News & Views | Published:

Evolutionary biology

Caught right-handed

Are two penises better than one? Not so, implies a study of doubly endowed earwigs. An ancestral behavioural preference for the right penis might have facilitated the loss of the left in species that arose later.

Human males may sometimes wonder about the size of their penis, but they rarely fret about which one to use. Not so for many arthropods, among them fairy shrimp1, dragonflies2 and spiders3, some of which face a delicate choice before each tryst: “Left or right tonight?” Doubly endowed lizards4 and snakes5 seem ambivalent, although they tend to alternate between the right and left of their two members. Males of two theridiid spider genera (Tidarren and Echinotheridion) have a particularly radical solution to their quandary: they voluntarily rip off a palp at random and eat it, leaving only one behind.

Writing in the Journal of Morphology6, Yoshitaka Kamimura describes his investigations of the private life of the doubly endowed male of the earwig species Labidura riparia (Fig. 1a). He shows that this earwig has a strong preference for its right penis: nearly 90% of field-collected and laboratory-reared males hold their intromittent organs in the 'right-ready' state (right side extended backwards, ready to mate; Fig. 1b) when not mating, as well as when in flagrante delicto. Curiously, the earwig's two penises are morphologically indistinguishable and fully functional. They connect to equivalent testes, and individuals with an injured or experimentally ablated right penis readily revert to using the left one. This right-ready asymmetry is therefore largely — if not entirely — behavioural.

Figure 1: Earwig penis number, orientation and evolution.

a, A male and female Labidura riparia, the object of Kamimura's studies6, showing the pronounced cerci for which earwigs are best known. b, The genitalia of the doubly endowed male L. riparia earwig showing two different penis orientations while at rest (not mating): 'right ready' (left penis folded anteriorly) and 'left-ready' (right penis folded anteriorly). Nine out of ten L. riparia prefer right-ready. (Figure modified from ref. 6.) c, Tentative phylogenetic relations among the families of the earwig suborder Forficulina7,8, with asterisks indicating nodes with weak or inconsistent support. The inconsistencies are resolved here following ref. 7. Number of penises, and 'ready' and 'used' side is shown (0.5R = right-side penis ready or used in 50% of cases, and so on; data from ref. 6). The coloured lines represent different combinations of behaviours as indicated by the data above each line.

On its own, right-handed penis preference in doubly endowed earwigs would qualify only as amusing natural history. But two other observations noted by Kamimura6 increase its significance. The first is that not all earwig taxa have two penises. Some have only one, and it is indeed on the right side. The second is that not all earwigs with two penises have a right-side preference. Some (such as certain species of the families Diplatyidae and Anisolabididae) use the right or left indifferently.

It is the evolutionary relations among these earwigs6,7,8 that elevates a genitalic curiosity to a higher plane (Fig. 1c). First, the earliest two lineages of the earwig suborder Forficulina, the families Diplatyidae and Pygidicranidae, have two penises, both of which point forward in the 'not-ready' position when not mating. Second, the three families Apachyidae, Anisolabididae and Labiduridae, in which doubly endowed males always hold one penis, either left or right, in the 'ready' position when at rest, are closely related, and lie evolutionarily between the primitive doubly endowed earwigs and the more recently evolved singly endowed earwigs. Third, these later earwigs (the three families Forficulidae, Spongiphoridae and Chelisochidae) that possess only one penis — the right — form a monophyletic group (they are all descendants of a common ancestor) derived from doubly endowed earwigs. Although morphological7 and molecular8 data yield somewhat different placements, the family Labiduridae, to which the right-leaning L. riparia belongs, is closely related to this common ancestor of single-penis earwigs.

Kamimura suggests6 an intriguing evolutionary scenario that stems from these phylogenetic relations. Male earwigs evolved from a primitive state with both penises held in the 'not-ready' orientation when not mating, first through a stage where they always held one penis — either the right or left at random — in the 'ready' orientation. The next evolutionary step was males that still possessed two morphologically indistinguishable penises, but which preferentially held the right in the 'ready' orientation. Finally, the less-preferred (left) penis disappeared altogether, leaving only traces of a closed, non-functional ejaculatory duct. Thus, a purely behavioural asymmetry might have facilitated the evolution of a fully-blown morphological asymmetry.

The current view of the Forficulina phylogeny is in places tentative (Fig. 1c), as is the exact placing of L. riparia in its family, so the possibility that right-handed behaviour and left-side penis loss evolved independently cannot be discounted. Nevertheless, the observations seem to support a still controversial and underappreciated mode of evolution, known as the Baldwin effect, that was advanced more than a century ago9,10. This theory of 'organic selection', named after the experimental psychologist James Mark Baldwin, emphasized the impact of behaviour on evolution. Baldwin argued, as had Jean-Baptiste Lamarck nearly a century before11, that new behaviours, learnt or otherwise, can expose an organism to novel conditions of growth to which it may then respond. But unlike Lamarck, Baldwin observed that altered behaviours can also yield altered morphologies through the inherent variation in possible outcomes (the 'plasticity') of an organism's development. If these behaviour-induced morphological differences affect performance positively, natural selection should favour genetic variants that increase or fix the plastic response.

Learnt handedness, coupled with developmental plasticity, provides an attractive mechanism for inducing right–left morphological asymmetries: excess use of one side of the body can induce its overdevelopment, as it does in the upper arm bones of professional tennis players12. The right-handed L. riparia seems to be one of those rare missing links where a right-side behavioural preference has been 'caught' genetically before any morphological differentiation. (As Kamimura shows6, L. riparia's right-side preference develops within four days of adult emergence, before any mating experience, so it involves no learning.)

The progression of the story would be deliciously complete if we also knew that individual males of the ambidextrous families Apachyidae and Anisolabididae actually learn to use one side preferentially with increased mating experience. Then the learnt handedness of an individual earwig would clearly have preceded evolutionarily the genetically captured handedness seen in L. riparia.

But a puzzling question remains: why a right-side preference? A minor anatomical asymmetry in the female L. riparia reproductive tract might favour right-handed males6, rather as it does in birds. Most male birds lack an intromittent organ, and mate by pressing together their posterior opening, known as a cloaca, with that of the female. In two species, this 'cloacal kiss' comes most often from the left side13,14, probably because the female's solitary ovary, and the opening to the oviduct, lie to the left15. But Kamimura shows that right-handed and left-handed L. riparia have equal mating success6.

Alternatively, right-side penis preference might somehow be connected to the unknown factor that favours a stronger curvature of the right cerci in some related wingless earwigs, for example Anisolabis and Euborellia16. But the cerci in L. riparia are symmetrical (Fig. 1a), so the puzzle remains.

Clearly, the earwig penis system warrants more study. It could become a textbook example of how a possibly learnt lateralized behaviour bred a lateralized morphology evolutionarily. It already qualifies as a fine example of a phenotype-precedes-genotype mode of evolution because the right-ready and left-ready penis variants, which are equally common in evolutionary intermediates (Fig. 1c), and therefore probably not heritable17, clearly existed before the genetically captured right-ready phenotype seen in L. riparia. Who would have ever thought you could learn so much from earwig penises?


  1. 1

    Rogers, D. C. & Fugate, M. West. N. Am. Nat. 61, 11–18 (2001).

  2. 2

    Bechly, G., Brauckmann, C., Zessin, W. & Groning, E. J. Zool. Syst. Evol. Res. 39, 209–226 (2001).

  3. 3

    Knoflach, B. & van Harten, A. J. Nat. Hist. 40, 1483–1616 (2006).

  4. 4

    Tokarz, R. R. & Slowinski, J. B. Anim. Behav. 40, 374–379 (1990).

  5. 5

    Shine, R., Olsson, M. M., LeMaster, M. P., Moore, I. T. & Mason, R. T. Behav. Ecol. 11, 411–415 (2000).

  6. 6

    Kamimura, Y. J. Morphol. 267, 1381–1389 (2006).

  7. 7

    Haas, F. & Kukalova-Peck, J. Eur. J. Entomol. 98, 445–509 (2001).

  8. 8

    Jarvis, K. J., Haas, F. & Whiting, M. F. Syst. Entomol. 30, 442–453 (2005).

  9. 9

    Baldwin, J. M. Am. Nat. 30, 441–451 (1896).

  10. 10

    Baldwin, J. M. Am. Nat. 30, 536–553 (1896).

  11. 11

    Lamarck, J. B. Philosophie Zoologique (Dentu, Paris, 1809).

  12. 12

    Trinkaus, E., Churchill, S. E. & Ruff, C. B. Am. J. Phys. Anthropol. 93, 1–34 (1994).

  13. 13

    Nyland, K. B., Lombardo, M. P. & Thorpe, P. A. Wilson Bull. 115, 470–473 (2003).

  14. 14

    Petersen, A. D., Lombardo, M. P. & Power, H. W. Anim. Behav. 62, 739–741 (2001).

  15. 15

    King, A. S. in Form and Function in Birds Vol. 2 (eds King, A. S. & McLelland, J.) 107–148 (Academic, New York, 1981).

  16. 16

    White, R. E., Borror, D. J. & Peterson, R. T. A Field Guide to Insects: America North of Mexico (Houghton Mifflin, Boston, 1998).

  17. 17

    Palmer, A. R. Science 306, 828–833 (2004).

Download references

Author information

Rights and permissions

Reprints and Permissions

About this article

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.