silk bolting cloth, and mounting small samples 
on microscopic slides. Ten random spots on 
each slide were studied at 100X magnification, 
and the percentage volume of each plant was 
estimated (Keith et al. 1959; Meyers and 
Vaughan, 1965; Ward and Keith 1962). This 
method was also described by Williams (1962) 
for use in studying microtine food habits. 
Hayden (1966) used a similar technique to 
examine black-tailed jackrabbit (Lepus cali- 
fornicus) stomach contents. He estimated per- 
centages of each plant material on 80 fields of 
vision at 100X on five slides containing one- 
third ec. contents. Values for three slides were 
averaged for each animal, and individuals were 
averaged to make a composite monthly sample. 
Bear and Hansen (1966) used a similar tech- 
nique in their food habits study of white-tailed 
jackrabbit (Lepus townsendit). 
Malechek (1966) overcame the problem of 
having many different sizes of food particles in 
steer rumen samples by first grinding oven- 
dried rumen samples in a Wiley mill equipped 
with a 1-mm. screen. He then mounted small 
subsamples of the finely ground material on 
standard microscopic slides, and analyzed them 
under a compound binocular microscope at 
125X. The botanical composition of each sam- 
ple was determined by the relative number of 
epidermal fragments of each species recog- 
nized in 100 microscope fields. 
Sparks (1967) used the microtechnique to 
study the food habits of black-tailed jackrab- 
bits in Colorado. Contents were washed and 
mixed in warm water, and then dried and 
ground in a Wiley mill over a 20-mesh screen. 
The material was washed again over a 200- 
mesh screen to insure mixing and to remove 
dirt and very small fragments. One slide was 
prepared from a sample of each stomach con- 
tent. The percentage of each food item in the 
diet was estimated by examining 20 syste- 
matically located fields on each slide with a 
binocular-compound microscope at 125X mag- 
nification. Average frequency percentages 
were computed for all species in the composite 
sample of 25 jackrabbits (500 fields). The fre- 
quency percentages were then converted to 
density per field (Curtis and McIntosh 1950), 
and percent composition of each food item was 
calculated for the sample. 
The use of microscopic slide techniques has 
been evaluated, but usually in connection with 
large-scale food habit studies. The first work 
with pocket gophers by the author was done 
very cautiously. Known plants were fed to 
caged gophers, and then the animal was killed 
and the stomach was examined to see if the 
plant could be identified. A few animals were 
fed mixed diets of known plant material, and 
the validity of the microscopic method was 
checked. 
152 
Several recent studies have confirmed the re- 
liability of the microtechnique for determining 
the vegetative composition of fistula and stom- 
ach content materials. Although he was not 
able to obtain samples from fistulated steers, 
Denham (1965) compounded forage samples 
from six plant species to simulate fistula for- 
age samples, and had them examined by micro- 
techniques. Samples were prepared for analy- 
sis by first oven drying, and then by grinding 
through a 1-mm. mesh. A small subsample of 
the material was mounted on a microscope 
slide. The analysis involved observing and 
identifying material at approximately 100 loca- 
tions on each of two slides. When all six spe- 
cies were correlated with the percentage 
weight of each expected in the sample, a very 
significant correlation of r = 0.97 resulted. 
Denham concluded that equally high degrees 
of accuracy should be achieved from analysis 
of samples collected through esophageal fistu- 
ae. 
Sparks and Malechek (1968) accurately esti- 
mated percent composition by dry weight for 
15 hand-compounded mixtures of plants that 
are found in the diets of some herbivores. The 
mixtures were sampled by recording the fre- 
quency of occurrence of each species in 100 mi- 
croscope fields under 125X magnification, con- 
verting frequency to density, and calculating 
relative density as an estimate of percent com- 
position by dry weight. Dry weight percent- 
ages could be predicted directly from relative 
density. The two requirements that had to be 
met before frequency percentage could be con- 
verted to density were (1) that the plant frag- 
ments must be distributed randomly over the 
slide, and (2) that the density of particles 
must be such that the most common species 
does not occur in more than 86 percent of the 
microscope fields. They concluded that the mi- 
croscopie technique they used is an accurate 
means of determining the dry weight composi- 
tion of stomach samples, esophageal samples, 
rumen samples, and clipped herbage. 
Dusi (1949) found that fecal pellet material 
must be mounted on microscope slides for anal- 
ysis. Only fresh droppings were used. In gen- 
eral, he found that the contents of each of the 
several fecal pellets of rabbits in a dropping 
pile were like others in the pile unless the pel- 
lets varied greatly in color, shape, and size. 
Further statistical testing showed that an ade- 
quate sample consisted of about one-eighth of 
an average size pellet broken into particles, 
spread evenly over a microscope slide, and cov- 
ered with a three-fourths-inch-square coverslip. 
Each slide was examined under a compound 
microscope. The identified particles were re- 
corded and treated by frequency of occurrence 
methods. 
Stewart (1967) examined the qualitative and 
a end 
