Resumido por el autor, Edward Phelps Allis, Jr. Sobre el origen de la hiomandibula de 10s Teleostomos. En Polypterus y otros Ganoideos las dos filas de branquias de cada uno de 10s arcos branquiales anteriores estitn reforzadas por radios branquiales cartilaginosos, y en Polypterus, las bases de dichos radios se han fusionado para fonnar una barra branquio-radial. Estas barras branquio-radiales se proyectan dorsoantero-mesialmente en cada arc0 formando un itngulo considerable con el epibranquial y faringobranquial del mismo arc0 y estitn dirigidas hacia puntos situados dorsalmente a la vena yugular. En 10s Selacios, una de estas barras ha dado lugar a 10s extrabranquiales de cada arco. En 10s Teleostomos barras semejantes han dado lugar, en el arc0 hial, a las cabezas articulares anterior y posterior de la hiomandibula, y en el arc0 mandibular a 10s procesos ascendente. y 6tico del palatocuadrado. El simpl6ctico es, probablemente, una parte de la barra branquio-radial anterior del arc0 hial y el interhial 10s elementos epales y faringeos del arco, fusionados y relativamente muy reducidos. Translatlon by Dr. JoSe Nonidez, Columbia University AUTHOR’B AB8TRACT OF T H I S PAPER E 5 U E D BY TEE BIBLIOQRAPEIC SERVICE ON THE ORIGIN OF THE HYOMANDIBULA OF THE TELEOSTOMI EDWARD PHELPS ALLIS, JR. Menton, France ONE FIGURE In a work published in 1914, I came to the conclusion that there must be, in fishes, “a primarily somewhat independent mass of mesoderm cells lying lateral to the neurocranium and dorsal to the dorsal ends, of the mandibular and premandibular arches, in the position of the pharyngeal elements of the branchial arches, which pharyngeal elements are wanting, as independent structures, in the mandibular and premandibular arches of all fishes.” These cells were assumed to be capable of chondrification and to have given rise both to the ascending and otic processes of the palatoquadrate of the Dipneusti, Amphibia and Reptilia, and to the lateral wall of the trigemino-facialis chamber of fishes and mammals. Similar cells related t o the hyal arch were said t o have possibly given rise to some portion of the otic capsule, and its derivative the operculum, and probably also to the toleostean hyomandibula. In a later work, published in 1915, I came to the conclusion that the cartilages derived from the mesoderm cells above referred to had, in 311 probability, their serial homologues in the extrabranchials and interarcual cartilages of the branchial arches of the Selachii, the posterior articular head of the teleostean hyomandibula being derived from the dorsal extrabranchial of the hyal arch and the anterior articular head from the interarcula cartilage between that arch and the mandibular arch. The symplectic was said to probably be a primarily independent cartilage, and probably an hypertrophied middle one or ones of the branchial rays of the mandibular arch. The single articular 257 258 EDW.ARD PHELPS ALLIS, JR. head of the hyomandibula of the Chondrostei was said to apparently correspond to the anterior articular head of the teleostean hyomandibula. The hyomandibula of Polypterus was left largely out of consideration, but it was said that the suprapharyngobranchials of' van Wijhe's ('82) descriptions of that fish and certain others of the Ganoidei were quite certainly represented in the extrabranchials of the Selachii, the suprapharyngobranchials accordingly being serial homologues of the posterior articular head of the hyomandibula. Since the publication of the work last above referred to, I have had occasion to examine the branchial arches in Polypterus, and I not only find that the suprapharyngobranchials of van Wijhe's descriptions are simply the epibranchiale of their respective arches, but that there are, in each arch, cartilages that quite certainly represent the special mesoderm cells that were assumed, in the two works above referred to, to have given rise, in the hyal arch, to the hyomandibula, and in the mandibular arch to the ascending and otic processes of the palatoquadrate. Van Wijhe, in the work above referred to, described, in the dorsal half of the first branchial arch of Polypterus, a small cartilage that he considered to represent the epibranchial of the arch, and two bones that he called the supra- and infrapharyngobranchials, the infrapharyngobranchial apparently being considered by him t o correspond to the typical selachian pharyngobranchial, and the suprapharyngobranchial to be a fifth element of a complete and typical branchial arch. The epibranchial is said to be almost completely concealed in a ligament that envelops both it and the infrapharyngobranchial, and that has its insertion on what Van Wijhe considered to be a part of the prootic covered by the thin lateral edge of the parasphenoid. The suprapharyngobranchial is said to be relatively large, to abut against a cartilaginous portion of the lateral wall of the neurocranium, and t o have its distal portion deeply grooved to lodge the efferent artery of the arch. In the second and third branchial arches there is said to be no epibranchial, and it is said that the upper ends of the supra- and infrapharyngobranchials of those arches may be fused to form a short tube which encloses the efferent HYOMANDIBULA O F THE TELEOSTOMI 259 artery of the arch. In the corresponding part of the fourth arch there is said to be a small pharyngobranchial, but in the figure given it is index-lettered as an infrapharyngobranchial. In a 75-mm. specimen of Polypterus senegalus that I have examined in serial transverse sections, there is no trace of the independent so-called epibranchial cartilage described by van Wijhe in the first branchial arch, and I also find no trace of it in adult specimens of Polypterus bichir and Polypterus ornatipinnis. In the first branchial arch of the 75-mm. specimen, the supraand infrapharyngobranchials of van Wijhe’s descriptions are found as independent cartilages, and they are certainly simply, respectively, the normal epibranchial and pharyngobranchial of the arch. The epibranchial articulates by its distal end with the ceratobranchial of its arch, and by the anterior corner of its proximal end with the pharyngobranchial, and the posterior corner of its proximal end has been prolonged to form a stout process which has acquired articular relations with the lateral wall of the bulla acustica,there lying ventral to thevena jugularis and the truncus facialis. The pharyngobranchial articulates with the epibranchial, as above described, and, running anteromesially and somewhat ventrally, enters the angle between the lateral and ventral (horizontal) plates of the ascending process of the parasphenoid, and there has its attachment. In its course it lies imbedded in the lateral surface of the thymus, dorsal to a stout ligament that extends from the angle of the ascending process of the parasphenoid to the dorsal end of the ceratobranchial of the first branchial arch, this ligament being the one that is said by van Wijhe to envelop his epibranchial and infrapharyngobranchial. A groove on the external surface of the epibranchial (suprapharyngobranchial of van Wijhe), between it and the pharyngobranchial (infrapharyngobranchial of van Wijhe) , lodges the efferent artery of the arch. In the adult specimens of both Polypterus bichir and Polypterus ornatipinnis, I find strictly similar conditions; but the epibranchial and pharyngobranchial have apparently fused with each other, and each has undergone extensive ossification, the two so-formed bones being in contact at their distal ends and 260 EDWARD PHELPS ALLIS, JR. there immovably connected with each other. The process on the posterior corner of the proximal end of the epibranchial has been completely ossified, excepting the articular cap by which it articulates with the lateral wall of the bulla acustica, and this process of the bone is strictly similar to that shown by van Wijhe in his figures of Amia and Lepidosteus, and by me in my figures of Amia and Scomber (Allis, '97, '03). In Amia T did not find the suprapharyngobranchial described by van Wijhe in that fish, but in Scomber, in a, corresponding position, I found an independent piece of cartilage that articulated with the pharyngobranchial of the second branchial arch and that I called a suprapharyngobranchial. In the second and third branchial arches of all my specimens of Polypterus, the 75-mm. one as well as the adults, the epibranchial and pharyngobranchial of each arch are completely fused with each other, :and the anterior (lateral) and posterior (mesial) edges of the so-formed piece have been produced dorsally so that they touch, or fuse with each other, dorsal to the efferent artery of the arch, thus enclosing it in a short tube. The nerve of the arch runs posteriorly dorsal to this tube, not traversing it. Each of the first three branchial arches of this fish is furnished with two rows of branchiae, each row supported by cartilaginous branchial rays the bases of which have fused t o form a practically continuous bar of cartilage, as shown in the accompanying figure. The rays thus form a comb-shaped structure the base of which is arched in a curve that corresponds approximately to that of the branchial bar of the arch when the mouth is opened and the branchial chamber expanded. When the mouth is closed and the branchia'l chamber contracted, the levator muscles have pulled the distal (ventral) end of the epibranchial upward, and that element and the pharyngobranchial are then directed ventro-antero-niesially a t a marked angle to the ceratobranchial. The comb-shaped structures formed by the branchial rays of each arch can not undergo a corresponding change of form, because of the relative rigidity of their basal bars, and the dorsal portions of those bars, the parts related to the epibranch- HYOMANDIBULA O F T H E TELEOSTOMI 261 ial and pharyngobranchial of each arch, project dorso-anteromesially, but more mesially, and less anteriorly, than the dorsal portions of the branchial bar, this leaving a wide space between themselves and that portion of the branchial bar. The basal bars of the rays, projecting dorsally, lie external t o the levator muscles of the arch, and their dorsal ends, which lie dorsal t o the vena jugularis, are there attached by connective tissues. Fig. 1 Lateral view of the posterior portion of the neurocranium of Polypterus, showing t h e first branchial arch in place, with the related anterior series of cartilaginous branchial rays, but the posterior branchial arches and the hyal arch removed. X 2. br., branchial rays of first branchial arch; cbr., ceratobranchial; ehr., epibranchial; $r., facialis foramen; g . , groove for vena jugularis; pbr ., pharyngobranchial ; ps., parasphenoid. The nerve and efferent artery of the arch now pass ventral to the anterior one of these two branchial-ray bars, and thenonward between the two bars onto the external surface of the ceratobranchial of the arch. In Amia, Lepidosteus and Polyodon cartilaginous branchial rays similar to those of Polypterus are found, and their bases are in contact with each other, but not so completely fused as in Polypterus; and in these fishes, also, the branchial-ray bars project dorsal to the epibranchial and pharyngobranchial of the arch T H E ANATOXICAL RECORD, VOL.15, NO. 5 262 EDWARD PHELPS ALLIS, JR. to which they are related. There is accordingly every reason to believe that similar conditions existed in the immediate ancestors of these fishes, and that in those fishes branchial rays, capable of fusing with each other at their bases, were found also in the hyal and mandibular arches. The dorsal ends of the branchial-ray bars of the latter arches would then lie close t o the bulging auditory portion of the neurocranium, dorsal to the vena jugularis, and hence in a position to form, in the hyal arch, a hyomandibula with one or two articular heads, and in the mandibular arch the ascending and otic processes of the palatoquadrate. The hyomandibula, thus formed, would lie in a plane somewhat inclined to that of the branchial bar of its arch, and the articulation of that bar with the hyomandibula would naturally be with its postero-internal surface, as is actually the case in Polypterus. The hyomandibula of Polypterus is always said to have but a single articular head, and the nervus hyoideus facialis runs outward posterior to that head, and the nervus mandibularis facialis anterior to it. I however find, in my 75-mm. specimen of this fish, a small and independent bit of cartilage lying immediately posterior to the cartilage that caps the actual articular head of the hyomandibula, and the so-called accessory hyomandibula is developed in relation to it. On one side of one adult specimen that was also examined, I find the dorsal edge of the accessory hyomandibula capped with cartilage, that cap forming a posterior ext,ension of the cap on the articular head of the hyomandibula ; and Traquair apparently shows similar conditions in his figure of this fish ('70, fig. 6, pl. 6). On the other side of the head of my specimen, this edge of the accessory hyomandibula is covered with bone, this being as van Wijhe ('82) found it in the specimen described and figured by him. A stout ligament always extends from the head of the accessory hyomandibula to the dorsal edge of the opercular process of the hyomandibula, and lies posterior to the nervus hyoideus facialis; this ligament and the accessory hyomandibula thus quite certainly representing the posterior articular head of the teleostean hyomandibula. The efferent artery of the arch lies postero-internal to HYOMANDIBULA O F T H E TELEOSTOMI 263 this ligament, thus corresponding, in its relations to the branchial rays, to the posterior efferent artery of the Selachii. The position of the nervus mandibularis facialis, anterior to the anterior articular head of the hyomandibula, is doubtless due to this nerve having separated from the nervus hyoideus shortly after the truncus facialis issued from its foramen, thus permitting it t o slip over the dorsal end of the anterior branchial-ray bar before that bar had acquired articulation with the cranial wall. The ventral end of the hyomandibula presents two angles, or processes, one of which articulates with the interhyal and is apparently formed by the posterior branchial-ray bar of the arch, while the other articulates with the quadrate, forms the so-called symplectic process of the hyomandibula, and is doubtless formed by the anterior branchial-ray bar. The interhyal must then represent the coalesced and relatively geatly reduced epal and pharyngeal elements of the arch. In the Holostei and Teleostei the conditions are similar to those in Polypterus excepting in that the posterior articular head of the hyomandibula is more fully developed, and in that the nervus mandibularis facialis does not separate from the nervus hyoideus until after the truncus facialis has passed between the two heads of the hyomandibula. In the Chondrostei, the posterior articular head of the hyomandibula is wholly wanting, .this indicating that the posterior branchial-ray bar has more or less completely aborted, and, doubtless in correlation with this, the interhyal has acquired articulation with the symplectic which must be either a detached portion of the anterior branchial-ray bar or be derived, as suggested in my earlier work, from branchial rays of the mandibular arch. In the Plagiostomi, the hyomandibula articulates with the cranial wall ventral to the vena jugularis, and is formed, in the Selachii, by the epihyal, and in the Batoidei by the pharyngohyal (Allis, '15). There is in these fishes but a single row of branchial rays, a posterior one, and it is found in the hyal as well as in the branchial arches. Associated with these rays there are so-called dorsal and ventral extrabranchials, which are currently considered to be simply modified dorsal and ventral ones of the 264 EDWARD PHELPS ALLIS, J R . branchial rays actually found in these fishes. Braus (’06), however, considers them to belong to an independent category of skeletal elements, for, in embryos of Heptanchus, he found them lying not only at a considerable distance from the branchial rays, but also at right angles t o those rays and parallel to the inner branchial bars. These relations t o the branchial rays at once suggest a branchial-ray bar that has been developed either in relation to those dorsal and ventral ones of the posterior row that were primarily related to the pharyngeal and hyal elements of the arch. If the extrabranchials have this latter origin, which seeins probable, then the dorsal one would be of similar origin to that here ascribed to the posterior articular head of the telecistean hyomandibula. The conclusions arrived at in my earlier works, and briefly statred in the opening paragraphs of the present article, would then have to be modified simply by the substitution of the “anterior branchial-ray bar of the hyal arch” in place of “an interarcual cartilage that lay between that arch and the mandibular arch,” and the symplectic would be derived from the anterior branchisl-ray bar instead of from the branchial rays of the mandibular arch. The conditJionshere found in recent Teleostomi and Plagioscami could evidently not be derived the one from the other without reversion to a type from which they both must have descended, and as there are no indications of any such reversion having taken place, the separation of the two lines here indicated must have taken place in very early geological times, for even in the crossopterygian Tristichopterus, remains of which are found in lower Devonian rocks, the hyomandibula must have articulated with the neurocranium dorsal t o the vena jugularis, for Traquair (’75) says that the posterior margin of the palatosuspensory apparatus, “apparently corresponding to the hyomandibular element,” gives articulation externally to the preopercular cheek-plate, and is itself connected dorsally with the squamosal region of the cranium; each of which conditions indicatcs that the hyomandibula lay external t o the vena jugularis and articulated with the cranium dorsal to that vein. Palois de CarnolBs, Menton, France April 13. 1918 HYOMANDIBULA OF THE TELEOSTOMI 265 LITERATURE C I T E D ALLIS, E. P., JE. 1897 The rranial muscles, and cranial and first spinal nerves in Amia calva. Jour. Morph., vol. 12, Eoston. 1903 The skull, and the cranial and first spinal muscles and nerves in Scoinber scomber. Jour. Morph., vol. 18, Lancs. 1914 ‘The pituitary fossa and trigemino-facialis chamber in Ceratodus forsteri. Anat. Anz., Bd. 46, Jena. 1915 The homologies of the hyomandihula of the gnathostome fishes. Jour. Morph., vol. 26. Philnda. B R ~ U SH. , 1906 Fher d r n cmhryonalen Kiemenapparat von Heptanehus. h a t . .inz., Ed. 29. Jena. TRAQUAIR, R. H. 1870 On the cranial osteology of Polypterus. Journ. Anat. and Physiol., vol. 5, London. 1875 Structure and affinities of Tristichopterus alatus. Trans. Royal Soc., vol. 27, Edin. WIJHE, J. W. VAR’ 1882 fiber das Visceralskelet und die Nerven des Kopfes der Ganoiden und von Ceratodus. Niederl. Archiv fur Zool., Bd. 5, H. 3. Leiden.