MATERIALS. 321 
sumptive test is only of value in suggesting suspicion, but insufficient to 
state the presence of sewage contamination. 
No. 28. Bacteria from the Root Tubercles of Legumes. The bacteria 
that have been commonly supposed to cause root tubercles may be obtained 
as follows: 
Add 5 gm. of wood ashes to 1,000 c.c. water, heat in steam, boil for 
one minute and filter. To the filtrate add 10 agar and 4 grm. maltose 
(or some other sugar) ; heat in steam till dissolved, boil one minute and 
filter. Place in test-tubes and sterilize by steaming on three successive 
days, or in an autoclave for one hour at 10 pounds pressu^g. 
Select some vigorously growing legume, not too large, and with a 
spade dig up its mass of roots still embedded in soil. Wash away the soil 
from the roots and probably plenty of tubercles will be found attached. 
Remove a small tubercle with forceps, wash in tap water and then immerse 
in a solution containing 500 c.c. water, i gram corrosive sublimate, and 2.5 
c.c. HC1. It should be immersed in this for two to three minutes to steril- 
ize the surface. Then the tubercle is held between folds of filter-paper that 
has been moistened in the sublimate solution and a gash is cut in it with a 
hot knife. A platinum needle is then sterilized and some of the central 
mass of the tubercle removed and placed in a drop of sterile water on a slide. 
Place a drop of sterile water in each of several Petri dishes and transfer a 
small loopful of the drop on the slide containing the tubercle contents, into 
the water drop in each of the Petri dishes. Pour into each a tube of the 
maltose agar, allow to harden and incubate at 70. The colonies will appear 
in three to four days and may be isolated and studied in the usual manner. 
No. 29. Bacteriological Analysis of Market Milk. Collect in sterile 
bottles a number of samples of market milk from different milkmen. 
Keep cool with ice until they are brought to the laboratory and then plate 
at once. Dilute 1,000, place i c.c. in each of two or three Petri dishes and 
add a tube of ordinary agar. Incubate at 98 for twenty-four hours, 
count the bacteria in each plate and determine the average. This is 
the procedure commonly used in making bacteriological analyses of mis- 
cellaneous samples of market milk. If there is any reason for thinking 
the milk is very old a higher dilution is necessary. If, on the other hand, the 
examination is to be made of milk known to be good and in cool weather , 
a dilution of 100 is better. 
No. 30. Bacteria in Sour Milk. Allow some milk to stand till sour, but 
not curdled. Dilute it 100,000 times (two 99 c.c. and one 9 c.c. water 
blanks) and make plates in litmus gelatin or litmus agar. Incubate at 
70 for two or three days and count the number of bacteria. Count the 
number of acid colonies. 
No. 31. Bacteria in Cream. Make plates as in No. 30 from fresh cream, 
diluting 1,000 times and using litmus gelatin and litmus agar. Make other 
plates from some ripened cream in gelatin and in agar and diluting 100,000 
times. Incubate at 70 for three days and compare the plates from the 
two lots of cream as to number of bacteria and relative proportion of 
acid bacteria. 
