window, and two doors. It was plastered, but not free from cracks. Dimensions, 11 X 12 X 15.

The dimensions of the other room at Middletown, Conn., were as follows:

From this, however, must be deducted several heavy pieces of shelving, cases, etc., estimated roughly at 150-200 feet. It will be seen that the room, therefore, compares very well as to shape and contents with the room at Newton. But it had plastered walls, two doors, and two windows, and not in the best repair, and stood upon an exposed corner. On the whole, however, it was a tolerably close but not a tight room; less tight, if anything, than the room at Newton. The dimensions of the room at Athol, Mass., were as follows:

From this, however, must be deducted about 100 feet occupied by a chimney, a bench, a large case, etc.

Different animals, usually of several species, were placed in different parts of the room on the floor, on the table, on the shelves, etc. before the experiment began, and their symptoms carefully noted as the experiment went on. Samples of the atmosphere which they breathed were taken from time to time by entering the room and emptying into a vessel demijohns (generally holding one gallon) previously filled with water. These were carefully stopped with solid corks, then taken from the room and immediately sealed with melted paraffine. The time the samples were taken was carefully noted, and they were afterwards analyzed.

The following is a condensed table showing approximately the results of the experiments upon animals:

:

None.

Drowsiness. Discomfort. Slight effects.

Slight effects.

I will now give the results to which our experiments have led us, and also certain practical conclusions which naturally follow:

I. With ordinary gas-fixtures it is generally difficult to get more than three per cent of illuminating gas into an ordinary room. By using one burner alone, it is difficult to exceed one per cent.

II. With coal gas it is a matter of some difficulty to get into an ordinary apartment, through the ordinary burners, gas enough to produce upon healthy animals distinctly poisonous effects. With water gas, on the contrary, it is comparatively easy to get into an ordinary apartment, through the ordinary burners, gas enough to produce poisonous and even fatal effects.

III. It does not follow that because one illuminating gas contains three, four, or five times as much carbonic oxide as another it is therefore only three, four, or five times as dangerous to life.

IV. Our experiments confirm the work of Gruber and others, who claim that carbonic oxide is not a cumulative poison,- that is, the breathing of a small quantity for a long time is not equivalent to the breathing of a large quantity for a short time. A similar conclusion may be drawn for all the constituents of illuminating gas.

We may now illustrate the foregoing conclusions by examples drawn from our own experiments. And, first, as to the difficulty of charging rooms heavily with illuminating gas. (Expt. III., page 18.)

A room containing 1140 cubic feet of space was supplied with four ordinary burners. Through these there entered the room at a tolerably constant rate during twenty-four hours 1200 feet of coal gas. Yet, at the end of the twenty-four hours, the top of the room just above the burners contained a mixture of gas and air of which the former composed only three per cent, while the lower portions of the room showed less than one per cent. Again (Expt. V.), a room holding about the same amount of air, received fifty-five feet of water gas during one and a half hours. At the end of that time the largest amount discoverable in the room was 1.1 per cent of gas in the whole mixture of gas and air.

To illustrate the second conclusion, viz., that it is somewhat difficult to get in enough gas by the ordinary fixtures to kill, if the gas be coal gas, but relatively easy if it be water gas, it is only necessary to note the effects of the two experiments just quoted. In the former (coal gas), after twenty-four hours, the animals, though somewhat

drowsy and stupefied, were not seriously affected, while in the latter, after one and a half hours only, similar animals showed most alarming symptoms, and one was dead from the effects of the gas. From other experiments it is certain that, had this experiment been long continued, others, and probably all the animals, would soon have perished.

Similar considerations illustrate the third conclusion, for it is impossible to say that in the latter case the animals were only four or five times worse off than in the former. It is plain that, as their lives were in imminent danger, and as they were vomiting and in distress, it is not possible to express their relative danger mathematically. The first experiment just mentioned also indicates that carbonic oxide is not cumulative, for exposure to a small amount for twenty-four hours led to no serious consequences.

As to the time required to produce poisoning: this seems to be merely the time required to attain a poisonous percentage of carbonic oxide; and this clearly depends on the rate of inflow, the size of the room, the leakage, etc.

In view of the foregoing conclusions, based upon experimental evidence, it seems to us that it must be admitted by all that water gas, with its thirty per cent, more or less, of carbonic oxide, is a more dangerous substance than coal gas with its six per cent or seven per ceut of carbonic oxide.

MEETING 337.

Recent Progress in Under-Ground Wires.

BY MR. W. W. JACQUES.

The 337th meeting of the SoCIETY OF ARTS was held at the Institute on Thursday, October 22nd, Prof. C. R. Cross in the chair.

After the reading of the minutes of the last meeting, the chairman introduced Mr. W. W. Jacques, who read a paper on "Recent Progress in Under-Ground Wires."

Mr. JACQUES said: Last winter I had the honor of appearing

before this Society, and reading a paper on under-ground telegraphy. That paper was chiefly historical, and described the various attempts that had been made to lay wires under ground in Europe from 1840 down to that time.

This evening I propose to call your attention more especially to the progress that has been made in methods of laying electrical wires under ground during the past few years.

Ten years ago the number of wires in use in our American cities was small compared with those in use today. The telephone-the wires connected with which perhaps outnumber those used for all other purposes in cities was entirely unknown.

Electric lighting had not come into practical use, and the number of telegraph wires was far less than that in use at the present day.

In our American cities, at that time, none of the wires were placed under ground, as electric cables were used only for the crossing of rivers and streams, and such cables were made almost exclusively of gutta-percha-covered wires, because gutta-percha, when kept continually wet, is an excellent insulator, and very durable.

The introduction of the telephone and the electric light, together with the increase of telegraphic communication, has given rise to a great cobweb of wires, extending over the business portions of most of our large cities. So long as there were comparatively few wires overhead, little objection was made to them; but the immense increase of the last ten years, together with the probable large increase in the future, makes it desirable, on the part of the companies operating these wires, as well as on the part of the public, to have them gathered together in some systematic way, and, if possible, to place them under ground.

The objections to overhead wires are that, where they are numerous, they are continually coming in contact with each other; or, as it is technically called, "crossing up," and it is well known. that, when this takes place, one or both of the wires is rendered entirely useless.

In the case of the telephone wires, a large force of men is continually employed in running over the roofs to detect and remove such crosses, and even then the telephone often fails to work just when it is wanted, simply because the wire is crossed with some other