After this, he heated up the virulent disease to kill it and then injected it into a mouse. This time the animal survived as predicted. Last he injected non-virulent pneumonia and virulent pneumonia that had been heated and killed, into a mouse. This time the mouse died.
Griffith speculated that the killed virulent bacteria had passed on a characteristic to the non-virulent one to make it virulent. He believed this characteristic was in the inheritance molecule. This passing on of the inheritance molecule was what he called transformation.