The evolution of segmentation of centipede trunk and appendages

The segmentation of centipedes is interpreted in the light of a biphasic model of segmentation (holomeric plus meromeric). The mid‐body anomaly (e.g. in the alternating short and long terga, or in the sequence of segments with and without spiracles) is regarded as due to an early patterning of the embryo, occurring before the onset of meromeric segmentation and affecting a level within the fourth eosegment of the trunk. Comparisons with the Diplopoda suggest that genital structures such as millipede gonopods did probably develop originally at this spot, whose position remained marked even after the transition from a putatively progoneate to the current opisthogoneate condition of centipedes, perhaps following gene duplication and divergence of expression patterns of the paralogues. A new lower limit for the number of leg‐bearing segments [27, in a male specimen of Schendylops oligopus (Pereira, Minelli & Barbieri,1995)] is established for Geophilomorpha. Coevolutionary trends involving the segmentation of the trunk, the segmentation of the appendages (especially the antennae), the postembryonic developmental schedule and the presence or absence of regeneration ability supports a recent view of the appendages as evolutionarily divergent duplicates of the main body axis.