It has long been recognised that the neuroglia presents difficulties of a very special kind in the way of its histological study. These difficulties are due mainly to the remarkable lack of affinity shown by its extra-nuclear part for the usual staining reagents. No histological method that has yet been described serves in a satisfactory manner for the complete investigation of this tissue in its normal and pathological states. The most generally serviceable at present in use is, I think, beyond all question the fresh method of Bevan Lewis. It at least furnishes us with a reliable test for the presence or absence of hypertrophy and sclerosis; but it does not give a complete or clear view of the normal tissue, the fibres remaining for the most part invisible (except in tissues from some of the lower animals), and the nuclear structure being quite obscured. Golgi's method gives us only a silhouette of a small proportion of the neuroglia-cells and fibres. It does not aid us in the solution of questions regarding nuclear and protoplasmic structure. Weigert's new method gives us a very clear view of the fibres and nuclei, but it leaves the protoplasm invisible in the normal state. Its chief deficiency, however, consists in the fact that its use is restricted to the fresh human brain. It is of little value for ordinary postmortem-room tissues, or for those of the lower animals. Other methods that might be mentioned are likewise imperfect in their results, or extremely limited in their utility. In these circumstances it is scarcely surprising that there is still the widest difference of opinion among the recognised authorities regarding the normal structure of the neuroglia. Yet until this question is carried beyond the stage of controversy it is obvious that all our views of the pathological changes occurring in the supporting tissue of the central nervous system must remain vague and unscientific.