The genitals of male insects bear a huge number of intricate bits and pieces. The genitalic terms for orthopterans alone (crickets and allies) include phalli, epiprocts, paraprocts, cerci, gonotremes and even titillators (from the Latin titallo, ‘tickle’).
But why have the male's genitals evolved to be so complex? Two answers have been proposed. First, the species-diagnostic nature of these male characteristics suggests that species differences in genitals may have evolved as barriers to insemination, preventing the production of low-quality hybrid offspring. So, in zones of species overlap, the female genital ‘locks’ of each species diverged, imposing Darwinian selection among conspecific males for a proper fit of each genital ‘key’. The second answer is that genital structures diverged through the sexual selection that occurs after insemination for any device that increases fertilization success. Possible examples are claspers, which hold the struggling female until insemination occurs1, or titillating devices that deliver ‘internal courtship signals’ to the female. Sperm from males with the best signals are used in preference to stored sperm from the females' previous mates, and such copulatory discrimination is favoured in females because they pass on the advantageous genital trait to their sons2.
On page 784 of this issue, Göran Arnqvist3 provides a rigorous test of the lock-and-key and sexual-selection hypotheses. The rigour is in his methods. First, he measures structural divergence geometrically, thereby overcoming the inevitable subjectivity problems that occur when researchers assess morphological complexity (see, for example, ref. 4). Then he compares these divergence measures for matched pairs of insect groups that share a common ancestor but differ in the pattern of female mating. One group contains species reported to have females that mate only once (monandrous species), whereas the other has multiply-mating females (polyandrous species). If post-insemination selection is important, the genitals should be more divergent in the polyandrous groups because competition between the ejaculates of rivals can occur only when females take on additional partners. Conversely, the lock-and-key hypothesis predicts greater genital divergence in monandrous species, because of the potentially large cost to the female if her single copulation is with the wrong species.
Arnqvist's results overwhelmingly supported post-insemination selection. In 18 of 19 pairs sampled from mayflies, flies, beetles and lepidopterans, the shape of male genitals from species in the polyandrous groups has diverged more than in single-mating species. Moreover, this divergence was found only in genitals, and not in other body parts such as legs.
Could any other selective forces explain the differences in genital divergence between the monandrous and polyandrous groups? After all, the two groups do differ with respect to other aspects of their life history — variables that are not controlled for in a comparative study such as Arnqvist's. For example, compared with polyandrous species, monandrous females tend to have a lower fecundity5 and they copulate immediately after emerging as adults6. Perhaps complex gadgetry is not needed to engage the pliant reproductive parts of a newly emerged female whose integument has not yet hardened. However, hypotheses such as this are unlikely to provide general explanations for the patterns observed by Arnqvist.
One hypothesis that may contribute to these patterns involves a combination of the two models considered by Arnqvist. Hybridization costs do exist in species with polyandrous females, and some post-insemination selection may be a response to these costs. In a beetle, lacewing, fruitfly, grasshopper and cricket (7,9 and refs therein) there is clearly no lock-and-key isolating mechanism — sperm from the wrong species (or, in one case, subspecies) can find its way into the sperm storage organs of females, and even successfully fertilize eggs. But this occurs only when it is the sole ejaculate present, and such heterospecific sperm usually lose the fertilization battle when in competition with sperm from the same species as the female. There is some evidence that this outcome is partly due to the ability of the female's reproductive tract to recognize and favour sperm from her own species. Such cryptic communication of species identity is not well understood9.
So here's the alternative hypothesis. Could male genital modifications in some polyandrous species include devices that signal to the female that conspecific sperm is being delivered (similar to the way in which genital structures are thought to signal the quality of a conspecific male2)? This idea is worth investigating, even though it is not a general explanation. For example, it does not explain why divergent genitals exist when there is no threat of mis-mating — such as in isolated island species or in parasites that do not share hosts with related species2. To work out the importance of genital signalling (of any sort) on the evolution of genitalic complexity, we need many more experimental and observational studies on individual species10.
Finally, it is worth noting that Darwin, who was a very competent entomologist, was keenly aware of “the complex appendages at the apex of the abdomen in male insects”, structures that he thought could function as mate-holding devices1. He recognized that sexual selection could produce obscure genitalic traits as well as the more widely acclaimed sexual structures such as the peacock's tail. As Arnqvist concludes, the fact that male genitals are under sexual selection certainly blurs the traditional dichotomy11 between primary reproductive organs — those that deliver ejaculates — and the secondary sexual structures of males such as fancy tails. In considering the intricate genital appendages of male insects, Darwin had a similar take on this issue when he concluded that “it is scarcely possible to decide which ought to be called primary and which secondary”.
Darwin, C. The Descent of Man and Selection in Relation to Sex 2nd edn (John Murray, London, 1874).
Eberhard, W. G. Sexual Selection and Animal Genitalia (Harvard Univ. Press, Cambridge, MA, 1985).
Arnqvist, G. Nature 393, 784–786 (1998).
Proctor, H. C., Baker, R. L. & Gwynne, D. T. Can. J. Zool. 73, 2010–2020 (1995).
Ridley, M. Biol. Rev. 63, 509–549 (1988).
Ridley, M. Anim. Behav. 37, 535–545 (1989).
Albuquerque, G. S., Tauber, C. A. & Tauber, M. J. Evolution 50, 1598–1606 (1996).
Price, C. S. C. Nature 388, 663–666 (1997).
Howard, D. J. & Gregory, P. G. Phil. Trans. R. Soc. Lond. B 340, 231–236 (1993).
Eberhard, W. G. Am. Nat. 142, 564–571 (1993).
Hunter, J. Observations on Certain Parts of the Animal Oeconomy(London, 1786).
How to Know the Grasshoppers, Crickets, Cockroaches and Their Allies (Dover, New York, 1987).
The Male Genitalia of Orthopteroid Insects Smithsonian Misc. Collections 96, 1-107 (1937).
About this article
Inhibition of female mating receptivity by male-derived extracts in two Callosobruchus species: Consequences for interspecific mating
Journal of Insect Physiology (2010)
Haldane's rule in Carabus: interspecific mating between Carabus punctatoauratus and Carabus splendens using experimental tests and molecular markers
Entomologia Experimentalis et Applicata (2006)
Evolutionary Ecology (2005)
A comparative study of interspecies mating of Phratora vulgatissima and P. vitellinae using behavioural tests and molecular markers
Entomologia Experimentalis et Applicata (2004)