23 I 



THE AMERICAN NATURALIST [Vol. LII 



in past geologic ages by study of the survivors, but it is 

 self-evident that there should be a perfect knowledge of 

 these survivors, morphological and physiological, so that 

 we may not be led astray by superficial resemblances 

 when there exist, in fact, deep-seated and fundamental 

 distinctions. 



The geologic evidence that has been adduced as to the 

 character of the primitive bacteria is of but the slightest 

 value. Speculation as to the primitiveness of nitrogen 

 fixers, for example, based upon the geologic evidence 

 introduced is scarcely convincing. 



It should also be noted that it is quite possible that the 

 bacteria do not constitute an homogeneous group in the 

 sense that all are descended from a single primitive type 

 of bacterium. It may be that there have been included 

 in the group bacteria forms which have assumed similar 

 morphological or physiological characters without having 

 a common ancestry. Botanists, for example, at the 

 present day are by no means convinced that seed plant- 

 have all had a common origin; in other word-, the ability 

 to produce seeds may have arisen independently in two 

 or more groups of the fern plants. It is possible that 

 some of the forms we term bacteria have been derived 

 from the fungi, others from the blue-green alga? or pos- 

 sibly some even from the protozoa. In short, it may be 

 that the actual relationships existing between various 

 bacteria may be very distant. 



A study of these modern bacteria will reveal relation- 

 ships such as those just indicated. The possibility that 

 the bacteria are a derived group must be constantly held 

 in mind. To prove that they are primitive it must be 

 shown that no group from which they might have sprung 

 or to which they seem to be related can be regarded as 

 more primitive. This has not been satisfactorily accom- 

 plished in certain cases. 



Modern systematic bacteriologists are in fair agree- 

 ment that there should be recognized five or six distinct 

 groups or orders of bacteria. Of these, the Eubacteriales, 



