Continuation of the chapter Three

4. PRINCIPLES FOR THE DETERMINATION OF RELATIONSHIPS
What are the characters on which the recognition and separation of natural groups may be based? In the classification of higher plants and animals, systematists have relied almost exclusively on morphology. Nevertheless, some exceptions to this rule may be found, particularly in the treatment of the thallophytes, where an increasing reliance appears to be placed on physiological characters. For example, Smith (1938) in breaking up the algae into seven new divisions, places as much weight on physiological characters (reserve products, nature of the cell wall, pigments) as on morphological ones. It is above all in bacterial systematics that extensive use of physiological criteria has been made. This is understandable
enough in view of the paucity of morphological data but, as pointed out by Kluyver and van Niel (1936), the injudicious use of physiology without a clear understanding of what constitute important physiological characters, has led to much confusion. A good example of this is the order Thiobacteriales of Buchanan, which is based on the presence of "bacteriopurpurin and/or
sulfur granules" in the cells. In addition to uniting exceedingly heterogeneous morphological groups, these characters also bring together two radically different physiological groups, the photosynthetic purple bacteria and some of the chemosynthetic colorless sulfur-oxidizing bacteria. Furthermore, extreme physiological systems have often neglected obvious morphological relationships, with the result that natural morphological groups have been split up or forced into assemblages with which they have little in common save certain aspects of metabolism. The Thiobacteriales again are a good example of this; other even more glaring ones may be found in the system of Orla-Jensen (1909) where, for example, the genera Mycobacterium, Corynebacterium and Actinomyces are placed among the cephalotrichous bacteria in one family with the genus Rhizobium! The chief stumbling block in attempting to draw up a phylogenetic system on a primarily physiological basis is the necessity of making a large number of highly speculative assumptions as to what constitute primitive and advanced metabolic types.Orla-Jensen, for example, regarded the chemosynthetic bacteria as the most primitive group because they can live in the complete absence of organic matter and hence are independent of other living forms. This overlooks the fact that a chemosynthetic metabolism
necessarily presupposes a rather highly specialized synthetic ability such as one would not expect to find in metabolically primitive forms. Furthermore, this reasoning was based at least
in part on the hypothesis that living forms arose at a time when the earth was devoid of organic matter, an hypothesis which has been effectively challenged by Oparin (1938) in his book on the
origin of life. According to Oparin, it is probable that a long period of chemical synthesis of organic material preceded the emergence of life, and that consequently the earliest living forms
were heterotrophs. On this reasoning, the development of autotrophism was a later adaptation to an environment in which organic materials had become scarce through the activities of heterotrophs.Thus, the basic assumption used by Orla-Jensen in erecting aphysiological phylogenetic system has been rendered, to say the least, highly doubtful. The physiological reasoning on which the further development of the system is founded is also open to serious
criticisms.
In spite of the comparative simplicity of bacteria it is rather naive to believe that in the distribution of their metabolic characters one can discern the trend of physiological evolution. For these reasons, a phylogenetic system based solely or largely on physiological grounds seems unsound. It is our belief that the greatest weight in making the major subdivisions in the Schizomycetes should be laid on morphological characters, although correlative physiological characters may also be used. What, then, are the basically important morphological characters which we can use? Clearly paramount is the structure of the individual vegetative cell, including such points as the nature of the cell wall, the presence and location of chromatin material, the functional structures (e.g., of locomotion), the method of cell division, and the shape of the cell. A closelyallied character is the type of organization of cells into larger structures. In addition, the nature and structure of reproductive or resting cells or cell masses deserve due consideration. In the following sections we will examine the major bacterial groups which we can discern by the application of these criteria, elucidating as far as possible the relationships of these groups to each other and to other microorganisms.