The tumor produced by the injection of a Berkefeld filtrate of our transmissible chicken-sarcoma is first noticed as a minute nodule at some point in the track of the injecting needle;1 and only a small proportion of the fowls injected develop the growth. On the other hand, when the causative agent has been introduced in the form of dried and powdered tumor-tissue, suspended in Ringer's solution, the sarcoma appears as a more or less diffuse mass at the site of injection; and it is found in many of the fowls. These facts have led us to suppose that the filterable causative agent requires for its action a cell-derangement or proliferation, such as the needle-prick or the presence of dried tissue induces. Experiment shows that this is indeed the case.
A number of susceptible fowls were injected, in one "breast" with a large quantity of active filtrate, in the other