Natural History Museum of Denmark (Zoological Museum)

MYRIAPODA

Henrik Enghoff

Natural History Museum of Denmark (Zoological Museum)

2005

page and fig. references, “RFB”, refer to

Ruppert, Fox & Barnes: Invertebrate Zoology, 7th. ed. 2004.>

There are four classes of Myriapoda, viz. Chilopoda (centipedes, DA: Skolopendre), Diplopoda (millipedes, DA: Tusindben), Pauropoda (pauropods, DA: Pauropoder) and Symphyla (symphylids, DA: Symfyler).

THE POSSIBLE MONOPHYLY OF MYRIAPODA AND THE RELATIONSHIPS BETWEEN THE MYRIAPOD CLASSES

(Also read “Phylogeny of Tracheata” in RFB p. 718-720”)

Myriapod phylogeny is unsettled, but possibly the four myriapod classes plus the class Hexapoda constitute a monophyletic 'subphylum' Uniramia, also known as Atelocerata and Tracheata.

The presence of numerous legs, which has given the myriapods their name, is obviously a symplesiomorphy. Within the framework of a monophyletic Uniramia, several sister-group relationships have been proposed:

·  (Hexapoda) – (Myriapoda)

·  (Hexapoda) – (Symphyla)

·  (Hexapoda) – (Chilopoda)

·  (Chilopoda) - ((Hexapoda) – (Symphyla+Pauropoda+Diplopoda))

The following traits have been mentioned as possible synapomorphies for the four myriapod classes, i.e., as arguments for a monophyletic group Myriapoda (numbering continued from first part of compendium):

1.  No mandibular abductor muscle; abduction of mandibles effectuated indirectly by the movable anterior tentorium (Fig. 1)

2.  Median eyes never present (in Crustacea, the median eyes are represented by the nauplius eye, in Hexapoda, they are represented by the ocelli .

3.  No perforatorium in the spermatozoa (Fig.2: "1-laget acrosom" (one-layered acrosome), in contrast to the two-layered acrosome in "Insecta" which is considered to be the original state)

All these have been considered reductional characters and therefore of limited value. The absence of an abductor muscle may, however, be primary, i.e., a plesiomorphy, and the same may be true of the movability of the tentorium (Klass & Kristensen i Deuve ed., 2000). The presence of the anterior tentorium in itself appears as an apomorphy immediately below the myriapod level, i.e. as an apomorphy for Uniramia.

The characters invoked in favour of the hypothesis of a sister-group relationship between Symphyla and Hexapoda are possibly symplesiomorphies or convergencies:

4.  Three pairs of buccal appendages (mouthparts) (RFB: fig. 20-7B-C, p. 710). Ruppert & Barnes follow Manton in rejecting this similarity based on differences between the buccal appendages of symphylids and hexapods: this suggests that the presence of three pair of buccal appendages could be convergent.

5.  Second pair of maxillae forming a labium (‘lower lip’). This character, too, may be regarded as a convergence. The morphological interpretation of the ‘lower lip’ in Diplopoda and Pauropoda is still under debate (cf. below).

The basis of the hypothesis of a sister-group relationship between Chilopoda and Hexapoda is tenuous:

6.  In both groups the gonopores are located at the posterior end of the body, unlike the situation in Diplopoda, Pauropoda and Symphyla (‘Progoneata’, cf. below) where they are placed near the anterior end. Out-group comparison (with Crustacea and Chelicerata) is of no help here, but the posterior gonopore is most likely the original condition within Uniramia and thus cannot be used as an argument in favour of a Chilopoda-Hexapoda sister-group relationship.

There are several similarities between Hexapoda on the one hand, and Diplopoda+ Pauropoda+Symphyla on the other. They are all quite subtle, and none is persuasively synapomorphic. One example is:

7.  The eversible coxal sacs which are found in Symphyla (RFB: p. 711), certain Diplopoda (Penicillata, Colobognatha, Nematophora) and certain (‘primitive’) Hexapoda. The coxal sacs are probably symplesiomorphic and have become lost in Chilopoda, Pauropoda, many Diplopoda and most Hexapoda).

Wheeler et al. (1993) analysed arthropod relationships using one morphological and two molecular (DNA) data sets. Their analysis included Diplopoda and Chilopoda (but not Symphyla and Pauropoda) as well as numerous groups of Hexapoda, Crustacea and Chelicerata. Both molecular data seta indicated that Diplopoda and Chilopoda were more closely related to each than to any other the other included groups and thus support the hypothesis of myriapod monophyly. (The analysis of the morphological character set – a character set with several flaws – did not resolve the relationship between Diplopoda, Chilopoda and Hexapoda.)

It was suggested above that Chilopoda differ from the other myriapod classes. The latter, viz., Diplopoda, Pauropoda and Symphyla, together constitute the group Progoneata. This name (‘pro’: in front, ‘gon-’: having to do with sexual organs) refer to a possible synapomorphy between the three classes, viz.:

8.  The gonopore is placed anteriorly on the body (cf. the discussion under character 6).

Other possible synapomorphies between the three classes (i.e., autapomorphies for Progoneata) are:

9.  clypeus and labrum are fused (in Chilopoda and Hexapoda the labrum is separated from the clypeus by a suture and is ± movable. RFB: fig. 20-2C, p. 705, shows the clypeus in a chilopod; the labrum is not visible, but is placed in ’the black hole’ between clypeus and first maxilla.

10.  The body segments have ventral apodemes (Fig. 3). In Diplopoda and Pauropoda these are the tracheal apodemes (see below). In Symphyla there are no tracheae connected with the apodemes – the tracheae may be secondarily reduced like in most Pauropoda.

11.  Trichobothria basally swollen. Trichobothria are a special type of sensory hairs which are set in a complicated socket. Trichobothria occur in numerous groups of terrestrial arthropods (especially in arachnids, but unfortunately not in Chilopoda), but in Progoneata the trichobothria are characterised by having the base of the hair swollen (Fig. 4). In Symphyla there is one such pair of trichobothria in the posterior end of the body (RFB: fig. 20-7D, p. 710). In Pauropoda there are five pairs laterally on the body (RFB: fig. 20-13, p. 718). In Diplopoda trichobothria are known only from the Penicillata where there are 3 pairs on each side of the head. In spite of the different body parts where the trichobothria are situated, the characteristic structure of the progoneate trichobothria be regarded as apomorphic.

Within Progoneata, Symphyla are sister-group to Diplopoda + Pauropoda. Synapomorphies between Diplopoda and Pauropoda include:

12.  Immediately behind the mandibles there is a complex ‘lower lip’ (Fig. 5), the composition of which is controversial. Some regard it as a composite of the two pairs of maxillae, others maintain that it is formed exclusively by the first pair of maxillae, the second pair being entirely suppressed; at least part of the dorsal part of the second maxillary segment is represented by the ‘collum’ (RFB: fig. 20-9, p. 712, fig. 20-13, p. 718). Cf. character 5 above.

13.  In both groups, the spiracles are situated on the sterna, near the leg bases (RFB: fig. 20-10B, p. 713). Both groups have tracheal apodemes associated with the spiracles (cf. character 10). Tracheae have arisen several times independently in arthropods. The ventral system in Diplopoda + Pauropoda undoubtedly represents one ‘invention’ of tracheae, whereas it is more doubtful if the lateral system in non-scutigeromorph centipedes (RFB: fig. 20-2E, p. 705) is homologous with that of Diplopoda + Pauropoda. The tracheal system in Symphyla (RFB: fig. 20-7, p. 710) is even more dubious in this respect. The dorsal tracheal system of scutigeromorph centipedes (RFB: fig. 20-2B, p. 705) certainly represents an independent ‘invention’.

Based on the characters discussed above, the relationships of the myriapods can be illustrated as in the cladogram, Fig. 6.

Alternative relationships between the myriapod classes have been suggested. Fig. 2 thus suggests a sister-group relationship between Pauropoda and Chilopoda, based exclusively on the morphology of spermatozoa; this similarity is probably a symplesiomorphy. Symphyla + Chilopoda have also been regarded as sister-groups, mainly because both classes lack limbs on the last two body segments. Balanced against the autapomorphies for Progoneata suggested above, this character is, however, not convincing.

CLASS CHILOPODA – CENTIPEDES (DA.: SKOLOPENDRE)

RFB: 703-710.

The numbers of apomorphies in this section refer to Fig. 7.

Chilopoda autapomorphies:

1.  First pair of body limbs transformed into poisons fangs (‘forcipules) (RFB: fig. 20-2, A, C, p. 705).

2.  Second maxilla in the embryo provided with an ‘egg-tooth’.

3.  Nucleus of spermatozoon spiral-shaped

Systematic review of Chilopoda – centipedes

More than 3000 species of centipedes have been described. Thirty-two species have been recorded in Denmark.

Order Scutigeromorpha

Ca. 130 species from the warmer parts of all continents. Easy to recognize by the 15 pairs of extremely long, multi-segmented legs, and equally extremely long antennae. Composite eyes which may be inherited from the common ancestor of all centipedes although some structural details suggest that the eyes of scutigeromorphs may be derived from single eyes of the type found in Lithobiomorpha and Scolopendromorpha; in the latter case the secondarily composite eye in Scutigeromorpha is an autapomorphy. Certain autapomorphies include:

16.  The dorsal spiracles (RFB: fig. 20-2B, p.705)

17.  The strongly sub-segmented tarsi

Scutigera coleoptrata is common in southern Europe and has been found occasionally in houses in Denmark (introduced).

Order Lithobiomorpha

Ca. 1500 species from all parts of the World. Fifteen pairs of walking legs like in Scutigeromorpha, but the tarsi are not multi-segmented, and the eyes are not composite; the spiracles are lateral and are present only on some of the segments. An autapomorphy is:

18.  The single testicle (RFB: fig. 20-4, p. 707). Two testicles are formed in the embryo, but one is later reduced. (Scutigeromorpha have paired testicles; see below on Scolopendromorpha and Geophilomorpha, character 12).

A further, possible autapomorphy is:

19.  coxal pores (Fig. 8, cf. also below) on at least two pairs of legs, not only on the last pair as in Scolopendromorpha and Geophilomorpha.

Lithobius forficatus (to 3 cm long) is extremely common in Denmark, like several smaller species of the same genus. Lamyctes emarginatus is the only myriapod known from Greenland; it is parthenogenetic and is also known from Denmark.

Order Craterostigmomorpha

Only one species, Craterostigmus tasmanianus, from Tasmania and New Zealand. Hatches from the egg with 12 pairs of legs, the adult number of 15 pairs is reached after one moult; the number of leg-pairs is certainly a symplesiomorpy with the two orders above. C. tasmanianus superficially resembles a Lithobious.

Order Scolopendromorpha

Ca. 500 species in all parts of the World, mainly in warm regions. They resemble Lithobiomorpha superficially, but they have 21 or 23 pairs of walking legs. An autapomorphy is:

20.  The tergum of the poison fang segment is fused with that of the following segments (that carrying the first pair of walking legs). (The fused tergum is seen in RFB: fig. 20-1A, p. 704 as a trapezoid plate just behind the head).

Family Scolopendridae: four eyes on each side of the head. The very large centipedes belong here, genus Scolopendra and others in the tropics and subtropica including southern Europe.

Fam. Cryptopsidae: no eyes. Cryptops hortensis (2-3 cm) in Denmark. Scolopocryptops (= Otocryptops).

Order Geophilomorpha – Da: jordskolopendre

Ca. 1000 speices in all parts of the World. From 29 to almost 200 pairs of legs. No eyes. Good autapomorphies include:

21.  the earthworm-like burrowing technique (RFB: 707)

22.  the constant number of 14 antennal articles. In the other orders, the number is large and variable.

The high number of legs and the lack of eyes are other possible autapomorphies.

Several families, a dozen species in Denmark including Geophilus carpophagus, which is bioluminescent (glows in the dark) and occurs in old houses, and Strigamia maritima, which is sometimes abundant under seaweed on the beach.

Centipede phylogeny

The phylogeny of centipedes has been subject of much debate. The discussion illustrates very well the difficulties with deciding which characters are original, plesiomorphic, and which are derived, apomorphic. The Geophilomorpha have been regarded as sister-group to the other centipede orders, and so have the Scutigeromorpha. The class has also been divided into two groups: Anamorpha (= Scutigeromorpha + Lithobiomorpha) and Epimorpha (= Scolopendromorpha + Geophilomorpha). Based, among other things, on increased knowledge of the fifth order, Craterostigmomorpha, Dohle (1985) and Shear & Bonamo (1988) were able to present a convincing phylogenetic analysis, according to which relationships are as shown in Fig. 7.

The group Anamorpha was named after the mode of postembryonic development in its members, viz., hemianamorphosis: The juveniles hatched from the egg have 4 pairs of legs (excluding the poison fangs) in Scutigeromorpha, 6-7 pairs in Lithobiomorpha. The adult number (15 pairs) is gradually attained during growth, and when it has been reached, no new leg-pairs are added during succeeding moults. Hemianamorphosis occurs in all Pauropoda and Symphyla and also in primitive Diplopoda; this character is thus a clear symplesiomorphy for the members of ‘Anamorpha’.

Several synapomorphies are shared by Lithobiomorpha, Craterostigmomorpha, Scolopendromorpha and Geophilomorpha, including:

4.  the head is flattened

5.  the tentorium is reduced in a characteristic way

6.  the sternum and coxae of the poison fangs are fused to a coxosternum (RFB: fig. 20-2, p. 705, “coxosternite plate of forcipule”). This is clearly apomorphic, since the poison fangs are derived from normal walking legs

7.  the last leg-pair at least has coxal organs of a characteristic structure. They open to the surface through coxal pores (Fig. 8) and probably serve a water-regulatory function

8.  the spermatophore is placed on a web produced by a spinneret in the rear end of the male (RFB: fig. 20-6A p. 709).

Craterostigmomorpha, Scolopendromorpha og Geophilomorpha share further synapomorphies:

9.  eggs and young juveniles are protected by parents (RFB: fig. 20-6C, p. 709)

10.  juveniles hatch with the full (or almost full) adult number of segments and legs.

Scolopendromorpha og Geophilomorpha finally share the following synapomorphies:

11.  juveniles hatch with full number of adult number of segments and legs

12.  the testicles are fusiform, with vasa efferentia originating from both ends (Fig. 9, cf. RFB: fig. 20-4, p. 707)

13.  tracheae from different segments anastomose

14.  there is no median suture on the coxosternum (cf. character 6)

15.  there is a direct articulation between first and fourth article of the poison fang’s telopodite (Fig. 10, cf. RFB: fig. 20-2C, p. 705).

One character is in strong conflict with the cladogram, Fig. 7, viz., the heterotergy. In Geophilomorpha all terga are of similar size (homotergy). In the other orders there is a more or less pronounced alteration between short and long terga: heterotergy. In RFB: fig. 20-1, p. 704, the heterotergy is clearly seen in Lithobius, not so clearly in Otocryptops. In Scutigera the short terga are entirely hidden under the long ones. Intuitively one would believe homotergy to be original, heterotergy thus being a synapomorphy for all centipedes except Geophilomorpha. Balanced against characters 4-15 above, the heterotergy, however, must be regarded as convergent or plesiomorphic, in the latter case the homotergy in Geophilomorpha would be secondary.