Several forms of point analysis have been used 
to provide estimates of volume, weight, and 
composition of food samples. 
The most appropriate method of assessment 
and presentation of data has been one of the 
controversial issues in the study of animal 
foods (Hartley 1948). Diversity of opinion 
began early (Beal 1897; McAtee 1912) and, to 
a lesser extent, continues (Jensen and Korsch- 
gen 1947; Brown 1961; Dirsch] 1962; Scotter 
1966). Various methods have been rejected 
because these imperfectly depict food consump- 
tion by means of a single quantitative expres- 
sion. 
Number Tabulation 
Tabulation of numbers of food items alone 
has been criticized as not being a sufficiently 
comprehensive indication of diet (McAtee 
1912). Certain foods, such as carrion, fruit 
pulp, sap, and fragmented vegetable material, 
cannot be assessed numerically, nor can allow- 
ance be made for different sizes of food items. 
A record of food items taken, however, re- 
mains an appropriate and acceptable method of 
appraisal for some purposes. Numbers of prey 
taken are often emphasized in studies of preda- 
ceous species (cf. Gross 1944). 
Occurrence Tabulation 
A statement of the frequency of occurrence 
of food items has been used by many investiga- 
tors and is considered an important interpre- 
tive index. In some studies it has been used ex- 
clusively (cf. Johnson 1961), but it is more 
often associated with other descriptive expres- 
sions. Gilfillan and Bezdek (1944) believed 
that a record of food-item occurrences pro- 
vided a more accurate picture of ruffed grouse 
food preferences than did volumetric data. 
Volume Measurement 
Volumetric methods of appraisal were em- 
ployed early in the development of food-habits 
research (McAtee 1912), and volume continues 
to be perhaps the most consistently applied 
descriptive statistic. Volumes may be estimated 
visually by assigning relative percentage pro- 
portions to sample materials or may be mea- 
sured directly by water displacement in vari- 
ous sizes or modifications of graduate cylinders 
(Martin 1949). The method employed depends 
on the nature of the sample material. Volume 
of discrete and readily segregated items may 
be measured by water displacement. Percent- 
age calculations based on visual estimates are 
made when food items are so small, mixed, and 
fragmented as to make segregation impracti- 
cal, or are too large or irregular to fit into 
glass graduates. 
136 
Martin et al. (1946) compared the “aggre- 
gate percentage” and “aggregate volume” 
methods of summarizing data. Aggregate per- 
centage methods are used when volumes are es- 
timated and percentages are the only figures 
available. Aggregate volume summaries can be 
used when volumes are actually measured and 
expressed in cubic centimeters. Most investiga- 
tors have used one of these two methods, and 
some (cf. Ward 1964) have used both. 
Hartley (1948) and Beck (1952) discussed 
some of the shortcomings of volumetric analy- 
ses and, along with others (McAtee 1912, 
Swanson 1940, Martin 1949, Brown 1961, Tal- 
bot and Talbot 1963, Martin and Korschgen 
1963), pointed out the limitations of any single 
quantitative description and the desirability of 
using two or more complementary expressions. 
Weight Measurement 
Assessment by weight has been considered 
the more laborious and more refined method of 
presenting food-habits data. Apparently until 
comparatively recently, only Rorig (1903) had 
expressed results by weight. Recent literature 
indicates more interest in using weight ap- 
praisals (Norris 1943; Jensen and Korschgen 
1947; Saunders 1955; Dirschl 1962 and 1963; 
Scotter 1966 and 1967). Objections to gravime- 
tric methods have been summarized by Hartley 
(1948) and Beck (1952). Martin and Korsch- 
gen (1968, p. 327) point out that weighing seg- 
regated materials after drying under condi- 
tions of controlled temperature and humidity 
“cannot eliminate all sources of error and has 
the handicap of requiring additional time and 
expense,”’ as compared to less refined methods 
of appraisal. 
Norris (19438, p. 246), however, used weight 
as “a more logical measure than volume in ad- 
dition to eliminating error in estimating volume 
percentage.” Scotter (1966) concluded that 
gravimetric methods offered several advan- 
tages over volumetric analyses in a caribou 
food habits study. In addition to being more 
rapidly determinable, weight values could be 
more easily related to forage yield and utiliza- 
tion studies and to nutritive content values ob- 
tained by chemical analysis of forage, since 
these values are usually expressed in terms of 
air-dry weights. 
Point Analysis 
Chamrad and Box (1964) describe an adap- 
tation of the point quadrat, often used in sam- 
pling vegetation, to estimate volumetric compo- 
sition of rumen contents. Comparisons with 
known volumes were within acceptable limits 
of error provided sample material was ade- 
quately mixed and there were no large items 
with unusual surface texture in the mixture. 
