For many decades, scientists thought that multicellular life didn't exist before the dawn of the Cambrian period, 542 million years ago. Then, in 1967, a careful geological study of pre-Cambrian rock beds turned up frond-like organisms called Charnia. This was the first time that definitive pre-Cambrian metazoan fossils had been found, and it created a sensation. Since then, over 18 fossil beds from the Ediacaran period (before the Cambrian) have been found, with over 100 different types of organisms dating back to 600 million years ago. Scientists get into deep arguments over whether these ancient forms are related to modern-day animals, and the consensus is that some of them are.

Though common knowledge states that modern phyla have their origin in the Cambrian explosion about 520 million years ago, recent findings have confirmed representatives of modern phyla in the Ediacaran period. Some of these likely represent stem groups that bear little to no relation to living forms. The phyla that have representatives in the Ediacaran period are Porifera (sponges), Cnidaria (Inaria.), Ctenophora (comb jellies), probable Mollusca (Kimberella), probable Onychophora (Xenusion), probable Arthropoda (Parvancorina), probable Annelida (Cloudina), probable Echinodermata (Arkarua), and members of an extinct proposed phylum of bilateral animals, Proarticulata (Dickinsonia).

Only three modern phyla (Porifera, Cnidaria, and Ctenophora) are mostly accepted by scientists as having existed in the Ediacaran period, and there is substantial reason to believe that Mollusca, Onychophora, Arthropoda, Annelida, and Echinodermata were represented as well. Early mollusks are among the oldest of the smally shelly fauna that mark the start of the Cambrian, and the Ediacaran fossil Kimberella has features like that of a mollusk, such as a univalve shell, and is found in conjunction with scrapings that strongly suggest a radula, the distinguishing feature of mollusks. Recent images of Xenusion strongly suggest that it is an early onychophoran (velvet worm), while Parvancorina clearly has a head, and despite the absence of fossilized limbs, looks like a stem group arthropod. Annelida and Echinodermata are more uncertain, but the numerous tube-shaped skeletons found from the Ediacaran period are suggestive of polychaete worms, and the five-fold symmetry of Arkarua makes it a probable echinoderm.