BUREAU OF ANIMAL INDUSTRY. 605 



casioiially form/? are observed longer than this (1.8 micromillimetcr) 

 and somewhat more slender ; in these also there is a noticeable differ- 

 ence between the central and the peripheral portions in the intensity of 

 the stain. (This difference is not an optical effect, for it is only seen after 

 the bacteria have been stained and are examined without a diaphragm. 

 It is well known that large pencils of light efface details of structure and 

 bring out those depending on differences in the intensity of the color.) 



In liquid media the bacterium is motile; its movements recall tliosc 

 oihactcnnm termo. When moving to and fro the pairs of bacteria are 

 apt to revolve about the point of division in such a way that each indi- 

 vidual describes the surface of a cone. That it is not hacterium icrmo is 

 proved by the fact that it does not liquefy gelatine and that not the 

 sliglitest putrefactive odor is emitted from any culture containing it. 



The bacterium under cultivation shares, with many other bacteria, 

 the property of varying slightly in form in different media, and at dif- 

 ferent periods in the same medium. Cultivated in meat infusion with 

 1 per cent. i)eptone for forty-eight hours, the average size of a bacterium 

 was .9 micromillimetcr in length, and .4 to .5 micromillimetcr in width. 



In meat infusion peptone gelatine they are larger than in cover-glass 

 preparations from the spleen. Occasionally a very long filament may 

 he seen lying amongst the ovals and short rods. These vary from .8 

 micromillimeter to 1.8 micromillemeter in length, and are about .7 micro- 

 millimeter broad in cultures forty-eight hours old. 



The value of dimensions must not be overestimated. They are at 

 best somewhat variable. The size of a microbe which is constantly 

 undergoing division must vary with the activity of multiplication, 

 w^hich in turn depends on the amount of nutritive material at hand. 

 Hence in a rich medium, at an early period of growth, microbes mul- 

 tiply faster, and the forms are relatively smaller than in less nutritive 

 media, or in such in which the food material is nearly exhausted. This 

 at least agrees with our own observations. In cultures of the swine 

 plague bacterium in chicken broth, probably the poorest of meat ex- 

 tracts, slender filaments of considerable length were frequently observed. 

 The culture was suspected of being impure, but when tested on jilates 

 no difference among the colonies could be observed. Moreover, these 

 forms had an abnormal appearance never seen in vigorously gi-owing 

 bacilli. They often bore dilations and constrictions, and their extremi- 

 ties were poorly defined. They were also characterized by very slug- 

 gish movements. 



In all cultures we have determined the purity of cultures not by one 

 but by several characters. The most important of these are: (1) The 

 slow growth and absence of liquefaction in nutritive gelatine ; (2) the 

 a])pearance of the colonies of the bacterium on gelatine plates; (3) its 

 form and motility in liquid media. 



In neutralized liquids, such as extracts or infusions of beef with or 

 without peptone, the bacterium multiplies very rapidly, so that tubes, 

 inoculated with a minimum quantity of blood, &o., containing it, are 

 invariably turbid on the following day. This turbidity is greatest in 

 liquids rich in nutritive matter, such as those containing iieptone. In 

 si ni pie meat extracts the liquid remains merely opalescent. On shaking 

 the tube a considerable flaky deposit is seen in a few days after inocu- 

 lation. There is at no time a distinct membrane observable on the sur- 

 face of the liquid, although, when tubes remain very quiet for a time, a 

 narrow whitish ring is ajjt to be deposited on the glass at the surface 

 of the liquid. This band, consisting solely of bacteria, is sometimes en- 

 tire, sometimes limited to a small portion of the circumference of the 



