<table> <tr> <th>Years.</th> <th>Declination.</th> <th>Observers.</th> </tr> <tr> <td colspan="3">Well.</td> </tr> <tr><td>1683</td><td>4 3°</td><td></td></tr> <tr><td>1692</td><td>6 0</td><td></td></tr> <tr><td>1700</td><td>8 0</td><td></td></tr> <tr><td>1717</td><td>10 42</td><td></td></tr> <tr><td>1723</td><td>14 17</td><td></td></tr> <tr><td>1748</td><td>17 40</td><td>Graham.</td></tr> <tr><td>1760</td><td>19 12</td><td></td></tr> <tr><td>1765</td><td>20 0</td><td></td></tr> <tr><td>1770</td><td>20 35</td><td></td></tr> <tr><td>1773</td><td>21 9</td><td>Heberden.</td></tr> <tr><td>1775</td><td>21 30</td><td></td></tr> <tr><td>1780</td><td>22 10</td><td></td></tr> <tr><td>1785</td><td>22 50</td><td></td></tr> <tr><td>1787</td><td>23 19</td><td>Gilpin.</td></tr> <tr><td>1790</td><td>23 34</td><td></td></tr> <tr><td>1795</td><td>23 57</td><td>Gilpin.</td></tr> <tr><td>1800</td><td>24 7</td><td></td></tr> <tr><td>1802</td><td>24 6</td><td>Gilpin.</td></tr> <tr><td>1805</td><td>24 8</td><td></td></tr> </table>

From this last table it appears that when the declination was first observed, the north pole of the magnetic needle declined to the eastward of the meridian of London, that since that time it advanced continually towards the west till 1657, when the needle pointed due north and south, and that ever since it has continually declined more and more towards the west, in which direction it appears to be still advancing.

At Paris, in different years, the declination has been observed as follows:

In 1550 - 8° of East. 1640 - 3 0 1660 - 0 0 1681 - 2 2 West. 1759 - 18 10 1760 - 18 20

From 1792 to 1794 21° 54' Stationary.

<table> <tr> <th>In 1798</th><th>- 22 17</th> </tr> <tr><td>1799</td><td>- 22 0</td></tr> <tr><td>1800</td><td>- 22 12</td></tr> <tr><td>1801</td><td>- 22 1</td></tr> <tr><td>1802</td><td>- 21 45</td></tr> <tr><td>1803</td><td>- 21 59</td></tr> <tr><td>1804</td><td>- 22 10</td></tr> <tr><td>At Jamaica 1805</td><td>- 6 30 E.</td></tr> </table>

At Alexandria in Egypt,

In 1761 - 11° 4' W. 1798 - 13 6

At Cairo,

In 1761 - 12° 25' W. 1798 - 12

The declination of the magnetic needle has been found to be different, even at different hours of the day. The following table contains the result of some observations made by Mr Canton on the daily variation, and on the mean variation of each month.

The declination observed at different hours of the same day. June 27. 1759.

<table> <tr> <th rowspan="2">H. M.</th> <th rowspan="2">Decl. W.</th> <th rowspan="2">Degrees of the Therm.</th> <th colspan="12">The mean Variation for each Month in the Year.</th> </tr> <tr> <th>January,</th><th>February,</th><th>March,</th><th>April,</th><th>May,</th><th>June,</th><th>July,</th><th>August,</th><th>September,</th><th>October,</th><th>November,</th><th>December,</th> </tr> <tr><td>0 18</td><td>12° 2'</td><td>62°</td><td>7° 8'</td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td></td></tr> <tr><td>6 4</td><td>18 58</td><td>62</td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td></td></tr> <tr><td>8 30</td><td>18 55</td><td>65</td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td></td></tr> <tr><td>9 2</td><td>18 54</td><td>67</td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td></td></tr> <tr><td>10 20</td><td>18 57</td><td>69</td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td></td></tr> <tr><td>11 40</td><td>19 4</td><td>68½</td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td></td></tr> <tr><td>0 50</td><td>19 9</td><td>70</td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td></td></tr> <tr><td>1 38</td><td>19 8</td><td>70</td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td></td></tr> <tr><td>3 10</td><td>19 8</td><td>68</td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td></td></tr> <tr><td>7 20</td><td>18 59</td><td>61</td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td></td></tr> <tr><td>9 12</td><td>19 6</td><td>59</td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td></td></tr> <tr><td>11 40</td><td>18 51</td><td>57½</td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td></td><td></td></tr> </table>

Table of the Mean Monthly Variation of the Magnetic Needle for 20 Years at London*.

<table> <tr> <th>Year.</th> <th>January.</th> <th>February.</th> <th>March.</th> <th>April.</th> <th>May.</th> <th>June.</th> <th>July.</th> <th>August.</th> <th>Septemb.</th> <th>Octobe.</th> <th>Novemb.</th> <th>Decemb.</th> </tr> <tr><td>1786</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td><td>0</td></tr> <tr><td>1787</td><td>23 19.2</td><td>23 19.8</td><td>23 20.3</td><td>23 18.5</td><td>23 17.0</td><td>23 18.3</td><td>23 19.6</td><td>23 21.9</td><td>23 22.8</td><td>23 24.5</td><td>23 25.0</td><td>23 25.5</td></tr> <tr><td>1788</td><td>23 25.6</td><td>-</td><td>-</td><td>-</td><td>-</td><td>23 28.9</td><td>23 29.3</td><td>23 29.8</td><td>-</td><td>23 32.1</td><td>-</td><td>-</td></tr> <tr><td>1789</td><td>-</td><td>-</td><td>-</td><td>-</td><td>-</td><td>23 34.2</td><td>-</td><td>-</td><td>-</td><td>-</td><td>23 41.2</td><td>-</td></tr> <tr><td>1790</td><td>23 38.9</td><td>-</td><td>-</td><td>-</td><td>-</td><td>23 39.0</td><td>-</td><td>-</td><td>-</td><td>-</td><td>-</td><td>-</td></tr> <tr><td>1791</td><td>23 35.6</td><td>-</td><td>23 36.0</td><td>-</td><td>-</td><td>23 36.7</td><td>-</td><td>-</td><td>-</td><td>-</td><td>-</td><td>-</td></tr> <tr><td>1792</td><td>23 41.1</td><td>-</td><td>-</td><td>23 41.9</td><td>-</td><td>23 43.6</td><td>23 43.9</td><td>23 45.6</td><td>23 45.9</td><td>23 45.2</td><td>-</td><td>-</td></tr> <tr><td>1793</td><td>23 46.9</td><td>23 48.3</td><td>23 48.8</td><td>23 46.2</td><td>23 47.3</td><td>23 48.5</td><td>23 50.5</td><td>23 48.6</td><td>23 52.6</td><td>23 52.3</td><td>23 51.9</td><td>23 52.3</td></tr> <tr><td>1794</td><td>23 54.2</td><td>-</td><td>-</td><td>-</td><td>-</td><td>23 54.4</td><td>23 57.2</td><td>23 58.1</td><td>-</td><td>-</td><td>-</td><td>-</td></tr> <tr><td>1795</td><td>-</td><td>23 57.5</td><td>-</td><td>-</td><td>-</td><td>23 57.1</td><td>23 57.1</td><td>24 0.4</td><td>-</td><td>-</td><td>23 59.4</td><td>-</td></tr> <tr><td>1796</td><td>24 1.1</td><td>-</td><td>-</td><td>-</td><td>-</td><td>23 58.7</td><td>23 59.2</td><td>24 0.1</td><td>-</td><td>-</td><td>24 1.5</td><td>-</td></tr> <tr><td>1797</td><td>24 1.5</td><td>-</td><td>-</td><td>-</td><td>-</td><td>24 0.2</td><td>24 0.3</td><td>24 1.4</td><td>-</td><td>-</td><td>24 1.3</td><td>-</td></tr> <tr><td>1798</td><td>24 0.6</td><td>-</td><td>-</td><td>-</td><td>-</td><td>24 59.4</td><td>24 0.0</td><td>24 1.4</td><td>-</td><td>-</td><td>24 1.4</td><td>-</td></tr> <tr><td>1799</td><td>24 1.1</td><td>-</td><td>-</td><td>-</td><td>-</td><td>24 0.6</td><td>24 1.8</td><td>24 2.0</td><td>-</td><td>-</td><td>24 2.3</td><td>-</td></tr> <tr><td>1800</td><td>24 3.6</td><td>-</td><td>-</td><td>-</td><td>-</td><td>24 1.8</td><td>24 3.0</td><td>24 3.6</td><td>-</td><td>-</td><td>24 3.3</td><td>-</td></tr> <tr><td>1801</td><td>24 5.2</td><td>-</td><td>-</td><td>-</td><td>-</td><td>24 2.8</td><td>24 4.1</td><td>24 3.8</td><td>-</td><td>-</td><td>24 5.4</td><td>-</td></tr> <tr><td>1802</td><td>24 6.9</td><td>-</td><td>-</td><td>-</td><td>-</td><td>24 5.3</td><td>24 6.0</td><td>24 8.7</td><td>-</td><td>-</td><td>24 6.8</td><td>-</td></tr> <tr><td>1803</td><td>24 8.0</td><td>-</td><td>-</td><td>-</td><td>-</td><td>24 7.0</td><td>24 7.9</td><td>24 10.5</td><td>-</td><td>-</td><td>24 10.7</td><td>-</td></tr> <tr><td>1804</td><td>24 9.4</td><td>-</td><td>-</td><td>-</td><td>-</td><td>24 6.0</td><td>24 8.4</td><td>24 8.9</td><td>-</td><td>-</td><td>24 9.6</td><td>-</td></tr> <tr><td>1805</td><td>24 8.7</td><td>-</td><td>-</td><td>-</td><td>-</td><td>24 7.8</td><td>24 7.8</td><td>24 10.0</td><td>-</td><td>-</td><td>24 9.4</td><td>-</td></tr> </table>

* Phil. Trans. 1806. p. 416. Charts have been constructed for shewing the declination of the needle in various parts of the earth by means of curve lines. Respecting these charts and several other circumstances with regard to this subject, see VARIATION of the COMPASS.

It may not be improper here to point out the general method of applying the polarity of the magnet to the useful purposes of navigation, mining, &c.

A mariner's compass, or magnetic needle in a cafe, is so placed as to be as little as possible disturbed by the motion of the vessel, person, &c. In a ship, it is placed in the binnacle (see BINNACLE), or suspended from the upper deck in the cabin. Then the head of the vessel is kept by the helm in such a direction as to make any required angle with the line of the needle, or the person (in mining or travelling) advances in a similar manner. Thus, supposing that a vessel sets out from a certain part, in order to go to another place that is exactly westward of the former; as for example, from the Land's End in Cornwall to Newfoundland on the coast of North America. The vessel must be directed in such a way, as that its course may be always at right angles with the direction of the magnetic needle, or so that the part of the needle or compass card, which points to the northward, (allowing for the variation) may be always kept to the right hand of the man at the helm, or to the starboard side of the vessel. The reason of this is evident; for, supposing the needle to point duly north and south, the direction of east and west being perpendicular to it, this must be the true course of the vessel. From this example, a little reflection will easily point out how a vessel may be steered in any other course (a).

The declination of the magnetic needle is disturbed by the near approach of a ferruginous body, especially if this be of considerable size.

On holding the extremity of a pretty large iron rod, such as a poker, near one end of a magnetic needle properly suspended, the needle will be found to turn considerably from its usual direction. This circumstance, though proper to be mentioned here, will be better understood when we have considered the attractive power of the magnet. The fact is useful, as it teaches us to keep magnetic needles in such a situation as not to be acted on by any considerable body of iron.

A magnet, whether natural or artificial, has a greater effect in disturbing the polarity of a magnetic needle than is produced by iron.

Magnetic polarity seems also to be affected by changes in the state of the atmosphere; and the following axioms respecting this effect on the declination of the needle, collected by M. la Cotte, are deserving of attention.

1. The greatest declination of the needle from the north towards the west, takes place about two in the afternoon; and the greatest approximation of it towards the north, about eight in the morning; so that from the last-mentioned hour till about two in the afternoon, it endeavours to remove from the north, and between two in the afternoon and the next morning, to approach it.

2. The annual progress of the magnetic needle is as follows:—Between January and March, it removes from the north; between March and May it approaches it; in June it is stationary; in July it removes from it; in August, September, and October it approaches it; its declination in October is the same as in May; in November and December it removes from the north; its greatest western declination is at the vernal equinox, and its greatest approximation to the north, at the autumnal equinox.

3. The declination of the magnetic needle is different, according to the latitude; among us, (i.e. in France) it has always increased since 1657; before that period it was easterly.

4. Before volcanic eruptions and earthquakes, the magnetic needle is often subject to very extraordinary movements.

5. The magnetic needle is agitated before and after the appearance of the northern lights: its declination on these occasions is about noon greater than usual.

So much has already been said respecting the phenomena, &c. of the dipping needle, under the article DIPPING NEEDLE, that it is unnecessary here to add much more on the subject. It was there noticed, that at the equator the dipping needle lies quite horizontal, and that one of its extremities inclines more towards the earth, according as the instrument is carried farther from the equator. We may here add, that from some late observations made by experimentalists with balloons, it appears that the higher we ascend above the surface

(a) In reply to some inquiries respecting the mode of employing the compass in mining, we were favoured by an ingenious friend, who is manager of one of the most extensive coalworks in this island, with the following remarks: "The compass is used in all mines where great accuracy is required. In some coal mines the cleats or faces of the coal are the guides to the miners in excavating the mine, and the compass is used to ascertain the situation and extent of the excavations. In other coal-mines the courses of the excavations are at first directed by the compass. In doing this, the compass is placed in a given situation, and is made to point the desired course. Then from the centre of one fight a perpendicular line is conveyed to the roof of the mine, and a small mark is there made with chalk; then a person looks at a candle (placed so as nearly to touch the roof), through the lower part of the sight of the compass nearest to him, and through the upper part of the opposite sight. The candle at the roof is moved in any direction until he sees it through both sights of the compass. It is then in a proper place, and a chalk mark is made in the roof immediately above it. A line struck with a chalked chord, between these two marks upon the roof, marks the proper course, by which the workmen are directed in making the excavation. By applying one part of a chalked cord along part of the course or white line thus begun on the roof, and extending the other part of the cord past it to any required distance, and then striking the cord, the course may be continued from time to time as the excavation advances." Magnetism.

In an aerostatic voyage made at St Petersburgh in Experiment 1804 by M. M. Sacharof and Robertson, it was observed that the south pole of a magnetic needle, balanced on a pin, dipped below the horizon nearly 10 degrees.

The following table shows the magnetic dip as observed at several different places at various times.

<table> <tr> <th>Latitude.</th> <th>Longitude.</th> <th>N. Pole below the Horizon.</th> <th>Years of Observation.</th> <th>Latitude.</th> <th>Longitude.</th> <th>N. Pole below the Horizon.</th> <th>Years of Observation.</th> </tr> <tr> <td colspan="4">North.</td> <td colspan="4">South.</td> </tr> <tr> <td>53° 55'</td> <td>193° 39'</td> <td>69° 10'</td> <td>1778</td> <td>0° 3'</td> <td>27° 38'</td> <td>30° 3'</td> <td></td> </tr> <tr> <td>49 36</td> <td>233 10</td> <td>72 29</td> <td></td> <td>4 40</td> <td>30 34</td> <td>22 15</td> <td></td> </tr> <tr> <td colspan="4">West.</td> <td>7 3</td> <td>33 21</td> <td>17 57</td> <td></td> </tr> <tr> <td>44 5</td> <td>8 10</td> <td>71 34</td> <td>1776</td> <td>11 25</td> <td>34 24</td> <td>9 15</td> <td></td> </tr> <tr> <td>38 53</td> <td>12 1</td> <td>70 30</td> <td></td> <td>16 45</td> <td>208 12</td> <td>29 28</td> <td></td> </tr> <tr> <td>34 57</td> <td>14 8</td> <td>66 12</td> <td></td> <td>19 28</td> <td>204 11</td> <td>41 0</td> <td></td> </tr> <tr> <td>29 18</td> <td>16 7</td> <td>62 17</td> <td></td> <td>21 8</td> <td>185 0</td> <td>39 1</td> <td>1777</td> </tr> <tr> <td>24 24</td> <td>18 11</td> <td>59 0</td> <td></td> <td>35 55</td> <td>18 20</td> <td>45 37</td> <td>1774</td> </tr> <tr> <td>20 47</td> <td>19 36</td> <td>56 15</td> <td></td> <td>41 5</td> <td>174 13</td> <td>63 49</td> <td>1777</td> </tr> <tr> <td>15 8</td> <td>23 38</td> <td>51 0</td> <td></td> <td>45 47</td> <td>166 18</td> <td>70 3</td> <td>1773</td> </tr> <tr> <td>12 1</td> <td>23 35</td> <td>48 26</td> <td></td> <td colspan="4">Prince of Wales's Island.</td> </tr> <tr> <td>10 0</td> <td>22 52</td> <td>44 12</td> <td></td> <td>5 10</td> <td></td> <td></td> <td>1799</td> </tr> <tr> <td>5 2</td> <td>20 10</td> <td>37 25</td> <td></td> <td colspan="4"></td> </tr> </table>

Table of the Magnetic Dip at London from 1786 to 1805.*

<table> <tr> <th rowspan="2">Poles Reversed.</th> <th colspan="2">Face east.</th> <th colspan="2">Face west.</th> <th colspan="2">True dip.</th> </tr> <tr> <th>Face east.</th> <th>Face west.</th> <th>Face east.</th> <th>Face west.</th> <th>True dip.</th> </tr> <tr> <td>1786 September</td> <td>72 28,7</td> <td>72 1,4</td> <td>71 57,3</td> <td>72 5,1</td> <td>72 8,1</td> </tr> <tr> <td>October</td> <td>72 29,9</td> <td>71 59,0</td> <td>72 0,4</td> <td>72 1,2</td> <td>72 7,6</td> </tr> <tr> <td>November</td> <td>72 7,6</td> <td>72 17,6</td> <td>72 2,4</td> <td>71 46,7</td> <td>72 3,6</td> </tr> <tr> <td>December</td> <td>72 10,6</td> <td>72 2,2</td> <td>72 2,2</td> <td>71 58,4</td> <td>72 3,4</td> </tr> <tr> <td>1787 January</td> <td>72 11,4</td> <td>72 1,8</td> <td>72 1,0</td> <td>71 56,0</td> <td>72 2,5</td> </tr> <tr> <td>February</td> <td>72 19,4</td> <td>72 10,8</td> <td>72 1,5</td> <td>71 55,8</td> <td>72 6,9</td> </tr> <tr> <td>March</td> <td>72 19,1</td> <td>72 11,9</td> <td>72 0,5</td> <td>71 52,2</td> <td>72 5,9</td> </tr> <tr> <td>April</td> <td>72 24,4</td> <td>72 9,5</td> <td>72 0,5</td> <td>71 52,2</td> <td>72 6,6</td> </tr> <tr> <td>May</td> <td>72 24,4</td> <td>72 9,6</td> <td>72 4,2</td> <td>71 52,9</td> <td>72 7,8</td> </tr> <tr> <td>June</td> <td>72 22,6</td> <td>72 7,9</td> <td>72 4,2</td> <td>71 52,9</td> <td>72 6,8</td> </tr> <tr> <td>July</td> <td>72 22,6</td> <td>72 7,9</td> <td>71 59,9</td> <td>71 55,1</td> <td>72 6,4</td> </tr> <tr> <td>August</td> <td>72 22,3</td> <td>72 6,7</td> <td>72 59,3</td> <td>71 55,2</td> <td>72 5,9</td> </tr> <tr> <td>September</td> <td>72 22,3</td> <td>72 6,7</td> <td>72 2,9</td> <td>71 51,0</td> <td>72 3,7</td> </tr> <tr> <td>October</td> <td>72 23,1</td> <td>72 2,5</td> <td>72 2,9</td> <td>71 51,0</td> <td>72 4,9</td> </tr> <tr> <td>November</td> <td>72 23,1</td> <td>72 2,5</td> <td>72 2,7</td> <td>71 50,3</td> <td>72 4,7</td> </tr> <tr> <td>December</td> <td>72 22,8</td> <td>72 2,0</td> <td>72 2,7</td> <td>71 50,3</td> <td>72 4,4</td> </tr> <tr> <td>1788 January</td> <td>72 22,8</td> <td>72 2,0</td> <td>72 2,6</td> <td>71 48,8</td> <td>72 4,0</td> </tr> <tr> <td>1789 January</td> <td>72 16,0</td> <td>72 0,9</td> <td>71 51,9</td> <td>71 31,1</td> <td>71 54,8</td> </tr> <tr> <td>December</td> <td>72 17,5</td> <td>71 59,4</td> <td>71 38,9</td> <td>71 42,8</td> <td>71 54,6</td> </tr> <tr> <td>1790 January</td> <td>72 16,9</td> <td>71 57,7</td> <td>71 49,2</td> <td>71 49,2</td> <td>71 53,7</td> </tr> <tr> <td>1791 January</td> <td>71 43,9</td> <td>71 36,1</td> <td>71 37,2</td> <td>71 17,5</td> <td>71 23,7</td> </tr> <tr> <td>1795 October</td> <td>71 12,8</td> <td>71 9,5</td> <td>71 13,9</td> <td>71 9,4</td> <td>71 11,4</td> </tr> <tr> <td>1797 October</td> <td>71 4,9</td> <td>71 10,9</td> <td>70 56,3</td> <td>70 44,7</td> <td>70 59,2</td> </tr> <tr> <td>1798 April</td> <td>71 4,7</td> <td>71 14,5</td> <td>71 2,3</td> <td>70 19,8</td> <td>70 55,4</td> </tr> <tr> <td>October</td> <td>70 55,6</td> <td>71 14,5</td> <td>71 7,7</td> <td>70 22,2</td> <td>70 55,0</td> </tr> <tr> <td>1799 October</td> <td>70 56,0</td> <td>71 13,5</td> <td>71 11,5</td> <td>70 7,9</td> <td>70 52,2</td> </tr> <tr> <td>1801 April</td> <td>70 47,4</td> <td>71 5,6</td> <td>70 52,4</td> <td>69 38,2</td> <td>70 35,6</td> </tr> <tr> <td>1803 October</td> <td>70 30,9</td> <td>71 9,9</td> <td>70 40,5</td> <td>69 46,7</td> <td>70 32,0</td> </tr> <tr> <td>1805 August</td> <td>70 25,2</td> <td>70 55,7</td> <td>70 26,9</td> <td>69 36,3</td> <td>70 21,0</td> </tr> </table>

* Phil. Transf. 1806. p. 491. To what was said under DIPPING Needle, respecting the construction of that instrument, we may add, that notwithstanding the great improvements that have been lately made in the arts, the making of a dipping needle is one of the most delicate and difficult tasks that an instrument-maker can undertake. The needle must be made of tempered steel which we are certain has no magnetism before it is touched; it must be poised so nicely, and with such a perfect coincidence of its centre of gravity and axis of motion, that it will retain any position (before being magnetized) that is given it. A good dipping needle cannot be had below twenty guineas.

SECT. II. On Magnetic Attraction and Repulsion.

A magnet attracts iron, and all bodies, into the composition of which iron enters in any considerable degree. This principle is illustrated by very simple experiments, which will readily occur to every reader. It is of consequence here to observe, that the purer and hotter the iron to which the magnet is presented, the stronger will be the attraction; thus, a magnet attracts a piece of soft and clean iron much more strongly than it attracts any other ferruginous body of the same shape and weight. Hard steel, or the harder ores of iron, are less forcibly attracted than soft steel, and still less than soft iron; and all pieces of iron are less forcibly attracted in proportion as they are more oxygenated.

The attractive power of a magnet is not equally strong on every part of its surface. It is most powerful at the poles of the magnet, and it is found to diminish in proportion as the part of the surface is more distant from the poles. Thus, in an oblong magnet, the attraction is least at about its middle, where it is often very trifling.

It is by this property of the magnet that we are enabled to discover the poles of a magnet, where they are not yet ascertained; a circumstance which is often necessary with respect to natural magnets, in which, when of an irregular shape, it would otherwise be difficult to discover the poles. The usual method of ascertaining the poles of a magnetic body is, to present various parts of the body to be examined, successively to the poles of a magnetic needle, when it will soon be discovered which parts of the body have most influence on the needle, by the pole of the latter standing perpendicularly to that part of the body. It will presently appear, that in this way it may also be ascertained which of these poles is the north, and which the south, as the south pole of the body under examination will have most influence on the north pole of the needle, and vice versa.

A good magnet should have no more than two poles, and these should be situated in the extreme surface of the magnet; but it sometimes happens, especially in natural magnets, and in artificial magnetic bars, if they be very long, that there are more than two poles, or that the poles are very confused. For example, in a very long magnetized bar, there may be a strong north pole at one extremity, a south pole a little farther on, then a weaker north pole, and so on to the extremity, which will be found possessed of a still weaker south polarity. These poles are to be discovered by presenting to the several parts of the bar one or other of the poles of a magnetic needle; for, as we shall immediately experiment, each pole of the needle will be attracted towards that part of the rod which is possessed of the contrary polarity.

The attractive power of the magnet and the iron is most forcible when the two bodies are in contact, and it diminishes as they are made to recede from each other. The exact law according to which this diminution takes place, has not yet been completely ascertained. We shall see in the next chapter, what approximation has been made to it.

A magnet is not capable of lifting above a certain weight of iron; and all magnets of the same form and size are not able to lift the same weight. Among the natural magnets the smallest seem in general to possess a greater attractive power in proportion to their size, than those of larger dimensions. Mr Cavalcot mentions a small magnet that weighed not more than 6 or 7 grains, and was capable of lifting about 300 grains; and Sir Isaac Newton possessed a magnet that he wore in a ring, weighing but about 3 grains, which is said to have lifted 746 grains, or nearly 250 times its own weight. The larger natural magnets are very weak in proportion to these. Those of two pounds scarcely lift more than ten times their own weight. It seems extraordinary, that a piece cut off from a large magnet is sometimes much stronger in respect of its attractive power, than the magnet from which it was taken.

It has been said that the attractive power of magnets is greatest at their poles. Both poles, however, are seldom equal in this respect; and it appears, that in these northern parts of the world, the north pole of magnets is more powerful than the south. In the southern hemisphere the contrary effect is said to take place. The attractive power of the magnet is most forcible when both poles are made to act conjointly; hence an armed magnet, or one of the horse-hoe form, is best adapted for experiments on the force of magnetic attraction.

It is of little consequence whether the iron that is presented to the magnet be in one piece, or consist of several pieces. The attraction is indeed stronger in the former case; but if several pieces of iron are presented to the magnet, they will either all adhere about the pole of the magnet, or will adhere to each other, so as to form a sort of chain. If a small iron ball be made to adhere to the pole of a magnet, this ball will support a second; and this latter, if the magnet be pretty strong, will support a third. If the magnet be of the horse-hoe form, and have these three balls hanging by one ball, if two others be suspended from the other pole, all the five may be made to adhere, so as to form a curved chain. It will be evident, that pieces of iron which present a greater extent of surface than the above spherical balls, will be more powerfully attracted.

One of the most pleasing experiments on the attraction of the magnet for iron, is shewn by means of iron filings.

Exper.—Let a paper be placed above a bar magnet, and let iron filings be shaken on the paper through a gaule sieve. They will arrange themselves round the magnet in a very beautiful manner, forming curves and arches of curves, as represented in fig. 6. At the two ends of the magnet, as a a, there are chains of filings standing out nearly perpendicular; and along the sides they form complete curves, bending outwards away from the magnet towards its middle, and having their extremities bounded by the poles of the magnet; and at the corners there are a number of arches that seem to form imperfect curves.

A similar effect may be produced by sowing iron filings on a piece of paper, so as to leave a vacancy in the middle, capable of receiving a bar magnet. When the magnet is placed on the paper, and the paper gently tapped, so as to agitate the filings, these will arrange themselves about the magnet, in curves, as above described.

The form of these curves will be better defined if the magnet be laid at the bottom of an earthen or glass vessel of water, and the iron filings be sifted over it so as to pass through the water.

The attraction between a magnet and a ferruginous body is mutual.

Exper.—Place a piece of iron or other ferruginous body upon a piece of cork or wood, so that it may float on the surface of water in an earthen or wooden vessel. Bring a magnet within a moderate distance of the floating body, and the latter will approach the former, and may be drawn by it in any direction. Again, place the magnet on cork or wood, so as to float on the water, and present to it a piece of iron, or other ferruginous body. The magnet will now approach the iron, and may be drawn by it as the iron was before. Lastly, Place both the magnet and the iron on separate pieces of wood or cork, within a moderate distance of each other, on the surface of the water. They will gradually approach each other, with a velocity that becomes greater in proportion as they approach nearer each other.

Magnetic attraction is not sensibly impeded by the interposition of bodies of any kind, that do not contain iron in their composition.

Exper.—Suppose that a magnet, placed at the distance of an inch from a piece of iron, exerts a certain degree of attraction, it will be found that the attraction is not sensibly weakened by the interposition of a plate of glass, a sheet of paper, a piece of copper, or any other similar substance. A needle, inclosed in a glass globe, will be still attracted by the magnet.

It is not easy to ascertain correctly the degree of attractive force exerted between a magnet and a ferruginous body. The usual method of observing this is, to fasten a magnet to one arm of a balance, and placing the body to be attracted at different distances below the magnet, to counterpoise the attraction with weights placed in the opposite scale of the balance. Proceeding in this way, then, if we find that it requires the weight of an ounce to counterpoise the attractive power of a magnet, when presented immediately to a piece of iron, it will be found that it requires the same counterpoise, if a plate of any matter that is not ferruginous be interposed.

Not only is iron attracted by the magnet, but under certain circumstances, one piece of iron exerts an attractive power on another piece of iron.

Exper.—Let an oblong piece of iron be fixed in a spherical piece of wood or cork, so as to float in water in the true magnetic line, as in Exper. 2. of No. 16. When the iron is nearly in the magnetical position, bring the extremity of a large iron rod, as the point of a new poker, holding it in a position not very different from that of the iron wire, within about a quarter of an inch of the upper extremity of the floating iron, and hold it there for some time, a little towards one side. The floating wire will gradually approach the iron rod with an accelerated motion, will at length touch it, and may be drawn through the water in any direction. A similar attraction will take place between the head of the poker and the extremity of the wire that is below the water.

The attractive power of the magnet is increased by the near approach of a piece of iron.

Exper. 1.—Suppose we have a magnetic bar that is capable of supporting a small key by one of its extremities, but which will not lift a key somewhat larger. If we bring a considerable oblong piece of iron near the opposite extremity of the bar, it will be found capable of supporting the larger key, or at least of lifting a weight somewhat greater than it sustained before.

Exper. 2.—Let an oblong magnetic bar be supported in a horizontal position, and let a piece of iron wire, about an inch in length, be hung by a short thread, so that its extremity is just opposite one of the poles of the magnetic bar, but so far out of the reach of the bar's attractive power as not to be brought from the perpendicular. Now, if a considerable iron bar be brought with one end within a moderate distance of the opposite pole of the magnet, the suspended wire will be drawn towards the magnet, thus shewing that the power of the latter has been increased by the juxtaposition of the bar of iron. If the bar of iron be brought till nearer the opposite pole of the magnet, the suspended wire will be drawn still nearer its adjacent pole; but if the bar of iron be drawn back from the magnet, the wire will fall into its original position.

This fact leads to many important practical conclusions in the management of magnets. As the juxtaposition of iron to the poles of a magnet improves its powers, we may infer, that if we keep a piece of soft iron in contact with the poles, the magnet will be improved by it; and this is in fact the case, and it shews the utility of the armature and conductor mentioned in No. 13. But of this more hereafter.

The attractive power of a magnet may be improved by increasing the weight appended to it.

This is best shown by a horse-hoe magnet, having a conductor of soft iron attached to its two poles, and a brafs ring at the convex part by which it may be suspended. If a small bag be hung to the conductor, and if the magnet is capable of sustaining a certain weight at any particular time, it will, by adding a little more, suppose a few shot, to the bag, at moderate intervals, be made to support gradually a much greater weight. If the magnet, on a first trial, was able to lift a small key, it will soon be able to lift a larger one, &c. How far this increase of power may be carried, has not, we believe, yet been ascertained.

It sometimes happens that a magnet does not shew any great attractive power, as exemplified in its power of lifting a considerable weight of iron, though it may have a great effect in exciting or in altering magnetic polarity. This was observed by Dr Gilbert, who re- marks that the directive power of a magnet extends farther than its attractive power.

The contrary poles of two magnets attract each other; that is, the north the south, and vice versa.

Exper. 1.—Place two oblong magnets on cork or wood, so as to float in water, or suspend each by a pretty long thread, with the north pole of the one opposed to the south pole of the other. They will gradually approach, and will at length rush together.

A similar effect will be produced, if the north pole of a bar magnet be held near the south pole of a magnetic needle; the latter will be attracted, and the same thing will happen if the south pole of the bar is brought near the north pole of the needle.

Exper. 2.—Take two semicircular magnets, and dip their extremities into iron filings. The filings will of course adhere to the extremities of the magnets, and will appear as if radiating from them. Now, present the two magnets with their adhering filings to each other, so that the north and south pole of the one is opposite to the contrary poles of the other, and the iron filings at their extremities will approach each other, and coalesce, as represented in fig. 7.

The attraction exerted between two magnets is not so strong in proportion, as between a magnet and a piece of soft iron in contact; but it has been found to commence at a greater distance.

The corresponding poles of two magnets repel each other; that is, the north the north, and the south the south.

Exper. 1.—Make the two magnets float on water, or suspend them by threads, so that the north or south pole of the one may be opposite to the north or south pole of the other. They will recede from each other; and the repulsion will evidently be greater, in proportion as they are brought nearer together.

Exper. 2.—Take two semicircular magnets, and dip their ends in iron filings, as mentioned above. Present them to each other, so that their corresponding poles may be mutually opposed. The filings at their extremities will start back, and leave a vacancy between the opposed poles of the magnets, somewhat like what is represented in fig. 8.

It sometimes happens that the corresponding poles of two magnets do not repel each other, but either mutually attract, or are quite indifferent. In this case, it will, in general, be found that one of them is stronger than the other; and the reason of the phenomenon will appear hereafter.

The repulsive power of a magnet is generally in a less proportion than its attractive power.

It is by the attractive power of the magnet that we usually ascertain whether any substance be magnetic; that is, whether the magnet possess any attractive power for it. If the body contain any considerable quantity of iron in its composition, its magnetism is easily ascertained, by approaching it with the pole of a pretty strong magnetic bar. If, however, the magnetism is too feeble to be discovered in this way, it may be ascertained by placing the body on a piece of cork or wood, so that it may float on the surface of water or mercury, in an earthen or wooden vessel, and bringing the pole of the magnet within a small distance of the floating body. It will sometimes be necessary to bring the magnet within one-tenth of an inch of the body, Exper. when, if it possesses any magnetism, it will gradually approach the magnet. This experiment is most satisfactory when the body to be examined is made to float on mercury; but in that case the vessel containing the mercury must not be too small, otherwise the natural convexity of the surface of the mercury will cause the floating body perpetually to fall down towards the sides of the vessel. A common soup plate will answer the purpose very well. It is also necessary that the mercury be very pure, and as clean as possible. To insure this, it will be proper, before using the mercury, to pass it through a conical piece of writing paper, rolled up so as to terminate in a very small aperture; or, what is better, to squeeze it through a pretty thick piece of flamous leather. It need scarcely be remarked, that in these delicate experiments, the air of the room should be kept as still as possible.

By the above methods, Mr Cavallo and others discovered, that the following substances are in some measure affected by the magnet. Most metallic ores, especially after their having been exposed to a fire; zinc, bismuth, and particularly cobalt, as well as their ores, are almost always attracted. Of the earths, the calcareous is the least, if at all, and the siliceous is the most frequently, attracted. The ruby, the chrysolite, and the tourmalin, are attracted. The emerald, and particularly the garnet, are not only attracted, but frequently acquire a permanent polarity. The opal is weakly attracted, especially after combustion. Most animal and vegetable substances, after combustion, are attracted. Even foot, and the dust which usually falls upon whatever is left exposed to the atmosphere, are sensibly attracted by the magnet.

"It has long ago been remarked, that platina, nickel, and several other bodies, acquire a sensible degree of magnetism; but some philosophers attribute this property only to a portion of iron not easy to be separated, and conclude, that by obtaining a greater degree of purity, we might succeed in rendering them perfectly indifferent to the action of the magnetic bar.

"The new experiments which Citizen Coulomb has made and repeated before the institute, lead us on the contrary to think, that the action of magnetism extends through all nature; for none of the bodies he has yet tried was found to resist this power.

"But however real this action may be, it is not alike in all bodies, and in most of them it must be necessarily very small, to have escaped the attention of philosophers to this time. In order therefore to exhibit and to measure these results, we must begin by placing the bodies in a situation which shall allow them to yield to the weakest action.

"For this purpose, Citizen Coulomb fashioned his subjects into the form of a cylinder or small bar; and in this state he suspended them to a filken thread, such as is drawn from the silk-worm's cone, and in this state he placed them between the opposite poles of two magnetic bars of steel. The single thread of silk could hardly bear the weight of a quarter of an ounce without breaking, consequently it became necessary to form small bars very light and thin. Citizen Coulomb made them about seven or eight millimetres in length (or less than half an inch), with three-fourths of a millimetre

"In his experiments he placed the steel bars in the same right line, their opposite poles being five or fix millimetres farther afarther than the length of the needle intended to oscillate between them. The result of the experiment shewed, that whatever might be the substance of the needle, it always disposed itself according to the direction of the two bars; and that if they were turned from this direction, they always recovered it, after oscillations of which the number was often more than 30 per minute. It was therefore easy in every case to determine, from the weight and figure of the needle, the force which had produced the oscillation.

"These experiments were successfully made with small needles of gold, silver, copper, lead, tin, small cylinders of glass, a piece of chalk, a fragment of bone, and different kinds of wood.

"Citizen Coulomb has proved, that the force of torsion of the silk thread is so slight, that in order to draw it round the entire circle, it would require a force scarcely equal to the one hundred thousandth part of a gramm, (or about one seven hundredth part of a grain). A quantity so minute cannot therefore sensibly derange the measure of magnetic force in the different bodies; and its effect, even if it were admitted to be of perceptible magnitude, may also be urged in proof of the general conclusion of Citizen Coulomb, because the magnetic power must overcome this resistance of the thread in order to manifest itself. Our author gives, in the third volume of the Memoirs of Natural Philosophy and Mathematics of the National Institute, a very simple formula to determine the magnetic force of a body from the time of its oscillations; and he means to shew in another memoir, the method of determining this result in different bodies of the same figure placed between the poles of two bars. He thinks it now proved, that all the elements which enter into the composition of our globe are subjected to the magnetic power, and that the whole mass collectively forms one single magnet.

"In favour of those who might be desirous of repeating his experiments, and rendering them very sensible, the author remarks, that the method of succeeding consists in diminishing the size of the oscillating bodies. From some essays, of which the results terminate this memoir, it seems to follow, that the accelerating forces are inversely as the masses, or very nearly in the direct proportion of the surfaces; but Citizen Coulomb gives this rule only as a first deduction, which requires to be confirmed*."

The opinion of the general influence of magnetism on all terrestrial bodies was, as we shall see hereafter, maintained by our countryman Dr Gilbert, though Coulomb has certainly the merit of having put it to the test of experiment.

Besides the experiments which we have related, there are some that depend on the attractive power of the magnet, and which are ranked among scientific amusements. We shall here describe a few of these.

Before we relate the manner of making these experiments, it may be proper to describe an instrument that is employed in some of them. This, from its form and apparent use, is called the magnetic perspective glas, and is thus constructed.

Provide an ivory tube about 2 1/2 inches long, and of such a form as is expressed in fig. 9. The sides of this tube must be so thin as to admit a considerable quantity of light. It is to open at one end with a forew, and at that end must be placed an eye-glas of about two inches focus, and at the other end any glas you please. Have a small magnetic needle like that in a compass. Fig. 9. It must be strongly touched, and to placed at the bottom of the tube that it may turn freely round. It is to be fixed on the centre of a small ivory circle C, of the thickness of a counter, placed on the object-glas D, and painted black on the side next it. This circle must be kept fast by a circular rim of pasteboard, that the needle may not rise off its pivot, in the same manner as in the compass. This tube will thus become a kind of compass sufficiently transparent to show the motions of the needle. The eye-glas serves more clearly to distinguish the direction of the needle, and the glas at the other end, merely to give the tube the appearance of a common perspective glas. It will appear, from what has been already stated, that the needle in this tube, when placed over and at a small distance from a magnet, or any machine in which it is contained, will necessarily place itself in a position directed by that magnet, and consequently show where the north and south pole of it is placed; the north end of the needle constantly pointing to the south end of the magnet. This effect will take place, though the magnet be enclosed in a case of wood, or even metal. You must observe, however, that the attracting magnet must not be very far distant from the needle, especially if it be small, as in that case its influence extends but to a short distance. This tube may be differently constructed, by placing the needle in a perpendicular direction, on a small axis of iron, on which it must turn quite freely, between two small plates of brass placed on each side of the tube; the two ends of the needle should be in exact equilibrium. The north and south ends of the needle will, in like manner, be attracted by the south and north ends of the magnetic bar. The former construction, however, appears preferable, as it is more easily excited, and the situation of the needle much more easily distinguished.

Exp. 1. The Communicative Piece of Money.

Take a crown or dollar, and drill a hole in the side of it, in which place a piece of wire, or a large needle well polished, and strongly touched with a magnet. Then close the hole with a small piece of pewter, that it may not be perceived. Now, the needle in the magnetic perspective before described, when it is brought near to this piece of money, will fix itself in a direction corresponding to the wire or needle in that piece. Desire any person to lend you a crown piece or dollar, which you dexterously change for one that you have prepared as above. Then give the latter piece to another person, and leave him at liberty either to put it privately in a snuff-box, or not; he is then to place the box on a table, and you are to tell him by means of your glas, whether the crown is or is not in the box. Then bringing your perspective close to the box, you will know, by the motion of the needle, whether it be there or not; for as the needle in the perspective will always keep to the north of itself, if you do not perceive it has any motion, you conclude the crown is not in the box. It may happen, however, that the wire in the crown may be placed to the north, in which case you will be deceived. Therefore, to be sure of success, when you find the needle in the perspective remain stationary, you may, on some pretence desire the person to move the box into another position, by which you will certainly know whether the crown-piece be there or not. You must remember that the needle in the perspective must here be very sensible, as the wire in the crown cannot possibly have any great attractive force.

Exp. 2. The Magnetic Table.

Under the top of a common table place a magnet that turns on a pivot, and fix a board under it, that nothing may appear. There may also be a drawer under the table, which you pull out to shew that there is nothing concealed. At one end of the table there must be a pin that communicates with the magnet, and by which it may be placed in different positions; this pin must be so placed as not to be visible to the spectators. Strew some steel filings or very small nails over that part of the table where the magnet is. Then ask any one to lend you a knife or a key, which will then attract part of the nails or filings. Then placing your hand in a careless manner on the pin at the end of the table, you alter the position of the magnet, and giving the key to any person, you desire him to make the experiment, which he will then not be able to perform. You then give the key to another person, at the same time placing the magnet, by means of the pin, in the first position, when that person will immediately perform the experiment.

Exp. 3. The Mysterious Watch.

You desire any one in company who has a watch with a steel balance (b), to lend it you for a few minutes, asking him whether it will continue to go when laid on the table. He will probably say it will. To prove to him that he is wrong, you lay it on that part of the table below which you have previously placed a strong bar-magnet (as in Exp. 2.), so that the watch may be above one of the poles. It will immediately stop. Now, if you shift the position of the magnet, and give the watch to another person to lay it on the table, it will not stop; but replacing the magnet, and desiring a third person to try the experiment, he will succeed. All this, to those who are not acquainted with the secret, will appear very extraordinary.

Exp. 4. The Magnetic Dial.

Provide a circle of wood or ivory, of about five or six inches diameter, as fig. 10., which must turn quite free on the stand B (fig. 11.), in the circular border A : on the circle must be placed the dial of pasteboard C (fig. 10.), whose circumference is to be divided into 12 equal parts, in which must be inscribed the numbers from 1 to 12, as on a common dial. There must be a small groove in the circular frame D, to receive the pasteboard circle; and observe that the dial must be made to turn so freely that it may go round without moving the circular border in which it is placed. Between the pasteboard circle and the bottom of the frame, place a small artificial magnet E (fig. 12.) Fig. 12. that has a hole in its middle, or a small protuberance. On the outside of the frame place a small pin P, which serves to shew where the magnetic needle I, that is placed on a pivot at the centre of the dial, is to stop. This needle must turn quite freely on its pivot, and its two sides should be in exact equilibrium. Then provide a small bag, that has five or fix divisions, like a lady's work-bag, but smaller. In one of these divisions put small square pieces of pasteboard on which are written the numbers from 1 to 12, and if you please you may put several of each number. In each of the other divisions you must put 12 or more like pieces, observing, that all the pieces in each division must be marked with the same number. Now the needle being placed upon its pivot, and turned quickly about, it will necessarily stop at that point where the north end of the magnetic bar is placed, and which you previously knew by the situation of the small pin in the circular border. You therefore present to any person that division of the bag which contains the several pieces on which is written the number opposite to the north end of the bar, and tell him to draw any one of them he pleases. Then placing the needle on the pivot, you turn it quickly about, and it will necessarily stop, as we have already said, at that particular number.

Another experiment may be made with the same dial, by desiring two persons to draw each of them one number out of two different divisions of the bag; and if their numbers, when added together, exceed 12, the needle or index will stop at the number they exceed it; but if they do not amount to 12, the index will stop at the sum of those two numbers. In order to perform this experiment, you must place the pin against the number 5, if the two numbers to be drawn from the bag be 10 and 7; or against 9, if they be seven and two. If this experiment be made immediately after the former, as it easily may, by dexterously moving the pin, it will appear the more extraordinary.

Exp. 5. The Divining Circles.

On the top of a thin box, as AB fig. 13. paste two Divining circles drawn on paper, as F, G, each of which is divided into compartments. In those of one circle, as G, Fig. 13. are written questions, and in those of the other, as F, appropriate answers. Through the centre of the circle G an axle passes, carrying a toothed wheel, and which works into the pinion d, to the axis of which is fixed another pinion, and this receives the teeth of another wheel g, whose axis is passed through the centre of the circle F. On the axis of the wheel c is to be fixed an index a above the paper circle, and to the axis of the wheel g, just below the cover of the box, is fixed a bar magnet q q, turning together with the axis; while on the part of the axis that projects above the circle F a loose needle x x is balanced, so as to move independently of the axis. A carton of strong paper, of the size

(b) The balance of a watch is sometimes, though very seldom, made of brass, when it is scarcely susceptible of magnetic influence. perimen- of F should cover the pasted circle, and turn easily on the centre &c; and it should have a triangular piece as F cut out, in order to see the answers. If now the needle be taken off its point, and a person be desired to ask some of the questions on the circle G, the index must be turned to the question, and then the needle placed on its pivot, giving it a whirl round. When it stops, its point will stand over the proper answer, which may be seen by turning the open part of the carton to that place.

SECT. III. Of the Communication and Production of Magnetism.

The whole of this important part of the subject may be said to depend on one general fact, which we shall therefore first lay down and illustrate.

Any piece of iron when in the neighbourhood of a magnet, is itself a magnet, and possesses all the material properties of that body.

Exp. 1.—Let there be a large and strong magnet properly supported in the horizontal direction, at a distance from iron or other similar bodies, and with its poles perfectly free. Take also any small piece of common iron, not more than two or three inches long, such as a common small key, and take another piece of iron, as a smaller key, or short piece of wire about the size of a goose quill.

In the first place hold the key in a horizontal position, with one end opposite one of the poles of the magnet, but so as not to be in contact with it. Then bring the other piece of iron to the other end of the key, and it will hang by the key, and will so continue to hang, though we withdraw the key from the magnet horizontally, till there is a certain interval between the key and the magnet, when the former will be no longer able to support the piece of iron. Even at this distance the key will, however, be found capable of supporting a piece of iron considerably smaller than the former, till its distance from the magnet be increased.

Again, hold the key with one extremity below one of the poles of the magnet, and touch the other extremity with the small piece of iron, the latter will adhere till the key be removed too far below the magnet.

Thirdly, Hold the key with one of its extremities above one of the poles of the magnet, but at such a distance that there is room for the small piece of iron to go between the key and the magnet, without touching the latter. The piece of iron will be supported by the key, as in the two former instances.

Fourthly, Let the magnet be placed in a vertical position, and hold the key with one extremity immediately below or above one of the poles. The piece of iron will be supported in a similar manner, in the former case by the extremity of the key that is most remote from the magnet, and in the latter by that which is adjacent.

If, instead of approaching the magnet with the key, we reverse the circumstances, the effect of the magnet in rendering the key magnetical will be still more evident. Suppose the piece of iron to be lying on the table; let one end of it be touched with the key, and there will be found no attraction between them: but if while we hold the key very near one extremity of the wire, we bring the pole of the magnet near the other end of the key, we shall see the wire rise from the table, and adhere to the key.

In all these cases the attractive power of the key, that is, its magnetism, is evidently derived from its juxtaposition to the magnet.

Exp. 2.—Let two pieces of iron wire be suspended by separate ends of a piece of thread, so that they may be hung from a pin in the wall in a situation parallel to each other, or in contact. Now bring one end of a bar magnet a little below the wires, and they will repel each other. If these wires are of soft iron, they will collapse immediately on the magnet being withdrawn; but if they are formed of hard iron or of steel, they will continue apart for a considerable time.

Here the two wires are, by the proximity of the magnet, become magnets, and the extremities next the bar have each acquired a similar polarity, i.e. both contrary to that of the adjacent pole of the bar. They, therefore, repel each other.

Exp. 3.—Let a bar-magnet, such as N, S, fig. 14. Fig. 14. be laid in a horizontal position, and let a small key, as B, C, be held near the north pole of the magnet, in the direction of its axis. Let a very small magnetic needle, supported on a sharp pivot, be brought near that end of the key C, which is most remote from N. The needle will immediately turn its south pole towards C, as is indicated by the feathered part of the arrow c. Hence it appears that the key has acquired a directive power like a magnet, and that its remote extremity performs the office of a north pole, as it attracts the south pole of the needle, and repels its north pole. If it be said that the magnetic needle in this case is affected directly by the directive power of the magnet, as it would take the above position though the key were not present; to shew that the effect is produced through the medium of the key, remove the needle into another situation as b, and it will still arrange itself with the same pole opposite C, and if it be carried to the proximate extremity of the key, as at a, it will turn round, and present its north pole to B, thus shewing that it is, at least in some measure, influenced by the key.

In general, when a piece of iron is presented to the pole of a magnet, the extremity next that pole is pol-acquires a fied of the contrary polarity, and the remote extremity has acquired a similar polarity. The situation of the poles, however, depends much on the form of the piece of iron, and on the part of its surface which is presented to the pole of the magnet. If the form be magnet, that of an oblong bar, one extremity of which is presented to the pole, which is the most usual case, the circumstances will be as we have just mentioned. If the oblong bar be presented to the pole in a perpendicular direction, with its middle very near the pole of the magnet, this middle point will be possessed of a polarity contrary to that of the adjacent pole, while the two extremities have acquired the same polarity. If the presented iron be in the form of a circular plate, and its centre be held near the pole of the magnet, this centre will have the contrary polarity, and every point of the circumference the same polarity. If the plate have its circumference fashioned into points, each of these points will acquire a very strong polarity, contrary to that of the pole near which the centre of the plate is held.

The communication of magnetic power from the magnet magnet to the key in the foregoing experiments, will be still more strongly illustrated by holding another piece of wire to the wire that is already suspended by the key. The new piece of wire will also be suspended, and so several more may be suspended by one another, like the links of a chain, according to the strength of the magnet. This fact was known to the ancients, who speak of a loadstone causing an iron ring to carry another ring; and that a third, till the whole puts on the appearance of a chain.

It will be found that the magnet has lost none of its power by producing magnetism in the iron, and of course, that nothing has been transferred from the magnet to the iron. The magnetism of the iron thus caused by its juxtaposition to a magnet is called induced magnetism, or magnetism by induction.

There is an apparent exception to the universality of the above proposition. If the key be held in such a position as that it shall be perpendicular to the magnet, with one extremity either opposite one of the poles, or a little above the centre of the magnet, the bit of wire will not be attracted by that extremity, and we may hence suppose that the key has acquired no magnetic power by its proximity to the magnet. But if we bring a needle or a piece of iron wire near its remote end, it will be strongly attracted, and shew that end to have the same polarity with the nearest pole of the magnet. Now, the ends both of the key and the wire that are next the magnet, having the same polarity with the pole of the magnet nearest them, cannot attract each other, but on the contrary will repel each other, and therefore the wire cannot adhere to the key, though by the change produced by the other extremity, it is evident that the key has acquired magnetic power.

There is, however, one exception. If the key in the first experiment, with the wire hanging to it, be carried from any of the situations there described, towards the middle of the magnet, the wire will fall off as soon as it arrives very near the middle. If we suppose a plane to pass through the centre of the magnet in a direction perpendicular to its axis, so as to form the magnetic equator, a slender piece of iron held anywhere within this plane can acquire no sensible magnetism, which is demonstrated by its shewing no signs of polarity, and not being attracted by the magnet. Now it is well known that the greatest activity of a magnet resides in its two poles, and that those magnets are the best in which this activity is least diffused. A certain circumference of every magnet is entirely inactive, as we see in the experiment with the iron filings described in No 26, where the filings collect themselves principally on two points of the surface, between which there is a space all round, to which no filings are attached. Many circumstances shew that the two poles of a magnet have contrary actions; the north pole producing a strong northern polarity in the remote end of an iron bar brought near it, and a south polarity in the proximate end, while an opposite effect is produced by its south pole. Now, adopting this principle, that the actions of the two poles are opposite, it follows that if these actions are equal, and act in a similar manner, each must counteract and prevent the action of the other, and produce what may be called a magnetic equilibrium. Therefore if a slender iron rod or thin plate be placed so that every part of it lies within the magnetic equator, it will exhibit no magnetism, will not be attracted by the magnet, and will not attract iron. This will be seen more satisfactorily when we have explained the theory of magnetism.

The consideration of the above important facts will enable us to explain, especially after what will be stated in the next chapter, the production or communication of magnetism in all the methods by which these are usually effected.

Magnetism may be produced artificially in a piece of iron or steel, by various methods.

1. By touching the iron or steel either with a natural magnet, or with a steel bar already magnetized.

The process of communicating magnetism by natural or artificial magnets, or by what has been called touching, has undergone various improvements and modifications, which we shall endeavour briefly to trace.

The most simple method of magnetizing a bar of old steel is to apply the north pole of a magnet to that extremity of the bar which we wish to acquire a fourth polarity. In this way, merely by contact, a slight degree of magnetic power will, after some time, be imparted to the bar, and the communication will be expedited by striking the bar so as to make it found. Only a slight degree of magnetism can, however, be communicated in this way, and unless the steel bar be very short, its poles will be much confused.

Another method of communicating magnetism to a bar of this kind is, to apply the pole of a magnet to one end of the bar, and pass it on to the other end, giving a moderate degree of pressure. This is repeated several times on both sides of the bar, taking care always to begin the stroke at the same end as at first, and instead of drawing the magnet back along the bar, lifting it up every time that we come to the other end. The following description will best explain the mode of communicating magnetism in this way, by one or two magnetic bars.

When only one magnetic bar is to be made use of, one of its poles must be applied as represented fig. 15. Fig. 15, where CD represents the needle or steel bar to be impregnated. The magnet AB is then to be drawn all along the surface of it, till it reaches the extremity D. The magnet being then removed, must be applied to the extremity C, and drawn over the needle as before. Thus the needle must be rubbed several times, by which means it will acquire a considerable degree of magnetism. In this method, the other extremity of the needle which the magnet touched last acquires the contrary magnetism; that is, if B be the north pole of the magnet, C will be the north pole and D the south of the needle. This method, however, is never found to be equally effectual with that in which two magnets, or both poles of one magnet, are made use of.

To communicate magnetism by means of two magnetic bars, place the bar or needle AB, fig. 16, upon a table; then set the two magnetic bars CD, EF, straight upright upon it at a little distance, equal on both sides from the middle of the bar AB, and in such a manner that the south pole D of one of the bars may be nearest to that end of the bar AB which is to become the north pole, &c. These two bars must then be slid gradually towards one extremity of the bar, keeping them constantly at the same distance from each other; and when one of them, for instance CD, is ar- Chap. II.

Experiments rived at A, then they must be slid the contrary way, till EF arrives at B; and thus the bar AB must be rubbed a greater or smaller number of times, till it will be found by trial to have acquired a considerable power. When the magnetic bars are powerful, and the bar AB of very good steel, and not very large, a dozen of strokes are fully sufficient; but when the bars are to be removed from the bar AB, care must be taken to bring them to the same situation where they were first placed; viz. a little and equal distance from the middle of the bar AB, from which they may be lifted up.

The mode of employing two bars instead of one was an improvement, and the method was still farther improved by placing them in an inclined position, with their extremities C, E, remote from each other, and sliding them contraryways from the middle towards each extremity of the bar AB, lifting them up when they come to the extremities, and replacing them on the middle of the bar, thus repeating the operation as often as required.

Horse-hoe bars, or those of a semicircular form, may be magnetized in a similar manner, except that the magnetic bars employed for the purpose must follow the curvature of the bar to be impregnated. The following is the method usually employed for magnetizing bars of this kind. The crooked bar is laid flat upon a table, and to each of its extremities is applied a straight magnetic bar, as DF, EG, fig. 17. and the remote extremities of these bars FG, are joined by the conductor or piece of soft iron FG. Then to its middle are to be applied two magnetic bars, with their opposite poles at a little distance from each other, H, I, and with these the crooked bar is to be stroked from end to end, following the direction of the crooked bar, so that on one side of it the magnetic bars may stand in the direction represented by the dotted lines at K and L. When in this manner the piece of steel ABC has been rubbed a sufficient number of times on the one side, it is to be turned, and the same operation repeated on the other side, taking care that the adhering magnetic bars, and the conductor of soft iron, be preserved in the same situation as at first. It must be observed that in this process the magnets DF, DG, as well as the magnets H, I, must be placed so that their south poles shall be towards that extremity of the bar which is to be made a north pole.

A material improvement in the process for communicating magnetism from artificial magnets to steel bars, was introduced by Duhamel. He formed a right-angled parallelogram, two of the sides of which were made by two equal bars of steel, that were intended to be magnetized, while the other two were formed by joining the extremities of the steel bars by two pieces of soft iron, also equal to each other in length, but much shorter than the steel bars. Then taking two parcels of bars already magnetized, he brought together their opposite poles towards the middle of one of the steel bars forming the parallelogram, and inclining the parcels as in fig. 18. he made them glide gently, separating them from each other towards the extremities of the bar; and this operation was repeated as often as required, when the inclined parcels of magnetic bars were carried to the opposite bars of the parallelogram, and this was rubbed in a similar manner. After the bars were rubbed sufficiently on one side, they were, as in former experiments, turned on the other.

This method is one of the best that we can employ for magnetizing the needles of compasses, and such steel bars as are of a moderate thinness, especially if we employ magnetic bars strongly impregnated for the purpose of rubbing the steel bars.

Much about the time that M. Duhamel contrived the above method, the same object was employing the attention of experimental philosophers in England, where the process of magnetizing bars was much improved by Mr Mitchell and Mr Canton.

Mr Mitchell employed two parcels of bars already strongly magnetized, joined together in a parallel direction, with their opposite poles united at each extremity, but in such a manner, that there remained between the two parcels a small interval. He then placed a number of equal steel bars in a straight line, and made one extremity of the magnetized bars slide over the line formed by the steel bars at right angles; and this he repeated as usual. In this way he found that the intermediate bars in the straight line acquired a great degree of magnetic power.

Mr Canton placed the bar which he wished to magnetize, so as to form part of a parallelogram, as in the method of M. Duhamel, and then employed the same means as Mr Mitchell for impregnating the bar, after which he separated the two parcels of magnets, and inclining them to each other in the manner of Duhamel upon the bar, he made them slide from the needle towards the extremities. This last method considerably augmented, according to Mr Canton, the magnetic power of the bar; but by Coulomb it is considered as the only effectual part of the process. These methods of Mitchell and Canton constitute what has been called the double touch, which was still farther improved by the celebrated Æpinus.

This philosopher, after having formed a parallelogram of steel bars, and pieces of soft iron, in the manner of Duhamel, placed upon the bar to be magnetized, two parcels of magnetic bars inclined in such a way that each of them formed on its own side an angle of 15° or 20° with the steel bar on which it was placed; their opposite poles being at a very small distance from one another. Keeping the parcels of magnetic bars in the same relative situation with respect to each other, he made both parcels slide along alternately from the middle of the bar towards each extremity, beginning at every renewal of the operation from the middle of the bar. This method has a very great advantage over the former, as by it we may magnetize bars of considerable length and thickness, by means of magnetic bars that have no great magnetic power.

In all these processes it must be remarked, that, in order to proceed properly, it is necessary to employ a considerable degree of pressure. A parallelogram of steel bars and soft iron should be kept firm by wedges, somewhat in the manner of printers types, and the extremities of the magnetic bars should be perfectly cleaned. Dr Robison supposed, that wetting these extremities considerably aided the process; but he found that the least particle of oil between the bars greatly obstructed it, as did the smallest piece of the thinnest gold leaf. He found that bars which were rough acquired a more powerful powerful magnetism than those which were moderately polished; but that, if moderately rough, they acquired the first degrees of magnetism more expeditiously than smooth bars, but did not receive so strong an impregnation as the latter.

The method of making artificial magnets has been greatly improved by M. Coulomb, who in a series of memoirs, printed in the Memoirs of the Academy of Sciences, and of the National Institute of Paris, has communicated a number of valuable observations and experiments, that have contributed, perhaps more than any preceding labours, to the advancement both of the theory and practice of magnetism. Many years ago he published his process for making very powerful artificial magnets.

In his operations he uses four very strong magnets previously impregnated. He placed his two strongest magnets, (as NS, NS, fig. 19.) on a horizontal plane in one right line, at such a distance that they might be a few lines nearer to each other than the length of the needle n s intended to be magnetized. He afterwards took the two magnets N' S', and inclining them as in the method of Æpinus, he placed them first on the middle of the needle, or with their poles nearly in contact. He then drew each magnet, without changing its inclination, to the extremity of the needle, and repeated this operation 5 or 6 times on each face of the needle. It is clear, that in this operation the poles of the needle n s remain fixed and invariable at the extremities of the needle, by means of the two strong magnets NS on which it rests. The effect produced by these can only be augmented by the action of the two superior magnets, which concur in magnetizing all the particles of the needle in the same direction.

He found likewise, that in this method of magnetizing there is a greater certainty of giving to both surfaces of needles intended to determine the magnetic meridian, an equal degree of magnetism; a circumstance deserving of the greatest attention in the construction of compasses, if the needle be suspended with its broadest surface parallel to the horizon.

After these previous preceles, he took 30 bars of steel hardened and tempered to the temper of a spring, five or six lines broad, two or three lines thick, and 36 inches long. The blades of fencing foils, such as are found in the shops, make pretty good magnets. English sheet steel cut into pieces one inch wide, hardened and lowered to spring temper, is preferable. When each compound magnet is to contain no more than 15 or 20 pounds of steel, it is sufficient to make the bars 30 or 36 inches long.

He magnetized each bar singly, according to the method already described. He then took two rectangular parallelopipeds of very soft iron, well polished, fix inches in length, between 20 and 24 lines broad, and 10 or 12 lines thick. With these two parallelopipeds, represented fig. 20, at N and S, he formed the armour of his magnet, by enveloping one extremity of each parallelopipedon with a stratum of his magnetic bars, so that the extremities of the parallelopipeds may project beyond the extremities of the bars 20 or 24 lines, and the other end may be enveloped by the ends of the set of bars. On this first layer of steel bars of three or four lines thick he places a second, three inches shorter than the first, so that the first projects beyond the second about 18 lines on each side. The whole is secured at the ends by two binding pieces of copper, which taints the bars close together, and prevent the armour from escaping.

Fig. 20. represents two artificial magnets composed Fig. 20. according to the method just described. N and S are the extremities of the two iron parallelopipeds. The two other extremities are inclosed by the bars. Each magnet thus compounded is solidly connected together by the copper pieces marked a, b, d', b'. The pieces of contact A, R, join the opposite poles of the magnets.

He found by experience, that with an apparatus of this form, each part weighing 15 or 20lbs. a force of 80 or 100lb. will be required to separate the pieces of contact; and that when an ordinary needle of the compass is placed on the two extremities of the compound bars, fig. 20. they become magnetized to saturation, without its being necessary to rub them with the upper pair. When magnets of greater force are desired, it is necessary, in proportion as the number of bars is increased, to augment their length also, and the dimensions of the parallelopipeds of iron which serve for the armour. It would be easy to ascertain the different dimensions which the magnets ought to have, in a manner sufficiently accurate for practice, from the laws of magnetism, and the position of the centre of action of the bars of steel of different lengths and thicknesses.

2. Iron or steel is rendered magnetical by being placed in a position corresponding to the magnetic meridian.

It has been often observed, that a bar of iron which has stood for a long time exactly or nearly in the magnetic direction, has acquired a degree of magnetic power, the extremities possessing opposite polarity. In this and other northern parts of Europe, old vanes of turrets, window bars, and even pokers that have stood long inclined in the chimney corner, are often very sensibly magnetic, their lower extremity becoming a north, and the upper a south pole. In the highest part of the steeple of St Giles's church in Edinburgh, on the north side, the upper bar of a hand-rail leading to a stair is very magnetical. It is worthy of remark, that those parts of such old bars which have become foliated and crumbly by exposure to the air are the most magnetic. This magnetic state of perpendicular iron bars was, as we are informed by Dr Gilbert, first observed in the vane spindle of the Augustine church at Mantua.

3. A bar of steel long hammered or exposed to violent By hammering or friction, while lying in the magnetic meridian, becomes magnetic.

This fact was well known to Dr Gilbert, who in a plate represents a blacksmith hammering a bar of steel in the magnetic position. Many smiths tools, such as long drills, that receive great pressure while in motion, broaches that are worked with a long lever, so as to act very fast, become very sensibly magnetical; the lower end, in the latitudes, being always a north pole. When a fleet punch is driven hard into a piece of iron, the punch has sometimes been rendered magnetical by a single blow. There is scarcely a cutting or boring tool in a smith's shop that does not possess some degree of magnetic power. Even soft steel and iron will acquire it by being violently twisted or exposed to great friction, and the magnetism thus acquired required is commonly permanent. From this circumstance it is difficult to procure for nice experiments pieces of iron that do not possess some degree of magnetism, and hence these experiments do not always succeed. It is therefore convenient to know how to deprive iron and steel of magnetism, and the method of doing this will appear from what will be said in the next section.

The steel balances of watches are often magnetic, sometimes even shewing evident polarity; a circumstance which is found to have some effect in disturbing the proper going of such watches or time-pieces. Hence it is recommended to make the balances of brass. See a paper on this subject by Mr Varley, in the first volume of the Philosophical Magazine.

4. Magnetism may be induced on substances that are susceptible of it, by heat.

Dr Gilbert remarks that such ores of iron as are in that particular metallic state, which he considers as most susceptible of magnetism, will acquire this power by being kept long in a red heat, while in a magnetic direction; and that their polarity corresponds to their position, that end of the mass which is opposite the north becoming a north pole. By many experiments made both by Dr Gilbert, and since his time by Dr Hooke, on iron and steel bars, it appears that these acquire permanent magnetism by being exposed to a strong heat, and suffered to cool gradually while lying in the magnetic direction; but that the magnetism thus acquired by steel rods is much stronger and more durable, if they are suddenly quenched with cold water, so as to give them a very hard temper. Dr Hooke found that the end of the bar next the north, or the lower end of a vertical bar, always became its permanent north pole, and the upper end, even when quenched, while the rest was suffered to cool gradually, became a very sensible south pole. If these operations were performed on bars placed in a position at right angles to the magnetic direction, no magnetism was acquired. Dr Gilbert makes a remarkable observation respecting the position of a magnetic needle brought near an ignited bar of iron, which was some years ago repeated in the Philosophical Transactions as a new discovery. "Bacillum ferreum, validè ignitum appone versorio excito: fiat veriturum, nec ad tale ferrum converitur: sed statim ut primum de condore aliquantulum remiserit, confluit illico." Thus it appears that iron is not susceptible of magnetism when red hot, but that it acquires magnetic power during its cooling. Dr Gilbert ascertained the degree of heat most favourable to the production of magnetism, but from his want of proper thermometers he did not succeed. Dr Robison found that though a bright red or a white heat does not make iron susceptible of magnetism while it is exposed to such a heat, it predisposes it for becoming magnetical. He found that when a bar of steel was made to acquire magnetism by being tempered in the magnetic direction, the acquired magnetism was much stronger when the bar was first made very hot, even though allowed to acquire its most magnetical state before being quenched, than if it had been heated only to this latter degree. Nay, he always found it stronger if quenched while red hot.

He also found that when he heated a small steel bar red hot, and quenched it while lying between two magnets, it acquired a much stronger magnetic power than it would acquire in any other way.

Mr Canton contrived the following method of producing magnetism in steel bars, without the assistance either of natural or artificial magnets.

Take twelve bars, fix of soft, and fix of hard steel, the former three inches long, one-fourth of an inch broad, and one-twentieth of an inch thick; with two pieces of iron, each half the length of one of the bars; but of the same breadth and thickness. The fix hard bars should be each five inches and a half long, one-half inch broad, and three twentieths of an inch thick, with two pieces of iron of half the length, but of the same breadth and thickness as one of the hard bars; and let all the bars be marked with a line quite round them at one end; then take an iron poker and tongs, or two bars of iron, the larger they are, and the longer they have been used, the better; and fixing the poker upright, or rather in the magnetical line between the knees, hold to it, near the top, one of the soft bars, having its marked end downwards, by a piece of sewing silk, which must be pulled tight by the left hand, that the bar may not slide; then grasping the tong with the right hand, a little below the middle, and holding them nearly in a vertical position, let the bar be stroked by the lower end from the bottom to the top about ten times on each side, which will give it a magnetic power sufficient to lift a small key at the marked end; which end, if the bar were suspended on a point, would turn towards the north, and is therefore called the north pole, and the unmarked end, for the same reason, is called the south pole. Four of the soft bars being impregnated after this manner, lay the other two parallel to each other, at a quarter of an inch distant, between the two pieces of iron belonging to them, a north and a south pole against each piece of iron; then take two of the bars already made magnetical, and place them together so as to make a double bar in thickness, the north pole of the one even with the south pole of the other, and the remaining two being put to these, one on each side, so as to have two north and two south poles together, separate the north from the south poles at one end by the interposition of some hard substance (1, fig. 21.), and place them perpendicularly with that end downward on the middle of one of the parallel bars AC, the two north poles towards its south end, and the two south poles towards its north end. Slide them three or four times backward and forward the whole length of the bar; then removing them from the middle of this bar, place them on the middle of the other bar BD as before directed, and go over that in the same manner; then turn both bars the other side upwards, and repeat the former operation: this being done, take the two bars from between the pieces of iron, and placing the two outermost of the touching bars in their stead, let the other two be the outermost of the four to touch these with; and this process being repeated till each pair of bars have been touched three or four times over, will give them a considerable magnetic power.

When the small bars have been thus rendered magnetic, in order to communicate the magnetism to the large bars, lay two of them on the table, between their iron conductors as before; then form a compound magnet with the fix small bars, placing three of them with the north poles downwards, and the three others with the south poles downwards. Place the two parcels at an angle, as was done with four of them, the north extremity of the one parcel being put contiguous to the south extremity of the other, and with this compound magnet stroke four of the large bars, one after another, about twenty times on each side, by which means they will acquire some magnetic power.

When the four large bars have been so far rendered magnetic, the small bars are laid aside, and the large ones are strengthened by themselves, in the manner followed with the small bars.

To expedite the operation, the bars ought to be fixed in a groove, or between brafs pins, otherwise the attraction and friction between the bars will be continually deranging them when placed between the conductors.

This whole process may be gone through in about half an hour, and each of the large bars, if well hardened, will lift about 28 ounces troy, and they are fitted for all the purposes of magnetism in navigation and experimental philosophy. The half dozen being put into a case in such a manner, as that no two poles of the same name may be together, and their irons with them as one bar, they will retain the virtue they have received; but if their power should, by making experiments, be ever so much impaired, it may be restored without any foreign assistance in a few minutes.

These bars must be kept in a wooden box, arranged in such a manner that their opposite poles may lie together, as represented at fig. 22.

There are various methods of communicating a permanent magnetism to ferruginous bodies, by means of a bar rendered magnetic, by position, of which the most simple is that described by Mr Marcel, whose experiments were made in the year 1726. Being employed in making some observations on the magnetic power which he found in great pieces of iron, he took a large vice weighing 90 pounds, in which he fixed a large anvil weighing 12 lbs. The steel to which he wished to give the magnetic power was laid upon the anvil in a north and south position, which happened to be the diagonal of the square surface of the latter. He then took a four-cornered piece of iron an inch thick every way, 33 inches long, weighing about 8 lbs, having one end rounded and brightly polished, the other being tapered. Holding then the steel fast upon the anvil with the one hand, he took the iron bar in the other, and holding it perpendicularly, he rubbed the steel hard with the rounded part towards him from north to south, always carrying the bar far enough round about to begin at the north. Having thus given 10 or 12 strokes, the steel was turned upside down, and rubbed as much on the other side. Proceeding in this manner till it had been rubbed 400 times, the steel was as strongly magnetic as if it had been touched by a powerful lodestone. The place where he began to rub was always the north pole. In these experiments it sometimes happened that the virtue was imparted by a few strokes; nay, by a single stroke a small needle was made to receive a very considerable power. Thus he imparted to two compass needles such a degree of magnetic power, that one lifted three fourths, and another a whole ounce of iron, and although these needles were anointed with linseed oil to keep them from rusting, and a hard coat was thus formed upon them, they nevertheless retained their power. Thus also a knife was made so strongly magnetic, that it would take up an ounce and three-fourths of iron. Four small pieces of steel, each an inch long, and one-twelfth of an inch broad, as thin as the spring of a watch, were thus impregnated with the magnetic power, and then joined into a small artificial magnet; which at its first formation took up eight times its own weight of iron; and after being six years kept in the most careless manner, was found to have gained rather than lost any thing of its power. In the course of his experiments, Mr Marcel found, that the end at which he began to rub was always the north pole, whatever position the steel was laid in. On rubbing a piece of steel from one end to the middle, and then from the other end to the middle, it acquired two north poles, one at each end, the middle being a south pole. Beginning to rub from the middle towards each end, he found a north pole in the middle, and a south pole at each extremity.

Magnetism may be communicated to a small piece of soft steel in the following manner: take two iron bars of about an inch square, and upwards of three feet in length; keep them in the magnetic line, or in a perpendicular posture, as represented fig. 23. Let the piece of steel CB be either fastened to the edge of a table, or held by an affiant; and placing the lower extremity of the bar AB, and the upper extremity of the bar CD, on opposite sides, and in the middle of the steel, stroke the latter from the middle towards its extremities, moving both bars at the same time. When both are arrived at the extremities of the steel, remove them from it, and apply them again to the middle. Do so for 40 or 50 times, and the steel will be found to have a considerable degree of magnetic power. Care, however, must be taken, in removing the bars, not to draw them along the surface of the steel, or the experiment will not succeed, because the magnetism is destroyed by the contrary strokes.

The late Dr Gowin Knight possessed a surprising skill in magnetism, being able to communicate an extraordinary degree of attractive or repulsive power, and to alter or reverse the poles at pleasure; but as he refused to discover his methods upon any terms whatever (even as he said, though he should receive in return as many guineas as he could carry), these curious and valuable secrets have died with him. In the 69th volume of the Philosophical Transactions, however, Mr Benjamin Wilfon has given a process, which at least discovers one of the leading principles of Dr Knight's art, and may perhaps be a means of discovering the whole to those who shall be less reserved. The doctor's process, according to Mr Wilfon, was as follows. Having provided himself with a great quantity of clean iron filings, he put them into a large tub, that was more than one-third filled with clean water; he then, with great labour, worked the tub to and fro for many hours together, that the friction between the grains of iron by this treatment might break off such smaller parts as would remain suspended in the water for a time. The obtaining of these very small particles in sufficient quantity seemed to him to be one of the principal desiderata in the experiment. The water being by this treatment rendered very muddy, he poured the same into a clean iron vessel, leaving the filings behind;

Experiments—hind; and when the water had flood long enough to be clear, he poured it out carefully, without disturbing such of the sediment as still remained; which now appeared reduced almost to an impalpable powder. This powder was afterwards removed into another vessel in order to dry it; but as he had not obtained a proper quantity thereof by this one step, he was obliged to repeat the process many times. Having at last procured enough of this very fine powder, the next thing was to make a paste of it, and that with some vehicle which would contain a considerable quantity of the inflammable matter; for this purpose he had recourse to linseed oil in preference to all other fluids. With these two ingredients only he made a stiff paste, and took particular care to knead it well before he moulded it into convenient shapes. Sometimes, while the paste continued in its soft state, he would put the impression of a seal upon the several pieces; one of which is in the British Museum. This paste was then put upon wood, and sometimes on tiles, in order to bake or dry it before a moderate fire, at about the distance of a foot. He found that a moderate fire was most proper, because a greater degree of heat made the composition frequently crack in many places. The time required for the baking or drying of this paste was generally about five or six hours before it attained a sufficient degree of hardness. When that was done, and the several baked pieces were become cold, he gave them their magnetic power in any direction he pleased, by placing them between the extreme ends of his large magazine of artificial magnets for a few seconds or more as he saw occasion. By this method the power they acquired was such, that when any of these pieces were held between two of his best ten guinea bars, with its poles purposely inverted, it immediately of itself turned about to recover its natural direction, which the force of those very powerful bars was not sufficient to counteract.

In the 66th volume of the Philosophical Transactions we have the following account from Dr Fothergill, of Dr Knight's method of imitating natural magnets, but which is by Mr Cavallo supposed to be some mistake or misinformation. "I do not know," says he, "that ever the doctor (Dr Knight) left behind him any description of a composition he had made to form artificial loadstones. I have seen in his possession, and many other of his friends have likewise seen, such a composition, which retained the magnetic virtue in a manner much more fixed than either any real loadstone, or any magnetic bar, however well tempered. In the natural ones he could change the poles in an instant, so likewise in the hardest bars, but in the composition the poles were immovable. He had several small pieces of this composition which had strong magnetic powers. The largest was about half an inch in breadth, very little longer than broad, and near one-fourth of an inch thick. It was not armed, but the ends were powerfully magnetic; nor could the poles be altered, though it was placed between two of his largest bars, and they were very strongly impregnated. The mass was not very heavy, and had much the appearance of a piece of black lead, though not quite so shining. I believe he never divulged the composition, but I think he once told me, the basis of it was filings of iron reduced by long-continued attrition to a perfectly impalpable flake, and then incorporated with some pliant matter to give it due consistence.

From these accounts it appears that the basis of Dr Knight's artificial loadstones was the black powder to which iron filings are reduced by being shaken with water, or the black oxide of iron, formerly called martial æthiops. Hence Mr Cavallo supposes that the following receipt for imitating the natural magnets will answer the purpose.

Take some martial æthiops, reduced into a very fine powder, or, which is more easily procured, black oxide of iron, the scales which fall from red-hot iron when hammered, and are found abundantly in smiths shops. Mix this powder with drying linseed oil, so as to form it into a very stiff paste, and shape it in a mould so as to give it any form you require, whether of a terrella, a human head, or any other. This done, put it into a warm place for some weeks, and it will dry so as to become very hard; then render it magnetic by the application of powerful magnets, and it will acquire a considerable power.

SECT. IV. Of the Circumstances which tend to impair or destroy the Magnetic Power.

The magnetic power in all its modifications, whether Magnetism of attraction, repulsion, or polarity, is in general temporary or perishing. The best magnets, whether natural or artificial, unless carefully preserved, with attention to certain circumstances that will presently appear, are observed to have their magnetic power diminished. Natural magnets, and artificial magnets made of steel tempered as hard as possible, retain their power most obstinately, and seldom entirely lose it except under circumstances which we know to be unfavourable to its durability. Magnets of steel of a spring temper, are much sooner weakened, lose more of their force merely by keeping, and finally retain little or none of it. Soft steel and iron seldom retain magnetic power when removed from the magnet where they acquired it, unless their metallic state undergoes some change.

The following circumstances have been observed to be most powerful in diminishing or destroying the power of magnets.

1. Improper position. Nothing has so much effect in impairing the power of a magnet as keeping it in an improper position, that is, too far from the magnetic line. If the axis of the magnet be placed in a direction that is at right angles with the magnetic meridian, that is, in this latitude nearly E. N. E. and W. N. W. it will soonest lose its magnetic power; and if it be placed in the magnetic line, but in a contrary position, or with the north pole where the south pole should be, if permitted to vibrate freely, it will gradually become weaker every day, and unless it be a natural magnet, or an artificial one made of very hard tempered steel, it will, in no very long time, entirely lose its magnetic power.

2. Heat. The dissipation of magnetic power is greatly promoted by heating the magnet. The heat of boiling water has a sensible effect in this way; but if the magnet be exposed to a red heat, its power is entirely destroyed, as has been long known. Dr Gilbert observed that the power of magnets was destroyed by a heat that was not sufficient to make the metal visible in the dark; and Mr Canton found that the heat of boiling water weakened the power of a magnet, but that the greatest part of this was recovered as the magnet cooled. If the heat be applied when the magnet lies in a position most favourable to the diffipation of magnetism, the power is soonest destroyed; hence, the best way to deprive iron or steel of accidental magnetism is, to heat it red hot, and allow it to cool while lying in a direction perpendicular to the magnetic line.

M. Coulomb has ascertained that at 200 degrees of heat, two-fifths of the magnetism of a magnet is diffipated, and that at 500 degrees the whole is lost.

3. By violent treatment. It is very extraordinary that the power of a magnet is impaired by rough usage. Dr Gilbert observed that a magnet which he had powerfully impregnated was greatly weakened by a single fall on the floor; and since his time it has been observed that when a magnet falls on a stone, or receives any concussion that makes it ring, it is injured much more than by being beaten with any thing soft and yielding. When a natural magnet is ground with coarse powders, in order to bring it to any required form, it is considerably weakened. This shows the propriety of altering the natural form of loadstones as little as possible, and when this is necessary, of doing it as expeditiously as may be, by cutting them briskly in the thin disks of a lapidary's wheel.

4. Placing them near each other with their similar poles being opposite. Magnets situated in this way always weaken each other, and when a powerful magnet is placed near a weaker, with their similar poles opposed, the polarity of the weaker is frequently reversed, that is, if the pole were north it becomes south, and vice versa. When the weaker magnet is a natural loadstone, or has been made of hard tempered steel, its original polarity is restored when the improper position is changed; but if it has been made of spring-tempered steel, the alteration is generally permanent, and often as complete as while the magnets were in the neighbourhood of each other.

CHAP. III. Theory of Magnetism.

RESPECTING the notions which the ancient philosophers entertained about the cause of magnetic phenomena, we know very little. One curious opinion which they entertained of the reason why a magnet was improved by the contact of iron, is worth noticing. They conceived that the magnet fed upon the iron, and hence acquired additional attractive power; and when deprived of this pabulum, it grew weak and languid.

—— "Nam ferro nurunt vitam, ferrique vigore Vescitur; hoc dulces epulas, hoc pabula novit; Hinc proprias renovat vires, hinc sula per artus Afera secretem fervant alimenta vigorem. Hoc absente perit, tristi morientia torpent Membra fame, venaque fitis consumit apertas."

CLAUDIAN.

In the 16th century, the philosophers of modern times first began to speculate about the cause of magnetic polarity, a phenomenon which then became interesting on account of the difference of declination observed by navigators. Various trifling opinions were published on the subject. Some said that the needle was directed by a certain point in the heavens, which was little more than saying that it pointed one way. Others ascribed the direction of the needle to vast magnetic rocks situated in the earth; but as to the exact situation of these rocks, they did not give themselves the trouble to inquire, till Fracasteri observed, that, if those rocks are supposed to be situated in any part of the globe yet visited by navigators, and if, as we must suppose, they act like loadstones, they will cause the direction to be very different from what is observed. He therefore placed them somewhere in the inaccessible polar regions, though not immediately at the poles. Norman, who, as we have seen (DIPPING Needle), discovered the dip of the magnetic needle, and observed that in every part of Europe, the north pole pointed very far below the horizon, was naturally led to ascribe this effect to the influence of the earth, though he does not express himself as if he thought that the needle was attracted by any point within the earth, but only that it was always directed to such a point.

From comparing the different positions of the compass needle, as described by Norman, with the positions which he had himself observed small needles to assume in relation to a magnet, Dr Gilbert was naturally led to consider the earth as a great loadstone, or else containing a great loadstone within it, which arranged the dipping needle, or the needle of the compass, in the same manner as he observed a small needle poised on its pivot, to be arranged by a large magnet. Dr Gilbert has explained his theory at large in his Physiologia Nova de Magnete, et de Tellure Magno Magneti. It may be briefly expressed in the following terms. All the appearances of natural magnetism are similar to what would be observed in the earth, were a large magnet with its poles situated near the poles of the equator, viz. the north pole not far from Baffin's bay in North America, and the south pole in about the opposite part of the globe. If a dipping needle were exposed to the influence of such a large magnet, it must arrange itself in a plane passing through the magnetic poles, a position indicated very nearly by the mariners needle; and the more we recede from the equator of the great magnet, the more must the dipping needle be inclined to the horizon.

Dr Gilbert's theory was equally ingenious and important, and affords, if firmly established, a complete explanation of all the phenomena of magnetism. At the time it was first published, however, observations were neither sufficiently numerous, nor sufficiently accurate, to enable the author to assign the real position of the great magnet, nor to ascertain its laws of action. The theory was chiefly founded on observations made by the dipping needle, and though those instruments made by Norman were more accurate than might have been expected at so early a period of the science, the observations made with them cannot, from many circumstances, be implicitly relied on. We are still in want of a numerous collection of observations on the dip, in order to perfect our knowledge of the magnetic poles. We can only say that the earth acts on the compass needle in the same manner as a large magnet would act; but the appearances do not seem to resemble the effects of what we should consider as a good loadstone having two vigorous gorous poles, but rather such as would result from the action of a very irregular loadstone with its poles very much diffused.

It is unfortunate that our most numerous observations of the dip have not been made in those places where they would be the most instructive. Dr Robison was of opinion that a series of observations should be obtained, extending from New Zealand northward, across the Pacific ocean to Cape Fairweather on the western coast of North America, whence it should be continued through that part of the continent. A second series might extend from the Cape of Good Hope along the western coast of Africa to the tropic of Capricorn; thence across the interior of the African continent through Sicily, Italy, Dalmatia, the eastern part of Germany, the gulf of Bothnia, Lapland, and the western part of Greenland. This series would be nearly in a plane passing through the probable situations of the poles. A third series might extend at right angles to the last, so as to form a small circle crossing the former, passing near Japan, through the island of Borneo, and the western part of New Holland; near Mexico, and a few degrees west of Easter island. Here and at Borneo there would be a considerable inclination of the magnetic plane to the horizon, though this cannot be found out. There are, however, other points of this circle in which the dip is considerable, where the inclination may be discovered. In short, all circumstances seem to indicate a multiplicity of poles, or, what renders calculation most difficult, an irregular magnetism in which the polarity is very much diffused.

Philosophers are very much divided respecting the situation of the magnetic poles of the earth. We shall here state only a few of their opinions, referring a fuller account of some of them for the article Variation of the Compass.

Dr Halley thought that the north magnetic pole was near Baffin's bay in North America.

Professor Kraft (see Peterburgh Comment. vol.xvii.) places the north pole in N. Lat. 70° and W. Long. 23° from London; and the south pole in S. Lat. 50° and E. Long. 92°.

Wilcke of Stockholm places the north pole in N. Lat. 75° near Baffin's bay, and in the longitude of California, while he fixes the south pole in S. Lat 70° in the Pacific ocean.

Churchman supposes the north pole to be in N. Lat. 59°, and W. Long. 135°, a little inland from Cape Fairweather; and the south in S. Lat. 59°, and E. Long. 165°, directly south of New Zealand. (See Variation).

Euler (Memoirs of the Acad. of Berlin, vol. xvi.) places the north pole in N. Lat. 75°. Lemouvier (Lois du Magnetisme) in N. Lat. 73°. Buffon in N. Lat. 71°.

La Lande places it in N. Lat. 77° 4', and in about W. Long. 98° from Paris. (See Connoissance des Temps, an. xii.).

However ingenious this hypothesis of Dr Gilbert was, it appears to have been nothing more than a sagacious conjecture. The hypothesis, however, is confirmed into a rational theory by many observations and experiments which were unknown or unthought of in Dr Gilbert's time.

Mr Hindshaw's beautiful experiment on the effect of an upright iron bar on the opposite ends of a compass-needle, according as one end or the other of the bar is next the earth (see Variation of the Compass) is an abundant proof of the justness of this theory.

We can imitate that experiment in a very satisfactory manner by artificial magnetism; thus forming a just comparison between the action of the earth and that of a magnet.

Let a large bar magnet, as SAN (fig. 24.) be supported so as to have its ends detached from surrounding bodies. Then place a small needle nicely poised, as B, about three inches below N, the north pole of the magnet, and so that its directive power for the magnet may be very weak. Now take a small piece of soft iron, and hold it in such a position as is represented at C; its lower end becoming a north pole, will attract the south pole of the needle. Now, while the needle is kept in the same position, turn round the piece of iron into the position D; the south pole of the needle will be seen to avoid it, and the north pole will be attracted. Here the magnet may be compared to the earth, and the small piece of iron to the iron bar in Mr Hindshaw's experiment.

Again, it has been seen that magnetism may be produced in iron or steel by hammering or heating them while in a determinate position with respect to the earth. The same effect will be produced by the same processes while the iron or steel is in the neighbourhood of a powerful magnet.

Lastly, the circumstance of the magnetic inclination of the north pole of the dipping needle being diminished, and the horizontality of the compass needle destroyed, as we ascend above the earth, is an additional and certain evidence of the truth of this theory.

In short, we may consider it as demonstrated, that the earth is a great magnet, or contains a great magnet, by the influence of which the direction of the needle and all the magnetic power acquired by iron, when placed in a proper position, are produced.

A further illustration and application of this theory will be given presently, when we have considered some other hypotheses posterior to that of Dr Gilbert.

It was very early an object with philosophers to assign the immediate cause of magnetic attraction and repulsion, and of that faculty of mutual impregnation which so remarkably distinguishes iron from all other substances. In particular, the curious arrangement of iron filings strewed round a magnet forcibly attracted their attention. It is scarcely possible to observe this arrangement without conceiving the idea of a stream of matter issuing from one of the poles of the magnet, moving round it, entering by the other pole, and again issuing by its former outlet. Accordingly, such an idea was entertained in the earliest times; but very different notions prevailed as to the manner in which such a stream produced the effects observed. One of the simplest methods was, to conceive it acting by impulsion, like any other stream of fluid matter. This idea was entertained by Lucretius, who supposed the surrounding air to be swept out of the way by the impulsion of the fluid, which thus rushing round the magnet carried the iron filings towards it.

In the last century Euler framed an hypothesis of Euler's hy-magnetism on this theory of impulsion. He supposes, pothesia, that the two principal causes which concur in producing the the wonderful properties of a magnet, are, First, A particular structure of the internal pores of the magnets, and of magnetical bodies; and, Secondly, An external agent or fluid, which acts upon, and passes through these pores. This fluid he supposes to be the solar atmosphere, or that subtle matter called ether, which fills our system.

Indeed, most writers on this subject agreed in supposing that there are corpuscles of a peculiar form and energy, which continually circulate around and through a magnet; and that a vortex of the same kind circulates around and through the earth.

"A magnet, besides the pores which it has in common with other bodies, has also other pores considerably smaller, destined only for the passage of the magnetic fluid. These pores are so disposed as to communicate one with the other, forming tubes or channels, by which the magnetic fluid passes from one end to the other. The pores are so formed, that this fluid can only pass through them in one direction, but cannot return back the same way; similar to the veins and lymphatic vessels of the animal body, which are furnished with valves for this purpose: So that the pores of the magnet may be conceived to be formed into several narrow contiguous tubes, parallel to each other, as at A, B, fig. 25, through which the finer part of the ether passes freely from A to B, but cannot return back on account of the resistance it meets with at a, a, b, b, nor overcome the resistance of the grosser ether, which occasions and continues the motion." For supposing the pole A of a magnet, filled with several mouths or open ends of similar tubes, the magnetic fluid, pressed by the grosser part of the ether, will pass towards B with an inconceivable rapidity, which is proportionable to the elasticity of the ether itself; this matter which, till it arrives at B, is separated from the tubes by the grosser parts, then meets with it again, and has its velocity retarded, and its direction changed; the stream, reflected by the ether, with which it cannot immediately mix, is bent on both sides towards C and D, and describes, but with less velocity, the curves DE and CF e, and approaching by the curves d and c, falls in with the effluent matter m m, and again enters the magnet; and thus forms that remarkable atmosphere, which is visible in the arrangement of steel filings on a piece of paper that is placed over a magnet*.

We have already had occasion (see the article IMPULSION) to make some observations on the general doctrine of impulsion, and these need not be here repeated. Respecting the explanations afforded by the canals and door-gates in Euler's hypothesis, opening in one direction and shutting in the other, we may observe, that as these constructions are altered in a moment in a bar of soft iron, merely by changing the position of the magnet, it is astonishing that they should ever have been conceived by so acute a philosopher. Even supposing such circumstances to take place, the effects resulting from them should be the reverse of what are actually observed, as the impelling stream should move those bodies least which afford the readiest channels for its passage. If the iron filings were arranged by this impelling stream, they should be carried along with it, and if they are carried towards one pole of the magnet, they should be driven away from the other.

Æpinus, of the academy of Petersburgh, whose theory of electricity we have explained and illustrated at considerable length, was led by the analogy observed between the phenomena of electricity and those of magnetism, and in particular from the resemblance between the attractions and repulsions of the tourmalin and those of a magnet, to conceive the idea that both classes of phenomena might be explained in a similar manner, or that the phenomena of magnetism, like those of electricity, were to be attributed to the motions of a certain fluid existing in all bodies susceptible of magnetism. This conjecture was confirmed by observing, that when magnetism was induced on a piece of iron by its proximity to a magnet, the power of the magnet is sensibly diminished. The following is an abstract of Mr. Æpinus's hypothesis.

1. There exists in all magnetic bodies a substance which may be called the magnetic fluid, the particles of which repel each other with a force that decreases as the distance increases.

2. There is a mutual attraction, varying according to the same law, between the particles of the magnetic fluid, and the particles of iron.

3. There is a mutual repulsion among the particles of iron, following the same law.

4. The magnetic fluid is capable of moving through the pores of iron, and soft steel, without any considerable difficulty: but its motion is more and more obstructed as the steel receives a harder degree of temper; and in steel of the hardest temper, and the ores of iron, it moves with the greatest difficulty.

5. From the supposed attraction between the magnetic fluid and iron, the latter may contain a certain determinate quantity of the former, and this quantity will be such that the accumulating attraction of a particle of it for the whole of the iron, balances the repulsion between the particles of the whole fluid contained in the iron; supposing the quantity of fluid competent to a particle of iron to be such, that the repulsion between it and the fluid competent to another particle of iron, is also equal to its attraction for that particle of iron. Therefore the attraction between the fluid in one iron bar A, and the iron of another bar B, is just equal to the repulsion between the iron in A and the iron in B. This determinate quantity of fluid in the iron is called its natural quantity.

6. From the mobility of the fluid through the pores of iron, it may, by the agency of a proper external force, be abstracted from one end of an iron bar, and condensed in the other end. This, however, is a violent state, and the mutual repulsion between the particles of condensed fluid, together with the attraction between the fluid and that part of the iron which it has quitted, tend to produce a more uniform distribution. It is evident that something of this tendency must take place in every state of condensation and rarefaction, and that a perfect equilibrium can be produced only when the fluid is diffused with perfect uniformity. This state of uniformity may be called the natural state of the body.

7. The production of such a uniform distribution will depend on the nature of the resistance to the motion of the fluid, opposed by the iron in its various states. If this resistance arises merely from the communication of motion, like that which perfect fluids oppose to the mo- tion of solid bodies, such resistance may be overcome by the weakest tendency to uniform diffusion; but if, as seems most likely, the obstruction is like that of a clammy fluid, or of a soft plastic body like clay, after the accumulation arising from the action of an external force, it may remain after that force is removed; and the diffusion will cease when there is a perfect equilibrium between the obstruction and the diffusing force.

As the illustration of this theory in general cases is precisely similar, mutatis mutandis, with that of electricity, so fully detailed under the article ELECTRICITY, from No 299. to 348, we need not repeat it here, but may refer the reader to that treatise, requesting him to consider the illustration as relating to the magnetic fluid.

It is proper, however, to remark here, that the phenomena of magnetism are limited by this circumstance; that magnets always contain their natural quantity of fluid. Of course, their action on iron, and on each other, depends entirely on its unequal distribution.

The most important part of this theory is that which explains the induction of magnetism on iron and steel by juxtaposition to a magnet; but before we can properly enter on that, we must notice some other particulars respecting the theoretical part of our subject.

A very material point in magnetism, as in electricity, is to ascertain the law of action, according to which this power acts on the particles of iron and other matter; and accordingly this has long been an object of attention with philosophers. The difficulty of ascertaining this law is extremely great, as will readily appear by the following consideration.

In the action of two magnets on each other, as A and B, there are four different actions to be considered that act at the same time, though with different degrees of force, and in different directions. Thus the north pole of A repels the north pole of B, and attracts its south pole, while the south pole of A exerts a repulsion on the south pole of B, and an attraction on its north pole. Now the force, which we attempt to measure, is compounded of these four forces; and these we cannot measure separately. The attraction observed is the excess of two attractions that are unequal above two unequal repulsions, and &c. with respect to the observed repulsion. Further, if we reflect that it is possible for a mutual action to exist between every two particles of the different magnets, and that the intensity of this action may vary, not only at different distances, but at the same distance, the difficulty will be greatly increased.

Numerous experiments have been made with a view of ascertaining this law. Mr Cavallo has detailed many of those made by Muhlenbroeck; but their results are so anomalous, that their inaccuracy is apparent. Indeed, the attempt to ascertain this law by observing merely the attractions and repulsions, was very unphilosophical. The method employed by Mr Hawkebee and Dr Brook Taylor, viz. observing how far the action of a magnet made a compass needle deviate from the meridian at different distances, was much more scientific, as this deviation is occasioned by the difference of the two sums of the same forces; and this may be made many times greater than the other, and must of course be much more sensible. The shape of the magnets employed by them was, however, very improper. Some experiments made by Mr Lambert of the academy of Berlin, were very judicious. He placed a magnetic needle at various distances from a magnet, but in the direction of its axis, and marked the declination from the magnetic line produced by the action of the magnet, and the obliquity of the magnet to the axis of the needle. Thus the action of the magnet and the natural polarity of the needle were placed in opposition and equilibrium; but the great difficulty was to discover the proportional change of these forces by their obliquity of action on this small lever.

Mr Lambert observed, that when the obliquity of the magnet to the axis of the needle was =30°, the needle was made to decline 15°; and when the obliquity was =75°, the needle declined 30°. Let us call the obliquity o and the declination d, and let us put f for that function of the angle which is proportional to the action. Also let us call the natural polarity of the needle p, and the force of the magnet m. Then it is evident that \( p \times f : 15 = m \times f : 30 \); and \( p : m = f_3 : 30 : f_1 : 15 \); and for the same reason \( p : m = f_5 : f_3 ; 30 \), and therefore \( f_1 : 15 : f_3 : 30 = f_5 : 75 \). But since \( 15 : 30 = \) sine \( 30 : \) sine \( 75 \); hence Mr Lambert concluded, that the sine was that function of the angle which was proportional to the action of magnetism on a lever. As this point, however, could not be determined by one experiment, he compared several other obliquities and declinations with the same distances, and with different distances of the magnet, and fully proved that he was right in his conjecture.

The result of Mr Lambert's experiments fully proves the fallacy of the theories of impulsion, which pretend to explain magnetic action by the impelling power of a stream of fluid, or by pressure produced by the motion of such a stream; as in such a case the pressure on the needle must have diminished in the duplicate ratio of the fine; or with the angle 90° the directive power must have been four times as much as with the angle of 30°, whereas it is shewn by observation to be only twice as much.

When Mr Lambert had ascertained the effect of obliquity, he proceeded to examine that of distance; and he found, that if we put f for the force of the magnet, and d for the distance of the nearest pole of the magnet from the centre of the needle, and a for a constant quantity nearly equal to two-thirds of the length of the needle, f will be proportional to \((\frac{d}{a})^2\).

Dr Robison endeavoured to investigate this law in a Dr Robison's very simple manner. He caused to be made some magnets consisting of two balls connected by a slender rod, By a particular mode of impregnation (which we suppose to be quenching them, after being red hot, between two magnets) he gave them a pretty good magnetism; and the force of each pole appeared to be nearly confined to the centre of the ball, which was his object in making them of such a shape, as it reduced the examination of their attractive and directive power to a very easy computation. The result of his experiments was, that the force of each pole varied inversely as the squares of the distances, or at least the error arising from such an hypothesis was very small, amounting only to one-fifteenth of the whole.

Dr Robison made a near approximation to the law of action, by supposing that the function of the distance expressing that law, represented by the ordinates of a curve similar to the hyperbola, referred to its asymptote as an axis, towards which its curve was of course always convex. On this supposition he explained the attractions and repulsions of magnets nearly in the following manner:

Let there be two magnets, A and B (fig. 26.) placed so that their four poles, S, N, s, n, may be in a straight line. Now, on the straight line O q take O m, O p, O n, O q = N s, N n, S s, S n; and let MPNQ be a curve line, whose asymptotic axis is the said line O q. Draw the ordinates m M, p P, N n, q Q to the curve, and these will represent the intensities of the forces exerted between the poles of the magnets. The distance between m, n, or between p and q—= the length of the magnet A, and m p or n q—= that of B, and M m, P p, N n, and Q q, are pairs of ordinates that are equally distant. Now, it is easy to see from the figure, that in whatever situation the pairs of equidistant ordinates may be, M m + Q q will always exceed P p + N n, or the sum of the attractions will be always greater than that of the repulsions.

Let the chords MQ, PN, MP, NQ be drawn. Bisect them in B, D, E, F, and join EF. Draw the ordinates E e, F f, and BD b (cutting EF in C). Draw P u parallel to the axis, cutting E e in t. Draw also Q i parallel to the axis, cutting F f in φ. Also draw F H L parallel to the axis, and P o t parallel to Q N; and draw PL l, and P e x, cutting M m in l and x. Let each ordinate be represented by the letter at its intersection with the axis. Thus, the ordinates M m and Q q may be represented by m and q, &c. Because MP is bisected in E, M t is double of E e, M l is double E L, and M x double of E e. Again, P t being parallel to Q n, and P u to Q i, t u equals N i.

If these ordinates are supposed to represent the mutual action of the magnetic poles, their tendency to or from each other, that is, their attractions or repulsions, may be expressed by (m + q) — (n + p) which represent the excess of the sum of the actions of the nearest and most remote poles above the sum of the action of the intermediate distant poles. This tendency may often be conveniently represented by (n — p) — (n — q) or the excess of the difference of the actions exerted by the nearest pole of A on the two poles of B, above the difference of the actions of the remote pole of A on the same poles of B. Now, 1. If we suppose the dissimilar poles of A and B to front each other, m + q will represent attractions, and p + m repulsions; but m + q is greater than p + n, therefore A and B will attract each other. Again (m + q) — (p + n) equals M t, = 2 E e = 2 BD = 4 CD.

The above action will be increased by any one of four circumstances, as, 1. By increasing the strength of either magnet. 2. By lessening the distance between the two magnets. 3. Increasing the length of A, the distance between it and B remaining the same. 4. By increasing the length of B, the distance between it and A remaining the same.

2dly, Let us place the magnets, so that their similar poles front each other. Here it is evident that the ordinates which in the former case represented attractions, will now represent repulsions, and that the repelling forces of the magnets are equal to the former attracting forces at the same distances. As magnets are seldom perfect, the repelling forces are, however, usually weaker than the attracting.

To explain the directive power of magnets, Dr Robison supposed the magnet A not to be at liberty to approach B or recede from it, but to be supported at its centre B, so as to turn round it. Now, its south pole s being more attracted by N than it is repelled by S, B is on the whole attracted by A, and by this attraction would vibrate like a pendulum supported at the centre B. Again, the north pole n being repelled by N more than it is attracted by S, will be on the whole repelled, and B n would also vibrate round B. Thus B would be kept in the position s B n. This will be more evident if we suppose the magnet B arranged at right angles to the line AB, as in the dotted representation s' B' n'; for now s' and n' are urged in opposite conspiring directions with equal forces, which, if the magnet be very small, will act nearly at right angles to n' s'. If the position were oblique, the forces would be somewhat unequal; and allowances must be made for the obliquity of the action, that we may know the precise rotative momentum. This modification of the action of A on B, we call the directive power of A; and the modification of B, by which it tends to or from A, we call the polarity of B.

Now, the directive power of A and the polarity of B may be increased, 1. By increasing the strength of either A or B, or both; 2. By diminishing the distance between A and B; 3. By increasing the length of A; and, 4. By diminishing the length of B, the distance between them remaining the same.

We may remark, that the directive power of A is always greater than its attractive power, by a certain measure which we may represent by the formula 2 (p — q) which is thus derived. The difference between them may be expressed by t l = 2 o L; but t e = P p — p, and t L = P p — F f = P p — Q q — F q — P p — Q q — o ; therefore o L = P p — Q q, and t l = 2 (P p — Q q) = 2 (p — q).

This picture of the forces, attentively examined, will suggest to the reader many interesting and instructive particulars. Dr Robison used to relate a curious and instructive phenomenon that he was long puzzled to explain, respecting the mutual action of large magnets. Amusing himself with some experiments on magnetism, with two large strong magnets, as A.B. fig. 27. which were placed at about the distance of three inches with their opposite poles fronting each other, he had placed a small needle balanced on a point between them as at D, which arranged itself in the same line with the magnets; but happening to set it off to a considerable distance on the table, as at F, he was surprised to see it instantly turn round on the point, and arrange itself in an opposite direction. When brought back to D, it reaffirmed its former position, but when he carried it out gradually along the line DF, perpendicular to N s, he found it grow sensibly more feeble, vibrating more slowly; and when arrived at a certain point E, it shewed no polarity towards either A or B, but retained any position given it: but when carried farther out, it again acquired polarity to the magnets, though in a contrary direction, arranging itself parallel to NS, with its north pole next to N, and south pole next to S. Being interrupted interrupted in the prosecution of this experiment, but having marked the line DF on the table, he afterwards replaced the magnets and needle, placing the latter at E, where he expected it to be neutral; but it now turned its north pole towards B, and did not become neutral till carried further out. When standing there, something happened to move the magnets A and B, which instantly rushed together, and at the same instant the needle turned itself briskly, and arranged itself as before at F. In short, by gradually withdrawing the magnets from each other, he found that the needle first became weaker, then neutral, and then turned into the opposite position.

Dr Robison explained this curious phenomenon by what he calls primary and secondary magnetic curves, such as NHM, NEL, and SGK, SIE; but our limits do not permit us to enter here on the investigation of these curves.

From all Dr Robison's experiments and calculations, he appears to have been fully convinced, that the true law of magnetic action is in the inverse duplicate ratio of the distances, and his opinion is still farther strengthened by the ingenious experiments of M. Coulomb related in the Memoirs of the Academy of Sciences at Paris for 1786 and 1787, or the Jour. de Phy. vol. xliii.

We are now prepared to examine the induction of magnetism in iron or steel by juxtaposition to a magnet, the general facts of which are mentioned and illustrated in No 44.

It was remarked in No 46, that the induction of magnetism in the iron by being near a magnet was not produced by a transference of something from the magnet to the iron. It follows that there must be some inherent property in iron, which is only excited, as it were, or routed into action, by the proximity of the magnet.

It has been remarked, that the magnetism of iron is momentary; but this must be understood only of the finest and purest iron, as when this metal is in the state of ore, or has undergone any change, as by exposure to the air, or by cementation, its magnetism becomes permanent, in proportion to the hardness of the metal.

It is of great importance to observe that the acquisition of induced magnetism is gradual and progressive, and that this gradation is more perceptible according as the iron is in a harder state. In soft iron the induction appears to be instantaneous throughout, unless the bar be exceedingly long; but when a magnet is brought near a bar of tempered steel, the near end acquires a contrary polarity long before the remote end appears affected, and it is a long time before the remote end acquires the same polarity with the proximate end of the magnet.

From what has been said we may infer, that a piece of iron brought near a magnet, is attracted only because it becomes magnetical by induction, and that the attraction of a loadstone for iron, or the tendency of iron to the loadstone, is the consequence of the proper disposition of the magnetism induced in the iron. It has already appeared, that this phenomenon arises from the excess of two attractions above two repulsions, and this is farther proved by the following considerations: 1. That the magnetism of the two poles is evidently of an opposite nature, the one attracting what the other repels, and vice versa. If a piece of iron is attracted by one, it ought therefore to be repelled by the other; but each pole, by inducing on the near end of the iron a magnetism opposite to its own, and on the remote end a similar magnetism, and its action diminishing as the distance increases, the attraction must always be in excess, and the iron must on the whole be attracted. 2. When we have two magnets placed in a parallel position, with their opposite poles together, if a piece of common iron be brought near their extremities, the different poles counteracting each other, the piece of iron will not be supported by the two magnets together, unless there is an inequality of action; but it is evident that either of them alone would be capable of supporting the iron. 3. In all the cases where the induction of magnetism is slow, the attraction is proportionally weak, and the attraction increases exactly according to the increase of the progressive induction. 4. An ore of iron that is not capable of acquiring magnetism, is not attracted by the magnet, and on the other hand it is an universal fact, that no substance which is not attracted by the magnet, can be rendered magnetical.

The induction of magnetism by juxtaposition affords a complete explanation of the curious arrangement of iron filings round a magnet. Let us suppose a great many small oblong pieces of iron to be lying near each other on the surface of mercury, and that a strong magnet be brought into the midst of them. They are all immediately rendered magnetical by induction; any one that is nearest the north pole of the magnet acquiring two poles, one a north and the other a south pole, turns the south pole towards the north pole of the magnet, and the north pole away from it; a similar effect is produced on another piece or filing that lies near the first, and so on of the rest. All those that lie near each other must mutually attract, as the magnetism of each is so disposed that both ends of it are in a state of attraction towards one or other of its neighbours. They will therefore arrange themselves by coalescence in a particular manner; if they are near enough, they will unite by their extremities, and if they are at some distance they will point towards each other, forming curved lines.

It is found that the magnetism of magnets, whether natural or artificial, is continually tending to decay, Now as we find that this magnetism may be induced merely by the approach of a magnet, and as we know that in producing magnetism, magnets may oppose each other, it is reasonable to conclude, that when a flight though permanent magnetism has been acquired by a piece of iron by its vicinity to a magnet, it may be destroyed, and the contrary magnetism induced, by applying a magnet in the opposite direction. Accordingly it is a well-known fact, that the poles of magnets made of soft steel can be reversed at pleasure.

This explains why magnetic repulsion is always weaker than attraction at the same distance, as magnets, when placed with their familiar poles fronting each other, in order to try their repulsion, are thereby weakened; whereas, on the contrary, magnets applied with their opposite poles, so as to attract each other, are thereby improved, and their attractive powers are made to appear greater than they really are.

It has been observed that a magnet is not weakened by inducing magnetism on iron. In fact, it is rather improved by such induction, and this will increase the effect; for as the magnet is improved, the induced magnetism of the iron will be thereby increased, and thus the magnet will be thus farther improved.

After what has been said, we need not enter further into an explanation of the phenomena, or of the processes employed in making artificial magnets. They are all referable to this one fact of the induction of magnetism by juxtaposition, and explanations will readily suggest themselves to readers who carefully consider the preceding facts, and compare them with Dr Gilbert's theory of terrestrial magnetism.

It is now time for us to return to Dr Gilbert's hypothesis, and consider an objection that has been strongly urged against it.

There is observed no tendency in the magnetic needle towards the great terrestrial magnet, that is, though, when made to float on water, it speedily acquires directive power, it does not in these latitudes approach the north side of the vessel, nor does an iron bar appear heavier when its south pole is uppermost, as ought to be the case on account of the attraction of the great magnet. Dr Gilbert saw this objection, and it appears to have given him some concern. He attempted to get rid of it by observing that the directive power of a magnet is greater than its attractive force; a fact in support of which he brings many experiments. A much more satisfactory answer may be derived from what has been stated respecting the actions of the four poles. We thence find, that the polarity of the needle depends on the difference of the sums of the actions of each pole of the magnet on both poles of the needle; whereas its tendency towards the magnet arising from the attraction between them, depends on the difference of the differences of the same actions. Hence the former may be very great, while the latter is very small. We find that small iron filings are much less forcibly attracted by magnets than coarse ones, and, if we consider that the largest magnets which we employ do not bear so great a proportion to the earth, as the finest iron filings to an ordinary magnet, we shall not wonder that the attractive power of the earth is not very sensible.

As this objection is one of the strongest that can be brought against the theory, and as we may consider this as done away, we may now receive the theory as just so far as it goes. We must remark, that though we call that pole of a magnet which inclines towards the earth in the northern latitudes, a north pole, it is properly speaking a south pole; for as we must call that pole of the great magnet the north pole which is in the north, and as this pole produces the contrary polarity in the proximate end of a needle, that end must be possessed of south polarity. We shall return to this subject in the article VARIATION.

Some valuable observations on terrestrial magnetism have lately been made in France by M. M. Humboldt and Biot, and as they would suffer materially by abridgement, we shall present our readers with the greatest part of the memoir nearly as translated in the Philosophical Magazine, vol. xxii.

After explaining the object of the memoir, and giving an account of the share that he had in conducting the observations, M. Biot proceeds as follows.

It is necessary to consider the action of terrestrial magnetism under different points of view, corresponding to the different classes of the phenomena which it produces.

If we consider it first in general, we find that it acts on the whole surface of the globe, and that it extends beyond it. This fact, which was doubted, has been lately proved by M. Guy-Lussac, during his two aerostatic voyages. And if these observations, made with all the care possible, have not shewn the least sensible diminution in the intensity of the magnetic force, at the greatest height to which man can attain, we have a right to conclude that this force extends to an indefinite distance from the earth, where it decreases, perhaps, in a very rapid manner, but which at present is unknown to us.

If we now consider magnetism at the surface even of the earth, we shall find three grand classes of phenomena which it is necessary to study separately, in order to have a complete knowledge of its mode of action. These phenomena are, the declination of the magnetic needle, its inclination, and the intensity of the magnetic force, considered either comparatively in different places or in themselves, paying attention to the variations which they experience. It is thus that, after having discovered the action of gravity as a central force, its variation, resulting from the figure of the earth, was afterwards ascertained in different latitudes.

The declination of the magnetic needle appears to be that phenomenon which hitherto has more particularly fixed the attention of philosophers, on account, no doubt, of the assistance which they hoped to derive from it in determining the longitude; but when it was known that the declination changes in the same place, in the course of time, when its diurnal variations were remarked, and its irregular traversing occasioned by different meteors, in a word, the difficulty of observing it at sea, within one degree nearly, it was necessary to abandon that hope, to consider the cause of these phenomena as much more complex and intricate than had been at first imagined.

In regard to the intensity of the magnetic power in different parts of the earth, it has never yet been measured in a comparative manner. The observations of M. Humboldt on this subject have discovered a very remarkable phenomenon; it is the variation of the intensity in different latitudes, and its increase proceeding from the equator to the poles.

The compasses, indeed, which at the departure of M. Humboldt gave at Paris 245 oscillations in 10 minutes, gave no more in Peru than 211, and it constantly varied in the same direction; that is to say, the number of the oscillations always decreased in approaching the equator, and always increased in advancing towards the north.

These differences cannot be ascribed to a diminution of the force in the magnetism of the compass, nor can we suppose that it is weakened by the effect of time and of heat; for after three years residence in the warmest countries of the earth, the same compass gave again in Mexico oscillations as rapid as at Paris.

There is no reason to doubt the justness of M. Humboldt's observations, for he often observed the oscillations in the vertical plane perpendicular to that meridian; and by decomposing the magnetic force in the latter plane, and comparing it with its total action, which is exercised in the former, we may from these data calculate its direction, and consequently the direction of the needle (c). This inclination, thus calculated, is found always conformable to that which M. Humboldt observed directly. When he made his experiments, however, he could not foresee that they would be subjected to this proof by which M. Laplace verified them.

As the justness of these observations cannot be contested, we must allow also the truth of the result which they indicate, and which is the increase of the magnetic force proceeding from the equator to the poles.

To follow these results with more facility, it will be proper to set out from a fixed term; and it appears natural to make choice for that purpose of the points where the inclination of the magnetic needle is null, because they seem to indicate the places where the opposite action of the two terrestrial hemispheres is equal. The series of these points forms on the surface of the earth a curved line, which differs very sensibly from the terrestrial equator, deviating from it to the south of the Atlantic ocean, and to the north in the south sea. M. Humboldt found this equator in Peru about 7° 1' S. Lat. which for that part of the earth places it nearly in the spot where Wilke and Lemounier had fixed it.

The places situated to the north of that point may be divided into four zones, the three first of which, being nearer the equator, are about 4° of latitude, while the latter, more extensive and more variable, is 14°. So that the system of these zones extends in America from the magnetic equator to 23° of north latitude, and comprehends in longitude an interval of about 50°.

The first zone extends from 7° 1' of south latitude to 2° 54'. The mean number of the oscillations of the needle in the magnetic meridian in 10' of time is there 211.9: no observation gives less than 211, or more than 214. From M. Humboldt's observations one might form a similar zone on the south side of the magnetic equator, which would give the same results.

The second zone extends from 2° 13' of south latitude to 3° 15' of north latitude. The mean term of the oscillations is there 217.9: they are never below 210, nor above 226.

The fourth zone, broader than the other two, extends from 9° 15' to 23° 8' of north latitude. Its mean term is 237: it never presents any observation below 229, nor above 240.

We are unacquainted, in regard to this part of the earth, with the intensity of the magnetic force beyond the latitude of 23° north; and on the other hand, in Europe, where we have observations made in high latitudes, we have none in the neighbourhood of the equator; but we will not venture to compare these two classes of observations, which may belong to different systems of forces, as will be mentioned hereafter.

However, the only comparison of results, collected in America by M. Humboldt, appears to us to establish with certainty the increase of the magnetic force from the equator to the poles; and, without wishing to connect them too closely with the experiments made in Europe, we must remark, that the latter accord so far with the preceding as to indicate the phenomenon.

If we have thus divided the observations into zones parallel to the equator, it is in order that we may more easily shew the truth of the fact which results from them, and in particular to render the demonstration independent of those small anomalies which are inevitably mixed with these results.

Though these anomalies are very trifling, they are however, so sensible, and so frequently occur, that they cannot be ascribed entirely to errors in the observations. It appears more natural to ascribe them to the influence of local circumstances, and the particular attractions exercised by collections of ferruginous matters, chains of mountains, or by the large masses of the continents.

One of them, indeed, having carried to the Alps the magnetic needle employed in an aerial excursion, he found that its tendency to return to the magnetic meridian was constantly stronger in these mountains than it was at Paris before his departure, and than it has been found since his return. This needle, which made at Paris 83.9° in 10' of time, has varied in the following manner in the different places to which it was carried.

(c) Let HOC (fig. 28.) be the plane of the magnetic meridian passing through the vertical OC; let OL be the direction of the needle situated in that plane, and OH a horizontal. The angle LOH will be the inclination of the needle, which we shall denote by I. If F represent the total magnetic force which acts in the direction OL, the part of this force which acts according to OC, will be F sine of I: but the magnetic forces which determine the oscillations of the needle in any plane, are to each other as the squares of the oscillations made in the same time. If we denote them by M, the number of oscillations made also in 10' of time in the magnetic meridian, and by P, the number of oscillations made also in 10' in the perpendicular plane, we shall have the following proportion:

\[ \frac{F \sin. I}{F} = \frac{P^2}{M^2} \]

from whence we deduce

\[ \sin. I = \frac{P^2}{M^2}. \]

The inclination then may be calculated by this formula, when we have oscillations made in the same planes.

In like manner, by making a needle oscillate successively in several vertical planes, we might determine the direction of the magnetic meridian. Places of Observation. Number of Oscillations in 10' of Time.

Paris, before his departure, 83.9 Turin, 87.2 On Mount Genève, 88.2 Grenoble, 87.4 Lyons, 87.3 Geneva, 86.5 Dijon, 84.5 Paris, on his return, 83.9

These experiments were made with the greatest care, conjointly with excellent observers, and always employing the same watch verified by small pendulums, and taking the mean terms between several series of observations, which always differed very little from each other. It appears thence to result, that the action of the Alps has a sensible influence on the intensity of the magnetic force. M. Humboldt observed analogous effects at the bottom of the Pyrenees; for example, at Perpignan. It is not improbable that they arose from the mass of these mountains, or the ferruginous matters contained in them; but whatever may be the cause, it is seen by these examples that the general action of terrestrial magnetism is sensibly modified by local circumstances, the differences of which may be perceived in places very little distant from each other. This truth will be further confirmed by the following observations.

It is to causes of this kind, no doubt, that we must ascribe the diminution of the magnetic forces observed in some mountains; a diminution which, on the first view, might appear contrary to the results obtained during various aerial voyages. This conjecture is supported by several observations of M. Humboldt. By making his needle to oscillate on the mountain of Guadaloupe, which rises 338 toises above Santa-Fé, he found it in 10' of time give two oscillations less than in the plain. At Silla, near Caracas, at the height of 1316 toises above the coast, the diminution went so far as five oscillations; and on the other hand, on the volcano of Antisana, at the height of 2467 toises, the number of oscillations in 10 minutes was 230; though at Quito it was only 218, which indicates an increase of intensity. A similar effect was observed on the summit of Mount Geneve, at the height of 800 or 900 toises, as may be seen from the numbers already given; and on this mountain M. Biot found the greatest intensity of the magnetic force. He saw on the hill of La Superba, in the neighbourhood of Turin, an example of these variations equally striking. Observing, with Valfai, on this hill, at the elevation of 300 toises, they found 87 oscillations in 10 minutes of time. On the side of the hill they had 88.8 oscillations, and at the bottom, on the bank of the Po, they obtained 87.3. Though these results approach very near to each other, their difference is, however, sensible, and fully shews that their small variations must be considered as slight anomalies produced by local circumstances.

This examination leads us to consider the intensity of magnetism on the different points of the surface of the globe, as subject to two sorts of differences. One kind are general; they depend merely on the situation of the places in regard to the magnetic equator, and belong to a general phenomenon, which is the increase of the intensity of the magnetic forces in proportion as we remove from the equator; the other kind of variations, which are much smaller and altogether irregular, seem to depend entirely on local circumstances, and modify either more or less the general results.

If we consider terrestrial magnetism as the effect of an attractive force inherent in all the material particles of the globe, or only in some of these particles, which we are far from determining, the general law will be, the total result of the system of attraction of all the particles, and the small anomalies will be produced by the particular attractions of the partial systems of the magnetic molecules diffused irregularly around each point; attractions rendered more sensible by the diminution of the distance.

It now remains to consider the inclination of the magnetic needle in regard to the horizontal plane. It has been long known that this inclination is not everywhere the same; in the northern hemisphere the needle inclines towards the north; in the southern towards the south; the places where it becomes horizontal form the magnetic equator; and those where the inclination is equal, but not null, form on each side of that equator curved lines, to which the name of magnetic parallels has been given, from their analogy to the terrestrial parallels. One may see in several works, and particularly in that of Lemourier, entitled Lois du Magnétisme, the figure of these parallels, and their disposition on the face of the earth.

It evidently results from this disposition, that the inclination is in proportion as we recede from the magnetic equator; but the law which it follows in its increase, has not yet, as far as appears to us, been given. To ascertain this law, however, would be of great utility; for the inclination seems to be the most constant of all magnetic phenomena, and it exhibits much fewer anomalies than the intensity. Besides, if any rule well confirmed could be discovered on this subject, it might be employed with advantage at sea to determine the latitude, when the weather does not admit an observation of the sun; which is the case in various places during the greater part of the year. We have some reason to expect this application, when we see the delicacy of that indication in the observations of M. Humboldt, where we find 35' 6'' of difference between two towns to near each other as Nîmes and Montpellier. These motives have induced us to study with great interest the series of observations made by M. Humboldt in regard to the inclination; and it appears to us that they may be represented very exactly by a mathematical hypothesis, to which we are far from attaching any reality in itself, but which we offer merely as a commodious and sure mode of connecting the results.

To discover this law, we must first exactly determine the position of the magnetic equator, which is as an intermediate line between the northern and the southern inclinations. For this purpose we have the advantage of being able to compare two direct observations, one of La Perouse, and the other of M. Humboldt. The former found the magnetic equator on the coasts of Brazil at 10° 57' of south latitude, and 25° 25' of west longitude, counted from the meridian of Paris. The latter found the same equator in Peru at 7° 1' of south latitude, and 80° 41' of west longitude, also reckoning from the same meridian. These data are sufficient cient to calculate the position of the magnetic equator, supposing it to be a great circle of the terrestrial sphere; an hypothesis which appears to be conformable to observations. The inclination of this plane to the terrestrial equator is thus found to be equal to 10° 58' 56", and its occidental node on that equator is at 120° 2' 5" west from Paris, which places it a little beyond the continent of America, near the Galapagos, in the South Sea; the other node is at 59° 57' 55" to the east of Paris, which places it in the Indian seas (d).

"We do not give this determination as rigorously exact; some corrections might no doubt be made to it, had we a greater number of observations equally precise; but we are of opinion that these corrections would be very small, and it will be seen afterwards that, independently of the confidence which the two observations we have employed deserve, we have other reasons for entertaining this opinion (e)."

"It is very remarkable that this determination of the magnetic equator agrees almost perfectly with that given long ago by Wilke and Lemounier. The latter in particular, who for want of direct observations had discussed a great number of corresponding observations, indicates the magnetic equator in Peru towards 7° 20' of south latitude, and M. Humboldt found it in the same place at 7° 1'; besides, Lemounier's chart, as well as that of M. Wilke, indicates for the inclination of the magnetic meridian about 11°, and they place the node about 140° of west longitude, reckoned from the meridian of Paris.

"Can it be by chance, then, that these elements, found more than 40 years ago, should accord so well with ours founded on recent observations? or does the inclination of the magnetic equator experience only very small variations, while all the other symptoms of terrestrial magnetism change so rapidly? We should not be far from admitting the latter opinion, when we consider that the inclination of the magnetic needle has changed at Paris 3° in 60 years since it has been observed; and that at London, according to the observations of Mr Graham, it has not changed 2° in 200 years, while the declination has varied more than 26° in the same interval, and has passed from east to west: but on the other hand the observation of the inclination is so difficult to be made with exactness, and it is so short a time since the art of measuring it with precision was known, that it is perhaps more prudent to abstain from any premature opinion on phenomena, the cause of which is totally unknown to us."

To employ the other observations of M. Humboldt in regard to the inclination, the terrestrial latitudes and longitudes reckoned from the magnetic equator were first reduced. The latter, being reckoned from the node of that equator in the South Sea, M. Biot first perceived by these calculations that the position of that plane determined by preceding researches was pretty exact; for some of the places, such as Santa-Fé and Java, where M. Humboldt observed inclinations almost equal, were found nearly on the magnetic parallel, though distant from each other more than 60° of longitude.

When these reductions were made, M. Biot endeavoured to represent the signs of the inclinations observed, and to leave as little to chance as possible. He first tried a mathematical hypothesis conformable enough to the idea which has hitherto been entertained in regard to terrestrial magnetism.

He supposed in the axis of the magnetic equator, and at an equal distance from the centre of the earth, two centres of attractive forces, the one auroral and the other boreal, in such a manner as to represent the two opposite magnetic poles of the earth. He then calculated the effect which ought to result from the action of

(d) To calculate this position, let NEE' (fig. 29.) be the terrestrial equator; NHL the magnetic equator, supposed also to be a great circle, and HL the two points of that equator, observed by Messrs Humboldt and La Perouse. The latitudes HE, LE', and the arc EE', which is the difference of longitude of these two points, is known; consequently, if we suppose HE=b, LE'=b', EE'=v, EN=x, and the angle ENH=y, we shall have two spherical triangles NEH, NE'L, which will give the two following equations:

\[ \sin x = \frac{\tan b \cot y}{R} \sin (x+v) = \frac{\tan b' \cot y}{R} \]

from which we deduce

\[ \sin \frac{(x+v)}{\sin x} = \frac{\tan b'}{\tan b} \]

and developing

\[ \cot x = \frac{\tan b \sin v}{\tan b'} - \frac{\cot v}{\sin v} \]

Let us now take an auxiliary angle φ, so that we may have

\[ \tan g = \frac{\tan b \sin v}{\tan b'}, \text{ and we shall have} \] \[ \tan x = \frac{\sin v \sin \varphi}{\sin (v-\varphi)} \]

By these equations we may find x, and then y, by any of the first two.

(e) La Perouse, after having doubled Cape Horn, fell in a second time with the magnetic equator in 13° north latitude, and 119° 7' of longitude west from Paris. He was therefore very near the node of the magnetic equator, such as we have deduced it from observations. This fact establishes in a positive manner two important consequences: First, that the preceding determinations require only very slight corrections; and the second, that the magnetic equator is really a great circle, if the earth, if not exactly, at least very nearly. of these centres in any point of the surface of the earth, making their attractive force reciprocally vary as the square of the distance; and in this manner he obtained the direction of the result of their forces, which ought to be that also of the magnetic needle in that latitude.

He supposes that the point B (fig. 30.) is the north magnetic pole of the earth, and that the point A is the south magnetic pole; he supposes also that there is in the point M, at the surface of the earth, a molecule of the austral fluid which is attracted by B and repelled by A in the inverse ratio of the square of the distance; and he requires what will be the direction of the power resulting from these two forces acting on that molecule. It is evident that this direction will be that also which would be assumed in the point M by the needle of a compass freely suspended; for, in consequence of the smallness of the needle in comparison of the radius of the earth, the lines drawn from its points to one centre, B or A, may be considered as parallel, especially if the points A and B are near the centre of the earth, which is the case with nature, as may be seen.

He first supposes that the earth has a spherical figure, and that the two poles A and B are equal in force, and he then examines how far the latter supposition agrees with the results observed.

Let AM then = D', BM = D, CP = x, PM = y, the angle MCP = u, CA = CB = a. He then makes a = Kr; r being = the radius of the earth, and K a constant but indeterminate quantity.

Let X, Y, also be the forces which attract M in the direction of the axes of the co-ordinates, and β the angle which the resulting force makes with the axis ABC.

\[ \tan \beta = \frac{\text{fin. } u}{\text{col. } u - K \left( \frac{D'^3 + D^3}{D'^3 - D^3} \right)} \]

\[ K \left( \frac{D'^3 + D^3}{D'^3 - D^3} \right) = \frac{(1 + 2K \text{ col. } u + K^2)^{\frac{1}{2}} + (1 - 2K \text{ col. } u + K^2)^{\frac{1}{2}}}{(1 + 2K \text{ col. } u + K^2)^{\frac{1}{2}} - (1 - 2K \text{ col. } u + K^2)^{\frac{1}{2}}} K. \]

These equations determine the direction of the magnetic needle in regard to each point M, the distance of which from the magnetic equator is known; but it is seen that this direction depends on the quantity K, which represents the distance of the magnetic centres from the centre of the earth; this distance being expressed in parts of the terrestrial radius, we must therefore first determine this quantity from observations.

To do it in the manner of approximation, and thus acquire a first idea of the value of K, M. Biot chose an observation made by M. Humboldt at Carichana in 6° 34' 5" of north latitude counted from the terrestrial equator, and 70° 18' west longitude reckoned from the meridian of Paris, which gives 14° 52' 25" of longitude counted from the magnetic equator, and 48° 51' 53" of west longitude, proceeding from the node formed by that equator with the equator of the earth. The inclination of the magnetic needle was observed in that place by M. Humboldt in the month of Messidor, year 8, and found to be equal to 33.78° of the centigrade division. A comparison of this result with the other observations of M. Humboldt, shews that it may indeed be considered as agreeing to that latitude.

He then gives the following equations, in which F is the magnetic force, at a distance equal to unity.

\[ x = \frac{F}{D^3} \text{ col. MBD} - \frac{F}{D'^3} \text{ col. MAD}; \] \[ D'^2 = y^2 + (x + a)^2 = r^2 + 2 \text{ axis } + a^2; \] \[ Y = \frac{F}{D^3} \text{ fin. MBD} - \frac{F}{D'^3} \text{ fin. MAD}; \] \[ D^2 = y^2 + (x - a)^2 = r^2 - 2 \text{ axis } + a^2, \] or, by putting for the coines their values:

\[ X = \frac{F(x - a)}{D^3} - \frac{F(x + a)}{D'^3} \] \[ Y = \frac{Fy}{D^3} - \frac{Fy}{D'^3}, \] and as we have

\[ \tan \beta = \frac{Y}{X}, \] we shall have also

\[ \tan \beta = \frac{Y}{X} = \frac{Y}{D^3 - D'^3} \] \[ = \frac{x + a}{D^3} - \frac{x - a}{D'^3} = \frac{y(D'^3 - D^3)}{(D'^3 - D^3) - a(D'^3 + D^3)}, \] and by putting for x, y and a, their values, col. u; r fin. u, K r;

\[ \tan \beta = \frac{\text{fin. } u}{\text{col. } u - K \left( \frac{D'^3 + D^3}{D'^3 - D^3} \right)} \] \[ D'^2 = r^2(1 + 2K \text{ col. } u + K^2); \] \[ D^2 = r^2(1 - 2K \text{ col. } u + K^2); \] which gives the system of the two equations,

\[ K \left( \frac{D'^3 + D^3}{D'^3 - D^3} \right) = \frac{(1 + 2K \text{ col. } u + K^2)^{\frac{1}{2}} + (1 - 2K \text{ col. } u + K^2)^{\frac{1}{2}}}{(1 + 2K \text{ col. } u + K^2)^{\frac{1}{2}} - (1 - 2K \text{ col. } u + K^2)^{\frac{1}{2}}} K. \]

To make use of it, M. Biot successively gave to K different values in the formula; he calculated the inclinations resulting from that latitude; and comparing these results with that which M. Humboldt really observed, the progress of the errors naturally led him to the most proper supposition. The following is a table of these trials.

<table> <tr> <th>Values of K.</th> <th>Inclinations of the Needle.</th> <th>Errors.</th> </tr> <tr> <td>K=1</td> <td>7.73°</td> <td>26.04</td> </tr> <tr> <td>K=0.6</td> <td>18.80</td> <td>14.97</td> </tr> <tr> <td>K=0.5</td> <td>22.04</td> <td>11.73</td> </tr> <tr> <td>K=0.2</td> <td>29.38</td> <td>4.39</td> </tr> <tr> <td>K=0.1</td> <td>30.64</td> <td>3.13</td> </tr> <tr> <td>K=0.01</td> <td>31.04</td> <td>2.73</td> </tr> <tr> <td>K=0.001</td> <td>31.07</td> <td>2.7</td> </tr> </table>

The first value of K would place the centre of the magnetic forces at the surface of the earth and the poles of the magnetic equator. It is seen that this supposition cannot be admitted, because it would give an increase of inclination much less rapid than that indicated by observations. The case is the same with the following following results, which place the centres of action on the terrestrial radius at different distances from the centre of the earth; but it is seen also in general, that they approach more and more to the truth in proportion as this distance becomes less; which evidently shews that the two centres of action of the magnetic forces are situated near the centre of the earth. All the other observations of M. Humboldt would also lead to the same consequence.

The most proper supposition would be to make K null, or so small that it would be needless to pay attention to it; which amounts to the same thing as to consider the two centres of action placed, as we may say, in the same molecule. The result, indeed, obtained in this manner is the most exact of all; it is =31.0843°; this value is still a little less than that which M. Humboldt observed, and the difference is =2.69; but it must be considered also that the formula from which we derive these values supposes the position of the magnetic equator is perfectly determined; but it may not be so with the utmost exactness, according to the two only observations of La Perouse and Humboldt, which we have employed. It is therefore by studying the progress of the formula, and comparing it with the observations, that we are able to appreciate it judiciously; after which we may think of remedying the small errors with which it may be accompanied.

To obtain the result here mentioned, and which is, as it were, the limit of all those which may be obtained by giving to K different values, it is to be remarked that the quantity

\[ K \left( \frac{D^3 + D^3}{D^3 - D^3} \right), \]

or,

\[ K \frac{(1 + 2K \cot u + K^2)^{\frac{3}{2}} + (1 - 2K \cot u + K^2)^{\frac{3}{2}}}{(1 + 2K \cot u + K^2)^{\frac{3}{2}} - (1 - 2K \cot u + K^2)^{\frac{3}{2}}} \]

becomes \( \frac{0}{0} \) when K is null; but by applying to it the methods of known quantities, it will be found that its value in this supposition is really determinate and

\[ = \frac{1}{3 \cot u}. \]

By substituting this in the formula we shall have

\[ \tan \beta = \frac{\sin u}{\cot u - \frac{1}{3 \cot u}} \]

an equation which may be reduced to this form:

\[ \tan \beta = \frac{\sin u}{\cot 2u + \frac{1}{3}}; \]

which will easily give the value of \( \beta \); and when this value is known, we shall have the inclination I, by the following formula:

\[ I = 100 + u - \beta, \]

which will serve throughout the whole extent of the two hemispheres.

From the progress thus traced out, it is seen that the preceding formula is not merely an empirical construction of observations; on the contrary, it is totally independent, and only supposes the inclination of the magnetic needle to be produced by a magnet infinitely small, placed in the centre of the terrestrial surface; but by calculating from this formula the inclination for the different latitudes, M. Biot found precisely the same numbers as M. Humboldt observed either in Europe or America; and it is not his observations only that are represented in this manner; but those which have been made in Russia, and at Kola in Lapland, during the last transit of Venus, are also comprehended under the same law.

It is seen that the results of the formula deviate very little from the observations; but these differences may be rendered still smaller. By examining, indeed, the progress of the errors, it is seen that the numbers given by calculation are a little too small in America for the low latitudes, and a little too great for the high latitudes, which shews that the whole may be allowed, with some slight modifications, either by changing, however little, the node and inclination of the magnetic equator, which two observations cannot determine with the utmost exactness, or by displacing ever so little our small magnet, leaving, however, its centre in the plane of the magnetic equator, and placing it in such a manner that it shall be a little nearer America than Europe. It is by these observations themselves, when we shall have a greater number, that we must be guided in these small corrections.

In a word, it must not be expected that we can represent in a rigorous manner, by a mathematical law, all the inclinations observed; for the phenomenon of the inclination, though more regular than the other magnetic effects, is not free from some anomalies; this may be easily seen on constructing the curve given by the observations themselves. Thus, for example, the inclination observed at Popayan is 8° 10' greater than at St Carlos del Rio Negro, though the magnetic latitude of the latter is 3° 7' greater. The case is the same with observations made at Java and Santa Fé. Other anomalies are discovered in the comparative progress of the observations and formula. This is the case in regard to Carichana, St Thomas de la Guyane, and Cartagena. The increase of the inclination from the first to the second of these points is by no means in harmony with the increase from the second to the third; and if we compare together the intensities observed in these different places, the anomalies they exhibit announce in some measure those which the inclination ought to experience.

The cause of these anomalies becomes evident from what has been already remarked; they are merely the effect of local circumstances, and arise from the small systems of attraction by which the general phenomena are modified. This must be sensible in particular for that part of America which M. Humboldt travelled over, and which is traversed throughout its whole length by the grand chain of the cordillera of the Andes. It is also in these places that the most considerable differences exist. Popayan, for example, is situated near the volcanoes of Sotara and Paurace; it is joined to bafatic mountains abounding with magnetic iron. Near Sulumito, to the east of Popayan, these bafatic columns have very striking poles; in like manner Mexico is situated at the height of 1160 toises on the ridge of the grand cordillera of Lenichtitlan; the ground there is covered with porous bafaltes and amygdaloids, daloïds, which are almost all charged with magnetic iron. Must not all these causes have a sensible influence on the inclination of the magnetic needle; and must not the different dispositions of the ferruginous masses, or their change of state, in consequence of the action of nature, produce also variations? M. Humboldt made on this point a decisive observation: the earthquake of the 4th November 1799 changed at Cumana the inclination of the needle. On the 16th of November it was \(43^\circ\ 65'\); on the 7th it was only \(42^\circ\ 75'\), and ten months after it returned to \(42^\circ\ 85'\), but it did not regain its former value; the intensity of the magnetic force was not changed by the effect of this earthquake.

It is proved, then, by these observations, that local circumstances may have on the inclination a sensible influence; and this influence is remarked in the countries traversed by M. Humboldt.

It appears, therefore, that the mathematical hypothesis which we have employed really expresses the law of nature, at least to the north of the magnetic equator; for, though the first results observed towards the south seem to bend to it also, the uncertainty under which we are, in regard to the true cause of these phenomena, must stop our conjectures, and prevent us from extending too far the consequences of the laws which we observe (r).

From the preceding results, we may calculate the points where the axis of the magnetic equator pierces the terrestrial surface; for their latitudes are equal to the complements of the obliquity of that equator, and their meridian is at \(100^\circ\) of longitude from its nodes. The north magnetic pole is found also at \(79^\circ\ 1' 4''\) of north latitude, and at \(30^\circ\ 2' 5''\) of longitude west from Paris, which places it to the north of America. The other magnetic pole, symmetric to the preceding, is situated in the same latitude south, and at \(149^\circ\ 67' 53''\) of longitude east from Paris, which places it amidst the eternal ice; indications entirely analogous to those of Wilke and Lemounier.

If we could reach these poles, the compass would be seen vertical; but if any confidence can be placed in the law which we have discovered, this would be the only difference which would be observed in regard to the inclination, and we should be still as far distant as in Europe from the real centres which produce it. This result might appear to be of such a nature as to diminish the interest one might have in visiting these horrid regions, had we not also the hope of discovering there new phenomena in regard to the intensity of the magnetic force, and the influence of meteors.

These consequences do not entirely accord with the opinion pretty generally received, and which ascribes the increase of the magnetic effects towards the north to the great quantity of iron dispersed throughout these regions; but it appears to us that this opinion is not agreeable to the truth. The cordillera of the Andes contains an enormous quantity of magnetic iron; the native iron of Chaco, that problematic mass analogous to that of Pallas, and those of Xacatares in Mexico, is found even under the tropics.

On seeing the inclinations of the compass so exactly represented in the hypothesis, they endeavoured to discover whether it could be applied to the intensities observed by M. Humboldt; but they found that it did not apply. It gives indeed, an increase of the magnetic forces from the equator to the pole; but this increase, which at first is too slow, becomes afterwards too rapid. M. Biot has not yet been able to try whether the small displacement of the terrestrial magnet will contribute towards representing them better; but it must be remarked, that the series of the intensities is extremely whimsical, and contains an infinite number of anomalies, so that local phenomena may have on this phenomenon a much more sensible influence than on the inclination.

On reviewing the results which have been given, it is seen that we have first determined the position of the magnetic equator by direct observations, which had never been done before; we have then proved that the magnetic force increases in proceeding from that equator to the poles; in the last place, we have given a mathematical hypothesis, which, when reduced to a formula, satisfies all the inclinations hitherto observed.

Supposing, as has been done in this formula, the final corrections of which it is susceptible, its utility becomes evident, either for making known, in the course of time, the variations which may take place in the action of the terrestrial magnetism, or to ascertain or even foresee the value of the inclination, which in a great many cases is of the utmost importance.

For example, near the magnetic equator, the increase or diminution of the inclination will indicate to a vessel on a voyage whether she has gained or lost in latitude by currents. This knowledge of the latitude is sometimes as important as that of longitude. On the coasts of Peru, for example, the currents tend from Chilié to the north and north-east with such force, that one may go from Lima to Guayaquil in three or four days, and two, three, and sometimes five months are necessary to return. It is consequently of the greatest importance for vessels coming from Chili which stretch along the coast of Peru, to know their latitude. If they go beyond the port to which they are bound, they must work to the southward, and every day's progress requires often a month of return. Unfortunately, the fogs which prevail during four or five months on the coast of Peru, prevent

(r) Observations made at the Cape of Good Hope, Cape Horn, and New Holland, by different navigators, are very exactly represented by the above-mentioned formula; and it follows, that it extends also to the austral hemisphere. We hope soon to have numerous and very exact observations on the inclination of the needle in that part of the earth. But we have thought it our duty to add to our table such results as relate to it, and which we have been able to procure. We have inserted also two observations on the intensity, made with great care by M. Roffel, during the expedition of d'Entrecasteaux, which are very important, as they prove that the terrestrial magnetic force increases also in the austral hemisphere in proportion as one removes from the equator. prevent navigators from distinguishing the form of the coast; nothing is seen but the summits of the Andes, and that of the peaks which rise above that stratum of vapours; but the figure of it is so uniform that pilots fall into mistakes. They often remain 12 or 15 days without seeing the fun or stars, and during that interval they come to anchor, being afraid of overhooting their port; but if we suppose that the inclination of the magnetic needle in the ports to the south of Lima is known, for example at Chancay, Huaura, and Santa, the dipping needle will show whether it be, in regard to Lima, to the south or to the north. It will show at the same time opposite what point of the coast a vessel is; and this indication will be attended with more exactness that one could hope for, because in these seas the inclination varies with extraordinary rapidity. M. Humboldt, to whom we are indebted for these remarks, observed in these seas the following values.

<table> <tr> <th>Places.</th> <th>South Latitudes.</th> <th>Inclinations.</th> </tr> <tr> <td>Huancey</td> <td>10° 4'</td> <td>6.80°</td> </tr> <tr> <td>Huaura</td> <td>11° 3'</td> <td>9.00°</td> </tr> <tr> <td>Chancay</td> <td>11° 33'</td> <td>10.35°</td> </tr> </table>

These observations prove that the error of three or four degrees in the inclination in these seas would produce but a degree of error in latitude; and, on account of the tranquillity of the Pacific ocean, the inclination may be observed to within a degree nearly. Frequent instances of such results may be seen in books of voyages. In like manner, if one knew exactly the inclination at the mouth of the Rio de la Plata, it would be very useful to navigators, who, when the Pamperos blow, remain 15 or 18 days without seeing the heavenly bodies, and go on different tracks for fear of losing the parallel of the mouth of that river.

In a word, the inclination may indicate also the longitude in these seas; and this method may be employed when others fail. A vessel which fails there in the direction of a parallel could not find its longitude either by a chronometer or the declination of Halley, unless a star could be seen, in order to take a horary angle or the magnetic azimuth. The dipping needle then throws light on the longitude amidst the thickest fogs. We point out this method as one of those which have only a local application; but hitherto little attention has been paid to it. These ideas may be extended and rectified by able navigators.

In general, if the inclination of the needle, and the law we have tried to establish, could be depended on, to observe the inclination and the terrestrial latitude would also be sufficient to determine the longitude; but we have not yet examined the extent of the errors of which this method may be susceptible, and consequently we confine ourselves to a mere indication of it.

The phenomenon of the inclination has in maritime observations a peculiar and very remarkable advantage, namely, that of not being subject to those great progressive variations which affect the declination. Without repeating what we have already said above on the supposed constancy of this phenomenon, it may be remarked that our formula even affords a new proof that it may comprehend in the same law the observations made many years ago in Lapland, those which Lacaille brought back in 1751 from the Cape of Good Hope, and those which M. Humboldt has lately made in America.

In short, when we tried to represent the inclinations in different latitudes by the supposition of a magnet infinitely small, very near the centre of the earth, and perpendicular to the magnetic equator, we did not pretend to consider that hypothesis as any thing real, but only as a mathematical abstraction useful to connect the results, and proper to ascertain in future whether any changes exist. In regard to the declination and intensity, we freely confess that we are entirely unacquainted with their laws or their causes; and if any philosopher is so fortunate as to bring them to one principle, which explains at the same time the variations of the inclination, it will no doubt be one of the greatest discoveries ever made. But this research, exceedingly difficult, requires, perhaps, before it be attempted, more observations, and in particular more precise observations, than have hitherto been collected. For this reason we have presented the preceding researches, imperfect as they are, hoping our readers will receive them with indulgence*.

We would willingly have entered into a more full illustration of the theory of Æpinus, and compared it with the phenomena noticed in CHAP. II. but the important paper just given has taken up so much room, that this article is already extended to very nearly the utmost limits assigned to it. We must, therefore, content ourselves with giving some idea of the induction of magnetism by juxtaposition according to Æpinus's hypothesis, and must refer for the rest to his Tentamen Theorie Eletricitatis et Magnetismi, or to the abridgement of it in Van Swinden's work Sur l'Analogie de l'Electricité et du Magnetisme, tom. ii.

Let NAS (fig. 31), be a magnet, of which the part Induced next the north pole AN is overcharged, and let a bar magnetum of iron s B n be brought near the north pole of the magnet, so that their axes are in the same straight line, explained. Now, in this theory, the overcharged pole N acts on Fig 31. the iron only by its redundant fluid, for that part of the fluid which is merely sufficient to saturate the iron will repel the fluid in B as much as the iron in AN attracts it, and of course can produce no change in B. In the same way SA acts on B merely by its redundant iron. Now, were the fluid in s B n immovable, no sensible effect would be produced on it; but as it is supposed to be easily moveable, the redundant fluid in AN will have the effect of repelling it towards n, till the resistance met with there, added to its own tendency to diffuse itself uniformly, just balances the repulsion of AN. In the mean time, however, an attraction exists between the redundant iron in AS, and the fluid in B, by which the latter would be drawn from B n, and condensed in B r, the attraction opposing the repulsion above mentioned; but since AS is more distant from every point of B than AN from the same point, the redundant fluid will prevail, and on the whole the fluid will be condensed towards n, and rarefied towards r. The more diffused we suppose the fluid and iron in the magnet to be, the more removed will be the centres of effort of its poles from their extremities, the smaller will be the action, and the difference of action of AN and AS, and of course the smaller the condensation towards n, and the rarefaction towards r. From this we learn, that, according as the poles of a magnet are more counteracted, the greater will be its power of action; and as this is agreeable to observation, it gives additional credit to the hypothesis.

Now, we see that the piece of iron n B s is attracted in consequence of its fluid being repelled towards its remote extremity, and diffused something like the fluid in NAS. In this hypothesis magnetism is supposed to depend entirely on the diffusion of magnetic fluid. The iron B has become a magnet, and by having magnetism induced on it, is attracted by the magnet A. In a similar way we might explain the action of the magnet, if its south or deficient pole were presented opposite to B.

When the notion of a magnetic fluid was once entertained, it is not surprising that philosophers, reasoning from the analogy between electricity and magnetism, and the different effects arising from the south and north pole of a magnet, should be led to the idea of the magnetic fluid being compounded of two fluids. Accordingly the hypothesis of two magnetic fluids has long been a favourite on the continent, where it has been chiefly supported by Coulomb and Hauy. As the experiments and observations of the former philosopher entitle him to the highest respect, we shall here give a sketch of his theory of magnetism.

1. Coulomb admits of two magnetic fluids, one of which may be called the northern, and the other the southern fluid.

2. The particles of each of these two fluids are mutually repulsive of each other; that is, the particles of the fluid N mutually repel each other, and the particles of the fluid S repel each other.

3. There is a mutual attraction between the particles of one of these fluids and the particles of the other; or the particles of the fluid N attract and are attracted by the particles of the fluid S.

4. In the ordinary state of iron not magnetized, these two fluids are found mixed together, and hence a piece of ordinary iron under the usual circumstances exhibits no signs of magnetism.

5. In a magnetized body these two fluids are separated, and this separation takes place as soon as we begin to magnetize the body; one of the fluids N, retiring towards one extremity, and the other fluid S to the other extremity of the magnetized body.

6. The attraction and repulsion of two magnetic bodies, when they approach each other, is the result of the mutual action of the two fluids.

Suppose we have two needles A and B. If we make them approach each other on the side of the two poles of the same name, N or S, they will repel each other; but if they are made to approach on the side of different poles, as when the needle A presents its north pole to the south pole of the needle B, they will attract each other. Here there are four forces in action; 1. the fluid N of the needle A repels the fluid N of the needle B. 2. The same fluid N of the needle A attracts the fluid S of the needle B. 3. The fluid S of the needle A repels the fluid S of the needle B; and, 4. The fluid S of the needle A attracts the fluid N of the needle B. Now, if the extremity N of the needle A be very near the extremity S of the needle B, the mutual attraction between the two fluids N and S, will be stronger than the mutual repulsion between the two fluids N, N, and the two fluids S, S, and consequently the two needles will approach each other.

7. The attraction and repulsion of the two magnetic fluids is in the direct ratio of the masses, and in the inverse ratio of the distances.

This important part of the theory Coulomb deduces from a series of very delicate experiments made with his magnetic bars, similar to those by which he proved the same law to take place with respect to the action of the electric fluid. See Electricity, Part IV. chap. ii.

8. The magnetic fluid is entirely in the interior of magnetic bodies, for as the magnetic fluid moves with difficulty in the interior of a magnetic body, it cannot diffuse itself over the surface, which is the reason why filings of iron brought near a magnetic bar, remain attached to it.

9. Consequently magnetic bodies can have no magnetic atmosphere.

10. In a magnetic needle, the centres of magnetic action are near the extremities of the needle.

11. A magnetic needle being broken in any place, each of its parts is found to have two poles.

12. The forces which attract a needle towards one pole, are equal to those which draw it toward the other pole.

13. Magnetic bodies do not act on other bodies susceptible of magnetism, in any other way than by attraction or repulsion; for the magnetic fluid remains entirely within the interior of these bodies.

14. Magnetic attraction ought to be regarded as a particular power, analogous, however, to the power which we call universal gravitation, the only difference being, that gravitation acts very sensibly on all bodies, whereas magnetism acts most powerfully on iron.

15. This magnetic power or attraction is therefore a particular power produced neither by impulsion, nor by the action of any other fluid.

Though the instrument which is usually employed to measure the inclination of the magnetic needle is very mode simple in its construction, it is nevertheless liable to great errors, which in general arise from the almost absolute impossibility of placing the needle in all the positions it can take in equilibrium with regard to the effect of gravitation, that is to say, so that its centre of gravity may always exactly agree with the point on which it turns. When the dimensions are considerable, a new inconvenience arises from a degree of flexure, which, though scarcely sensible, is nevertheless productive of very great effects from the slightest displacement of the centre of gravity producing a combination of the power of gravitation with that of magnetism.

To obviate these difficulties, Citizen Coulomb, instead of endeavouring to ascertain immediately, as has been hitherto done, the direction of the magnetic needle in the vertical plane which passes through the magnetic pole, conceives the force of this pole to be decomposed or resolved into two others in the same plane, the one acting in a horizontal, and the other in a vertical direction. He determines separately the intensity of each of these last forces, and the result gives the direction in which the magnetic force acts, and which a needle governed singly by this force would take. Citizen Coulomb has proved, in the Memoirs of the Academy of Sciences for the year 1789, that the magnetic needle suspended by its centre of gravity is incessantly brought back to its true direction by a constant force at the same place and time. It thence follows, that by observing the number of oscillations made in a given time by a needle horizontally suspended, the ratio of the horizontal component part of the magnetic power with gravity may be obtained. As to the vertical component part, it is measured by determining with care the weight necessary to be added to the southern part of the magnetic needle, to maintain it in a perfectly horizontal position. That being done, if A and B represent the respective measures of the horizontal and vertical component parts of the magnetic power, \( \frac{B}{A} \) will be the tangent of the angle made by their result with the horizontal force, and consequently, it will be the inclination of the magnetic needle.

In the experiments made by Citizen Coulomb, the needle had the form of a right-angled parallelopipedon, very thin in proportion to its breadth, and always suspended so that its breadth was kept in a vertical plane. Let P represent the weight of the needle, l the half of its length, \( \lambda \) the length of a pendulum that performs its oscillations in the same time as the needle when it obeys the magnetic power in a horizontal plane. Coulomb then gives the formula \( \frac{P l^4}{3 \lambda^3} \) to calculate the momentum of the magnetic force referred to the arm of a lever of one millimeter in length. The length of the needle was 427 millimeters, its breadth 13, and its weight 88,753 milligrammes. It was suspended horizontally by a thread of silk in a box well closed, and it made 30 oscillations in 286 seconds, and by applying these data to the preceding formula, Coulomb found that the logarithm of the momentum of the horizontal magnetic force is 4.1740.

Coulomb having placed his needle in a clip, having knife edges, which rested on two cylinders of glass, in the manner of the beam of a balance, endeavoured first to bring it to an equilibrium in a horizontal situation coinciding with the magnetic meridian, by placing the edges in a proper manner, and when they were sufficiently near the point where the equilibrium took place, he completed it by the addition of small weights. He then reversed the poles of the needle by the magnetic touch, but without altering the position of the clip, and again bringing it to an equilibrium in this new state, the sum of the momenta of the additional weights placed in these two operations gave him the double of the momentum of the vertical component parts of the magnetic force, valued at 74467. The result of this force, and of the horizontal force, is inclined 68° 9'.

In repeating these operations three times, Coulomb found successively 68° 9', 68° 13', and 68° 11'. Though the differences of these results are very trifling, he thinks they are to be entirely attributed to errors in the observation; for he is assured they do not amount to so much. It is possible that the needle is subject to variations in the vertical similar to those which are known to take place in the horizontal plane.

Daniel Bernoulli contrived an ingenious dipping needle that may answer the purpose of an universal instrument for making accurate observations on the dip. It depends on the following principle. If a dipping needle be made by an ordinary workman, and balanced needle, with some care, so that when impregnated with magnetism, it may show nearly the true dip, and if it be touched, and the dip observed, then its magnetism destroyed, and its balance so altered, that without any magnetism it will take nearly the inclination of the true dip; and if it be then touched again, giving it the same polarity as it had before, it is evident that it will now approach very nearly to the true dip, since, by its want of perfect equilibrium, it was deranged only a few degrees from its proper direction. If the second observation of the dip should, from the inaccurate formation of the needle, differ considerably from the first, the operation must be repeated; and in this third observation there will very seldom be an error of more than half a degree.

Bernoulli's instrument is as follows. A very light graduated brafs circle EFG (fig. 32.) is fixed on one side of the dipping needle, so as to be concentric with its axis, and the whole is balanced with as much nicety as may be, before being impregnated. CD is a very light index fixed to the axis in such a manner as to turn on it with some difficulty. By this the equilibrium of the needle will be destroyed. If great care has been taken in forming the instrument, and if it has been balanced with great accuracy, it will, by the addition of the index, be made to settle so as to have the index perpendicular to the horizon, at whatever degree of the circle the needle may happen to point. As such accuracy, however, is scarcely to be expected, let the index be set to several different degrees of the circle, and note the inclination taken by the needle before being magnetized, corresponding to each position of the index, and let all these be written down. For example, let us suppose that when the index is at 50°, the needle inclines 46° from the horizon; if we observe at any place that the needle, after being magnetized, inclines 46°, when the index is at 50°, we may be sure that the former is the true magnetic dip at that place, as the needle is not deranged by the magnetism that has been given it, from the situation it would assume by gravity alone. We usually know something of the dip that may be expected at any place. If we set the index accordingly, and if the needle does not then point out the expected dip, change the position of the index, and again observe the dip; examine whether this second position of the index and the second dip form a corresponding pair of numbers, such as we have written down; if they do, we have got the true dip, but if not, another position of the index must be tried. Thus, by noticing whether the agreement of this last pair be greater or less than that of the former pair of numbers, we learn whether we are to change the position of the index in the same or in the opposite direction.

A close analogy has long been remarked between the phenomena of magnetism and those of induced electricity, especially those of attraction and repulsion. The mechanical composition of these actions produces a directive and magnetic power and polarity, both in electrical and magnetic bodies. It is easy to form an electrical needle that will arrange arrange itself with respect to the overcharged and undercharged ends of a body electrified by position, just as a magnetic needle arranges itself with respect to the magnet. A stick of sealing wax may be touched in a manner similar to the double magnetic touch, so as to acquire poles of considerable force, and very durable. Again, melted sealing wax, when cooled in the neighbourhood of a positive and negative electric, acquires permanent poles, just as a red-hot steel bar acquires them by being quenched near a magnet. Lastly, lightning sometimes gives polarity to needles, sometimes destroys it, and sometimes reverses their polarity.

From these various circumstances of resemblance, some have supposed that both phenomena originate from the same cause, but there are several circumstances which show their original causes to be different. Thus, we find that electricity is common to all bodies, and can be excited or induced on all in a degree that is pretty nearly equal. Magnetism, on the contrary, though from Coulomb's experiments, it appears in some degree to affect all terrestrial bodies, acts, however, very imperceptibly on all but iron and its compounds. The action of lightning must not be considered as a proof of their identity, since that is accompanied with a great degree of heat, and we have already seen that this power, under favourable circumstances, is a very active agent, both in producing and destroying magnetism. Again, there is nothing in magnetism like a body being entirely overcharged, or entirely undercharged, as in electricity; but a magnetic body having two poles, must always be overcharged at one extremity, and undercharged at the other. There is nothing in magnetism resembling that inconceivably rapid motion which we see in electricity. In fine, the only perfect resemblance is between the induced magnetism of common iron, and the induced electricity of a conductor. On the arguments that have been employed for and against the identity of magnetism and electricity, our readers may consult Van Swinden, Sur l'Analogie de l'Electricité et du Magnetisme, and a tract by Æpinus De Similitudine Electricitatis et Magnetismi.

Some late experiments of Ritter tend to show a greater analogy than has yet been supposed, between magnetism and that modification of electricity which we call galvanism.

Mr Ritter's first experiments with the magnet were on frogs. He found that a magnetic iron wire, with another not magnetic, excited a galvanic palpitation in these animals. Presently he observed, that the south pole excited stronger palpitations, and the north pole weaker, than the iron not magnetic. Having constantly noticed, that the metals most susceptible of oxidation excited the strongest palpitations, he inferred, that the south pole possesses a greater affinity for oxygen than simple iron, and the north pole less.

This supposition he confirmed by means of several chemical reagents. He placed a magnetic iron wire on pieces of glass in a plate of earthen ware, and poured upon it a very weak nitric acid. The south pole was attacked by the acid much more powerfully than the north; and was soon surrounded by a deposition of oxygen, the quantity of which greatly exceeded that of the other pole.

The different oxidability of the magnetic pole is very well exhibited likewise, by taking three small bottles of equal size, filled with water, either pure or slightly acidulated, and putting into one the fourth polar end of a magnetic wire, into a second the north polar end of a similar wire, and into the third the end of an equal wire not magnetic; the south pole will first begin to deposit oxide, the unmagnetic iron a little after, and the north pole last. This experiment requires considerable care. The surface of the water must be covered with very fresh oil of almonds, to exclude all access of air. Care must be taken too, that one of the bottles is not more exposed to the sun than the others, because light accelerates oxidation. Ritter convinced himself of this by direct experiments; exposing two iron wires in water to the sun, but covering one of the phials with black paper, when that in the phial left uncovered was oxidated much more quickly.

If infusion of litmus be substituted instead of the water in the three phials in the preceding experiment, the relative oxidations will be the same, but they will be attended with a change of colour, showing that an acid is produced proportional to each oxidation; so that the fourth pole not only undergoes the greatest oxidation, but likewise reddens the infusion of litmus most. The action that takes place in this experiment is very feeble, and frequently requires a week to produce a distinct effect; and indeed to accelerate it so much as this, it is necessary to add, previously to the infusion, as much acetic acid as will incline it to red, without completely changing its colour. The infusion reddened in this experiment resumes its blue colour on exposure to the air; but we must not hence conclude, that the acid produced by the action of the magnet is very volatile, for infusion of litmus reddened by phosphoric acid, or any other, exhibits the same phenomenon.

The following experiment exhibits some things peculiar, and therefore we shall give it more at large. It has not been repeated, but the harmony of its results is in favour of its accuracy. Sixteen magnetic wires, of equal size and power, were placed in fix vessels, all equally full of a mixture of one part nitric acid, and 36 parts water, in the following manner: in the first glas were placed two wires, one with the north pole immered in the fluid, the other with the fourth, and not more than half a line afarnder: in the second, the same, but the wires an inch and three-fourths apart: in the third and fourth were each three wires, with the south poles of all immersed, but their distances in the two glasses different, as in the first and second: in the fifth and sixth were wires similarly arranged, but with the north poles immered. Different quantities of oxide were gradually depoed, and to express the whole in few words, we will call the fourth pole S, the north pole N, their greater distance g, and their lefs p, and we will express the order of oxidations as follows: SNg→SNp→3 Sp→3 Sg→3 Np→3 Ng→. On the nineteenth day it was observed, that the loss of fluid by evaporation had not been equal in all the vesseles, but took place in the inverse order of the oxidations. All the magnetic wires were weakened in power; NSg left; NSp more: of the wires 3 Sp, two had lost lefs power than the third; and in like manner 3 Sg, 3 Np, 3 Ng, had each two left more powerful than the third; the strongest were equal to NSg.

PLATE CCXCVIII. 2d

Fig. 1. Fig. 2. Fig. 3. Fig. 4. Fig. 5. Fig. 6. Fig. 7. Fig. 8. Fig. 9. Fig. 10. Fig. 11. Fig. 12. Fig. 13.

W. Archibald Sculp.

PLATE CXCIX.

Fig. 14. Fig. 15. Fig. 16. Fig. 17. Fig. 18. Fig. 19. Fig. 20. Fig. 21. Fig. 22. Fig. 23.

W. Archibald sculp.*

PLATE CCC.

Fig. 24. S A N C H D

Fig. 25. E D B C F m m a a m m

Fig. 26. S A N B n n

Fig. 27. M R L K F I G E H N D S N B N

Fig. 28. O H L C N

Fig. 29. E E L H

Fig. 30. A C B P D E M

Fig. 31. S A N S B

Fig. 32. F G A e O B E In another experiment, where two little vessels filled with infusion of litmus were employed, one of them containing two magnetic wires, the fourth poles of which were immersed in the fluid; the other two similar wires, of which the opposite poles were immersed; the oxidation was greatest in the latter vessel.

The analogy between galvanism and magnetism is still farther proved by other experiments of Ritter on galvanizing metals, which he does by placing them in a stream of galvanic fluid proceeding from a strong pile. He found that a golden needle thus galvanized and balanced on a pivot, exhibited, like a magnetized iron needle, both directive power and horizontal inclination.

Some late experiments of Ritter, referring still more directly to the analogy between magnetism and galvanism, were communicated to the Royal Academy of Sciences of Munich, and the following are their general results.

1. Every magnet is equivalent to a pair of heterogeneous metals united together; its different poles represent as it were different metals.

2. Like them, it gives electricity; that is to say, one of the two poles, the positive electricity, and the other the negative.

3. By following the same process a certain number of magnets, as well as a certain number of pairs of metals, afforded electricity; and in this manner the electricities afforded by the poles of different magnets, have been successfully indicated by the electrometer.

4. By means of these electricities, one of these batteries of magnets, accordingly as it is more or less strong, produces upon dead and living bodies, all the phenomena which are produced by a pile of Volta, of the common kind, and of the same force.

5. The experiments which prove this, show, that in magnetized iron, the south pole gives positive electricity, and the north pole negative electricity; but that on the contrary in magnetized steel, the north pole affords the positive, and the south pole the negative.

6. The same inverse disposition is also observed with regard to the polar oxidability of the magnetized body in which this change is produced by magnetism. In magnetized iron the south pole is most oxidable, and the north pole least; whereas in magnetized steel the north pole is most oxidable, and the south least.

7. Mr Ritter thinks, that by considering the earth as an immense magnet, these results might serve to explain various phenomena of nature, such as the physical difference between the two hemispheres, the aurora borealis and aurora australis. In fact, after what has been just stated, the earth considered as a magnet, may be taken as an equivalent to an immense pile of Volta, of which the poles are on one side sufficiently closed by the waters of the ocean. And the action of this pile must produce, and has produced the greatest chemical changes, in the materials of the earth; changes which must have differed according to the poles; and of which pile the poles at the other extremity have always such an abundance of electricity as to cause its splendour to appear by radiations in the vast spaces of the heavens.*

* Nicholson's Journal, xv. 78.

The foregoing experiments appear to prove that magnetism has some effect in producing chemical changes, and thence we may infer that perhaps it would not be altogether inactive in the animal economy.

INDEX.

A. Action, magnetic, law of, No 80, 89 investigated by Lambert, 83 by Robison, 84 Spinus's method of touching bars, 55 theory of magnetism, 79 incients, opinions of, respecting the cause of magnetic attraction, 70 apparatus, magnetical, p. 362 nature of magnets, No 13 attraction, magnetic, p. 368 between iron and the magnet neutral, No 27 attraction, magnetic, not sensibly diminished by interposition of non-ferruginous bodies, 28 how usually measured, 29 increased by iron, 31 by an increased weight, 32 explained, 85 vis of a magnet, 7 Bernoulli's dipping needle, 105

B. Biot's observations on terrestrial magnetism, No 95

C. Canton's method of touching bars, 54 of making artificial magnets, 61 Communicative piece of money, 39 Cotte's axioms respecting the magnetic declination, p. 366 Coulomb's experiments on universal magnetism, No 36 method of making artificial magnets, 57 theory of magnetism, 103 method of finding the magnetic dip, 95

D. Declination, magnetic, varies, 19 tables of, p. 364 axioms respecting, No 366 Dial, magnetic, No 42 Dip, magnetic, tables of, p. 367 diminishes as we ascend above the earth, 366

E. Dip, magnetic, Coulomb's method of ascertaining, No 95 Dipping of the magnet, needle, 4 Bernoulli's, 105 Directive power, what, 2 explained, 87 Divining circles, 43 Duhamel's method of touching bars, 53

Earth, magnetism of, first rationally explained by Gilbert, illustrated, 71 objection answered, 94 Electricity and magnetism, analogy between, 106 Equator of a magnet, magnetic, determined by Humboldt, 98 Euler's theory of magnetism, 77 Experiments illustrating magnetic polarity, on magnetic attraction, 363 repulsion, No 34 entertaining, 37 Experiments,

Experiments, Ritter's, No 107 F. Forces, magnetic, picture of, 85 G. Galvanism and magnetism, analogy between, 107 Gilbert's theory of magnetism, 71 proofs of, 75

H. Hawkbee's attempts to investigate the law of magnetic action, 82 Hindbaw's experiment illustrating terrestrial magnetism imitated, 75 Humboldt's observations on terrestrial magnetism, 95 determination of the magnetic equator, 98

I. Inclination of the needle increases as we proceed from the magnetic equator. See Dip. Induced magnetism, what, gradual, 46 Iron naturally arranges itself in a determinate manner, 16 polarity of, temporary, 17 attracted by the magnet, 22 filings, action of a magnet on, 26 attracts iron, 39 becomes magnetical by proximity to a magnet, 44 by position, 58 by hammering or friction, 59 by heat, 60 attracted only because it becomes magnetical by induction, 91 filings, arrangement of, explained, 92

K. Knight's artificial lodestones, 64

L. Lambert's investigation of the law of magnetic action, 83

M. Magnet, what, 1, 12

Magnet, declination of, No 4 dipping of, 3 artificial, 6, 13 modes of making, 51—64 axis and equator of, 7 armature of, 13 polarity of, permanent, 18 poles of, how found, 23 attractive power of, varies, 24

Magnetism, general idea of, 1 utility of, 9 works on, list of, 10 how applied to use, p. 366 how usually discovered, No 35 universal, Coulomb's experiments on, 36 communication of, p. 373 induced, what, No 46 artificial, produced by touching with a magnet, 50 artificial, produced by position, 58 artificial, produced by friction, 59 artificial, produced by heat, 60 impaired by improper position, 66 by heat, 67 by rough usage, 68 by opposition of similar poles, 69 theories of, p. 380 terrestrial, of Gilbert, No 71 observations on, by Biot and Humboldt, 95 Marcel's method of touching bars, 62 Mining, use of the compass in, p. 366 Muichenbroeck's attempt to investigate the law of magnetic action, No 81

N. Needle, magnetic, 14 dipping, 15

P. Perspective glafs, magnetic, 38 Polarity, magnetic, what, 2 disturbed by the approach of iron, 20

Polarity, magnetic, affected by the atmosphere, contrary, induced on iron by a magnet, Poles of magnet, how found, contrary, attract each other, corresponding, repel each other, Power, magnetic,

R. Ritter's experiments on the analogy of magnetism and galvanism, Robison's investigation of the law of magnetic action, explanation of magnetic attraction, of directive power,

S. Steel, soft, how magnetized,

T. Table, magnetic, Taylor's attempts to find the law of magnetic action, Terrestrial magnetism, observations on, by Humboldt and Biot, magnetism acts on the whole surface of the earth, magnetism increases from the equator to the poles, magnetism modified by local circumstances,

Theory of Gilbert, Euler, Æpinus, Coulomb, Touch, double, what, Touching magnetic bars, old methods of, of curved bars, how improved by Duhamel, by Michell and Canton, by Æpinus, by Coulomb,

W. Watch, mysterious,

Animal Magnetism, a sympathy supposed by some persons to exist between the magnet and the human body; by means of which the former, it was thought, possessed the property of curing many diseases.

The notion of animal magnetism appears to have originated, in 1774, with a German philosopher named Father Hehl, who greatly recommended the use of the magnet in medicine. M. Mefner, a physician of the same country, by adopting the principles of Hehl, became the direct founder of the system; but afterwards deviating from the tenets of his instructor, he lost his patronage, as well as that of Dr Ingenhouz, which he had formerly enjoyed. Mefner had already distinguished himself by "A dissertation on the influence of the Stars upon the human body," which he publicly defended in a thesis before the university of Vienna; but he he was so unable to stand before the opposition of Hehl and Ingenhousz, that his system fell almost instantly into disrepute. Mesmer appealed to the Academy of Sciences at Berlin; but they rejected his principles as destitute of foundation, and unworthy of the smallest attention. He then made a tour through Germany, publishing everywhere the great cures he performed by means of his animal magnetism, while his enemies everywhere pursued him with detections of the falsehood of his assertions.

Mesmer, still undaunted by so many defeats, returned to Vienna; but meeting there with no better success than before, he retired to Paris in the beginning of the year 1778. Here he met with a very different reception. He was first patronized by the author of the Dictionnaire des Merveilles de la Nature; in which work a great number of his cures were published, Mesmer himself receiving likewise an ample testimony of his candour and solid reasoning. Our physician soon collected some patients; and in the month of April 1778 retired with them to Creteil, from whence he in a short time returned with them perfectly cured. His success was now as great as his former disappointment. Patients increased so rapidly that the doctor was soon obliged to take in pupils to assist him in his operations. These pupils succeeded equally well as Mesmer himself; and so well did they take care of their own emolument, that one of them named M. Delon realized upwards of 100,000l. sterling. In 1779 Mesmer published a memoir on the subject of Animal Magnetism, promising afterwards a complete work upon the same, which should make as great a revolution in philosophy as it had already done in medicine.

The new system now gained ground daily; and soon became so fashionable, that the jealousy of the faculty was roused, and an application concerning it was made to government. In consequence of this a committee was appointed to inquire into the matter, consisting partly of physicians and partly of members of the Royal Academy of Sciences, with Dr Benjamin Franklin at their head. This was a thunderstroke to the supporters of the new doctrine.—Mesmer himself refused to have any communication with the committee; but his most celebrated pupil Delon was less scrupulous, and explained the principles of his art in the following manner:

1. Animal magnetism is an universal fluid, constituting an absolute plenum in nature, and the medium of all mutual influence between the celestial bodies, and betwixt the earth and animal bodies.

2. It is the most subtle fluid in nature; capable of a flux and reflux, and of receiving, propagating, and continuing all kinds of motion.

3. The animal body is subjected to the influences of this fluid by means of the nerves, which are immediately affected by it.

4. The human body has poles and other properties analogous to the magnet.

5. The action and virtue of animal magnetism may be communicated from one body to another, whether animate or inanimate.

6. It operates at a great distance without the intervention of any body.

7. It is increased and reflected by mirrors; communicated, propagated, and increased by sound; and may be accumulated, concentrated, and transported.

8. Notwithstanding the universality of this fluid, all animal bodies are not equally affected by it; on the other hand, there are some, though but few in number, the presence of which destroys all the effects of animal magnetism.

9. By means of this fluid nervous disorders are cured immediately, and others mediately; and its virtues, in short, extend to the universal cure and preservation of mankind.

From this extraordinary theory, Mesmer or M. Delon, had fabricated a paper, in which he stated that there was in nature but one disease and one cure, and that this cure was animal magnetism: and, lastly, M. Delon engaged, 1. To prove to the commissioners, that such a thing as animal magnetism existed; 2. To prove the utility of it in the cure of diseases; after which he was to communicate to them all that he knew upon the subject. The commissioners accordingly attended in the room where the patients underwent the magnetical operations. The apparatus consisted of a circular platform made of oak, and raised about a foot and a half from the ground; which platform was called the baquet. At the top of it were a number of holes, in which were iron rods with moveable joints for the purpose of applying them to any part of the body. The patients were placed in a circle round, each touching an iron rod, which he could apply to any part of the body at pleasure; they were joined to one another by a cord passing round their bodies, the design being to increase the effect by communication. In the corner of the room was a piano forte, on which some airs were played, occasionally accompanied with a fong. Each of the patients held in his hand an iron rod ten or twelve feet long; the intention of which, as Delon told the commissioners, was to concentrate the magnetism in its point, and thus to render its effects more sensible. Sound is another conductor of this magnetism; and in order to communicate the magnetism to the piano forte, nothing more is necessary than to bring the iron rod near it. Some magnetism is also furnished by the person who plays it; and this magnetism is transmitted to the patients by the sounds. The internal part of the platform was said to be so contrived as to concentrate the magnetism, and was the reservoir whence the virtue diffused itself among the patients. Its structure, however, is not mentioned; but the committee satisfied themselves, by means of a needle and electrometer, that neither common magnetism nor electricity was concerned.

Besides the different ways of receiving the magnetism already mentioned, viz. by the iron, cord, and piano forte, the patients also had it directly from the doctor's finger, and a rod which he held in his hand, and which he carried about the face, head, or such parts of the patient as were diseased; observing always the direction of what he called the poles. The principal application of magnetism, however, was by pressure of the hands or fingers on the hypochondria or lower regions of the stomach.

The effects of these operations upon Delon's patients were very different. Some felt nothing, neither had the magnetism any effect whatever upon them. Some Some spit, coughed, sweat, and felt, or pretended to feel, extraordinary heats in different parts of the body. Many women, but very few men, had convulsions, which Delon called their crisps, &c.—The commissioners at last found that they could come to no satisfactory conclusion while they attended in this public way, and therefore determined to try the experiments themselves privately. As the fluid itself, however, was totally imperceptible by any of the senses, they could only ascertain themselves of its existence by ultimately curing diseases, or by its observable effects upon the human body. Being well assured, however, that though many diseases were cured, it would not amount to any proof of the existence of animal magnetism, they determined to observe its effects on the animal economy. For this purpose they made the following experiments:

1. They tried it upon themselves, and felt nothing.

2. Seven of Delon's patients were magnetized at Dr Franklin's house, four of whom felt nothing; three felt, or affected to feel, something.

3. Several persons in a higher sphere of life were magnetized, and felt nothing.

4. The commissioners, now determined to discover what share imagination had in this business, blindfolded several of the common people, and made them sometimes think that they were magnetized, at other times they magnetized them without letting them know that they did so; the consequence was, that when they supposed themselves magnetized, the patients likewise thought they felt something, and vice versa.

5. A magnetized tree was said to produce convulsions; a young man, blindfolded, fell into convulsions when he imagined himself near the tree, though he was really at a considerable distance from it. Delon accounted for this on the principle of all trees being magnetic: but in this case, every one, susceptible of magnetism, would be seized with convulsions when he approached a tree. The same influence of imagination was observed in a woman accustomed to have convulsions when magnetized. They came on when nothing was done to her, on being told, when blinded, that she was magnetized.

Other instances are given, from which it was evident, either that the patients were impostors, or in such a most wretched state of debility both of mind and body, that the most trifling effects of the former had the most powerful effects on the latter. The commissioners therefore entirely disapproved of the whole. The touch, imitation, and imagination, they concluded, were the great causes of the effects produced by M. Delon's operations; and by means of these they supposed, that convulsions, which in themselves are a very violent disorder, might be spread much farther than could be wished, even through a whole city. It was observed that the operator sometimes pressed strongly, and for a length of time, upon different parts of the body, particularly the hypochondria and pit of the stomach; and it is well known that a strong pressure on these parts will produce disagreeable sensations in those who enjoy perfect health.

It is needless to add more upon this subject, than that Mesmer complained of the report of the commissioners, petitioned parliament, was by them commanded to discover the mysteries of his doctrine; and that it is now exploded by every man of sense.—The conclusion of the academicians concerning it was, that it is not entirely useless even to philosophy; as it is one fact more to be configned to the history of the errors and illusions of the human mind, and a signal instance of the power of imagination.