20 THE AMERICAN MONTHLY [Jannary, 



5 to lo times ; bouillon cultures, 40 to 50 times, and when the cover-glass prepar- 

 ations are made directly from gelatine or agar cultures a dilution of 100 times is 

 necessary. He employs boiled water for making the dilutions. A drop of a 10 

 per cent, solution of alcohol is placed on the preparation and the whole allowed 

 to dry in the air. The dried preparations, without being heated, are placed in 

 the mordant, where they are allowed to remain for from & to 12 hours or longer. 

 They are then thoroughly rinsed in water, after which they are placed m the 

 iodine water, where they should remain for about one hour. At the end of that 

 time the}' are again rinsed in water and transferred to the staining solution. 

 This is allowed to act for about one-half hour, when the preparations are rinsed 

 and mounted for examination, either directly in water or allowed to dry and sealed 

 in balsam. With some germs better results are obtained by varying the quantity 

 of hydrochloric acid in the mordant. 



DowdeszvelVs Method. — In a comparatively recent publication Mr. 

 Dowdeswell (10) states, after referring to the method employed by 

 Neuhauss to demonsti"ate the flagella on the comma bacillus, that it is 

 not difficult to detect the flagella on the comma forms if appropriate 

 but ordinary means are adopted. The optical apparatus required are 

 simply a normal retina, a good objective with a inoderate angle of 

 aperture, and good light. He recommends as a staining fluid an aque- 

 ous solution of gentian violet, although other aniline dyes will answer 

 quite as well, and for mounting purposes the acetate of -potash. There 

 are no special directions given excepting that the specimens must be 

 mounted in the fluid mentioned and not in Canada balsam He also 

 aflirms that no difficulty will be experienced in staining the flagella on 

 microbes as small as the bacteriufn terfno. 



From what has already been said it will be observed that the flagella 

 differ to a greater or less degree on the various bacteria on which they 

 have been demonstrated. Dallinger (3) figures the flagella on the bac- 

 terium termo as two thread-like appendages, one projecting from each 

 end of the rod-shaped germ. Their length is much greater than the 

 long diameter of the germ, but their thickness one-tenth that of the 

 bacterium. He gives the mean average diarneter of the flagellum of 

 this germ as 1-304700 inch, which he determined from fifty measure- 

 ments with each of four diflerent lenses. The flagella on the sfirillu?n 

 volutans and several other forms are similarly figured by the earlier in- 

 vestigators. From the plates and descriptions given by Loeffler and 

 Trenkmann it seems that the spirilla and a few bacilli possess one or 

 more thread-like filaments projecting only from the extremities of the 

 organism, but that the greater number of the inotile bacteria on which 

 these appendages have been demonstrated are provided with a greater 

 or less number of long, delicate, thread-like filaments. They appear to 

 be given off' either single or in tufts both at the extemities and at short 

 and varying intervals along the entire length of the rod-shaped organ- 

 ism. In the stained preparation the ffagella appear either straight, 

 curved, or in the form of spirals. Many of them are usually detached 

 from the germs, presumably during the process of preparation, and ap- 

 pear as extremely long, delicate bacilli or spirilla lying between the 

 bacteria. This fact renders it difficult to determine with any degree of 

 accuracy the number of flagella belonging to any particular germ. 



The question may naturally arise, which one of the different methods 

 presented is best adapted for staining the flagella on any particular mo- 

 tile germ that may be under consideration .? In answer to such a ques- 

 tion I would say that Dr. Theobald Smith successfully employed 



