The stria terminalis is the current name of the former pathway of Foville² or “taenia of Tarin”³ and became over time the semi-circular tenia (tenia semi-circulaire) according to French authors. It has been studied in depth as early as 1895 by J Déjerine (1901) and in parallel by A Koelliker (1896). It is a band of white substance, which appears as a flattened, very thin, 2–3 mm-wide band, with a fairly homogeneous diameter before it divides. It is very fragile and therefore difficult to dissect. Its length (6.6 cm, on average) slightly varies from one individual to another. It connects the amygdala to the septal region and to the hypothalamus. It is formed of a pathway of long fibres among which several are myelinated and whose origins are at the level of the nuclei of the amygdala, such as has already been demonstrated in animals, by CA Fox in 1943, T Ban and F Omukai in 1959, E Hall in 1960 and WJH Nauta in 1961. According to WM Cowan et al. 1965, only cortical and medial nuclei appear to be involved in the stria terminalis. Our observations made with humans have shown that, in fact, all of the nuclei are involved (Fig. 6.1) with a predominance for the medial nucleus. Yet, the stria terminalis appears to laterally leave the amygdala, which is paradoxical at first glance, since it emerges from the lateral slope of the superior pole (where is located the medial nucleus!). Such as Ch Foix and J. Nicolesco (1925) had already noted, the partitions between the different nuclei of the amygdala or surrounding them, as well as the peripheral capsule from which depart these partitions, are white substance lamina, formed by the neurofibres of neurons forming the nuclei. The stria thus results from the assembly of all the axons of the amygdaloid neurons, which are joined by the axons of the neurons from the centres, which it connects to the amygdala. The stria terminalis then follows the trajectory of the concave edge of the caudate nucleus. It initially runs along the medial surface of the tail of this formation. It will firstly progress to the rear, pass in the depth of the endorhinal sulcus, at the level of the uncus of the hippocampus, to reach the roof of the temporal lorn of the lateral ventricle, parallel to the fimbria from which it is only separated by a few millimetres. At the junction of the body and head of the hippocampus, the tenia of the fimbria and that of the stria terminalis unite (H.M. Duvernoy, in 1988), thus forming a small crescent moon-shaped lamella: the velum terminale of Aeby⁴ (Fig. 6.2). The choroidal web of the temporal horn of the lateral ventricle is attached to the above tenia and leaves the head of the hippocampus free and therefore not covered by the choroid plexi, unlike the other parts (body and tail) of this formation. The stria terminalis penetrates the opto-striated groove (or thalamo-striate sulcus), at the junction of the posterior thalamic convexity and concavity of the caudate nucleus. It therefore passes in the embryological joining area between the caudate nucleus (forebrain) and the thalamus (diencephalon). The stria terminalis is located in this groove successively in front of the thalamo-striate superior vein, at the level of the body of the caudate nucleus and below this vein, at the head of this nucleus. It is recalled that the superior thalamo-striate vein (terminal vein) drains the veins of the septum pellucidum and of the caudate nucleus and joins the homolateral superior choroidal vein to form one of the two internal cerebral veins, which are the roots of the great cerebral vein. The stria terminalis and its satellite vein will jointly form the bend with an anterior concavity of the thalamo-striate groove and thus reach the floor level of the frontal horn of the lateral ventricle, under the head of the caudate nucleus. They are covered, at this level, by the lamina affixa (lamina cornea), a thin epithelial layer, on the medial border of which are attached the choroid plexi of the lateral ventricle. Thus, the stria terminalis, over its arch-shaped trajectory, forms a 3/4 circle whose diameter is of 2.2 cm on average. Over this trajectory, the stria terminalis not only develops some thin branches, which target the caudate nucleus and the thalamus, but also a significantly sized branch which penetrates the latter at the level of the anterior third of the dorsal surface (Fig. 6.3). According to our chosen description approach, we can say that these branches contain input fibres targeting the thalamus. However, the opposite is also true and we can also say that these branches contain fibres from the thalamus, which are going to merge with the stria terminalis.

When the stria reaches the vicinity of the anterior commissure, it will be divided into three main pathways, which can be accessed for dissection (Fig. 6.2): the supra-commissural pathway (or precommissural component), the commissural pathway (commissural component) and the sub-commissural pathway (or retro-commissural component). These three pathways will develop as follows: