They present a great variety of characters with variously arranged ciliated bands. Most of these forms can be more or less satisfactorily derived from a larval form, like that of Serpula (fig. 139 B) or Polygordius (fig. 142); and the constant recurrence of this form amongst the Chætopoda, combined with the fact that it presents many points of resemblance to the larval forms of many Rotifers, Molluscs, and Gephyreans, seems to point to its being a primitive ancestral form for all these groups. The important characters of this larval form are (1) the division of the body into a large præ-oral lobe and a relatively small post-oral region containing the greater part of the alimentary tract; (2) the presence of a curved alimentary canal divided into stomodæum (oesophagus), stomach and intestine, and opening by a ventrally placed mouth, and an anus near the hind end of the body. To these may be added the frequent presence of (1) a ganglion at the apex of the præ-oral lobe, (2) a large cavity between the wall of the gut and the skin, which is the remnant of the segmentation cavity, and is usually traversed by muscular strands, of which one connecting the apex of the præ-oral lobe and the stomach or oesophagus is very commonly present (fig. 142). The arrangement of the ciliated bands presents great variations, though in some instances it is constant through large groups. In Chætopods there is a widely distributed præ-oral ciliated band, which is similarly placed to the ring constantly found in the larvæ of Molluscs, Rotifers, etc. In many of these forms the band is practically double, the opening of the mouth being placed between its two component rings (_vide_ fig. 142). The best introduction to the study of the Chætopod larval forms will be the history of the changes of a typical larval form in becoming converted into the adult. [FIG. 142. POLYGORDIUS LARVA. (After Hatschek.) _m._ mouth; _sg_. supra-oesophageal ganglion; _nph._ nephridion; _me.p._ mesoblastic band; _an._ anus; _ol._ stomach.] For this purpose no better form can be selected than the interesting larva of Polygordius (_vide_ Agassiz, No. 332, Schneider, No. 352, and Hatschek, No. 339), which was first discovered by Lovén, and believed by him to be the larva of an ordinary Chætopod. Its true nature was determined by Schneider. At a very young stage the larva has the form (fig. 142) of a flattened sphere, with a small conical knob at the posterior extremity. At the equator are situated two parallel ciliated bands[137], between which lies the ventrally placed mouth (_m_). The more conspicuous ciliated band is formed of a double row of cilia, and is situated in front of the mouth. The thinner ciliated band behind the mouth appears to be absent in the American species. [137] These two rings are at first (Hatschek) not quite closed dorsally, calling to mind the early condition of the Echinoderm larvæ with a præ-oral and post-oral ciliated area.