TWO EARLY ARABIC SOURCES

ON THE

MAGNETIC COMPASS

 

Petra G. Schmidl

 

Frankfurt

 

In this paper two previously unpublished texts on the magnetic compass from the medieval Islamic world will be discussed, the first by the Yemeni Sultan al-Ashraf (ca. 1290) and the second by the Cairene astronomer Ibn Simʿūn (ca. 1300). These two treatises constitute the earliest known evidence attesting the use of the magnetic compass for the determination of the qibla, the sacred direction of Islam. A brief introduction glimpses at the history of the magnetic compass in Europe and China and mentions previously known early Arabic sources on the instrument and its use. This is followed by some remarks on the authors and the manuscripts, the Arabic texts with English translations, and comments on problems encountered while working on the texts.

 

1. Survey of the history of the magnetic compass

The Yemeni astronomer-prince al-Ashraf (ca. 1290) and (apparently) also a Cairene astronomer called Ibn Simʿūn (ca. 1300)[1] both wrote treatises on the magnetic compass. These two previously unpublished texts will be presented in this paper, prefaced by a brief survey of the knowledge on and use of the magnetic compass in the European and Islamic Middle Ages in order to put these new sources in the context of previous knowledge on the subject.

Although the magnet and its attractive property were known in Antiquity, there is no mention of its directive potential in the sources.[2] In the late nineteenth [82] and early twentieth century the so-called south-pointing chariot was understood as an early indication of the use of the magnetic compass in China, but it seems to have been a mechanical device to maintain a specific direction, for the earliest reference to the magnetic compass in China dates from the eleventh century.[3] The first known written reference to a magnetic compass in the Occident dates back to the year 1187, when Alexander Neckam reported the use of a magnetic compass for the region of the English Channel.[4] In 1269 Petrus Peregrinus of Maricourt, in his well-known Epistola de magnete, described a floating compass for astronomical purposes as well as a dry compass for seafaring.[5] For the former instrument a magnetic [83] stone is enclosed in a small waterproof wooden box and put in a bowl filled with water, the rim of which is divided into 360 degrees. In addition, this box carries a sort of alidade in the form of a diametrical rule, for taking bearings.[6] For the latter, two needles of copper and iron fixed crosswise on one another are put into an axle with two holes perpendicular to the axis and perpendicular to each other (probably one slightly higher than the other). This device is squeezed into a small round box between the bottom and the transparent lid in such a way that it can rotate freely. The lid is divided into 360 degrees, and in addition a kind of alidade is attached to it. The iron needle is magnetized by bringing it close to a magnetic stone. In Europe and in China, a third use of the magnetic compass was the alignment of sacred architecture.[7] Further information is provided by the earliest European instruments fitted with a magnetic compass, but they date only from the second half of the fifteenth century.[8]

[84] Of concern for us here is a part of the area between China and Europe, namely, the world of Islam. Eilhard Wiedemann, whose essays on the history of Islamic science and technology are a gold-mine of information, greatly contributed to our understanding of the history and use of the magnetic compass in Islamic civilization. The earliest secure evidence attesting knowledge of the magnetic compass[9] is found in the Persian anthology Jāmiʿ al-ikāyāt by Sadīd al-Dīn Muḥammad ibn Muḥammad Bukhārī, known as ʿAwfī, where an event during a voyage in the Red Sea or the Persian Gulf in the year 630H (123233) is related:[10] a fish made of iron is rubbed with a magnetic stone and then put in a bowl filled with water; it rotates until it stops pointing to the south.[11] The first full description of the [85] use of the magnetic compass for nautical purposes in the Islamic world is presented by Baylak al-Qibjāqī in his Kitāb Kanz al-tujjār fī maʿrifat al-aḥjār written in 681H (1282).[12] There he describes the use of a floating compass during a sea voyage from Tripoli in Syria to Alexandria in the year 640H (124243).[13] An iron needle is joined crosswise with a rush and put in a bowl filled with water. Then a magnetic stone is brought close to this device, and the hand holding the magnetic stone describes a circle clockwise above it. The cross of the needle and the rush follows this move. When the magnetic stone is suddenly removed, the needle is supposed to be aligned with the meridian. In addition al-Qibjāqī reports that on the Indian Ocean floating compasses with a hollow floating fish made of sheet‑iron were used.[14]

Further references to the magnetic compass in the Islamic world during the Middle Ages and the early Renaissance can be found in Taqī al-Dīn al-Maqrīzīs al-Mawāʿiẓ wa-ʾl-iʿtibār fī dhikr al-khiṭaṭ wa-ʾl-athār, a topography of Fusṭāṭ and Cairo and history of Alexandria and Egypt in general [86] written about 800H (1400);[15] in Ibn Abī ʾl-Khayr al-usnīs al-Nujūm al-shāriqāt, a technical cookery-book on the production of colors and inks, the process of soldering, the handling of iron, etc., probably written in the second half of the sixteenth century;[16] and in treatises on navigation by Ibn Mājid from the second half of the fifteenth century[17] and by Sulaymān al-Mahrī from the first half of the sixteenth century.[18] Particularly important in this context is the treatise Zahr al-basātīn fī ʿilm al-mashātīn written by the Egyptian Muḥammad ibn Abī Bakr al-Zarkhūrī about 802H (13991400),[19] where two different kinds of magnetic compass are described. One instrument is a fish made of willow wood or pumpkin, into which a magnetic needle is inserted and which afterwards is sealed with tar or wax to prevent the penetration of water; this device swims on water. Instead of the fish a disk with the image of a miḥrāb or prayer-niche can be fixed on the magnetic needle.[20] The other instrument is a dry compass. The upper side of a disc of paper is decorated with the figure of a miḥrāb; two magnetized needles are fixed on the bottom and in the middle a thing like a funnel. This funnel rotates on an axis which is pivoted in the middle of a box sealed with a plate of glass to prevent the disc of paper from dropping.[21]

[87] Several kinds of Islamic instruments featuring a magnetic compass were unknown to Eilhard Wiedemann and should be mentioned here. The ṣandūq al-yawāqīt made by ʿAlāʾ al-Dīn Ibn al-Shāṭir, an astronomical compendium or multifunctional device seems to be the earliest Islamic instrument fitted with a dry magnetic compass. Only one device made by Ibn al-Shāṭir himself in 767H (1366) survives.[22] Further instruments fitted with a magnetic compass to orient them in the cardinal directions are the dāʾirāt al-muʿaddal, equatorial (semi-)circles. Devised in Egypt in the fifteenth century, they were used in the Ottoman world.[23] Of greater historical interest is a ceramic bowl for a floating compass made in Damascus, datable about 1520. The inscriptions on this piece include qibla values for 40 cities, and prove that it belongs to a Persian tradition that predates this particular instrument by at least two centuries.[24] Further Islamic instruments fitted with a magnetic compass are known, but all of them postdate 1600.[25] Summing up, it may be said that the previously known Islamic sources on the magnetic compass first mention the floating compass, followed by the dry compass. Originally magnetic compasses are described primarily as nautical instruments, later as qibla indicators or components of astronomical instruments.[26]

[88] Against this background we now introduce two new textual sources on the magnetic compass; for each the Arabic text with a translation will be presented, followed by a commentary.

 

2. Al-Ashrafs treatise

Al-Ashraf ʿUmar ibn Yūsuf was the third Rasulid sultan of the Yemen, reigning from 694H (129495) to his death in 696H (129697). He followed his father, the sultan al-Muẓaffar Yūsuf on the throne.[27] He was of minor importance for the political history of his realm, but is of considerable importance to the history of science. According to the statement of his teachers,[28] he wrote several important and sophisticated scientific works,[29] including one on the construction of astrolabes and sundials, and, in addition, he constructed several astrolabes, one of which is preserved in the Metropolitan Museum of Art in New York.[30] At the end of his treatise, we find two short statements, one on the water-clock and one on the magnetic compass and the determination of the qibla.[31]

[89] Three complete manuscripts of al-Ashrafs treatise on the magnetic compass are known, as well as a manuscript containing merely a part of the chapter on the magnetic compass.

The most significant manuscript is preserved in the Egyptian National Library (Dār al-Kutub al-Miṣrīya) in Cairo with the signature TR (Taymūr riyāḍa105 henceforth labeled C.[32] It was copied in the Yemen in 692H (1293).[33] The text bears the title Muʿīn al-ṭullāb fī ʾl-ʿamal bi-ʾl-asṭurlāb, which is probably not original. Altogether it has 149 folios;[34] the section on the magnetic compass Dhikr risālat al-ṭāsa, which interests us here, begins on fol. 143r and ends on fol. 146r.

Another manuscript preserved in the library of the Majlis al-Umma al-Īrānī (No. 150) in Tehran,[35] consists of three parts, each being separately paginated, and written in two different hands.[36] It contains the chapter on the magnetic compass twice, henceforth abbreviated as T1 and T2. The first part of the Tehran manuscript is datable about 900H (1500).[37] The Dhikr risālat al-ṭāsa (T2) is contained in pages 3 to 8. The third part of the Tehran manuscript is dated 7 Dhū ʾl-Qaʿda 888H (7 December 1483), and bears the title Minhaj al-ṭullāb fī ʾl-ʿamal bi-ʾl-asṭurlāb,[38] which is probably not original.[39] The Dhikr risālat al-ṭāsa (T1) is contained in pages 159 to 163.[40]

[90] In a manuscript in Berlin (Staatsbibliothek Stiftung Preußischer Kulturbesitz, Orientabteilung, Ahlwardt Nr. 5811, 2, Sprenger 1870, 1), an anonymous addition to the treatise on the magnetic compass is found, henceforth abbreviated as B. It dates from the year 1114H (17023). The complete manuscript of 19 folios has the title Risālat asṭurlāb, but this has been added later; the text contains a treatise on the astrolabe.[41] The text on fol. 12a deals with the compass bowl under the heading ʿAmal al-ṭāsa li-maʿrifat ikhrāj al-qibla wa-ʾl-jihāt. A German translation was published by Wiedemann,[42] and it was his translation that enabled me to identify this fragment as being al-Ashrafs work. Neither Ahlwardt nor Wiedemann were aware of the origin of the treatise, nor King of the manuscript.

The basis of the following Arabic text and the translation is C. It is written in a clear and clean naskhī hand. Moreover it is the earliest and most complete copy and is without noticeable mistakes in content. Hamzas and diacritical points missing in the original have been silently added in my edition. Occasional differences between T1, T2 and B are noted in the critical apparatus. The punctuation of the translation follows the English usage. Additions facilitating understanding, but without equivalents in the texts, are in parentheses. Uncommon Arabic terms are likewise given in parentheses. These are also included in the glossary.

 

2.1. Arabic text

 

 1  1 2 .

v 3 . 4   4 v  6 v 7   8 v 9   10  11  12 . :

v 13 v . (144r) v 14 [91]  .  15  15  .  16  .  17  17        .                .  18  19            .   (144v)  20  21  22  23  .

[92] 24  25  25  26  27  26 .  28  29  30    31  .  32  32    33  34      []    35 -|36 .  37  38    39  40  41     [  ]    42  43    44  .  45  46  46    47 [  ]  48 .  49  .

 50 (145r) [93]          .    . ][  51  .  52  53    .  54  54    55  . [  ]  56   57  57    ][    49  58  [94]   [  ] (145v)  59 .  60  61 v 61 :

 

The diagram with the inscription follows. There are three circles; the outer circle is divided into four parts marked: al-janūb (above), al-maghrib (right), al-shimāl 62 (below), al-mashriq (left), qiblat Taʿizz63 (from north to west near 20), qiblat ʿAdan64 (from north to east near 20). The middle circle is divided into 72 equal parts marked for every five degrees. The innermost circle is divided into 360 parts marked for every degree with a short stroke. In the center are two lines from south to north and from east to west marked: wuqūf al-samāra fī ʾl-ṭāsa65 muqābila khaṭṭ al-mashriq wa-ʾl-maghrib (by the east-west line), wuqūf al-ibra fī ʾl-ṭāsa muqābila khaṭṭ niṣf al-nahār (by the north-south line).

 

 66  6667     68      (146r)    69  .    . 5    5 . 5    .       5 70   71 .

[95] 72  72 73  74v v  75  . v 76 .  77  77 v  v 78  v 79 . v 80   81v v  82 v v .  83.

  84  85 86 .

 

Notes on al-Ashrafs Text

 

1-1 illegible in T2

2 v follows in T1

3 missing from T2

4 in T1

5 missing from T2

6 follows in T1, T2

7 missing from T1, T2

8 in T1, T2

9 missing from T2

10 in T1, T2

11 in T2 [96]

12 in T1, T2

13 missing from T1, T2

14 missing from T2

1515 - illegible in T2

16 in T1, T2

1717 - missing in T1, T2

18 repeated in T1, T2

19 in C, in T1, T2

20 v follows in T1, T2

21 v follows in T1, T2

22 v follows in T1, T2

23 v follows in T1, T2

24 B begins, entitled .

2525 - missing in B

2626 - missing in B

27 T1, T2 (not punctuated)

28 marginal note in C; in B

29 in T1, T2, B

30 in B

31 in B

32 illegible in T2

33 missing from B

34 in T1, T2

35 missing from B

36 illegible in T2

37 in B

38 missing from B

39 in B

40 (not punctuated) in T2 [97]

41 in B

42 follows in T1

43 in B

44 in B

45 in B

4646 - missing in B

47 (not punctuated) in B

48 B ends.

4949 - missing in the copy of T1 at the authors disposal

50 illegible in T2

51 crossed out, then repeated in C

52 in T2

53 follows in T2 (cf. previous sentence)

5454 - illegible in T2

55 in T2

56 in T2

5757 - illegible in T2

58 (not punctuated in) T2

59 in T1, T2

60 below the line T2

61 missing from T2

62 near it T1, T2

63 mark not inscribed T1, T2

64 in T1, T2

65 in T2

6666 - illegible in T2

67 illegible in T2

68 illegible in T2 [98]

69 missing from T1, T2

70 illegible in T2

71 v follows in T1

7272 illegible in T2

73 v follows in T1, T2

74 marginal note in C

75 missing from T1, T2

76 missing from T2

77 in T1; illegible in T2

78 missing from T1, T2

79 missing from T1, T2

80 missing from T2

81 in T1

82 (first letter not punctuated) in T1

83 colophon follows in T2 (illegible in the copy of T2 at my disposal)

84 over the line in T1

85 follows in T1

86 follows in T1

 

2.2. Translation

On the Use of the Compass Bowl (ṭāsa)[43] for the Determination of the Qibla.[44] In the name of God, the Merciful and Compassionate, praise be to God, the Compassionate and Merciful,[45] the Giver and Benefactor. May His [99] blessings be upon His Prophet to whom He revealed the Koran, and His most excellent salvation and forgiveness upon his family.

This is a treatise presented in the clearest possible terms on the knowledge of the use of the needle (ibra)[46] and the direction (inḥirāf) of the qibla for all localities. I seek help from God [144r] and place my trust in him. I rely on Him in that which is obscure and difficult. You begin by making a bowl of silver or brass of medium size and with a broad rim, level (ṣaḥīḥa) so that (the tip of) the ruler moves on its rim evenly, like the rim (ḥujra)[47] of the astrolabe. Then you fill the bowl with tar or wax,[48] which is melted and poured into the bowl until it is full and (the liquid) is evenly level with the rim (maʿa masḥat al-shafa sawāʾ). Then you take a plate of brass and put it in the middle of the bowl, so that it sinks into the tar or the wax. The center will be on it. Then you draw with the ruler a straight line across the middle of the bowl to the other rimwhich is the north‑south lineand another perpendicular straight line from the rim of the bowl to the (point on the) rim opposite the beginning point of the linewhich is called the east-west line. At this point the two lines will intersect each other exactly in the middle of the plate. The center of the compass will be at the point of intersection. You divide the breadth of the rim into four parts and describe a circle on the part [100] next to the inner rim of the bowl. Then you obliterate two parts of its distance. You describe a circle on the fourth part next to the outer rim of the bowl, and then you divide the circlewhich is called the degree scale (lit. circle of degrees)which is next to the inner rim of the bowl into three hundred and sixty equal parts, each of which is called a degree. You divide the next two circles which follow this circle namely, the circle of the five (degree) arguments (dāʾirat al-aʿdād al-akhmās)into seventy-two parts, each part of which consists of five degrees, and you write the individual numbers on it. This will be mentioned in the diagram (mithāl) of the circle on the rim of the bowl, God Almighty willing. Then you divide the (so‑called) circle of the quadrant outside this circle and next to the outer rim of the bowl into four equal parts and write [144v] north, south, east, and west, respectively, on each of its parts, whereupon (the bowl) is finished.

If[49] you wish to determine the meridian (lit. line of midday, khaṭṭ niṣf al-nahār) as nearly and as easily as possible, then fill the bowl with water and place it in a level spot on the ground which slopes neither upwards nor downwards but which is even and well protected against the wind, lest the needle be disturbed. Then you take a steel needle and rub its head well with a magnetic stone. The best (species) is the black and bright[50] (variety), which has especially sympathetic qualities (fa-innahu min al-khawāṣṣ).[51] Then you take a (slender) rush-like stem,[52] be it a stalk of grass, or saffron,[53] or straw, the (last mentioned) being best. Its length should be the same as the length of the needle. You insert the needle into the rush-like stem at half length and push it through until the stem reaches the middle of the needle. (The rush-like stem) and the needle will then form a cross, thus: | . Now the needle is placed on the water, and it rotates until it stops approximately [101] on the meridian (khaṭṭ wasaṭ al-nahār). Sometimes the sharp head (of the needle), which you have rubbed on the stone, faces north, and sometimes it faces south, whereas the place for inserting the thread, which you did not rub, faces north. There is no way to predict which will occur. But whether its head faces north or south, for one of the needle's two heads will always be attracted to the north or to the south,[54] and whatever direction it points to, you will know from how it stops that the meridian, approximately, is parallel to the length of the needle. When you have ascertained the meridian, which is from the north point to the south point, you will know (that) one of the two heads of the rush-like stem points east and the other west.[55] If the needle moves away from the middle of the bowl to one of its sides, push it a little bit with your hand or bring the magnetic stone close to it, until its head is opposite the north point and it is back in the middle of the bowl.

The essential thing in determining the direction of the qibla is to place the needle [145r] exactly over the center of the bowl in such a way that the intersection of the needle with the rush-like stem is directly above the center of the bowl and does not deviate from it, not even a little. Otherwise there might result an error in the direction of the qibla equivalent to this deviation (mayl). The intersection of the needle and the rush-like stemthat is, the stalkon the water should be precisely over the center of the bowl without any divergence (zaygh). Then the meridian will be facing the head of the needle. Let us imagine that a straight line is drawn from the head of the needle so that it passes through one of the divisions on the rim of the bowl. This division opposite the head of the needle will then represent approximately the deviation of the meridian, which is the north point. From the north point, you subtract towards the east (the amount) necessary for the deviation of the qibla in the 2020 table (al-jadwal al-ʿishrīnī). If the needle rests in the middle of the bowl and its head is pointing in one of the intermediate non-cardinal directions (al-jihāt al-n-k-b) and is not facing the north point, then [102] turn the bowl until the head of the needle that is in the middle of the bowl is opposite the north point. If it is opposite it, then the four (cardinal) directions are right according to the best approximation, as we have shown. This instrument does not require sighting the sun or the stars, but is adequate in itself, showing the meridian and all directions under a cloudless or a clouded sky, and by night and day. If you have determined the north line with it, and you next wish to determine the qibla for every locality according to the values found in the 2020 table for establishing the qibla, and you want to determine the qibla, then you count 27 on the rim of the bowl from (the point) opposite the head of the needlewhich is the north pointeastward, that is to say, twenty-seven degrees. The twenty-seventh degree is the qibla for the middle of the Yemen, for Aden, Taʿizz and Zabīd. Then you draw a line on the ground parallel to the north point and a line from the center to the twenty-seventh degreewhich is the qibla for the middle of the Yemen. Pray in that direction, because, when the north line is known, you know the qibla [145v] of every locality of the climates according to (its) deviation and according to all the surrounding degrees on the circle of the entire rim of the bowl that are associated with it. Here is the diagram of the circle you make on the rim of the bowl and the individual arguments. The diagram of the needle is in the middle.

 

(The diagram with the inscription follows [see fig. 1 at the end]. There are three circles; the outermost is divided into four parts marked: south [up], west [right], north [down], east [left], qibla of Taʿizz [20 degrees west of north], qibla of Aden [20 degrees east of north]. The middle circle is divided into 72 equal parts lettered every five degrees. The innermost circle is divided into 360 parts, each degree marked with a short stroke. In the center are two lines from south to north and from east to west marked: where the rush-like stem comes to rest in the bowl in line with [lit., opposite] the east-west line and where the needle comes to rest in the bowl in line with [lit., opposite] the meridian.)

 

The order (tartīb) of these notations is that you write 5 (hāʾ), that is, (the number) five, in the first division to the right of the north line, 10 (yāʾ), that is, (the number) ten, in the second division, and likewise with the remaining divisions, in the same manner, [146r] in units and tens respectively, such that the number of the last division, which is at the east line, is 90 (ṣād). Then you write another 90 (ṣād) in the division which follows it on the south side, so that the east line lies between the two 90s (ṣādayn). Then after the 90 you write 5, and you continue to the last division, which is next to the south [103] line, where the number is 5. Then you write 5 in the division which is next to it, and you continue to the last division, which is next to the west line, where the number is 90. You write another 90 on the division which is next to it, so that the west line lies between the two 90s, and you continue to the last line of the section, which is at the line of the mid-heaven (khaṭṭ wasaṭ al-samāʾ), where (the number) is 5. All (the markings) are now complete.

Here ends the text of the two treatises on the astrolabe, the water‑clock and the sundial, and the treatise on the magnetic compass concerning the determination of the qibla. We have accomplished our task to the best of our ability and after investigating the matter with the leading scholars in this field, from discussions with them and looking at the images they have drawn after the discussion, until we came to share with them the beneficial knowledge we had attained. (Only) then did we write down the operations we have recorded in this book of ours. Whoever studies it, let him not judge us too harshly. We are (only) one of those who have participated (in the search for knowledge) and not one of those who claim (to know much). We belong to those who have endeavored (to come from something known to something new), not to the speculative debaters. We ask God Almighty for assistance, increase (in knowledge)[56] and inspiration to reach our desire through His benevolence, His might, His strength and His power. He is all I need and sufficient. Herewith the treatises are concluded.

Praise be to God, to Him alone, and may His blessings and His favor be upon His messenger, our Lord Muḥammad the Prophet, and upon his family.

 

2.3. Commentary

The manuscript begins with the customary eulogy,[57] followed by a short summary of the subject and several paragraphs on the construction of the bowl (ṭāsa), the determination of the meridian (khaṭṭ niṣf al-nahār) and a main section on the determination of the qibla. It ends with a reference to investigations with the leading scholars in this field and further praise of God.

[104] The description a broad rim level (ṣaḥīḥa) so that (the tip of) the ruler moves on its rim evenly, like the rim (ḥujra) of the astrolabe is not to be taken to mean that the compass bowl is fitted with a device like an alidade for taking magnetic bearings.[58] The text seems to introduce the ruler only in order to explain that the rim should be absolutely flat. But there are also technical problems: the cross of the needle (ibra) and the rush-like stem cannot support an alidade in the middle of the bowl,[59] and there is no reference to an axle or guide rails to stabilize an alidade. Further, there is no hint of an alidade in the detailed figure in the manuscript. Surely the flat and carefully divided rim suggests an alidade, but why is it not mentioned in the text, since al-Ashraf describes every detail so carefully?

In an analogy to al-Ashrafs mention of the use of tar or wax, al-Bīrūnī, in his treatise on the construction of the astrolabe, describes a brass ring to be used as a mold for construction purposes. This ring is filled with tar or wax so that the plate laid on it is on an equal level with the rim of the ring.[60]

Al-Ashraf uses steel needles magnetized by rubbing with a magnetic stone. These needles keep their magnetic property longer than iron needles of the same size. While Gerland argues against the use of steel needles, Wiedemann is convinced that steel needles actually were in use and refers to experiments of his own on the magnetic behavior of steel nails.[61]

[105] Explanations of the magnetic properties of the needle follow.[62] Al-Ashraf does not seem to know whether the end rubbed with the magnetic stone or the other one turns north, but he knows that each head retains its attraction (ṭilba, taliba) or desire to turn north or south.[63] In his comments on manuscript B, Wiedemann here understands that al-Ashraf refers to the fact that the head of the needle which is not rubbed has also changed its behavior.[64]

In the following part, on the determination of the qibla with the compass bowl, al-Ashraf first treats the determination of the north point. But there are some obscurities in the manuscript which raise questions about al-Ashrafs knowledge of the magnetic declination, that is, the deviation of magnetic north from geographical north. But how is magnetic declination to be determined? Provided the variation is sufficient,[65] a comparison between the [106] astronomical cardinal point observed and the direction marked by the magnetic needle will give a deviation.[66] But this observation does not have to lead to the assumption that the magnetic meridian differs from the astronomical one. The problem could be explained by the inaccuracy of the observations and the deficiency of the magnetic compass used.[67] In his text on the determination of the north point al-Ashraf always adds bi-taqrīb or taqrīban, that is, approximately, which can reflect magnetic variation or the insufficiency of the compass bowl or the observers incompetence.[68] More importantand more questionableis the sentence: This division (on the rim of the bowl) opposite the head of the needle will then represent approximately the deviation (inḥirāf) of the meridian (khaṭṭ niṣf al-nahār), which is the north point. It could refer to the magnetic variation, but not necessarily, and it could also be a terminus technicus. In the figure there are marks at 20 degrees east and west of the north point, which are probably not the qibla values mentioned in the text;[69] but on the other hand the cross of [107] needle and rush-like stem is aligned in the cardinal directions. Further, there are no hints at a corrected north point in the manuscript. The most likely conclusion would seem to be that al-Ashraf did not know of magnetic variation, even if he had an idea that the magnetic needle does not point to the north point every time. The expression inḥirāf khaṭṭ niṣf al-nahār remains unexplained.

Before turning to the the essential thing in determining the direction of the qibla, one remarkable term has to be discussed. Speaking of the case in which the needle is not directed to the north point, al-Ashraf uses al-jihāt al-n-k-b, where the verb nakaba can mean to turn aside. Further it has the same root as nakbāʾ, meaning a wind that blows obliquely, taking a direction between (the directions of) two (cardinal) winds.[70] At first sight it seems to make sense to equate the four cardinal directions with our four cardinal winds.[71] However, they do not correspond to them in the Islamic world,[72] and in addition the Arabs used another form of the compass card.[73]

Al-Ashraf states that qibla values should be taken from the table called al-jadwal al-ʿishrīnī. This qibla table contains 2020 or 400 entries for each degree of latitude and longitude difference from Mecca from 1 degree to 20 degrees, hence the name jadwal ʿishrīnī (ʿishrīn = 20). There are two further references to 2020 tables in medieval Arabic treatises, one in some miscellaneous notes on spherical astronomy, copied in Cairo about 1200, and another in Ibn Simʿūns treatise,[74] which is our second source. But it is difficult [108] to identify the qibla table they deal with. Possibly they refer to an Abbasid qibla table, which also contains 2020 entries and is preserved in nine manuscripts, three of them of Yemeni provenance.[75] Al-Ashraf gives the qibla for the middle of the Yemen (Aden, Taʿizz, Zabīd) as 27 degrees east of north. But in the figure, a mark 20 degrees west of north is referred to as qiblat Taʿizz and one 20 degrees east of north as qiblat ʿAdan. In his extensive astrological compendium al-Tabṣira fī ʿilm al-nujūm, preserved in a single copy in Oxford, al-Ashraf gives coordinates of these localities:[76]

 

locality

L

ϕ

Mecca[77]

67;0/60;0

21;0

Aden

65;30

13;0

Taʿizz

66;30

13;43

Zabīd

62;0

14;0

 

These coordinates together with several methods and tables enable us to recompute al-Ashrafs qibla values.[78] The first of the following tables gives the values according to al-Ashrafs treatise on the magnetic compass. The quadrant of the deviation is given because the Oxford manuscript has two values for the longitude of Mecca (LM), and the geographical longitudes of Aden, Taʿizz and Zabīd are between the two.

 

 

Aden

Taʿizz

Zabīd

text

27 NE

27 NE

27 NE

figure[79]

20 NE

20 NW

 

[109] The second table gives the calculated values of the qibla according to the correct modern formula[80] and according to two approximate methods. The first of the approximations uses recomputed values based on a standard approximate method, and the second uses the Abbasid qibla table mentioned above:[81]

 

 

LM

Aden

Taʿizz

Zabīd

L=63;30

ϕ=14;30

calculation

60;0

32;33 NW

39;30 NW

14;57 NW

 

67;0

09;57 NE

03;41 NE

33;33 NE

approximation

60;0

34;29 NW

41;45 NW

15;59 NW

(first method)

67;0

10;37 NE

03;56 NE

35;34 NE

approximation

60;0

32;44 NW

39;55 NW

14;57 NW

26;54 NW

(second method)

67;0

09;55 NE

03;40 NE

33;45 NE

 

The values of the deviation taken from the figure correspond neither with the values given in the manuscript nor with any of the calculated values. Nor can approximative methods explain the value of 27 degrees east of north given by al-Ashraf. But one explanation is possible. Earlier in al-Ashrafs text, in his treatise on the sundial, which precedes the treatise on the magnetic compass,[82] the author also gives a qibla for the Yemen of 27 degrees and a fraction, but without mentioning the quadrant. He uses for Mecca a longitude of 60;0 and for the Yemen of 63;30.[83] For these values and an assumed latitude of 14;30,[84] approximately the value given in the treatise on [110] the magnetic compass is obtainedhowever in the wrong quadrant. This could be explained by the lack of the quadrant in the earlier passage, and it provides an explanation for the values in the diagram as well. The copyist was not aware which quadrant was the right one.

In the final paragraph of our text the scope of the entire work is mentioned: the two treatises on the astrolabe, the water-clock and the sundial and the treatise on the magnetic compass concerning the determination of the qibla. This makes it clear that the treatise on the magnetic compass is part of the main astronomical work.

Al-Ashraf concludes with the remark that he had put forth his best efforts in writing the treatise and had consulted the leading scholars in this field. It is difficult to imagine that he did not also consult an earlier treatise on the magnetic compass.

To sum up, while the earliest known sources on the magnetic compass in the Islamic world describe its use at sea, in al-Ashrafs treatise on the construction of astronomical instruments, written at the end of the thirteenth century, the compass has an entirely different function. We can not [111] only determine the meridian under any weather conditions and at any hour of day or night, but we can also use it as a basis for the determination of the qibla. Furthermore, al-Ashrafs treatise constitutes on the one hand the first reference to the magnetic compass in an astronomical work[85] and on the other the first full description of its construction in the Arabic sources known to us.

 

3. Ibn Simʿūns treatise

A treatise on time‑keeping written about 1300 by an Egyptian astronomer and muezzin called Ibn Simʿūn contains a chapter on the magnetic qibla indicator which is our second source. It is based mainly on the work of Abū ʿAlī al-Marrākushī, who compiled a compendium of spherical astronomy and astronomical instruments, which is perhaps the most valuable single source for the history of Islamic astronomical instrumentation,[86] as well as on the Muṣṭalaḥ zīj, the most popular zījas medieval Islamic astronomical handbooks are calledin medieval Egypt.[87] Only one copy of Ibn Simʿūns treatise, preserved in the Universiteitsbibliotheek Leiden (Or. 468, 192 fols.), is known. The work is entitled Kanz al-yawāqīt fī ʾstīʿāb al-mawāqīt. Fol. 190r contains the Chapter on the Use of the Qibla Instrument for Every Locality.[88] This source is henceforth labeled L.

The Arabic text given here is provided with hamzas and diacritical points. On the punctuation and use of parentheses, see above.

 

3.1. Arabic text

 

[112]

     .    .

     .  .    .  .  .     ][    .    . [113]  .  .  .    .

 

The diagrams with the inscriptions follow. There are two figures, on the right hand side the qibla instrument, labeled: wa-hādhihi ṣūrat al-ḥuqq al-madhkūr (headline), al-jām al-muṭabbaq ʿalā fam al-ḥuqq (above), dhāt al-maḥārīb (center), al-turays (below, left side), al-ibra (below, right side), ṣifat al-ḥuqq (right). The second figure, on the left hand side, is the disk with the Kaaba provided with the cardinal points, in the center the Kaaba: wa-hādhihi jadāwil maḥārīb jumla min al-buldān bi-ḥasab istidāratihā ḥawla ʾl-Kaʿba fī dhāt al-maḥārīb (headline, repeated under the figure). Al-jadwal al-awwal fī ʾnḥirāf maḥārīb al-buldān allatī ʾl-Kaʿba minhā sharqīya janūbīya (headline of the following table).

 

3.2. Translation

Chapter on the Use of the Qibla Instrument (ālat al-qibla) for every locality you wish by means of the magnet, the description of the Kaaba and a table which contains the qiblas (inḥirāf al-qibal) for a number of localities and the direction of the qibla in any given horizon quadrant.

Know that we have not found (a description of) the use of the magnet by any of the early (scholars) of this discipline, although it is well known amongst a number of more recent (scholars), especially in our own time. Al-Marrākushī did not mention it in his (Kitāb jāmiʿ) al-mabādiʾ wa-ʾl-ghāyāt (fī ʿilm al-mīqāt). (The use of the magnet) belongs to the diversions (milḥ) of this science, not to its core (matīn). Reliance on it is dangerous, because of (the possibility) of the loss (salb) of its sympathetic qualities and because of the quickness with which the cone (turays) changes on which the qibla instrument turns, owing to its considerable hesitating motion.

(Scholars) are in agreement that it does not display hesitating motion in a glass container nor is it led (fa-lam yuqad), nor does it do so in (a container) of [114] silver, nor in one of deep-red crystal (billawr qāniʾ),[89] for (in all these cases) it recoils (yastankir) and stops. Therefore the early scholars neglected it, because of the two problems we have mentioned. Now let me mention of the construction of the instrument and how one uses it to determine the qibla according to the opinion of those more recent scholars who are concerned with the diversions (of our science). We say, and with God is success:

The manner in which the qibla instrument is contracted is that you make a round form of paper pasted together (waraq mutamāsik) or qarʿ[90] or the like. You divide its circumference into 360 degrees, and you mark (the direction of) the prayer-niches of the localities you choose, according to their qibla, distinguished by an inscription indicating the locality. Sometimes a horizontal sundial (basīṭa) for a specific latitude is made on some of these with a little gnomon (shakhṣ laṭīf), appropriate to this instrument in accordance with its size.[91] A piece of magnet is put on its side, either on the meridian or inclined from it in the direction in which you want to perform your religious obligations (al-jiha allatī aradta ilayhā al-taḥrima), according to the deviation (inḥirāf) of the magnet from the line (of midday). Two needles (ibratayn) are put on its back in order to balance it. Sometimes a diagram of the exalted Kaaba is made in the middle of the instrument. This instrument is put on something like a cone of glass or brass or other material. The smoother and freer the rotation of the instrument the better (the material). (Then) this is put on a needle implanted in a box of ebony or brass or other material as may be appropriate. A lid (jām) of mirror-glass is put on the opening of the box, through which one can see when the decorative prayer-niche stops, which is what is intended when it is moved in the direction of the qibla (yurā minhu wuqūfu ʾl-miḥrābi ʾl-marsūmi ʾl-maṭlūbu ḥāla taḥrīkihi fī jihati ʾl-qibla). The less the weight of the attractive (element) [115] of magnetic material and the greater the weight of the prayer-niche, the higher the quality of that particular kind of magnet. Let garlic be kept away (from the magnet), for it neutralizes its sympathetic qualities. But these are restored by dipping the magnet in goats blood.[92] And God knows best.

 

(The diagrams with the inscriptions follow [see fig. 2 at the end]. There are two figures, on the right side the qibla instrument labeled: This is the diagram of the box mentioned [headline], the corresponding lid placed above the opening of the box (al-jām al-muṭabbaq ʿalā fam al-ḥuqq) [above], the prayer-niches [center], the cone [below, left], the needle [below, right], description of the box [right]. The second figure, on the left hand side, is the disk with the Kaaba labeled with the cardinal points, in the center the Kaaba. The headline repeated under the picture reads: These are tables of prayer-niches for a collection of localities according to their place on a circle around the Kaaba on the prayer-niche scale in this figure. The headline of the table that follows[93] reads: The first table of the qiblas of the prayer-niches of the localities for which the Kaaba is in the south-east.)

 

3.3. Commentary

The Chapter on the Use of the Qibla Instrument includes construction of the qibla indicator, a description of the Kaaba and a table of the deviation of qiblas (inḥirāf al-qibal). The name used for the instrument described (ālat al-qibla) is remarkable in that it is not determined only by its form but by its purpose, namely, to indicate the qibla.

[116] The first substantive section of the passage begins with a short general introduction on the directive power of the magnetic stone. The statement that the use of the magnet is not known to any of the early (scholars) of this discipline, although it is well known amongst a number of more recent (scholars), agrees with our assertion that the magnets directive power was unknown in Antiquity but was known later in the Islamic world. The author specifically mentions al-Marrākushī, the major contemporary for astronomical instruments, who does not mention the use of the magnetic compass in his opus magnum.

Two flaws of the magnetic compass are described: loss of magnetic properties and attrition of the cone (turays).[94] Because of these two flaws the early scholars (al-awwalūn) did not use the magnetic compass and probably in the opinion of the more recent scholars (al-mutaʾakhkhirūn) it belonged to the diversions of the subject (milḥ). Unfortunately, neither group has been identified. But it is remarkable that this statement refers to the earlier knowledge of a dry compass.

The second section of the passage contains instructions for the construction of the qibla indicator. The construction of the round plate is more or less clear, but the materials described as paper pasted together (waraq mutamāsik) and pumpkin or gourd (qarʿ) are a problem. They should be firm and light in weight. This is all we can say. Three further matters arise within the next few lines. Firstly, is the piece of magnetic stone fixed upon the round paper plate or used to magnetize the needles (ibratayn) mentioned later? Secondly, is the direction in which you want to perform your religious obligations (al-jiha allatī aradta ilayhā al-taḥrima) an unusual way of referring to the qibla with some sort of miḥrāb indicating the direction towards Mecca? Thirdly, does the author know of the magnetic declination when he speaks of the deviation of the magnet from the line (of midday) (inḥirāf dhālik al-maghnāṭīs ʿan al-khaṭṭ)?[95] As mentioned above, in the commentary on al-Ashrafs treatise, knowing that the needle does not point to the true north-south line is not necessarily the result of an observation of the magnetic declination.

Especially difficult to understand is the word cone (turays), which is written without diacritical points in the manuscript. As yet I have been able [117] to find no reference for it.[96] But the author himself provides some clues. Firstly, reliance on it (the compass) is dangerous, because . . . of the quickness with which the cone (turays) changes on which the qibla instrument turns, owing to its considerable hesitating motion. Secondly, this instrument is put on something like a cone of glass or brass or other material. The smoother and freer the rotation of the instrument the better the material. Thus it is likely that the cone was thought to be something like a little plate or dome fixed under the round plate of paper in order to diminish the friction between the plate and the needle implanted in the box. This seems to be similar to al-Zarkhūrīs description of a dry compass.[97] A comparison between the Arabic texts of al-Zarkhūrī and Ibn Simʿūn could provide further information, especially regarding vocabulary.[98]

The treatise of the Egyptian Ibn Simʿūn thus provides the earliest evidence of a dry compass in the Islamic world. In the headline the author uses ālat al-qibla for his instrument rather than any of the several terms known to us from medieval Arabic sources (ḥuqq al-qibla, ṭāsa, ḥikk, bayt al-ibra[99]). He explains the purpose of the instrument at the beginning. Some aspects of the construction are only implicit in the text, but the introduction and the end clearly indicate that the author is familiar with the traditions of the magnetic stone. Apart from al-Zarkhūrī this qibla indicator can also be compared with the ṣandūq al-yawāqīt of Ibn al-Shāṭir, which is an astronomical compendium[100] from the second half of the fourteenth century, on the construction [118] and use of which two manuscripts are preserved.[101] The magnetic compass fitted with the instrument is now lost. In the two texts about the instrument the compass is called south-pointer (murī al-janūb).[102] There are only general hints at the construction of the compass, and these are not sufficiently detailed to make possible a comparison with Ibn Simʿūns qibla indicator.

 

4. Summary

The importance of the two treatises discussed here lies in the description of the determination of the qibla by means of the magnetic compass. To this day al-Ashraf and Ibn Simʿūn are the earliest textual sources we know for this religious use of the directive power of the magnet. The Yemeni Sultan al-Ashraf, at the end of the thirteenth century, describes a floating compass mentioned also by other authors, but he gives detailed information on its construction and use. His text is an integral part of an astronomical treatise, the first reference to a floating compass in a medieval Islamic scientific treatise. The Egyptian astronomer Ibn Simʿūn describes another kind of qibla indicator, a dry rather than a floating compass, with specific markings. His is the earliest known reference to this instrument in the Islamic world.

Nearly a century ago, Wiedemann, referring to ʿAwfī and al-Zarkhūrī, stated that the Arabs were aware of magnetizing by rubbing with a magnetic stone at the beginning of the thirteenth century, that it was known that iron with a content of steel may be made into a permanent magnet, that the technique was practiced in the fourteenth century, and that the magnetic compass was in general use.

 

Die mitgeteilten Stellen zeigen, daß man bei den Arabern schon am Anfang des 13. Jahrhunderts die Magnetisierung durch Streichen kannte und daß man diese auch im 14. Jahrhundert im Orient anwandte. Beide Stellen zeigen auch, daß bekannt war, daß das stets etwas stahlhaltige Eisen dauernd magnetisch wurde, und daß diese Erscheinung experimentell geprft wurde; ferner, daß zu dieser Zeit der Kompaß allgemeine Anwendung fand. (Wiedemann, 1907)[103]

 

Now we can add that the use of the magnetic compass in the service of [119] religion is attested in two medieval Arabic treatises, and in two different varieties: one floating, described by a Yemeni astronomer-prince, and one dry, described by an Egyptian mosque astronomer, both writing around 1300.


[120] 5. Appendix

 

5.1. Glossary [104]

 

Arabic

English

Source

 

ebony

L

 

horizon

L

 

-

clime (in the sense of classical geography), also region

C

 

qibla instrument, qibla indicator

L

 

horizontal sundial

L

 

deep-red crystal

L

 

(geometrical) compass

C

 

straw

C

 

water-clock

C

 

little shield; cone, dome, funnel

L

 

vessel; lid

L

 

-

table

L, C

 

outer rim of the astrolabe

C

 

()

angle between the meridian of any localitiy and the great circle passing through this locality and Mecca

L, C

 

-

sympathetic qualities, virtus: describing an inexplicable force inherent in animate and inanimate nature

L, C

 

line of midday, meridian

L, C

 

line of the middle of the heaven, meridian

C

 

-

degree

L, C

 

sundial

C

 

divergence

C

 

glass

L

 

[121]

 

 

 

tractability; something being or becoming smooth

L

 

being opposite, facing

C

 

sort of blade (of rush?)

C

 

gnomon

L

 

wax; candle

C

 

bowl, drinking vessel; magnetic compass bowl

C

 

silver

L, C

 

steel, cleaned and improved iron

C

 

pitch, tar

C

 

-

qibla, the sacred direction of Islam towards Mecca

L, C

 

pumpkin

L

 

-

part, portion

C

 

blade

C

 

   

magnet, magnetic stone; magnetism

L, C

 

deviation, declination (astron.)

C

 

copper; brass

L, C

 

-

wind, blowing from a direction between two of the cardinal directions

C


 

[122] SELECTED BIBLIOGRAPHY

 

Works mentioned in the footnotes are listed here. They are quoted according to the original edition; reprints are only listed.

 

Ahlwardt, W. Die Handschriften-Verzeichnisse der Königlichen Bibliothek zu Berlin. Vol. 17. Verzeichnis der arabischen Handschriften. Vol. 5. Berlin, 1893.

al-ʿAzzāwī, A. Tārīkh ʿilm al-falak fī ʾl-ʿIrāq wa-ʿalāqātihi bi-l-aqṭār al-islāmiyya wa-l-ʿarabiyya fī ʾl-ʿuhūd al-tāliya li-ayyām al-ʿabbāsiyyīn. (History of astronomy in Iraq and its relations with Islamic and Arab countries in the post-Abbasid period). Baghdad, 1959.

Balmer, Heinz. Beiträge zur Geschichte der Erkenntnis des Erdmagnetismus. Veröffentlichungen der Schweizerischen Gesellschaft fr Geschichte der Medizin und der Naturwissenschaften 20. Aarau, 1956.

van Bemmelen, Willem. De Isogonen in de XVIde en XVIIde Eeuw. Proefschrift ter Verkrijging van den Graad van Doctor in de Wis- en Natuurkunde aan de Rijks-Universiteit te Leiden. Utrecht, 1893.

. Die Abweichung der Magnetnadel. Beobachtungen, Säcular-Variation, Wert- und Isogonensysteme bis zur Mitte des 18. Jahrhunderts. Observations Magnet. Meteorol. Observatory Batavia 21, supp. Batavia, 1899.

al-Bīrūnī. The Book of Instruction in the Art of Astrology by al-Bīrūnī. Trans. with facsimile R. Wright. London, 1934.

Bobinger, Max. Alt-Augsburger Kompaßmacher. Abhandlungen zur Geschichte der Stadt Augsburg. Schriftenreihe des Stadtarchivs Augsburg. Augsburg, 1966.

Brice, W. C. Compasses, Compassi and Kanābīṣ. Journal of Semitic Studies 29 (1984): 16978.

Brockelmann, Carl. Geschichte der arabischen Litteratur. 2nd edition. Leiden, 1943.

Bromehead, C. N. Alexander Neckam on the Compass-Needle. Terrestrial Magnetism and Atmospherical Electricity 50 (1945): 13940.

Chandler, Bruce. Three Nurnberg Compassmacher. Metropolitan Museum Journal 2 (1969): 21116.

Clment-Mullet, Jacques. Essai sur la minralogie arabe. Chapitre XIV: Laimant, al-Māgnāṭīs ou al-Māgnīṭis. Journal asiatique, 6ime srie, 11 (1868): 17078. Reprint, IG 15:5, pp. 15664.

Dizer, Muammer. The Dāʾirat al-Muʿaddal in the Kandilli Observatory, and Some Remarks on the Earliest Recorded Islamic Values of the Magnetic [123] Declination. Journal for the History of Arabic Sciences 1 (1977): 25762.

Dozy, R. Supplment aux dictionnaires arabes. 2 vols. 2nd edition. Leiden, Paris, 1927.

DSB = Dictionary of Scientific Biography. 16 vols. New York, 197080.

EI1 = The Encyclopaedia of Islam. First Edition. 9 vols. Leiden, London, 191336. Photomechanical Reprint. Leiden, 1987.

EI2 = The Encyclopaedia of Islam. New Edition. 8 vols. to date. Leiden, 1960.

Ferrand, Gabriel. Instructions nautiques et routiers arabes et portugais des XVe et XVIe sicles. Tome III: Introduction lastronomie nautique arabe. Paris 1928. Reprinted as Gabriel Ferrand, Introduction lastronomie nautique arabe, Nachdruck der Ausgabe Paris 1928. Ed. Fuat Sezgin. Frankfurt, 1986.

Finsch, H. Geschichte der Magnetnadel und der Änderungen des Erdmagnetismus. Programm der Lateinischen Hauptschule in Halle fr das Schuljahr 18781879, no. 196. Halle, 1879.

Focus Behaim‑Globus [Nuremberg GNM Catalogue]. Ed. Gerhard Bott. 2 vols. Ausstellungskataloge des Germanischen Nationalmuseums. Nuremberg, 1992.

Flgel, Gustav. Concordantiae Corani arabicae. Leipzig, 1898.

Gerland, Ernst. Der Kompaß bei den Arabern und im christlichen Mittelalter. Mitteilungen zur Geschichte der Medizin und der Naturwissenschaften 6 (1907): 919. Reprint, IG 15:5, pp. 28292.

. Zur Geschichte der Magnetisierung von Kompaßnadeln mit Hilfe von natrlichen Magneten. VdDPG 10 (1908): 37792.

Hackmann, W. D. Jan van der Straet (Stradanus) and the Origins of the Mariners Compass. In Learning, Language and Invention: Essays Presented to Francis Maddison, ed. W. D. Hackman and Anthony J. Turner, pp. 14879. London, 1994.

Harrandon, H. D. Some Early Contributions to the History of Geomagnetism I. Terrestrial Magnetism and Atmospherical Electricity 48 (1943): 317.

. Some Early Contributions to the History of Geomagnetism  VIII. Terrestrial Magnetism and Atmospherical Electricity 50 (1945): 6368.

Hartner, Willy. The Principle and Use of the Astrolabe. In Willy Hartner, Oriens‑Occidens, Ausgewählte Schriften zur Wissenschafts- und Kulturgeschichte: Festschrift zum 60. Geburtstag, ed. Gnther Kerstein et al., [123] pp. 287311. Hildesheim, 1968.

Hashimoto, Masukichi. The Origin of the Compass. Memoirs of the Research Department of Toyo Bunko 1 (1926): 6992.

Hawkins, G. S., and David A. King. On the Orientation of the Kaʿba. Journal for the History of Astronomy 13 (1982): 102109. Reprint, King, ASI, XII.

Hellmann, Gustav. Über die Kenntnis der magnetischen Deklination vor Christoph Kolumbus. Meteorologische Zeitschrift 23 (1906): 14549.

. Über die Kenntnis der magnetischen Deklination vor Christoph Columbus. Meteorologische Zeitschrift 25 (1908): p. 369.

Heß, Jean-Jacques. Кαλαμίτης Magnetnadel. In Festgabe Adolf Kaegi, pp. 18990. Frauenfeld, 1919.

al-Ḥibshī, ʿAbd Allāh Muḥammad. Maṣādir al-fikr al-ʿarabī al-islāmī fī ʾl-Yaman. Sanʿāʾ, n.d.

Hill, Donald R. Arabic Water-Clocks. Sources & Studies in the History of Arabic-Islamic Science. History of Technology Series 4. Aleppo, 1981.

IG 15:5 = Sezgin, Fuat et al., eds., Reprint of Studies on Nautical Instruments. Islamic Geography 15: Mathematical Geography and Cartography 5. Frankfurt, 1992.

Janin, Louis, and David A. King. Ibn al-Shāṭirs Sandūq al-Yawāqīt: An Astronomical Compendium. Journal for the History of Arabic Sciences 1:2 (1977): 187256. Reprint, King, IAI, XII.

Jomard, Francois. Rapport sur une notice insre au Journal de la Socit asiatique de Calcutta, et relative aux instruments des Arabes. Bulletin de la Socit de Gographie. 2me srie 10 (1838): 28992. Reprint, IG 15:5, pp. 15255.

Kennedy, Edward S. Al-Bīrūnī on Determining the Meridian. Mathematics Teacher 56 (1963): 63537. Reprinted in Edward S. Kennedy, Studies in the Islamic Exact Sciences, pp. 61820. Ed. Mary H. Kennedy and David A. King. Beirut, 1983.

. A Survey of Islamic Astronomical Tables. Transactions of the American Philosophical Society. New Series 46:2. Philadelphia, 1956. Reprinted ca. 1990.

King, David A. On Arabic Water-Clocks. History of Science 15 (1977): 29598. Reprint, King, IAI, XXI.

. Fihris al-makhṭūṭāt al-ʿilmīya al-maḥfūẓa bi-Dār al-Kutub al-Miṣrīya (A catalogue of the scientific manuscripts in the Egyptian National Library). 2 vols. Cairo, 198186.

. Astronomical Alignments in Medieval Islamic Religious Architecture. Annals of the New York Academy of Sciences 385 (1982): 30312. Reprint, King, ASI, XIII.

. The Astronomy of the Mamluks. Isis 74 (1983): 53155. [125] Reprint, King, IMA, III.

. MAY = Mathematical Astronomy in Medieval Yemen. A Biobibliographical Survey. American Research Center in Egypt. Malibu, 1983.

. The Sacred Direction in Islam. A Study of the Interaction of Religion and Science in the Middle Ages. Interdisciplinary Science Reviews 10:4 (1985): 31528.

. The Medieval Yemeni Astrolabe in the Metropolitan Museum of Art in New York City. ZGAIW 2 (1985): 99122. Reprint, King, IAI II. Supplement in ZGAIW 4 (198788): 26869.

. IMA = Islamic Mathematical Astronomy. London, 1986. 2nd edition: Aldershot, 1993.

. A Survey of the Scientific Manuscripts in the Egyptian National Library. Winona Lake, Indiana, 1986.

. The Earliest Islamic Mathematical Methods and Tables for Finding the Direction of Mecca. ZGAIW 3 (1986): 82149. Reprint, King, ASI, XIV. Supplement in: ZGAIW 4 (198788): p. 270.

. IAI = Islamic Astronomical Instruments. London, 1987. Reprint, Aldershot, 1995.

. The Earliest Islamic Mathematical Methods and Tables for Finding the Direction of Mecca. ZGAIW 3 (1986): 82149. Reprint, King, ASI XIV. Supplement in: ZGAIW 4 (198788): p. 270.

. Weltkarten zur Ermittlung der Richtung nach Mekka. In Focus Behaim-Globus (1992), vol. 1, pp. 1677, and vol. 2, pp. 68691.

. ASI = Astronomy in the Service of Islam. Aldershot, 1993.

. Astronomical Instruments between East and West. In Kommunikation zwischen Orient und Okzident, Alltag und Sachkultur: Internationaler Kongreß, Krems an der Donau 6. bis 9. Oktober 1992, pp. 14398. Österreichische Akademie der Wissenschaften, Phil.-hist. Klasse, Sitzungsberichte 619. Wien: Verlag der Öst. Akademie der Wissensch., 1994.

. Lastronomie en Syrie lpoque islamique. In Syrie (1994), pp. 38695, 44041.

King, David A., and Gerald LE. Turner. The Astrolabe Presented by Regiomontanus to Cardinal Bessarion in 1462. Nuncius 9:1 (1994): 165206.

King, David A. On the Role of the Muezzin and the Muwaqqit in Medieval Islamic Society. In Tradition, Transmission, Transformation: Proceedings [126] of Two Conferences on Pre-modern Science Held at the University of Oklahoma, ed. F. Jamil Ragep and Sally P. Ragep, pp. 285345. Collection de traveaux de lAcadmie Internationale dHistoire des Sciences 37. Leiden, 1996.

. Two Iranian World Maps for Finding the Direction and Distance to Mecca. Imago Mundi 49 (1997): 6282.

. World-Maps for Finding the Direction and Distance to Mecca: Innovation and Tradition in Islamic Science. Leiden, Boston, 1999.

. Ciphers, Monks and Astrolabes. A Forgotten Number Notation of the Middle Ages. Forthcoming.

Klaproth, M. J. Lettre M. le Baron A. de Humboldt, sur linvention de la boussole. Paris 1834. Reprint, IG 15:5, pp. 1138.

Körber, Hans-Gnther. Zur Geschichte der Konstruktion von Sonnenuhren und Kompassen des 16. bis 18. Jahrhunderts. Veröffentlichungen des Staatlichen Mathematisch-Physikalischen Salons. Forschungsstelle Dresden-Zwinger 3. Berlin, 1965.

Kunitzsch, Paul. Glossar der arabischen Fachausdrcke in der mittelalterlichen europäischen Astrolabliteratur. Nachrichten der Akademie der Wissenschaften in Göttingen, 1. Phil.-Hist. Klasse 11. Göttingen, 1982.

Lane, Edward William. An Arabic-English Lexicon. 8 vols. London, Edinburgh, 18631893. Reprint, Beirut, 1968.

Lee, Samuel. Notice of the Astronomical Tables of Mohammed Abibekr Al Farsi, Two Copies of Which are Preserved in the Public Library of the University of Cambridge. Transactions of the Cambridge Philosophical Society 1 (1822): 24965.

LexMA = Lexikon des Mittelalters. 7 vols. to date. Munich and Zurich, 1980.

von Lippmann, Edmund O. Geschichte der Magnetnadel bis zur Erfindung des Kompasses (gegen 1300). QSGNM 3:1 (1932): 149. Part 4: Die Araber. Reprint, IG 15:5, pp. 44651.

Löfgren, Oscar, and R. Traini. Catalogue of the Arabic Manuscripts in the Biblioteca Ambrosiana. Vol. 1. Vicenza, 1975.

Mayer, L. A. Islamic Astrolabists and their Works. Geneva, 1956.

Maddison, Francis. Medieval Scientific Instruments and the Development of Navigational Instruments in the XVth and XVIth Centuries. Agrupamento de Estudos de Cartografia Antiga, Secão de Coimbra 30. Coimbra, 1969.

Michel, Henri. Trait de lastrolabe. Paris 1947. Reprint, Paris, 1976.

Mitchell, A. Chrichton. Chapters in the History of Terrestrial Magnetism. Part 1: On the Directive Property of a Magnet in the Earths Field and the Origin of the Nautical Compass. Terrestrial Magnetism and Atmospherical Electricity 37 (1932): 10546. [127]

. Chapters in the History of Terrestrial Magnetism. Part 2: The Discovery of the Magnetic Declination. Terrestrial Magnetism and Atmospherical Electricity 42 (1937): 24180.

Needham, Joseph. Science and Civilisation in China. 6 vols. Cambridge, 195486.

. The Chinese Contribution to the Development of the Mariners Compass. Scientia 55:7 (1961): 22533. (Supp., pp. 11624.)

Nippoldt, Alfred. Ein Beitrag zur Frage der Ausrichtung der Kirchenachsen mit dem Magneten. Archiv fr Geschichte der Naturwissenschaften und der Technik 7 (1916): 10914, 23644.

North, J. D. The Astrolabe. Scientific American 230:1 (1974): 96106. Reprinted in J. D. North, Stars, Minds, and Fate. Essays in Ancient and Medieval Cosmology, pp. 21120. London, Ronceverte (West Virginia), 1989.

QSGNM = Quellen und Studien zur Geschichte der Naturwissenschaften und der Medizin.

Paulys Realencyclopädie der classischen Altertumswissenschaft. Neue Bearbeitung. Stuttgart 1893 .

The Planispheric Astrolabe. National Maritime Museum. Greenwich, 1976.

Prinsep, James. Note on the Nautical Instruments of the Arabs. The Journal of the Asiatic Society of Bengal 5 (1836): 78494, and 2 plates. Reprinted with comments in Gabriel Ferrand, Instructions nautiques, pp. 130. Reprint, IG 15:5, pp. 13951.

Radl, Albert. Der Magnetstein in der Antike. Boethius 19. Stuttgart, 1988.

Rara Magnetica 12691599. Neudrucke von Schriften und Karten ber Meteorologie und Erdmagnetismus. Ed. G. Hellmann. No. 10. Berlin, 1898.

Sarton, George. Introduction to the History of Science. 3 vols. Carnegie Institution of Washington, Publication No. 376. Baltimore, 192748. Reprint, 1962.

de Saussure, Lopold. Lorigine de la rose des vents et linvention de la boussole. Archives des sciences physiques et naturelles 5 (1923): 14981, 25991. Reprinted with comments in Gabriel Ferrand, Instructions nautiques, pp. 31127. Reprint, IG 15:5, pp. 311407.

Schleif, W. Geschichtliches vom Magnetkompaß. Der Seewart 14:6 (1953): 2428.

Schck, Albert. Erwähnung eines Vorgängers des Kompasses in Deutschland um die Mitte des 13. Jahrhunderts. Mitteilungen zur Geschichte der Medizin und der Naturwissenschaften 13 (1914): 33343.

Sdillot, Jean-Jacques. Trait des instruments astronomiques des Arabes compos au treizime sicle par Aboul Hhassan Ali de Maroc intitul Jāmiʿ al-mabādiʾ wa-l-ghāyāt. 2 vols. Ed. Louis-Amlie Sdillot. Paris, 1834-35. Reprinted as Louis-Amlie Sdillot and Jean-Jacques Sdillot, Trait des instruments astronomiques des Arabes compos au treizime sicle par Aboul Hhassan Ali de Maroc intitul Jāmiʿ al-mabādiʾ wa-l-ghāyāt. Veröffentlichungen des Institutes fr Geschichte der Arabisch-Islamischen Wissenschaften. Reihe B: Nachdrucke 2. Ed. Fuat Sezgin. Frankfurt, 1984. [128]

Sdillot, Louis-Amlie. Mmoire sur les instruments astronomiques des Arabes. Mmoires de lAcadmie Royale des inscriptions et belles-lettres de lInstitut de France 1 (1844): 1229. Reprinted in Veröffentlichungen des Institutes fr Geschichte der Arabisch-Islamischen Wissenschaften. Reihe B: Nachdrucke, Abteilung Astronomie 3. Ed. Fuat Sezgin. Frankfurt, 1989.

Siggel, Alfred. Mitteilungen ber das Buch der aufgehenden Sterne [al-Nujūm al-shāriqāt] des Muḥammad b. Abī ʾl-Khair al-Ḥasanī. QSGNM 8 (194142): 43557.

SPMSE = Sitzungsberichte der physikalisch-medizinischen Societät zu Erlangen.

Storey, C.A. Persian Literature. Volume 1, Part 2, Biography, Additions and Corrections, Indexes. London, 1953.

Suter, Heinrich. Die Mathematiker und Astronomen der Araber und ihre Werke. Abhandlungen zur Geschichte der Mathematischen Wissenschaften mit Einschluß ihrer Anwendungen 10 (1900). Reprinted in Heinrich Suter, Beiträge zur Geschichte der Mathematik und Astronomie im Islam, Nachdruck seiner Schriften aus den Jahren 18921922, pp. 1285. Veröffentlichungen des Institutes fr Geschichte der Arabisch-Islamischen Wissenschaften, Reihe B: Nachdrucke, Abteilung Mathematik 1. Ed. Fuat Sezgin. Frankfurt, 1986.

Suter, Heinrich. Nachträge und Berichtigungen zu Die Mathematiker und Astronomen der Araber und ihre Werke. In Heinrich Suter, Beiträge zur Geschichte der Mathematik und Astronomie im Islam. Abhandlungen zur Geschichte der Mathematischen Wissenschaften mit Einschluß ihrer Anwendungen 14 (1902): 15785. Reprinted in Heinrich Suter, Beiträge zur Geschichte der Mathematik und Astronomie im Islam. Nachdruck seiner Schriften aus den Jahren 18921922, pp. 286314. Veröffentlichungen des Institutes fr Geschichte der Arabisch-Islamischen Wissenschaften, Reihe B: Nachdrucke, Abteilung Mathematik 1. Ed. Fuat Sezgin. Frankfurt, 1986. [129]

Syrie. Mmoire et civilisation. Institut du Monde Arabe. Paris, 1994.

Taylor, Eva Germaine Rimington. The South-Pointing Needle. Imago Mundi 8 (1951): 17.

Tibbetts, G. R. Arab Navigation in the Indian Ocean before the Coming of the Portuguese. London, 1971. Reprint, 1981.

Ullmann, Manfred. Die Natur- und Geheimwissenschaften im Islam. Handbuch der Orientalistik 1. Abt. VI:2. Leiden, 1972.

Varisco, Daniel Martin. Medieval Agricultural Texts from Rasulid Yemen. Manuscripts of the Middle East 4 (1989): 15054.

. Medieval Agriculture and Islamic Science. The Almanac of a Yemeni Sultan. Seattle and London, 1994.

VdDPG = Verhandlungen der Deutschen Physikalischen Gesellschaft.

Voorhoeve, P. Handlist of Arabic Manuscripts in the Library of the University of Leiden. Leiden, 1957.

Wiedemann, Eilhard. Beiträge zur Geschichte der Naturwissenschaften II. Part 3: Über Magnetismus. SPMSE 36 (1904): 32231. Reprint, IG 15:5, pp. 24655.

. Zur Geschichte des Kompasses bei den Arabern. VdDPG 9 (1907): 76473. Reprinted in Eilhard Wiedemann, Gesammelte Schriften, vol. 1, pp. 22635, and IG 15:5, pp. 25665.

. Zur Geschichte des Kompasses bei den Arabern. VdDPG 11 (1909): 26266. Reprinted in Eilhard Wiedemann, Gesammelte Schriften, vol. 1, pp. 28286, and IG 15:5, pp. 26670.

.Beiträge zur Geschichte der Naturwissenschaften XXV. Ueber Stahl und Eisen bei den muslimischen Völkern. SPMSE 43 (1911): 11431. Reprinted in Eilhard Wiedemann, Aufsätze, vol. 1, pp. 73148.

. Über den indischen Kreis. Mitteilungen zur Geschichte der Medizin und der Naturwissenschaften 10 (1912): 25255. Reprinted in Eilhard Wiedemann, Gesammelte Schriften, vol. 2, pp. 66669.

Wiedemann, Eilhard, and Fritz Hauser: Über die Uhren im Bereich der islamischen Kultur. Nova Acta, Abhandlungen der Kaiserl. Leop.‑Carol. Deutschen Akademie der Naturforscher 100:5 (1915): 1272. Reprinted in Eilhard Wiedemann, Gesammelte Schriften, vol. 3, pp. 12111482.

Wiedemann, Eilhard. Zur Geschichte des Kompasses bei den Arabern. VdDPG 19/20 (1919): 66567. Reprinted in Eilhard Wiedemann, Gesammelte Schriften, vol. 2, pp. 88385, and IG 15:5, pp. 27173.

. Magnetische Wirkung nach der Anschauung der Araber. ZP 3 (1920): 14142. Reprinted in Eilhard Wiedemann, Gesammelte Schriften, vol. 2, pp. 93132. [130]

Wiedemann, Eilhard, and Josef Frank. Beiträge zur Geschichte der Naturwissenschaften LXI. Allgemeine Betrachtungen von al Birūnī in einem Werk ber die Astrolabien. SPMSE 52/53 (19201921): 97121. Reprinted in Eilhard Wiedemann, Aufsätze, vol. 2, pp. 51640.

. Vorrichtungen zur Teilung von Kreisen und Geraden usw. nach Bīrūnī. Zeitschrift fr Instrumentenkunde 41 (1921): 22536. Reprinted in Eilhard Wiedemann, Gesammelte Schriften, vol. 2, pp. 96374.

Wiedemann, Eilhard. Zur Geschichte des Kompasses. ZP 13 (1923): 11316. Supplement in ZP 14 (1923): p. 240. Reprinted in Eilhard Wiedemann, Gesammelte Schriften, vol. 2, pp. 10371041, and IG 15:5, pp. 27478.

. Zur Geschichte des Kompasses und zu dem Inhalt eines Gefäßes in verschiedenen Abständen zum Erdmittelpunkt. ZP 24 (1924): 16668. Reprinted in Eilhard Wiedemann, Gesammelte Schriften, vol. 2, pp. 11071109, and IG 15:5, pp. 27981.

. Beiträge zur Mineralogie usw. bei den Arabern. In Studien zu Geschichte der Chemie: Festgabe fr Edmund O. von Lippmann von der Deutschen Gesellschaft fr Geschichte der Medizin und der Naturwissenschaften, ed. Julius Ruska, pp. 4854. Berlin, 1927. Reprinted in Eilhard Wiedemann, Gesammelte Schriften, vol. 2, pp. 120410.

. Aufsätze zur arabischen Wissenschaftsgeschichte. Ed. Wolfdietrich Fischer. 2 vols. Collectanea 6. Hildesheim, New York, 1970.

. Gesammelte Schriften zur arabisch-islamischen Wissenschaftsgeschichte. Ed. Fuat Sezgin et al. Veröffentlichungen des Institutes fr Geschichte der Arabisch-Islamischen Wissenschaften. Reihe B: Nachdrucke 1:13. Frankfurt am Main, 1984.

ZGAIW = Zeitschrift fr die Geschichte der arabisch-islamischen Wissenschaften.

Zinner, Ernst. Leben und Wirken des Johannes Mller von Königsberg, genannt Regiomontanus. Schriftenreihe zur bayerischen Landesgeschichte 31. Munich, 1938. Reprint, Osnabrck, 1968.

. Deutsche und niederländische astronomische Instrumente des 11.18. Jahrhunderts. 2 vols. Munich, 1956.

ZP = Zeitschrift fr Physik.


 

[131]

 

 

 

 

Fig. 1. Al-Ashrafs diagram of the compass bowl (taken from MS Cairo TR 105, fol. 145v; courtesy of the Egyptian National Library).


 

[132]

 

 

 

 

Fig. 2. Ibn Simʿūns qibla indicator (taken from MS Leiden Or. 468, fol. 190r; courtesy of the Universiteitsbibliotheek).



[1] Prof. David A. King informs me that the author is probably an Egyptian astronomer and muezzin called Ibn Simʿūn; see further King, article Ṭāsa in EI2; Suter, Mathematiker und Astronomen, 1900, p. 162, no. 398; King, Survey, 1986, p. 60, no. C24; King, On the Role of the Muezzin and the Muwaqqit, 1996, pp. 298f. (Since the present article was completed there has appeared in King, World-Maps, 1999, a facsimile of Ibn Simʿūns treatise [p. 113] as well as fol. 145v of al-Ashrafs treatise, with the diagram [p. 111].) The research for this article was supported by a grant from the Hans A. Jenemann-Stiftung (Gesellschaft Deutscher Chemiker).

[2] See Rommel, article Magnet in Pauly; Radl, Magnetstein in der Antike, 1988; Balmer, Geschichte des Erdmagnetismus, 1956, pp. 47ff.; Klaproth, Lettre, 1834, pp. 9f. Hackmann, Stradanus, 1994, p. 158, mentions a section in Pliny which he interprets as a hint at the knowledge of the directive power of the magnet.

[3] On the south-pointing chariot as a magnetic compass see, for example, Klaproth, Lettre, 1834. Mitchell, Terrestrial Magnetism and Atmospherical Electricity 37, 1932, pp. 105ff., gives a survey of authors advocating this opinion. On the correction showing the south-pointing chariot to be a mechanical device and the earliest established proof see Hashimoto, Memoirs of the Research Department of Toyo Bunko 1, 1926; Needham, Scientia 55:7, 1961, pp. 225ff.; Needham, Science and Civilisation, 19541986, 4:2, pp. 286ff. (south-pointing chariot), 3, pp. 310ff. (compass-sundials), 4:3, pp. 562ff. (use of the magnetic compass for navigation), 4:1, pp. 229ff. (magnetism, magnetic stones and their use).

[4] The magnetic needle is mentioned in his De utensilibus, and further in his De naturis rerum, probably written at the end of the twelfth century (see Sarton, Introduction, 19271948, 2, pp. 385f.; Mitchell, Terrestrial Magnetism and Atmospherical Electricity 37, 1932, p. 125, and Bromehead, Terrestrial Magnetism and Atmospherical Electricity 50, 1945, pp. 139ff.; Schck, Mitteilungen zur Geschichte der Medizin und der Naturwissenschaften 13, 1914, pp. 335f., with the two short Latin texts). On Alexander Neckam see Dchting, article Alexander Neckam in LexMA, who ignores the mentioning of the magnetic needle in Neckams treatises.

For an overview of the European sources on the magnetic compass see, for example, Mitchell, Terrestrial Magnetism and Atmospherical Electricity 37, 1932, pp. 123ff.; Schnall, article Kompaß in LexMA; Klaproth, Lettre, 1834, pp. 40f.

[5] On Petrus Peregrinus see Grant, article Peter Peregrinus in DSB, with a detailed list of manuscripts, editions and translations. On the Epistola de magnete see further Balmer, Geschichte des Erdmagnetismus, 1956, pp. 24377, and p. 627. On the use of the magnetic compass at sea and on land see Schnall, article Kompaß in LexMA; Grant, article Peter Peregrinus in DSB, who assumes the use of the floating compass also for seafaring and of the dry compass for astronomical purposes; and Taylor, Imago Mundi 8, 1951, pp. 1f., who also distinguishes between its use in navigation and astronomy.

[6] See, for example, the edition of Peregrinuss text in Rara Magnetica, 1898, pp. 36ff.; the German translation in Balmer, Geschichte des Erdmagnetismus, 1956, pp. 271ff., and the English translation in Harrandon, Terrestrial Magnetism and Atmospherical Electricity 48, 1943, pp. 3ff.; see further Schleif, Seewart 14:6, 1953, p. 25.

[7] On China see Klaproth, Lettre, 1834, p. 93, who cited San thsa thou hoei: Dans les annes Yan yeou (de 1314 1320) on voulut orienter le monastre de Yao mou ngan, et on sen [dune figure sculpte en jade, et dont la main montrait toujours le sud] servit pour en dterminer lemplacement (my additions in square brackets). Also other buildings in Peking have probably been aligned in the meridian with a magnetic compass because they deviate from the meridian by 2 degrees according to the magnetic declination (see Klaproth, Lettre, 1834, p. 70). On Europe see Nippoldt, Archiv fr Geschichte der Naturwissenschaften und der Technik 7, 1916, p. 110, who cited a treatise written in 1516 probably by Lorenz Lacher, one of the masters of the Straßburger Bauhtte: Item so du wildt ein Khor an das Hochwerkh anleg wo er stehn sol, der abmerckung, der sonen aufgang, so nimb ein Khumbast, setz den auf ein winkelmaß, und laß den magnad auf die mitdaglinie stehn, usw.

[8] On the earliest European compass-sundials see Zinner, Instrumente, 1956, p. 93; Zinner, Regiomontanus, 1938, fig. 6265; Hellmann, Meteorologische Zeitschrift 23, 1906, pp. 145ff.; Hellmann, Meteorologische Zeitschrift 25, 1908, p. 369, mentioning one from 1451, another from 1463, both with a mark of the magnetic variation. King, Ciphers, section 6.3 (forthcoming) dates a medieval French compendium with a magnetic compass to the fourteenth century, early or late, depending on the date of the astrolabe with ciphers and continues: Note the similarity of the inscriptions to those on the two Northern French astrolabes. On this instrument see further Maddison, Medieval Scientific Instruments, 1969, pp. 14ff., dating it ca. 1500. A German astrolabe fitted with a magnetic compass in the throne probably from the second half of the fifteenth century is mentioned in King and Turner, Nuncius 9:1, 1994, pp. 190, 193, and fig. 18.

[9] In his History of Africa and Spain (al-Bayān al-mughrib fī ʾkhtiṣār] akhbār mulūk al-Andalus wa-ʾl-Maghrib), written in 712H (131213), Ibn ʿIdhārī (on whom see Bosch-Vilá, article Ibn ʿIdhārī in EI2) reports of a battle in the year 239H (854). Dozy (see Dozy, Supplment, 1927, 2, pp. 337f.) found the word qaramīṭ in this context, which he translated as magnetic needle. Even if we take into consideration that Кαλαμίτα, as also in some European idioms (Italian, Greek and others) calamita, means magnetic needle (see Klaproth, Lettre, 1834, p. 13; Heß, Кαλαμίτης, p. 189), this interpretation seems to be very doubtful (see Wiedemann, VdDPG 9, 1907, p. 765). It becomes apparent in the discussion on this reference that qaramīṭ cannot be interpreted as a first hint at the magnetic compass in Islamic sources (see Plessner, article Baylak al-Qibjāqī in DSB, p. 533b; von Lippmann, QSGNM 3:1, 1932, p. 16); von Lippmann, QSGNM 3:1, 1932, p. 20, mentions a further early reference to the compass from the twelfth century by ʿUmar al-Khayyām (on whom see Youschkevitch and Rosenfeld, article al-Khayyāmī in DSB; Minorsky, article ʿOmar Khaiyām in EI1), which seems questionable to von Lippmann himself.

[10] On this author see Nizamuddin, article ʿAwfī in EI2; Storey, Persian Literature 1:2, 1953, pp. 781ff. On the passage see Wiedemann, VdDPG 9, 1907, p. 765. The confusing datesdedicated in 625H (1228), or again, dedicated to the wazīr of Iltutmish, Sultan of Delhi, Qiwām al-Dīn al-Junaydī not earlier than 628H (123031), but on an event in the year 630H (123233)originate from the fact that the one is taken from Nizamuddin, article ʿAwfī in EI2, p. 764a, and from Storey, Persian Literature 1:2, 1953, pp. 782f., while the other is from Wiedemann, VdDPG 9, 1907, p. 765.

[11] See Wiedemann, article Maghnāṭīs, 2. The Compass, in EI1 and EI2, p. 1169a; Wiedemann, VdDPG 9, 1907, pp. 765f.; von Lippmann, QSGNM 3:1, 1932, p. 20.

[12] Also Kitāb Kanz al-tujjār fī maʿrifat al-aḥjār, see Plessner, article Baylak al-Qibjāqī in DSB; edition in Klaproth, Lettre, 1834, p. 59; with translation in Clment-Mullet, Journal Asiatique (srie 6) 11, 1868, p. 174; translation in de Saussure, Lorigine, 1928, pp. 80f. (reprint Ferrand); Wiedemann, Beiträge II, 1904, pp. 330f.; Wiedemann, VdDPG 1920, 1919, p. 665. On the contents see Ullmann, Natur- und Geheimwissenschaften, 1972, p. 128. On its integration into history see von Lippmann, QSGNM 3:1, 1932, pp. 17ff. On a further reference to the Arabic sources from the mid-thirteenth century see Mitchell, Terrestrial Magnetism and Atmospherical Electricity 37, 1932, p. 119 and note 180, citing the Leyes de las partidas (Las siete partidas del Sabio Rey don Alfonso el X, Madrid 1829, 1, p. 473), a compilation of the legal knowledge of Alfonsos time.

[13] Soucek, article Milāḥa, 2. In the Later Medieval and Early Modern Periods, in EI2, p. 46a, ignores this passage when he writes: The mariners compass is first attested for 1187 [probably Alexander Neckam for the region of the Channel]. . . . The term, of Latin derivation, only marginally appeared in Arabic. . . . This term does not appear to have been current in pre-modern Arabic, a symptom of how little use the instrument received on the part of seafaring Arabs in the Mediterranean (my addition in brackets). Mitchell, Terrestrial Magnetism and Atmospherical Electricity 37, 1932, p. 119, takes into consideration that . . . it is carefully noted that he [al-Qibjāqī] said nothing as to the nationality of the vessel in which he sailed (my addition in brackets).

[14] On the different nautical traditions in the Mediterranean and the Indian Ocean see Soucek, article Milāḥa, 2. In the Later Medieval and Early Modern Periods, in EI2, p. 48b.

[15] On the author see Rosenthal, article al-Maqrīzī in EI2. In this treatise a floating compass similar to al-Qibjāqīs is described. According to Wiedemann, ZP 24, 1924, p. 166, there are two editions: Būlāq, 1270H (on the magnetic compass 1, p. 210) and Cairo, 1324H (on the magnetic compass 1, p. 337).

[16] On the author see Suter, Mathematiker und Astronomen, 1900, p. 200, no. 511; Wiedemann, ZP 13, 1923, pp. 113f.; Brockelmann, GAL, 1943, S II, p. 485; King, Survey, 1986, p. 89, no. C124.

This treatise is translated in Siggel, QSGNM 8 (194142): pp. 43557.

[17] On Shihāb al-Dīn Aḥmad ibn Mājid, the most famous Arab navigator in the Middle Ages, see Maqbul Ahmad, article Ibn Mādjid in EI2; Tibbetts, Arab Navigation, 1971, pp. 7ff. On the use of a magnetic compass for determining the qibla in one of his poems, see Ferrand, Instructions nautiques, 3, 1928, pp. 209f. On the magnetic compass in his Kitāb al-fawāʾid fī uṣūl [ʿilm] al-baḥr wa-ʾl-qawāʾid, see Tibbetts, Arab Navigation, 1971, pp. 291ff.

[18] On Sulaymān ibn Aḥmad al-Mahrī see Tibbetts, article Sulaymān al-Mahrī in EI2. On the magnetic compass in his Kitāb sharḥ Tuḥfat al-fuḥūl fī tamhīd al-uṣūl see Ferrand, Instructions nautiques, 3, p. 245; see further Tibbetts, Arab Navigation, 1981, pp. 293f.

[19] On the author see Brockelmann, GAL, 1943, 2, p. 174; partial translation in Wiedemann, VdDPG 9, 1907, pp. 766ff.

[20] See Wiedemann, VdDPG 9, 1907, pp. 767ff.

[21] See Wiedemann, VdDPG 11, 1909, pp. 264f. Unfortunately he givesas usual in his writingsonly a translation; a comparison with the Arabic terms of Ibn Simʿūns treatise discussed below would be useful and worthwhile.

[22] On the author see King, article Ibn al-Shāṭir in DSB. For a discussion of the instrument, and two medieval texts on the use of the compendium, likewise incomplete, see Janin and King, Journal for the History of Arabic Science 1:2, 1977. On the magnetic compass see especially ibid. pp. 192, 195f., 204, 209, 215ff.

[23] See, for example, Dizer, Journal for the History of Arabic Science 1, 1977, pp. 25762, and the literature there cited.

[24] See King, Imago Mundi 49, 1997, pp. 6282, a summary of King, World-Maps for Finding the Direction and Distance to Mecca: Innovation and Tradition in Islamic Science, 1999 (in press at the time of this writing).

[25] There are, for example, a Persian prayer compass, made by one Muḥammad Ṭāhir in Isfahan in the seventeenth century and inscribed with qibla values of localities as well as with instructions for its use (on the instrument see Körber, Sonnenuhren und Kompasse, 1965, pp. 98f.; on the maker see Mayer, Islamic Astrolabists, 1956, p. 78), and two cartographic plates, made in Isfahan about 1700 and fitted with compass box up in the lower middle (on the one with a now empty compass box see King, Weltkarten, 1992, 1, p. 170 with fig. 4 and 2, pp. 686ff.; King, Astronomical Instruments, 1994, p. 173 and fig. XVIa; on both see King, Imago Mundi 49, 1997, pp. 6282; King, World-Maps [in press at the time of this writing]).

[26] On the use of the magnetic compass at sea see Tibbetts, article Milāḥa, 3. In the Indian Ocean, in EI2, p. 51b, mentioning a further nautical instrument, the khashaba. On the rare appearance of the magnetic compass in navigational texts see Tibbetts, Arab navigation, 1971, p. 290. On other nautical instruments in the Islamic world see also Prinsep, Journal of the Asiatic Society of Bengal 5, 1836; Jomard, Bulletin de la Socit de Gographie 2:10, 1838.

On the sacred direction in Islam, the qibla, see King, article Ḳibla, 2. Astronomical Aspects, in EI2; King, Interdisciplinary Science Reviews 10:4, 1985, pp. 315ff.

[27] On Sultan al-Ashraf see Varisco, Almanac, 1994, pp. 12ff., and the literature cited there; see further Suter, Mathematiker und Astronomen, 1900, pp. 160f., no. 394; King, MAY, 1983, pp. 27f.; King, ZGAIW 2, 1985, p. 100. On the Rasulids see Smith, article Rasūlids in EI2, and the literature cited there.

[28] See King, ZGAIW 2, 1985, p. 113.

[29] A list is given in Varisco, Almanac, 1994, pp. 14ff.; see further King, MAY, 1983, pp. 27ff.

[30] See Mayer, Islamic Astrolabists, 1956, pp. 83f. A detailed description is given in King, ZGAIW 2, 1985, pp. 99f.

[31] On astrolabes in general see Hartner, The Principle and Use of the Astrolabe, 1968; North, The Astrolabe, 1974; Michel, Trait de lAstrolabe, 1947; Planispheric Astrolabe, 1976. On sundials in Islamic civilization see King, article Mizwala in EI2. On water-clocks see Hill, Arabic Water-Clocks, 1981; King, History of Science 15, 1977. On the magnetic compass in the Islamic world in general see Wiedemann, article Maghnāṭīs, 2. The Compass, in EI1 and EI2; King, article Ṭāsa in EI2.

Al-Ashrafs text was treated by S. Banerjee and A. I. Sabra in a paper entitled A Thirteenth-Century Magnetic Compass Described by Sultan al-Ashraf of Yemen presented at the Second International Symposium on the History of Arabic Science, Aleppo, 1979. The proceedings of that conference were never published and the paper has not been published elsewhere. The authors used photos provided by Prof. King.

[32] See King, Catalogue, 198186, 1, p. 581, and 2, pp. 362ff.; King, Survey, 1986, p. 132, no. E8; with more detailswith reference to the Tehran manuscriptKing, MAY, 1983, pp. 28f.; King, ZGAIW 2, 1985, pp. 101f.

[33] On the date see King, Catalogue, 1981, 1, p. 581; King, ZGAIW 2, 1985, p. 101, note 8. On the provenance see King, ZGAIW 2, 1985, p. 107. On the copyist see King, ZGAIW 2, 1985, p. 101.

[34] See King, Survey, 1986, p. 132.

[35] See King, ZGAIW 4, 198788, pp. 268f.; King, ZGAIW 2, 1985, p. 108; Al-ʿAzzāwī, Tārīkh ʿilm al-falak, 1959, p. 234, mentions two manuscripts in Tehran: one in the library of the Majlis al-Umma al-Īrānī and another in a private library.

[36] Part 1 (hand A): pp. 18, part 2 (hand A): pp. 1126, part 3 (hand B): pp. 1163 (see King, ZGAIW 4, 198788, p. 269).

[37] See King, ZGAIW 4, 198788, p. 268.

[38] Also manhaj, see King, ZGAIW 2, 1985, p. 108. On the contents of the manuscript see King, ZGAIW 4, 19878, p. 269.

[39] See King, MAY, 1983, p. 28; King, ZGAIW 2, 1985, p. 108.

[40] Consequently, between p. 161 and p. 162 one page of text is missing, the pagination, however, is continuous. The question is whether the pagination is on the microfilm or whether it has been added later (corresponding references in King, ZGAIW 4, 19878, p. 269).

[41] See Ahlwardt, Verzeichnis, 1893, pp. 240f.

[42] See Wiedemann, VdDPG 1920, 1919, p. 666. It is also mentioned in Wiedemann, article Maghnāṭīs, 2. The Compass, in EI1 and EI2.

[43] Al-Ashraf describes both the making of the compass bowl and its use using the term ṭāsa for a (drinking) bowl (of metal) (see Lane, Lexicon, 18631893, 5, p. 1890, sub ṭ-w-s), which is to translateas the manuscript showsas a pars pro toto with magnetic compass.

[44] The Arabic Dhikr risālat al-ṭāsa fī maʿrifat al-qibla, at first sight rather awkward, is probably best rendered in this way. The use of the word risāla apparently corresponds to the use of the word bāb for method in scientific Arabic (see King, ZGAIW 3, 1986, p. 103).

[45] The order of C here seems to be required by the sajʿ. T1 and T2 have the normal order the Merciful and Compassionate.

[46] Bayt al-ibra is one of the terms for the magnetic compass in medieval Arabic sources (see, for example, Wiedemann, article Maghnāṭīs, 2. The Compass, in EI1 and EI2, p. 1169a).

[47] See Wiedemann and Frank, Zeitschrift fr Instrumentenkunde 41, 1921, p. 226; Kunitzsch, Glossar, 1982, pp. 523f.: Die Vokalisierung des arabischen Wortes ist unsicher. In der hier eigentlich anzunehmenden Bedeutung Rand (Zaun, Einfriedung) wäre al-ḥajra (mit a) zu vokalisieren. . . . Sehr viel öfter haben die Hss. (und Drucke) dagegen die Vokalisierung mit u: al-ḥujra . . . ; al-ḥujra bedeutet jedoch eigentlich ein abgegrenztes Feld oder einen geschlossenen Raum (modern: Zimmer), und wrde demnach eher auf die von dem Außenrand umschlossene Innenfläche bzw. den Innenraum als auf diesen wulstigen Außenrand selbst zielen.; Hartner, The Principle and Use of the Astrolabe, 1968, p. 293, vocalizes ḥajra: The front [of the astrolabe] consists of a circular outer rim (Arabic: ḥajra, meaning side, or ṭawq, meaning ring, or kiffa, meaning curve; Latin: limbus or margo).

[48] Qār was used to seal up vessels and wine tubes and also to rub camels having the mange (see Lane, Lexicon, 18631893, 7, p. 2557, sub q-y-r and 7, p. 2621, sub k-f-r; shamaʿ describes also wax candles. Ullmann, Natur- und Geheimwissenschaften, 1972, p. 127, mentions that al-Qazwīnī in his cosmography attaches tar to a group of viscous substances, and so does Ḥamd Allāh Mustawfī (see Ullmann, Natur- und Geheimwissenschaften, 1972, p. 131). Tar or pitch is mentioned in several stone books (see Ullmann, Natur- und Geheimwissenschaften, 1972, pp. 95ff.).

[49] The translation of B in Wiedemann, VdDPG 1920, 1919, p. 666, begins at this point.

[50] Wiedemann, VdDPG 1920, 1919, p. 666, translates: Der beste ist der Löwe und der mit glänzendem Auge (baṣāṣ).

[51] See Ullmann, article Khāṣṣa in EI2; Ullmann, Natur- und Geheimwissenschaften, 1972, pp. 393ff. Magnetite (Fe3O4) is a black, metallic bright, non-transparent, cubic mineral. Larger pieces are natural magnets. Wiedemann, Beiträge II, 1904, p. 328, mentions a passage in al-Tifāshī also describing the magnetic stone as a black stone.

[52] Wiedemann, VdDPG 1920, 1919, p. 666, does not translate samāra, whereas in his Beiträge II, 1904, p. 330, he uses Binse, i.e. rush; Dozy, Supplment, 1927, 1, p. 682, sub samār, simār or sumār mentions several sorts of rush.

[53] Wiedemann, VdDPG 1920, 1919, p. 666, reads ḥaṣīr, and freely translates Binse, i.e., rush.

[54] Wiedemann, VdDPG 1920, 1919, p. 666, translates: Es ist in Bezug hierauf nicht abzusehen, wie ihr (der Nadel) Ende der Linie des Nordens oder der Richtung nach Sden gegenberliegt, wenn nicht eines ihrer Enden eine solche Veränderung erfahren hat, daß es nicht (auch) nach dem Norden oder nicht (auch) nach dem Sden hinstrebt. And he comments: Der Verfasser meint, daß, wenn beide Seiten beim Reiben so verändert wären, daß sie dem Norden bzw. dem Sden zustrebten, eine Einstellung in die Nord-Sd-Richtung nicht eintreten wrde; es mußte also das nicht geriebene Ende eine solche Veränderung erfahren haben, daß dies nicht geschieht. Cf. below in the commentary, pp. 1045.

[55] The translation of B in Wiedemann, VdDPG 1920, 1919, p. 666, ends here.

[56] See Koran 20:113 [114].

[57] See Sellheim, article Kitāb in EI2, p. 207b. In Welch, article al-Ḳurʾān, 4.c. The Basmala, in EI2, as well as in Gardet, article Asmāʾ in EI2, we only find the first arrangement of the adjectives (al-raḥmān al-raḥīm), not the other way round, as is found in C, perhaps transposed by reason of the rhyme. To the ninety-nine most Beautiful Names of God belongs also al-wahhāb, the constant Giver, but not al-wāhib (see Gardet, article Asmāʾ in EI2, further Flgel, Concordantiae, 1898, p. 216).

[58] So Brice, Journal of Semitic Studies 29, 1984, pp. 169ff., quoting an unpublished text of Banerjee and Sabra.

[59] The floating compass described by Petrus Peregrinus is fitted with a device like an alidade for taking magnetic bearings, but there is a little floating box supporting the alidade (see, for example, Rara Magnetica, 1898, pp. 36ff. (with figure); Grant, article Petrus Peregrinus in DSB, p. 536b).

[60] Kitāb fī ʾstīʿāb al-wujūh (al-mumkina) fī ṣanʿat al-aṣṭurlāb; partly translated and commented in Wiedemann and Frank, Beiträge LXI, 192021, p. 97121. On al-Bīrūnī, the greatest scientist of the Middle Ages, see Boilot, article al-Bīrūnī in EI2; Kennedy, article al-Bīrūnī in DSB. On this passage see Wiedemann and Frank, Zeitschrift fr Instrumentenkunde 41, 1921, pp. 22536.

[61] On iron and steel production in general see Ruska, article Ḥadīd in EI1 and EI2, also al-Hassan and Hill, article Maʿdin in EI2; Ullmann, Natur- und Geheimwissenschaften, 1972, p. 114. On the magnetic properties of steel needles see Dietrich, article Maghnāṭīs in EI2, p. 1167a; Wiedemann, Beiträge II, 1904, pp. 329f.; Wiedemann, VdDPG 11, 1909, pp. 262f. On the arguments against steel needles see Gerland, VdDPG 10, 1908, p. 384; Gerland, Mitteilungen zur Geschichte der Medizin und der Naturwissenschaften 6, 1907, pp. 15f. (in his opinion there were iron needles with a small content of steel), and on the arguments in favor of steel needles see Wiedemann, VdDPG 9, 1907, pp. 769f.; Wiedemann, VdDPG 11, 1909, pp. 265f.; Wiedemann, Beiträge XXV, 1911, p. 130. On experiments with steel nails similar to needles used by the Muslims see Wiedemann, VdDPG 11, 1909.

[62] On the knowledge of magnets and magnetism in the Islamic world in general see Dietrich, article Maghnāṭīs, 1. The Magnetite and Magnetism, in EI2, mentioning numerous legends on the magnet and its attractive power; Wiedemann, Beiträge II, 1904, pp. 322f.; Ullmann, Natur- und Geheimwissenschaften, 1972, pp. 396f. See further Wiedemann, ZP 3, 1920, p. 141, note 1: Mit der Wirkung des Magneten auf das Eisen und umgekehrt haben sich die muslimischen Gelehrten vielfach beschäftigt. Sie verglichen häufig dabei deren Anziehung mit derjenigen zwischen dem Liebenden und der Geliebten, ähnlich wie dies auch bei der Anziehung des geriebenen Bernsteins auf das Stroh geschah.

[63] Al-Ashraf could not say whether the needle would point south or north because he did not know in advance if the sharp or the blunt end would be aligned opposite the north. Mitchell, Terrestrial Magnetism and Atmospherical Electricity 37, 1932, p. 121, refers to the north-pointing needle rather in the manner of a European compass, to the south-pointing needle rather in the Chinese manner. On this topic in Europe see, for example, Taylor, Imago Mundi 8, 1951, pp. 1f.

[64] See Wiedemann, VdDPG 1920, 1919, p. 666.

[65] A statement on the magnetic variation in the Yemen during the late thirteenth century is as yet impossible. In van Bemmelen, Isogonen, 1893, the variation charts begin only in 1540. In van Bemmelen, Abweichung, 1899, the earliest values collected go back to the second half of the fifteenth century, and a magnetic variation for Aden in 1610 is given as 1240 W (p. 76). Dizer, Journal for the History of Arabic Sciences 1, 1977, p. 260, starts his values for Istanbul in 1500. Harrandon, Terrestrial Magnetism and Atmospherical Electricity 50, 1945, p. 68quoting Simon Stevinuss Portuum investigandorum ratiospecifies values for around 1600. Finsch, Geschichte der Magnetnadel, 1879 gives only values for the nineteenth century. The Egyptian astronomer ʿIzz al-Dīn al-Wafāʾī (on whom see Suter, Mathematiker und Astronomen, 1900, p. 177, no. 437; King, Survey, 1986, pp. 70ff., no. C61) determines a value for the magnetic variation for the first time in the Islamic world in the fifteenth century (Dizer, Journal for the History of Arabic Sciences 1, 1977, p. 260, gives al-Wafāʾīs value as 7 degrees east of north; see further Janin and King, Journal for the History of Arabic Sciences 1:2, 1977, p. 204, note 6; King, Lastronomie, 1994, p. 389).

[66] Using an Indian Circle, for example, the meridian is easy to determine. Cf., for example, the description in Wiedemann, Mitteilungen zur Geschichte der Medizin und Naturwissenschaften 10, 1912, pp. 252f., based on a passage in the Tafhīm of al-Bīrūnī (trans. Wright with facsimile, London 1934), or the geometrical solution given by the same author (see Kennedy, Mathematics Teacher 56, 1963, and the literature there cited.).

[67] On Europe see Mitchell, Terrestrial Magnetism and Atmospherical Electricity 42, 1937, p. 241: For at least three hundred years before Gilberts time, it had been noticed that the suspended magnet did not, always and everywhere, point to the exact geographical north. At first, this was explained as being due to the lodestone, by which the compass-needle was magnetised, having different properties in different parts; later on, it was attributed to imperfections in the method of magnetising the needle, or to errors in the observation of its direction relative to the geographical meridian.

[68] See further Wiedemann, VdDPG 1920, 1919, p. 666, note 4: Ich glaube kaum, daß hier eine Beobachtung der Deklination vorliegt, sondern daß nur der Meinung Ausdruck gegeben wird, daß die Einstellung nicht ganz genau ist.

[69] In the text al-Ashraf gives the deviation of the qibla for Aden, Taʿizz and Zabīd as 27 degrees. On the problems relating to the qibla in this treatise see, further, King, ZGAIW 3, 1986, pp. 131f.

[70] See Lane, Lexicon, 18631893, 8, p. 2845, sub n-k-b.

[71] In the anwāʾ traditions, the compass rose could be based on four cardinal winds (see Forcada, article Rīḥ in EI2; on the anwāʾ traditions in general see Pellat, article Anwāʾ in EI2).

[72] See King, Annals of the New York Academy of Sciences 385, 1982, p. 307, fig. 2; King, Journal of the American Oriental Society 104:1, 1984, p. 119, and p. 120, fig. 7.

[73] See, for example, de Saussure, Archives des sciences physiques et naturelles 5, 1923, especially p. 91 (reprint Ferrand). But al-Ashraf does not mention a compass card in his treatise. See further King, Annals of the New York Academy of Sciences 385, 1982, p. 309, relating to the root of qabūl, one of the four cardinal winds from the direction of the summer solstice, and to that of qibla, the direction towards Mecca.

[74] See King, Survey, 1986, p. 192, no. Z25; King, ZGAIW 3, 1986, pp. 130f.: the first reference is to be found in a text preserved in MS Paris B.N. ar. 2506, fols. 42r42v; the second reference purports to present the table itself, but a blank page (fol. 189v) follows.

[75] See King, ZGAIW 3, 1986, pp. 118ff.

[76] These cities are listed in a geographical table in the last part of the treatise. After Mecca, Aden, Sanʿāʾ, Taʿizz, and Zabīd follow another 40 cities. On this manuscript, Bodleian Huntington 233 (Uri 905), and its contents, see Suter, Mathematiker und Astronomen, 1900, p. 161, no. 394; King, MAY, 1983, p. 28; Varisco, Manuscripts of the Middle East 4, 1989, p. 152; Varisco, Almanac, 1994, especially pp. 16ff. On the geographical coordinates see King, ZGAIW 3, 1986, p. 132.

[77] The two values for the longitude of Mecca are based on a note in the manuscript inserted after the first value saying that several manuscripts have the more correct value 60 degrees (see, further, King, ZGAIW 3, 1986, p. 132).

[78] For an overview of early methods and tables for finding the direction to Mecca, see King, ZGAIW 3, 1986, pp. 82ff.

[79] Labels are marked as qiblat Taʿizz and qiblat ʿAdan.

[80] The modern formula is

q = arccot

with q: deviation (inḥirāf), ϕ: latitude, ϕM: latitude of Mecca, ΔL: difference of the longitudes.

[81] On the first approximative value see the table in King, ZGAIW 3, 1986, pp. 108f., and on the second, the table pp. 120f. Values with fractions of degrees are calculated by linear interpolation, for ΔL=0 is assumed q=0.

[82] C, fol. 143r.

[83] C, fol. 143r, writes latitude for the middle of the Yemen instead of longitude, as do T1, p. 158, and T2, p. 2.

[84] These are exactly the coordinates underlying al-Fārisīs tables in his Zīj al-Muẓaffarī (see Lee, Transactions of the Cambridge Philosophical Society 1, 1822, p. 260). There are further relations between al-Fārisī and al-Ashraf. Al-Fārisī was an astronomer at the Rasulid court and dedicated his zīj to al-Ashrafs father, the sultan al-Muẓaffar; and in al-Ashrafs astrological compendium al-Tabṣira fī ʿilm al-nujūm there are correspondences to al-Fārisīs folk astronomical treatise Tuḥfat al-rāghib wa-turfat al-ṭālib fī taysīr al-nayyirayn wa-ḥarakāt al-kawākib, for example, the very similar qibla schemes. See, in general, King, MAY, 1983, p. 23. On the author see especially Brockelmann, GAL, 1943, 2, p. 474, and S, 1, p. 866f.; Suter, Mathematiker und Astronomen, 1900, p. 218, note 72, and Nachträge, p. 175, no. 349 On the Zīj al-Muẓaffarī see Kennedy, Survey, 1956, p. 132, no. 54; Lee, Transactions of the Cambridge Philosophical Society 1, 1822, pp. 249ff.; Ḥibshī, Maṣādir, n.d., p. 483; on the Tuḥfa see Hawkins and King, Journal for the History of Astronomy 13, 1982, pp. 102ff.; on the manuscript [Milan, Ambrosiana X 73 sup.] see Löfgren and Traini, Catalogue, 1975, p. 142, no. CCLXX A; on the qibla scheme see King, article Makka, 4. As the Centre of the World, in EI2, p. 183b and fig. 4.

See further King, ZGAIW 3, 1986, pp. 131f.: Al-Ashraf himself in the treatise on the compass states that the qibla for the central Yemen (Aden and Taiz) is 27 degrees east of north, without giving any details about the geographical coordinates with which this value was derived, but implying that it was taken from a jadwal ʿishrīnī. However, some notes at the end of this treatise (fol. 143r) state that the qibla is 27 and a fraction east of north, and that this value is for the Yemen with L=63;30 and Mecca with L=60;0. Now, if we assume that ϕ =14;30, which is one of the values used by al-Ashraf for the Yemen, and enter in the table with arguments Δϕ =6;30 and ΔL=3;30, then using linear interpolation between the values: we obtain q=26;54.

[85] See King, Lastronomie, 1994, p. 389; King, ZGAIW 2, 1985, p. 112; King, MAY, 1983, p. 29.

[86] On the author see King, article al-Marrākushī in EI2. Al-Marrākushīs compendium of spherical astronomy and astronomical instruments Kitāb Jāmiʿ al-mabādiʾ wa-ʾl-ghāyāt fī ʿilm al-mīqāt was edited by father and son Sdillot (Sdillot, Trait des instruments astronomiques, 183435, contains the first part on spherical astronomy and sundials; Sdillot, Mmoire sur les instruments, 1844, summarizes the second part on astronomical instruments).

[87] On zījes see in general Kennedy, Survey, 1956. On the Muṣṭalaḥ zīj, see idem, p. 131, no. 47; King, Isis 74, 1983, pp. 535f.

[88] See Voorhoeve, Handlist, 1957, p. 153, who gives 282 folios; King, ZGAIW 3, 1986, p. 131.

[89] On rock crystal see Ruska [C. J. Lamm], article Billawr in EI2; see further Wiedemann, Mineralogie, 1927, p. 53, mentioning the construction of glass (zujāj) and rock crystal (billawr) vessels.

[90] The material first mentioned in the manuscript is a sort of pasteboard; qarʿ properly means pumpkin (see Ullmann, Natur- und Geheimwissenschaften, 1972, pp. 264f., translating with Cucurbit [i.e. pumpkin]; flask). This material is indeed connected with the making of a floating compass (in al-Zarkhūrīs treatise); probably it is a sort of wood.

[91] This device reminds one of the European compass-sundials (see, for example, Zinner, Instrumente, 1956, 1, pp. 92f.; Körber, Sonnenuhren und Kompasse, 1965, pp. 14ff.; Chandler, Metropolitan Museum Journal 2, 1969; Bobinger, Alt-Augsburger Kompaßmacher, 1966, pp. 26ff.).

[92] This topic appears frequently in antique and medieval Arabic and Latin sources (see Ullmann, Natur- und Geheimwissenschaften, 1972, p. 398, referring, for example, to Demokritos, ʿAlī ibn Rabbān, Jābir and al-Bīrūnī). Dietrich, article Maghnāṭīs, 1. The Magnetite and Magnetism, in EI2giving the Arabic sources (among others al-Qazwīnī and al-Dimashqī)mentions among the substances that diminish the power of the magnet the spittle of a fasting man and onions, and among those that enhance it the blood of a freshly-slaughtered he-goat. The dipping in goats blood is also found in al-Ḥusnī (see Siggel, QSGNM 8, 194142, p. 443; Wiedemann, ZP 13, 1923, p. 114). For Europe Balmer, Geschichte des Erdmagnetismus, 1956, p. 44, mentions among others Marbodaeus Gallus referring to the enhancing effect of the he-goats blood on a magnets power, and Albertus Magnus, Matthiolus and Georg Agricola citing the vitiating effect of garlic.

[93] A table with qibla values follows on the next two pages. It is merely a list of names of localities and the associated qibla values; the geographical coordinates are missing.

[94] See below on this page and the following, and note 96, on the possible meaning of the word turays.

[95] No statement is yet possible on the magnetic variation in Egypt during the middle of the fourteenth century. See further above, note 65, where some early values with references are given.

[96] My reading turays, the diminutive of turs, is based on Wiedemann, Nova acta 100:5, 1915, p. 211, who interprets turs as a part of a clockalso an instrument. Ibn Mājid in any case renders this part of the magnetic compass by qubba, or dome (see Tibbetts, Arab Navigation, 1971, p. 293). In another passage Ibn Mājid describes the use of the lodestone over the compass-box by tartīb al-maghnāṭīs ʿalā ʾl-ḥuqqa (see Tibbetts, Arab Navigation, 1971, p. 292)in Arabic the orthography of the word tartīb is rather similar to turays.

[97] On al-Zarkhūrī see above. Wiedemann renders this cone in his article on al-Zarkhūrīs text with Trichter, or funnel.

[98] A further problem is that the text does not explicitly mention a marking outside the round form, for example, on the box to orientate the magnetic needle correctly (observation for which I am indebted to Dr. Helga Dittberner, Frankfurt).

[99] See further Wiedemann, article Maghnāṭīs, 2. The Compass, in EI1 and EI2, p. 1169a. The passages, given only in translation, are not reproduced (for example, Wiedemann, VdDPG 9, 1907, p. 766, renders the term for the vessel used by ʿAwfī with the German word Teller, or plate).

[100] On Ibn al-Shāṭir and his instrument, see above.

[101] MS Berlin Ahlwardt 5845, fols. 1r2v and 4r7r (according to Janin and King, Journal for the History of Arabic Science 2:1, 1977, pp. 191, 196).

[102] On murī al-janūb, also mūrī al-janūb see Janin and King, Journal for the History of Arabic Science 2:1, 1977, p. 195, in the appendix pp. 243, 247; on the construction pp. 192, 204; on the use pp. 195f., 204, 209.

[103] Wiedemann, VdDPG 9, 1907, p. 773.

[104] There is no vocalization in the texts.