|
Genetic divergence for characters pertaining to reproductive isolation is
of considerable interest in evolutionary biology. There has been substantial
progress in the last decade in understanding the evolution of reproductive is
olation. A main conclusion is that tie genetic divergence underlying postmati
ng isolation between sibling species is very extensive. On the other hand, we
know little of the genetics of premating isolation due to the lack of appropr
iate materials. Since molecular studies were widely used in population geneti
c studies, several species have been found to contain populations differentiat
ing at the molecular level. These help us to find a possible candidate for th
e study of genetic variation at an incipient stage of speciation. In this stu
dy, I first investigate the population geneticdifferentiation within Drosophil
a formosana. Both esterase and mitochondrial DNA (mtDNA) RFLP showed D. f
ormosana and D. immigrans to have very low genetic variation in Taiwan populat
ion. This suggested that these two species may have been through population b
ottleneck recently. Surprisingly, a survey of the cytochrome oxidase II (COII
) sequence polymorphism found extensive genetic differentiation between the Th
ailand and Taiwan population of D. formosana. Furthermore, complete premating
isolation between these two populations was observed. These two "populations
" have apparently evolved to the subspecies or species level, although very fe
w morphological differences are detectable between them. To demonstrate the
genetic architecture of species divergence, I chose to genetically dissect se
xual isolation in D. melanogaster. The females of D. melanogaster collected f
rom Zimbabwe and nearby regions (the Z-type) would mate only with males of the
same geographical origin but not with the cosmopolitan D. melanogaster males
(the M-type). The variations in mating preference among natural isolates of D
. melanogaster, represented by the cosmopolitan and Zimbabwe sexual races, hav
e been measured. Surveyingfour populations in southern Africa and several cos
mopolitan lines, we observe extensive genetic variation in sexual characters a
s well as a positive correlation between sexes. The populations are highly di
fferentiated and represent various stages of evolution between the African and
the cosmopolitan type of sexual behaviors. The genetic variation for these se
xual characters coupled with their pattern of evolution have implications for
models of speciation by sexual selection. The genes for the behaviors are m
apped to all three major chromosomes with the same ranking and comparable magn
itude of effects for both sexes: III > II >> X ( 0 (III, II and X were designa
ted as the effects of the three chromosomes). The genes for Z-maleness are nu
merous and somewhat redundant since males carrying only partial Z-genotypes ca
n completely monopolize mating with Z-females when competed against M-males.
In contrast, Z-femalenesshas less redundancy as partial genotypes never exhibi
t full phenotypic effect.Whole-chromosome effects for Z-maleness appear nearly
additive and show little dominance. Chromosomal effects for Z-femaleness are
more complex－epistatic interactions are evident in certain genetic background
s and Z-dominance, though incomplete, can often be seen. Furthermore, the chr
omosomal effects are strong enough to allow further characterization at the ge
nic level. Here I show that at least four different chromosomal segments for
either male or female behavior can be identified on the third chromosome alone
. Apparently, extensive genetic divergence underlying sexual isolation can ev
olve between races or nascent species before any form of postmating isolation
has evolved.
|