The material for this problem was selected from control sections used for previous physiological studies on dogs (Allen, '38, '40). It consisted of many Marchi and Nissl stained sections of Ammon's horn and several Marchi series through the hypothalamus. These structures were removed 2 to 6 weeks after the fornix had been transected.
In transecting the fornix, a hemisphere was retracted to one side to e-qose the corpus callosum, through which a narrow straight-backed scalpel was inserted in midline below the level of the fornix and then moved sufficiently lateral to sever the fornix. One hippocampus and fornix were removed in two pieces after slitting the ectosylvian fissure into the ventricle and lifting up the cephalic portion with a spatula. The microtechnic has been described in an earlier paper (Allen, '19).
TERMINATION O F THE FORNIX I N THE HYPOTHALAMUS
Probst ('01) after cutting the column of the fornix, traced Marchi degenerated fibers to the mammillary body and adjacent olfactory field.
A rich connection was described as taking place in the mammillary body through the Vicq d'Azyr bundle to the median nucleus of the thalamus and Gudden 's Haubenbundel to the midbrain. Yamagata ('27, 27a) after extirpating one hippocampus traced the fornix by Marchi sections to the lateral part of the mammillary body. Koelliker (1894,1896)' Ram6n y Cajal ( '11)' Smith (1896), Malone ('12)' Winkler and Potter ( '14), Hilpert ( '21), Greving ( '23)' Griinthal ( '29)' Rioch ( '29), Loo ( '31), Krieg ( '32), Fuse ( '38) and others have shown the termination of the fornix in the mammillary body or in the median nucleus of the same with Nissl, silver or Weigert preparations. Koikegami ('38) noted from Weigert sections that the mammillothalamic tract arose from the nucleus magnocellaris corporis mamillaris, while the fornix terminated in the parvocellular nucleus. A 283