T. M. SONNEBORN 285 



occur in other asexual organisms, including some in which 

 serologic traits are used to define species. Failure to appreciate 

 that organisms with the same genotype can express serologically 

 unrelated antigens could easily lead to assigning organisms witli 

 the same genotype to different species. Only by knowing the full 

 array of potentially expressible antigens in a clone and by taking 

 into account the grades of difference in cross reactions between 

 corresponding serotypes in the arrays producible in different 

 clones could a sensible beginning be made in establishing sero- 

 typic criteria for distinguishing species. Lederberg (1955) points 

 out other weaknesses of the use of serologic species criteria in 

 bacterial taxonomy based upon his work on gene transduction. 



Physiological differences such as virulence and host specificity 

 are used as species differentials in Trypanosomes and other para- 

 sitic asexual organisms. Both of these criteria may be objected to 

 on the ground that comparable traits in other organisms (e.g., 

 bacteria and viruses) are known to arise as single mutational 

 steps and to be inherited as single gene traits. So simple a genetic 

 difference would surely not be acceptable as a species differen- 

 tial in sexual organisms, and there is no good apparent reason 

 why this should be done in asexual organisms. It is sometimes 

 argued that confinement to different hosts is equivalent to 

 geographic isolation. But this is only true if the adaptation to 

 different hosts is outside the range of simple genetic differences. 

 Otherwise the mutational array of host specificities within each 

 single line of descent will yield a common array of potential 

 hosts. 



Restriction of species differentials to visible differences does 

 not necessarily avoid this difficulty. Some visible differences also 

 are of course simple in genetic basis. For example, certain species 

 of Trypanosomes are known to differ only with respect to the 

 presence or absence of a kinetoplast. This is a self-reproducing 

 cytoplasmic granule. Under the action of certain chemicals, the 

 kinetoplast can be irreversibly lost. Sublines of the same clone 

 can differ in this way. Yet they must be assigned to different 

 species if this is accepted as a valid species differential. The sit- 

 uation is closely parallel to the existence of killer and sensitive 



