Woodpile wanderers and cohort-splitting

The value of dead wood as a habitat for invertebrates is well known (one of my 'static pages' covers this), so yesterday, while rummaging through our garden woodpile, I found what looked initially like a centipede about 20mm long, I felt compelled to put fingers to keyboard...

The first thing that leapt out was that it wasn't a centipede - it had two pairs of legs per segment (rather than one) and was therefore a millipede. The dorsal plates are splayed out sideways (the shape helps it push through the soil) and are coarsely sculptured along the back, and the legs and antennae are longer than in many other millipedes. Side on, it's clear why it's called 'flat-backed'.

Flat-backed Millipede

Side view showing the flat back

It's a common species, and is indeed known as the Common Flat-Backed Millipede (Polydesmus angustus). Found on roots and fruits (especially strawberries), it's often associated with all sorts of decaying plant material; dead leaves, compost, dead wood and so on. It's a native of north-west Europe, but has also been accidentally introduced to the south-east USA.

Although a common species, it does have some interesting life-cycle characteristics. Mating occurs from late spring into summer, then again from late summer to mid-autumn. Males usually mate only once, but females store the sperm from a single mating to produce several batches of eggs. Research by David (2009) showed that females born between May and August have a one-year life cycle while those born from late August onwards have a two-year life cycle (a strategy known as 'cohort-splitting', hence the title). A third type of life cycle ('interseasonal iteroparity' where iteroparity simply means reproducing more than once in their lifetime) was seen in a few females born late in the season.

Results from looking at the reproduction of individual females indicated that only annual females produced an appreciable proportion of biennial offspring from late August onwards; this means that life-cycle duration can't be genetically determined - cohort-splitting must therefore be driven non-genetically, supporting previous research by David et al. (2003) showing the effect of photoperiod (day length) on the life cycle of this species. As individual females reproduce for about two months, this automatically leads to a cyclical pattern of life-cycle duration (annual/biennial/annual) in the long-term progeny of any female.

Personally I find this fascinating, not least as an example of how a common species in an everyday habitat can surprise us with unuusal aspects of its biology or ecology. Wonderful! 

The telson and other posterior segments AKA 'the end'

Reference

David, J.-F. (2009). Female reproductive patterns in the millipede Polydesmus angustus (Diplopoda: Polydesmidae) and their significance for cohort-splitting. European Journal of Entomology 106(2): 211–216.

David, J.F., Geoffroy, J.J. & M. L. Célérier, M.L. (2003). First evidence for photoperiodic regulations of the life cycle in a millipede species, Polydesmus angustus (Diplopoda: Polydesmidae). Journal of Zoology (London) 260: 111–116.