that recovered the head was more or less white, and continued so for years afterwards; and even to this day, every now and then, a white head, as the gillies called the diseased fish, comes up from the sea.

"It is a curious and interesting fact that the condition of the fish was not affected in even the far advanced cases. Nutrition did not appear to have been interfered with. The body was as plump and fat and the pink colour as high as usual. I did not eat of those very far gone in the disease; of those less so I did eat, and found their flavour as in the healthy salmon. You will observe from what I have said that our disease, whatever might have been its cause, was a disease of the head, and confined to the head.

"So much for the form of our disease; now as to its origin. Whatever may have been the predisposing or its immediate cause, it is certain that the fish brought it with them from the sea, or, as in my opinion, acquired it in the tide-way in Loch Roch-Roag. They did not take it down with them when they went to the sea as kelts or smolts, but they brought it up from the sea in summer as grilse and fresh-run salmon. After mature consideration of all the attendant circumstances, I have come to the conclusion that the disease arose from the fish being kept moving so long up and down between the salt and brackish waters. With each flood tide they moved up in dense masses toward the mouth of the river, vainly looking for water sufficient to carry them into it, and, when the ebb came, going down again for two or three miles into the deep and comparatively salter water. This continuing for weeks, with the water in the bay becoming daily more shallow, the heat and bright sun during the day was sufficient, in my opinion, to account for the disease. I have already said the sea trout did not suffer, because very little water was sufficient to take them into the river, and they were kept outside for but a short time. Again, the fish in the Blackwater, a river within two miles of the Grimasto, had no disease—at least, I did not hear of any having been seen in it; the reason, as I think, being that at all times, except in the lowest neaps, the tide came up so near its mouth as to allow the fish to get into its lowest pools without much difficulty. Against my theory there is this to be said: as already mentioned, the island of Lewis has been subject within the last fifteen years, to my knowledge, to

VOL. X.

2 F

many dry seasons; and notably in 1863 there was a very longcontinued drought, proving so destructive to the salmon at Grimasto that it was said some 1500 fish were picked up dead on the shores of the bay and mouth of the river. I and my friends were not in the Lewis that year, and therefore I cannot speak as to the symptoms of that disease; but inquiry afterwards failed to elicit any evidence that they resembled the outbreak of 1868. . . . It may be that the fish in 1868 were in some peculiar abnormal condition before coming up from the sea, predisposing them to disease of the head; but at any rate I can give no other cause for the outbreak than those I have mentioned."

3. On the Form and Structure of the Teeth of Mesoplodon Layardii and Mesoplodon Sowerbyii. By Professor Turner, M.B.

The author in the first instance described the characters of the teeth of Mesoplodon Layardii from two specimens which had been collected during the expedition of H.M.S. "Challenger," under the scientific superintendence of Sir C. Wyville Thomson. The one specimen, a young animal, under 14 feet in length, was obtained at Port Sussex, East Falkland Islands, by Mr H. N. Moseley, F.R.S.; the other, an adult skull, was procured at the Cape of Good Hope.

The teeth of the younger animal, two in number, were imbedded in their alveoli in the lower jaw. Each tooth consisted of a small triangular denticle or crown projecting outwards, and slightly upwards from the middle of the upper border of the fang. The denticle measured 4-10ths inch in its longer diameter, the fang was 2 inches by 8-10ths. At the base of the fang was a cleft 2-10ths inch wide, which communicated with a pulp cavity that was prolonged almost to the apex of the denticle.

The denticle was invested by enamel, subjacent to which was a well-defined mass of dentine, which was prolonged as a thin layer almost down to the cleft at the root of the fang. The fang was invested by cement, which was separated from the dentine

by an opaque layer, consisting of a granulated matrix containing numerous branched and anastomosing vascular canals, like the Haversian canals of bone. A similar layer was prolonged into the pulp cavity, so as to line the dentine on its inner surface. This layer is apparently to be regarded as a modified form of vasodentine.

The teeth in the adult mandible were formidable tusks, which curved up the sides of the beak on to its dorsum, where they decussated across the middle line. Each tooth was 14 inches long, 7 inches of which had protruded beyond the gum. It consisted of a triangular denticle and a strap-shaped curved shaft. The denticle was somewhat smaller than in the young tooth, and the enamel was almost entirely worn off its surface. The size of the tooth was therefore due to the enormous development of the fang which formed the strap-shaped shaft. The shaft consisted for the most part of a cortical layer of cement investing an opaque central band, which had the structure of the modified vaso-dentine of the younger tooth. 7-10ths of an inch from the summit of the shaft was a minute mesial chink 1-10th inch long, which represented the pulp cavity, but the rest of the shaft was solid throughout. The summit of the shaft was more complicated in structure, and consisted from without inwards of the following layers :-cement, opaque modified vaso-dentine, opaque vaso-dentine, dentine, opaque modified vasodentine. When traced from the summit to the sides of the shaft the dentine and vaso-dentine disappeared, and then the two layers of modified vaso-dentine blended with each other and formed the opaque central band of the rest of the shaft. The size of the fang is due to the great growth of the cement and the tissue of the opaque central band. The teeth of this specimen are larger than in any of the previously recorded specimens, and the animal from which they are obtained was probably an old male.

The structure of the teeth of Mesoplodon Sowerbyii was examined from the skull described by the author in the "Transactions of the Royal Society, Edinburgh," 1872, vol xxvi. Each tooth was laterally compressed, and formed almost an equilateral triangle, and the crown was not separated from the fang by any sharp line of demarcation. The tooth consisted in great part of dentine, which in the crown was invested by a layer of not very strongly marked enamel.

The dentine extended down the fang to the sides of the narrow cleft at its base, which communicated with the pulp cavity. This cavity, bounded by the dentine, was contracted at the base of the fang, but dilated into a considerable space in the body of the tooth. The fang was invested by cement, but between the cement and dentine a layer of modified vaso-dentine was situated which increased in thickness in the lower part of the fang, whilst the dentine became thin. The structure of this tooth was then compared with that of the adult Mesoplodon Sowerbyii described by Professor Ray Lankester.*

The teeth both of M. Layardii and Sowerbyii in their non-erupted stage do not materially differ in structure from the ordinary human or carnivorous teeth, for the crown is covered by enamel, and the fang by cement, whilst the great body of the tooth consists of dentine, in which is a well-marked pulp cavity. The exceptional structure of these teeth in the erupted stage is due to the disappearance of the enamel from the crown, to the cessation in development of the ordinary dentine, and to the excessive formation in the adult Sowerbyii of osteo-dentine, and in Layardii of modified vaso-dentine, which cause the fang to assume unusual dimensions.

The following Gentleman, having been duly recommended and balloted for, was elected a Fellow of the Society:JAMES ABERNETHY, V.P. Inst. C.E., Prince of Wales Terrace, Kensington Garden, London.

Monday, 16th June 1879.

PROFESSOR MACLAGAN, Vice-President, in the Chair.

The following Communications were read:

1. Atomicity or Valence of Elementary Atoms: Is it constant or variable? By Professor Crum Brown.

*

"Quarterly Journal of Microscopic Science," 1867, vol. vii.

2. Action of Heat on some Salts of Trimethylsulphine. By Professor Crum Brown and J. Adrian Blaikie, D.Sc. No. IV.

I. The carbonate of trimethylsulphine is obtained by the action. of carbonate of silver on the iodide of trimethylsulphine. The solution of the salt may be evaporated to a syrup in the water-bath. On standing for some weeks over sulphuric acid in vacuo it crystallises out in exceedingly hygroscopic prismatic crystals, containing water of crystallisation, and having a strong alkaline reaction.

Heated in the air to 100° the salt gives off water, sulphide of methyl, and carbonic acid. Heated in a sealed tube to 100° C. for about eight hours it was almost entirely decomposed, gave off a gas consisting entirely of carbonic acid, and yielded two layers of liquid -the upper, sulphide of methyl; the lower, water and methylic alcohol. The decomposition is expressed by the equation—

{(CH3)3}2CO3+H2O = 2(CH3)2S+ CO2+2(CH3)OH.

II. The metaphosphate of trimethylsulphine is obtained by the action of metaphosphate of silver on the iodide of trimethylsulphine. The metaphosphate of silver was made from glacial metaphosphate of soda (Graham's salt) by precipitation with nitrate of silver. The metaphosphate of trimethylsulphine does not crystallise, but on evaporation leaves a colourless hygroscopic glass, containing some water.

The salt, when acted upon by heat, gives off sulphide of methyl, and the resulting product is at the same time decomposed, leaving free metaphosphoric acid. On further heat being applied the mass slightly chars.

III. The ferrocyanide of trimethylsulphine is obtained by the action of ferrocyanide of silver on the iodide of trimethylsulphine. On evaporation of the solution the salt crystallises out in pale-green transparent plates; they are not hygroscopic, and the salt gives all the reactions of an alkaline ferrocyanide. On drying over sulphuric acid or phosphoric acid, the crystals lose their water of crystallisation. Analysis leads to the formula {(CH3)3S} Fe2Cy 12+18H2O. The salt when heated to 220° C. gives off sulphide of methyl along with other products, including hydrocyanic acid, but does not