Another bottle was filled with distilled water and tube 20 filled up again with that.

631] The tube 20 filled with the same distilled water mixed with 200 of spirits of wine.

Same mixture mixed with more of spirits of wine, id est, sp' wine in 18 of distilled water.

Therefore there is not much difference between the resisting power of the above distilled water and spirits of wine and mixtures of the 2: but of the 2, spirits of wine resists least.

After it was measured 7 inches were cut off from straight end, and the numbers in the first col. are the distances from the shortened end. A new bend was also made 12.8 from new end, and the part where the tube is equal to tube 14 is at 10.3 from D°.

633] Jan. 1781. The following tubes were measured over again by introducing a col. and measuring its length in 3 different places, the beginning of the 1st col. being at inch from bend, and the beginning of the second at the end of the first. Another column was afterwards introduced whose length was pretty nearly equal to the sum of the 3 former, and weighed.

634] The two following tubes were measured by stopping up the end near bend and weighing them with different quantities of in them, and measuring the distance of the top of the column from straight eud, whence it was found that in

635] In the following result the column whose length is given in the 2nd column is supposed to begin at inch from the bend.

By the resistance of each is meant

1

grains in each inch*

[The resistance of a column of mercury one inch long weighing one grain is 13 Ohms, and the resistance of saturated solution of salt at to Centigrade is to that of mercury as 10 is to 2015+ 451 (t - 18). Hence the resistance as given by Cavendish must be multiplied by 6907 + 82-2 (59-T) to convert it into Ohms when the tube contains saturated solution at To Fahrenheit.]

COMPARISON OF RESISTANCE OF COPPER WIRE WITH THAT OF SAT. SOL.

636] The wire was wound on reel on bars of glass about inch broad, the distance of one round of wire from the next on same bar being '6.

The mean circumference of reel = 46.7 × 2 √2*.

There were 8 rows of glass bars, and 28 rounds of wire on each row, and on one row there was round over. Therefore whole length of wire = 93.4 × √2 × 8 × 28 + 1 = 29623 inches.

*

X

[The reel was probably square, with glass bars at the corners, the length of the diagonal being 46.7 inches.]

This weighed 2967 grains, consequently there are 9.984 inches to 1 grain.

N.B. There were many knots in the wire.

637] The resistance of this wire was attempted to be compared with that of sat. sol. in tube 17 by shock melter* as in former experiments, but without success. It was therefore compared by the sound of the explosion by discharging the jars by a wire without its passing through my body; but in this there was considerable difficulty, as the light of the spark passed through the wire was very different from that passed through the water, the first being reddish and the latter white. The sound also was of a different kind, the latter being sharper.

638. Two Leyden vials were made of barometer tubes filled with and coated on outside with tinfoil. The quantity of electricity in them was found to be very nearly the same, but that in N° 1 rather the greatest.

The charge of each of these tubes is about 714 inches, and that of the large jars about 6100, and that of the three jars 1, 2 and 4 together is also 6100*.

The shock of these tubes was received through my body in the same manner as in trying the large jars, either by making the shock pass through the copper wire or through the sat. sol. or receiving it in the simple manner without passing through either: the experiment being tried as usual by charging both tubes from the same conductor and receiving the charges of one one way and the other the other.

639. It was found by repeated trials that the shock received through the copper wire was plainly greater than the simple shock. When received through the sat. sol. with wires

640] It was also found by the small jars 1, 2 or 4 that the shock received through the wire was stronger than the simple shock.

The shock through the wire was also much greater than the simple shock when the covering which was put over the wire to defend it from accidents was taken away.

It was also plainly greater when the shock passed through only 3 rows of the wire instead of the 8.

If the shock was received through 166 inches of the same wire not stretched upon glass, without any knots in it, it seemed not at all greater, but if anything less than the simple shock. It was the same if received through a piece of wire of about the same length with 37 knots in it.

641] Some more of the same wire was stretched by silk into 32 x 12 rows, each 78.7 inches long; consequently the whole length was 30220 inches. It weighed 3272 grains, id est, 9-24 inches to a grain.

[Probably globular inches. The numbers do not agree with those in Art. 583.]