96 J. B. JOHNSTON, 
lobus lineae lateralis and runs through the acusticum probably to 
the base of the medulla. This bundle has been described above 
(page 83). Farther forward another large bundle takes a similar 
course toward the base of the medulla, passing on the external sur- 
face of the cerebellar crest (Pl. 13, bundle marked?). But these 
bundles account for only a small part of the decrease which takes 
place in the molecular layer. No considerable number of fibres leave 
the molecular layer to join the granular layer, but instead the mole- 
cular layer continually receives large numbers of fibres from the 
eranular layer. All these considerations make it seem almost cer- 
tain that the greater number of the fibres of the molecular layer end 
in relation with the rough dendrites of the PURKINJE cells which 
lie imbedded in the molecular layer. I turn now to the description 
of the nerve elements of the lateral lobes. 
The granular layer. — The granule cells of the lateral — 
lobes measure 8—12 by 12—18 u, and are in every way like those 
of the body. They send their neurites into the molecular layer, 
where they turn caudad. These cells are extremely numerous and 
closely packed around the free caudal border of the lateral lobe 
and in the thin band which connects it with the lobus lineae lateralis. 
The cells which form this connecting band send their neurites, at 
least in part, forward along the dorsal surface of the cerebellar crest. 
It is probable that they turn into the crest and bend backward to 
run in the same direction as the other fibres of granule cells. I 
have found a few cells which send their neurites caudad among the 
ascending fibres of the VIII. These are to be classed with the cells 
in the body whose neurites run in the granular layer. Cells of 
the II type are much more numerous near the junction with the 
body than elsewhere in the lateral lobes, and those whose neurites 
present the club-like thickenings I have found only in this region, 
The PuRKINJE cells present the same appearance and the same 
variations in form and size as do those of the body (Phots. 40, 41). 
The differences in form affect the measurements of these cells in 
the same manner as in the body, and they may be assigned to two 
groups measuring 16—20 by 32—48 u and 8—12 by 96—288 u. 
The cells are arranged with some regularity at or near the periphery 
of the granular layer and the dendrites ramify in the molecular layer. 
The parts of the lateral lobe in which the molecular layer is wanting 
is without PURKINJE cells also. There seems to be no constant 
relation between the disposition of the dendrites and the planes of 
