ZOOLOGY AND BOTANY, MICROSCOPY, ETC. 
445 
are a large number of them, and they give the appearance of an 
extremely close parallel striation to vertical longitudinal sections. The 
large granular cells are more scattered and rarer ; they have numerous 
ramified protoplasmic processes, which pass into the molecular layer, 
and also into the medullary lamellae. The nervous processes of the 
cells of Purkinje give off a moderate number of fine lateral branches, 
some of which return to the molecular layer. The smaller cells of the 
molecular layer are external or internal ; the former have richly branched 
protoplasmic processes, which often extend for a considerable distance, 
and a nervous process, the exact relations of which are unknown. The 
latter have very long and well-branched protoplasmic processes, some of 
which reach to the outermost parts of the molecular layer. The nervous 
process is very long, and extends as a transverse fibre over the bodies of 
the cells of Purkinje, and gives off, from time to time, vertical processes 
which pass inwards; these divide and surround the cell-body like 
basket work. 
The medullated fibres of the cerebellum of adult animals divide in 
the molecular layer only ; they form a thick plexus in the granular 
layer. In the brains of embryonic and young mammals the medullary 
lamellae of the cerebellum exhibit a certain number of undoubted nerve- 
fibres, which divide and become lost in the two layers of the grey 
substance, where they form anastomosing arborescent divisions. None 
of the fibrous structures revealed by Golgi’s methods give certain indi- 
cations of anastomoses, and as yet there is no fact that justifies us in 
believing in the presence of a nervous network in the grey substance. 
Does a Magnet affect Karyokinesis ? * — M. L. Errera, like many 
other observers, has been impressed by the resemblance between some 
karyokinetic figures and magnetic curves. He was led to try whether 
an electromagnet had any influence on the dividing nuclei in the staminal 
hairs of Tradescantia virginica. But the currents of protoplasm per- 
sisted, and the karyokinesis proceeded quite normally, so that the result 
of the experiment was distinctly negative. 
y. General. 
Origin of Nerve-centres of Coelomata. f — M. L. Roule discusses 
this question, and comes to the conclusion that in the Trochozoa 
(Mollusca and Annelida), and, without doubt, in the Chordata also, the 
nerve-centres of the adult, which are arranged in a bilaterally sym- 
metrical manner, are always derived from simple and median rudiments, 
which are subsequently divided into two lateral symmetrical halves, and 
that they are not formed from the junction of two primitively distinct 
rudiments. When the larva has a proper nervous system, this is 
sometimes arranged radially (Trochozoa), and sometimes longitudinally 
(Chordata). In the former case the greater part of the system dis- 
appears, while what remains becomes the rudiment of the nerve-centres 
of the adult, or put themselves into relation with rudiments formed 
directly by the ectoblast; in the latter case the nervous system is 
preserved entire, or parts disappear, as in the tail of the caducichordate 
Tunicata. 
* Bull. Soc. R. Bot. Belg., xxix. (1890) pp. 17-21. 
f Arch. ZLool. Exper. et Gen., viii. (1890) pp. 83-100. 
