200 J. B. JOHNSTON, 
found in all Vertebrates investigated (VAN GEHUCHTEN, P. Ramon, 
S. Ramon), except Mammals. Their neurites take a horizontal course 
and without doubt leave the tectum. These cells are probably closely 
allied to those last mentioned. 
Thus it seems possible to reduce all the cells described by the 
CAJALS to three or four generalized types such as are found in Aci- 
penser. With regard to the large number of layers described by these 
authors, it must be said that they have no physiological significance 
and are rather confusing than otherwise. VAN GEHUCHTEN de- 
scribes only three layers in fishes and birds, and KÖLLIKER recog- 
nizes six in the chick and five in man. These layers probably have 
some physiological significance and are at least an aid to de- 
scription. . 
In all Vertebrates there are three main classes of cells in the 
tectum: II type cells, cells sending neurites to the retina, and cells sen- 
ding neurites to other parts of the brain. The II type cells are the . 
chief ones to receive impulses from the optic fibres, and they are con- 
sequently very numerous in lower Vertebrates. In the higher Verte- 
brates, where many of the optic fibres end in the thalamus nuclei, 
the II type cells are relatively few. The cells giving rise to efferent 
fibres are the chief ones to receive impulses from the secondary 
sensory tracts. These have become relatively much more numerous 
in the higher Vertebrates and have doubtless been differentiated 
into several varieties with reference to their functions. The collaterals 
on the efferent fibres are much better developed in the higher Verte- 
brates than in the lower, so that they must to a considerable extent 
take the place of the II type cells in the function of association. 
The efferent fibres have shifted their position from the middle zone 
in fishes to the internal zone in higher Vertebrates, displacing out- 
ward the II type cells and those whose neurites go out in the optic 
nerve. 
Torus longitudinalis. — Sata (95) has studied the torus. 
longitudinalis of Teleosts (Tinca) by the method of Gouer. The cells 
are small, 10—14 u, irregularly disposed, and have each an irregular 
dendrite which divides and subdivides. The neurites form a bundle 
running upward and outward, dorsal to the dorsal decussation. The 
bundle breaks up into several bundles which enter the outer (optic) 
fibre layer. The fibres. probably go to the optic nerve. I have de- 
scribed these cells above (page 103) as if they were common cells of 
the tectum, although restricted to the mid dorsal region. The cells 
