the cords unite, is found to be in direct connection with the olfactory lobe of the brain forming a cap over the bulbus.

13. When established, the nerve is seen to consist of two portions, a lateral and a more median one.

From observations upon swine embryos, there seems to be no reason to question the peripheral origin of the cellular mass which is connected with the olfactory epithelium by cords, chiefly cellular. At no time, in the development, were any structures observed arising from the brain, which might concontribute to the formation of this cell mass or to the cords.

MARSHALL, VOn Kölliker, BEARD, HIS, and others have considered these cellular structures to constitute the anlage of the olfactory nerve. That these cellular cords represent, in position, the olfactory filaments that arise later, there can be no doubt. Every stage can be traced from the first appearance of the cords as elevations of the epithelium up to the time of the presence of the olfactory filaments connecting the epithelium with the brain.

It is quite evident, however, that the majority of the cells of the cords are not the cells from which the olfactory fibers arise. The latter arise from cells the majority of which retain their original position within the olfactory epithelium. Only a few of the neuroblasts migrate from the epithelium.

tral processes of the neuroblasts, located within the epithelium, are seen to be directed toward the cellular cords. In the chick, DISSE has been able to trace the processes into the developing cords and finally to the bulbus where they end in glomeruli. My preparations show also some neuroblasts in the act of migrating from the epithelium into the adjacent mesoderm. GOLGI preparations do not show within the bulbus any indication of the development of neuroblasts which send processes peripherally to form olfactory filaments. Since it is known that nerve fibers are continuous with only one nerve cell, we must conclude that the neuroblasts observed in the olfactory epithelium and which in some cases migrate into the mesoderm are the

cells of origin for the fila olfactoria. Consequently it must be affirmed that the olfactory nerve has a peripheral origin. This, view is supported by observations made by EHRLICH, RETZIUS,2 GRASSI and CASTRONOVO,3 RAMÓN Y CAJAL, JAGODOWSKI' and others upon the adult nasal epithelium of mammals and other animals.

The majority of the cells, therefore, that form the primitive cellular anlage of His and others do not develop into nerve cells. DISSE has suggested that they constitute the cellular sheath of the nerve, that is, are comparable to neuroglia cells. Their fate I have not been able to ascertain. The majority of them, however, accumulate in the cap formed over the end of the bulbus.

His's contention that this cellular anlage constitutes a true ganglion is not supported by my observations. My conclusions on this point harmonize with those of DISSE, arrived at for the chick. At most, there is only a partial formation of a ganglion located between the olfactory epithelium and the brain.

According to this view, the belief that the fibers of the olfactory nerve have the same relation to the olfactory cells as do fibers to sensory cells in the taste buds and tactile corpuscles is not tenable. The olfactory cells constitute the cell bodies of neurones of the first order.

1 EHRLICH. Ueber die Methylenblaureaktion der lebenden Nervensubstanz. Deutsche Med. Wochenschrift, 1886.

haut.

2 RETZIUS. G. Zur Kentniss der Nervenendigungen in der RiechschleimBiol. Untersuchungen, Neue Folge, Bd. IV, 1892.

* GRASSI and CASTRONOVO. Beitrag zur Kentniss des Geruchsorgans des Hundes. Archiv f. mikros. Anatomie, Band XXXIV.

* RAMÓN Y CAJAL. Origen y Terminacion de las fibras, nerviosas olfactorias. Gazeta sanitaria municipal di Barcelona, Dec., 1890.

5 JAGODOWSKI, K. P. Zur Frage nach der Endigung des Geruchsnerven bei der Knochenfischen. Anat. Anz., XIX, pp. 257-267.

THE RELATION OF THE CHORDA TYMPANI TO

THE VISCERAL ARCHES IN MICROTUS.

By VICTOR E. EMMEL.

(Contributed from the Biological Laboratory of Pacific University, under the direction of G. E. COGHILL).

The mammalian chorda tympani is a branch of the facial nerve which passes over the tympanic cavity, underneath the auditory ossicles and joins the lingual branch of the trigeminus. It is generally accepted that the tympanic cavity and auditory ossicles are derivatives of the spiracular cleft and visceral arches of fishes. It would seem a natural conclusion, therefore, that the chorda tympani is also homologous with the pre-spiracular branch of the facial nerve of fishes and amphibians. Upon this point, however, authorities are not agreed. On the one hand, a large number of investigators regard the chorda tympani as the homologue of the pre-spiracular (pre-trematic) nerve of fishes, as, for example, BALFOUR, in describing the anterior branch of the seventh nerve of Elasmobranchii, says: "This branch forms the prae-spiracular nerve of the adult and is homologous with the chorda tympani of mammals" (Comparative Embryology, Vol. II, p. 459). STRONG, in his work on the cranial nerves of Amphibia, interprets the r. mandibularis internus of fishes and Anura, and the r. alveolaris of Urodela, as homologous with the mammalian chorda tympani. On the other hand, DRÜNER denies that the r. alveolaris is the homologue of the chorda tympani (Zool. Jahrb., XV, 3); while COGHILL, in his work on the cranial nerves of Amblystoma, interprets the r. alveolaris of Urodela as pre-spiracular, and takes the tentative position that the "most complete morphological and physiological representative (of the chorda tympani) in the

over and in front of the tympanic cavity, so that, in view of the confusion of ideas as noted above, the question now is whether the pre-tympanic position is primary and maintained throughout the embryonic life, or whether it is a position secondarily acquired in the development of the tympanum. Of course, the further questions of homology cannot be settled till this point in mammalian embryology is determined. It was for the purpose of contributing something to the solution of this question that the following study in the embryology of Microtus was undertaken.

The embryos used in this study were killed and preserved in formalin. To insure a correct conception of the relations of the nerves to all parts of the head a model of a 2.3 mm. embryo, magnified 50 diameters, was made by the BORN method. Reproductions of two older embryos, also, were made by KASTCHENKO'S method of graphic projection. My observations were made from the same series of serial sections as were employed for the model and projections, and from several other series of slightly different ages and cut in different planes.

First Embryo.

The youngest of the embryos was used for the construction of the model. The model and the sections from which it was made demonstrate clearly all the structures which are of importance for this study: the brain and all its flexures, the roots and ganglia of the fifth, seventh and eighth nerve, the mandibular and hyoid arches, and the posterior visceral arches as they are modified to form the sinus cervicalis. At this period the visceral arches are united only by a membrane composed of the two layers of epithelial cells, and the nerve trunks can be traced only a short distance into the mesenchyme of the arches.

Second Embryo.

From this embryo, more advanced than the first, two graphic projections were made, one of the exterior of the head and the other of the brain, fifth and seventh nerves and pharyngeal cavity. The mandibular and hyoid arches are still conspicuous