The development of the nervous system presents many interesting problems as a developing system with numerous parameters of differentiation as well as from the point of view of the establishment of adult structure and function. With our growing understanding of developmental processes in general, and interactions at various stages of development in particular, it should be profitable to study more closely events of each period in a developing system, looking for information concerning their immediate control and their relation to events of other periods. In the nervous system, one phase which should be investigated much more thoroughly—especially from the point of view of the control of cellular differentiation—is that of the initial appearance of neuroblast cells and formation of the first nerve processes. Most studies of normal embryos which have included the period of initial differentiation have been primarily concerned with tracing the origins of definitive nuclei and fiber tracts, though possible mechanisms controlling various aspects of their development have of course been discussed.

The present study is concerned specifically with the period of initial differentiation of cells and fibers in the diencephalon and mesencephalon of the chick embryo. This region has been chosen because it is among the early areas of differentiation, and it contains a number of different centers, which are not continuous with other areas of differentiation at first. This study was begun as part of a thesis (Lyser, 1960) and reconsidered in the light of recent work in related fields.

In the chick embryo, neuroblasts with processes appear first in the hind brain and shortly thereafter in the diencephalon and mesencephalon, where the first neuroblasts with processes have been reported at 17- or 18-somite stages (Tello, 1923; Windle & Austin, 1936). The initial differentiation of neurites thus takes place quite early in the development of the central nervous system. In the spinal cord of the chick embryo (Hamburger, 1948), and presumably in the brain also, initial neuroblast differentiation begins while or before proliferation has reached its peak (cf. Hamburger, 1948; Tello, 1923; Windle & Austin, 1936) and so overlaps this phase. It is of course continuous with the later phases of development, including histological differentiation, but it begins well before these become apparent. In the diencephalon and mesencephalon the mantle layer does not become distinct from the inner cell layer nor can the longitudinal columns of cells be distinguished until about the fourth day (Palmgren, 1921; Rendahl, 1924; Kuhlenbeck, 1937).

Forty-five 13-through 30-somite chick embryos were studied. All the embryos were serially sectioned, either sagitally or transversely, and stained for nerve fibers with silver. Eighteen embryos were from the collection of Professor Leigh Hoadley. These embryos had been fixed in 95 % ethanol and stained with pre-war German Protargol by a modified Bodian (1936) method. The other twenty-seven embryos were prepared for this study. A number of fixatives recommended for the embryonic nervous system and several silver stains were tried in various combinations. Staining by a modified Holmes’s (1942) method was most satisfactory. For these young embryos, the following fixatives were found to be useful : 95 % ethanol, Bodian’s fixative no. 2 or no. 4 (Bodian, 1937), Mahdissen’s fixative as given by Gray (1954, p. 192), Lavdowsky’s mixture as given by Guyer (1953, p. 236), or Lavdowsky’s mixture modified by substituting formic acid (1-6 ml) for acetic acid (2-0 ml). Embryos remained in ethanol for 112 h or in one of the other fixatives for approximately 24 h. They were stored in 70 % or 80 % ethanol, dehydrated in ethanol, cleared in cedar-wood oil, and embedded in 60-63 °C Tissuemat (Fisher). Serial sections were cut at 10 or 12μ.

Graphic reconstructions of some of the younger embryos were made by drawing neuroblast cells with processes and other segments of fibers on camera lucida tracings of each section and then tracing these on to an outline of the brain (Text-figs. 1-3). All cells and fibers which could be seen were recorded. In addition, diagrams of the pattern in some of the older embryos were made by sketching representative cells and fibers on an outline of the brain as the sections were studied (Text-figs. 4-6). In these drawings the actual number of cells present is not indicated; only a few are shown, illustrating the locations and orientations of the neuroblasts and fibers observed.

Text-fig. 1.

Group A: 16-somite embryo, sagittal sections. Graphic reconstruction of neuroblasts with processes (a-d, f-h) and segments of fibers (e, others unlabelled) in the diencephalon and mesencephalon of one of the least-advanced embryos. O, Location of optic stalk. Arrows indicate boundaries of telencephalon (T), diencephalon (D) and mesencephalon (M). A, Right side; B, left side. × 110.

Text-fig. 1.

Group A: 16-somite embryo, sagittal sections. Graphic reconstruction of neuroblasts with processes (a-d, f-h) and segments of fibers (e, others unlabelled) in the diencephalon and mesencephalon of one of the least-advanced embryos. O, Location of optic stalk. Arrows indicate boundaries of telencephalon (T), diencephalon (D) and mesencephalon (M). A, Right side; B, left side. × 110.

There is some variation in the development of the neuroblasts and nerve fibers among embryos of the same stage as determined by somite count. This appears to be due to a difference in the time at which differentiation of axons begins in this area of the brain in relation to the development of the somites. However, development is fairly regular in general location and arrangement of cells and fibers and in the order of their appearance. The embryos can be arranged in sequence by considering the numbers and distribution of cells and fibers together with the number of somites and incubation time.

The earliest stage at which fibers were identified in the diencephalon and mesencephalon in the group of embryos studied was the 14-somite stage. For purposes of description, development from the earliest appearance of nerve fibers in this region through the 30-somite stage has been divided arbitrarily into five periods: (A) 14-to 16-somite embryos in which axon differentiation in this region is just beginning, (B) 16-somite embryos in which differentiation is slightly more advanced, (C) 17-to 18-somite embryos, (D) 19-to 22-somite embryos and (E) 23-to 30-somite embryos.

At the beginning of the period of development under consideration, the wall of the neural tube is essentially a pseudostratified columnar epithelium, the cells of which may be referred to as neural epithelial cells. Those which are undergoing division move toward the neurocoel; the nuclei of interphase cells are at various levels (Sauer & Walker, 1959; Sidman, Miale & Feder, 1959; Fujita, 1963). At the outer edge is a nucleus-free zone, consisting of the outer ends of the epithelial cells, where the marginal layer will subsequently form. This will be referred to here as the ‘peripheral zone’; the area between the peripheral zone and the neurocoel will be called the ‘nuclear zone’. Neuroblasts and fibers that are parallel to the surface of the neural tube and oriented in a dorso-ventral direction or obliquely will be referred to as ‘circumferential’, those which are parallel to the longitudinal axis of the neural tube as ‘longitudinal’, and those which are perpendicular to the margin of the neural tube as ‘radial’.

In these preparations, neuroblasts with processes stand out because the cytoplasm is more darkly stained than the cytoplasm of adjacent epithelial cells. Their processes and other segments of fibers are black. It usually is not possible to trace a fiber that extends through several sections from one section to the next, even in the younger embryos where there are only a few fibers. As with other methods for identifying nerve cells and fibers, the question of whether all neuroblasts are stained can be raised. It seems likely that with this method most neuroblasts are recognizable, but it cannot be definitely determined that all are stained. Silver methods may demonstrate only nerve cells with neuro-fibrillae (Guillery, 1965; Gray & Guillery, 1966), but there is no evidence at present that there are nerve processes in the early embryo which do not contain neuro-fibrillae or neuro-tubules. In electron micrographs of motor neuroblasts of chick embryos, for example, all the axons which could be definitely identified contained fibrillar structures (Lyser, 1964).

Two other problems encountered in studying silver-stained sections should be remembered in regard to the intent and the basis of interpretation of observations. As indicated above, in well-stained sections nerve fibers and the cell bodies of neuroblasts with processes are usually distinct. Sometimes, however, the edges of epithelial cells are dark and difficult to distinguish from axons, or fibers do not show up well and neuroblasts from which processes arise cannot be clearly distinguished. In descriptions of individual embryos the lower number of fibers recorded includes only those which can be identified with certainty; the higher number also includes those cellular structures which are thought to be neuroblasts or fibers but which are not clearly identifiable. Both have been included in the figures.

The neuroblasts which can be seen in any one embryo represent less than the total number present, since the plane of section must be nearly parallel to the long axis of a neuroblast in order to see the origin of the process from the cell body. It is difficult to see cross-sections of individual fibers if they are scattered singly, even though groups of transversely sectioned fibers show up well. To obtain an adequate picture of the pattern of nerve cell bodies, which are oriented in various directions, both transversely and sagittally sectioned embryos must of course be studied. Also, deviation of the plane of section from a true sagittal or transverse plane must be taken into account. In 16-to 18-somite embryos in particular, the plane of section is often at an angle to transverse or sagittal in part of the diencephalon and mesencephalon because the cranial flexure is beginning and the head of the embryo is turning to the right at the same time.

Individual cells and fibers of each embryo, and the numbers present in each case, have been analysed in order to obtain as much information as possible on the pattern of differentiation and on the way development proceeds. Specific numbers, etc., are not intended to have any other significance per se. To repeat, it is felt that study of a series of embryos sectioned in different planes, and including several embryos of each stage, gives meaningful information of this sort.

A. 14-through 16-somite stage: initial appearance of neuroblasts with processes

The first group includes the least advanced embryos in which nerve fibers were found in the diencephalon and mesencephalon. In each of these embryos a few neuroblasts with processes and a few additional segments of fibers could be seen on each side in the posterior half of the diencephalon. The embryo illustrated (Text-fig. 1), which was sectioned sagittally, has at least 2, and possibly 4, neuroblasts with processes and 7 other nerve fibers on the right. There are at least 2 and possibly 5 fibers on the left, 3 of which seem to come from cells in the same sections. In other embryos studied a few more fibers are visible; 1-4 neuroblasts and 4-20 other fibers were found on each side.

These neuroblasts and fibers are located within an area including the lower part of the dorsal half and the upper part of the ventral half of the lateral wall, and extending from the junction of the diencephalon and mesencephalon to about the middle of the diencephalon. The area covered is not quite as large in the least-advanced embryos (Text-fig. 1) as in those with slightly more neuroblasts and fibers. The individual neuroblasts and fibers are scattered singly among the neural epithelial cells. There is no indication of a specific cell by cell pattern of distribution within this area. The more ventrally placed neuroblasts in this group appear comparable to those identified by other authors (Tello, 1923; Windle & Austin, 1936) as belonging to the future nucleus of the medial longitudinal fasciculus. The more dorsal neuroblasts, certainly in slightly older embryos if not at the earliest stages, are farther dorsal than the early cells described by these authors, and apparently correspond to cells they identify as thalamo-tegmental or thalamo-bulbar.

The neuroblasts in these embryos are oriented circumferentially with processes extending ventrally (Text-fig. ïa:c, d; lb:f, g; Plate 1, fig. A), or radially, with their long axes parallel to adjacent epithelial cells and their processes extending laterally (Text-fig. 1 a : b; 1 b : h). There are also some neuroblasts at an angle between dorsal-ventral and medial-lateral orientations (Text-fig. 1 a:a). Due to the orientation of the cell body and axon arising from it relative to the plane of section, the radially oriented neuroblasts can be seen best in transverse sections and the circumferential neuroblasts in sagittal sections. All of the cell bodies of the radially oriented cells and most of those of the circumferential cells are in the nuclear zone. A few circumferential cells are just outside the nuclear zone.

The axons of almost all of the neuroblasts grow ventrally or slightly obliquely in a ventral and posterior direction. Those of the radially oriented cells, which initially grow laterally, turn ventrally within the nuclear zone, at the border between the nuclear zone and the peripheral zone, or in the peripheral zone (Text-fig. la:b). Fibers that extend ventrally within the nuclear zone tend to enter the peripheral zone eventually. The fibers in the peripheral zone lie in the inner or middle part; very few are found along the outer edge. Occasionally branching fibers are seen (Text-fig. 1 b:e). In these embryos the fibers whose cell bodies are not seen appear as short, straight or slightly wavy segments, radially or circumferentially oriented. In the more advanced embryos some are slightly longer than those of the younger embryos. Occasionally a few fibers oriented in a longitudinal direction are seen in a ventral position at the junction of the diencephalon and mesencephalon.

B. 15-through 16-somite stage: more advanced embryos

During this period more neuroblasts are added to the original area of differentiation and longitudinal fibers appear ventrally. In the more advanced embryos of this group (Text-fig. 2), cells with processes and fibers can be seen farther dorsally, anteriorly and ventrally in the diencephalon than previously and also in the anterior mesencephalon, especially in the ventral part. The distribution of neuroblasts and fibers is more dense than before, particularly at the center of the area, where the first neuroblasts were located.

Text-fig. 2.

Group B: 18-somite embryos, sagittal sections. Graphic reconstruction of neuroblasts with processes (a-g, i-j) and segments of fibers (h, others unlabelled) in the diencephalon and mesencephalon, right side. O, Location of optic stalk. Arrows indicate boundaries of telencephalon (T), diencephalon (D) and mesencephalon (M). × 150.

Text-fig. 2.

Group B: 18-somite embryos, sagittal sections. Graphic reconstruction of neuroblasts with processes (a-g, i-j) and segments of fibers (h, others unlabelled) in the diencephalon and mesencephalon, right side. O, Location of optic stalk. Arrows indicate boundaries of telencephalon (T), diencephalon (D) and mesencephalon (M). × 150.

The neuroblasts in the lateral diencephalon are oriented in a radial or circumferential direction as the first ones were. The cell bodies are usually located in the nuclear zone, and most of the processes extend laterally, ventrally or obliquely in a posterior and ventral direction (Text-fig. 2; Plate, 1 fig. B). Some of these processes turn. An axon arising from the posterior-lateral side of the cell body may turn ventrally, immediately (Text-fig. 2: a) or at the outer edge of the nuclear zone (Text-fig. 2 : c). A ventral process may turn posteriorly (Text-fig. 2:d) or laterally (Text-fig. 2:e). A few axons can be seen to branch (Text-fig. 2:b,f, g). For example, one of these (Text-fig. 2:f) ends at the edge of the nuclear zone in a triangular enlargement with what seem to be very fine branches curving out dorsally and ventrally. Another (Text-fig. 2:g) divides just before reaching the outer edge of the nuclear zone; the two branches curve around the opposite sides of another cell. One cell (Text-fig. 2:b) extends laterally to the outer edge of the nuclear zone, where it runs ventrally a short distance, then turns in a lateral direction and ends in a Y-shaped branch. Most of the segments of fibers which can be seen in this area (Text-fig. 2) are oriented in a dorsal-ventral direction or obliquely from anterior and dorsal to posterior and ventral.

At the ventral edge of the original area in the diencephalon and in the anterior mesencephalon, longitudinal fibers, as well as a few which are radially or circumferentially oriented, are seen. Some of the longitudinal fibers seem to be processes of neuroblasts located in the lateral part of the diencephalon. This is suggested by the fact that a few of the most ventral circumferential fibers turn posteriorly among the longitudinal fibers (Text-fig. 2: h) or occasionally anteriorly. Other longitudinal fibers can be seen to arise from neuroblasts located at this level, often in the nuclear zone. The processes extend posteriorly, or posteriorly and laterally, from the cell bodies and pass into the peripheral zone, where they continue in a posterior direction (Text-fig. 2:i). A few of the neuroblasts in the mesencephalon have processes which extend in a ventral direction (Text-fig. 2:j). As in other areas, a few of the fibers branch.

C. 17-through 18-somite stage

Between the 16- and 18-somite stages, there is further development of the first area of axon differentiation, and neuroblasts in a second area, the dorsal mesencephalon, begin to send out processes. The latter are tectal cells.

The number of neuroblasts in the lateral diencephalon continues to increase and new cells with processes appear anterior and posterior to those seen previously. By the 18-somite stage, nerve fibers and neuroblasts are found from the level of the optic stalk to the anterior edge of the mesencephalon (Text-fig. 3). In the anterior part of the diencephalon, short, scattered fibers are seen. These are oriented radially or circumferentially. In the posterior diencephalon some longer fibers are seen and they are more densely distributed. Many of the fibers in the lateral part of the posterior diencephalon are oriented in an anterior and dorsal to posterior and ventral direction. Similarly oriented fibers are seen in the anterior mesencephalon, constituting the posterior part of this area of differentiation. Neuroblast cells are oriented so that processes extend laterally, ventrally, or from those in the ventral diencephalon and mesencephalon, posteriorly, or occasionally anteriorly (Text-fig. 3). In some cases the processes extend ventrally and then turn posteriorly.

Text-fig. 3.

Group C: 18-somite embryo, sagittal sections. Graphic reconstruction of neuroblasts with processes and segments of fibers in the diencephalon and mesencephalon, left side. O, Location of optic stalk. Arrows indicate boundaries of telencephalon (T), diencephalon (D) and mesencephalon (M). × 150.

Text-fig. 3.

Group C: 18-somite embryo, sagittal sections. Graphic reconstruction of neuroblasts with processes and segments of fibers in the diencephalon and mesencephalon, left side. O, Location of optic stalk. Arrows indicate boundaries of telencephalon (T), diencephalon (D) and mesencephalon (M). × 150.

Text-fig. 4.

Group D: 21-somite embryo, sagittal sections. Diagram of the pattern of neuroblast cells and nerve fibers in the diencephalon and mesencephalon. The positions and orientation of representative cells and fibers are indicated; not all of the cells and fibers actually visible in the embryo are shown (see Plate 1, figs. C, D). O, Location of optic stalk. Arrows indicate boundaries of telencephalon (T), diencephalon (D) and mesencephalon (M). × 40.

Text-fig. 4.

Group D: 21-somite embryo, sagittal sections. Diagram of the pattern of neuroblast cells and nerve fibers in the diencephalon and mesencephalon. The positions and orientation of representative cells and fibers are indicated; not all of the cells and fibers actually visible in the embryo are shown (see Plate 1, figs. C, D). O, Location of optic stalk. Arrows indicate boundaries of telencephalon (T), diencephalon (D) and mesencephalon (M). × 40.

The number of longitudinal fibers in the ventral diencephalon and mesencephalon becomes greater and the area in which they are found increases in size. By the 18-somite stage, longitudinal fibers are present from the area just posterior to the base of the optic stalk to the middle of the mesencephalon (Text-fig. 3). There are also circumferential fibers at this level in the diencepha-lon and a few in the ventral mesencephalon. The longitudinal fibers appear to be processes of longitudinally oriented neuroblasts in the ventral diencephalon and mesencephalon and of neuroblasts whose processes extend ventrally and turn at this level.

Nerve fibers appear in the dorsal mesencephalon by the 18-somite stage. At this time the area covered by these fibers is discrete from the first area of differentiation. In the embryo in Text-fig. 3, segments of fibers can be seen scattered along and to each side of the midline from the posterior boundary of the diencephalon to the anterior end of the rhombencephalon.

D. 19-through 22-somite stage

The most striking feature of this period is the elaboration of the fiber pattern within the lateral area of differentiation. Neuroblasts and processes become considerably more numerous, and the segments of fibers visible in each section are longer. In addition, a new center of differentiation develops just posterior and ventral to the base of the optic stalk.

The lateral area of differentiation continues to expand. By the 21-somite stage (Text-fig. 4) it extends from just posterior to the optic stalk into the anterior part of the mesencephalon. At the anterior edge of the mesencephalon, in the synencephalon (approximately the posterior third of the diencephalon, which is demarcated for a time in the embryo from the anterior portion of the diencephalon, or parencephalon, by a transitory constriction) and in the posterior parencephalon, there are circumferential fibers from the dorsal part of the lateral wall to the level of the longitudinal fibers. Anterior to this they are not found as far dorsally; just posterior to the optic stalk circumferential fibers can be found from approximately the middle of the diencephalon to the level of the longitudinal fibers. The most dorsal fibers in the posterior diencephalon and anterior mesencephalon have a dorsal-ventral orientation, axons extending more or less directly ventrally from the cell bodies (Plate 1, figs. C, D). Farther ventrally some are at an angle from anterior-dorsal to posterior-ventral and some from posterior-dorsal to anterior-ventral. Most of the fibers in the anterior part of the parencephelon are oriented obliquely from anterior-dorsal to posterior-ventral.

The ventral longitudinal fibers begin to form a more clearly outlined fasciculus. At the 21-somite stage they are seen from the area posterior to the optic stalk to the posterior border of the mesencephalon, where they merge with the longitudinal fibers of the hind brain. In the posterior parencephalon and in the synencephalon the fasciculus is quite wide, extending from the middle of the diencephalon to the ventral part of the lateral wall. The more dorsal longitudinal fibers curve ventrally as they approach the mesencephalon where they form a narrower fasciculus. There are fewer fibers in the fasciculus in the mesencephalon than in the posterior part of the diencephalon; in the embryo represented in the diagram (Text-fig. 4) the number decreases to about five at the posterior end of the mesencephalon. In the anterior part of the parencephalon the dorsal edge of the fasciculus also curves ventrally and the fibers decrease in number until there are just a few longitudinal fibers at the level about one fourth of the distance from the floor plate of the diencephalon to the roof.

Cells seen at the level of the longitudinal fasciculus have processes extending ventrally, at an angle anteriorly or posteriorly, or directly anteriorly or posteriorly. Sometimes two adjacent cells have processes extending in different directions; in a few instances the processes can be seen to cross near the cell bodies.

At the level of the optic stalk is another group of longitudinal fibers. These are oriented at an angle from anterior-ventral to posterior-dorsal. No fibers are visible between this group and the first longitudinal fibers and none can be seen crossing the mid line. No neuroblast cell bodies have been seen in this area in the embryos examined; it is not clear where the neuroblasts which give rise to these fibers are located.

The fibers in the dorsal mesencephalon have a pattern similar to that at the previous stage; they do not extend much farther ventrally during this period. In the embryo illustrated (Text-fig. 4), a few short pieces of fibers are visible on each side of the mid line.

E. 23-through 30-somite stage

During this period the areas of axon outgrowth which have appeared during the previous stages are enlarged and neuroblasts of the oculomotor nucleus begin to send out processes.

In the lateral diencephalon more neuroblasts send out processes ventrally or obliquely, but their distribution does not change markedly (Text-figs. 5, 6; Plate 2, figs. A, B). At the end of this period (Text-fig. 6, 28-somite embryo) circumferential fibers can be seen throughout the lateral wall of the posterior diencephalon. In the anterior diencephalon they are quite numerous in the ventral part of the lateral area, but much more sparsely distributed dorsally. At the dorsal edge of the lateral fiber area in the synencephalon and posterior part of the parencephalon there are a few fibers oriented longitudinally. There are also a few which cross the posterior part of the synencephalon at an angle from dorsal and posterior to anterior and ventral; that is, they seem to run from the dorsal longitudinal fibers to the longitudinal fasciculus. At the anterior end of this area, posterior to the optic stalk and just above the center of the lateral wall, there are also a few fibers oriented longitudinally. At the posterior end of the synencephalon fibers oriented more or less transversely seem to cross the dorsal mid line. These fibers are not separated anteriorly and ventrally from those of the lateral area or posteriorly from the fibers of the dorsal mesencephalon. Neuroblasts that have processes extending ventrally or obliquely can be seen throughout the lateral diencephalon in sagittal sections; in transverse sections, cells with processes extending laterally are apparent.

Text-fig. 5.

Group E: 24-somite embryo, sagittal sections. Diagram of the pattern of neuroblast cells and nerve fibers in the diencephalon and mesencephalon. O, Location of optic stalk. Arrows indicate boundaries of telencephalon (T), diencephalon (D) and mesencephalon (M). × 40.

Text-fig. 5.

Group E: 24-somite embryo, sagittal sections. Diagram of the pattern of neuroblast cells and nerve fibers in the diencephalon and mesencephalon. O, Location of optic stalk. Arrows indicate boundaries of telencephalon (T), diencephalon (D) and mesencephalon (M). × 40.

Text-fig. 6.

Group E : 28-somite embryo, sagittal sections. Diagram of the pattern of neuroblast cells and nerve fibers in the diencephalon and mesencephalon. O, Location of optic stalk. Arrows indicate boundaries of telencephalon (T), diencephalon (D) and mesencephalon (M). × 40.

Text-fig. 6.

Group E : 28-somite embryo, sagittal sections. Diagram of the pattern of neuroblast cells and nerve fibers in the diencephalon and mesencephalon. O, Location of optic stalk. Arrows indicate boundaries of telencephalon (T), diencephalon (D) and mesencephalon (M). × 40.

The number of fibers in the juxta-optic group increases also. In 24-somite embryos (Text-fig. 5) more fibers than at the previous stages can be seen on both sides posterior and ventral to the base of the optic stalk. As before, they are oriented predominantly in an anterior and ventral to posterior and dorsal direction. However, they are no longer separate from fibers of other areas. Circumferentially oriented fibers are found as far anterior as the level of the juxta-optic fibers, just dorsal to them. Posteriorly, the juxta-optic fibers merge with the longitudinal fasciculus. In this embryo fibers cannot be seen crossing the ventral mid line. By the 28-somite stage (Text-fig. 6) the fibers are more numerous still, especially just posterior to the optic stalk, and there are some fibers ventral and medial to it. In transverse sections, fibers crossing the mid line at the level of the posterior edge of the optic stalk can be seen.

At the 24-somite stage, the area of differentiation in the mesencephalon is a little more extensive than in embryos of the 19-through 22-somite group. In addition to circumferential fibers at the anterior edge of the mesencephalon, there are now fibers in the ventral half of the lateral wall as far posterior as the middle of the mesencephalon. In this area some neuroblasts with fibers extending ventrally or obliquely posteriorly and ventrally can be seen. As before, there are circumferential fibers in the dorsal midbrain. Occasionally a fairly long fiber extends ventrally to the middle of the mesencephalon. In the dorsal part there are also a few longitudinal fibers.

By the end of this period (Text-fig. 6), circumferentially oriented fibers are present throughout the lateral wall of the midbrain. They form one continuous group, which also includes transversely oriented fibers in the mid-dorsal region. Most of the lateral fibers are dorsal-ventral or slightly oblique. Another new feature which appears during this period is longitudinally oriented fibers at the middle of the lateral area of the mesencephalon. The ventral ends of some of the circumferential fibers can be seen to turn posteriorly at this level, thus contributing to the longitudinal group. Neuroblasts with processes extending ventrally or obliquely can be seen in the mesencephalon.

The number of fibers in the ventral longitudinal fasciculus and the number of cells with processes located at this level also continue to increase. In the younger embryos of this group (Text-fig. 5), there are only a few longitudinal fibers in the posterior mesencephalon and not many in the anterior diencephalon. By the 28-somite stage (Text-fig. 6), the ventral longitudinal fasciculus contains a considerable number of fibers throughout. The fibers are more or less parallel to each other, though there is some crossing. In the mesencephalon the fasciculus is located at the ventral-lateral corner of the neural tube. In the synencephalon and posterior parencephalon it is in a more dorsal position and is wider; the most dorsal fibers are at about the middle of the diencephalon. Anterior to this the fasciculus becomes narrower and is continuous with the fibers of the juxtaoptic area. There are circumferential fibers at the level of the longitudinal fasciculus, that is, crossing the longitudinal fibers, particularly in the posterior diencephalon. A few circumferential fibers which turn anteriorly or posteriorly at the level of the longitudinal fibers are visible. Neuroblasts with processes extending posteriorly, or occasionally anteriorly, posteriorly and ventrally or anteriorly and ventrally, can be seen among the longitudinal fibers.

Fibers of the oculomotor nerve are first recognizable in 25-somite embryos. They are located near the anterior end of the mesencephalon at the medial edge of the longitudinal fasciculus. The oculomotor area is not completely separate from other areas of differentiation, since by this time neuroblasts are recognizable at the level of the longitudinal fasciculus, just lateral to the oculomotor area. No specific pattern of arrangement within the oculomotor area is dis-cernible during the period under consideration. In the youngest embryos in which oculomotor fibers can be identified, a few can be seen emerging from or just outside the ventral mesencephalon on each side. For example, in one 25-somite embryo, sectioned transversely, all the oculomotor fibers can be seen in four sections (six sections distant from the anterior edge of the mesencephalon and eighteen sections from the posterior edge). They are located at the medial edge of the longitudinal fasciculus. In the first section there are three fibers ventral to but not touching the neural tube on the right side and one on the left which is touching the edge of the neural tube but cannot clearly be seen to come from within the mesencephalon. In the second section there are two fibers emerging from the mesencephalon on the right and one on the left. In the third section there is one fiber emerging on the right. In the fourth section there is one fiber on the right which is touching the edge of the mesencephalon but not definitely emerging in this section. The more lateral emerging fiber on the right in the second section turns out of the section just inside the edge of the neural tube. The other emerging fibers seem to come from within the nuclear zone in the same section; the cell bodies cannot be identified. Just anterior to the fibers of the oculomotor nerve there are some radial fibers medial to or at the level of the longitudinal fibers which may also be oculomotor. Some are in the nuclear zone, some pass from the nuclear zone into the peripheral zone.

In slightly older embryos, more fibers and some neuroblasts of the oculomotor nucleus can be seen anterior and posterior to the first fibers as well as among them (Text-fig. 6; Plate 2, fig. C). They are located at the medial part of the longitudinal fasciculus and just medial to it. A few of the neuroblasts which give rise to the oculomotor fibers can be identified near the edge of the neural tube at the medial side of or just medial to the longitudinal fasciculus. They have processes that extend ventrally and emerge from the mesencephalon. In embryos sectioned sagitally, neuroblasts can be seen which have processes that extend posteriorly and then turn ventrally to leave the neural tube.

In summary, during the period of initial neuroblast differentiation in the diencephalon and mesencephalon the following sequence has been observed. The first neuroblasts with processes were seen in 14-somite embryos in the lateral part of the posterior diencephalon. Their axons extend laterally or ventrally from the cell bodies; the lateral fibers turn ventrally. By the 15- or 16-somite stage the beginning of a ventrally located longitudinal group of fibers is indicated. The longitudinal fibers are apparently processes of laterally located cells which turn posteriorly at this level and of cells located among the longitudinal fibers. At about the 18-somite stage a new area of differentiation appears in the dorsal mesencephalon with fibers extending circumferentially. The next area of differentiation is visible by the 21-somite stage in the juxta-optic region. All of these areas are enlarged so that by the 30-somite stage neuroblasts and fibers are present, fairly evenly distributed, over most of the lateral wall of the diencephalon and mesencephalon. The more dorsal and lateral fibers are mainly circumferential or oblique. A prominent group of ventrally located longitudinal fibers extends from the juxta-optic area through the mesencephalon, and some longitudinal fibers are also seen dorsally in the posterior diencephalon and along the middle of the lateral wall of the mesencephalon. Oculomotor fibers, from neuroblasts in the ventral mesencephalon, are first recognizable at about 25 somites.

The general pattern of differentiation as described above agrees for the main part with previous studies which include early stages of nerve-fiber development (Tello, 1923; Windle & Austin, 1936; Van Campenhout, 1937). Observation of fibers in slightly younger embryos in the present study may be due to differences in staining procedures, as well as possible variations in embryos and determination of stages. The present study is concerned primarily with details of development during this period and does not include study of older stages which would be necessary for more extensive discussion of the identity of neuroblasts and fiber groups in terms of adult nuclei and tracts.

From these observations of the development of the diencephalon and mesencephalon in 13-to 30-somite chick embryos, the following generalizations can be made about the way in which the initial phase of differentiation proceeds.

  1. Several different areas of differentiation can be recognized in the diencephalon and mesencephalon, which appear in the embryo in regular sequence.

  2. In each area the first neuroblasts are distributed in a scattered fashion, the area is progressively enlarged by the differentiation of additional cells at its edges, and at the same time new cells differentiate within the old area. (3) Processes of the cells in each area, or part of an area, grow out in a generally consistent and characteristic direction, with small irregularities and variations of the courses being typical.

The sequence of initial neuroblast differentiation in the diencephalon and mesencephalon is more complex than the anterior-posterior, dorsal-ventral pattern seen in the development of the embryo in general. Differentiation begins in the diencephalon and mesencephalon after it has started farther posteriorly, in the hind brain. Within the mid- and forebrain it occurs first in the lateral wall of the posterior diencephalon and anterior mesencephalon, followed by the more posterior and dorsally located dorsal mesencephalon, and then by the more anterior juxta-optic area. Furthermore, in each area, differentiation spreads from the initial location in various directions. Such a characteristic pattern implies a specific regional differentiation within the neural tube at the time of initial differentiation of neuroblasts. This organization may result, directly or indirectly, from the very early regional differentiation of the medullary plate and neural tube. The latter is demonstrated by the distinctive gross form of various parts of the brain and by developmental capacities under experimental conditions, such as experimental regional induction or development of isolated regions, including development of characteristic patterns of function (Corner, 1964). Separation of parts of the neural tube in vivo during initial neuroblast differentiation demonstrates the ability of these regions to differentiate further, morphologically and functionally, without continuity with one another (Rhines & Windle, 1944; Hamburger & Balaban, 1963; Hamburger, Balaban, Oppenheim & Wenger, 1965). This does not necessarily imply complete independence of differentiation in the neural tube. There is some evidence from organ-culture studies that differentiation in the central nervous system is influenced by the surrounding environment, including adjacent tissues (Szepsenwol, 1940 a, b;Lyser, 1966). Though certain adjacent tissues affect histological organization in the spinal cord (Holtfreter, 1939; Holtfreter & Hamburger, 1955), the effect could be somewhat non-specific in terms of cellular differentiation, comparable to certain cases of mesenchymal-epithelial interactions (Grobstein, 1953, 1962; McLoughlin, 1961). The influencing factor could support development of a cell type determined by the specificity of the region of the neural tube.

The characteristic, specific pattern observed pertains to the location and sequence of differentiation of groups of neuroblasts. No evidence of a specific pattern of individual cells has been found. This possibility cannot be ruled out by observations such as those in this work, where all the cells cannot be seen because of their orientation relative to the plane of section, but it seems unlikely that such would occur.

The scattered distribution of the first cells within an area can probably be correlated with their early differentiation. As regards each individual neuroblast cell, the sequence of proliferation followed by differentiation, and especially the absence of division once morphological differentiation has begun, is adhered to. Since proliferating cells are distributed all along the neural tube, and since much proliferation is still to take place in all areas, it is not surprising that scattered individual cells from among the neural epithelial cells differentiate first. This leaves the rest to divide until a sufficient number have been produced to form a mantle layer and an inner neural epithelial layer.

What is responsible for the initiation of differentiation in certain cells is a question still to be answered. The explanation must cover the localization of the process in specific areas and also selection of a limited number of cells in each area out of a large population, many of which would presumably be able to differentiate in the same way.

The sequence of differentiation in the earliest neuroblasts appears to be different from that of the classical description, applicable to later stages; but it is consistent with the description of Windle & Austin (1936; see Windle & Baxter, 1936, and Lyser, 1964). The presence of radial cells, circumferential cells and cells with an orientation intermediate between these suggests that the neuroblasts begin to send out processes while still having the position and orientation of neural epithelial cells. They subsequently shift to a position where the mantle layer will form and sometimes, as in the lateral diencephalon, change their orientation.

The characteristic pattern of development pertains to the orientation of neuroblast cells and the courses of their axons as well as to the location and sequence of differentiation. Radial orientation of cell bodies is not necessarily included in this category, but rather may be a reflexion of the sequence of differentiation in the early neuroblasts. Otherwise, cell bodies generally have fairly consistent orientations which can be considered characteristic of each area. The courses of individual fibers suggest a specificity in regard to the general direction, with small variations probably due to a number of other factors influencing the pathway at the same time. Small deviations in the courses of individual fibers suggest that they are following a path of least mechanical resistance around various obstacles. Mechanical factors have been shown to be important in determining the pathways of nerve fibers under various experimental conditions, and could exert some effect on an outgrowing fiber in the embryo also. For example, nerve fibers need a surface or interphase along which to grow (Lewis & Lewis, 1912; Harrison, 1914). Nerve processes grow along the fibrils in a plasma clot that have been oriented by stroking (Weiss, 1934). In regenerating tadpole tails, nerve fibers grow along the surface of fibroblasts and of other nerve fibers and also can be blocked by fibroblast cells and processes when these structures form obstructions in the paths of the fibers (Speidel, 1933).

However, it seems that the overall pattern would be much more random, instead of fairly consistently the same for the fibers in a given area, if there were not some more specific factor directing the fiber in a particular direction. No explanation of such a directive mechanism is apparent, but initial out-growth of fibers may be comparable to experimental situations, including regeneration of fibers, where there is evidence of very specific selection of pathways. For example, if the hind brain is reversed or if an obstruction is placed in its path, the axon of Mauthner’s cell eventually assumes a position in the cord at, or fairly near, its normal location (Piatt, 1943, 1947; Stefanelli, 1950, 1951). Also, after removal of spinal ganglia in tadpoles previous to the development of the hind limb there is an almost normal motor pattern and there are no nerves in the sensory pathways. After removal of ventral horn cells, a normal sensory pattern develops (Taylor, 1943, 1944). The most striking example of the growth of particular fibers to particular end structures is the regeneration of amphibian and teleost optic fibers from each part of the retina to a specific area of the optic tectum, even after the eyeball is rotated (Sperry, 1945, 1948).

Since a consistent pattern of bifurcating fibers is not apparent, in the diencephalon and mesencephalon the branching of axons observed once in a while may represent temporary branches from fine processes of the growth cone, one of which will be established as the next segment of the fiber and the others withdrawn. The absence of many random fibers argues against formation and retraction of any but very short branches. Selection from among fibers reaching various other cells, as seen, for example, in regenerating tadpole tails, where cutaneous fibers sometimes grow toward muscles instead of toward the surface and are eliminated by retraction or degeneration (Speidel, 1942), is unlikely to occur here.

The extent of cellular differentiation, or determination, with respect to neuron type at the time of initial outgrowth of axons, is not known. The morphological pattern suggests differentiation at least of cells in one major area of development as distinct from those in another, or as embarked on a different course of development. The neuroblasts in the lateral diencephalon area, with processes extending ventrally and posteriorly, are distinct from those of the oculomotor nucleus, with processes passing out of the neural tube. However, within each area cells may or may not be differentiated into more specific types, which will later be included in various nuclei or parts of nuclei and have characteristic connexions. One of the simpler examples of differentiation within a group of neuroblast cells related to the present observations is the oculomotor nucleus. There is no indication of separate groups of cells at the beginning of differentiation, though the cells innervate four different extrinsic eye muscles, and hence are four functionally different groups of neurons, which are probably arranged in a particular way in the adult. (This has not been demonstrated in birds but there is some information on functional localization in mammals; see Warwick, 1953).

Regardless of the extent of specificity present in the neuroblasts at the time they begin to send out processes, the pattern of the location and the sequence of initiation of differentiation does not parallel that of future nuclei. For example, the first neuroblasts in the diencephalon-mesencephalon area seem to be scattered so as to include thalamo-tegmental and thalamo-bulbar cells as well as those of the future nucleus of the medial longitudinal fasciculus in one continuous area. This pattern suggests that the control of the initiation of differentiation may be separate from factors determining particular pathways and connexions specific for each type of neuron; cells are somehow ‘triggered’ to differentiate but the particular type of differentiation depends on some mechanism already set within the cells, or on other factors influencing it at the same time.

The pattern of differentiation of the first neuroblast cells and their processes in the diencephalon and mesencephalon of the chick embryo thus points out several problems: What is responsible for the initiation of differentiation in specific locations? How is the specificity of each individual cell acquired? What is the mechanism of directional growth of a nerve fiber? Does the pattern of initial differentiation influence subsequent formation of nuclei? It is hoped that this study of the pattern in the normal embryo will be the basis of further investigations to obtain information on some of these questions.

  1. The pattern of cells and nerve fibers in the diencephalon and mesencephalon during the initial stages of neuroblast differentiation has been studied in silver-impregnated sections of 13-through 30-somite chick embryos.

  2. The first neuroblasts were seen at the 14-somite stage, located in the lateral part of the posterior diencephalon with axons extending laterally and ventrally. Ventral longitudinal fibers appear by the 15- or 16-somite stage. Centers of differentiation appear subsequently in the dorsal mesencephalon, and the juxta-optic area. Oculomotor fibers appear at about the 25-somite stage.

  3. The differentiating neuroblasts are scattered; the initial areas are extended by differentiation of additional neuroblasts among the first cells and at the edges of original areas.

  4. The orientation of neuroblast cell bodies and the directions of fibers are characteristic for each area.

  5. These observations demonstrate a specific pattern of development within the nervous system and emphasize the need for further investigation of the factors controlling the various aspects of differentiation.

Le développement du diencéphale et du mésencéphale d’embryon de poulet au cours des phases initiales de différenciation des neuroblastes

  1. On a étudié la disposition des cellules et des fibres nerveuses du diencéphale et du mésencéphale au cours des stades initiaux de la différenciation des neuroblastes sur des coupes d’embryons de poulet de 13 à 30 somites, imprégnées à l’argent.

  2. Les premiers neuroblastes ont été observés au stade 14 somites, localisés dans la partie latérale du diencéphale postérieur avec des axones s’étendant latéralement et ventralement. Des fibres longitudinales ventrales apparaissent aux stades 15 ou 16 somites. Des centres de différenciations apparaissent par la suite dans le mésencéphale dorsal et la région juxta-optique. Les fibres oculomotrices apparaissent aux environs du stade 25 somites.

  3. Les neuroblastes en différenciation sont dispersés; les zones initiales s’étendent par différenciation de neuroblastes additionnels au milieu des premières cellules et sur les bords des zones d’origine.

  4. L’orientation des corps cellulaires des neuroblastes et la direction des fibres sont caractéristiques pour chaque zone.

  5. Ces observations mettent en évidence un plan de développement spécifique dans le système nerveux et soulignent la nécessité de nouvelles recherches sur les facteurs qui contrôlent les divers aspects de la différenciation.

The author wishes to express her appreciation to Professor Leigh Hoadley for all of his help in many ways. Part of this work was done during the tenure of a National Science Foundation Pre-doctoral Fellowship.

Bodian
,
D.
(
1936
).
A new method for staining nerve fibres and nerve endings in mounted paraffin sections
.
Anat. Rec
.
65
,
89
97
.
Bodian
,
D.
(
1937
).
The staining of paraffin sections of nervous tissues with activated pro-targol: the role of fixatives
.
Anat. Rec
.
69
,
153
62
.
Corner
,
M.
(
1964
).
Localization of capacities for functional development in the neural plate of Xenopus laevis
.
J. comp. Neurol
.
123
,
243
56
.
Fujita
,
S.
(
1963
).
The matrix cell and cytogenesis in the developing central nervous system
.
J. comp. Neurol
.
120
,
37
42
.
Gray
,
E. G.
&
Guillery
,
R. W.
(
1966
).
Synaptic morphology in the normal and degenerating nervous system
.
Int. Rev. Cytol
.
19
,
111
82
.
Gray
,
P.
(
1954
).
The Microtomisfs Formulary and Guide
.
New York
. :
The Blakiston Co
.
Grobstein
,
C.
(
1953
).
Epithelio-mesenchymal specificity in morphogenesis of mouse sub-mandibular rudiments in vitro
.
J. exp. Zool
.
124
,
383
414
.
Grobstein
,
C.
(
1962
).
Interactive processes in cytodifferentiation
.
J. cell. comp. Physiol. (Suppl. 1)
,
60
,
35
48
.
Guillery
,
R. W.
(
1965
).
Some electron microscopic observations of degenerative changes in central nervous synapses
.
Progr. Brain Res
.
14
,
57
76
.
Guyer
,
M.
(
1953
).
Animal Micrology. University of Chicago Press
.
Hamburger
,
V.
(
1948
).
The mitotic patterns in the spinal cord of the chick embryo and their relation to histogenetic processes
.
J. comp. Neurol
.
88
,
221
83
.
Hamburger
,
V.
&
Balaban
,
M.
(
1963
).
Observations and experiments on spontaneous rhythmical behaviour in the chick embryo
.
Devl Biol
.
7
,
533
45
.
Hamburger
,
V.
,
Balaban
,
M.
,
Oppenheim
,
R.
&
Wenger
,
E.
(
1965
).
Periodic motility of normal and spinal chick embryos between 8 and 17 days of incubation
.
J. exp. Zool
.
159
,
1
14
.
Harrison
,
R. G.
(
1914
).
The reaction of embryonic cells to solid structures
.
J. exp. Zool
.
17
,
521
44
.
Holmes
,
W.
(
1942
).
A new method for the impregnation of nerve axons in mounted paraffin sections
.
J. Path. Bact
.
54
,
132
6
.
Holtfreter
,
J.
(
1939
).
Gewebaffinität, ein Mittel der embryonalen Formbildung
.
Arch, exp. Zellforsch
.
23
,
169
209
.
Holtfreter
,
J.
&
Hamburger
,
V.
(
1955
).
Amphibians
. In
Analysis of Development
, pp.
230
96
. Ed.
B. H.
Willier
,
P. A.
Weiss
and V. Hamburger. Philadelphia : Saunders
.
Kuhlenbeck
,
H.
(
1937
).
The ontogenetic development of the diencephalic centers in a bird’s brain (chick) and comparison with the reptilian and mammalian diencephalon
.
J. comp. Neurol
.
66
,
23
75
.
Lewis
,
W. H.
&
Lewis
,
M. R.
(
1912
).
The cultivation of sympathetic nerves from the intestine of chick embryos in saline solutions
.
Anat. Rec
.
6
,
7
31
.
Lyser
,
K. M.
(
1960
).
The early development of nerve pattern in the diencephalon and mesencephalon of the chick embryo. Thesis, Radcliffe College, Cambridge, Massachusetts
.
Lyser
,
K. M.
(
1964
).
Early differentiation of motor neuroblasts in the chick embryo as studied by electron microscopy. I. General aspects
.
Devl Biol
.
10
,
433
66
.
Lyser
,
K. M.
(
1966
).
Différenciation du tube neural de l’embryon de poulet en culture organotypique
.
Archs Anat. microsc. Morph, exp
.
55
,
37
54
.
McLoughlin
,
C. B.
(
1961
).
The importance of mesenchymal factors in the differentiation of chick epidermis. II. Modification of epidermal differentiation by contact with different types of mesenchyme
.
J. Embryol. exp. Morph
.
9
,
385
409
.
Palmgren
,
A.
(
1921
).
Embryological and morphological studies on the mid-brain and cerebellum of vertebrates
.
Acta zool., Stockh
.
2
,
1
94
.
Piatt
,
J.
(
1943
).
The course and decussation of ectopic Mauthner’s fibres in Amblystoma punctatum.
J. comp. Neurol
.
79
,
165
83
.
Piatt
,
J.
(
1947
).
A study of the factors controlling the differentiation of Mauthner’s cell in Amblystoma.
J. comp. Neurol
.
86
,
199
236
.
Rendahl
,
H.
(
1924
).
Embryologische und morphologische Studien über das Zwischenhirn bein Huhn
.
Acta zool. Stockh
.
5
,
241
344
.
Rhines
,
R.
&
Windle
,
W. F.
(
1944
).
An experimental study of factors influencing the course of nerve fibers in the embryonic central nervous system
.
Anat. Rec
.
90
,
267
90
.
Sauer
,
M. E.
&
Walker
,
B. E.
(
1959
).
Radioautographic study of interkinetic nuclear migration in the neural tube
.
Proc. Soc. exp. Biol. Med
.
101
,
557
69
.
Sidman
,
R. L.
,
Míale
,
I. L.
&
Feder
,
N.
(
1959
).
Cell proliferation and migration in the primitive ependymal zone: an autoradiographic study of histogenesis in the nervous system
.
Exp. Neurol
.
1
,
322
33
.
Speidel
,
C. C.
(
1933
).
Studies of living nerves. II. Activities of ameboid growth cones, sheath cells, and myelin segments, as revealed by prolonged observation of individual nerve fibers in frog tadpoles
.
Am. J. Anat
.
52
,
1
79
.
Speidel
,
C. C.
(
1942
).
Studies of living nerves. VII. Growth adjustments of cutaneous terminal arborizations
.
J. comp. Neurol.
76
,
57
69
.
Sperry
,
R. W.
(
1945
).
Restoration of vision after crossing of optic nerves and after contralateral transplantation of eye
.
J. Neurophysiol
.
8
,
15
28
.
Sperry
,
R. W.
(
1948
).
Patterning of central synapses in regeneration of the optic nerve in teleosts
.
Physiol. Zool
.
21
,
341
61
.
Stefanelli
,
A.
(
1950
).
Studies on the development of Mauthner’s cell
. In
Genetic Neurology
, pp.
161
5
. Ed. P. Weiss.
University of Chicago Press
.
Stefanelli
,
A.
(
1951
).
The Mauthnerian apparatus in the ichthyosids; its nature and function and correlated problems of neurohistogenesis
.
Q. Rev. Biol
.
26
,
17
34
.
Szepsenwol
,
J.
(
1940a
).
Influencia de los somites sobre la diferenciación y el trayecto de los nervios in vitro.
Revta Soc. argent. Biol
.
16
,
608
15
.
Szepsenwol
,
J.
(
1940b
).
El trayecto de las fibras nerviosas reticulares y fasciculares en cultivos in vitro.
Revta Soc. argent. Biol
.
16
,
589
97
.
Taylor
,
A. C.
(
1943
).
Development of the innervation pattern in the limb bud of the frog
.
Anat. Rec
.
87
,
379
—413.
Taylor
,
A. C.
(
1944
).
Selectivity of nerve fibres from the dorsal and ventral roots in the development of the frog limb
.
J. exp. Zool
.
96
,
159
85
.
Tello
,
J. F.
(
1923
).
Les différenciations neuronales dans l’embryon du poulet, pendant les premiers jours de 1’incubation
.
Trab. Lab. Invest, biol. Univ. Madr
.
21
,
1
93
.
Van Campenhout
,
E.
(
1937
).
Le développement du système nerveux cränien chez le poulet
.
Archs Biol. Paris
.
48
,
611
66
.
Warwick
,
R.
(
1953
).
Representation of the extra-ocular muscles in the oculomotor nuclei of the monkey
.
J. comp. Neurol
.
98
,
449
504
.
Weiss
,
P. A.
(
1934
).
In vitro experiments on the factors determining the course of the out-growing nerve fiber
.
J. exp. Zool
.
68
,
393
448
.
Windle
,
W. F.
&
Austin
,
M. F.
(
1936
).
Neurofibrillar development in the central nervous system of chick embryos up to 5 days’ incubation
.
J. comp. Neurol
.
63
,
431
63
.
Windle
,
W. F.
&
Baxter
,
R. E.
(
1936
).
The first neurofibrillar development in albino rat embryos
.
J. comp. Neurol
.
63
,
173
87
.
Plate 1

Fig. A. 16-somite embryo, group A. Ventral part of the posterior diencephalon, sagittal section. A neuroblast cell with a process extending ventrally (N) and two segments of fibers (F) can be seen. × 2000.

Fig. B. 16-somite embryo, group B. Lateral diencephalon, dorsal to the middle, sagittal section. The neuroblast (N) has a process extending ventrally and slightly posteriorly along the border between the nuclear and peripheral zones. × 2000.

Fig. C. 21-somite embryo, group D. The same embryo as in Text-fig. 4, mesencephalon and diencephalon, sagittal section. Fibers and neuroblasts in the lateral diencephalon (D) and longitudinal fibers in the ventral mesencephalon (M) can be seen in this section. × 240.

Fig. D. 21-somite embryo, group D. The same section as fig. C. Neuroblasts in the posterior diencephalon with processes extending ventrally and obliquely can be seen. × 1000.

Plate 1

Fig. A. 16-somite embryo, group A. Ventral part of the posterior diencephalon, sagittal section. A neuroblast cell with a process extending ventrally (N) and two segments of fibers (F) can be seen. × 2000.

Fig. B. 16-somite embryo, group B. Lateral diencephalon, dorsal to the middle, sagittal section. The neuroblast (N) has a process extending ventrally and slightly posteriorly along the border between the nuclear and peripheral zones. × 2000.

Fig. C. 21-somite embryo, group D. The same embryo as in Text-fig. 4, mesencephalon and diencephalon, sagittal section. Fibers and neuroblasts in the lateral diencephalon (D) and longitudinal fibers in the ventral mesencephalon (M) can be seen in this section. × 240.

Fig. D. 21-somite embryo, group D. The same section as fig. C. Neuroblasts in the posterior diencephalon with processes extending ventrally and obliquely can be seen. × 1000.

Plate 2

Fig. A. 28-somite embryo, group E. The same embryo as in Text-fig. 6, lateral part of the posterior diencephalon, sagittal section. This section shows cells and fibers at the middle of the lateral wall, including oblique fibers which cross. × 800.

Fig. B. 28-somite embryo, group E. This is the next section medial to that in fig. A, showing more ventrally located cells and fibers in the diencephalon (D) and mesencephalon (M), including longitudinal fibers. × 800.

Fig. C. 28-somite embryo, group E. The same embryo as figs. A and B, mesencephalon, sagittal section. This section is at the level of the oculomotor nerve. Its fibers can be seen emerging from and outside of the neural tube. Sections of longitudinal fibers (L) can be seen posterior to the oculomotor nerve. × 1000.

Plate 2

Fig. A. 28-somite embryo, group E. The same embryo as in Text-fig. 6, lateral part of the posterior diencephalon, sagittal section. This section shows cells and fibers at the middle of the lateral wall, including oblique fibers which cross. × 800.

Fig. B. 28-somite embryo, group E. This is the next section medial to that in fig. A, showing more ventrally located cells and fibers in the diencephalon (D) and mesencephalon (M), including longitudinal fibers. × 800.

Fig. C. 28-somite embryo, group E. The same embryo as figs. A and B, mesencephalon, sagittal section. This section is at the level of the oculomotor nerve. Its fibers can be seen emerging from and outside of the neural tube. Sections of longitudinal fibers (L) can be seen posterior to the oculomotor nerve. × 1000.