[Skip to Navigation]
February 11, 1905


JAMA. 1905;XLIV(6):478-479. doi:10.1001/jama.1905.02500330046006

The relation which exists between certain forms of bacterial infection and the increase in certain varieties of leucocytes in the blood has been known for many years, and has been taken advantage of in diagnosis. We know that septic infection is generally associated with a polynuclear leucocytosis, that in typhoid fever the mononuclear leucocytes are relatively increased, and that infection with intestinal parasites often causes eosinophilia. In recent years, the use of cytodiagnosis has brought out the fact that tuberculous effusions usually show an excess of lymphocytes, and has suggested that the tubercle bacillus elaborates a poison which is positively chemotactic to lymphocytes.

This suggestion is not a new one; in fact, a number of observers have claimed a lymphocytosis following the use of tuberculin. These observations are not numerous, and recently Fauconnet2 has deemed the subject worthy of more extended investigation. He states that in the early stages