i26 THOMPSON YATES AND JOHNSTON LABORATORIES REPORT 
in that respect. Experimenting now with one of these cultures, non-motile and 
non-flagellate, I attempted to restore its motility and flagella, and I grew them for a 
long time on glycerine and dextrose solutions. But so far the motility has not been 
restored. I then passed this culture through a guinea-pig, once, and isolated the 
bacillus from the hearts blood. The bacillus examined was now motile, and had all 
the cultural characters of typhoid. On staining for flagella the result was positive. 
These observations seem to me to prove not only that we have non-flagellate 
typhoid bacilli, but that the flagella can be restored by ' passage ' through animals. I 
think this is of importance in many respects. 
1. It shows that non-motility and the non-flagellate condition may not be 
characteristic of the bacillus, but may be an acquired character due to laboratory 
conditions of growth. How far other bacilli can lose their flagella remains to be seen. 
For instance, a very old strain of B. pyocyaneus examined by me was found to be 
actively motile, and, of course, to possess flagella. 
2. It will be necessary in certain cases before we can state that a bacillus is 
non-flagellate to examine it as soon as isolated from the body, or to pass it through 
an animal. Thus in the case of B. dysenteriae and other bacilli and cocci there is a 
certain conflict of evidence as to the presence or absence of flagella, which may be 
due to the age of the culture. 
3. If this bacillus 'reacted normally' in the agglutination test of typhoid serum 
then the presence of flagella cannot be an essential factor in agglutination. 
4. Biologically, the experiment is of interest as showing how profoundly 
environment can modify structure. 
I may refer, in conclusion, again to the method used by me — it is a simplifica- 
tion and modification of Van Ermengem's silver method. 
1. The mordant is osmic acid and tannin (I find also that FeSO + and tannin will 
give positive results ; but I have, so far, only made one observation). This is allowed 
to act for one-half to one hour or more. 
2. Silver solution, O" 1 to 0'3 per cent. 
3. Ammonium tannate solution : — 
2/10 c.c. 20 per cent, tannin ) 
/TT „ , y y t r , v 1 freshly made. 
1 c.c. (H 2 0 and NH 3 rortiss : equal parts) - J ' 
This solution is a yellowish red colour. A little silver solution is placed on the 
slide. Then some tannate. It is rocked to and fro for a short time and then 
washed off with water. The process is repeated three or four times until the film 
appears deep brown or black. This formula is somewhat different from my previous 
ones, but the exact proportions do not seem very important, except that it is easier 
to obtain absolutely clean specimens with the strong ammonia solutions. The 
flagella are beautifully stained. The emulsion used should be just faintly milky, and 
I find it best to flood the slide with the solution and then allow it to drain away. 
