1967b) while others devoted most of the day to 
the job (Brown 1961; Healy 1968) (Marchin- 
ton and Baker *) The governing factors were 
the behavior of the deer and the length of time 
they spent actively feeding, as well as the par- 
ticular research needs. 
Where the research objectives call for semi- 
tame animals which are approachable in a 
large enclosure or on open range, special train- 
ing is required. Marchinton and_ Baker ° 
achieved this with a white-tailed buck fawn by 
initial bottle feeding followed by a period of 
minimal human contact and isolation from 
other deer, first in a box stall and then in a 
1.6-acre pen. This deer developed as an ap- 
proachable but independent animal well suited 
3 food-habits observations of free-ranging 
eer. 
Dunkeson (1955) combined free movement, 
approachability, and control over sampling 
area by releasing a pet deer in a 90-acre enclo- 
sure. Dcieciolowski (1966) observed his tame 
red deer hind in an enclosure including 
various forest conditions and permanent water. 
Where such enclosures exist or can be built in 
a suitable habitat, handling is minimal and 
training may consist only of bottle feeding 
until the animal is weaned. However, the cost 
of fencing sample areas of adequate size will 
often be prohibitive. Use of movable paddocks 
(Smith and Gaufin 1950) provides one solution 
to this problem. 
The ultimate achievement in close observa- 
tion of big game animals without interference 
might be that of Graf (1955), who was able by 
frequent contacts to condition wild Roosevelt 
elk to his presence at close range. This pro- 
gram was carried out in an Oregon State park 
where the elk had never been hunted. Brown 
(1961) observed black-tailed deer that had 
been raised by hand but that had been allowed 
to roam free most of their lives. These animals 
were essentially wild in habits, but still ap- 
proachable. Marchinton and Baker® located 
their deer in the woods daily by radio-tracking 
procedures. Such methods permit close observa- 
tion without frequent manipulation, but leave 
the choice of study area entirely up to the ani- 
mals. 
There is no apparent reason why elk, moose, 
or other big game animals should not be tamed 
and trained for studies of food habits and for 
other studies. Bison have been successfully 
trained as draft animals; however, they are 
somewhat untrustworthy (Garretson 19388). 
The use of reindeer as saddle, pack, and draft 
animals is well known (Zhigunov 1968). Tamed 
Aoudad sheep, sika deer, and axis deer are 
being used in Texas in the evaluation of food 
_*Marchinton, R. L., and Baker, M. F. Direct observa- 
tion & feeding behavior of radio instrumented deer. 
Unpublished personal communication, 1968. 
108 
habits of exotic ungulates (Ramsey 1968). 
With larger animals, however, much greater 
caution in handling will be necessary than has 
been the case with deer. 
EXPERIMENTAL CONTROL 
The paramount advantage of using tractable 
wild animals for observing forage intake is 
that observations can be planned in relation to 
specific interests—vegetation types, pasture 
treatments, seasons, etc. McMahan’s (1964) 
work is the outstanding example of the applic- 
ability of the technique in a complex experi- 
mental design. On the Kerr Wildlife Manage- 
ment Area (Texas Parks and Wildlife Depart- 
ment), a series of ten 96-acre pastures was 
used to study how different intensities of graz- 
ing by cattle, sheep, goats, and deer affected 
deer production. Pronounced differences in re- 
sponse of deer were obtained (McMahan and 
Ramsey 1965), but measurements of vegeta- 
tion or of utilization did not explain the rela- 
tionship of diets of various classes of animals to 
grazing treatments. To obtain this relationship, 
McMahan used tame animals of each class, 
grazing them daily for a year, according to a 
time-and-place design through four seasons in 
four pastures, which represented the range of 
conditions existing in the 10 experimental pas- 
tures. He used a factorial analysis of variance, 
the pertinent variables being animals, pastures, 
and seasons, to determine between and within 
differences in amounts taken of species and 
classes of forage in all combinations. From 
these analyses, he was able to define the nature 
of interspecific competition in pastures and 
seasons, and to conclude that the variety and 
abundance of forbs was the principal factor 
influencing the performance of deer. 
Watts (1964) obtained clip-and-weight sam- 
ples of total available forage in six arbitrary 
seasons, and graphically compared relative 
availability with relative use on a 4-acre area of 
hardwood forest in Pennsylvania. 
Work currently being conducted by the sec- 
ond author, Neff, is concerned with the effect 
of various kinds of vegetation manipulations 
on the welfare of game populations. To deter- 
mine the effect of treatments on the forage 
supplies of deer, treatments are applied and 
controls are retained, the response of under- 
story vegetation is measured, and the feeding 
of tame animals is observed. 
SUMMARY 
Numerous researchers have used tame or 
semitame pronghorn antelope, red deer, white- 
tailed deer, or mule deer to study forage con- 
sumption by these species. In a few cases, such 
