THE RESEARCH DOG: RANDOM SOURCE OR COLONY REARED? 
S. Potkay and J. D. Bacher* 
Although unparalleled advances have been made in 
research laboratory instrumentation and methodology 
during this century the standard research dog of today 
is essentially the same as its counterpart of 50 or 100 
years ago. As an alternative to the continued use of 
dogs of unknown genetic background, age, tempera- 
ment and disease experience, The National Institutes of 
Health and various other facilities have undertaken to 
breed dogs selected to meet specific investigational re- 
quirements. Although such animals are generally more 
expensive on an individual basis than those obtained 
from vendors, it has been shown that overall needs for 
dogs can be reduced by their use. This presentation will 
summarize the experiences of the NIH Animal Center 
in breeding and rearing American foxhounds for re- 
search as compared with quarantining and conditioning 
random source dogs. The use of a canine blood donor 
colony as a means of reducing the total needs for dogs 
will also be discussed. 
INTRODUCTION 
On a national basis, more than a quarter of a 
million dogs are used in research annually 
(Table I).^-^ Of these, at least 85% originate 
from either pounds or animal dealers or both. 
Relatively few (1.3%) are donated to medical 
institutions for research. These data also indi- 
cate a growing tendency for dealers and re- 
search institutions to breed dogs specifically for 
research. It is the purpose of this communica- 
tion to discuss previous and current practices at 
The National Institutes of Health as they relate 
to procuring, quarantining, and conditioning of 
random source dogs and to suggest alternatives 
to this method of supplying the needs of the re- 
search community for dogs. 
RANDOM SOURCE DOG COLONY 
Random source dogs, for the most part, con- 
sist of unowned "wild" dogs, rejected hunting 
* Veterinary Resources Branch, Division of Research Services 
National Institutes of Health, Bethesda, Maryland. 
dogs and strayed or unwanted pets.'* They may 
be mongrel or purebred, and are in various 
states of health. The various steps required to 
bring the random source dog from its original 
setting into the laboratory are shown in Table 
II. The minimum period of time involved in 
these multiple transfers is approximately 10 
days, but the average is probably about 30 days. 
Exposure of the dogs to a variety of pathogens 
at each facility involved and stress stemming 
from frequent transportation and changes in 
social grouping, diet, husbandry and environ- 
ment are considered principal factors responsi- 
ble for the morbidity and mortality observed 
among random source dogs at the research fa- 
cility. These observations are illustrated by the 
data shown in Figure 1. During fiscal year 1965 
through 1967, dogs accepted for quarantine and 
conditioning suff'ered a combined mortality/ 
morbidity of about 65%."' It was recognized 
that 30 days were frequently insufficient to-a4^ 
quately condition dogs for investigational use. 
Consequently, during fiscal year 1968 through 
1970, the quarantine and conditioning pe- 
riod was extended to 45 days; coincidentally, 
criteria for accepting dogs into the facility were 
reformulated and adhered to more strictly. 
These actions were associated with a slight de- 
crease in morbidity and a substantial increase 
Table I 
APPROXIMATE NUMBERS OF DOGS USED FOR BIOMEDICAL 
RESEARCH IN THE UNITED STATES*: 1967-1969 
Year 
Soldi' 
Bredt 
From Pounds^ 
Donated 
Total 
1967 
247,753 
17,599 
105,672 
371,024 
1968 
269,584 
32,417 
107,471 
409,472 
1969 
220,546 
43,723 
103,028 
4,932 
372,229 
Totals 
737,883 
93,739 
316,170 
4,932 
1,152,725 
* Adapted from ILAR News 
' Based on an average 63.2% return of questionnaires 
t Based on an average 59.9% return of questionnaires 
1061 
