48 CLASSIFICATION OF BACTERIA 



Genus IV. Phragmidiothrix cells at first united in unbranched 

 threads, dividing in three planes, thus forming a rope of cells; later 

 some of the cells may penetrate through sheath and thus give rise 

 to branches. 



FAMILY V.Beggisatoacea cells united in sheathless threads; 

 division in one direction of space only; motility by undulating mem- 

 brane as in Oscillaria. 



Genus L Thiothrix unbranched, non-motile threads, inclosed 

 in fine sheaths; division of cells in one plane only; cells contain 

 sulphur granules. 



Genus II. Beggiatoa cells with sulphur granules. 



The difficulties inherent in this classification and especially the 

 needs of reform to the* agricultural bacteriologist are seen from the 

 following: 



"Many workers in medical bacteriology and in other special 

 fields of applied microbiology, who deal with only a few well- 

 recognized species, may perhaps feel no need for any change in 

 current practice. Few can deny, however, that it is a serious 

 inconvenience for such names as B. welchii, B. spprogenes, B. per- 

 fringens to be used by various workers, sometimes for the same, 

 sometimes for different organisms, or for the same form to be 

 described as Bacterium lactis aerogenes or Streptococcus lacticus 

 when it is isolated from milk and as Streptococcus salivarius or 

 Sir. fecalis when it is isolated from the human mouth or intestine." 



"When one passes from a study of the practical effects of the 

 activity of some particular microbe to a consideration of its relation- 

 ship to other forms it becomes essential not only to have a name 

 for each kind of organism but to have also a system of nomenclature 

 which will make it possible to express such relationship with reason- 

 able clearness and accuracy. 



"This need is met by the Linnaean system of classification uni- 

 versally adopted by all biologists outside our own limited and sys- 

 tematically undeveloped fields. According to this Linnaean system 

 each recognizable kind of plant or animal receives a binomial 

 Latinized name, the first half designating the genus or group to 

 which it belongs and the second half the particular kind or species 

 to which the name applies. The genera in turn are grouped in 

 tribes, the tribes in families, the families in orders, and the orders in 

 classes. These divisions will often be artificial and often of greatly 

 unequal size and importance in different groups. They make it 

 possible, however, to express in a simple manner the essential facts 

 of biological relationship the fact that A, B, and C are more 

 nearly related to each other than are any of them to D, E, and F; 

 and that the series A-F exhibits common relationships closer than 

 any similarities which its members bear to G or H. 



"If such a system is accepted it is in the next place important to 



