CINIDARIA AS THE ONLY COELENTERATA D 



unparalleled in other animals. In addition to this comes the 

 special kind of ontogeny. 



Finally, when we study attempts to construct the phylogeny 

 of Spongiae as if they had the nature of Coelenterata (Remane, 

 Jagersten)— these attempts must be considered as quite un- 

 successful—then it actually becomes clear that Spongiae are 

 really something "exceptional," and that they have developed 

 independently from Protozoa, most probably out of Choano- 

 flagellata. (Hadzi, 1917). Yet even here it remains impossible 

 to get ahead without a grain of imagination; we must use it to 

 construct the phylogenetic processes of sponge evolution since 

 it is completely hopeless to expect that we will ever run into 

 any new documents which will help us to construct an actual 

 phylogeny of Spongiae. Naturally enough, such an artificial 

 construction, too, must be based on the actual facts that are 

 now accessible. The scheme must be as close as possible to 

 reality. 



Since the Spongiae with their structure, their way of life, 

 and their individual development represent something different 

 from all other Metazoa, it must be expected, and actually 

 reckoned with, that their phylogenetic evolution, too, was a 

 special one. It is therefore quite improbable that the ancestors 

 of Spongiae could be similar to gastraea. Here we run into 

 very great difficulties when we try to construct the origin of 

 Spongiae. We cannot expect any palaeontological evidence. 

 We must limit ourselves to comparative morphology where, 

 too, we are completely unable to make any progress without 

 some imagination. Nevertheless, our scientific imagination- 

 let us call it so, though it is usually called speculation— must 

 be kept in check and within the limits of probability. The 

 reconstruction must correspond to the factual material that is 

 available. 



On the basis of my studies on Spongiae (Hadzi, 1917) I con- 

 structed a phylogeny of Spongiae and submitted this to the 

 International Congress of Zoologists at Budapest in Hungary 

 (Hadzi, 1929). This construction was later somewhat improved 

 2 



