196 PROTOZOA 



types, and having the knowledge and skill required to get all 

 the types in sexually reactive condition. This is a colossal task 

 and beset with many difficulties. Beyond question, it is too 

 much to expect of a zoologist who merely wishes to conform 

 to the mores by attaching the correct species name to the or- 

 ganism about which he wishes to publish his observations. The 

 only alternative is for him to send his culture to an expert for 

 identification. Although this could be done now — it is no easy 

 job even for the expert and could take weeks or months of work 

 in the case of varieties with long immature periods — the solu- 

 tion is one of limited dependability. There is no guarantee that 

 there will always be a laboratory maintaining a full battery of 

 live standard cultures. If identification depends upon such a 

 hazardous procedure, this basis for setting up species is much 

 too insecure. Type specimens would have to be kept perpetually 

 alive. I therefore reject mating type specificity as a basis for de- 

 fining named species. 



At present, serologic specificity cannot serve the purpose either. 

 However, this may be largely due to insufficient knowledge and 

 to unavailability of enough diagnostic antisera. If these lacks 

 were ever supplied and the diagnostic antisera were routinely 

 maintained by a number of established institutions well dis- 

 tributed around the world, perhaps identification of the varieties 

 might become practicable and within the limits of routine taxo- 

 nomic procedure. The situation would then be comparable to 

 the one that now exists for certain microorganisms such as Salmo- 

 nella. But P. aurelia is not Salmonella. There is no reason to be- 

 lieve that the demand for these means of identification would 

 ever justify the great labor and expense of maintaining a supply 

 of the 300 or more necessary antisera. I therefore also reject 

 serologic characteristics as a basis for defining named species 

 in this material. 



To aid in judging whether there arc any combinations of 

 characteristics which could routinely be used to achieve success- 

 ful identification, Table III presents the differential traits of the 

 varieties in summary fashion. Most of the differential characters 

 involve for their recognition prolonged study and highly elabo- 



