ArticlePDF Available

Evidence of Heterosis Associated with an Enzyme Locus in a Natural Population of Drosophila

Authors:
  • Retired from Humboldt State University

Abstract

A wild population of Drosophila paulistorum (Andean semispecies) was analyzed by starch gel electrophoresis for a sex-linked enzyme locus, Tetrazolium oxidase. The results of the analysis of F(1) progenies from 106 single female lines from São José do Rio Prêto, Brazil, has allowed us to determine the genotype of 106 wild-collected females and 104 wild-collected males. The population is polymorphic for two alleles with frequencies of 0.56 and 0.44. The observed frequencies among the females of the two homozygotes were 0.23 and 0.13 and of the heterozygote 0.64. The deviation from the frequencies expected according to the Hardy-Weinberg formula is statistically significant (P < 0.005). There is a substantial excess of heterozygotes over the expected number. This observation is interpreted to mean that this polymorphism is maintained by heterosis. This is probably the first reported case of heterosis associated with an enzyme locus in a natural population of Drosophila.
Procedings
of
the
National
Academy
of
Sciences
Vol.
67,
No.
3,
pp.
1264-1267,
November
1970
Evidence
of
Heterosis
Associated
with
an
Enzyme
Locus
in
a
Natural
Population
of
Drosophila*
Rollin
C.
Richmondt
and
Jeffrey
R.
Powell
THE
ROCKEFELLER
UNIVERSITY,
NEW
YORK,
N.Y.
10021
Communicated
by
Theodosius
Dobzhansky,
August
7,
1970
Abstract.
A
wild
population
of
Drosophila
paulistorum
(Andean
semispecies)
was
analyzed
by
starch
gel
electrophoresis
for
a
sex-linked
enzyme
locus,
Tetra-
zolium
oxidase.
The
results
of
the
analysis
of
F1
progenies
from
106
single
female
lines
from
Sao
Jose
do
Rio
Preto,
Brazil,
has
allowed
us
to
determine
the
genotype
of
106
wild-collected
females
and
104
wild-collected
males.
The
population
is
polymorphic
for
two
alleles
with
frequencies
of
0.56
and
0.44.
The
observed
frequencies
among
the
females
of
the
two
homozygotes
were
0.23
and
0.13
and
of
the
heterozygote
0.64.
The
deviation
from
the
frequencies
expected
according
to
the
Hardy-Weinberg
formula
is
statistically
significant
(P
<
0.005).
There
is
a
substantial
excess
of
heterozygotes
over
the
expected
number.
This
observa-
tion
is
interpreted
to
mean
that
this
polymorphism
is
maintained
by
heterosis.
This
is
probably
the
first
reported
case
of
heterosis
associated
with
an
enzyme
locus
in
a
natural population
of
Drosophila.
It
has
now
been
unequivocally
shown
that
natural
populations
of
organisms
as
diverse
as
man
and
several
species
of
Drosophila
have
many
enzyme
poly-
morphisms.1-3
How
these
polymorphisms
are
maintained
is
an
open
question.
Lewontin
and
Hubby
have
pointed
out
that
the
current
mathematical
formula-
tions
of
selectional
processes
in
natural
populations
cannot
account
for
the
ob-
served
abundance
of
enzyme
polymorphisms
in
Drosophila
populations.4
Sev-
eral
authors
have
suggested
more
realistic
models
of
selectional
processes
which
would
allow
populations
to
contain
many
polymorphic
loci.5-7
Kimura
and
Crow
propose
that
many
polymorphisms
are
present
in
natural
populations
because
of
the
selective
neutrality
of
the
alleles
involved.8'
The
evidence
presented
below
indicates
that
natural
selection
can
and
does
alter
the
frequencies
of
some
allozyme2
alleles
in
natural
populations.
The
polymorphism
we
have
studied
appears
to
be
maintained
in
the
population
by
selection
in
favor
of
individuals
heterozygous
for
two
alleles.
Materials
and
Methods.
Single
Drosophila
paulistorum
females
inseminated
in
the wild
were
used
to
start
106
strains.
All
females
were
collected
at
a
single
locality,
Mirassol,
near
Sao
Jose
do
Rio
Preto,
Brazil.
As
D.
paulistorum
is
a
complex
of
six
partly
reproductively
isolated
semispecies,
the
strains
were
crossed
to
standard
tester
stocks
to
determine
to
which
semispecies
they
belong.10
All
strains
from
Rio
Preto
used
in
this
study
were
of
the
Andean
semispecies.
No
other
semispecies
has
been
found
in
this
locality.
One
F1
male
and
six
F1
female
progenies
from
each
strain
were
analyzed
for
the
allo-
zymes
of
Tetrazolium
oxidase
(To),
which
is
(R.
C.
Richmond,
unpublished
data)
a
sex-
1264
VOL.
67,
1970
HETEROSIS
IN
DROSOPHILA
1265
linked
locus
in
D.
paulistorum.
From
the
genotypes
of
the
progenies,
the
genotype
of
the
parents
can
be
determined
with
high
probability.
The
probability
of
not
detecting
one
of
the
two
alleles
of
the
female
parent,
so
that
the
female
would
be
erroneously
scored
as
a
homozygote
is
only
(0.5)6.
Starch
gel
electrophoresis
was
used
to
detect
the
various
forms
of
the
enzyme.
The
gel
was
prepared
from
electrophoresis
starch
(Sigma)
in
87
mM
Tris-boric
acid
buffer
(pH
9.0)
containing
1
mM
EDTA
and
0.1
mM
3-NAD+.
Each
fly
was
homogenized
in
buffer
of
this
composition;
the
homogenate
was
absorbed
by
a
small
piece
of
filter
paper
and
placed
in
a
slot
in
the
gel.
A
voltage
gradient
of
20-25
V/cm
was
applied
for
6
hr.
The
gel
was
sliced
and
stained
in
100
ml
of
50
mM
Tris-HCl
buffer
(pH
8.5)
contain-
ing
20
mg
of
Nitro
Blue
Tetrazolium
salt
(Sigma),
25
mg
of
O-NAD
+,
and
5
mg
of
phen-
azine
methosulfate
(Sigma).
The
gel
in
the
stain
was
exposed
to
light
for
1
hr,
the
stain
was
then
removed,
and
the
gel
was
fixed
in
acetic
acid-methanol-water
1:5:5.
Light
catalyzes
reduction
of
the
tetrazolium
salt
to
a
blue
precipitate
except
where
there
is
an
enzyme
(To)
capable
of
oxidizing
the
tetrazolium
salt."
The
in
vivo
function
of
tetrazolium
oxidase
is
not
known
and
this
name
is
given
only
to
describe
its
in
vitro
action.
These
bands
do
not
stain
with
any
other
common
stain
and
variants
behave
as
single
sex-linked
Mendelian
factors.
Results.
Fig.
1
shows
a
zymogram
of
females
homozygous
and
heterozygous
FIG.
1.
Tetrazolium
oxidase
zymogram
of
Drosophila
females
homozygous
and
heterozygous
for
two
allelic
variants.
The
arrow
indicates
direction
of
Ad
G
migration.
Genotype
F
F/
F
/S
S
for
the
two
To
alleles
present
in
the
Mirassol
population.
The
two
variants
are
designated
"F"
(fast
migrating)
and
"S"
(slow
migrating).
From
the
pattern
of
the
heterozygote,
it
is
obvious
that
this
enzyme
is
composed
of
at
least
two
sub-
units.
Table
1
gives
the
genotype
frequencies
of
the
106
females
in
our
sample
as
TABLE
1.
Genotype
frequencies
of
wild
females
of
D.
paulistorum
at
the
Tetrazolium
oxidase
locus.
Expected
values
are
calculated
for
a
Hardy-Weinberg
equilibrium
with
frequencies
of
the
S
allele
being
0.554
and
the
F
being
0.446.
These
frequencies
were
determined
from
the
sample
of
106
mothers
and
104
fathers.
F/F
F/S
S/S
Total
Observed
14
68
24
106
Expected
21.1
52.4
32.5
XI
=
9.26
P
<
0.005
inferred
from
the
genotypes
of
their
progenies.
As
this
locus
is
sex-linked,
no
similar
data
can
be
given
for
males.
The
expected
genotype
frequencies
were
calculated
using
the
gene
frequencies
in
the
natural
population
(from
the
sample
of
106
females
and
104
males
as
there
was
no
difference
between
sexes)
and
as-
suming
a
Hardy-Weinberg
equilibrium.
Laboratory
crosses
did
not
show
any
segregation
distortion
associated
with
this
locus.
In
three
crosses
of
F/S
females
and
males,
27
F
and
28
S
males
were
scored.
Among
the
female
progenies
a
slight
excess
of
heterozygotes
was
de-
tected.
1266
GENETICS:
RICHMOND
AND
POWELL
PROC.
N.
A.
S.
Five
sets
of
reciprocal
crosses
between
strains
homozygous
for
the
two
alleles
were
made
to
provide
heterozygous
females.
Salivary
gland
smears
were
pre-
pared
from
five
larvae
from
each
cross
and
examined
for
the
presence
of
hetero-
zygous
inversions
in
either
arm
of
the
X
chromosome.
Although
the
X
chromo-
some
did
contain
several
inversions,
no
consistent
association
between
any
in-
version
and
the
To
alleles
could
be
established.
Discussion.
Some
possible
explanations
for
the
excess
of
females
heterozygous
for
the
To
locus
are
the
following.
First,
the
gene
frequencies
in
the
population
may
not
be
at
equilibrium,
that
is,
one
allele
may
be
in
the
process
of
replacing
the
other
and
therefore
our
sample
may
be
at
an
intermediate
stage
in
this
directional
selection
process.'2
This
seems
unlikely
because
of
the
magnitude
of
the
heterozygote
excess.
Also,
Richmond
has
noted
(unpublished
observations)
that
several
widely-distributed
populations
of
the
AndeAn
semispecies
are
segregating
for
these
two
alleles.
It
seems
unlikely
that
a
directional
selection
process
would
be
occurring
simulta-
neously
over
so
wide
an
area.
Second,
it
has
been
shown
by
Dobzhansky and
Pavlovsky'3
that
flies
hetero-
zygous
for
a
chromosomal
inversion
are
sometimes
present
in
excess
of
Hardy-
Weinberg
equilibrium
frequencies.
If
the
To
locus
were
associated
with
such
a
supergene,
the
heterozygous
excess
would
not
necessarily
be
due
to
the
locus
per
se,
but
rather
to
its
location
within
an
inversion.
As
mentioned
above,
no
systematic
relationship
between
heterozygosity
for
the
To
locus
and
heterozy-
gosity
for
an
X
chromosome
inversion
was
detected
in
the
material
studied.
Third,
since
the
genotypes
of
the
wild-collected
females
were
inferred
from
their
F,
progenies,
any
systematic
bias
in
the
segregation
ratio
of
the
To
alleles
might
produce
the
results
observed.
However,
laboratory
crosses
showed
no
significant
segregation
distortion,
although
a
slight
heterozygote
excess
was
noted
for
certain
crosses.
Fourth,
heterosis
may
be
acting
at
this
locus.
There
is
evidence
that
"hetero-
zygous
enzymes"
may
have
an
advantage
over
"homozygous
enzymes."'4
Such
an
advantage
may
result
from
association
of
different
subunits
giving
rise
to
an
enzyme
with
different,
favorable
properties.
Our
observations
are
most
consistent
with
heterotic
balance
maintaining
the
To
polymorphism
in
the
population.
To
our
knowledge
this
is
the
best
evidence
to
date
for
this
type
of
mechanism
acting
at
an
enzyme
locus
in
a
natural
pop-
ulation.
However,
the
possibility
that
heterosis
is
acting
not
at
the
To
locus
but
rather
at
some
tightly
linked
locus
or
loci
cannot
be
ruled
out.
The
authors
thank
Olga
Pavlovsky
for
preparing
chromosome
slides,
Boris
Spassky
for
classifying
the
original
wild
collection,
and
Dr.
A.
J.
Gallo
of
the
FFCL
of
Sio
Jos6
do
Rio
Preto,
S.P.
who
kindly
collected
the
original
material.
We
also
thank
Professor
Th.
Dobz-
hansky
and
Dr.
F.
J.
Ayala
for
their
helpful
discussion
and
criticism
of
the
manuscript.
*
Supported
by
NSF
grant
GB-12562
(International
Biological
Program),
AEC
contract
AT-(30-1),
NIH
predoctoral
fellowship,
1-Fl-GM33,
and
an
NSF
Graduate
Fellowship.
t
Present
address:
Department
of
Zoology,
Indiana
University,
Bloomington,
Ind.
47401.
1
Harris,
H.,
Proc.
Roy.
Soc.
Ser.
B.,
164,
298
(1966).
2
Prakash,
S.,
R.
C.
Lewontin,
and
J.
L.
Hubby,
Genetics,
61,
841
(1969).
VOL.
67,
1970
HETEROSIS
IN
DROSOPHILA
1267
3
Ayala,
F.
J.,
C.
A.
Mourio,
S.
Perez-Salas,
R.
Richmond,
and
Th.
Dobzhansky,
Proc.
Nat.
Acad.
Si.
USA,
67,
225
(1970).
4Lewontin,
R.
C.,
and
J.
L.
Hubby,
Genetics,
54,
595
(1966).
6
Sved,
J.
A.,
T. E.
Reed,
and
W.
F.
Bodmer,
Genetics,
55,
469
(1967).
6
King,
J.
L.,
Genetics,
55,
483
(1967).
7Milkman,
R.
D.,
Genetics,
55,
493
(1967).
8
Kimura,
M.,
and
J.
F.
Crow,
Genetics,
49,
725
(1964).
9
Kimura,
M.,
Nature,
217,
624
(1968).
'0
Dobzhansky,
Th.,
and
0.
Pavlovsky,
Genetics,
55,
141
(1967)..
Baur,
E.
W.,
and
R.
T.
Schorr,
Science,
166,
1524
(1969).
2
Kettlewell,
H.
B.
D.,
Ann.
Rev.
Entonol.,
6,
245
(1964).
13
Dobzhansky,
Th.,
and
0.
Pavlovsky,
Proc.
Nat.
Acad.
Sci.,
41,
289
(1955).
14
Schwartz,
D.,
and
W.
J.
Laughner,
Science,
177,
626
(1969).
... Although there is no direct evidence for this conclusion, the pattern of isozyme variability seen in D. paulistorum is most readily explained by the balanced selection hypothesis (Richmond, 1972). This indicates that the frequencies of allozyme alleles in the species are subject to selectional forces (see also Richmond and Powell, 1970). ...
... Studies in Drosophila provide evidence for heterosis. Hybrid vigor regarding fitness parameters, sex linked enzyme locus and H 1 histone proteins has been documented (Brncic, 1954;Anderson, 1968;Richmond and Powell, 1970;Martinez and Mcdaniel, 1981;Fry et al., 1998;Vaiserman et al., 2013). ...
Article
Full-text available
Reciprocal cross effects with respect to larval salivary secretory protein levels were studied in the interspecific fertile reciprocal hybrids by crossing Drosophila nasuta nasuta, and Drosophila nasuta albomicans. These proteins are produced copiously during the third larval instar stage and are believed to play a role in the attachment of pupa to the substratum prior to pupariation as well as in insect immunity. Quantitative variations were encountered among the reciprocal hybrids. Significant heterosis was observed between D. n. nasuta and the F1 hybrid female of a cross between D. n. albomicans female and D. n. nasuta male (21.39%) while the F1 hybrids of a cross between D. n. nasuta female and D. n. albomicans male showed a marginal increase (4.24%) from the mid parent level. The glue secretions were correlated to total cell number but independent of gland size. SDS PAGE revealed a considerable heterosis with respect to X-linked protein fractions. Here we report sex specific biochemical heterosis. However the X-linked fractions undergo dosage compensation in both parents and hybrids indicating strict regulatory control.
... These changes may include a decrease in attractiveness to males (Wolfner 1997), reduced receptivity to further mating (Fuyama 1995) as well as decreased lifespan (Chapman et al. 1995). Remating is common in females of numerous Drosophila species under both natural (e.g., Richmond and Powell 1970;Anderson 1974;Cobbs 1977) and laboratory conditions (e.g., Gromko and Pyle 1978;Singh and Singh 1999;Singh et al. 2002). Frequency of female remating, latency to remating, and duration of copulation during the first and second matings varies considerably among species (e.g., Smith 1956;Richmond and Ehrman 1974;Newport and Gromko 1984;Letsinger and Gromko 1985;McRobert et al. 1997;Singh and Singh 1999). ...
Article
Full-text available
Female multiple mating, known as polyandry, is ubiquitous and occurs in a wide variety of taxa. Polyandry varies greatly from species in which females mate with one or two males in their lifetime to species in which females may mate with several different males on the same day. As multiple mating by females is associated with costs, numerous hypotheses attempt to explain this phenomenon. One hypothesis not extensively explored is the possibility that polyandrous behavior is captured and “fixed” in populations via genetic processes that preserve the behavior independently of any adaptive benefit of polyandry. Here, we use female isolines derived from populations of Drosophila pseudoobscura from three locations in North America to examine whether different female remating levels are associated with patterns of chromosome inversions, which may explain patterns of polyandry across the geographic range. Populations differed with respect to the frequency of polyandry and the presence of inversion polymorphisms on the third chromosome. The population with the lowest level of female remating was the only one that was entirely comprised of homokaryotypic lines, but the small number of populations prevented us investigating this relationship further at a population level. However, we found no strong relationship between female remating levels and specific karyotypes of the various isolines.
Article
Using the method of computer densitometry we have determined the expression level of the electrophoretically separated S- and F-allozymes of β-specific esterase (E.C. 3.1.1.2) in Drosophila melanogaster homozygotes and heterozygotes for β-Est locus. α-naphtylacetate, β-naphtylacetate and α-naphtylpropionate were used as substrates. Expression intensity of the esterases was estimated using the quantity of the reaction product which is created as a result of simultaneous azocoupling between naphtol and diazonium during 4, 24, 44 and 64 min of incubation time. We have established the reliable differences between the S-and F-allozyme expression depending on the β-Est locus structure. In all the variants the higher level of summary activity S- and F-allozymes of β-esterase of the heterozygotes comparing to those of two types of homozygotes was demonstrated independently of the Drosophila sex. We compared the characteristics of the expression dynamics of the allozymes of dominant homozygotes (β-Est S/β-EstS), heterozygotes (β-Est Sβ-EstF) and recessive homozygotes (β-Est F/β-EstF). We also consider some possible mechanisms of heterosis of S- and F-allozyme expression according to the theory of biochemical enrichment of heterozygote genotypes.
Article
Given that relatively simple electrophoretic methods have repeatedly proven their usefulness in sorting out difficult species complexes, it seemed a promising approach to apply them to various local populations of E. viridis in order to test whether this species name is in fact being used for more than a single species. Specimens were collected from geographically separated local populations in the Channel region (Saint-Efflam, Granville) and the North Sea (Newcastle, Norderney, Helgoland, Helsingør and Tjärnö). Particular enzymes and general protein patterns of individual animals were compared by means of isoelectric focusing (IEF), and the animals were examined for morphological differences. Distinct differences between the IEF band patterns of specimens from the western and eastern sites were indeed observed, and these were found to be correlated with differences in the morphology and size of the dorsal cirri and the size of the proboscideal papillae. Therefore, it became necessary to subdivide the material into two species, E. viridis for that found in Sweden, Denmark and Germany and E. clavigera for that from France and England.
Article
In this study genetic variation was determined with isozyme analysis in 8 natural house fly (Musca domestica L.) populations from the western and southern coasts of Turkey, a part of the southern Palearctic region. For this purpose, 7 isozyme systems (ACP, AOX, CAT, HBDH, HXK, MDH, and SOD) were resolved using the native polyacrylamide gel electrophoresis (native-PAGE) technique. A total of 9 loci and 20 alleles were detected. The proportion of polymorphic loci in populations ranged from 44.44% to 66.67%. The overall mean number of effective alleles per locus (Ne) was 1.314 +/- 0.454. The mean estimated expected heterozygosity (He) of populations was 0.177 +/- 0.228. A very high proportion (93.70%) of genetic diversity was due to variation within populations while a relatively small portion (6.30%) was due to variation among populations. The average level of gene flow (Nm) was calculated to be 3.7182 per generation among all populations studied. The gene flow detected among the Aegean populations (Nm = 6.4451) was 2 times higher than the gene flow detected among the Mediterranean populations (Nm = 3.3331). Our results confirmed the high levels of genetic diversity in house flies reported earlier, supporting the notion that the southern Palearctic region, particularly the Middle East, is the most likely location for the origin of M. domestica.
Chapter
Introduction Assays Isoenzymes and Electromorphs Structure and Evolutionary Relationships Mechanisms Biosynthesis Conclusions
Article
The degrees of genetic similarity among the semispecies of D. paulistorum have been compared using the indices of Hedrick (1971). These analyses show that the Centroamerican, Transitional and Andean-Brazilian semispecies are closely related and that the Orinocan, Interior and Amazonian semispecies form a similar group. Approximately 9-16% of the genotypic differences between populations of different semispecies can be attributed to semispecific differences. No correlation between the degree of reproductive isolation among semispecies and their genetic similarities exists. Other studies of D. paulistorum have shown that the Amazonian semispecies is probably the most distinct of the semispecies. This investigation reveals that the Orinocan semispecies is the most distinct as judged by enzyme-gene criteria. This finding is discussed in the light of other recent studies of the relationships among the semispecies.
Article
Full-text available
We have studied by gel electrophoresis the variability of 14 structural genes in four sibling species, Drosophila willistoni, D. paulistorum, D. equinoxialis, and D. tropicalis. Samples of about 30 populations from different parts of the distribution areas of each species were examined. Genetic variants are found at every locus; 67% of the loci are polymorphic, having two or more alleles, the rarer of which has a frequency of 5% or higher. The gene frequencies are fairly uniform over the distribution area of each species, but considerably different in different species. It is estimated that individuals which belong to the different species differ on the average in somewhat more than one half of their gene loci. The morphological similarity of the four sibling species contrasts with the extensive diversity in their genetic materials.
Article
The failure to respond normally to parathyroid hormone (PTH) administration has been reported in patients with severe hypomagnesemia. A patient with hypoparathyroidism and a markedly decreased serum concentration of magnesium (0.7 mEq/liter), but a normal red blood cell magnesium level, is described who increased serum calcium concentration and decreased per cent renal tubular reabsorption of phosphate when parathyroid extract was given. It is suggested that PTH responsiveness in hypomagnesemic patients may, at least in part, be dependent upon the adequacy of intracellular magnesium stores. This interpretation is supported by the normal cellular (red blood cell) magnesium concentrations observed in this patient and in comparable studies in which PTH responsiveness in the presence of hypomagnesemia was demonstrated. In addition, a failure of optimal renal conservation of magnesium was noted to occur in this patient since, despite hypomagnesemia, urinary magnesium excretion was greater than the 1 mEq/day loss that is seen when magnesium conservation is induced by means such as dietary restriction.
There are a large number of different enzymes synthesized in the human organism, and many of these probably contain more than one structurally distinct polypeptide chain. If current theories about genes and proteins are correct we must suppose that the primary structure of each of these different polypeptides is determined by a separate gene locus, and that there are probably also other loci which are specifically concerned with regulating the rate of synthesis of particular polypeptides or groups of polypeptides. Furthermore, we may expect that genetical diversity in a human population will to a considerable extent be reflected in enzymic diversity. That is to say, in differences between individuals either in the qualitative characteristics of the enzymes they synthesize, or in differences in rates of synthesis. The work I am going to discuss was largely aimed at trying to get some idea of the extent and character of such genetically determined enzyme diversity among what may be regarded as normal individuals. When my colleagues and I started on this line of work about three years ago the information available about this aspect of the subject was very limited. It had of course been recognized for quite a long time that there are many rare metabolic disorders, the so-called 'inborn errors of metabolism', which are due to genetically determined deficiencies of specific enzymes (Harris I963). These conditions can in general be attributed to mutant genes which result either in the synthesis of an abnormal enzyme protein with defective catalytic properties, or in a gross reduction in rate of synthesis of a specific enzyme protein. By and large such genes appear to be relatively uncommon and have frequencies of between 0.01 and 0-001 in the general population. Heterozygotes often show a partial enzyme deficiency though they are usually in other respects quite healthy. A few cases are also known where a specific enzyme deficiency occurs quite commonly in certain populations. The most extensively studied example of this is glucose 6-phosphate dehydrogenase deficiency, and it seems likely that in this particular case the relatively high incidence in certain populations is attributable to a specific selective advantage which the deficiency may confer in situations where endemic malaria is an important selective agent (Motulsky 1964). Virtually all these enzyme deficiencies have been identified in the first instance because of some more or less striking clinical or metabolic disturbance of which they were the cause. They therefore represent a highly selected group of mutants,
Article
Calculating the rate of evolution in terms of nucleotide substitutions seems to give a value so high that many of the mutations involved must be neutral ones.
Article
S pointed out in the first paper of this series (HUBBY and LEWONTIN 1966), A no one knows at the present time the kinds and frequencies of variant alleles present in natural populations of any organism, with the exception of certain special classes of genes. For human populations we know a good deal about certain polymorphisms for blood cell antigens, serum proteins, and metabolic disorders of various kinds but we can hardly regard these, a priori, as typical of the genome as a whole. Clearly we need a method that will randomly sample the genome and detect a major proportion of the individual allelic substitutions that are segre- gating in a population. In our previous paper, we discussed a method for accom- plishing this end by means of a study of electrophoretic variants at a large number of loci and we showed that the variation picked up by this method behaves in a simple Mendelian fashion so that phenotypes can be equated to homozygous and heterozygous genotypes at single loci. It is the purpose of this second paper to show the results of an application of the method to a series of samples chosen from natural populations of Drosophila pseudoobscura. In particular, we will show that there is a considerable amount of genic variation segregating in all of the populations studied and that the real variation in these populations must be greater than we are able to demonstrate. This study does not make clear what balance of forces is responsible for the genetic variation observed, but it does make clear the kind and amount of varia- tion at the genic level that we need to explain. An exactly similar method has recently been applied by HARRIS (1966) for the enzymes of human blood. In a preliminary report on ten randomly chosen enzymes, HARRIS describes two as definitely polymorphic genetically and a third as phenotypically polymorphic but with insufficient genetic data so far. Clearly these methods are applicable to any organism of macroscopic dimensions.