Phylogeny of Animals 517 



mass; above this, in Loxosoma as in some rotifers, a pair of 

 simple, but in this instance temporary, eyes arise. Below 

 the ciliated ring, and ventrally placed, is the mouth. The 

 ventral area between it and the foot (v) is greatly shortened; 

 the alimentary canal has become bent and opens well forward 

 on the restricted dorsal surface; while the supposed excretory 

 tubes of Verworn and Cori occupy appropriate position for 

 such, though greatly reduced in size. This however is to be 

 expected, when the capacity for excretion alike by the arms 

 and the atrium, as well as the sluggish character of the adult, 

 are considered. 



The so-called Cyphonautes embryo of many marine genera 

 undergoes striking change owing to enlarged forward growth 

 of the supra- or epi-oral region, by ingrowth and then deep 

 invagination of the atrium, and by transfer of the alimentary 

 canal in the process, while the mouth and anus both open into 

 the atrium (Fig. 22 b-e.) 



A study of the forms of polyzoan embryo therefore shows 

 that all evidently started from a common ancestry with the 

 Rotifera, and that gradual but striking modification has been 

 effected by slow evolutionary processes. Such simple and 

 soft-bodied forms as branched off from older rotiferan ances- 

 tors continued mainly if not wholly amid fresh-water sur- 

 roundings. Remnants of these are seen in Urnatella amongst 

 Entoprocta, and in the phylactolsemids amongst Ectoprocta. 

 By degrees adopting the colonial habit, they seem to have 

 migrated or been carried seaward, probably during late ar- 

 chsean times. There by chitinous growth, or by secretion of 

 lime salts from the ocean, as with the seaward migrating Spon- 

 gida and Coelenterata, they had by silurian times established 

 a varied and rich assemblage of species that are in part known 

 to us in the fossil state. 



In frequent past systems of classification the Polyzoa and 

 Brachiopoda have been united as the Molluscoidea. But, 

 whether such be accepted or not, many zoologists regard both 

 as fairly well related. So far as we are aware, however, no 

 attempt has hitherto been made to trace the Brachiopoda 

 back to a rotiferan ancestry, though the trochophore character 

 of the larva has usually been recognized. 



