S. POTKAY AND J. D. BACKER 
1063 
three weeks.' On such a schedule, each donor 
can supply 17 units of blood per year. Based on 
an average yield of two units of blood per ex- 
sanguinated random source dog, each permanent 
donor thus represents a savings of 8.5 dogs that 
would otherwise have to be exsanguinated. The 
current colony of 240 donors, therefore, can 
produce a quantity of blood equal to 2040 exsan- 
guinated dogs per year. 
CANINE BREEDING COLONY 
Dissatisfaction with the quality of random 
source dogs led to establishing a production col- 
ony of purebred foxhounds to provide standard 
research dogs for the NIH researcher. Some of 
the advantages of the colony bred research dog 
include : known age and genetic background and 
acceptable temperament. In addition, morbidity 
and mortality associated with common canine 
diseases, including helminth and protozoan par- 
asitisms, can largely be eliminated. Production 
of uniform, purebred dogs also lends itself to 
ease in assembling baseline reference data per- 
taining to anatomy, physiology hematology and 
biochemistry for specific sex, age or other 
groups. Finally, litter-mates or other closely re- 
lated dogs can be made available to meet specific 
investigational needs. 
The argument generally employed against the 
use of dogs specifically bred for research is that 
their initial purchase cost is more than that for 
random source dogs. While this is true, it is an 
oversimplification of the situation, for it fails to 
take into consideration all of the expenses asso- 
ciated with utilizing dogs in research. 
Itemized operative and postoperative expen- 
ses for open heart experiments at one research 
facility indicated that the cost of the dog was 
only 6.7 % of the total expense.^ Considering 
that overhead, professional salaries and related 
expenditures (e.g., clinical pathology, necropsy, 
photography, etc.) were not included in arriv- 
ing at this figure, the cost of the dog becomes an 
even less significant factor. Another important 
consideration is the stamina and viability of the 
research dog. Reported observations that the 
survival rate of Laborador retrievers was sub- 
stantially higher than that of healthy, condi- 
tioned random source dogs undergoing open 
heart surgery^ are similar to opinions ex- 
pressed by NIH researchers using purebred 
foxhounds and random source dogs. In this re- 
gard, arguments against using purebred dogs 
become even less tenable. 
Acceptance of the American foxhound as a 
standard research dog at the NIH seems well 
established and current demands exceed produc- 
tion. The original pilot canine breeding colony 
has been gradually expanded and presently con- 
sists of 130 bitches and 7 stud dogs. Production 
has increased to the extent that the number of 
colony bred foxhounds issued exceeds that of 
random source dogs. 
CONCLUSIONS/SUMMARY 
The interaction of the three programs de- 
scribed above is illustrated in Figure 2. The 
graph, which shows the numbers of dogs issued 
and those maintained for production purposes, 
demonstrates the sharp decline in the use of 
random source dogs and in the overall numbers 
of dogs required to meet the total needs of the 
NIH. In evaluating these figures, it should also 
? 3 
I I Random Source Dogs 
Conine Blood Donor Colony 
Conine Breeding Colony 
[ j Colony Reared Dogs 
1964 1965 1966 
1967 1968 1969 
FISCAL YEAR 
1970 1971 1972 
Figure 2. — Total numbers of dogs maintained and issued 
for biomedical research at the National Institutes of 
Health; fiscal years 1964-1972. 
