J. COCHIN 
703 
of radiant heat on the skin of the dog. These 
methods are the direct descendents of the Har- 
dy-Wolff-Goodell technique^ in man — methods 
that did not prove terribly successful in man 
and that are not much better in animals. 
The last thermal technique that I will men- 
tion is that developed in its final form at NIH 
by Eddy and co-M^orkers.** This is the hot-plate 
procedure. In this assay, rats or mice are placed 
on a plate that is heated to a constant 
temperature — 55 °C, for instance. The time to 
end point is either a lifting of the hind leg, or a 
licking of the front pavi^s depending on the labo- 
ratory using this assay. In the hands of a com- 
petent, trained observer the results obtained by 
this method can be reproduced within a number 
of laboratories and are comparable between lab- 
oratories. It also has the advantage of a rela- 
tively long cut-off time since there is minimal 
tissue injury and the same animal can be tested 
a number of times — a procedure that is neces- 
sary in long-term studies of tolerance and de- 
pendence. Time-effect curves can be calculated 
over a period of many hours after injection of 
an analgesic, and changes in drug sensitivity 
can be followed with repeated testing for weeks 
or months. 
In the course of studies of long-term effects 
of analgesic drugs and of the mechanisms in- 
volved in tolerance^^'i^ thousands of hot-plate 
assays have been carried out by the author, his 
technicians, and his graduate students. The 
agreement of values obtained for areas under 
the time-effect curves between observers and 
over a period of years has been extremely good 
and has served to substantiate our opinion of 
the reliability and reproducibility of the assay 
in both the rat and the mouse. Since we are con- 
tinually doing long-term studies of tolerance, it 
is important that the assay method itself should 
not add a variable in the form of tissue injury 
which would confound the results and confuse 
the experimenter. The hot-plate technique is one 
which results in minimal tissue damage even 
when used repeatedly on the same animals and 
which makes it possible to use a sufficiently long 
cutoff time so that small changes in drug effect 
can be measured. Its drawbacks, shared by the 
other thermal techniques, are that it is not sen- 
sitive to the effect of analgesics of the agonist- 
antagonist type nor will it give positive results 
with the non-narcotic analgesics of the salicy- 
late class. 
Heat is not the only noxious stimulus that can 
be used to elicit a pain-like response from ani- 
mals. Pressure on the skin or distension of hol- 
low viscera have been used for years. The tail 
pinch described by Haffner^^ and by Eddy^^ 
has the advantage of extreme simplicity of 
both apparatus and observation, but this is 
offset by the disadvantages of tissue injury and 
lack of precision of end point and stimulus, 
making repeated observations difficult. The end 
point is vocalization and there is some argu- 
ment as to whether this is a better end point 
than the reflex movements elicited by other 
techniques. 
One of the most ingenious variations on the 
mechanical technique was devised by Randall 
and Selitto.^'* These investigators increased the 
sensitivity of the rat to the pressure stimulus 
by injecting a yeast suspension into the hind 
paw and applying pressure to the inflamed area 
produced by the yeast injection. Both the nar- 
cotic analgesics and the salicylate-type analge- 
sics increase the struggle threshold significantly 
— a struggle reaction being used as the end 
point rather than vocalization. The interpreta- 
tion of the results is confounded by the fact that 
most of the non-narcotic analgesics have antiin- 
flammatory properties as well, and there is no 
easy way of determining whether the drug 
acted by virtue of its antiinflammatory proper- 
ties or its analgesic properties. Nevertheless, 
this technique is capable of detecting com- 
pounds of the salicylate type and the mecha- 
nism by which this is done is of theoretical 
rather than practical interest. 
The chemical methods might be considered as 
logical outcomes of the yeast injection men- 
tioned above. These assay procedures measure 
the writhing induced by the intraperitoneal 
injection of chemicals such as 2-phenyl-l,4- 
benzoquinone,!^ acetic acid,^*' bradykinin,^'^ 
and hydrochloric acid^^ and the inhibition of 
this writhing by various analgesics. The "writh- 
ing test", which is also euphemistically called the 
"mouse peritoneal test", has caused a great deal 
of furor since writhing seems to offend the sensi- 
bilities of many investigators. Despite this it is 
