THE MAIN LINES OF ANIMAL EVOLUTION 



duction is the rule for such colonial flagellates. The sperm are at once 

 similar to some simpler, noncolonial flagellates and to the typical sperm 

 of Metazoa. Such colonies may also show anteroposterior differentiation. 

 These large, highly specialized colonies occur principally among the plant 

 flagellates, yet colonies of a highly suggestive character also occur among 

 the animal flagellates. None of these reasons is conclusive, yet collectively 

 they carry considerable weight. 



However much the origin of other Metazoa may be disputed, it seems 

 almost certain that the Porif era ( sponges ) were derived from choanoflag- 

 ellates. It is a short step from the structure of the colonial choanoflagellate 

 Proterospongia to that of the simplest sponges. Proterospongia (Figure 

 45 ) consists of a small mass of gelatinous material in the surface of which 

 are imbedded choanocytes (collared, flagellate cells), and in the interior 

 of which are ameboid cells. The choanocytes can withdraw the collar and 

 flagellum and move into the interior to become ameboid cells. In order to 

 change this to the structure of a simple sponge it would be necessary only 

 to develop a system of channels through the gelatinous mass, to let these 

 channels be lined with choanocytes, and to let the outer surface be cov- 

 ered by a simple epithelium. No organ systems are present, the functions 

 of the sponge being carried on by the component cells individually. 



Phylum Porifera. Cooperative activity is at a minimum, and it has often 

 been debated whether a sponge is a true organism or a colony of unicellu- 

 lar organisms. There are enough evidences of cooperative activity to swing 

 the balance of zoological opinion in favor of the former alternative. Thus, 

 while some sponges have no definite shape, many have rather complex 

 shapes. Most sponges produce skeletal elements in the form of spicules 

 which may be quite complex, and so exactly formed in every case that 

 they are among the best taxonomic characters. Further, the beat of the 

 flagella in the canals is not obviously coordinated, yet the water current 

 is unidirectional, and so there must be at least a limited control of the 

 flagellar beat. 



Sponges can reproduce asexually by formation of groups of cells, the 

 gemmules, but they also reproduce sexually by means of typical eggs and 

 sperm. The zygote develops into a free-swimming larva which is flagel- 

 lated. After swimming for a brief time, it settles down to become a sponge. 

 Because this larva has typical ( not collared ) flagella, some zoologists be- 

 lieve that the sponges were derived from some typical flagellates, with 

 choanocytes being a secondary development. Yet this is not unexpected in 

 view of the recapitulation principle, for the choanoflagellates themselves 

 must have been derived from typical flagellates. 



Evolution within the Porifera has taken the forms of elaboration of 

 the canal system and of the supporting spicules or fibers. As relationships 

 within the phylum are not at all clear, it may be more profitable to go 

 directly to the problem of the relationships of the Porifera to other ani- 

 mals. The Porifera are a terminal group. Further, they are so diflFerent 

 from all other Metazoa in the absence of unified tissues, in the physiologi- 

 cal independence of the individual cells, and in their embryology and 

 adult anatomy, that it is generally believed that their origin from the 



133 



