114 • Alternatives to Animal Use in Research, Testing, and Education 
• reduced ability to study behavior; 
• reduced ability to study the recovery of 
damaged tissue; 
• reduced ability to study interaction between 
the organism and its environment; 
• reduced ability to study idiosyncratic or 
species-specific responses; 
• reduced ability to distinguish between male- 
and female-specific phenomena; and 
• a handicap to probing the unknown and phe- 
nomena not yet identified. 
This general listing of advantages and disadvan- 
tages provides a framework for examining the use 
of alternatives in specific disciplines of biomedi- 
cal and behavioral research. Many of these pros 
and cons are cited in this chapter’s detailed descrip- 
tion of alternatives. 
CONTINUED, BUT MODIFIED, USE OF ANIMALS 
IN BIOMEDICAL RESEARCH 
Animal use in biomedical research can be modi- 
fied in a number of ways, including strengthen- 
ing experimental design to use fewer animals, re- 
ducing the degree of experimental insult, and 
substituting one organism for another. In the 
case of substitution, cold-blooded vertebrates may 
supplant warm-blooded ones. 
Reduction in the Number 
of Animals Used 
Up to half the animals used in research protocols 
may be untreated, or control, animals. The impor- 
tance of using parallel, internally controlled de- 
signs for experimentation may be one of the first 
lessons learned by science students whose results 
are rejected for not providing comparable data 
from treatment groups of animals matched for size, 
age, sex, and dosage. Studies with investigator- 
initiated, internal controls support substantially 
stronger inferences than those without them. 
Common and Historical Controls 
Fewer animals may be used in an experiment 
by sharing a control group with other investiga- 
tors or by not using a concurrent control group. 
In both cases, all the physical and genetic charac- 
teristics of the treatment group (s) must be matched 
to those of the control group, and the conditions 
under which the data are collected must be as pre- 
cisely duplicated as possible. There are difficul- 
ties unique to each method. Investigators may en- 
counter constraints on their particular study when 
sharing controls. For example, sharing may be im- 
possible if one group needs to extend its studies 
beyond the time agreed for termination and au- 
topsy of the shared animals, or if the actions of 
one group adversely affect the other, as might hap- 
pen by the inadvertent spreading of a parasite or 
pathogen. In the case of historical controls, the 
difficulty rests in exactly duplicating earlier con- 
ditions. Use of such controls must be carefully doc- 
umented and justified (82). 
Animal Sharing 
Another way to use fewer animals is to share 
individual experimental animals or their tissues 
between research groups. Although this method 
may encounter the same types of difficulties de- 
scribed for the sharing of controls, it appears to 
be gaining in popularity among compatible groups . 
At the University of Virginia, investigators in en- 
docrinology (Department of Internal Medicine) and 
in the molecular genetics of heme synthesis (De- 
partment of Biology) use the pituitaries and livers 
of the same rats even though the two departments 
are on opposite sides of the campus (54). 
Research animals may also be shared among dif- 
ferent sites. This is especially practicable in the 
case of long-lived primates. As long as sequential 
protocols are not deemed inhumane or scientifi- 
cally conflicting, primates may be shipped from 
one research site to another. The Primate Research 
Institute (PRI) of New Mexico State University , for 
example, will loan chimpanzees and rhesus and 
cynomolgus monkeys to qualified U.S. scientists. 
PRI currently has 240 chimpanzees on campus, 
with another 150 animals on loan. 
