BEHAVIORISTIC STUDY OF THE RAT 
9 
also ensured an accurate control of the amount of food eaten 
from day to day. Actual weighing of the food showed that the 
amount eaten in twenty minutes remains almost constant from 
day to day. The animals developed normally on this method of 
feeding. 1 
For the first part of the experiment eight animals were used. 
Normal activity records were taken for five days preceding the 
beginning of starvation. Four of the animals of this group were 
also deprived of water after the five days in order to obtain addi¬ 
tional information on the relation of activity to water intake. 
The amount of activity per day was obtained by counting the 
marks made on the smoked paper by each movement of the 
animal in the activity cage described above. The kymograph 
was set running just fast enough to enable the marks to be 
recorded individually. 
Three of the animals starved but permitted to have water all 
the time showed a definite increase in activity for the first two 
to three days after the beginning of starvation and then a steady 
marked decrease to a point of almost complete inactivity on the 
eighth day. All of the animals deprived of both food and water 
showed a steady marked decrease in activity immediately. This 
group reached the point of complete inactivity already on the 
fifth day. The results of this experiment are shown in figure 4 
where the days of the experiment are given on the abscissae and 
the total amount of activity per day in activity units (single 
1 In this present work a synthetic diet was used after the formula of Dr. E. V. 
McCollum of the School of Hygiene, Johns Hopkins University. This is an excel¬ 
lent diet. For behavior work on the rat it is ideal. 
Flour (graham). 72.5 
Casein. 10.0 
Milk powder (skimmed). 10.0 
Calcium carbonate. 1.5 
Salt. 1.0 
Butter fat. 5.0 
The importance of using a good diet for all animal behavior work was con¬ 
vincingly demonstrated by the findings of Dr. E. V. McCollum in connection with 
work on feeding problems. He found that the maternal reactions, nest-building, 
caring for young, nursing, are absent in animals fed on poorly balanced diets 
especially diets low in proteins. These animals often eat their young. 
