Fiber connections in the central nervous system are studied both in normal material (especially with the aid of the Golgi technique) and in experimental material. Following an experimentally induced lesion of a nerve fiber or its cell body, the fiber peripheral to the point of the lesion starts to degenerate, a process termed anterograde degeneration. The first morphological sign of degeneration is an irregular outline of the fiber followed by varicosities and drop-like disintegration of the axon. The degenerated nerve fiber can be impregnated, and thus traced to its termination, with silver-impregnation methods, most of which are modifications of the reduced silver-impregnation technique described independently by Bielschowsky and Ramon y Cajal at the turn of the century. Although the silver-impregnation technique is extremely demanding, it can, when conditions are optimal, identify the synaptic fields of fiber tracts with a clarity equaled by no other method. The electron microscope technique has recently been introduced into experimental neuroanatomy, and we are currently witnessing one of the most dramatic developments in the study of interneuronal connections. The ultra-structural features of anterograde degeneration--including, among others, a general increase in electron density of the axoplasm--are specific enough to allow for an electron microscope diagnosis of degenerating terminal end-structures, and electron microscope identification of axon degeneration is rapidly becoming a routine procedure in neuroanatomic laboratories. It must be emphasized, however, that successful application of the electron microscope in experimental neuroanatomy requires a continuous feedback from adequate light microscope observations. Moreover, the serial combination of light microscope analysis, with Golgi or reduced silver techniques, and electron microscope analysis of one and the same specimen has very recently opened up some most promising lines of inquiry. Although remarkable progress in the study of interneuronal connections has been made during the last few decades, there is a never-ceasing need for more refined and more sensitive neuroanatomic techniques. Clearly, technical improvements, both in the design and performance of neuroanatomical methods, must be given highest priority if neuroanatomy is to remain abreast of neurophysiology in the joint attempt to elucidate the composition of neuron circuits.