gated the plesiosaurian neck. He compared the chelonian plastron with the plesiosaurian girdle-bones, to show the dependence of form in the chelonian type upon the potential energy due to the conditions of locomotive activity, and used this consideration in elucidation of chelonian and plesiosaurian resemblances and differences. The plesiosaurian interclavicle he regarded as homologous with the chelonian and lacertian interclavicle; but the chelonian clavicles he believed to be epiphysial, potential representatives of the precoracoids, and therefore bones of which no analogues should be sought in Plesiosaurs on the theory of their chelonian affinities. He believed that the clavicles were ocasionally distinct from the interclavicle, though usually blended with it, just as all these ossifications become obliterated by scapular extension. The restorations and interpretations of the plesiosaurian pectoral girdle given by Conybeare, Hawkins, Owen, Huxley, Cope, and Phillips were discussed, and reasons given for dissenting from their views. The old genus Plesiosaurus was divided into two families-the Plesiosauridæ, containing the genus Plesiosaurus, and the Elasmosauridae, with Eretmosaurus, Colymbosaurus, and Muranosaurus. A new type was taken for the genus Plesiosaurus, which showed distinct clavicles. Eretmosaurus has neither clavicle nor interclavicle; and the scapulæ, concave in front, are blended in the median line, and blended laterally with the coracoids. Its type is Plesiosaurus rugosus of the Lias. Colymbosaurus has for its type Plesiosaurus megadeirus of the Kimmeridge Clay. It has no interclavicle; the scapulæ are prolonged forward in a wedge, and backward so as to meet the coracoids in the median line and enclose two coracoscapular foramina. Muranosaurus is founded on a new type from the Oxford Clay. It has no interclavicle; but the scapulæ are prolonged forward to meet in the median line; they are not prolonged backward to meet the coracoids; hence but one coraco-scapular foramen is formed. A similar condition marks the pelvic girdle. 2. "Murænosaurus Leedsii (Seeley), a Plesiosaurian from the Oxford Clay."-Part I. By Harry G. Seeley, Esq., F.L.S., F.G.S. All parts of the animal, except teeth, ribs, and hind limbs, were described. The premaxillary bones extend bird-like between the nares to the frontals. The foramen parietale is between the parietal and frontal, and directed backward. The cerebral lobes of the brain have a chelonian form, are prolonged in olfactory nerves, like those of Teleosaurus, and have the optic lobes moderately developed. The exoccipital bones do not enter into the occipital condyle. The basisphenoid is perforated by the carotids, as in Ichthyosaurus. The hypoglossal nerve does not perforate the exoccipital bone. There are 44 cervical, 3 pectoral, 20 dorsal, 4 sacral, and the first 8 caudal vertebræ preserved. Atlas and axis are anchylosed. The zygapophyses are semicylindrical, being concave in front and convex behind. A process of the neural spine is prolonged between the anterior zygapophyses, so as to divide the posterior zygapo physes and lock between them. The caudal vertebræ have the facets for the chevron bones free from the articular margins, forming large prominent unequal tubercles on the base of the centrum. The coracoids are as broad as long, nearly square, not prolonged in front of the humeral articulation. The scapulae are sickle-shaped, thin towards the median line, and without any suprahumeral process. The pubic bones are transversely oblong, and the ischial bones more than usually triangular, owing to there being no forward prolongation to meet the pubes. The iliac bones are less expanded at the proximal end, and relatively longer than usual. In the limbs the ulna and radius approximate in form to those bones in Pliosaurus, and the phalanges resemble those of Pliosaurus in not being compressed from side to side. There are six polygonal carpal bones. 3. "On the remains of Labyrinthodonta from the Keuper Sandstone of Warwick, preserved in the Warwick Museum." By L. C. Miall, Esq. Communicated by Prof. Huxley, F.R.S., F.G.S. In this paper the author first noticed the remains of Labyrinthodonts from the Warwickshire Keuper, acquired by the Warwick Museum since 1842, the date of Prof. Owen's memoir on this subject, and then reviewed the determinations of Prof. Owen. He considered that Labyrinthodon ventricosus, Owen, is not a distinct species, and that L. scutulatus, Owen, has not been proved to be a Labyrinthodont. The species as identified by the author are as follows: : Mastodonsaurus Jægeri, Von Meyer. pachygnathus, Owen. Labyrinthodon leptognathus, Owen. In conclusion the author described some of the structural peculiarities of the teeth of various Labyrinthodonts, XXVIII, Intelligence and Miscellaneous Articles. ON MAGNETISM. BY A. TRÈVE. IF [F we place between the poles of the large electromagnet of Ruhmkorff the two extremities of the thick wire into which the current passes from the pile-in other terms, if the circuit be closed between the two poles, we get neither spark nor report; but when we open it a violent detonation is produced, "almost as powerful as the report of a pistol," says M. A. de la Rive, who first made the discovery. The learned Genevan physicist adds, "it seems that the intensity of the extra current is greatly augmented in this case by the influence of the poles of the magnet." This is the phenomenon which I have essayed to fathom. Is it necessary to interrupt the current between the two poles in order to obtain this, effect? No. If we increase the distance between the two poles of the electromagnets in order to study only their isolated action, we shall quickly verify : (1) That the phenomenon announced by De la Rive is equally produced within the sphere of attraction of either of those poles; (2) That the phenomenon is not inherent in the inducing current alone, but that the current from any independent pile, interrupted within that sphere of attraction, gives rise to the same effects. (3) That the extra current receives a very real and considerable augmentation of tension. Oxygen being magnetic (as M. Edm. Becquerel has proved), there was room to ask if some action of condensation or separation of the constituent elements of air did not take place in the magnetic field of the pole. I therefore collected, by means of ordinary aspirators, air from several points in that field; and analysis of it revealed none of these peculiar actions. To give still more precision to this research, M. Duboscq and I investigated the interference-phenomena, by submitting one of the two luminous rays, or both at once, to the action of a powerful electromagnet. The apparatus employed was the interferential reflector of M. Jamin. We caused the two rays, or one of them, to pass successively into air, oxygen, nitrogen, hydrogen, carbonic acid; and in each of these different cases we could never observe the slightest displacement of the fringes. The hypothesis of an atmosphere of vibrating ather around the poles of magnets (a conclusion to which I have thought I could come) would perhaps permit us to understand a little better than it has yet been comprehended the phenomenon of induction by magnets, discovered by Faraday. I seize this opportunity to say that I have repeated the experiment of induction in vacuo, and at all pressures, without ascertaining the least alteration, either to more or less, in the intensity of the current produced. To this communication the author adds a Note relative to "the magnetic atmosphere of magnets." From some experiments, which are to be realized with greater precision by means of apparatus which is now being constructed, he believes he can infer the mode of action of an electromagnet upon a cylinder of soft iron placed along its axis.-Comptes Rendus de l'Acad. des Sciences, vol. Ixxx. pp. 310-311. ON A NEW ELECTROMAGNET, FORmed of coNCENTRIC TUBES OF IRON SEPARATED BY LAYERS OF CONDUCTING WIRE. BY J. CAMACHO. The new electromagnet, of which I have the honour to forward a sketch to the Academy, was constructed for the purpose of obtaining very considerable dynamic effects with relatively feeble currents. Each of the cores consists of a series of concentric tubes, the intervals between them being nearly equal to their thickness; on each of the tubes an insulated wire of copper is wound, always in the same direction, the thickness of the layer of wire being greater on the outermost tube. The extremities of the wire belonging to each tube traverse the breech and are united so as to form only one single conductor arranged in the following manner: the wire, after being wound round one of the two outside tubes, passes on to the inner tube next to this, then upon the tube concentric to the preceding, and so on to the central tube of this core; then, having run along the breech, it is wound round the central tube of the second core, within which core it follows a course the reverse of that indicated for the first core; that is to say, after winding round each of the concentric tubes successively and in the same direction, passing from the smallest to the largest, it issues finally after enveloping the outside tube of the second core*. The following is the result of some experiments made with this electromagnet : Employing the current of ten Bunsen elements of the usual size with bichromate of potass, the attractive force of the electromagnet at a distance of 12 millims. is 713 kilogrammes, and the time required for the development of the magnetism to raise this weight is 1.33 second. If we cut the wires which pass from one core to the other, then connect the four free ends crosswise (that is, connect the lower end of the wire of the right core with the upper end of the wire of the left core, and the upper end of the wire of the right core with the lower end of that of the left core), and then pass the current of the same ten elements, but combined in two parallel series of five elements each, the power of the electromagnet is unchanged, but the time necessary for the magnetization is reduced to one fourth, or to 0.33 second. If, finally, each of the two coils of the electromagnet be covered with a round shield of soft iron, thus binding the upper part of the four concentric cores which constitute it, the electromagnet loses some of its power, and is reduced to the condition of an ordinary electromagnet with a solid core. After indicating the theoretical considerations which led to the arrangement adopted, the author continues: 66 Experiment has shown that, if we cover the polar extremities of the tubes constituting each core with a round shield of iron, the electromagnet loses its great power and is again in the same condition as an ordinary electromagnet. In fact, the magnetism received by the shields will have been developed by the influence of *The diameter of the outside tube is 12 centims., and the thickness of the concentric tubes, which are four in number, is about 6 millims. ; the insulated copper wire is 1.8 millim. in diameter; and the number of helices of wound wire is seven to the outermost, and only two between each of the series of concentric tubes forming a core. In these conditions, the height of the cores being 20 centims., the total length of the wire is about 600 mètres; its weight, corresponding to this length, is 11.5 kilogrammes; and the total number of turns is 2000. the polar extremities of all the tubes in contact with them: b these polar extremities cannot develop a magnetism greater tha that which they possess, and that only on those atoms of th shields with which they are in contact; so that the magnetis of the atoms situated on the other side of the shields (that is, a the outside) will be very feeble, in consequence even of the thick ness of the latter. 66 Moreover, as the free extremities of the tubes of which eac core consists have each of them the same magnetic pole, o uniting them to one another by a round shield of iron, betwee these poles reactions are developed which diminish the magneti force of the system, just as happens in bundles formed of perma nent magnets."-Comptes Rendus de l'Académie des Sciences, Feb. 8 1875, vol. lxxx. pp. 382-384. ON THE EXPERIMENTAL DETERMINATION OF DIAMAGNETISM BY ITS ELECTRICAL INDUCTIVE ACTION. BY PROF. A. TOEPLER OF GRATZ. As is well known, Faraday and Weber have shown that diamagnetic bodies, when brought nearer to or moved further from a closed spiral in a powerful magnetic field, generate induction-currents. These certainly very feeble currents produced by the motion of the diamagnet have even been made use of by Weber in order, by means of a very sensitive apparatus, to compare the polarity of bismuth with that of iron. The electrical-induction currents, however, can be observed and measured without moving the diamagnet, by means of the diamagnetism which is produced and vanishes-whereby the process is in many respects simplified. I use for this purpose a differential inductor with a system of commutators, in the following manner :— Let two spirals (A and B) of thick wire be inserted one behind the other in the circuit of a series of constant intensity, and two induction-spirals (a and b) nearly identical in constitution placed in their cavities. Let the latter be likewise one behind the other, but closed at opposite ends by a galvanometer. At the opening and closing of the principal circuit the galvanometer is acted on by only the difference of the two inductions; and this difference is quite eliminated by adding to the more feebly acting inductionspiral (for instance, b) a small auxiliary spiral, which is inserted with it in the galvanometer-circuit, and shifted by a micrometerscrew towards the principal spiral B until the galvanometric effect of the closing and opening vanishes even with a principal current of great intensity. If now a magnetic or diamagnetic body be placed in the middle of the other coil (a), the galvanometer again gives, on the closing and opening of the principal circuit, the induction of the moment which is produced and vanishes. But this method (which, besides, was in like manner recommended by Dove for feebly magnetic bodies such as nickel &c.) is quite in |