The blood picture in uncomplicated influenza is a normal one for the red cells and the hæmoglobin, but the white cells are characteristically altered. We have made many observations on the red blood cells, and from all aspects the picture appears to be normal. Similarly, there is nothing significant about the hæmoglobin estimations. Where we have slight alteration in the red count and in the hæmoglobin it is probably safer not to attribute the change to the epidemic. We have no records showing a secondary anæmia due to the initial epistaxis. A leucopenia or a normal count is what one should see in most of the uncomplicated influenzal cases. We are almost ready to say that any estimation above normal limits means secondary bacterial invasion. The count may remain low throughout the illness, rising to the normal rapidly as the temperature falls. We do not regard a leucocytosis at the end of an epidemic case as part of the blood picture. Our experience is that with convalescence the normal count returns and remains within normal bounds. Very often hidden sinus infection is responsible for some of the post-influenzal leucocytoses. The leucopenia may vary from a slightly subnormal count to a point well below 2,000. Most of the simple epidemic cases showed some degree of leucopenia. As far as we have been able to estimate, we are led to believe that one should not lay any special stress on the grade of leucopenia as being of prognostic significance in uncomplicated influenza. Many of the mildest clinical types showed very low counts, and _vice versa_. There is, however, a prognostic relation to be noted with reference to a falling white count in the pneumonia, but this we shall mention again later. The onset of the leucopenia corresponds to the onset of the disease. It was present with the earliest cases we examined, and remained fairly stationary, although we have records of its fluctuating slightly one way or the other. But one must remember in this regard the personal error in blood counting, and also particularly the error of the apparatus. For careful work only those counting chambers and pipettes should be used that have a Bureau of Standards certificate. The duration of the leucopenia was fairly close to the duration of the disease.