704 DR ALEXANDER BRUCE AND DR JAMES W. DAWSON ON 
and states that the first rudiments of the nerve roots appear as protoplasmic columns 
uniting the spinal cord to the muscle segments before any differentiation of the cells 
of the early neural tube has taken place. In these protoplasmic columns were found 
numerous nuclei which Dorn believed had migrated from the primitive medullary 
tube : the protoplasmic column later individualises into separate fusiform cells, which 
unite by their tapering extremities to form moniliform bands, and the protoplasm 
afterwards undergoes a differentiation into fibrils. The cells nearest the medullary tube 
fuse with the centrifugal process of the nerve cells in the cord. In this first stage there 
is no enveloping of the bundles by mesodermic cells; the latter penetrate later with the 
vessels and have no réle in the formation of the nerve tube. V. Visux likewise con- 
siders that the nerve fibres are produced through a differentiation of the protoplasm 
of cell strands, which he, like Donry, saw extending from the cord to the muscle 
segments in Selachians. It is necessary to add that Donry, in continuing his work on 
Selachian embryos, saw appearances which he felt might be used for or against his 
previous views. The mesodermic elements lie so close to the emerging ganglion cell 
process that any distinction between the two is lost, especially as mesodermic cells 
predominate. Dourn also thinks that an early penetration of mesodermic elements 
into the emerging zone of the motor fibres may have misled him. 
Brarp (1892), in embryos of Raja batis, has shown that the transient ganglion cells 
are the first-formed ganglionic elements from the neuro-epithelium, and that their 
nerves are mere transformations of chains of nerve-forming cells, 2.e. migrated ganglion 
cells. The motor end-plates are also derived from migrated ganglion cells. He also 
early noted the great resemblance between muscle fibre development and nerve fibre 
development. Brarp has also shown that the lateral nerve in Raja is formed by a 
chain of nerve-forming cells arisen from the neuro-epithelium at the level of the lateral 
line, and that axis-cylinder and myelin are differentiated in these cells. He thinks that 
the histogenesis of motor spinal nerves simply repeats the history of such a nerve as the 
lateralis if the region which will become the anterior horn of the spinal cord be looked 
upon as the parent neuro-epithelium. The chain of cells leaves the cord in the same 
manner, and the terminal cells form the motor end-plates and must therefore also be 
looked upon as ganglionic in character. The attachment of the spinal ganglia to the 
cord takes place by a chain of undifferentiated ganglion cells from the spinal ganglionic 
anlage developing into nerve-forming cells. These short. cell-chains reach upwards 
from the ganglion and form a continuous chain of several rows of cells along the route 
of the future posterior columns, and Brarp’s observations lead him to conclude that 
whenever a column or tract of fibres arises in the nervous system its development, 
as in this case, is initiated by the laying down of a chain of nerve-forming cells. 
Apatuy (1892-1907) believes that there is a primary differentiation of neuroblasts 
into central and peripheral groups; the latter migrate and develop into the nerve 
fibres, and the continuity between centre and periphery takes place later. Hach 
individual cell, central and peripheral, consists, according to Aparuy, of protoplasm 
