752 
MR. A. SANDERS ON THE ANATOMY OF THE 
longitudinal section are seen to be as follows. Commencing at the posterior end of the 
optic lobe, with wdrich they are connected by a narrow isthmus (the sole representative 
of the structure so highly developed in Teleostei), they all pass forward to form 
the ventral wall of the anterior lobe; arrived at the anterior point they turn back and 
constitute the dorsal wall of the anterior and posterior lobes, between which there is 
often found a small fissure ; from the posterior end they again pass forward and form 
the ventral wall of the posterior lobe. Arrived at the foramen of communication with 
the fourth ventricle they turn towards the ventral surface, and then backward, supplv- 
ing the posterior part of the roof of that passage, and finally turn upward to end close 
beneath the lower surface of the posterior lobe ; here they form a curved free process 
which may be termed the inferior lobe. Busch # did not recognise the cerebellar 
structure of this lobe, and considered it as merely an appendage to the restiform 
bodies, “ Fimbria ex corpore restiforme orta.” 
The four layers are not equal either in superficial extent or in mass (fig. 17). The 
external or molecular layer is deficient both at the anterior and posterior end for a 
small space; behind it continues from the under surface of the posterior lobe into the 
upper surface of the inferior lobe ; at the point of junction of the two it forms a fold 
of its own, not participated in by the other layers (except the intermediate), w 7 hich 
fills up the angle between the lower surface of the one and the upper surface of the 
other lobe. 
The intermediate or Purkinje cell layer is scarcely separable from the molecular, and 
may be looked upon as part of the same, it follows all its windings, and is deficient 
where that is deficient. 
The third or fibrous layer may be traced at the junction of the anterior end of the 
cerebellum with the optic lobe into the posterior end of the latter, where it enters 
the third layer close to the endothelium ; it consists of a longitudinal bundle of fibres, 
which passes along the ventral wall of the posterior lobe, and is dispersed and dis¬ 
appears in the anterior end of the granular layer. It may, perhaps, be looked upon 
as the representative of the crura cerebelli ad cerebrum. In the dorsal wall this 
layer forms two subordinate strata, the external of which consists of transverse or 
oblique, while the internal is formed of longitudinal fibres ; in some parts there is a 
longitudinal layer external to as well as internal to the transverse fibres; these latter 
are derived directly from the crura cerebelli ad medullam, whereas the longitudinal 
gain their position more indirectly ; they disappear in the granular layer anteriorly 
and posteriorly. Both sets of fibres are traceable into the inferior lobe. 
On inspecting a transverse section of the cerebellum (fig. 13), it is seen that the 
granular layer consists of four longitudinal tuberosities or tori, which project into the 
ventricle of the cerebellum ; they are placed one on each side of the central line of 
the dorsal wall, and in a corresponding position on the ventral wall. In the interval 
between these tori this layer is absent, and the wall at these parts is formed onty by 
* Op. cit., Tab. 1, fig. 4. 
