Every animal is a dark continent under the skin. Beneath the surface of even the most humdrum worm lurks the information-rich world of the genome – the genetic sequence, or program, that encodes the basic features of that organism. Genomic research is the hottest area of biology.

Although most attention is focused on the genomes of humans, and laboratory animals such as mice, research on the genomes of some of the Earth's lesser-known inhabitants is turning up some remarkable surprises. One of the more remarkable – from one of the Earth's least-known – is presented in a report in Nature1 (21 October 1999.)

Nobody would mind if you hadn't heard of tiny creatures called dicyemids. These creatures, barely discussed outside a small band of specialists, are microscopic parasites only found in what passes for kidneys in squid and octopi. The bodies of dicyemids are among the simplest of all multicellular animals. They lack guts and nervous systems: in fact, the entire body consists of a single large 'axial' cell surrounded by 10-40 outer cells. In fact, the dicyemid body is about as simple as an animal can possibly be without being reduced to a single cell.

Parasites often undergo dramatic simplifications in body form, as they exploit the bodies of their hosts for everyday needs. Tapeworms, for example, lack guts and absorb nutrients from the host. Sometimes the bodies are so reduced that it is extremely hard to determine the evolutionary relationships of parasites. Some barnacles, parasitic on other crustaceans, look more like amorphous fungoid blobs than animals, their status being revealed only in the features of their larvae.

The wholly parasitic dicyemids are no exception to this general rule, that parasites are abbreviated versions of free-living ancestors. The problem is that dicyemids are so abbreviated – just a ball of cells – that they could have evolved from virtually anything. Some researchers have suggested that they are just fancy versions of single-celled protozoa – others that dicyemids are dramatically reduced flatworms.

Many flatworms inhabit the kind of internal spaces occupied by dicyemids today, and it is easy to imagine that, millions of years ago, a flatworm crawled up inside an octopus, made itself comfortable, and proceeded to shed every vestige of its flatworm heritage. The traditional zoological view is a compromise, placing dicyemids in a separate group called the 'Mesozoa', an evolutionary limbo between single-celled protists and properly multicellular animals.

Enter Peter Holland of the University of Reading in the UK. Holland has made something of a career investigating the genetics of creatures only ever discussed in the less-frequented pages of zoology textbooks. He and his colleagues have turned to dicyemids, and have been looking in particular at a family of genes called 'Hox' genes. These genes are implicated in determining the overall layout or 'body plan' of animals from flatworms to humans.

The evolution of animal complexity can be read from the Hox genes. In particular, there is a subgroup of Hox genes only found in animals more complex than jellyfishes: it turns out that the Hox gene isolated from the dicyemid Dicyema orientale corresponds most closely to a gene in flies that falls into this subgroup. Even more specifically, the dicyemid Hox gene bears a tell-tale motif that groups it with the so-called lophotrochozoa – a large group of animals that includes annelids (segmented worms), molluscs and some flatworms such as the planaria of ponds and streams. This research shows just how far dicyemids have fallen, making them among the most dramatic cases of secondary simplification in the animal world.