Page:Blood examination.djvu/22

10 into error of diagnosis, for, although in both diseases we see absence of leucocytosis, yet in malaria we should find the parasites and the large mononuclear increase, whilst in typhoid we should expect a relative increase of the lymphocytes, which progressively go up as the disease advances. The Bacillus Typhosus can also be cultured in bouillon from the the blood in 78 per cent, during the first week on the average of 700 cases, and in severe cases for longer.

The Widal serum reaction, if obtained, will finally settle the diagnosis, for Cabot has collected 4339 cases of undoubted typhoid, divided amongst eighteen physicians, all of whom had performed at least 100 Widal reactions, and he finds that the error is only 2 per cent. The reaction can usually be first observed from the sixth to the eighth day—a dilution of 20 should give the characteristic clumping in ten minutes—inquiry should be made as to whether the patient has had typhoid before, for if so the reaction may occur as a result of that attack.

A rapid diminution of hæmoglobin is in favour of malaria as opposed to typhoid or pneumonia. In pneumonia, also, more often than not, there is a leucocytosis.