Download Functional Morphology of the Male Genitalia of Four Species of

MORPHOLOGY, HISTOLOGY, AND FINE STRUCTURE
Functional Morphology of the Male Genitalia of Four Species of
Drosophila: Failure to Confirm Both Lock and Key and
Male-Female Conflict Predictions
WILLIAM G. EBERHARD1
AND
NATALIA RAMIREZ
Escuela de Biologṍa, Universidad de Costa Rica, Ciudad Universitaria, Costa Rica
Ann. Entomol. Soc. Am. 97(5): 1007Ð1017 (2004)
ABSTRACT Several hypotheses attempt to explain why male genitalia typically diverge rapidly over
evolutionary time. Predictions of the lock and key and the sexually antagonistic coevolution hypotheses were tested by studying the functional morphology of several male genitalic traits by freezing
copulating pairs of four species of Drosophila: saltans, willistoni, melanogaster, and malerkotliana.
Contrary to the predictions of the hypotheses, there were no species-speciÞc differences in female
morphology that corresponded to interspeciÞc differences in the morphology of the male surstylus
(which pressed and perhaps spread the distal tip of the oviscape), the epandrium (which grasped the
external surface of the extended oviscape), and the aedeagus and paraphyses (which, in willistoni,
clamped the oviscape). Antagonistic coevolution could possibly explain the diverse male genitalic
morphology if female resistance behavior rather than morphology has coevolved with male morphology, but there are reasons doubt this explanation.
KEY WORDS genitalia, cryptic female choice, sexually antagonistic coevolution, lock and key
MALE GENITALIA OFTEN DIVERGE relatively rapidly compared with other body parts, and there are several
hypotheses that attempt to explain this pattern (Eberhard 1985, Shapiro and Porter 1989, Alexander et al.
1997, Simmons 2001). The lock and key and sexually
antagonistic coevolution hypotheses predict that male
and female morphology will often coevolve. The lock
and key hypothesis (Shapiro and Porter 1989) proposes that female genitalic structures are in effect
complex mechanical locks, into which only males of
same species can Þt. The sexually antagonistic coevolution hypothesis (Alexander et al. 1997, Holland
and Rice 1998, Chapman et al. 2003) proposes that
males and females are engaged in coevolutionary arms
races over control of copulation and insemination.
Female genitalic structures, in particular those that
interact mechanically with species-speciÞc male
structures, are supposed to constitute antimale devices that impede the manipulative action of male
genitalia. Such female defenses result in selection on
males favoring further modiÞcations to overcome
them. In contrast, a third hypothesis, that male genitalia are under sexual selection by cryptic female
choice (Eberhard 1985), does not predict that female
morphology will necessarily change when that of the
male changes. It is compatible with both female morphological coevolution with male structures, and lack
of such evolution.
1 Smithsonian Tropical Research Institute (e-mail: archisepsis@
biologia.ucr.ac.cr).
Male and female morphological coevolution has
been documented recently in Gerris water striders
(Arnqvist and Rowe 1995, 2002a,b), but the lack of
species-speciÞc female adjustments to males in several
earlier studies of lock and key (Shapiro and Porter
1989) renders the generality of such coevolution uncertain, making further studies desirable. The large
genus Drosophila offers advantages for study of interspeciÞc differences in genitalia. The probable relationships between different species groups are well
studied (e.g., Pitnick et al. 1999), and male genitalia
are generally complex and species-speciÞc. One previous study found a relationship between the form of
the ventral process of the male epandrium and the
degree of overlap between the femaleÕs eighth and
ninth tergites, where the ventral process rests during
copulation, in Drosophila melanogaster Meigen and
three closely related species; morphological differences were greater between species that showed the
greatest mating asymmetries (Robertson 1988). Another study documented male-female coevolution,
but between different, internal structures: sperm size
and internal female morphology (seminal receptacle)
(Pitnick et al. 1999). The current study addresses the
external mechanical genitalic mesh between the genitalia of males and females in four species, D. melanogaster, Drosophila saltans Sturtevant, Drosophila
malerkrotliana Parshad & Paika, and Drosophila willistoni Sturtevant. All are in the subgenus Sophophora;
D. melanogaster and D. malerkotliana are in different
subgroups of the melanogaster species group, whereas
0013-8746/04/1007Ð1017$04.00/0 䉷 2004 Entomological Society of America
1008
ANNALS OF THE ENTOMOLOGICAL SOCIETY OF AMERICA
Vol. 97, no. 5
Fig. 1. Posterior views of the resting genitalia of noncopulating males of D. saltans, willistoni, melanogaster, and malerkotliana.
the other two belong to the saltans and willistoni
species groups.
Materials and Methods
Copulating pairs of virgin ßies were frozen in liquid
nitrogen and immediately transferred to an open container of liquid N before they thawed. This container
was placed in a ⫺20⬚C freezer, where the N was
allowed to boil off, and where the frozen pairs were
then ßooded with absolute ethanol at ⫺20⬚C. The
pairs were thus not allowed to thaw before being Þxed.
After a week in ethanol at ⫺20⬚C, when their frozen
tissues had presumably become Þxed, the pairs were
brought to room temperature and placed in 80% eth-
anol. This treatment guaranteed that the ßiesÕ positions at the moment they were frozen were faithfully
preserved (Huber 1993, Eberhard and Pereira 1996).
Samples of ⬎10 noncopulating ßies of each sex were
prepared the same way, or by direct Þxation in 80%
ethanol.
The genitalia of males frozen in copula were exposed by carefully dissecting away the female in three
to seven pairs of each species. Similar numbers of pairs
(four to six) were left intact, except in D. melanogaster,
in which 14 intact pairs were examined. Specimens
were processed for study with the scanning electron
microscope by sublimation drying and coating with
gold. In the descriptions of copulating ßies, orientations refer to the position of the female. Flies were
September 2004
EBERHARD AND RAMIREZ: FUNCTIONAL MORPHOLOGY Drosophila GENITALIA
1009
Fig. 2. Posterior close-up views of genitalia of noncopulating males of D. saltans (resting), willistoni (resting), melanogaster (directed posteriorly), and malerkotliana.
identiÞed by D. Grimaldi (American Museum of Natural History, New York).
Results
The positions of the genitalia of noncopulating
males varied. In the “resting” position, the aedeagus
tip was directed dorsally in D. melanogaster but
posteriorly in the other species (Fig. 1). Positions of
the epandria varied intraspeciÞcally; they were
sometimes relatively spread (e.g., D. saltans in Fig. 1)
and sometimes more closed (e.g., D. willistoni in
Fig. 1). In some D. melanogaster, the aedeagus tip
was lowered and directed posteriorly (Fig. 2). In all
but D. malerkotliana the phallotreme was visible as a
curved slit just anterior to the tip of the aedeagus
(Figs. 1 and 2).
Some aspects of copulation did not vary intraspeciÞcally and were similar in all four species. The fe-
male oviscape was extended distally, a position also
seen occasionally in noncopulating females killed in
ethanol (Figs. 3 and 4) (in D. melanogaster extension
of the oviscape, which was hidden during copulation,
was conÞrmed by dissections). The epandria of the
male were ßexed medially (Fig. 5), and their distal
surfaces pressed against the membranous sides of the
extended oviscape (Figs. 6 and 7) (this detail was
conÞrmed in D. melanogaster by dissections). The
male surstyli, whose species-speciÞc arrays of stout
setae (Fig. 8) were directed medially or posteriorly
when at rest (Figs. 1 and 2), were ßexed medially
⬇180⬚C, so that their stout setae were directed
more or less dorsally. In intact pairs, the stout setae
were nearly (D. saltans in Figs. 6, 7, and 9) or completely hidden from view (D. willistoni, D. malerkotliana in Figs. 6 and 9). In D. willistoni and D. saltans
(and possibly the other species), these setae pressed
Fig. 3. Ventral views of exerted oviscapes of D. saltans, willistoni, melanogaster, and malerkotliana.
Fig. 4. Lateral views of exerted oviscapes of D. saltans, willistoni, melanogaster, and malerkotliana.
September 2004
EBERHARD AND RAMIREZ: FUNCTIONAL MORPHOLOGY Drosophila GENITALIA
1011
Fig. 5. Ventral views showing spread oviscape valves of females that were frozen during or immediately after copulation
ended in saltans, willistoni, and malerkotliana (compare with Fig. 2).
on or near the distal tips of the oviscape valves (Figs.
6 and 9).
There were also differences between species. In
D. saltans, an aedeagal process emerged from the
tip of the ejaculatory duct in two males killed
during copulation (Fig. 10), but it was not visible in
Þve other copulating males and or in noncopulating
males (Figs. 1 and 2). No similar aedeagal processes
were seen in the other species. The curved plates of
the aedeagus of D. malerkotliana (Fig. 9) were unique
in being realigned during copulation to form a pair
of tubes, one of which apparently contained the
phallotreme (Fig. 10). In D. melanogaster, the tips
of the male and female abdomens pressed together
tightly and hid the oviscape and the positions of
the male genitalia relative to it (Fig. 11) (two intact
pairs of D. willistoni were also tight together). The
distal portions of the relatively prominent ventral
lobes of the epandrium in D. melanogaster and
D. malerkotliana meshed with the female in different
ways. The medial surface of the ventral lobe of the
epandrium of D. malerkotliana pressed against the
exposed, membranous side of the female oviscape
(Fig. 7). In D. melanogaster, in contrast, the distal
portion of the ventral lobe of the epandrium was
hidden from view under the rear margin of female
tergite 8 (Figs. 7 and 11).
In two intact pairs of D. saltans, the edge of the
surstylus and its stout setae pressed against a process
of the paraphysis, apparently trapping the tip of the
femaleÕs genital valve between them (Fig. 6). The
processes of the paraphysis seemed to cause the tips
of femaleÕs genital valves to diverge from each other.
In one pair of D. saltans, however, these processes of
the paraphyses were not visible (Fig. 9). In D. willistoni, the dorsally curved, sharply pointed paraphyses
pressed on the outer, dorsal wall of the oviscape
(Fig. 11), evidently trapping the wall in the pincer
formed by these points and the sharp ventral tips of the
processes of the aedeagus and the aedeagal sheath
(Fig. 11).
1012
ANNALS OF THE ENTOMOLOGICAL SOCIETY OF AMERICA
Vol. 97, no. 5
Fig. 6. Ventral views of copulating pairs (male above, female below) of D. saltans, willistoni, melanogaster, and malerkotliana, showing grasp with epandria, male cerci, and the medially ßexed sustyli of saltans and willistoni.
Discussion
In all four species, the male surstyli ßexed medially
⬇180⬚ during copulation. Their stout setae, whose
numbers, form, and distribution differ both among the
species of this study (Fig. 8) and in other Drosophila
(e.g., Pitnick and Heed 1994), apparently pressed
against the femaleÕs oviscape, probably on or near the
tips of the oviscape valves. They seem to serve to
clamp the oviscape and also may spread the oviscape
valves apart (Fig. 5), perhaps to allow intromission.
Contrary to the predictions of mechanical lock and
key and male-female conßict of interests hypotheses,
the tips of the oviscape and of the oviscape valves did
not show species-speciÞc modiÞcations that corresponded to differences in the setae of the male surstyli
(Fig. 3).
The male-female conßict hypothesis might escape
these criticisms if female genitalic defenses against
males involve behavior of the oviscape, rather than its
morphology (Eberhard and Pereira 1996). This possibility cannot be ruled out deÞnitively because we did
not observe the behavior of the oviscape. Nevertheless, it seems unlikely, because it is difÞcult to imagine
what oviscape behavior could have selected for the
observed differences in surstylus design.
In D. willistoni, the paraphyses press on the outer
surface of the femaleÕs oviscape and act as a clamp in
opposition to ventral processes on the aedeagus and
the aedeagus sheath, which press on the inner surface
of the vagina at nearly the same point (Fig. 11). No
grasping device of this sort occurred in the other
species. There was no special female morphological
Fig. 7. Lateral and ventro-lateral views of copulating pairs (male on right, female on left), showing mesh of male
epandrium and surstyli with the female oviscape in D. saltans, willistoni, melanogaster, and malerkotliana.
Fig. 8. Surstyli in noncopulating males of D. saltans, willistoni, melanogaster, and malerkotliana.
1014
ANNALS OF THE ENTOMOLOGICAL SOCIETY OF AMERICA
Vol. 97, no. 5
Fig. 9. Close-up views of male genitalia and the distal portion of female oviscape during copulation (male above, female
below) in D. saltans, willistoni, and malerkotliana.
courterpart to this male grasping mechanism on the
oviscape of female D. willistoni. There are no sclerites
or obvious irregularities on the outer surface of the
female in this region that could mesh with the male
structures in a lock and key manner or that could
impede male grasping as expected by the male-female
conßict hypothesis. Female D. willistoni shake energetically from side to side during copulation more
than the other species (approximately one-quarter of
the 8.1 ⫾ 2.0-min copulation) (mean ⫾ SD, N ⫽ 30)
(N. Ramirez, unpublished data); but the males were
never dislodged, even though they always released
their hold with all their legs, folding them immobile
and hanging on only with their genitalia (N. Ramirez,
unpublished data). Perhaps the female shaking, which
seems not to serve to dislodge the male, serves instead
to allow the female to sense or test the maleÕs genitalic
grip.
The positions of our copulating pairs of D. melanogaster were similar to those observed by Robertson
(1988) except that both the male epandrium and the
female tergite 8 protruded less. Perhaps this difference
was due to our Þxing pairs before they thawed. Robertson argued that the mesh between the ventral lobe
of the epandrium and the overlap in female tergites
eight and nine represented a lock and key mechanism
of species isolation. But neither his data nor ours
support this idea convincingly: 1) the female tergites
are joined by a ßexible membrane, and thus the
amount of overlap is not rigid like a mechanical lock
but varies with the degree of extension of the femaleÕs
abdomen; and 2) there is no physical barrier in the
September 2004
EBERHARD AND RAMIREZ: FUNCTIONAL MORPHOLOGY Drosophila GENITALIA
1015
Fig. 10. Genitalia during copulation (female dissected away) of males of D. saltans, willistoni, melanogaster, and malerkotliana.
female that would prevent a cross-speciÞc male from
seizing her oviscape with his ventral lobes. Data from
other Drosophila suggest instead a role for genitalic
stimulation of the female. Female D. simulans respond
differently when coupled with males of the probable
derived sister species D. mauritiana (an island endemic) in which the ventral processes of the epandrium have a different form (Coyne 1993), suggesting
that females are able to sense differences in the form
of the epandrium. The 19 quantitative trait loci that
contribute to the genitalic difference between these
species nearly all act in the same direction (Zeng et al.
2000), indicating that the differences are the result of
selection; this suggests, in turn, that females can apparently distinguish very small differences (perhaps
on the order of 1/19 of the difference between these
species).
We are unable to assign possible functions to many
portions of the male genitalia that enter the female
rather than remaining on her outer surface during
copulation, including the aedeagus and its processes
(in species other than D. willistoni), the gonopods,
and the paraphyses (in species other than D. willistoni). Some of these structures, such as the aedeagus,
are often species-speciÞc in form in Drosophila. Study
of sections of copulating pairs would probably help
clarify their functions.
It is important to note several limitations of this
study. Male and female genitalia may change positions
during copulation, as suggested by the differences in
the position of the paraphysis and aedeagal processes
in D. saltans (see Lachmann 1996 and Eberhard and
Huber 1998 for changes in other acalyptrate ßies).
Position changes may be especially likely in the very
early stages of coupling. Our small sample sizes and
lack of information on the stage of copulation of each
pair did not allow us to examine these possibilities.
Rhythmic movements, such as apparent male scraping
on the sides of the oviscape with his epandrium in D.
saltans (N. Ramirez, unpublished data) and possible
female movements, also were not covered by this
study (see Eberhard 2001a,b for rhythmic movements
of male genitalia in other ßies).
1016
ANNALS OF THE ENTOMOLOGICAL SOCIETY OF AMERICA
Vol. 97, no. 5
Fig. 11. Top, lateral view of male epandrium (right) and its mesh with the female in a copulating pair (left; male is on
the right) of D. melanogaster; the ventral process of epandrium in inserted beneath the female tergite 8. Bottom, lateral view
of the close apposition of processes of the maleÕs aedeagus and aedeagal sheath with the tips of his paraphyses during
copulation in D. willistoni (female dissected away) (right) and of one male paraphysis pressed against the external dorsal
surface the femaleÕs oviscape during copulation (left).
Acknowledgments
We thank David Grimaldi for kindly identifying ßies and
for help with genitalic terminology, Jerry Coyne (University
of Chicago) for helpful comments, and Maribelle Vargas
(Universidad de Costa Rica) for outstanding technical assistance. Financial support was provided by the Smithsonian
Tropical Research Institute.
References Cited
Alexander, R. D., D. C. Marshall, and J. R. Cooley. 1997.
Evolutionary perspectives on insect mating, pp. 4 Ð31. In
J. Choe and B. Crespi [eds.], The evolution of mating
systems in insects and arachnids. Cambridge University
Press, Cambridge, United Kingdom.
Arnqvist, G., and L. Rowe. 1995. Sexual conßict and arms
races between the sexes: a morphological adaptation for
control of mating in a female insect. Proc. R. Soc. Lond
B. 261: 123Ð127.
Arnqvist, G., and L. Rowe. 2002a. Antagonistic evolution
between the sexes in a group of insects. Nature (Lond.)
415: 787Ð789.
Arnqvist, G., and L. Rowe. 2002b. Correlated evolution of
male and female morphologies in water striders. Evolution 56: 936 Ð947.
Chapman, T., G. Arnqvist, J. Bangham, and L. Rowe. 2003.
Sexual conßict. Trends Ecol. Evol. 18: 41Ð 47.
Coyne, J. A. 1993. The genetics of an isolating mechanism
between two sibling species of Drosophila. Evolution 47:
778 Ð788.
Eberhard, W. G. 1985. Sexual selection and animal genitalia. Harvard University Press, Cambridge, MA.
Eberhard, W. G. 2001a. Species-speciÞc genitalic copulatory courtship in sepsid ßies (Diptera, Sepsidae, Microsepsis). Evolution 55: 93Ð102.
Eberhard, W. G. 2001b. Genitalic behavior during copulation in Hybosciara gigantea (Diptera: Sciaridae) and the
evolution of species-speciÞc genitalia. J. Kans. Entomol.
Soc. 74: 1Ð9.
Eberhard, W. G., and B. A. Huber. 1998. Copulation and
sperm transfer in Archisepsis ßies (Diptera: Sepsidae) and
the evolution of their intromittent genitalia. Studia Dipt.
5: 217Ð248.
Eberhard, W. G., and F. Pereira. 1996. Functional morphology of male genitalic surstyli in the dungßies Achisepsis
September 2004
EBERHARD AND RAMIREZ: FUNCTIONAL MORPHOLOGY Drosophila GENITALIA
diversiformis and A. ecalcarata (Diptera: Sepsidae). J.
Kans. Entomol. Soc. 69: 43Ð 60.
Holland, B., and W. R. Rice. 1998. Chase-away selection:
antagonistic seduction vs. resistance. Evolution 52: 1Ð7.
Huber, B. A. 1993. Genital mechanics and sexual selection
in the spider Nesticus cellulanus (Araneae: Nesticidae).
Can. J. Zool. 71: 2437Ð2447.
Lachmann, A. D. 1996. Copulation and engagement of male
and female genitalia in Þve Coproica Rondani species
(Diptera: Sphaeroceridae). Ann. Entomol. Soc. Am. 89:
759 Ð769.
Pitnick, S., and W. B. Heed. 1994. New species of cactusbreeding Drosophila (Diptera: Drosophilidae) in the nannoptera species group. Ann. Entomol. Soc. Am. 87: 307Ð
310.
Pitnick, S., T. Markow, and G. S. Spicer. 1999. Evolution of
multiple kinds of female sperm-storage organs in Drosophila. Evolution 53: 1804 Ð1822.
1017
Robertson, H. M. 1988. Mating asymmetries and phylogeny
in the Drosophila melanogaster species complex. Pac. Sci.
42: 72Ð 80.
Shapiro, A. M., and A. H. Porter. 1989. The lock-and-key
hypothesis: evolutionary and biosystematic interpretation of insect genitalia. Annu. Rev. Entomol. 34:231Ð245.
Simmons, L. 2001. Sperm competition and its evolutionary
consequences in insects. Princeton University Press,
Princeton, NJ.
Zeng, Z.-B., J. Lin, L. F. Stam, C.-H. Kao, J. M Mercer, and
C. C. Laurie. 2000. Genetic architecture of a morphological shape difference between two Drosophila species.
Genetics 154: 299 Ð310.
Received 31 July 2003; accepted 3 March 2004.