The Rate of Tumor Growth in underfed hosts. 129 
experiment animals. Preliminary observations were made on 
several series of healthy rats to determine how little of the bread 
would sustain life while permitting of a gradual emaciation, and 
then no young, growing animals, carrying the Flexner-Jobling 
tumor, were submitted, half to food restriction, while the other 
half were fed full and used as controls. The food was carefully 
measured out, the exact size of the tumors charted each week, and 
the rats weighed twice a week. Moreschi observed the effect of 
food restriction from the time of the tumor's implantation or 
shortly thereafter until the host's death. For our purpose animals 
were mainly used in which the tumor had attained a large size before 
the restricted diet was begun. Contrary to expectation it was 
found that these large tumors continued to grow with the same 
rapidity in hosts emaciating on a restricted diet as in the controls, 
many of which were still gaining weight. With yet larger tumors 
and hosts already cachectic food-restriction had also no effect 
on the primary growth, and none on the frequency of metastasis 
formation. 
Moreschi's findings were confirmed so far as regards the in- 
fluence of poor nourishment of the host to hinder the develop- 
ment of tumor grafts. And when food restriction was begun a few 
days (four) after the introduction of the grafts it was rioted that 
the tumors developed a little more slowly than in the controls- 
It is easy to understand how, given such a retarding influence on 
the tumors early in their career, the experiment animals might 
outlive the controls, as Moreschi found to be the case. The 
sarcoma he used was probably more susceptible than our carcinoma 
to changes in the host's nutrition. 
The fact that large tumors take their course irrespective of the 
condition of nourishment of the host does not at present admit of 
direct explanation any more than does the phenomenon of re- 
generation in starving animals. It is to be correlated with the 
clinical observation on malignant disease that the tumor often 
grows with great rapidity despite a marked progressive emaciation 
of the patient. Our experiments have not yet answered the impor- 
tant question as to whether in animals from which the primary 
tumor has been removed the metastases will behave like tumor 
grafts in that their growth is slower in ill-nourished hosts; or 
