August, 1955 
For the three bunch grasses, 4d ndropo- 
gon scoparius, A. gerardi, and Bouteloua 
curtipendula, the shoots that grew as in- 
dividuals or in small bunches were counted 
and recorded. In the Phegley prairie, the 
shoots were small and mostly distinct and 
separate. In the Sampson prairie, the shoots 
were mostly aggregated into “tufts,” and 
these into bunches of varying sizes. In 
this prairie, 4. scoparius occurred also in 
several large patches within the plots. 
The number of shoots in such patches was 
estimated as follows: the average number 
of shoots per square inch in the smaller 
bunches was determined by counts; this 
number was multiplied by the number of 
square inches in each of the larger patches. 
During later stages of preparation of 
this report, it was evident that an esti- 
mate of numbers of grass-plant individuals 
per unit of area could be of great value 
in finding average plant densities in hill 
prairies. Other workers have made such 
estimates. Steiger (1930), in his study of 
high and low prairies of Nebraska, ap- 
parently counted each occurrence, whether 
a single shoot or a bunch, as a grass-plant 
individual. From the quadrat maps in his 
report, the bunches of grass appear not 
so large as those in hill prairies in Illi- 
nois. Korstian & Coile (1938), in a 
study of plant competition in forest stands, 
found the most densely covered forest- 
floor milacre had about 10 grass plants 
per 0.01 milacre. “Thus each 0.01 mil- 
acre occupied by a colony was regarded as 
fully stocked, even though it contained 
few or no individual plants.” Neither 
Steiger’s procedure nor the one followed 
by Korstian & Coile seemed adequate for 
Illinois hill prairies. The estimate of 
numbers of grass-plant individuals for 
these prairies was therefore made by a 
different method. 
Two bunches of Andropogon scoparius 
were obtained, one from the Northeast 
Meredosia prairie, one from the basal 
slope at Reavis Spring prairie. Each 
bunch was taken from an unpastured 
prairie strip that was separated from the 
adjacent pastured prairie slopes by a fence. 
These bunches were carefully dissected 
to find how many shoots were connected 
by living stems in what might be consid- 
ered as individual plants. For convenience, 
Evers: Hitt Prairies or ILLrNots 381 
each aggregate of shoots considered to be 
an individual plant was called a tuft. Be- 
fore separation, each bunch was mapped 
with a pantograph to show foliage area or 
crown cover, area at ground surface, lo- 
cation of solitary live shoots, and what 
appeared to be aggregates of shoots or 
“apparent tufts.” (It was recognized 
that separation of a bunch into tufts might 
give very different results from those of 
the preliminary surface examination.) It 
was found that the average number of 
shoots per tuft of 4. scoparius was 3.375 
for the Northeast Meredosia sample and 
3.793 for the one from Reavis Spring. 
The characteristic tuft (grass-plant indi- 
vidual) of 4. scoparius was found to be 
an aggregate of 3 or + shoots, average 3.5. 
The same method was used by H. A. 
Moore and A. G. Vestal to determine the 
number of shoots per plant individual for 
Andropogon gerardi. In a clump of big 
bluestem, collected by Moore from a rail- 
road trackway east of Urbana, Moore and 
Vestal found by separation that the aver- 
age number of shoots per tuft or grass 
plant was 1.739. The characteristic tuft 
(grass-plant individual) of big bluestem 
consisted of 1 or 2 shoots, average 1.75. 
As a preliminary step in estimating the 
number of plants of Bouteloua curtipen- 
dula, the writer obtained a small sample 
from Mud Creek prairie and carefully 
dissected the bunches to determine the 
number of shoots per plant or tuft. It 
was found that the average number of 
shoots per tuft was 3.437; the individual 
plant of B. curtipendula was an aggre- 
gation of 3 or 4 shoots, average 3.5. 
For each of the bunch-grass species, the 
estimated number of tufts in bunches was 
found by dividing the number of shoots 
in bunches by the average number of 
shoots per tuft, or by a factor based upon 
this average. 
Density of Vegetation.—The meas- 
ure of vegetation density in this study is 
the number of plants per milacre. ‘The 
mapped central milacre of the Sampson 
and that of the Phegley prairie were used 
in determining density for this report. 
Grasses.—In the central milacre of the 
Sampson prairie, 4,051 shoots of dndro- 
pogon scoparius grew in bunches and 
patches. The 4,051 shoots formed an es- 
