Onychophora - Phylogeny

Phylogeny

In their present forms, the velvet worms are probably very closely related to the arthropods, a very extensive taxon that incorporates, for instance, the crustaceans, insects and arachnids. They share, among other things, an exoskeleton consisting of α-chitin and non-collagenous proteins; gonads and waste-elimination organs enclosed in true coelom tissue; an open blood system with a tubular heart situated at the rear; an abdominal cavity divided into pericardial and perivisceral cavities; respiration via tracheae; and similar embryonic development. Segmentation, with two body appendages per segment, is also shared. However, antennae, mandibles and oral papillae are probably not homologous to the corresponding features in arthropods, i.e., they probably developed independently.

Another closely related group are the comparatively obscure water bears (Tardigrada); however, due to their very small size, these lack some characteristics of the velvet worms and arthropods, such as blood circulation and separate respiratory structures. Together, the velvet worms, arthropods and water bears form a monophyletic taxon, the Panarthropoda, i.e., the three groups collectively cover all descendants of their last common ancestor.

Due to certain similarities of form, the velvet worms were usually grouped with the water bears to form the taxon Protoarthropoda. This designation would imply that both velvet worms and water bears are not yet as highly developed as the arthropods. Modern systematic theories reject such conceptions of "primitive" and "highly developed" organisms and instead consider exclusively the historical relationships among the taxa. These relationships are not as yet fully understood, but it is considered probable that the velvet worms' sister groups form a taxon designated Tactopoda, thus:

For a long time, velvet worms were also considered related to the annelids. They share, among other things, a worm-like body; a thin and flexible outer skin; a layered musculature; paired waste-elimination organs; as well as a simply constructed brain and simple eyes. Decisive, however, was the existence of segmentation in both groups, with the segments showing only minor specialisation. The parapodia appendages found in annelids therefore correspond to the stump feet of the velvet worms.

Within the Articulata concept developed by Georges Cuvier, the velvet worms therefore formed an evolutionary link between the annelids and the arthropods: worm-like precursors first developed parapodia, which then developed further into stub feet as an intermediate link in the ultimate development of the arthropods' appendages. Due to their structural conservatism, the velvet worms were thus considered "living fossils". This perspective was expressed paradigmatically in the statement by the French zoologist A. Vandel:

Onychophorans can be considered highly evolved annelids, adapted to terrestrial life, which announced prophetically the Arthropoda. They are a lateral branch which has endured from ancient times until today, without important modifications.

Modern taxonomy does not study criteria such as "higher" and "lower" states of development or distinctions between "main" and "side" branches—only family relationships indicated by cladistic methods are considered relevant. From this point of view, several common characteristics still support the Articulata concept—segmented body; paired appendages on each segment; pairwise arrangement of waste-elimination organs in each segment; and above all, a rope-ladder-like nervous system based on a double nerve strand lying along the belly.

An alternative concept, most widely accepted today, is the so-called Ecdysozoa hypothesis. This places the annelids and Panarthropoda in two very different groups: the former in the Lophotrochozoa and the latter in the Ecdysozoa. Mitochondrial gene sequences also provide support for this hypothesis.

Proponents of this hypothesis assume that the aforementioned similarities between annelids and velvet worms either developed convergently or were primitive characteristics passed unchanged from a common ancestor to both the Lophotrochozoa and Ecdysozoa. For example, in the first case, the rope-ladder nervous system would have developed in the two groups independently, while in the second case, it is a very old characteristic, which does not imply a particularly close relationship between the annelids and Panarthropoda.

The Ecdysozoa concept divides the taxon into two, the Panarthropoda into which the velvet worms are placed, and the sister group Cycloneuralia, containing the threadworms (Nematoda), horsehair worms (Nematomorpha) and three rather obscure groups: the mud dragons (Kinorhyncha); penis worms (Priapulida); and brush-heads (Loricifera).

Particularly characteristic of the Cycloneuralia is a ring of "circumoral" nerves around the mouth opening, which the proponents of the Ecdysozoa hypothesis also recognise in modified form in the details of the nerve patterns of the Panarthropoda. Both groups also share a common skin-shedding mechanism (ecdysis) and molecular biological similarities. One problem of the Ecdysozoa hypothesis is the velvet worms' subterminal mouth position: unlike in the Cycloneuralia, the mouth is not at the front end of the body, but lies further back under the belly. However, investigations into their developmental biology, particularly regarding the development of the head nerves, suggest that this was not always the case and that the mouth was originally terminal (situated at the tip of the body). This is supported by the fossil record.

The "stem-group arthropod" hypothesis is very widely accepted, but some trees suggest that the onychophorans may occupy a different position; their brain anatomy is more closely related to that of the chelicerates than to any other arthropod.