THE GENERAL SHAPES OF ANIMALS. 177 



partly determined by the habitual mode of gemmation and 

 partly by the surrounding conditions in each case, is in 

 great measure indefinite. We can say no more about it than 

 that it is not obviously at variance with the laws alleged. 



E vidence of a more positive kind occurs among those com- 

 pound Molluscoida which are most like the compound 

 Ccelenterata in their modes of union the Polyzoa. Many of 

 these form groups that are more or less irregular spreading 

 as films over solid surfaces, combining into sea-weed-like 

 fronds, budding out from creeping stolons, or growing up 

 into tree-shaped societies ;' and besides aggregating ir- 

 regularly they are irregularly placed on surfaces inclined in 

 all directions. Merely noting that this asymmetrical dis- 

 tribution of the united individuals is explained by the 

 absence of definiteness in the relations of the aggregate to 

 incident forces, it concerns us chiefly to observe that the 

 united individuals severally exemplify the same truth as do 

 similarly- united individuals among the Ccelenterata. While 

 their internal organs, though said to have a trace of bi- 

 lateralness, cannot be said to display any definite symmetry ; 

 their external organs are completely radial. Averaging the 

 members of each society, the ciliated tentacles they protrude 

 are similarly related to prey on all sides ; and therefore 

 remain the same on all sides. This distribution of tentacles 

 is not, however, without exception. Among the fresh-water 

 Polyzoa there are some genera, as Plumatetta and Crystatella, 

 in which the arrangement of these parts is very decidedly 

 bilateral. Some species of them show us such relations of 

 the individuals to one another and to their surface of attach- 

 ment, as give a clue to this modification ; but in other species 

 the meaning of this deviation from the radial type is not 

 obvious. 



249. In that somewhat heterogeneous assemblage of 

 animals now classed, perhaps provisionally, as Annuloida, we 

 begin again with simple aggregates of the second order, and 



