.
1978; Barnes et al. 1989; Van Delden and Kamping 1989)
and, at lower temperatures, aGpdhFF flies have a lower rate
202 J. VAN ?T LAND ET AL.
FIG. 1. Map of the 15 sampling sites.
of development than aGpdhSS flies (Barnes et al. 1989). In
addition, experimental evidence indicates that the aGpdhSS
genotype shows a larger flight output at low temperature
(Barnes and Laurie-Ahlberg 1986). However, there are a
number of studies that failed to demonstrate significant
aGpdh effects on metabolic rate, high temperature resistance,
or ethanol tolerance (for references, see Oudman et al. 1992).
Laboratory experiments showed that independently of the
Adh and aGpdh loci homozygotes for In(2L)t have a selective
advantage at high temperatures. Moreover, these homozy-
gotes are smaller and have a longer development, whereas
the heterozygotes for the inversion were found to have a
higher viability under high-temperature conditions (Stalker
1980; Knibb 1982; Van Delden and Kamping 1989, 1991;
Singh and Das 1992). In addition, seasonal variation in
In(2L)t frequency has been observed, with higher frequencies
during or at the end of the warmer season (Langley et al.
1977; Stalker 1980; Zacharopoulou and Pelecanos 1980;
Aguade? and Serra 1987; Kim and Sung 1988; Sanchez-Re-
fusta et al. 1990; Kamping and Van Delden 1999).
Many studies have found a direct correlation between wing
length (which is a measure for body size) and flight activity,
as well as inversion and allozyme polymorphism (Jones 1974;
Pieragostini et al. 1979; Stalker 1980; Pfriem 1983; Serra
and Oller 1984; Barnes and Laurie-Ahlberg 1986; Hasson et
al. 1992; Bitner-Mathe? et al. 1995). As with inversion and
allozyme frequencies, clear latitudinal clines have been re-
vealed for body size and, particularly, wing length. Smaller
wings are found in equatorial regions (Capy et al. 1993; Im-
asheva et al. 1994; James et al. 1995), but this cline is not
always monotonic (Long and Singh 1995). Previous papers
have reported conflicting conclusions about the degree of
linkage disequilibrium among Adh, aGpdh, and In(2L)t
(Voelker et al. 1978; Malpica and Vassallo 1980; Yamaguchi
et al. 1980; Knibb 1983; Inoue et al. 1984; Oakeshott et al.
1984; Alonso-Moraga and Mun?oz-Serrano 1986; Anderson
et al. 1987; Van Delden and Kamping 1989). These con-
trasting results may reflect differences among continents in
the degree of linkage disequilibrium or even differences in
research methodology, which demonstrates the complexity
of the association between Adh, aGpdh, and In(2L)t. For this
reason, and because Central and South America have only
scarcely been sampled with respect to the existence of a cline
for Adh, aGpdh, and In(2L)t, we sampled wild populations
of D. melanogaster in Panama and along the west coast of
South America. We have estimated the frequency of Adh and
aGpdh alleles, scored In(2L)t genotype, determined gametic
disequilibrium between Adh and aGpdh, and measured wing
length. Here, we report on the associations among these poly-
morphisms and discuss their adaptive significance and lati-
tudinal distribution.
MATERIALS AND METHODS
Populations
Between June and November 1991, flies were collected at
five locations in Panama. During February and March 1995,
flies were collected at one location in Ecuador and nine lo-
cations in Chile (Fig. 1). The 15 sampling sites cover ap-
proximately 40 latitudinal and 12 longitudinal degrees (Table
1). Flies were caught by sweeping a net over boxes containing
rotten fruit in fruit markets. If necessary, additional traps with
a mixture of fermenting bananas and yeast were used. Several
authors have found an association between altitude and Adh
allele frequencies and morphological characters in Drosoph-
ila (Grossman et al. 1970; Pipkin et al. 1976; Louis et al.
1982; Bitner-Mathe? et al. 1995). Accordingly, all our sam-
plings, except Linares (140 m) and Copiapo? (350 m), were
done at locations with an altitude below 100 m. No effect of
altitude on allele or inversion frequency was observed in this
study.
Allozyme Electrophoresis
Wild females collected in the sampling sites were placed
individually in a separate vial containing instant Drosophila
medium and allowed to produce eggs. Horizontal polyacryl-
amide gel electrophoresis was performed on the wild caught
flies as described by Van Delden and Kamping (1989) to
determine the joint genotype for Adh and aGpdh (cytological
locations 35B3 and 26A, respectively). The offspring were
used for further tests.
Wing Length Measurement
Left wings of wild males were embedded in a drop of
Euparal on a microscopic slide and measured under a light
microscope (10 3 8) with an ocular micrometer. The absolute
length of the anterior crossvein to the wingtip was taken as
a measure for wing length (Prout 1958). For Panama, male
wing length data were derived from female wing lengths by
using a transformation factor of 0.875 (conforming to data
203LATITUDINAL VARIATION IN D. MELANOGASTER
TABLE 1. Location and climatic data of the sampled populations. For further information on climatic data see the Materials and Methods
section.
Population
Latitude
(S/N)
Longitude
(W)
Tyear
(8C)
Ryear
(mm)
Syear
(h)
Hyear
(%)
Guayaquil, Ecuador (GU)
Las Tablas, Panama (LT )
Chiriqui Grande, Panama (CG)
Panama City, Panama (ST )
Barro Colorado Island, Panama (BC)
Bocas del Toro, Panama (BT )
Arica, Chile (AR)
Iquique, Chile (IQ)
Antofagasta, Chile (AN)
Copiapo?, Chile (CO)
Coquimbo, Chile (CQ)
Valpara??so, Chile (VA)
Linares, Chile (LI)
Valdivia, Chile (VD)
Puerto Montt, Chile (PM)
28139S
78459N
88569N
98009N
98089N
98199N
188289S
208139S
238389S
278209S
298569S
338059S
358489S
398489S
418309S
798549
808159
828079
798309
798509
828159
708199
708109
708249
708219
718249
718409
718369
738149
728509
25.3
26.5
26.5
25.6
25.5
26.5
18.7
17.9
16.4
15.2
13.6
14.0
12.7
11.0
10.1
950
1200
3750
2328
2250
3300
0.5
0.6
1.7
12
79
373
967
1871
1803
1614
n.a.
n.a.
2337
n.a.
n.a.
2258
2816
2966
n.a.
2185
2120
2463
1876
1593
77
84*
90*
86
87*
82*
74
74
77
74
83
82
76
83
85
n.a., data not available.
* Data from 1991 only (see text).
of Robertson and Reeve 1952; Prout 1958; Zwaan et al. 1992;
David et al. 1994).
Detection of In(2L)t
The frequency of In(2L)t (cytological limits: 22D2?E1,
34A8?9) was determined by performing single-pair matings
between wildtype males and virgin females from a laboratory
stock fixed for the second chromosome markers dumpy (dp:
II, 13.0) and black (b: II, 48.5). Tests were done in the third
generation after collection. Ninety pair-matings were done
for each Panamanian sample site and 50 for the Chilean and
Ecuadorian sites. Eight F1 females from each cross were in-
dividually backcrossed to dp b males. The presence of flies
with only the dp or only the b phenotype in the F2 generation
indicated the occurrence of recombination of the second chro-
mosome and, consequently, absence of In(2L)t (Van Delden
and Kamping 1989). The average number of F2 individuals
scored per cross was 50. This procedure enabled us to de-
termine the frequency of heterozygous and homozygous flies
for In(2L)t.
Climatic Data
In Chile, Ecuador, and Panama City, climatic data were
recorded by weather stations adjacent to the collection sites
and were kindly provided to us by Direccion Meteorologica
de Chile, Santiago de Chile, Chile, and the Koninklijk Ned-
erlands Meteorologisch Instituut, De Bilt, the Netherlands.
Climatic data from the Panama locations were obtained from
the Smithsonian Tropical Research Station. Climatic data in-
cluded Tyear (average annual temperature), Tmin (average
monthly minimum temperature), Tmax (average monthly max-
imum temperature), Ryear (average of total yearly rainfall),
Syear (average of total yearly sun-hours), and Hyear (average
of annual relative humidity). All figures used are long-term
(30-year) averages, except when indicated differently (Table
1). Because of the strong correlation of Tyear with both Tmax
and Tmin (r 5 0.99 and 0.96, respectively), only Tyear was
used in the analyses.
Estimation of Gametic Disequilibrium
Joint gametic frequencies for Adh and aGpdh could not be
determined directly from the observed dilocus genotypes be-
cause it was not possible to distinguish between the coupling
and repulsion heterozygotes. Therefore, the maximum-like-
lihood method for codominant loci was used to estimate ga-
metic frequencies from the observed genotypes (Hill 1974).
Assuming that D is normally distributed, the variance of D
when D 5 0 can be approximated by V(D) 5 p1p2q1q2/n,
where p1 and p2 are the allele frequencies at the first locus
and q1 and q2 are the allele frequencies at the second locus,
and n is the sample size. The hypothesis of D 5 0 can be
rejected at the 0.05 significance level if zDz . 1.96(V[D])
(Hedrick 1983).
Changes in allele frequencies affect maximum potential
gametic disequilibrium (Dmax). Therefore, absolute values of
gametic disequilibrium may give rise to misleading conclu-
sions; for this reason, a relative measure (D/Dmax) is pref-
erable (Hedrick 1983, 1987). The latter value ranges from
zero to one, and is independent of allele frequencies (Hedrick
1987). However, under circumstances where both D and Dmax
are small, the D/Dmax ratio may take unrealistically high value
and samples either with only two or three of four possible
gamete types will give a D/Dmax value of one.
Statistical Analysis
To normalize the distribution, wing length data were nat-
ural-log transformed. Frequencies of AdhS, aGpdhF, In(2L)t,
population heterozygosities, and annual relative humidity
data were angularly transformed. Total yearly rainfall was
square-root transformed. Dilocus heterozygosities were ob-
tained by averaging the heterozygosities for Adh and aGpdh
for each population.
RESULTS
Latitudinal Variation in Adh and aGpdh Allele Frequencies
AdhS frequencies varied from 0.99 to 0.07 and aGpdhF
frequencies between 0.97 and 0.43 (Table 2; Fig. 2). Both
204 J. VAN ?T LAND ET AL.
TABLE 2. Number of collected flies (n), Adh and aGpdh allele frequencies in the collected samples, and In(2L)t frequencies in the third
generation. Significant deviations from Hardy-Weinberg equilibrium and significant gametic disequilibria are indicated by asterisks. See text
for the calculation of D and Dmax (gametic disequilibrium) and H (heterozygosity) and Table 1 for population abbreviations.
Population n
Allele frequency
AdhS aGpdhF
Frequency
In(2L)t D D/Dmax H
GU
LT
CG
ST
BC
BT
AR
IQ
AN
CO
CQ
VA
LI
VD
PM
130
81
31
122
345
17
49
28
72
103
22
334
194
192
43
0.99
0.91
0.81
0.85
0.96
0.85
0.25*
0.34
0.26
0.17
0.16
0.14***
0.07
0.11
0.13
0.62
0.90
0.92
0.86
0.93
0.97
0.66
0.63
0.66
0.47
0.45
0.69
0.43
0.47
0.64
0.18
0.18
0.30
0.13*
0.18*
0.44
0.46
0.40
0.43*
0.17
0.27
0.18
0.10*
0.04
0.15
0.001
20.007
20.025
20.005
0.002
0.004
20.051
20.074
20.075*
20.068*
20.087
20.041*
20.029*
20.040*
20.046
0.12
0.08
0.38
0.04
0.73
1.00
0.62
0.58
0.86
0.76
1.00
0.94
0.70
0.68
1.00
0.26
0.17
0.18
0.25
0.11
0.18
0.30
0.39
0.42
0.41
0.39
0.31
0.32
0.33
0.37
* P , 0.05; ** P , 0.01; *** P , 0.001.
FIG. 2. Relation between AdhS (C) and sGpdhF (1) frequencies
and latitude.
AdhS and aGpdhF frequencies were negatively correlated with
latitude when all 15 samples from Panama and South America
were considered (Table 3). A negative correlation between
latitude and AdhS frequency was also found for the ten sam-
ples from South America (r 5 20.89, P , 0.001) and for
the nine samples from Chile (r 5 20.86; P , 0.01). No
significant correlation between latitude and aGpdhF frequen-
cy was present in the latter two cases. No correlations be-
tween latitude and AdhS and aGpdhF were found for the five
samples from Panama. Given that latitude correlates with
Tyear (r 5 20.97, P , 0.001), its correlation with allele
frequencies (Table 3) can be attributed to the well-known
correlation of allele frequencies with temperature. Using the
equation given by Anderson et al. (1987), we calculated the
frequencies for AdhS and aGpdhF in Standard (i.e., no In[2L]t
carrying) chromosomes. Table 3 shows that, although most
r-values decrease marginally, all significant correlations be-
tween allele frequencies and latitude or climatic variables
remain significant. In other words, the latitudinal clines of
Adh and aGpdh are largely independent of the presence of
In(2L)t.
Latitudinal Variation in In(2L)t Frequencies and
Wing Length
Estimated In(2L)t frequencies ranged from 0.46 to 0.04
(Table 2). Four populations showed a smaller number of het-
erozygotes than would be expected if the populations were
in Hardy-Weinberg equilibrium.
The average percentage of dp or b recombinants among F2
progeny in which In(2L)t was absent varied between 31.1%
and 33.2%. These percentages are less than expected (35.5%).
This is probably due to presence in the samples of chro-
mosomes with reduced recombination caused by small in-
versions (Van Delden and Kamping 1989). In(2L)t appeared
always to be associated with the AdhS/aGpdhF allele com-
bination.
For the total dataset, no latitudinal correlation was found
for the frequency of In(2L)t (Table 3, Fig. 4), if frequency
was calculated as percentage of chromosomes carrying
In(2L)t among all chromosomes tested. However, all chro-
mosomes carrying In(2L)t had an AdhS aGpdhF genotype.
When the frequency of inversion-carrying chromosomes
within all AdhS aGpdhF chromosomes was calculated, a high-
ly significant positive correlation with latitude was found
(Table 3), with the frequency of AdhS aGpdhF chromosomes
carrying In(2L)t approaching 100% at 208 latitude (Fig. 5).
When only the nine Chilean samples were taken in consid-
eration, a negative correlation of latitude and In(2L)t fre-
quency (r 5 20.90, P , 0.001) was found.
Wing length varied from 1.08 mm to 1.45 mm (Fig. 3) and
was positively correlated with latitude (Table 3).
Gametic Disequilibrium
It can be seen from Figure 4 that, although there is a sig-
nificant decline in In(2L)t frequency between 208 and 408
latitude (r 5 20.90, P , 0.001 for this latitudinal range only),
the leveling of this decline is primarily due to the rarity of
AdhS aGpdhF chromosomes in this latitudinal range. Figure
4 also shows the strong latitudinal cline of AdhS aGpdhF
chromosomes (r 5 20.93, P , 0.001 for the correlation of
205LATITUDINAL VARIATION IN D. MELANOGASTER
TABLE 3. Correlation coefficients (r) of AdhS, aGpdhF, and In(2L)t frequencies and wing length on latitude and four climatic variables. For
AdhS and aGpdhF, frequencies for all chromosomes (population) and frequencies corrected for the presence of In(2L)t (ST chromosomes) are
given. A distinction is also made for In(2L)t frequencies: The proportion of all chromosomes carrying In(2L)t (population) and the proportion
of AdhSaGpdhF chromosomes with In(2L)t (AdhSaGpdhF chromosomes) are presented. Sample sizes are shown in parentheses.
Trait Latitude Tyear Ryear Syear Hyear
AdhS (population) 20.93***
(15)
0.95***
(15)
0.53*
(15)
20.19
(10)
0.40
(15)
AdhS (ST chromosomes) 20.88***
(15)
0.91***
(15)
0.61*
(15)
20.34
(10)
0.42
(15)
aGpdhF (population) 20.72**
(15)
0.83***
(15)
0.62*
(15)
0.11
(10)
0.59*
(15)
aGpdhF (ST chromosomes) 20.64**
(15)
0.76**
(15)
0.59*
(15)
0.01
(10)
0.45
(15)
In(2L)t (population) 20.36
(15)
0.29
(15)
20.36
(15)
0.58
(10)
20.29
(15)
In(2L)t (AdhSaGpdhF
chromosomes)
0.81***
(15)
20.87***
(15)
20.59*
(15)
0.19
(10)
20.42
(15)
Wing length 0.97***
(15)
20.98***
(15)
20.43
(15)
20.04
(10)
20.29
(15)
* P , 0.05; ** P , 0.01; *** P , 0.001.
FIG. 3. Relation between male wing length (mm) and latitude.
Vertical bars indicate standard error of the mean. Linear regression:
r2 5 0.96, F 5 303.66, P , 0.0001.
AdhS aGpdhF chromosome frequency and latitude), where in
tropical populations the AdhS aGpdhF chromosomes reach
0.90?0.95. This is clearly due to gametic disequilibrium be-
tween AdhS and aGpdhF (see below).
Both allozyme loci were tested for deviations from Hardy-
Weinberg equilibrium (Table 2). For aGpdh, no deviations
were detected. Adh, however, did show significant deviations
for two populations (Arica and Valpara??so), caused by short-
age of heterozygotes.
To test for independence of segregation between Adh and
aGpdh, D-values for gametic disequilibrium were calculated
and the resulting values were tested against the hypothesis
that D 5 0. The results are summarized in Table 2. It is clear
that D tends to become increasingly negative when the dis-
tance of the population from the equator becomes larger (r
5 20.61, P , 0.05 for the correlation between D and lati-
tude), and this points to a relative excess of chromosomes
with the AdhS aGpdhF and the AdhF aGpdhS combinations at
higher latitudes. This holds also for D/Dmax, a measure for
the degree of gametic disequilibrium that is more independent
of allele frequencies than D (correlation with latitude: r 5
0.65, P , 0.01).
Heterozygosity
Heterozygosity (H) was estimated as a measure for genetic
diversity. Contrary to the results presented by Oakeshott et
al. (1984) and Singh et al. (1982) (who screened populations
from several continents for 10 and 26 loci, respectively) our
results showed a positive correlation between heterozygosity
and latitude (r 5 0.69, P , 0.01). Oakeshott et al. (1984)
found no significant relationship between H and latitude,
while Singh et al. (1982) found a negative correlation. One
of the reasons for this discrepancy may be the fact that our
data are based on two loci only: Adh and aGpdh, which are
both probably under some (direct or indirect) selection pres-
sure. In addition, Table 2 shows that the correlation with
latitude is mainly caused by the markedly lower H-values for
the Panamanian populations. Heterozygosity based on In(2L)t
karyotypes did not produce any indication of a correlation
with latitude (r 5 0.36, ns).
DISCUSSION
We found latitudinal clines in Central and South America
for Adh, aGpdh, wing length and, to a certain extent, for
In(2L)t as well. All clines have a general correspondence to
those reported for other continents (e.g., Voelker et al. 1977;
Knibb 1982; Oakeshott et al. 1982; Anderson et al. 1987;
David et al. 1989; Capy et al. 1993).
A point for discussion is the degree to which the latitudinal
clines are actually the result of natural selection. Drosophila
melanogaster is supposed to have originated in East Africa,
spread out to Eurasia, and colonized Australia and the Amer-
icas only recently (Lemeunier et al. 1986; David and Capy
1988; Hale and Singh 1991; Capy et al. 1993). However, the
history of South American populations remains unclear. As
a result of the commensal relationship between D. melano-
206 J. VAN ?T LAND ET AL.
FIG. 4. Relation between frequencies of In(2L)t (C) and AdhS/
aGpdhF chromosomes (1) with latitude.
gaster and humans, it is possible that the Chilean populations
have descended from European populations, whereas the Pan-
amanian and Ecuadorian flies could have originated from a
mix of descendants from African and European flies (see
Veuille et al. 1998). The clines presented in this paper could
then be the result of migration and gene flow between ad-
jacent source populations (Krimbas and Loukas 1980; Coyne
et al. 1987). This could lead to a gradient of allozyme and
inversion frequencies or morphological characters between
the tropical African populations around the equator to the
temperate European populations in Chile.
There are several arguments against this hypothesis. First,
extensive research on inversion polymorphism in recently
colonizing populations of Drosophila subobscura in South
and North America has provided strong support for the adap-
tive value of inversion polymorphisms and for the speed at
which these latitudinal clines can be formed (Prevosti et al.
1988, but for conflicting results on quantitative traits, see
Pegueroles et al. 1995). Second, latitudinal correlations are
now known to exist for five continents. These are strong
arguments for the hypothesis that the allele frequencies of
Adh and aGpdh and the frequency of In(2L)t in wild popu-
lations of D. melanogaster are under natural selection. Fi-
nally, extensive laboratory research on Adh, aGpdh and
In(2L)t has provided experimental support for selection.
With respect to the maintenance of the Adh polymorphism,
evidence is accumulating that balancing selection is involved.
Strong indications come from analyses of DNA polymor-
phisms in and around the Adh region. Berry and Kreitman
(1993) investigated populations in a north-south cline along
the east coast of North America. They found a distinct cline
of the S/F polymorphism and the ,1 insertion/deletion poly-
morphism (located in the 59 adult intron), but not for nu-
merous silent DNA polymorphisms (see also Kreitman and
Aguade? 1986; Simmons et al. 1989). Berry and Kreitman
concluded that despite high levels of homogenizing gene
flow, the S/F and ,1 polymorphisms were under clinal se-
lection. Kreitman and Hudson (1991) found an excess of
silent nucleotide variation around the F/S site indicating bal-
ancing selection. This fits with the results of Bijlsma-Meeles
and Bijlsma (1988), who estimated fitnesses of Adh genotypes
under laboratory conditions without the presence of ethanol.
They found overdominance for female fecundity and lower
male virility for AdhSS compared to AdhFF and AdhFS. These
results agree with the finding of converging allele frequencies
over generations in experimental populations started with dif-
ferent initial frequencies (Van Delden et al. 1978; Bijlsma-
Meeles and Bijlsma 1988), thus pointing to the existence of
an equilibrium allele frequency.
We observed that the correlation between latitude and AdhS
frequency is not linear, but shows a steeper part between 108
and 188 (Fig. 2). David et al. (1989) described a ??Mediter-
ranean instability?? in populations sampled between 308 and
428 latitude, which results in a nonlinear relation between
latitude and Adh allele frequencies when these frequencies
are plotted over a broader range of latitudinal degrees. This
seems not to be the case for the Mediterranean-like region
in Chile, which is located between 328 and 378. Also, the
location and the climate of an instability region varies be-
tween studies (308 to 428 in the Mediterranean, David et al.
1989; 208 to 288 in India, Parkash and Shamina 1994; 108 to
188, this study), indicating that it is not the typical Mediter-
ranean climate (dry, hot summers and mild rainy winters)
that, as claimed by David et al. (1989), is the direct cause
of the high variation in Adh allele frequencies.
However, the exact nature of the selection forces on all
the polymorphisms considered here is still not clear. Our
results show that Adh and aGpdh alleles become more strong-
ly linked the further the population is from the equator and
that the significant negative D-values and the large D/Dmax
values are caused by the fact that virtually all AdhS alleles
in these southern populations are found in combination with
aGpdhF alleles. When the frequency of the AdhS/aGpdhF
gamete is calculated on the basis of the respective one-locus
allele frequencies, it becomes clear that in all populations
from Copiapo? southward it is approximately two times higher
than expected. It is very likely that this is due to the presence
of In(2L)t. Given this, we can conclude: (1) that there is a
strong association between AdhS and aGpdhF at temperate
latitudes that is caused by In(2L)t; and (2) that even at these
higher latitudinal degrees, there exists some selective force
that maintains In(2L)t and therefore the AdhS aGpdhF com-
bination as well at a relative high frequency. Such strong
gametic disequilibrium due to association with In(2L)t has
also been observed in laboratory populations kept at high
temperature (Van Delden and Kamping 1989, 1991) and a
seminatural population in a tropical greenhouse (Kamping
and Van Delden 1999).
A negative latitudinal cline for In(2L)t was found for the
Chilean populations (Fig. 4). The correlation between latitude
and inversion frequency is not significant if the tropical pop-
ulations are also taken into account. The fact that the In(2L)t
frequencies from our tropical populations are high, but do
not reach values higher than 46%, may be due to overdom-
inance. Veuille et al. (1998) have also reported high In(2L)t
frequencies in tropical African populations: from 0.23 in Ma-
lawi to 0.73 in the Ivory Coast. Van Delden and Kamping
(1989, 1991, 1997) found that In(2L)t homozygotes devel-
oped more slowly and had a lower weight than the other two
karyotypes, with the result that the In(2L)t frequency de-
creased in cultures reared at temperatures of 208C and 258C.
At high temperatures (298C or 338C), however, In(2L)t-car-
207LATITUDINAL VARIATION IN D. MELANOGASTER
FIG. 5. Relation between In(2L)t frequency within AdhS/aGpdhF
chromosomes and latitude. The solid line represents an empirically
fitted exponential curve: y 5 1.08 2 1.14e(2x/16.78) (r2 5 0.76, F
5 19.42, P 5 0.0002).
rying genotypes were more heat resistant than ST homozy-
gotes and overdominance was observed. Consequently the
inversion frequencies were kept stable at equilibrium values.
In addition, lower fitness of individuals carrying an In(2L)t
chromosome was found on ethanol-supplemented food. Het-
erosis of In(2L)t was also described for other populations
(Watanabe and Watanabe 1973; Stalker 1976). The combi-
nation of overdominance with local differences in ecoclimatic
conditions (i.e., amount of ethanol in food and mean tem-
peratures) may give rise to variable tropical In(2L)t equilib-
rium frequencies. However, four populations showed signif-
icant deviations from Hardy-Weinberg equilibria (Table 2),
resulting from a shortage of In(2L)t heterokaryotypes, so we
cannot confirm the presence of overdominance for this in-
version in the Central and South American populations.
Adh, aGpdh, and inversion polymorphisms have all been
related to body size and wing length (Pieragostini et al. 1979;
Serra and Oller 1984; Barnes and Laurie-Ahlberg 1986; Oud-
man et al. 1991; Van Delden and Kamping 1991; Hasson et
al. 1992). In our study, wing length was found to be strongly
correlated with latitude and temperature. Many studies have
reported a latitudinal cline for wing length in wild popula-
tions of D. melanogaster on other continents as well (e.g.,
Imasheva et al. 1994; James et al. 1995; Long and Singh
1995). A high rearing temperature causes shorter develop-
ment time and smaller adult body size. However, the lati-
tudinal cline for wing length still persists in our stocks when
reared in the laboratory under standard conditions, which
indicates a genetic influence on wing length (Van ?t Land et
al. 1999). Stalker (1980) found that flies adapted to high
temperatures have a relative large wing-load index, i.e., rel-
atively small wings and a rapid wing beat. This may relate
to Adh, aGpdh, or In(2L)t. There exists experimental evidence
that flies with either an aGpdhFF or an AdhSS genotype are
smaller and have shorter wings, and that aGpdhFF flies have
a higher flight output at high temperatures (Pieragostini et
al. 1979; Serra and Oller 1984; Barnes and Laurie-Ahlberg
1986; Oudman et al. 1991). Our findings of higher frequen-
cies for AdhS and aGpdhF and shorter wings in the tropics
are consistent with these results. However, the mechanism
responsible for the connection between the polymorphisms
at these loci and a polygenic trait like wing length remains
a matter of further investigation (see Van ?t Land et al. 1999).
ACKNOWLEDGMENTS
This research was funded by the Netherlands Foundation
for the Advancement of Tropical Research (WOTRO) by
grant W 82?182. We acknowledge the essential help of the
Smithsonian Tropical Research Station, Panama, for provid-
ing the facilities for the collecting and measuring of the Pan-
amanian flies. R. Choez and her students from the University
of Guayaquil, Ecuador, kindly assisted with collecting the
Ecuadorian flies. E. Zouros and two anonymous reviewers
gave useful suggestions for improvement of the text. We
thank E. Hartman for her help in preparing the manuscript.
LITERATURE CITED
Aguade?, M., and L. Serra. 1987. Analysis of microdifferentiation
in a Spanish cellar population of Drosophila melanogaster. Ge-
netica 75:3?9.
Alahiotis, S. N., S. Miller, and E. Berger. 1977. Natural selection
at the a- GDH locus in Drosophila. Nature 269:144?145.
Alonso-Moraga, A., and A. Mun?oz-Serrano. 1986. Allozyme poly-
morphism and linkage disequilibrium of Adh and aGpdh loci in
wine cellar and field populations of Drosophila melanogaster.
Experientia 42:1048?1050.
Anderson, P. R., W. R. Knibb, and J. G. Oakeshott. 1987. Obser-
vations on the extent and temporal stability of latitudinal clines
for alcohol dehydrogenase allozymes and four chromosome in-
versions in Drosophila melanogaster. Genetica 75:81?88.
Barnes, P. T., and C. C. Laurie-Ahlberg. 1986. Genetic variability
of flight metabolism in Drosophila melanogaster. III. Effects of
Gpdh allozymes and environmental temperature on power out-
put. Genetics 112:267?294.
Barnes, P. T., B. Holland, and V. Courreges. 1989. Genotype-by-
environment and epistatic interactions in Drosophila melano-
gaster: the effects of Gpdh allozymes, genetic background and
rearing temperature on larval developmental time and viability.
Genetics 122:859?868.
Be?nassi, V., and M. Veuille. 1995. Comparative population struc-
turing of molecular and allozyme variation of Drosophila me-
lanogaster Adh between Europe, West Africa and East Africa.
Genet. Res. 65:95?103.
Berger, E. M. 1971. A temporal survey of allelic variation in natural
and laboratory populations of Drosophila melanogaster. Genetics
67:121?136.
Berry, A., and M. Kreitman. 1993. Molecular analysis of an allo-
zyme cline: alcohol dehydrogenase in Drosophila melanogaster
on the east coast of North America. Genetics 134:869?893.
Bijlsma-Meeles, E., and R. Bijlsma. 1988. The alcohol dehydro-
genase polymorphism in Drosophila melanogaster: fitness mea-
surements and predictions under conditions with no alcohol
stress. Genetics 120:743?753.
Bitner-Mathe?, B. C., A. A. Peixoto, and L. B. Klaczko. 1995. Mor-
phological variation in a natural population of Drosophila me-
diopunctata: altitudinal cline, temporal changes and influence of
chromosome inversions. Heredity 75:54?61.
Bubli, O. A., and A. G. Imasheva. 1997. Geographic variation of
Adh-F allele frequency in Drosophila melanogaster populations.
Russ. J. Genetics 33:780?788.
Capy, P., J. R. David, R. Allemand, Y. Carton, G. Febvay, and A.
Kermarec. 1986. Genetic analysis of Drosophila melanogaster
in the French Indies and comparison with populations from other
parts of the world. Genetica 69:167?176.
Capy, P., E. Pla, and J. R. David. 1993. Phenotypic and genetic
variability of morphometrical traits in natural populations of
Drosophila melanogaster and D. simulans. I. Geographic vari-
ations. Genet. Sel. Evol. 25:517?536.
208 J. VAN ?T LAND ET AL.
Cavener, D. R. 1983. The response of enzyme polymorphisms to
developmental rate selection in Drosophila melanogaster. Ge-
netics 105:105?113.
Coyne, J. A., S. H. Bryant, and M. Turelli. 1987. Long-distance
migration of Drosophila. 2. Presence in desolate sites and dis-
persal near a desert oasis. Am. Nat. 129:847?861.
Das, A., and B. N. Singh. 1991. Genetic differentiation and inver-
sion clines in Indian natural populations of Drosophila melan-
ogaster. Genome 34:618?625.
David, J. R., and P. Capy. 1988. Genetic variation of Drosophila
melanogaster natural populations. Trends Genet. 4:106?111.
David, J. R., A. Alonso-Moraga, F. Borai, P. Capy, H. Merc?ot, S.
F. McEvey, A. Munoz-Serrano, and S. Tsacas. 1989. Latitudinal
variation of Adh gene frequencies in Drosophila melanogaster:
a Mediterranean instability. Heredity 62:11?16.
David, J. R., B. Moreteau, J. P. Gauthier, G. Pe?tavy, A. Stockel,
and A. G. Imasheva. 1994. Reaction norms of size characters
in relation to growth temperature in Drosophila melanogaster:
an isofemale lines analysis. Genet. Sel. Evol. 26:229?251.
Endler, J. A. 1977. Geographic variation, speciation, and clines.
Princeton Univ. Press, Princeton, NJ.
Gibson, J. B., A. Cao, J. Symonds, and D. Reed. 1991. Low activity
sn-glycerol-3-phosphate dehydrogenase variants in natural pop-
ulations of Drosophila melanogaster. Heredity 66:75?82.
Grossman, A. I., L. G. Koreneva, and L. E. Ulitscaya. 1970. The
variability of ADH locus in a natural population of Drosophila
melanogaster. Genetika 6:91?96.
Hale, L. R., and R. S. Singh. 1991. A comprehensive study of genic
variation in natural populations of Drosophila melanogaster. IV.
Mitochondrial DNA variation and the role of history vs. selection
in the genetic structure of geographic populations. Genetics 129:
103?117.
Hasson, E., J. J. Fanara, C. Rodriguez, J. C. Vilardi, O. A. Reig,
and A. Fontdevila. 1992. The evolutionary history of Drosophila
buzzatii. XXIV. Second chromosome inversions have different
average effects on thorax length. Heredity 68:557?563.
Hedrick, P. W. 1983. Genetics of populations. Science Books In-
ternational, Boston, MA.
???. 1987. Gametic disequilibrium measures: proceed with cau-
tion. Genetics 117:331?341.
Hill, W. G. 1974. Estimation of linkage disequilibrium in randomly
mating populations. Heredity 33:229?239.
Imasheva, A. G., O. A. Bubli, and O. E. Lazebny. 1994. Variation
in wing length in Eurasian natural populations of Drosophila
melanogaster. Heredity 72:508?514.
Inoue, Y., and Y. Igarashi. 1994. On the category of naturally oc-
curring inversions of Drosophila melanogaster. Jpn. J. Genet.
69:105?118.
Inoue, Y., and T. K. Watanabe. 1980. Inversion polymorphism in
some African, New Guinean, and Philippine populations of D.
melanogaster. Annu. Rep. Natl. Inst. Genet. (Jpn.) 30:88?90.
Inoue, Y., Y. N. Tobari, K. Tsuno, and T. K. Watanabe. 1984.
Association of chromosome and enzyme polymorphisms in nat-
ural and cage populations of Drosophila melanogaster. Genetics
106:267?277.
James, A. C., R. B. R. Azevedo, and L. Partridge. 1995. Cellular
basis and developmental timing in a size cline of Drosophila
melanogaster. Genetics 140:659?666.
Johnson, F. M., and H. E. Schaffer. 1973. Isozyme variability in
species of the genus Drosophila. VII. Genotype-environment
relationships in populations of D. melanogaster from the eastern
United States. Biochem. Genet. 10:149?163.
Jones, M. D. R. 1974. Inversion polymorphism and circadian flight
activity in the mosquito Anopheles stephensi (Diptera, Cullici-
dae). Bull. ent. Res. 64:305?311.
Kamping, A., and W. van Delden. 1999. A long-term study on
interactions between the Adh and aGpdh allozyme polymor-
phisms and the chromosomal inversion In(2L)t in a seminatural
population of D. melanogaster. J. Evol. Biol. 12:809?821.
Kim, D. I., and K. C. Sung. 1988. Chromosomal polymorphism of
Drosophila melanogaster according to the seasonal changes and
geographical clines in Korea. Kor. J. Gen. 10:17?36.
Knibb, W. R. 1982. Chromosome inversion polymorphisms in Dro-
sophila melanogaster. II. Geographic clines and climatic asso-
ciations in Australasia, North America and Asia. Genetica 58:
213?221.
???. 1983. Chromosome inversion polymorphisms in Drosophila
melanogaster. III. Gametic disequilibria and the contributions of
inversion clines to the Adh and Gpdh clines in Australasia. Ge-
netica 61:139?146.
Knibb, W. R., J. G. Oakeshott, and J. B. Gibson. 1981. Chromosome
inversion polymorphisms in Drosophila melanogaster. I. Lati-
tudinal clines and associations between inversions in australa-
sian populations. Genetics 98:833?847.
Kreitman, M., and M. Aguade. 1986. Excess polymorphism at the
Adh locus in Drosophila melanogaster. Genetics 114:93?110.
Kreitman, M., and R. Hudson. 1991. Inferring the evolutionary
histories of the Adh and Adh-dup loci in Drosophila melanogaster
from patterns of polymorphism and divergence. Genetics 127:
565?582.
Krimbas, C. B., and M. Loukas. 1980. The inversion polymorphism
of D. subobscura. Evol. Biol. 12:163?234.
Langley, C. H., K. Ito, and R. A. Voelker. 1977. Linkage disequi-
librium in natural populations of Drosophila melanogaster: sea-
sonal variation. Genetics 86:447?454.
Lemeunier, F., J. R. David, L. Tsacas, and M. Ashburner. 1986.
The melanogaster species group. Pp. 147?256 in M. Ashburner
and E. Novitski, eds. The genetics and biology of Drosophila.
Vol. 3e. Academic Press, New York.
Long, A. D., and R. S. Singh. 1995. Molecules versus morphology:
the detection of selection acting on morphological characters
along a cline in Drosophila melanogaster. Heredity 74:569?581.
Louis, J., J. David, J. Rouault, and P. Capy. 1982. Altitudinal var-
iations of Afro-tropical D. melanogaster populations. Dros. Inf.
Serv. 58:100?101.
Malpica, J. M., and J. M. Vassallo. 1980. A test for the selective
origin of environmentally correlated allozyme patterns. Nature
286:407?408.
McKenzie, J. A., S. W. McKechnie, and P. Batterham. 1994. Per-
tubation of gene frequencies in a natural population of Dro-
sophila melanogaster: evidence for selection at the Adh locus.
Genetica 92:187?196.
Mettler, L. E., R. A. Voelker, and T. Mukai. 1977. Inversion clines
in populations of Drosophila melanogaster. Genetics 87:
169?176.
Oakeshott, J. G., J. B. Gibson, P. R. Anderson, W. R. Knibb, D.
G. Anderson, and G. K. Chambers. 1982. Alcohol dehydrogenase
and glycerol-3-phosphate dehydrogenase clines in Drosophila
melanogaster on different continents. Evolution 36:86?96.
Oakeshott, J. G., S. W. McKechnie, and G. K. Chambers. 1984.
Population genetics of the metabolically related Adh, Gpdh and
Tpi polymorphisms in Drosophila melanogaster. I. Geographic
variation in Gpdh and Tpi allele frequencies in different conti-
nents. Genetica 63:21?29.
Oudman, L., W. van Delden, A. Kamping, and R. Bijlsma. 1991.
Polymorphism at the Adh and aGpdh loci in Drosophila melan-
ogaster: effects of rearing temperature on developmental rate,
body weight, and some biochemical parameters. Heredity 67:
103?115.
???. 1992. Interaction between the Adh and aGpdh loci in Dro-
sophila melanogaster: adult survival at high temperature. He-
redity 68:289?297.
Parkash, R., and Shamina. 1994. Geographical differentiation of
allozymic variability in natural Indian populations of Drosophila
melanogaster. Biochem. Genet. 32:63?73.
Pegueroles, G., M. Papaceit, A. Quintana, A. Guille?n, A. Prevosti,
and L. Serra. 1995. An experimental study of evolution in pro-
gress: clines for quantitative traits in colonizing and Palearctic
populations of Drosophila. Evol. Ecol. 9:453?465.
Pfriem, P. 1983. Latitudinal variation in wing size in Drosophila
subobscura and its dependence on polygenes of chromosome O.
Genetica 61:221?232.
Pieragostini, E., S. Sangiorgi, and S. Cavicchi. 1979. ADH system
and genetic background: interaction with wing length in Dro-
sophila melanogaster. Genetica 50:201?206.
Pipkin, S. B., E. Franklin-Springer, S. Law, and S. Lubega. 1976.
209LATITUDINAL VARIATION IN D. MELANOGASTER
New studies of the alcohol dehydrogenase cline in D. melano-
gaster from Mexico. J. Hered. 67:258?266.
Prevosti, A., G. Ribo, L. Serra, M. Aguade?, J. Balan?a, M. Monclus,
and F. Mestres. 1988. Colonization of America by Drosophila
subobscura: experiment in natural populations that supports the
adaptive role of chromosomal-inversion polymorphism. Proc.
Natl. Acad. Sci. USA 85:5597?5600.
Prout, T. 1958. A rapid method for measuring winglength. Dros.
Inf. Serv. 32:170?171.
Robertson, F. W., and E. C. R. Reeve. 1952. Studies in quantitative
inheritance. I. The effects of selection on wing and thorax length
in Drosophila melanogaster. J. Genet. 50:414?448.
Sanchez-Refusta, F., E. Santiago, and J. Rubio. 1990. Seasonal
fluctuations of cosmopolitan inversion frequencies in a natural
population of Drosophila melanogaster. Genet. Sel. Evol. 22:
47?56.
Serra, L., and J. M. Oller. 1984. Analysis of allozymic and quan-
titative variation produced by artificial selection in Drosophila
melanogaster. Genetica 63:39?47.
Simmons, G. M., M. Kreitman, W. F. Quattelbaum, and N. Mi-
yashita. 1989. Molecular analysis of the alleles of alcohol de-
hydrogenase along a cline in D. melanogaster. Evolution 43:
393?409.
Singh, B. N., and A. Das. 1992. Further evidence for latitudinal
inversion clines in natural populations of Drosophila melano-
gaster from India. J. Hered. 83:227?230.
Singh, R. S., D. A. Hickey, and J. R. David. 1982. Genetic differ-
entiation between geographically distant populations of Dro-
sophila melanogaster. Genetics 101:235?256.
Stalker, H. D. 1976. Chromosome studies in wild populations of
D. melanogaster. Genetics 82:323?347.
???. 1980. Chromosome studies in wild populations of Dro-
sophila melanogaster. II. Relationship of inversion frequencies
to latitude, season, wing-loading, and flight activity. Genetics
95:211?223.
Van Delden, W. 1982. The alcohol dehydrogenase polymorphism
in Drosophila melanogaster: selection at an enzyme locus. Evol.
Biol. 15:187?222.
Van Delden, W., and A. Kamping. 1980. The alcohol dehydrogenase
polymorphism in populations of Drosophila melanogaster. IV.
Survival at high temperatures. Genetica 51:179?185.
???. 1989. The association between the polymorphisms at the
Adh and aGpdh loci and the In(2L)t inversion in Drosophila
melanogaster in relation to temperature. Evolution 43:775?793.
???. 1991. Changes in relative fitness with temperature among
second chromosome arrangements in Drosophila melanogaster.
Genetics 127:507?514.
???. 1997. Worldwide latitudinal clines for the alcohol dehy-
drogenase polymorphism in Drosophila melanogaster: what is
the unit of selection? Pp.97?115 in R. Bijlsma and V. Loeschcke,
eds. Environmental stress, adaptation, and evolution. Birkha?user
Verlag, Basel, Switzerland.
Van Delden, W., A. C. Boerema, and A. Kamping. 1978. The al-
cohol dehydrogenase polymorphism in populations of Drosoph-
ila melanogaster. I. Selection in different environments. Genetics
90:161?191.
Van ?t Land, J., P. van Putten, B. Zwaan, A. Kamping, and W. van
Delden. 1999. Latitudinal variation in wild populations of Dro-
sophila melanogaster: heritabilities and reaction norms. J. Evol.
Biol. 12:222?232.
Veuille, M., V. Be?nassi, S. Aulard, and F. Depaulis. 1998. Allele-
specific population structure of Drosophila melanogaster alcohol
dehydrogenase at the molecular level. Genetics 149:971?981.
Vigue, C. L., and F. M. Johnson. 1973. Isozyme variability in spe-
cies of the genus Drosophila. VI. Frequency-property-environ-
ment relationships of allelic alcohol dehydrogenase in Dro-
sophila melanogaster. Biochem. Genet. 9:213?227.
Voelker, R. A., T. Mukai, and F. M. Johnson. 1977. Genetic var-
iation in populations of Drosophila melanogaster from the west-
ern United States. Genetica 47:143?148.
Voelker, R. A., C. Clark Cockerham, F. M. Johnson, H. E. Schaffer,
T. Mukai, and L. E. Mettler. 1978. Inversions fail to account
for allozyme clines. Genetics 88:515?527.
Watanabe, T. K., and T. Watanabe. 1973. Fertility genes in natural
populations of Drosophila melanogaster. III. Superiority of in-
version heterozygotes. Evolution 27:468?475.
Yamaguchi, O., M. Ichinose, M. Matsuda, and T. Mukai. 1980.
Linkage disequilibrium in isolated populations of Drosophila
melanogaster. Genetics 96:507?522.
Zacharopoulou, A., and M. Pelecanos. 1980. Seasonal and year-to-
year inversion polymorphism in a southern Greek Drosophila
melanogaster wild population. Genetica 54:105?111.
Zwaan, B. J., R. Bijlsma, and R. F. Hyoekstra. 1992. On the de-
velopmental theory of ageing. II. The effect of developmental
temperature on longevity in relation to adult body size in D.
melanogaster. Heredity 68:123?130.
Corresponding Editor: E. Zouros