506 
Journal of Agricultural Research 
Vol. XXXI, No. 6 
In making the suspensions, the even dispersion of the soil particles 
by the vibrator was very evident. The suspensions did not contain 
any of the large particles that were common in the hand-shaking 
series, and on the whole the suspensions seemed more uniform. This 
might be expected, since there was no current to carry over the 
heavier particles and only those particles which were dispersed by 
the vibration could pass into the receiving flask. 
The residue in the vibrator cup naturally varied with the sample. 
Of 1 gm. of soil No. 109, 0.58 gm. remained in the cup, 0.42 gm. 
having become suspended in the diluent. For soil No. 260, the 
figures were 0.38 gm. and 0.62 gm., respectively. 
MODIFIED MEDIA. FOR PLATING 
Preliminary tests made some years ago with ComTs asparaginate 
agar (1), variously modified media as recommended by Lipman and 
Brown (4), and soil-extract agar (5, p . 101), indicated that the 
highest and most uniform counts were obtained on soil-extract agar. 
This latter medium was therefore adopted in this laboratory for 
making routine counts of the microorganisms in the soil. The 
mannite-salts agar used by Whittles was originally recommended by 
Thornton (10). Three advantages were ascribed to its use—that it 
inhibits spreading colonies, that it is uniform in composition, and 
that the results can be produced with different batches of the medium. 
Unfortunately, no comparative results obtained with a known medium 
have been presented, so that its efficiency, aside from its ability to 
repress spreading colonies, remained in doubt. The few tests 
recorded m Table III were not very promising, but it seemed desir¬ 
able to use this medium along with others in making routine platings 
of a series of soil samples taken from small squares at frequent 
intervals. 
COMPARISON OP MANNITE-SALTS, EGG-ALBUMEN, AND SOIL*EXTRACT AGARS 
Dilutions for plating were made by adding 10 gms. of soil to 250 
c. c. sterile tap water, and, after vigorous shaking for about three 
minutes, 10 c. c. of the suspension were added to 90 c. c. of sterile 
water. This was repeated until a dilution of 1: 250,000 was reached. 
One cubic centimeter of this dilution was then pipetted into the 
Petri dish, melted agar was poured in, and the plates were incubated 
for seven days at 28° C. 
The second part of Table IV contains the results of the platings 
made August 13 to 27 from squares 21 to 46, inclusive, on soil-extract 
(S. E.) and mannite-salts agars (M. S.). 8 Four plates were poured 
with each medium, the figure for each square being the average of 
these four plates in millions of microorganisms per gram of dry soil. 
The counts on soil extract agar (S. E.) are from 2 to 6 times higher 
than those on the mannite-salts agar (M. S.), the average of the 26 
tests being 30.1 and 8.0, respectively. 
8 The mannite-salts agar is made as follows: K 2 HPO 4 , 1 gm., KNO 3 , 0.5 gm., asparagine, 0.5 gm. Dis¬ 
solve in water, and add 0.2 gm. MgS0 4 , 0.1 gm. CaCl 2 , 0.1 gm. NaCl, and 0.002 gm., FeCh in the form of 
standard solutions. Fifteen grams agar is then added and dissolved, the volume is made up to 1,000 c. c., 
and the agar is filtered at 100° C. through cotton. One gram of mannitol is added to the filtrate, and this 
is cooled to 60° C., and adjusted to Ph 7.4. Sterilization in the autoclave. 
